DELHI UNIVERSITY LIBRARY Cl. No. Ac. No. ■ / \ 1 This book »iooId be returned on or before the date Iasi stamped below. An overdue charge of 5 Paise will be collected for each day the book is kept overtitno Dale of release for loan SCIENCE OF THE SEA SIR jOHN MURR'^'V SCIENCE OE THE SEA AN tLEMiNTARY HANDBOOK OF PRACTICAL OCEANOGRAPHY FOR TRAVELLERS, SAILORS, AND YACHTSMEN PREPARED BY THE CHALLENGER SOCIETY tOR THE PROMOTION OF THE STUDY OF OCEANOGRAPHY ORIGINALLY EDITED BY G. HERBERT FOWLER, B.A., Ph.D. SECOND EDITION EDITED BY E. J. AiLEN, D.Sc., F.R.S. MARINE BIOLOCICAL ASSOCIATION OF THE UNITED KINGDOM PLYMOUTH OXFORD AT THE CLARENDON PRESS 1928 OXFORD OMVERSirY PRESS I ON DON VMEN HOUSE, F C 4 LDINUTJRGII GLASGOW lEIP/IG ^OUK iORONTO MELBOLXNC LAPETOWV BOMBAY CALCl TTA UADRAS SHASGHil HUMPHREY mLrORD PUBLISHER TO TlIC LMX CRSITV Printed m Great Britain Fig. I. H.M.S. 1873-6 , PREFACE TO THE FIRST EDITION Not seldom those who are engaged in oceanographic work are asked by some one about to start on ^ long voyage: ‘How can I do some work for science.? I don’t know anything about it; where can I “read it up’’.?’ Such seekers after knowledge might often do most valuable work, even in waters which they would expect to be quite well known. In this study yachtsmen, and officers of the Royal Navy, who are apt to find time heavy on their hands in port or on a foreign station, could enjoy an ever-fresh interest in the sea, its workings, and its inhabitants; even officers of the Merchant Service could find time to help, especially in meteorological and surface-water work. To all such this handbook is primarily addressed: to those who are willing to take observations, but fear that a lack of scientific training will render their work or their collec- tions worthless. Any one who follows the lines here laid vi Preface down may feel certain that his work will have real value, that he is building a sound brick into the pyramid of human knowledge. In order still further to promote the work, the Challenger Society has named an advisory committee of specialists in various branches of oceano- graphy, to whom may be referred through the Honorary Secretary inquiries upon points which are beyond the elementary scope of the handbook. But all who propose to take observations, in whatever branch, are strongly advised to spend a few days at one of the marine stations (pp. 468-74), in order thoroughly to understand the practical work. However valuable the information in this handbook may prove to be, it must remain elementary. It is no ‘royal road’ ; in no sense will it do away with the serious study which all those must undergo who, not content merely to record observations, wish also to grasp the use- fulness and meaning of what they are doing. But the book may fairly claim to have brought between two covers instru'~tions, of which many are not in print, and the rest are scattered and hard to find. The indiscriminate use by the writers of Fahrenheit and Centigrade, metres and fathoms, will prepare the reader for what he will find in the literature of the subject; an oceanographer must be able to think in either scale. The selected list of manufacturers and tradesmen (p. 486) should prove of value to the beginner. The object of the volume is severely practical — ^what to do, and how to do it — ^w'ritten in language which, as far as possible, takes no previous scientific training for granted. The headings of the various chapters sufficiently explain the general plan. With a single exception, they have been written by Fellows of the Society; to the excep- tion — Mrs. Dr. A. Weber-van Bosse, the wife of an Preface vii Honorary Fellow, and herself distinguished for researches on Algae — the Society desires to express its special in- debtedness. To those who have allowed the use of existing illustrations (notably to Sir John Murray, who placed the blocks of the Reports of H.M.S. Challenger at our dis- posal ; and to Mr. S. Henshaw, Curator of the Museum of Comparative Zoology at Cambridge, Mass., who allowed the use of blocks from ‘Three Cruises of the Blake'), as also to numerous manufacturers who have freely supplied valuable information, the Society also renders its thanks. The first edition of such a work as this is necessarily of a tentative nature, and the Editor will gratefully receive corrections and suggestions of all kinds for a future issue. G. H. F. Asplly Guisi , Bedfordshirf, A ^ nl , 1912. PREFACE TO THE SECOND EDITION In preparing the Second Edition of ‘Science of the Sea’ for the Challenger Society, the general plan of the First Edition has been followed, though several of the chapters have been entirely rewritten and others considerably re- vised to bring them abreast of modern developments. Chapter I, on the Air, has been rewritten by Mr. Brunt and Commander Garbett, the superintendents respectively of the Army and Navy Meteorological Services at the Air Ministry. Mr. D. J. Matthews of the Hydrographic De- partment of the Admiralty has contributed in Chapter II a new and up-to-date account of the physical and chemical problems of sea-water, to which are added sections on alkalinity and on the minor constituents of the water by Dr. W. R. G. Atkins. an en- viii Preface Professor Stanley Gardiner has also provided tirely new account of tropical shore collecting and of coral reefs and islands for Chapter III. Of the two sections of Chapter IV, that on floating plants, by Dr. Marie Lebour, is new, whilst that on fixed plants, by Dr. Weber-van Bosse, has been rewritten. Chapter V, on floating animals, has been revised by Mr. E. T. Browne. The late Sir John Murray’s chapter (VI) on the sea floor has been reprinted as originally written by him, with the addition of an occasional foot- note referring to later work. Chapter VIII has been revised by the authors; and Chapter IX, where necessary, by Mr. F. M. Davis and myself; Chapter X, by the original author, Mr. L. W. Byrne; and Chapter XI, on the preservation of Marine Organisms, by Mr. E. T. Browne and myself. Professor D’Arcy Thompson has largely rewritten Chapter XII, on Whales, Seals, and Sea-Serpents. The last Chapter (XIII), on Logs, Notes, and Labels, which was conjibuted by Dr. G. H. Fowler to the first edition, has been reprinted without change. Dr. L. A. Borradaile has provided a revised Classification Table of Animals, the corresponding table of Plants being reprinted as before. To all those who have contributed to this revised edi- tion of the handbook the thanks of the Challenger Society are due, as well as to Mr. F. S. Russell, who with great skill and care has assisted the editor in seeing the book through the press. E. J. A. Marine Biological Laboratory, Plymouth. May 1928. CONTENTS List of Illustrations . . . . xi List of Charts ...... xviii Introduction . . . xix By G. H. Fowler I. The Air ....... i By D. Brunt, M.A., B.Sc., Superintendent of the Army Meteorological Services, Air Ministry; and L. G. Gar- BETT, Commander R.N. (retired). Superintendent of the Navy Meteorological Services, Air Ministry. II. The Water ...... 33 By D. J. Matthews, Hydrographic Department of the Admiralty ; and W. R. G. Atkins, Sc.D., F.I.C., O.B.E., F.R.S., Director of the Department of Phy- siology at the Plymouth Laboratory of the Marine Biological Association. III. 1 'ropical Shores A. TROPICAL shore COLLECTING . -93 RESEARCH PROULEMS OF TROPICAL SHORES . 1 12 B. Coral Reefs AND Islands . . .121 By J. Stanley Gardiner, M.A., F.R.S., Professor of Zoology in the University of Cambridge. IV. The Plants FLOATING PLANTS (pHYTOPLANKTOn) AND borderland forms . . . . .150 By Marie V. Lebour, D.Sc., F.Z.S., Naturalist to the Marine Biological Association of the United Kingdom, Plymouth. FIXED PLANTS . . . . ■ By Mrs. A. Weber-vah Bosse, Ph.D. V. The Floating Animals .... 206 By G. H. Fowler and Edward T. Browne, M.A. VI. The Sea Floor ...... 241 By the late Sir John Murray, K.C.B., LL.D., F.R.S., Naturalist on the Expedition of H.M.S. Challenger, 1873-6. ConteiUs VII. Animais o>’ the Sea Floor .... 270 By G. P. Farran, B.A., Depjrlmunt of Agriculture and I'cchnical Instruction for Ireland (Fisheries Branch); and \V. T. Calman, D.Sc., F.L.S., F.Z.S., F.R.S., British Museum (Natural History) VIII. Equipment ...... 307 By Stanley VV. Kemp, M.A., Director of the ‘ Dis- covery’ ’ Expedition of the Falkland Islands Depen- dencies ; and the Editor IX. Dredging, Trawling, and other Methods of Fishing ...... 328 By E. J. Allen, D. Sc., F.R.S., Director of the Ply- mouth Laboratory, Secretary of the Marine Biological Association; and Stanley W. Kemp. X. Fishes ....... 366 By L. W. Byrne, MJV., Barrister-at-Law. XI. Preservation of Marine Organisms . . 386 By E. J. Allen, D.Sc., F.R.S., and Edward T. Browne, M.A. XII. Whales, Seals, and Sea-Serpents . . . 427 By D'Arcy Wentworth Thompson, C.B., F.R.S., ^ Professor of Zoology in the University of St. Andrews. XIII. Logs, Notes, and Labels, &c. . . 460 By G. H. Fowler. List of Marine Stations .... 468 Classification Tables . . . -475 Literature . . . .482 Classified List of Recommended Firms . . 486 Index ....... 491 LIST OF ILLUSTRATIONS Sir John Murray. By courtesy of Lady Murray . Frontispiece r. H.M.S. Challenger, 1873-6 ..... v а. Oceanic or Water Hemisphere. Reproduced, by permission of Messrs. Macmillan, from Huxley’s Pliysiography . . xxi 3. Continental or Land Hemisphere. Reproduced, by permission of Messrs. Macmillan, from Huxley’s Physiography . . xxi 4. The South Pole on the ‘Star Projection' of Steinhaus. From Krummel’s Handbuch der Ozeanographie, by permission of J. Eugelhorn’s Verlag in Stuttgart .... xxi 5. Diagram of air pressures, winds, and currents . . . xxii б. Sounding Balloon as used in meteorological work. Drawn by Com- mander F. E. B. Haselfoot, D.s.o., R.N. . . -23 7. The Nansen-Pcttcrsson Bottle . . . -59 8, g. Reversing Deep-sea Thermometer. By permission of Messrs. Negretti & Zambra . . . 6i to. Knudsen's Reversing Frame for Deep-sea Thermometers. By per- mission of the International Council for the Exploration of the Sea 64 1 1 . Reversing Water-Bottle. From Dictionary of Applied Physics, by per- mission of Messrs. Macmillan . . . . • <>S 12. Ekman Current Meter. After Gardiner . . . • 71 13. Chart of the World, showing the supposed distribution of .Wnd in the Cretaceous Period, ^fter Neumayr 123 14. Diego Garcia Atoll. By permission of Liniiean Society of London 125 15. Salomon, Chagos. By permission of Linnean Society of London . 123 16. Section of fringing reef, Rotuma Island, Pacific Ocean . . 126 17. St. Joseph, Amirante Group. By permission of Linnean Society of London ....... 127 18. A typical Atoll; Peros Banhos, Chagos Group. From the Admiralty Chart ......■■ 128 ig. Peros Banhos Atoll. By permission of the Linnean Society of London ....... 129 20. Egmont Atoll. By permission of the Linnean Society of London . 129 2 1 . Lithothamnioneae. Gonilithon frutescens and Lithophyllum Gardineri. By permission of the Linnean Society of London . ■ * 3 ^ 22. Fulanga, Fiji. By permission from Proceedings of the Cambridge Philosophical Society, vol. ix . . . ■ -133 23. Kambara, Fiji. By permission from Proceedings of the Cambridge Philosophical Society, wo\, \\ . . . • • ^33 XI 1 List of Illustrations 24 Sectinml dngT im tn illustnte formition of Atoll ind Barrier Rctf by sub^idtriCL Aftei Dirwm t Coir;/ IJ7 25 Viriua Mb lilt 11 Fiii lit pu iniiiioii fit ni P>ocl ilmt’s of t/u Cum biuigt Phtlosopfiual Son ty, lol i\ ^39 26 Sections to shott form Uion of Rei-fs by \b ision of Land 142 •’7 Sectional diagrim to illustnte formition of an Atoll on the site of in isl nd of loos" mitcriil which has been partly washed away 143 "S Sectionil diigrim to illustritc formation of in Atoll on 1 Mound of the Sei flooi i 4 S ->9 Dil\ s Sections to illustrate the deielopment of Birriei Reefs ind Atolls according to the glaciileontrol 'heorj From A G Mayor Memoiial Volume, tol xi\, bv permission of the Carnegie Insti tute of ^^ashlngton 147 30 Pmnuliria Viridis From Muiiaj, Intiotu Uon to the Stue/y of Sea Cl is, bi permission of Messrs Macmillan 153 3 1 Planktonielli 154 3'’ Skelelomci 1 From Schutt, PflansscnLben da Hochsee, b) permis- sion of Messrs Lipsius ind 7 ischer 155 33 Chietoceros fnm Ph mouth Sound 155 34 Bicteiiastrum 155 3^ Tjpes of Gentium Fiom Schutt, Pflantoenleben det Hochsee, by permissi m of Messrs I ipsius and 1 ischer iy6 36 Peiidinium, from Plymouth Sound ij7 37 Ornithocercus From Schutt, Pflanzcnleben der Hochsee, by per mission of Me srs Lipsius and Tischer 159 33 a, b c, PlucoCjStis, d, Dictiocha, e, Distephanus, f tuglenoid I rom Ph mouth Sound 161 39 Hilrsphieri From Murra), Iniioduction to the Study of Seasweeds, bj permission of Messrs Macmillan 163 40 Rhibdosphaera {Blale) 163 41 Coccosphiera {Blale) 163 4"’ Trichodesiiiium {Blah) 163 43 Chaetomorphi lyg 44 Cladophora After Kutzing 17S 45 Ulva (Young) 178 46 Calothnx. After Bomet and Thuret 179 47 Halimcda macroloba 179 48 Hahmeda {Blake) 180 49 Udothea {Blale) 18 1 List oj Illustrations 50. Cymopolia and Acctabularia xiii rS3 51. Bornetella ..... 183 52. Caulerpa racemosa .... 183 53. Caulerpa prolifera .... 1S3 54. Macrocystis. After Hooker aud Harvey . 183 55. Saccorhiza ..... 184 56. Lessonia. After Bory de St. Vincent 185 57 - Egregia ...... 185 58. Ecklonia ..... 187 59. Nereocystis. After Setchell 187 60. Dictyota dichotoma. After Bornet and Thuret . 1S9 61. Section through Frond of Fucus. After Bornet and Thuret 189 62. Sargassum ..... 189 63. Himanthalia. After Murray igt 64. Durvillaea. After Lemmermann . 191 65. Hormosira and Notheia .... 191 66 . Sargassum [Blake) .... 193 67. Polysiphonia . ... 19s 68. Lithophyllum ..... 197 69. Archaeolithothamnium .... 197 70. Claudea ...... • • 197 71. Rhodymenia ..... i ' 197 72. Zinc box for carrying bottles and plan of stand for loose bottles . 200 73. Press for drying Algae .... 205 74. Globigerina [Challenger) .... zoS 75. Heliosphaera. After Hertwig 209 76. Conchidium. After Borgert 209 77. Stylospira. After Hertwig 209 78. Eucyrtidium. From Lankester’s Zoology, by permission Black ...... of Messrs. 209 79. Acanthometra. After Hertwig . 209 80. Coelodendrum. After Borgert 211 81. Tuscarora ..... 2tl 82. Euphysetta. After Borgert 211 83. Noctiluca ..... 2II 84. Two species of Challengcron . After Haeckel 2ri XIV List of Illustrations 85. Phialidium. After Browne ..... 86. Euphysa. After Browne ..... 213 87. Colony of Bougainvillea attached to a floating bit of wood. From Lankester's Treatise on Zoology, by permission of Messrs. Black . 213 88. Obelia. From Lankester’s Treatise on Zoology, by permission of Messrs. Black ....... 89. Pan of Colony of the Hydroid Bougainvillea. From Lankester's Treatise on Zoology, by permission of Messrs. Black 90. Rhizostoma. After Milne Edwards .... 91. Pelagia. After Forbes ...... 92. Ephyra, or Young Aurelia. After Dclap . 215 93. Velella (Blaie) ....... 216 94. Physalia {Blake) ....... 217 95. Diphyes ....... 217 96. Porpita {Blake) . . . . . • zt? 97. Arachnactis . . . . , .217 98. Cestus or Venus’s Girdle. After Chun . . . .218 99. Pleurobrachia. After Chun ..... 219 too. Tomopteris. From The Cambridge Natural History, by permission of Messrs. Macmillan . . . . .219 101. Beroe . . . -2*9 102. Typesfof Chactopod Larvae ..... 219 103. Spadella ... ... 219 104. Sagitta ........ 219 105. Evadne ........ 221 106. Conchoecia. Left shell and Antenna 221 107. Conchoecia. Shell removed ..... 221 108. Caligus ..... , 221 109. Calanus. After Claus .... . 221 no. Temora. After Claus .... 221 in. Sapphirina . . .... 221 112. Euthemisto. After Sars ...... 221 113. Phronima ....... 223 1 14. Cirolana. After Stebbing 223 115. Lucifer. After Claus ...... 223 116. Nyctiphanes. From Lankester’s Treatise on Zoology, by permission of Messrs. Black ...... 222 List of Illustrations XV 1 17- Scrgestes Larva. After Claus 223 ii8. Phyllosoma. After Gardiner 223 1 19. Nauplius Larva of a Barnacle. After Claus 223 120. Nauplius Larva of Penacus. After Claus . “5 121. ZoCa Larva of Penaeus. After Claus 225 122. Zoea of Callianassa, After Claus 223 123. ZocaofThia. After Claus 223 124. Megalopa Larva of a Crab. After Claus . 223 125. Early Larva of a Squilla. After Claus 2^3 126. Late Larva of a Squilla. After Brooks 223 127. Oyster Larva. After Hoek 227 128. Atlanta ...... 227 129. Carinaria ..... 227 130. Veliger Larvae ..... 227 13 1. Larva of a Polyzoan .... 227 132. Clione ...... 227 133. Shelled Pteropods .... 227 134. Halobates {Blake) .... . 227 135-142. Auricularia Larva of a Hoiothurian; the same later, the young Trepang showing; Pluteus Larva of an Echinoid; the same later, the young Sea-Urchin prominent; Bipinnaria Lai-va of an Aste- roid; the same later, carrying the young Starfish; Pluteus Lajva of an Ophiuroid; the same later, with the young Brittle-Star. Modified from Johannes MQllcr .... 227 143. Appendicularia. After Lohmann 229 144. Pyrosoma {Blake) .... 229 145. Doliolum. The nurse form which buds 229 146. Doliolum. The spoon form 229 147. Doliolum. The wandering form . 229 148. Doliolum. The sexual form 229 149. Salpa. The sexual form .... 229 150. Salpa. The budding form 229 151. Baillie Sounding Machine as used on board H.M.S. Challenger 241 152. Set of Sieves for grading the contents of Dredge or material gathered on shore {Challenger) Trawl, or 243 133. Scalpellum on a Manganese Nodule (flhalleyiger) . 258 154. Spherule with metallic nucleus coated with black Magnetite {Chal- lenge) ....... 239 xvi List of Illustrations 155. Diatom Oi)7C. After Chun, from Kriimmel''! Orrnnngrnphk, by per- mission of J. En^elhorn's Verlag in Stuttgart . . . 260 156. Dead shells of Globigcrina (C4«2/&nge»‘) .... 261 157. Globigerina Ooze, consisting largely of Foraminiferan shells (Blake) 261 158. Pteropod Ooze, consisting mainly of Pteropod and Foraminiferan shells (Blake) ....... 262 159. Nodosaria, drawn by Messrs. Heron- Allen and Earland 275 160. Spiroloculina ,, „ ,, „ . . 275 161. Polystomella ,, ,. „ „ . . 275 162. Astrorhiza „ ,. „ „ . . 275 163. Mlliolina „ „ „ „ . . 275 164. Holtenia, a Deep-sea Glass Sponge (Challenger) . . 277 165. Cladocarpus, a Hydroid (Blake) ..... 279 166. Pennatula, a Sea-Pen (Blake) ..... 281 167. Gorgonia, a Sea-Fan. From Lankester's Treatise on Zoology, by per- mission of Messrs. Black . . . . .281 168. Gorgonia, showing the Polyps in contraction and expansion. From Lankester’s Treatise on Zoology, by permission of Messrs. Black . 281 1(39. Loptoponus, a Deep-sea Coral (Challenges) . . . 283 170. Ophiura, a Brittle Star (CW/enger) .... 285 171. Archaster, a Starfish (Blake) ..... 287 172. Serpula, a Tube-dwelling Bristle-worm. From The Cambridge Az/araf by permission of Messrs. Macmillan . . 287 173. Salenia, a Sea-Urchin (Challenger) .... 289 174. Hyalinoecia, a Bristle-Worm (Blake) . . . . . 289 175. Amphinomc, a Bristle-Worm (Blake) . . . 289 176. A Sea-Cucumber or Trepang (Challenger) . . 291 177. Rhizocrinus, a Feather-Star (Blake) .... 293 178. Willemoesia, a Deep-sea Lobster (Challenger) . . ■ ^9 i 179. Epimeria, an Amphipod (Blake) ..... 297 iSo. Caprella, a Ghost Shrimp. After WyviUc Thomson . . 297 181. Catapagurus, a Hermit-crab (Blake) .... 297 182. Rorinela, an Isopod (Blake) ..... 297 183. Eolis, A Sea-slug. From The Cambridge Natural History, by per- mission of Messrs. Macmillan ..... 297 184. Munida, a Galatheid (Blake) ..... 299 185. Chiton. From The Cambridge Natural History, by permission of Messrs. Macmillan ...... 299 185a. Terebratula, a Lamp-shell (Challenger). Enlarged . . 299 List oj Illustrations xvii 186. Eledone (Blaie) ....... 301 187. Benthoteuthis, a Deep-sea Squid (Jilake) .... 301 188. Cellularia, one of the Polyaoa {Blake) .... 301 189. Cellularia, magnified {Blake) ..... 301 190. Steam Capstan for use with Sailing Yachts. By permission of Messrs. Brook & Co. ...... 309 191. Commercial Trawling Winch. By permission of Messrs. Good & Menzies . . . . . . .311 192. Section of wire rope. By permission of Messrs. Bullivant . 312 193. Nipper for heavy trawling. By permission of Messrs. Bullivant . 315 194. Small Nipper Rendering. By permission of Messrs. Bullivant 315 195. Small Nipper Gripping. By permission of Messrs. Bullivant 315 196. Sounding-Leads, with Snapper for collecting bottom sample . 318 197. Lucas Sounding Machine. From Krummel’s Ozeono- graphie, by permission of J. Engelhorn's Verlag in Stuttgart . 32 1 198. Ball Dredge. After Wyville Thomson .... 329 199. Beam Trawl. After Holdsworth ..... 335 200. Agassiz Trawl, Blake pattern ..... 340 20:. Traces ........ 349 202. Long Line, as used by Portuguese Fishermen . . 333 203. Hand Net Frame ...... 353 204. Frame of Hand Net for use among Rocks 353 205. Ring Nets . . . . . . 334 206. Part of a Train of Drift Nets ..... 336 207. Seine Net ....... 339 208. End of a Seine Net ...... 359 209. Petersen’s Yngel-Trawl or Young-Fish Net . 363 210. Collapsible Pot ....... 364 21 1. Descriptive Topography of a Fish .... 369 212. Descriptive Measurements of a Fish .... 371 213. Horny Egg Capsule of (a) Chimaera, {b) Scyllium . 377 214. Demersal Ova of a Shore Fish (Gobius minutus) in an empty shell 377 213. Pelagic Eggs, Demersal Egg of Goby, and Demersal Egg of Spotted Sucker. After Holt ...... 379 216. Pelagic Larva of a Sole. After Holt .... 381 217. The Greenland Whale. Vrom Fauna Arctica . . 426 218. The Bottlenosc ....... 426 219. The White Whale, or Beluga ..... 426 220. The Narwhal . . . . ... . 426 3166 b LIST OF CHARTS I. Mean Pressure for January .... page 9 II. Mean Pressure for July . . . . . „ 1 1 III. Mean Temperature for January . . . ■ j. 13 IV. Mean Temperature for July . . - » 15 V. Synoptic Weather Chart . . . • » 25 VI. Mean Surface Isotherms . . . . ■ >» 37 VII. Mean Surface Currents . . . . „ 39 VIII. Mean Surface isohalines, showing the weight of salt in parts per thousand by weight. After Gerhard Schott . . „ 41 IX. Depths. After Murray . . bei’vceen pages 24S-9 X. Bottom Deposits .... i> „ 266-7 XI. Areas of Marine Distribution . . » >, 3°4~5 Introduction xxiii Between these extremes of scale — between the gigantic and world-embracing movements of Air and Water, and the brief life of the minutest organism — lie, intricately interwoven, the problems with which the student of oceanography must wrestle. But w'hile his proper work is the orderly marshalling of facts into groups, as a basis for explanations — or at* least for hypotheses — ^the facts themselves can be gathered in quantity by any traveller who chooses to take the trouble of doing so. I THE AIR By D. Brunt and L. G. Garbett The atmosphere in which the earth is embedded is a mix- ture of gases, the main constituents being nitrogen (78 per cent.) and oxygen (21 per cent.). There are other constituents present in small quantities, but except for water vapour and possibly for ozone, they are not of importance in the study of the motions of the atmosphere. It is a relatively rare thing to observe air to be at rest for any considerable period. Its normal condition is one of motion, and air in motion is wind. Over the surface of land the motion of air is complicated by the effects of contour and of the general roughness of the ground, which give rise to swirls and eddies in the same way that a rock in a stream does. Over the sea the motion of air is not complicated by contour, though it is likely that waves on the surface of the sea have a similar effect to that'of the general roughness of ground in level country. It is cer- tain, however, that over the sea the motion of air is of an essentially simpler nature than irs motion over land. There is an additional reason for this apart from the lack of contour. The surface of the sea is a uniform flat sur- face, having everywhere the same capacity to absorb, reflect, and radiate heat, whereas on land the power of absorption, radiation, or reflection differs for different types of surfaces. The power to absorb the heat rays of the sun differs very considerably for desert sand, grassland, bare soil, woodland, &c., and these inequalities give rise to considerable complexities in the motion of air over the earth. The motion of the air over the sea is free from such complications, and it might appear at first sight that at sea 316b B 2 The Air the meteorological observer should find the most favour- able conditions for the study of the air and its ways. In practice, however, this is not the case, for the simple reason that instrumental observations of all kinds are far more difficult to carryout on board ship than on land. The movement of the ship makes the observation of the wind with extreme accuracy practically impossible. Allowance has to be made for the course and speed of the ship, and, if the wind is measured by means of an anemometer, the rolling and pitching of the ship introduces errors into the records. The measurement of temperature is also trouble- some, as care has to be taken that the air which reaches the thermometers has not come into earlier contact with any part of the ship. In general it can be stated quite de- finitely that the greater apparent simplicity of the problem of meteorology over the sea is more than counter-balanced by the difficulties in the way of accurate observation. It may not be an exaggeration to say that civilization began when man began to take note of the weather, to observe the regular changes of the seasons, and to arrange his life in accordance with these. The men of the open air have in all ages been forced to take special note of the weather and its changes. The sailor and the farmer have been habitually weatherwise. But so long as the weather remained an individual study the observer’s weather knowledge was limited by his own horizon. He could only take note of what he actually saw. The invention of the barometer led to the formulation of a number of rules for forecasting the weather, but real progress in meteoro- logy came with the introduction of the synoptic chart about the middle of the nineteenth century. The develop- ment of telegraphy made the synoptic chart a practical possibility, and it was almost immediately recognized that meteorology should be developed from an international Properties oj Air 3 standpoint, 'i'he progress made was rapid, and the chief features of the weather chart as we know it to-day had become established by about 1870. At the same time it was recognized that charts showing the conditions over the oceans would confer enormous benefits on the seafaring interests of the world. At the instigation of Admiral (then Lieutenant) Matthew Fon- taine Maury, U.S.N., an International Conference was held at Brussels in 1853, and a plan of observations to be followed on board ship was recommended. This plan was adopted by practically all the seafaring nations of the world. Maury was the author of a well-known book, ‘Physical Geography of the Sea’, a book which is still well worth reading. It is a book which is thoroughly imbued with the spirit of scientific research, and marked by the boundless enthusiasm of its author, and though some of its facts are out of date, and the theoretical treatment appears fantastic in the light of the knowledge acquired since its appearance, it remains a model for writers of popular science. Some Properties of Atmospheric Air It will be well, before we try to describe the motions of air over the globe, to fix in our minds some of the physical properties of atmospheric air. Dry air, i.e. air from which water vapour has been removed, no matter where it is sampled, except near factory chimneys or other sources of pollution, is made up of the same gases in the same pro- portion. Normally, however, air contains a quantity of water vapour, and the amount of water vapour changes from place to place, and from time to time in any one place, and it is no exaggeration to say that the larger part of meteorology is concerned with the study of the vagaries of the water vapour in the atmosphere. 4 The Air The amount of water vapour which a sample of air can hold depends solely on the temperature of the air, and is independent of the pressure. Now water vapour is lighter than dry air, in the ratio 5 to 8. When water vapour is introduced into air, it displaces an equivalent amount of dry air, so that the density of the resulting damp air is less than that of dry air at the same temperature and pressure. Damp air then is lighter than dry air, just as warm air is lighter than cold air. Air is relatively very light com- pared with water. (Air at ordinary pressures and temperatures has a density about gj^th of the density of water.) The weight of air above a square inch of ground is about 14 lb. This is equal to the weight of a column of mercury a square inch in cross-section and about 30 inches high. We thus commonly speak of normal atmospheric pressure as being roughly equivalent to 14 lb. per square inch, or to 30 inches of mercury. The measure- ment of pressure by means of a mercury barometer simply consists in balancing the pressure of the air against the weight of a column of mercury, and measuring the length of the column. This is true whether we express the pres- sure in inches, millimetres, or millibars. The pressure of air falls off with increasing height, and at 1 8,000 feet it is only half the pressure at the ground, while at 36,500 feet it is only about one quarter the pressure at the ground. This means that of the total mass of the atmosphere one half is below the level of 18,000 feet, and three-quarters is below the level of 36,500 feet. These figures are average values. Where the temperature is high, and the air therefore less dense than the average, the rate of decrease of pressure with height is less than the average. If a portion of air is heated to a higher temperature than the surrounding air, it will become lighter than the Tra7isjer of Heat 5 surrounding air, and will tend to rise, its place being taken by cooler air. The rising air carries its excess of heat with it. This method of transferring heat from one place to another by the bodily transfer of air is called the convec- tion of heat, and rising currents produced in this way are called convection currents. Convection .is not the only process by which heat can be transferred by the air. When the earth is heated by the sun, it radiates upward some of the heat it receives and this radiation is to some extent absorbed by the water vapour in the air. The air in turn can radiate out in all directions at least a' portion of this heat. Only to a minute extent does air transfer heat from one place to another by ordinary conduction, in the sense in which heat is con- ducted from one end of a hot poker to the other end. This type of transfer is carried out by molecules handing on some of their energy to their neighbours, and in a gas the distance between neighbouring molecules is too great to allow of this being done to any considerable extent. The earth’s atmosphere can be regarded as a colossal machine, whose motive power is derived from solar radia- tion. The complexity of the motion of the air over dif- ferent regions of the earth is due in the first place to the variation with latitude of the amount of solar radiation received at the ground, and in the second place to the differences in the capacity for absorbing or radiating heat possessed by different portions of the earth’s surface. Again, the amount of heat necessary to raise the tem- perature of a given mass of water through one degree is from four to five times the amount of heat required to raise the same mass of dry soil through one degree. This is one of the reasons why changes of temperature over the sea are so much less than those over land. The differences between land and sea are so funda- 6 The Air mental and so far-reaching in their effects that it is desirable to form a clear idea of their causes. When the sun is low, its rays are almost entirely reflected from the surface of the ocean, and so sent back into space without having any heating effect upon the sea. When the sun is high, its rays penetrate into the water to a considerable depth. The heating effect is thus spread through a large mass of water, and this, added to the fact that a given amount of radiation produces a relatively small heating effect in water, accounts for the absence of any consider- able heating of the sea surface during the day. At night the sea surface radiates heat quite readily, but when the surface layer is cooled it tends to descend, and so spreads the cooling effect through a considerable depth, with the result that the actual surface temperature of the sea does not show any marked drop during the night. Thus the temperature of the sea surface should show no marked diurnal variation such as is shown by a land surface. There must naturally be a small diurnal varia- tion, but it is too slight to be readily measured. In the same way, the differences of sea temperatures in summer and winter are relatively slight by comparison with those of land. The temperature of the sea then is far less respon- sive to solar radiation than the temperature of land. It is for this reason that oceanic climates are in general less subject to violent extremes of temperature than continental climates. It must not be overlooked, however, that a continental climate may at times, and in places, extend out to sea. For example, in winter the winds blowing out of the Siberian, anticyclone across the China Seas are intensely cold. In the British Isles, on the other hand, the prevailing winds are oceanic winds in summer and winter. Hence the cool summers and mild winters of the British Isles, by com- Isobars and Isotherms 7 parison with the more extreme climates of the middle of a continent. When the air over a large area is heated from below, it expands, and produces a lower pressure over that area than over surrounding regions. The inverse effect holds when the air is cooled over a large area. The explanation of these phenomena is not quite so simple as to justify its inclusion here. The effect, however, is to tend to produce anticyclones over the continents in winter, and over the oceans in summer. In Charts I, II, III, and IV are given the mean isobars (lines drawn through points of similar atmospheric pres- sure) and isotherms (lines drawn through points of similar temperature) for January and July. A very cursory in- spection of these maps shows that in middle and high latitudes there is a definite association of anticyclones with continents in winter, and with oceans in summer, and of low pressure with oceans in winter, and continents in summer. The most striking individual effect is the forma- tion of a very large and intense anticyclone over North Eastern Asia in winter, as shown in Chart I, in a position corresponding very closely to the pole of the most intense cold shown in Chart III. There is another centre of very low temperature in winter over Greenland, but this does not produce so marked a pressure effect as the Siberian centre, on account of the nearness of the North Atlantic. The "Fundamental Meteorological Observations The features of the weather which are ordinarily observed are: (1) The state of the sky, whether clear or cloudy, and whether rain is falling. (2) The temperature of dry and wet bulb thermo- meters. From these two readings, by the use 8 The Air of tables, we can readily estimate the relative humidity, which expresses as a percentage the ratio of the amount of water vapour actually in the atmosphere, to the maximum amount of water vapour that air of that temperature can hold in the form of true v'apour. (3) Pressure or height of the barometer. (4) The wind direction and force. (5) The extent of horizontal visibility, and the pres- ence of mist or fog. These observations are all of a simple kind and require no special skill. Instruments The meteorological instruments usually found on board ship are as follows: Mercurial Barometer, Aneroid, Barograph, Thermometers; other instruments occasion- ally used at sea are Thermograph, Psychrometer, Rain- gauge, and Anemometer. I. The most important of these instruments is un- doubtedly the mercurial barometer which measures pres- sure. It is mounted on gimbals, so that it can remain vertical even when the ship is rolling. It must be placed in a position which will allow it to do this and where there is an equable temperature, free from draughts, in a good light and at a convenient height so that the eye can be brought easily to the level of the top of the mercury. The chart house is often the most convenient place for the barometer, but this is not always the case owing to its exposed position and its distance from the centre of gravity of the ship. The ‘pumping’ of the mercury caused by the movement of the ship in a seaway is reduced in a Marine Barometer by a constriction of the tube at the cistern end. The rise and fall of the ship in a seaway lo The Air produces a small irregularit}’-, and when it is remembered that roughly speaking the barometer falls one thousandth of an inch for every foot of increase of height, it is plain that the vertical rise and fall of the ship in rough weather may affect the readings by several hundredths of an inch. It is necessary therefore to take the mean of the highest and lowest readings when the ship is in the trough and on the crest of successive waves. It must be remembered that a high degree of accuracy is required in the reading of the barometer when the reading is to be used for plotting on a Synoptic Chart. Barometers are graduated in inches or millibars, some- times with both scales. The reading must be reduced to sea level and corrected for temperature and latitude if it is to be plotted on a synoptic chart or is to be transmitted by wireless telegraphy for the information of other ships. (For correction of barometer see ‘Barometer Manual’.) An ingenious device, known as the ‘Gold Slide’, for autonjatically correcting the Barometer, is fitted to some barometers and saves reference to the correction tables. 2. The aneroid also measures pressure and is a par- ticularly useful instrument in ships as it can often be placed in a position immediately under the supervision of the Officer of the Watch. It is in many ways a more convenient instrument than the barometer but is not so reliable in measuring absolute pressures. The aneroid should be frequently compared with the barometer. The readings do not require correction for temperature or latitude but only for index error and height above sea level. 3. The barograph is a self-registering form of aneroid and is a valuable supplement to the barometer, since baro- grams show minor fluctuations of atmospheric pressure which are seldom noticeable in the action of mercurial Chart II 12 The Air barometers. It should be carried in a cradle, sling, or spring suspension bracket and placed in the position least likely to be affected by vibration or movements of the ship. 4. Thermometers measure temperature. They should be of standard type, tested at Kew or some other national testing establishment, and supplied with a certificate of accuracy. This condition is insisted on for all instruments supplied by the Meteorological Office. The wet and dry bulb thermometers must be exposed in a louvred screen, quite in the open, free from all arti- ficial heat, and protected from sea-spray and from all risk of injury. The best exposure is usually to be found on the weather side of the bridge. The bulb of the dry thermometer must be kept as free from moisture as pos- sible. The bulb of the wet bulb thermometer is kept moist by means of a covering of fine muslin to which is attached a piece of cotton wick whose lower end dips into a srnall vessel containing water. This water must be changed frequently to prevent any crustation of salt. A weekly change of muslin is also advisable and the use of distilled water is recommended when accurate results are to be aimed at. For taking the surface temperature and specific gravity of the sea vide p. 54. 5. The thermograph is a self-recording thermometer and gives a continuous record of temperature. It should be exposed on deck in a louvred screen and its readings should be compared with the reading of the dry bulb thermometer at each time of observation. 6. The psychrometer measures the temperature and humidity of the air. Assmann’s aspiration psychrometer consists of a wet and dry bulb thermometer placed side by side in metal tubes over which air is drawn by a quick- 14 The Air running fan actuated by clockwork. The Whirling Psy- chrometer consists of a wet and dry bulb mounted side by side on a boxwood frame. A spindle and handle are pro- vided for whirling the instrument in the air. Observa- tions with the psychrometer should be taken on the wind- ward side of the deck over the ship’s side. 7. The raingauge for use at sea is attached to an iron frame and is hoisted to a suitable heieht above the deck where it will be clear of the sheltering effects of deck houses, &c., and exempt from spray. Further observa- tions of rainfall at sea are highly desirable as at present there are only a few observations available for comparison with the rainfall over the land. A number of ships have, however, in recent years been equipped with raingauges. 8. The anemometer registers the relative force of the wind. An electric cup anemometer is fitted in some ships. The cups turn a contact maker which completes the cir- cuit and rings a bell once for every 25 turns of the cups. The interval between two rings corresponds with a wind run of 139 ft. The exposure of the anemometer is a' very important point to be considered; when possible the fore- mast head is the most suitable position, the recording portion of the instrument being on the bridge. When, as is usually the case, there is no anemometer available the direction and force of the wind on the Beaufort Scale is obtained by observing the appearance of the sea and the effect of the wind on the surroundings generally. When there is a following wind and the smoke rises vertically, the speed of the ship gives the velocity of the wind. With winds in other directions the course and speed of the ship must be taken into account unless the estimation is made by observing the effect of the wind on the waves. Chart IV ( i6) The Circulation oj the Atmosphere The largest scale motion of the atmosphere is its rota- tion with the earth. At the equator, for example, the velocity due to rotation of any point on the earth is about 3 miles per second or over i,ooo miles per hour. In any other latitude the velocity due to rotation decreases in pro- portion to the cosine of the latitude, i.e. in proportion to the distance from the earth’s axis. Thus the motion of the atmosphere due to its rotation with the earth is very much greater than the deviations from this value. What we generally call still air is air that is moving around with the same speed as the solid earth below it, and it is the devia- tion of the motion of air from the motion of the ground beneath it that we observe as wind. We must not overlook this ‘solid rotation’ of the atmosphere with the earth. It is a motion which repre- sents an enormous store of energy, and it is possible for a restricted region to draw upon this store of energy in the formation of a cyclone. In general, however, when we speak of air in motion we shall refer only to that motion relative to the earth which is known as wind. We next have to consider the question ‘Do the winds of the globe at any given time conform to some general scheme, at least so far as their main features are con- cerned i ’. In other words, is there a general circulation of the atmosphere regarded as a whole.^ We can most readily get at the answer to this question by considering for a moment the relation betw'een wind and the distribution of pressure. This can be stated in the form known as Buys Ballot’s law, w'hich may be inter- preted as follows : An observer who stands with his back to the wind has lower pressure to his left than to his right, or in other words an observer who faces into the Pressure Distribution 17 wind has lower pressure to his right than to his left. The advantage of adopting this rule at once is that we can reduce the question of the circulation of winds to that of patterns of pressure distributions. The Charts I and II show the mean pressure distribu- tions for January and July. It will be seen that in both these months there is at the equator a belt of relatively low pressure, and calm or light winds. There is a belt of calm around the greater part of the equator, known as the ‘Doldrums’, whose mean position is between 5° and 10° N. It swings to north and south of this mean position during the year, tending to follow the sun to north in the northern summer and to south in the northern winter. Beyond the belt of calms, the wind blows from an easterly direction, with a component of motion into the belt of low pressure, so that the actual winds are north- east in the northern hemisphere, and south-east in the southern hemisphere. These winds circulate around the equatorial side of belts of high pressure (the sub-tropical highs), whose mean position in the southern hemisphere is centred about latitude In the central region of the belts of high pressure winds are light, and these regions are known to sailors as the ‘Horse Latitudes’. The wind circulation in the southern hemisphere is fairly regular and retains the same main features throughout the year. There the southerly side of the belt of high pres- sure is characterized by westerly winds, which cover a wide belt in which pressure decreases southward. In the northern hemisphere, however, the distribution of land affectsveryconsiderably theprevalling conditions. Wehave already referred to the influence of land and sea, and the tendency of high pressures to develop over the oceans in summer and over the continents in winter. We see from the January chart of mean pressure that while In the 1 8 The Air northern hemisphere a narrow belt of relatively high pressure persists over the ocean, over North America there is an extensive region of high pressure, and over Asia there is a very large anticyclone which extends down into, and merges with, the sub-tropical high pressure. To the north of this arrangement of antic}’^c]ones and high pressure belts is a region of westerly winds and diminishing pres- sure, with well-marked low pressures over the North Pacific and North Atlantic. The low pressure region shown south of Greenland extends in a long trough across to the north of Scandinavia. North of this region pressure again increases and winds are easterly. The above description of the mean pressure distribu- tion and of the corresponding wind distribution gives a reasonably good picture of the day to day situation so far as the southern hemisphere is concerned, and also for low latitudes in the northern hemisphere. But over the North Atlantic the day to day variations are very considerable. This is a region of travelling cyclones and anticyclones at all times of the year. The cyclones of middle latitudes are usually centred near the boundary between the warm westerly current originating in low latitudes and the cold easterly current of polar latitudes. During the northern summer the continental anti- cyclones of winter are replaced by vast regions of relatively low pressure, while high pressures develop over the oceans. The Atlantic anticyclone is on the average centred near the Azores, but it is subject to large oscillations of posi- tion. The Asiatic low pressure extends down over India to the Equator and over the eastern portions of Africa. Around the southern fringe of this region of low pressure a broad south-westerly current blows into the low pres- sure. This south-westerly current, which is called the ‘South-west Monsoon’, originates south of the Equator Winds of the Upper Air 1 9 and dominates the entire situation over the Indian Ocean. The belt of doldrums disappears over this region, being wiped out by the monsoon, which blows with considerable intensity, often attaining gale force in places. During this period the wind over the China sea is southerly, and that over the Mediterranean (the Etesian wind) is westerly, all in accordance with the pressure distribution, and all show- ing that the monsoon winds form part of the greatest cyclonic circulation of the globe. What is known in the east as the north-east monsoon in winter is the trade wind, reinforced by the develop- ment of the Asiatic high pressure, but it never attains the strength of the south-west monsoon. It is not possible to give an equally definite picture of motion in the upper air. The trade winds are reversed aloft, and a south-westerly wind blows above the north- east trade, and a north-westerly wind above the south-east trade. These currents form a part of the return flow from the equator. The depressions and anticyclones of middle latitudes must not be regarded as mere incidents in the general circulation of the atmosphere. On the contrary, they form an integral part of that circulation, acting as means of transport of heat and momentum from equatorial to polar regions. They convey warm air from tropical or sub- tropical regions to higher latitudes, and cold air from high latitudes back to low latitudes. In midwinter when the sun is low even at noon, the temperature at places in middle and high latitudes depends far more on the direc- tion of the wind than on whether the sun is shining. In December in England a cloudy day with a south-westerly wind is usually much warmer than a sunny day with a northerly wind. ( 20 ) The Tipper Air While meteorology has made great strides since the days of Maury, we are still far from having arrived at a complete understanding of the physical processes at work in the atmosphere, an d there are many types of observations which are to-day urgently needed to help in solving some of the problems of the atmosphere. First among these comes the observation of upper winds and temperatures. Our knowledge of atmospheric conditions over the land has been greatly extended in recent years and systematic observation of upper winds up to considerable heights above the ground are carried out in most of the principal countries of the world, but there still remains the sea area amounting to ^ths of the globe over which comparatively few observations have been recorded. From the information obtained over the land we can say that in general the wind veers and increases in strength up to about 1,500 feet, above this it may still increase with height but follows no general rule, and its variation will be mainly controlled by the distribution of temperature in the horizontal. The variations of wind with height are much less regular than the variation of temperature with height. The latter can be estimated with considerable accuracy and on the average falls off i°F. for every 300 feet in all latitudes. Observation of winds and temperatures in the upper air over the sea are urgently required and the co-operation of those who go down to the sea in ships is necessary in order to obtain them. Two forms of observation have been proved to be practicable at sea. (i) The observation of a track of a pilot balloon by means of a sextant and compass. The balloon is of india- Sounding Balloons 21 rubber and inflated with hydrogen and released from the ship. Its rate of ascent depends upon its weight and free lift; a usual rate of ascent is 700 feet per minute, so its height can be found approximately at any time by noting the number of minutes that have elapsed since the time of starting. The altitude and azimuth are obtained by means of a sextant and compass, and from data so obtained the velocity and direction of the horizontal current are found at any time. (2) The other form of observation is by sounding balloons. These balloons are of larger size than those which are used as pilots. They are sent up in pairs in tandem and carry a sea anchor float and a self-recording instrument, known as a meteorograph, which records simultaneously the pressure, temperature, and sometimes also the humidity. The height at any moment is deduced from the observations of pressure and temperature. On attaining a great height, say 40,000 or 50,000 feet, one of the balloons will burst and the remaining balloon, being unable to support the instrument, descends until the sea anchor float reaches the water. The balloon is now able to support the instrument some feet above the surface of the water and is prevented from drifting by the sea anchor float until recovered by the ship which has followed it. The balloons take about an hour to go up and an hour to come down. If the sea room is limited or if time is not available for taking observations up to the maximum height, a device can be attached by which the upper balloon can be re- leased at a pre-arranged height. The device consists of an aneroid box which expands as the pressure on it decreases and by so doing releases a catch which holds the upper balloon. A sketch of the apparatus after release is given in Fig. 6. 22 The Air In cases where the pilot balloon method is not practic- able, observations of the motion of cloud are still useful. It would be of great assistance to have observations of cloud anywhere near a tropical cyclone, as these might help to throw light on the origin of the cyclone. Weather Charts The modern method of forecasting weather is based on the use of a synoptic chart, i.e. a chart on which are plotted a number of observations all taken at the same time over a wide area. The observations show the actual barometric pressure, direction and force of the wind, temperature, weather, &c., at the time for which the chart is compiled. Lines are drawn through the places where the pressure is the same and are known as ‘isobars’. The isobars are the most important feature of the chart and give a pictorial representation of the pressure distribution at sea level. The wind always takes a definite direction relative to the trend of these lines and its velocity is approximately inversely proportional to their distance apart. If the isobars are drawn over a large area in middle latitudes, it will be found generally that some of the iso- bars are closed, and run round a centre of low pressure, like the contour lines of a hollow on a map, while others run round a centre of high pressure like the contour lines of a hill on a map. These two types of pressure distribu- tion are called a depression and an anticyclone, or ‘low’ and ‘high’, respectively. Regions on a synoptic chart are usually classed as one of seven types, but the seven types are in reality included in the depression or anticyclone, or in the regions between pairs of these. It has been found that certain definite distributions of weather are associated with each particular type of pressure distribution. A full RELEASING DEVICE METEOROGRAPH/ SEA ANCHOR FLOATS Fig. 6. Sounding Balloon as used in meteorological work 24 The Air description of these types will be found in any text-book of meteorology. The aim of drawing synoptic charts at sea is to make it possible to forecast the weather for some time in advance. The central problem of weather forecasting may be stated in a few words. It is the problem of forecasting the changes in the pressure distribution, since the winds and weather are associated very closely with the pressure dis- tribution. And further, since the changes in pressure distribution in middle latitudes are mainly brought about by the motion of depressions, the main problem of the forecaster is to determine the future motion of the depres- sions, which appear or are likely to appear on the chart. Chart V shows a synoptic chart drawn at sea from information obtained by wireless telegraphy from shore stations and ships. The direction of the wind is repre- sented by an arrow flying with the wind, the point of the arrow being on the position of the station, the force of wind is represented on the Beaufort Scale by the number of fleches on the arrow. The pressure is given in milli- bars’ and inserted alongside the point of the arrow. The air temperature is in degrees Fahrenheit and is' noted below the pressure. Sea temperature is given in degrees Fahrenheit in brackets alongside the air temperature. The weather in letters of the Beaufort notation is given below the figures for temperature. The figure above the pressure indicates the number of half-millibars the pressure has changed during the last three hours, indicating the rise or fall of pressure. Weather messages giving synoptic data are issued from stations in all parts of the world and in addition a certain number of Atlantic liners send regular reports to the London Meteorological Office. These messages which are usually in code (see Admiralty List of ‘ 1,000 millibars = 29-53 Chart V 26 The Air W/T Signals) can be intercepted with a -wireless set by any one requiring them. Full instructions for drawing synoptic charts are given in ‘Forecasting Weather’ by Sir Napier Shaw and other Meteorological publications. Some Phenomena observed at Sea The tropical cyclone. This is the revolving circular storm of low latitudes, whose name varies with locality. It is known as a cyclone in the Indian Ocean and Mozam- bique Channel, as a hurricane in' the West Indies, and as a typhoon in the China Seas. Its area is small, but within that area pressure falls off very steeply towards the centre, and the winds which occur in it are extremely violent. Its motion is so slow, that when a falling barometer has indicated the approach of such a storm, by taking note of the ‘Law of Storms’ it should be possible to navigate the ship away from the centre.' The central region of the tropical cyclone is a region of calm, often giving a glimpse of blue sky. The approach of the cyclone is usually heralded by motion of cirrus away from the centre. The tornado is similar in its nature to the tropical cyclone. It is, however, of much more limited horizontal extent, but is capable of doing imlnense damage on ac- count of the colossally high winds it produces. Tornadoes frequently appear to form at the boundary of two opposing currents, much after the manner in which eddies form as dimples in a mill pond at the boundaries of an inflowing stream of water. It is, however, certain that tornadoes can be formed by violent convection of heated air. Many were observed to form over the fires which devastated Tokyo after the great earthquake of September 2, 1923, and a ■ Vide Barometer Manual, 1925 edition, p. 76. hand and Sea Breezes 27 vivid description of the formation over a great fire of oil of a succession of whirls, one of which carried a cottage some fifty yards and then dropped it as a complete wreck, will be found in the ‘Scientific American’ for December, 1926. The waterspout is a small tornado over the sea. The whirl is supposed to be set in motion by the interaction of opposing currents. The whirling motion leads to a diminution of pressure within the core, with a consequent lowering of temperature. If the temperature falls below the dew point, condensation takes place and the core is made visible by the water drops. It frequently happens that the core is visible below a cloud, and again at the sea level, but invisible at middle heights. The reason for this is the relative dryness of air at middle heights in con- trast with the nearer approach to saturation of air near cloud level and above the sea. There may be a certain amountof suckingup into the core of water drops churned up from the sea surface. The winds in the whirl are extremely violent, but the horizontal extent of the whirl is very slight and usually no effect is observed at 100 yards from the centre. Waterspouts frequently occur in groups of two or three or more. Land and sea breezes occur over most coasts, though they are frequently masked by other conditions. The sea breeze, blowing towards the land, sets in during the morning, and the land breeze, blowing towards the sea, sets in during the evening. Along tropical coasts, the sea breeze prevails during the greater part of the day, reaching its maximum intensity about 2 p.m., and the land breeze prevails during the greater part of the night, reaching its maximum intensity about i a.m. These winds are shallow, extending only to a height of 500 to 1,000 feet, where they die away, and are presumably .replaced by a return 2 8 The Air current. The direction of motion at the surface is always from cold to warm, i.e. from cold sea to warmer land by day, and from cold land to warmer sea at night. Clouds may be observed to perfection at sea, where the whole sky is visible down to the horizon. The very large number of forms of clouds may be divided into four main classes, cirrus, cumulus, stratus, and nimbus, following the classification proposed by Luke Howard at the begin- ning of the nineteenth century. Cirrus clouds are very high clouds of feathery or wisp form. Cumulus clouds or heap clouds may occur at almost any height from i,ooo to 10,000 feet. Stratus cloud is a flat cloud sheet or layer in appearance like a high fog. Nimbus cloud is the heavy rain cloud. For our present purpose this classification is sufficient, but it may be added that intermediate types occur between these main classes. An international classi- fication of clouds is described in a small pamphlet pub- lished by the Meteorological Office, entitled ‘Cloud Forms according to the International Classification’. In this pamphlet twelve types of cloud are described. The observation of clouds is the first step in the study of meteorology. Cirrus clouds moving from the centre give the first indication of the approach of the depressions of middle latitudes, as well as of the approach of the tropical revolving storm. Cirrus clouds are frequently to be observed above the north-east trades, moving north- eastward in a south-westerly current. Cumulus clouds predominate in the doldrums and in the south-east trade belt. Rainfall. Observations of rainfall at sea are unfor- tunately few, largely on account of the fact that its measurement is not quite so simple as on land since it necessitates a special raingauge equipment. Much valu- able information could be obtained even without a rain- Fog 29 gauge if the time when the rain began and ceased were recorded in the log, with a descriptive note as to whether the fall was light, heavy, or torrential. Fog is caused in general by the cooling of damp air until condensation takes place. On land this happens most frequently through the cooling of the ground by radiation at night, but that cause is not effective at sea to an appreciable extent. A more potent cause of fog at sea is the passage of a warm damp current of air over cold sea. Much of the fog of the Great Banks of Newfound- land occurs in currents of air which first traverse long distances over warm sea, thereby becoming warm and damp, afterwards reaching the cold currents of sea off the Great Banks, where they are cooled from below by the cold sea surface. Sea fogs are generally formed in this way, by the passage of warm damp air over colder water. For this reason they are more prevalent in summer than in winter. Land fogs which are mostly associated with winter anticyclonic conditions may drift out over the sea, as frequently happens around the coasts of the British Isles in autumn and winter. True sea fogs, formed over the sea, are more frequent in summer than in winter. Coastal fogs may partake of the nature of either land fog or sea fog. There is apparently another class of sea fog, formed when cold air passes over a warmer sea. The physical conditions of the formation of fog in these cases is not clear. Probably a layer of warm saturated air is formed near the sea surface, and the mixing of this layer with the lowest layers of the cold air produces fog. Coastal fog may be expected whenever there is a change of wind current, whether it be the replacement of a cold current by a warm current, or vice versa. When an unusual phenomenon is observed, it is im- portant to give as full details as possible of all the 30 The Air conditions prevailing at the time. Thus observation of fog may be usefully accompanied by observations of sea and air temperature, and a note should always be made of the depth of fog when this can be observed. Now that wireless telegraphy has become universal, observations of sudden occurrence, or sudden cessation, of atmospherics, may well lead to increased knowledge in unexpected ways. Changes in ease of wireless reception with the approach or receding of a tropical cyclone are of interest, since it has been stated by certain writers that atmospherics are always associated with the clash of warm and cold currents. This view is not given as correct or incorrect, but is stated merely in order to show that if no atmospherics are observed in a tropical cyclone, then on that view, the tropical cyclone is made up of warm air, and is not due to the interaction of warm and cold currents. Observations of atmospherics may therefore be of great value in meteorology. Optical Phenomena. There are innumerable oppor- tunities at sea for the observation of optical phenomena. Such phenomena, being of only occasional occurrence, should always be carefully observed and recorded, not only because they often give the first indication of an approaching storm, but because of their intrinsic interest. Among the optical phenomena most frequently ob- served at sea we may include rainbows, coloured rings round the sun and moon (known as coronae, halos, &c.), sun pillars and the green ray. The sun shining on the drops of water in a passing shower produces the rainbow by double internal reflection. Frequently a second and even a third rainbow may be seen concentric with the first, but much fainter than the primary. When an observer stands facing a rainbow he has the sun behind him. Electrical Phenomena 3 1 Rainbows are frequently to be seen at night, when the moon is bright. Coronae are rings of coloured light seen around the sun or moon when shining through clouds of the cumulus or stratus variety. Halos are due to sunlight or moonlight shining through high clouds of cirro-stratus, in which ice crystals take the place of waterdrops. Halos present very striking and beautiful effects, sometimes of a very com- plex form. Careful observation should be made of all such phenomena, which are of considerable intrinsic interest. The observations should include the height of sun or moon above the horizon, the state of the sky as regards clouds, the hour of the day, the diameter of the main ring measured by a sextant, and the diameters of any frag- mentary circles estimated by their visible arcs. A sun pillar is a pillar of light which, as the sun sets, shoots vertically upward through a height of 10 to 15 degrees above the point of setting of the sun. It usually remains stationary for some time. The green ray at sunset and sunrise is the brilliant green colour which the line bounding the upper limb of the sun assumes when it is just touching the water horizon. This phenomenon is plainly visible to the naked eye, but the old navigators do not appear to have noticed it. Electrical Phenomena. Amongst the electrical phe- nomena we have — apart from lightning — the aurorae and St. Elmo’s fires. At times the aurora appears as pulsating waves of white light, flashing across the sky in inde- pendent action. An aurora is generally white, but the display is often accompanied by a decidedly red appear- ance of the sky. St. Elmo’s fire, which appears as a sort of electrical discharge, occurs in the height of storms, not always thunderstorms. These fires indicate a highly electrical state of the atmosphere, the light seeming to 32 The Air play round the mastheads and yards of a ship with a faintly white glow. It would be worth noticing whether the steel masts of a steel ship with wire rigging would show this effecr. Finally, it is extremely desirable that any one aspiring to make observations of value in marine meteorology should consult the instructions issued by the Meteorological Office and such standard works as are cited in the List of Literature given on p. 483. THE WATER BY D. J. MATTHEWS Our knowledge of the physical geography of the sea, of its depths, composition and temperature, and the direc- tion and speed of its movements, has been derived almost entirely from observations made from ships at the surface. Surface observations are made directly with simple appa- ratus, but observations in the depths must be made by means of apparatus which can be lowered to the required level on a line of known length. A complete set of physical observations at any position would include the following at least: 1. Position, date, and time. 2. Depth. 