''■r%^ : ;• V ■:- / .^r f ■^ <•* % V '^^ '^^ ^x,r i ^^#:#l 4 f ■ _D = o So so DISCOVERY REPORTS VOLUME XIV Cambridge University Press Fetter Lane, London Neiu York Bombay, Calcutta, Aladrai Toronto Mdcmillan Tokyo Maruzcn Company. Ltd All rights reserved ^ DISCOVERY REPORTS ^,^,^ Issued by the Discovery Committee Colonial Office, London on behalf of the Government of the Dependencies of the Falkland Islands T VOLUME XIV CAMBRIDGE AT THE UNIVERSITY PRESS 1937 PRINTED IN GREAT BRITAIN BY WALTER LEWIS, M.A., AT THE UNIVERSITY PRESS, CAMBRIDGE CONTENTS ON THE DEVELOPMENT AND DISTRIBUTION OF THE YOUNG STAGES OF KRILL (EUPHA USIA SUPERB A) (published 7th December 1936) By F. C. Fraser, B.Sc. Introduction page 3 Development 17 Distribution 109 Summary ' . . . 166 List of Literature Cited 167 Appendix I 169 190 Appendix II ic THE SOUTHERN SPECIES OF THE GENUS EUPHAUSIA (published 14th December, 1936) By D. Dilwyn John, M.Sc. Introduction page 195 Material 107 The Surface Waters of the Southern Oceans 200 The Species as Adults 203 The Relations of the Species to one another 239 The Later Larval Development of Five Species 253 References Appendix I Appendix II 304 307 319 THE REPRODUCTIVE SYSTEM OF EUPHAUSIA SUPERB A (published 2Sth June, 1937) By Helene E. Bargmann, Ph.D. Introduction External Characters Reproductive System Fertilization Summary Bibliography . . 327 328 329 346 348 349 Plates I-V jollowing page 350 4O!0n:2 vi CONTENTS LARVAE OF DECAPOD CRUSTACEA. PART IV. HIPPOLYTIDAE (published 25th June, 1937) By Robert Gurney, D.Sc. The Larval Genus Eretmocaris Bate page 353 The Genus Tozeuma Stimpson 377 Chorismus antarcticus Pfeffer 384 Sarmon marmoratus (Olivier) and Allied Forms 390 Latreutes mucronatus (SriUPSON)} 39° Discussion and Summary of Generic Characters 399 Literature 4°2 [Discovery Reports. Vol. XIV, pp. 1-192, December, 1936.] ON THE DEVELOPMENT AND DISTRIBUTION OF THE YOUNG STAGES OF KRILL {EUPHAUSIA SUPERB A) By F. C. FRASER, B.Sc. Department of Zoology, British Museum (Natural History) CONTENTS Introduction page 3 Sources of material 3 Methods 4 Acknowledgments 16 Development 17 The egg 17 First Nauplius 19 Second Nauplius 19 Metanauplius 20 First Calyptopis 23 Second Calyptopis 26 Third Calyptopis 29 Summary of previous knowledge relating to the Furcilia and Cyrtopia stages 32 Early Furcilia stages 3^ Interpretation of dominant stages 3^ Intermediate Furcilia stages 4^ Later Furcilia stages 49 Description of Furcilia stages 68 First Furcilia 09 Second Furcilia 7" Third Furcilia 84 Fourth Furcilia °° Fifth Furcilia 92 Sixth Furcilia 9^ Adolescent forms 99 Anomalous length frequencies in adolescent forms 103 Average length of larvae loo Average growth rate i°7 Distribution i°9 Eggs 109 Regional io9 Vertical m Time of spawning '^3 Nauphi 114 Metanauplius ^'4 Regional ^H Vertical I'S Time of occurrence 117 Calyptopis and Furcilia stages and adolescents 117 Regional distribution ii7 Vertical distribution ^5^ Time distribution ^"^ General remarks on distribution 103 Summary '"6 List OF literature CITED 167 Appendix I ^69 Appendix II ^9° ON THE DEVELOPMENT AND DISTRIBUTION OF THE YOUNG STAGES OF KRILL (EUPHAUSIA SUPERBA) By F. C. Fraser, B.Sc. Department of Zoology, British Museum (Natural History) (Text-figs. 1-76) INTRODUCTION THIS report deals with the development of Euphaiisia siiperba during the period covered by the first fifteen months of its growth. The distribution of this young euphausian with reference to locality of occurrence, time of occurrence and vertical position is also discussed. It has been possible to come to certain conclusions regarding the life history of the animal, but it is felt that there are many points mentioned upon which more research will be required in order to understand their proper significance. It is hoped, however, that the results set down here will be useful in future analyses of E. superba material, especially in connection with the very large collections made by the vessels of the Discovery Committee, and that in this way they may help in the final elucidation of the biology of this animal which is such an important constituent of the antarctic fauna. In a report to the Discovery Committee in 1930, Dr J. F. G. Wheeler demonstrated that there is a two-year developmental period in E. superba as opposed to a one-year period in other euphausians whose life history is known. He measured krill from the stomachs of whales throughout a whaling season, and from the analyses of length frequencies was able to distinguish the two-year groups. Dr Johan T. Ruud in 1932 gave the first published account of this phenomenon. SOURCES OF MATERIAL The stations mentioned below, from which plankton samples were examined, are nearly all located in the Falkland Islands sector of the Antarctic. The area round South Georgia was concentrated upon in the earlier stages of the investigations, for at that time ice-edge whale fishing was in its infancy and South Georgia was one of the chief centres of the industry. In addition to observations in the South Georgia area many were made in the Bransfield Strait and on lines of stations in the Scotia Sea. It is perhaps un- fortunate that the main source of the data presented is thus in a portion of the Antarctic where hydrographic and bathymetric conditions are so very highly specialized. The great ridge of the Scotia Arc, the proximity of Graham Land to the southern end of South America, and the system of currents associated with the Weddell Sea, all con- 4 DISCOVERY REPORTS tribute to the production of a complex the hke of which is encountered in no other region of the Antarctic. Apart from the stations in the area of the Dependencies of the Falkland Islands, much valuable material was obtained from a series of ice-edge stations extending from east- wards of Bouvet Island to South Georgia: also from stations in the Bellingshausen Sea and from stations made during the first circumpolar cruise of the R.R.S. ' Discovery II ' in 193 1. Respecting the last series of stations I am indebted to Mr D. Dilwyn John, who was senior scientific officer on the ship at that time, for examining the plankton samples and sending me the young euphausians. I have also drawn very greatly on the informa- tion contained in his reports to the Discovery Committee relating to this circumpolar cruise. A list of the samples in which eggs or young E. superba were obtained is given in Table I. The 70-cm. nets yielded the main supply of younger developmental stages, the i-m. oblique and horizontal nets older larvae and adolescents. The i-m. oblique and horizontal hauls of the circumpolar cruise were at stations most of which were made during the southern winter. Both in time distribution and in the development of the E. superba recorded, the results of the circumpolar cruise form a convenient link between the main body of 70-cm. net results and those of the i-m. nets. The plankton samples of the circumpolar cruise were examined by John and only the picked out krill was deah with by the writer. John's search for E. superba covered all stations made between the Antarctic convergence and the ice-edge. The positions of all stations from which samples were examined are marked on the maps showing distribution, and indication is provided of the presence or absence of krill. METHODS A full description of the plankton nets used in obtaining the samples is given in Vol. I of the Discovery Reports (Kemp, Hardy and Mackintosh, 1929) and it is therefore un- necessary to recapitulate it here. The whole catch was examined in the small samples and a portion of the whole was used when very large amounts of plankton were taken. In the search for eggs, Nauplii and Metanauplii the macroplankton was picked out and the residue inspected under a binocular microscope. The Calyptopis and Furcilia stages are sufficiently distinctive to enable them to be identified by eye alone. Drawings of appendages were made from unstained material with the aid of a camera lucida. For the measurement of the larvae two scales were used of which the smaller, having 62 divisions to i mm., was employed for eggs and small larvae and the larger, having 24 divisions to i mm., for more advanced stages of development. In analyses where actual length in millimetres is not important, results have been classified primarily according to the original units of measurement. The corresponding dimensions in millimetres have been added. In order to measure the larvae as speedily as possible a shallow trough, having the two long sides at an acute angle, was constructed by cementing strips of glass on a DEVELOPMENT OF YOUNG EUPHAUSIA SUPERB A Table I. Statmis at which Euphausia superba was taken Rtation Position Date Net Depth Time kJ Lu \, l\J 1 1 m. Shot Hauled R.R.S. 'Discovery' 1927 i6i 57° 2' 20" S, 46° 43' 30" W 14. ii N70V 100-50 250-100 0905 1 100 167 169 60° 50' 30" S, 46° 15' w 60° 48' 50" S, 51° 00' 20" w 20. ii 22. ii N70V N70V 250-100 500-250 750-500 1045 0905 1130 1210 171 187 193 62° 07' S, 57° 03' W 64° 48' 30" S, 63° 31' 30" W 63° 24' S, 61° 33' W 25. ii 18. iii 28. iii N70V N70V N70V 250-100 100-50 500-250 0905 0945 1 100 1 100 1020 194 62° 57' 30" S, 60° 22' w 28. iii N70V 750^-500 50-0 100-50 250-100 2100 1240 196 197 62° i7'3o"S, 58°2i' W 62°27'S, 58° ii'3o"W 3. iv 3. iv N70V N70V 500-250 50-0 50-0 100-50 750-500 1055 1620 2220 1240 198 62° 38' S, 58° 04' W 3-4. iv N70V 1000-750 50-0 100-50 750-500 2220 1805 199 62° 49' S, 57° 56' 30" W 4. iv N70V 1000-750 50-0 250-100 500-250 0900 0042 200 201 62° 59' 30" S, 57° 49' W 63° 00' 30" S, 59° 06' 30" W 4. iv 5-iv N70V N70V 700-500 300-250 50-0 1445 0820 1030 1545 202 62° 48' S, 60° 05' W S-iv N70V 250-100 50-0 1720 0910 203 62° 56' S, 59° 50' W 5-6. iv N70V 500-250 50-0 100-50 2240 1915 204 63° 05' S, 59° 42' W 6. iv N70V 250-100 500-250 0800 0030 205 206 209 63° 14' S, 59° 34' W 63° 26' S, 59° 28' W Port Foster, Deception I., S. Shetlands R.R.S. 'Discovery 11' 6. iv 6. iv 14. iv 1930 N70V N70V N70V 225-100 250-100 150-100 1200 1 545 0900 0907 1217 1615 0940 300 302 303 52° 25' 30" S, 37° 14' w 52° 46' 30" S, 37° 12' W 53°oo'S, 37°ii'W 20. i 21. i 21. i N70V N70V N70V 500-250 250-100 500-250 500-250 750-500 1730 0415 1048 2036 0648 1250 In the column headed "Net", N 70 is the 70-cm. net, N stramin bag. B = hauled obliquely, H = hauled horizontally. 100 the i-m. net, and TYF the 2-m. net with V = hauled vertically. DISCOVERY REPORTS Table I {cont.) Time Station Position Date Net Depth m. Shot Hauled 1930 3°4 53° 06' S, 37° 14' W 21. i N70V 500-250 1555 1823 305 53° 17' 8,37° ID' W 21-22. i N70V 100-50 250-100 500-250 2130 2353 312 53° 39' 45" S, 35° 37' 3°" W 24-25. i N70V 220-100 2350 0345 313 53° 32' 30" S, 35° 24' 30" W 25. i N70V 250-100 750-500 0530 0720 319 53° 37' S, 38° 39' 3°" W to 53° 33' 3°" S, 38° 37' W 29-30. i N70V 50-0 100-50 250-100 1000-750 2253 0116 320 53° 10' 30" S, 39° 44' 30" W 30. i N70V 250-100 500-250 1610 1806 321 53° 17' S, 39° 31' W 30-31- i N70V 100-50 1000-750 2119 2352 323 53° 28' S, 38° 55' W to 53° 29' S, 38° 55' 30" w 31- i N70V 500-250 1000-750 1 105 1247 324 54° 56' S, 39° 57' W to I. ii N70V 250-100 0855 — 54° 56' S, 40° 01' W 500-250 1215 332 54° 44' 30" S, 39° 09' W 2-3. ii N70V 250-100 2300 233° 335 55° 33' S, 36" 49' 30" W to 55° 31' 30" 8,36° 49' 30" w 4-5- ii N70V 50-0 2203 0315 337 55° 09' S, 36° 48' W 5. ii N70V 250-100 1223 1440 338 55° 00' 30" S, 36° 46' W S-" N70V 225-100 1614 1646 342 55°47'S,34°ii'W 7. ii N70V 500-250 1210 1505 343 55° 40' S, 34° 23' W 7. ii N70V 250-100 1724 1850 344 50° 33' S, 34° 35' 30" W to 55° 29' 30" S, 34° 34' w 7-8. ii N70V 100-50 2135 2350 345 55° 20' S, 34° 47' 30" w 8. ii N70V 180-100 1155 1257 353 54° 17' 30" S, 35° 06' W 9. ii N70V 250-100 1216 1405 354 54° 15' 30" S, 34° 47' 30" W to 54° 13' 15" S, 34° 46' W 9. n N70V 500-250 1548 1800 356 54° II' S, 33° 49' W to 10. ii N70V 250-100 0610 — 54° 08' 45" S, 33° 47' 3°" W 500-250 — 0743 357 53° 07' S, 34° 48' W to 10. ii N70V 100-50 1835 53° 07' 30" S, 34° 45' 30" W 250-100 — 2217 358 53° 16' 30" S, 35° 02' 30" W to II. ii N70V 50-0 0113 53° 17' S, 34° 58' w 100-50 — 0357 360 55° 53' S, 32° 33' W to 24-25. ii N70 V 50-0 2020 55° 50' S, 32° 26' 30" W 100-50 — 233° 361 55° 53' 3°" S, 30° 46' W to 55° 52' 30" S, 30° 44' W 25. ii N70V 50-0 100-50 250-100 0909 1220 362 56° 04' S, 29° 15' W to 56° 03' 15" S, 29° 20' W 25. ii N70V 50-0 100-50 250-100 1835 2146 365 56° 55' S, 27° 02' W to 56° 53' S, 26° 59' w 2. iii N70V 50-0 100-50 250-100 500-250 1305 1505 DEVELOPMENT OF YOUNG EUPHAUSIA SUBERBA Table I (cont.) T me Station Position Date Net Depth m. Shot Hauled 1930 368 Douglas Strait, Southern Thule, S. Sandwich Is., one mile N. of Twitcher Rock 8. iii N70V 50-0 100-50 250-100 500-250 1 1000 1050 369 59° 17' 15" S, 26^ 57' W 9. iii N70V 50-0 250-100 0855 1053 372 57° 57' S, 29° 53' W 18-19. 'ii N70V 50-0 100-50 250-100 2214 0004 373 58° 00' S, 33° 44' W 19. iii N70V 50-0 1959 2204 374 57° 55' S, 37° 30' W 20. iii N70V 50-0 100-50 1847 2135 375 57° 47' S, 40° 49' W 21. iii N70V 50-0 500-250 1343 1624 378 62° 21' 30" S, 6o°36'W 13. iv N70V 50-0 100-50 1 142 1210 383 60° 32' S, 62° 42' W 14. iv N70 V 100-50 250-100 0703 1035 393 A 54° 17' S, 35° 30' W 7.V N70V 150-75 1321 1420 393 D 54° 17' s, 35° 30' w 8.V N70V 300-150 0255 0420 453 54°05'3o"S, 3°57'i5"Eto 16-17. X N70V 50-0 2040 2341 54° 07' S, 4° 03' E N 100 B 165-0 2353 0014 454 53° 42' S, 4° 42' E 17. X NiooB 192-0 2147 2203 455 53° 55' 30" S, 4° 47' E 18. X NiooB 1 16-0 0327 0347 459 55° 09' 15" S, 2°oo'E 19. X NiooB 183-0 2110 2130 460 56°46'S, o°4i'45"W 20-21. X NiooB 155-0 2049 2109 461 A 56° 44' S, 2° 23' 45" W 21. X NiooB 80-0 170-80 270-170 385-270 510-385 650-510 1434 1434 1434 1434 1434 1434 1454 1455 i455i 1456 1456! 1457 N70V 50-0 750-500 1603 1741 461 B 56° 44' S, 2° 23' W 21. X NiooB 75-0 160-75 255-160 345-255 440-345 520-440 1830 1830 1830 1830 1830 1830 1848 1849 1849I 1850 1 850 1 1851 N70V 50-0 1956 2127 461 C 56° 44' S, 2° 22' W 21-22. X NiooB 95-0 200-95 310-200 420-310 535-420 660-535 2236 2236 2236 2236 2236 2236 2256 2257 2257J 2258 2258* 2259 N70 V 50-0 250-100 2344 0130 DISCOVERY REPORTS Table I {cont.) Time Station Position Date Net Depth ■^-' V(4 L I V^l 1 M. WOlLlWll IN CL m. Shot Hauled 1930 461 D 56° 41' S, 2° 24' W 22. X NiooB 85-0 180-85 280-180 385-280 490-385 600-490 0237 0237 0237 0237 0237 0237 0257 0258 0258I 0259 0259I 0300 N70V 50-0 100-50 0355 0516 461 E 56° 41' S, 2° 24' W 22. X NiooB 75-0 245-160 33c^245 420-330 515-420 0629 0629 0629 0629 0629 0649 0650I 0651 065 1 1 0652 N70V 50-0 0725 0842 461 F 56° 44' S, 2° 22' W 22. X NiooB 80-0 1025 1046 N70V 50-0 100-50 250-100 1130 1315 461 G 56''44'45"S, 2°2i':io"Wto 22. X NiooB 95-0 1429 1449 56° 44' 30" S, 2° 21' W 315-205 (-0) 700-560 (-315) 1429 1429 1456* 1458 N70V 50-0 1540 1736 462 56° 01' 8,7° 28' W 23.x NiooB 90-0 2159 2219 463 55° 42' S, 10° 54' W 24.x NiooB 132-0 2042 2102 464 56°o3'S, i2°i8'W 26. x NiooH 67(-o) 1028 1048 465 56''49'S, 14° 02' 15" W 26. X NiooB 1 13-0 2059 2117 466 55° 35' S, 16° 31' 30" W 27.x N 100 B 79-0 2053 2113 467 54° 39' 30" S, 19° 05' 30" W to 54° 40' 30" S, 18° 58' w 28-29. " NiooB 143-0 2039 2059 469 54° 07' 15" S, 22° 01' 45" W 29. X NiooB 151-0 2141 2201 470 54° 42' 30" S, 26° 36' 45" W 30. X N 100 B 91-0 2118 2138 471 54° 57' S, 27° 59' 30" W 31. X- N70V 50-0 1750 1937 I. xi NiooB 168-0 2356 0016 472 53° 23' 8, 30° 29' 30" W I. xi N 100 B III-O 2030 2050 475 53° 30' 15" S, 42° 44' 30" W 12. xi NiooB 165-0 183s 1855 477 53° 35' 30" 8,41° 25' 45" W 13. xi N70V 50-0 100-50 0029 0055 NiooB 132-0 0134 0153 480 53° 40' 30" 8,39° 54' W 13. xi NiooB 161-0 1827 1849 481 53° 44' 15" S, 39° 29' 30" W 13-14. xi NiooB 139-0 375-140 0000 0000 0020 0030 482 53° 46' 45" 8, 39° 04' 45" W 14. xi NiooB 168-0 0519 0538 483 53° 54' 15" S, 38° 25' 30" W 14. xi N70V 50-0 100-50 1315 1350 NiooB IIO-O 1418 1438 484 53° 52' 15" S, 37° 05' 30" W 16. xi N70V 50-0 100-50 150-100 0738 0814 N 100 B 73-0 0825 0845 DEVELOPMENT OF YOUNG EUPHAUSIA SUPERB A Table I (cont.) Time Station Position Date Net Depth ^ V.'k./A VA V.'X A m. 8hot Hauled 1930 48s 53° 37' 15" S, 37° 18' W 16. xi N70V 50-0 180-100 1021 1055 N 100 B 133-° 1 124 1 144 486 53° 24' 15" S, 37° 29' 45" W 16. xi N70V 50-0 1326 1514 NiooB 124-0 375-124 1603 1603 1623 1637 487 53° 1 1' 30" S, 37° 41' 45" W 16. xi NiooB 108-0 2210 2230 488 52° 59' 45" S, 37° 48' W 17. xi NiooB III-O 370-1 1 1 0052 0052 0112 0120 N70V 50-0 0138 0331 489 52° 48' 15" 8,38° 04' W 17. xi NiooB 167-0 400-167 0950 0950 lOIO 1020 490 52° 35' 8,38° 13' 30" W 17. xi NiooB 140-0 1540 160I 491 52° 22' 8, 38° 22' 45" W 17. xi N70V 250-100 500-250 1849 2050 N 100 B 164-0 2121 214I 492 53° 12' 45" 8, 37" 04' 15" W 18. xi N 100 B 148-0 1312 1335 493 52° 35' 45" S, 35° 26' 15" W 18-19. xi NiooB 155-0 365-155 0135 0135 0155 0207 494 52° 50' 30" 8, 35° 35' 45" W 19. xi N 100 B 160-0 0826 0846 495 53°°4'45"S. 35' 43' 45" W 19. xi NiooB 186-0 1413 1437 496 53° 17' 45" S, 35° 56' W 19. xi N 100 B 155-0 2019 2039 498 54° 00' 30" 8, 36° 22' W 21. xi NiooB 137-0 1450 I5IO 499 53° 44' 30" 8, 36° 16' W 21. xi N 100 B 155-0 1804 1824 500 53° 30' 8, 36° 09' 30" W 21. xi N70V 500-250 2004 2250 NiooB 142-0 2308 2328 502 53° 47' S, 33° 51' 45" W 22. xi N 100 B 132-0 2129 2149 503 53° 53' 45" S, 34° 12' 45" W 23. xi N70V 50-0 0016 0212 NiooB II 5-0 0316 0336 504 54° 00' 45" S, 34° 33' 30" W 23. xi N70V 1000-750 0604 0740 506 54° 14' S, 35° 15' 30" W 23. xi NiooB 1 61-0 1803 1823 507 54° 19' 30" 8,35° 33' 30" W 23. xi N 100 B 75-0 2012 2032 508 55° OS'S, 33° 35' W 24. xi NiooB 181-0 1440 1500 509 55° 05' 30" 8, 34° 01' W 24. xi NiooB 1 12-0 2003 2023 510 4-8 miles N 70° E of Gierke Rocks, 8. Georgia 25. xi NiooB 149-0 0817 0837 511 54° 58' 15" S, 34° 51' 15" W 25. xi NiooB 71-0 1115 II35 513 54° 56' 8,35° 17' W 25. xi N 100 B 137-0 1410 1430 513 54° 53' 45" S, 35° 42' 30" W 25. xi N70V 50-0 1606 1626 NiooB 71-0 1651 17II 514 55° 51' S, 35° 32' W 26. xi NiooB 155-0 0540 0600 N70V 50-0 100-50 0625 0755 516 55° 25' 30" 8, 35° 52' 45" W 26. xi NiooB 123-0 1520 1540 517 54° 42' 45" S, 36° 36' 45" W 26. xi N70V 50-0 100-50 2123 2154 N 100 B 102-0 2216 2236 518 54° 58' 8, 36° 23' W 27. xi N 100 B 90-0 01 19 0139 519 55° 13' 8, 36° 09' 30" W 27. xi NiooB 137-0 0531 0551 520 55° 49' S, 39° 07' 15" W 28. xi NiooB 155-0 0253 0318 N70V 1000-750 0338 0504 DISCOVERY REPORTS Table I {cont.) T me Station Position Date Net Depth m. Shot Hauled 1930 521 55° 34' 30" S, 38° 43' w 28. xi N 100 B 164-0 1200 1220 522 55° 20' 30" S, 38° 19' w 28. xi N 100 B 95-0 1937 1957 523 55° 08' 45" S, 37° 59' 45" W 29. xi N 100 B 157-0 0413 0433 525 54° 36' 15" S, 37° 23' 30" W 29. xi NiooB IIO-O 1533 1553 528 55° 33' S, 30° 15' W 12. xii NiooB 173-0 2125 2145 537 61° 07' 30" S, 54° 26' w 19. xii N70V 250-100 500-250 750-500 1000-750 0440 0634 NiooB 137-0 0645 0705 538 61° 29' S, 54° 44' 15" W 19. xii N70V 750-500 1000-750 1003 1203 539 61° 48' S, 54° 51' 30" W 19. xii N70V 250-100 500-250 750-500 1000-750 1445 1652 540 62° 06' 15" S, 55° 08' 30" W 19. xii N70V 50-0 100-50 250-100 500-250 1926 2010 541 62° 22' S, 55° 23' W 19-20. xii N70V 50-0 100-50 250-100 2248 2320 542 62° 08' S, 57° 28' 30" W 20. xii N70V 500-250 0700 0845 543 62° 16' S, 57° 20' W 20. xii N 100 B 178-0 1205 1225 546 62°46'i5"S, 57°ii'i5"W 20. xii N70V 250-100 500-250 1919 2000 547 62° 59' 15" S, 57° 03' W 20. xii N 100 B 37-0 2316 2336 558 65°3i'S, 67° 07' 45" W 29. xii 193 1 N70V 100-50 200-100 2123 2154 585 67° 08' 30" S, 70° 15' 30" W 13- i N70V 50-0 0346 0425 618 59° 42' 45" S, 43° 57' 45" w 18-19. ii N70V 50-0 100-50 250-100 500-250 750-500 2235 Olio 620 59° 12' S, 40° 23' 30" w 19-20. ii N70V 50-0 100-50 250-100 750-500 1000-750 2229 2359 622 59° 05' 30" S, 36° 25' W 20-21. ii N70V 50-0 100-50 2225 2353 624 58° 34' 45" S, 31° 21' 30" W 21-22. ii N70 V 50-0 2235 0007 635 57° 42' 45" S, 50° 06' 15" W 7. iii N70 V 50-0 100-50 250-100 2017 2250 DEVELOPMENT OF YOUNG EUPHAUSIA SUPERB A Table I (cont.) Time Station Position Date Net Depth m. Shot Hauled 1931 636 59° 01' 45" S, 49° 18' 30" W 8. iii N70 V 250-100 500-250 750-500 0941 1230 637 60° 00' 15" S, 49° 28' 15" W 8. iii N70V 100-50 250-100 500-250 750-500 1000-750 1907 2028 638 61° 00' 30" S, 49° 48' 30" W 9. iii N70 V 250-100 500-250 750-500 1000-750 0537 0730 639 61° 57' 45" 8,51° 59' W 9. iii N70V 50-0 100-50 250-100 500-250 750-500 2107 2240 644 61° 20' 30" S, 56° 40' w II. iii N70V 100-50 250-100 0700 0732 646 60° 22' 30" S, 57°43' W II. iii N70V 250-100 500-250 1537 1770 647 59° 29' 15" 8,58° 39' 45" w 12. iii N70V 50-0 100-50 250-100 500-250 750^500 1000-750 Olio 0328 648 58° 30' 45" S, 59° 41' 15" w 12. iii N70V 50-0 100-50 250-100 500-250 1 106 1239 661 57° 36' S, 29° 54' 30" W to 57° 36' S, 29° 35' W 2. iv 1932 TYFB 360-0 0136 0226 852 58° 39-5' S, 40° 03-9' E 18. iv N 100 B 1 19-0 0429 0449 853 61° 00-2' S, 43° ii-i'E 19. iv NiooB 1 19-0 0435 °455 854 63° 30-2' S, 46° 24-9' E 20. iv NiooB 1 19-0 248-94 0342 0255 0402 0330 855 65° 10-4' 8,48° 43-7' E 20. iv NiooB 125-0 280-154 2310 2310 2330 2340 861 56° 28-9' S, 79° 18-2' E 28. iv NiooB 109-0 254-110 0019 0019 0039 0049 862 55°33-8'S, 83°oo-4'E 28. iv NiooB 102-0 220-98 2313 2313 2333 2343 887 63° 41-4' S, 130° 07' E 27. V NiooB 86-0 235-115 120-0 2119 2119 2202 2139 2149 2222 NiooH °-5 2213 2233 DISCOVERY REPORTS Table I {cont.) Time Station Position Date Net Depth m. Shot Hauled 1932 888 63° 23-2' S, 130° 29-7' E 28. V NiooB 98-0 0655 0715 240-90 0655 0725 912 6i°02'S, 158° 26' E 24. vi NiooH 0-2 1240 1310 954 62''i8-2'S, I28°i6-2'W 9. ix NiooH 0-2 1115 1 145 955 62° 17-2' S, 158° 13-2' W 9. ix NiooH 0-2 1210 1240 956 62° 12-8' S, 158° 11' W 9. ix NiooB 97-0 280-100 1351 1351 1411 1421 957 61° 56-3' S, 155° 49-6' W 10. ix NiooH 0-2 1046 1 1 16 959 6i°o7'S, 153° 57-2' W lo-ii. ix NiooH 0-5 2350 0012 NiooB 91-0 240-110 0012 0012 0032 0042 994 66° 457' S, 80° 19-8' W 29. X NiooH 0-5 2155 2255 995 67° o6-2' S, 79° 55-8' W 30.x N70B 320-120 0342 0412 996 66° 53-8' S, 78° 52-6' W 30.x NiooH 0-5 1735 1755 998 66° 407' S, 75° 137' W 31. X NiooH 3-4 1158 1230 999 66° 55-8' S, 73° 51-5' W 31.x N70B 1 51-0 2207 2227 NiooH o-S 2207 2237 1000 65°o6-6'S, 7i°397'W I. xi N 100 B 128-0 1 108 1128 lOOI 64° 53-8' S, 68° 43-9' W I. xi N 100 B 95-0 230-66 2143 2143 2203 2213 N 100 H 0-5 2147 2207 1002 64° 23-4' S, 65° 44-5' W 2. xi N70B 86-0 0952 1012 N 100 B 86-0 0952 1012 N70B 230-94 0952 1022 1003 63° 407' S, 63° 077' w 2. xi N70B 1 1 5-0 2050 — NiooB 1 15-0 ~ 2112 R.R.S. ' William Scoresby ' 1928 WS139 53° 00' S, 49° 50' w 10. ii N70V 500-250 1840 2240 WS 141 53° 32' S, 44° 52' W 13. ii N70V 250-100 500-250 1250 1543 WS147 53° 50' S, 35° 50' W 21. ii N70V 250-100 2120 2310 WS 182 55° 30' S, 34° 50' W 8. iii N70V 250-100 0355 0715 WS 197 56° 00' S, 40° 50' w 17. iv N70V 100-50 250-100 500-265 750-500 1000-800 0530 0915 WSI98 57° 31' S, 42° 52' W 19. iv N70V 250-100 500-250 1310 1610 WS 199 58° 10' S, 44° 10' W 20. iv N70V 500-250 1025 1255 WS200 59° 05' S, 46° 32' W 21. iv N70V 500-250 750-500 1220 1420 WS20I 59°57'S, 50° 12' W 22. iv N70V 50-0 100-50 250-100 500-250 1847 2255 WS202 60° 23' S, 52° 52' W 23. iv N70V 100-50 250-100 500-250 1845 2115 DEVELOPMENT OF YOUNG EUPHAUSIA SUPERB A Table I (cont.) 13 Time Station Position Date Net Depth m. Shot Hauled 1928 WS254 53° 03' S, 46° 58' W 22. viii NiooB 91-0 1339 1400 WS255 53° 23' S, 44° 10' W 22-23. vi" N70V 50-0 2335 0310 NiooB 96-0 0329 0350 WS2S6 53° 42' S, 40° 33' w 23-24. viii NiooB 1 00-0 0150 021 1 WS257 54° 04' S, 36° 18' W 27. viii NiooB 84-0 1728 1748 WS258 53° 56' S, 36° 06' W 27. viii NiooB 91-0 2055 2118 WS259 53° 49' S, 35° 53' 30" W 27-28. viii NiooB 103-0 0332 0353 WS260 53° 42' S, 35° 41' W 28. viii Nioo B 86-0 0515 0535 WS261 53° 34' 30" S, 35° 28' 30" W 28. viii N 100 B 104-0 1031 1051 WS262 53° 27' S, 35° 17' w 28. viii NiooB 85-0 1200 1223 WS263 53° 20' S, 35° 04' W 28. viii N70V 500-250 1320 1630 NiooB 87-0 1710 1730 WS264 53° 13' 30" S, 34° 51' W 28. viii NiooB 97-0 1848 1908 WS265 52° 40' S, 37° 05' W 29. viii N looB 130-0 1230 1250 WS266 52° 50' S, 35° 05' W 29. viii N70V 750-500 1000-760 1420 1640 NiooB 89-0 1713 1733 WS267 53° 01' S, 37° 05' W 29. viii NiooB 79-0 1845 1907 WS 268 53° II' S, 37° 05' W 29. viii N70V 250-100 2020 2300 NiooB 106-0 2321 2341 WS269 53° 21' S, 37° 05' W 30. viii NiooB 1 00-0 0100 0120 WS270 53° 31' S, 37° 05' W 30. viii N70V 50-0 0240 0415 N 100 B 1 1 6-0 0500 0520 WS271 53°42'S,37°05'W 30. viii NiooB 91-0 0721 0741 WS272 53° 52' S, 37° 05' W 30. viii NiooB 87-0 0955 1015 WS273 53° 24' S, 39° 00' W 4. ix N 100 B 124-0 0635 0655 WS274 53° 30' S, 38° 47' W 4. ix N70V 250-100 0810 1035 NiooB 1 13-0 1132 1152 WS275 53° 38' S, 38° 34' W 4. ix N70V 1 00-0 750-500 1305 1535 WS276 53° 45' S, 38° 22' W 4. ix N looB 104-0 2200 2220 WS278 54° 23' S, 35° 52' W 12. ix N 100 B 124-0 1938 1958 WS279 54° 23' S, 35° 35' W 13. ix NiooB 1 14-0 1818 1838 WS280 54° 23' S, 35° 18' W 17. ix NiooB 92-0 1506 1526 WS281 54° 23' S, 35° 00' W 17. ix NiooB 102-0 1646 1706 WS 282 54° 22' 30" S, 34° 43' W 17. ix N70V 50-0 1820 1919 NiooB 137-0 2052 2112 WS283 54° 22' S, 34° 25' W 17. ix NiooB 78-0 2320 2340 WS284 54° 21' 45" S, 34° 08' W 18. ix N70 V 50-0 0625 0732 NiooB 1 00-0 0848 0908 WS285 54° 21' 30" S, 33° 53' 30" W 18. ix NiooB 89-0 1021 1041 WS286 55° 00' S, 32° 55' W 18. ix N70V 50-0 1835 2050 NiooB 1 00-0 2202 2222 WS287 54° 54' S, 32° 08' W 19. ix N70V 50-0 0815 i°55 NiooB 164-0 1 1 50 1220 WS288 54° 52' S, 32° 19' w 19. ix NiooB 102-0 1429 1449 WS29I 54° 53' S, 35° 31' W 2. X NiooB 120-0 1227 1247 WS292 55° 02' S, 35° 16' 40" W 3-x NiooB 78-0 1604 1620 WS293 55° 09' S, 35° 05' W 3-x NiooB 64-0 1806 1823 WS294 55° 17' S, 34° 53' W 3-x N 100 B 98-0 1951 201 1 14 DISCOVERY REPORTS Table I {cont.) Time Station Position Date Net Depth m. Shot Hauled 1928 WS295 55° 23' 40" S, 34° 41' w 3-4- X N70V 50-0 2132 0108 NiooB 97-0 0140 0200 WS296 55° 31' S, 34° 29' W 4.x NiooB 1 1 0-0 0338 0359 WS297 55° 34' 05" S, 33° 14' W 4.x N70V 50-0 1020 1245 N 100 B III-O 1330 1350 WS298 55° 27' 30" S, 32° 21' 40" w 5-x NiooB 94-0 1008 1028 WS299 55° 18' 30" S, 32° 05' 20" W 5-x NiooB 1 00-0 1205 1225 WS 300 55° 07' 30" S, 31° 56-55" W 5-x NiooB 1 00-0 1736 1756 WS 301 55° 00' 05" S, 32° 08' 25" w 5x NiooB 88-0 1913 1937 WS302 54° 57' 20" S, 31° 49' 35" W 6.x NiooB 98-0 0522 0542 WS 303 54° 51' 25" S, 31° 20' 10" W 6.x NiooB 109-0 1312 1332 WS304 54° 54' 40" S, 30° 21' 20" W 6.x NiooB IIO-O 1843 1908 N70V 50-0 1920 2205 WS30S 54° 44' S, 29° 49' w 7.x NiooB 1 00-0 0945 1005 WS306 54° 41' 50" S, 30° 49' 35" W 7.x NiooB 105-0 1427 1447 WS307 54° 19' 30" S, 30° 31' 30" w 7.x NiooB 120-0 1751 1811 N70V 50-0 1840 2125 WS308 54° 04' 05" S, 30° 18' w 8.x NiooB 97-0 0830 0850 WS309 53° 58' 50" S, 28° 50' 10" W 8.x NiooB 91-0 1610 1630 WS3I0 54° 00' S, 28° 38' W 8.x NiooB 1 00-0 1800 1820 N70V 50-0 1835 2055 WS3II 54° 45' S, 35° II' W 10. x NiooB 1 16-0 1811 1839 WS323 53° 38' 30" S, 38° 35' 10" W 20-21. xii 1929 N70V 500-250 2205 0000 WS377 58° 34' S, 44° 47' W 9. ii N70V 100-50 1445 1710 WS380 60° 22' S, 50° 33' W 12-13. 11 N70V 250-100 500-250 2115 0015 WS382 62° 15' 35" S, 58° 18' 30" W 15. ii N70V 400-250 1335 1423 WS385 62° 32' S, 57° 55' W 16. ii N70V 500-250 0005 0215 WS394 62°5i'S, 6o°4o'W 18. ii N70V 200-100 0759 0830 WS427 53° 34' S, 40° 10' w 28-29. iv N70V 50-0 100-50 2155 2248 WS464 53° 40' S, 37° 33' W 31. X NiooB 135-0 2200 2220 WS468 55° 52' 8,56° 53' W 9-10. xi N 100 B 193-0 0055 0115 WS474 61° 03' S, 56° 42' w 13. xi NiooB 1 00-0 2301 2321 WS476 62° 16' S, 58° 18' w 14. xi N 100 B 91-0 2054 21 15 WS477 62° 20' 30" S, 58° 14' w 14-15. xi N70V 1000-500 2150 0140 N 100 B 140-0 0210 0230 WS478 62° 24' 30" S, 58° 06' 30" w 15. xi NiooB 146-0 1535 1555 WS480 62° 51' 30" S, 57° 47' 30" W 16. xi N70V 200-100 500-200 0430 0714 N 100 B 1 00-0 0729 0749 WS48I 62° 59' S, 57° 28' W 16. xi NiooB 109-0 1046 1 108 WS482 63° 10' S, 57° 16' 30" w 16. xi NiooB 54-0 1325 1336 WS483 62° 46' 45" S, 59° 37' 30" W 21. xi N70 V 500-240 0705 0905 WS484 62° 54' S, 59° 28' W 21. xi N70 V 750-500 950-740 1155 1352 DEVELOPMENT OF YOUNG EUPHAUSIA SUPERB A Table I [cont.) IS Station Position Date Net Depth m. Time Shot Hauled WS 485 63° 02' 30" S, 59° 17' W WS486 63° II' 30" S, 59° 13' W WS 487 63° 17' S, 59° 20' W WS 488 63° 51' 30" S, 62° 30' W WS 489 63° 38' S, 62° 32' W WS 490 63° 24' 30" S, 62° 35' 30" W WS491 63° 12' S, 62° 26' w WS 496 67° 14' S, 70° 12' w 1929 21. xi 21. xi 22. xi 22 xi 22 XI 22- 23. XI 23 XI 30 xu N70V NiooB N70V NiooB N70V NiooB NiooB N 100 B NiooB NiooB N70V NiooB 500-250 750-500 164-0 500-250 700-500 109-0 500-250 lOI-O 107-0 97-0 98-0 96-0 50-0 100-50 600-250 106-0 1635 — 1755 1900 1840 2020 2215 0430 0618 1755 2125 0010 0334 0950 1249 2135 2235 0545 0638 1815 2145 0030 0354 1125 1309 WS 497 67° 05' S, 70° 40' w WS498 66° 21' S, 69° 01' W WS 499 65° 45' S, 67° 18' W WS 501 64° 52' S, 63° 58' W WS 505 70° 10' 30" S, 87° 46' w WS 508 69° 04' S, 77° 40' W WS 524 53° 36' S, 43° 00' W WS 527 57° 30' S, 45° 35' W WS 529 56° 05' S, 53° 45' W 1930 1. i 2-3. i 3-i 3-i 4. ii 10. ii 2. iii 30. iii 2. IV NiooB NiooB NiooB N 100 B N70V NiooB N70 V N70V N70V 97-0 lOI-O 76-0 109-0 100-50 250-100 91-0 500-250 50-0 100-50 250-100 500-250 750-500 1344 0132 0903 2248 1118 0204 0750 0812 0930 1404 0152 0925 2308 1308 0224 0932 WS534 54° 17' 30" S, 35° 39' W WS538 57° 03' 30" S, 24° 32' w WS540 57° 55' S, 21° 21' W WS542 58° 39' S, 18° 13' W WS 560 56° 27' S, 28° 59' w WS 570 52° 22' S, 38° 23' w WS572 53° II' 30" S, 37° 41' W WS 573 52° 59' 45" S, 37° 48' W 193 1 22-23. ' 26. i 27-28. i 28. i 9. ii 8. iii 24. iii 25. iii N70V NiooB N70V N 100 B NiooB N70V N70V N70V 50-0 100-50 97-0 50-0 500-250 750-500 77-0 130-0 250-0 750-0 250-100 50-0 2315 — — 0000 1335 1355 2245 — 2235 2140 1700 1600 0127 0230 2255 2200 2015 1855 0350 i6 DISCOVERY REPORTS micro slide. Each larva was placed on the slide with the tail projecting into the angle, and was held up by the sides of the trough so that the total length from tip of rostrum to end of telson could be read quickly. The measurement of adolescents was done directly on a scale marked off in 0-5 mm. divisions. A difficulty was encountered as a result of the preservation of the specimens. The greater mass of the material was preserved in formalin, but nearly all the samples taken by the 'William Scoresby' were in 70 per cent alcohol. A comparison of specimens preserved in alcohol with others from the same stations preserved in formalin showed that a considerable shrinkage had taken place in the specimens kept in the former fluid. Table II Average length, mm. Station Numbers averaged A as percentage ofB Factor for expression of J as 5 (A) Preserved (B) Preserved in spirit in formalin WS263 WS282 WS286 94 100 100 9-99 13-23 II-20 10-94 14-47 12-47 91-4 91-4 89-8 1-096 1-094 1-113 WS288 WS296 WS487 100 50 100 12-04 14-75 15-98 12-99 15-79 17-28 92-7 93-4 92-4 1-079 1-070 1-085 The actual figures are expressed in Table II. In order to make the specimens more nearly comparable as regards length the factor i -09 was employed and records of length of spirit-preserved larvae were multiplied by it to bring them into line with the rest of the measurements. ACKNOWLEDGMENTS Whilst the work on this paper was in progress Dr Rustad published a paper on " Euphausiacea, with notes on their biogeography and development" (1930), and another "On the Antarctic euphausiids from the Norvegia Expeditions, 1929-30 and 1930-31 " (Rustad, 1934). Dr Johan T. Ruud (1932) also published his paper " On the biology of southern Euphausiidae ". I wish to make full acknowledgment to both of these authors for the use I have made of their publications, which are constantly referred to here. Acknowledgments are due to Dr S. W. Kemp, F.R.S., for much helpful criticism and advice, to Dr W. T. Caiman, C.B., F.R.S., to my colleague Mr D. Dilwyn John and former colleagues Dr N. A. Mackintosh and Dr H. E. Bargmann for the assistance they have given me. I am also greatly indebted to Mr G. E. R. Deacon for his valuable help where hydrographical problems were involved. 17 DEVELOPMENT THE EGG The eggs found in the plankton were identified as those of E. superba by comparison with others shed by gravid females, which from time to time were kept in tanks on board the R.R.S. ' Discovery II '. The appearance of the eggs is sufficiently distinct to enable them to be picked out with comparative ease from the plankton samples. They measure o-6 mm. in diameter when unpreserved, but after formalin has been used there is a decrease in size; for example the average diameters of two batches of a hundred each from two different catches at St. 540 were respectively 0-56 and 0-59 mm. The average sizes obtained from smaller numbers agree closely with these figures. The eggs are opaque and rather densely granular ; in reflected light their milky white appearance helps greatly to distinguish them. There is a thin, transparent, unsculp- tured membrane investing the contents of the egg, the latter completely filling the egg-shell. The average size of the eggs and the depths at which they are found do not seem to be correlated. At St. 356, 10. ii. 30, 250-100 m., ninety-five adult E. superba were obtained in the 70-cm. net, although none were caught in other nets either vertical or oblique. Most of the females were gravid, with spermatophores inserted in the thelycum. Four gravid females were placed in an aquarium, and on the following day the distension of the cephalothorax had subsided and a number of eggs were found in the water. On dis- secting a female many eggs were found floating in a milky fluid which appeared to con- sist of an emulsion of tiny oil globules. The eggs were of the same size as those in the aquarium, namely o-6 mm. diameter. At St. 548, 21. xii. 30, 102-0 m., eighty E. superba were taken, among them several gravid females. Two of these females were placed in a vessel containing sea water on the upper bridge of the ship, and on December 23, between 2 and 4 a.m., both shed their eggs. The females did not cast their skins beforehand and one still had the spermato- phores in situ. On December 29 the female without spermatophores died and the other died on the following day ; neither had moulted. Six gravid females were kept alive from St. 602, 19. i. 31, i lo-o m., and in due course eggs were deposited, but attempts to induce segmentation were unsuccessful. The eggs were divided into batches and subjected to varying degrees of temperature. Some were kept in an aquarium at laboratory temperatures, some in a plunger jar on deck at air temperature, and others in a vessel surrounded by lumps of ice. In all the experiments the water was filtered before the gravid females were placed in it. Eggs occurred in the plankton showing all stages of development, culminating in the clearly distinguishable form of the ist Nauplius, this latter completely filling the interior of the egg capsule. i8 DISCOVERY REPORTS Table III. Occurrence of eggs Station Date Depth m. Number Average diameter* mm. Remarks 356 10. ii. 30 250-100 — — Eggs from gravid female 537 19. xii. 30 250-100 2 0-55 (2) 500-250 8 0-57 (8) I adult in sample 750-500 27 0-58 (27) I adult in sample 1000-750 II 0-57 (10) 538 19. xii. 30 750-500 2 0-55 (2) 4 adults in sample and cast skin of adult 1000-750 8 0-58 (8) 539 19. xii. 30 250-100 3 0-57 (3) 500-250 7 0-59 (7) 750-500 6 0-57 (6) 1000-750 9 0-58 (9) 540 19. xii. 30 50-0 4 0-57 (4) 100-50 52 0-58 (51) 250-100 478 0-56 (100) 500-250 2496 0-59 (100) 5 adults in sample 541 19. xii. 30 50-0 I 0-56 (I) 100-50 5 0-55 (5) 250-100 I 0-56 (I) 542 20. xii. 30 500-250 I 0-56 (I) 546 20. xii. 30 500-250 4 0-58 (4) I adult picked out 558 29. xii. 30 100-50 I 0-58 (I) 3 other eggs not E. stiperba 585 13- i- 31 50-0 I o-6o (i) 618 ■18/19. ii. 31 250-100 I 0-58 (I) 637 8. iii. 31 250-100 I 0-55 (i) 644 1 1, iii. 31 100-50 4 0-59 (4) 250-100 6 0-59 (6) WS 147 21. ii. 28 250-100 I o-6o (i) WS323 20. xii. 28 500-250 I 0-58 (i) WS377 9. ii. 29 100-50 I 0-56 (1) At St. WS 376, 750-500 m., 4 adult males, 2 adult and gravid females with spermatophores attached WS380 12. ii. 29 250-100 2 0-56 (2) 500-250 2 o-6i (2) VVS382 15. ii. 29 400-250 2 0-55 (2) WS385 16. ii. 29 500-250 I o-6o (i) WS394 18. ii. 29 200-100 I o-6o (i) WS477 14. xi. 29 1000-500 I 0-58 (I) WS480 16. xi. 29 200-100 13 0-57 (12) 500-200 70 o-6o (10) 2 females with enlarged ovaries in this sample WS483 21. xi. 29 500-250 I 0-56 (i) WS484 21. xi. 29 750-500 I 0-56 (i) 950-75° I 0-55 (i) WS48S 21. xi. 29 500-250 3 0-54 (3) 750-500 2 0-58 (2) WS486 21. xi. 29 500-250 I 0-56 (i) 750-500 I 0-52 (i) WS487 22. xi. 29 500-250 I 0-56 (i) WS496 30. xii. 29 100-50 I 0-58 (I) 600-250 I o-6i (i) Shell broken WS505 4. ii. 30 250-100 2 0-56 (2) * Number from which the average was taken is shown in brackets. DEVELOPMENT OF EUPHAUSIA SUPERB A 19 FIRST NAUPLIUS As shown in Table IV there are only two records of free-swimming ist Nauplii in the material examined, but at some of the stations where eggs were found, notably at St. 537, many of the eggs contained well-developed naupliar forms. A Nauplius measuring 0-62 mm. in length (Fig. i a) was dissected from an egg, the interior of which it completely occupied. This differs somewhat from Ruud's description (1932, pp. 47, 49, fig. 11), in which the Nauplius in the egg is stated to be 0-51 mm. long and does not completely occupy the space within the egg-shell. Q b Fig. I. First Nauplius ( x 46). a, specimen dissected from egg; b, free-swimming specimen. Table IV. Occurrence of First Nauplius Station Date Depth m. Number Length mm. 537 539 19. xii. 30 19. xii. 30 750-500 250-100 I I 0-63 0-66 Of the two free-swimming larvae one measured 0-63 and the other o-66 mm. They were presumed to be E. superba from their relatively large size compared with other Nauplii in the plankton, from their similarity to nearly hatched Nauplii dissected from eggs of E. superba, and from their occurrence at stations containing recognizable eggs of this species. The body of the Nauplius is in dorsal aspect oval in shape, broader posteriorly than anteriorly (Fig. i b). It is in form a normal ist Nauplius with three pairs of swimming appendages. The first pair of limbs is uniramous with the two succeeding pairs biramous. SECOND NAUPLIUS All the 2nd Nauplii occurred at stations where eggs were present ; because of this, but particularly on account of their large size, they were assumed to be E. superba. The average length of the 2nd Nauplius is 0-67 mm. with a range of from 0-63 to 0-70 mm. 3-2 20 DISCOVERY REPORTS Table V. Occurrence of Second Nauplius Station Date Depth m. Number Length mm. Remarks 54° 546 558 WS480 WS 483 WS484 WS486 19. xii. 30 20. xii. 30 29. xii. 30 16. xi. 29 21. xi. 29 21. xi. 29 21. xi. 29 250-100 250-100 200-100 500-200 500-250 750-500 500-250 I I I I I I I 070 0-68 0-65 0-65 0-63 0-66 0-69 Female with enlarged ovaries in net A damaged specimen The outline of the body in dorsal aspect is roughly oval (Fig. zb), truncated at the posterior end and tending to be pointed, rather than rounded, at the anterior end. At each postero-lateral corner of the body there are two spines, a longer inner one about one-quarter to one-third of the length of the body, and a shorter outer one about one- tenth of the length of the first-mentioned spine. Three pairs of appendages are de- veloped (Fig. za). The antennules are uniramous with two terminal setae; the antennae Fig. 2. Second Nauplius ( x 46). a, lateral aspect; b, dorsal aspect. are biramous, having six terminal setae on the inner ramus and three terminal and one lateral on the outer. Posterior to the antennae are the biramous mandibles, each ramus having three setae distally. METANAUPLIUS In appearance the Metanauplius is very robust and of great size in comparison with other euphausian Metanauplii found in the same region. The average lengths of the larvae from different stations are given in Table VI. A hundred Metanauplii from St. 647, 750-500 and 1000-750 m. respectively, were measured and give normal length DEVELOPMENT OF EUPHAUSIA SUPERBA zi frequency curves (Table VII), with the two averages closely similar to one another, namely 0-97 and 0-92 mm. The length ranges between 0-90 and i -05 mm. in the first group and between 0-84 and 0-98 mm. in the second. The extreme range of size is be- tween 0-84 and I -08 mm. Ruud (1932, p. 47) states that the length of the body in the Metanauplii of £■. superba examined by him is from 0-72 to 0-82 mm. He also states that the carapace of all the specimens was rather swollen, so that his larvae must have been considerably smaller than any found in the present material. Table VI. Occurrence of Metanaiiplius Station Date Temperature range °C. Depth m. Number Average length mm. 193 28. iii. 27 — 0-26 to -0-66 500-250 I I -08 618 620 18/19. ii- 31 19/20. ii. 31 079 to 0-89 0-43 to i-oi 750-500 750-500 20 2 0-92 0-98 636 8. iii. 31 0-70 to 1-46 1000-750 750-500 4 17 0-90 0-91 (13)* 637 8. iii. 31 0-03 to —0-22 750-500 24 0-94 638 9. iii. 31 -0-19 to -0-85 1000-750 250-100 10 It 0-93 750-500 3 0-93 639 9. iii. 31 -0-76 to -0-99 1000-750 50-0 4 2 0-91 0-95 647 12. 