3. Nature of the bottom. 4. Temperature at surface, intermediate depths, and bottom. 5. Collection of samples of water, as in 4, for the deter- mination of: a. Salinity. Specific gravity. c. Dissolved gases. ■ d. Other minute constituents such as phosphates, silicates, arsenites, nitrates, nitrites, ammonia, &c. e. The hydrogen ion concentration (/'H). /. Any other special objects of investigation. 6. Currents at as many depths as possible. 7. Transparency, or depth to which light can pene- trate. 8. Colour. 34 The Water The list is a formidable one and only a large staff in a well-equipped ship could hope to deal with it. But it must not be supposed that if many of the observations are omitted the others are of no value. The position, date, and time, must of course be given. But a yachtsman pro- vided with the gear necessary for determining temperature and collecting samples of water down to a depth of a few hundred fathoms need not hesitate to use it because he cannot measure the currents or make a sounding in the two or three thousand fathoms of water which may lie beneath his keel. Even surface observations alone are of value, especially in little frequented waters. History The serious study of physical oceanography may be said to have had its origin in the work of Maury. He extracted from ships’ logs the scattered observations of winds, currents, temperatures, and so forth, and published the results in charts of mean values. It was found that the use of these charts by navigators led to such a marked shortening of passages that the collection of data was organ- ized and now many hundreds of vessels make regular returns of observations to national institutions such as the Meteorological Office and the German Seewarte. . The next great impetus was received when deep-sea cables raised for repair were found to be encrusted with living animals, and the belief that the depths of the oceans were lifeless wastes, or at least inhabited only by the lowest forms, was disposed of for good and all. Following on this discovery many expeditions were fitted out for the study of oceanography, the most famous of which was that in the Challenger. Since that date many research ships have been sent to sea, some for pure oceanographical investigation, some chiefly as surveying ships, and some Oceanographical 'Exploration 35 for polar exploration in the first place. At the beginning of this century the International Council for the Exploration of the Sea was instituted, with its head-quarters at Copen- hagen. Though founded primarily to study the fisheries of northern Europe it has conferred the greatest possible benefit on physical oceanography by drawing up tables showing the relation between chemical constitution, salinity, pressure, temperature, and density, and by pro- viding tubes of ‘normal water’, that is, carefully analysed seawater, to be used as standards in the determination of salinity. The work of the Council led to detailed study of the conditions in the fishing-grounds and their varia- tions, both periodic and unperiodic, and has provided a model for such work elsewhere. At the time at which this is being written great advances are being made in physical oceanography, so great indeed as to lead to a faint regret that a revised edition of this book must be prepared just now. The oceanographical work of the Gauss expedition in the southern Indian and the Antarctic Oceans and of the Sealark in the Indian Ocean proper has just been published; two great expeditions have recently returned from the South Atlantic Ocean in -the Discovery and the Meteor, and the detailed investigation of the Pacific Ocean is being taken in hand by the Pan-Pacific Conference. In one direction, however, it is to be feared that oceanography will suffer. The success of the echo-sounding gear makes it not unlikely that nearly all deep soundings will be made with it in the future, with a consequent serious decrease in the number of bottom temperatures and of specimens of bottom deposit collected. ( 36 ) The Teinpcraturc and Salinity at the Surface We must first consider, in the broadest outlines only, what is known of the general distribution of temperature, salinity, and currents, fo r these are the foundations of o ceanoeranhical investigation . Only two assumptions need be made for the present ; one, that sea water is pure water in which is dissolved more or less of a mixture of constant composition, over three-quarters of which is common salt, sodium chloride; and the second, that the density of sea-water increases as the salinity and pressure rise and the temperature falls. The average distribution of surface temperature is shown on Chart VI, which is founded chiefly on data con- tained in the meteorological logs kept in ships for the national institutions in various countries. In areas far from the traffic lanes records are naturally scanty and here the yachtsman can do valuable work. The chief points to be noticed are the broad belts of maximum temperature crossing each ocean between the tropics, in which there is little change in a north and south direction, the more rapid fall polewards outside the tropics shown by the crowding of the isotherms (lines of equal temperature), and the irregular course of many of the latter. If the temperature of the sea depended upon the altitude of the sun we should expect the isotherms to be parallel to the equator, but actually they often follow a very different course. The isotherm for 5° C. for instance in the North Atlantic runs for a space nearly north and south off the coasts of Labrador and Newfoundland, and that for 10° C., from about 40° N. lat. on the Atlantic coast of North America to nearly 60° N. lat. off the west of Scotland, a difference of latitude of nearly 1,200 sea miles. These irregularities are due to the Gulf Stream ' (Chart VII), I Or North Atlantic Drift (sec p. 51). The mean surface isotherms (Centigrade). The lines are drawn through points of equal temperature 38 The Water which carries tropical water to the British Isles, the coasts of Norway, and even to Spitzbergen, and to the Labrador Current, which brings southward ice-cold water, with bergs and pack ice during the early part of the year. The warm water loses a great deal of its heat on its long drift, but still retains enough to keep the Norwegian harbours always free from ice and make Spitzbergen approachable on its western side during the summer months. Much of the heat lost in the Norwegian Sea is carried to Norway by the prevailing on-shore winds and gives it a comparatively open climate; without these winds the warm waters would have little influence except on a narrow strip along the coasts. In the South Atlantic Ocean the Brazil Current and the Falkland Island Current are the causes of the sharp bends in isotherms between 25° C. and 10° C. on the American coast. Off the Japanese coasts the warm Kuro- siwo, the counterpart of the Gulf Stream, gives rise to similar irregularities but on a far smaller scale. The average minimum temperature of sea water is its freezing- point, between — 1° C. and — 2° C., but there is no natural upper limit. In the open sea temperatures above 28° C. are uncommon, but very high readings have been ob- tained in the enclosed waters of the Persian Gulf, reaching even 35° C., nearly the temperature of the human body. The distribution of surface salinity (Chart VIII) is less simple than that of temperature. The average salinity of all the oceans is about 35 parts per thousand by weight. The upper limit of over 40 per thousand is reached in the northern part of the Red Sea, where the evaporation is intense and the inflow of fresh water negligible. Salinities of 39 per thousand are found in the neighbouring waters of the eastern Mediterranean. In the open oceans 1 the highest salinities occur near the centres of high baro- metric pressure, where rapid evaporation occurs under J Chart VII The mean surface currents 40 The Water clear skies with nearly constant breezes and only weak surface currents. The most famous of these is in the centre of the North Atlantic Current whirl, the Sargasso Sea, with a salinity reaching nearly 38 per thousand. Similar centres occur in the South Atlantic, in the Pacific, and in the Southern Indian Oceans. In the Northern Indian Ocean the conditions are somewhat different be- cause it is barred by land north of 20° N. lat., but a large area of high salinity occurs in the Arabian Sea. The effect of the great surface currents on the distribution of salinity is as clear as in the case of temperature. The Gulf Stream and North Atlantic Drift carry great masses of water of 35 per thousand salinity to Norway and Spitzbergen, and near this the East Greenland Current brings southward water from the Polar Sea, the salinity of which has been reduced to 32 or less by the outflow from the great rivers which enter it and by snowfall and the melting of ice. Heavy rain and large supplies of river water are also found in low latitudes, and salinities of less than 34 per thousand occur near the equator in the open sea, but here they are associated with high temperatures. It is difficult to assign a lower limit to the salinity of true ocean water: perhaps it may be put conveniently at about 32 per thousand. In enclosed seas it may be much less, and in the Gulf of Bothnia the water is almost drinkable. There is another factor affecting surface conditions of which mention must be made; that is upwelling from below. Where an off-shore wind or even a wind along the shore removes the surface water more quickly than it can be replaced by surface flow, water wells up from below and marks its presence by relatively low temperatures. Such areas of upwelling occur in many places, such as off the Atlantic coast of Morocco, in Walvis Bay, in the waters north of Sokotra, under the lee of the Galapagos O jD 80 120 i6ol i6ow i; Chart VI II The mean surface isohalmcs showing the weight of salt in parts per thousand after Schott 42 The Water Islands, and in the Peru or Humboldt Current on the western coast of South America. Part of the low tempera- ture of the latter is due to its origin in the cold easterly drift of the Southern Ocean, but its extraordinary uniformity over many degrees of latitude is almost cer- tainly caused by its origin in a deep layer the temperature of which hardly changes with increase of distance from the Antarctic Continent. Seasonal Changes in Temperature The seasonal change of surface temperature varies greatly from place to place. There is, for instance, a small area in about 5° N. lat. in the North Atlantic Ocean where the mean yearly range is less than 1° C., and a large one on both sides of the equator in which it is only 2° C. ; such a condition is easily explained by the facts that the midday sun is never low and that the meteorological conditions are fairly constant. South of Newfoundland the mean range is 1 5° C. or more, owing to the variation in the layer of cold low salinity water from the Labrador Current and from the St. Lawrence River, combined with the greater change in the altitude of the sun. The seasonal changes of salinity are also considerable, and due largely to changes in the strength of currents and in the outflow from rivers. The seasonal change decreases with the depth and is very small below 200 m. ; it can be detected in still greater depths, but it is difficult to decide how much is due to a change proceeding downwards from the surface and how much to variation in the strength of deep currents. An important result of the decrease of seasonal varia- tion downwards is that observations at depths of 400 m. or more may be used for drawing charts and sections of temperature and salinity, regardless in most places of the time of the year at which they are made. ( 43 ) The Subsurjace Circulation oj the Oceans It has long been known that in the open sea both temperature and salinity decrease continuously, on the whole, with increase of depth. On the bottom, in depths greater than 1,000 m., the salinity is not far from 35 per thousand and the temperature 3° C. or less, falling to below 1° C. in some areas in free communication with high southern latitudes. It has also long been known that the rate of decrease varies from place to place. In the Sargasso Sea, for instance, water of high salinity and temperature penetrates to great depths, in the corresponding southern latitudes to depths which though smaller are still con- siderable, while under the equator relatively cold and fresh water rises to near the surface. A very simple theory of the deep circulation was founded on these observations. It was supposed that the surface water of subtropical and temperate zones sinks to moderate depths in both northern and southern hemispheres, and then turns towards the equator, where it rises to replace the water carried away by the surface equatorial currents. According to the theory it receives at the same time a considerable addition of very cold water which has sunk from the surface in high latitudes and has made its way equatorwards along the bottom ; the mixture of these two waters accounts for the low temperatures at small depths under the equator. This theory divides the deep circulation into two nearly sym- metrical parts. More recent observations and the rediscussion of the older ones have shown that the true circulation is pro- bably far more complicated. In the Atlantic Ocean, for instance, there appear to be at least four different layers of water. In the equatorial regions there is first the sur- face layer of a depth of a few hundred metres only. 44 The Water Below it, from about 500 m. to 1,200 m , there is a layer of lower salinity which has been formed along the shore of the Antarctic Continent and spread northwards to 45° or 50° S. lat., where it has sunk and continued its way to at least 15° N. lat. Below this again there is an increase of salinity due to a current which has sunk from the sur- face in the subtropical waters of the North Atlantic Ocean and flows southwards to 50° or 60° S. lat., where it rises a little and appears to mix partly with the current of low salinity going northwards and partly with cold and dense water which sinks along the edge of the continent and covers the larger part of the bottom of the Atlantic Ocean. The differences of temperature and salinity on which this theory of the deep circulation is founded are comparatively small and demand the most accurate thermometers and water-bottles for their observation. It would be advisable too in the future to space the observations below 1,000 m. somewhat more closely than has been usual in the past. The maximum-minimum thermometer is of course useless under such conditions, unless it is so slow that it can be lowered through layers of considerable thickness without taking their temperature, as appears to have been the case with those used in the Challenger. But a slow thermometer leads to great uncertainty and to loss of time. The complete truth of this theory of oceanic circulation cannot be said to have been proved as yet, largely because the question of currents in other than meridional direc- tions has not been sufflciently considered, but it fits the facts as known at present better than any other, and the recently published observations in the southern Indian Ocean provide additional support. The investigations of the South Atlantic Ocean now in progress should show decisively whether the theory is tenable or not. The circulation in the Indian Ocean appears to be Faroe-Shetland Channel 45 identical if allowance is made for its limitation in a northerly direction, and the circulation of the Pacific Ocean is probably on the same model. The Faroe-Shetland Channel The Norwegian Sea and Arctic Ocean are cut off from the North Atlantic Ocean by a rise stretching from the north of Scotland to the Faroe Islands, Iceland, and Greenland, with a maximum depth of about 600 m. The part of the rise which lies between the Faroe and the Shetland Islands has a maximum depth of not quite 500 m. and a width of about 12 miles; the Faroe-Shetland Channel which crosses it is the most famous region in the history of oceanography. On the south side we have the warm water of the North Atlantic Drift (Gulf Stream Drift), with nearly uniform temperature of 7° C. or more and a salinity of 2 S ' 3 per thousand down to a depth of 1,000 ni., and on the north side the cold arctic water, with a temperature only a little above 0° C. and a salinity of 34' 94 per thousand. The Drift water streams through this Channel, which it fills nearly completely and covers the southern Norwegian Sea with warm water to a depth of 400 m. or more. A little of the cold northern water flows southwards over the ridge on the bottom, but its amount is insignificant. Thus there is one continuous body of water from the surface down to the level of the Ridge, while below this level there are within a few miles of one another two perfectly distinct water masses each with its peculiar temperature, salinity, and animal life. Conditions in Enclosed Seas Many enclosed seas are connected with the oceans or w'ith the larger seas by narrow entrances nearly blocked by relatively shallow bars, so that they are completely cut 46 The Water off from the cold bottom waters which are formed in high latitudes. Such seas will be filled from the depth of the bar down to the bottom with the densest water which can gain access to them. At most places, but not everywhere, the temperature is the decisive factor and the densest water is the coldest; the temperatures in such seas will depend upon that in the outer ocean at the greatest depth over the bar, the ‘threshold depth’, and upon the average minimum surface temperature. In the Sulu Sea the threshold depth is 700 m. and the temperature of the Pacific Ocean at this depth is 10-2° C., while the minimum surface temperature is much higher. The densest water which can gain access to the sea is that of the outer ocean at 700 m., and accordingly the temperature from this depth to the bottom in more than 4,000 m. is 10-2'’ C. In the Mediterranean Sea the mean minimum surface temperature in the winter is between 12° C. and 13° C., and this extends to all depths. The water is therefore slightly warmer than that of the Atlantic Ocean at the ‘threshold depth’, but at the same time it has been so concentrated by evaporation, to a salinity of 38 per thou- sand or more, that it is the densest water of the region. Accordingly it fills the Mediterranean from two or three hundred metres downwards and flows out as a bottom current over the Gibraltar rise, while on the surface the Atlantic water flows in to take its place. As a third example of an enclosed sea we may take the Black Sea, which is connected with the sea of Marmora by the Bosporus, with a minimum depth of a little under 50 m. It is divided into an upper layer of about 200 m. in thickness- in which the salinity is reduced to 18 per thou- sand by dilution with river water, and a lower layer from 200 m. to the bottom, the water of which is derived from the Sea of Marmora and has a salinity reaching 22 to 23 External Causes oj Currents 47 per thousand at 2,000 m. with a temperature of about 9° C. Owing to the great differences of salinity, the sur- face water, however much it may be cooled, can never become heavy enough to sink through the bottom layer, and vertical circulation stops at 200 m., with the result that the minimum temperature occurs at a depth which varies with the position from 50 m. to about 100 m. An even more striking result of the cessation of convection at 200 m. is that below this depth all the higher forms of life are missing owing to the lack of oxygen, and the water smells of sulphuretted hydrogen set free by the action of bacteria upon the sulphates present in solution. Finally, we may refer to a peculiar kind of enclosed sea which is found in its most striking form in the Western Pacific Ocean. Such seas are deep steep-sided depressions in a relatively level ocean floor with free access to the cold bottom water. This bottom water as it sinks into them is heated by compression so that the temperature increases downwards. Ocean Currents and their Causes i. External Causes The chief cause of ocean currents is the direct friction of the wind upon the surface of the sea. In proof of this we have the experience of all seafarers, the agreement between the charts showing the permanent wind systems and the permanent currents, and the result of direct ob- servations of the effect of local winds of short duration. These observations have been made in various places, and the experiments carried out in the Baltic in lightships are probably the most satisfactory because they are prac- tically unaffected by tides and because care was taken to measure the speed a short distance below the surface in 48 The Water order to avoid the direct action of the wind on surface drifters. In comparing the directions of the currents with those of the winds which gave rise to them, it must be re- membered that the rotation of the earth causes every moving body on its surface to drift ‘with the sun’ if it is not prevented from doing so by other forces, so that a current tends to deviate to the right of its path in the northern hemisphere and to the left of it in the southern. And this is not a purely theoretical deduction. The charts of the permanent systems show it, and in the Baltic devia- tions to the right of the wind are more frequent than those to the left. Further, the surface current must act like a wind on the layer immediately beneath it, which is in turn set in motion with a further deflection cum sole^ with the sun, and at a slower speed. It is plain that at a certain depth the direction of the current must be reversed and its speed very small. The depth at which this occurs cannot be calculated because we do not know the mag- nitude of the friction due to eddies between the moving layers, but actual observation has shown that the depth of a great wind drift in the open sea where it is not con- strained by the neighbourhood of land is surprisingly small, perhaps too to 1 50 metres (50 to 80 fathoms), and the reversal of direction must take place at some depth not very different from this. On the other hand there is reason to believe that where the current as it swings round is banked up against other bodies of water its movements may extend to much greater depths. It is ofte n state d tha t I the rotation of the earth can set up a current, but this is ^a fallacy. The rotation can only influence a body already in motion, and the magnitude of the deflecting force is directly proportional to the velocity; when velocity is nil there is no deflection. If rotation could set up a current it Internal Causes of Currents 49 would be impossible for a sphere to reinain motionless upon a level surface in the absence of some other force constraining it to do so. An important consequence of this deflection is that an ocean current in the open sea has little momentum; as soon as the driving force is removed it swings from its path and tends to return on its tracks. It is also often stated that the areas of permanent or semi-permanent high barometric pressure can cause per- manent currents. If this were so the sea bottom should be dry beneath these ‘highs’ as a result of currents flowing away from them for longages. What really happens is this. A sudden increase of barometric pressure sets up a current beneath it which carries water away until the surface has been lowered by an amount corresponding to this pressure. Suppose, for instance, that the barometer suddenly rises by one inch; mercury is about 13-6 times as heavy as water, so the current will continue until the level of the sea has been lowered by 13-6 inches and then cease. Observations on the question are difficult to make because it happens that in the localities otherwise suitable for the purpose a change in barometric pressure sets up winds which also give rise to currents and it is not easy to separate the two effects. ii. Internal Causes If two bodies of water are in free communication with one another and the density of one is increased by decrease of temperature or increase of salinity, then the heavier water will sink and flow under the lighter, and the lighter water will flow on the surface to replace the heavier, so that there will now be two currents in opposite directions. The currents in the Straits of Gibraltar are due to such a cause, as has been explained on p. 46. Currents of similar origin are formed in many parts of the world, and 3166 E 50 The Water the great subsurface current systems of the Atlantic and Indian Oceans, to which reference has already been made, are due to causes of this nature. The direct observation of deep currents is a matter of considerable difficulty, but they may be often traced with relative ease by means of observations of temperature and salinity. Temperature and salinity also give us a great deal of information about such currents in another way. A density current is deflected by the earth’s rotation in the same way as a wind current, and the force acting on a given unit volume of water will vary directly as its density and velocity and the sine of the latitude. Now it often happens that when the density at a number of places in a current is calculated from the temperature and salinity it is found that it varies from point to point and depth to depth in such a way that the water is in unstable equilibrium. In the Labrador Current, for instance, light water is banked up against the coast instead of spreading everywhere over the surface of the sea, and the obvious reason for this is that it is deflected to the right by the rotation of the earth. We can therefore calculate from the observations of density what the relative velocity at various depths must be to produce the arrangement found. We cannot calculate the absolute velocities, but only the differences between the velocities at various depths. If we know on other grounds the absolute velocity at any depth, then we can calculate it at all the rest. iii. The Surface Currents In each of the three oceans the north-east and south- east trade winds blow towards the equator and carry the surface waters with them as the Equatorial Currents, which flow westwards until they meet the land in their path. They split here into two branches which turn to Atlantic Currents 51 the right and left respectively, and set up great circular movements ‘with the sun’, to the right in the northern hemisphere and to the left in the southern. In the Indian Ocean only the southern circle is developed, because the northern part is limited by the Asiatic continent and because during the northern summer the direction of the wind is reversed when the north-east trade wind is re- placed by the south-west monsoon. The equatorial cur- rents are the mainspring of the whole circulation. In the Atlantic Ocean the Southern Equatorial Cur- rent divides on Cape San Roque. The northern branch flows north-westwards along the coast to the West Indies, picking up in its course a tributary from the Northern Equatorial Current. The larger part of it flows between the islands of the Lesser Antilles and through the Caribbean Sea to the Straits of Yucatan and so into the Gulf of Mexico. The current is very strong every- where in the Caribbean Sea, but the Gulf of Mexico acts rather as a reservoir and is almost free from current except where the flow from the Straits of Yucatan turns sharply to the right round the western end of Cuba and escapes through the Straits of Florida as the Gulf Stream with a speed which may reach 120 miles a day. In about 30'’ N. lat. it is joined by that part of the Equatorial Current which has passed outside the West Indian Islands. It continues its course in a northerly and easterly direction, skirting the coastal shallows of the United States and the Grand Banks of Newfoundland, and off the latter it turns almost due east. In mid-Atlantic it divides into two por- tions, and as its speed and breadth are now so different from what they were near the Straits of Florida it may be better called the North Atlantic Drift. The southern branch turns southwards in the latitude of the north of Spain, and then still more to the right until as the Canary 52 The Water Current it joins the Northern Equatorial Current again. The northern part spreads fanwise and throws off branches to the English Channel and the western coasts of the British Isles, to the Faroe-Shetland Channel, and past the south of Iceland to the east coast of Greenland. The branch which passes through the Faroe-Shetland Channel and the Norwegian Sea finally sinks in the neighbourhood of Spitzbergen ; it has no counterpart in the other circular systems in either geographical or economic importance. The whole of the Gulf Stream System is relatively warm. In marked contrast is the East Greenland Current of cold water which overflows from the North Polar Sea and carries great masses of pack ice south-westwards to Cape Farewell, round which it turns to Baffin Bay. Here it meets other outflows from the Polar Sea and the com- bined waters flow south-eastwards as the Labrador Cur- rent, which carries great quantities of pack ice and bergs with it as far as the Atlantic traffic lanes in the first half of the year. Off the southern edge of the Newfoundland Banks it meets the Gulf Stream at right angles and is in part deflected by it and in part dives under it. In the South Atlantic Ocean the southern branch of the Southern Equatorial Current, the Brazil Current, the Southern West Wind Drift, and the Benguela Current correspond to the Northern Equatorial Current, the Gulf Stream, the North Atlantic Drift and the Canary Current, but on the whole the circulation is less powerful than in the northern hemisphere. The Falkland Island Current is the counterpart of the Labrador Current and like it carries great numbers of bergs. Eastwards of the thirtieth meridian the Atlantic Counter Current sets into the Gulf of Guinea between the Equatorial Currents. It is in part a compensation current replacing the water carried away by the latter and in part Pacific and Indian Oceans 53 due to the south-west monsoon which is fairly strongly developed in this region. The currents of the Pacific Ocean on the whole resemble those of the Atlantic in that two great circular systems are developed. In the northern hemisphere the Kuro-Siwo corresponds to the Gulf Stream, and off the coasts of Japan is a rapid and powerful current, but its prolongation eastwards to the American coast and the southerly flow along the latter are far weaker than their Atlantic counterparts. The cold current which flows south-westward from Bering Sea and the Sea of Okhotsk corresponds to the Labrador Current, but it carries no bergs with it. In the southern Pacific Ocean there is a circular flow corresponding to that of the Atlantic; but there is no polar current corresponding to the Falkland Island Current. The Peru Current is the feeding ground of the guano birds, and when as sometimes happens it is displaced by a warm current from the north, El Nino, they die in their thousands from starvation and weakness, and even as far south as Callao the sulphuretted hydrogen given off by decaying fish darkens the white paintwork of ships in harbour there. Both the Equatorial and the Counter Currents are well developed. In the southern half of the Indian Ocean the circular s)'stem is found again. The Southern Equatorial gives off branches which turn south-westwards on both sides of Madagascar. The one which flows to the west- ward of the island, the Mozambique Current, increases greatly in strength as it advances along the coast, and off the southern part of the African continent it is known as the Agulhas Current. Here it meets the West Wind Drift of the South Atlantic Ocean and after penetrating it to a certain extent is turned back eastwards together with the current which flows southwards outside Madagascar. 54 The Water In the northern winter the north-east trade wind and the north-east monsoon strengthen one another and the general movement of the water in the northern Indian Ocean is westwards with a strong easterly counter current, but the land masses prevent any circular system as in the other oceans. In the northern summer the south-west monsoon blows with considerable force and the general flow is to east and north-east. A very rapid current flows north-eastwards along the Somali coast into the Arabian Sea and escapes south-eastwards round Ceylon into the Bay of Bengal. As in the winter there is no circular current system. In the region of the ‘Brave West Winds’, south of the American, African, and Australian continents there is a continual drift to the east which throws off branches into each of the oceans, and south of this, close to the shores of the Antarctic Continent, a slow set to the west. METHODS OF INVESTIGATION Temperature and Salinity Temperature is observed at sea, but the determination of salinity with the necessary accuracy is not possible except in a large and well-equipped ship, and it will generally be necessary to preserve samples of the water for examination on shore. For this purpose the bottles known commercially as ‘6-oz, milk bottles with porcelain stopper, rubber washer and swing catch’, holding about 1 70 c.cm., are very convenient and not expensive. A single determination of chlorine by the volumetric method re- quires at most 40 c.cm. inclusive of the water used for washing out the pipette, so that a bottle holds sufficient for two titrations, and a determination of density by weighing as well if this is to be done. Each batch should be Temperature and Salinity 55 examined carefully when received from the makers and any bottle in which the washer appears to fit loosely should be rejected. Nothing but the best rubber should be used for the washers. All glass is slightly soluble in sea water on long standing ; the brown variety used for beer bottles appears to be less affected than the other kinds. With good glass the solubility decreases with use and bottles previously used for holding sea water are to be preferred. New bottles may be improved by steaming followed by washing with hot distilled water. For all ordinary hydro- graphic purposes the solubility of the glass does not affect the determination of the chlorine by chemical methods, but the specific gravity becomes appreciably higher with time, and this would have some effect on the determina- tion of density by weighing and on any method involving the refractive index. When the greatest accuracy is re- quired bottles of special resistance glass should be used ; they should be fitted with rubber stoppers. Suitable boxes for packing the bottles can be had from the glass-makers. They should be divided into compart- ments so that each bottle is separated from the others, and lid and bottom should be lined with felt. No other packing is required. Hasps and staples should be fitted and rope handles; hand holes give admission to rats which eat labels. The number of bottles in a box may be from 30 to 42; a box of 42 bottles measures about 15 by 20 by ir in. and weighs about 60 lb. empty. Care should be taken to rinse out the bottle with some of the water and not to fill it within an inch of the stopper. Frost should be guarded against. Bottles with glass stoppers are not reliable, but sound, well-washed corks may be used in an emergency. The necessary particulars, date, time, position, depth, temperature, and ship, should be written in pencil on stout Manilla labels, which should 56 The Water be tied to the bottle in such a way as to prevent the catch springing open. There is one rule of the greatest importance, that is, that no thermometer should be used unless its errors are known. All patterns of thermometer used for marine in- vestigation can be bought with a certificate of examina- tion. In this country the certificates issued by the National Physical Laboratory at Teddington give the correction to be applied to the reading, but the Physikalisch-Technische Reichsanstalt at Charlottenburg shows the errors of the scale. Thus if the reading is 10° and the correction is -o°-i, the true temperature is 9°' 9; but if the error is -o° - 1, then the temperature is io“ - 1 . The glass of which a thermometer is made continues to contract fora longer or shorter time after manufacture and so causes a slight rise of the zero-point. For this reason the instrument should be tested at intervals, either by comparison with another of known accuracy, or better by means of melting ice. The thermometer is immersed in pure ice which has been crushed and washed with distilled water, so that the zero point is just visible. The vessel containing the ice should have an opening to allow the water to drain away, and sufficient time should be allowed for the temperature to become steady; the best way is to read at intervals, tapping the thermometer lightly each time, until no further change is seen. The experiment must be made in a room with a temperature above the freezing-poinr in order that the ice may melt. An error due to rise of the zero applies equally everywhere on the scale. It is as a rule not possible to divide a small division by eye more closely than to one tenth; it follows that no correction need be considered which is less than one tenth of the smallest division shown. ( 57 ) SurJdCe Observations Surface observations are usually made by drawing a bucket of water over the side well clear of any discharge from the ship and reading its temperature with a thermo- meter, after which a sample of water is bottled off. Surface temperatures have little meaning in the second place of decimals, and a thermometer divided to half degrees is sufficient for the purpose if no subsurface observations are being made at the same time. If a vertical series is being made, it is better to make the surface observation with the same apparatus as is used in the depths if the condition of the sea allows ; otherwise the surface temperature should be read to the nearest tenth of a degree C. Before reading, the thermometer must be well stirred about in the bucket and it must not be taken out of the water. The most useful kind of thermometer is that in which the graduations are drawn on a flat scale of milk-glass enclosed in an outer glass tube and are thus protected from the water; the bulb is not so enclosed and the instrument is very rapid. In this country this type is sometimes called an ‘insulated thermometer’ ; care should be taken when giving an order not to confuse it with the thermometer used in the ‘in- sulating water bottle’ which is described below. They have no metal guard round them and can be easily cleansed from dried salt, so that it is permissible to use the same bucket of water for reading a temperature and preserving a sample. If a thermometer with a metal frame is used, such as the British Meteorological Office pattern, a fresh bucket of water must be drawn. ( 58 ) ObservaUons below the Surface A thermometer which has done much service in the past for measuring temperatures in the depths, but is no longer used, is the maximum-minimum thermometer, which depends for its recording upon the movements of two small indices which are pushed along the bore by the movement of the mercury. One registers the highest and the other the lowest temperature encountered by the thermometer while it is being lowered. It is plain that unless the temperature changes continuously in one direc- tion as the depth increases the readings must be inac- curate. Modern investigations have shown that there is reason to believe that this is not generally the case in the open sea, so the maximum-minimum thermometer must be discarded. In any case it is liable to give false readings owing to movements of the indices caused by the vibra- tion of the line during hauling in. It is usual to combine the observation of temperature and collection of water samples in one operation, and so many instruments have been designed for the purpose that it is impossible to describe them all. The Insulating Water-Bottle The most convenient apparatus for measuring the temperature and collecting samples of water at depths not greater than about i,ooo m. is the insulating water-bottle, the Copenhagen model of which is shown in Fig. 7 ; the original Nansen-Pettersson model was closed by weights instead of by springs and was more lightly built. It is a device by means of which a specimen of the water may be brought up to the surface without undergoing any change of temperature. The body of the instrument con- sists of a series of concentric cylinders a, of which only 6o The Water the outer one can be seen in the drawing; it slides freely on the guides b, which are connected rigidly at the bottom by the circular plate c, which is fitted with rubber washers D and a tap e, and at the top by a cross-bar f, which carries the releasing mechanism c. Another circular plate H,with washers and a thermometer guard k, also slides on the guides. The thermometer, which is divided in tenths of a degree Centigrade, has a long narrow bulb and is sealed up in a strong glass sheath to protect it against pressure; it is kept permanently fixed in h with its bulb projecting some distance below it. The bottle is opened for lowering by raising the thermometer guard until the ball at the top is held by the releasing mechanism, and closed when at the required depth by allowing a messenger to slide down the line. The top plate h falls on to the body of the bottle, and the two together on to the bottom plate; the bottle is kept firmly closed by springs in the tubular guides. The temperature remains constant to o°-oi C. for six minutes, even when the outside temperature is raised by I 2 ° C., but begins to rise rapidly after seven minutes. When there is little or no change of temperature vertically the insulating period is longer, but there is another source of error which prevents the use of this apparatus in great depths. This is the cooling which it undergoes on de- compression on hauling up. The correction can be cal- culated for the water contained, but not with any accuracy for the materials of which the bottle is made. The table on p. 6 1 contains the full amount of the correction due to the expansion of the water, but only half that due to the bottle itself. It has been found to be satisfactory by direct comparison with temperatures given by the reversing thermometer. The insulating water-bottle is quick in operation because the thermometer can come to equili- brium while it is being hauled up, but on the other hand Temperature Corrections 6i only one can be used on the wire at a time. It weighs 12 kg., holds 1,700 c.cm. of water, and is supplied by Professor Knudsen, of the Polytekniske Lacreanstalt at Copenhagen. RKMAN’S TABr.X: OI COPRECTIONS lO BE ADDED TO TEMBLRA'I ORES MEASI'RED BY THE INSULATING WATLR-BonLF. l'en)pi'rdtun: of Wjter Sample 100 1 Depih in Mttics <cc j 6CiO 7CC 800 0 0 0 I, 0 C '5 1,100 1,200 -2 •CO •Of 1 •cr •c? •02 -03 "•3 •04 ■°s •c-S ' -06 •07 -I ' •or •CI 1 TX 1 *02 •03 -.3 -04 •05 •05 •c6 1 -07 •08 o ' •C i •OI ' *02 *'2 •-'3 1 ■'74 i-t'S ■35 1 •'j6 •07 •08 •09 I •or •01 *C2 •-3 , •-4 -4 •c6 •07 •08 1 -09 •10 •or •02 •02 1 •=4 -'-5 •0- 1 •'S •09 1 •10 •ii 3 •oi •02 .r -3 •03 1 -0^ -3; -7 1 •09 • 10 1 •!! •12 4 •or •C 2 •33 •04 1 ••“5 •c^ 1-07 •08 1 -9 • 10 1 •“ •13 5 •"I •02 •03 •'4 ' •-5 -06 •07 •09 •I'- • 1 1 •12 •14 6" •ci •02 •C 3 •04 •c6 -07 •08 *09 •11 •12 •13 •H •CI •C2 •04 1-05 •c6 ‘oj •09 •10 •J 1 • 12 •‘4 'IS 8 •CI •01 •04 1-05 •06 tS •'9 •10 1 •12 •13 •M •ifi 9 ' •CI •03 • •=4 •=■5 •07 •cS •10 •11 •12 •14 •‘S •17 10'’ •OI •03 • 04 1 Tfi 1 •cy •'“9 •10 *12 1 •'3 •'S •:(> •18 ii‘’ •OI •03 •'■4 •CO •07 -09 ^ •10 •12 •14 •'S •17 •19 12“ •02 •03 ■c6 • * 1 -=9 1 ■j -3 •J 4 1 •:6 •18 ■19 13- ■02 •03 ■03 ' *06 •c8 *10 •II ■13 •'5 ■17 •18 •20 14° •02 •=3 •05 1 -07 •c8 ' •lo • 12 'H, •1': •17 •19 •21 15° •02 •03 •05 ■07 1 *t 9 1 'lO •12 '14, ■16 ' • la •20 ■22 16" 1-02 •04 •07 •C9 •II •13 ’15 •17 •19 ■’O ■22 17° ■02 •04 •c6 •C 7 ■09 •!! •■3 I •J 7 •19 •21 ■23 18 ■02 •t ^4 •o''- ’c8 •10 •iz •14 ■16 •18 ■ i-i •22 •24 19° ■02 •04 •06 •c8 •10 ^12 •M •16 1 •18 •70 ■22 ; -*5 20° ■02 •04 •06 ■08 •10 1 •'4 ■17 1 •19 ■21 1 -23 i Reversing Water-Bottles and Thermometers The reversing thermometer, Fig. 8, records the tem- perature by separating the mercury which stands above a certain point on the stem at the required depth so that it can be read on a special scale after hauling up. I’he bulb is long and narrow; above it there is an irregularity in the bore of the capillary, which may be either a simple 62 The JVater 8 9 Fig. 8. I'lLgrttti and Zam- bra’s Reversing Deep sea Thermometer Fig. 9 The Appendix and loop bend of thermo- meter contraction or ‘knife-edge’, or more commonly a short branch ending blindly, the ‘appendix’, as in the figure. Above this again there is an enlargement bent into an S-curve, or U-curve, or into a complete circle. The capillary has an unusually large cavity at the upper end and the amount of mercury is so great that It partly fills it when the thermo- meter is held with the bulb down- wards. A small thermometer of the common pattern is sealed up together with the reversing thermometer in a strong glass sheath, and the space between the main bulb and the sheath contains mercury to hasten the interchange of temperature. The thermometer is lowered with the main bulb downwards, and after a wait of seven minutes for the attainment of equilibrium, it is re- versed or turned upside down by a suitable mechanism. The mercury breaks at the appendix or knife- edge and runs to the other end of the capillary, where the temperature can be read off on the scale which is graduated for the reversed position. As a rule the thread broken off will have expanded after reversal, and a small correction must be made, which IS calculated as follows. Let a be the apparent coefficient of expansion of mercury in Reversing Thermometers 63 the glass of which the thermometer is constructed, gene- rally about eo'^DQj the volume of mercury broken off expressed in degrees of the scale, and T and r the readings of the main and auxiliary' thermometers respectively. The volume n is the reading of the main thermometer plus the volume which would, have been broken off if the reversal had been made at 0° C. ; this ‘volume at 0° ’ is engraved on the stem. The correction is then an (T-t)-, if it is large it should be recalculated, using the new value of T given by applying the first correction. Reversing thermometers are the only means of measur- ing the temperature at great depths. They can be made so sensitive as to show differences of o°-oi C., but they are liable to develop errors suddenly owing to a trace of residual air finding its way to the breaking-off point. When the error is large it is apparent at once and the reading can be repeated with another instrument, but if it is only a fraction of a degree there may be no reason to suspect it, and for this reason reversing thermometers should always be used in pairs. The makers provide in- structions for dislodging the air and passing it back to the upper cavity. Other defects are locking of the mercury in the capillary after reversal, and a tendency for mercury which passes the breaking-off point after reversal owing to rise of temperature to join the rest of the thread instead of being held in the curved enlargement. Reversing Frames and Water-Bottles It is usual to collect a sample of water whenever the temperature is observed, so that there is little use for a frame without a water-bottle in combination with it. A reversing frame for two thermometers with messenger release is illustrated in Fig. 10; it can be fixed either at the end or on the side of the line. A similar frame with 64 Fig. 10. Knudsen’s Rc\crsii'g Frami' for Deep-sea Thermometers. One- fifth natural size JFater a propeller release is also made. There are several types of reversing water-bottles to be had fitted to carry two ther- mometers and with either propeller or messenger re- lease; they fit on the side of the line and as many as six can be used at once on a good steel rope of 3 mm. diameter, with great saving of time. The one shown in Fig. 1 1 is the latest Copenhagen model. The covers shut down parallel to the ends of the bottle, a great improvement on the earlier pattern in which they were carried on hinges, and are drawn together tightly by means of a strong spring passing through the centre of the bottle. The weight is 53 kg. and the capacity i litre. The pattern released by a messenger has a fitting by means of which each bottle as it is reversed drops a mes- senger on to another bottle fixed on the line below it. Another pattern of revers- ing bottle which has been much used by its inventor is the Nansen bottle, which also Water- Bottles 65 carries two thermometers. A much lighter model, that of Dr. Richard of Monaco, carries one only. Method oj Using Water-Bottles, &c. ^Vll modern instruments are designed for use on steel wire, which is generally made up in the form of a very flexible cord with a diameter of 2 to 4 mm. and a breaking strain of 400 lb. or more. It is sometimes called sashcord. Single wire such as is used for sounding is liable to kink and is seldom used for water- bottles. Water-bottles may be used from a davit or from a swing- ing boom; the latter requires one extra hand to work it, but it is far better as it makes it possible to swing the gear m for reading temperatures and taking samples of water, so ^ , . Fig II. Rcicrsmg Water-bottle that work can be carried out in decidedly rough weather. The wire paid out is measured by passing it over a block provided with a counting mechanism. This should be graduated in metres, as all the tables for making dynamical calculations of the speed of currents are on the metric system; a very convenient pattern is that showing single metres up to 1,000 m., with a circumference of 50 cm. The reading of the block should be corrected for the diameter of the wire by adding half the latter to the radius of the sheave. The radius of a sheave 50 cm. in circumference is 79-58 mm.; if the diameter of the wire is 2 mm., then the radius of the sheave 3166 ^ I 66 The Water must be considered as increased to 80-58, and all readings corrected by adding one metre in each eighty shown. In most cases the drift of the ship will make the actual depth reached uncertain by at least this amount. The wire is preferably stored on a drum with handles for heaving in and provision for a power drive if work in deep water is contemplated. A good brake is imperative. The depths at which observations should be made can- not be laid down beforehand in unknown waters with any certainty. A good rule is to make observations at the surface and at 5, 10, 20, 30, 50, 75, 100, 150, 200, 500, 1,000 m., and then at each 500 metres and just clear of the bottom ; the depth should have been first obtained by a sounding. The temperatures should be plotted on squared paper as they are obtained, and if this shows that there is a sudden change in the rate of decrease, additional observations should be made to find the depth at which this occurs. If the conditions are known beforehand to a certain extent, the list of depths may be modified con- siderably. The distance apart at which observing stations should be spaced will also vary with circumstances; in unknown waters the only guide is a careful study of pre- vious work in similar areas and of the temperatures and soundings as they are obtained. The spacing of surface observations made without stopping the ship will also depend upon the area. In the western part of the English Channel they might be made at intervals of ten miles, on the liner tracks between Eng- land and America at each hundred or two hundred miles, or at definite times in the day, which would be far more convenient. ( 67 ). The Salinity and Constitution oj Sea-Water Sea-water is pure water holding in solution a variable quantity, from about 40- 5 parts per thousand downwards, of a mixture which we may call ‘sea salt’, and which is of practically the same composition everywhere as regards its main constituents. This mixture contains about 78 per cent, of sodium chloride (common salt), and smaller quantities of the chlorides, sulphates, bromides, and car- bonates of magnesium, calcium, and potassium. In solu- tion these substances are largely dissociated or ionized, that is, split up into two or more parts bearing electric charges. It is therefore not strictly correct to say that sea- water contains certain quantities of sodium chloride, mag- nesium sulphate and so on ; we ought rather to say that we can make an artificial sea-water which is nearly the same as natural sea-water by dissolving certain substances in appropriate proportions in pure water containing suffi- cient carbonic acid gas (carbon dioxide) to keep the car- bonates of calcium and magnesium in solution. The true composition of an average sea-water of a salinity of 35-00 per thousand is Substance Sodium Magnesium . Calcium Potassium Chlorine Sulphate ion Carbonic ion Bromine Total . Symbol Parts per thousand Na 10-72 Mg 1-32 Ca 0*42 K 0 oj 00 Cl 19-32 so^ 2-70 CO3 0-07 Br 0-07 35-00 An artificial sea-water may be made by dissolving in pure water containing carbon dioxide r 2 68 The Water Sodium chloride NaCl gram 27-2 Magnesium chloride MgCl, 3-8 Magnesium sulphate MgSO, 1-6 Calcium sulphate . CaSO^ 1*3 Potassium sulphate K,SO^ 0-9 Calcium carbonate CaCOj C‘l Magnesium bromide MgBr, o-i and making up to i ,000 grams. All the weights are cal- culated for the anhydrous substances. Such an artificial water would serve to fill an aquarium in which to keep the hardier fish, but it is biologically unlike the true sea-water in that it lacks certain minute constituents, phosphates, silicates, nitrogen compounds, and so on, which are essential for the production of the minute plants which form the food of the smaller animals and indirectly of the whole of the animal population of the sea. True sea-water also contains in all probability traces of a substance resembling vitamins in its action. Traces of many other elements are also present; it is not known whether they have any influence on marine life, but on the analogy’ for instance of the necessity for traces of boron for the groAvth of beans the question is well worth in- vestigation. Having collected a sample of sea-water and observed the corresponding temperature, we have to determine in the first place the salinity, which may be defined as the total weight of salts contained in a thousand parts by weight of the water. The direct determination of the salinity by evaporating a known weight of water and weighing the residue is so difficult and tedious an opera- tion that it is never used as a routine method. But since the composition of sea-salt is constant for all practical purposes, we may determine any one of the constituents and calculate the salinity from it by multiplying it by the Chlorine Determination 69 proper factor. The only constituent which is convenient for the purpose is the chlorine, or more strictly the sum of the chlorine and bromine, and fortunately this can be determined with ease and accuracy by Mohr’s titration method in which sodium chromate is used as an indicator. The analysis should be carried out under the supervision of a skilled chemist, not because the operation is difficult, since any one with good colour vision should be able to learn it in a week, but because there are so many small possible sources of error in all chemical operations which cannot be guarded against in the instructions and yet would be obvious to any one with chemical training.- The silver nitrate solution used is standardized against the ‘normal water’ supplied by the Laboratoire Hydro- graphique at Copenhagen in order to ensure that all results shall be comparable wherever the analyses are made, and the burette readings are converted to chlorine, salinity, and density at any temperature by means of Knudsen’s Hydrographical Tables, 1901 (see p. 483). If the density in situ, that is, corrected for the pressure due to depth, is required, use must be made in addition of one of the tables drawn up by Bjerknes, Ekman, or Hesselberg and Sverdrup, to which reference is made in the bibliography at the end of this book (p. 483). The method of carrying out the titration is described in the article on ‘Physical Oceanography’ in Glazebrook’s ‘Dictionary of Applied Physics’, vol. iii, p. 677. We may also begin by determining the density at an accurately known temperature and then calculate the salinity from this by means of Knudsen’s Tables. As a general rule this method is not to be recommended, as it is slower than titration and the possibilities of error are greater. In skilled hands, however, and in a properly equipped laboratory it gives excellent results. 