111. 31 1-34 to 2-01 750-500 3702 0-97 (100)* 648 12. iii. 31 i-ii to 3-76 1000-750 50-0 100-50 577 0-92 (100)* 0-85 0-93 (47)* WS 197 17. iv. 28 1-33 to 1-27 250-100 500-250 1000-750 4 9 115 0-93 0-94 (3)* 0-96 (88)* * The figure in brackets is the number from which the average length was calculated when all specimens were not measured, •j- Larva damaged. Table VII. Length frequencies of Metanaiiplius from St. 647 750-500 m. 1000-750 m. Micro- divisions Length mm. Frequency Micro- divisions Length mm. Frequency 56 57 58 59 60 61 62 63 64 65 0-90 0-92 0-94 °-95 0-97 0-98 i-oo 1-02 1-03 1-05 4 2 10 22 41 5 7 3 4 2 52 53 54 55 56 57 58 59 60 61 0-84 0-87 0-89 0-90 0-92 0-94 0-95 0-97 0-98 2 5 16 21 25 17 7 5 2 Av. length 0-97 mm. Av. length 0-92 mm. 22 DISCOVERY REPORTS The carapace (Figs. 3 a, b) which envelops the anterior portion of the body and part of the abdomen is without marginal spines or processes. It is tucked in under the body posterolaterally, so that when it is dissected and flattened out (Fig. 3 d) the pos- terior portion is wider than the anterior. At the anterior margin of the body underneath the carapace a small pair of frontal sense organs can be distinguished. There are traces Fig. 3. Metanauplius. a, dorsal aspect ( x 46) ; d, extended carapace ( x 46) ; b, lateral aspect ( x 46) ; e, telson ( x 100) ; c, ventral aspect ( X 46); /, antenna ( x 100). In Figs. 3 b and c the bristles on the appendages have been omitted. of the paired eyes of the adult and of the median eye, although the latter is not easily seen. The compound eyes are in the form of two roughly spherical masses, and each contains a bunch of fibres within its tissues which is the developing luminous organ of the ocular peduncle (cf. Metschnikoft', 1869, pp. 479, 481, fig. xxxvi). The abdomen is DEVELOPMENT OF EUPHAVSIA SUPERB A 23 short, tapering in lateral view and slightly bilobed in dorsoventral aspect. The posterior margin (Fig. 3 e) is armed with twelve spines, of which the pair on either side of the emargination are very short and incompletely separated from one another. The antennules are unsegmented and uniramous, bearing three long bristles at their tips. The antennae (Fig. 3/) are biramous and also furnished with long bristles. The mandibles are in this stage reduced and bud-like in appearance. Maxillae I and II are rudimentary single-lobed buds, and the first thoracic legs are small bilobed processes without setae. Anterior to the mandibles lies the labrum and posteriorly between the mandibles and the ist maxillae the bilobed labium (Fig. 3 c). FIRST CALYPTOPIS The average lengths and the ranges of length for varying numbers of ist Calyptopis are stated in Table VIII. The extreme range for 500 individuals from an oblique net fished at St. 647 is between 1-33 and 1-92 mm. The frequency distribution of these larvae is as follows: Length, mm. 1-33 1-38 1-41 1-46 1-50 1-54 1-58 1-63 1-67 171 1-75 179 1-83 i-88 1-92 Frequency 3 i i 3 7 19 41 51 78 62 78 73 67 15 i Their average length is 171 mm. The lengths of all the other larvae fall within the limits of the stated range, as do the lengths of those examined by Ruud (1932, p. 47). The carapace is without teeth or projections (Figs. 4 «, ^). The median eye is present, and anterior to it are indications of the compound eyes. The rudiment of the ocular luminous organ can clearly be distinguished as a dark mass of fibres within the de- veloping compound eye. There is also a pair of frontal sensory organs anteriorly. The appendages, with one or two small diff'erences in detail, are as described by Sars (1885) in Nyctiphanes australis. The two sensory appendages on the antennules are difficult to see in unstained specimens. The antenna (Fig. 4 c) is composed of a two- segmented peduncle and, distally, an inner and outer ramus as in the Metanauplius. In the mandible (Fig. 4 d) the dentiform projection between the dentate part and the molar protuberance is present ; also, as in N. australis, the denticulated plate — the lacinia mobilis. At the base of the projection, where the plate is inserted, there is a bunch of setae surrounding it. The palp of the ist maxilla consists of two segments as mentioned by Ruud (1932) ; it bears five spines as opposed to six in A^. australis, three being at the tip of the distal segment and two on the inner margin of the proximal segment. The inner masticatory lobe is furnished with seven spines of varying structure and the outer lobe with three (Fig. 4 e). The larval exopod is as in Sars' description. The 2nd maxilla (Fig. 4/) diff'ers only from Sars' description in having three instead of four bristles on the terminal segment. The ist thoracic limb (Fig. 4^) differs from that figured by Sars for N. australis in having a short bristle at the articulation of the exopodite and the basipodite on the external margin. The posterior portion of the body, behind the wrinkling of the integument which foreshadows the segmentation of later stages, is shorter than the anterior part. The abdomen is furnished with a pair of spines situated m9rginally but rather ventrally about 24 DISCOVERY REPORTS one-third of the abdominal length from the distal end (Fig. 4 //). At the posterior end of the abdomen there are three pairs of postero-lateral spines increasing in length from the outer to the inner, and six terminal spines which increase in length from the inner- most pair outwards. The longest terminal spine is about half the length of the longest postero-lateral. a, dorsal aspect ( x 20) ; b, lateral aspect ( x 25) ; c, antenna ( x 65); Fig. 4. First Calyptopis. d, mandibles ( x 335) ; g, telson ( x 70) ; e, masticatory lobes of the first maxilla ( x 315); /, second maxilla ( x 70); h, first thoracic (limb^ x 70. DEVELOPMENT OF EUPHAUSIA SUPERB A Table VIII. Occurrence of First Calyptopis 25 Station Date Depth No. of No. Length range Average length m. larvae measured mm. mm. i6i 14. ii. 27 100-50 I I 171 1-71 169 22. ii. 27 500-250 3 3 I -54- 1 -79 1-65 194 28. iii. 27 100-50 3 3 I-79-I-88 1-85 250-100 3 3 I-75-I-88 1-79 198 3. iv. 27 50-0 I 1-92 1-92 202 ■ 5- iv. 27 50-0 I 1-88 1-88 302 21. i. 30 500-250 I 1-75 1-75 305 21. i. 30 100-50 2 I-63-I-67 1-65 313 25- i- 30 750-500 I 171 1-71 319 29-30. i. 30 100-50 I 1-54 1-54 320 30. i. 30 500-250 I 1-63 1-63 332 31- '■ 30 250-100 5 i-5o-f63 I '54 356 10. ii. 30 250-100 I 1-58 1-58 357 10. ii. 30 100-50 3 1-67-171 1-70 250-100 2 175-1-88 1-82 365 2. iii. 30 250-100 I 175 1-75 378 13. iv. 30 50-0 I 1-79 1-79 100-50 I 1-75 1-75 383 14. iv. 30 100-50 122 50 1-63-1-79 1-69 250-100 24 6 1-63-1-88 1-72 618 18-19. "■ 31 50-0 142 99 I-58-I-88 I-7S 100-50 196 90 I -46-1 -88 1-70 250-100 5 5 1-67-1-79 1-74 750-500 4 4 1-58-1-67 1-63 620 19-20. ii. 31 50-0 59 59 1-46-1-79 1-66 100-50 148 100 1-50-1-83 1-70 250-100 II II 1-46-1-88 1-64 750-500 10 10 1-42-1-75 1-58 1000-750 6 6 I -50-1 -67 1-58 622 20-21. ii. 31 50-0 30 5 1-67-1-75 1-70 100-50 II II 1-58-1-79 1-68 636 8. iii. 31 250-100 5 5 1-58-1-75 1-69 500-250 31 28 I -46-1 -88 1-64 750-500 I I 1-58 1-58 637 8. iii. 31 100-50 2 I 1-54 1-54 250-100 21 21 I -42-1 -75 1-59 500-250 66 63 1-46-1-83 1-68 750-500 2 — — — 638 1000-750 4 4 1-58-1-71 1-67 9. iii. 31 500-250 4 4 I-58-I75 1-67 639 750-500 2 2 1-54 1-54 9. iii. 31 50-0 2 2 175-179 1-77 100-50 117 100 1-42-1-88 1-63 250-100 91 86 I-46-1-88 1-68 500-250 3 3 1-58-175 1-65 646 750-500 5 5 1-67-1-83 1-74 II. iii. 31 250-100 I I 1-79 1-79 500-250 2 2 1-71-1-75 1-73 647 12. iii. 31 50-0 361 100 I -46-1 -83 1-63 100-50 20 20 1-46-1-75 1-62 250-100 94 87 1-46-1-88 1-69 500-250 150 100 1-42-1-79 1-60 750-500 282 50 1-42-1-88 1-68 648 1000-750 7 7 I -50-1 -79 1-67 12. iii. 31 100-50 7 7 1-58-171 1-64 WS139 500-250 3 3 I-63-I-67 1-66 10. ii. 28 500-250 2 2 1-54-175 1-65 WS 141 13. ii. 28 250-100 I I 1-75 1-75 WS 182 8. iii. 28 250-100 2 2 1-58-1-67 1-63 WS 197 17. iv. 28 250-100 36 36 1-50-1-88 1-70 500-250 28 28 1-50-1-88 1-71 750-500 2 2 1-58-1-88 1-73 WS 198 19. iv. 28 250-100 I I i'7S i'75 WS572 8. iii. 31 250-100 I I 1-75 1-75 26 DISCOVERY REPORTS SECOND CALYPTOPIS The average lengths and ranges of length for varying numbers of 2nd Calyptopis are stated in Table IX. The range of forty-eight larvae, the largest number measured in one sample, is between 2-38 and 2-96 mm. and the average is 2-71 mm. The smallest larva measured was 2-13 mm. and the largest 3-33 mm. The carapace is evenly rounded and there are no lateral denticles (Figs. 5 a, b). The abdomen is segmented, and in lateral view the developing uropods can be seen within the integument. The telson (Fig. 5 c) is furnished with three postero-lateral spines on Fig. 5. Second Calyptopis. a, dorsal aspect ( x ig); b, lateral aspect ( x 25); c, telson ( X 80) ; d, antennule ( x 80) ; e, masticatory lobes of the first maxilla ( x 287). each side of the seven terminal spines. A spine is situated laterally on each side of the telson, as in the ist Calyptopis. Both lateral and postero-lateral spines carry a diminutive dorsal spinule just beyond the middle of their length, as described by Rustad in Euphausia frigida. Posterior to the ist pair of thoracic appendages are situated ventrally and side by side two roughly hemispherical lobes enclosed by the integument containing the buds of the following five pairs of thoracic legs. These lobes are a distinguishing feature of the 2nd Calyptopis of E. superba, and their presence is, of course, associated with the exceptionally early development of the thoracic appendages II-VI in this DEVELOPMENT OF EUPHAUSIA SUPERB A Table IX. Occurrence of Second Calyptopis 27 Average Station Date Depth No. of Number Length range length m. larvae measured mm. mm. 161 14. ii. 27 100-50 6 6 2-92-3-2I 2-98 167 20. ii. 27 250-100 I I 271 271 169 22. ii. 27 500-250 3 3 2-50-275 2-6l 187 iS. iii. 27 100-50 I I 2-92 2-92 193 28. iii. 27 500-250 1 1 7 2-92-3-17 3-07 750-500 I I 3-00 3-00 194 28. iii. 27 100-50 29 26 2-42-3-13 2-80 250-100 17 17 2-42-3-00 2-72 197 3. iv. 27 100-50 I I 2-92 2-92 750-500 I I 2-92 2-92 1000-750 2 2 2-92-2-96 2-94 198 3. iv. 27 50-0 22 21 2-63-3-13 2-94 100-50 5 5 2-71-3-08 2-92 1000-750 I I 2-92 2-92 199 3. iv. 27 500-250 2 2 3-00-3-08 3-04 202 5. IV. 27 50-0 3 3 2-54-2-79 2-68 204 6. iv. 27 500-250 3 3 3-00-3-13 3-07 205 6. iv. 27 215-100 2 2 2-92 2-92 206 6. iv. 27 250-100 2 2 3-00-3-13 3-07 209 14. iv. 27 150-100 I I 2-75 2-75 302 21. i. 30 250-100 2 2 2-50-2-58 2-54 500-250 6 6 2-71-2-92 2-79 303 21. i. 30 500-250 33 25 2-50-2-96 2-69 750-500 I I 271 271 304 21. i. 30 500-250 2 2 2-63 2-63 305 21-22. i. 30 100-50 5 5 2-46-2-63 2-54 250-100 4 4 2-58-2-71 2-6l 500-250 2 2 2-63-2-83 2-73 313 25- i- 30 250-100 I I 2-63 2-63 319 29-30. 1. 30 50-0 5 I 2-58 2-58 320 30. i. 30 250-100 I I 2-42 2-42 500-250 28 17 2-42-2-67 2-54 321 30-31- »• 30 100-50 3 3 2-46-2-50 2-49 323 31- '• 3° 500-250 5 5 2-50-2-67 2-59 1000-750 I I 2-67 2-67 332 2-3. ii. 30 250-100 2 2 2-42-2-50 2-46 344 7-8. ii. 30 100-50 2 2 2-42-2-71 2-52 345 8. ii. 30 180-100 I I 2-58 2-58 356 10. ii. 30 250-100 8 8 2-58-2-96 2-74 500-250 I I 2-63 2-63 357 10. ii. 30 100-50 I I 2-29 2-29 250-100 2 2 2-42-2-58 2-50 358 II. ii. 30 50-0 II II 2-58-2-75 2-67 100-50 3 3 2-50-2-71 2-61 361 25. ii. 30 100-50 I I 2-79 2-79 362 25. 11. 30 50-0 48 48 2-38-2-96 271 250-100 I I 2-79 2-79 365 2. iii. 30 50-0 2 2 2-96-3-00 2-98 250-100 44 44 27I-3-33 2-99 500-250 5 5 2-92-3-17 3-04 378 13. iv. 30 50-0 6 6 2-67-2-96 2-84 100-50 2 2 2-88-2-92 2-90 4-2 28 DISCOVERY REPORTS Table IX {cont.) Average Station Date Depth No. of Number Length range length m. larvae measured mm. mm. 383 14. iv. 30 100-50 23 23 2-50-2-96 2-71 250-100 80 20 275-3-I3 2-95 477 12. xi. 30 100-50 I 271 2-71 618 18-19. ii. 31 50-0 160 31 2- 1 3-2-67 2-41 100-50 21 21 2- 17-2-63 2-41 620 19-20. ii. 31 750-500 I 2-50 2-50 1000-750 I 2-38 2-38 622 20. ii. 31 100-50 I 2-58 2-58 63s 5. iii. 31 50-0 6 6 2-46-2-79 2-60 100-50 6 6 2-33-2-63 2-49 250-100 2 2 2-25-2-67 2-46 636 8. iii. 31 250-100 3 3 2-38-2-46 2-42 500-250 4 3 2-50 2-50 639 9. iii. 31 50-0 I I 2-67 2-67 100-50 39 33 2-38-2-92 2'58 250-100 25 22 2-21-2-83 2-52 500-250 I I 2-21 2-21 750-500 I I 2-88 2-88 646 1 1, iii. 31 250-100 I I 2-67 2-67 500-250 4 4 2-63-2-71 2-67 647 12. iii. 31 50-0 6 6 2-63-2-92 2-74 500-250 I I 2-54 2-54 854 20. iv. 32 1 19-0 5 5 2-75-2-83 2-78 248-98 I I 2-92 2-92 855 20. iv. 32 125-0 I I 3-04 3-04 125-0 50 10 2-83-3-I3 2-94 280-154 4 4 2-71-3-00 2-88 WS139 10. ii. 28 500-250 I I 2-88 2-88 WS141 13. ii. 28 250-100 I I 2-75 275 WS 173 6. iii. 28 100-50 I I 2-63 2-63 WS 182 8. iii. 28 250-100 2 2 2-42-2-79 2-61 WS 197 17. iv. 28 250-100 13 13 2-50-2-92 2-67 500-250 8 8 2-33-2-7I 2-56 WSS27 30. iii. 30 250-100 1 I 2-42 2-42 species. The size of the lobes varies with the size of the larvae, being inconspicuous in the very small larvae and large and easily noticeable in the larger specimens. The com- pound eyes are more distinctly defined but do not project beyond the margin of the carapace. The fibres of the developing luminous organ at the base of the eyes can be dis- tinguished in the posterior region of the ocular mass. The antennule consists of a three-jointed peduncle with two rudimentary flagella distally, the one rather larger than the other (Fig. 5 d). The sensory filaments on the outer flagellum are situated terminally; there are none on the inner flagellum. The antennae and mandibles are as in the previous stage. The first maxilla (Fig. 5 e) is also similar to that found in the ist Calyptopis except that the outer masticatory lobe bears five spines instead of three and the palp is indistinctly two-segmented. The 2nd maxilla and the ist thoracic appendage are, in form, as in the previous stage. DEVELOPMENT OF EUPHAUSIA SUPERB A 29 THIRD CALYPTOPIS The average lengths of varying numbers of individuals are stated in Table X. The extreme range of size is between 3-17 and 4-83 mm. The range for the largest number (fifty-eight) measured from one sample is between 3-50 and 4-58 mm. The lower limit of length is almost i mm. less than that given by Ruud (1932, p. 48), but the upper limit agrees closely with his. Fig. 6. Third Calyptopis. a, lateral aspect ( x 18); /, mandible ( x 165); b, dorsal aspect ( x 14); g, first maxilla ( x no); c, rostrum and antennules, flagella omitted ( x 35); /;, first thoracic limb with limbs II-VI in situ ( x 35); d, uropod ( X 46); J, thoracic limbs II-VI ( x 35). e, telson, ventral aspect ( x 35); The carapace is without lateral denticles (Fig. 6 a). In dorsal aspect (Fig. 6 b) it com- pletely covers the compound eyes, which are globular in shape and show traces of pig- 30 DISCOVERY REPORTS Table X. Occurrence of Third Calyptopis Station Date Depth m. No. of larvae Number measured Length range mm. Average length mm. i6i 14. ii. 27 250-100 I I 3-96 3-96 169 22. ii. 27 500-250 7 7 3-67-4-38 4-00 750-500 I I 4-79 479 187 18. iii. 27 100-50 I I . 4-17 4-17 193 28. iii. 27 500-250 77 50 4-08-4-79 4-48 750-500 4 4 4-17-4-46 4-40 194 28. iii. 27 100-50 63 58 3-50-4-58 4-06 250-100 59 48 3- 54-4-42 3-84 500-250 I I 4-04 4-04 197 3. iv. 27 50-0 I I 4-29 4-29 100-50 7 7 3-96-4-46 4-29 1000-750 2 I 3-92 3-92 198 3. iv. 27 50-0 34 33 3-83-4-83 4-42 100-50 3 3 4-25-4-38 4-32 750-500 I I 4-42 4-42 1000-750 I I 4-17 4-17 199 3. iv. 27 500-250 4 4 4-04-4-58 4-28 75c^5oo 3 3 4-o8-4-2i 4-15 202 5. iv. 27 50-0 7 7 4-00-4-33 4-15 500-250 2 2 4-04-4-25 4-15 203 5- jv. 27 250-100 I I 3-83 3-83 204 6. iv. 27 500-250 2 2 4-58-4-7S 4-67 205 6. iv. 27 215-100 2 2 4-04-4-08 4-06 300 20. i. 30 500-250 15 3 3-58-3-88 376 302 21. i. 30 250-100 2 2 3-96-4-17 4-07 500-250 2 2 3-96-4-04 4-00 303 21. i. 30 500-250 90 50 3-67-4-29 4-01 304 21. i. 30 500-250 22 22 3-63-4-33 4-05 305 21-22. i. 30 100-50 I I 3-54 3-54 250-100 I I 4-29 4-29 313 25- i- 30 250-100 5 4 4-00-4-29 4-10 319 29-30. i. 30 250-100 3 3 3-67-4-17 3-92 320 30. i. 30 500-250 60 40 3-67-4-04 3-87 321 30-31- i- 3° 100-50 I I 375 375 1000-750 I I 3-96 3-96 323 31- '• 30 500-250 5 4 3-54-3-83 3-64 324 I. ii. 30 250-100 I I 3-5° 3-50 500-250 I I 3-42 3-42 332 2-3. ii. 30 250-100 3 3 3 -54-3 75 3-67 335 4-5. u. 30 50-0 I I 3-63 3-63 344 7-8. ii. 30 100-50 I I 3-54 3-54 345 8. ii. 30 180-100 5 5 3-54-4-08 3-80 356 10. ii. 30 250-100 8 8 4-25-4-54 4-41 357 ID. ii. 30 100-50 I I 3-83 3-83 358 II. ii. 30 50-0 6 6 3-79-3-96 3-88 100-50 I I 4-04 4-04 360 24. ii. 30 100-50 I I 379 379 361 25. ii. 30 50-0 2 2 3-9M-38 4-17 100-50 I I 3-96 3-96 250-100 18 9 3-54-4-2I 4-°3 362 25. ii. 30 50-0 138 50 3-33-4-25 378 100-50 2 2 3-58-4-08 4-33 250-100 50 50 3-38-4-46 3-94 DEVELOPMENT OF EUPHAUSIA SUPERB A Table X (cont.) 31 Station Date Depth m. No. of larvae Number measured Length range mm. Average length mm. 365 2. iii. 30 50-0 II II 4-08-475 4-40 250-100 441 50 4-00-4-75 4-39 500-250 10 10 4-04-479 4-43 368 8. iii. 30 50-0 2 2 3-96-4-38 4-17 250-100 II II 4-00-4-33 4-13 500-250 I* — — — 369 9. iii. 30 50-0 I I 4-13 4-13 250-100 12 12 4-08-4-38 4-27 378 13. iv. 30 50-0 I I 4-17 4-17 383 14. iv. 30 100-50 17 14 3 •46-4- 17 3-84 250-100 384 50 3-88-4-58 4-17 618 18-19. "■ 31 50-0 2 2 3-50 3-50 635 7. iii. 31 50-0 9 9 3-I7-3-50 3-33 100-50 2 2 3-29-3-50 3-40 250-100 I I 3-25 3-25 636 8. iii. 31 500-250 I I 3-29 3-29 853 19. iv. 32 1 1 9-0 3 3 3-96-4-2I 4-07 854 20. iv. 32 1 19-0 12 7 3-67-3-96 3-84 248-98 2 2 375-379 377 85s 20. iv. 32 125-0 23 23 4-o8~475 4-35 125-0 420 50 3-88-4-58 4-27 280-154 6 6 4-00-4-38 4-II WS141 13. ii. 28 500-250 I I 3-96 3-96 WS 197 17. iv. 28 100-50 I I 4-17 4-17 250-100 3 3 3 •96-4- 17 4-06 500-250 3 3 3 •96-4- 1 3 4-03 WS 198 19. iv. 28 250-100 2 2 4-o8-4-i7 4-13 500-250 I I 4-04 4-04 WS 199 20. iv. 28 500-250 II II 3-83-4-38 4-13 WS200 21. iv. 28 500-250 10 10 3-88-4-33 4-10 WS 201 22. iv. 28 50-0 8 8 4-o8-4-46 4-25 100-50 2 2 4-17 4-17 250-100 I I 4-04 4-04 WS527 30. iii. 30 250-100 I I 3-17 3-17 * Specimen damaged — unmeasurable. ment. In lateral view the carapace anteriorly is more angular than in the 2nd Calyptopis, and in preserved specimens the eyes may project a very little way beyond its margin. In some of the larger 3rd Calyptopis a clearly defined hyaline area can be distinguished just above each eye in the carapace, a foreshadowing of the emarginations which are present in all later stages. The abdomen is more elongate than in the 2nd Calyptopis, seven-segmented and longer than the anterior portion of the body. Each uropod (Fig. 6 d) consists of a short basal segment bearing two rami, of which the outer is longer than the inner. The outer ramus carries a large spine externally at the tip and two rather small weak spines in- ternally. The inner ramus has a small spine distally. The arrangement of spines at the end of the telson (Fig. 6 e) is similar to that of the 2nd Calyptopis. 32 DISCOVERY REPORTS In the antennule the peduncle consists of three segments, of which the basal is pro- longed externally into a strong spine, extending to, or a little way beyond, the distal margin of the distal peduncular segment (Fig. 6 c). The flagella are not greatly altered from the form found in the 2nd Calyptopis ; the outer flagellum bears two sensory fila- ments terminally. The antenna, mandible, ist and 2nd maxilla and the ist thoracic ap- pendage (Figs. ()f,g, h) are unchanged. Behind the ist thoracic appendages are five pairs of bilobed "sausage-shaped" thoracic rudiments (Fig. 6^). In each limb the endopod is longer than the exopod. The posterior limbs are shorter than the anterior ; there is no trace of segmentation or setae on the limbs. In Fig. 6 // the ist thoracic limb has been shown along with the succeeding five limbs to compare the difference in size. The luminous organ at the base of the eye, recognizable as a bunch of fibres, is the only one that can as yet be distinguished. SUMMARY OF PREVIOUS KNOWLEDGE RELATING TO THE FURCILIA AND CYRTOPIA STAGES In the report on the Schizopoda collected by H.M.S. ' Challenger' (Sars, 1885), in the section dealing with development of Euphausiidae, it is stated that the three schizo- podous genera, Calyptopis, Furcilia and Cyrtopia of Dana were proved by Claus (1863) to be different stages in the development of Euphausiidae. Metschnikoff (1869, 1871) describes still earlier stages, which Sars in accordance with earlier authors designates the Nauplius and Metanauplius stages. For the three succeeding stages Sars applies the generic denominations suggested by Dana. It is the last two of these stages, namely Furcilia and Cyrtopia, which are to be discussed here. They are described by Sars as follows : Furcilia stage. Compound eyes more fully developed (than in Calyptopis), mobile, and projecting beyond the sides of the carapace. Antennae still retaining their original structure, natatory. Anterior pairs of legs and pleopoda successively developing. Cyrtopia stage. Antennular flagella becoming elongate and distinctly articulate. Antennae trans- formed so as not to serve the purpose of locomotion. Posterior legs and gills successively appearing. Within the division which he terms Furcilia, Sars in his description of Euphausia pelhicida, Dana (= E. superba), recognizes three stages which he names respectively first, intermediate and last Furcilia. Similarly in the Cyrtopia of this species he recognizes two stages, a first and last. The distinction between different stages of both Furcilia and Cyrtopia is based mainly on the degree of development of the appendages. Referring to the development of Thysanopoda tricuspidata Sars states (p. 167) that there are several successive stages of Furcilia in the collection. He only refers to one Cyrtopia stage. In Nematoscelis rostrata, G. O. Sars, he recognizes two Furcilia stages and one Cyrtopia. Brook and Hoyle (1888), describing the metamorphosis of British Euphausiidae, state that in the Furcilia of one species there are eleven moults judging from the comparative development of the pleopods, thus : DEVELOPMENT OF EUPHAUSIA SUPERB A 33 Stage I. No rudiments of pleopods. 2. First pair of pleopods as simple rudiments 3- Second „ „ „ „ „ 4- Third ,, „ „ „ „ 5- Fourth „ „ „ „ „ 6. rittn ,, ,, ,, ,, ,, 7- First pair of pleopods biramous and setose 8. Second „ „ „ „ ,, 9- Third „ „ „ „ „ 10. Fourth „ „ „ „ „ II. Fifth „ „ „ „ They indicate also that there is more than one Cyrtopia stage, and remark that before the adult stage is reached a large number of ecdyses must take place. Lebour (1926 c) in a paper entitled "A general survey of larval euphausiids, with a scheme for their identification", recognizes a succession of Furcilia forms commencing with that in which no pleopods are present and followed by others during which the pleopods develop successively until all are setose and biramous. It is stated in this paper that "the pleopods develop in different ways in several different genera, of which the first three or four (forms) are the same and the last two or three are the same, but in be- tween there are different orders of development some of which appear to be character- istic of certain genera". In Thysanoessa, for instance, all five pleopods are simple buds before any are setose. In Euphausia krohnii the first is setose whilst four are simple. The commonest group found, it is stated, is one which occurs in five genera — Nycti- phanes, Meganyctiphanes, Thysanopoda, Euphausia, and Nematoscelis — in which the first pleopod is setose with the three following simple and the last not yet formed. In all the genera with the exception of Euphausia the second pleopod becomes setose before the last bud appears. Finally in Stylocheiron the first pleopod is setose with only two buds behind. This paper later states that the distinctive stages nearly always seem to occur as though certain stages were dominant. Thus in Stylocheiron it is Furcilia 7, that is, according to the pleopod development, a form having three pairs of pleopods one pair of which is setose. In Thysanoessa it is Furcilia 6, that is, a form having five pairs of non-setose pleopods. In Euphausia krohnii it is Furcilia 9, a form in which the pleopod arrangement is one pair setose and four pairs simple. Lebour's paper draws attention to the jumping of stages in some species; it is stated that " Mr Elmhirst and Mr Mac- donald from Millport tell me that Meganyctiphanes may jump several stages but they always jump into a stage known for that genus. Mr Elmhirst has provided me with some notes on Meganyctiphanes (reared in aquaria) in which one jumped from one to three pairs of simple pleopods and one which jumped from three pairs of simple pleopods to three pairs setose and two pairs simple." Macdonald (1927 b, p. 785), in his paper on "Irregular development in the larval history of Meganyctiphanes norvegica", states that the Furcilia is to be recognized by having the eyes no longer covered by the carapace and by the appearance of the pleopods. He distinguished eleven stages of Furcilia, certain of which tended to be dominant and others to be suppressed. He also found that " those Furcilia stages which were observed 34 DISCOVERY REPORTS to be sometimes omitted during development in captivity were less frequent m the plankton than the other Furcilia stages ". "On the other hand ", he states, "there were certain stages which predominated in numbers in the plankton." The graph of the fre- quency of distribution of his eleven Furcilia stages for 302 larvae shows a distinctly bimodal curve, having the one maximum at stage 4 (i.e. three pairs of simple pleopods) and the second at stage 9 (three pairs of setose pleopods and two pairs simple). It is also interesting in this paper to note that of twenty-six larvae kept in aquaria more than half moulted irregularly. Attention is drawn by Macdonald to further evidence of curtailed larval history pro- vided by another euphausiid, Nemntoscelis microps, writing of which Lebour states (1926 d, p. 766): "The youngest stage found is presumably the second furcilia, mea- suring 2-4 mm. in length and having one pair of simple bud-like pleopods The next seen has one pair of pleopods setose and three pairs simple. It is striking that these stages, together with the tenth,i seem to be dominant, as no intermediate stages were found whilst these were abundant. Moreover, the sixth and tenth are the stages described by Sars." Again, on p. 770 Lebour states with reference to Euphmtsia krohiii: " Sars describes the second and seventh furcilia- having one pair of simple pleopods, and one pair setose and four pairs simple respectively. No stages between these two have been found in the Alexandria samples." "The Furcilia with all pleopods setose is presumably the last." On pp. 768 et seq. of this paper on euphausiids from the Mediterranean, Lebour de- scribes the larvae of Thysanopoda aeqtialis, Hansen. " The first furcilia ", she states, " the earliest stage seen, measures 2-6 mm. in length. It has no pleopods. The next stage present, measuring 2-9 mm. in length, is probably the third Furcilia having two pairs of simple pleopods. Then apparently several stages are absent and the next seen, measuring 3-5 mm. and probably the seventh Furcilia, has two pairs of setose pleopods and two pairs simple." The following Furcilia stage (text-fig. 2, p. 769) is that in which there are four pairs of setose pleopods and the last pair non-setose. Rustad (1934, p. 15) shows that in Euphausia frigida the Furcilia stages 5, 13 and 14 of Lebour 's pleopod diagram are the only ones to be found ; these stages are, respectively, forms having four pairs of non-setose pleopods, four pairs setose and one pair non- setose, and all five pairs setose. In the same publication, under Thysanoessa macrura, G. O. Sars, and Th. vicina, Hansen (the two species are not separated), Rustad states that he found numbers i, 6 and 14 of the Furcilia stages of Lebour, representing larvae having respectively no pleopods, five pairs of non-setose, and five pairs setose. Lebour suggests the dominance of certain stages in her earlier paper (1926 c) and demonstrates it in the species Nematoscelis microps, Euphausia krohiii and Thysanopoda aeqtialis (1926 J). Macdonald shows that in a sample of Meganyctiphanes norvegica, where eleven different Furcilia forms occur, two stages are dominant, set apart from one another by an interval of several intermediate stages. 1 The tenth stage referred to is the tenth for this species, not the tenth stage of Lebour's scheme. The pleopod arrangement is four pairs setose and one pair non-setose. 2 The seventh stage here is the 9th Furcilia stage of Lebour's scheme. DEVELOPMENT OF EUPHAUSIA SUPERBA 35 It has escaped notice that of the stages which are dominant in the six species just mentioned, the earher dominant stage bears a direct relationship to the later. This relationship is significant in the consideration of Eiiphaiisia siiperba which follows. The less advanced dominant stage becomes in each species the more advanced by the provision of setae on the non-setose pleopods present in the earlier stage and the addi- tion of one or more pairs of non-setose pleopods. In other words, each pleopod appears first in the non-setose form, and is provided with setae at the next ensuing moult. Thus, in Nematoscelis as indicated in Fig. 7 below, the one pair of simple pleopods gives rise NEMATOSCELIS MICROPS IT 5?^ □ roS? /K/V/T\yX A\ EUPHAUSIA KROHNM ■^^ ^W^ 9BS THYSANOPODA AEQUALIS corn (^qrrn 55^ 555? 5gg EUPHAUSIA FRIGIDA W^ 9ffl? SSS THYSANOESSA VICINA AND MACRURA WWv ^^ ^ ^ /K MEGANYCTIPHANES NORVEGICA (DOMINANT STAGES ONLY) Fig. 7. Diagram to show the relationship of pleopod development and successive dominant stages. In euphausians where the development is known the pleopods appear first in the non-setose form and become setose in the ensuing dominant stage. to a form having one pair of setose pleopods and three pairs simple, and in the next conspicuous stage these three simple become setose and one non-setose pair is added. A similar method of development takes place in Eiiphaiisia krohnii, Thysanopoda aeqtialis, Euphausia frigida and Thysanoessa macnira and vicina. Meganyctiphanes norvegica shows by its dominant stages that a similar process of development is taking place in it also, for the first dominant stage, that having three pairs of non-setose pleopods, is followed by one having three pairs setose and two pairs non-setose. In a very recent paper by Frost (1935) there is still further evidence of a succession of early Furcilia stages in which non-setose pleopods become setose in the next ensuing moult. The pleopod arrangement in Nematoscelis megalops is: Two pairs non-setose — two pairs setose, three pairs non-setose — all five pairs setose. In Stylocheiron longicorne 36 DISCOVERY REPORTS it is: no pleopods — one pair non-setose — one pair setose, two pairs non-setose — three pairs setose, two pairs non-setose — all five pairs setose. Thus in each instance where the number of pleopods at any stage is known it is possible to predict the number of setose pleopods after the next ensuing moult. EARLY FURCILIA STAGES Turning now to Enphausio superba in the samples used for the identification of Fur- cilia stages, namely those from 70-cm. nets covering a number of seasons, and from the oblique nets of the circumpolar cruise, fourteen different forms of Furcilia are recog- nizable by the variation of development in the pleopods. These forms and their frequency of occurrence are shown in the table below, in which is expressed the number of stations at which the various Furcilia forms were found. Number of Stations at which the form was found (I) No Pl eopods I (2) 2 pairs of simple pleopods I (3) 3 , )> M 5 (4) 4 » J) j» 23 (5) 5 » )» »» 39 (6) 3 , pleopods, 2 setose, i non-setose I (7) 4 I >. 3 >» I (8) 4 2 „ 2 >» I (9) 4 3 .. I >) I (10) 4 , ,, all setose 2 (II) 5 , pleopods, 2 setose, 3 non-setose I (12) 5 3.-2 >» 2 (13) 5 4 .. I ji 12 *(i4) 5 , , , all setose, A and B forms 45 * In this form there are two types of larva recognizable : A, a smaller less developed form which normally moults again into B, and B, larger and more developed and generally moulting into a form with five terminal spines on the telson. Of the fourteen diff"erent Furcilia identified seven occur at one station only, and for six of them the station is the same, namely WS 527. The remaining forms are found at varying numbers of stations, but it will be noted that those occurring at the greatest number of stations are as follows: (i) forms having four or five pairs of non-setose pleo- pods, and (2) forms having four pairs of setose pleopods and one pair non-setose or all five pairs setose. In other words these are dominant forms comparable to those in Meganyctiphanes described by Macdonald. Analyses of the plankton samples taken with vertical 70-cm. nets at St. WS 527 show that Euphausia superba larvae were present in a variety of forms, stated in Table XI below. The number of different kinds of larvae and the frequency of occurrence are shown and the average length of each stage is given. For completeness there have been included Calyptopis stages and Furcilia stages with reduced numbers of spines ter- DEVELOPMENT OF EUPHAUSIA SUPERB A 37 minally on the telson, but it is the earlier Furcilia forms up to the point where five pairs of pleopods, all setose, are found which are to be discussed here. Within these limits thirteen different kinds of Furcilia occur. As in the frequency expressed by the number of stations at which they were present, so here, by the frequency of individuals, two maxima are recognizable: (i) larvae having four or five pairs of non-setose pleopods, and (2) larvae having either four pairs setose and one pair non-setose, or all five pairs setose. These are represented in far greater abundance than other combinations of setose and non-setose pleopods and are the dominant stages. Table XI. Euphausia superba larvae from St. WS 527 Numbers of larvae Average length 50-0 100-50 250-100 Total mm. alyptopis m. m. m.* 2ndC 4 4 2-42 3rd C alyptopis — — 4 4 3'i7 Furcil ia with no pleopods — — 8 8 3'5o , 2 pairs of non-setose pleopods — — 4 4 3-88 , 3 pairs of non-setose pleopods — — 8 8 3-8i , 4 pairs of non-setose pleopods — — 36 36 4-37 , 5 pairs of non-setose pleopods — — 20 20 4-55 , 3 pairs of pleopods, 2 setose, i non-setose — — 8 8 4-00 , 4 pairs of pleopods, i setose, 3 non-setose — — 4 4 4-58 , 4 pairs of pleopods, 2 setose, 2 non-setose — — 4 4 4-58 , 4 pairs of pleopods, 3 setose, i non-setose — — 4 4 4-50 , 5 pairs of pleopods, 3 setose, 2 non-setose — 2 4 6 4-81 , 4 pairs of pleopods, all setose — — 32 32 4-84 , 5 pairs of pleopods, 4 setose, i non-setose — 4 116 120 5-o6 ' , 5 pairs of setose pleopods. Smaller, Form A 8 10 224 242 5-58 , 5 pairs of setose pleopods. Larger, Form B 12 I 16 29 6-65 , 6 terminal spines 4 — — 4 7-38 , 5 terminal spines 80 2 — 82 779 , I terminal spine 8 — — 8 10-19 * In this sample only a quarter of the total was analysed. Inspection of the varying forms of pleopod found amongst the Furcilia at this station shows that without retrogression in development, either in the number of pleopods or in the formation of setae, the larvae cannot moult successively into all the different forms found. Thus those with four or five pairs of non-setose pleopods cannot moult into forms having three pairs of pleopods, two pairs of which are setose and one pair non-setose, without diminishing the number of pleopods. Or if this three-pleopod form be con- sidered less advanced than that having four or five pairs of non-setose pleopods, it can- not moult into one of these latter forms without losing the setae on the first two pairs of pleopods. For the same reasons larvae having two pairs of setose and one pair of non- setose pleopods cannot follow or be followed by forms having one pair setose and three pairs non-setose, nor can larvae having four pleopods, all setose, follow or be followed by forms having five pleopods, three pairs of which are setose and two non-setose. 38 DISCOVERY REPORTS Anatomical examination of these larvae shows that the progress of development can- not be through a series in which at first non-setose pleopods are successively added, and then by these becoming setose until five setose pleopods result. On the contrary practi- cally every one of the larvae having non-setose pleopods has within the integument the rudiments of setae, so that larvae having two pairs of non-setose pleopods would on moulting become forms having two pairs of setose pleopods and presumably one, two, or three pairs of non-setose pleopods. Larvae with three pairs of non-setose pleopods on moulting become larvae having three pairs of setose pleopods and one or two pairs non- setose, and so on. Thus one would expect to find that the relative abundance of the dif- ferent kinds of non-setose larvae would be reflected in the setose section and this is what actually happens, for scarce forms such as those with two or three pairs of non-setose pleopods are represented by equally scarce setose forms with two or three pairs setose and one or two pairs non-setose. Likewise those with four or five pairs of non-setose pleopods, which occur abundantly, are followed by equally abundant forms having four pairs setose and one pair non-setose or all five pairs setose. The suggestion that the larvae moult from forms with non-setose pleopods directly into forms in which those pleopods are setose is supported by the frequency distribution of larvae at stations where a large number occur in the samples. Table XII below shows the larvae from such stations classified according to the condition of the pleopods. The Calyptopis have been included to show that by far the greater number of larvae moult from the third Calyptopis into such as have four or five pairs of non-setose pleopods, and then into larvae having five pairs of setose pleopods or four pairs setose and one pair non-setose. It is fair to conclude then that in Euphausia superba the progress of Furcilia develop- ment is from forms in which four or five non-setose pleopods become directly changed on ecdysis into larvae recognizable by having the four or five pairs of non-setose pleo- pods changed into setose pleopods, and in the former alternative by the addition of one pair of non-setose pleopods. Smaller numbers of non-setose pleopods should be re- garded merely as variations of the dominant numbers and, likewise, other combinations of setose and non-setose pleopods should be regarded as infrequently occurring variants of the later dominant form. Thus up to the point when the Furcilia has five pairs of setose pleopods only two Furcilia stages of E. superba are recognizable. INTERPRETATION OF DOMINANT STAGES Comparing what happens in E. superba with what has been observed in Nemato- scelis microps, Euphausia kroJmii, E. frigida, Thysanopoda aequalis, Meganyctiphanes norvegica, Thysanoessa macrura and Th. vicina, I would suggest that the stages recog- nized as dominant in these species should be regarded as actual stages of that part of the developmental history. Thus in Nematoscelis tnicrops the stages referred to as 2nd, 6th and loth, based on the successive addition of pleopods and by these becoming setose, should be regarded as stages i, 2 and 3, and so with the other species mentioned above, except Meganyctiphanes, in which conditions are somewhat different. In this DEVELOPMENT OF EUPHAUSIA SUPERBA 39 last-mentioned genus Macdonald, as already stated, recognized eleven Furcilia, but of these the Furcilia having three pairs of non-setose pleopods and that having three pairs setose and two pairs non-setose are dominant. On analogy with what happens in Eiiphansia siiperba, particularly with respect to the larvae from St. WS 527, I think it reasonable to suggest that these two " dominant " forms are in fact the first two Furcilia stages, that stages identified by having no pleopods, one pair, two pairs and three pairs of simple pleopods are variants of the early stage and that the remainder with setose pleopods are variants of the 2nd stage. Table XII. Frequency of occurrence of larvae of varying form at stations where representative samples were taken Furcilia Calyptopis Pairs of pleopods Apex of telson D u 1 u (U (U V Station Ol. a, a 0- B B en a. B en 0^ B 'tn setose ; spines Total TJ U 1) N IH N ro m -* 'i- "TJ O) ^O 10 193 — 12 81 — — 26 : I 120 194 6 46 123 — 2 13 — — — I — 191 198 I 28 39 — — — 4 — — • 2 9 S 88 303 — 34 90 — — 7 — — — 2 133 304 — 2 22 — — 4 — — — 2 — 30 305 2 II 2 — — I — — — I — 17 320 I 29 60 — ^ 4 — — — — 99 332 5 2 3 — — — I — — — — — — II 361 — I 12 — I 80 — — — 74 8 I 177 362 — 49 190 — — 4 90 — — — 9 2 — 344 36s — 51 462 — — 192 — — — . 4 — 709 368 — — 14 — IS 22 — — — 14 I 66 369 — — 13 — 14 12 — — — 9 2 — 50 383 146 43 401 — — 5 132 — — — — 727 63s — 12 II — 2 5 4 — — — I — — 35 636 32 4 I — 2 — — — — — 39 WS199 — II — — — 31 — — — 17 3 62 WS200 — 10 — — — 79 — — — ';2 13 154 WS201 — — II — — 4 9 — — I 3 I 29 WS527 4 4 8 4 8 36 20 8 4 4 4 6 32 120 242 29 4 82 619 The institution by the early writers of a succession of Furcilia stages, recognized by the addition of pleopods commencing with the pair on the first abdominal segment and later by the provision of setae on these pleopods, is apparently based neither on direct evidence of moulting nor (until Macdonald 's work on Meganyctiphanes) on any quantitative analysis of the material. The result of Macdonald's quantitative investi- gation has already been stated, and his observations of moulting show that more than half do so "irregularly"; of the remainder which moult "regularly" he does not 40 DISCOVERY REPORTS indicate from which stage they moult. Whatever they are, the evidence they provide, in conjunction with the "irregularly" moulting individuals, shows in the clearest way that in this species development is not by a well-defined progression of stages in which pleopods are successively added. It may be well here to quote Rustad (1930, p. 45) in his description of the larval stages of Euphausia frigida and Thysanoessa macrura. He says: " Development seems to pro- ceed rather schematically up to and including the Calyptopis stages. From the Furcilia stage and upwards it seems on the other hand that there is greater room for action of in- dividual variation. Certainly we observe a clearly pronounced main line which finally leads up to the fully developed adults, but it is impossible to demonstrate any absolute or fixed relation in degree of development between the different organs, and each ' stage ' therefore covers a rather broad range of variation in degree of development and form of the single appendage or organ." These remarks should be borne in mind, for they are equally applicable to the development of Euphausia superba both with regard to the Furcilia stages already discussed and those which are to be considered in the next section. Macdonald (19276) quoting Gurney ( 1 924) states that ' ' Among Crustacea, continuous larval development is a primitive feature, whereas a marked metamorphosis is charac- teristic of more highly developed forms." For this reason Macdonald says "the above observations are interesting as they suggest tentative steps in an evolutionary progress in the order Euphausiacea towards reduction in the number of larval stages. They also suggest that in those arthropods in which the life history consists of a few pronounced stages these are to be regarded not as having evolved independently from a continuous life history, but rather as the survivors of a once greater number of successive stages." In the light of the evidence provided by Euphausia superba and the interpretation of known larval histories in relation to this I consider that it is misleading to talk of a ten- tative reduction in the number of larval stages. This conception is based on the assump- tion that each variety of pleopod development found indicates a larval stage. It implies that as a primitive phase in the evolutionary history of this group the larvae went through this succession of numerous well-marked changes in form, which is of course contrary to the idea of continuous development. In primitive development one would expect to find no well-marked coincidence of ecdysis with a fixed degree of development. Metamorphosis, as already stated, is a characteristic of highly developed forms. The Euphausiacea should be regarded as coming somewhere between the two extremes, they are arthropods in which "continuous" development is giving way to metamorphosis — already well defined in the Nauplius, Metanauplius and Calyptopis stages, less well de- fined in early Furcilia by the presence of dominant forms, and still less intelligible in later development. This developing metamorphosis is not essentially a reduction in the number of successive stages, but is being brought about by the larvae tending to moult into forms showing a certain degree of development in preference to a lesser or a greater degree. Continuous development is not an altogether appropriate term to apply to the de- DEVELOPMENT OF EUPHAUSIA SUPERB A 41 velopment of any arthropod, since the course of development is in all cases marked off into stages by the occurrence of ecdyses between which no marked change in form usually occurs. Where ecdyses are numerous and the morphological changes occurring at each are slight, the development is of the type to which the term "continuous" has been applied. Where ecdyses are reduced in number while the total change to be achieved remains the same, the amount of change at each moult will obviously increase till it may reach a degree deserving the name "metamorphosis". Larval specializations requiring elaboration of the straightforward course of develop- ment may further increase the extent of change that has to be accomplished at a single moult. There is a further consideration to which attention may here be drawn. In the more specialized forms with fewer ecdyses the position of each moult on the scale of develop- ment, and therefore the changes occurring at each, becomes as it were standardized. In the more primitive cases with so-called "continuous" development the position of the moult may vary, and the changes effected at each are not always identical. The series of moults may, in fact, be regarded as a kind of "grid" superposed on a course of actually continuous development. In the more primitive cases this grid may still shift slightly backwards and forwards; in the more specialized cases the "grid" has become fixed and all individuals show the same changes at each moult. INTERMEDIATE FURCILIA STAGES In Table XII above it is indicated that there are two different forms of larvae having all pleopods setose and having seven terminal spines. Examination of the appendages of these Furcilia shows that there is a very great amount of variation in the degree of de- velopment. This is most noticeable in the mandibular palp and ist thoracic limb. There are also differences in size in the antennular fiagella, in the postero-lateral spines of the telson and in the shape of the rostrum and telson. Nine selected larvae of differing sizes from one sample were examined to discover the range of variation occurring in the mandibular palp, ist thoracic limb and telson. It was seen that the mandibular palp varied between the small bud-like process typical of earlier stages and an elongated three-segmented setose appendage not unlike that found in the adult. The endopod of the ist thoracic limb could be from twice to three times as long as the exopod and be composed of from two to five segments. The innermost postero-lateral spine on the telson varied in the breadth of the proximal portion ; in the smaller larvae the base of the spine was narrow, in the larger ones it was broadened. It is possible in some of the larvae to determine the number of terminal spines which are to be present when the larva moults (cf. Macdonald, 1928, pi. iv, fig. 10) and it is found that some are to be seven-spined and others five-spined. Before ecdysis the soft tissues draw away from the enclosing integument, and in the telson region particularly the form of the succeeding stage is defined. It cannot be stated for certain in all the larvae, but in those where it is possible the length frequency distribution correlated with the number of spines in the succeeding stage indicates that the smaller forms clearly 42 DISCOVERY REPORTS tend to moult into larvae, which again have seven spines, while the larger ones moult directly into five-spined forms. This is shown in Table XIII for two samples of seven- spined larvae. Table XIII. Length frequencies of larvae with seven terminal spines, arranged according to the number of spines which will be present in the following stage Length classes mm. Following stage Total 7-spined 5-spined 6-spined Uncertain 5-00-5-24 S-2S^5-49 5-5c^5-74 575-5-99 6-oo-6'24 6-25-6-49 6-50-6-74 675-6-99 7-0O-7-24 7-25-7-49 7-50-774 2 7 I 2 2 I I I I 2 10 8 16 4 2 I I 2 I 6 II 7 3 2 7 2 6 3 9 22 16 19 4 2 There is a considerable overlap in the size ranges of the two groups of larvae in which the number of spines in the succeeding stage can be determined. The normal reduction of the number of terminal spines proceeds by the disappearance of the two outermost, and larvae having six spines should be regarded as exceptional. The nine larvae mentioned above, when arranged in order of increasing development of the appendages (Fig. 8), furnished a rough scale which was used in the analyses of the samples of seven-spined larvae recorded in Tables XIV and XV. These samples were examined to discover what correspondence, if any, exists between size, degree of de- velopment and the numbers of terminal spines in the teison subsequent to moulting. The letters a toj in the columns of the analyses indicate the form of appendage of the nine larvae which that of the larva in the sample most closely resembled, a being the least and j the most developed. In Table XIV the results of an examination of thirty-one larvae are given. It is seen that the rostrum, although it may have a spine in quite small individuals, is in these more often without, whereas in larger larvae there is a greater proportion with a spinous rostral plate. The antennular flagellum increases in length with increasing size of larva. The degree of development of the mandibular palp, ist thoracic leg and teison within broad limits also correspond with the size of the larva. Here again the fact is demon- strated that smaller larvae moult into seven-spined forms, the larger into five-spined forms, but that there is a great amount of overlapping in size and degree of development. Table XV shows a similar analysis of larvae from St. 639. The average size of the larvae here is much less than at the previous station cited, and in correlation with this, development is much less advanced ; nearly all the larvae indicate that in the following stage they will be seven-spined. E o ,_, ^ t^ ■*-» vO -a ■^ « X) F n fc t> <> <) U-, « vO o ^ J3 lo *-• •a rn r; M « -« 0) e ^ « s txo E: -c - OJ o B 6-2 44 DISCOVERY REPORTS Table XIV. Analysis of larvae with seven terminal spines on the telson from St. 374 Length of antennular flagellum Length Rostral Form of Telson No. of spines in succeeding Remarks mm. spine mm. Micro Mandi- bular ist thoracic Telson units palp leg stage 5-08 0 0-19 4-5 a h a 7 5-08 0 0-25 6 b b a-c 7 5-29 0 0-21 5 b b a 7 5-29 0 0-21 5 b-c b a 7 5-42 0 0-21 5 a b a 7 5"7i 0 0-25 6 c d b-c 5 Note I 5-88 0 0-31 7-5 c d d 6 Note 2 5-88 + 0-36 8-5 f f f 5 Note I 6-17 0 0-33 8 d e d-e 7 6-29 0 0-42 ID g e e 7? 6-33 0 0-42 10 h h h 7? 6*50 0 0-33 8 c-d e e 6 Note 3 6-50 0 0-54 13 S — f 7? 6-50 0 0-42 10 h h h-j 7? 6-66 0 0-40 9-5 f g g 7? 6-66 + 0-46 II g g g-h 5 Note 4 6-71 + 0-46 1 1 f f f 7? 6-75 + 0-46 1 1 f f f 7? 6-79 + 0-42 10 d e f 5 6-79 0 0-50 12 g h f 7 6-79 + — — g g g 7? 6-92 + 0-54 13 h-i h-j J 5 7-00 + 0-50 12 g g f-g 7? Note 5 7-00 0 0-52 12-5 g h h 5 7-po 0 0-50 12 h h h 7? 7-08 0 0-50 12 H g a 5 Note 4 7-08 0 — — h^j h-j H 5 7-08 + 0-58 14 y+ J+ i 7? Note 6 7-46 0 0-58 14 H h h-j 5 7-50 0 0-58 14 i H H 5 7-58 0 0-58 14 i J j 5 The letters a-j\n columns 5-7 afford reference to the nine larvae (Fig. 8), which were arranged according to increasing development of their appendages. Note I. The two specimens measuring 5-71 and 5-88 mm. respectively will definitely have a five-spined telson in the succeeding stage, although they are of small size and have the less developed appen- dages generally found in larvae which will again moult into a seven-spined form. Note 2. Into the outermost left-hand terminal spine of the telson there projects a very thin filament, whilst the corresponding spine on the right side has projecting into its interior a definite process similar to those found within the remaining spines. Note 3. There will be six well-developed spines in the next stage, the projection into the outermost right- hand spine being completely absent. Note 4. Within the integument of the telson there are definite projections into five of the terminal spines. The outermost spines on each side have tiny processes projecting into them. The latter would probably atrophy and the succeeding stage would almost certainly be five-spined. Note 5. The outermost spines have processes projecting into them which are almost as well defined as the others, but preparation for the following stage has not gone far enough to say whether the succeeding stage will be seven- or five-spined. Note 6. The formation of seven spines is indicated within the integument of the existing terminal spines, but the outermost on each side is much less robust than the remaining five and would possibly have degenerated before the larva moulted. DEVELOPMENT OF EUPHAUSIA SUPERBA Table XV. Analysis of larvae with seven terminal spines on the telson from St. 639 45 Length of antennular flagellum Form of Telson. Length Rostral No. of spines in succeeding Remarks mm. spine mm. Micro Mandi- bular I St thoracic Telson units palp leg stage 5-46 0 0-21 5 a a a 7 5-91 0 0-25 6 a a a 7 Note I 6-00 0 0-27 6-5 a a a~b 7 6-o8 0 0-25 6 a-b a-b a-b 7 6-o8 0 0-29 7 a-b a a-b 7 6-12 0 0-29 7 b a a-b 7 6-13 0 0-29 7 a-b a a 7 6-21 0 0-27 6-5 — a a-b 7 6-21 0 0-27 6-5 b a a-b 7 6-25 0 0-31 7-5 b a a-b 7 ' 6-25 0 0-31 7-5 b a a-b 7 6-25 0 0-29 7 b a a-b 7 6-38 + 0-29 7 b a a-b 7? 6-46 + 0-42 10 f e d-e 7? 6-75 + 0-38 9 g d d-e 7 The letters a-j in columns 5-7 afford reference to the nine larvae (Fig. 8), which were arranged according to increasing development of their appendages. Note I. The 5th pleopods are non-setose. Table XVI gives the results of ex- amination of sixty-one seven-spined larvae from St. 374. In addition to the degree of development as in the two previous tables, the lengths of the mandibular palp, ist thoracic append- age, and inner antennular flagellum are given both in millimetres and in the units (24 units = i mm.) actually em- ployed in measurement. The length frequency of all the seven-spined larvae from St. 374 is given in Table XVII, in which also is shown the number of specimens which will have seven and five spines in the . length classes ,. __, , Fig. 9. Length frequency of seven-spined larvae from St. succeedmg stage. The results are \, ,wu • ji r> ■ a^ ° o 374. ^, all the seven-spmed larvae, is, seven-spmed larvae shown graphically in Fig. 9. in which the telson form in the succeeding stage can be dis- The length frequency distributions tinguished. succeeding stage seven-spined. of the mandibular palp, ist thoracic succeeding stage five-spined. appendage and inner antennular flagellum, are stated below in Table XVIII and shown 46 DISCOVERY REPORTS Table XVI. Analysis of larvae with seven terminal spines on the telson from St. 374 Length mm. 5-25 5-30 5-30 5-42 5-5° 579 5-83 5-83 5-92 6-00 6-13 6-25 6-25 6-38 6-42 6-42 6-46 6-46 6-50 6-50 6-54 6-54 6-54 6-54 6-58 6-58 6-63 6-63 6-63 6-63 6-67 6-67 6-71 6-71 6-7S 6-75 6-75 6-75 679 6-83 6-88 6-88 6-88 6-92 6-96 7-00 7-00 Rostral spine o o o o o o o o + o o + o o + o + + o o o + + + o o + o o o o + o o o + + + o + + o + o o o + Length of antennular flagellum 0-23 0-23 0-21 0-25 0-21 0-25 0-25 0-48 0-29 0-29 0-29 0-29 0-31 0-38 0-50 0-42 0-38 0-38 0-38 0-42 0-38 0-42 0-42 0-42 0-38 0-40 0-38 0-33 0-38 0-44 0-42 0-38 0-42 0-44 0-42 0-42 0-46 0-46 0-46 0-42 0-40 0-42 0-50 0-42 0-46 0-44 0-48 Micro units 5-5 5-5 5-0 6 5 6 6 7 7 7 7 7-5 9 12 10 9 9 9 10 9 10 10 10 9 9-5 9 8 9 10-5 10 9 10 10-5 10 10 II II II 10 9-5 10 12 10 II 10-5 ii-S Mandibular palp Form b-c b-c b a b c b g-h e-f f f f h h g d-e f f-g f g-h S-h h g-h g-h h-j f f e-f g-h f-g g h f-g J J h h g-h S-h 0-42 0-44 0-44 0-38 0-54 0-75 0-46 i-o8 0-83 0-58 0-67 0-92 I -08 1-04 1-04 I-2I 0-83 1-04 0-71 1-04 0-96 1-04 I-I7 I-I7 i-ii 1-17 0-83 I-I3 1-17 1-29 I -00 1-25 0-96 I -00 0-96 1-04 I-I3 I-2I I-I3 I-2I i-o8 I-2I 1-46 I-I7 1-17 I-2I 1-25 Micro units 10 lo-S 10-5 9 13 18 1 1 26 20 14 16 22 26 25 25 29 20 25 17 25 23 25 28 28 26-5 28 20 27 28 31 24 30 23 24 23 25 27 29 27 29 26 29 35 28 28 29 3° I St thoracic leg Form c b b-c b a d c g d d e f-g e-f e h g e-f g-h e f-g a / g g-h f-g f-g h h g f g f f-g J g g-h h g-h h a S-h Micro units 2-71 2-63 275 2-46 2-63 2-92 2-42 379 3-38 2-96 375 3-58 3-33 3-63 3-96 3-88 3-88 3-92 3-50 3-67 3-96 4-17 3-92 4-08 3-96 3-54 371 4-08 3-96 3-96 4-04 4-08 4-17 3-88 4-38 4-00 4-13 379 4-25 4-21 4-08 3-92 4-17 3-92 65 63 66 59 63 70 58 91 81 71 90 86 80 87 95 93 93 94 84 88 95 100 94 98 95 85 89 98 95 95 97 98 100 93 105 96 99 91 102 lOI 98 94 100 94 Telson Form a b a-b b a, b, c b-c b-c No. of spines in suc- ceeding stage 7 7 7 7 7 7 7 to c 1 ■ 7? b-c 7 e 7-' e V- b-c 7 d-e 7? f 5? f 5 f 5 f-g 7-' e 5-' f 5 e-f 7? e-f 5 g 5 h 5-' g-h 7? d-e 7? g 7? f 7? f 7? g 5 f-g V- f 5 e-f 5 g-h 7? f-g 5 g-h 7? g-h 5 f 5 J r- h 5 f 5 g-h 7-' J 5 g-h 7? h 7? h 5 h 5 Remarks Note I Note 2 DEVELOPMENT OF EUPHAUSIA SUPERBA Table XVI (cont.) 47 Length of Telson antennular flagellum Man dibular palp 1st thoracic ■leg Length mm. Rostral spine No. of spines Remarks Micro Micro Micro Form m suc- mm. units Form mm. units Form mm. units ceeding stage 7-00 + 0-46 II J 1-25 30 — — — j 5 7-00 + 0-48 II-5 J 1-25 30 h 4-29 103 H 5 7-00 + 0-44 10-5 '•-J 1-29 31 h-j 4-29 103 h-j 5 7-04 + — — J 1-33 32 — — — h-i 5 7-08 + 0-46 II H i-o8 26 g-h 3-88 93 h 5 7-08 + 0-46 II j 1-46 35 H 4-75 114 h-j 5 7-13 + 0-42 10 g 1-04 25 g-h 3-88 93 R-h 5 7-13 + 0-48 II-5 j I-2I 29 J 4-04 97 ] 5 7-13 + 0-50 12 H 1-25 30 J 4-42 106 J 5 7-13 + 0-54 13 ] 1-33 32 J 4-46 107 H 5 7-i6 + 0-52 12-5 ] 1-33 32 J 4-38 105 J 5 7-25 0 0-42 10 h I-2I 29 h 4-33 104 h 5 7-38 + 0-50 12 J 1-54 37 J 4-42 106 J 5 7-41 + 0-52 12-5 J i-3« 33 J 471 113 J 5 See footnote to Table XIV (p. 44) for explanation of the letters denoting the form of the mandibular palp, ist thoracic leg and telson. Note I. The left 5th pleopod is non-setose, the right is setose. Note 2. The mandibular palp is unsegmented. The endopod of the ist thoracic leg is five-segmented. Table XVII Length classes mm. Total no. of larvae Following stage 7-spined 5-spined Uncertain 5-25-5-49 5-50-5-74 575-5-99 6-00-6-24 4 I 4 2 4 I 2 I — 2 I 6-25-6-49 6-50-674 6-75-6-99 7-0O-7-24 7-25-7-49 7 16 II 13 3 I 2 6 6 13 3 4 10 5 Total 61 9 30 22 graphically in Fig. 10. The graphs for the length frequencies of these appendages are bimodal, a less conspicuous maximum representing the smaller number of larvae of small size in this sample and a more conspicuous one for the bulk of the sample which is made up of large average size. DISCOVERY REPORTS t i 1 lU < > tr 25 _ < _1 B feao - or UJ CO 15 - 2 Z 10 LENGTH CLASSES Fig. II. Length frequencies of larvae with seven-spined telson arranged according to their degree of development. A, seven-spined larvae in which the telson form in the succeeding stage can be distinguished. succeeding stage seven-spined. suc- ceeding stage five-spined. B, seven spined larvae arranged with reference to nine selected larvae (see Fig. 8). — — — degree of development a-d. degree of development e-j. 8 e 10 II LENGTH (UNITS) 13 UNITS Fig. 10. Length frequencies of : ^, mandibular palp; B, first thoracic appendage; C, antennular flagellum. Table XVIII. Length frequencies of mandibular palp, ist thoracic appendage and antennular flagellum Mandibular palp ist thoracic palp Antennular flagellui n Length class No. Length class No. Length class No. Units mm. Units mm. Units mm. 6-1 1 11-16 16-21 21-26 26-31 31-36 36-41 0-25-0-46 0-46-0-67 0-67-0-88 0-88-1-08 1-08-1-29 I-29-I-50 1-50-1-71 4 3 6 13 26 8 I 50-60 60-70 70-80 80-90 90-100 lOO-IIO 1 10-120 2-08-2-50 2-50-2-92 2-92-3-33 3'33-3-75 3-7S-4-I7 4-17-4-58 4-58-5-00 2 5 2 8 27 10 2 5-6 6-7 7-8 8-9 9-10 lO-II 11-12 12-13 13-H 0-21-0-25 0-25-0-29 0-29-0-33 0-33-0-38 0-38-0-42 0-42-0-46 0-46-0-50 0-50-0-54 0-54-0-58 4 3 5 I II 17 13 5 I DEVELOPMENT OF EUPHAUSIA SUPERB A 49 If the larvae are arranged in two groups according to their average development as expressed by their correspondence to the scale provided by the nine larvae mentioned above, the result (Table XIX) is two sets of figures which correspond in their rano-e with those of the larvae which are to be seven-spined or five-spined respectively in the following stage. This is shown graphically in Fig. 11, Table XIX. Larvae with seven-spined telson from St. 374 arranged in length groups according to their degree of development Length classes mm. Degree of development a-d H S-25-5-49 4 — 5'5°-574 I — 575-5-99 6-00-6-24 3 2 I 6-25-6-49 6-50-6-74 7 16 675-6-99 — II 7-00-7-24 — 13 7-2S-7-49 — 3 Attention is drawn here to Table XV for specimens from St. 639 in which the average size of the larvae is much less than at St. 374. The larvae compare with the group of larvae of small size at St. 374, and with the exception of two they can, like these latter, be included in the range of development a to d. From these analyses, therefore, it is possible to reach the conclusion that in the develop- ment of Eiiphaiisia superba the larvae possessing five pairs of setose pleopods and seven terminal spines can be divided into two groups: {A) those having a succeeding seven- spined stage and {B) those having a succeeding five-spined stage. The larvae may be identified as follows : Form A Smaller average size. Antennular flagellum usually shorter than the two distal segments of the antennule. Mandibular palp sausage-shaped, not generally multiarticulate. Endopod of ist thoracic limb not developed as a swimming foot, about twice as long as exopod. Telson not having innermost postero-lateral spine greatly modified. Telson length not fully three times the width. Form B Larger average size. Antennular flagellum usually longer. Mandibular palp distinctly three-segmented and with terminal spine developed. Endopod of ist thoracic limb five-segmented, much more than twice as long as exopod. Telson with proximal portion of the postero- lateral spine widened greatly in comparison with the distal portion. Telson length more than three times the width. LATER FURCILIA STAGES For the development of the larvae after they have the full complement of setose pleopods and less than seven terminal spines on the telson, a random sample of 243 larvae from St. 374 was examined. JO DISCOVERY REPORTS The following observations were made : Total length of larva. Telson : (a) Number of terminal spines. (b) Number of postero-lateral spines. Number of spines on the telson foreshadowed for the succeeding stage. Antenna. Number of segments in inner ramus. Mandibular palp : {a) Number of segments. (b) Number of spines on the terminal segment. I St thoracic appendage. Number of segments in the endopod. Presence or absence of a spine on the rostrum. In addition to the evidence from these observations, the records of larvae from the 70-cm. net samples generally, and from the oblique and horizontal net samples from the circumpolar cruise have been utilized. Before proceeding with the discussion of this evidence it is proposed to consider the term "Cyrtopia" applied to euphausian larvae. As stated above (p. 32) the name was applied by Dana to distinguish a schizopodous genus which was afterwards shown by Claus to be a stage in the development of Euphausiidae. Later workers have adopted this division of larval development, and recognize the Cyrtopia by the alteration in form and function of the antenna of earlier stages. Rustad states (1930, pp. 68 et seq.) : " The Furcilia stages are characterized by having the antennae retained as swimming ap- pendages whereas in the Cyrtopia stages they are directed forwards and evidently have no importance for swimming. In preserved material, however, where the antennae as a rule are directed forwards in late Furcilia stages one must have recourse to morpho- logical characteristics and concerning this Lebour writes (1925, p. 816) 'The best way to distinguish a late Furcilia from an early Cyrtopia is by the flagellum of the antenna which is unjointed in the Furcilia and jointed in the Cyrtopia'. " Considering the indefinite nature of later euphausian development which has already been demonstrated and anticipating the conclusions derived from the sample of larvae from St. 374, it seems that the recognition of the Cyrtopia places too much stress on the alteration of form and function of one particular appendage. The larvae designated Cyrtopia are not recognizable from Furcilia by any sudden increase in size as they change from the old to the altered form. The change in the antenna does not necessarily coincide with equally significant changes in the form of other appendages, for example, the mandibular palp and the ist thoracic appendage. It varies also in relation to the de- velopment of the telson: for the number of terminal telson spines in the first Cyrtopia of various species of euphausians is given as : Meganyctiphanes norvegica\ ^^^^.^^^ ^^.^^^^ Thysanoessa raschn J Eiiphausia krohnii 5 ,, Nyctiphanes couchii 3 „ Nematoscelis tnicrops ] Thysaiiopoda aequalis J It is apparent, therefore, that the change in the antenna is not of the significance formerly attributed to it, and the altered form does not merit the distinction of a division DEVELOPMENT OF EUPHAUSIA SUPERB A 51 in the larval history. It is merely one of several changes in form, and presumably in function, which take place in the larval appendages in the process of development. These changes are probably intimately connected with the provision of other appen- dages to do the work performed in earlier stages of the life history by those more anteriorly placed. The name Furcilia will therefore be used here to describe all the later developmental stages in Eiiphansia super ba. The larvae in a sample of 243 individuals from St. 374 have been arranged in Table XX in order of ascending size, and when of the same size roughly according to the degree of development of various appendages. Length and telson. In Table XXI the larvae have been arranged according to their length frequency. The succession of peaks presented is to some extent made intelligible by distinguishing the larvae according to the number of spines on the telson. There is a correspondence in position of the maxima in the length frequency of all the larvae with those identified according to the telson spine number. Increase in size is accompanied by reduction in the number of telson spines, but there is a great amount of overlapping in the length range of any one telson spine group and the length ranges which succeed and precede it. The reduction in the number of telson spines is brought about normally by the dis- appearance of the outermost on either side so that the reduced numbers are 5, 3 and i. This is not invariable, as the telson spine numbers 6, 4 and 2 indicate in Table XXI ; but these forms are present in such small numbers in relation to those with 5, 3 and i that they must be regarded as exceptional. Independent evidence in favour of such a conclusion is provided by the frequency, as expressed by the number of stations at which they were found, of larvae having the different forms of telson recognized. Thus: Larvae with 6 terminal spines occur at 5 stations. s } )} 20 4 » »» 4 3 > j» II 2 )» ) y> 2 I »> » >J 32 Several aberrant telson forms are referred to in the notes below the list on p. 53, and some of these are figured (Fig. 12). In three of the figures of unusual telson form the number of spines in the succeeding stage is shown — all three indicate that the aberration of telson form persists when the larva moults. Telson spine numbers. In Table XXII are stated the length frequencies of larvae arranged according to the number of telson spines in the succeeding stage. Here again, in the seven-spined larvae as in earlier analyses, the smaller ones tend to moult again into seven-spined forms and the larger into five-spined forms. In the five-spined larvae similarly a few moult again into five-spined forms, and there is a surprisingly large number with four spines, but the greater number moult directly into a three-spined form. 7-2 53 DISCOVERY REPORTS In the three-spined larvae it is more difficult to distinguish the number of spines to be found in the succeeding stage, but in those in which it is possible, a succeeding one- spined stage predominates. One larva has two spines in the following stage and there is no sign that any are to be three-spined. f iCit^i fT\ -ni \ /: A ^1 f\ 1 1 w f 1 Fig. 12. Normal and abnormal types of telson tip in Furcilia stages (x 85). a, the normal seven-spined telson. b-e, abnormalities. In Table XX, b is referred to under Note 10, c under Note 12, (/under Note 13, e under Note 14, /under Note 15. In«,i,c and/, the form of the telson in the succeeding stage can be distinguished. It would appear therefore that in the succession of moults which follow on the larvae having five setose pairs of pleopods and the B form of telson, the ecdyses generally coincide with a reduction in the number of terminal spines in the telson from seven to DEVELOPMENT OF EUPHAUSIA SUPERB A Table XX. Characters of late Fur cilia larvae from St. 374 S3 Telson spines Mandibular palp Terminal Total length mm. 13 1 1 0 Rostral spine spines of telson in following Segments of inner ramus of antenna 0 OJ Segments of 1st thoracic limb Remarks ■g 2 stage s U5 ^ c S i-< (/) bO •-^ c V 0 0) Q.