70 The Water At one time densities were determined by means of the common hydrometer in which a part of the stem projects above the surface. Unfortunately it has been found that its real accuracy is far less than its theoretical sensitivity, chiefly owing to the influence exerted upon the surface tension by a trace of grease, and it is now obsolete for oceanographical purposes. Accurate determinations of density may be made by weighing a known volume of water at an accurately known temperature, or by weighing a suitable heavy body in pure water and in the sea water, also at accurately known tem- peratures, or finally by means of the total immersion hydrometer. The first two require a good balance and weights and facilities for keeping the temperature con- stant. The total immersion hydrometer requires a far simpler outfit and is accurate. It resembles an ordinary hydrometer with a very short stem and is adjusted to float in midwater, partly by dropping small ring-shaped weights over the stem and partly by small changes of temperature; the corresponding density is taken from the calibration tables which are supplied with the instrument. Its chief disadvantages are that it is slow and requires a rather large quantity of water. Several other pieces of apparatus have been designed upon the same principle. Current Measurement The measurement of surface currents by means of sur- face drifters thrown overboard to be picked up perhaps thousands of miles away requires considerable skill and an organization which is generally beyond the reach of those to whom this book is addressed. If, however, any yachts- man who is visiting little frequented parts of the world feels inclined to experiment in this direction he should Current Meter 7 1 apply to the Challenger Society for advice, which will be given gladly. Surface currents may be measured from a ship at anchor by means of a log-ship of the kind used before patent logs, but made much larger than usual. It is simply a flat piece of wood shaped as the sector of a circle and weighted along A, vane; B, propeller, C, first messenger; D, arm to unlock and lock, c, end of D carrying slot which engages with B; r, box containing revolution-counter and shot-store; below this is the compass-box. its curved edge with lead so that it floats upright in the water. It is allowed to drift astern at the end of a light line, which is marked at every 10 feet, and the length of line carried out in a given time is recorded. For the measurement of currents below the surface the Ekman meter is probably the most convenient apparatus (Fig. 12). It consists of a hollow spindle turning freely by means of ball-bearings on a wire passing through its 72 The Water centre, and bears on one side two vanes a, inclined at a small angle to each other, and on the other a framework carrying devices for recording the velocity and direction. The velocity is measured by the revolutions of the pro- peller B, which turns on centres supported in agate cups and is protected by a guard ring not shown ; it is connected by gear wheels with the pointers moving on the dials on the cover of the box f. These pointers are sprung on to their pivots, and can be set to zero before each observa- tion. The velocity of the current is calculated from the number of revolutions in a given time by a formula which must be determined experimentally for each instrument. Beneath the box f is a compass, which can be quickly removed and replaced, and is kept in position by the thumb-screw shown below it. The needle is heavy, and the arms are inclined downwards, and one arm is deeply grooved on its upper surface. In the box f there is a store of small shot, which are released one by one by the rota- tion of the propeller, and fall through a funnel on to the centre of the needle, whence they run down the grooved arm on to the compass card. The latter is fixed and divided by raised partitions, and the shot falls into the compart- ment below the end of the arm, and so records the direction in which the apparatus was pointing at the moment. The card, really the bottom of the metal ‘bowl,’ is divided at every lo degrees, in such a way as to record the direction from which the current comes, and not that in which it goes. The meter is started and stopped by means of mes- sengers. When it is being lowered to the required depth, the messenger c is not on the wire, and the arm d is more nearly vertical than shown in the drawing. The other end, E, is therefore slightly raised, and the projection on the propeller blade butts against it, so that the apparatus is locked. To release the propeller, the first messenger c Current Meter 73 is allowed to slide down the wire; it forces d outwards into the position shown in the figure ; the slot in e is now opposite to the projections, and the mechanism is freed. After a measured interval a second messenger is sent down ; this is larger than the first, and hollowed out below so as to pass over it. It forces d still farther outwards, and again locks the propeller. The direction and velocity of the current then can be read off from the mean position of the shots in the compass card and the number of revolutions shown on the dial. The wire should be as fine as possible to avoid stray, and if of steel any deviation which it may cause in the compass should be determined and allowed for. The sinker may weigh from i o pounds upwards, according to the strength of the current. A torpedo-shaped sinker sup- ported horizontally on a short vertical rod with fins to keep it head to current is far more efficient than other kinds. With currents of a mile an hour and over, in depths greater than thirty fathoms, the writer has found that it was not possible to feel and time the arrival of the messengers on account of the vibration of the wire. In such a case one must determine the time of fall of each messenger for various depths and plot the result on a curve. It will be seen that they soon reach a constant velocity, so that the time may be calculated from the depth. A simple telephone, consisting of a stretched string fastened to the centre of the bottom of a small box, is often useful. The Ekman meter, of course, demands some fixed point, such as an anchored boat or ship, from which to work it, and in rough weather the velocities are not re- corded accurately; under suitable conditions, however, it is very easy to use, and the results agree well among them- selves. When tested at the surface from a boat, especially 74 The Water if the current is slow, it is seen to wave from side to side continually, but this appears to be due to eddies; below the surface in currents of over a quarter of a mile per hour it points constantly in one direction. The interval during which the propeller should be allowed to rotate depends on the strength of the current ; when this is not known approximately, four or five minutes may be tried in the first instance. The magnetism of an iron ship will deflect the compass of a current meter at considerable depths. Appreciable errors have been noted at 33 ft. below a vessel 76 ft. long and at 80 ft. below one 150 ft. long. The corrections can be calculated by the well-known theory of the compass in an iron ship. Additional apparatus has also been made by means of which it is possible to work in rough weather from a ship that is not anchored, but the writer has not had any experience in its use. An anchor or heavy weight is made fast to the end of the wire; above this at the requisite dis- tance the meter is fixed, and above this again, but below the surface, is a buoy, the floating power of which is not great enough to lift the anchor from the bottom, though it keeps the line taut. The buoy contains clockwork which drops the two messengers down the line to start and stop the mechanism. The wire above the buoy is slack, and the ship steams up to it. Later models of the Ekman meter have been made, one of which is much heavier and stronger than that described, and another is designed to show the order in which the observations have been made. There are many other current meters, some on entirely different principles, but they are useful to the professional oceanographer rather than to the yachtsman to whom this book is addressed. ( 75 ) Tidal Observations As to the daily movements of ocean water, the subject of the theory of the tides is for the astronomer and mathe- matician rather than for the oceanographer. It will be enough to say that if a stay of any length be made in an out-of-the-way port or harbour accurate measurements of tidal movements would be found useful to the naval authorities and of real value. A tide pole, well guyed, marked in feet and decimals, will do all that is necessary. A board carrying a light scale-rod, floating freely in a square tube of four planks knocked together, with holes in the bottom to admit the water, will be found easier to read as avoiding the wash of the waves. Both these require reading at frequent, say hourly, regular intervals, and if the observations are to be of real value they should be continued for a minimum of 15 successive days. An exact comparison between some of the observations and a permanent fixed mark on shore should be obtained and noted in the tidal record. Note added 1st October 1^28. A preliminary examination of some of the results obtained by the Meteor has shown that the system of the ciiculation of the Atlantic Ocean, sketched on pp. 43 and 44, is on the whole correct. The interchange of water is, however, hr more active on the western side of the Central Ridge than on the eastern, since the eastern trough is largely blocked below 3,000 m. by cross ridges. II (contd.) The Alkalinity of Sea-water BY W. R. G. ATKINS If one takes a dilute solution of washing soda (sodium carbonate) or of caustic soda (sodium hydroxide) and to it adds a few drops of an alcoholic solution of phenol phthalein it is seen that the solution becomes a deep red. Taking a little of this solution in a test-tube and blowing down into it through a tube one finds that the red colour becomes less and less intense, passing finally through a faint pink to colourless. Now since sodium hydroxide and sodium carbonate are alkalis, and phenol phthalein is found to give a red colour with them, it is said to be an indicator of alkalinity, and the depth of the colour enables one to judge of the intensity of the alkalinity, which is not the same thing as the total amount of alkalinity, as will be explained later. Thus we see that alkalinity dis- appeared when air from the lungs was bubbled through, the solution being neutralized by the carbonic acid formed in the water when the carbon dioxide given off in respira- tion was passed into it. It is easily seen, therefore, that if we have a solution such as sea-water in which animals are living, and if we can get a measure of its alkalinity, we thereby have obtained information as to whether the animals in it are rapidly altering its conditions by their respiration. Now, sea-water taken from a rock pool, in which there are abundant algae during the summer months, is usually found to give a fairly deep colour with phenol phthalein. Sea-water from the open sea gives a much less intense colour if drawn from the surface or near it; if drawn from a depth of lOO metres it will probably give no colour at all with phenol phthalein. In the winter Alkalinity of Sea-Water 77 it is found that even the surface water gives little or no colour with this indicator. These phenomena are ex- plained by the fact that during the summer the abundant algae on the coasts and the diatoms of the plankton have sufficient light to enable them to carry on photosynthesis, abstracting from the sea-water more and more carbon dioxide, so that it becomes increasingly alkaline. A limit is set to this change by the growth and multiplication of the animals that feed upon the plants and are, of course, continually respiring, as are also the plants when not in good illumination. To be accurate, the plants too respire all the time, but when in good light, their respiration is masked by the amount of carbon dioxide they remove by photosynthesis. A study of the alkalinity of sea-water affords us, therefore, an idea of the relative amounts of plant and animal activity, indeed by an exact chemical estimation of the difference in alkalinity between the sea- water at any place in winter and in summer it is possible to arrive at a minimum value for the quantity of carbo- hydrate formed by photosynthesis. The value is only a minimum one, because, of course, the animals present in the sea have been respiring all the time and removing the algal plankton. It is found, using phenol phthalein as an indicator and running in from a burette centi-normal sulphuric acid, that the sea- water of the English Channel off Plymouth requires i c.c. more acid to neutralize 100 c.c. of water in summer than it does in winter. Since this was the value obtained early in October, when the fall in temperature had resulted in a thorough mixing of the sea-water from top to bottom, it is possible to calculate the amount of carbohydrate formed in a column of water having a base of one square metre and extending from top to bottom. Taking the depth found near the International Hydro- 78 The Water graphic Station E 2, viz. 83 m. (half-way across the English Channel on the line Plymouth to Ushant), it is found that i kg. of carbohydrate, reckoned as dextrose, is formed per 4 sq. m. This amount may not seem very large, since it is distributed throughout a volume of water of 4 by 8 3 cubic metres, but it gives us the somewhat start- ling result that 250 metric tons of carbohydrate are formed under each square kilometre of the surface of the sea. There is, however, a simpler method of following the changes of alkalinity in the sea-water by means of noting the change in colour of an indicator added in carefully measured proportions to a measured volume of water and by comparing it with standards carefully prepared to be of a known alkalinity. For such comparisons phenol phthalein has been used, but it is not the most suitable indicator, because owing to its low solubility in water it soon separates out of the standards, so that they fade in colour. Furthermore, it gives no information about the condition of the sea-water once the latter has more than a very slight excess of carbon dioxide above what it would retain if shaken up thoroughly with the air, so as to be brought into equilibrium with the small amount of carbon dioxide contained in the air. Instead of phenol phthalein it is advisable to use another indicator known commonly as cresol red, which covers a range of alkalinity found in sea-water drawn from all but the greatest depths of the ocean, and in all save the most alkaline rock pools. It is, for example, quite suitable for the study of water in the tanks of the Plymouth Aquarium, which are always rather less alkaline than the water of the open sea, and are never sufficiently alkaline to give any colour at all with phenol phthalein. For these less alkaline regions it is instructive to confirm the results by using another indicator, phenol red, which is yellow in very faintly acid water, and a deep Alkalinity oj Natural Waters 79 red at the intensity uf alkalinity at which phenol phthalein becomes colourless. We see, therefore, that it is not as suitable as cresol red for the more alkaline sea-water. For highly alkaline water such as that of the rock pools in summer, it is often advisable to use an indicator, thymol blue, which is a light yellow colour in the water of the open sea, but blue in water in which photosynthesis has brought about an increased alkalinity. It may be added that by the use of suitable indicators it is possible to study the condition of all natural waters, from the most highly acid, such as in a bog pool, to the slightly acid, such as those of mountain rivers in districts where the rocks are granite or slate — ^very poor in lime — down to the alkaline waters (quite as alkaline as that of the sea) found in streams in limestone districts and ex- tending to the very highly alkaline water which may be met with in ponds filled with calcareous water (viz. hard water) in which an abundance of plant life is actively developing ; in these it is by no means uncommon to find the water giving an intense blue with thymol blue. It would be a great mistake to consider sea-water alone without having any ready means of comparing it with fresh waters, and such a comparison is conveniently effected by referring waters acid, neutral, or alkaline to one scale, viz. that of the concentration of hydrogen ions found in them. An outline of how this scale is arrived at and of how the measurements may be made is given in the following pages, but as this book is intended to deal with marine life only the applications of the method will be limited to those necessary for the study -of sea-water. (8o) Hydrogen Ion Concentration In acidimetry the alkali is added to the acid till an indicator reaches a certain tint which is said to indicate neutrality. Gravimetric work proves that this is the stage at which a salt is formed, such as sodium chloride from hydrochloric acid and sodium hydroxide, no trace of either acid or alkali remaining over. Along with the salt, water, in addition to that already present, is produced. Neutrality then is the point at which only water and the salt are left. The salt gives sodium ions and ions of the chloride radicle, both in quantity, since its solution is a good conductor of electricity. The water gives very few"^ ions, as is shown by the fact that it is a poor conductor. It does, however, give off a minute number, and from its conductivity Kohlrausch and Heydweiller calculated what their concentration must be. It is necessary at this stage to consider pure water and neutrality. Pure Water and Neutrality. Pure water dissociates into hydrogen and hydroxyl ions, and the product of the ionic concentrations is constant at a constant temperature, namely, [H"^] x [OH“] =^[H,0]. Because the dissocia- tion proceeds only to such a small extent, theconstant^may be replaced by another constant, thus [H ' ] x [OH“] = K^,. Since the ions are produced by pure water in equal numbers, the concentration, reckoned as usual in gram-equivalents, is the same for both. A solution is accordingly termed neutral when the hydrogen and hydroxyl ions are present in equivalent amounts, as in pure water. With an acid solution the hydrogen ion is in excess, and the hydroxyl is correspondingly reduced, since the product is constant. In alkaline solutions hydroxyl ions preponderate; but as the product is constant, it is possible and convenient to state the reaction of the solution in terms of hydrogen Neutrality^ Acidity, and Alkalinity 8i ion concentration rather than in terms of the hydroxyl. One scale is therefore obtained, the hydrogen ion con- centration scale, instead of two diverging in opposite directions from the concentrations of the two ions in pure water, namely, from neutrality. It may be stated that the value given by Michaelis is Kj£,= I X at 22° C. Since the two ions are present in equal concentrations, each is present to the extent of 1X10“^ gram-equivalents per litre, for (ixio~^)x (i X io“^) = I X Since the equivalent weight of hydrogen is i and hydroxyl 1 7, the actual weights are for hydrion 1x10“^ and for hydroxylion 17x10“^ grams per litre. Each litre of pure water therefore contains one ten-thousandth of a milligram of hydrogen ion, and seventeen ten-thousandths of a milligram of hydroxyl ion. The pure water mentioned is, of course, the purest obtainable, free from traces of carbon dioxide, of ammonia and of salts derived from the solution of the glass vessel containing it. Acidity and Alkalinity. Having now defined neutrality as the condition in which the hydrogen and hydroxyl ions are present in equal numbers, we may define acidity as that state in which there are present a greater number of hydrogen ions than of hydroxyl, alkalinity as that in which the hydroxyl ions are the more numerous. In future we shall, as a rule, consider only the hydrogen ion concentra- tion, for the product of the two ions is constant. Thus if the hydrogen ions are increased tenfold, the hydroxyl are reduced tenfold, water being produced. At this stage it is necessary to consider the difference between hydrogen ion concentration and the total replace- able hydrogen as measured by titration. The former is a measure of the hydrogen ion effective at any instant, and it is this quantity, rather than the titratable acidity, that is 82 The Water concerned in regulating the precipitation of hydroxides — and therefore their solubility — the action of enzymes and many biological processes. For example, equal volumes of N/ioo hydrochloric and acetic acids require identical amounts of alkali for their neutralization, but the former solution has a far greater concentration of hydrogen ions than has the latter, since in dilute solutions hydrochloric acid is almost completely ionized, whereas acetic is not; it is accordingly said to be a weaker acid. If it be assumed that the ionization of hydrochloric acid is complete at this dilution, its concentration in terms of hydrogen ions is C,j or [H'^] = I X io“^ grams per litre, namely, lo mg. per litre. This may be written [H‘^] = , or pH = = 2. This symbol, pH, is thus defined as the logarithm of the reciprocal of the hydrogen ion concentra- tion, expressed in grams per litre, or as the negative power to which ten has to be raised to show the concentration in grams per litre. It may at first sight appear to be both cumbersome and unnatural to use such an expression as pH, viz. — /o^[H ' ], to denote hydrogen ion concentration, but in practice it is extremely convenient, and gives simple numerical values which are easily remembered. More- over, since an increase of unity in a pH value denotes a decrease to one-tenth in the hydrogen ion concentration, it is obvious that for any graphical presentation of results difficulties arise when the changes are of the order of 10“^ or io~®, yet these are quite commonly found. For certain purposes it is at times convenient to use [H''"] or C„ or even C„„ values ; as a rule, however, for sea-water, the use of Co„ values is to be considered undesirable as it neces- sitates the conversion of the data into pH values if any comparison is to be made with similar researches in fresh Bujfer Action 8 3 waters of low calcium content. In general the units used should be equally applicable to limnology and to marine investigations. An additional reason for using the notation is that in the fundamental electrometric determination of hydro- gen ion concentration the relation deduced is equated to log [H+J- It may be explained that by the measurement of the electromotive force of a cell which has different concentrations of hydrogen ions at each electrode it is possible to determine the hydrogen ion concentration of certain mixtures of acids, salts, and alkalis, so that these may be used as standards. By the use of a series of these ‘buffer’ mixtures with appropriate indicators, it is pos- sible to have a very convenient and instructive method for measuring hydrogen ion concentration. Buffer Action. If acid is added to alkali, after a certain amount has been run in, a neutral solution is obtained. With a strong acid, such as hydrochloric, the addition of the alkali in successive portions results in a progressive diminution in hydrogen ion concentration, as may be seen from the fact that the acid was largely ionized at the start. But with a weak acid the neutralization of the hydrogen ions existing at any instant results in a new equilibrium being attained by the remaining undissociated molecules of the acid, which become ionized to about the same per- centage as before. The alteration in the hydrogen ion concentration is therefore much less over a considerable range. If the results are plotted with />H values as or- dinates and cubic centimetres of alkali added as abscissae, the slope of the curve will be less steep for the weak acid than for the strong, except quite close to the neutral point. Curves illustrating this are given by W. M. Clark in his book, ‘The Determination of Hydrogen Ions’, which is 84 The Water the authoritative work on this subject. It contains direc- tions for making up standard buffer solutions, and has a useful coloured chart showing the tint assumed by various indicators at different hydrogen ion concentrations. This can be purchased separately, and suffices for approximate work, for which its use may replace that of standard buffer tubes. A weak acid thus has a considerable ‘buffer action’ in preventing rapid alterations in pYi value; the same is true of weak bases. Colorimetric Determination of Hydrogen Ion Con- centrations. As previously mentioned, various mixtures of weak acids or alkalis or of the corresponding salts are standardized by means of the electrical method. The mixtures usually made up cover a range of C„i x 10“^ to C,j I X 1 0"'° or I o~“, viz. />H i to pH i o or 1 1 . For this purpose various mixtures have been devised, the most widely used being those of Soerensen and of Clark and Tubs. The latter are made up at intervals of o-z in terms of pH. Methods have also been devised for colorimetric determinations without recourse to standard buffer solu- tions, by employing, in varying number, drops of indi- cators in pairs of test-tubes. In one set the full acid tint is developed ; in the other the full alkaline tint. On looking through the pairs of tubes intermediate tints are seen. The salts used for preparing buffer mixtures include phthalates, phosphates, and borates. Citric acid, boric acid, glycocoll, carbonates, and acetates have also been used. These are mixed in varying proportions either among themselves, with hydrochloric acid, or with sodium hydroxide. The most serviceable indicators are in the main mem- bers of the sulphone phthalein series. Of these dibromo thymol sulphone phthalein changes from yellow at pH6 to deep blue atpH7-6, being a clear green atpH7. It is Indicators 85 known as hrom-thymol blue for short, and is much used, as it covers the neutral point. On the alkaline side phenol red, cresol red, and thymol blue are reliable indicators; on the acid brom-cresol purple, brom-cresol green, brom- phenol blue and thymol blue (acid range) may be used. Methyl red covers the same range as brom-cresol green, but the latter is more permanent, and on this score has advantages. These indicators are made up in o-o2 or 0-04 per cent, solution; the addition of one drop per cubic centimetre of test solution is about right as a rule. With unbuffered solutions, such as ‘soft’ natural waters, the indicator should be reduced to a minimum so as not to alter the reaction of the water. It is well to keep the indi- cator at about the middle of its range, so as to minimize errors from this source. Thus brom-thymol blue made up ready for use should be neither yellow, in which case it may be too acid, nor blue, when it may have excess of alkali, but green — the colour which proves it has no excess of either. Colorimetric Determination oj Hydrogen Ion Concentration of Sea-W ater For this purpose the buffer mixtures of Clark and Lubs may be used, but a correction must be made owing to the effect of the salt upon the colour of the indicator. For the indicators recommended, viz. those of the sulphonc phthalein series, the effect is to give a result showing the water to be more alkaline than in reality it is. For the water off Plymouth in the open sea, o- 1 8 is to be sub- tracted from the pH value as found by comparison with Clark and Lubs buffer mixtures; for example, water which appeared to be pH 8-42 was in reality 8-14 using cresol red as indicator. This correction is made because the standard electrometric method is not thus affected by 86 The Water the salt. It is, however, convenient to have a set of buffer tubes specially made up for use with sea-water; such a set has been prepared by McClendon as follows. Using distilled water, boiled to remove carbon dioxide, a solution is made containing i8-6 gm. of boric acid and 22-5 gm. of sodium chloride to the litre. The boric acid used is recrystallized and dried in a desiccator, not by heat. The other solution contains 28-67 gm. of borax and 19 gm. of sodium chloride to the litre and has the same salt action on the indicators that the first solution has. The borax should be recrystallized and dried in dry air short of efflorescence, or it may be dried in a desiccator over a saturated solution of sodium bromide. If kept at a low temperature, borax may crystallize out of the stock solu- tion, and this must be made uniform before mixtures are made. The same applies to the mixtures. The propor- tions in which the two solutions should be mixed to give a series of standards suitable for use with sea-water are shown in the table. It is convenient to work with 10 c.c. of sea-water which should be as freshly drawn as possible, to this should be added 0-50 c.c. of the indicator, such as cresol red. The latter should be 0-02 per cent, and the solution should not be a clear yellow or a deep red, but Intermediate in colour between the two. The measurement of the amount of indicator should be carried out with care, using a pipette divided into hundredths of a cubic centimetre. The addition of drops of indicator which suffices for approximate determinations accurate to o- 1 on the pH scale, does not suffice for determinations carried out upon sea-water in which accuracy of o-oi should be aimed at. This is desirable because the annual variations are small in the open sea, and because the data already available have conformed to this standard. Resistant Glass Tubes 87 It should be mentioned that in making a comparison the sea-water tubes should be allowed to attain to the same temperature as those of the standard buffer tubes before a comparison is made. It facilitates comparison to examine the tubes side by side standing on a piece of white paper, across which parallel lines have been ruled, and to slant the tubes so that a greater thickness of liquid inter- venes between the eye and the lines ruled on the back- ground. The glass used for the tubes should be hard glass such as is prepared by makers in England and Scot- land. It is now quite unnecessary to import foreign glass for this purpose, although this is sometimes advocated — a superstition devoid of foundation. A simple way to test whether the glass is, or is not, sufficiently resistant is to wash it with distilled water, fill it with the same and place it in a bath of boiling water for one, or even two hours. On cooling and adding the indicator no pink colour should appear, whereas with soft glass the distilled water will have become markedly alkaline as shown by the intensity of the pink tint assumed by the cresol or phenol red. The tubes containing the standard buffers may be pro- tected against the air by closing them with small rubber caps, and have been known to keep for several months when thus protected. If desired, however, they may be drawn out and sealed up. To prevent the growth of moulds it is advisable to add a few drops of toluene to the boric and borax stock solutions from which the buffer tubes are prepared. The table (p. 8 9) gives in the first column the percentage of the boric mixture. In the second that of the borax. In the third column is shown the/>H of sea-water, which is 0-4 normal with respect to chloride. This value for pH is correct whatever indicator is used. In the fourth and fifth columns similar pH values are given for sea-water, 0-5 8 8 The Water normal and o-6 normal, corrections having been applied for the salt error of the sulphone phthalein indicators. In finding the normality of sea-water, allowance must be made for the fact that by the addition of o-fo c.c. of indicator to lo c.c. of water the original concentration has been altered somewhat. The water off Plymouth is not far from 0-5 normal, so that the buffer tube in the 0-5 column and shown as 8-15 is labelled 8- 14 for use with the water here. The tubes may be marked with a yellow grease pencil or more permanently with ink made by adding a spirit solution of crystal violet, or some such spirit soluble dye, to Friars’ balsam, as ordinarily sold at a pharmacy. If too much spirit is not added a mixture is obtained which flows evenly from an ordinary pen. Should too much spirit be added, if the stopper is allowed to remain out of the bottle a suitable consistency may be obtained. One further rather obvious 'precaution may be men- tioned, viz. that the diameter of the tubes used should be closely the same, 12 mm. is a suitable size. For ordinary use with water from the open sea, 0-5 normal as regards chloride, it suffices to make up the standards only between pVL 8- 00 and 8.30, viz. a set of eight. The table, however, ranges from />H 7-45 to 9-05, and for the extreme values it is necessary to use thymol blue at the alkaline end, and though cresol red maybe used at the acid end, phenol red is preferable. As examples of the pH. values met with the following may be quoted: the Faroe Islands results were obtained in May and June and the Mediterranean results between June and September. (See opposite page.) These results obtained by Palitzsch show clearly that the alkalinity decreases with depth. This is due to the fact that photosynthesis is carried on only in the upper of sea water ‘Boric* ‘Borax’ 0-5 N.; 0-6 N.; r.-if N j all sulphone sulphone per cent, per cent. indicators phthalcins phtbaleins 79*5 20-5 7-50 TAS 7-40 78-0 22*0 7-55 7-50 7-45 76-0 24-0 7*6o rss 7-50 74-0 26-0 7-63 7-60 7 - 5 ^ 72-0 0 ob rl 7.70 7-65 7-60 70*0 30-0 7-75 7-70 7-65 68*0 32-0 7-80 7'75 7-70 66-0 340 r»s 7-80 T7S 64*0 36-0 7.90 7 - 8 j 7-80 62'0 38-0 7-95 7.90 7-85 6o*o 40-0 8-00 7-95 7-90 58*0 42-0 8-05 8 00 7-95 56-0 44.0 8-10 8-05 8-00 54-0 46-0 8-15 8-10 8-05 52*0 48 -o 8*20 8-ij S-io 51-0 49-0 8*22 8-17 8-12 49*5 50-5 8-25 8-20 8-15 47-0 53-0 8-30 825 8-20 44*5 ss-s S-SS 8-30 825 42*0 58-0 8-40 8-35 8-30 39*5 6o'5 8-45 8-40 8-35 37*0 63-0 8-50 8-45 8-40 34'5 65-5 8-SS 8-50 8 - 4 J 32-0 68-0 8 -60 8-55 8-50 29-0 71-0 8-65 8-6o 8-55 260 74-0 8-70 865 8 60 23*0 77.0 8-75 8-70 8-65 20-0 8o-o 8-8o 8-75 8-70 I7'0 83-0 8-85 g-8o 8-75 14*0 860 8-90 885 8-8o 11*0 89-0 8-95 8-90 885 8-0 92-0 9*00 8-95 8 ,90 4*5 95-5 9*05 9*00 8-95 1*0 99-0 9*10 905 9*00 North Sea Atlantic Mediterranean Depth in East of West of between Sardinia Black Sea Metres Faroe Islands Portugal and Italy pH pH pH pH 0 8-13 8-22 8-23 8-34 100 8-09 813 S-2I 7-86 400 8-03 8-04 8-19 7 -S 3 t 1000 7 - 98 » 8-01 8-14 7-26 2000 T9S 8 -09 3200 8-07 700 metres. f H,S below 180 metres. 90 The Water and well-ilJuminated portions of the sea, though as the season progresses there is a considerable amount of vertical mixing. To illustrate the seasonal change in alkalinity and the increase in this variation as the shore is approached the following table for temperature, and salinity may be quoted for the Plymouth district. /H Temp. °C. Salinity % Aquarium tanks, normal 7 - 60 - 7 - 4 S tS-g-g-i 36- 14-36-85 do. abnormal 8*00-7*05 — — Rock pool 8*57-8*01 2I-4-8-2 — Shallow water 8*42-8*01 ig-z-S'Z — Li. Plymouth Sound 8-29—8*01 i6-3-8-2 35-00-30-69 Lz. do. Breakwater £i. About 20 miles out to sea 8-27—8*07 i6'i-8'8 35-I7-32-25 from the Breakwater 8*27—8*14 i 6 ' 2 - 9-9 35-40-35-I3 Minor Constituents oj Sea-Water In addition to the salts which make up the greater part of those dissolved in the water of the sea there are a large number of others. Of these certain ones, such as those of gold, are of no ascertained importance. Others, such as the salts of phosphoric, nitric, silicic, and arsenious acid are of very great importance, especially the phosphate and nitrate. These are taken up by plants, microscopic or fixed algae, for all of which phosphate and nitrate are essential; by some, the diatoms, silicate is used to form their outer covering and many take up arsenic, though it has not been proved that they cannot grow without it. Iron salts, too, are a necessity for plants, though only present in sea-water in minute amounts. Delicate colori- metric methods have been devised for estimating the traces of these indispensable salts, of which phosphate is present in winter around England to the extent of only about forty parts and nitrate nitrogen about two hundred parts per thousand million, or, what is the same, milli- Light 9 1 grams per cubic metre of water. In summer no phosphate or nitrate remains in the surface water, and these sub- stances may be much reduced in amount down to 70 m. or more. Since they are essential for plant growth, and since the entire animal population of the sea depends (just as on the land) ultimately upon plants for its food, we see that the abundance of life in the sea is regulated by the minute amounts of these salts — provided there is suffi- cient light for the plants to use them up completely, which there is not in winter ; neither is there enough light in the deeper waters of the ocean, which consequently act as storage houses for phosphate, nitrate, &c. Where rocks or banks occur in deep water the bottom water may be caused to well up, bringing its abundance of essential salts. Accordingly microscopic plants flourish, the zoo- plankton becomes enriched, and fish abound. Light, Transparency, and Colour The measurement of the actual illumination at any depth below the surface may be carried out in a variety of ways, such as by chemical methods, by the exposure of photographic plates at given depths, and by photoelectric cells. Mention of some of the results obtained by the latter method and their bearing on the distribution of plants is made on p. 168. It must be remembered, how- ever, that the composition of the light varies with depth, and that different methods may be most sensitive for dif- ferent regions of the spectruin, so that only a partial picture of the true light conditions is furnished by any one method. Interesting observations of the transparency of sea- water in different localities may be made by the simple method of lowering a white disk to the depth at which it disappears from sight. A circular disk, 20 to 30 cm. in 92 The Water diameter, coated with a fiat white paint, should be used. This should be lowered on the shady side of the ship to the depth at which it disappears; the depth at which it reappears on being hauled up should also be noted. Care must be taken to keep the line quite vertical, and a water telescope should be employed if feasible. The colour of the water should also be noted. Ill TROPICAL SHORES; CORAL REEFS AND ISLANDS BY J. STANLEY GARDINER A. TROPICAL SHORE COLLECTING The naturalist on his first visit to a tropical land is often as much disappointed with what he sees on the ocean shores, as he would be if he had transferred his activities from the relatively rich waters of Cornwall to the barren beaches of Picardy. He usually arrives at some con- tinental port, and he probably wanders to the sandy beach nearest to the open ocean. The tidal rise and fall is rela- tively small, and he is confronted with lines of breaking rollers, each swirling foam-flecked water several yards up the beach. Spring tides seldom rise and fall more than 24 in. beyond neaps, so that in the highest springs there are often no great flats exposed; even in dead calm weather the perpetually breaking waves. He, who wades out a few yards, feels the undertow, and he who is venturesome is swept off his feet to swim in on the surface waters, a ‘game’ only to be played on a rising tide. In any case he need not think he can collect in spite of difficulties, for there is almost nothing living to collect as the bottom is of sand in such a state of motion that it is suitable to few forms of life. The most the naturalist can do is to try and get some idea from the material which forms the beach, of what life exists in the waters outside. He will find there an epitome both of the complete bottom life to to fathoms at least and of the larger pelagic organisms (plankton) as well. The basal material of the beach depends on the 94 Tropical Shores neighbouring lands, but frequently more than half of its substance is derived from organisms still living in the sea. The colour is often extraordinarily white, due to organic remains, and spectacles of Crooke’s glass with toric lenses are called for. Omitting what is microscopic, the shells of the pelagic Foraminifera, which form globigerina ooze, may be picked out, all polished and spineless. Only about 30 species share in the formation of the 50 million square miles of this ooze, and generally the shells of about half of these can be found; several species of Globigerina with their joined series of chambers; Hastigerina, similar but set with thick rods of lime; Orbulina, the coiled shell en- closed in an outer sphere up to 2 mm. across ; Pulvinulina, a flat spiral of firmly joined chambers; Sphaeroidina, two visible chambers only; and Candeina a lumpy little mass of chambers differing in appearance in every view. Much broken fragmentary material will prove on examination to belong to the shells of various pelagic molluscs, Ptero- poda and Heteropoda, also ooze formers, but often pecu- liarly abundant near coasts. These are recognizable at once as being thin and fragile, almost transparent, but perfect forms of a few mm. long of the coiled Limacina, of the Creseis tube, of the Clio ‘vase’, and of the Cavolinia ‘seed husk’, each of the last three pointed at the base, may be sought. If the day is lucky, there will also be secured odd shells of the heteropods, Atlanta and Oxygurus, and high up the beach, probably blown over the crest, will assuredly be found an odd shell or two of lanthina, a thin coiled form of ultramarine colour 3 cm. across, accustomed usually to float on the surface of the ocean by means of a bubbly mucous mass secreted by the foot gland. The search for the latter generally reveals the coiled shell of Spirula about i J cm., its coils not touching each other and its surface quite polished and showing no trace of its Plant Remains on the Beach 95 chambers. Then there are ‘pens’ of various other cephalo- pods, calcareous to horny, and a very occasional shell of the female Argonauta, a float organ not homologous with any other shell. The above shells all belong to the ocean fauna, but are the first organic remains to be noticed, as, being of light texture, they lie upon the surface and are swept up the beaches. With them, especially on the upper ridge, if the land allows such to exist, will be found light plant remains, leaves of Cymodocea, Halophila, and Zostera, marine phanerogams, which materially help to bind together sand and mud below tide level; the first, by the accumulation of its grass-like dark leaves, often forms the ridge on pro- tected beaches. Conversely, there are few or no remains of Laminaria, Fucus, or any of the similar algae that are characteristic of temperate zones, but the Sargasso weed (Sargassum)may be found on any tropical beach. In place of these seaweeds we can collect small blue-green and green algae, made up of hair-like growths of filaments and of stems perhaps branching, of flat ‘leaves’ to low spread- ing plates, and of more or less tubular structures. All are small, some being naturally so, while others are stunted by the rush of the waves over them. There are numerous species of Caulerpae of many aspects, and then there are the Siphoneae, which are encrusted with carbonate of lime, this skeletal matter persisting as a hollow shell. They are of any shape, but all small, and all easily break up ; the species of Halimeda, bunches of flattened stems with disk- like ‘leaflets’ attached end to end, are the most abundant. Of other plant remains, those of the Lithothamnioneae, brown-red algae which deposit carbonates of lime and magnesium in their substance so that they form stone, will be found in the lower levels of the beach, unless cast up by a gale, and then generally bedded below lighter material. 96 Tropical Shores They show no structure, their lime being laid down densely, and they are peculiarly white and opaque and vary from tiny rounded braiichlets and marbles to large balls. They are not peculiar to the tropics, being indeed more abundant in temperate seas, where banks are often covered loosely with their little rounded but densely branched growths, and where the flat rocks at low spring tides are made brilliant by the red, orange, pink, and violet incrustations of the Squamariaceae. They are, however, more conspicuous here on account of the absence of sea- weeds, and for the reason that they grow to great size and thickness and resist the heaviest breakers. Animal remains from the bottom fauna cast upon our beach reveal the existence of all the same groups of animals as live in similar seaward positions in temperate zones, except that there is much greater variety in species, seldom that dominance of a very few kinds of bivalve and spiral molluscs, and of sea urchins and starfishes, often so characteristic of the temperate zones. Filamentous and incrusting Polyzoa are never so common, and starfishes are always everywhere relatively scarce in species ; bivalves are as numerous as in temperate zones, but spirally coiled gastropods are enormously abundant, especially the tiny forms, of which often the veriest novice may easily find over 50 species in a bucket of the sand. The larger shells are brilliantly coloured cowries, elongated gastropods equally gaudy, spider shells, and so on, either buried or cast up by gales above the tide; limpets and similar shells are rare. Then there will be found dried up leathery masses of various alcyonaceans and sponges, often gritty owing to their contained spicules, and fragments of all sorts of coral skeletons. The latter are mostly white, but some have dark red basal markings, and some violet tips. They comprise true colonial corals with plates (septa) in Animal Remains on the Beach 97 the cups which house their anemones (Madreporaria), flat plates (Millepora) and various branching species quite similar (Hydrocorallinae), with tiny round holes and much denser skeletons, blue plates (Heliopora), masses of tiny parallel red tubes fused together (Tubipora), and branch- ing and often anastomosing brown, red, yellow, or black branches, almost like dried plant remains (Gorgonaceae and Antipathariae). Unless the beach be a peculiar meeting-place of winds, tides, and currents, all of these can never be present in the same place in the old world, for the existence of true corals and Millepora implies coral flats extending some distance seawards, and of gorgonians steeper ocean slopes. Crustacean remains are all too much broken up to be of use, but there are few beaches on which crabs do not scamper, and the semi-land forms (Ocypode) of the upper dry part are interesting diggers, running down occasionally to the water to wet their gills. Worm remains are rare, dry or moist, as compared with our coasts, but in any case fresh animal remains are at once seized by these burrowing crabs or by Anomura (Hippa, Remipes, and Albunia) which live in the sand, following the edge of the tide as it ebbs and flows. This study of the beach is no waste of time on the part of the naturalist, for from his deductions he can, aided by charts, lay out his plan of campaign in accordance with his desires. He will have a fair idea of the bottom off his beach to 10 fathoms or more, and he will be able to judge whether the beasts he wants can or cannot occur there. He will, however, not be satisfied with his sandy beach alone, and doubtless he will wander to rocky points and into deep hays. Beaches of loose stones, pebbles, and rocks are the most hopeless of collecting grounds, and the pools below the same, unless protected from the waves, appear superficially to be almost barren of life. In bays 98 T ropkal Shores he may find upgrowing banks formed mainly of carbonate of lime skeletons, either of corals or of Lithothamnioneae. Both of these leave in their skeletons a quantity of organic matter, capable of digestion, and in consequence the rock they make is riddled by all sorts of boring organisms. The actual surface of rock is seldom entirely flat or rounded, and all sorts of nooks and crannies are left in which animals shelter. Where there is much branching of corals, the amount and variety of shelter is enormously enhanced, and a new fauna of fish and crustaceans is revealed, darting and crawling in and out of the stems. There is always much decay on organically formed shoals, and all sorts of crevices are found under the grow- ing masses of the building organisms, many of which will ultimately topple over; similar cavities are also formed by coral masses growing together. In all such places sand and bits of coral accumulate, while organic detritus gets swept in. The result is seen in a new set of animals that oecologically is divisible into four faunas. The first of these consists of sedentary animals which exist in the cracks owing to accident rather than to any adaptations, scanty and usually stunted growths of a very few species of hydroids, half a dozen solitary and colonial actinians, about as many Polyzoa, a few dark red or black colonial corals (Dendrophyllia) that shun the light, as well as a few tiny solitary corals, a vast number of little sponges with every kind of skeleton and of tiny colonies of encrusting tuni- cates, here and there little stony branching red skeletons of the foraminiferan Polytrema, a few tubicolous worms usually of small size, and finally odd pairs of alphaeid prawns, characterized by one large immense chela which snaps, its fellow quite tiny, sedentary because they here live in a matted tube of fine fibres. The second fauna may be termed the sand feeders. Sand Feeders 99 animals which obtain their nutrition by sending almost continuous streams of sand through their alimentary canals, this sand probably passing out of their bodies in a much finer state than that in which it entered. Holothurians and sea urchins belong here, but they have such strong methods of holding on that they can live in the open waters and need no protection. Some sea urchins are covered in the sand which they steadily pass through them, while others rasp off the growing surfaces of seden- tary animals; some live quite in the open and obtain pro- tection from the waves by wedging themselves bv thick spines, and others possess elongated pointed and poi- sonous spines. The larger wandering holothurians are what the Chinese call trepang and use for soup, these often having little fish (Fierasfer) darting in and out of their cloacae ; but in crevices under rocks and corals are many smaller and less coloured species, and there are a few Synapta as well, these characterized by their thin sticky skins, often variegated with every shade of colour, some of the most striking species up to 6 ft. in length by an inch thick but capable of any amount of expansion and con- traction. Then in the sand accumulations are sipunculids, all easily recognizable and colourless. Thalassema is rare, but is always worthy of attention on account of its fre- quently gaudy coloration and by its peristaltic waves of contraction, giving rings of rounded swellings in the meshes between its longitudinal and transverse muscles. Lastly, various enteropneusts may be expected, Prycho- dera, Spengelia, and so on, all colourless forms and all giving to the sand a peculiar iodoform smell, which is quite diagnostic of their presence, its use perhaps to keep other forms of life away out of their muddy sand. The third fauna consists of animals which live in sand but do not feed upon it, such as polychaetes, often species I oo Tropical Shores with highly objectionable spines, a few kinds of bivalves, nemerteans always scanty in numbers, but this the best site in which to find them, a few peculiar species of stoma- topods usually in pairs, and finally any of the sand-living crustaceans, here more by accident than choice. Fourthly may be recorded the regular free-living animals of all kinds, the home especially characteristic of certain minute fish which enjoy a ‘menage a deux’, certain short free- living chaetopods, some genera of chitons (Cryptoplax), and many crabs and prawns, some peculiarly flattened. Always difficult to collect in variety but here more diverse than elsewhere are the brittle stars, of which the species are few, the nudibranchs and the Turbellaria. Many of the latter groups are peculiar in possessing colorations of extraordinary brilliance, set in the most bizarre patterns, while often living in an environment of almost total darkness. All the above forms tend to be associated together wherever carbonate of lime is being deposited by organ- isms in sufficient quantity to form shoals of some sort and necessarily may occur on coral reefs. The surface of the latter to seaward is so pounded by waves and incrusted by Lithothamnioneae that it is not favourable to the collector. There are, however, areas where dead coral colonies tend to be piled together, and there are usually beaches with beach sandstone formations that readily split off revealing crevices rich in life. Only the corals themselves, the various calcareous algae, and the fauna associated with the spaces between their living stems are more varied on actual coral reefs. Here the association of all animals is essentially based on the corals, which may consist of a regular tropical jungle of many species of coral anemones growing in great luxuriance inside a lagoon to little stunted growths on outer reefs, or to isolated colonies fighting the sand and Corals I o I mud which tries to envelop them. These reef corals feed mainly by the help of their commensal algae, and therefore they are necessarily striving like plants to grow towards the light. Massive corals when they reach the surface die in the centre and grow outwards on the sides forming spreading masses. Plate-like forms grow outwards and upwards especially in crevices, and some are quite cup- like. Branching forms where exposed to the waves have broad tops formed of multitudes of small low stems of equal height, set on a central pedestal; staghorn species prefer to live where less exposed. Acropora (Madrepora) and some species of Montipora are the commonest branching forms, their skeletons perforate and hence pre- senting rather a sandy appearance when broken ; many of the branches of the former have violet tips, this colora- tion as well as the green due to the algae. Pocillopora is next in quantity among the branching species, its skeleton imperforate and much firmer and smoother with plates across its tiny tubes, its colonies never of great size but found everywhere, even in the heaviest waves, their colour green, light brown, and pink. Stylophora and branching Porites have more rounded stems, and often form round masses up to 3 ft. across, their bases bedded in sand. All these corals have small anemones, seldom more than i'5 mm. across and so contracted into their cups, when re- moved from the water, as to make their whole colonies appear almost lifeless. The largest polyps are those of such genera as Mussa and Euphyllia up to 2 or 3 in. across, the former a rough and spiny skeleton, while of intermediate size are a whole host of astreids and other forms mostly of massive facies. Fungia lie about on the smooth underlying rock, and their minute strobilating fixed stages may be found at the bases of large corals or in crevices. Of massive forms Porites colonies may be 102 Tropical Shores found up to 20 ft. or mure across, dead in the centre and even hollowed out, with a living edge of polyps. Usually present are yellowish plate-like growths, very smooth and stinging, set in masses several feet across (Millepora), and the blue coral (Heliopora) has a similar form, its surface film of living tissues dark green. Then there are great masses of the leathery corals (Sarcophytum, Lobophytum, &c.), whose side growths flop about like pigs’ ears, these sometimes living in quite muddy water, and occasional colonies of zoanthids and of Xenia, in places the latter almost covering the bottom. These latter show at once, even at low tide, their actinian afiinities, all other forms having their polyps sunk in their skeletons, their tentacles all retracted. Towards low tide this retraction is usually the case, but on a dull day at high tide some colonies can be seen expanded, then presenting an appearance of rich and luxuriant life. It is not a matter of light and darkness alone, for they show little difference in day and night. Next there are the free-living coral associates. There are on an average reef about two dozen species of tiny fish that never live anywhere save between the stems of the living and branching coral colonies, and there are at least as many more who are indifferent as to whether the corals are alive or not. Many of these fish are laterally flattened, the better to penetrate the spaces, and all are provided with markings, stripes, splashes, and iridescences that look ex- traordinarily conspicuous to the human watcher in the air, but which may appear quite otherwise, if viewed from below the surface of the water. It is amusing to see them darting in and out of the corals, pairs or small shoals after bits of broken up fish or other food, or to catch them on tiny hooks dipped in the ink of the cuttlefish, preferably dashed with some strong scent. There are twice as many kinds of little crabs and they too are much coloured; Estuarine Shore 103 Melia is never common, but it is peculiarly interesting as it carries about in its great claws a pair of small anemones, doubtless to make use of their stinging powers. A fauna of peculiar prawns and alphaeids, there is too, many quite civilized by being apparently paired for life. In addition, all forms of free-living animals, whether of the surface, or of the holes in the reef, may be found, but they prefer the semi-protected spaces at the bases of the corals and do not seek the living stems. Occasional caprellids, pycnogonids, and crinoids are visitors from deeper waters, where their association is with branching gorgonians and forms of similar growth. As a last environment there is the estuarine shore, which has its own peculiar physical and chemical condi- tions, and which is very profoundly influenced by the generally luxuriant vegetation of the country behind. There is a tendency for great flats here, and for these flats, unless controlled by man, to be overgrown with man- groves, which ever march by their stilt-like roots further and further out into the water, holding up the mud. The mangroves above water have their own peculiar fauna of birds, insects, and even small mammals, while the mud beneath at low tide is the home of countless burrowing crabs (especially the fiddler crab, Gelasimus, with its one huge chela), bivalves and worms, while fish move up be- tween their roots with the rising tide. Frequently excel- lent oysters grow on the roots and gastropods cling to them, but nothing eats the mangrove bark or bores into the wood. The mud itself is usually stinking and has strong tanning properties, and how any animal can live is as difficult to understand as Arenicola and Echiuridae in the muds of our east coast estuaries. The species are few in number, and the interest is the study of what love it — and how they are adapted to it — and of what tolerate 1 04 Tropical Shores it. The same species of Diogenes-like hermit crabs are found here as in every other environment, these scavengers being the least specialized animals in this respect. The naturalist, having completed a reconnaissance of all the environments to be found on his tropical shore and having perhaps spent a week thereon, is now in the posi- tion to undertake the job for which he has come and to collect and observe for his own satisfaction. Let him once and for all understand that the scientific day is passed for miscellaneous collections of all forms of life, merely to record the topographical distribution of marine organisms. Such collections will not be welcomed in museums, and most of them will simply be allowed to rot, unless they have attached to them notes, as to how and where the separate beasts lived, of such excellence as to attract systematists. These notes to be so characterized may have to comprise the date and time of collection, the nature of the environment, the temperatures of the water and sand, the exposure to sunlight, the salinity and alkalinity or acidity (pH) of the water, &c., while the dates of new moons may be mentioned, since it is clear that many marine animals shed their generative organs according to a certain lunar periodicity. Many of such observations savour of pure research on individual animals or associa- tions of such, and this is so true that we would only add, for the encouragement of acquisitive naturalists, that any exhaustive collection of any phylum of organisms is of great and permanent value to science, especially if date, locality, depth, and nature of the bottom is recorded on each label. The equipment for collecting on the shore is of the simplest character, a knapsack full of corked tubes of dilferent sizes and with two pockets for full and empty Equipment Jor Collecting 105 tubes, a little vial of neutral formic aldehyde, so that pre- serving fluids for delicate specimens may be made on the spot by pouring a few drops on to the beast, when it is in a tube of sea-water; Bouin’s fluid is also most valuable. In the belt is a sheath knife, one or two pairs of strong forceps and of course a hammer, and in the hand a light crowbar of fine steel, with one end clawed and the other chiselled, used to turn over masses of rock and to delve into holes. Bone forceps are often most useful. On open shores the naturalist will bring his (native) boy with him, providing him with a bucket or two and some form of hand net or other. As soon as all vessels are filled up with animals and plants, go home and at once sort the bucket into three or more enamelled basins — coloured are often preferable to white — filling up with fresh sea-water. In any case, fill up one bucket tight with weed, decaying coral masses, &c., and treat it in the same way. Now pick out after an interval all that is visible, and on the weed left in each basin throw a handful of epsom salts and a fresh crop of animals will, in two or three hours, have revealed itself. The contents of the basins can be put to- gether and may be further dosed and left overnight, but the organisms of this lot will be ready to disintegrate and must before bottling be well hardened in 90 per cent, alcohol. It is often advisable to bottle all the ‘dirt’ to- gether, not sorting at the time, gradually decanting off the alcohol and adding fresh until the preserving fluid is of sufficient strength. The most hopeless looking material when narcotized often yields rare forms. Each bottle or tube must have its label in the fluid, the carbon duplicate of which remains in the book. Many of the organisms whose skeletons were found on our exposed beach may not be secured, or the coast may be so entirely unsuitable to such collecting that other io6 Tropical Shores means may have to be devised.^ Such means entail various forms of nets, dredges, and swabs, but there is perhaps no tropical land where these may not be made on the spot and as cheaply as at home. The natives in every part of the world use seine nets and in most weighted nets, cast from the shoulder; these are indispensable methods of collecting fish and all free-living swimming animals. If small forms are desired, baby seines of mosquito netting may readily be run up with the needle, weighted with bits of stone and floated by any light dry wood in the absence of cork. The principle of stake nets may be developed, and natives too usually have some forms of movable traps. Heaps of stone and coral may be made on sandy flats, their tops six inches below the surface at low tide, and baited with pounded up crabs and dead fish remains; free-living animals, especially fish and crusta- ceans, after a few days find out such heaps, and all these will be found ‘at home’ at low tide. They are then all readily caught in a large basket of palm leaves, which is kept open by a few masses of branching corals; this is placed, mouth open, on the opposite end of the heap to that which the collector is turning over. The strands of hempen ropes may be unravelled, to make swabs such as are used for drying up the decks of vessels. These, swung on to the bottom, with a few fathoms of rope to spare, over the stern of a vessel at anchor and swinging to the tide, if taken up and cleaned every twelve hours or so, will often bring up a great variety of organisms, pieces of rock with a couple of dozen kinds of animals attached or clinging to them, numerous crusta- ceans, starfishes, molluscs, worms, and other forms. A I There is usually a rich fauna to be found attached to harbour or break- water piles, and harbourmasters are usually sympathetic to their investigation, while their men will often for a few coins pull up the suspended chains of moorings; note the period since their last cleaning. The Dredge 107 swab or a dredge may he dropped out from the ship or shore as far as the length of warp will allow and hauled in. The dredge had better be, for such hand work, rectangular or triangular, its swords not more than 18 in. long by 23 in. across, set at angles of about 30° ; each sword has to be perforated at its thick edge for the attachment of the net, which may be of cord of about as small mesh as the local netter can manage. In all cases where rock material is thus obtained, it should be anaesthetized for boring worms and hidden organisms. A pair of plankton nets floating out from the ship at the same time, in one-third and two- thirds of the depth of the anchorage, will indicate the floating fauna, on which the animals of the bottom so largely feed. When the anchor is raised, mud or sand is often found clinging to it, and a good sample should cer- tainly be secured and dried, after putting any living organisms that may be found into some preserving fluid. A preliminary washing in fresh water and treatment with a little weak spirit, which otherwise would have been cast away, will prevent any objectionable smell from the sand when in process of drying. Our naturalist will doubtless visit and search every possible habitat. If he is studying some particular groups or phyla of animals, he will be adopting a converse method to the author by considering his animals from the point of their structure, thinking out what kinds of bottom and of conditions are suited to them and so should be sought. Thus Amphioxus of different genera all live in sand, but such sand must be clean (sharp, as it is termed) and pre- ferably in such a position that it is slightly stirred at every rise and fall of tide. If present, Amphioxus is generally local, but where it occurs it is enormously abundant. Large numbers may be readily obtained by sifting the sand through one or a series of copper-bottomed sieves of io8 Tropical Shores different mesh, fitted into one another; these are held, while being shaken, just submerged in the sea-water, and this method is one which may also be used for all small sand and mud-loving animals. Equally sand loving are the enteropneusts, but they prefer either rather dirty sand flats into which they burrow (throwing up great ‘mole’ hills of the sand which has passed through their bodies) or shallow accumulations of sand under stones and living corals; their iodoform smell has already been mentioned. The author’s recipe for these and all sand-loving animals is to find a flat of sand with rock a foot or two below. The collector will see at once if there are any mounds or sand- castings, and he will probably dig with success, using shovel or spade, while, if the sand is deep, he would follow his animals’ burrows until he himself is submerged. After capture the enteropneusts should be kept in the shade in basins of sea-water, changed every three or four hours, until, by next day, they have shed the sand of which their bodies were full. They will then be killed and hardened and got into preserving fluid. The latter, what- ever is used, must be subsequently changed at least twice, since these beasts possess in their bodies some substances, which in the tropical heat acts on the usual preservatives tending to decompose them, their own bodies disinte- grating at the same time. Worms (spiny Chaetopoda, Echiuroidea and Sipun- culoidea) are found in still, muddy sand, and the mounds of many sipunculoids are difficult to distinguish from those of enteropneusts. Tunnel living stomatopods, such as the large Lysiosquilla, may also be occasionally dug. In addition there are many Chaetopoda which form tubes on rocks and in sand, and members of this order and of the Sipunculoidea are among the most important of boring organisms. Polychaete worms need handling with special Collecting 1 09 care to secure as many unbroken specimens as possible; the head in all cases must be present. The slender thread- like Drilonereidae are not uncommon but break up into separate segments. Forms in coral can only be extracted after narcotization. Amphinomidae, large forms with white bristles, which in the fingers cause sores, should be handled with forceps and never bottled with soft bodied animals. The outsides of echinoderms should be examined for polynoids, short scaled worms coloured as their hosts. The bug-like Myzostoma lives on crinoids. The adult Bispira with its double spiral plumes has the mouth of its tube level with the surface of Porites or is in the stem of a branching coral ; it is not to be confused with the proso- branch Vermetus, which has a transparent brown or pinkish tube. The tubes of many sabellids and terebellids are inconspicuous, level with the surface of the rock. There is no royal road to collecting, save only work. It is well to half fill a boat with masses of rotten calcareous rocks and of dead and living corals. Then, when the rise of the tide makes further work in the sea impossible, sit down with hammer, chisels, and bone forceps and break up these masses into fragments, carefully chipping out each animal. Any promising pieces may be thrown into a bucket and left for the night, and all free-moving animals will soon get starved for oxygen and come out. Around will be a circle of basins, for the catch will be a varied one. There will be Crustacea and Mollusca of all kinds, a few small fish, some flatworms (Turbellaria), and an odd nemertean perchance, all living in the crannies of the rock masses; a few sedentary animals will also be picked off, and good samples of these should be put in tubes at once, for species and colonies are difficult to sort apart subsequently, especially from this environment of stunted growths. Then in the rock or coral itself will be I lo Tropical Shores more crustaceans, such as Hapalocarcinus and Cryp- tochirus, which have been grown over by the coral, and fleshy barnacles. There are also true borers, multitudes of worms, molluscs such as the ‘date stone’ (Lithodomus) and a few gastropods, masses of brown boring sponges, and many more animals of odd groups, such as octopods, or cuttle-fish, which have annexed convenient spaces; large forms of the latter live in holes in the reef rock, from which they are hooked out with difficulty. Of other free-living animals there are the sea spiders or Pycnogonida, which should be placed in small tubes of strong spirit, not more than two or three together. They crawl on the fine branches and stems of weeds, hydroids, and especially Polyzoa. The Turbellaria dwell mainly on the under surfaces of stones, crawling between the seden- tary sea-squirts (Tunicata) and taking refuge in any hollows of the rocks. Most are small, but Pseudoceros is sometimes six inches in length. Each can be induced to creep into a clean tube held under the water against the surface of the stone ; by this means they can be brought in alive, for, in the present state of our knowledge, coloured drawings of these animals are necessary. Their preserva- tion is always difficult as they tend to crinkle up, but a few drops of formic aldehyde poured into each tube, which should not have more than one animal, is as good as any- thing. The same remarks apply to the nudibranch, Chromodoris, which has been best preserved by being made to crawl on to wet paper, which is then plunged into a fixative; this method should be developed for Turbel- laria and means might be found for keeping them flat. Bottom living Foraminifera can be seen creeping up the glass sides of any tumbler cr beaker, into which some clean sand from below tide level has been placed. We know little of their life histories even in temperate seas. Sedentary Animals 1 1 1 and tropical forms are mainly of interest for comparison of their shells with those in various elevated rocks ; we get their shells in sand samples, which we have probably dried for more general purposes. Of insects we may mention the bug, Halobates, often found on rock pools, though a wanderer over all tropical oceans ; its eggs and larvae are found on floating cuttle bones and pumice, frequently being confused with young barnacles. Needless to say any insects which creep down into salt water, or live in brackish water marshes, should be carefully collected and preserved in 70 per cent, alcohol. Lastly, we have snakes (Hydrus), dangerous pit vipers, but withal timid beasts, incapable of harm so long as the body is covered and the thin parts of the hands are kept well above the surface of the water. They must not be confused with the smaller eels of the coral reefs, of which there are about half a dozen forms, some mottled. Of sedentary animals we might make a catalogue of hydroids, Alcyonaria, anemones, corals, tunicates, Polyzoa, and sponges, and remark that pieces placed in 70 per cent, alcohol, if possible anaesthetized and expanded, with small slips both in 90 per cent, alcohol and in 4 per cent, formic aldehyde, are likely to be useful to specialists. We are, however, beset with difficulties at every turn in re- spect to the determination of the species of these forms. They have for the most part floating eggs or larvae or young, which can be carried by ocean currents over wide stretches of sea, and accordingly present little variation of species over each of our two great tropical oceans, the Atlantic and Indo-Pacific. Let the naturalist collect a few species only, but let him study the growth of these few carefully on the spot in relation to their environment, its detailed topography, the incidence of the sun’s rays, the amount of the currents, and how far the water is clean and 1 12 Tropical Shores carries plankton and land detritus, &c. It does not matter whether he knows at the time the names of his genera and species or not, for really intensive scientific studies of tropical sedentary animals have never yet been made. He will have the interest of knowing that he is adding some- thing of high permanent value to biological science and he will be undertaking a study, the interest in which grows as the work continues.