-- H Ph CO !/) 5-00 7 3^ ... 7 I Rud. 2 Pleopods V non- setose 5-29 7 3A 0 Rud. 3 or 4 5-42 7 3^ 0 7 Rud. 2 PleopodsV non- setose 5-54 7 3^ ... 7 3 I 2 Pleopods V non- setose. Note I 5-63 7 3^ 0 7 I 0 2 Pleopods just setose. Note 2 5-67 7 3^ 0 7 3 0 3 or 4 571 7 3A 0 7 3 0 3 Note 3 575 7 3A 7 3 0 3 575 7 3A 0 7? 3 Rud. 3 Note 4 5-82 7 3A 0 7 I 0 2 6-00 7 3A + 3 I Missing 6-04 7 3A 0 7 I 0 3 6-04 7 3A 7 2 0 3-4 6-04 7 3A 7 2? 0 5 Pleopods V non- setose 6-o8 7 3A 0 7 I 0 3? 6-21 7 3A-B 0 7? 3 5 6-29 7 3B + 3 5 6-42 7 3A-B + 7'? 3 5 6-42 7 3A-B 0 3 5 6-42 7 3A 3 5 6-46 7 3B + 5 3 5 6-50 7 3B 0 5 3 5 6-58 7 3A-B 0 6 Rud. 4 or 5 Note 5 6-58 7 3B? 0 3 5 6-58 7 3A-B 0 7'? 3 5 6-67 7 3 0 3 5 6-67 7 35 0 3 5 6-67 7 3B + 5'? 3 5 Note I. The mandibular palp is definitely three-segmented with a short terminal spine on the distal segment. Note 2. The first thoracic appendage is of the primitive form with two-segmented endopod. The distal segment is indistinctly divided into three within the integument. Note 3. The larva is very near ecdysis, so that the mandibular palp appears six-segmented — the three existing segments and the three of the following stage. The existing cuticle was dissected from the telson leaving the seven terminal spines of the following stage. Note 4. The one mandibular palp consists of three segments with a very small spine at the tip of the distal one ; the palp as a whole is small and sausage-shaped. The other palp is indistinctly three-segmented with no spine at the end. Note 5. The antennular flagella are only slightly longer than the distal peduncular segment. The mandibular palp is sausage-shaped. The inner ramus of the first thoracic appendage is little more than twice the length of the exopod : it is impossible to state certainly whether it is composed of four or five segments. 54 DISCOVERY REPORTS Table XX (cont.) Telso n spines Mandibi, lar palp Terminal Total , spines of telson in following Segments c *-» Segments length mm. C S 0 Rostral spine of inner ramus of antenna 0 OJ c/3 of I St thoracic limb Remarks 'b 0 stage c 2 Ui CA W) • 3 C ZJ 0 V D.-- H P- m CT! 6-67 7 3fi 0 5 3 5 671 6 3 + 3 5 Note 6 671 7 3B 0 6 3 5 671 7 3 5 3 5 675 7 3B + 5 3 5 675 7 3 0 7 3 5 675 7 3fi + 5 3 5 675 7 3^ 0 5 3 5 679 5 3 4 3 5 6-83 7 3B + 3 5 6-83 7 3B 0 5 3 5 6-83 7 3B 0 5 3 5 6-83 7 3B 0 5 3 5 6-83 7 3B 0 5 3 5 6-88 7 3 3 5 6-88 5 3 0 3 5 6-88 7 3B 0 3 5 6-88 7 3B + 5 3 5 6-88 7 3B 0 5 3 5 6-88 7 3B + 6 3 5 6-88 7 3B + 5 3 5 6-88 7 3B 0 5 3 5 6-92 5 3 0 3 I 5 6-92 5 3 + 3 5 6-92 7 3B 3 5 6-92 7 3B + 7? 3 5 6-96 7 3B 0 7? - 3 5 6-96 7 3A-B 0 5 3 5 6-96 7 3B 0 5 3 5 6-96 7 3B 0 5 3 5 6-96 6 3 0 5 3 5 Note 7 7-00 5 3 0 3 5 7-00 7 35 0 3 5 7-00 7 3B 0 5 3 5 7-00 7 3B 0 5 3 5 7-00 7 3B 5 3 5 7-00 6 3B + 3 5 7-00 5 3 3'? 3 5 7-04 7 3B 0 5 3 5 7-04 5 3 5 3 5 7-08 7 3 0 5? 3 5 Note 6. The five normal terminal spines of the telson are subequal in length ; on the extreme left is a small spine less than half the length of the others. There are clearly defined projections into the five larger spines indicating that the succeeding stage will be five-spined. Note 7. There will be five telson spines in the succeeding stage, there is no projection into the existing spine on the right. DEVELOPMENT OF EUPHAUSIA SUPERB A Table XX (cont.) ss Telson spines Terminal Segments of inner ramus of antenna Mandibular palp Segments "2 la Total length mm. "a C 'b 2 Rostral spine spines of telson in following U5 c B 0 OJ 0 of I St thoracic limb Remarks u 0 stage •^•s h Ph c« Cfi 7-o8 7 3B + 5 I 3 I 5? I St thoracic limb broken 7-o8 7 3 5 3 5 7-o8 6 35 0 5 3 5 Note 8 7-o8 5 3 ... 3 5 7-o8 5 3 0 5"? 3 5 7-o8 5 3 + 3 3 5 7-13 6 3 + 5 3 5 Note 9 7-13 5 3 + 4 3 5 Note 10 7-17 7 3B 5 3 5 7-17 5 3 5 3 5 7-17 5 3 + 4 3 5 7-17 5 3 + 3 3 ^ 5 7-21 7 35 + 5 3 5 7-21 5 3 + 5 3 5 7-21 5 3 + 5? 3 5 7-21 5 3 0 3 3 5 7-21 5 3 + 3 3 5 7-21 5 3 0 3 3 5 7-25 7 3B + 5 I 3 5 7-25 5 3 0 5? 3 5 7-29 7 3B ... 5 3 5 7-29 7 3B + 5 3 5 7-29 6 3 0 5 3 5 Note II 7-29 5 3 + 5 3 5 7-29 S 3 0 4 3 5 7-29 5 3 + 3 3 5 7-33 5 3 0 5 3 5 7"33 6 3B 4 + 3 5 Note 12 7-38 7 3B 0 5 3 5 Note 13 Note 8. Telson with six terminal spines, but there will be five terminal spines in the following stage. There is no projection into the existing spine on the extreme left. As in Note 6. On the telson the outermost terminal spine on the right is a very small vestige, about one-tenth the length of the spine next to it. Within the integument there are four projections into the spines; there is no projection into the small spine on the right nor into the outermost on the left (see Fig. 12 b). As in Note 6. Of the six terminal spines on the telson the outermost on the left is half the length of that next to it. The number of spines in the following stage is four large normal spines and a very small fifth one on the extreme right. The latter may atrophy altogether, so that only four spines may be present in the succeeding stage (see Fig. 12 c). Of the seven terminal telson spines, the outermost on the right is one-fifth the length of the next to it, the outermost on the left is about three-quarters the length of that next to it. The remaining spines are subequal with a slight reduction in the size from the outermost to the central (see Fig. 12 d). Note Note Note Note 9- 10. II. 12. Note 13. 56 DISCOVERY REPORTS Table XX {cont.) Total length mm. Telson spines Rostral spine Terminal spines of telson in following stage Segments of inner ramus of antenna Mandibular palp Segments of I St thoracic limb Remarks 'a C "2 C — ■ U c 1) •|.i h |2h c/: CD 7-38 7 3B + 5 3 5 7-38 6 3B + 5? 3 5 7-38 6 3 5? 3 1 5 Note 14 7-38 5 3 + 5 3 5 7-38 5 3 0 ... 3 5 7-38 5 3 0 3 3 5 7-42 5 3 + 5 3 5 7-42 5 3 + 5? 3 5 7-42 5 3 + 3? 3 5 7-42 5 3 + 3 3 5 7-42 5 3 0 3 3 5 7-42 5 3 3 3 5 7-46 5 3 0 5 3 5 7-46 5 3 + 4? 3 5 7-46 5 3 0 3? 3 5 7-46 5 3 0 3 3 5 7-5° 7 2B ... S 3 5 7-50 5 3 0 4 3 5 7-5° 5 3 + 3 5 7-5° 5 3 0 3 3 5 7-50 5 3 3 3 5 7-54 7 35 + 5 3 5 7-54 7 35 0 5 3 5 7-54 7-58 5 7? 3 3B 0 0 3 4 3 3 5 5 Note 15 7-58 6 2B + 5 3 5 Note 16 7-58 5 3 + 4? 3 5 7-58 5 3 0 5? 3 5 7-58 5 3 0 3 3 5 7-58 5 3 0 3 3 5 7-58 5 3 3 3 5 7-58 5 3 + 3 3 5 7-58 5 3 0 3 3 5 7-58 7 3 0 5 3 R2, Li 5 7'63 7 3B 0 S 3 5 Note 17 7-63 7 ZB 5 3 5 Note 15. Note s very small, about one-quarter to about one-sixth the length of the here being no projection into either Note 14. Of the six terminal telson spines the outermost on the left one-fifth the length of the next spine to it (see Fig. 12 e). Of the seven terminal telson spines, the outermost are only remainder. There are four projections into the existing spine, of the small spines nor into the large spine on the left (see Fig. 12/). 16. There are six terminal spines on the telson and there will be five in the following stage; there is no projection into the spine on the extreme right, so that in the existing stage it is the outermost spine on the left which is assumed to be missing. 17. There are seven terminal telson spines; the outermost on each side only about three-quarters the length of the remaining five, which are subequal. Note DEVELOPMENT OF EUPHAUSIA SUPERBA 57 Table XX {cont.) Total length mm. Telso n spines Rostral spine Terminal spines of telson in following Segments of inner ramus of antenna Mandibular palp Segments of I St thoracic limb Remarks "2 c OJ 1) "r- 3 0 u 1 0 stage E en " U --2 u c/3 tiO •- C r" 0 u a,-- h di cc C/D 7-63 6 2B 0 5 3 5 Note 18 7-63 5 3 3 3 5 7-67 5 3 + 4 3 5 Note 19 7-67 5 3 + ... 3 5 7-67 5 3 + 3 3 5 7-67 5 3 3 3 Ri, L2 5 771 7 35 + 5 3 5 771 5 3 + 3 3 5 779 7 35 0 5 3 5 779 5 3 + 4 3 5 Tl^ 5 3 + 3 3 5 779 5 3 3 3 5 779 5 3 0 3 3 5 779 5 3 + 3 3 Ri, L2 5 7-83 5 3 0 3 3 5 7-83 5 3 + 3 3 5 7-88 5 3 3 3 R2, Li 5 7-92 5 3 + 4 3 2 5 7-92 5 3 0 3 3 2 5 7-96 5 3 + 3 Prob. I 3 2 5 8-00 5 3 0 3 3 2, I 5 8-00 5 3 3 3 2 5 8-04 5 3 4 2 + 1 3 I 5 8-13 5 3 3 i + i 3 2 5 8-17 5 3 3 3 2 5 8-21 5 3 3 3 2 5 8-25 5 3 3? 3 I 5 8-25 5 3 + 3 3 I 5 8-29 5 3 + 3 3 I 5 8-29 5 3 ... 3 ... 3 2 5 8-33 5 3 + 3 I 3 2 5 8-38 5 3 3 i + i 3 Li, R2 5 Note 20 8-38 3 3 + 2 + 1 3 2 5 8-46 5 3 3 3 I S 8-54 3 3 I + I 3 2 5 8-58 4 3 + I + I 3 2 5 8-58 3 3 0 I 3 2 5 Note 18. There are six terminal telson spines and there will be five in the next stage. There is no projection into the spine on the extreme left, so that in the existing stage it is the outermost spine on the right which is probably wanting (cf. Note 16 supra). Note 19. Within the integument of the telson four quite well-defined spines are visible. There is no spine projection into the existing spine on the extreme left. Note 20. There is asymmetry in the development of the appendages of this larva. The left mandibular palp has one large seta, the right two setae on the terminal segment. The left antenna is more primitive than the right, the outer ramus is not scale-like and the inner ramus is very indistinctly divided into two, whereas the left antenna has a definite scale and a distinctly two-segmented flagellum. 58 DISCOVERY REPORTS Table XX {cont.) Total Length mm. Telson spines Rostral spine Terminal spines of telson in following Segments of inner ramus of antenna Mandibular palp Segments of I St thoracic limb Remarks "3 "2 1 6 « I- c U 1) S M 0 (U en 2 stage i-< f/i M ■S ^ u 0 U Oh-" H cS Cfi CO 8-63 5 3 3 I 3 2 5 Note 21 8-75 3 3 0 I + 1 3 2 5 879 5 3 + 3 2 5 8-79 3 3 ... 2 2 + 1 3 2 5 8-88 3 3 3 3 5 8-88 3 3 2 + 1 3 3.2 5 8-92 4 3 + 3 i + i 3 2 5 8-92 3 3 + 2 + 1 3 R3, L2 5 8-96 5 3 0 3 3 I 5 8-96 5 3 I + 1 3 Ri, L2 5 Note 22 8-96 3 3 + 3'? i + i 3 2 5 8-96 3 3 0 3 4 5 9-08 5 3 + ... 3 I 5 9-17 3 3 + 7? 3 R3, L2 5 9-21 3 3 + 2 + 1 3 3 5 9-25 3 3 3 2 5 9-25 3 3 + 2 + 1 3 3 5 9-25 3 3 2 + 1 3 4.3 5 9-33 5 3 ... 3'? 2 + 1 3 3 S 9-38 3 3 0 2 + 2 3 2 5 9-38 3 3 + I? 3 R3, L2 5 9-38 3 3 I? 3 3 5 9-38 3 3 + ... 2+ I 3 3 5 9-38 2 3 + I 2 + 3 3 3 5 9-38 3 3 + 3? 2 + 3 3 4 5 9-46 3 3 + 2+1 3 2 5 9-46 3 3 + 2 + 1 3 3 5 9-46 3 3 2 + 1 3 4.3 5 9-54 3 3 I 2 + 3 3 2 5 9-54 3 3 I + I 3 3 5 9-58 3 3 I 2+1 3 3 5 9-58 5 3 ... 2 + 3 3 4 5 9-63 4 3 0 ... 2 + 1 3 2 5 9-63 3 3 + 2+1 3 2 5 9-63 3 3 0 2 + 3 3 3 5 9-67 3 3 2+1 3 3 5 9-67 3 3 2 + 1 3 4 5 971 5 3 + 3'? 2+1 3 2 5 Note 23 971 3 3 + I ... 3 R3, L2 5 Note 21. The inner ramus of the antenna is unsegmented, but within the integument there are definite signs of segmentation indicating that when the animal moults there will be 2 + i segments. Note 22. The inner ramus of the antenna is distinctly divided into two segments, of which the distal gives indication that it becomes three-segmented later. Note 23. On the telson the terminal spine on the right is only about half the length of the remaining spines. DEVELOPMENT OF EUPHAUSIA SUPERB A Table XX (cont.) 59 Total Length mm. Telso n spines Rostral spine Terminal spines of telson in following Segments of inner ramus of antenna Mandibular palp Segments of ist thoracic limb Remarks « ■3 1 c 1 U V § ^ G 0 u stage £ So 01 " r'" 0 u a.-- ^ p; c« Cfi 971 3 3 + 2 + 1 3 3 5 979 3 3 -f i + i 3 3 5 979 3 3 -f 2-f I 3 3 5 9-83 3 3 + 2 + 1 3 2 5 9-83 3 3 2 + 1 3 3 5 9-83 3 ... 2 + 3 3 3 5 9-92 3 3 + 3 R2, L3 5 9-92 3 3 + I 2 + 1 3 3 5 10-00 5 3 + 3 2+1 3 3 5 10-00 3 3 0 I? 24-2 3 3 5 lo-oo 3 3 0 2 + 3 3 3 5 10-00 3 3 -f 2 + 3 3 3 5 10-00 3 3 + 2 + 3 3 4 5 lo-oo 3 3 + 2 + 3 3 3 5 10-00 3 3 + I 2 + 6 or 7 3 3 5 10-04 2 3 0 I 2 + 3 3 3 5 ro-08 3 3 0 2 + 3 3 3 5 io-o8 3 3 + 2 + 3 3 3 5 10-21 3 3 2 + 3 3 4 5 10-33 3 3 ... ... 2 + 3 3 4 5 10-38 3 3 + I 2 + 3 3 3 5 Note 24 10-46 3 3 0 I 2 + 3 3 4 5 10-50 3 3 I? 2 + 3 3 3 5 10-54 I 3 + 2 + 3 3 4 5 10-63 3 3 + 2 + 3 3 3 5 10-63 3 3 + I 3 3 5 10-67 3 3 -f I 2 + 3 3 R3, L2 5 10-71 3 3 + I 2 + 3 3 4 5 10-79 3 + 11-21 3 + 2 + 7 3 5 5 11-46 3 0 2 + 9or 10 3 4 5 11-58 3 + 3 4 5 11-71 3 + 2 + 4 + * 3 4 5 * Broken 12-38 3 3 4 5 12-92 3 + 3 6 5 Note 24. The antennal inner ramus consists of two basal segments and a three-segmented flagellum. Within the integument of the flagellum can be distinguished a much greater number of segments— si.x or eight in number at least. five, five to three and three to one, but that, as might be expected in arthropods whose development is so undefined, by no means all the larvae conform to this scheme. PosTERO-LATERAL TELSON SPINES. In none of the larvae in this sample is the number of postero-lateral spines reduced from three to two as eventually happens. In the seven- spined larvae the state of development of the innermost postero-lateral is shown in 8-2 6o DISCOVERY REPORTS Table XX by the letters A or B, these letters corresponding respectively to the forms with narrow and broadened proximal portions. As in the description already given, the form with narrow-based spine is restricted to larvae of smaller size, whereas the broadened form is found in the larger seven-spined larvae. As all later larvae have the broadened form of innermost postero-lateral, the letter J5 has been omitted in descrip- tions of them. Table XXI. Length frequency of larvae from St. 374 arranged according to the number of terminal telson spines Terminal telson spine number Length classes mm. 1 Total 7 6 5 5 3 2 I . s-oo- 5-24 I I — — — — — — 5-25- 5-49 2 2 — — — — — — 5-50- 574 4 4 — — — — — — 575- 5-99 3 3 — — — — — — 6-00- 6-24 6 6 — — — — — — 6-25- 6-49 5 5 — — — — — — 6-50- 6-74 II 10 I — — — — — 675- 6-99 27 22 I 4 — — — — 7-00- 7-24 28 10 3 15 — — — 7-25- 7-49 27 5 4 18 — — — — 7-50- 774 28 8 2 18 — — — — 775- 7-99 12 I — II — — — — 8-00- 8-24 6 — — 6 — — — — 8-25- 8-49 8 — — 7 — I — — 8-50- 874 4 — — I I 2 — — 8-75- 8-99 1 1 — — 3 I 7 — — 9-00- 9-24 3 — — I — 2 — — 9-25- 9-49 13 — — I — II I — 9-50- 9-74 12 — — ■ 2 I 9 — — 975- 9-99 7 — — — — 7 — — IO-00-IO-24 II — — I — 9 I — IO-25-IO-49 3 — — — — 3 — — 1 0-50- 1 0-74 6 — — — — 5 — I IO-7S-IO-99 I — — — — — — I 1 1 •00-11-24 I — — — — — — I II-25-II-49 I — — — — — — I II-50-II-74 2 — — — — — — 2 1175-11-99 — — — — — — — — I2-00-I2-24 — — — — — — — — I2-25-I2-49 I — — — — — — I I2-50-I2-74 — — — — — — — — 1275-12-99 I — — — — — — I Totals 245 77 II 88 3 56 2 8 Antenna. The development of the antenna from its primitive form as a swimming appendage into a specialized scale and flagellum takes place within the range of larvae found at this station. The way in which this is brought about is shown in Fig. 13. The change of the outer ramus into a flattened scale follows the normal course de- scribed in other euphausians. DEVELOPMENT OF EUPHAUSIA SUPERB A 6i In the inner ramus the unsegmented condition is seen in Fig. 13 a. In Fig. 13 b this has become divided into two. In Fig. 13 c the ramus is still divided into two, but of these the proximal shows a foreshadowing of the two-segmented peduncle of later stages while the distal segment is still unjointed. From this and from the examination of other larvae at this station it is evident that the proximal segment divides into two, forming the peduncle, before segmentation begins in the distal segment. In Fig. 13 d the ped- uncle is distinctly segmented into two and the flagellar portion into four. Within the Table XXII. Length frequency of larvae from St. 374 arranged according to the number of telson spines in the following stage Length classes mm. Telson with 7 spines Telson with 5 spines Telson with 3 spines Following stage Following stage Following stage •a 'S. •a c 'Eh I •a u c "S. CO c "a u u 0 C P 'Hh T3 U c 'Eh ffi 4- ■a c 'E. c -a c 'S. eft N •a c 'S. ■*-» u 0 c D 5-00- 5-24 5-25- 5-49 5-50- 574 575- 5-99 6-00- 6-24 6-25- 6-49 6-50- 674 675- 6-99 7-00- 7-24 7-25- 7-49 7-50- 774 775- 7-99 8-00- 8-24 8-25- 8-49 8-50- 874 875- 8-99 9-00- 9-24 9-25- 9-49 9-50- 974 975- 9-99 IO-00-IO-24 IO-25-IO-49 10-50-1074 I I 4 2 3 I 2 I I 3 14 8 5 8 I I I 3 4 5 6 2 2 5 I 2 I 2 2 I 5 6 12 9 5 6 I I I 3 6 6 3 I 2 I I 2 I 3 I I 2 3 I 2 6 2 II 6 6 8 I 2 integument the state of the flagellum in the following stage can be discerned as having a much greater number of segments. Rustad (1930, p. 72) describes two different forms of development of the inner ramus of the antenna in euphausians. In the one (Fig. 14 b) the flagellum and peduncle are already separated in the two-segmented stage — the E. frigida type. In the other (Fig. 14 a) the peduncle and flagellum part in the three-segmented stage — the Thysanoessa macnira type. Euphaiisia siiperba (Fig. 14 c) follows the E. frigida type but differs from it in that the peduncle divides again into two before segmentation of the flagellum 62 DISCOVERY REPORTS proceeds, whereas in E. frigida division of the peduncle coincides with an increase in the number of flagellar segments. Fig. 13. Development of the antenna (x 46). a, larva with 5 telson spines; c, larva 9-71 mm. long with 3 telson spines; b, larva 9-16 mm. long with 3 telson spines; d, larva 11-30 mm. long with 3 telson spines. In the statement of the number of segments in the inner ramus of the antenna of the larvae from St. 374 (Table XX) the number has been expressed as the sum of two, of which the first indicates peduncular and the second flagel- lar segments. In Table XXIII the length frequency of the larvae according to the number of terminal spines in the telson and the state of the inner ramus of the antenna are given. None of the seven- or six-spined larvae possess segmented antennal inner rami. Among five-spined larvae 89-7 per cent have unsegmented and 10-3 per cent have segmented antennal inner rami. The number of four-spined larvae is incon- siderable but all have segmented inner rami. By the time the larvae are three-spined, only a very small minority are unsegmented, and in the two- and one-spined larvae no unsegmented antennae occur. The analysis shows therefore that the change in the form of antenna is not restricted to one stage but may take place in five-spined or Fig. 14. Segmentation of antennal inner ramus. three-spined larvae. It also shows that, within a,Thysanoessa maoura; b,Eiiphausiafrigida; c, E. siiperba. Diagrammatic, a and b after Rustad. DEVELOPMENT OF EUPHAUSIA SUPERB A 63 narrow limits at all events, the size does not determine the point at which the antennae should assume the segmented form ; for instance, in the five-spined group there is an overlap of i mm. in the length distributions of the unsegmented and segmented forms. At the same time, however, it should be noted that of the five-spined larvae which Table XXIII. Length frequency of Fur cilia larvae from St. 374 divided into groups according to the numbers of spines on the telson and the form of the inner ramus of the antenna Length classes mm. Seven Six Number of telson Five Four spines Three Two One Total unsegmented Total segmented -0 1) c u a BO % C P ■a u c u a t>0 u -i|m'-'|NNi-i|||||||l| 0 IT) II 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 " 1 1 1 1 1 1 1 1 1 1 1 1 1 - ro 0 H 11 1 1 1 1 1 1 1 1 [ 1 1 1 ^^ ^ t^^ ^OONHHTt-j 1 1 1 1 j 1 [ 1 CO N 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 " 1 1 " M 1 1 1 1 1 1 1 1 1 N ro 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 c-^ N u-i^o ^t^""^!!!!!!!!! CO to 1 1 1 1 1 1 1 1 1 1 II 1 1 1 1 1 " 1 1 " 1 1 1 1 1 1 1 1 1 1 1 N N 1 1 1 1 1 1 1 1 |iHrou-,vn^ Ti--o I'^'^'-'llllllllllll C<1 ro lllllllllllll"'^'^l'^^"llllllllllll T|- 1 M 1 1 1 1 1 1 1 II 1 1 "" 1 1 " 1 1 1 1 1 1 1 1 1 1 1 1 1 f^ 10 ||||l|l[|MNu-)inrOMN||i-i||||||||||||| 0 M l^ 1 1 II 1 1 1 1 1 1 " 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 " M 'a I 1 " 1 M m 0 t^OO ^>Ot^"Tt-l«i-H||l||l|ll|||ll| II 1 "NNNN 1 lllllllllllllll 00 10 1 1 1 1 1 1 1 -^ "O t^O vDmt)-|p-ii-(|||I||II|II|II( lllllll tHHH^H 1 lllllllllllllll 10 vO 1 1 1 1 1 1 " " ">'*-^ 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 )-H r>- l|w|N"-)avNOiot^Ml|l|ll|l|||l|||l|||| 0 c 0 t^ " ^ --^1 " i 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2" Mandibular palp spine number u c 0 c ^ ^0^'+0^^a^'t-0^^0^'^-0^-f-O^^G^'*-C^'^-C^■^-C^-1-0^■^0^-^0^'t•0^^0^ « N -^r^ONN -^r-^ON '^t^<^N '^r^ON -^t^ONN ^I^ON -^r^ONN rf-i>.ON J2 . 10 10 10 u-j>o vjb 0 0 r^ r^ t^ f^ob obobobo^a^6^6^6666'HlHMlHN^JN^^ 0 ^Siiiiiiiiiiiiiiii[i!iTTTTTVTi!iii '5 C 0 LOO »00 u-iO i^O»OOtoO loO 100 100 >J-)0 100 »00 100 u-iOi^O 10 C io io u-i io^ ^O ^6 sO r^ t^ JC^ r^ob obobobONOvONONOOOO'^'^'^'^NNNN ^ ^„„„». 68 DISCOVERY REPORTS also given. The numbers of two- and one-spined larvae are too small to be really in- dicative of what happens, but from the remaining groups it is clear that the change in form of the rostrum can take place in any one of a number of larval stages, the spinous form tending to be more common in the later than in the earlier stages. Table XXVIII. Presence or absence of the rostral spitie in late Fiircilia larvae Number of spines on the telson Rostral spine Present Absent Number °/ /o Number 0 ' / 0 7 6 5 4 3 2 I 21 5 36 2 29 I 4 34 55 57 67 76 55 67 40 4 27 I 9 I 2 66 45 43 33 24 50 33 The examination of these larvae shows that development in Euphausia superba in the later Furcilia stages is not a rigidly fixed process in the degree of development of par- ticular appendages in relation to one another. It is only possible to indicate in the most general way the track that development takes. The analysis also shows what has been referred to at the beginning of this section, namely that in this species the change in form of the antenna is preceded by changes equally fundamental in character in the mandible and the first thoracic appendage. This furnishes a strong argument against the use of the term Cyrtopia for larvae with altered form of the antenna. It may be added that John in his paper on the southern species of Euphausia, about to be published, shows that in different species the change in form of the antennae can precede, or accompany the changes in the mandibular palp and ist thoracic limb. DESCRIPTION OF FURCILIA STAGES For the purpose of this description the Furcilia forms are classified in the following stages : First Furcilia (p. 69). Pleopods non-setose. In this stage are included all larvae, in advance of the 3rd Calyptopis, having simple non-setose pleopods. Second Furcilia (p. 76). Pleopods setose, postero-lateral and terminal telson spines unaltered. This stage includes larvae with setose pleopods, up to five pairs in number, having seven terminal spines on the telson and the postero-lateral spines unaltered. Third Furcilia (p. 84). All pleopods setose, seven terminal spined telson, postero- lateral spines altered. Fourth Furcilia (p. 88). Five terminal spines on the telson. Fifth Furcilia (p. 92). Three terminal spines on the telson. DEVELOPMENT OF EUPHAUSIA SUPERB A 69 Sixth Furcilia (p. 96). One terminal spine on the telson, postero-lateral spines not reduced in number. This account excludes description of larvae with six-, four-, or two-spined telson; these are rarely occurring aberrant forms which in development are intermediate be- tween those larvae whose telson spine numbers are respectively one greater and one less. Tables giving particulars of these forms are inserted in their appropriate positions. FIRST FURCILIA Included in this stage are the following larvae in advance of the 3rd Calyptopis: (a) Larvae with no pleopods. (6) ,, I pair of non-setose pleopods. V') It 2 ,, ,, ,, {d) .. 3 (e) >. 4 (/) .. 5 All these forms, with the exception of (6), are represented in the material examined. In Table XXIX the non-setose larvae have been separated into groups according to the number of pleopods. At the end of each subdivision the number of larvae, their range of size and average length are stated. An examination of the number of larvae recorded in each subdivision demonstrates the dominance of the {e) and (/) forms, particularly the latter, and the comparative rarity of the others. It will be noted that there is a direct correlation between the number of pleopods and the average sizes in the different groups. For all the ist Furcilia the average length is 5-30 mm., and the range is from 3-50 to 6-50 mm. Larvae having two, three, four and five pairs of non-setose pleopods were examined in detail. The eyes project beyond the edge of the carapace which has two emarginations an- teriorly between which is the rostral plate (Figs. 15^, h,j). In the two-pleopod form the anterior edges of the rostral plate form a very obtuse angle, but in the larvae with more pleopods the rostrum is evenly rounded, reaching not quite to the distal margin of the basal antennular segment in the smaller larvae and still more distant from it in the larger larvae. The smaller larvae have no spine on the postero-lateral margin of the carapace ; it is present, but inconspicuous, in the larva with four pairs of pleopods and is quite a distinct denticle in the largest larva in this group (Figs, i^ a, b, c). The telson (Figs. 15 4 it ) > 2 ,, 2 ,, id) >) 4 yy ) 3 »» " (e) T) 4 yy , all setose. if) „ 5 yj , 2 setose, 3 non-setose. ig) „ 5 ») y 3 >' " {!') ■)■) 5 ») y *+ " »> U) ») 5 »> „ all setose and having the spines of the telson unaltered. The above combinations of setose and non-setose pleopods actually occur in the samples examined. It is possible, however, that other variations may be found, for in- stance, five pleopods of which one pair is setose and four pairs non-setose. In Table XXX the frequency of occurrence of these larvae with respect to the number of stations and the number of individuals is given. The average lengths are stated in the same manner as in the previous Furcilia stage. For the 2nd Furcilia the average length is 6-1 1 mm. and the range between 3-96 and 7-92 mm. The description of this stage is based chiefly on detailed examination of larvae of the forms (/), (g), (h) and (j) in the list given above. The larva having the (/) form of pleo- pod development was of small size with the fifth pair of legs not nearly so advanced as in the larva having five pairs of non-setose pleopods described in the previous Furcilia stage. In the (/) larva the pleopods of the fifth pair were small hardly distinguishable buds, yet the first two pairs were definitely setose. The degree of development of the thoracic limbs indicated that the larva was certainly E. siiperba. In this stage the rostrum is still evenly rounded without any median spine and does not reach nearly to the distal margin of the basal peduncular segment of the antennule (Fig. 17 b, c, d, e). There is a small lateral denticle on the postero-lateral corner of the carapace (Fig. 17 a). The telson is seven-spined terminally and has the postero-lateral spines unaltered (Fig. 18 a); that is to say, the innermost postero-lateral has not the a, lateral aspect of larva with all pleopods setose (x 12); h, dorsal aspect of larva with all pleopods setose ( x 14); c, antennule and rostrum of larva with 3 setose, 2 non-setose pleopods ( x 35); d, antennule and rostrum of larva with 4 setose, i non-setose pleopods ( x 35); e, antennule and rostrum of larva with 5 setose pleopods ( x 35); /, uropod and telson of larva with 3 setose, 2 non-setose pleopods ( x 35); g, uropod and telson of larva with 4 setose, i non-setose pleopods (X35); h, uropod and telson of larva with 5 setose pleopods ( x 35); j, antenna of larva with 5 setose pleopods ( x 35); k, mandibular palps of larva with 5 setose pleopods ( x 83) ; /, first maxilla of larva with 5 setose pleopods, plumose condition of exognath bristles and spinules on terminal bristles of palp not indicated (x 165); m, second maxilla of larva with 5 setose pleopods ( x 35). Fig. 17. Second Furcilia. 78 DISCOVERY REPORTS Table XXX. Occurrence of Second Furcilia Station Date Depth No. of No. Length range Average length m. larvae measured mm. mm. {a) 3 pleopods : 2 setose, i non-setose WS 527 30. iii. 30 250-100 {b) 4 pleopods : i setose, 3 non-setose 8 2 3-96-4-04 4-00 WS 527 30. iii. 30 250-100 (f ) 4 pleopods : 2 setose, 2 non-setose 4 I 4-58 4-58 WS 527 30. iii. 30 250-100 (d) 4 pleopods : 3 setose, i non-setose 4 I 4-54 4-54 WS 527 30. iii. 30 250-100 {e) 4 pleopods : all setose 4 I 4-50 4-50 WS527 30. iii. 30 250-100 32 8 4-63-5-29 4-86 WS529 2. IV. 30 500-250 2 2 4-58-5-00 479 (/) 5 pleopods : 2 setose, 3 non-setose^ 34 10 4-58-S-29 4-84 661 2. iv. 31 360-0 ( cr) 5 pleopods : 3 setose, 2 non-setose I I 4-79 4-79 WS527 30. iii. 30 100-50 2 I 4-38 4-38 250-100 4 I 5-25 5-25 (/;) 5 pleopods : 4 setose, i non-setose 6 2 4-38-5'25 4-82 198 3. iv. 27 50-0 2 2 6-63-6-84 6-74 375 21. iii. 30 500-250 3 3 5-2I-579 5-51 639 9. 111. 31 50-0 I I 5-92 5-92 853 19. iv. 32 1 19-0 I I 5-96 5-96 861 27. IV. 32 109-0 3 3 5-00-5-88 5-53 887 27. V. 32 120-0 2 2 5-08-5-33 5-21 WS201 22. iv. 28 500-250 I I 6-00 6-00 WS202 23. iv. 28 500-250 2 I 5-50 5-50 WS527 30. iii. 30 100-50 4 4 S-04-5-I3 5-10 250-100 116 29 4-58-5-58 5-06 WSS29 2. iv. 30 750-500 I I 5-33 5-33 137 49 4-58-6-84 5-47 1 This pleopod arrangement was found in one larva only, taken from a young Fish Trawl net and not from a 70-cm. vertical net. DEVELOPMENT OF EUPHAUSIA SUPERB A Table XXX (com.) 79 Station Date Depth m. No. of larvae No. measured Length range mm. Average length mm. ij) 5 setose pleopods; spines on telsoD unaltered 199 3. iv. 27 500-250 5 5 6-33-6-83 6-62 200 4. IV. 27 300-250 I I 6-83 6-83 201 5- iv. 27 50-0 2 2 6 -46-7 -00 6-73 250-100 I I 6-67 6-67 . 203 5. iv. 27 250-100 I I 5-33 5-33 204 6. iv. 27 500-250 8 5 5-92-7-00 6-58 302 21. i. 30 250-100 I I 6-92 6-92 303 21. i. 30 500-250 2 2 6-08-6-25 6-17 305 21-22. i. 30 250-100 I I 6-46 6-46 319 29-30. i. 30 100-50 2 2 5-63-5-7S 5-69 323 31- i- 30 500-250 2 2 5-63-6-00 5-82 335 4-5. ii. 30 50-0 I I 6-00 6-00 338 5- "• 30 225-100 I I 6-o8 6-o8 342 7. ii. 30 500-250 3 3 6- 17-6-83 6-6i 343 7. ii. 30 250-100 I I 6-46 6-46 353 9. ii. 30 250-100 I I 6-63 6-63 354 9. ii. 30 500-250 I I 6-75 675 360 24. n. 30 50-0 I I 5-83 5-83 361 25. ii. 30 50-0 I I 575 575 100-50 3 3 5-96-6-67 6-24 250-100 140 5° 5-83-7-33 6-62 362 25. ii. 30 50-0 7 7 6-46-7-29 675 250-100 2 2 6-42-6-79 6-61 365 2. iii. 30 50-0 2 2 7-25-7-42 7-34 250-100 2 2 7- 17-7-29 7-23 368 8. iii. 30 50-0 I I 7-00 7-00 100-50 I I 6-50 6-50 250-100 12 12 6-25-7-08 6-56 369 9. iii. 30 50-0 I I 6-92 6-92 250-100 8 8 6-54-6-92 6-70 372 18. iii. 30 50-0 7 7 5-58-6-04 5-81 100-50 3 3 5-63-5-92 574 373 19. iii. 30 50-0 I I 6-17 6-17 374 20. iii. 30 50-0 I I 5-67 5-67 375 21. iii. 30 500-250 8 8 5-21-6-67 5-91 635 7. iii. 31 50-0 I I 5-92 5-92 855 20. iv. 32 125-0 8 8 6-54-6-96 675 125-0 5 I 6-25 6-25 861 27. iv. 32 109-0 61 61 5-00-6-71 5-88 270-138 44 44 5-25-7-00 606 862 28. iv. 32 102-0 6 6 5-63-6-38 5-99 887 27. V. 32 120-0 9 9 5-79-6-75 6-19 86-0 19 19 5-92-6-71 6-30 5-0 6 6 6-17-6-83 6-38 WS 199 20. iv. 28 500-250 17 15 6-29-7-21 6-74 WS200 21. iv. 28 500-250 49 42 6-33-7-92 ' 6-87 750-500 3 2 6-54-7-17 6-86 WS201 22. iv. 28 50-0 2 2 6-13-6-42 6-28 250-100 I I 5-42 5-42 WS202 23. iv. 28 100-50 4 2 5-00-6-00 5-50 250-100 2 2 5-29-6-46 5-88 WS524 2. iii. 30 500-250 2 2 5-92-6-00 5-96 WSS27 30. iii. 30 50-0 8 2 5-2I-575 5-48 100-50 10 10 5-08-6-21 5-55 250-100 224 56 4-63-6-33 5-59 WS529 2. iv. 30 500-250 3 3 5-63-6-21 5-86 750-500 3 3 5-63-6-I3 5-95 722 439 4-63-7-92 6-22 8o DISCOVERY REPORTS basal portion greatly broadened as in subsequent stages. The telson is from 2| to 3 times as long as its greatest width. The uropods extend to, or a very little way beyond, the lateral spines of the telson. The setae on the margins of the exopod and endopod are in- creased in number — as many as twelve being found on the outer ramus and ten on the inner ramus (Fig. 17/, |', h). Antennule. The antennules are not altered in general structure, consisting as they do of three peduncular segments with two flagella distally (Fig. i'] c, d, e). The latter are unsegmented except in the largest larvae included in this stage (Fig. 17 e); in these the flagella may show faint indications of division into a proximal and distal segment. The sensory filaments of the flagella are situated away from the tip of the inner side of the outer flagellum. The forward spinous lateral projection of the basal peduncular seg- ment does not extend as far as the distal margin of the distal peduncular segment. As in the previous stage it is armed with spinules along the inner margin. There are also several spinules around the base of this spine on the basal peduncular segment. The flagella are shorter than the combined lengths of the two distal peduncular segments. Antenna. The antennae are still unchanged (Fig. 177). Mandible. In this stage the first signs of the changes in the mandible leading to the adult form become noticeable. The palp tends to be more elongate in the larger larvae ; for example, in the larva with all pleopods setose it is an unsegmented digitiform pro- cess. The specimen figured shows segmentation within the integument and a long spine embedded in the soft tissues (Fig. 17 k). In the smaller larvae the palp is a short conical process similar to that found in earlier stages. There are alterations in the tooth portion of the mandible which are most pronounced in the largest larvae. Sars (1885), referring to Nyctiphanes australis (p. 153), says in his description of the Calyptopis mandible : The armature of the cutting edges is well developed though somewhat dissimilar to that in the adult animal, resembling rather the armature found in some Mysidae. As it occurs in the latter the anterior part is divided into several strong, and very closely arranged, teeth forming together two partly superposed portions which in the two mandibles exhibit a somewhat different shape. Between this dentate part and the molar protuberance occurs a short dentiform projection, and at the base of the latter is affixed on both mandibles a very peculiar narrow plate expanded at the extremity and having the apical edge finely denticulate. This plate, wanting entirely in the adult animal, would appear to be movably connected with the mandible since it is very easily disengaged. That part of the incisor process which could in the earlier stages be recognized as a pedunculated projection bearing small denticulations on the inferior border distally becomes reduced in size in the smaller larvae belonging to this stage (Fig. 186) and in the larger disappears altogether. The lacinia mobilis also is less conspicuous in the larger larvae of this furcilia stage. First maxilla. The general structure is as in 3rd Calyptopis, that is, consisting of a palp, outer and inner masticatory lobe and an exognath (Fig. 17ft). The advance in development from the previous stage is evident in the palp, which is now composed of a single segment, and in the masticatory lobes, where an increase in the number of spines has taken place. In the larva with three pairs of setose pleopods the outer lobe DEVELOPMENT OF EUPHAUSIA SUPERB A 8i has eight marginal spines and one submarginal ; the inner lobe spine number is un- changed. In the larva with four pairs of setose pleopods there are at least nine marginal spines on the outer masticatory lobe and a similar number on the inner. In the larva with all pleopods setose there are only eight marginal bristles on the outer masticatory lobe, but there are three submarginal bristles; on the inner there are nine marginal bristles. Second maxilla. The 2nd maxillae are as in the previous stage (Fig. 17 m). First thoracic limb. In the smaller larvae in this group the ist thoracic appendage (Th. I) is as in previous stages, but in the larger larvae there is an advance in develop- ment marked by the elongation of the inner ramus and by its further segmentation into more than two. The endopod is never more than about twice as long astheexopod(Figs. iS c, d, e). Thoracic limbs II-VII. The thoracic limbs II-VI are much larger than Th. I. Also in this stage the flexion of the limbs which is found in adults becomes noticeable ; that is to say, in the endopod the two proximal segments are directed forwards and downwards, and the three distal segments reflexed so that they point downwards and backwards. Of the larvae examined that having three pairs of setose pleopods has the thoracic limbs IV-VI of the form shown in Fig. 18 c. The 2nd thoracic limb endopod is longer than the ist but shorter than Th. Ill or Th. IV which are the most developed of these appendages. The fifth limb endopod is indistinctly five-segmented, and the sixth, which is much the shortest, is still "sausage-shaped" and not so densely setose as the more anterior limbs. The setae on the endopods, which are chiefly situated along the inner margin of the limb, are of two kinds — short plumose setae projecting inwards, and longer plumose setae directed anteriorly. This applies to the proximal segments of the endopod; the distal three segments are not so heavily setose. The exopods, with the exception of the Vlth, are more lamellar in form than in the pre- vious stage ; setae are present on the outer and inner margins as well as on the tips. In Th. Ill, IV and V the exopod shows the beginning of differentiation into two portions, a distal natatory part and, proximally, a strongly muscular portion. The distinction is noticeable in the outer margin by the formation of a shoulder-like projection about the middle of this edge, and on the inner margin by a curved projection near the base. In the larvae with four pairs of setose pleopods there is an increase in the number of setae on exopods and endopods (Fig. 18 d). The endopod of Th. VI is segmented into four and the exopod is more lamellar in form than in the smaller larva with three pairs of pleopods. Situated behind Th. VI is a very small inconspicuous protrusion — the first rudiment of Th. VII. In the largest larva — that having five pairs of setose pleopods — the endopods of the limbs II-VI are all distinctly five-segmented and the exopods are clearly divided into two portions, a distal natatory part and a proximal muscular part (Fig. 18 e). Setae extend along both the outer and inner margins to the projections on these described above. Setae on the endopod are greatly developed. Fig. 18 e shows the inner margin of Th. V to illustrate the arrangement of the two sets of setae found in the two proximal DISCOVERY REPORTS Fig. i8. Second Furcilia (continued). a, telson of larva with 5 pairs of non-setose pleopods, spinules omitted ( x 85); b, mandible of larva with 3 setose, 2 non-setose pleopods ( x 165) ; c, thoracic limbs I-VI of larva with 3 setose, 2 non-setose pleopods ( x 35); d, thoracic limbs I-VI of larva with 4 setose, i non-setose pleopods, bristles on limbs omitted except on Th. VI (X 35); , ,r, , 1 • ., thoracic limbs I-VI of larva with 5 pairs of non-setose pleopods, bristles omitted except V lateral view of Th. V, drawn to show inner aspect of limb ( x 35). DEVELOPMENT OF EUPHAUSIA SUPERB A 83 segments ; it also shows the natural flexion of the limb which is typical of thoracic ap- pendages from this stage onward to the adult. Between Th. VI and the first abdominal segment, in the larva having five pairs of setose pleopods, there is a small process on each side, which is flattened antero-posteriorly and is the vestige of Th. VII (Fig. 19 a). In this aspect it appears roughly semicircular in outline with a small emargination near the external border. The ill-defined outer lobe is the developing gill, while in the inner part is the luminous organ belonging to this appendage which will develop in the next stage. LN \ s '^ VI " I II III IV V Fig. 19. Second Furcilia (continued). a, lateral view of bases of thoracic limb VI and pleopod I with, between them, thoracic limb VII, a Th. VII seen from behind ( x 83); b, gills on thoracic limbs II-VI in larva with 4 setose, i non-setose pleopods ( x 83); c, gills on thoracic limbs II-VI in larva with 5 setose pleopods ( x 83); d, pleopods I-V in larva with 3 setose, 2 non-setose pleopods ( x 35); e, pleopods I-V in larva with 4 setose, i non-setose pleopods ( x 35); /, pleopods I-V in larva with 5 setose pleopods (x 35); g, appendix interna ( x 360); h, setae within the integument of a non-setose pleopod ( x 165). Gills. In the smaller larvae included in this stage the gills are at much the same stage of development as in the larva having five pairs of non-setose pleopods, the only ad- vance noted in the larva having three pairs of setose pleopods is that the gill of Th. V is now bilobed. In the larva with four pairs of setose pleopods the branchiae of Th. II-VI are all distinctly bilobed (Fig. 19 b). The gills of Th. II-VI in the larva with all pleopods §4 DISCOVERY REPORTS setose have increased in size and show the beginnings of a third lobe at the base of the inner lobe (Fig. 19 c). The gill of Th. VII has already been described in connection with the limb rudiment. There is no sign in this Furcilia stage of a gill on the ist thoracic limb. Pleopods. The variety of pleopod forms within this stage is indicated in the list at the beginning of this section. The appearance of the pleopods in larvae having respectively three, four and five pairs setose is shown in Fig. ig d, e, f. The non-setose pleopods are typically as in the larger larvae in the preceding stage, that is, having the exopod marked off' from the protopodite and the endopod merging into it. Within the integu- ment of the non-setose pleopods there can be distinguished the setae with which the exopod will be furnished when the larva moults (Fig. 19 h). The setose pleopods have six to eight setae on the exopod, which is now divided off from the protopodite, and from one to three setae on the endopod. At this stage the appendix interna appears on the inner margin of the endopods of those pleopods which are setose (Fig. 19 ^). In the smaller larvae it is represented by a small hook-like process on the endopod, whilst in the larva with five setose pleopods a number of these hooks are borne on a digitiform projection from the inner margin of the endopod. Luminous organs. In addition to those on the ocular peduncle and at the base of Th. II there are the following unpaired luminous organs distinguishable between the pleopods : The larva having 2 setose pleopods and 3 non-setose has luminous organs on abdominal segments i and 2. 3 setose pleopods and 2 non-setose has luminous organs on abdominal segments i, 2 and 3. 4 setose pleopods and i non-setose has luminous organs on abdominal segments i, 2, 3 and 4. 5 setose pleopods has also four abdominal luminous organs. The development of the luminous organs on the abdominal segments thus coincides with the development of setae on the pleopods. The luminous organ on Th. VII cannot yet be distinguished. THIRD FURCILIA The reasons for recognizing this form of larva as a separate stage are given above on pp. 41-9. The frequency of occurrence of these larvae and their average lengths at dif- ferent stations are stated in Table XXXI. For all the 3rd Furcilia the average length is 7-32 mm., and the range lies between 579 and 875 mm. The two previous Furcilia stages are recognizable by the number and degree of de- velopment of the pleopods. In this stage other criteria must be adopted, as the pleopods have reached their full number and are all setose as in the largest larvae of the preceding stage. The most convenient character for this purpose is the form and number of the telson spines ; the number is the same as in the earlier Furcilia stages, but the innermost postero-lateral spine, as shown in Fig. 20 c, is altered, being greatly widened at the base. Other distinctions in development in this Furcilia stage will be mentioned below. a, lateral aspect ( x 9) ; b, rostrum and antennules ( x 35) ; c, telson, lateral spinules ( x 85); d, uropod and telson ( x 35) ; e, antennule ( x 35) ; /, mandible ( x 83) ; Fig. 20. Third Furcilia. g, lacinia mobilis ( x 360); h, thoracic limb I showing arrangement of bristles ( x 35) ; j, thoracic limb III (X35); k, gills I-VII, base of Th. VI with Br. VI shown entire, Th. VII consisting of luminous organ and gill ( x 83); /, pleopods I-V ( X 35). 86 DISCOVERY REPORTS Table XXXI. Occurrence of Third Furcilia 5-setose pleopods: spines on telson altered Station Date Depth No. of No. Length range Average length m. larvae measured mm. mm. 194 28. iii. 27 50-0 I I 7-29 7-29 197 3. iv. 27 100-50 7 7 7-29-8-I7 7-70 1000-750 I I 8-00 8-00 198 3. iv. 27 50-0 4 3 7-42-8-I3 774 100-50 5 4 7-42-7-83 7-64 199 3. iv. 27 50-0 I I 7-50 7-50 250-100 I I 7-58 7-sS 500-250 19 17 7-0O-7-92 7-45 750-500 21 21 6-75-8-00 7-41 200 4. iv. 27 300-250 2 2 7-58-7-7I 7-65 . 201 5. iv. 27 50-0 3 3 7-08-7-58 7-32 203 5. IV. 27 50-0 6 4 7-04-779 7-43 100-50 I I 8-00 8-00 250-100 I I 771 7-71 204 6. iv. 27 500-250 13 13 6-92-7-67 7-28 206 6. iv. 27 250-100 I I 7-00 7-00 304 21. i. 30 500-250 2 2 7-58-7-63 7-6i 337 5- »• 30 250-100 I I 7-54 7-54 360 24. 11. 30 50-0 I I 7-38 7-38 361 25. ii. 30 250-100 16 8 8-00-875 8-46 362 25. ii. 30 50-0 I I 8-04 8-04 368 8. iii. 30 250-100 I I 8-25 8-25 369 9. iii. 30 50-0 I I 8-o8 8-08 250-100 I I 8-25 8-25 372 18. iii. 30 50-0 3 3 7-08-7-29 7-11 100-50 I I 6-96 6-96 373 19. iii. 30 50-0 I I 7-58 7-58 374 20. iii. 30 50-0 9 9 6-38-7-58 6-99 100-50 I I 7-63 7-63 375 21. iii. 30 50-0 I I 6-67 6-67 500-250 10 10 6-25-7-63 6-92 85s 20. iv. 32 125-0 2 2 7-58-7-67 7-63 861 27. iv. 32 109-0 29 29 6-67-7-83 7-27 270-138 28 28 671-8-15 7-26 862 28. iv. 32 102-0 3 3 7-04-7-38 7-20 220-98 I I 6-71 6-71 887 27. V. 32 120-0 4 4 6-58-6-96 675 86-0 5 5 6-58-8-04 6-96 5-0 7 7 6-67-7-21 6-87 888 28. V. 32 98-0 23 18 6-92-7-50 7-29 240-90 I I 7-13 7-13 WS 199 20. iv. 28 500-250 3 3 7-33-8-04 774 WS200 21. iv. 28 500-250 13 13 7-75-8-So 8-21 WS201 22. iv. 28 50-0 I I 7-33 7-33 WS427 28. iv. 29 50-0 I I 7-o8 7-08 WS 527 30. iii. 30 50-0 12 3 6-75-7-08 6-94 100-50 I I 6-58 6-58 250-100 16 4 5-79-6-79 6-46 WS529 2. iv. 30 500-250 13 13 6-33-7-42 7-02 750-500 29 29 5-92-7-67 6-99 329 289 5-79-875 7-32 DEVELOPMENT OF EUPHAUSIA SUPERB A 87 The carapace is, with minor modifications, as in the previous stage (Fig. 20 a). The posterior margin is more concave and the rostrum sUghtly more elongate (Fig. 20 b), but the latter is still rounded anteriorly except for a minute spine which may or may not be present on the apex of the anterior border. The telson is seven-spined terminally and has the modification (Fig. 20 c) of the postero-lateral spine just mentioned. The length of the telson is about 2,3 times the width. The uropods extend beyond the lateral spines of the telson and are more setose, having in the specimen figured thirteen setae on the outer ramus and twelve on the inner (Fig. 20 d). Antennule. The flagella of the antennules are multi-articular: in the specimen figured five segments are found (Fig. 20 b). They are roughly equal in length to the two distal peduncular segments. The spine on the outer margin of the basal peduncular segment is shorter than in the preceding stage, extending not more than about two- thirds of the length of the distal peduncular segment (Fig. 20 e). Antenna. The antennae are still unchanged. Mandible. The palp (Fig. 20 f,g) is three-segmented with a long terminal spine. The variations in form of this appendage in this stage are dealt with in another section (p. 65). The lacinia mobilis is still present with fine hairs surrounding the base. First maxilla. The palp of the ist maxilla is unsegmented and has two extra bristles on the inner margin, making seven bristles in all. The outer masticatory lobe in the specimen examined has nine marginal and two submarginal bristles. The inner masti- catory lobe has ten bristles, five of which are heavily setose and more robust, the re- mainder slighter, smaller, and only finely plumose. The exognath is unchanged. The divisions separating the ist and 2nd segments of the protopodite (sympod) cannot be distinguished in this or in earlier stages, although a suggestion of such segmentation was indicated in the 3rd Calyptopis. Second maxilla. The 2nd maxillae are unchanged. First thoracic limb. The endopod of this appendage is definitely five-segmented, armed with numerous bristles and about three times as long as the exopod (Fig. 20 Ii). The joint between the basal segment and the ist endopod segment is very indistinct in this as in all the other limbs. The exopod is still primitive in form, bearing four bristles distally. Thoracic limbs II-VII. The thoracic appendages II-VI are proportionately more elongate than in the previous stage. The exopods are rather longer in relation to the most proximal segment of the endopod and more setose than in the previous stage. Only Th. Ill has been figured to show the size compared with Th. I (Fig. 20 j). Th. II is still very slightly shorter than Th. III-VI. Th. VII is represented by two portions made up of a branchial lobe externally and a luminous organ internally (Fig. 20 k). There is no trace of Th. VIII at this stage. Gills. The gill on Th. I is small and inconspicuous. Those on Th. II-V are trilobed, having the central branch longer than the remaining two. The gill of Th. VI is bilobed 88 DISCOVERY REPORTS with the beginnings of a third at the base of the inner lobe, and that on Th. VII is a single lobed process bearing the luminous organ on the peduncle (Fig. 20 k). Pleopods. All the pleopods are setose, proportionately longer than in the preceding stage and have the endopod longer in relation to the exopod (Fig. 20 /). The pleopods are more setose, with eight to ten setae on the exopod instead of six to eight as in the largest larvae of the preceding stage. Pleopod V is more nearly similar to the remainder, but the inner ramus is still less developed than the others. The appendix interna is a digitiform process furnished with hooks. Luminous organs. In addition to the light organs present on the largest larva belonging to the previous stage, the organ of Th. VII can now be distinguished (Fig. 20 k vii). Table XXXII records the occurrence of larvae having six terminal spines on the telson. As stated on p. 51 above, larvae with an even number of terminal spines are in their general structure intermediate between those having respectively one more and one fewer telson spines. The infrequency of occurrence of the six spined form compared with those having seven and five shows that the former number must be regarded as a rare abnormality. Table XXXII. Occurrence of Fiircilia larvae with six terminal spines on the telson Station Date Depth m. No. of larvae No. measured Length range mm. Average length mm. 197 374 375 862 WS527 3. iv. 27 20. iii. 30 21. iii. 30 28. iv. 32 30. iii. 30 100-50 50-0 500-250 102-0 50-0 2 7 2 3 I 2 7 2 3 I 8-54-8-92 6-46-7-54 6-88-7-08 7-29-7-63 7-38 8-73 7-19 6-98 7-45 7-38 15 15 6-46-8-92 7-43 FOURTH FURCILIA The frequency of occurrence and average lengths of larvae belonging to this stage are stated in Table XXXIII. The average length of all the larvae is 8-oi mm., and the range between 6-67 and 9-92 mm. The rostrum (Fig. zi a,h) is more elongate and narrower than in the previous stage and may have a small denticle at its apex. The lateral denticle is present. The telson (Fig. 21 d)is about four times as long as broad, bearing five terminal spines and three postero-lateral spines. The innermost postero-lateral is more modified than in the previous stage (Fig. 21 c)\ the broadened portion extends much nearer to the tip and the setae are missing from the inner margin proximally. The uropods extend beyond the lateral spines on the telson and may have one or two more setae than in the previous stage. DEVELOPMENT OF EUPHAUSIA SUPERB A 89 Antennule. The flagella are multi-articulate and rather longer than the two distal peduncular segments. The spine on the outer margin of the basal peduncular segment extends to about the middle of the distal segment (Fig. 21 b). A lateral view of the peduncular segment is shown in Fig. 21 e. Fig. 21. Fourth Furcilia. a, lateral aspect ( x 9) ; /, antenna ( x 35) ; b, rostrum and antennules ( x 35) ; a, mandible ( x 83) : g', tip of palp ( x 83) ; c, telson ( X 83); fi, first maxilla, spinules on bristles omitted ( x no); d, uropod and telson ( x 35) ; y, second maxilla ( x 35). e, antennule, lateral view ( x 35); Antenna. There is only a faint indication in the antennae of differentiation into a scale and flagellum in the specimen figured (Fig. 21 /), but the variations in the form of this appendage are dealt with on p. 60. In this specimen the endopod is broader in proportion to its length than in earlier stages, but the number of bristles is unchanged. 9° DISCOVERY REPORTS Table XXXIII. Occurrence of Fourth Fur cilia 5 terminal spines on telson Average length Station Date Depth No. of No. measured Length range m. larvae mm. mm. 196 3. iv. 27 50-0 I I 8-i6 8-16 197 3. iv. 27 100-50 6 6 8-58-9-46 9-12 198 3. iv. 27 50-0 5 4 8-38-875 8-55 199 3. iv. 27 50-0 I I 8-58 8-58 500-250 2 2 8-54-9-I7 8-86 203 5. iv. 27 50-0 5 3 8-13-8-88 8-65 204 6. iv. 27 500-250 3 I 8-75 8-75 361 25. ii. 30 100-50 I I 9-75 975 372 18. iii. 30 50-0 9 9 7-04-9-38 8-04 100-50 2 2 7-00-7-38 719 373 19. iii. 30 50-0 6 6 7-o8-9-29 7-92 374 20. iii. 30 50-0 21 21 7-I3-9-92 774 100-50 . 3 3 7-38-9-I7 8-02 375 21. iii. 30 50-0 2 I 771 771 500-250 19 19 7-00-9-25 7-96 861 27. iv. 32 109-0 19 19 775-9-33 874 270-138 3 3 775-9-29 8-50 862 28. iv. 32 102-0 38 38 6-67-8-08 7-41 220-98 2 2 7-50-8-I3 7-82 887 27. v. 32 120-0 I I 7-50 7-5° 86-0 8 8 7-67-8-83 8-19 0-5 2 2 7-67-9-04 8-36 235-115 I I 8-33 8-33 888 28. V. 32 98-0 I I 8-42 8-42 912 24. vi. 32 2-0 16 2 7-63-871 8-17 WS427 28. iv. 29 100-50 2 2 7-79-7-88 7-84 WS527 30. iii. 30 50-0 80 20 7-42-8-46 7-8i 100-50 2 2 6-83-8-33 7-08 WS529 2. iv. 30 500-250 18 18 7-42-8-83 8-28 750-500 10 10 7-08-8-58 7-93 289 209 6-67-9-92 8-01 Mandible. The lacinia mobilis is still present in the mandible. The palp is three- segmented, and in the specimen figured has two spines on the apex of the distal segment of the one, and only one on the other (Fig. 21 g). The variation in the number of spines in the distal segment is dealt with on p. 65. First maxilla. The palp of the ist maxilla is unsegmented with seven bristles from the apex along the inner margin. There are two less conspicuous bristles on the outer side of the palp. The masticatory lobes are very similar to those found in the previous stage but with ten marginal bristles on the outer lobe. The pseudexopod rudiment is visible but not conspicuous in the region of the larval exognath (Fig. 21 //). Second maxilla. The terminal palp of the 2nd maxilla (Fig. 21 j) has four bristles extending from the tip along the inner margin. On the outer side of the maxilla, where the palp is articulated, there is a little rounded process bearing two setae : this is the exopod. The segmentation of this appendage into three basal segments {vide Hansen's Studies on Arthropoda, 11, 1925, pi. xii) is not clear. DEVELOPMENT OF EUPHAUSIA SUPERB A 91 First thoracic limb. The form of the ist thoracic appendage (Fig. 22 a) is not much altered, the endopod is five-segmented and heavily setose, and the exopod, as before, primitive in shape and bearing four bristles distally. The joint at the base of the endopod is not clear. The length of the limb is about three-quarters that of the succeeding thoracic appendage. e III a, thoracic limb I ( x 35); b, thoracic limb II ( x 35); ii-viii) are more developed than previously, those of Th. II and III are five-lobed, Th. IV is also five- lobed with indications of a further lobe developing. The gills of Th. V and VI are similar in development to those of the preceding stage. The gill of Th. VII has four well-developed lobes and two more short lobes developed in proportion to the size of the larva: two diff"erent conditions of this gill have been figured. The gill of Th. VIII has three large lobes and two small ones near the peduncle. Pleopods. There is an increase in the number of marginal bristles (Fig. 26 c i, ii, v) and the appendix interna (Fig. 26 d) has four or five hooks. ADOLESCENT FORMS Rustad (1930, p. 68) gives reasons for the abandonment of the term "post-larval " to describe the stages which succeed those formerly classed as Cyrtopia. He states that the latter stages merge gradually into the adult and points out, what Dr Bargmann has also 13-2 DISCOVERY REPORTS VMI Fig. 26. Sixth Furcilia (continued). a i, thoracic Hmb I, bristles on endopod omitted ( x 35); a vi, thoracic limb VI, bristles on endopod omitted ( x 35); b \-b viii, gills I-VIII, two forms of gill VII showing Th. VII, & viii showing Th. VIII ( ■ 83); ci, f li, fv, pleopodsl. Hand V(:: 35); rf, hooks on appendix interna ( x 165). DEVELOPMENT OF EUPHAUSIA SUPERB A loi observed in Eiiphausia superba, that sexual maturity may be reached before the adult form is attained. It is proposed therefore in the present paper to use the term "adoles- cent" to describe that portion of development between the 6th Furcilia stage and the part of the life history where changes take place associated with sexual maturity. The scope of the paper, so far as adolescents are concerned, is limited to the younger indi- viduals in this phase and does not include reference to those that from their size require special examination to decide whether they are adolescent or adult. By the time the euphausiid reaches the 6th Furcilia stage the major developmental changes have been effected and in appearance it is characteristically a euphausian. Such alterations as take place subsequently chiefly involve the elaboration of existing struc- tures, for instance, increase in the number of setae on the antennal scale, in the number of segments in the antennular flagellum, in the number of lobes on the branchiae, and so on. These changes are gradual and show great individual variation. Attempts which were made to arrange the adolescents in groups or stages dependent on the number of setae on the antennal scale did not yield any satisfactory results, and it was made obvious by inspection of the animals that in other appendages as well no hard and fast pattern of development exists. It should be emphasized again that this generalized development is not an exclusive feature of the adolescents, but that it is incipient, in some characters at any rate, in early Furcilia stages and becomes more and more evident as development goes on. Thus although Furcilia 6 is recognized in the main by having one terminal and three postero- lateral spines on the telson, as opposed to two postero-lateral spines in the adolescent, yet the distinction between the two stages is diffuse and ill-defined. In Appendix I the length frequencies of late Furcilia and early adolescents taken in the i-m. nets have been set out. The time of year when the larvae were obtained, the total number and the number examined, the average length for each phase of develop- ment, the percentage of the total in each phase and the general average have been stated. The first point to be noted from this analysis is that Furcilia 5, representing larvae about 8 mm. in length, although occurring in August and September is in such very small numbers that it may fairly be concluded that the vast majority of the larvae reach the Furcilia 6 stage before the end of the southern winter. Inspection of the table shows that the total number of E. superba at different stations varies very greatly, and that the small numbers at some of the stations tend to give anomalous results. It is possible, however, to get an approximate indication of the re- lative abundance of the two phases of krill principally represented, namely Furcilia 6 and adolescent. In Table XXXVIII the average percentage occurrence of Furcilia 6 is stated. The percentages from which these averages are derived are of catches where the number of euphausians examined was considered sufficiently high to give a trustworthy idea of the proportions in which Furcilia 6 and adolescents were present.^ 1 The arbitrary number of 48 was selected. I02 DISCOVERY REPORTS Table XXXVIII. Average percentage occurrence of the sixth Fur cilia 1928-29 1929-30 1930-31 August 75 September 49 — — October 35 — 19 November 10 16 December — 0 0 January — 0 0 It is seen that there is a decrease from August to December from a stock three-quarters of which are Furcilia 6 to one in which this stage is not represented. 26 24 - A • 22 ■ A A -A 0 I20 A A • t 1 A 0 . I 1- ^18 - 0 0 ^ ^2*^ A u « •«' t) ^,6 - 0 £>. A A6 • A A IX ^A e^ .'^ bJ e A £k A* • 14 . 3 °'^°°® A A • e 0° 0 ® 0 ' A 0* • 12 0 "^ ^ 0 0 ® ° ego 0 0 0 A • A 00 -A .0 -1927-28 -1928-29 -1929-30 CLASS. 10 A 0 0 1 J I 1 1 1 1 AUG SEPT OCT NOV DEC JAN Fig. 27. Average length of second season krill. The smaller signs express average lengths of individual samples, the larger ones the half-monthly average lengths of all the young euphausians taken. An impression of the length range of Furcilia 6 is easily obtained by reference to the appendix. The way in which the upper limit of length overlaps the lower limit of the adolescent size range emphasizes the indefinite nature of the division between the two phases of developmental history. The length range of Furcilia 6 covers 9 mm., extending from 8 to 17 mm. with a mean length of about 12 mm. But it is possible to get ado- lescents 10-5 mm. long, so that there is an overlap in the two length ranges of 6-5 mm. In Fig. 27 the average lengths of the krill from the samples referred to above are de- picted. The smaller signs express average lengths of individual samples and the larger ones the half-monthly average lengths of all the young euphausians taken. The three seasons' catches have been given the designation of classes, that is to say the 1927-8 class represents larvae which originated in that season. They would be adolescents in the 1928-9 season and adult in the succeeding season. DEVELOPMENT OF EUPHAUSIA SUPERB A 103 The figure shows the great range of average size which can be encountered at any one period. For instance, in the month of November the range in average length (not individual length) is between ii-8 and 22-5 mm.— this from samples of the same year class. This range is almost equal to the increase in average length for the whole period under discussion, namely the second half of August to the first half of January. Although there is this great range in size for any one period, there is a definite and fairly regular increase in the half-monthly average size. In the second half of August the average length of the larvae is 1070 mm.; in the first half of January it is 22-18 mm. These figures involve results from two diflferent year classes, but inspection of the figure shows that in those instances where observations were made over the same period in diflFerent seasons, the half-monthly average length shows such a small variation that it seems justifiable to assume that the rate of growth is approximately the same in different years. ANOMALOUS LENGTH FREQUENCIES IN ADOLESCENT FORMS In his reports to the Discovery Committee on the results of the circumpolar cruise, John drew attention to the length grouping of the young krill in the catches obtained. His remarks have been incorporated in the account of the circumpolar cruise on p. 137 below. The larvae from St. 954 fell into two well-defined groups having their maxima at 13 and 18 mm. respectively: this was in September. Later, in October-November, John depicts (Fig. 61) the difference in constitution between krill taken in the Bellings- hausen Sea, in the ice-free water near South Georgia and in the Weddell Sea. The two sets of data involve two different problems : (a) the occurrence of two distinct length groups simultaneously in the one locality, and (b) the occurrence, in different localities at approximately similar times, of young E. superba differing widely in length frequency. There is no obvious explanation for the occurrence of two different length groups in the same locality simultaneously, and the statements made here must be regarded as conjectural. John's report of this occurrence led to the reconsideration of certain records from i-m. net catches where groups of £■. superba had been measured and discarded from the material used in this paper. They were regarded as being outside the range of larval and adolescent lengths here dealt with and were given the provisional designation of "small adults ". Thus at St. WS 279, 13. ix. 28, including the " small adults ", a bimodal curve was obtained having maxima at 12-37 ^^^ 27-4 mm. respectively. There were 163 larvae in the first group and forty-eight in the second, the two groups being separated by 8 mm., from 16 to 24 mm., within which no larvae were found. Again at St. WS 288, 19. ix. 28, the euphausians were arranged in two groups, the one with average length of 13-13 mm. and the other with average length 25-18 mm. There were over 20,000 of the smaller larvae and only thirty-seven of the " small adults " ; the two groups were separ- ated by a space of 5-5 mm. If the three sets of results detailed above from Sts. 954, WS 279 and WS 288 are 104 DISCOVERY REPORTS considered together the peculiar nature of the larger of the two groups is evident. The average growth curve on p. io8 below shows that the average length of the larger group is in each instance much greater than that represented in the graph. Thus in September the normal average is about 12 mm. as opposed to 18 mm. in John's larger group. In October the normal average is about 13 mm. as opposed to 27 and 25 mm. in the "small adults" from Sts. WS 279 and WS 288. Ruud's mean length diagram (1932, p. 41, fig. 8) shows that in October the mean length of adult krill is about 50 mm., so that our "small adults" fall somewhere between the normal adolescent average length and fully adult average length. The origin of these anomalous forms is not clear, and what the factors are that in- fluence their difference in size is open to speculation. It may be that they are derived from batches of eggs laid very early in the summer season that have pushed ahead with development, obtaining the full benefit of the summer season, whereas the majority of the larvae encountered only obtain benefit from the latter part of the summer, having been derived from eggs laid well on in that season. It may be, as John suggests, that specially favourable environmental conditions, for instance a rich diatom flora, may have been the lot of some of the larvae for a greater period than the remainder. Whatever be the true explanation, the fact that forms occur having this peculiarity of length leads to enquiry as to when they become sexually mature. If the growth rate in these eu- phausians is similar to that in normal circumstances there is every possibility that they reach a size at which they are sexually mature before the coming of the second winter. If this does occur the complications to be encountered in unravelling the constitution of the adult krill population are greatly increased. The occurrence of young E. superba differing widely in length frequency in different localities at approximately similar times was described by John in a comparison of larvae from the Bellingshausen Sea, from the Weddell Sea ice-edge and from stations in open water west of South Georgia. The period covered was little more than a month and the contrasted resuhs are referred to on p. 141 and shown in Fig. 61. He suggests that better conditions for growth, such as the more abundant diatom flora, may account for the larger size of some of the larvae as compared with the others. Apart from one instance, mentioned below, no attempt has been made in this paper to correlate the occurrence of diatoms with the krill results. It is regretted that it has not been possible to fulfil the promise implied in Hart's Phytoplankton Report (1934, p. 11). The detailed investigation of this very important field of research will, it is felt, clear up many of the problems connected with the growth and distribution of krill. The correlation is inserted as it has a bearing on John's suggestion concerning varia- tion in richness of the food supply. In Fig. 28 is shown the average length of adolescent E. superba and the abundance of Thalia ssiosir a antarctica at the stations indicated. The presence, but not the abundance, of Clwetoceros socialis is also represented in the figure ; this small colonial form occurred in such great numbers that its abundance could not be estimated satisfactorily. The stations from which the data were taken were all made in DEVELOPMENT OF EUPHAUSIA SUPERB A 105 the South Georgia plankton survey of November 1930. An account of the phyto- planktonic conditions prevailing is given by Hart {loc. cit., pp. 41 et seq.). Concerning Chaetoceros socialis he writes (p. 51): " Chaetoceros socialis .. .was taken at fourteen stations only, all of which were grouped towards the southern extremity of the island, some close inshore, and the others in the neighbourhood in which the pack-ice lingered longest during this spring, and all in the water which from the general nature of the phytoplankton as a whole, almost certainly originated in the western Weddell Sea." Concerning Thallassiosira antarctica he writes: "It also reached its greatest abundance in the western Weddell Sea surface water round the southern end of the island and was comparatively rare to the north and north-west. It was, however, very widely distri- 5- Z4 STN5 a3 < 12 e A 0 0 O 0 © ®- 1926-27 CLASS e -19 27-28 A-1928-29 0-19 29-30 H -1930-31 B -193 1 -32 NOV DEC JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV Fig. 29. Half-monthly average length of Euphausia superba larvae obtained from 70-cm. vertical nets and from the i-m. nets fished during the circumpolar cruise. AVERAGE LENGTH OF LARVAE In Fig. 29 there is represented the half-monthly average lengths of the Euphausia superba larvae obtained from the 70-cm. vertical nets, except for the 193 1-2 series which was taken in the i-m. nets fished during the circumpolar cruise. The scattered distribu- tion of the average lengths gives some indication of the variation which can take place. It is in part due, no doubt, to the different seasons involved and to local environmental DEVELOPMENT OF EUPHAUSIA SUPERB A 107 dissimilarities which must accelerate or retard the growth rate in accordance with favourable or unfavourable conditions. Comparison of size with locality of occurrence gave no satisfactory result, nor is there any clear suggestion that larval development periodically commences sooner in one area than another. There is perhaps a hint that in the South Georgia region this may be so, but it is so ill-defined in the data as to be of little importance. Notwithstanding the great variations in average length which exist at any one time, the figures as a whole show an upward trend from early in the summer season to the end of it, so that by June the larvae have attained an average length of io-6 mm. The June figure is the only record for that period of the year, but it is from a reasonably large number of measurements and may be considered fairly representative. There is not another record until the second half of August when the length of a small number of larvae in a different season is io-02 mm. The averages from August onwards are of small numbers of Furcilia 6 and adolescents and for these a better indication of what happens is obtained from the results of the i-m. net samples set out above. It may be concluded then that the larvae generally reach a length of about 10 mm. by the time the winter sets in and that this size has been achieved in a period of six months. AVERAGE GROWTH RATE Fig. 30 is constructed on the data shown in Figs. 27 and 29 above. It attempts to give some impression of average growth rate during the first year or so of larval life. From November to August the figures represented are the half-monthly average lengths of all the first season larvae taken during that period. From August to January the graph is a repetition of Fig. 27. The interpretation of the curve of average growth is straightforward in this second portion, but the first part requires some comment. The figure for the first half of No- vember is derived from a single individual. The figures for the second half of December and second half of January are from four and five measurements respectively. In the beginning of January and February numbers have increased to 368 and 112 respectively, and in the second half of February and first half of March they are 1966 and 6602. The next three half-months have roughly comparable figures in the thousands. The average lengths during this period show no well ordered progression in size. In January and February these figures are very much greater than in the first half of March, where there is a large population with greatly reduced average length. In the second half of March the average length increases again by about 4 mm. The position of the first portion of the average length curve, as shown in the graph, is suggested by these figures. The large number of very small larvae in the first half of March counteracts the effect of the lesser numbers of greater length before and after that time and makes the upward tendency of the growth curve less steep. It is possible that the conjectured earlier development of larvae in the South Georgia region, referred to in the previous section, is responsible for the differences between the January-February period and March. In the former of the two periods the average 14-2 io8 DISCOVERY REPORTS lengths are derived, in two of the three half-months, from larvae taken near South Georgia and in the third half-month from the South Shetland-South Sandwich area. The low figure for the first half of March is caused by the great numbers of Meta- nauplius and ist Calyptopis taken at stations just north of the South Shetlands in that half-month. But this indefinite division into a north and south area breaks down alto- gether in the second half of March and first half of April, where the contrasted