* Research Problems of Tropical Shores The dedication of the altar of science is to ‘the unknown god’. This implies that the devotee has made acquaint- ance with what has been revealed and is willing to see what lies obviously in his field of vision. How has the shore and the submarine slope off it been formed.? The contour of the land should be carefully noted, particularly its valleys and slopes, since the underlying rock of the sea floor is bound to be correlated with that of which the land is composed, masked probably by a superficial deposit of sand or mud. How far the scenery is due to rain, wind, heat and cold will give a clue to the antiquity of the land. If it is recent volcanic country, the steep slope, which commences off nearly all lands at about 50 fathoms and extends down to about 150 fathoms, will be convex; if it is ancient or its slope due to coralline agencies, it will be concave. The ‘steep’ may have been directly or indirectly ‘ The statements made in this section are largely based on the author’s experience in the Indo-Pacific, but have been checked by Mr. F. A. Potts, and more particularly by Dr. Cyril Crossland. Exceptions to nearly every fact of topographical distribution could be given, for instance localities where the naturalist can wade through gorgonians with their associated crinoids, or through thick masses of attached Sargassum.or over reefs where corals are almost non-existent. As to the West Indies, almost every paragraph requires some punctuation by exceptions, the oecological conditions of this region being clearly different from those of any part of the Indo-Pacific; this matters less as it pos- sesses an abundant literature, mostly emanating from the United States. Research Problems 113 due to faulting, and the land-slope down to it may have once been part of the land, absolutely cut out of it or separated by a coast line due to faulting, tilting, or slipping. ‘ This area may have been extended, or be extending, further and further seaward by land sediments or coral growths, or alternatively perhaps be washing away. The earth has by no means an unbreakable crust, but the latter has a fluid layer beneath it, and its vertical position is influenced by alterations in the weights that overlie it, and perpetually recurring elevations and subsidences are patent to any traveller in most tropical island groups. These observations give the key to the land slope, and it is essential to hold this key, if the secrets of geographical distribution are to be unlocked, this also being an essential primary consideration in the study of the organisms of any particular shore environment. The ocean is one mass of salt water, not a series of isolated masses. Its organisms either in the egg, young, or adult condition have the widest powers of dispersal. Nearly all groups of them are of inconceivable antiquity, and isolations affected by land — the Indian Ocean was open to the Atlantic in the cretaceous period and the latter to the Pacific in the miocene — are quite recent as com- pared with the aeons of their existence. There are polar, temperate, and tropical areas separated by temperatures, with certain species of organisms apparently peculiar to each. There are the east and west coast areas of con- tinents, supposed to be characterized respectively by being warm and cold water areas, and by having reef growths and almost bare shores. The western tropical coasts are partly regions, where the counter equatorial ^ The possibility of ihe fli^aling apart of rontinenis and islands (We};cncr*s hypothesis) need not cojiceni usj the explanation of the facts of geographical distrbutlon of organisms does not require any such theory. 3166 1 1 14 Tropical Shores currents of great heat approach the shore; a full investi- gation of the whole biology (fauna, flora, physical, and chemical conditions) of one of these regions would be of great value, preferably the north-west coast of South America, comparison being possible with the relatively well-known fauna and flora of the West Indies. The slope off Walfisch Bay would be of interest in connexion with the faunistic work of the Fisheries Survey of South Africa. There is a belt of relatively cool water across the bay of Aden causing an isolation of the Red Sea, an area extraordinarily rich in life and reefs; this is in contrast to the shores of Somaliland, of South Arabia, and the whole Persian Gulf, while the west shores of India and Malay would seem to be richer than the east. Why does the reef growth on the east side of Africa stop long before the Mozambique Channel is reached.? Can it be only tem- perature here with the warm Agulhas current constantly flowing down from the north after crossing the hottest of the oceans.? Assuming that these statements are correct — and we have not precise facts enough to be certain that they are — can we believe that temperature is the main factor, while the brethren of the same organisms are quite tolerant of great ranges of temperature in temperate seas.? The best-known case, that of corals requiring a tempera- ture of at least 20° C., seems to us to require definite investigations, not only by inquiries as to the conditions where these animals occur, but also by experimentation, especially since their chief assistants as reef builders (Lithothamnioneae) live happily at lats. 60° N. and S. Why are there no reefs at Galapagos and off Panama, since, according to Dr. Crossland, Pocillopora, Porites, and other corals in smaller quantities, as well as Lithotham- nioneae and Vermetus all live there.? And why alone in the tropics are there dense shore algae oft' Galapagos.? Are Depth Distribution 1 1 5 there cold currents from north or south or an upwelling of deep cold water? We probably cannot answer the question about currents without the use of a ship, but we can attempt to determine such factors as are present in the water itself, and presumably govern the life, growth, and reproduction of our organisms. Another distribution of marine organisms is by depth, and here the top zone is in connexion with the photosyn- thetic power of the sun’s rays acting by means of chloro- phyll. Marine plants, except phanerogams, will not grow on pure mud at any depth, but this is largely a question of the difficulty of attachment. In muddy waters the in- sufficiency of light prevents the growth of algae below 10 fathoms, while in clearer waters, especially off oceanic coral reefs, they may be found in vigorous growth at 30 fathoms, and apparently existing quite happily at 40 fathoms. The plankton is beyond our scope, but its im- portance is obvious so far as the nutrition of our beasts is concerned. First in importance, as probably in antiquity after the bacteria, are the flagellates, and these often cloud the sea near tropical continental shores, many not to be distinguished as animals or plants save by their green colouration, some conceivably at times with, and at others without, chlorophyll. Less coastal are the peridinians, and who can prove whether these are animals or plants, or an evolution independent of either of the main stems? Why do they swarm at times in incredible numbers? All we can say is that this is not meaningless, but whether it is cor- related with unknown physical or chemical conditions or is possibly cyclical we cannot say. Similar causes pro- bably affect many other organisms, directly or indirectly, and in many places any failure in the plankton organisms concerned would convert a shore luxuriant in animal life into a barren waste. Then there are the diatoms, less 1 1 6 Tropical Shores numerous in species in the tropics than in temperate seas. They are in distribution some more coastal, some more oceanic, and some peculiar to each region. Quantitative studies of plankton show relatively few in the tropics, but occasionally their frustules are so numerous in the digestive cavities of such marine animals as do not pre- sumably select their food, that the observer is inclined to doubt the present beliefs. The Radiolaria belong here; observations are required on every phase of their life-his- tories, on their reproduction and on their physiology. The author has never found Ciliates abundant in clear oceanic tropical waters, though often swarming in harbours or near city drainage. The initiatory phases of much of the work required belongs to the fixed laboratories of tem- perate zones, but the tropical naturalist can scarcely escape the considerations that are entailed in dealing with his shore fauna and can accept almost nothing as deter- mined. Below the light zone comes a mixed assemblage of all sorts of bottom animals feeding on the detritus of the organisms above and continuing downwards to lOO fathoms or more. It has often been referred to as a specialized fauna, but it commences at whatever depth the light-loving organisms cease to exist. Many of its or- ganisms have been stated to be confined to specific depths, but continued research has increased the ranges of very many. The bottom in some places is so rough that a dredge or trawl can only be hauled for a few yards. Off continental shores it is often smooth and then hauls of even miles are possible; the contents of the bags then make widely scattered and rare animals appear common, and thus quite erroneous ideas arise as to the biological condi- tions of the area. Away from the land or reef the bottom becomes with depth progressively covered with smaller Food oj Marine Animals 1 1 7 sediments,' and this of course is supposed to inhibit the fixation of many sedentary animals, these being presum- ably absent from a continental ‘steep’, but certainly helpful in maintaining the similar ‘steep’ off coral reefs. The problem is largely that of Sir John Murray’s mud line, but it is not a simple problem. There must be determined by observations, not by deductions, the physical and chemical conditions monthly for a year over whole sectors, from the shore to at least 200 fathoms, and the fauna at each depth station must be stated with accuracy, not only qualitatively but quantitatively as well. Trawling and dredging, while furnishing useful checks, are not accurate enough, and recourse must be had to Petersen’s or other grabs of known area. These should throw a flood of light on the question of seasonal migrations in depth, and interesting comparisons will be possible with temperate regions, the quantity of life obtained being enumerated or otherwise measured. The study of the food of marine animals in relation to the general economy of the sea can well fill up all intervals of time. The preservation of specimens for this purpose and the subsequent examination of their gut contents is of little use, because the gut contents are often ejected on preservation, or, if retained, so mass together as to be indeterminable; indications as to the food are often re- quired on the spot, so as to enable any investigator of the continental slope to conduct his researches along reason- able lines. It is so evident that a study of the food of each group of animals is necessary that any selection by the author of special organisms for such research purposes is perhaps unwise. However, on tropical shores, especially in coral regions, boring and sand-feeding organisms may I Considerable chemical activity may be visible in the deposits of this area as instanced by phosphatic concretions, glauconite casts, and manganese nodules. 1 1 8 Tropical Shores play a conspicuous part in the formation of the coral mud. In coral skeletons the alga Achyla and the sponge Clione are of almost universal distribution, and these are suc- ceeded in turn by numerous chaetopods, sipunculids, Lithodomusand other molluscs, and certain animals which enlarge the holes made by these. Many forms bore right up to the living polyps, and their advance is so rapid that they clearly have an efficient nutrition. The coral may be broken off, rolled about, killed, and so rotted that it is easily broken down into fragments, or it may be crushed up by flat-jawed fish, or rasped to pieces by sea-urchins and molluscs. Much of it is now in the condition in which it can be utilized by sand-feeding organisms, such as other sea-urchins, holothurians, sipunculids,’ echiurids, en- teropneusts, and chaetopods, most of whom show adapta- tions to this food in ciliated grooves or siphons in the digestive organs. How far is sand dissolved or rubbed down into mud when thus eaten? How much is con- sumed daily? What are the estimated numbers of these beasts in different areas? And what is their estimated effect? Why do some holothurians shed and regenerate their alimentary canals ? Is there a regular autotomizing of the hind part of the trunk in enteropneusts? Or is this due merely to man’s interference? What function does the iodoform smell of these last indicate? In the regions affected by sand feeders are many free-living animals, and their feeding is interesting too as well as their adaptations by swimming or otherwise so as to avoid burial. The feeding and keeping clean of sedentary animals are other * This group is so little known that it would probably repay attention and be worth collecting thoroughly. In doing so a few nemerteans will be obtained, and we require knowledge as to the function of the proboscis of the Hetero- nemertea, the method of feeding and the food of all forms, and the method of contraction of the body down to lo per cent, or less of the length of the living beasts, some of which are over lOO ft. long. Breeding 1 1 9 prublems. How fur do they select their food? Do they feed at particular times and is their feeding affected by light and darkness, rain, temperatures, &c.? Reef corals are a special case, in that they often have their body spaces crowded with commensal algae, in some anemones the stomodoea being completely closed. Do they take in solid food at all, or are they by this curious association prac- tically holophytic? The latter might account for their limit in depth, viz. to the light zone, but not for the well- marked and numerous granular gland cells in their sto- modoea and mesenterial filaments. What are the functions of the nematocysts which crowd these two organs? And lastly what are the especial characters which enable Millepora, Pavona, Pocillopora, and certain branching species of Porites to resist mud longer than all other forms? In all these observations the action of mud as a sediment and as an organ of rasping function, when being carried in suspension, must be carefully distinguished. To recommend embryological problems is beyond our scope, but observations on the breeding of all organisms of our tropical shores are urgently needed. Are sedentary ‘colonial’ animals unisexual, protandrous, or protogynous? Do all forms ripen their eggs at fixed seasons or is the ripening merely correlated with food and growth? What is the actuating cause in bringing about the dehiscence of the eggs and sperm? How far do all forms of a species breed simultaneously even if in different environments? The advantage of doing so is obvious, but why does Eunice select a particular full moon for her operations and why do certain echinids exhibit similar phenomena’ Where there are marked seasons, hot and cold, abundance and absence of food, much may be understood, but are marine forms of the tropics, which are not seasonal, more prolific than temperate forms? They are supposed to 120 T Topical Shores grow faster, but the accurate measurements of their growth rates is worthy of attention. The setting of cleaned stones, piles, and floats is useful and may be under- taken especially in connexion with specialized embryo- logical problems. Fishery authorities require a technique by means of which to determine the age of tropical fish. Lastly, how far do fish, Crustacea, nudibranchs, &c., exhibit seasonal breeding migrations.? To the naturalist the little-known economy of the tropical organisms is so bewildering that his chief danger lies in a lack of concentration. He sees a vast variety of species, and no one form is so dominant as to compel attention. Probably three-quarters of the species of animals he brings home were caught in habitats other than those in which they flourish at their optima. Almost every sedentary beast he looks at has its associates, crabs and other Crustacea with all, pycnogonids, caprellids and crinoids, ophiurids, fish, and crawling worms with most, occasional galls in the stems, &c. Holothurians have their fish, crabs, gastropods, and worms. Hermit crabs have their epizoanthus, hydroids, and anemones. Hetero- psammia has its sipunculid. Can we be content with merely recording such associations without thinking how all concerned come together and may be affected.? Then there are the wonderful colours of many forms. The fish of very shallow waters are most conspicuous, and explana- tions have been offered depending on the incidence of direct and reflected lights in relation to the environment, suggesting that they subserve certain functions of protec- tion. Of what use is the extraordinary coloration of the mantle edges of Tridacna, or that of turbellarians and nudibranchs, so many of which live in darkness, and actively shun the light.? But why go on further.? An intensive study of the daily life of almost any living Islands 1 1 1 animal will yield secrets valuable to science. Let the research worker see what interests him, and having selected his problem in relation to his own technical knowledge, let him pursue it to the exclusion of all else; he will, however, save time in the end by ‘wasting’ a few days in trying to visualize the interrelationships of his shore beasts, before he sets to work on his own more specific theme. B. CORAL REEFS AND ISLANDS Islands divide themselves into three classes, con- tinental, volcanic, and coralline, in accordance with the composition of their rocks, though it is not always easy to separate them. A continental island is one which, from the nature of its rocks, or more generally from the con- siderable variety of the animals and plants which live upon it, must be supposed to have been at one time connected with some continental mass of land, upon the surface of which, according to our present views, a large variety of animals and plants may be supposed to have been evolved. Thus, for example, all the greater islands of the East and West Indies must have been connected at one time with their neighbouring continents. There is little room for doubting (see Fig. 13) that Madagascar was at one time joined to Africa, and New Zealand to Asia and Australia through New Guinea; New Zealand may have had also a polar connexion with South America. The Seychelles Islands are formed of granite, a type of rock essentially associated with the immense early solidifications of erup- tions which formed great continents. They doubtless were a part of the Indo-African Continen't,' whicl^ is 122 Coral Reejs and Islands believed by many to have once extended between Africa and India — ‘Lemuria’, as it is often termed. The Fij i Archipelago is a more doubtful case ; its rocks are of recent volcanic types, overlaid by ‘soapstone’, com- posed of submarine deposits formed largely by Foramini- fera and raised limestones. Its fauna and flora, however, are, comparatively sneaking, rich and varied, but it must always be remembered that the numbers of species of animals and plants on any ancient land vary proportion- ately to its size, temperature, varieties of environment and rainfall; the influences of each of these cannot, of course, be exactly stated. In the case of islands formed by sub- marine eruptions, their proximity to other lands, and the presence or absence of currents and winds, suitable for the transport of the seeds and germs of plants and animals, are of primary importance. Hawaii, Samoa, and Tahiti were probably formed by eruptions, and to some degree they illustrate the influence of distance and size already mentioned. Mauritius, Reunion, and Rodriguez are pro- bably further examples; but the former may have been connected by intervening islands to the Seychelles, and so to continents; Rodriguez was partially formed by an upheaval of the sea floor. The Atlantic islands — the Canaries, Cape Verdes, and Azores — are probably all of recent volcanic origin, though some of them have been supposed to exist on part of the site of a continent once connecting South America to Africa. It will be now clear that the importance of the study of islands to a large degree depends on the light which they throw on the existence of former continents — on the dis- tribution of land and sea in past geological periods. An island may exist where we are practically certain that there was formerly continental land, but it by no means follows that the island was ever part of that continent, since it Coral Islands 123 may have been formed subsequently to the submergence of the continent. Yet it is in islands that ve might expect to find our best evidence of the existence of former con- necting land masses. The interest of our third class of islands — coral islands — is again different, and lies in the question, how far do they indicate the sites of former land masses which have disappeared owing to subsidence or to other causes.^ Some Fig. 13. Chan of the World, showing the supposed distribution of Land (dotted) in the Cretaceous Period of these lands have been elevated to great heights, 1,000 feet or more, and are surrounded by reefs of a later forma- tion. Such islands may be isolated, rising out of depths of 1,000 or 2,000 fathoms. Examples of these are Christmas Island in the Indian Ocean, Metia and some of the Pelews in the Pacific, and Bermuda in the Atlantic. All elevated ‘coral’ islands are not correctly so termed, for example Bermuda is formed of shell sand, the Bahamas of a kind of oolite, and Barbados of a comparatively deep sea ooze succeeded by miocene and pliocene deposits of medium depth. Indeed, there is much variation in the elevated limestones of the West Indies, while those of the Indo- 124 Coral Reejs and Ishfnds Pacific arc largely composed of coralline rocks.’ In Fiji high elevated coral islands are abundant, associated with volcanic islands and with landless coral reefs ; every such island may have been formed by a separate and rapid upheaval, but Davis postulates a wave elevatory movement which affected the group. Many other islands, such as are ordinarily called ‘coral islands’, are quite low, rising only a few feet above the sea-level. They rest upon the surface of broad reefs, coral plateaux at the surface of the sea ; and the method of formation of these coral reefs is the important question. Coral reefs are found between lat. 30° N. and lat. 30° S., where the surface temperature of the sea does not fall below a minimum of about 20° C. Within these limits they exist in every land area of the Indo-Pacific Ocean, from the Red Sea to the Paumotuas, and, in addition, form a vast number of small isolated rings, most of which are crowned with land and garlanded with coconut palms. In the Atlantic they are found in the West Indies, off Florida, and in places off the coast of Brazil, but they do not form isolated peaks in the ocean bed or exist around its other islands, except Bermuda. Further, on the eastern shores of both oceans typical reefs do not exist, while on their western shores, as to the east of Africa and Australia, they attain great dimensions. Their shape and appearance wherever they exist is quite characteristic, a flat, almost level with the surface of the sea at low tide, on the seaward edge of which the ocean pounds down, even in the calmest weather, with great breaking waves. The water at low tide rushes over the flat for some distance, only to flow back by countless fissure-like channels in the edge of the reef. This area, the reef flat, as it is termed, is marked ^ It is most important that biologists visiting such islands should attempt to obtain fossils upon which their geological age can be determined. 126 Coral Reefs and Islands off on its inner side by a band of boulders, generally small masses of reef rock or coral colonies, which have been hurled up from inside or outside of the reef; hence is derived its name of the boulder zone. Instead of these boulders we may have a land, perchance formed by them, perchance of elevated coral rock, or of granite as in Seychelles, or of volcanic material. The reef here forms a narrow shelf around the land, being termed a fringing reef (see Fig. 1 6). Sometimes the land behind is separated from the boulder zone by a channel; this may be only a few feet in depth, when it is termed a boat channel, or it Fig i6. Section of fringing reef, Rotuma Island, Pacific Ocean a, land; b, b, boat channel or lagoon flat; c, c, boulder zone, d, d, reef-flat, e, crest or edge of reef;/, /, seaward slope. may be some fathoms deep, sufficient for a ship’s anchor- age, when it is commonly called a lagoon. To a consider- able degree the depth of this lagoon depends on the dis- tance of the land from the surface reef outside, but it seldom exceeds 50 fathoms. Inside of most reefs, within the boulder zone, there is almost invariably (if no island stands within the limits of this zone) a broad flat, with i or 2 feet of water at low tide, more or less covered with sand, part bare, part rich in coral growth. This is termed the lagoon flat, and may be the flat beneath our boat channel, or a flat before the water deepens to form the lagoon. Such a reef, with a deep lagoon within it around the land, is termed a barrier reef, but there is no sharp transition between it and the fringing reef. Lastly, the reef may exist as an isolated structure by itself, with its reef flat, boulder zone, and lagoon flat. It 1 ir 17 jobcph, \mirantc Group VicAV of cist shore of Pclicin ind, showing h\t fcrmcr bca.(hc<( inaikcd by bcich ‘«and tone, indicaliii^ piuscs 111 the washing awa^ of the liud 128 Coral Reefs and Islands may be (i) a small mass completely swept by the breakers ; or (2) a larger reef with a mere pile of boulders in its centre; or (3) with the smallest puddle of water in a ring of boulders; or (4) with a great shallow lagoon flat a mile or more across; or, lastly, (5) with a lagoon flat and a deep lake, like the barrier reef lagoon above. In the latter Fig. 18. A typical atoll; Peros Banhos, Chagos Group. Scale, 4 millimetres = i mile case it is termed an atoll (Fig. 1 8). Then it is essentially a ring of reef, bent as you will, surrounding a lake, the atoll lagoon, which may be a mere pool or a great enclosed sea 30 or 40 miles across), which, however, never attains a greater depth than 60 fathoms. A barrier reef may extend along a thousand or more miles of a continent’s shore, as the Great Barrier Reef off Queensland, or, again, it may surround an island or a mere speck of land. Were the latter removed we would have an atoll, and the 130 Coral Reejs and Isla?ids obviousness of this fact was, pending the detailed examina- tion of the reefs themselves, the cause of the prevalence for several decades of the single view that the ring-reefs were formed on the disappearance of the land which they once surrounded, first as fringing and later as barrier reefs. Theories to account for the formation and origin of coral reefs centre on the atoll, since it is easy to see that any bold theory must be capable of explaining the formation of both barrier and atoll reefs. A caution may be at once issued to the effect that it is unlikely that any one theory will explain the formation and origin of all coral reefs. As to the formation of fringing reefs and isolated reef masses in shallow waters, there is no dispute, but to explain their building we require briefly to consider the organisms, the skeletons of which enter into their composition. It must not be forgotten that it is the biology of these which makes possible the formation of such reefs, and that we are not yet in the position to ‘assume’ these organisms, either in the present or in past ages, as Agassiz has done for the whole Indo-Pacific. Coral reefs are formed principally of the limestone skeletons of animals and plants consolidated into rock mainly by sedentary organisms. Sand, formed from the skeletons of pelagic Foraminifera, or from the broken down shells of molluscs and echinoderms, or from the fronds of calcareous algae, always enters largely into their com- position. In addition, there may be concretions of an oolitic appearance and muddy precipitations, for the sea may become supersaturated with calcium carbonate by a rise of its temperature, so that less carbon dioxide is held in solution, or by evaporation, or by the action of certain bacteria or of green plants, these using up the COj,- In addition, there is much loose stone and sand material Fig 21 Lithothamnioneae upper figure, Gmuhthm frutescens, lower figure, Ltthopl^llum Gardvnen 132 Coral Reefs and Islands formed by the decay and breaking up of corals and Litho- thamnioneae, which are the great organisms in the con- solidation of the whole into the rock that coals the sur- faces of such shallow faces as are exposed to the force of the ocean. Any single ingredient may be found forming by itself a deposit, where conditions are suitable, but all kinds of organic remains enter into the composition of the massive barriers, upon which falls the full weight of the waves. Reefs in lagoons may have grown in protected situations and may consist mainly of corals, their inter- stices filled with sand, while the ocean rim may seem to have been built entirely by the red-brown incrustations of the Lithothamnioneae. In elevated coral rock these algal remains exhibit little structure and so are difficult to deter- mine; the same remark applies to the remains of Halimeda and similar algae, the importance of which is much under- estimated. Corals, as the most widely distributed organisms in the formation of coral reefs, must first claim attention. The skeleton or corallum is mainly a deposit of carbonate of lime outside the ectoderm cells of the basal parts of anemones. These anemones have free floating planula stages, which in captivity have been known to live as such up to twenty-three days. They attach to hard objects and form anemones, which bud out other anemones from or near their mouth disks forming ‘colonies’, in which the digestive cavities of all the members still retain their con- nexions. With growth the deposition of corallum goes on until senile decay sets in, and the colony spreads side- ways either by side budding, or by the intercalation of fresh mouths surrounded by tentacles as the surface be- comes domed by the formation of skeleton. In this way great massive rounded corals are formed, quite commonly several tons in weight. In some forms surface areas of the Fig 23 Kambara, Fiji An cxampk of an elevated atoll, peiforatedin Delai Yaloi by a subsequent eruption of the volcanic foundations on which it was built Subsequent to elevation it has been clifted by sea action, and fringing iccfs merging to biriier have been formed 134 Coral Reejs and Islands quite small colonies are stimulated to growth so as to give rise to branched masses (Pocillopora, Stylophora, Monti- pora, &c.), or individual anemones may grow out in a similar manner, budding off side mouths (Acropora). In yet other species plates (Pavona) and cups (Turbinaria) may result, but Fungia with certain allies becomes free- living. The reef corals, which are a mixed assemblage of all these forms of growth, are classified into about forty genera of reef building importance, of which more than half are now confined to the Indo-Pacific region, about five genera being peculiar to the Atlantic. So far as can be seen the existing species of the reefs would appear to be of quite recent origin, most certainly not earlier than the pleistocene. They can withstand temperatures between 1 8° and 36° C. and salinities of 25 and 45 per mille, but exposure of an hour to the sun is usually fatal unless some part of their living tissues is still in the water. Mud they do not like, and off Seychelles all the corals were found to have been killed on a large area of reef after a heavy rain storm. Mayor for Samoa and the author for the Maldives have independently calculated that their rate of growth is such that they might by their own skeletons alone give rise respectively to reefs of 13-5 and 14-5 fathoms in 1,000 years, but the calculated rate for the West Indies is only a fraction of this. Recent work has confirmed Darwin in placing the extreme depth of flourishing banks of surface reef corals at 25 fathoms; the extreme limit of any of these forms is about twice this and the optimum depth for them as builders is 5 fathoms or less. Their feeding by commensal algae has already been referred to (p. 1 1 9), and experiments have been made showing that light is essential to them. Corals, assisted perhaps by Millepora, Heliopora, Tubipora, and to some degree Gorgonaceae, Polytrema, Growth of Coral Reefs 135 the alga Halimeda and sponges, may build reefs and indeed frequently do so in protected situations. Against the open ocean they cannot build in shallower depths than about 5 fathoms, unless consolidated by Lithothamnioneae, of which there are three important reef genera. These have fine branching species on sand flats, rounded nodules in passages through reefs, and massive attached growths on the seaward faces of reefs; they are not common lagoon forms. Their skeleton consists of about 74 per cent, of CaCOj and 15 per cent, of MgCO^, while there is twice as much Mg as Ca in sea-water. The maximum depth at which these and other algae can live is about j' 5 fathoms. Their importance lies in the fact that they are more re- sistant to any change in the environment than corals, and their incrustations can resist the heaviest breakers, so that they always dominate the seaward parts of reefs. Here they are the chief agents by which reefs grow up to reach the surface and, when once there, spread outwards. In the Cape Verde Islands reefs have been described as formed solely of them and the tubes of Vermetus. Ordi- narily on the outer edges of reefs little colonies of massive corals or of Pocillopora grow in hollows left by them, later to be incorporated into the solid mass. They spread out over channels, gullies, and pits and enclose them and their contents, leaving hollow spaces in the coral rock. The massive incrusting forms seem to like the rushing waters after the breakers crash; with increase in depth they be- come less important, coral largely taking their place, but this area to seaward is difficult to study on account of the surge and the roughness of the bottom, which prohibit dredging (Figs. 15, 20). From the above paragraphs it will be clear how in seas of suitable temperature reefs may grow up on any shoal or round any land from a depth of 30 or 40 fathoms. 136 Coral Reejs and Islands massive coral and algal structures with their interstices filled in with sand. The reef will be a flat on the surface, with the corals dead on the top or represented only by small colonies of poor growth. The sea will be surging up around, bringing abundant ‘food’ to its seaward face. The rollers will be constantly pounding down upon it and breaking off pieces, a few of which may be cast upon the surface of the reef to form a boulder zone, but most of which will be washed outwards to bring fresh areas within suitable depths for the reef builders. As this process con- tinues, the reef will be spreading farther and farther sea- ward, broadening its surface, and making life less and less possible on its inner zone as the animals become more distant from their chief source of food-supply, the open waters of the ocean. This process may continue indefi- nitely, the rate of outward growth depending on the steepness of the outer slope of the ocean floor. The outgrowth must be assumed to result finally in the (almost universal) gradual slope for about 200 yards to about 40 fathoms, followed by the steep slope, often with an angle of over 30 degrees to about 140 fathoms. Fur- ther observations on the currents, bottom conditions, and fauna of the outer slope, are urgently needed. The steep is clearly a ‘talus’ formation, resulting from the outwash of loose masses of coral from the reef above, its angle of slope being the angle of rest of such rocks in compara- tively calm water. The theory of the formation of atoll and barrier reefs propounded by Charles Darwin in his ‘Coral Reefs’, a book of genius in its line as great as his ‘Origin of Species’, depends on the assumption that the land, round which a fringing reef has formed, is sinking at such a rate that the organisms upon the reef can, by their growth, keep it at the surface of the sea (Fig. 24). Such organisms will grow Darwin's Theory of Subsidence 137 more vigorously at the ocean edge of the reef, where they will have more food. Thus it will be easily understood that, as sinking proceeds, a channel (at first a boat channel, and then a lagoon) might be formed between the reef and the land. This lagoon will attain its breadth by the gradual submergence of more and more of the land, which finally might be completely lost, the formerly fringing reef be- coming successively a barrier reef and an atoll reef, the contours of these roughly representing the former exten- -A" Fig. 24. Scction.ll diagram to illustrate foimation of Atoll and Barrier Reefs by subsidence. A, A, sea-level of island, with fringing reef in black; A', A', the same after subsidence — island with barrier reef shaded; A'', A", the same after the island has been submerged — atoll reef dotted. The vertical scale is many times the horizontal scale sion of the land mass. It is a theory delightful in its sim- plicity, but it depends on the assumption that the land continues to submerge slowdy for periods of time even geologically large. There are many objections to this premiss, and there is no evidence that such a general submergence is going on as would be required by the theory, i. e. from Africa across the East Indies to Hawaii and the Paumotuas. In this area many atolls arise pre- cipitously from 1,000 to 2,500 fathoms, and, whether on this theory general or local sinkings are postulated, there should be similar thicknesses of coral rock.' There are * It would possibly be an exceptional case where the added weight of coral formations would initiate the sinking of an island, but, if this were established, the constantly increasing mass of material, even if of relatively low specific gravity, might quite conceivably cause its continuance so as to give such thick- nesses. The expected mounds and ridges of compensation are not known aiound 138 Coral Reefs and Islands many elevated reefs in coral reef regions, and none show such vertical dimensions, the vast majority indeed being less than 300 ft. (50 fathoms) thick. Most are formed of very recent organisms, while such reefs as Darwin postulated might be expected to have been forming con- tinuously through several geological periods. The eleva- tions, which raised some coral reefs above sea level, were relatively rapid and conversely subsidences might be expected to have been similar. Furthermore, our knov/- ledge of geology gives us no clear parallel to the slow and extensive bending or sinking of the earth’s crust such as is required. In addition 60 fathoms is the maximum depth for the lagoons of atolls and barrier reefs — ^Vanua Mbalavu (Fig. 25) in the Fiji group alone showing a bight of 50 to 100 fathoms extending for several miles into a lagoon enclosed by a barrier reef — not a tenth of the sinking that must be postulated to form most atolls. The subsidence view of the successive formation of barrier reefs and atolls is strongly opposed by all those who have in the last thirty or forty years devoted them- selves to the biological examination of series of reefs, mention in particular being made of the late Sir John Murray, Professor A. Agassiz, and Professor Semper, who, in addition to having made explorations themselves, have had the benefit of the experience of numerous other expeditions. It has in the first place been shown that fringing reefs may owe much of their breadth to the abrasion of the land round which they form a belt. Tall cliffs and masses of rock, overhanging at their base, are often found bounding the shoreward edges of the reefs. The land in such a position is evidently being rapidly removed between tide-marks, and, as it recedes, it leaves atolls> but the process of sinking would be so slow that ample time would be allowed for the dispersal of stresses over wdde areas of the surrounding ocean. Fringing Reejs 139 at low-tide level a flat platform, indistinguishable from the shoreward part of a reef. If organisms commence to grow on the outer edge of such a platform, its surface appear- ance is in every respect that of a true reef (cp. Zanzibar and Pemba). This is especially the mode of formation Fig 25 Vanua Mbalavu, Fiji The barrier lecf is about 8o miles round The islands are formed of \ olcanic, or limestone oi mixed rocks The elet ation is at least 700 ft and the 100 fathom line extends deeply into the lagoon where the land is composed of elevated coralline limestone, as is frequent in island groups where fringing and barrier reefs are abundant. Again, as a fringing reef broadens, neither the reef flat nor the boulder zone increase greatly in breadth, the inner part (boat channel or shore flat) broad- ening and taking up the extra breadth at their expense. The shore or lagoon flat is obviously formed where once 140 Coral Reejs and Islands was boulder zone, and it should therefore be strewn with boulders. On a broad fringing reef such loose masses are not found on the lagoon flat,* it being generally a bare or slightly hollowed-out area, which is alternately covered and uncovered by the tides. The latter must be the ulti- mate agents which effect the removal of the boulders, and this they are supposed to do by removing the coral, either in solution or as fine mud in suspension in the water, or by both means. The solvent action of sea-water on coralline material is relatively slight, but organisms are known to break up coral rock into fragments — gravel, fine sand, and, lastly, mud. The removal of material in suspension is evidently of great importance, hundreds of thousands of square miles of ocean floor near coral reefs being charted as covered with coral mud (see p. 2 65). How the latter is produced requires further investigation, but it is difficult to determine whether its fine particles were originally formed out of reef building organisms or are the result of precipitation. By such methods as the above the inner rocks of the boulder zone might be removed, while the zone itself continues to extend outwards on to the reef flat as the reef grows seawards. It is easy to imagine the processes going further and hollowing out the inner part of the reef to form a boat channel and finally a lagoon. Should the reef start as a small mass in mid-ocean, it would pass through similar stages — a flat in the centre, then a shallow pool, and finally a lagoon. As a barrier the reef should in some degree conform to the contour of the land, but as an isolated reef it should be to a large extent more or less rounded in its early stages. Later, both the barrier and the I Boulders formed of the skeletons of the blue coral, Heliopora, often persist for long periods of time, resisting most of the ordinary agencies which in such positions cause decay. The same is also true to some degree of Millepora. Lagoon Formation 14 1 atoll would show the effect of currents in carrying and depositing mud and other materia], so as to make the extension of the reef in certain directions easier. From the first formation of any inner pool, channels would form through the boulder zone to allow the escape of the tidal water. The latter would especially rush out over the lowest parts of the encircling reef. Being charged with much sediment in suspension, it would kill any corals and other sedentary organisms in its path, and further would prevent any fresh corals from commencing to grow in its channels. Where dead rock occurs boring organisms enter, but where a surface, such as the seaward face of a reef, is covered with living tissues they cannot penetrate. The rock under a channel, being bare, will be bored and rotted like the rocks in the lagoon, with the result that its depths may increase more or less in proportion to that of the lagoon to which it gives entrance.' Pending further research, it is quite impossible to say that this view of lagoon formation ‘can be eliminated from further consideration’, even though the solution of lime- stone postulated by Murray may not be the chief agent in the process. We know lagoons that are filling up and again others where the lagoons are as certainly growing, since their seaward faces are surrounded by small sub- marine cliffs, and the lagoon flats above the same appear to have been formed of exactly the same kinds of coral as are now found to seawards, cemented together by massive algal growths. Separate fragments of the land rocks have been found bedded in the coral rock, which forms the floor of boat channels, and, as already said, the whole ‘ Sometimes rounded growths of Lithothamnioneae are common in such a position. Often the sides of the-c channels arc rich dredging-grounds for sedeniars' organisms, especially Cjorgonians, I’olylrcma, and Hydrocorallines. Where the lagoriti is large and there are scseral channels, this question of depth becomes complicated. It should be reinvestigated in each atoll. 142 Coral Reejs and Islands question of the formation of coral mud requires investiga- tion. On the other hand, there is no adequate explanation of the shallow depth of lagoons, even though boring and sand-feeding organisms become of decreased importance with increase of depth. We. want sections across some atoll and barrier reefs with quite shallow borings, even 5 or 10 ft., with precise information as to what is building the reef to seaward, which must not be assumed to be growing outwards without direct evidence of the same. Fig. 26. Sections to show formation of Reefs by Abrasion of Land I, Ongea Levu, Fiji, an elevated coral island; 2, Wakaya, Fiji, a volcanic island ; A, supposed former sections of land ; B, present sections of land and reef. The vertical scale is many times the horizontal These should be reinforced with precise information as to the topography of the lagoon, as to the condition and composition of its lagoon flats and shoals, as to sand deposition (exemplified by samples), as to rates and nature of currents, with measurements of material carried in suspension in the same. The last theory of the formation of lagoons offers no explanation to account for the submarine foundations on which atoll reefs are built up. It is conceivable that islands, which were originally surrounded by fringing reefs, may have been completely removed by abrasion and solution, the reef becoming successively of barrier and atoll type. Lagoon formation \ 4,3 This mode of formation is seen in the Lau group of h’iji (Fig. 26), where the islands are mostly formed of raised limestone; but atoll formation by this means is probably not of wide occurrence. Secondly, an island may be cut down by the action of wind, waves, and ocean currents to form a submarine bank, no fringing reef forming while the process is taking place, owing to sediment produced by abrasion. On such a bank an atoll would be built up owing to the more rapid growth of suitable sedentary or- ganisms on its rim, where they would be better nourished. Fig. 27. Sectional diagram to illustrate formation of an Atoll on the site of an island of loose material which has been partly washed awayj such an island as it is farmed by a submarine eruption A, original island; o, bank after the top of the island has been washed away to 20- JO fathoms; C, atoll formed by the direct and greater upgrowth of reef organisms on the edge of the bank. The vertical scale is many times the horizontal Submarine eruptions may occur anywhere, and the con- tact of the molten lava with the water often produces a mound of cinders and ash, which may form an island or never appear above the surface of the sea (Fig. 27). Such a mound would rapidly be cut down to the limits of wind- wave action, even to 40 fathoms or more; on the bank an atoll would arise directly, in the way above suggested. Such a mound was Falcon Island, 'I'onga, erupted in 1 8 H 3 to a height of about 250 ft., but later a submarine plateau covered by a few fathoms of water. Direct evidence of this method of formation of coral reef foundations has been obtained by the dredging of masses of semi-consoli- dated volcanic ash from 744 fathoms off J^rovidence Fcef, and Christmas Island on the other side of the Indian 144 Coral Reejs a?id Islands Ocean would seem once to have been an atoll so formed. Vulcanism and coralline organisms are clearly to-day the great visible builders of land, and regions of elevated coralline limestone are areas of eruptive action. The lines of coral atolls in the Pacific almost irresistibly suggest themselves as being lines of vulcanism, and the topo- graphy of the many submerged shoals that are known, often called incipient atolls, is frequently explicable on this theory of the formation of their foundations. Thirdly, mounds might be built up beneath the sea to suitable depths for the reef builders by the remains of various organisms. On any slight elevation on the ocean floor sedentary and other animals congregate to an extra- ordinary degree. Chief among these are the deep-sea corals, enormous banks of which have been discovered on the floor of the Atlantic by dredging and sounding expedi- tions. Their remains, assisted by the remains of pelagic organisms, build up the surface of any elevation to an extent out of all proportion to the upgrowth of the sur- rounding area of the ocean floor, so that the mound ultimately approaches the surface (Fig. 28). The reef builders obtain sway, and a reef soon reaches the surface, later to spread out on its own talus and to be hollowed out in its centre, forming a typical atoll. Atolls and banks built on such foundations, as postulated here and in the last method, should show some rounding off against and some extension in the direction of flow of the most con- stant currents which lave them, whereas the actual shapes of relatively few can he so correlated. None of the above theories take full account either of the original mode of formation of the coast lines, or of the action of the sea on the same in periods of rest. Waves cut terraces into the land and build terraces, to continue the same further seaward. The movement of the Basement Platforms 145 waters is largely lesponsible for the shelf, which extends to about 50 fathoms, before a steeper slope commences. This shelf is an area of great motion and of peculiar suita- bility to corals and calcareous algae, and to the consequent fauna attracted by them and by the abundance of algal food. Plenty of such shelves are known in temperate seas, and in some places sand banks are found on them parallel Fig 38 Sectional diagram to illustrate formation of an Atoll on a Mound of the Sea-Floor A, original mound j b'-b4, building up of same bv remains of decp-sei animals, bj pelagic deposits, and, lastly, by reef organisms to the sea-leiel, C'-C>, out- ward extension by accumulation of talus and other materials on the slopes and further growth of reef organisms, hollowing out of the suiface by solution and remotal of mud in suspension to the coasts. Were conditions on these to become suitable for coralline organisms, it would be easy to imagine the direct upward growth of barrier reefs, but on most shores the barrier would only be formed after passing through antecedent fringing reef conditions. Daly, who regards most existing coral reefs as still in the growth stage and as having been built on antecedent basement platforms either off land or as upstanding shoals, postulates such a slow con- tinuous submergence of these as would allow of the direct upgrowth in their present forms of barrier and atoll reefs in general. This submergence is due to the recent melting of the great pleistocene ice caps that covered the two polar regions, even extending down into what arc now temperate 146 Coral Reejs and Islands latitudes. These caps were many thousand feet in thick- ness, and they further lowered the equatorial oceans by the attraction of their mass on the waters. The general lowering of temperature in the sea would prevent or- ganisms protecting the sea faces of lands and the surfaces of shoals, and great submarine plains and terraces would be formed off them. As these polar ice caps melted, the conditions would become more suitable for the requisite reef building organisms, and all possible areas would be- come populated by corals and massive algae. As the water gradually flowed back from the poles, the growth of these organisms would keep the reefs they had formed level with the surface of the sea, the method of growth of the same being that postulated by Darwin in the formation of barrier reefs and atolls (Fig. 29). At the present day in all tropical lands we see an emergence perhaps still in progress of 10 to 30 ft., but the pleistocene emergence and subsequent submergence of Daly is calculated to have been 200 to 250 ft. On emergence the sea would have been beating on mud and relatively loose material, not rocky coasts, and the forma- tion of platforms would have been rapid. The fact that lagoons attain an average depth of 200 ft. agrees well. Many tropical lands in coral regions show some evidence of submergence, and this, not being represented in the topography of the surrounding ocean, is explained. So far as the biology of living organisms is concerned there would seem to be little objection to this theory, but to estimate its value as an explanation of the varied features of coral formations requires the comparative study of a series of tropical islands with barrier reefs. The same process of emergence and submergence has not taken place only in the Pleistocene and later, but in previous geological periods, and we require to know the relations of their iSBiiliili^ liliiiiiiiiiiii 3 7 z 0 I 2 3 4 5 ly KILOMETERS Fig. 29. Daly’s sections to illustrate the det'clopment of Barrier Reefs aiul Atolls according to the glacial control theory Section i, A normal volcanic island. Section 2. The same island largely penepkined, with the necessary formation of an encircling embankment of detritus (stippled). Section 3. The same island cxtcnshcly benched by waves, involving some increase of enibankment. Such benching is expected in very old islands which have been exposed to active abrasion, cither because of the Pleistocene chilling of the ocean, or because of temporary failure of reef protection in prc-Glacial times. Section 4. Complete truncation of the same island by continued marine abrasion, with a slight broadening of the embankment. This i.s a stage that, in many instances, was possibly attained in pre-Glacial periods as well as during the Pleistocene. In Sections 2, 3, and 4 the si/e of the embankment, as drawn, corresponds merely to the bulk of purely inorganic detritus. If iiitermixed mef and other organic material were allowed for, the embankment must be represented as broader. After the abrasion, the fringing, barrier, and atoll mefs would be favourably located at X, Y, z, respectively. Shifts of sea-level are not shown. The sections are drawn to scale and are also intended to show the great areal extent of the weak embankment materials, laid down around old oreanle vol- canoes in pre-Glacial times. About half of the platform represented in Section 4 is underlain by these materials, which must have offered little resistance to the benching surf of the Pleistocene iieriod. 148 Coral Reejs and Islands coralline deposits. Is the surface of the earth so elastic as to allow of the removal of this water from the equator and its piling up in at least 1,000 fathoms thickness of ice at some parts of the poles without breaking and faulting beyond what we at present know? It is supposed that it may be, but we are as yet only laying the foundations of knowledge as to the history of the earth. We deal here with wide alterations of loads, but it has often seemed to the author that every alteration of load must be felt. The increased weight due to submarine erupted materials and to coral formations or even to glacial ice at the poles must be reflected in the earth’s crust beneath it, and thus adjustment by sinking may help to a limited degree in the formation of both barrier and atoll reefs. Having now briefly stated the chief theories of the formation of atoll and barrier reefs, it appears that widely divergent views are held as to their peculiar features. It will be particularly interesting if the traveller will note any parallel cases of barriers such as the sand banks, which frequently are found lying parallel to the neighbouring shores. Ring-shaped reefs forming parts of barrier and atoll reefs and similar shaped shoals within lagoons (faro)’ deserve careful examination. Without direct examination it must not be assumed that either lagoon or any other shoals are growing upwards or that reefs in general are increasing to seaward, for according to Dr. Crossland such is not the case off Tahiti or Zanzibar. The one constant factor we have lies in our organisms, their reactions to different environments and their relations to one another. So far as we can see these reactions are not mightily dif- ferent in our varied reef regions and ‘therefore it is evident ' The word atoll is derived from the Maldivaii word ‘atolu’, and we may as well borrow a second word from the same language, especially since these faro are a most important feature of the Maldivan coral region. formation oj Atolls and Reejs 149 that the form of the reefs produced are not so much dependent upon the architectural desit^n of the builders, as upon the form of the base they may have to build upon’. (Stutchbury, 1832.)' ’ The above article was in type before publication of The Cora! Reef Problem, by VV. M. Davis. It is a most important and valuable study of .ill knowledge in respect to tropical roasts from a geographical standpoint. He mainuins that the bays, large indentations and estuaries in such, if covered by reefs, merge into the land valleys at their heads. Where this is the case, they were clearly produced by subaerial degradation, and can only be explained by subsidence, or a positive movement of water-level. All methods postulated above are possible, and each is probably partiallv correct for certain reefs. Reefs generally are growing surfacewards or laterally, or both, the organisms upon them accumulating heavy material as compared with water. This means alterations in the weight supported by the e<irth's crust beneath them, and it may well be that these are ultimately met by local depres- sions of the same. The alterations in weight are continuous, and such sinking might well be equally continuous (see Review, Geogr. Joum. Ixxii, pp. 168-74., 1928). Falcon Island (p. 143) has recently been in further eruption and a high island has again formed. The dimensions were 1,730 by 1,430 yds. by 305 ft. high (7 October, 1927). IV THE PLANTS Floating Plants {Phytoplankton') and Borderland Forms BY MARIE V. LEBOUR Plants in the sea are represented by the seaweeds of the coast on the one hand, and on the other by an enormous number of floating microscopic forms. This floating flora is known as the phytoplankton. In the open sea, over any great depth, floating is the only means of living for the ordinary plant, as the sea bottom is too deep for any fixed form to live, light being a necessity for its feeding pro- cesses ; therefore no plants exist on the ocean bottom and the only large seaweed found in the ocean is the floating Sargasso Weed described on page 1 8 8 . The study of the phytoplankton is thus the study of microscopic and unicellular plants. The chief groups represented are the Diatoms, Flagellates, Silico-flagellates, Coccolithophores, green algae and red algae, the two last with few species, but these may be enormously abundant. The groups mentioned are true plants, but there is another large group, the Dinoflagellates, which are usually re- garded by botanists as plants and by zoologists as animals (see Chapter V). They are here regarded as on the border- land, for most certainly some of them are holophytic and others holozoic, that is to say, feed as true plants or true animals. Even the same individual may feed either as a plant or as an animal. There are also large numbers of Bacteria in the sea. These are plants related to the fungi and are of an exceed- ingly simple kind. They are so small and require such special methods of study that they are here disregarded. Methods oj Collecting ’ 1 5 1 although they are of extreme importance in the economy of the sea. All these form an enormous mass of available food for the sea animals, beginning with the unicellular species from dinoflagellates to infusorians throughout the Animal Kingdom, for the unicellular plankton serves as food some- times for even large vertebrates. The plant' feeds by utilizing the salts and gases in the sea, breaking up carbon dioxide in the presence of sunlight by means of the colour- ing matter, chlorophyll or some closely related substance, contained in special bodies, the chromatophores, and so building up the complex protoplasm of which it is com- posed. It therefore occupies the base of the food chain, all animals being dependent on the plant, eating it either directly or indirectly. Thus a small sea animal eats the plant and is itself eaten by a larger animal, and so the great food cycle always goes on. The phytoplankton is sometimes aptly termed 'the meadows of the sea’. Methods of Collecting. To collect these microscopic organisms special methods are necessary, although a pail or bottle dipped into the sea, and a sample of the water so obtained centifuged, will usually yield a fair amount of material. For the larger forms the same methods as those used for the small floating animals are applicable (see p. 230). A ‘very fine’ tow net (180 meshes to the inch) will catch a considerable quantity of diatoms, dinoflagellates, and flagellates, but the smaller species of these groups and practically all the coccolithophores and silicoflagellates usually escape through the meshes of the finest net. A water-bottle designed to fish at any depth is used in these cases (p. 59), and samples of the water thus secured filtered or centrifuged. The centrifuge is now usually * The term plant is used here to indicate the typical plant which contains either chlorophyll or some closely related substance. 