results Fig. 30. Graph showing average growth rate of Euphausia superba during the first fifteen months of larval life. of the two periods are both derived from larvae from the Bransfield Strait and South Sandwich Islands-Burdwood Bank Line. The growth curve can be divided into four parts: a low, gently increasing portion from November to March, a steeper portion between March and June, a part having little or no upward tendency between June and August, and a steeply ascending portion from August onwards. The first part of the curve coincides with the observed period of spawning. The average length of the larval krill population will be kept down by the constant addition of very small developmental forms; it is a period of production and growth. The second part extends from the end of the spawning period to the time when winter conditions become eflFective. During this period the average length is not aflFected by influx of very young DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERBA 109 larvae to the stock — it is exclusively a period of growth. The third division of the graph coincides with the depth of the southern winter when conditions for growth will be unfavourable, and although the June average depends on one sample it may be taken that it represents, roughly at any rate, what actually happens at that time of year. With the coming of spring, environmental conditions become favourable again and growth and development go ahead. The larvae pass from Furcilia 6 to the adolescent phase, so that by the end of the year the former stage of development is no longer encountered in the samples. This curve of average length corresponds fairly closely with the comparable portion of Ruud's Total Growth Curve (Ruud, 1932, p. 45, fig. 10), but in the present figure the first period referred to above is continued for at least a month longer than Ruud's figure indicates. The period of rapid growth prior to winter covers the period April-June, as compared with February-March depicted in Ruud's figure. This last difference affects the length of the winter period of reduced rate of growth. On the whole, however, the results are in fair agreement, and what differences there are probably originate from the widely separated sources from which the present material was obtained. The distribution of records throughout the year with the exception of the month of July helps toward a better understanding of the development and growth of this euphausian than was hitherto possible, and although many of the results cannot be regarded as final and conclusive, they at any rate furnish a basis for future corroboration or contradiction. DISTRIBUTION EGGS REGIONAL DISTRIBUTION In the samples examined there are forty-eight records of eggs from twenty-eight dif- ferent stations. Most of the catches are of small numbers of eggs, but at one station, St. 540, relatively enormous numbers were obtained, of which the maximum occurred in the 500-250-m. net. In the Falkland sector, from which all the data relating to eggs have been obtained, they were found in great abundance near Graham Land and the South Shetlands or in the path of currents which have a component towards these places (Fig. 31). They were also found near the South Orkney Islands and South Georgia and above the Scotia Arc in the vicinity of the South Shetlands. They were also found close to the continental shelf west of Graham Land in the south of the Bellingshausen Sea. The area of greatest abundance of eggs is near the northern end of Graham Land, particularly in the Bransfield Strait; at each group of stations made there between November and February eggs were found in greater or less abundance. There are probably two main reasons why eggs occur in the regions described. Along the west coast of Graham Land and the Scotia Arc, as far as the South Sandwich Islands at all events, the temperature and salinity are roughly the same (vide Deacon, 1933, figs. 12, 13, 24), and these conditions are probably the optimum for eggs to be laid. no DISCOVERY REPORTS These same conditions are continued with small modifications towards the north near South Georgia and the Shag Rocks. The second reason is probably to be found in the bathymetric conditions. West of Graham Land and in the neighbourhood of the South Shetlands, South Orkneys, South Sandwich, South Georgia and the Shag Rocks there are extensive areas of comparatively shallow water, and on the Scotia Arc joining these regions {vide Herdman, 1932, p. 214) the sounding is generally less than 2000 m. If, as will be suggested later, the eggs de- velop near or at the bottom in comparatively deep water, the depths in these regions are probably most suitable. Fig. 31. Distribution of eggs of Euphausia superba. The reason for the particular abundance in the Bransfield Strait area may be that tem- perature and salinity or the bathymetric conditions are particularly suitable. For, in the Bransfield Strait and separated from the surrounding seas by submarine ridges rising to within 250-600 m. of the surface, there are extensive basins with soundings from 500 to 2000 m. These basins may be particularly favourable for the production and develop- ment of eggs. One other striking feature of this region which may make it suitable is the homogeneous condition of the water column. Over the shallow ridges the water is almost completely mixed, and even in the deep basins there are only small changes of temperature and salinity with depth. The eggs were found in nets fished through water of which the temperature varied from — I -So to 2-02° C, but the greater number of catches was in water of temperature DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERBA iii less than o" C, and the greatest numbers of eggs were taken from water of this low temperature. In salinity the conditions found at St. 540 seem to indicate some approximation to the optimum for the occurrence of eggs. At this station the salinity is between 34-36 and 34-62 parts per thousand, and at others, where the eggs are comparatively plentiful, the salinity corresponds more or less closely to these limits. VERTICAL DISTRIBUTION Eggs occurred in all six nets of the vertical series fished from the surface down to 1000 m., but the bulk of them were found below 250 m. At stations where more than one net yielded eggs, and where large numbers were found, the greater proportion was in the lower nets. The depths of the nets in which eggs were found is given in Table III. At all stations a series of closing nets was fished from the surface downwards, but above the shallowest net mentioned no eggs were obtained. The percentage of the total number of eggs per 50 m. unit in each of the six hauls is as follows : 50- o m. 0-8 % 100- 50 m. 8-3 % 250-100 m. 22-0 % 500-250 m. 67-2 % 750-500 m. i-o % 1000-750 m. 0-7 % These percentages are calculated from the total number of eggs in each net haul, cor- rected to make the numbers comparable. The numbers are very greatly influenced by the great abundance of eggs at St. 540. At this station the sounding was 510 m., so that the deepest net fished reached to 10 m. from the sea-bottom. It suggests a concentration of eggs near the bottom and a possible explanation of their relative paucity in our catches as a whole. If the eggs are laid at or near the bottom, their occurrence in the surface nets might easily be accounted for by surface mixing of the water column, which in the Bransfield Strait region is known to be particularly effective. Rustad (1930) referring to Thysanoessa macrura suggested that spawning in that species might take place below 400 m., and that eggs and sexually mature animals might be concentrated in circumscribed shoals. He thought it unlikely, however, that "a sur- face form like Eiiphaiisia superba would spawn in deep water". Ruud (1932) discounted the explanation of circumscribed shoals and, referring to E. superba, postulated the hypothesis that it spawns close to, or right under, the drifting ice. In a later paper Rustad (1934, p. 36) replies to Ruud's criticism and states "that the physical conditions in the deeper layers, as compared with those right under the ice should be unfavourable to the spawning and development of the youngest larvae, is hard to see. The chief dif- ference is found in the salinity but this seems to be of less importance, our finds demon- strating that the larvae may develop at the higher salinity in the deeper layers ; nor does the smaller content of oxygen seem to be unfavourable." 112 DISCOVERY REPORTS Rustad admits that he has not sufficient material to get a trustworthy idea of the dis- tribution of the eggs and points out the Hmitations due to making only two vertical net hauls — the lower from 400 m. — in the interpretation of his results. If eggs of E. superba occur in abundance in the proximity of drifting ice, it is rather surprising that so very few have been taken over a period of four years in the surface nets at stations made by the vessels of the Discovery Committee at the edge of, or actually in, pack-ice. Many of these stations were made at a time of year when E. superba is known to spawn, and if the number of eggs at St. 540, for instance, is typical of a spawn- ing region the eggs could not fail to have been conspicuous in the plankton of the surface nets. From discussion with Dr T. J. Hart it is learned that in his examination of the samples analysed for his Phytoplankton Report (Hart, 1934) he did not find any eggs recog- nizable as belonging to E. superba. The samples were from 50-cm. vertical nets (N 50 V) fished from 100 m. to surface, and, as the Report eff^ectually demonstrates, used very frequently and over a wide area in the Antarctic. This indication of distribution of eggs in other than the first hundred metres, although admittedly negative, is based on such a large number of observations that it cannot be neglected. Evidence in favour of deep spawning, additional to that supplied by the vertical dis- tribution of the eggs themselves, is the occasional occurrence of fully adult animals in the deep hauls of the 70-cm. vertical nets. It should be emphasized that this net, like others, is selective in its fishing, and it is unusual to find adults of E. superba in it at any time. That they should be found deep down at the same time as eggs are taken, as in- dicated in Table III, leads one to infer that their occurrence at these depths is connected with spawning. Their condition as regards maturity also indicates that departure from the surface is associated with the deposition of eggs. The capture, referred to on p. 17, of ninety-five adults including many gravid females in a 250-100 m. haul lends still further support to the hypothesis of deep spawning in this species, and the fact that they were the only adults to be taken in any of the vertical or oblique nets hauled at this station is in favour of Rustad 's idea that the animals congregate in circumscribed shoals for spawning. It is remarkable that comparatively few eggs have been found in the samples analysed, and although Rustad 's suggestion of circumscribed shoaling "and the consequent con- centration of the eggs in relatively small areas " may be partly the reason I do not think that it is completely satisfactory. E. superba must be reproduced in immense numbers to hold the key position it has in the ecology of Antarctic life. The yield of the 70-cm. vertical nets, with the possible exception of those at St. 540, is surely not indicative of the normal concentration. Is it possible that the development of eggs of E. superba takes place in water which is deeper than the lower limit of the vertical nets and that the eggs obtained are the scattered product of dispersal of a much greater mass situated in still deeper water.? It may be that in this species the eggs, when laid, usually sink below 1000 m., and that the great abundance at St. 540 is due to the net coming within 10 m. of the sea-bottom DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERBA 113 at a place of concentration where eggs are in somewhat shallower water than is usual. Much more positive evidence is required on this problem before any definite solution can be put forward; conclusions drawn purely from negative evidence or from out- standing exceptions to the normal are very apt to be misleading, if not altogether wrong. TIME OF SPAWNING It is possible to get information about the duration of spawning in at least two ways. The first by observation of adults for indication of sexual maturity, and the other by consideration of the records of eggs in the plankton. It is not proposed in the present paper to deal with the first method beyond referring to the spawning of females in aquaria on the ship. The earliest record of this is on December 21 in the season 1930-1, and it is followed by another almost a month later on January 19. The latest record of eggs hatching is on February 10 in the season 1929-30. There is, then, from this source evidence of spawning taking place over about seven weeks at the height of the southern summer. Conclusions based on the presence of eggs in the plankton involve the assumption that the time between the laying of eggs and the development into Naupliar and Calyptopis forms is very brief. The occurrence of all stages of development up to the clearly distinguishable ist Nauplius in eggs in one catch, and the fewness of ist and 2nd Nauplii in any of the catches, suggest that this is probably the case. Table XXXIX shows in half-monthly periods for the years 1928-31 the time at which eggs were found. Table XXXIX. Showing the half-monthly periods in which eggs have been found Season 1927-8 1928-9 1929-30 I930-I Nov. 1-15 — — X — 16-30 — — X — Dec. 1-15 — — — 16-31 — X X X Jan. 1-15 16-31 Feb. 1-14 — — — X X X _ 15-28 X X — X Mar. 1-15 — — — X In the season 1927-8 there were no observations before February, but from the latter half of this month there is one record of eggs. In the season 1928-9 eggs were found be- tween the second half of December and the second half of February, in the 1929-30 season between the first half of November and the first half of February and in the 1930-1 season between the second half of December and the first half of March. The longest period of spawning in one season is three and a half months, and for all the seasons under consideration the range is four and a half months. In the 1 930-1 season the station from which the first eggs were recorded was not more than 120 miles distant 15 114 DISCOVERY REPORTS from the station nearly three months later at which the last eggs were found, so that the extended egg-laying period cannot be attributed to great difference in the locality from which the plankton samples were taken. In E. stiperba, therefore, it may be concluded from the available evidence that spawning is not restricted to one short period but is diffused over most of the months of the southern summer. There are indications that the greatest production of eggs is in November-December, but as regards abundance a distorted impression is obtained because certain plankton stations were taken at times and places close to each other while others were in these respects spaced widely apart. NAUPLII It is not possible to come to any definite conclusion regarding distribution of Nauplii when dealing with such small numbers as were found in our plankton samples. Both ist and 2nd Nauplii were taken only at stations from which eggs were also recorded, so that the general remarks regarding the regional distribution of eggs apply equally to the Nauplii. Both the ist Nauplii were found below loo m., the one in the 250-ioo-m. net and the other in the 750-500-m. net. It may be noted, too, that both Ruud's records are from nets fishing in subsurface water. At three stations the 2nd Nauplii were in the 250-ioo-m. net, at three in the 500- 250-m. net, and the remaining one was in the 750-500-m. net. There is no record from above 100 m. At two of the three stations where the Nauplii occurred in the 250- loo-m. net eggs were found in the shallower nets, and it is possible that the Nauplii de- veloped from such eggs. As mentioned above, the rarity of ist and 2nd Nauplii and the smallness of the num- ber where records exist may indicate that these stages are passed through very rapidly in this species, as in other euphausiids where the development is known (vide Lebour, 1926 c, pp. 520, 521). In the 1929-30 season all the 2nd Nauplii were taken in the second half of November, and in the 1930-1 season in the latter part of December. They were not found outside the range of the spawning period. METANAUPLIUS REGIONAL DISTRIBUTION The distribution of the bulk of the Metanauplii differs in a marked manner from that of the eggs. The latter were concentrated within the Bransfield Strait, but, with the exception of those taken at St. 639, all the Metanauplii were taken north of the ridge formed by the Scotia Arc, and the exception accounts for two only out of a recorded total of over 4000 of this stage. Within the area of their distribution in the Scotia Sea this stage is found in the north-east at St. WS 197, not far distant from South Georgia. At this station 115 Metanauplii were taken between 1000 and 750 m. In the north-west they were found as far north of the South Shetlands as the vicinity of the Antarctic convergence (St. 646) from surface down to 500 m. to the number of fifty-six. The DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERBA iis greatest concentration of larvae was at St. 647, the one before that just mentioned; here 4279 MetanaupUi were taken. The distribution of the stations where MetanaupUi were taken is shown in Fig, 32. VERTICAL DISTRIBUTION The majority of the larvae belonging to this stage were found in the two deepest nets between 500 and 1000 m., although their occurrence is recorded also in the four upper nets at certain stations. At St. 647, where the larvae occurred in greater abundance than at all the other stations added together, they are entirely confined to depths between 500 Fig. 32. Distribution of Metanauplius of Euphansta superba. and 1000 m. In areas of open water away from the influence of the ridge of the Scotia Arc, and, in the case of St. 648, away from the influence of the Antarctic convergence, the larvae are entirely restricted to depths below 500 m. It is at such depths that the bulk of MetanaupUi are found, and it is important to consider the conditions of their environment. They are found in the warm deep water and not in cold Antarctic surface water. The properties of this warm deep-water layer are described by Deacon (1933, pp. 222 et seq.), and it is not necessary to recapitulate them here beyond pointing out that he states that the depth of the maximum temperature in 57° 30' S is 600 m. and of maximum salinity 700 m., and that the usual component of movement is from warm to cold regions and not from cold to warm. Now if the temperature range within which Metanauplii were taken is examined, it will be seen that although the larvae are occasionally found in water of temperature below 0° C. they apparently favour the warmer water. Thus at St. 647 the range is be- 15-2 ii6 DISCOVERY REPORTS tween 1-34 and 2-01° C, at St. WS 197 it is 1-33-1 -27° C, and with like positive tem- peratures for Sts. 618, 620, 636, 637 and 648. With regard to the last station, at which the larvae were found in each of the four nets from surface down to 500 m., it may be that the warm surface temperature, due to the proximity of the station to the convergence, produced the same favourable conditions for spawning and subsequent development as are more generally to be sought for in the warm deep layer below 250 m. The significance of the vertical distribution of the Metanauplii is considered in con- junction with that of later larval forms on p. 163. Later stages of development were present at all the stations where Metanauplii occurred, and it is interesting to compare their vertical distribution with that of the Metanauplii. In this comparison Sts. 639 and 648 have been regarded as exceptional and the distribution of the larvae taken there is not represented in Fig. 33. At the former station two Metanauplii were found, both in the surface net, and at the latter this stage was present from the surface down to 500 m., with greatest concentration between 100- STATfON 637 620 618 647 W5 197 638 636 TIME 1907-2026 2214-2359 2225-0100 0059-0328 0530-0915 0537-0730 0935-1230 Fig. 33. Diagram showing the vertical distribution of the Metanauplius and later stages of development at the stations indicated. Areas shown in outline represent the Metanauplius, areas in black represent later stages. 50 m. In the remaining seven stations, as shown in Fig. 33, the Metanauplii are con- centrated in the lowest nets fished. With the exception of two individuals all are in the 750-500 m. and 1000-750 m. nets. It will be seen that there is considerable variation in the vertical distribution of the older larvae and that this can be connected with the time of day at which the net hauls were made. The stations have been arranged in a sequence from midday to midday, the first starting at 1907 and the last finishing just after midday. The figure shows that in the larvae of E. siiperba subsequent to the Meta- nauplius stage there is a definite diurnal migration, so that during the hours of darkness they are concentrated between the surface and 100 m., whereas at other times they are massed at 500-250 m. The vertical migration of larvae will be referred to more fully later ; it is sufficient here to mark the contrast between the behaviour of the Metanauplius and subsequent larval stages. DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERBA 117 Ruud (1932, p. 33) says: "We only find larvae in any quantities in our samples at the stage when they begin to feed, i.e. the first Calyptopis stage (Taube, 1915). The younger larvae would naturally be found with the eggs." With respect to the last statement the present observations on the vertical distribution of the Metanauplius furnish very favourable evidence for the hypothesis that the eggs are to be found in deep water rather than at the surface. TIME OF OCCURRENCE All the Metanauplii were taken during the months of February, March and April and, while it is possible that about the middle of April may be the latest time for the occur- rence of this form, the range in the other direction must be extended considerably be- yond the date of the earliest record. It will be seen presently that there is an isolated record of a 2nd Calyptopis in the first half of November, which would indicate a period for the production of Metanauplii prolonged to at least five and a half months. But if this 2nd Calyptopis be regarded as an unusual exception — and that this may be so is indicated by there being no record of 2nd Calyptopis again until two and a half months later — then the normal time of occurrence of Metanauplii may be reduced to a period of about three months, extending from mid-January to mid-April in the second half of the southern summer. CALYPTOPIS STAGES, FURCILIA STAGES AND ADOLESCENTS REGIONAL DISTRIBUTION It has been mentioned in the Introduction to this paper that the 70-cm. vertical nets yielded the main supply of earlier developmental forms and the i-m. nets older larvae and adolescents. In discussing the distribution of stages subsequent to the Metanauplius it is convenient to classify the material examined in three divisions as follows: Material caught in the 70-cm. nets. Material obtained during the circumpolar cruise. Material caught in the i-m. nets. Of these three divisions the first is concerned mainly with first season larvae, that is to say larvae taken during the summer season in which they were hatched ; and the third division deals with second season forms, including late Furcilia and adolescents, hatched in the summer previous to that in which they were caught. The circumpolar cruise was made chiefly in the winter months and only material from the i-m. nets was examined. As regards both time of year and state of development the larvae in the second of the above sections are placed intermediately between those of the other two sections. Material caught in the 70-CM. nets In the discussion of the 70-cm. net catches the stations have been divided into the following areas: (i) Falkland Islands to South Georgia. (2) South Georgia surveys. ii8 DISCOVERY REPORTS (3) South Georgia to South Sandwich Islands. (4) Scotia Sea. (5) Drake Strait. (6) Bransfield Strait surveys. (7) Palmer Archipelago and Bellingshausen Sea. (8) East of South Georgia (ice-edge stations). (9) Weddell Sea. For each area the groups of observations have been arranged with reference to the southern summer according to the time of year during which they were made and not in strict chronological order. (i) Falkland Islands to South Georgia Stations Year Time of year {a) WS 251-256 1928 August {b) WS 314-320 1928 December (c) WS 137-143 1928 February (d) WS 518-526 1930 February-March (e) 655-659 1 93 1 March (/) WS 427-433 1929 April-May (a) August 1928 (Sts. WS 251-256). Fig. 34.* Only one larva was recorded at St. WS 255, west of the Shag Rocks: it was a Furcilia 6. (b) December 1928 (Sts. WS 314-320). Fig. 35. No larvae were taken. (c) February 1928 (Sts. WS 137-143). Fig. 36. The first three stations were made in sub- Antarctic water; at the last of these, St. WS 139, four Calyptopis were found in the 500-250 m. net in temperatures of 2-66-3-09° C. At the surface at this station the temperature was 6-15° C. The larvae were carefully examined as it was considered unusual to get E. siiperba to the north of the Antarctic convergence, but they were undoubtedly this species. One Calyptopis i and one Calyptopis 2 were found at St. WS 141, 250-100 m., and one Calyptopis 3 at 500-250 m. (d) and (e) February-March 1930 and March 193 1 (Sts. WS 518-526 and 655-659). Figs. 37 and 38. No E. superba. (/) April-May 1929 (Sts. WS 427-433). Fig. 39. Larvae were taken at the first station only, situated between the north-west end of South Georgia and the Shag Rocks, and then only in small numbers, five in all representing Furcilia 3, 4 and 5. In the Falkland Islands-South Georgia region larvae occur, but in small quantity, from the Antarctic convergence eastwards to South Georgia. A late Furcilia was ob- tained early in the season and Calyptopis and early Furcilia after the new year in the second half of the southern summer. * In this and in the subsequent distribution charts the larger circular marks indicate stations at which krill was found. The smaller marks indicate stations where the nets under consideration were examined with negative results. DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERBA 119 5^ FALKLAND Is. WSS52 ..■.{/ / ' WSE54 WS255 • wsase S. GEORGIA 40' Fig. 34. Distribution of young Euphausia siiperba between Falkland Islands and South Georgia (70-cm. net hauls), August 1928. 5^ WS3I9 ..r FALKLAND Is. WS3I8 .•/.- .'X' WS3I7 /^ r ' WS316 WS3IS WS3S0 **WS3I4 S. GEORGIA 40' Fig. 35. Distribution of young Euphausia superba between Falkland Islands and South Georgia (70-cm. net hauls), December 1928. FALKLAND Is. // wsisa ^ / /,^WSI39 ^ •■ WSI40 WSI4I WSI42 " WSI43 S.GEORGIA 45° Fig. 36. Distribution of young Euphausia superba between Falkland Islands and South Georgia (70-cm. net hauls), February 1928. DISCOVERY REPORTS "= 60* 55° 50' 45* +0° 35- ■' ^ 51 =^ -I* WS5I6 WSSI9 Cj 51° ^^ c<«j*- WSS20 -.O' . ,b WS52I •>^ FALKLAND Is. / WS522 WS5E3 53- O^ WSSa4 ^5525 . WS52S /s/ ■«y ~^S. GEORGIA 55 \i- 5^ 57 57° 60° 55° SO' 45° -t-O" 35° Jl Fig. 37. Distribution of young Euphausia superba between Falkland Islands and South Georgia (70-cm. net hauls), February-March 1930. 60° 55° 50° 45° 40° 35° 51 51" -m^? " • ■^ ■il' FALKLAND Is. ,•■/ 656 53' ■5- '655 6S7 Bsa ■'^ •559 -^•-*, ■■^■; ^ S. GEORGIA / \ ^ s< 57° 57 60° 55° 50' 45° 40' 35° II Fig. 38. Distribution of young Euphausia superba between Falkland Islands and South Georgia (70-cm. net hauls), March 193 1. S%5.^ VVS433 FALKLAND Is. 4& ■■„v ,■0- WS43I ■J:' f WS43Q WS429 WS428 WS427 ■ '"tsJ^S.GEORGlA 5^ 40* Fig. 39. Distribution of young Euphausia superba between Falkland Islands and South Georgia (70-cm. net hauls), April-May 1929. DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERBA (2) South Georgia surveys Stations Year Time of year («) WS 257-296 1928 August-September-October (b) 475-525 1930 November-December (c) WS 321-365 1928-9 December-January (d) 300-358 1930 January-February (e) WS 144-195 1928 February-March (/) WS 567-575 1931 March (g) WS 417-426 1929 April (h) 393 1930 May (a) Survey of August-September-October 1928 (Sts. WS 257-296). Fig. 40. As shown in Fig. 40 krill was present at several stations in this survey. All the larvae belonged to the previous year class and with the exception of one adolescent at St. WS 282 and WS 295 all were in the Furcilia 6 stage. The number of larvae recorded is remarkably small, never more than three at any one station. (b) Survey of November-December 1930 (Sts. 475-525). Fig. 41. With the exception of one Calyptopis 3 at St. 477 all the young krill from this survey belonged to the previous year class and were either Furcilia 6 or small adolescents. As in the previous survey the number of larvae recorded is very small except at the innermost station near Bird Island where 169 were found. The i-m. net hauls, however, give a better indication of the abundance of the later and larger developmental forms. The majority of the records are confined to the two uppermost nets, but it is interesting to note that there are two from 1000-750 m. net hauls. The larvae do not show any indication of restriction to the areas beyond the shallow coastal water, and the largest number taken was close inshore. It is likely that this distribution is intimately bound up with the presence of pack- ice in the vicinity of South Georgia during this survey and just before it was made. The predilection of adolescents for the vicinity of pack-ice will be demonstrated later; it is mentioned here to give a possible explanation of larvae so close inshore. (c) Survey of December-January 1928-9 (Sts. WS 321-365). Fig. 42. No larvae were caught during this survey. The only evidence of the presence of E. superba was one egg taken at St. WS 323, the first deep-water station near Bird Island. {d) Survey of January-February 1930 (Sts. 300-358). Fig. 43. The stations at which larvae were obtained are shown in Fig. 43. To demonstrate that practically all the larvae were taken at stations made in deep water the 250-m. contour has been inserted in the map. The larval stages found were Calyptopis i, 2 and 3 and Furcilia i, 2 and 3, with Calyptopis 3 pre- dominating. At one station (St. 356) gravid females were taken in the 250-100 m. net, but neither eggs nor stages younger than Calyptopis 2 were found in the plankton samples. The larvae were fairly abundant in number with the maximum at St. 303, where 133 were taken. Ninety-nine were taken at St. 320, four other stations had more than twenty larvae, and eighteen stations had less than ten. At one station or another where the young euphausians were taken the net hauls from surface down to 1000 m. yielded larvae, and although the 500-250 m. net appears to be the deep limit for the majority it is not so for all of them. There is no reason to suppose that the larvae from the deeper nets were dead specimens sinking to the sea-bottom, as it is easy to distinguish dead E. superba in the samples. The occurrence of the larvae at stations made in water deeper than 250 m., as opposed to shallower water stations, may be connected with their vertical distribution. It has already been remarked D .XIV 16 122 DISCOVERY REPORTS Fig. 40. Distribution of young Euphausia superba. South Georgia survey (70-cm. net hauls), August-October 1928. 40° 38° 36° 3-*° 191 • 430 ■493 4BE 494 0 53 488 53 •487 492 • 495 495 "7^ 479 480 483 • • 4B& • 46 500 • S03 504 • 501 502 0 477 • 1 499 (..(J 506 BOS • 55 525 524 518 ^1 9 507 511 55 528 S23 519 516 ft, • rr 5*0 ^S 3 508 52 r 515 seo 514 • 40° 38° 36° S-t" Fig. 41. Distribution of young Euphausia superba, South Georgia survey (70-cm. net hauls), November-December 1930. DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERBA 123 38° 36" 34." 53 34 W5323 •WS32e • WS344 - .WS343- I •WS342 I . WS34I • WS340 I . WS339 1 WS33e ^^Cj-fei;; WS320 WS329 54 ■ KjA W5348 ■* ^ f • • S f :^: WS359 -W536I 55 • WS3&3 .W5365 38" 36° 34- Fig. 42. Distribution of young Euphausia superba, South Georgia survey (70-cm. net hauls), December-January 1928-9 40° 38° 36° 3-* • 301 ) •301 • 302 53 63 •320 • 321 • 304 •305 \ •3S7 35a 5* 55 .322 • 323 "3.5 .306 -^«-~-'35i^..- • 313 • 312 .311 54 55 ''' \ 332,^ . 333 • 325 ,.<^ 35,:'-,f/ • 0 /,-> > S- .346 3S4 35S • 356 327 32G/-32! 331 ^'330 320 •341 i__^ •""' , -' •337 \^ •336 • 335 .334 -34 7 /' 3.46-' ' ^345 • 344 • 343 •3 12 4.0° 38° 36" ■■^'* Fig. 43. Distribution of young Euphausia superba. South Georgia survey (70-cm. net hauls), January-February 1930. 16-2 124 DISCOVERY REPORTS (P- 115) that the larvae later in development than the Metanauplius, undergo a vertical migration diurnally through a column of water greater in depth than 250 m. That they should be absent from water shallower than 250 m. in this survey argues that for reasons connected with light intensity it is essential for them to have greater depths to recede to in the daytime. This will be dealt with in the section on vertical migration. (e) Survey of February-March 1928 (Sts. WS 144-195). Fig. 44. Very few larvae were taken during this survey; one Calyptopis 2 at St. WS 173 just beyond the shallow water on the Vakop Line and two Calyptopis 1 and 2 at St. WS 182 in the first deep-water station on the Clarke Rocks Line. It was noted above that one egg was taken at St. WS 147 in the 250-100 m. net. This station was on the edge of the continental shelf with a sounding of 274 m., but the previous station (WS 146), taken in almost the same position three days before, gave a sounding of 1096 m., so that both must have been on the very edge of the continental shelf. 38" 53 54 55 .^ „I wsias •WSI67 ■ W5 1 79 I -wsiao wsiai 53 55 38° 36" 3-t" Fig. 44. Distribution of young Euphausia superba, South Georgia survey (70-cm. net hauls), February-March 1928. (/) Prince Olaf lines, March 1931 (Sts. WS 567-575). Fig. 45. Two lines of stations to the northward from Prince Olaf Harbour were made in March 1931, and at two stations, both in deep water, one Calyptopis i and one Furcilia 6 were recorded. {g) South-west side of South Georgia, April 1929 (Sts. WS 417-426). Fig. 46. No larvae were taken at any of the stations. (/?) East side of South Georgia, May 1930 (St. 393). At this station four flights of nets, each from 300 m. to the surface, were made, but only one larva, a Furcilia 5, was taken. Two points are made apparent by these surveys; first that Calyptopis and early DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERBA 125 Furcilia stages are to be found chiefly in offshore deeper water, and secondly that older larvae, Furcilia 6 and adolescents, are not so restricted but come close to the island. It should be noted, too, that there is an ordered arrangement in the occurrence of the larvae. Up to the end of the year, that is until about the middle of the summer season, only larvae and adolescents of the previous year class are obtained. In the new year Calyptopis and early Furcilia appear, indicating that spawning must commence about the end of the year. From March onwards later Furcilia are obtained but with Calyptopis i still occurring. 38° 36° • WS570 • WSS75 .WSS74 53 £,3 • WSS7S • WS569 .wssee •ve567 54 ' ''^^r-'-^A^^ 38° 36° 54 36° W54e3 •WS422 38° Fig. 45. Distribution of young Euphausia superba. Fig. 46. Distribution of young Euphausia superba. Prince Olaf lines (70-cm. net hauls), March 1931. south-west side of South Georgia (70-cm. net hauls), April 1929. (3) South Georgia to the South Sandwich Islands, February-March 1930 (Sts. 360-369). Fig. 47 The line of stations from South Georgia to the South Sandwich Islands was made immediately after the completion of the South Georgia survey of January-Feb- ruary 1930 described above. It will be recalled that more larvae were taken dur- ing this survey than in any of the others carried out in that neighbourhood. This abundance was increased at the stations of the South Sandwich line. Larvae were taken at each of the six stations where the 70-cm. vertical nets were fished, and at most of these the numbers were greatly in excess of those in the net hauls of the above-mentioned South Georgia survey. Sts. 360, 361 and 362 were made be- tween South Georgia and the northern- most of the South Sandwich Group. At St. 360 one each of Calyptopis 3 and Furcilia i, 2, 3 and 6 were taken; at 3 0° 5b J% 360 • 3G • '3S2 • ^365 / 0 / e / S5 60° (^369 / 66 T 40° 30° 20° Fig. 47. Distribution of young Euphausia superba, South Georgia to South Sandwich Islands (70-cm. net hauls), February-March 1930. 