152 The Plants preferred to the filter. Those oceanic ascidians known as Salps and appendicularians (see p. 228) are extremely useful for securing very minute planktonic material as they feed by means of a fine natural filter which only admits very small organisms into the stomach. An examination of these may bring to light many rare or new forms. When examining the water-bottle samples by centrifuging, the best kind of tube to use is one drawn out to a thin blunt point, so that, after centrifuging, the super- ficial water may be poured away, leaving the required deposit at the bottom, which can be removed by a pipette and examined on a slide. There is nothing so good as examining such plankton alive. If preservation is neces- sary, however, formalin (5 per cent, solution) neutralized by borax is excellent, as it preserves the calcareous as well as the siliceous and cellulose skeletons. Fleming’s Fluid, strong (seep. 400), loc.c. added to 250 c.c. of the sample, is exceedingly good for everything except the calcareous forms, and will preserve the shape of the delicate pelagic flagellates and naked dinoflagellates perfectly. The Diatoms. The diatoms are unicellular algae with siliceous skeletons having brown or greenish-brown chromatophores containing substances closely related to chlorophyll. A typical diatom cell is formed like a box of two parts or valves, one fitting over the other, bound together by a girdle. Fig. 30, p. 153, shows one of the bottom forms with the typical diatom structure. This outer covering is formed chiefly of silica and is usually marked by pits or punctures arranged in exceedingly beautiful patterns. Inside is the protoplasm forming a thin lining layer and attached to the central mass by strands, the spaces between being full of a clear cell fluid. Embedded in the protoplasm are the chromatophores and nucleus. The individual cells may live singly or in Diatoms 153 colonies. They are very variable in shape — round, oval, square, oblong, needle-like, hair-like, and may be twisted in various ways. All the planktonic diatoms are adapted in some way to a floating life. Thus if the cell live singly and be shaped like a disk or pill-box (Coscinodiscus), it is lightened by large masses of cell fluid of almost the same specific gravity as the sea-water and may a, ^ahe vle^^ , b, gudio mcw, f, the new vahe has. forced the girdle apart contain light fatty oils. Many disk-shaped forms are held together by means of protoplasmic threads (Thalassio- sira), others may cling together by spines (Skeletonema, Fig. 32) or hooks, or by pads of protoplasm coming from special apertures. The large genus Chaetoceros (Fig. 33) has the corners of each cell drawn out into long setae which intertwine and form a large surface to prevent sinking. Bacteriastrum (Fig. 34) has similar spines all round the margin of the cell. Some may be in colonies embedded in a mass of protoplasm. Planktoniella (Fig. 31) 154 The Plants has a broad membrane all round the cell. All these are devices for keeping the organism in those upper layers of light, below which they must die. They are able to live as far down as about 200 m.,but are more abundant above this, and it is found that in the ocean usually a maximum occurs at about 50 m. The planktonic forms generally have thin skeletons and the markings on them may be very faint compared with the bottom and littoral species. Fig. 31. Planktoniellax ca 500 Diatoms are by far the most important of the floating plants in the sea and may occur in enormous numbers, affording food for an innumerable quantity of animals. The fatty oils contained in them make them very nourish- ing, although so small. They may be found anywhere in the sea or fresh water, but the marine forms are usually more abundant near land. This is probably because much food is brought down by the rivers into the sea, and this is taken by the coastal diatoms, leaving little for those in the ocean. Oceanic diatoms flourish best where there is plenty of vertical circulation or ascending currents, when food which has accumulated below is brought to the upper layers. Nitrogen and phosphorus are probably the most 156 The Plants essential of their food-stuffs, silica next. They are usually more numerous and larger in cold and temperate climates than in hot regions and are never evenly distributed. Large areas may be free from them, whilst they abound elsewhere. Regular periods of maximum and minimum occur, in northern temperate regions there being two maxima, one in early spring and the other in autumn, each caused usually by different species. Diatoms reproduce by cell division, one new valve being added to each old one, or they may form auxospores, a large swollen mass of protoplasm bursting from the cell and forming new larger valves within. Spores have occasionally been observed. In adverse circumstances dia- toms may form resting spores, which have hard coverings quite unlike the parent cell, sometimes of beautiful and peculiar form and often armed with spicules or spines. These when liberated from the parent sink to a lower level and are capable of great resistance and prolonged Dinoflagellates 157 inactivity. They are usually very minute and pass through the meshes of the finest net, but may be found by centri- fuging the water samples, especially in coastal areas. The Dinoflagellates. Next to the diatoms in importance come the dinoflagellates or peridinians. As already men- tioned, these are on the borderland between plants and animals. Several well-known forms, some of which were formerly placed in separate groups, are included here. Amongst them are many brightly luminous organisms. Here we have Noctiluca (Fig. 8 3), which, although not the most important light-giving cell in the sea as was for some time supposed, still amongst the neritic species is very important in this respect. The different forms of Pyro- cystis, perhaps the brightest of all, which light up the warmer seas to an enormous extent, are now known to be phases in the of certain dinoflagel- lates, and no longer occupy a group to themselves. Ceratium (Fig. 35), the first and best-known genus, is also highly luminous. Dinoflagellates are lovers of warmth and are to be found universally distributed, although more frequent in the warmer seasons, sometimes colouring the sea in large patches. They may be naked or provided with a covering or theca, composed of a substance related to cellulose, which is usually divided into plates. Typical forms have a groove running equatorially round the body, in which vibrates a whip-like thread, the flagellum (Fig. 36). A second flagellum is lodged in a less conspicuous longi- 158 The Plants tudinal groove and usually trails behind. In the naked forms the latter groove serves as a mouth which may engulf diatoms or its own relatives in a truly animal way. Noctiluca, which is the largest known dinoflagellate, may even eat small multicellular animals. Many dinoflagellates are well adapted for life in the open sea, often having long drawn-out horns, spines, and mem- branes, which help to keep them afloat. They are almost always slightly asymmetrical, which helps them to adjust themselves easily and prevent sinking. Ornithocercus (Fig. 37) is extreme in having an enormous wing-like membrane, and this is perhaps the most truly oceanic of all. The genus Ceratium, with its three horns and anchor- like shape, is of world-wide distribution. Peridinium (Fig. 36) is also very widely distributed. Both Ceratium and Peridinium, with many others, are covered by a theca com- posed of plates arranged in a regular order. The holophytic forms may have green, yellow, or brown chromatophores, and there are many colourless or with the plasma itself coloured pale pink or pale yellow, which have no chromatophores, and are probably sapro- phytic, living on decayed organic matter. All the species of Ceratium have yellow or brownish-yellow chromato- phores, and some species of Peridinium and related genera also possess these, but many have none. The thecate forms are presumably holophytic or saprophytic, as their armature prevents the taking into the body of food of any size. It is quite possible, however, that many naked forms are phases in the life-history of these, and in such phases may be holozoic. The naked dinoflagellates may possess chromatophores, yellow, greenish, or brown, or may be colourless, or the plasma may be pink, yellow, or even crimson, purple, or blue. Those without chromatophores are usually holozoic, and engulf food as large or larger Reproduction oj Dinoflagellates 159 than themselves. The chromatophure bearing furms nia) be either holozoic or holophytic, sometimes both. Those which are highly coloured are genet ally oceanic, and amongst them we find species which possess eves with elaborate lenses and curious retractile prods coming from the longitudinal groove. Sometimes the body is twisted round the mam axis, so that the grooves wind spirally tie ■J7. Ormthocercus X 310 round the cell, which may be greatly elongated. Poly- krikos seems to be well on the way towards the multi- cellular animals, for its cells, usually eight, hang together permanently and contain nematocysts, or stinging bodies, of a similar character to those belonging to the coelen- terates. Dinoflagellates usually reproduce by cell division. One or more spores may be formed in the thecate forms, the theca opening to set them free. The naked species gene- rally divide obliquely, and so also does Ceratium, each side of the theca forming a new half. Newly divided cells often i6o The Plants hang together for some time, forming chains. Division appears always to take place at night. Complicated life- histories involving different phases of spore formation seem frequently to take place. As food for other organisms the dinoflagellates may be individually even more valuable than the diatoms, although not so numerous, for their cell content is highly nutri- tious and the sizes of many of them far surpass the diatom. For those plankton-eating fishes, such as the pilchard, they form a regular food. The Flagellates. Flagellates form an important part of the microplankton, but many of these are extremely minute and are very difficult to see in preserved material. Certain forms, such as Euglena, are claimed by the zoo- logist, and it is difficult or impossible to draw the line here between plants and animals. Many of the minute flagel- late forms in the sea are probably spores of Foraminifera or other unicellular organisms. Those usually included in the phytoplankton have greenish-brown chromatophores and one or more flagella, or they may be colourless and possibly saprophytic. The most important of the flagel- lates in the sea is Phaeocystis (Fig. 38). This is a colonial form, frequently occurring in spherical, sausage-shaped, or large bulging shapeless masses of a gelatinous sub- stance in which are embedded small greenish-brown cells. These masses appear periodically in enormous quantities, usually in coastal regions, and clog up any nets that may be fishing. Thus they are a great pest, both to collector and fisherman. Flagellated spores are formed inside these masses, which escape and swim away. There are other small species in this group, occurring singly or in colonies, perhaps the most important of these, although exceedingly minute, being Dinobryon, a branched colonial form. Euglenoid forms occur sparingly. A Euglenoid (Fig. 38/) Green Algae 1 6 1 has amouth from which protudes a flagellum or two, green chromatophores being usually present and a red stigma. Green Algae Often classed with the flagellates, the only important planktonic member of the Green Algae is Halosphaera (Fig. 39). This occurs singly but is often very abundant, its bright green globules showing up con- Fig 3S a, 6, f, Phaeoc^ stis a, spherical form x ca 48, 6, large mass, r, swarm spores and resting spore X ca 1,000 d, Dictyocha x ci 400 f , Cistephanus X ca 350 y, Euglcnoidx ca 1,700 From Plymouth Sound spicuously among the brown diatoms. It repioduces by flagellated moving spores and by resting spores, and being very light is usually to be found in the surface la}ers, although occasionally it lives in quite deep waters Several other minute relatives of Halosphaera occur, but are so small that they are frequently overlooked. Tiochiscia is a spherical form beaiing spines all round the cell, sometimes occurring in masses with the spines clinging together. Silicoflagellates. Closely related to the flagellates, 3160 1 62 The Plants several minute silicoflagellates occur in the plankton. As the name implies, these have skeletons of silica, variously shaped, the commonest having hexagonal (Distephanus), or four-rayed (Dictyocha) skeletons (Fig. 38 e). In- side is a mass of protoplasm containing yellow chroma- tophores and bearing a flagellum. When about to divide the skeleton becomes double. Coccolithophores. This is a very interesting group of exceedingly minute organisms related to the flagellates and having calcareous skeletons composed of plates of the most wonderfully varied patterns. These plates were discovered on the ocean floor long before the live cells were known. Those of an oval shape were termed coccoliths and of a rod shape rhabdoliths. Sir John Murray dis- covered their real nature and succeeded in finding the live coccospheres and rhabdospheres, as they are now called, by allowing the water to stand in glasses in which were dipped fine silk threads to which the minute cells clung and could readily be examined. The protoplasm contains yellow or brownish-yellow chromatophores and there may be one flagellum or two. In the tiny Pontosphaera com- mon in coastal water it is sometimes absent. The plates may be oval, round, rod-shaped, club-shaped, or even barrel-shaped and trumpet-shaped, sometimes sticking out from the cell like the heavy spines of a tropical sea- urchin (Fig. 40). They are sometimes pierced with holes as in Coccosphaera (Fig. 41), the best and earliest known species whose plates were first discovered on the Atlantic floor. This is common both in the ocean and in coastal waters and forms the food of many small animals. Rare species may be found inside Salps. The coccolithophores form an important, if minute, part of the oceanic plankton, where they frequently replace the diatoms and are some of the most beautiful of all unicellular organisms. .J.'gite 163 Blue-green Algae. I'herc .uc not iiian\ njembcr;- of this group in the plankton, but certain spe^-ies iua\ ocxur in enormous numbers so as to give a colour to the sea for miles, the so-called 'water bloom'. The cells arc thread' Fig. jq H.'l'sphierixcii In the sphere on the Ictt the protopin-'n his .rat head a' j .t -h rtuiiei linirrg the wall; that on the ri^ht has shed the I'Utcr ni'nbrt'K. in r*! d iui^hter tells base separated from the wall; below ere iiur 't.>.re> o- th. d . 'in r i t zoospores Fig. 40. Rhabdosphaera Fit, 41. Cocoosphatra hit, 4. Irvl ' X 1,400 X 1,103 ni'Ui'ixa4 like and cling together in bundles (Fig. 42). Irichodes- mium, which is reddish-brown, occurs periodically in the Red Sea in huge quantities. In tropical seas it sometimes spreads out for miles in large floating masses, and is ot- casionally to be found in temperate regions also. All these minute organisms which make up the ph% to- plankton, together with those which may be regarded as M 2 1 64 The Plants on the borderland between plants and animals, form a collection of living material always more or less present in the sea, never evenly distributed, but occurring at various depths and in various localities, their appearance and disappearance governed by laws which are being in- vestigated continually and which ever produce new prob- lems to solve. The importance of detailed notes on any group cannot be overestimated, and the microplankton, both animal and vegetable, offers a field for investigation where new organisms and new facts will be continually discovered. IV ^(.nnld.) F.'.Wi/ P Units ‘ UY MRS. DR. A. \MUtR-\AN liOSSi-. After a storm heaps of plants are oast ashore bv the sea, especially if the coast is rocky. On closer examination, these entangled masses are found to consist of green, brown, purple, or bright red plants. What are these : Where do thev live ? These bright-coloured plants are algae, or dowerless plants, distinguishable bv their colour and various shapes from marine flowering plants; the latter have, as a rule, narrow, long, linear leaves of a dark green colour, as for instance the common sea grass, Zostera marina, so well known on English coasts. In various parts of the Mediter- ranean we find the sea grasses Ct'modocca nodosa and Posidonia oceanica, while Zostera nana is the only marine flowering plant of the Caspian Sea. In the tropics the genera Thalassia, Halophila, and Enhalus will often form green belts along the coast mixed up with various algae. But however interesting the study of the ‘Sea Grasses’ may have become since Ostenfeld published his paper on their geographical distribution,- this section treats only of attached or fixed marine algae. The mostly microscopical drifting algae, belonging to the plankton, have been dealt with in the previous pages. A few words will be said of algae growing in brackish water, its interesting flora showing the influence of the changing medium ; for some marine algae migrate through brackish water into fresh water, which has, besides, a rich flora of its own. ' The original drawings which illuscralc this section were made under the author’s direction by Mr. J. Ohbes. = Ostenfeld, On t)ie Geographical Distribution of Sea Grasses, Ptoceedingt Ray. Soe. Victoria, vol. xjtvii, N.S , pt. 2, Melbourne, 1915. 1 66 The Plants Algae can assume the most different shapes; their fronds may be either simple or branched filaments (Chaeto- morpha, Cladophora, Polysiphonia), or expanded mem- branes of various colours, composed of one or more layers of cells (Monostroma, Ulva, Porphyra); they may have the form of a smaller or larger bush with stem and branches (Splachnidium, Desmarestia, Delesseria) ; or look like a delicate bit of lace (Vanvoorstia, Claudea); or be com- posed of a long, flexible stalk, bearing long linear blades at its swollen top (Ecklonia buccinalis).' Again, others form crusts, adhering closely to the substratum (Ralfsia, Hildenbrandtia), or have crust-like . or ramified fronds entirely solidified with carbonate of lime (Lithothamnium, Corallina). The diversity of their ‘thallus’ (as their fronds are usually called) is enormous; some are so small that they nearly escape our notice, and others so large that stem and leaves have a length of hundreds of feet. But however great their outward diversity may be, their inner structure is comparatively simple, and built up of either thin- or thick-walled cells. Nowhere do we find a differentiation of these cells into vascular bundles, as in flowering plants. Neither have algae any true roots that can be compared to the roots of flowering plants; their holdfasts are in many cases a few rootlets or a whorl of so-called hapteres ; other algae fix themselves with a smaller or larger disk on the substratum, and algae growing in sandy bottom have long, hair-like, much-branched rootlets which fasten them- selves to the grains of sand, and enclose so many of them that, when the plant is withdrawn from the sand, its rootlets present the appearance of a large lump. Algae have various colours, and in accordance with these they are divided into: I. Blue algae — Myxophyceae or Cyanophyceae. i67 Algae 2. Green algae — Clilorophyceae. 3. Brown algae — Phaeoph}'ceae. 4. Red algae — Rhodophyceae or Florideae. We must bear in mind, however, that, though this division is right in the main, there are many algae that do not show the colour of the group to which they belong. Blue algae living in deep water may become red ; red algae may turn purple, brown-red, green, or even yellow, on being more or less exposed to. the glare of the sun; but with a little practice the group to which a certain alga belongs is easily recognized; for differences of structure and fructification correspond with the differences of colour. Algae are found on all coasts in both hemispheres if the outward conditions are favourable for their growth. We find them in the Arctic and Antarctic as well as in the tropics, forming a more or less well-defined belt along the coast. Some algae grow so high up that only the spray of the sea, as it is dashed upon the beach or against the rock, can reach them; others live in deeper water, from which only a dredge can bring them up. This belt of algae is divided into two regions; the littoral region, comprising that tract of the coast which lies dry at every ebb-tide ; and the sublittoral region, comprising all vegetation below low-water mark. The algae of the littoral region are adapted to a period of exposure at every ebb, the algae of the sublittoral region to a continually submerged exist- ence. Both the littoral and the sublittoral regions are again divided into an upper and a lower zone, according to the different algae found in each; for, though many algae have a wide distribution, yet almost every species has its own privileged locality In a vertical, as well as in a horizontal, direction. In this connexion it will be worth the student’s while 1 68 The Plants to study Cotton’s ‘Clare Island Survey’, which gives an elaborate ecological and distributional survey of that island. The internal organization and the external agencies determine the place a given alga will occupy. The external agencies are light, the nature of the bottom, the salinity of the water, the strength of the currents, &c. To under- stand their influence well, one has to study different localities, for agencies exercising a certain influence on the coast of Greenland will have an entirely different effect in the tropics, according to their different combination. Light is the chief factor; every plant wants light to reduce the carbonic acid needed for its existence. Some algae may want more, others less; but without light no vegetation can exist. Different men of science have studied the problem of how far light penetrates in sea water; for the old method of lowering a white disk and registering the depth at which it was still visible, has long been felt to be unsatisfactory. Photo-electric cells are far more reliable, and Gail and Shelford published in 1922 the results of their observations made with the Kunz photo- electric cell on the penetration of light in the waters of the Puget Sound. They made their study with particular reference to the distribution of algae. Poole and Atkins carried on these studies in the English Channel near Ply- mouth by a telephone method of measuring photo-electric currents, which enables measurements of submarine illu- mination to be made in a small vessel by means of photo- electric cells. The description of the method and the apparatus must be studied in the elaborate papers of the authors.^ * H. H. Poole, On the Photo-Electric Measurement of Submarine Illumina- tion; Sclent. Proc. R. Dublin Soc., 1925, vol. xviii, No. 9. H. H. Poole and W. R. G. Atkins, On the Penetration of Light into Sea Water; Joum. cf the Mar. Biol. Association, Plymouth, 1926, vol. xiv, No. i. . Light 169 The results Poole and^^Jcins arrived at show some minor differences from the results Gail and Shelford ob- tained. Both found that the absorption coefficients of liEjht are greater when the water contains much floating material. In Plymouth water the least loss is w'ith a glassy sea and a grey sky, and in Puget Sound waters the maximum of photosynthesis takes place with a clear skv and smooth water. In Puget Sound 35 m. depth is about the lower limit at w'hich photosynthesis takes place. In the English Channel water with glassy surface transmits 0-54 per cent, of the vertical illumination to 34- 8 m. The greatest depth at which encrusting calcareous algae were found in the Channel is 50 m. (Mme Lemoine). Interesting as the results of Poole and Atkins are, we must not forget that their coefficients of absorption are found for northern seas. In tropical oceans with an almost ever clear sky, the illumination will be quite different, and a botanist may say ‘I have found algae at a depth of 1 20 m. (Lithophyllum Philippi in the Mediterranean) and there- fore some rays of light must penetrate so far’. But this actual limit gives us no idea of the quantity of light each alga needs to thrive well. Each alga has three limits with regard to light; an optimum, a maximum, and a mini- mum limit. The optimum limit is that degree of light in which a given alga thrives best; the maximum limit indicates the strongest; and the minimum limit the feeblest degree of light in which the same plant contrives to live. The four groups of algae just mentioned have different optima with regard to light. The blue and green algae live mostly in the upper part of the littoral region; the brown algae love the sunny places of the lower littoral and the upper part of the sublittoral region ; and red algae are often found in the sublittoral region or in localities 1 70 The Plants sheltered by overhanging rocks or corals. But, true as this division may be in a very general way, there are many exceptions : some red algae can even live in places exposed to the glare of the sun on reefs in the tropics, though they loose their bright red colours in these exposed localities. Green alga again may go down to a depth of lOO m. (Dry Tortugas). Each alga thrives best in the place of its optimum limit and will chiefly be found in it. If we study the alga belt of a given coast, we see that algae live in zones often over- lapping one another, and that each zone has its own par- ticular inhabitants. These vary in different parts of the world: in Iceland the zones consist of different algae from those of the coast of England or of France, not to mention the different vegetation in southern or tropical seas. Light is the chief factor in determining the place a given alga will occupy. But there remain still many puzzling facts and questions open to controversy. These cannot be treated here, for it would lead us too far. It may be mentioned that according to Lubimenko* each species of red algae may diminish or add to the total quantity of its pigments (Chlorophyll and Phycoerythrine), according to the depth of its habitat and that the proportion of Phycoerythrine is variable in the different species and becomes greater with the depth of the habitat. Depth. When speaking of the penetration of light into sea-water, the depth at which some algae occur has already been mentioned. The greatest depths from which algae have been recorded are near Capri at 66-5 to 71 fathoms; near Minorca, 90 fathoms; near Spitzbergen, 150 fathoms; in the Barents Sea, 175 fathoms; in the Antarc- tic, 55-5 to 85-5 fathoms; and south of the Cape, at 40 » Lubimenko, Sur Tadaptation chromatique chez les algues marines, Comptc Rendus Ac. Sc. clxxxxi, 1925, p. 730. Nature oj Bottom 171 fathoms. Probably, however, the algae found near Spitz- bergen, and in the Barents Sea, will have sunk into these depths from shallower localities. The Bottom is another factor of great influence. A rocky and protected shore is a favourite haunt of al^ae, though many species prefer the high surf of the exposed shore. It makes a great difference whether rocks rise perpendicularly or whether they form big boulders or flats with rockpools filled with fresh sea-water at every flood and lying exposed at ebb-tide. The perpendicular rocks afford no shade to algae and only light-loving species will inhabit them, whereas broken rocks afford shelter from the sunlight and species loving shade will be found there. Rockpools are a wonderful place for collecting algae, and collectors should bear this in mind when studying sea charts in search of a fit place where they may expect algae. A sandy bottom is poor in algae, on account of the shifting of the sand by the movement of the vater, and only species with long hair-like rootlets, as for instance species of Caulerpa, can contrive to live in it. For the same reason, algae avoid places covered with fine mud, though Caulerpas may also be found there. If, however, a bit of stone or wood rises out of the mud or the sand, it will soon be covered by algae. Sea grass (Zostera marina), which is really a true flowering plant, thrives well in a sandy bottom where it can cover whole stretches and form submarine meadows. Such Zostera fields carry a rich alga] vegetation, mostly of small species, attached to the Zostera leaves. Salinity. Sea-water contains, on the average, in the open sea from 3-5 to 4 per cent, of salts; but on approaching the coast this percentage decreases in different degrees, ac- cording to the quantity of fresh water brought, by land 172 The Plants drainage, into the sea. Marine algae flourish luxuriantly in this medium, but if the percentage of salt falls con- siderably — for instance, when big streams pour their fresh water into inland seas (Black Sea, Baltic Sea, White Sea), and the water becomes brackish — true marine algae grow scarce. The transition from a rich algal vegetation to a poor one is, however, gradual, and such inland seas or big estuaries are interesting spots for investigation. Gobi wrote two memoirs on the brown and red algae of the Finnish Gulf, where the salinity of the water gets very low. The results he obtained are interesting, for with lessening salinity Phyllophora underwent such changes that, the modified forms would be taken for autonomic species if one had not seen the intervening specimens. Marine algae have an optimum, maximum, and minimum limit, with regard to salinity as well as with regard to light, and for some species those limits lie far apart. The Movement of the water also influences the growth of algae. Some like places where the tidal waves rush along the coast; others prefer exposure to the high surf of the sea; and, again, others thrive well only in quiet pools. Algae are admirably adapted to profit by, or to withstand, the strong force of the waves. Some surf-loving algae have long flexible stalks that follow the movement of the water, while the blades at their top swing to and fro in the waves ; other algae in the same locality can withstand the action of the waves, because their tissues are encrusted with carbonate of lime. Ocean currents are an important factor in the distribution of algae, and tidal currents often favour the vegetation of algae. Temperature. Some algae can endure great diflFerences of temperature without damage. Kjellman observed in Nameless Bay, Nova Zembla, Enteromorpha minima, a green alga, in fresh water of 0° C., and in sea-water of 4“ T tmperatarc i ~j? to C. In the tropics the algae arc exposed to the glare of the sun during ebb-tide, and :n the arctic ^egiol^.^ Kjellman found Laminariac thriving welt in a temrerature of C., and on being exposed at ebb-tide, they even withstood a temperature of— cc C. Alsrae want, however, as a rule a more congenial tem- perature, and if the coast of Norway has >uch a decidedlv different vegetation from the coast of adjacent Polar land>, this is mainly due to the North Atlantic Drift, the in- fluence of which is felt even on the \\esr coa'^i of No^m Z embla. Still more striking perhaps is the di^erence on the west and east coast of South Africa. On the latter, bathed by the Mozambique current coming froni the tropics, tropical and subtropical algae flourish; while or, the west coast, with its Benguela current partb conning from the Antarctic, we find a totally dirterent vegetarior:. Many more examples might be cited, ail showing clearly the great influence which the temperature of the sea has on vegetation. Professor Farlow describes how Cape C od, in North America, is a boundarv between rw o vegetat'ons. To the north of Cape Cod, under the innuence of the Labrador current, Arctic and northern European forms occur; to the south, with its Gulf Stream, species charac- teristic of w^armer seas. Professor Setchell was able to show' that a change in the kelp flora of western North America takes place w'ith the increase of ever)- 5 C. of surface temperature. We touch here upon the question of the geographical distribution of algae. This is a question wanting caretul investigation, for so many agencies exercise an influence on the occurrence of algae in a given locality. The Suiwey of Clare Island by Cotton has already been mentioned; the researches of Berthold in the Gulf of Naples, of Kjellman in the Arctic Sea, of Koiderup Rosenvinge on the coast of 174 Plarits Greenland, of Borgesen at the Faroe Islands, of Svedelius, Kylin, Skottsberg and many others, have opened a rich field of study with regard to the conditions of algal life and the physiognomy which algae give to the coast. When searching for algae we observe that large stretches of the coast are characterized by one or two — sometimes more — algae living, in large numbers, under the same outward conditions. Other algae may be mixed with them, but they are always in smaller numbers, or less conspicuous; they do not put a stamp on the vegetation, as the others do. Thus, for instance, Ecklonia buccinalis is a most characteristic alga of the Cape, covering large stretches of the coast in the sublittoral region. Little red algae live parasitically on its stem, but they are entirely hidden under the mass of brown blades floating on the sea. Borgesen used the term ‘association’ for such a community of algae, and when several associations live under the same outward conditions he called it a ‘formation’. Borgesen distinguishes, for instance, a formation of encrusting algae on the coast of the Faroe Islands, and all encrusting algae belonging to it and live under the same external condi- tions. Borgesen calls it the Hildenbrandtia formation, after the little red alga, Hildenbrandtia, which predomi- nates in it. Cotton distinguishes a Fucaceae formation in the British Islands consisting of: ‘Pelvetia canaliculata, Fucus spiralis, Ascophyllum nodosum (moderate shelter necessary), Fucus vesiculosus, Fucus serratus, Fucus cera- noides (admixture of fresh water necessary), occurring in the order given.’ Such formations and associations can vary endlessly in the different seas, according to the nature of the bottom, the salinity of the water, &c.; Svedelius has even described, for the Baltic, two detached associations of Phyllophora and Fucus at a depth of 4 to 5-5 fathoms, of which the detached specimens are Geographical Distribution 175 •absolutely sterile and have fronds far narrower than the attached plants, growing a considerable distance away, from which the loose ones have been derived. A thorough knowledge of the formations and their components on a given coast is the corner-stone for the study of the geo- graphical distribution of algae, so interesting in itself because it may help us to reconstruct the former aspect of the globe. By a careful study of the algae of the Baltic and Arctic Seas, Svedelius arrived at conclusions which strengthen the already existing views of zoologists, about a connexion of those seas in the Glacial Period, and in his later paper on the discontinuous geographical distribution of some tropical and subtropical marine algae, he shows that the majority of the older genera of algae have evidently had their main distribution in the Indo-Pacific Ocean, whence they have migrated into the Atlantic, in Tertiary times. A. Forti* has in a recent paper, illustrated with most interesting photographs, described fossil algae from Bolca (Verona) that confirm an analogy between the ter- tiary flora of the Mediterranean and the still existing flora of the Indo-Pacific. Borgesen has given beautiful photographs of various formations occurring on the coast of the Faroes, and so has Cotton for Clare Island. How much might be done in this line if some of the many photographers of our day took to photographing algae! The field is still new, and might tempt a spirit keen to surmount difficulties. It is impossible to give in a few pages a systematic survey of so large a group as the algae; their systematic arrangement can be studied in books written for that purpose. The names of the principal divisions only are given here with the most striking characteristics that are of value for the collector. Figures of very common or I A. Forti, Alghe del Pdicogenc di Bolca, Padova, 1926. 176 The Plants very striking algae are inserted, and a few families are dealt with in more detail. The questions touched upon will show how interesting those families are, but other families are equally interesting from similar or different points of view. Therefore collectors of algae, who are not botanists, may feel sure that they can render good service to science if they will only collect carefully, as only well- prepared and duly labelled collections are of value. Two little groups of algae may be mentioned first: the Parasites and the algae living in symbiosis with other organisms. True symbiosis is of profit to both partners, whereas parasites live more or less at the cost of their host. Symbiosis may take place between two algae or between algae and sponges. A beautiful instance of symbiosis is the green alga Spongocladia vaucheriaeformis ; its long ramified filaments are entirely interwoven in the tissue of the sponge. The author of this chapter could follow the transformation into Spongocladia-like filaments of the leaf-like green Struvea delicatula while living with a sponge. Ceratodictyon is a red alga always found living in symbiosis with a sponge, and different species of Phormidium (Myxophyceae) inhabit different sponges. Parasitic algae live on and in other algae. They may be so small that only the microscope will show their pres- ence, or large and visible to the naked eye. The best- known parasite of the English coast is Polysiphonia fas- tigiata occurring on Ascophyllum nodosum. Its parasitic nature was first revealed by Sauvageau (1921)' and after- wards confirmed by L. Batten (1923).* But the number of known parasites is fast increasing. They occur in temperate and in tropical oceans and are often of the same ' Sauvageau, Observations biologiques sur le Polysiphonia fastigiata, Recueil des tra’v. hot. NSerlandais, 1921, p. 213. > L. Batten, The genus Polysiphonia, Linneans Soc. Joum. Bot., vol. xlvi, 123. Blue-green Algae 177 family as their host. The parasitic nature of Actinococcus, Colacolepis, and Steryocolax, living on Gymnogongrus, Phyllophora, and Ahnfeldtia, is however doubtful, since R. W. Phillips suggested that investigation will probably prove that they are the tetrasporangia-bearing form of their host. The M3rxophyceae, or blue-green algae, are small, and the least organized of the algal tribe; they are either uni- or multi-cellular. The latter form long filaments enclosed in a common cell-sheath, and dividing, for the sake of reproduction, into several portions called ‘hormogonia’, which move out of the sheath. The young hormogonia secrete a new cell-sheath, and develop by division of their cells into a new filament. Myxophyceae, after having been collected, should, as soon as possible, be spread out on the paper on which they are to dry, and not left to stand over- night. If this is neglected, the hormogonia will, in all probability, have moved out from the sheaths and the specimen with be worthless. Myxophyceae live, as a rule, in the upper littoral zone, attached to stones, pebbles, or other algae. Lyngbya majuscula — the mermaid’s hair — is sometimes found floating on the sea. Some species of Calothrix (Fig. 46) form patches on the rocks; species of Rivularia have a hemispherical thallus, either solid, from as big as a pin’s head to J in. in diameter, or hollow and forming soft expansions. Sometimes Scytonema and Lyngbya will cover the rocks with a slippery coating, disagreeable, and even dangerous, to the collector when the waves are strong. Myxophyceae are found in all temperate and tropical seas; their genera and species have, as a rule, a very wide distribution. In the Arctic and Antarctic they seem to be rare. Perforating Algae. Several Myxophyceae and a few 178 The Plants Chlornphyceae live in the shells of molluscs, in the skeleton of corals, or in calcareous algae, where they bring about bluish, green, or violet spots. They may even attack and bore into the surface of calcareous rocks, where they biing about a blue-greenish colour. These ‘per- forating algae’, as they are called, on account of their power to bore into calcareous tissues, are known from the Chaetomorpha temperate and tropical seas. By their incessant labour they destroy many shells, corals, and calcareous algae, setting free carbonate of lime. The Chlorophyceae, or green algae, abound in the littoral, and are also found in the sublittoral, region. Their usual way of reproduction is by spores, which move by the aid of cilia. Some of these spores conjugate before germination, others germinate without conjugation. Of Caulerpa no reproductive organs are known. Many formations of green algae are known from the temperate and tropical seas. The filamentous Urospora, Chaeto- Fig 46 Calothrix The hormogonia have left the sheath or are moYing out of it Fig 47 Halimeda maoroloba i8o The Plants morpha (Fig. 43), Cladophora (Fig. 44), the membrane- like Monostroma and Ulva (Fig. 4 c). the tube-like En- teromorpha, prevail in the temperate seas, while the large group of the Siphoneae occurs chiefly in the tropics or in warmer temperate seas. The Siphoneae attain a very high Fig. 4S Ildlimcda outer difFerentiatiorij but each member of this group, how- ever complicated the structure of its frond may be, con- sists of only one multinucleated cell ; therefore they are called Siphoneae from their tubular nature. Many Siphoneae : Flalimeda (Figs. 47, 48), Neomeris, Aceta- bularia (Fig. 50), Udothea (Fig. 49), Bornetella (Fig. 5 r), Penicillus, &c., are encrusted with carbonate of lime, and to encrustation Flalimeda owes its importance as a reef- Brown Algae 1 8 1 forming organism. The Caulerpae (Figs. 52, 53) cover the coral reefs with numerous species of multifarious aspect, or live in a depth of several fathoms. Outside the tropics only a few representatives of this family are known, e.g. Caulerpa prolifera occurs plentifully in the Mediter- ranean ; three species occur on the coast of New Zealand ; Fig. 4.9 Udothca and five or six species along the coast of Japan, in the warm waters of the Kuro Shiwo, the ‘Gulf Stream’ of the Pacific. Formerly the Phaeophyceae were divided into Phaeo- sporeae, Dictyotaceae and Fucaceae, but since the first edition of this book the systematic classification of the Phaeosporeae has undergone great changes. Sauvageau was the first to discover the alternation of generations in Laminaria. Not only many brown algae, but also many 1 82 The Plants Rhodophyceae have an alternation of generations; this consists in a generation producing only asexual motile or motionless spores (zoospores) which germinate and develop sexual female egg-cells and male antherozoids (spermatia). After fertilization of the egg-cell by an an- therozoid this grows out again into a plant producing only zoospores. The alternation is, however, not always so regular, above all not in the Rhodophyceae. The sexual and asexual plants of the brown algae are often of very different sizes. The sexual plants of the Laminariales are so small that they escaped notice till Sauvageau detected them. Other algologists discovered them in various members of the Laminariales, and Kylin cut up the old group of the Phaeosporeae into Ectocar- pales, Sphacelariales, Cutleriales, Laminariales, Tilopteri- dales, to which afterwards Taylor added the Dictyosi- phonales, and Sauvageau the Sporochnales. In the northern seas we find numberless Ectocarpales, either as slender filaments (Ectocarpus) or forming crusts on stones and shells (Ralfsia) or bladders on other algae (Leathesia), and many more. Sphacelariales have erect shoots springing from a basal disk and a large top cell visible by naked eye. They live mostly epiphytic, but also endophytic and parasitic on stems of Fucus and Sargassum in almost all seas. The alternating generations of Cutleria were formerly considered as distinct genera under the names of Aglao- zonia, the asexual and Cutleria, the sexual generation. The blades of the Laminariales bear unilocular zoo- sporangia, and small microscopical plants born from the zoospores produce the antheridia and the egg cell that after fertilization develops into the enormous alga. The Laminariales are all alike in their youngest stages, but afterwards the blades and stripes undergo many changes Caulerpa prolifera 184 The Plants in the different genera that play such an important part in the physiognomy of some coasts. They live in the littoral and sublittoral zone, though it is reported that Macrocystis pyrifera (Fig. 54), with a frond attaining a length of 200 m., goes down to a great depth in the southern hemisphere. They all love exposed places, and Darwin wrote in his ‘Narrative of the Beagle’: ‘I know few things more surprising than to see the Kelp, or Macrocystis pyrifera, grow- ing and flourishing amongst those great breakers of the western ocean, which no mass of rock, let it be ever so hard, can long resist’. The Laminariales contain much mucilaginous substance in their tissue, and the following fact, recorded by Mrs. Vallentin, from the Falkland Islands, is interesting. After describing the enormous quantities of Kelp washed ashore after a heavy gale, she continues: ‘When the ‘Kelp’ is being torn up and the fronds and stems are being broken by the fury of the elements, the muci- laginous substance exuded from these broken seaweeds is so great, that it has almost the effect of oil in smoothing the crests of the waves. This is at times so markedly the case that the rollers lose much of their danger.’^ Laminariales avoid the tropics, but occur on almost all coasts in the temperate and Polar seas. The latter are rich in representatives of Laminaria, Alaria, Agarum ; only three Laminariae are known from the south-west coast of Africa. ' Mrs. Vallentin in Cotton, On the Cryptogams of the Falkland Islands, p. 141. Fig. 56 Lessonia Fig 57 Egregia 1 86 The Plants Saccorhiza bulbosa (Fig. 55), common on the coast of England and Norway, is the largest European alga, with a height of 2 to 4-60 m., and a blade 2 to 4 m. in length. Lessonia (Fig. 56) and Ecklonia (Fig. 58) are prin- cipally inhabitants of the southern hemisphere, only three Eckloniae being known from Japan. Lessonia fuscescens assumes, on the south-west coast of America, the habitus of a tree with a height of 3 to 4 m., and a trunk as large as a man’s thigh. From California are known Dictyoneuron, Postelsia, Egregia (Fig. 57), Eisenia, and the gigantic Nereocystis (Fig. 59). I should like to give some details about the Dictyosi- phonales, Tilopteridales, and Sporochnales, but space is wanting. Their mode of sexual reproduction and their anatomical structure show that they are the representatives of different groups. Sauvageau urged the question that these groups, some of which have only very few actual living members, are the remnants of a rich vegetation in former geological periods. The Dictyotales (Dictyotaceae) have light- or dark- brown ribbon-like or fan-shaped fronds, these latter often splitting up into narrow, long pieces. Their reproductive organs are motionless eggs, and motile antherozoids con- tained in antheridia, which form small groups, called ‘sori’, visible to the naked eye, and scattered over the surface of the frond. Padina pavonia has a fan-shaped frond, encrusted with carbonate of lime; it occurs on the English coast with Dictyota dichotoma (Fig. 60), Taonia and Halyseris ; but Dictyotaceae otherwise prefer warmer seas, and flourish especially well in the tropics. Lobospira is exclusively an Australian genus. Fucales or Fucaceae. The large family of the Fucaceae is characterized by their non-motile eggs and motile antherozoids being contained in special conceptacula, 1 8 8 The Plants embedded either in the frond (Fucus, Fig. 6 1) or in special fruit-bearing organs (Sargassum, Fig. 62). The form of their fronds is widely different; many of them have air- bladders in the continuation of the frond (Fucus vesicu- losus) or special air vesicles, allowing the plant to float on the sea, when it has been torn away from its support (Sargassum). In the arctic and northern temperate regions Fucus, Pelvetia, Ascophyllum, Himanthalia (Fig. 63), and Halidrys, cover the coast with a dense vegetation. In the southern region the Laminaria-like Durvillea flourishes (Fig. 64). Australia and New Zealand are very rich in Fucaceae; we must mention from there the necklace-like Hormosira, with its parasite Notheia (Fig. 65). Scyto- thalia occurs in the Antarctic. Turbinaria is strictly tropical; Cystoseira, Cystophyllum, and Sargassum (Fig. 66) belong to the tropical and warm temperate seas. Of this latter genus a few words more must be said, as it has a right to claim the special attention of collectors. In each complete Sargassum we can distinguish two parts — a short principal axis, with a tuft of rather broad basal leaves, and long branching shoots of a secondary order, with, as a rule, leaves of different size, air vesicles, and special fruiting branches. When basal and upper leaves differ from each other the transition is gradual. The fruiting branches bear conceptacula which are highly differentiated in the different species. In some species the branches with male conceptacula seem to be cylindrical, and those with female conceptacula dentate, but this ques- tion must still be carefully investigated. It is important, because J. G. Agardh divided the Sargassa into various groups, according to the form of the conceptacula-bearing branches. The genus Sargassum has approximately 230 species, according to Grunow’s latest census, and is distributed Fig. 6o. Dictyota dichotoma The sorus above with motionless eggs, below with ajitheridia Fig. 6i. Section through frond of Fucus y a, Air vesicles; b, cylindrical conceptacula; c, leaf 190 The Plants over the warmer temperate and tropical seas; it is especially frequent on the coast of Australia. One species has for four centuries attracted the attention of men of science — Sargassum bacciferum — the alga of the famous Sargasso Sea. Discovered by Columbus, de- scribed by Humboldt, the origin of these enormous quan- tities of drifting seaweed is, after four centuries, still practically unknown. There are two opinions about the origin of S. bacciferum. Humboldt, Forbes, Piccone and A. Agassiz, allow the possibility of the present S. bac- ciferum being the floating form of one or more Sargassa that lived attached during a former period of the earth’s history. If this were true, S. bacciferum must have pro- pagated itself vegetatively for long ages, because it has been collected almost always in a barren state. Kriimmel is of another opinion. In his beautiful work, ‘Die Reise- beschreibung der Plankton-Expedition’, he gives a very interesting chart of the distribution of S. bacciferum in the Atlantic, thereby dispelling Humboldt’s error, who be- lieved in the existence of two fixed banks of Sargassum. Krtimmel assures us farther, on the authority of Kuntze, that S. bacciferum and S. vulgare are identical, and he be- lieves that S. bacciferum is the floating form of S. vulgare, which occurs plentifully on the coast of the Bermudas. But against this opinion Sauvageau, and afterwards Bor- gesen, have protested most energetically; they maintain that S. bacciferum and S. vulgare are not identical. By Sauvageau’s kindness, for w^hich I hereby wish to express my sincere thanks, I am allowed to make use of letters written to him by the American algologists, Farlow and Collins, who have studied the flora of the American seas, and agree that S. bacciferum and S. vulgare are different plants. Mr. Collins, in speaking of S. vulgare, says: ‘The plant in question grows attached in shallow water; I do 192 The Plants not think it occurs floating, except as any other alga that is washed ashore when torn from its fastening, but not continuing to live in a floating state.’ According to Professor Farlow, it might be possible that S. bacciferum is a floating form of S. lendigerum, common in the Ber- mudas, but this is only a supposition ; and Collins, again, says, speaking of S. lendigerum, that its shoots soon perish after having been detached. Farlow, Collins, and Howe agree thatS. bacciferum does not occur on the coast of America growing attached. The question would be solved if S. bacciferum could be forced to grow fertile specimens. This might perhaps be brought about in artificial cultures. Fertile specimens would show if S. bacciferum is only the floating form of an existing species or whether it is an autonomic species. It is clear that intelligent collectors could do much to advance our knowledge of the Sargassa. They should try, above all, to bring home entire plants, with rootlets and basal leaves as well as with the long secondary shoots, and not forget that fruit-bearing branches are of great interest. Rhodophyceae. In genera and species, though not in bulk, the greater part of marine algae are Rhodophyceae ; they are usually small in size, with the exception of some species of Halymenia, Gigartina, Iridaea, &c. They are, as a rule, soft and slippery to the touch, and of a bright red colour, except in places exposed to the glare of the sun. The beauty of red algae is something exquisite, and every collector will consider his trouble well rewarded if he has been fortunate enough to collect some of these lovely algae. The organs of fructification of this great group are essentially alike, and consist of asexual cruciate or tripar- tite motionless or slowly sliding spores, antherozoids, and a carpogonium with a basal part containing the egg- Rhodophyceae 193 cell, and a hair-like top, called the ‘trichogyne’. 'I'he small antherozoids adhere to the trichogyne, pierce its cell-wall and their contents, and entering the trichogyne, fertilize the Fig. 66. Sargassum egg-cell. After this fertilization, a complicatedprocess takes place, the result of which is the cystocarpic fruit, consisting of a globular mass of spores, either borne free (Nemalion), buried in the tissue of the frond (Halymenia), or contained 194 Plants within special conceptacles. These latter are the typical cystocarpic fruits proper to the group of the Rhodo- melaceae. Polysiphonia urceolata (Fig. 67), common on the English coast, is a good example to show the cysto- carpic fruit. To the Rhodophyceae we may also reckon at least par- tially the Bangiales, formerly called Proto-Florideae, since the beautiful investigationsof R. W. Phillips on Gracilaria.' In Cracilaria and in the genus Porphyra (Bangiales) the trichogyne is reduced to a simple papilla on a peripheral cell, which in this case is the Carpogonium. In Porphyra the carpogonium after fertilization may become a spore or divide into several spores; in Gracilaria fertilization ‘is followed by the formation of the cystocarpical swelling and the egg-cell embedded at the base of the proliferated tissue, gives rise ultimately to the carpospores’. The Rhodophyceae, or Florideae, are found in all seas, but they prefer the temperate and tropical regions. Many of their genera have a limited, others a wide, distribution, but it does not follow that the species of these latter have also a wide distribution. It may be so, but more often they are limited to smaller localities. The Rhodophyceae are divided into Bangiales (pr. p.), Nemalionales, Cryptonemiales, Ceramiales, Gigartinales, Rhodymeniales. Among the Bangiales, Porphyra is the most common and well-known alga of English shores. To the Cryptonemiales belongs an important group amongst the red algae, the Corallinaceae, so called on account of the carbonate of lime which they secrete and deposit in their cell membranes, with the exception of the cells of the reproductive organs and those in the joints of the Corallineae. By this process they grow as hard as stone, and have for this very reason contributed largely to the * R. W. Phillips, Cystocarps of Gracilaria, Ann. of Bot. 1925, p. 39 Fic. 67. Poiysiphonia 196 The Plants formation of the earth’s crust. They are known from very early periods in the earth’s history, for Lithothamnium jurassicum is already reported from Jurassic strata ; in Cre- taceous deposits they appear plentifully, and in Tertiary times they were so numerous that entire mountains were builtup by them. Thewell-known Latomien, near Girgenti, in Sicily, are quarries in rocks consisting of Lithothamnia. In our present day they continue their work, and occur in enormous quantities, forming banks in the polar, tem- perate, and also in the tropical seas, where they contribute largely to the building up of coral reefs. The late Mr. Foslie, the well-known authority on Lithothamnioneae, kindly told me, in a letter which he allowed me to quote, that the number of known species was 276. Of these, approximately 1 2 are strictly arctic or sub-arctic ; 8 2 are limited to the tropics; and 182 occur in the temperate zones. Since his death the work on the Melobesieae has been pursued by Mme Lemoine, who besides giving us a good classification, has described many new genera and species. The Corallinaceae are divided into Corallineae and Melobesieae. From Corallineae are best known the slender and bush-like Corallina, Amphiroa, and Arthro- cardia. To the Melobesieae belong Melobesia and the encrusting or free-branching Lithophyllum (Fig. 68) and Lithothamnium (Fig. 69). The encrusting forms often act like mortar to cement loose material together ; the free- branching forms develop into nodules even bigger than a man’s fist, and, cemented together, they can form masses of more than a metre in diameter. All Corallinaceae are fond of strong currents ; Corallina and allied genera love places exposed to the surf of the sea; Lithophyllum and Lithothamnium prefer the sublittoral zone, and go down to depths even of 30 to 40 fathoms. The author has seen a bank of Lithothamnium dry at spring-tide, near the Coral/inene 197 island of Timor, but this is an exceptional case. It is a curious fact that, however large a Lithothamnium bank may be, it consists chiefly of only one species. Such banks figure frequently on sea-charts as ‘corals’. For the Litho- Fic 68 Llthoph^llum Frc 69 Archaeolithothamnium Frc. 70 Claude J Fic. 71. Rhod\meni thamnioneae the rule holds good as for other algae, from warm to boreal or arctic seas the number of species de- creases, but the number of individuals increases. The red algae are so numerous that it is difficult to choose which deserve special mention. To the Ceramiales 198 The Plants belong Ptilota, common in northern seas ; Ceramium, Poly- siphonia (Fig. 67), Laurencia, plentiful both in temperate and tropical seas; the latter being the favourite haunt of Claudea (Fig. 70) and Martensia. Delesseriaceae are found in all seas, but seem to prefer colder water; Skotts- berg and Kylin have described many new genera and species from the An tarctic. The genus Eucheuma, belong- ing to the Gigartinales, abounds in the tropics, and Rhodymenia palmata (Fig. 71) is common on Irish and Scotch shores. I fear that I have perhaps already men- tioned too many. The reader may be reminded of Harvey’s Quaker, who said: ‘Brother, if thou knewest what I keep in, thou wouldst not grumble at what I let out.’ In 1899 there appeared from the hand of Professor W. A. Setchell a paper containing directions for collecting and preserving marine algae (‘Erythea’, vol. vii, No. 3). Whatever the learned Professor advises the collector to do, as to the best methods of collecting and preserving algae, is the best that can be done. I can only add a few details, and it is therefore with the kind permission of Professor Setchell that I transcribe here some passages out of his paper. For full information on the subject I refer the reader to that paper. Collecting and Preserving Methods. Wherever a col- lector goes to seek for algae, he must first find out the time of the tides from the local authorities. The collecting ground should be reached from one to three hours before the occurrence of extreme low water, in order that the collector may follow the tide as it goes down. In this way the littoral region can be explored ; the upper part of the sublittoral can only be reached by wading; its lower part lies too deep for wading, and recourse must be had to a boat. From the boat many species may be seized, and Collecting Methods 199 hauled into it by means of an ordinary garden rake, if the water be quiet enough; otherwise it will be necessarv to use the dredge, and that mu&t always be used as a matter of course where the water gets deeper. If the bottom is rough, care should be taken to protect the bag of the dredge on both sides with strong sailor’s canvas. If this is not done the meshes of the net will be easily caught by corals or stones; protected by canvas, the net will often glide over these impediments. The success attending the use of the dredge will depend upon local circumstances. Farlow, in his ‘Marine Algae of New England’, called a day spent in dredging a wasted day ; but during the Dutch Siboga Expedition the dredge has several times brought up a rich harvest, and, used by the Belgian Arctic Expedi- tion, the results w'ere often ‘a fine vegetation’. The author once had in the tropics, at a depth of 6 to 7 fathoms, the help of divers, and it proved a great success. Berthold attained much by diving himself. The new diving apparatus makes diving more attractive than it formerly was and may in time become a great help in collecting specimens not attainable by hand. Collectors should not forget to search the coast carefully after a stormy day, for at the time of maturity the algae of deeper water are more or less readily torn away from their attachments; they rise to the surface, or near to it, and are drifted ashore, especially after storms. To the collecting apparatus belongs, first of all, a general receptacle for the transport of the collected material. A good canvas or rubber bag, with a flap for closing the mouth, and a broad strap for the shoulder; a basket, a canvas pail, or a large net with straps, are all very useful implements, out of which one may be chosen ac- cording to the collector’s tastes. It is also very desirable to have some bottles with preserving fluids. To carry 200 The Plants these we used a zinc box (Fig- 72), 35 cm. long, 25 cm. high, and 1 1 cm. broad; it was a little concave on the side turned towards the body of the bearer. Inside it contained, at some distance from the bottom, a stand for loose bottles of various sizes. Specimens for drying may be wrapped at once in news- paper, a large quantity of which should form part of a collector’s outfit. A knife, well anointed with vaseline to prevent rust- ing, and a fairly powerful pocket lens, are absolutely essential. A spoon or a pair of small forceps are of service in gathering from rocks small gregarious algae, and a small geological hammer and chisel are necessary if occasion arises to collect en- crusting species growing on rocks. A pair of fisherman’s boots with extensible legs reaching to the hips are necessary for wading in temperate seas ; ladies will do well to dress in a rather heavy skirt of pure wool and high boots; in the tropics a bathing- costume of thin woollen material will be the thing for ladies, but care should be taken to protect the feet against the spines of sea-urchins or the rough surface of coral reefs. In some localities it will be safer for the collector to go out with a rope tied round the waist and firmly held by friends in a safe position on shore, on account of treacherous holes in the rocks and reefs; in others the waves may, at the turn of the tide, run in so suddenly and violently that danger of getting drowned is imminent if Fig. 7z. Zinc Box for carrying bottles a. Plan of stand for loose bottles Preserving Seaweeds 2 o i proper care is not taken. Collectors should study carefully each special locality — above all, on exposed rocky shores — before risking themselves too far out at ebb tide. On the collecting-ground the best specimens available should always be selected, and, if the fruit can be detected, fertile specimens should be carried home in preference to sterile ones. Specimens covered with parasites should only be gathered for the sake of the latter. When botanizing on algologically poor and rarely visited roasts or reefs, it is best to take whatever can be found, in order that we may get an insight into the nature of the flora; but, as a rule, only specimens in a good state of preservation should be taken. The methods of preserving algae depend on whether they are required for research in the laboratory or for the herbarium. For laboratory work they must be put in various fluids, best studied in papers enumerated at the end of this book. For each special investigation there are almost always special methods of preservation. For general investigation, algae may be put at once in alcohol of 70 per cent., or in a solution of formalin (i to 2 c.c. formalin in 99 to 98 c.c. fresh water).* In alcohol algae will shrink a little and lose their colour; in formalin they do not shrink, and keep their colour a long time if kept in the dark. But, in the long run, a solution of formalin affects the tissues slightly. One per cent, chrome alum in dis- tilled water, carefully filtered through sand, is, according to Guignard and Lotsy, an aqueous solution that does very well for Myxophyceae, Chlorophyceae, and Rhodo- phyceae; Phaeophyceae do better in formalin. For the herbarium algae can be either roughly dried, or mounted and pressed, or salted, as desired. Large algae, ‘ Se^-water can also be used for making up solutions, but rusts all steel instruments used on the material afterwards in a most annoying way. 202 The Plants such as Kelps or species of Fucus, can be very well dried in the air, care being taken to shut out the glare of the sun, in which the algae dry too soon and become brittle. The method of drying successfully in the air will depend upon the moistness or dryness of each climate. In Nor- way Kelps dried beautifully on being exposed to the wind. To save space, one may fold or roll up big algae, care being taken to do this before they are too dry and too brittle. They should afterwards be allowed to dry further. It is absolutely necessary to let Kelps dry before folding them up, or before putting the small ones into the press; for if put into the press at once they easily become mouldy. The dried specimens can afterwards be moistened again ; the water absorbed in soaking them out will be very easily given up in the press. The plants to be pressed for the herbarium ought to be carefully washed and sorted in sea-water, in which they retain their colours better than in fresh water. In a few cases it may be advisable to wash the algae in fresh water before putting them into press. Travellers will prefer photographic dishes for this sorting and the subsequent spreading, but collectors living near the sea-side will find common earthenware dishes, white on the inside, prefer- able. For floating out the specimens, a method of the late Mr. Holden is to be recommended, which I quote from Professor Setchell’s paper. The utensils are very simple — a shallow dish and a rect- angular piece of zinc, just a little smaller than the bottom of the dish. The corners of the zinc (or two sides) are bent over at a sharp angle, all the same way. The piece of zinc is then placed in the dish, with the bent portions downwards, so as to raise the main portion of the zinc somewhat from the bottom of the dish. Water is then poured in — fresh or salt, as the case may be — until it is Pressing for the Herbarium 203 just the least bit above the surface of the piece of zinc. Now take a piece ot uhite paper or card, ]a\ it with one surface on the zinc, and turn it over, thorou^hlv wetting both sides. Then place the specimen to be mounted, care- fully cleansed, upon the paper; with the forefinger of the left hand gently depress the paper and zinc, until the specimen is floating freely. With the right hand spread out the specimen upon the paper, u>ing a pair of needles or a soft brush to assist in the process. W hen finalK spread out gnidually release the pressure upon the i'inc, allowing it to rise and to drain off some of the water. Then gently lift the zinc until the whole of the specimen is out of the water. When the paper is fairly dry, but before the alga has begun to show the effects, the specimen is placed upon a sheet of diw'ing-paper. a thin cloth is put over it, another drier upon that, a specimen upon that, and so on. A flat board w'ith some stones makes a very good press; better still is a press consisting of two iron frames, with a network or iron wfire in the middle (Fig. 73), and each frame provided with four little hooks. These little hooks are fastened together by four copper chains that can be made shorter or longer, as circum- stances require. The chains must be fastened loosely, and only a little weight placed on the board in the beginning. The specimens may be pressed more strongly the second day. It is absolutely necessary to change the driers every- day, and it is better still to change them twice a day in the beginning. For mounting the algae, white, unglazed, tough paper is the best; the driers must be of pretty thick drying paper, or sheets of blotting paper. The cloth must be a thin cloth or muslin, entirely free from starch. Most of the finer specimens will adhere sufficiently to paper, but the coarser ones must be afterwards fastened on to the mount with 204 Plants narrow strips of gummed paper. We have already said that blue algae must be spread out at once after having been collected. It is best to let them dry in the air. Each specimen should be provided with a label, on which must be mentioned the locality where it was collected, the date, the nature of the bottom, the depth, and a number, begin- ning with I, and so on. The numbers of the specimens must also be entered in a notebook wherein all details are to be mentioned that strike the collector, such as the different shades of colour which some algae have in and out of the water, and the periods of the year in which they occur; for many algae have regular periods of growth and rest just like the higher flowering plants. Utility of Algae. The commercial value of algae is relatively small. Formerly large quantities of Kelp and Fuci were gathered, and even grown, on the coasts of islands near Ireland and Scotland, for the manufacturing of carbonate of soda. This was, however, afterwards given up. Kelps are still collected for the manufacture of iodine. Many algae, especially the Kelps and the Fuci, are gathered as manure, and on the coasts of Ireland and France large quantities of calcareous algae known as ‘Maerl’ are col- lected for the same purpose. Cows are fed in winter-time in northern Norway on Rhodymenia palmata (Duke) if other fodder is scarce, and during the Great War experiments were made in France to feed horses on duly washed out Laminaria flexicaulis and L. saccharina. The horses that conquered their dis- like of this food throve very well and added to their weight, though doing their ordinary hard work. Unfortunately the number of horses that would take to this diet was small. In Iceland horses are fed on Kelp in winter-time. In the Malay Archipelago and in Japan various species of algae are consumed by the people, and species of Edible Seaiveeds 205 Eucheuma, Sphaerococcus, and Gelidium are gathered in enormous quantities for the manufacturing of the famous Agar-Agar, known in Japan as ‘Kanten’, which word signifies ‘cold weather’ and implies that Kanten can never be prepared in a hot climate, but only when the tempera- ture is rather low. The Japanese eat also a dish ‘Amanori’ made of Porphyra; ‘Kombu’, made of various Laminaria, Fig 73 Press for dr) mg Algjf is an ordinary food of the people. The commercial value of algae is rather high in Japan. On the west coast of America acetone, for the fabrica- tion of cordite and black powder, was extracted during the Great War from the Giant-Kelp growing along the coast. But it is doubtful whether this manufacture will be kept up now the demand for cordite has lessened. For manure great quantities of Kelp are gathered. The real importance of algae in oceanography lies in the fact that they form the basis of the ultimate food supply of the neritic or coastal fauna. V THE FLOATING ANIMALS BY G. H. FOWLER AND E. T. BROWNE To those animals and plants which float or drift in the water, either shallow or deep, ‘dead or living’, the term Plankton was originally applied by Hensen. Later on Haeckel applied it so as to include all pelagic animal and plant life as a whole, in contrast to bottom life, either fixed or crawling, as a whole, which he termed Benthos. Haeckel’s definition is now sometimes adopted, and it takes in all the powerful swimmers, such as fish, but in this chapter only the floating and drifting animals are dealt with. Plankton is also subdivided according to the size of the organisms into two groups ; the microplankton consisting of very minute organisms which require the use of a microscope for their determination, and the macroplankton for the larger organisms, but the division is rather an arbitrary one. None but zoologists appreciate the wealth of animal life in what appears to be clear, barren sea-water, and this is only revealed when a suitable fine meshed net has been hauled for a few minutes through the water, or when one realizes that the Greenland whale, 50 ft. long, lives solely on the smaller plankton. Most groups of the invertebrate or backboneless animals are represented in plankton, and the main types, which are likely to be captured near the surface, will now be briefly sketched. The figures, which in most cases have been a good deal simplified from the originals, and the text give only suf- ficient detail to help the beginner in finding out the group to which a specimen belongs. The exact position in the Protozoa 207 Animal Kingdom of each form named will be found in the classification (p. 475). In order to find the real names of the animals, or to learn something of their structure and life-history, the reader must consult the ordinary text- books, monographs, and special memoirs cited. Animals are divided into two great sub-kingdoms, ac- cording as they are composed of a single cell (Protozoa), or of many cells (Metazoa). Of the single-celled animals, several groups occur in the plankton, and are distinguished for their great beauty and interest. The pelagic Foraminifera form a shell of lime and are abundant in tropical and subtropical regions, where their dead shells form a deposit on the floor of the oceans known as the Globigerina ooze (see p. 261), They also occur in the temperate regions, but decrease in number as the water becomes colder. The shell, into which the animal can be completely withdrawn, consists of successive chambers, generally more or less spirally arranged, opening by a mouth from the largest chamber, and also perforated all over by fine holes. From the chambers radiate extraordinarily delicate spines of lime, which not only help to prevent the animals from sinking by increasine the frictional resistance to the water, but also allow the living matter (protoplasm) flowing out of the shell-mouth and fine holes to form a sort of bubbly lather between them, and to put out long threads for the capture of prey. The commonest genus is Globigerina (Fig- 74)- Another group is the Radiolaria, which usually have shells or skeletons of glass-like silica (Figs. 75, 76, 77, 78), which make most beautiful microscopic objects, or of a horny substance called acanthin, which is a complex sili- cate (Fig. 79), or of some other mineral (Strontium), hut never, like the Foraminifera, of lime. These hard frames. 2o8 The Floating Animals which are covered by cytoplasm, have developed in a complexity of forms and variety of detail that are found 1 ' IG 74 Globigerina (from a Living Specimen) (phaUenger) The central dark mass is the shell (compare Kig. 156, p. 261), from which radiate stiff spines of lime. Round the shell is a bubbly mass of living matter) from which fine granular threads extend between the spineS) generally straight) but some curved. in no other group of animals. Some have the skeleton in the form of two valves (Fig. 8o), free or hinged together, and others have a kind of projecting fringe to the shell 210 The Floating Animals (Fig. 82). A few, however, are without a skeleton, but may have spicules, others form small colonies (Collozoum), having the individuals embedded in a mass of jelly. Radiolaria occur in vast quantities in the tropical and warm regions, to a lesser extent in the temperate zones, but some prefer cold water (Fig. 84), and others only live at great depths (Fig. 81). Like the Foraminifera, their skeletons form an ooze at the bottom of the deep oceans. In most Radiolaria peculiar nucleated yellow corpuscles are found in abundance, and they are regarded as sym- biotic algae. The Ciliata have no skeleton or shell and swim by means of cilia. Many genera occur in the sea, but they require special searching for, and are seldom abundant. The Flagellata move about by the movement of one or more flagella, and some form colonies. In the group some naturalists place Ceratium (p. 156, Fig. 35) and Peri- dinium (p. 157, Fig. 36), whilst others consider them to be members of the Vegetable Kingdom. There is, how- ever, one conspicuous flagellate, Noctiluca (Fig. 83), which is enormous in size for this group. It is also brilliantly phosphorescent, and when occurring in vast shoals lights up the surface of the sea at night upon its being disturbed. In the course of evolution the Protozoa gave rise to the Metazoa, the many-celled animals, and these at the outset divided into two branches. One produced the Porifera or Sponges, and then remained stationary; the other gave rise to the Coelentera, and through them all the other groups of the Animal Kingdom. The Coelentera include the sea anemones, corals, and various groups of animals usually called ‘jelly-fish’, which are among the commonest and the most beautiful organisms in the plankton. Any one not familiar with the classification of marine Jelly-Jish 2 1 1 animals would call any animal a ‘jelly-fish’ which has the consistency and appearance of a jelly. But it is necessary to limit the name, even in its popular sense, so as to Fig bo Coelodendrum Fig 8i Tuscarora flG 82 Euphysctta Fic. 83. Noctiluca o Fig 84 Two Speuts of Challengeron exclude the swimming Salps (p. 228), which are most likely to be mistaken by the uninitiated for true jelly-fish. The former are quite different in shape and structure, and are nearly related to the Vertebrates (cf. Figs. 149 and 150 with Figs. 85 and 90). The jelly-fishes may be separated for the sake of 212 The Floating Animals convenience into four groups — (i) Hydromedusae, (2) Scyphomedusae, (3) Siphonophores, (4) Ctenophores — each of which has well-marked characters. The jelly- fishes belonging to the first two groups are commonly called simply Medusae. The Hydromedusae are small in size, seldom exceeding an inch in diameter, and are caught by means of a tow-net. They may be subdivided into two groups: (i) neritic or shore forms; (2) oceanic or blue-water forms. The neritic forms occasionally drift out into the ocean, but, as a rule, they become scarce beyond the loo-fathom line. The neritic Medusae (Figs. 85, 86) are budded off from small animals called ‘zoophytes’, or ‘hydroids’ (Figs. 87-9), which form colonies, and are permanently fixed at the bottom of the sea, or in some cases to floating objects. The hydroids, which thus bud off Medusae, do not develop eggs, and are, therefore, asexual. The little Medusae, after leaving their hydroids, begin an inde- pendent free-swimming career. In the course of growth and development they usually pass through a series of progressive stages, and finally become sexually adult. The eggs from the female Medusa do not develop into Medusae, but into hydroids, so that there is an alternation of generations. The geographical distribution of these littoral Medusae depends upon the habitat and distribution of their hydroids. Some have a very wide range; others are limited to definite localities. It is the bathymetrical limit of these hydroids, which seldom extend to a greater depth than the loo-fathom line, that confines this group of Medusae to the littoral waters. The oceanic Hydromedusae have not a fixed hydroid stage, and the eggs develop directly into Medusae. Their natural habitat is the ocean, but occasionally they drift ’Jelly-fish 213 into littoral waters. The species generally have a very wide geographical range, and some aie found at great depths. Fig 87 Colon) of Bougaim lUej attached to a Floating; Bit of Wood Fig SS Obelia ShotMn^ jell\-fibh being budded at the sides of a special poh p The large jelly-fishes, which are commonly seen from the deck of a ship slowly swimming at the surface of the sea, or found stranded on the shore by the receding tide, are representatives of the Scyphomedusae (Figs. 90, 91). 2T4 The Floating Animals A few have the power of stinging like a nettle, but most of them may be handled with impunity. One (Fig. 9;) is noted for its brilliant phosphorescence, and at night has the appearance of a globe of fire gliding through the water. Some of the Scyphomedusae have an alternation of generations. The eggs from an adult Medusa develop into tiny anemone-like animals, which are attached to stones, shells, &c., at the bottom of the sea. These fixed forms finally give off a number of little free-swimming rayed disks, called Ephyrae, which develop into adult Medusae. Others omit the fixed stage in the course of development, and the eggs develop directly into Ephyrae (Fig. 92). The Scyphomedusae are found all over the world; some are oceanic, and others are more at home near the shore. A few live deep down in the very cold zones of the oceans. The Siphonophores (Figs. 93-6) are closely related to the Hydromedusae, but most of them have not the appearance of a jelly-fish. Although some are fairly simple in their construction, others are extremely com- plicated. They usually have the appearance of a colony of animals composed of different kinds of individuals. This group contains some of the most beautiful, the most delicate, and the most fascinating animals that live in the sea. They are typically oceanic animals, and keep afloat by the aid of air-sacs or oil-globules. Nearly all swim, often at a good pace, by means of special organs called ‘swimming bells’. As a rule, they keep below the surface of the sea, and a few have been taken at a very great depth. They are found all over the world, even under the ice off the Antarctic continent, and the species have usually a very wide geographical range. 2 i 6 The Floating Animals Their life-history is very complicated, and in the course of development they pass through stages which are not at all like each other, either in structure or general appear- ance. To the Siphonophores belongs the well-known Portu- guese man-of-war (Physalia, Fig. 94). This is the largest Siphonophore, and it floats by the aid of an air-sac at the surface of the sea. Its stinging power is tremendous, pro- ducing a maddening pain which lasts for hours. Another Fig 93. Veleila common form is the little blue Veleila (Fig. 93), which also floats at the surface, and is not infrequently blown ashore. Adult sea-anemones (Actiniaria) are rarely found in plankton ; but larvae of anemones, which eventually affix themselves to or burrow in the bottom, are fairly common near shore, and one, at least, is found on the high seas north and west of Britain (Arachnactis, Fig. 97), The group known as Ctenophora, from their comb-like paddles, is common; Pleurobrachia (Fig. 99) may be taken off our own coasts; with its two graceful tentacles, it plays its prey exactly as an angler plays a salmon. Cestus, or Venus’s girdle (Fig. 98), is an inhabitant of 2i8 The Floating Animals warmerseas. Beroe (Fig. loi), a glove-finger of pink jelly, also occurs off our coasts. The various groups of flatworms are rarely met with in plankton, and consequently should be carefully preserved if found. But the group of roundworms (Chaetopoda), to which belong the common earthworm, and the ragworm and lugworm beloved of sea-anglers, require mention. Near shore swimming larvae, which ultimately grow into crawling or burrowing adult worms, are frequent at some seasons ; Fig. 1 02 shows some of the simpler forms. Many are provided with long bristles, such as give the name to this group (‘bristle-feet’). Of adult forms, perhaps the commonest is Tomopteris (Fig. 100), sometimes 2 in. long. Of the Nereids or ragworms, some are occasionally captured near shore, apparently at the breeding season. Of the curious Syllid worms (Autolytus), some reproduce asexually by forming a second or even several heads in the course of the body, and then dividing into as many worms as there are heads. Of the wheel-animalcules (Rotifera), so common in Fig 102 Types of Chaetopod Larvae From left to right Mitiaria, Nenne, Polygordius, Chaetopterus Fig 103 Spadella Fig 104 Sagitta 220 The Floating Animals fresh water, some seventy species are stated to be marine; little is known of them. The little glass-arrows, or Chaetognatha, are of con- stant occurrence, apparently in all seas ; they vary in length from \ in. to 4 in. (Figs. 103, 104). The group to which crabs, lobsters, and so forth belong — the Crustacea — are probably more plentiful than any other in plankton. Of the ‘water-fleas’ (Cladocera), minute animals with two shells, a large pair of rowing antennae, and one large eye, some are not infrequently taken (Evadne, Fig. 105). Another subdivision of Crus- tacea, also w'ith two shells, is termed the Ostracoda, one family of which is sometimes very abundant (Conchoecia, Figs. 106, 107). Probably the most widely spread and proliflc group of the whole Animal Kingdom, so far as plankton is concerned, is formed by the ‘oar-footed’ Copepoda (Figs. 108-1 1 1), which are only very rarely, if ever,absent from a haul ; generally minute, they are active swimmers ; they have been met with in such abundance as to give the sea a reddish tinge. When magnifled, many show brilliant colourings, due either to iridescence (Sapphirina,Fig. 1 1 1) or to the presence of drops of coloured fat. Many of the ‘fish-lice’ belong to this group; some are only temporary parasites, and can swim (Fig. 108); others embed them- selves in the flesh of fish and other creatures, and become so altered, in adaptation to a parasitic existence, that only a study of their development shows them to be Copepods, or, indeed, Crustaceans at all. The Barnacles, or Acorn- shells (Cirrhipedia), are not properly planktonic, but may be found attached to floating weed, driftwood, &c. These have free-swimming larval stages, enclosed in two shells, like an Ostragod. The subdivision of Crustacea, which includes the ‘sand- Crustacea 221 hopper’ and the so-called ‘fresh-water shrimp’ (the Am- phipoda), although far more plentiful on the bottom, has representatives (Fig. 1 12) in the plankton. Some of them Fig. 105 Evadne Fig. 106. Conchoccia Fig. 107. Conchoecia (Left Shell and Antenna) (Shell removed) Fig. 108. Caligus Fig. 109. Calanus Fig. I to. Temora Fig. III. Fig. 112. Euthemisto Sapphirina haunt the bells of jelly-fish, presumably for protection, while others (Phronima, Fig. 1 1 3) establish themselves in the hollow skin of Doliolum (see p. 229). Very closely allied to these is the subdivision which includes the com- mon wood-lice or pill-bugs; these are not insects, as is 222 The Floating Animals generally thought, but belong to the subdivision Isopoda; they may generally be distinguished from the previous subdivision by being flattened from above downwards, while the former are flattened from side to side. Few of them occur in plankton on the high seas, except as para- sites, either temporary or permanent (Fig. 1 1 4). The ‘opossum-shrimps’ (Schizopoda), so called from the brood-pouch present in some species, are frequently taken. Several have special luminous organs, provided with lens and reflector, in different parts of their bodies (Fig. 1 1 6). The subdivision, which includes the lobster and* shrimp, is termed Macrura — the ‘long-tails’; few of these are planktonic when adult, but some prawn-like forms are widely found. Mention must be made, how- ever, of the curious little Sergestids (Fig. 115), not only because they are often captured, but also because of their curious larvae (Fig. 1 1 7). Less adapted to a swimming life are the Brachyura, or ‘short-tails’, of which the com- mon crab is a member; some can swim a little, and one at least has been found in quantity about 200 miles from shore over very deep water. Crabs are also found attached to gulf-weed and other floating objects. The larvae of Crustacea are often even more numerous than the adults. Most marine Crustacea hatch out of the egg in the form called a Nauplius, which may be known by a shield-shaped body, a single eye, and three pairs of limbs (Figs. 1 19, 120). This larval stage moults or sheds its skin repeatedly, not only growing, but also changing its shape somewhat at each moult, until it attains the parent form. Many of the higher Crustacea, after passing through the Nauplius stage, gradually attain a Zoea sta^e, which may be recognized by the limbs of the forepart of the body having sprouted, by a pair of large eyes having Fig 113 Phronima Flc 114 Fic 115 Lucifer Cirolma Frc 1 16 Njctiphanes Fig 1 17 Sergestes Larva The Fig 118 Phyllosoma right side of this figure is of the dorsal or upper surface, on the left side IS draw n the ventral or under surface to show the limbs 224 'The Floating Animals appeared, and by the tail part of the body having grown out ; the latter carries no limbs, except sometimes the last pair (Figs. 12 1-3). A few families have special larvae, which are often com- mon near shore after the breeding season. Thus the crawfish, or spiny lobster (Palinurus), passes through the remark- able stage Phyllosoma (Fig. 118); the crabs, after their Zoea stage, assume the form known as Megalopa before tucking in their tails (Fig. 124); and the Stomatopoda have a set of larvae peculiar to themselves (Figs. 125-6). Of the group to which oysters, snails, and cuttle- fish belong, the Mollusca, some subdivisions occur in plankton. The oyster-forms, or Lamellibranchs, possessing two shells, are occasionally represented by larvae (Fig. 127). The snail-forms, or Gastropoda, are frequently met. Jan- thina, a warm-water snail, sometimes thrown on our western shores, has a purple shell; to its foot is attached a raft, on the underside of which the eggs are embedded. Atlanta (Fig. 128) has a spiral shell, with a narrow keel ; Carinaria, like a transparent slug, carries a shell on its hump (Fig. 129); very like it, but without shell or hump, are several other warm-water forms. The pteropod, or ‘wing-footed’ forms, are found in all seas; of these the majority have shells (Figs. 132, 133). Near-shore larvae of Gastropods are often plentiful ; the most characteristic is the ‘Veliger’ stage (Fig. 130). The cuttle-fishes, of which the octopus and squid are too well known to require a figure, are all more or less planktonic, although many, especially in shallow water, are bottom feeders; they sometimes attain a huge size. Some have a web between the arms, which thus form a great funnel. Any forms with an external shell (except the familiar Pearly Nautilus and Argonauta) should be Fig 1 1 9- Naupljus Larva of a Barnacle Fig. lao. Nauplius Larva of Penaeus Fig. 131 . Zoca Larva of Penaeus Fig. 124. Megalopa Larva of a Crab Fig 125. Early Larva Fig. 126. Late Larva of a Squilla of a Squilla 226 The floating A?iimals carefully preserved for study by experts, such as the body of the little Spirula, the dead shells of which are not uncommon in eastern seas. Of the group of Echinoderms — the starfish, brittle- stars, sea-urchins, trepangs, sea-lilies — a single adult genus only has as yet been taken in plankton, but their larvae are frequently captured ; these are even more unlike their adult form than in the Crustacea. The adult form in Crustacea was attained by a series of moults, but in the Echinoderms it rather grows out of the larvae, which, so to speak, shrivels down on to it. Each of the five sub- divisions has its own type of larvae, four of which are here shown (Figs. 13 5-42). Of the enormous group of Insecta, only one small family is truly planktonic (Fig. 134). Of the Polyzoa (they have no English name), none are found in plankton, except accidentally attached to drift- wood or weed. Near shore their larvae are common; of these, the most likely to strike the eye has a shell like a flattened funnel (Fig. 131). The foregoing groups are all taken from the inverte- brate animals, but there remain some creatures, very com- mon in plankton, which only a naturalist would dream of classing anywhere near vertebrates. Yet these animals, both by structure and development, are undoubtedly akin to vertebrates; they have (among other characteristics), either when larval or throughout life, a supporting rod in the back (the notochord), exactly like that of a sturgeon or lamprey, or the similar rod round which the vertebrae are moulded in the embryos of all vertebrates, including Man. The names of these groups refer to this notochord. The ‘sea-squirts’ and salps are included in the group Urochordata — i.e. with the notochord in the tail. Some of them retain it throughout life ; for instance. Append!- Fig. 127. Oyster Larvd Fig. 128. Atlanta Fig. 129. Carinaria Fig. 130. Veli- ger Larvae: above, Dorsal; below, Ventral Fig. 131. Larva of a Polyzoan Fig. 132. Clione Fig. 133. Shelled Pteropods: above, Hyalea; below, Styliola Fig. 139. Fig. 140. Fig. 141. Fig. 142 Figs. 135. Auricularia Larva of a Holotburian. 136. The same, later, the young Trepang showing. 137. Pluteus Larva of an Echinoid. 138. The same, later, the young Sea-Urchin prominent. 139. Bipinnaria I.arva of an Asteroid. 140. The same, later, carrying the young Starfish. 141. Pluteus Larva of an Ophiuroid. 142. The same, later, with the young Brittle-Star. 22 8 The Floating Animals cularia (Fig. 143), which can often be captured in this country in an estuary on a flowing tide; some species form temporary ‘houses’ of jeliy-like material. Others drop the tail and notochord on assuming the adult form; of these Pyrosoma (Fig. 144), brilliantly luminous, as its name ‘fire-body’ implies, is a colony of animals embedded in a common cylinder of jelly, mouths outwards. A solitary form, Doliolum, is remarkable for its complex life-history; the form, A (Fig. 145), itself developed from a fertilized egg, gives rise to little buds, which wander up on to the projecting horn behind their parent, and arrange them- selves in three rows. Those of the side rows, B, grow into a spoon-shape, and act as food and water pumps for the rest (Fig. 146) ; large buds of the middle row, C, presently break off (Fig. 147), carrying on their broken stumps tiny buds, which also belong to the middle row. These grow up into the sexual form, D (Fig. 148), and break away in their turn; their eggs begin the cycle anew. Even more complex, and not yet fully understood, is the life-cycle in two similar forms. Specimens of all such from tropical seas should be carefully preserved for study by competent hands. Another subdivision is represented by Salpa. In this, an animal which has developed from the egg, buds off a chain of little Salps; first the chain, then the separate individuals are detached, become sexual, and produce eggs, and the life-history begins again (Figs. 149, 150). A second group is formed by the Hemichordata, in which the notochord is only partly developed; none are planktonic, but one (Balanoglossus) has a planktonic larva, which is not uncommon. Lastly, in the group Cephalochordata (Amphioxus), in which the notochord extends to the tip of the head, the Fig. 148. Doliolum. Fig. 149. Salpa. Fig. 150. Salpa. The D. The sexual form. The sexual form. budding form with a chain of little salps hanging out Excluding the fish, which are the subject of a separate chapter, the animals above described may be taken to represent fairly completely types of the forms which are likely to be captured at or near the surface. ( 230 ) Methods oj Capture The next question is, how to catch them. The net, termed a tow-net, consists of the frame, the net, and the tin. The net about to be described is suitable for any one wishing to sample the contents of the sea, either from a rowing-boat or a small motor-boat. The frame is best made as follows; from an oil-shop procure a stout lo foot length of flexible cane, such as is used to poke down drains; a sharp bend in it may be taken out on soaking the bend in boiling water. When straight, twist this up into a circle of about 2 ft. in diameter and lash the ends in with strong cord or copper wire. The net is best made of Swiss silk (called bolting cloth, such as is used for sifting fine flour). This material is very expensive, and if cost has to be considered, a muslin net can be used, but it is not so lasting, neither are the meshes so regular or clear of fine fibres, and they are more liable to become clogged up. A muslin net is quite serviceable providing that it is well cleaned out after use; silk nets must be dried directly after use, or else they will very quickly rot. A useful all-round sort of net has 6o meshes in a linear inch. The small net with a glass tube at the end used by pond-life collectors is useless for sampling the sea. The organisms are not so concentrated as in small ponds. A much larger net has to be used and a convenient size for general purposes is about i \ ft. in diameter at the mouth, tapering gradually down to 4 in. at the bottom, where the tin is tied on, and about 5 ft. in length. Care should be taken to fold the material so that it makes, when cut out and sewn up, a perfect cone, if wrongly folded and sewn the net becomes lop-sided. At the top and bottom the net is sewn on to a stout band of broad tape or on a strip T om-netting 231 of calico; to the top band are sewn short pieces of tape to tie the net on to the cane frame ; through the bottom band should pass, in a plait, a running cord which will gather it securely over the tin. The ‘tin’ is best made of copper or zinc with a rounded flange round the mouth, below which the running cord grips, and large enough to hold about a pint of water. From the frame run three ‘bridles’ of stout line, about 3 ft. long and spliced together where they meet; to these is attached the tow-rope. For the latter a stout hemp line is best, if too thin it cuts the hands when wet. When tow-netting near shore from a small boat, a small net usually meets all requirements, it can be sent down to near the bottom by attaching a lead weight near the bridles and is easily hauled in by hand. For such work as is carried out at Marine Biological Stations and on Expedi- tions more elaborate methods of tow-netting are used, such as fastening several nets at different depths on a wire rope with a heavy weight at the end, or having large and special nets designed for research work. Such methods can only be carried out in a satisfactory manner on ships properly equipped with the necessary gear for lowering and hauling in heavy nets, and also having some one in charge who has had the required training, or the nets will soon be found round the propeller. The net should be towed very slowly, but both pace and mesh are decided by the kind of animals which one hopes to catch. With a net about i \ ft. in diameter one should be able to hold the tow-rope on one finger; a reasonable pull should be about three pounds against a spring balance. The secret of successful tow-netting is the pull on the net. A good catch depends much upon the slow filtering of the water through the meshes of the net. When the pressure of the water inside the net is much greater than the pressure outside it, the net becomes more 232 The floating Animals or less waterlogged and the organisms instead of entering the mouth of it shoot over its edge and thus escape cap- ture. The same result happens when a dirty net is used or the meshes have become choked with diatoms. The net ceases to filter properly and becomes waterlogged, then one might as well tow along an iron bucket. By tow- netting slowly with just enough pull to keep the net properly extended the animals are caught uninjured, for they pass slowly down inside the net into the tin. It sometimes happens that the pull is not sufficient to keep the net in a horizontal position or properly extended, it sags down owing to the w'eight of the tin. This can be easily remedied by tying some corks round the tin. A twenty minutes haul is generally long enough, after this the net should be hauled in very gradually, otherwise the pressure of the water against the meshes seriously damages the more delicate animals. As soon as it arrives on board, the contents of the tin should be emptied, with plenty of extra water, into a big glass bottle — a wide-mouth pickle-jar or sweet-jar is ex- cellent. The larger animals can be picked out with a lifter (an old tablespoon, the bowl bent at right angles to the shank, serves the turn). For those of smaller size, take a glass tube, about 8 in. long and | in. outside diameter, hold this between thumb and middle finger, stopping the upper end firmly with the first finger; lower it thus over the animal, remove the first finger, and animal and water will rush up into the tube. Replace the finger, and the tube with its contents can be lifted out if kept vertical. The animals can thus be transferred to glass dishes and examined, or preserved for future study (see Chapter XI). In temperate climates most planktonic animals will live for some hours in a glass bottle if absolutely clean and kept cool. In the tropics, where the bottle cannot be kept Pljitkton ::33 cool, thev die very soon, and should be preserved as soon as possible. For immediate examiiiation on board, as a ship is rarely still enough (except in harbour) to allo^v of the use of a microscope, a small ‘dissecting microscope’, with simple lenses magnifying about 6 and lo diameters, is absolutely invaluable. Naturally enough, the animals in plankton are not always and everywhere the same. In the first place the distance from shore makes a great diiference. Plankton from shallow water near land shows numerous larvae of bottom animals, and such jelly-fish as are budded from hydroids, but over deep water such larvae are hardh' ever present; there are larvae of planktonic adults, but they are not so numerous in proportion, and these adults are mostly different from those near shore. Hence we dis- tinguish the ‘neritic’, or coast-wise plankton, from the ‘oceanic’, or blue-water plankton; the loo-fathom line may be roughly taken as the boundary- between them, but they pass imperceptibly into one another. Secondly, the tem- perature of the water makes a great difference in the plank- ton ; it is fairly obvious that as a rule an animal which can stand arctic cold of 30° to 35° F. will not be happy in the tropics with a sea temperature of 80' to 90'' F. (although such cases do exist); but many marine animals seem to be influenced by quite small differences in temperature. Thirdly, if there be too little salt in the water — for example, in estuaries or the Baltic — many species will not live there. Lastly, the character of the plankton alters a good deal with the depth; it is not every animal which can thrive in absolute darkness, at a very low temperature, on such casual food as falls from the surface layers towards the bottom. This last statement opens up a fresh branch of the sub- ject. Plankton is by no means confined to the upper 234 The Floating Animals layers of water, but has been captured even at the greatest depth to which suitable nets have been lowered (2,734 to 2,187 fathoms). Such nets, termed ‘closing nets’, are lowered to a known depth, with the mouth closed; they are then opened by mechanical means, towed either ver- tically or horizontally for a known distance, closed again, and brought closed to the surface; they therefore only contain animals from the desired depth. This mid-water work has been systematically attempted only of late years, and forms a fascinating branch of study. Very little is known at present about the plankton of great depths, and any yacht with a steam winch can do most valuable work in this direction. A description of the necessary outfit is rather beyond the scope of these pages, but details can easily be obtained from the Challenger Society or the Director of the Marine Biological Association’s Labora- tory at Plymouth. Many of the animals from great depths seem to be confined to deep water, but some are known also from the surface. Of one branch of plankton we know almost nothing — that is, of those invertebrates which, although capable of swimming a little, are closely dependent on the bottom of the deep sea either for food or for resting. For their capture tow-nets are attached to the trawl-heads or dredge, streaming a little behind them so as to catch the animals disturbed by raking up the bottom. The plankton, then, is not always the same everywhere, but the reasons why it is different in different places are only partly understood; it seems, however, certain that the temperature has more to do with this than anything else. Some species seem able to stand much greater ex- tremes of temperature than others; but for every species there is a (maximum) temperature above which it cannot live, a (minimum) temperature below which it cannot live. Plankton and Temperature 235 and an (optimum) temperature at which it thrives best. To take a well-known case, characteristic animals of the Gulf Stream and North Atlantic Drift rarely reach as far east as our islands, although the actual water in which they lived travels as far as the North Cape; they die out as the temperature falls. The slaughter is particularly heavy where hot and cold currents meet. Supposing that we knew precisely the maximum and minimum temperatures of a species — say 60° and 45° F. for a north Atlantic species — then by drawing lines on the map through all the points where 60° and 45° F. occurred at the surface, we could run a surface boundary round the animal, and map its ‘horizontal distribution’. Further, if this species can tolerate so low a temperature as 45° F. at the surface, there seems to be no reason why it should not live at 45° F. at any point below the surface; consequently, in addition to the horizontal distribution, we want to be able to map also the distributions at different depths, and thus the space occupied by the species will form an irregular figure in three dimensions, with length, breadth, and thickness. Unfortunately, in the present state of knowledge, we are unable to draw this figure with certainty for any single species; hence it will be readily understood how important it is always to record the tem- perature of the water from which a haul is made, and, if practicable, also the salinity. One interesting result of the ability of many animals to live below the surface, provided that they keep above their minimum temperature, is that some arctic species are con- tinuously distributed from pole to pole in the deep cold water of mid-ocean; but have never been captured any- where near the surface in the tropics, where the tempera- ture is above their maximum. A comparison of the organisms taken in nets at the 236 The Floating Animals same time and place, but at different depths, has favoured the view that some come to the surface at night and sink to some depth during daylight as a protection against too bright light. It is only recently that series of obseiwations have been made to investigate thoroughly this interesting point. They have been carried out by the staff of the Marine Laboratory at Plymouth, with a special large net, about six feet in diameter and a reliable depth self-record- ing instrument used. The results clearly show that certain animals do come to the surface at night and sink to a definite depth during daylight. Some living close to the bottom ascend at night towards midwater and become more uniformily distributed, whereas others are quite indifferent to the intensity of the light. This vertical dis- tribution is found in nearly all the animal groups, but only certain species are affected. These show an ordered vertical distribution during a full summer’s day and dif- ferent species segregate at different depths, each having its own maximum, minimum, and optimum with regard to light intensity, showing that each prefers a definite light intensity where most specimens will be found (op- timum), and a decrease in number towards the light in- tensity, above which none are usually found, and the same in respect to the minimum light intensity. How far the downward movement goes is at present unknown, but probably not much below 50 fathoms. There is also the possibility of its being checked by the fall in the tempera- ture of the water, to which most organisms are sensitive. There is also a seasonal change in the depth due to the intensity of sunlight in the different months of the year, the optimum depth would thus be nearer the surface in winter than in summer. The depth therefore depends upon the intensity of the light prevailing at the time, varying day by day and month by month. Vertical Distribution 237 This vertical distribution of the plankton is an impor- tant point to keep in mind when tow-netting in the open sea and where the water is deep. The specimens wanted may not be near the surface, but say at 20 fathoms, so it is advisable when using only one net either to sink it deep and haul it in by stages to the surface, or to take several hauls at different depths. Where currents run strong the vertical distribution does not occur, as here there is too much of a general stir up taking place. Hauls may be readily compared if standardized — that is to say, if they are made with nets of similar mouth, length, and mesh, for the same time, and at the same speed of towing. In this case all that is necessary is to preserve the whole catch, and to count the specimens of any species under study; thus it can be readily shown in accurate terms that a species is rare at the one hour, depth, date, or place, and common at the other. Every haul made for comparative work (and that is pre-eminently the work now required) should be made with standard nets, time, and speed. The only thing difficult of control is the speed, except in a small launch, but this can be read by a current meter. Standards are suggested on p. 240. As a rule, the greater number of the species captured at or near the surface at any particular position at sea will be found to extend to about 100 fathoms at that position (not necessarily elsewhere, where the temperature and other conditions may be different). For a proper study of the alleged up-and-down movements of the plankton, and for many similar objects, it is desirable to know, not only whether it is rarer or more abundant at the surface under certain circumstances, but also where it goes to when not at the surface. The method of hauls with open nets at various depths will throw a good deal of light on this, but these nets are not to be trusted absolutely, because they catch 238 The Floating Animals specimens from higher levels while being hauled to the surface. For accurate work a closing net should be used horizontally, lighter and smaller than those used for true deep mid-water work; they should carry the same standard net, and be used for standard time and at standard speed. Other points requiring study by similar methods are the supposed movements of plankton due to rain, breaking seas, wind, fresh water from large rivers, floating ice, and so forth. A curious feature well worth attention is the way in which plankton becomes concentrated into swarms or streaks; it is by no means universally distributed, even over a small district. Such streaks should be surveyed, if possible for dimensions both horizontally and vertically. Particular attention should be paid to the comparative abundance of plant food inside and outside the area, as well as to their position as regards current and tide, for it has been suggested that they are formed exactly as a river throws raffle on its banks. All special adaptations of a planktonic organism to its conditions of life should be noted. A striking series of these adaptations is naturally directed towards preventing weak swimmers from sinking. This object is attempted in various ways, of which the chief are — either {a) a diminution of weight by the formation inside the animal of light fluids (in the bubbles of Foraminifera and Radio- laria), fats (Copepoda), jelly (Jelly-fish and Salps) or gas (Siphonophores) ; or {V) an increase of surface in order to increase the friction in falling through the water, pro- duced either by flattening the body (Phyllosoma, Sap- phirina) or the growth of long spines (Foraminifera, Crustacean larvae). Coloured sketches from life are always valuable away from home waters. Most of the larger invertebrate plankton Natural History oj Plankton 239 is nearly colourless and transparent, and any exceptions to this should be noted, and, if possible, sketched. Round our own coasts there is much to be learnt of the daily and seasonal movements of the larger plankton, and of its behaviour under changing conditions of its sur- roundings. Once clear of European coasts, a traveller may feel certain that his collection of plankton will be serviceable to science if accompanied by proper observa- tions and full notes. What the study of plankton requires nowadays is not the mere collection of specimens, but the collection of facts. Naturalists have described thousands of planktonic animals from all sorts of places, but know very little about how they came there, why they are there and not else- where, how they live, what they eat, and what eats them. Without such knowledge the study of plankton is about as interesting as to read a street directory containing solely names and numbers. Consequently every item of what is called ‘natural history’ should be observed and noted, and every possible detail recorded whenever a haul is made. The routine details are set out in the sample General Log (p. 462), subject, of course, to modification for special cases. Except when an observer is simply collecting plankton for his own instruction (if even then), no haul should be taken without a definite object; it should have a story to tell of the plankton of a particular place, hour, depth, or season, of the effect of rain, breaking sea, ebb and flow of tide, and so forth. For the study of all general questions, the hauls should always be standardized as suggested on pp. 237 and 240. It will be seen readily from this brief sketch that plenty of interesting work awaits any one who has opportunity and a tow-net. In harbour, from a pier or reef of rocks if the tide serves, from a small boat or a big ship, the tow- 240 The Floating Animals net can be streamed with the certainty of an interesting result; even if the catch be scanty, the result becomes of value if the reason for its scantiness be found. No observa- tion, however apparently insignificant, is scientifically worthless if it be securely based on fact. On however small a scale, valuable work may be done for science with- out attacking the more intricate problems. The simple tow-net haul in foreign parts, if accompanied by the neces- sary observations, will always be worth making, preserv- ing, and submitting to experts for report. Finally, a word of advice to the beginner who wishes to collect for his own instruction, and to do serious work; he will be at first bewildered with the wealth of life at his disposal, and depressed with the difficulty of finding what is already known, what not. Let him therefore confine himself strictly to the animals of one, or at most two, groups, and hand the rest over to an expert at home; let him select for his own study the group or work to which he feels most attracted and interested in. Then work is no longer a toil but a real pleasure. Note. — The International Council for the Exploration of the Sea have recently' recommended the adoption of standard nets for vertical and horizontal hauls. These nets are specially adapted for closing, but it will be sufficient for the purposes of the general collector if they are used as non-closing nets. The dimensions laid down are : i . 'Set, length 2 m., diameter at mouth 50 cm., diameter at tin 10 cm., breadth of calico at top 3-4 cm., at bottom 8 cm.j 2. Frame, a ring of 5 ^8 in. iron, 50 cm. in diameter; 3. Material, Bolting Silk No. 25, 77 strands to 10 mm., for the fine-meshed net; and No. 3, 23 strands to 1 o mm., for the coarse-meshed net.' ■ Ostenfeld, C. H., and Jespersen, P., 'Standard Net for Plankton Collections’. Conseil Permanent International pour I’Exploration de la Mer. Publications de Circonstance, No. S4, 1924. > The numbers given here are those used by Albert Wydler, the manufacturer of the silk at Ziirich. This material can also be obtained from Messrs. Stanniar of Manchester, and corresponds to their numbers zoo and 58. THE SEA FLOOR BY TUB LATE SIR JOHN MURRAY (1912) Methods The oldest method of sounding was by means of the hand-lead, usually 12 to 14 lb. in weight, ‘armed’ with lard or tallow, to which a sample of the bottom adheres, and with a line marked in fathoms and fractions of a fathom. Practically all the coasts of the world have been surveyed by instruments of this kind used from rowing- boats. When attempts were made to sound in 100 or 200 fathoms, heavier weights and more carefully prepared hempen lines were employed, and the leads were provided with cups, valves, or snappers, to bring up samples of the sands, gravels, and muds. Sir James Clark Ross, during his Antarctic cruise (1839 to 1843), made most praise- worthy attempts to sound the greater depths of the ocean with ordinary sounding lines and heavy weights from small boats. He succeeded in recording depths down to 3,600 fathoms, but, although the time each 100 fathoms left the reel was noted in the usual way, he was uncertain when the weights reached the bottom, and the results were not altogether trustworthy. About the year 1854 Lieutenant Brooke, of the United States Nav)', intro- duced the method of detaching a heavy weight from the sounding-tube on its striking the bottom; henceforth deep soundings were recorded much more frequently and with greater accuracy, while good samples of the deposit were obtained. During the Chullenger Expedition (1872 to 1876) sounding with wire had not passed the experimental 242 The Sea Floor stage; all that was known for certain was that the same wire could not be used for many soundings without breaking. The Challenger made use of a specially made hempen sounding line. It had to carry on each occasion a load of valuable instruments, and in this it was brilliantly successful. Wire rope is suitable for sounding pure and simple; indeed, the systematic investigation of the form of the bottom of the ocean can only be rapidly carried out with wire. For deep sea researches, where many valuable thermometers and other instruments must be sent to the bottom in deep water, hemp rope is much more trust- worthy than wire. However, during the recent Michael Sars Expedition in the north Atlantic a six-strand wire 3 or 4 mm. in circumference was used, with excellent results, to carry several thermometers and water-bottles down to depths of over 1,000 m. In the ordinary routine on board the Challenger the soundings were made with the Baillie machine (see Fig. 15 1), the sinker weighing 336 lb., the tube 25 lb., and the attached water-bottle 20 lb. The effective sinking-weight was 333 lb. in water; every 100 fathoms of line used in sounding added 6 lb. to this effective sinking-weight. In very shallow water there was acceleration in the rate of descent, but in deep water there was continual and pro- gressive retardation. With wire there is continuous ac- celeration if allowed to run free like the hemp line. In recent years soundings have been made by various types of wire sounding machines, such as the Le Blanc, the Sigsbee, and the Lucas. The Lucas machine (Fig. 1 97, p. 321) is the one most frequently in use, and is extremely satisfactory. It seems unnecessary to describe these various instruments in detail ; their great advantage is that they indicate at once when the weight strikes the bottom. In well-appointed cable ship^ soundings in 2,000 Fig 151 Baillie Sounding Machine as used on board H M S Challenger A, in elevation, B, m section The heavy weights, e, drive the central tube, a, into the bottom deposit, and "^he ooze is retained in the tube by the butterfly vahe, / As soon as the suspension is relaxed, the shoulder carrying the wire which holds up the heavy weights becomes sheathed, the wire and weights being left at the bottom 244 Floor fathoms can be made in an hour from start to finish, but in order to do this the wire must be hove in as fast as possible, and breakages of the wire frequently occur. One cannot, therefore, work at this rate with any regard to the safety of instruments attached to the wire. With hempen line the maximum rate of working can be observed, even when the line carries instruments. On cable ships there is usually plenty of old or damaged wire, and where the depth only is required, this old wire is sometimes used and cut adrift when the bottom has been reached in order to save the time of heaving it in.* A great many arrangements have been used with the sounding apparatus to bring up samples of the deposits forming on the bottom in deep water. The ‘arming’ of the hand-lead with lard and tallow has already been men- tioned; snappers, cups, and tubes with valves are also used for this purpose. Experience alone can show the best appliance to adopt in special circumstances. On board the Challenger, when sounding over Red Clay areas, the Baillie sounding tube, 2 ^ in. in diameter, without butterfly valves, has been forced from 1 8 in. to 2 ft. into the deposit, and has brought up a section of that depth, the total quantity of clay being more than enough to fill a quart bottle. So far as obtaining a sample of the bottom is concerned when sounding, this is the most successful record. Sometimes, however, the whole of the mud would slip out of the tube while being hauled up. To prevent this, valves and balls at the upper end of the tube were ' Recently a method of echo-sounding has been introduced, whereby sound waves are sent to the sea bottom, from which they are reflected back to the surface. The speed of sound through water is approximately 4,900 ft. per second, varying slightly with salinity and temperature. By this means soundings may be taken very rapidly compared with the old methods, and a ship can steam at full speed sounding as she goes. The German research cruiser Meteor has recently undertaken a very large number of such soundings in the South Atlantic. — Ed. Obtaining Samples 245 used with varying success. In positions where a topical Globigerina Ooze w'as expected, a butterfly valve was generally inserted in the lower end of the tube, and this arrangement w^as generally successful in retaining some of the deposit. To place three or four small tubes side bv side, without valves, has been found a good arrangement. Gravel and small stones are the most difficult of all samples to recover, and for these the snapper (Fig. 196, p. 318) Fig. 15Z. Set of Sietes for grading the contents of Dredge or Trawl, or material gathered on shore For use the whole set is immersed m a tub of sea-water and the ‘Rendle tube’ are the most successful. During the Michael Sars Expedition (1910) a heavy contrivance of Nansen’s was successfully used to bring up a section of the deposit, but this entailed a stranded wire and a separate sounding operation. In the dredge and trawl large quantities of the deposits were sometimes brought to the surface, and a careful examination of these threw a great deal of light on the samples procured by the sounding machine. Large samples of mud, ooze, or clay, were carefully passed through wire sieves (Fig. 1 52) with different sized meshes, and these coarse and fine ‘washings’ were carefully 246 The Sea Floor preserved for future examination. It has often been re- gretted that larger samples procured in the dredge were not preserved and brought home without being passed through sieves. The samples of deep-sea deposits which are procured are best preserved by being at once placed in wide- mouthed glass-stoppered bottles, and a small quantity of methylated spirit can be with advantage added, although this is not essential. Cork-stoppered bottles and tubes can also be used. Spirit, added to the samples which contain sea-water, will throw down some of the sulphate of lime, in a gelatinous-like condition (Bathybius), but this is not a very serious objection. When a roll of the deposit is procured from the sounding tube, it should be surrounded with blotting-paper, or pushed out on to blotting-paper; the whole can then be enveloped by the paper, dried, and retained for future detailed examination. The colour of the sample and any other peculiarities should be carefully noted in the wet condition. A label should be put on the outside of the bottle or tube (not on the cork), as well as inside. The Depth of the Ocean The ocean has now been sounded in nearly all direc- tions, and our views as to the relief of the sea floor are slowly becoming more and more definite. It is often asserted that we know very little concerning the depth of the ocean, so few are the soundings compared with its vast extent. It is possible to underrate the extent of our know- ledge in this respect, for if the extent of the ocean be great, it should be remembered that the uniformity of the conditions is also great over wide areas. Notwithstanding the great number of deep-sea soundings recorded since 1895, when the depth-hemispheres in the final volume of the Challenger reports were published, it is somewhat T)eep Soumlings 247 remarkable that the contours shown on these have been comparatively little altered in their broad general outlines, although many deeps and submarine elevations have been discovered in recent years, and in some regions the con- tour-lines now represent a much more complicated appearance. These depth-hemispheres have been kept up to date bv Sir John Murray and Mr. J. Cl. Bartholomew. An inspec- tion of the reduced chart given here will show at a glance where deep soundings arc most abundant. The greatest depth yet recorded in the ocean is 5,269 fathoms (9,636 m.), or 66 ft. less than six English miles.' This was a sounding taken by the U.S.S. A'ero in the Challenger Deep to the south of the Ladrones, where the Challenger recorded a depth of 4,575 fathoms. Depths greater than 5,000 fathoms are also known in the Aldrich Deep in the south Pacific. There are altogether ten deeps in which depths exceeding 4,000 fathoms have been recorded, two of these being in the Atlantic, and the remainder in the Pacific. The term ‘deep’ is limited to those parts of the ocean in which soundings of 3,000 fathoms or deeper have been taken. On the depth hemi- spheres above referred to, fifty-seven deeps are now shown — thirty-two in the Pacific, eighteen in the Atlantic, five in the Indian Ocean, and two in the Southern Ocean south of latitude 40° S. From the data at present available the depths over the total floor of the ocean are now estimated as follows : Between shore line and 100 fathom line 7 per cent. „ 100 fathom line and 1,000 fathom line 9 ,, ,, 1,000 ,, ,, 2,000 „ ,, 19 55 ,, 2,000 „ ,, )> 5^ )> Over 3,000 fathoms ..... 7 „ * The grcjtcst depth now rerorded is in the Pacific off Mindanao, in the Philippine Islands, and is 5,348 fathoms. — Ed., Oct. igsS. 248 The Sea Floor It will he seen from the above that the extent of the sea-bottom lying between the shore line and a depth of 100 fathoms (600 ft.) is very nearly equal to the area lying between 100 fathoms (600 ft.) and 1,000 fathoms (6,000 ft.). The first area is known as the continental shelf, and is regarded as belonging to the continental areas. Many portions of it were evidently dry land at no very remote geological period. Deep ravines frequently cross the con- tinental shelf, and are believed to be submerged river valleys, which in some cases can be associated with still existing rivers — for instance, the Rhine and the Congo. The materials which cover the continental shelf are to a large extent under the influence of waves, tides, and cur- rents, and are constantly subjected to transport by these agencies. It is believed that waves or currents do not transport other than the very finest particles at depths greater than 100 to 1 50 fathoms (600 to 900 ft.). In cup- shaped depressions, or sheltered bays within the 100- fathoms line, fine mud may be deposited in relatively shallow depths, but on coasts facing the great ocean basins muddy deposits commence to form at an average depth of 100 fathoms. Beyond that depth the whole floor of the ocean may be regarded as an area of deposition. We have as yet no evidence of transport by currents at these greater depths.' The area between 100 fathoms and the next 900 fathoms, it will be observed, covers only 9 per cent, of the whole floor of the ocean. This area is called the continental slope. It represents the sides of the ocean basins, and shows a relatively rapid descent into deep ■ On the W'yville Thomson Ridge, between the North of Scotland and the Faroe Bank, mud is not deposited at depths of 250 and 300 fathoms. The tidal wave, being here confined, apparently sweeps the top of the ridge at these great depths, and we have evidence of the same phenomenon on the saddle-backs between contiguous oceanic islands, and on very deeply submerged ridges. Further observations in this direction arc much desired. CHART IX.— DEPTHS Depths of less than r,ooo fathoms are dottedj beyond these a contour line is drawn at every thousand fathoms, the figure i meaning ‘over 1,000 fathoms*, 2 meaning 'over 2,000 fathoms', and so on. The extreme North is left blank (except in the Greenland area) for want of detailed observation Facivg page S4li Oceanic Deposits 249 water; it is the only area of the ocean’s floor which can be regarded as concave upwards. Beyond 1,000 fathoms the floor of the ocean becomes more and more flat and uniform : 84 per cent, lies below 1,000 fathoms 65 ,, ,, 2,000 ,, and ” ,, ,, 35®®® 55 These great submerged plains sink, as we have seen, in some instances to depths of nearly six English miles, and at times they are interrupted by cone-like elevations, some of which do not rise to the surface of the ocean, while others do, and then form groups of volcanic or coral islands. Although a great many soundings have been recorded exceeding 3,000 fathoms, and a great many sub- marine elevations have been also recorded in recent years, still the general result of recent explorations has been to increase the area of the ocean lying between 2,000 and 3,000 fathoms. Deposits on the Floor of the Ocean We have very few indications of the nature of the solid rocks forming the base on which oceanic deposits have been laid down, therefore all fragments of rock from a greater depth than 100 fathoms should be carefully pre- served for study. Along some continental coasts frag- ments have been dredged which indicate an outcrop of tertiary strata. Along some coral atolls solid calcareous rocks have been torn from larger masses, and off both coral atolls and volcanic islands fragments of igneous rocks thickly covered by peroxide of manganese have been broken from steep slopes, indicating an outcrop of volcanic rocks. In the north Atlantic, from a depth of 1,460 fathoms, the telegraph engineers have brought up, by means of the ‘Lucas Grapnel’, fragments of gneiss which had been 250 The Sea Floor broken olT larger masses. These may have come from outcrops of bare rock in place, but more probably they were broken from large boulders of gneiss transported to the position by icebergs. We have likewise very few indications of the thickness of the deposits now covering the floor of the ocean, or of the rate at which these are now forming. In the North Atlantic, telegraph engineers think there are reasons for supposing that about i inch of Globigerina Ooze accumu- lates in ten years. The rate at which this type of deposit accumulates, where cold and warm currents meet at the surface, is probably greater than elsewhere; and the rate at which terrigenous deposits are laid down near the em- bouchures of large rivers is evidently greater than that of any other type of deposit. The deep Red Clay of the Pacific Ocean far from land appears to accumulate at an exceedingly slow rate. The materials which make up oceanic deposits are now fairly well known, but are worthy of much more detailed study. They consist of inorganic and organic elements: 1 . Materials of all kinds borne to the ocean by rivers, winds, and floating ice, as well as what is torn from the coasts by waves, currents, and other agencies; these are termed ‘terrigenous’ or ‘earth born’. 2. Materials derived from volcanic eruptions, both submarine and subaerial. Volcanic dust is transported immense distances by the wind. Pumice of various types is worthy of special notice, because its areolar spongy structure, containing cavities filled with gas, admits of very wide distribution, by drifting on the surface of the ocean. 3. Materials of extra-terrestrial origin, which have fallen to the earth from interstellar space, such as cosmic spherules of iron and nickel, and chondrites with lamellar structure. Oceanic Deposits 251 4. Materials of organic origin may be carried to the ocean by rivers and winds, but the chief source is the hard parts — the shells and skeletons — of organisms which live in the surface, intermediate, or bottom waters of the ocean. These may be composed of lime (calcium carbonate), such as calcareous Algae, Foraminifera, Corals, Alcyonarian and Tunicate spicules, worm tubes, Ostracod and Cirriped shells, Echinoid spines, shells of Molluscs, Polyzoa, bones of fishes and whales, &c.; or they may be composed of flinty (siliceous) spicules and frustules, such as Diatoms, Radiolarians, and Sponge spicules. In these organic re- mains it is important to distinguish between those which belong to floating or plankton organisms, and those which belong to the bottom-living or Benthos organisms. Gene- rally speaking, it has been found that the remains of plank- ton organisms predominate in the deposits of deep water far from land, and the remains of bottom organisms pre- dominate in the deposits of shallow waters near shore. 5. In addition to the foregoing there are many sub- stances which may be termed ‘chemical’ or ‘secondary’ products. Clayey matter, which arises from the decom- position of pumice, felspars, and other minerals on the sea floor, may be regarded as a secondary product ; so also may the manganese nodules so abundant in some Red Clay areas, where the lapilli and bombs of basic rocks thrown out by volcanoes abound ; the palagonite and zeo- litic materials found in the Red Clay may be placed in the same category. In addition, ‘greensand’ (glauconite) grains, glauconitic casts of calcareous organisms, phos- phatic concretions, sulphate of barium nodules, iron and calcareous concretions, must be classed among the secondary products now forming in the deposits on the sea bed, many of them through the intervention of decaying organic matter and bacteria. 252 The Sea Floor The varied materials just enumerated are distributed very unequally over the floor of the ocean. The ter- rigenous materials are most abundant along the shores of continents and islands, and in all enclosed and par- tially enclosed seas. The geological structure of the adjacent land-masses determines the general character of the deposits along the shores and in the shallower waters, especially as regards the inorganic constituents. The deposits off coral-reef shores are, of course, chiefly made up of fragmentary particles derived from the growing reefs. The size of the fragments diminishes, as a rule, as the distance from the shore line and the depth of water increase; but this is not always the case, for in the very centre of the Pacific and in very deep water rounded frag- ments of pumice have been obtained of all sizes, up to more than i foot in diameter. Again, in regions now, or in recent geological times, affected by floating ice, large boulders, and varied rock-fragments have been frequently dredged up hundreds of miles from land. It is known that seals and penguins may transport rock-fragments in their stomachs to great distances. For these, among other reasons, the size of the particles is of little use for the pur- pose of the classification of marine deposits. While the detrital matter from continents and islands, and the shells and skeletons of bottom-living organisms, prevail in the deposits near shore and in shallow water, the remains of pelagic or plankton organisms, and chemical or secondary products, prevail in all the deep water de- posits far from land. It is a remarkable fact, however, that in areas where the depth is 3,000 fathoms or deeper, hardly a trace of calcareous pelagic shells, such as those of Fora- minifera and Pteropods, is to be found in the deposit, although these may be met with in great abundance in an adjacent area where the depth is only 1,500 or 1,000 Oceatlic Deposits 253 fathoms. In like manner the skeletons of Radiolaria and the frustules of diatoms are not nearly so abundant in some deposits as would be expected from the numbers captured in the surface-waters of the region. Both cal- careous and siliceous remains are dissolved by sea-water, and when they are not specially abundant in the surface waters, they may be very rare in, or wholly absent from, the deposits at the bottom. At other times these remains of pelagic organisms appear to be almost entirely masked by the terrigenous materials, or other organic remains, forming the deposit. From the above considerations it is evident that the materials which go to form marine deposits vary in nature, abundance, and size, according to the structure of the nearest land, the character of the submarine volcanic erup- tions, the distance from the shore line, the depth of water, and the nature of the organisms living on the bottom and in the surface and subsurface waters. After the detailed study of the marine deposits collected by the Challenger and other deep-sea expeditions, Murray and Rcnard pro- posed the classification on page IS S' It has frequently been pointed out, even by the authors themselves, that this classification is not perfect, but no better has been as yet proposed. It is founded on the dominant element or characteristic aspect of each type of deposit, and attempts to bring into prominence their origin and constitution. There is no sharp line of demar- cation between the different types; they all shade the one into the other, and it is often difficult to determine what name should be given to a deposit on the border line between two or three types.' * It is L'sscntini to be continually on one's ffuard against sources of error when examining deep-sea deposits. The object in view is to arrive at .1 true nriiion of the composition of the deposit as it lies at the bottom of the sea. Sometinifs the trawl or dredge may come up filled with fully a ton of manganese nodules, or 254 The Sea Floor Fot the information of the geologist, physical geo- grapher, and others interested in marine deposits, it is specially important that descriptions should give as nearly as possible the proportions of the various constituents of which the deposit is made up, as well as the size of the mineral fragments. This, it is believed, can best be ac- complished by the method adopted in describing the Challenger collections, as set forth in the following model : H.M.S. CHALLENGER Station, 338. March 2r, 1876. Lat. 21° 15' S.; Long. 14° 2' W. Depth 1,990 fathoms. large and small fragments of pumice, there may be hundreds of sharks’ teeth mixed with dozens of oar-bones of whales Some of these may have living Lepas, or Hydroids, or Scrpula, or Foraminifera attached to them It has sometimes been supposed that the whole deposit was composed of these remains, it is, however, almost certain that these various objects were embedded in clay, which passed through the meshes of the nets The same may be observed w ith reference to the phosphatic and glauconitic nodules found in relatively deep water along continental shores In using sounding tubes with butterfly valves, a large pro- portion of the finer parts of the deposit is sometimes washed out while the tulie IS being hauled in, and the larger calcareous organisms appear to be more abundant than the) really are When a large quantity of deposit is brought up in the dredge or trawl, and passed through sieves, the various constituents are separated, and the different portions have occasionally been described separatelv , in this waj have arisen erroneously the names Orbulma ooze, Biloculina cla), and Coprolitic mud In like manner the finer parts of a Radiulanan ooze have been described as Diatom ooze, and the coarser parts of a Diatom ooze as Radiolanaii ooze When the deposit is prcseived in water or spirit, the sample is sorted into different lavers b) shaking, if a portion be taken for Description from the upper lasers, and another from the lower lasers in the bottle, widely different results will be obtained The best sample for description is that which is taken direct from the sounding tube, and it should be noted whether the sample is taken from the superficial or the deeper lasers With the knowledge we now possess it is possible for an expert to tell approximately the latitude, longitude, and depth from which a deposit has been procured When, then, a sample bears a label showing a Globigerina ooze, for instance, from a depth or latitude which does not accord with experience, it raises suspicions of mislabelling This has frequently occurred with samples labelled on the corks, which have been interchanged Oceanic Deposits 255 GLOBIGERINA OOZE. White, with slight rose tinge, granular, homogeneous, resembling chalk when dr)'. Calcium Carbonate (92-54 per cent.). Globigeri- nidae, Pulvinulina (80 per cent.); Miliolidae, Textu- MARINE DEPOSITS* I . Deep-sea depo- sits beyond 100 - fathoms. Red clay Radiolarian ooze Diatom ooze Globigeriiia ooze Pteropod ooze Blue mud Red mud Green mud Volcanic mud Coral mud A. Pelagic deposits formed in deep water removed from land. B. Terrigenous de- posits formed in deep and shal- low water close to land-masses. 