126 DISCOVERY REPORTS St. 361 there were 339 larvae with Furcilia i predominating but with Calyptopis 2 and 3 and Furcilia 3 and 4 also present, and at St 362 there were 343 larvae, with Calyptopis 3 predominating but having in addition Calyptopis 2 and Furcilia i, 2 and 3. All these stations were in water over 3000 m. deep, but all were in the vicinity of the ridge connecting South Georgia and the South Sandwich Islands. St. 365 was situated between Visokoi and Candlemas Island. 601 larvae were taken there with Calyptopis 3 predominating. Calyptopis i and 2 and Furcilia i and 2 were present. At Sts. 368 and 369, both in the neighbourhood of Cook and Thule Islands, fifty and fifty-seven larvae respectively were taken, and at both stations Furcilia i predominated, but with Calyptopis 3 and Furcilia 2 and 3 also present. The most notable feature in this Une of stations is the increase in the number of larvae taken compared with that in the South Georgia area. The stages of development are in accordance with what was found in the South Georgia surveys. The record of Furcilia 6 so early in the year is noteworthy. The larvae do not show an ordered arrangement in the degree of development from north to south, for at the first station where larvae are abundant Furcilia i predominates, at the next two to the south Calyptopis i, and at the two southernmost Furcilia i again. (4) Scotia Sea (Fig. 48) Stations Locality Year Time of year («) 161-167 -170 South Georgia- South Orkneys- Elephant Island 1927 February (b) WS 374-381 South Georgia- South Orkneys- Elephant Island 1929 February (c) 618-629 South Orkneys- South Sandwich- South Georgia 1931 February id) 633-638 South Georgia- South Orkneys 1931 March (e) 372-375 WS 527-531 South Sandwich- Burdwood Bank 1930 March-April if) WS 196-202 South Georgia- South Shetlands 1928 April is) WS 203-209 South Shetlands- Burdwood Bank 1928 April The stations in this section were all taken in the area bounded to the north, east and south by the Scotia Arc and on the west by the Drake Strait. {a) South Georgia to the South Orkneys and Elephant Island, February 1927 (Sts. 161-170). Vertical nets were fished at Sts. 160 and 161 between South Georgia and the South Orkneys, but only at the latter were larvae present. This station was situated about midway between the Shag Rocks and the South Orkneys, and at it eight larvae, all Calyptopis with the 2nd predominating, were taken. A single Calyptopis 2 was found at St. 167, the next vertical station made in the vicinity of the South Orkneys. DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERBA 127 At St. 169, to the east of Clarence Island, the three Calyptopis stages and the ist Furcilia were represented in a total of seventeen larvae, with Calyptopis 3 predominating. (b) South Georgia to the South Orkneys and Elephant Island, February 1929 (Sts. ws 374-381). At St. 376, midway between South Georgia and the South Orkneys, four adult male and two adult and gravid females were taken in the 750-500 m. net. At the next station to the south, St. WS 377, a single egg of E. superba was taken in the 250-100 m. net. Nothing was found at the two following stations, but at St. WS 380, midway between the Orkneys and Elephant Island, four eggs were found, two in the 250-100 m. and two in the 500-250 m. net. There were no larvae at St. WS 381. Fig. 48. Distribution of young Euphausia superba, Scotia Sea (70-cm. net hauls). {c) South Orkneys to the South Sandwich Islands and South Georgia, February 1931 (Sts. 618-629). Vertical nets were fished at Sts. 618, 620, 622, 624 and 629. At St. 618, the first station on this line, situated to the north-east of the South Orkneys, there were 985 larvae distributed from the surface down to 750-500 m. Calyptopis i predominated, but there were present also Metanauplii, deep down, and Calyptopis 2. At this station also an egg was found in the 250-100 m. sample. At St. 620, larvae, totalling 243, were found in all the nets down to 1000 m. with the exception of the 500-250 m. haul. Calyptopis i was again the most plentiful with few Calyptopis 2 and Meta- nauplii, the latter being entirely confined to the lowest nets. At St. 622 the number of larvae was much reduced, only forty-two being taken. With the exception of one Calyptopis 2 all were in the Calyptopis i stage. No Metanauplii were obtained at this station. At St. 624 only one larva, a Furcilia i, was obtained from the quarter sample of the 70-cm. net analysed. Ten Furcilia i, four Furciha 2 and one Furciha 3 were taken in the i-m. net. There were no larvae at St. 629, between the South Sandwich Islands and South Georgia. Of these stations, all up to St. 626 were taken in the proximity of the pack-ice edge. 128 DISCOVERY REPORTS (d) South Georgia to South Orkney, March 193 1 (Sts. 633-638). St. 633, the first vertical station on this Hne, was situated to the eastward of the Burdwood Bank, and with the succeeding one was made in sub-Antarctic water. At St. 635, the first station south of the Antarctic convergence, twenty-eight larvae representing stages from Calyptopis 2 to Furcilia 3 were found, with Calyptopis 2 and 3 in greatest number. At St. 636 there were sixty-four larvae from Metanauplius to Furcilia i, with Calyptopis i predominating. The Metanauplii were all in the 750- 500 m. net, whilst the remaining stages were distributed between 100 and 750 m. An egg, Meta- nauplii and Calyptopis i were found at St. 637, the egg at 250-100 m., the Metanauplii at 1000- 500 m. and the Calyptopis between 100 and 1000 m. In all 129 larvae were taken, with Calyptopis i in greatest abundance. At St. 638 the number of larvae was smaller, fourteen being taken. This station was made midway between the South Orkneys and Elephant Island, and only Metanauplii and Calyptopis i were taken. The Metanauplii were eight in number, and with the exception of one in the 250-100 m. net were in the two lowest nets. The Calyptopis were found between 250 and 750 m. The last station (St. 639) on this line is better considered with the Bransfield Strait Stations, as it is situated midway between the South Orkneys and Joinville Island. (e) South Sandwich Islands to the Burdwood Bank, March-April 1930 (Sts. 372-375 and WS 527-531). This line of stations traversed the Scotia Arc in a diagonal course from east to west, crossing the Antarctic convergence between Sts. WS 529 and WS 530. The Discovery II stations were made between March 18 and 21, and those of the William Scoresby between March 30 and April 5. At the first station (372) Furcilia 5 was in greatest abundance, with Furcilia 2, 3, 4 and 6 present: there were forty-one larvae in all. Furcilia 5 was again predominant at St. 373, there being twenty-seven of this stage out of a total of thirty-seven larvae. Furcilia 2, 3, 4 and 6 were represented. At St. 374 Furcilia 4 was the predominant stage with twenty-three out of a total of sixty-one larvae in which Furcilia 2-6 were present. Furcilia 5 was once again the principal stage represented in the last Discovery II station: Furcilia 2, 3, 4 and 6 were also present. No less than 756 larvae, distributed between the surface and 500 m., were obtained at St. WS 527. The larvae have been described in another section of this paper, in which attention was drawn to the great variety of forms encountered. At this station, for the first time on this line, Calyptopis were present. Furcilia 2 was the principal stage, and in addition Furcilia i, 3, 4, 5 and 6 were represented. No larvae were recorded from St. WS 528. At St. WS 529 loi larvae were taken with Furcilia 3 predominating and Furcilia 2, 4, 5 and 6 present. Sts. WS 530 and WS 531 were in warm water and no E. superba were recorded. (/) South Georgia to the South Shetlands, April 1928 (Sts. WS 196-202). St. WS 196 was made in shallow water to the south-west of South Georgia and no larvae were taken. At St. WS 197 Calyptopis and Furcilia i were taken in the upper nets and nothing but the Metanauplius in the lowest net. Down to 750 m. Calyptopis i was the principal stage and below that only the Metanauplius was obtained. There were 209 larvae altogether. Very few larvae were obtained at St. WS 198, the three Calyptopis stages and Furcilia i being represented with the latter predominating in a total of nine. There were sixty-two larvae at St. WS 199, of which Furcilia i was the most prominent; in addi- tion Calyptopis 3 and Furcilia 2 and 3 were present. At St. WS 200 there were 154 larvae, with Furcilia i again the principal stage and Calyptopis 3 and Furcilia 2 and 3 also present. DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERBA 129 The number was reduced to twenty-nine at St. WS 201, with the same stages represented as before and the same principal stage. At St. WS 202, to the north-east of Elephant Island, twelve larvae were taken in which Calyptopis 2 and Furcilia i and 2 were present with Furcilia 2 the principal stage. (g) South Shetlands to the Burdwood Bank, April 1928 (Sts. WS 203-209). This line was made immediately after that just described, and no larvae were taken. The observations in the Scotia Sea emphasize the general and widespread distribution of E. superba larvae within that part which is south of the Antarctic convergence. The occurrence of the different developmental stages is broadly what would have been an- ticipated from the foregoing observations. In February, in three different seasons, Calyptopis 2 is the predominant stage, showing that spawning must have commenced at about the same time as was indicated in the South Georgia region. But although in February 193 1 Calyptopis 2 predominated there is a reversion of predominance to the previous stage in a line to the westwards made in March 193 1. In March-April 1930 Furcilia 2-5 are most prominent; in April 1928 there is a reversion of predominance to earlier stages. It is thus obvious that in the same season, but in different localities of the same area, there is quite considerable variation in the degree of development at com- parable times. It also demonstrates that different stages may predominate at the same time of year in different years. The series of stations 618-639 requires particular attention. At each of the vertical stations along the ice-edge, larvae were taken in considerable quantity, which diminished in amount from west to east. From St. 625 onwards no larvae were taken. Compare these results with those described by Mackintosh (1934, p. 128, fig. 45) for this hne of stations. He states: In February the ' Discovery II ' returned to the South Sandwich region working stations along the ice-edge between the South Orkney Islands and the Sandwich Group. These stations (618-629) ^re shown in Fig. 45. The ice line had retreated very little since December but the plankton had changed to a much warmer type None of the very "cold" group was present but such warm water species as Pareuchaeta, Pleuromamma and Euphaiisia frigida were included in the catches. As the ship ap- proached the South Sandwich Islands, however, signs of a "colder" plankton appeared. Vanadis occurred at St. 624 and Diphyes antarctica at Sts. 625, 626, 628 and 629. That is to say that the young krill, though admittedly taken at the ice-edge, were ob- tained exclusively in the region of Mackintosh's "warmer " plankton. Mackintosh states that the presence of ice does not necessarily entail the presence of cold-water plankton. This is a point to be emphasized, that the presence of the larval krill at the pack-ice edge does not mean that it is a "cold" species. This is borne out by the line from South Sandwich to the Burdwood Bank in March- April, 1930. The whole of this line was in a region where conditions for the occurrence of " warmer " plankton were prominent, and at each of the stations excepting St. WS 528, young larvae were present in an abundance which was conspicuous at St. WS 527 and WS 529, where the water was even warmer than at the preceding stations. D XIV 17 I30 DISCOVERY REPORTS e5° 60° 55° \ .ws4a7 1 ' 55 .WS406 1 . W5405 / 1 \ \ •WS403______- .WS468 / 1/ A •WS469 / / y .WS470 • WS47I • WS472 V^ — "1 1 .WS473 0 60 y^--^'^ \ .WS40H \ \ •W540I 1 •WS474 \ WS4 0G- L^ ■ \ \ A' cl ^^ ICf 65° 60° 55° 55 60 Fig. 49. Distribution of young Eiiphausia superba, Drake Straits (70-cm. net hauls), February and November 1929. 65° 60° 55° 55 ^i-: \ \ 0 55 / • 383 \ \ / •3B7 \ 1 / • 38S / 1 385-^ 1 \ 60 60 y- \ •383 1 \ .382 1 S'-'S \ \ 0^ - V \ A'M . ^ft f> 70° 65° 60° 55° Fig. 50. Distribution of young Eiiphausia superba, Drake Straits (70-cm. net hauls), March 193 1. Fig. 51. Distribution of young Eiiphausia superba, Drake Strait (70-cm. net hauls), April 1930. DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERBA 131 (5) Drake Strait Stations Year Time of year («) WS 468-475 1929 November (b) WS 400-407 1929 February (c) 644-651 1931 March id) 378-388 1930 April (a) and {b) yielded no eggs or larvae (Fig. 49). {c) March 193 1 (Sts. 644-651). Fig. 50. This line of stations was made rather to the east of the Drake Strait from a point west of Elephant Island towards Staten Island. Of seven stations at which vertical nets were fished four were south of the Antarctic convergence and at all of these eggs and larvae were taken. At St. 644 a few eggs only were taken and at St. 646 eight Calyptopis i and 2, with the latter stage predominating. 5200 larvae representing Metanauplius, Calyptopis i and 2 were taken at St. 647. The Metanauplii, 4279 in number, were confined to the two lowest nets between 500 and 1000 m. The Calyptopis were distributed throughout the six nets with greatest concentration in the surface net. There were eighty Metanauplii and Calyptopis i at St. 648, with the Metanauplii predominating. The conditions at this station have been mentioned above, p. 115. No E. stiperba were taken at the remaining stations north of the Antarctic convergence. (d) April 1930 (Sts. 378-388). Fig. 51. This line of stations was made immediately after the completion of the combined Discovery II- William Scoresby line (4(e)), from the South Sandwich Islands to the Burdwood Bank. It will be remembered that in that line Furcilia 5 predominated in the east and Furcilia 2 and 3 in the west. Vertical nets were worked at eight stations in the present line. Four of the stations were south of the Antarctic convergence and the remainder in warmer water. Larvae were taken at Sts. 378 and 383, the first and third stations of the line. There were few larvae at the former, eleven in all, representing the three Calyptopis stages with the 2nd predomin- ating. Larvae were once again abundant at St. 383, 787 being taken between 50 and 250 m. The three Calyptopis stages and Furcilia were represented, with Calyptopis 3 predominating. The Drake Strait lines are chiefly important for the yield of the Metanauplius stage which one of them presents. The regional occurrence of this larval stage has been men- tioned previously (p. 1 14), and its preference for warmer water pointed out. At both lines of stations from which we have records of larvae the greatest concentration is to the northward rather than the southward, and in the warmer rather than the colder water. These concentrations are of larvae having young stages predominant, in the one line Metanauplius and in the other Calyptopis 3. Both lines were made late in the summer season, yet such early stages as Meta- nauplius and Calyptopis i were still obtained, which shows that in this species, unless of course development is delayed in some of the young larvae, the time range of spawning is not restricted to a short, definite period. 17-2 132 DISCOVERY REPORTS (6) Bransfield Strait surveys Stations Year Time of year («) WS 476-493 1929 November (b) 537-555 1930 December (c) WS 382-399 1929 February id) I7I-I77 1927 February-March {e) 639. 644 193 1 March if) 193-209 1927 March-April (g) 376-377 1930 April (a) Survey of November 1929 (Sts. WS 476-493). Fig. 52. Eggs were taken at several stations in this survey. They were in greatest abundance between King George Island and Trinity Land, where eighty-three are recorded from St. WS 480 at depths be- tween 100 and 500 m.; at St. WS 477 there was one egg between 1000 and 500 m. At each of the stations in the middle line between Livingston and Astrolabe Islands eggs were taken in very small quantity, never more than five at any one station, in nets fishing between 250 and 950 m. None were found in the Smith Island to Brabant Island line, but four Nauplius 2 were recorded, one at St. WS 480 and the remainder in the middle line. (b) Survey of December 1930 (Sts. 537-555)- Fig- 53- Eggs were taken at all the stations on the east line from Elephant Island towards Joinville and at the northernmost and southernmost but one of the King George Island to Trinity Land line. Nothing is recorded from any of the stations in the western line from Snow Island to Trinity Island. (c) Survey of February 1929 (Sts. WS 382-399). Fig. 54. Only four eggs were taken in this survey, three of them in two stations on the King George Island line and one at St. WS 394 in the vicinity of Deception Island. {d) Survey of February-March 1927 (Sts. 171-177). There was only one vertical net station near Cape Melville, King George Island, at which a Furcilia i was recorded. (e) Survey of March 193 1 (Sts. 639 and 644). Fig. 54 (inset). St. 639, situated about midway between the South Orkney Islands and Joinville Island, yielded Metanauplii, the three Calyptopis stages and Furcilia 2. It was at this station that the Metanauplii, two in number, were found in the surface net. Calyptopis i predominated, there being 218 out of a total of 293 larvae. At St. 644 near Elephant Island two days later, ten eggs only were taken. (/) Survey of March-April 1927 (Sts. 193-209). Fig. 55. At St. 193 at the west end of the Bransfield Strait between Smith and Trinity Islands, 120 larvae were taken with Calyptopis 3 predominating and Calyptopis 2 and Furcilia i also present. There were 189 larvae taken at St. 194, with Calyptopis 3 again predominating, and Calyptopis and 2, and Furcilia i and 3 present. At St. 196, off King George Island, only two Furcilia 4 were taken; at St. 197 thirty-two larvae were taken in which Calyptopis 2 and 3 and Furcilia i, 2 and 3 were represented, Calyptopis 3 and Furcilia 3 were most prominent. At St. 198 there were eighty-seven larvae with Calyptopis 2 and 3 prominent and Calyptopis i and Furcilia i, 2, 3 and 4 present. There were fifty-nine larvae at St. 199 including Calyptopis 2 and 3 and Furcilia 2, 3 and 4, with Furcilia 3 the most prominent. At St. 200, the last station on the King George Island line, three Furcilia i and 2 were taken. DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERB A 133 F'g- 52. Distribution of young Euphausia superba, Bransfield Strait survey (70-cm. net hauls), November 1929. Fig- 53- Distribution of young Euphausia superba, Bransfield Strait survey (70-cm. net hauls), December 1930. 134 DISCOVERY REPORTS Fig. 54. Distribution of young Euphausia superba, Bransfield Strait survey (70-cm. net hauls), February 1929, and inset March 193 1. Fig. 55. Distribution of young Euphausia superba, Bransfield Strait survey (70-cm. net hauls), March-April 1927. DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERB A 135 At St. 201, between the east and west lines, there were seven lar\'ae with FurciHa 2 and 3 exceeding in number FurciHa 4. Larvae were taken at each of Sts. 202-206: sixteen with Calyptopis 3 predominating, at St. 202; seventeen, with FurciHa 3 predominating but no Calyptopis present, at St. 203; twenty-nine at St. 204 with FurciHa 3 again predominating and five at St. 205 with two each of Calyptopis 2 and 3, and one FurciHa i ; three larvae were found at St. 206, comprising two Calyptopis 2 and one FurciHa 3. St. 209 in the middle of Deception Harbour yielded one Calyptopis 2. (g) Survey of April 1930 (Sts. 376-377). These yielded no E. superha. The Bransfield Strait stations are noteworthy as the source of most of the records of eggs in the plankton examined. Their abundance at certain stations has been discussed in the section dealing with the distribution of eggs. Here it may be noted that in this area, so highly specialized as far as hydrographic conditions are concerned, there is, following the incidence of eggs early in the season, a widespread abundance of larvae in the March-April survey. In this last survey Calyptopis 3 predominates and FurciHa 3 is conspicuous, but there is no well-defined directional arrangement in the distribution of these stages. (7) Palmer Archipelago and Bellingshausen Sea Stations Year Time of year (a) WS 495-517 1929-30 December-February (6) 558-604 1930-31 December-January {c) 178-192 1927 March [a) December-February 1929-30 (Sts. WS 495-517). Fig. 56. Twenty-one vertical stations were made west of Graham Land in the Bellingshausen Sea: at two only was there evidence of E. superba. St. WS 496 in 63 1 m. oiT Adelaide Island gave one adolescent and two eggs and two eggs were found at St. WS 505 south-east of Peter I Island. {b) December-January 1 930-1 (Sts. 558-604). Fig. 57. Only nine vertical stations were made in the Bellingshausen Sea and only two eggs were found, at Sts. 558 and 585. The former station was off the Biscoe Islands and the latter on the line made off Adelaide Island. {c) Palmer Archipelago in March 1927 (Sts. 178-192). Fig. 58. Five vertical stations were made in the neighbourhood of Anvers and Brabant Islands, but only two Calyptopis were taken — at St. 187 in the Neumayr Channel. The scarcity of plankton in the Bellingshausen Sea has been noted by Mackintosh (1934), and E. superba is no exception to the general rule. Most of the stations, however, were taken at a time of year when it would not be expected that many larvae would be found, and although the adolescents, as will be seen later, were also few in number I do not think that the evidence obtained here is sufficient to form any general conclusion about the occurrence of very young larvae in this region. DISCOVERY REPORTS Fig. 56. Distribution of young Euphausia superba, Palmer Archipelago and Bellingshausen Sea (70-cm. net hauls), December-February 1929-30. 66° e^" 62"" 64 ^ t^. ° ^> >/ ^ "132 f^ ^ ^■y flO/ r' ^las* /I I c\ I -, J^'A V>' -'A 190 ^ fi< 65 r ^ ? ^. 66° G4° 62° Fig. 57. Distribution of young Euphausia Fig. 58. Distribution of young Euphausia superba, superba, Bellingshausen Sea (70-cm. net Palmer Archipelago (70-cm. net hauls), March 1927. hauls), December-January 1930-1. 55° w 30° 20' \ •'**m^^^^'^ — L / 3S° •/ .WSS37 / WS55B° WS540 . 1 •*' WSSS7 / 60 1/ WS554« 60° 65 WS5S3°/ JwSSSE' 65 .^^T^ \ 1 / r- — -, 40° 30° 20° Fig. 59. Distribution of young Euphausia superba, Weddell Sea (70-cm. net hauls), January-February 193 1. DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERBA 137 (8) East of South Georgia (a) Sts. WS 297-310, observations in pack-ice in October 1928. (b) Sts. 452-471, Cape Town-Bouvet-South Georgia in October 1930. (a) October 1928 (Sts. WS 297-310). At eight vertical stations made in the vicinity of pack-ice only six larvae were taken, three being Furcilia 6, two adolescent and one undetermined. They were found at Sts. WS 297, WS 307 and WS 310. The i-m. nets, however, give a better indication of the occurrence of these forms. (b) Cape Town-Bouvet Island-South Georgia, October 1930 (Sts. 452-471). St. 452 was the first south of the Antarctic convergence on the track between Cape Town and Bouvet Island. No larvae were taken there in the vertical nets, and at the following stations only E. superba of the previous year class were encountered. Here again the i-m. nets give a much better idea of abundance. Larvae were obtained at Sts. 453 and 471 only. At the former station one adoles- cent was taken in the surface net, and at the latter there were thirty-three Furcilia 6 and thirty-three adolescents, again in the surface net. This section is important, if only in a negative manner, so far as younger stages are concerned. The stations were all taken very early in the summer season and no eggs or first season larvae were obtained. (9) The Weddell Sea, January-February 1931 (Sts. WS 534-561). Fig. 59 A line of stations was made in January-February 1931 from South Georgia via the Sandwich Islands into the Weddell Sea as far south as 68° 53'. There were twenty-eight stations at eight of which vertical nets were fished. On the outward journey at St. WS 534 and WS 540 respectively, east of South Georgia and the Sandwich Group, Furcilia 6 and adolescents were taken in moderate quantity. On the return journey, at St. WS 557, one damaged juvenile was taken. No larvae or eggs were found in the Weddell Sea proper, and the i-m. nets bear out this scarcity. In these nets the juvenile E. superba taken are entirely restricted to the northern stations, and are all of the pre- vious year class. The numbers of E. superba in the 70-cm. vertical nets are as follows: St. WS 534 ninety-six; St. WS 540 twenty-five; St. WS 557 one. Material obtained during the circumpolar cruise A general account of this cruise of the R.R.S. ' Discovery II ' is given by John (1934). A series of voyages to and from the ice-edge on a course east-about from South Africa to South America was made during the winter months, and the importance of these, so far as the present investigation is concerned, is twofold. The observations were the first made in the winter months, and they were made in regions of the Antarctic from which we had not previously had plankton samples. John picked out the young E. superba and sent them to me, together with valuable notes of the environmental conditions in which the larvae were taken. I have also had access to the scientific reports sent by him to the Discovery Committee. D .\IV 18 138 DISCOVERY REPORTS The observations are derived from the records of the i-m. net hauls and not from the 70-cm. vertical net hauls. In a previous section attention was drawn to the selectivity of these nets, and it was pointed out that the larger nets give a better indication of the presence of larger larvae. At the time of year when the circumpolar cruise was made, larger, rather than smaller, larvae were to be found, so that the records obtained form a useful link both in time and developmental stages between those of the 70-cm. nets, concerned chiefly with larvae of the first year class, and those of the i-m. nets examined with special reference to larvae of the second year class. • The following is a list of the sections into which the stations of the cruise have been subdivided (see Fig. 60): ((?) Cape Town-Enderby Land-Fremantle, April 1932. (b) Fremantle-Adelie Land-Melbourne, May 1932. (c) Melbourne-ice-edge-New Zealand, June 1932. (d) New Zealand-ice-edge-subtropical convergence, September 1932. (e) Bellingshausen Sea, October-November 1932. (/) South Georgia-ice-edge-Cape Town, March 1933. In the section (e) incidental reference is made to catches of young krill from the vicinity of South Georgia and from the Weddell Sea, with which the krill from the Bellingshausen Sea is compared. (a) Cape Town-Enderby Land-Fremantle, April 1932 (Sts. 843-876). The first cruise began at Simonstown on April 8, and at St. S52, four stations north from the turning point at the ice-edge, krill was taken. Larvae were found in increasing amount until the ice-edge was reached. One Furcilia 2 was found at St. 852. At St. 853 six larvae representing Calyptopis 3, Furcilia i and 2 were taken, with the Calyptopis predominating. Nineteen larvae were taken at St. 854, with Caly- ptopis 3 predominating and Calyptopis 2 and Furcilia i present. At St. 855 close to Enderby Land, there were 575 larvae in the surface net with Calyptopis 3 predominating and Calyptopis 2 and Furcilia i, 2 and 3 represented. In the lower net there were four Calyptopis 2 and six Calyptopis 3. No larvae were taken again until St. 861, where deep and shallow nets were towed. In the shallow net were 131 larvae, and in the deep 86. No Calyptopis were taken but Furcilia 1-5 were present with Furcilia 2 predominating. At St. 862 most of the larvae (39) were again concentrated in the upper net; in the lower net there were five. Furcilia 1-6 were present with Furcilia 4 the most abundant. John suggests that the presence of larvae at Sts. 861 and 862 was due to the stations having been made where the surface layer was of cold Antarctic water deflected northwards along the west side of the Kerguelen-Gaussberg ridge. The majority of the larvae, taken in April off Enderby Land, were Calyptopis with a rise to dom- inance of Furcilia 2 and 4 at Sts. 861 and 862. (b) Fremantle-Adelie Land-Melbourne, May 1932 (Sts. 877-896). On the second visit to the pack-ice only small numbers of larvae were obtained. John states that they met young pancake-ice in 63" 41^' S as darkness fell. A station (887) was made and larval krill was taken. John had hoped that on the following day they would find the pancake-ice had formed a narrow fringe along pack-ice near which he anticipated finding large numbers of krill larvae: but during the night pancake-ice began to form around the ship, so she turned and steamed northwards making St. 888, twenty-five miles from the edge of the ice. DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERBA 139 150° "West 180 East 150 Fig. 60. Distribution of young Euphausia siiperba, circumpolar cruise. 18.2 I40 DISCOVERY REPORTS At St. 887 in the pancake-ice seventy-two larvae were taken in four different net iiauls. No stage earlier than Furcilia 2 was recorded, which was the most abundant form ; Furcilia 3, 4 and 5 were also present. Only one of the larvae was taken in deep water, the remainder being caught in horizontal nets towed at the surface or oblique nets fishing about 100 m. At St. 888 there were twenty-six larvae comprising Furcilia 2 and 3, with the former pre- dominating. The small number of larvae obtained on this second visit to the ice-edge led John to believe that large numbers of larvae were to be expected, not in the young newly formed pancake-ice, but along the edges of the older pack-ice. This influenced the observations made on the third visit to the ice- edge between Melbourne and New Zealand in late June. (c) Melbourne-ice-edge-New Zealand, June 1932 (Sts. 897-928). In June, on the third journey southward, the ice-edge was encountered in 61° S, farther north than on either of the two previous occasions. The ship met light pancake-ice on the night of June 22 and steamed through it in search of pack-ice, but none was found, and the pancakes became too tightly packed to make a full station in darkness possible. The ship therefore steamed to the edge of the ice and made a full station, no krill or krill larvae being obtained. In the morning the 'Discovery II' again steamed into the pancake-ice but could see no sign of pack-ice to the south so she turned east and steamed through the pancake-ice, making three oblique towing stations at four-hourly intervals. No krill was taken at any of these stations nor at a station made at night at the edge of the ice. Throughout the night the ship steamed eastward parallel to the edge of the ice and entered it to continue the search the following morning. Oblique nets were towed and small numbers of late Furcilia were taken in the upper nets. The i-m. and 70-cm. nets were fished on the same warp just below the surface and clear of the ice; they were closed before hauling in. The i-m. net was at a depth of 2 m. and caught large numbers of larvae; the 70-cm. net at 5-7 m. caught none. The nets were fished again in the same way with the same results. A full station was subsequently made in the ice but no young E. superba were taken in the vertical net hauls. They were caught again in large numbers in the i-m. net, and on this occasion in the 70-cm. net, towed just below the ice at the end of the station. Krill had been taken on the previous cruises some distance from the ice in the cold westward- flowing Antarctic water, but after leaving the ice on this last occasion, although oblique nets and a horizontal surface i-m. net were towed every four hours until the boundary of the cold water was reached, neither adult nor larval krill was taken. The sample of krill analysed, from the i-m. horizontal net towed at 2 m. at St. 912, consisted of one-eighth of the total number, 161 6. Furcilia 4, 5 and 6 and one adolescent were taken, with Furcilia 6 greatly predominating over the other stages found. John states that the horizontal surface hauls near the ice were enormous compared with the oblique hauls. (d) New Zealand-ice-edge-subtropical convergence south of mid-Pacific, September 1932 (Sts. 942-967). The search for young E. superba along the ice-edge on this cruise resembled that made on the previous visit, but the number taken was much smaller. Two days were occupied in the search, and the night between was spent steaming eastwards along the ice-edge. Twenty-one net hauls were made in the ice and along its edge; eight of them were the routine oblique hauls from 250-100 m. or from 100 m. to the surface and thirteen were horizontal surface hauls made particularly for the purpose of catching young krill. The total number taken was small, approximately 220; only one adult was obtained. At St. 954, 0-5 m., 108 Furcilia 6 and adolescents were taken, with the latter predominating. It was observed that, contrary to the usual length distribution of young E. superba, in this instance they fell into two well-defined length groups, the lower with an average about 13 mm. and the higher about DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERB A 141 17-5-18 mm. These two maxima were not connected with the stage of development, for in the lower group Furcilia 6 and adolescents were present in approximately equal numbers, while in the upper the larvae were entirely adolescent. It cannot be, therefore, that the second length group is the re- sult of the larvae moulting, with sudden increase of length from Furcilia 6 to adolescent. The first length group thus does not moult directly to the second. At St. 959 there was a similar distribution of length groups, but not quite so well-defined. The larger group is still too small to be regarded as the young adults of the ensuing season. Our figures for the average length of adolescents show that normally they reach about 20 mm. by the end of the first half of their second summer season, so that they would be much larger nine months later. It is more likely that at each of Sts. 954 and 959 we are dealing with two different broods belonging to the same year class. It may be that the larger group originates from eggs spawned very early in the pre- vious season or, as suggested by John in his report, concerning a similar length distribution of larvae late in the circumpolar cruise, the size difference may be due to better conditions for growth in cer- tain areas, such as a more abundant diatom flora. At St. 955 young E. superba representing Furcilia 6 and adolescents were taken, with the latter predominating. At St. 956 only two Furcilia 6 were taken and at St. 957 three undetermined stages. One adult and eighty young were taken at St. 959, with adolescents predominating. {e) Bellingshausen Sea, October-November 1932 (Sts. 968-1003). The second portion of the W-shaped track from New Zealand to Magellanes was made in condi- tions which prevented the ship from reaching the ice-edge. Fuel was short and a prolonged gale interfered with the programme arranged. The ship turned northwards again in 63° 57' S, loi " 16' W with clear water to the southward. No krill was taken. From Magellanes another V-shaped cruise was made at the end of October and beginning of November, south to the edge of the pack-ice and then north-east roughly parallel to the Graham Land Coast. Young krill, but no adults, occurred in very small numbers in nine of the ten stations made along and near the pack-ice — the total number from both the oblique i-m. and 70-cm. nets and the hori- zontal surface nets was only 170 individuals, 113 of which were taken at the northernmost station (1003). Furcilia 6 predominated, whereas south-east of New Zealand in September adolescents and Furcilia 6 were in about equal numbers if the second larger length group is ignored; adolescents greatly preponderated if the larger group is included. The Bellingshausen Sea krill were measured by John, and later the small numbers taken at Sts. 1027 and 1 03 1, and 200 young from a large haul of mixed old and young krill from St. 1029, were measured. These stations were to the north-west, south-west and west of South Georgia, far from the ice. A third batch of young krill, taken from six stations near to the Weddell Sea ice (Sts. 1034, 1035, 1039, 1041, 1044, 1047) was