2. Shallow-water | deposits between Sands, gravels, low-water mark muds, &c. and 1 00 fathoms. | 3. Littoral deposits | between high- Sands, gravels, and low-water j muds, &c. marks. I laridae, Lagenidae, Rotalidae, Nummulinidae (i per cent.) ; otoliths of fish, Gasteropods, Lamcllibranchs, Pteropods, Heteropods, Lepas valves, Ostracods, Echinoderm frag- ments, Polyzoa, Coccoliths, Rhabdoliths (11-54 percent.). Residue (7-46 per cent.). Reddish-brown. Siliceous Organisms (i per cent.). Sponge spicules, ‘ Krummcl has recently introduced the term ‘hemipclagic’ for the deep-sea terrigenous deposits, and ‘eupelagic’ for the pelagic deposits in the above classification (‘Handbuch der Ozeanographie’, vol i, p. 205, Stuttgart, 1907). These terms have, howeter, been employed already to express quite other facts. 256 The Sea Floor Radiolaria, imperfect casts of Foraminifera, Astrorhizidae, Lituolidae, a few Diatoms. Minerals (i per cent.). Mean diameter o-o6 mm. angu- lar; felspar, hornblende, magnetite, magnetic spherules, pumice, a few manganese grains, bronzite spherules. Fine washings (5-46 per cent.). Amorphous matter, with small mineral particles, and fragments of scoriae and siliceous organisms. Remarks. This is one of the purest Globigerina oozes obtained by the Challenger^ and is almost wholly composed of the dead shells of surface organisms. A few pumice fragments were found in the washings of a large quantity of deposit. When carbonate of lime is present, the shells and other calcareous organic fragments are determined by micro- scopic examination. The lime is then removed from the sample by weak hydrochloric acid, and the relative per- centages of lime and residue can be approximately esti- mated by inspection, or determined by a simple analysis. The nature and relative abundance of the clayey matter, mineral particles, and remains of siliceous organisms in the residue can be estimated by microscopic examination. A description of this kind can quite easily be carried out on board ship. The following is a short description of the principal types of deep-sea deposits: A. PELAGIC DEPOSITS I. Red Clay. This deposit is the most characteristic, and probably the most widely distributed, of all deep-sea deposits over the ocean’s floor, covering a very large por- tion of the deeper parts of the Pacific. The name is sufficiently expressive of the nature and appearance of ReJ Clay 257 this type of deposit, there being always a considerable proportion of amorphous clayey matter, usually of a reddish colour, passing in some regions into a dark choco- late colour from the abundance of small grains of peroxide of manganese. Usually the Red Clay contains very few, if any, remains of calcareous organisms, but occasionally there may be an appreciable admixture of the shells of pelagic and bottom-living Foraminifera, teeth and otoliths of fishes, fragments of Echinoderms, Molluscs, Ostracods, Polyzoa, &c., and on approaching shallower water in tropical and temperate regions the shells of pelagic Foraminifera become more and more numerous until the Red Clay passes into Globigerina Ooze. The remains of siliceous organisms (Radiolaria, Diatoms, Sponge spi- cules, arenaceous Foraminifera) may generally be detected ; and in some regions where the Radiolarian remains be- come abundant, the Red Clay passes gradually into Radiolarian Ooze, while in other regions towards the far north and far south the Diatom remains increase in number, and the Red Clay may pass into Diatom Ooze. Among the inorganic elements met with in Red Clay, the most constant and widely distributed is pumice, which occurs in larger and smaller fragments down to the most minute particles, and in all stages of decomposition and disintegration; the minerals found in pumice, like sani- dine, plagioclase, augite, hornblende, magnetite, &c., are also present, along with basic volcanic glasses frequently transformed into palagonite. The peroxides of iron and manganese are universally distributed throughout the Red Clay, sometimes as minute grains or coatings, sometimes deposited as concretions around organic remains, pumice fragments and other nuclei, forming manganese nodules of larger or smaller size, especially where the debris, ashes, and lapilli, of basic volcanic rocks are abundant and have 258 The Sea F/oor undergone decomposition. These manganese nodules vary in form and size in different localities; at one place they may be large subspherical, and smooth, resembling a lot of potatoes; at another place, smaller, like marbles; at another place large and spherical, but the external surfaces rough to the touch, owing to the numerous mammillations (Fig. 153); at another place flattened with one side rougher than the other; and at yet another place the nodules take the form of huge slabs. With the manganese no- dules are usually associated, especially in deep water far from land, numerous teeth of sharks and ear-bones of whales, im- pregnated and coated more or less thickly by the peroxides of iron and manganese. In the Red Clays, also, numerous minute magnetic spherules (Fig. 154), some with metallic nuclei, have been met with, and have also been extracted from the man- ganese nodules after these have been broken up in a mortar; these spherules are supposed to have fallen from inter- stellar space, and are hence called ‘cosmic spherules’. In some positions, again, there are small zeolitic crystals (phillipsite), as individuals, twins, stellate groups, and spherulitic aggregations; these are supposed to be secondary products, formed in situ, arising from the .'W v-'-S . 7 . ^ . •■ 7 ’ Fig. 153. Scalpellum on a Manganese Nodule Radiolartan Ooze 259 decomposition of the basic volcanic particles present in the deposits. Red Clay is estimated to cover an area of about 51.500.000 square miles (or 133,975,000 square kilo- metres) — i.e. about 40,800,000 square miles in the Pacific Ocean, about 5,800,000 square miles in the At- lantic Ocean, and about 4,900,000 square miles in the Indian Ocean. 2. Radiolarian Ooze. This deposit is distinguished by the abundance of the skeletons of Radiolaria, and is found typically in very deep water in the tropical regions of the Pacific and Indian Oceans. ‘ It is otherwise similar to the Red Clay just described, and may contain a few shells of pelagic Foraminifera and small angular volcanic mineral particles, frag- ments of pumice, augite, felspars, hornblende, magnetite, volcanic glass, frequently altered into palagonite, as well as man- ganese nodules, sharks’ teeth, and ear-bones of cetaceans. Radiolarian Ooze is estimated to cover an area of about 2.290.000 square miles (or 5,957,000 square kilometres) — i.e. about 1,160,000 square miles in the Pacific Ocean, and about 1,130,000 square miles in the Indian Ocean.' 3. Diatom Ooze. This deposit is distinguished by the prominence of Diatom frustules (skeletons), and is there- fore characteristic of those regions in which these pelagic plants swarm in the surface waters, as in the extreme northern part of the Pacific and far south in the neighbour- hood of the Antarctic circle. The skeletons of Radiolaria w /I ^ ■ ] L A FiC. 154. Spherule with metallic nucleus coated with black Magnetite, supposed to lie of cosmic origin > A small parch of Radiolarian Oorc has since been found by the Discovery Expedition in the South Atlantic at 46° S. 23° W., and 50° S. 30° W. — Ed. 26 o The Sea Floor and the shells of one or two species of pelagic Foraminifera are usually present, as well as continental mineral particles and ice-borne rock fragments, since this deposit occurs generally within the regions affected by floating ice (Fig- ^SS)- In some of the Red Clays and Radiolarian Oozes of the Central Pacific the frustules of the large diatom Cos- Fig. 15?. Diatom Oo7c 1-5, Coscinodiscus; 6, Asteromphalus; 7, Fragilaria antarcica; 8, 9, Synedra; 10, Rhizosolenia; ii, Chaetoceras; iz, Navicula (r); 13, Dictyocha; 14, a broken Radiolarian cinodiscus rex are so abundant that the deposit has some- times been described as a Diatom Ooze. Diatom Ooze is estimated to cover an area of about 10,880,000 square miles (or 28,300,000 square kilo- metres) — principally in the Southern Ocean, with a small area in the North Pacific Ocean. 4. Globigerina Ooze. This deposit is named from the predominance of the dead shells of Foraminifera, which lived in the surface waters of the ocean, the genus Globi- gerina (Figs. 74, 156) being the most characteristic. Globigerina Ooze 261 though the representatives of other genera are usually present in the tropics (Fig. 157). Associated with the Fig. 156. Dead shells of Globigerina, the spines of which have been dissolved in their fall to the bottom Fig. 157. Globigerina Ooze, consisting largely of Foraminiferan shells shells of pelagic Foraminifera are the shells of pelagic Molluscs (Pteropods and Heteropods), pelagic calcareous Algae (Coccospheres and Rhabdospheres, or their broken fragments — Coccoliths and Rhabdoliths), as well as the 262 The Sea Fioor reiiidinb uf cJcaicuus oiganisms, which habitually live on the bottom ot the sea, such as Molluscs, Echinoderms, Annelids, Coials, Polyzoa, and bottom-living Foramini- fera. The remains of siliceous organisms (Radiolaria, Diatoms, and Sponge spicules) may generally be detected, and a few small mineral particles, such as felspar, augite, hornblende, magnetite, and volcanic glass, with a small quantity of clayey matter coloured by the oxides of iron and manganese. This deposit varies greatly in composi- tion, both with respect to the species of organisms present and their relative abundance, and also in the abundance and nature of the mineral constituents. In all the great depths of the ocean, exceeding 2,500 or 3,000 fathoms, Globigerina Ooze gives place to Red Clay, even in those regions where pelagic Foraminifera inhabit the surface waters in great profusion; this is ascribed to the longer time during which the shells are exposed to the solvent action of sea-water, while falling through the greater depth of water, and while lying uncovered on the bottom. Globigerina Ooze is estimated to cover an area of about Pteropod Ooze 263 49.520.000 square miles (or 128,824,000 square kilo- metres) — i.e. about 22,500,000 square miles in the At- lantic Ocean, about 14,800,000 square miles in the Pacific Ocean, and about 12,220,000 square miles in the Indian Ocean. 5. Pteropod Ooze. This deposit differs from Globi- gerina Ooze only in the greater abundance of the shells of pelagic Molluscs (Pteropods and Heteropods), and occurs characteristically at lesser depths than the Globi- gerina Ooze (Fig. 158). Thus Pteropod Ooze may be said to attain its typical development at depths of 800 to 1.000 fathoms, while Globigerina Ooze occurs typically at depths of 1,500 to 2,000 fathoms. The reason why the shells of pelagic Molluscs are removed from the deposits at lesser depths than the shells of pelagic Foraminifera is believed to be that these thin and fragile shells present a larger surface to the solvent action of sea-water. Pteropod Ooze is estimated to cover an area of about 400.000 square miles (or 1,000,000 square kilometres), principally in the Atlantic Ocean. B. TERRIGENOUS DEPOSITS 6. Blue Mud. This deposit is the one most frequently met with in the deeper waters surrounding continental land and in enclosed and partially enclosed seas. It is principally composed of materials derived from the dis- integration of continental land, consisting largely of the fragments and minerals of continental rocks (the older crystalline and schistocrystalline rocks, quartzites, sand- stones and limestones) of various dimensions, but usually larger near shore and smaller as the deep sea is approached, except in the regions affected by floating ice. Quartz is the characteristic mineral species, associated with ortho- clase and plagioclase felspars, green hornblende, mica. 264 The Sea Floor &c.; glauconite is usually present, but not in such abun- dance as in the Green Mud. There is usually a considerable proportion of amorphous clayey matter, increasing in amount with increasing distance from land, so that some of the deeper samples have a decidedly clayey aspect, but the deposit, as a rule, may be described as earthy rather than clayey. In some situations the remains of bottom- living organisms may be present in considerable numbers, and in others the remains of pelagic organisms may be so abundant that the deposit resembles a Globigerina Ooze. Blue Mud is estimated to cover an area of about 14,500,000 square miles (or 37,700,000 square kilo- metres). 7. Red Mud. This is a local variety of Blue Mud, hitherto known only from the Yellow Sea in the Pacific, and off the Brazilian coast in the Atlantic, characterized by the presence of a large quantity of reddish ferruginous matter brought down by the large rivers in the vicinity. Red Mud is estimated to cover an area of about 1 00,000 square miles (or 260,000 square kilometres). 8. Green Mud. This is a variety of Blue Mud, found along bold exposed coasts where no very large rivers enter the sea, and distinguished by the greater or less abundance of green glauconite grains and glauconitic casts (internal models in green sand), of the shells of calcareous organisms, usually associated with a greenish amorphous (probably organic) matter. In the shallower depths the glauconitic grains and casts are sometimes associated with phosphatic concretions, while the amor- phous clayey matter is less abundant, and the deposits, being more granular and more incoherent, are called Green Sands. Green Mud is estimated to cover an area of about 1,000,000 square miles (or 2,600,000 square kilometres). Volcanic and Coral Muds 265 9. Volcanic Mud. This deposit occurs around oceanic islands and submarine elevations of volcanic origin, and is made up largely of volcanic rock fragments and volcanic mineral particles, such as lapilli of basaltic and andesitic rocks, especially the vitreous varieties, sanidine, plagio- clase, augite, hornblende, rhombic pyroxenes, olivine, and magnetite. In shallower waters the volcanic particles are larger, associated with less amorphous clayey matter, and the deposits, being less coherent, are called Volcanic Sands. The remains of floating or bottom-living cal- careous organisms may become so abundant that it is sometimes difficult to distinguish this deposit from a Globigerina Ooze on the one hand, or from a Coral Mud or Sand on the other. Volcanic Mud is estimated to cover an area of about 750.000 square miles (or 1,950,000 square kilometres). 10. Coral Mud. This deposit occurs off coral islands and coral reefs, and is chiefly made up of fragments of organisms living in the shallow waters and on the reefs, such as calcareous Algae, Corals, Molluscs, Polyzoa, Annelids, Echinoderms, and Foraminifera. In the shal- lower waters near the reefs these calcareous fragments are larger, and the more finely divided calcareous matter is less abundant than in the deeper waters further removed from the reefs; the deposits are then called Coral Sands. These deposits may contain, at times, much volcanic material. Coral Mud is estimated to cover an area of about 2.700.000 square miles (or 7,000,000 square kilometres). General Remarks The Black Sea is far removed from the great oceans, and its deeper water is cut off by submarine barriers at the Bosphorus from the Mediterranean, which in its turn is 266 The Sea Floor cut ofF from the Atlantic by an additional barrier at the Straits of Cibraltar. Vertical circulation is much re- stricted by this and by the fact that the surface layers are often much less saline than the deeper layers; hence the deeper waters of the Black Sea become more or less saturated with sulphuretted hydrogen ; no living organisms (other than bacteria) are met with in depths greater than loo fathoms. Amorphous carbonate of lime is preci- pitated from the water of the Black Sea, and makes up a considerable part of the deposits now in process of forma- tion ; this is the only place where this is known to occur in the seas of the present day. In all partially enclosed seas — like the Mediterranean and the Red Sea — circulation is generally cut off by submarine barriers, and this results in thermal and other conditions which are less favourable to organisms than the conditions existing in the open ocean. The deposits in those seas which interpenetrate the continental masses are terrigenous in origin, although there may be an approach to pelagic conditions towards the more central parts. It has already been observed that the continental shelf facing the great oceans is continually swept by waves, tides, and oceanic currents, down to an average depth of loo fathoms. Just beyond this depth all the minute mineral and organic fragments from the continental shelf come to rest on the bottom, and form what I have called the ‘mud line’. This constitutes the great feeding-ground of the ocean. Large numbers of Holothurians and other marine creatures here eat the mud to obtain the organic matter associated with it. Indeed, it is more than probable that all marine deposits are in this way passed through the intestines of organisms. Very many Crustaceans frequent this area to pick up the little particles of organic matter which are just settling on the bottom, and some of them — Only the Iirger areas aie here sho\t n Continental Slope 267 like Nyctiphanes — are provided with phosphorescent organs to enable them to do this more effectively. All these mud-eating creatures are in turn the prey of car- nivorous animals, both vertebrate and invertebrate. The continental slope extends from the mud line (100 fathoms) down to the mean sphere level (1,450 fathoms).' The continental slope and similar areas around oceanic islands present a greater variety of conditions than is found elsewhere on the ocean bed. At the upper limit the sun’s rays may produce twilight, but elsewhere there must be total darkness, except where this is relieved by the phosphorescent light of organisms. The temperature con- ditions are likewise widely different at the upper and lower limits of the region. At some points the descent from the loo-fathom line is known to be almost perpendicular; at other points outcrops of stratified and volcanic rocks are indicated. The deposits now being laid down over the continental slope vary greatly according to position. Off large rivers they are chiefly made up of detritus from the land ; at other places, especially where cold and warm currents alternately occupy the surface, pelagic conditions are more or less pronounced, and greensand and phosphatic deposits are being laid down; quartz and other continental minerals predominate. Generally it may be said that in enclosed seas, and along the continental shelf and slope, deposits are being formed quite similar to those which have made up the stratified rocks of past ages. Indeed, it seems as if inland seas and the borders of continental masses had again and again been pushed up into dry land, and again and again been torn down and transported to the ocean ' If all the elevated portions of the earth’s crust were cut away, and filled into the hollows till the whole surface was uniform, then the whole earth would be covered by an ocean 1,450 fathoms in depth — the mean sphere level. 268 The Sea Floor by the same denuding and disintegrating agents, the final result being that quartz particles accumulate on the con- tinental areas, rendering these rocks specifically lighter than the deposits on the oceanic areas. In the abyssal regions, beyond the depth of the mean sphere level covering about one-half of the earth’s surface, the physical conditions are uniform and widespread; the temperature everywhere approaches zero Centigrade ; the darkness is relieved only by phosphorescent light; motion of all kinds must be extremely slow, and there is little evidence of transport or erosion. In these cold and silent depths the deposits vary accordingly to the surface condi- tions ; where organisms which secrete carbonate of lime abound at the surface, we find Globigerina or Pteropod Oozes at the bottom; where silica-secreting organisms abound at the surface we find Radiolarian or Diatom Oozes at the bottom. In very great depths the calcareous organisms are removed by solution, and in some places the siliceous skeletons are also partly removed. From the deep water of the South Pacific Ocean, at points the farthest removed from continental land on the globe, the trawl brings up over wide areas hundreds of sharks’ teeth and dozens of ear-bones of whales, some of them belong- ing to extinct species, tons of manganese nodules and, mixed up in the clay, magnetic spherules of metallic iron and nickel, and chondres, which are only found in meteor- ites. All the indications go to show that the rate of ac- cumulation of the deposit is extremely slow, possibly not more than a foot since Tertiary times. The reason why these teeth and bones and extra-terrestrial spherules are found here more abundantly than elsewhere is because few other materials reach these remote and deep areas to cover up or mask them, as in other deposits. Throughout the abyssal area quartz fragments, when Radium 269 present, are rare, when compared with their abundance in the terrigenous deposits close to shore, and it is very doubtful if any stratified rock has yet been discovered which can be regarded as the equivalent of any deposit now forming in the deep water far from continental land. It is extremely difficult to imagine how a continent, as we understand the term, could ever have existed in the centre of any of the great ocean basins, which appear to be the most stable portions of the earth’s crust, while the con- tinental areas, especially the border regions, are the most unstable. From a recent examination of some Challenger de- posits sent to him from the Challenger Office, Professor J. Joly has been able to give the radium contents of deep- sea deposits. He finds that the Red Clays and Radio- larian Oozes contain much more radium than the Globi- gerina Oozes or terrigenous deposits; in fact, the radium content is greatest where the rate of deposition is believed to be the least. The idea at once suggests itself that this is connected with the greater amount of cosmic dust in the deep deposits far from land. This is well worthy of the attention of those who in future may dredge up large samples of deposits from the Red Clay areas. VII ANIMALS OF THE SEA FLOOR BY G. P. FARRAN AND W. T. CALMAN In dealing with the fauna of the sea bottom it is necessary to make use of some convenient and natural classification of the subject. It has long been recognized that, in passing from the shallow water surrounding the shore to* the abysses of the ocean, we may find many striking changes in the character of the inhabitants associated with the increasing depth ; and a classification based on the depth at which the animals live is a more or less natural one. While those areas or zones of depth which are accepted are marked out, rather on account of their physical features than because they possess separate faunas sharply defined from each other, still the assemblage of animals found in each zone, taken as a whole, differs con- siderably from those of the other zones, and may con- veniently be dealt with apart from them. The zones of depth usually distinguished in European seas are based on those originally (defined by Edward Forbes in the middle of the last century. These were the Littoral Zone, the Laminarian Zone, the Coralline Zone, and the region of Deep-Sea Corals. For the last two may be substituted the Continental Shelf and the Continental Slope, and to the list may be added the ocean floor, a region which in Forbes’s day was believed to be devoid of living inhabitants, but which more recent research has proved to have a varied fauna. These zones must be taken as applying primarily to the shores of the North Atlantic and the Mediterranean, but with some small alteration will be found to hold good for many other seas. The Littoral Zone, or the space between high and Zones of Depth 271 low-water mark, is dealt with elsewhere in this volume (Chapter III). It is necessarily inhabited by a fauna which is able to endure a certain amount of periodic exposure to the air. At its lower limit it passes into what is known in European waters as the Laminarian Zone, taking its name from the oarweed or Tangle (Laminaria), which, with other seaweeds, is its most striking feature. It extends to a variable depth, usually less than 1 5 fathoms. Though the name Laminarian is somewhat restricted in its applica- tion, yet the physical characters of this zone may be recog- nized almost universally. These may be summed up as the penetration of sunlight, and the consequent presence of abundant vegetable life in the form of brown and red seaweeds •, the exposure to violent wave action which may stir up sediment, and the lowered salinity from the surface drainage of the land, even in the absence of any large rivers. It is in this zone that we must look for the majority of the vegetable feeders of the sea, with the exception of those which live upon minute floating diatoms and similar organisms of the plankton. Below the region in which the brown seaweeds flourish, or (speaking very roughly) from 1 5 fathoms downwards, lies the region which we now deal with comprehensively as the Continental Shelf. As a general rule the land-masses of the globe are sur- rounded by an area of comparatively shallow water, vary- ing in extent, which has been called the ‘area of the Con- tinental Shelf’. This shelf or margin, sloping gradually to a depth of 100 or 200 fathoms, and extending for a distance of from 20 to 200 miles from the shore, is suc- ceeded by a relatively steep declivity, the Continental Slope descending to the ocean floor; this latter then stretches, at a depth of between 1,000 and 3,000 fathoms, over an area measuring about one-third of the whole 272 Animals of the Sea Floor surface of the globe, and is hollowed out in places into depths of 4,000 or even 5,000 fathoms (cf. Chapter VI). The edge of the Continental Shelf maybe taken as a divid- ing line between two faunal areas, the sloping descent of a continent to the ocean floor being regarded as a sort of neutral ground which, while having a few inhabitants that belong more properly to the Continental Shelf and a few peculiar to itself, is for the most part peopled by species which may be regarded as belonging to the fauna of the deep sea. It is to be remembered that this dividing line does not mark any sudden change either in the physical conditions or in the animals to be met with, and that any limit which may be chosen must in the nature of things be more or less an arbitrary one, and its position will be different in different regions. The physical conditions prevailing over the sea bottom are dealt with fully in other parts of this volume (Chapters II and VI), and it is only necessary to call attention to some of the chief differences between the areas which we are considering. The Laminarian Zone and Continental Shelf Throughout the area extending from low-water mark to the edge of the Continental Shelf we find that the vari- ability and liability to sudden alterations in physical condi- tions, which are so noticeable in the shallow water, become very much reduced. The changes in temperature in the deeper parts of the area are rather seasonal and gradual than diurnal or rapid, and are not dependent immediately on the alterations of the temperature of the air. The mean temperatures in various regions show, of course, very great differences. Continental ShelJ 273 The salinity, like the temperature, is locally fairly con- stant, the changes being slow, and the range comparatively small, though great differences may be noticed between different seas. The light is very much diminished, passing from a dim twilight in the more shallow parts to complete darkness in the deeper. The actual depth to which light can penetrate depends very much on the clearness of the water and its freedom from sediment; but it may be taken that at depths exceeding 600 fathoms practically no light can be received from above, while at depths considerably less the amount of light is so small as to have little in- fluence on life. The absence of light entails the absence of plant life and consequently of vegetable feeders, with the exception of such forms as subsist on microscopic floating vegetable matter. The greater part of the area is too deep to be troubled by wave action, and the force of tides and cur- rents is much diminished. The bottom is fairly uniform over large areas, though the actual types of bottom do not differ materially from those found immediately adjoining the shore, the materials of the sea floor in both cases being derived from the debris of the adjacent land. It is usually the case that the composition of the bottom becomes finer as the depth and the distance from shore increase, but this is by no means a universal rule, gravel, rocks, and stones being often met with even beyond the edge of the Con- tinental Shelf. The water overlying this zone contains in suspension a quantity of organic and mineral matter derived from the shore. The importance as a food supply of finely divided organic matter of vegetable origin, or ‘detritus’ as it is generally called, appears to be greater than was at one time suspected. Many mollusca and other bottom-living animals seem to use it in shallow land- locked waters, where the breaking up by wave action of 274 Animals oj the Sea Floor each year’s summer growth of seaweed and zostera fur- nishes an abundant and perennial supply. This detritus may drift out with the currents to considerable distances from the shore ; at what point it ceases to be a factor in the food-supply of the bottom-living animals it is difficult to determine. Murray indicated a limit, which he called the ‘mud-line’, at just beyond the lOO fathom line, and con- sidered the area included by it as ‘the great feeding-ground of the ocean’. Taken as a whole, the area from low-water mark to the edge of the Continental Shelf possesses a fauna which is perhaps richer and more varied than that of any other area on the surface of the globe. A catalogue of its inhabitants would include all the chief types of invertebrate animals (except the air-breathing Arthropods) and many fishes. It would be impossible within the space at our disposal to attempt even an outline sketch of this fauna and of its variations in different parts of the world, but for the guidance of those unfamiliar with zoology a few of the commoner types represented in the illustrations may be mentioned. The simplest animals (Protozoa) are represented especially by the bottom-living Foraminifera, which are abundant in certain localities, creeping on the surface of the mud or firmly adhering to stones or other solid bodies. The shells of the Foraminifera are often exceedingly beautiful (Figs. 159-63). They are generally composed of carbonate of lime, but some are built up of grains of sand cemented together. Most of them are of microscopic size, although in tropical seas some species may be as much as an inch across, and even in the cold waters of the deep Norwegian fjords the star-shaped sandy Astrorhiza is as big as a threepenny piece (Fig. 162). Sponges (Porifera) often form a conspicuous part of the Fig. 163. Miliolina TYPES OF FORAMINIFERA 276 Animals oj the Sea Floor con tenth of the dredge or trawl in waters of moderate depth, and deserve the attention of the collector, not only for their own sake, but also on account of the numerous Crustacea, worms, and other animals which burrow in them or shelter in the crevices. In passing from shallow to deeper water the shapeless encrusting sponges of the littoral region are replaced, in part at least, by types of definite and often beautiful form (Fig. 164). This is especially the case with the glass-sponges (Hexactinellida), of which the well-known Venus’s Flower-basket is an example. Below the depth at which true seaweeds flourish the dredge is often filled with masses of what the fisherman or trawler terms ‘weed’, but which the zoologist recognizes as plant-like animals. Even for the zoologist it is not always easy to distinguish by the unaided eye between the feathery tufts of Hydroids (Figs. 87, 89, 165) and the much more highly organized Polyzoa. Belonging to the same division of the animal kingdom (Coelenterata) as the Hydroids are the sea-pens, or Pennatulida (Fig. 1 66), the sea-fans, or Gorgoniacea (Figs. 167, 168), the fleshy masses of the ‘dead-men’s-fingers’ (Alcyonaria), the red or precious coral (Corallium), the sea anemones, and the true corals (Madreporaria), both colonial and solitary (Fig. 169); the true corals are by no means confined to the warmer seas, although it is only there that they display their full luxuriance. Starfishes (Fig. 171), brittle-stars (Fig. 170), sea- urchins (Fig. 173), and their kindred are abundant in all seas. Belonging to the same group (Echinodermata) are the sea-cucumbers, trepangs, or Holothuroidea (Fig. 176), and the feather-stars (Crinoidea), some of the latter free- living like the Antedon of our own seas, some attached by a stalk like the ‘stone-lilies’ of earlier geological epochs Fig. 164. Holtenia, a Deep-Sea Glass Sponge {Challenger). ^ J 278 Animals of the Sea Floor (Fig. 177). These stalked erinoids, confined for the most part to fairly deep water, were formerly among the most prized rarities of zoological collections, though the pro- gress of deep-sea exploration has now rendered them tolerably familiar. Under the name of ‘worms’ are included very many animals of widely dissimilar appearance. The Planarians are mostly flattened leaf-like forms, often of brilliant colours, which glide smoothly over the surface of stones or shells. The Nemertines may be recognized by their soft, very extensile bodies, and their long thread-like proboscis, which can be completely withdrawn within the animal. They are frequently richly coloured and of very varied form, some slender and of immense length, others short and thick and very fragile, breaking up into small fragments on the least provocation. Burrowing in mud or clay we find the Echiuroids, or spoon-worms, and the Sipunculoids, the latter with tough leathery bodies, blunt at the tail, and tapering to a small proboscis, which can be withdrawn into the body like the finger of a glove which is turned inside out. By far the most important and numerous group of worms is that of the bristle-worms, or Polychaetes, which are to be found practically everywhere in salt water (Figs, lyaand 174). They carry stumpy, foot-like lobes beset with bristles along each side of the body, by means of which they crawl or swim, and breathe by gills which are either situated on the lateral lobes, or else (in the tube-dwelling forms) crowded on the head of the animal. These Poly- chaetes may be found in most unexpected places, one species living in the topmost coils of a dead shell in- habited by a hermit-crab, another in the groove on the underside of the common sandstar, and yet another cling- ing closely to the underside of a species of sea-cucumber. 2 8 o Animals of the Sea Floor The tube-building forms display an endless variety in their habitations. The tough parchment-like tubes of Chaetopterus and Hyalinoecia (Fig. 174) are buried in sand or mud, others are made of sand or broken shells, some roughly put together, and some built up with ex- quisite neatness and symmetry. Many species secrete calcareous tubes of various shapes, usually fastened to stones or shells, and closed by a stopper which the in- habitant carries attached to his head. The feathery plumes which many of these tube-dwellers bear as gills are among the most attractive sights of an aquarium, where alone they can be seen to full advantage (Fig. 172). Perhaps the most graceful are those of Sabella and its allies, whose tubes, built largely of mud, have the appearance and tex- ture of india-rubber piping. Though most of the free- living Polychaetes are unable to swim, there are a few entirely pelagic species ; and some of the normally creep- ing forms, when the breeding season arrives, become entirely altered in appearance, developing swimming- lobes along the sides of their bodies, and come to the sur- face in immense swarms. The ‘Palolo’ of Pacific islands, which, leaving its shelter in the coral reefs, suddenly ap- pears in countless numbers on the shore, is a well-known instance of this change of habit. The huge group of the Crustacea is one which perhaps attracts more attention than any other. In number of species it far exceeds any other group of marine animals, but most of these are of very minute size. It is perhaps not out of place to suggest here that, in order to capture these smaller forms, the water in which dredged material has been put to stand should be strained off through a piece of fine silk. Very unlike the other Crustacea in appearance are the Cirripedes, including both the stalked barnacles and the sessile acorn- Fig i66 Pennatula, a Sea FiC i68 Gorgoma, showing the Pen yi Polyps in Contraction and Ex- pansion Enlarged 282 Animals oj the Sea Floor shells. The stalked forms are often found growing on floating objects, but some genera such as Scalpellum are found in deep water attached to Hydroids or other fixed objects (Fig. 1 53, p. 258). The sessile or acorn-barnacles are most abundant between tide-marks or in shallow water, but there are a few deep-sea species. Of the other Crustacea, the Branchiopods belong to the plankton (Chapter V), but the Copepods and Ostracods have many bottom representatives. There are two very large groups of Crustacea, number- ing some thousands of species, which we may conveniently illustrate by the sandhopper and the wood-louse. These are the Amphipods (Fig. 179) and the Isopods (Fig. 1 82). They may be found everywhere, and may, for the most part, be recognized by their resemblance to the types above mentioned. One section of the Amphipods, however, the ghost shrimps, or Caprellids, differs very markedly from the rest. The Caprellids may be recognized by their slender skeleton-like bodies, their thin legs, the two front pairs of which are armed with strong claws, and their ‘looper’-like movements (Fig. 1 80). The Cumacea, another group of small Crustacea, living for the most part buried in mud, may be known by their small nut-shaped bodies and their long slender tails ending in a fork. Resembling the Prawns in appearance are the so-called ‘Schizopods’, a name sometimes used to include two unrelated groups. Of these, the Euphausiacea (Fig. 1 16, p. 223), are almost entirely pelagic, but the Mysidacea are for the most part bottom-haunting forms, rarely more than an inch in length. Of the larger Crustacea the most unfamiliar are pro- bably the squills, or mantis shrimps (Squilla) (Fig. 126, p. 22 5), with small bodies and broad, flattened tails. They Crustacea 283 carry a poAverful pair nf claws, which they lift in a manner that recalls the mantis insect. They arc rarely found except in the warmer seas. The most highly organized, and the largest of the Fig 169 Lcpiopcnus, a Deep-Sea Coral .,4 Crustacea, the Decapoda, are usually divided into three groups: (i) the Macrura (or long-tailed), including the lobsters, prawns, and shrimps (Fig. 178); (2) theAnomura (or irregular-tailed) — these comprise, among others, the hermits (Paguridae), with soft bodies (Fig. 181), and the Galatheids, in which the tail is broad and flattened, and can either be turned up under the body or else straightened 284 Animals of the Sea Floor out and used for swimming (Fig. 184); (3) the Brachyura, or crabs proper — these include an immense number of very different forms; they retain, however, a sufficient family resemblance for recognition as crabs. Some of the more conspicuous types are the swimming crabs, in which the last pair of legs are paddle-shaped and the body flattened, as in the British genera Portunus and Bathy- nectes ; and the spider crabs, which have the legs long and slender, and apparently more designed for concealment than locomotion, the movements of these crabs being deliberate and sluggish, and their bodies often furnished with a crop of seaweeds, Hydroids, or sponges. The Ebalias, with their small round bodies and closely tucked-in claws, might easily pass for pebbles, and usually frequent ground where the bottom is of coarse gravel. The sea spiders, or Pycnogonida, are easily recognized by their resemblance to the animals from which they take their English name. In most species the small body scarcely affords room for the attachment of eight (very rarely ten) long slender legs, but in a few cases the legs are short and stout; as, for instance, in the common Pycnogonum littorale, which, in spite of its name, is often found in moderately deep water. Most of the Pycnogons are of small or moderate size, but in some deep-water species the legs may span a length of 2 feet. The commoner forms of the Mollusca, at any rate of the shelled species, are familiar to every one, but there are some shell-less species, and others with peculiar shells, which call for special note. The usual type of Gastropod, as exemplified by the whelk or periwinkle, bears an asym- metrically coiled shell, but there is reason to believe that the earliest forms of molluscs were symmetrical. This symmetry is still found in the Chitons, in which the shell 2 86 Animals of the Sea Floor is made up of eight plates, overlapping as in an armadillo or wood-louse (Fig. 1 85). Allied to the Chitons is a very remarkable group of molluscs, in which the shell is en- tirely absent. These are for the most part worm-like in form, and are usually found tightly coiled round the branches of Hydroids and Alcyonarians. The nudibranchs, or sea-slugs, are shell-less gastropods, which externally appear to be symmetrical. They start life, however, in a minute coiled shell. One section of them, the Dorididae, in which the gills are arranged in a circle on the hinder end of the animal, are found in their greatest development, both in number and size, in tropical seas, where, both along the shore and in shallow water, they attract attention by their large size and brilliant colours. Another section, the Eolididae, in which the gills are not concentrated on one spot, but scattered as papillae all over the body, are, on the contrary, very poorly repre- sented in warm seas, but are richly present on our own coasts (Fig. 183). The Lamellibranchs, comprising such molluscs as the oysters, mussels, cockles, scallops, and so forth, are too well known to need further mention. Among the Cephalopods, the Octopus and its allies (Fig. 186), with eight equal arms and a rounded body, must be distinguished from the squids and cuttle-fish, which have ten arms, one pair of arms being much longer than the rest. The ‘paper Nautilus’ belongs to the first of these groups. The second includes a great variety of forms — some short and squat, like the cuttle-fish (Sepia, Sepiola), others elongate, like Loligo and the other squids (Fig. 187). The pearly Nautilus belongs to a group apart from all the rest. It is the sole representative of a once very abundant Order, to which the Ammonites, so well known as fossils, also belonged. Most of the Cephalopods 288 Animals oj the Sea Floor can swim very well, and some are entirely pelagic, but most are to be found at the bottom at moderate depths, and some abyssal forms have been found so deep as 2,000 fathoms. If we look at the stones brought up by the dredge, we are sure to find many of them covered with a fine lace-like incrustation, which, upon close examination, seems to be made up of minute cells, each perforated by a small pore. These cells are formed by the Polyzoa, each cell having a separate inhabitant, which extrudes a tiny crown of tentacles through the opening. Besides the encrusting forms there are other branching ones (Figs. i8 8, 189), either soft and finely divided, like Hydroids, as already mentioned, or else thick and hard, like small corals. Allied to the Polyzoa, but differing from them entirely in appearance, is the group known as the Brachiopods, or lamp shells. The Brachiopods were once classed as Mol- lusca, on account of their external resemblance to Lamelli- branchs, but may generally be distinguished from them by the facts that their valves are unequal in size, and that they are attached by a small stalk, which passes out through a hole in the apex of the larger valve (Fig. 185a). They represent a group which was very abundant in early geo- logical times, and of which some forms have persisted almost unchanged to the present day. Many of the species are found at considerable depths. The Tunicates, or sea squirts, are generally particularly unattractive objects when found among the contents of the dredge. The solitary forms have shapeless sack-like bodies of a gelatinous or leathery texture, often encrusted with mud or sand, and having two openings through which, in the living animal, currents of water enter and leave the body; the colonial forms often encrust stones in large sheets. They are strangely degenerate animals, for 3i66 Fig 173 Salenia, a Sea Lrchin X4 Fig 174 Hjalmoecia a Bnstle-W orm Reduced Fig 173 Amphinome, a Brinde-Worm xj 290 Animals oj the Sea Floor they begin life as free-swimming, tadpole-like larvae, which approach somewhat to the Vertebrates in their structure. As regards the geographical distribution of the animals inhabiting moderate depths, it may be said, speaking very generally, that the character of the fauna is determined rather by temperature than by geographical position. Thus it is found that genera and even species of animals, which may be dredged at a depth of a few fathoms in the Arctic regions, occur at greater depths in the more southern parts of their range, where the shallow water is too warm for them. For example, a certain starfish, which is found commonly off the west coast of Ireland, between 300 and 400 fathoms, and which stretches as far south as the Bay of Biscay in deep water, occurs at a depth of 1 5 fathoms within the Arctic circle. Again, the current of relatively cold water, which runs northwards along the west coast of South America, allows many of the southern types to extend far to the north, and carries the character- istically subantarctic Isopod genus Serolis to the coasts of Southern California. It is possible to divide up the shallow water of the globe into regions whose faunae differ from each other, both in the actual species which compose them, and also in the general proportion in which the various groups of the animal kingdom are represented. Thus it is found that the fauna of the shores of East Africa has a much closer resemblance to that of India and of the Pacific than it has to that of South or of West Africa. Similarly, the fauna of the north-east coast of the United States has more species in common with that of the western shores of Europe than with that of the West Indies. Ortmann, in his ‘Grundziige der MarinenTiergeographie’, divided the coastal waters of the globe, down to 100 fathoms, into a Fig. 176. A Sea-Cucumber or Trepang carrying young ones on Its back. Natural size 292 Animals oj the Sea Floor number of faunistic regions which may be briefly men- tioned here. 1 . The Arctic Region, again subdivided into : a. A Circumpolar subregion, which includes Green- land, the north coast of Europe and Asia, and the north and north-east coasts of America; h. An Atlantic Boreal subregion, which includes the west coast of Europe ; and c. A Pacific Boreal subregion, which includes the northern shores of the Pacific Ocean from Japan to California. The tropical parts of the globe are divided into four regions : 2. The Indo-Pacific Region, a very large area stretch- ing from the east coast of Africa to the Islands of Oceania, and including the shores of India, the East Indies, China, and most of Australia. 3. The West American Region, which only includes the small stretch of coast line from California to the south of Ecuador. 4. The East American Region, from Newfoundland to Cape Hatteras in the United States. 5. The West African Region, which is subdivided into a, A Mediterranean subregion ; and b. A Guinea subregion. The whole of the Antarctic is included in a single region : 6. The Antarctic Region, which includes the south coast of Australia, the coasts of Cape Colony and German South-West Africa, and the greater part of South America, from the River Plate on the east to Ecuador on the west (this northern extension of the cold region being caused, as has been alluded to above, by the current of cold water running up the west coast of South America). 294 Animals of the Sea Floor The definition of these regions must not be taken as implying that there is any strict demarcation between their faunae. There is naturally a great deal of overlapping, and very many species are common to two or more regions; for example, a large number of the common marine animals of the British Isles are to be found in the Mediter- ranean, often at increased depths. A more detailed subdivision of the seas, on much the same lines as that of Ortmann, but including oceanic bottom areas, and suitable also for planktonic areas, has been adopted by the Challenger Society for bibliographical work, and is reprinted here as Chart XI (p. 304). The doctrine of ‘bipolarity’ (that the arctic and antarctic faunae are more closely allied to each other than they are to the inhabitants of the more equatorial waters) is based on the fact that identical or closely allied forms have been found in both north and south Polar seas, but have not been reported from the interven- ing waters. The number of undoubted cases of this is not large, and most of them may perhaps be explained as being instances of species which belong to a widespread deep-water fauna, which have been taken at the Poles at moderate depths, but have escaped notice in the abysses of the warmer regions where our knowledge of the deep- sea fauna is very limited. The local distribution of the fauna within the narrower limits of each region appears to depend largely on the varying character of the sea bottom. Within the limits of the zone which we are considering (from the lower limit of the Laminarian Zone to the edge of the Continental Shelf) it may be said that the differences of fauna due to depth are insignificant as compared with those associated with the different kinds of bottom, and therefore it is impossible to attach importance to the subdivisions of this Fig. 178. WiUemoesia, a Deep-Sea Lobster, of the Family Ervoniclac Natural size 296 Animals of the Sea Floor zone which have been proposed. The types of sea bottom usually met with may be arranged in a descending series, beginning with rock and passing through stones, gravel, sand and gravel, pure sand and muddy sand, to mud, varied by the occurrence of beds of dead shells or of the calcareous alga usually known as ‘coral’ (Lithothamnium). In any given district it is usual to find that each of the above-named types of bottom has its own special fauna, and that many species seem to recognize a more minute subdivision than man, whose observations are necessarily made from a distance, can appreciate. The reason for the preference for one type of bottom rather than another usually resolves itself into a question of support or of shelter, either for the species itself or for its prey. Of course, there are very many species that may be found indifferently on all sorts of bottom and under all kinds of conditions. The connexion between the nature of the bottom and its inhabitants may be illustrated by an example. In some spots on the west side of the Irish Sea, when a haul of a trawl is made, at least 90 per cent, of the catch consists of the Alcyonarian or ‘false coral’, known as ‘Dead Men’s Fingers’ (Alcyoniumdigitatum), while on the surrounding grounds only a few specimens are to be met with. The reason of this is clear. These spots where Alcyonium flourishes are beds of dead shells, mostly scallops and spiny cockles, which afford a foothold for the swimming larva of the Alcyonarian when seeking a place on which to settle down. Incidentally, the local range of at least three other species is marked out at the same time; two Nudibranchs which feed on Alcyonium are always to be found in the neighbourhood of these shell beds, and also a small Isopod Crustacean which lives on the same species. In the case of organisms which are free to move about. 298 Animals oj the Sea Floor the reasons for one kind of ground being more suitable than another are not so evident. The Norway lobster, or Dublin Bay prawn (Nephrops norvegicus), which abounds on the mud covering the bottom of a great part of the Irish Sea, rarely ventures on to the adjacent sandy grounds; while the masked crab (Corystes), which frequents a clean sandy bottom, is never found on the mud. The circular crab (Atelecyclus) is not found on either sand or mud, but should be looked for in coarse sandy gravel. In these instances the animals find shelter by burrowing beneath the surface (though in the case of the Norway lobster this burrowing habit must be inferred, as it does not seem to have been established by direct observation), and the medium chosen is probably the one best suited to their excavating powers, or breathing apparatus. The fauna of an area of soft mud affords many instances of animals, either specially modified to meet their peculiar conditions of life, or else adapting themselves to their cir- cumstances without any special structural alteration. Leaving aside those species which are capable of burrow- ing beneath the surface, we notice that the peculiarities of a mud fauna are due mainly to the absence of any solid points of attachment or support. This relationship between animals and their surround- ings and between different animals inhabiting the same type of ground has in recent years attracted a good deal of attention and has been studied in great detail by Dr. C. G. J. Petersen, the Danish zoologist, and his followers. Dr. Petersen has designed an efficient grab dredge, which lifts from the bottom- the mud or sand of a measured area with its contained fauna, and from the data furnished by a number of grabs the average distribu- tion per square metre can be estimated for the contained animals, both those living on the bottom and those which FiC 1853 Terebratula, a Lamp- shell Enlaiged '-/'ivy T Fig 1 85 Chiton 300 Animals oj the Sea Floor burrow not too deeply beneath it. By the aid of this instrument various communities of associated animals, usually, though not always, living at similar depths in or on soil of the same texture, have been defined, e.g. the community characterized by the echinoderms Bryssopsis lyrifera and Amphiura chiajii, living along with several characteristic molluscs in soft mud in deep water, or the Echinocardium cordatum — ^Amphiura filiformis commu- nity, living in coarser soil at lesser depths. This allotment of species to communities has been found to hold good with no great modifications in the North Sea, the Baltic, and the adjoining waters, but no doubt when the investiga- tions are extended to more distant seas other species with similar habits will be found to replace those already defined for a limited area. The Petersen Grab has also proved very useful in making clear the life stories of several common species and in showing how precarious a hold some of them have on existence. Mr. F. M. Davis, in his exploration of the fauna of the Dogger Bank, found that Spisula subtruncata, a bivalve mollusc which is the main food of the plaice on the rich feeding-grounds of the bank, forms there beds of vast extent in the shallower waters. Each bed is composed of individuals of the same age, and when first formed is densely populated, but in a few years is almost wiped out by the attacks of fish and other predaceous animals. Fresh beds are in the meantime formed by the settling down of the clouds of spat emitted from the older molluscs. The spat drifts for a week or so at the mercy of the current and then settles down wherever it happens to be. It is easy to see that some unusual set of the current might drift the spat from the bank to the surrounding deep water, where it is probable that it would perish and the bank would be left unstocked by that year’s generation. Fig. iS 8. Cellularia, one of the Polyzoa. x z Fig. 189. Cellularia. Mag-niiied 302 Animals of the Sea Floor The failure from time to time of the Ceylon Pearl Oyster fishery on the Paars or shallow rocky banks in the Gulf of Manaar seems to have a similar explanation. The Pearl Oyster, in reality a kind of mussel, needs a clean rock for attachment, and unless a good swarm of spat should happen to be in the water over the Paars when the time for settling down is reached, a failure of the fishing some years later will naturally result. The Deep Sea Shortly after the edge of the Continental Shelf is passed, we begin to meet with members of the true deep-sea fauna, which peoples the abysses of the ocean. With the rapidly increasing depth the proportion of shallow-water forms becomes less and less, till at a depth of i,ooo fathoms there is hardly a species left of the abundant life of the Continental Shelf. Their place has been taken by a fresh assemblage of animals, which, as a whole, present a very different appearance. If, however, we take into account the very much altered conditions under which they exist, these differences are perhaps not so great as might have been expected. In fact, except amongst the fishes, and, perhaps, some of the sea-cucumbers and Crustaceans, there is hardly an inhabitant of the deep sea which differs so markedly from its nearest allies which live in shallower water, that, were its origin unknown, it could be said without hesitation to be an inhabitant of the abyss. The physical conditions prevailing over the bottom of the deep sea are almost uniform, so far as we can perceive, through long periods of time and throughout all oceans. The darkness is absolute, or broken only by gleams of light from ‘phosphorescent’ animals. The temperature is very low, at most 3° or 4° F. above freezing, and constant. The pressure is enormous, upwards of 3 tons per square Deep Sea Fauna 303 inch at a depth of 3,000 fathoms. It is hardly necessary to point out that, once the tissues of an organism were adjusted to meet this pressure, it would be felt no more than is the pressure of the atmosphere by the inhabitants of the earth’s surface. The effect of this pressure, com- bined with the low temperature, on the metabolism of the animal (the chemical changes connected with its vital pro- cesses), must, however, be very great. The water above the bottom is still or in very slow movement (unless, per- haps, in some narrow straits), and is not rendered turbid by any river sediment or by mud stirred up by currents from the bottom. The bottom itself is covered with a fine ooze, usually without rocks or other solid bodies to afford support to sedentary organisms. In the absence of any vegetable growth the inhabitants of the deep sea must be carnivorous, and as it is not possible for them to sustain life by feeding mutually on each other, we must look for some external source of supply. This is to be found in the constant rain of dead organisms which have lived and perished in the upper layers of the ocean. The existence of this constant supply of organic matter is illustrated by the fact that the oozes which cover the ocean bed over a large part of its extent (Chapter VI) are formed from the shells or skeletons of the microscopic Radiolarians, Foraminifera, or Diatoms, which swarm in the waters above. This does not mean that the whole of the food supply consists of microscopic matter, for in the upper waters representatives of almost every order of marine animals, from whales downwards, may be found. Very little is known of the animal life in the deepest parts of the ocean floor, but the few hauls which have been made in depths of over 2,000 fathoms show that the fauna is comparatively poor both in species and individuals as compared with that to be found at about 1,500 fathoms. 304 Animals oj the Sea Floor where a rich invertebrate fauna, in which Echinoderms are predominant, may be trawled, though fishes are scarce. This is shown clearly by the results of the Michael Sars Atlantic Expedition, and the Challenger had a similar experience. In the early days of deep-sea exploration it was be- lieved by many naturalists that the unknown depths of the ocean would be found to harbour survivors of an earlier geological epoch which would fill gaps in the known, pedigree pi the Animal Kingdom. This expectation has not, save tA^a few instances, been realized, for, although an immens amber of deep-sea species have been dis- covered, they are, on the whole, either very like those from shallow water or else have been specially modified in adaptation to the changed conditions of life. On account of this general resemblance of the abyssal fauna to the animals inhabiting shallow water at the present time, rather than to those of a remote geological period, it is generally believed that the inhabitants of the deep sea have migrated thither at some comparatively recent geo- logical age, the migration probably beginning at the close of the Mesozoic Period, and continuing to the present day. It is hard to point out any features in which, as a whole, the animals of the deep sea differ from those of more shallow waters. Their colours are, as a rule, peculiar. Fishes are usually black, while Crustacea are mostly red or pink. The Echinoderms show a great range of colour, from yellow through orange and red to crimson antj purple. Uniform colours are the rule, rather than blotches or stripes, such as are often found in shallow water. These markings may perhaps be regarded as having a protective or warning significance, which would be of no advantage to an animal living in utter darkness. Many of the animals Z n E U m ^ C a « w bo Ji "O ^ C U E 0-1 Cl o E CL «j ■AREAS OF MARINE DISTRIBUTION Deep Sea Fauna 305 themselves are blind, and probably burrow in the soft ooze, or live a sedentary life, clinging to hydroids, false corals, or sponges. Others have large eyes especially adapted — at least, in some cases — for the perception of faint gleams of light. These are probably more active and predatory types, guided to their prey by its own luminosity, or sometimes having themselves luminous organs, which serve as veritable searchlights. While the deep-sea fauna, as a whole, is by no means of a primitive type, there are a few cases where it does seem that animals of an antique type have found in the deep sea a refuge from the more strenuous competition of the thickly inhabited shallow waters. This is notably the case with some Crustacea. The Eryonidae (Fig. 178), a family of flattened lobster-like animals, are the living representatives of a group of which the remains are abundant in the rocks of the Triassic and Jurassic Epochs; the Homolodromiidae of the deep sea appear to be the most primitive of the existing crabs, and resemble some found as fossils in Secondary Rocks ; and certain deep-sea prawns (belonging to the families Penaeidae and Acan- thephyridae) are probably of more ancient types than their relatives in shallow water. Finally, the extreme imperfection of our knowledge of the fauna of the sea bottom in nearly all regions of the world cannot be too strongly emphasized. Even in shal- low waters the smaller animals are almost unknown, ex- cept in parts of the North Atlantic and Mediterranean; and those who have the opportunity of dredging in depths exceeding 100 fathoms in any part of the world may be assured that any animals which they may find are likely to be of interest and value to specialists. The methods of dredging and trawling are explained in Chapter IX of this volume, and we need only mention here the importance 3o6 Animals oj the Sea Floor of carefully sifting the dredged material. The con- tents of the finest sieves should be preserved in bulk for subsequent examination in the laboratory. It is astonish- ing what wealth of living things may often be found by this method in mud which seems at first sight to be barren of life. It need hardly be said that the precise locality of each haul should be carefully noted, and, what is even more important, the exact depth and temperature recorded where possible. A rare opportunity of obtaining specimens from the greatest depths of the oceans is afforded in the course of repairing deep-sea telegraph cables. These are often encrusted with sponges, corals, barnacles, and other or- ganisms, while Crustacea, Mollusca, and Echinoderms are brought up clinging to them. The preservation of such specimens would yield much information regarding the fauna at depths which are otherwise only to be reached by costly and elaborately equipped expeditions. VIII EQUIPMENT BY S. W. KEMP AND THE EDITOR Practically all oceanographic work, except heavy dredg- ing and trawling at great depths, can be done from a sailing yacht, if furnished with a little steam capstan, such as is fitted to sailing trawlers. In a sailing yacht without any power whatever, the col- lection of water samples, plankton work down to at least 500 fathoms, and dredging and trawling in shallow water can be carried o/i without difficulty. But in a sailing yacht, whether with or without a cap- stan, the rig, tonnage, freeboard, and deck space are so important that it would be well to inquire beforehand from experts what branches of the work could be most suitably pursued. With power available for propulsion as well as hauling, oceanic work will be done more easily, and more can be got through in the same time. It is not only true that no time is wasted in tacking between stations, but the handling of the ship when at work is much facilitated. In using most deep-sea instruments, especially if driven by messengers (weights which slide down the rope), it is most important that the rope should be kept vertically up and down, and not allowed to stream away as the ship drifts. With plenty of steam on board, the capstan can be replaced by a steam winch or windlass, with the result that work can be carried on at far greater depths, and far more rapidly and safely. Supposing that there is no deck engine already on X z 3o8 'Equipment board, the owner will find that a good winch can be utilized for practically every kind of work — thermo- meters, water-bottles, soundings, tow-nets, dredges, and trawls, as well as for hoisting boats and anchor, cargo and coaling work. With winches (as with much other gear), the owner, master, and engineer must settle the exact specification, according to the work to be done. We can only give the most general indications as a guide. One may use (i) either a regular trawling winch, on the barrels of which the rope is stored ; (2) or a cargo winch, which will haul the rope but not store it. Of these two types, there is no doubt that the regular traw'ling winch is much to be preferred. It will not only serve for the or- dinary work of a cargo winch, but when scientific work is required it will save deck space and hands. We figure a commercial trawling winch, showing the usual two hauling and storing barrels in the centre of the engine. For really deep-water work this type requires some modification. In the first place, if required for trawling and dredging only, a single large barrel is better than two smaller barrels, the divisions between the latter having been known to burst and cut the rope under pres- sure. In the second place the brake wheels should be cast solid with the barrel ends, and strengthened by radiating ribs, in order to resist the enormous lateral thrust of a great length of rope wound under tension. But if it is pro- posed to use closing tow-nets and other apparatus with small messengers, which require a slighter rope (p. 234), in addition to trawl and dredge, it might be well to retain the double barrel, using one barrel for each rope, and strengthening the division between them. In this case the barrels should be clutched separately, which is not always the case with commercial trawling winches. A cargo winch can be used for all deep-water work if it Equipment 309 carries a largish, deeply grooved drum for winding, of (say) 18 or 24 in. diameter, in addition to the ordinary small barrels. This drum will only carry two or three turns, and the wire must be taken up, as it comes off the Fic. 190. Steam Capstan for use with Sailing’ Yachts: suitable for the largest Trawls drum, on to a storage reel of some kind. It is no small labour to reel up 1,000 fathoms or so of wire rope by hand power, and as a rule it cannot be taken up as fast as the winch can deliver it; the method is therefore wasteful of both time and labour, and the storage reel should, if pos- sible, be driven mechanically. If a fairly powerful cargo 310 Equipment winch is already on board, which it is not desired to dis- place, we should recommend that a one or two horse- power steam engine or motor should be fixed behind the winch, so that it could drive whichever storage reel was required at the moment. Another plan which has been suggested is to drive the storage reel by an endless rope (or swifter) from the small barrel of the winch, the rope being kept taut by passing through a block against which a weight pulls. When the reel tends to run too fast, the rope-drive will slip and so provide the necessary com- pensation. It should be remembered that with a cargo winch there is much more wear on the warp than with the trawling winch. In using either of the above methods care should be taken to reduce the slip on the driving drum to a minimum. Storage reels for the above purposes should be of extra solid construction and provided with a powerful hand brake. Most of the patterns on the market are only in- tended for occasional use, and are consequently of the very lightest build, with inferior bearings. A. Deep Sea I. DREDGING AND TRAWLING As this is the heaviest and in many respects the most difficult part of oceanic work, it should be the first to be considered. Next in importance to the winch is the wire rope to be employed; this must be flexible, lest it kink and snap when the strain is taken oflF, as, for example, when the trawl touches bottom. Flexibility implies several strands, each of numerous wires, and this unfortunately involves weight. Fig. 192 shows a section of a suitable rope, and Fig 191 Commercial Trawling Winch 312 Equipment displays six strands of twenty-four wires each (6x24); each strand has a hemp core, and the whole rope a hemp heart. Before selecting a rope, it is of great importance to select a manufacturer with care; it is not every firm which understands the special demands made upon a rope by deep-sea trawling. The selection of a rope will, of course, depend largely upon the depth to which it is proposed to work, the pro- bable character of the bottom, and the size of the dredge or trawl. If it is proposed to work down to but not over 1,000 fathoms, a single length of 1,500 fathoms of I -in. rope, with breaking-strain about 3 tons,' would probably serve the turn. Work at anything over r,ooo fathoms implies an ‘expedition’ rather Fig. 192. Section of than a yachting cruise, but for work Wire Rope _ , , - to 2,500 fathoms the following com- bination may be suggested: 1,500 fathoms of i-in. rope at the trawl end, 1,000 fathoms of li-in., and 500 fathoms of I i-in. at the winch end, all shackled together;' a special shackle is made for this work. The object of this com- bination is to strengthen the rope where the heaviest strain comes, and the total result is as follows : I It may be mentioned here that H.M.S. Challenger used hemp ropes of 2 in. and 2J in. circumference, with breaking-strains of i ton 12 cwt. and 2 tons 6 cwt. respectively. For shallow-water work was employed a rope of 3 in. circum- ference, with breaking-strain of 2 tons n cwt. But it must be remembered that in water a hemp rope carries much more of its own weight than a steel rope. Sir John Murray informs us that the rope was hauled at the rate of 1,000 fathoms per hour, and that during the three and a half years’ cruise, when trawlings and dredgings were made at 354 stations, there were only eleven cases of parting of the rope. Wire ropes were first used for such work by Agassiz in 1877. > Owing to the shackles (or splices) such a rope could not be used below 1,500 fathoms for closing-nets or similar apparatus worked by messengers. Wire Rope 313 Weight Breaking-strain At sheave . ton I cwt. 17 tons S cwt. 0 Above first shackle 1 7 !! 0 Below „ . . . I 7 4 14 Above second shackle . 0 13 4 14 Below „ „ . . . 0 13 3 II Whether the rope be stored on the winch-barrel or on a separate reel, it should be carefully and evenly spread as it comes in. In hauling heavy gear, especially where the warp is wound directly under considerable tension, the spreading is best done by means of a mechanical spreader ; this apparatus is sometimes supplied by the manufacturers of trawling-winches. In some patterns the wire is spread by a long steel fork, which moves, either mechanically or by means of a handle fitted to a long pivoted arm, across the face of the barrel. In this type the friction on the warp when hauling heavy weights would be very destructive; a carrier holding three rollers (two upright and one hori- zontal) should be substituted. In a simple type, which one of us is able to recommend, the carrier and rollers are made to travel on a slide by the rotation of a long screw- shaft, actuated by a handle from one side. The work can be done with more labour by using a snatchblock on a lanyard or spar; but while this method will be found efficient for light work, or for reeling wire that is not under tension, it is scarcely applicable when hauling heavy gear, for in such cases the strain is so great that the united efforts of two men and a ‘purchase’ will be required. Greasing is also an essential to the preservation of wire rope. As it comes in, it should pass through well-oiled cotton-waste. A mixture of equal parts of tallow and castor oil, applied warm, is strongly recommended, or one of the special oils supplied by the rope manufacturers ; the 314 Equipment ordinary fluid machine oils seem to disappear at once, and should be renewed daily if used at all. It should be stipulated with the makers that all wire ropes be sent properly coiled on a reel of some kind. As soon as received, if the rope is to go on the trawling-winch, it should be paid on to the winch barrel under a steady strain with the reel revolving; under no circumstances should it be paid on with the reel flat, or from a coil. Be- fore being used for trawling, the whole length should be run out at sea, and carefully reeled up under a fairly high strain. Two other pieces of gear are advisable — an accumu- lator and a recording sheave. The old type of accumulator consisted of a number of stout cylindrical rubber straps between two boards, the whole slung from a derrick or spar, and carrying the block over which the trawl rope runs out-board; the object of this was to lessen any sudden strain on the rope due to the trawl hitching, and to warn the man in charge of such strain. Nowadays the less bulky and more reliable steel spring accumulators are to be pre- ferred. If trawling from a derrick, the topping lift, passing over a block stropped to the mast, is brought down and made fast to the head of a long steel spring, rigged up against the mast. * A sudden strain on the rope pulls down the derrick head, the spring yielding. Such an indication will often enable the gear to be recovered before it is seriously damaged, while with very severe pitches the warp is (probably) saved from snapping. For really heavy trawling a nipper may be substituted for the accumulator, or used in conjunction with it. This nipper (Fig. 193) is attached by means of a hemp rope to the winch or some other solid deck fixture, and the • This arrangement is figured in Report of the Siboga Expedition, ‘Descrip- tion of the Ship’, by Lieutenant G. F, Tydeman, Plate I. Nipper jar Wire Rope 315 slack tripping line is also shackled to the same support; the accompanying figures will explain the method employed. Fic. 194. Small Nipper Rendering Fig. 195. Small Nipper Gripping Any undue strain causes the hemp rope to break, allowing the wire to pay out from the drum or reel (if the brake is left open), while the nipper is checked by the tripping 3 1 6 'Equipment line, and either falls to the deck or allows the wire to render through it, according to the particular make fitted. A small nipper will also be found useful for holding the warp in the event of anything going wrong; Figs. 194, 19^ illustrate a simple form which may be employed for this purpose. A recording sheave (or meter wheel) of some kind is necessary for work in deep water. This need only consist of a strong sheave of known circumference (as i meter), to the spindle of which is affixed an ordinary engine-room counter, but the specially made patterns with roller bear- ings and more durable recording mechanism are prefer- able. For working heavy vertical or mid-water nets this sheave may be bolted to the deck, but for work of a lighter character it may be slung from a davit in place of the ordinary snatchblock. For bottom trawling a recording sheave can often be dispensed with, for a sufficiently good estimate of the amount of warp paid out can usually be made by eye. When, however, long lengths of heavy wire rope are to be used for various purposes, it will often be convenient to mount a counter on a large revolving bol- lard securely bolted to the deck; with this arrangement shackles and bridles can, if desired, be wound on to the winch-barrel. The revolving bollard should be fitted with roller bearings and should be made of specially hardened steel. All rollers and sheaves should be brass-bushed if pos- sible, and they should be kept drenched in oil; it is as- tonishing how soon oil disappears when i ,000 fathoms of rope are being paid out, and what disaster attends a block jammed for want of it. ( 31 '- ) 2 . SOUNDING A sounding should always be taken before any gear is shot. In suiweying vessels the deep-sea sounding machine is generally rigged in the bows; it can be driven by a swifter from the deck engine. Some forms of deep-sea sounding machines — for example, the Lucas (used by the Hydrographic OtRce), the Sigsbee, and the Le Blanc — can be driven by their own little engines. For deep soundings single-strand piano-wire is nearly always used, but it is veni' stifi. and liable to kink and snap if not checked the moment that the lead touches bottom. To get good results a considerable amount of practice is necessar}' even vith the best mauhines. In the Lucas machines (both for deep and shallow watep the bt-sheave is mounted on a niovable arm which actuates a band-brake when it is eletated ''see Fig. 19"}. This arm :s held in position by Pro springs and the tension of these can be regulated by a screv. When sound" ng, the tension is slowly increased to balance the weight out-board, and ^sith experience in manipulation the machine can be so regulated that it stops automatically as scon as bottom is reached. For really deep-sea sounding it is advisable to use a heavy sinker of at least 50 lb., which is detached on strik- ing bottom and left there; this methoU enables the wire to be hauled rapidly, and without danger either of breaking the wire or of bursting the barrel on which it is reeled up. The sinkers are made of cast iron. Several types of deep-sea sounding leads are in use, their form being dictated by the fitting employed to bring up a sample of the bottom (cf. Chapter VI, ‘The Sea Floor’). For the latter purpose have been employed a cup (not now in use) ; a heavy tube, which, when being hauled, 3 1 8 'Equipment is closed below, either by a butterfly valve or a cock ; and a snapper, consisting of two tightly fitting buckets which grip a bottom sample. The second of these is represented in Fig. 151, p- 243, the third in Fig. 196. The wire should be well greased. Very great care should be exercised in spreading the sounding wire evenly on the barrel, and the same precautions taken in paying it on to the barrel from the manufacturer’s reel, as in the case of wire rope. Fig. 196. Sounding-Leads, with Snapper for Collecting Bottom Sample. (Lucas Snappers.) The lower lead has a weight which is detached on striking the bottom 3. OTHER WORK It is not economical either of time, labour, or the life of a rope, to use a heavy rope where a lighter one would do the work, and a i-inch rope is not well adapted for other uses than trawling, dredging, and big tow-nets. A lighter and cheaper type of rope can be used for re- versing thermometers, current meters, water-bottles, and tow-nets up to 2 ft. in diameter. A suitable rope for such work would be 4 mm. in diameter, composed of six strands each containing seven wires, with a breaking strain of 19I cwt. and a weight of 3-7 lb. per 100 ft. (in air). By reducing the number of wires a much stronger rope of the same diameter can be spun, but it will lack flexibility. ( 319 ) B. Shallow Water The hints already given relate to the most difficult and troublesome part of all oceanographic work, that which involves the most power, time, weight, storage-space, skill, care, and expense. If deep-sea work is the most ambitious, it is also the most interesting; if it is the most difficult, it is also the most needed, and the most productive of striking results. On the other hand, excellent and valuable work can be done at small depths in almost any part of the world. Even in home waters little has been done systematically outside the lOO-fathom line; beyond home waters (except in one or two favourite areas) any one with a rowing-boat and a little apparatus can add materially to knowledge. I. DREDGING AND TRAWLING In shallow water, down to (say) 200 fathoms near land, it must be remembered that if the net catch on a rocky bottom, a breeze acting on the freeboard of a large yacht or steamer will impose a very great strain on the rope; a trawl full of stones or mud is also a menacing possibility. For such work a i j-inch or ij-inch rope, with a breaking- strain of 4 to 8 tons, is desirable ; the lifting power of the deck engine can be somewhat reduced, but an accumulator or some similar protection is certainly required. With this gear a yacht can work in as little as 3 or 4 fathoms, if the water is known to be clear of dangers. But in uncharted waters near shore, and particularly in coral seas, the owner will naturally not risk his yacht close in shore. A small sailing boat or steam pinnace, using good tarred i^-inch hemp rope, hauled and coiled by hand, can easily work an 8 to 10 foot beam trawl, or a fairly heavy dredge; even a yacht’s punt can work a small dredge. Very excellent 320 Equipment work round foreign ports can be done in this most un- ambitious way by any one. One of the most distinguished marine naturalists used a light hand dredge about lo inches long, which he cast like a lead as far as possible, and hauled at once. 2. SOUNDING AND OTHER WORK In Fig. 197 is shown a small sounding machine, with capacity of 200 fathoms of piano wire, 22 gauge. A 20- pound sinker, not detaching its weight, carrying a snapper or a valved tube, will suffice. A small sounding machine could be utilized with a fine stranded wire for thermometers, light water- bottles, current meters, &c., down to 200 fathoms (if .the barrel will take so much) ; but valuable instruments should not be trusted to piano wire in a small ship with a short roll. A cheaper form of reel can be knocked up in plain galvanized iron. For some work it is enough to mark the wire by whippings, but if messengers are to be used this marking will not do, and a recording sheave of some kind will be needed; an excellent little sheave is made for the International Com- mission for the study of the North Sea, which records the length of wire out in metres. The reel or sounder can be fixed close to the ship’s rail, or if fixed amidships the wire can be led over the recording sheave slung from a davit or from a spar, which can be swung inboard when the gear is up, in order to enable the instruments to be conveniently attached and detached. Lighting. If there is a dynamo on board, a ‘cargo cluster’, with enamelled iron shade and flexible connexion, is very useful for working at night. Heavy nets should never be shot at night until the whole crew is thoroughly well practised in the use of the gear ; a man can easily lose Ship' s Laboratory 321 his life by getting a turn of the warp or spans round his leg. A portable ‘electric torch’ is very useful for reading the log, &c., at night. Fig. 197. Lucds Sounding Machine, hand-driven C. Laboratory Fitting Some sort of place will have to be used as a laboratory, where the scientific log can be written, collections pre- served, labelled, and packed, &c.; so that a few notes may be here added on fitting it up as a laboratory. It is perfectly possible, by a little forethought, to do solid work, and yet to keep the yacht as spick and span as any owner can desire. The position of the laboratory and its fittings will de- pend entirely on the rig and build of the vessel. It should be amidships and on deck, if possible. It cannot have too much light. A poop deck-house makes an excellent laboratory. A very important matter is the work-table ; this can be either fixed or swinging, and both have their supporters. A stout table, of the kitchen type, with a well-oiled top, 3166 Y 322 Equipment and a large drawer for charts and flat papers, and firmly bolted down, has this advantage over the swinging pattern — that the space below can be fitted up with pigeon-holes of various sizes to take apparatus and accessories of all kinds. Half at least of the surface should be divided up permanently, by fiddles about 3 in. high, into compartments suitable for the bottles or other apparatus required. If this is the only table in the laboratory, the other half should be left clear for writing, charting, drawing, &c., but a separate table, for writing and for microscopic work, is a great boon. If a swinging-table be preferred, a stout rail, level with the table and about I2 in. clear of its edge, fixed at the corners to stanchions from the floor, will bring up any one lurching against it in a heavy sea. If there is a dynamo on board, a cluster of lamps should be over the centre of the table; but a hanging oil-lamp should not be arranged to throw a shadow over the writing-table. The particular work which it is proposed to attempt will, of course, determine the character of the bottles and apparatus to be kept in the laboratory; but preserving jars, and glass tubes, are needed in all kinds of zoological and botanical work, and should be racked in quantities round the walls. Large bottles of killing and preserving fluids should be racked together. A 2-gallon milk-can, with tap below, and a filter of fine boulting-cloth at the top, is handy for the necessary store of clean water, whether salt or fresh (ship’s fresh water not being ideally clean); it should be emptied and refilled every day. An enamelled slop-pail, with a top, should also be racked in a handy position to receive mess and waste fluids of all sorts. I. Plankton. Special receptacles in the fiddles of the Ship's Laboratory 323 work-table should be made to take two or three jars or deep dishes into which the contents of the tow- net tin are emptied for sorting. Other compartments should take those smaller bottles in which the larger organisms are to be pickled. Empty the ‘tin’ into the jar or dish with plenty of filtered salt water, pick out the larger specimens with lifter and pipette, and pickle them with as much elaboration as time allows (see Chapter XI, ‘Preservation’). Add a few drops of undiluted formalin to the jars, and stir the contents gently till they are mori- bund ; they are then allowed to settle. Pour off superfluous water, and transfer them to a wide-mouthed, smaller bottle (say 16 oz.), and fill up with 2 per cent, formalin. Later on, pour this off, and transfer the catch to the storage bottle, and fill up with 7 per cent, formalin. The use of a compound microscope for live material is usually impossible, except in harbour, but a little ‘dissect- ing microscope’, with lenses multiplying 6 and 10, or 8 and 1 8 diameters, is extremely useful •, it can be clamped to a fixed table, or swing in gimbals. 2. Dredging and Trawling. In this somewhat messier work the contents should be emptied on to an old sail, to save the deck. It is advantageous to lay this over a fish- fiddle (say 4x3 ft.), fixed, when wanted, against a hatch, and to run a coaming 2 in. high round the hatch, which can then be used as a table. When the material obtained by the trawl or dredge has been deposited on deck, an immediate effort must be made to sort the heterogeneous assemblage of animals which is revealed. Scarce or fragile specimens should be trans- ferred at once to glass jars or enamelled dishes. Fish may be placed in baskets, and the commoner invertebrates similarly treated, or stacked in an enamelled fish-kettle or slop-pail in sea-water. 324 Equipment Fish should all be measured to the nearest centimetre, and entered in the scrap log-book along with an enumera- tion of the invertebrates. Unless the identity of the species is established beyond all doubt, samples should be preserved. It is evident that numbers of indeterminable forms will be found; these may be entered as Gadoid A, Asteroid A, &c. The same species occurring in subse- quent hauls should be referred to in the same way, but unless the identity of the two forms is certain, a sample must again be preserved. When possible the stomachs of fish should be ex- amined, for, apart from the light which is thus thrown on the food supply of the species, they occasionally contain invertebrates which are seldom or never found in any other way. Search should be made for hermit crabs and worms resident in the dead shells of molluscs, and for encrusting animals living on stones or other debris. As a rule it is best to avoid attempts at extraction or removal. The comparatively large animals, obtained in bags of sprat meshing, may be dealt with in the same way, but the sorting of the material collected by the fine-meshed nets usually demands so much time that immediate treatment is out of the question. In such cases it is best to transfer the whole to a jar containing 5 per cent, formalin, until a more favourable occasion arises. It should be remem- bered that formalin, even in a 5 per cent, solution, is sufficiently acid to act as a decalcifying agent; all animals which are liable to injury from this cause should be trans- ferred as soon as possible to 70 or 75 per cent, alcohol. The sand, mud, or ooze, which is sometimes brought up in considerable quantity by the trawl or dredge, generally contains a considerable number of animals. For the treatment of this material sieves must be employed; Chemical and Physical Wdrk 325 the process is rather laborious, but the results are as a rule eminently satisfactory. It is best to obtain a set of four sieves (with different sized meshes), which may be em- ployed successively, using the coarsest first. The sieves should be made of brass wire gauze, and are more con- venient if fitted with handles (Fig. 152). The coarsest should have meshes about I inch square, the finer pat- terns with eight, sixteen, and thirty-two meshes to the inch. It is advisable to preserve a sample of the ooze in alcohol, for many minute organisms will pass through even the smallest of the above meshes. A stream of water from a service pipe may be turned into the sieve, or it may be dipped repeatedly in a bucket of water. Shaking is very destructive and should be avoided. Samples of stones, sand, mud, or ooze should always be preserved ; small canvas bags are useful for this purpose. 3. Much valuable work on the chemical and physical conditions of the sea and the nature of the ocean currents can be carried out by quite small vessels. The auxiliary ketch Armaur Hamen, of only 60 tons gross, has made several summer cruises in the north-eastern Atlantic be- tween Norway and the Azores; from the data of tempera- ture and salinity of the water w'hich were collected at various depths the current system of this ocean has been mapped. The necessary equipment for this work is relatively simple — ^water-bottles, thermometers, meter- wheel and a drum of fine wire with some motive power for reeling it in, since it is desirable to be able to obtain samples down to a depth of 1,000 meters or more. The analysis of the collected samples is usually carried out on shore. Much could be added to the existing sum of our knowledge of ocean currents by a vessel so equipped, which had time at her disposal to stop for one or two hours at a series of positions in the open ocean. 326 'Equipment 4. Storage of Specimens at Sea.* The larger and rougher specimens obtained by dredge or trawl, after the necessary preservation (Chapter XI), should have their label tied on to them, and then be wrapped in cheese-cloth and stored in spirit or formalin. It is handy to have on deck for this purpose an enamelled tank, or one made of wood and lined with lead (zinc tends to pit with formalin). The lid should fit on to a thick rubber flange, and may be held down by wedges driven under the cross-stays of the tank. Trays of lath inside serve to distribute the pressure of the upper layers of specimens. A useful size of tank is 3 x i j x li ft. high. On a long cruise the contents of this can be occasionally transferred to barrels and headed up; to be sent home or stored in the hold. Large but more delicate specimens can be stored in stoneware jars with very wide mouths ; these are made in various sizes, and the lid is screwed down tightly on to a rubber flange by an iron clamp and thumbscrew. Two 4-gallon jars of this type, racked in a padded box, since they are brittle, will hold a very large quantity of material. Sweetmeat jars, with a cork ring, are not to be re- commended for prolonged storage where they cannot be constantly examined, unless tied over with bladder. Many kinds of fruit-preserving jar are on the market, ranging in capacity from 1 lb. to 7 lb., and some of these are very well adapted for the storage of specimens. The best have a glass lid resting on a rubber washer and kept tight by a screw-flange. This flange must be kept greased or it will rust rapidly. Wide-mouthed bottles (4, 8, and 1 2 oz.), fitted with corks, are often used for small speci- mens and plankton, but the cork, even though sealed with wax, is always a source of danger. Plankton may be * For further information sec pp. 404-9. Storage oj Specimens 327 preserved in small jars of the type mentioned above, and small specimens are best placed in unshouldered glass tubes. These tubes should be filled to the top with pre- serving fluid, and plugged (without leaving an air-bubble) with balls of cotton-wool wrapped in tissue paper. They may then be stored together in a large jar, also full of fluid. IX DREDGING, TRAWLING, AND OTHER METHODS OF FISHING ‘ BY E. J. ALLEN AND S. W. KEMP Dredging Dredges are suitable for the biological investigation of all grounds that are too rough to admit of trawling; they are very serviceable for obtaining samples of the bottom, and for testing the nature of unknown localities. Many different patterns of frame are used for dredging, but these, for the most part, only differ from one another in unimportant structural details. The best type for all ordinary purposes is known as the double-sworded oyster dredge. Fig. 198 shows an example of this type, which was used by the Porcupine Expedition nearly sixty years ago, and the pattern has not been modified to any great extent since that date. The essential features of this type of dredge are the obliquely set blades or swords, on both sides of the mouth, the rings by which the net is attached (these prevent the edge of the bag from chafing when working on the bottom), and the loose connexion, which allows free lateral movement, between the bridles or arms and the frame. The whole instrument should be strongly made of wrought iron. Frames about 2 ft. long and 6 in. broad are suitable for in-shore work from small boats. For work at depths from 10 to 100 fathoms patterns from 3 ft. to 3 ft. 6 in. long and 8 to 9 in. broad are useful, while for very great depths large heavy makes up to 5 ft. long and i ft. broad are recommended. The term naturalist’s dredge may be restricted to a ' The Editor is indebted to Mr. F. M. Davis for revising the portions of this chapter on trawling and other methods of fishing. Dredging 329 form of frame which differs from the foregoing only in the absence of swords. The effect of this is that the straight edge of the frame tends to skim the surface of the ground instead of biting into it. This type has been almost en- tirely superseded by that mentioned above. The conical dredge is designed to dig into the soil; of this it brings up a good sample, together with specimens of bur- rowing animals. The bag or net which is fastened to the dredge should be quite as long as, if not longer than, the frame, and should not be tapered behind. It may be made of strong manila trawl-twine of 2-inch mesh," but a most suitable bag may be constructed from the cod end of an old trawl ; this is usually easily pro- cured and fitted. Loose chafing pieces attached to the outside are very useful, and save a great deal of wear and tear. For catching small animals and for use on smooth ground, a lining of sprat netting of I -inch mesh, or even finer, may be fitted inside the bag (some firms stock an extra stout make of sprat netting, which is excellent for this purpose). The lining should be made fully as large as the outer bag, so that the strain is taken up by the latter. Two small bags may also be fitted in the tail end of the lining; one of these is made of shrimp netting, and is designed to catch some of the smaller organisms which ^ Meshes are measuied diagonally -when the net is stretched out; thus 2-in mesh means openings of i square inch. 330 Dredging and Tranvling would pass through the i-inch mesh, while the other is made of sacking or stout canvas, and preserves a sample of the sand or other fine bottom material. The two main bags — of trawl-twine and sprat netting — are lashed to rings at the back of the dredge, or they may be fastened directly to the frame by means of seizing wire. Each bag is opened or closed behind by means of a running string through the meshes. The small bags are stitched in position through both the main bags, and their ends are secured by a lashing. The nets for small frames for use in shallow water are generally of a simple pattern. For these a piece of the cod end of a shrimp-trawl or the extra stout sprat netting mentioned above is suitable, with or without a lining of mosquito netting. For the conical dredge a strong canvas bag is recom- mended. Dredges used in deep water should always have a piece of chain fastened to the end of the bag to prevent it getting foul of the frame while being lowered to the bottom, and this device is also useful at all depths when the net is new. A tow-net made of cheese-cloth, with a strong cane ring, is sometimes fastened to the dredge. It is usually attached to the shackle at the end of the warp, with a strop sufficiently long to allow it to lie with its mouth a little behind the frame. By this arrangement the tow-net is designed to catch the small animals disturbed by the swords of the dredge. This method has been known to produce most satisfactory results, but the safe recovery of the tow-net is decidedly uncertain. Tangles (or swabs) consist of lengths of hemp rope teased out, and fastened to the end of the bag or to an iron rod attached by chains to the frame. Starfish and other animals become entangled in the threads of the Dredging 331 hemp, and arc thus drawn to the surface. Tangles were very highly spoken of by the naturalists on the Challenger^ but in recent years they have not been much emploved. The dredge should be attached to the warp in the following way: The eye at the end of one of the bridles is secured to the eye at the end of the warp by means of a screw shackle, and the other eye is made fast merely by a strong strop or lashing of trawl-line. This arrange- ment is invaluable in the event of the dredge becoming fast among rocks, for in that case the lashing parts and the frame is drawn along sideways, thus usually clearing the obstruction and preserving the dredge from injury. In f o fathoms of water or less it is usually best to pay out warp to the extent of two or three times the depth; in deeper water less is required, and when working in soundings of 1,000 fathoms or more, 400 or 500 fathoms in excess of the depth will be found sufficient. In many localities, however, the nature of the bottom is so extremely rough that, if this amount of warp is given, the strop parts on every occasion. On such grounds it is advisable to pay out only a little warp in excess of the depth, and, after the ship has been stopped, to ascertain if the dredge is biting by holding the warp.‘ More warp may then be given cautiously if required. Shooting and Tou'ing Sailing Yachts. Always tow with the tide if possible. With a moderate amount of way on the boat throw the dredge over on the windward quarter, and pay out a sufficient length of warp. Make the warp fast with a stop of small rope, but be sure to have the end well secured inboard, so that if the stop parts the whole will not be lost. ' By holding the warp it is often possible to feel a heavy dredge working, even at depths of 500 fathoms. 332 Dredging and Trawling Tow slowly, feeling the warp from time to time to make sure that the dredge keeps on the bottom. If the dredge comes fast, lower sail if possible, pass the warp up to the bow of the boat, and haul in until it clears. If it still can- not be got free, try hoisting sail and beating back over it. In working from small boats it is safer always to lower sail before hauling. Steam Yachts. In small vessels it is best as a rule to tow from the warping-chock right aft, or from a set of special rollers (consisting of two vertical and one hori- zontal) placed in the same position. In the absence of a capstan the warp may be eased away from the bollards. For handling the dredge a purchase from the mizzen is useful, and with heavy patterns necessary. In larger yachts the dredge may be towed direct from the trawling winch, the warp being led through the gal- lows or trawl-port to a strong davit or snatch-block on the port quarter, but the exact arrangement will vary with different vessels. For shallow in-shore grounds, especially where very rough, manila warp is better than wire, as it is more easily handled and manoeuvred if the dredge becomes fast. In shooting, the dredge is slung overboard by the derrick or masthead tackle, and slowly lowered away. In very deep water the engines may be put at half speed, and the warp eased away more quickly. When towing from a davit on the port quarter, especially with twin-screw vessels, it is advisable to put the helm hard-a-starboard. In this case the dredge should be shot with the ship’s head about four points to starboard of the direction in which it is intended to tow, so that the ship may at once be brought up on her course when the gear is clear away. When a sufficient length has been paid out, the warp is Dredging 333 made fast by means of a rope-strop and rolling hitch; ' if the dredge fouls, this strop should break. When using manila a few turns of the slack warp should be taken round the bollards, and when towing from the winch barrel the brakes should be left open if plenty of wire remains. Tow dead slow, preferably with the tide, feeling the warp from time to time to make sure that the dredge is biting properly. If the dredge hitches and the strop parts, stop the ship at once, and if it does not clear itself immediately, haul it up. Should it still remain fast, lead the warp forward and go astern, steaming slowly over the dredge. The dredge is usually left on the bottom for from ten to thirty minutes; if longer time is given the net may fill right up and the strop on the bridle may part, involving the loss of the greater part of the contents. The procedure in hauling is obvious. The ship is stopped, the rolling hitch is cast off, and the warp hauled by hand, capstan, or winch. In the latter case it is well to see that the warp is evenly spread across the breadth of the barrel; otherwise loose bights appear which later on will cause considerable annoyance. With a considerable amount of warp out, it is allowable to go slowly and cautiously astern at the commencement of hauling, a careful watch, of course, being kept that the warp does not come foul of the propeller. If block and tackle are employed for getting the dredge inboard, it is advisable to keep the frame slung up, and if necessary lashed, until the contents of the bag have been emptied. ’ It is sometimes difficult to make a rolling hitch hold on well-oiled w ire-warp, especially when of small diameter; in such cases it will be found that a serving of trawl-twine or a canvas parcelling will materially assist. Alternatively, a nipper of the pattern shown in Figs. 194 and 195 may be used. ( 334 ) Trawling Trawling is the best and easiest way of obtaining ex- tensive samples of the bottom fauna, but the method can only be satisfactorily employed on smooth ground. Three types of trawl are available for biological in- vestigation — the beam trawl, Agassiz trawl, and otter trawl. The Beam Trawl. In this pattern the net is kept open horizontally by a long beam of greenheart or oak, and vertically by a pair of iron runners or heads ; Fig. 199 gives a good idea of the appearance of the frame. Beam trawls may be used at practically any depth, but in very deep water they show a tendency to capsize, the beam dragging along the bottom, and the foot rope com- pletely closing the net. In such cases the haul is naturally a complete failure. With a certain amount of experience in shooting this difficulty may be largely overcome; but the otter and Agassiz trawls will be found more efficient for work in greater depths than 300 fathoms. The beam trawl collects a larger number of inverte- brates than the otter trawl, but does not secure so many fish — active species of the latter are doubtless scared by the beam in its passage along the bottom. Trawls of this pattern range from 6 to 50 feet in beam- length. Small makes, with a 6 or 8 feet beam, are easily handled from a rowing-boat or launch. The commercial beam trawl is usually from 40 to 50 feet in length; these sizes are, however, unwieldy, and the procedure in shoot- ing and hauling is very complicated. Yachtsmen will find patterns with a 15 to 30 feet beam amply large enough. In these sizes the shooting and hauling does not present any considerable difficulty. Tie 199 Be im Trawl With enlarged view of the D-shaped head carrMnj, the end of the beam, the span, ind the foot rope 336 Dredging and Trawling The beam trawl is towed by a pair of spans which are shackled to eyes in the front of the heads. Each span should be three and a half or four times the length of the beam, and as stout as the warp. The head rope of the net is as long as the beam, and is laced to it. The foot rope, to which the lower edge of the net is attached, is twice the length of the beam. The ends are securely lashed to the eyes at the back of the heads. As it is intended to scour the bottom, it is necessary to make the foot rope heavy; for medium-sized trawls 8-inch rope is recommended, weighted by the attachment of lengths of chain. The net should be about one and a half times the length of the beam. It is made of hemp or manila twine of 2 to 22-inch mesh in the back and belly, and i-inch double mesh in the cod end. The lower part of the cod end, which comes in contact with the ground, is protected from chafe by an extra piece of stout netting attached on the outside. The trawl moves comparatively slowly over the sea bottom, and consequently fish which have entered the net frequently manage to swim out again and avoid capture. To obviate this, pockets are formed by stitching together the back and belly of the net at the sides, as indicated in Fig. 199, and by the inclusion of a ‘flapper’ or trap im- mediately below the pockets. This piece of netting, 3 to 4 feet in length, according to the size of the trawl, is made 50 meshes at top part and bated evenly on both sides to 25 meshes at bottom part. The meshes are of similar size to the meshes at that part of the net at which it is inserted. The ‘flapper’ should be put in its place before the pockets are laced up. The 50 mesh edge is sewn on the top or bating part of the net and the sides are then fastened on to the belly or lower part of the net, care being taken that Trawling 337 the flapper is not tighter than the net. The bottom part of the flapper is left unattached. The water passing down the net when the gear is being towed keeps the flapper open for the passage of the fish down into the cod-end. When the ship is stopped the flapper falls down, so that fish coming back along the net must go up into the pockets. The net is opened and closed behind by a run- ning string through the meshes. The 2- or 22-inch mesh is much too large to retain any small organisms. The whole net may be lined with sprat netting (i-inch mesh), or bags of the same material and of mosquito netting may be attached to the back, in the path of the swirl caused by the foot rope. The latter system has the advantage that the coarser part of the catch is kept in the trawl, while samples of the smaller animals are separately collected and saved from crushing. Tow-nets with cane or iron rings and a long cheese-cloth bag may also be attached to the beam or to the net. If placed too far back, they frequently fill with sand or mud, and burst. The very small makes of trawl, which are suited for use from rowing-boats or launches, are usually known as ‘shrimp trawls’. In these the net is made of stout sprat meshing with a lining of mosquito net. The length of warp required varies with the depth. The following table will give some indication of what is required : In soundings up to 50 fathoms — 3 times the depth. „ of ^ 0—200 ,, 22 „ ,] ,, of 200—500 ,, 2 „ ,, „ of over 500 „ — 500 fathoms in excess of the depth. Beam trawls may be towed from the warping-chock aft, 3166 z 338 Dredging and Trawling from a trawl-post, or from a wide-mouthed snatch - block * slung from a derrick or davit. Shooting over the stern is much the most convenient, but the method can only be used with very small patterns. As a general rule, on all steam trawlers, it will be found best to shoot and haul the gear from the starboard side, the propeller on modern trawlers turning to starboard. The following is an outline of the procedure : The trawl is laid on deck with the heels of the irons close to the rail, and with the net brought back over the beam. The spans are coiled down on deck, and the warp, after it has been passed through the trawl port or other fairlead, is shackled to them. It is best to use a separate shackle for each span. When these preparations are com- plete, way may be got on the ship, and the trawl shot with helm hard-a-port. The frame is lifted to the rail by hand or by tackle. The cod-end is put over the side first, fol- lowed by the heads and beam, the trawl drifting away on the port quarter as the spans uncoil. It is best to square the trawl by checking the after span, while the other is let run freely. This should be done while the net is still at the surface and can be seen. When the whole length of the spans has disappeared, the warp may be eased slowly away, and the ship brought up on her course. In most cases it will be necessary to trice the warp up to the after hawse pipe, to keep it clear of the propellers. It is advisable to start shooting with the ship’s head from four to eight points off the wind or tide, whichever is the stronger. The ship’s course should be as straight as possible when shooting, the ship being brought slowly up to her course only when the gear has set on the bottom. ^ A snatch-block for this purpose must be wide enough to allow the passage of the shackles connecting the warp and spans, and should have a deep V-shaped groove. Traisling 339 It is necessary to keep a steady strain on the warp and to check it repeatedly as it is being paid out. After each check the warp should be paid out gently at the beginning to prevent the beam and heads dropping and the net coming forward over the beam. Neglect of this precau- tion is the most frequent cause of a foul shot. When a sufficient length has been given, the warp is made fast (as in dredging) by means of a rope stop or by a nipper (P- 315)- The trawl is generally fished for from one and a half to two and a half hours. By holding the warp it is possible, in moderate depths, to feel the gear working its way along. With a little practice a useful estimate of the nature of the bottom can be formed by this means. If the trawl hitches and the stop parts, the net should be hauled at once. If it still remains fast, lead the warp to the bow, go astern and pull it backwards off the obstruction. Hauling small beam trawls is not difficult. The warp is reeled in until the spans appear, and each span is sepa- rately hauled by hand until the beam and irons are got on board ; the net may then be hauled in, foot rope first. This procedure cannot be applied to larger patterns (with beam more than 1 5 feet in length), for in these makes the weight is too great to admit of the spans being hauled by hand. The addition of a ‘beam rope’ is almost essential in such cases, and it is also advisable to step a stout davit with a snatch-block on the port quarter. The ‘beam rope’ is in reality a third span, used only iir hauling. It is six feet longer than the spans, and is made fast by one end to the beam, close to the after head, and by the other to the trawl warp just above the shackle. When the spans appear through the fairlead, the beam rope (which is, of course, slack) is taken off, passed through the snatch- block on the port quarter, and led back to the capstan, or 340 Dredging and Trawling to one of the small drums of the winch. Heave on the beam rope until the after span is slack and can be un- shackled and taken from the sheave. Then heave away on the fore span and on the after beam rope until the trawl head appears. The quicker this operation is completed the better, as the net will hang almost up and down from the ship’s side if there is no wind. Under these conditions the flapper does not operate and the more lively fish will tend to escape from the net. When the frame appears at the water’s edge, one head after another may be lifted in by the masthead or derrick tackle. The net can then be hauled in over the beam, taking the foot rope first to avoid the loss of any of the con- tents. The Agassiz Trawl. This is a small trawl useful for work in deep water. When a small trawl of the usual pattern is shot in deep water, it is liable to turn over, so that the beam lies on the ground and the net does not fish. In order to obviate this Professor Agassiz devised a trawl with two ground ropes and an iron frame, so arranged that the net would fish whichever way up the trawl fell on the bottom (Fig. zoo). The shape of the frame is shown in the figure. The following dimensions are given by Tanner: Beam: ii ft. Runners: length, 4 ft.; depth, 3 ft. 6 in.; width, 3 in. Trawl net: length, 17 ft.; mesh, i in. to -g- in. Otter Trawl 341 For working in water over 100 fathoms in depth, it has been found advantageous to weight the frame consider- ably, and this may be conveniently done by winding chain around the iron beams. The net should be of ‘shrimp’ or ‘sprat’ mesh. The Agassiz trawl is more useful for catch- ing invertebrates than for fish. Its disadvantages, as com- pared with beam and otter trawls, are that it is necessarily small and that fine-meshed nets cannot be attached to the back. The procedure In shooting and hauling does not differ in any essential way from that already detailed under ‘Dredging’ (pp. 331-3). Agassiz trawls of small pattern (5 to 6 feet) may be operated easily over the stern. The Otter Trawl. For work on small steamboats, especi- ally where a winch or capstan of any kind is available, a small otter trawl is more handy to use than a beam trawl of equal spread. If it is desired to convert a beam trawl net into an otter net, the square of the beam net must be so cut that the net at this point will take the crescent- shaped curve of the head rope. The square of the beam net is straight to fit the beam. It would be preferable, however, to purchase a new otter square to fit the net. In the case of the otter trawl, two oblong boards, fitted with iron shoes — the otter boards — are used to keep the mouth of the net open. The mouth of the net in both beam and otter trawl has a heavy ground rope below, which drags along the sea bottom, and a head line above, which in the beam trawl is attached to the beam, but in the otter trawl is floated up in the water by means of corks or (in deep water) glass floats attached to it. These two ropes, the head line and the ground rope, are made fast at each end to the hinder ends of the oblong otter boards, which, when the net is working, stand on their lower iron-shod edges on the sea bottom. 342 Dredging and Trawling The two otter boards act after the manner of kites which are ‘flown’ from the stern of the ship by means of two towing warps, one bearing away to starboard, the other to port. Each towing rope is fixed to its otter board by means of iron brackets or of four chains, in such a way that the point of attachment is behind the centre of the board, and the board therefore moves forward obliquely to the direction of tow. The two boards in this way keep a strain on the head line and ground rope, which increases with the speed at which the vessel is towing. It is pro- bable that under normal working conditions the distance which the boards actually keep apart is about two-thirds the length of the head line of the trawl. The head line, being floated with cork or hollow glass balls, rises up in an arch between the two boards, and gives a much higher opening to the mouth of the net than is yielded by the beam trawl, in which the height of the opening is fixed by the height of the beam above the ground. This increase in the height of the opening of the mouth of the net constitutes one of the chief advantages of the otter trawl as a fishing instrument, since the take of round fishes, which swim a little above the actual bottom, is thereby much increased. The size of the otter trawl usually worked on a steam trawler is as follows : Length of head line, 96 to 105 ft.; length of ground rope 1 20 to 1 30 ft. The relative dimensions of head line and ground rope will depend on the type of square, top wing, and lower wing used in the trawl. The standard otter boards suitable for nets of the above dimensions are 10 ft. by 4 ft. 3 in., thickness 2 ^ in. Smaller nets may be used with these boards, but it is preferable to have lighter boards to suit such small nets. Size of mesh, measured diagonally with mesh taut, diminishes from 53 inches in the square to 2| inches at cod-end. Otter Trtvwl 343 The fine-meshed nets already mentioned in connexion with beam trawling may also be fitted to otter trawls. On a commercial steam trawler two independent towing warps are used, one for each otter board, and special iron gallows are fixed to the vessel’s side, fore and aft, for getting in the heavy boards. It is not necessary to give here the details of the method of shooting and hauling these heavy trawls, as this should not be attempted ex- cepting with a specially fitted vessel and under expert direction. The following instructions may be useful for working a small otter trawl from a small vessel, without special fittings. Such a trawl is most conveniently worked with two ‘bridles’ and a single towing warp, as in the case of a beam trawl. Each otter board should not be heavier than can be conveniently handled by one man. The following dimensions apply to what is perhaps the largest otter trawl that can be conveniently worked in this way: Square: ii ft. long: 140 meshes down to no; mesh 2 in. Top wings: 17 ft. long; 36 meshes down to 6 at the board end; mesh 2 in. Bottom wings: 31 ft. long; 25 meshes down to 28; mesh 2 in. Eatings: no meshes down to 36 meshes; length 20 ft. The pockets may be short or long, as desired, but must be started from the head of the ‘flapper’. Belly: Braided same as batings. Cod-end: 56 meshes across, 6 ft. long. Mesh as desired. Head rope: 44 ft., 7 ft. to be allowed for bosom and 1 ft. for each top wing. Ground rope : 5 7 ft. No bosom, but brought to an apex. 344 Dredging and Trawling The circumference when rounded should be about 7 in. Length of spans used: 50 ft. of i-J- in. flexible wire. The main warp should not be too heavy as this has a tendency to pull the boards down. Size of otter boards: 3 ft. 6 in. by 2 ft. 3 in. A trawl of this pattern may be shot over the stern of the vessel. To prepare the trawl for shooting, attach the head line and ground rope to the two boards, shackle one bridle to each otter board, and then attach the other ends of the bridles to the towing warp by means of a freely working swivel. When the cod-end of the net has been tied, steam slowly ahead and pay the net overboard, starting with the cod-end. One man should be in charge of each otter board, and should stand on the outside of the bridles. When the net is trailing away all clear astern, the two boards are thrown overboard together, and a short length of the bridle is allowed to run out. The bridles are then checked for a moment to give the boards time to open the trawl. When this has happened, and the trawl is squared, the rest of the bridles are run out, being checked at intervals, until the towing warp is reached. The latter should be held for a moment, either by the brake of the winch, if one is being used, or round a bollard. The warp is then slowly paid out until the trawl reaches the bottom. The amount of warp required is the same as in the case of the beam trawl. The warp is made fast by a stopper of thin rope, which will part if the trawl comes fast. When hauling the gear the vessel should not be stopped, but eased down a little until the trawl is off the bottom. The speed of the vessel should then be slightly increased in order that the trawl may be kept well astern while heaving in. The rate of hauling should be compatible with the strength of the towing gear. This operation tends General Gear 345 to carry all the livelier fish down into the cod-end. There is perhaps more art in retaining the live fish than in catch- ing them, and if no lively fish are being got in the net, the speed must be accelerated to prevent their escape while the net is being hauled. There should be fairleads on each quarter of the ship, so that when the spans come in they may be parted one on each quarter and the trawl hauled in the centre. The trawl will then be clear and the flapper working. The vessel should only be eased down when the boards are up at the fairleads. The ground rope is then hauled in. The boards should be left hanging outboard, and the trawl hauled in between them. The boards are in position for the next haul. If there is a heavy catch in the bag, a rope may be passed round it, and the bag hauled in with a tackle from mast or derrick. General Gear Before dealing in detail with the various kinds of line and net likely to be used by those for whom this work is intended, a few prefatory remarks upon the general sub- ject may prove useful. 1 . All lines and nets, of whatever material, should be oil-dressed or barked, not tarred. 2. Hooks, swivels, shackles, and other metal articles, if made of iron or steel, should be galvanized. 3. Many of the lines and nets described take up a lot of room when mounted ready for use ; it is generally better in such cases to carry their component parts, and mount them as and when required. 4. All gear should be well dried after use before being stowed away. 5. Nets in particular constantly need mending, and any vessel carrying nets should also carry at least one person who can mend them. 346 Dredging and Trawling 6 . Plenty of spare lines, hooks, snooding, twine, ropes, lead (pipe and wire), corks (or glass floats), and of any- thing else likely to be required for repairs or renewals, should always be carried. 7. And lastly, never despise native gear, however un- promising in appearance; for one thing, it is always possible to find some one locally who can make it fish efficiently; and, for another, it has probably stood the test of long use in practice. For exploring any waters in which a commercial fishery is carried on, the hiring or subsidizing of those who carry it on will probably, in the long run, prove the best way of securing fish. LINES Hand Lines may be used either with or without a rod. Quite apart from its sporting aspects, a rod possesses two advantages: (i) it holds the line clear of the boat, pier, or rock from which it is used; and (2) it enables lighter tackle to be employed than if a hand line is used. For fuller information as to sea rods, reels, and lines, reference should be made to such books as the Badminton Library volume on sea fishing; but the following brief notes may prove useful : The rod should be stiff and not too long, its metal parts made of phosphor-bronze or some other rust-proof metal. It must be remembered that the weight required to keep a line on the bottom in a tideway is considerable, and that a long rod is a nuisance in a boat, and in no way more efficient than a short one. The reel should be large and of the ordinary Notting- ham pattern, with an optional check. We do not know that anything is really better than a strong wooden reel with an easily fitting line drum, preferably with a rust- proof metal flange and back plate; the working parts Line Fishing 347 should be kept well greased and the line-drum fit easily, unless backed with metal; otherwise the wood will swell when saturated with sea-water, and the drum jam in the flange. The cost should not exceed ioj. or \ 2 s. For lines plaited and barked hemp is quite satisfactory and inexpensive, costing from 2s. 6 d. to 4/. per 100 yards, according to weight; a twisted line is apt to kink, and should be avoided. Traces and snoods may be of plain or twisted gut, horsehair, wire, hemp, or raw hide, according to the probable weight and biting powers of the prospective fish. Swivels should always be used, whether for bottom fishing or trolling, between the running line and the trace. Leads for trolling should be boat shaped or drain-pipe shaped — preferably the former if at all heavy. For bottom fishing a conical lead is as good as anything else; this should be attached to the trace by a snood of such strength as to break before the trace, if the lead gets jammed in a crevice among rocks or wreckage. It is advisable in bottom fishing to have a swivel between the trace and the lead-snood. A line rigged as in Fig. 201 permits the lead to be kept on the bottom at whatever depths the hooks are fishing, and render it easy to detect any inability on the lead's part to hold the line straight against tides or currents. Spinning-baits, both natural and artificial, are number- less; we believe the best natural bait to be a strip of skin and flesh from the tail of some silvery and tough-skinned fish (e.g. a mackerel) or a small silvery fish used whole. Of the artificial baits, two standard and useful forms are (i) a piece of black, white, or red rubber tubing, and (2) a triangular or fish-tailed piece of bright metal, with its 348 Dredging and Trawling edges turned so as to make it spin. Red worsted, also, has its merits, especially if used in combination with a bright metal spinner. Fishes which swim and feed in shoals may often be caught by a ‘jigger’; this is a piece of lead shaped like an ordinary sounding lead, and 6 inches long or less, armed according to fancy with bare hooks. The lead must be scraped bright before use, and is then worked rapidly up and down, so as to hook any fish attracted by curiosity or the prospect of its proving edible. The above remarks apply generally to hand lines, but such lines must, of course, be considerably heavier and stouter than lines used with rods; also the use of plaited lines and swivels is not so necessary. When spinning with a hand-line, it is often more convenient to use several small leads than one large one (see Fig. 201). Hemp lines, 40 fathoms long, should cost from about 36J. to 74J. per dozen, according to weight. Long lines, spillers, or bolters, are suitable for use in water down to 500 fathoms or more. They possess two great advantages: that they can be used on ground too patchy for trawling, and that they can be left to themselves while the vessel setting them is otherwise occupied; also one very serious disadvantage — that they require a great deal of bait. The long line as ordinarily used by English fishermen for commercial purposes is 40 fathoms in length, and carries hooks on snoods of from 3 to 5 feet long, fastened on to the line at intervals of i-g to 2 fathoms. Any number of such lines up to twenty dozen (which is the full outfit of a modern line-fishing vessel) are fastened together end to end, and fished in one string. The ends of the string are made fast to anchors or weights and their positions marked by buoys ; and, in the case of a long string, further Long Lines 349 anchors are added at the end of every sixteen lines, and their positions also marked by smaller buoys. Before setting, the lines must be carefully coiled in boxes or shallow baskets (each ordi- narily holding eight lines), and the hooks baited and so arranged as to run out smoothly without fouling. Long lines should always be shot across a tide, and, where possible, hauled at slack water. No one who has not had considerable experi- ence of long lines and their ways should ever attempt to set or haul them out of anything but a rowing- boat; and it is well to re- member that if lines are set at low water or on a rising tide sufficient slack should be allowed on the buoy lines to prevent the buoys being drawn under the surface as the tide rises. By regulating the dis- tance from the buoys at Fig. 201. Traces 1. Spinning trace suitable for use with a hand line with several small leadsj 2. spinning trace for use with a rod or hand line; 3. trace for bottom fishing R. Running line; t. trace; s. swivels; Ij., lead-snood; L, lead 350 Dredging and Trawling which the lines are fastened to the buoy ropes, a long line may be set so as to fish at any desired distance from the bottom. By the courtesy of a well-known east coast firm we are able to give the following estimates