measured by John. The length-frequency distribution of these three samples of krill is shown in the graphs in Fig. 61. The young krill from the Bellingshausen Sea {A) were taken between October 29 and November 2 ; they were small. An analysis of the young krill taken south-east of New Zealand seven weeks earlier showed it to be composed of two size groups falling about the lengths of 12 and 20 mm. The majority of the Bellingshausen Sea krill were 12-14 mm. long and would have been a millimetre or two less in early September. The young euphausians from the open sea to the west of South Georgia {B) and those from the Weddell Sea ice-edge (C), all of which fall about the lengths of 24-25 mm., were taken between November 18 and 30, that is two to four weeks later than those from the Bellingshausen Sea. The period between the observations is not suflicient to account for the great difference in size. The similarity of the two populations represented by the curves {B) and (C), together with Deacon's dis- covery of the existence of a constant flow of water from the Weddell Sea towards the Shag Rocks, makes it certain that the krill in the open water to the west of South Georgia was of Weddell Sea origin. It is suggested by John that better conditions for growth in the ice-free water, such as the 142 DISCOVERY REPORTS more abundant diatom flora, may account for their larger size compared with those taken immediately afterwards along the ice-edge. The graph C, representing the Weddell Sea ice-edge yearling krill, is the only one of the three graphs having two peaks showing two widely separated size groups in the population; but, as John remarks, more knowledge is necessary to understand what it and the two smaller and less widely separated size groups in the Bellinghausen Sea yearling krill mean. LENGTH (UMI Fig. 6i. Analyses of length frequencies of samples of young Eiiphausia superba: A, Bellingshausen Sea; B, South Georgia; C, Weddell Sea ice-edge. (/) South Georgia-ice-edge-Cape Town, March 1933 (Sts. 1137-1167). The last of the V-shaped sections of the circumpolar cruise from South Georgia to the ice-edge far to the east and south, and from there to Cape Town was made in March 1933. I have not had the opportunity to examine the krill taken during this voyage but quote from John's report: "Although the places where we made many of our stations and the times at which we made them were places and times at which larval krill might have been expected we did not recognize any in our hauls. . . . Catches of large numbers of adult krill have been few although we continued to tow a surface metre net in addition to the oblique series at every station. We saw no krill patches. Near the ice-edge, in 69° S, 9° E, we followed our usual practice on such visits of towing net after net, until DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERBA 143 we obtained a quantity of krill. They were adolescents of an unexpectedly small size, the majority being a little smaller than those taken along the Weddell Sea ice-edge three months before." The observations made during the circumpolar cruise are of great help towards a better understanding of the development and distribution of E. superba. The dominance of successive stages is made clear; in April Calyptopis 3 and early Furcilia stages are most abundant, in May Calyptopis stages have disappeared and Furcilia 2 predominates, in June it is Furcilia 6 and in September adolescent forms which have the chief place. The concentration of the larvae at the ice-edge with the coming of winter is demon- strated. In April it is found that larvae occur both at the ice-edge and at a distance from it (Sts. 852-855), but in May, June and September they are concentrated exclusively at the ice-edge. It has been suggested above that at Sts. 861 and 862 the presence of larvae away from the ice-edge is connected with the deflection northwards of Antarctic surface water along the west side of the Kerguelen-Gaussberg ridge. In support of the sugges- tion that the larvae have been carried northwards in the surface water is the occurrence of typical ice-frequenting plankton at Sts. 861 and 862, and the absence of such plankton and young krill at five preceding stations, all of which, like Sts. 861 and 862, were made in the hours of darkness. Movement southwards of the krill is ultimately limited by the ice-edge (see p. 159), and accumulation of larvae in that region is brought about by this limitation and augmented by the expansion of the area of ice-covered water with the coming of winter. At stations made in May, June and September no krill was found except at the ice- edge : its absence elsewhere can be explained if advance in time of year is brought into account. The southward movement of larvae developed in more northerly regions will have been terminated by the ice-edge being reached, and accumulation there will have been speeded up by the northward spreading of the ice field with the advent of winter conditions. In the Falkland sector it was found that in the summer larval krill was diffuse in its distribution, and there is no apparent reason against assuming a similar distribution in other parts of the Antarctic. On the other hand, the observations made during the circumpolar cruise give clear evidence of concentration in the vicinity of the ice-edge in winter. It is likely that the changes of distribution are brought about by the factors mentioned above. Material caught in the i-m. nets For information about the occurrence of larvae in their second season the data sup- plied by the oblique hauls of the i-m. nets will be considered. The material is de- rived from stations made by R.R.S. 'William Scoresby' during the seasons 1928-9, 1929-30 and 1930-1 ; and by R.R.S. ' Discovery II ' during the season 1930-1. The time of year covered is from August to February, but the number of observations for the latter month and January is very small. It will be found convenient to take the stations in their chronological order, starting with the William Scoresby stations in August 1928. 144 DISCOVERY REPORTS Falkland Islands to South Georgia, August 1928 (Sts. WS 251-256). Fig. 62. The Antarctic convergence was crossed between Sts. WS 253 and WS 254, and at the latter the first station to the southward of it a small quantity of young krill was taken. At the succeeding stations towards South Georgia krill was also present in small quantities. The stations which were made in late winter were nevertheless in ice-free water, no pack was encountered. South Georgia survey, August-November 1928 (Sts. WS 257-285, WS 289-296, WS311-313). Fig. 63. The observations off South Georgia in the early part of the 1928-9 season were divided into three by two visits to the pack-ice to the east of South Georgia. In the first set of observations, Sts. WS 257-285, four lines from Bird Island, Prince Olaf Harbour, Cape Larsen and Cape Vakop were made, between August 28 and September 18. At every station, with the exception of two on the Bird Island line, young krill was taken in com- paratively small numbers for the most part, but with indication of a rather greater concentration eastward of South Georgia at the outer stations of the Larsen and Vakop lines. On the last station on the Vakop line (St. WS 285), the largest catch of 1600 individuals was made. From St. WS 285 the 'William Scoresby' steamed south to the edge of the not far distant pack-ice where, as will be seen in the next section, abundant krill was found. The line from Cooper Island was completed at the beginning of October, and at each of the stations from WS 291 to 296 krill was taken. At the inner stations the quantity was small, whilst the two outer stations yielded moderately abundant amounts. The three stations WS 31 1-3 13 to the eastward of South Georgia were made between October 10 and November 5, only one young euphausian being taken. The pack-ice stations, September-October 1928 (Sts. WS 286-288, WS 297-310). Fig. 64. On September 18, on the way from the outermost station on the Vakop line to the edge of the pack, St. WS 286 was worked, 900 young E. superba being taken. St. WS 287 was made inside light pack- ice a few miles from the edge; only a moderate number of krill was taken. St. WS 288, worked just outside the edge of the pack-ice, yielded over 20,000 young E. superba from one net haul. It is inter- esting to note that this last station was made during the day at 1410 hours and that the net was fished from 102 to o m., showing that in certain circumstances, if not invariably, the young E. superba do not obey the same rhythmic diurnal migration noted in earlier larval forms, but may remain at the surface during the hours of daylight. On the second visit to the ice east of South Georgia, nearly all the stations (WS 297-310) were worked in the neighbourhood of the ice, and at every station, whether actually in close proximity to the pack or so far from it that no note of its presence has been recorded in the station list, krill was found. It should be noted that at such stations as WS 304 and WS 305, where the pack-ice was at some distance, there were as many young euphausians taken as at stations worked at the edge of or actually in pack-ice. When it is realized that the pack-ice is capable of considerable and rapid local movement, due to wind, one would not anticipate that the greatest concentration of the animals would necessarily be in the immediate proximity of the pack-ice, but rather that they should be distributed within the area of the ice-edge movement. Falkland Islands to South Shetland, November 1929 (Sts. WS 468-474). Fig. 65. The next line of stations is that made in November of the following year between Falkland Islands and South Shetland. At St. WS 468 made close to the Antarctic convergence one Furcilia 6 was taken, and at St. WS 474, west of Elephant Island, two Furcilia 6 and one adolescent. No krill was taken at the intervening stations. DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERB A 145 .WS2SI FALKLAND Is. WS252 i' ■■ • WS254 ♦ W5E55 #WS255 S.GEORGIA 5^ Fig. 62. Distribution of young Euphausia superba, Falkland Islands-South Georgia (i-m. net hauls), August 1928. 30 ♦O' 30 39" 30 36* W 37* JO 36* AJ 3S' JO 34' 30* — x A/S2S5 » WS26G 3' 30 i ► WSS73 • W5274 • WSE75 •W5E • \^ 76 wsssa WSEG9 WS270 WS27I WS27a • • •W£ • WS5E • WSS60 »WS253 wsssa #wse64 wsesa ass 1 J 0 ■^■«% •ws \ S57 .^ # i' i- v^ -^ \ • ^V 1? ^'\ a '' •WSS9I S3II ■^ • WSB93 #WSe94 • WSP95 JO »■ 4 3* .W .1 J* 3 i* JD 3 7° JO 3 JO 3 •W529 5' to 3 6 V JO' Fig. 63. Distribution of young Euphausia superba, South Georgia survey (i-m. net hauls), August-November 1928. 19 146 DISCOVERY REPORTS Bransfield Strait survey, November 1929 (Sts. WS 475-494). Fig. 66. In the November 1929 survey of the Bransfield Strait young E. superba were only abundant at one station (WS 487) at the southern end of the middle line. 1424 were taken, whereas at the remaining stations comparatively small numbers were recorded. The krill was distributed throughout the area of the strait, with an indication of greatest concentration on the Graham Land side. M -^.&. 0WS3OB 0WS3O7 I »WS306 • WS305 - W5266 WS28B».^^ • ,WS304 >RF;« ^9 W5303 ^ , (5- •WS300 A? •WS299 WSS37 • WS29a Fig. 64. Distribution of young Euphausia superba, pack-ice stations (i-m. net hauls), September-October 1928. 55 Fig. 65. Distribution of young Euphausia superba, Falkland Islands to South Shetlands (i-m. net hauls), November 1929. Bellingshausen Sea, December 1929-February 1930 (Sts. WS 496-517). Fig. 67. The area covered by the stations in this section extends from Anvers Island westwards of Peter I Island to 100° W. The small quantity of krill taken is noteworthy. With the exception of a single individual at St. WS 508, north-west of Charcot Land, no krill was found to the west of Adelaide Island. The stations at which it was present, with the exception just mentioned, were all situated off DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERBA 147 ^ ■•y 1 .WS4a3 WS493 •W5490 Fig. 66. Distribution of young Euphausia superba, Bransfield Strait survey (i-m. net hauls), November 1929. W5S06 WS507 #W550a Fig. 67. Distribution of young Euphausia superba, Bellingshausen Sea (i-m. net hauls), December-February 1929-30. 19-2 148 DISCOVERY REPORTS the coast between Adelaide Island and Anvers Island, with greatest concentration off the former island. No krill was taken on the line of stations made to the north-west from Adelaide Island (Sts.WS 508-517). Weddell Sea, January-February 193 1 (Sts. WS 534-561). Fig. 68. Towards the end of January 193 1 the 'William Scoresby' sailed south-eastward from South Georgia, encountering pack-ice north of the South Sandwich Group. The ice was skirted to the eastward and was found to fall away to the south and east. The ship went southwards for about 600 miles in open water to 68° 53' S, 13" 03' W, where the pack was again encountered. 20" •W5537 S^W5536 " , |#WS53a WS559' .wsssa J^S539« WS540 WS557*' |*WSS4I O i WS556 • I ©WSS^a W5555 Fig. 68. Distribution of young Etiphausia superba, Weddell Sea (i-m. net hauls), January-February 1931. The stations where krill was taken are all concentrated on the northern part of the line between South Georgia and the point east of the Sandwich Group, where the ship turned southward into the ice-free water of the eastern Weddell Sea. At St. WS 534 ninety-six young E. superba were taken in the vertical nets. At Sts. WS 538 and WS 542 they were abundant in the i-m. nets, over 5000 individuals being taken at each of these stations, and the vertical nets at St. WS 540 contained a moderate number. E. superba was absent from all the stations southward from WS 539, including the ice-edge station WS 552, at which six sets of ob- servations were made over a period of twenty-four hours. Flights of five closing metre nets were fished, the lowest at about 500 m. ; in none of these was young E. superba taken. DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERBA 149 Returning northwards again it was not until the ship reached the region in which krill had pre- viously been found that one individual, at St. WS 557, was again recorded in the vertical net series. A small quantity of krill was taken at St. WS 560 in the i-m. net, to the westward of the north end of the South Sandwich Group. Cape Town-Bouvet-South Georgia, October-November, 1930 (Sts. 446-472). Fig. 69. In October 1930 the 'Discovery 11' sailed southwards on a line of stations from Cape Town to Bouvet and South Georgia. The Antarctic convergence was crossed between Sts. 451 and 452. The first catch of young krill was made at St. 453 where a large quantity was recorded. At Sts. 453-455, all in the neighbourhood of Bouvet, a considerable quantity of krill was taken, at the last station over 6000. From Bouvet Island to South Georgia the track was along the edge or through the outskirts of the pack-ice. Young E. superba were taken at each of the stations where the metre net was fished. At St. 461 a twenty-four hours station was made, the results of which are dealt with in a separate section. At St. 468 no lOO-cm. nets were fished. In his report on this series of stations Dr Kemp, referring to E. superba, states: "We found krill first of all to the N.E. of Bouvet where we took a quantity of adults and a great number of late larval and early post-larval forms, stages that we had scarcely seen hitherto. Along the pack adults were again taken at many points and at every station 30° 10° Fig. 69. Distribution of young Euphausia superba, Cape Town, Bouvet, South Georgia (i-m. net hauls), October-November 1930. young have been obtained, usually in great abundance. When making our way through the pack, either in the main body of it or when crossing the narrow tongues, krill was almost always to be seen and often it occurred in great quantity. The young stranded themselves in numbers on floes that we momentarily submerged in our passage, while the adults, with greater activity, often jumped clear of the water and landed kicking on ice some 10 or 12 inches above the surface. The quantity of krill to be found in and near the pack at this season is amazing and the reason why whales haunt the ice- edge is evident." South Georgia survey, November 1930 (Sts. 474-525). Fig. 70. After the completion of the Bouvet-South Georgia stations a plankton survey round South Georgia was commenced. At the beginning of the survey, as stated by Mackintosh (1934, p. 127), the pack- ice was close up to the island, but at the end of November when the survey was completed the ice had receded some way to the south-east. There can be no doubt that the abundance in which young krill was found in the vicinity of South Georgia during this survey was due to the proximity of the pack to the island, and it is likely that earlier in this season the ice-edge extended north and west to the positions shown on the map where krill was especially abundant. It is likely too that the surface currents {vide Deacon, 1933, p. 183, fig. 5) would tend to transport the krill northwards and east- wards in the tongue of water extending along the north-east coast of South Georgia. South Shetland survey, December 1930 (Sts. 537-555)- Fig. 71. In this survey the distribution of the young krill was much the same as that of the eggs described on p. 109. The greatest concentration was at St. 537, the first on the Elephant-Joinville line: this ISO DISCOVERY REPORTS 42" 40° 3«° 36° 3+° S43I • 490 1 • 493 • 489 • 494 53 «4se 53 • 487 492 • 495 • 496 54 475 • • ^78 -J79 450 463 • • 486 • 48 • • 500 •499 464 _49a ■ SOI •502 •503 i^na 1 54 ,-f{ ''"^■8^^437 1 -SOS •506 507 V ^\ \ 5e5« '"•\ -^ 55 .524 ^16 p-WS'^^IE ^" .510 55 522. • 523 5I9^ • 516 »S09 •508 •521 • SIS SEO* •sw 42° 40° 38° 36° 34° Fig. 70. Distribution of young Euphausia superba, South Georgia survey (i-m. net hauls), November 1930. Fig. 71. Distribution of young Euphausia superba. South Shetland survey (i-m. net), December 1930. DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERBA 151 station is situated between Elepiiant and Clarence Islands and over 7000 young E. superba were taken. Much smaller numbers were taken at Sts. 538-541, but some were present at each of the stations.^ On the Cape Melville-Trinity Land line krill was in very small quantity at Sts. 543 and 547, two individuals being obtained at each, and was absent from the remaining stations. On the Snow Island-Trinity Island line there were ten at St. 549 and one each at Sts. 552 and 553, the southern- most stations in that line. VERTICAL DISTRIBUTION In this section the vertical distribution and migrations of E. superba from Calyptopis i to adolescent will be described. It was shown previously (p. 116) that, whereas the Metanauplius tends to remain in the same position whatever the time of day, there was evidence in the later larval stages of a pronounced diurnal migration. Before going on to deal with this in detail, it may be well to mention the difference in our results from those which Ruud obtained from his material (Ruud, 1932). He states in the summary of his paper (p. 93) that E. superba is a surface species, and that both the adults and the young in his catches were taken, almost without exception, above 200 m. "In other words", he remarks, "it belongs to the Antarctic surface layer." That this statement does not apply to all stages of the life history is made evident by the examination of the present material. The distribution of the adults is not within the scope of the present paper except to the extent that they are associated with the distribution of eggs, but there is some reason for believing that the adults seek the deeper water for the deposition of eggs. So far as larval stages from Calyptopis to adolescent are concerned we can demon- strate that they are not by any means restricted to the first 200 m. The routine 70-cm. vertical net hauls made at plankton stations covered a range of 1000 m. from surface downwards or from surface to within a few metres of the sea- bottom if the depth were less than 1000 m. Six hauls were made with closing nets, as follows: 50-0, 100-50, 250-100, 500-250, 750-500 and 1000-750 m. The nets were hauled at a constant speed of i m. per second. The depth through which the two uppermost nets were hauled is one-third that of the next lower net and one-fifth that of the three remaining nets. In order, therefore, to make the actual number of individuals in any one net haul comparable with the re- mainder, it is necessary to employ the lowest common multiple, which is 15, to obtain corrected totals. In the diagrams used to demonstrate vertical migration (Figs. 72 and 73) the width in each column represents the numbers in a 50-m. vertical haul between the indicated levels expressed as a percentage of the total. The day is divided up into six periods of four hours each, and each station has been placed in the four-hour period during which the greater part or the whole of the vertical net series was fished. The larvae have been arranged in six groups of increasing development, the first three of which comprise the three Calyptopis stages; the fourth includes Furcilia i, i.e. larvae with non-setose pleopods; the fifth includes Furcilia 2-4 and the intermediate forms with six and four terminal spines ; the sixth group comprises Furcilia 5 and 6, including what were formerly recognized as Cyrtopia stages. 1 For some reason not recorded these larvae were not measured. The numbers were: St. 538 twenty- three, St. 539 twenty-seven, St. 540 three and St. 541 nine. 152 DISCOVERY REPORTS Between six and seven thousand larvae have been examined in this analysis of vertical distribution. As might be anticipated, Calyptopis i is in greatest abundance, and the group made up by Furcilia 5 and 6 has the smallest number of individuals in it. It is TIML 1400-iaOO FIRST CALYPTOPIS T cr 750-500 SECOND CALYPTOPIS IP THIRD CALYPTOPIS O 500-250 t 0200-0600 iTt T nT+ Fig. 72. Diagram showing diurnal vertical migration in Calyptopis 1-3 of Euphausia superba. difficult to say how greatly the difference in number is due to the superior efficiency of the 70-cm. net in catching Calyptopis as compared with the larger larvae, and to what DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERB A 153 extent the numbers taken may be an indication of the comparative abundance of the two forms. The six diagrams show that, with the exception of FurciHa stages 5 and 6, larval forms TIME 1400-ieOO 0200-0600 FURCILIA I. If FURCILIA ?- I- i FURCILIAS-6 ^ SOO-250 f Fig. 73. Diagram showing diurnal vertical migration in Furcilia stages of Euphausia stipcrba. The areas enclosed by broken lines indicate results obtained from inadequate data. were present in greater or less abundance, from the surface down to 1000-750 m., the lowest depth to which nets were fished. In the Furcilia 5 and 6 the 750-500 m. net is the deepest in which larvae were discovered. But while it is true that the 1000-750 m. 154 DISCOVERY REPORTS net, and to a greater extent the 750-500 m. net, yielded larvae, yet for all stages and whatever the time of day by far the greater majority was to be found in the four upper- most nets between 500 m. and the surface. The first three diagrams indicate very clearly the pronounced diurnal migration that is to be found in the Calyptopis stages. It is to be noted that in the period which in- cludes midday there are still larvae to be found in the uppermost net; in Calyptopis 2, for instance, 20 per cent of the total were found in this net. Even with such proportions, however, the three diagrams show that in the daylight hours the greater bulk of the larvae is away from the surface and below 100 m., while at night they are massed in the surface water above 100 m., with a marked concentration between 50 m. and the sur- face noticeable in Calyptopis 2 and 3. In the daytime the mass of the larvae are between 250 and 100 m., but the appreciable percentage in the 500-250 m. net indicates that the larvae are concentrated in the lower rather than the upper portion of the 150 m. traversed by the 250-100 m. net. If this is so it means that these larvae must migrate upwards through a distance of roughly 200-250 m. to get to the surface each night. Mention has already been made (p. 121 supra) of the distribution of the larvae — Calyptopis 3 predominating — in the South Georgia plankton survey in January- February 1930. It will be recalled that all the larvae were taken in the stations made beyond the 250 m. contour and that the 500-250 m. net was the deep limit for the majority. It seems likely that light intensity, rather than temperature or salinity, is the con- trolling factor in the vertical distribution of the larvae, for in their daily migrations the variation experienced by the larvae in the two latter factors is that to be found in the Antarctic surface water and the warm deep water. The absence of larvae in the shallow water in the neighbourhood of South Georgia seems to indicate that there is a necessity for the larvae to seek out greater depths, so that in the daytime they may recede to them in order to obtain that degree of light intensity which is most favourable for their existence. In the two diagrams representing Furcilia i and Furcilia 2-4, the paucity of data for the period 0200-0600 upsets the evenness of the curve showing diurnal migra- tion. Of Furcilia i for this period two larvae only were taken, and both were in the surface net: there was one larva of Furcilia 2-4 in the 250-100 m. net. It is probable that both groups of larvae, if they were adequately represented during this four-hour period, would have their greatest concentration between 250 and 100 m. On the whole the two diagrams show, although perhaps less clearly, the same diurnal migration as is witnessed in the Calyptopis stages. There is a noticeable difference in the last of the six diagrams, that representing Furcilia 5 and 6, compared with the five preceding it. The larvae at all times of the day, with the exception of the 1400-1800 period, are concentrated in the two uppermost net hauls. The diurnal migration of earlier stages is not evident unless the distribution of the 1 400- 1 800 period is indicative, but whether or not this is so, the diagram shows that in twenty out of the twenty-four hours the larvae are at the surface and that below 100 m. DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERBA 155 the frequency of occurrence is comparable to the frequency below 500 m. in earlier larval stages. In the section of this paper dealing with the circumpolar cruise it was shown that larvae of comparable stages of development to those last mentioned were almost in- variably concentrated, whatever the time of day, in the upper of two oblique nets fished. This localized distribution was most marked in the results of the June visit to the ice- edge. John states that of two nets fished, one, i m. in diameter, towed at a depth of 2 m., caught large numbers of larvae, whereas the other, 70 cm. in diameter, towed at 57 m., caught none. The process was repeated with like results. This was in close proximity to pack-ice and in the daytime. Allowance has to be made for the superior catching power of the i-m. net, but it is quite clear that the shoals of krill were limited vertically to a range of a few feet. It has also been mentioned already (p. 149 supra) that, on the course taken by the 'Discovery 11' from Bouvet to South Georgia in October-November 1930, when the ship was making its way along the edge of the pack, krill was almost always to be seen at the surface in great quantity, with larval and adolescent forms predominating. VERTICAL DISTRIBUTION AT ST. 461 An extract from the scientific log referring to this station says : Seven hauls were made at intervals of four hours with a series of six i-m. nets used obliquely. The oblique nets were put on the wire at intervals of 143 m. with 140 m. between the upper net and the surface. The total wire out was 855 m. The nets were hauled at 1.V-2 knots and when the upper one had reached the surface the messenger was dispatched to close those below. Depths were determined by a depth gauge at the end of the wire and Kelvin tube near the second net from the top, the depths of the individual nets being calculated The first six hauls covering a period of twenty-four hours were taken without any hitch, but in the seventh haul one of the messengers failed to release on im- pact of that above and some nets were in consequence hauled open for longer than was intended. The station was made in 56° 44' S, 2° 23!' W to the south-west of Bouvet Island and in the vicinity of pack-ice. As might have been anticipated from the time of year, October 20-21, young E. superba were represented only by Furcilia 6 and small ado- lescents. It is noteworthy that while over eighteen and a half thousand young krill were taken in seven hauls, only 147 adults were found; of these 122 were in the G flight in the net fishing between 750 and 560 m. and closed at 315 m. The remaining adults were scattered throughout the nets of the other flights in number never more than four in any one net. Table XL shows the quantities of young krill taken in the flights A-G. The depth of the nets and time of day are stated. The numbers show that a distinction can at once be made between the uppermost net catch of each flight and all the others. In five of the seven hauls relatively enormous quantities of krill were taken in the uppermost net, and in all seven more than 59 per cent of the total taken in each flight was in this net. In dealing with the numbers in the percentage column of the table, therefore, the figure given for the uppermost net is the percentage of the total caught in the flight of nets. For the remaining five nets the figures are the percentages of total number in those nets 156 DISCOVERY REPORTS but excluding the first one. The vertical distribution is shown graphically in Fig. 74, the G haul being omitted. Although the young euphausians preponderated at the sur- face they were taken in appreciable numbers in the lowest nets, 660 in the sixth net of the C series being the greatest. When the station was started the ship was in shoals of krill which increased in density to a maximum at the time when the C flight was taken. The shoals had come to an end by the time the E and F hauls were made and another one was encountered at G. Considering the lower nets, 1 1 -VI, it is seen that with the approach of night there is a distinct trend upwards of the main body of the larvae and a descent again in the morning. Thus in the A flight, between 2.30 and 3.0 p.m., the larvae are more or less evenly distributed below 100 m.: in the B flight between 6.30 and 7 p.m. they are con- centrated between 500 and 400 m.: in the C flight between 10.30 and 11 p.m. they are concentrated between 400 and 300 m.: in the D flight between 2.30 and 3.0 a.m. the subsurface krill has merged with the main body of the surface shoal, and in E, with the TIME 1434-1457 2236-2259 Fig. 74. Vertical distribution of young Euphausia superha at St. 461. approach of daylight, there is a withdrawal of some of the krill from the surface to an ill-defined point of concentration at about 300 m. To what are we to attribute this difference of behaviour in the larvae dividing them into the two groups, (a) those that remain at the surface and (b) those that undergo a vertical diurnal migration? There are several possible factors which might be effective, of which the following may be considered: (i) Differences due to ontogenetic variation in the two groups. (ii) Hydrographic differences. (iii) Seasonal changes. (i) As Russell (1927, p. 239) points out, a species may be found to occur at a certain depth when aduh, but while young it may have a different vertical distribution. Ex- amination of the larvae from St. 461 shows that no distinction can be made between the degree of development of the surface larvae and those taken deeper down ; they are a homogeneous population with Furcilia 6 and adolescents occurring in all the catches in proportions which preclude any conclusion that either stage prefers a particular water depth (see Appendix II). DISTRIBUTION OF YOUNG STAGES OF EUPHAUSIA SUPERB A 157 (ii) There is a well-marked discontinuity below 100 m. where the cold surface water and the intermediate warm water meet. It might be suggested that this is the barrier which concentrates so large a proportion of the larvae at the surface, but examination of Table XL shows that it is apparently not a sufficiently strong obstacle to prevent the migration of the subsurface larvae which move towards, and actually merge with, the surface group in the D series. In this series the discontinuity is at 230 m. Table XL. Catches of young Euphausia superba at St. 461 Haul ... A B C D Time ... 1434-1457 1830-1851 2236-2259 0237-0300 D ^ r l^ioi Ml 1 >-:mrimt^iritnTi-\ 1 1 m| | 1 fri 1 6 N JU3DS9]OpY 1 1 1 1 1 1 1 1 "" ^ 1 1 1 " 1 1 1 do " M M 9 ^FlP-ind Ml" "->-"^N « 1 1 1 1 1 1 1 M t^ 0 « K b S Bi[pjnj ■" M M 1 1 1 1 1 1 1 1 1 1 1 1 1 »n ro\b 00 00 M in . M ■> I«loi 1 1 1 1 1 1 1 1 1 1 1 1 ^ 1 i 0 in \0 juaosajopY 1 1 1 1 1 1 1 1 1 1 1 1 - 1 1 1 1 1 10 M M 9 Eijrajnj[ 1 1 1 1 1 1 1 1 1 1 1 1 1 in CO w 00 " 00 in . I^JOX 1 1 1 1 1 1 1 " 1 1 1 1 1 1 1 1 1 1 " 1 1 m b H - juaosajopv 1 1 1 1 M 1 1 M 1 1 1 1 1 1 1 1 1 1 1 9 Eiipjnj 1 M 1 1 1 1 " 1 1 1 1 1 1 1 1 M in « 0 N 2 2 00 N vD .„• lO;- rl ;, CO N FJOJ, I 1 1 N N m\0 -^ ■'t 1 ►-< 1 M N HI 1 Ml ?>1 1 0 in b 0 CO juaosajopY 1 1 1 1 M 1 1 1 M 1 " " " 1 " 1 in ^£> 0 N 9 Biipjnj 1 1 1 M N in\o * ■* 1 1 1 " 1 1 1 1 1 0^ 00 N M > N N FIOJL M 1 " 1 1 " 1 " 1 1 1 1 1 1 M 1 c 1 1 00 in CO juaasajopY 1 1 1 1 1 1 1 1 1 1 1 1 1 M 1 1 1 1 1 1 9 Biiiajn^ 1 1 1 " M " 1 - 1 1 1 1 1 1 1 1 1 00 « :3 en'? 1«J0X 1 1 1 1 1 1 1 1 " " ^ " " " 1 1 1 1 : 1 1 M M N juaosajopY 1 1 1 1 1 1 1 1 1 1 """" 1 1 1 1 fO vO *n fo ^ N m M 9 Bijraanj 1 1 11 1 M 1 " " " " 1 1 1 1 1 1 in tn Y' f in M c .2 a, 0 loO loO mO mOi'^OioO mO Lr*0 "^ K r^db do b^ <> b b *-> ^ n m coco^--tioio 1 1 1 1 1 iTTTTTTiTTi i i \b i>- i>6o 60 ONO^b b ^ i^ n n coco^Tj-irj •3 s s h CM < i c V 0 1) lyo 00 o . 00 •+ N ^ 00 00 vD . N :- ^ 00 00 N N \0 . N :3 00 00 O N vO . N :a c .2 „ DISCOVERY REPORTS PioX juaosajopv 9 EiiiDjrijj l^ioj. juaDssppv 9 BT]pjnj[ IBIOX 9 Biipin^j £ Bijpjn^ij j Ti-ir)u^o»nLor^f^| 1 wri N M M I 1 " M w ^m»r)\0 i/%mir)'-< f^O O I^N iriNCOO N {^ in -^ fn ^ r^vO O t^" c^X^'+r^" 1«J0X rol Nc^-i-f^ololrl^ 9 Bijpjrijj N M ■* t^ N N N S BIlI3jnj[ Fioj, JU30S3|0pV 9 Eijpjnj S Bt{pjnj[ pjox juaosaiopv 9 Eijpjnj io mino "^ o c^'^r^"^'^*^^ m "^ ro >-< ro rl ^ ir, in O -^-OO O -t-ro riroroMcnrlr^Nt-' M ro m rl r^ N M be C I^OO CO " N M inOO N "Kb I- M 6 CO M 00 00 « 9 " 00 CO in t^ o* t^ in 6 m t^ M N N M r^ CO in N ■-< r^ O i^ '-' 00 6 O in o y^ o lAi o "^ o »n o y"i o\ b 6 ^ I-* N N I in O in O m O m O in O in o CO rn t}- ^ in in \0 I I I O »n I I I I I I O m O in O "^ CO CO Tt- -^ Wi "^ h p- IBWX 1 — 1 1 " " r' 1 " " 1 1 1 i 1 1 2 1 1 b M }U33S3JOpV 1 1 1 1 M 1 1 1 M - " 1 1 1 1 1 1 o N O O N n 9 Eijpjnjf 1 — I 1 ^ " 1 N 1 1 1 1 1 1 1 1 00 0 o! 00 0> N CO l^ioj. 1 1 1 . 1 1 N» 1 « N« 1 1 1 1 1 1 1 ^ 1 1 0 CO M N juaosappv 1 1 1 1 1 1 " " 1 " ^' ^" M M M 1 9 Eijiojnj 1 1 1 " 1 1 " 1 1 1 1 1 1 1 1 1 1 1 1 00 00 N 0 . N PJox M M r>vo 00 r^oo >orot^r^iH«| 1 1 1 1 1 SI 1 b M o JU30S3|OpV 1 1 1 1 1 1 M 1 '*'^" " 1 1 1 1 1 1 « do N 9 Eiii3jn_j 1 w m^O 00 t^OO vOCOCONI 1 1 1 1 1 1 1 ■+ bv bv S Eijioanj " 1 1 1 1 1 1 1 1 1 M 1 1 1 1 1 1 1 "^ ^ 00 vO . PJox NNCOsOr^mOO^sO-^--1-MCOrOMN»H 1 ^ ° 1 1 o 1 1 N o o iuaosajopY 1 1 1 1 1 1 iMHNrONtOCONNMliH 1^ M M N M 9 Biipjnj " " >- 1 1 1 1 1 1 1 1 o^ o -^ 00 vO . N :- l> > OS N PIOX 1 N M COsO Tt-^O 00 1 rOMrO«inT(-f->| M 1 al 1 N M CO }U30S3IOpV 1 1 1 1 1 , 1 „ 1 „ „ „„ ^^„| „ 1 0 y^ CO M 9 EJlpJOJ 1 1 N m\D Ti-so r^ 1 <^ 1 1 1 1 1 1 1 1 1 N CO ^ OS S Bi]pjnj[ 1 ^ 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 M GO r- CO K 1. M Q u-jO m 0 mO u^O >00 i/^O u^O "^O u-iO 10 r^ob do ONC^O 6 *-* « NNroforl-Vio in sD \b 1 1 1 iTTTTTTTTTTTTTT O inO u^O lOO lOO mO inO iriO lOO lOO C C h ^ l 1 p rJ M 00 iri juaossiopv 1 1 1 1 1 1 1 1 M 1 ro^ Tl- r^ ro M 1 " 1 in 9 Bi[iojnj 1 1 M w M M CO ^^D T)- ro 1 M M 1 1 1 1 en — lEiOJ, 1 1 MMNTj-l/1001OOv->t-N^mTt-Ni-i 1 8 1 1 rn M CO JU33S3lopV 1 1 1 1 1 1 1 |M^rt-0010in-|-MMl IT) IT) IT) >" ro cn m 9 Bijioanj 1 1 M M N 't •nOO ro lO 0 ^ " 1 1 1 1 1 in "^ r^ CD •- ^2 PJOX 1 1 1 1 " " 1 " ^« -> " " 1 1 «•! 1 00 M vO }U9DS3IOpY 1 1 1 1 1 1 1 1 1 " 1 M M 9 Bi]pjnj 1 1 1 1 " " 1 " ^ " " " 1 1 11 1 1 Ov Ov 0 ^ ^ 4 PJOJL M 1 i-.>HT)-a)o^r^co^oo«<~i"" 1 '-''-' S 1 1 M 00 }U33S3]OpV 1 1 1 1 |Mi-iMMno>r-5M««l«i-c tn l/^ -^00 N N N 9 Bijpjnj wl wHTj-t^oOi/lMMMl "1 1 1 1 1 " r- 6 in M N . ^ 4 lEJOJ, 1 " 1 1 1 i 1 1 1 1 1 1 1 1 1 1 1 1 " 1 1 in N do M 9 Ei]pjnj 1 " 1 M 1 1 1 1 1 1 1 1 1 1 1 1 1 „ o „ M 00 rOOO lElox 1 1 1 1 1 1 1 " 1 1 1 1 1 1 1 1 1 1 " 1 1 1/1 " 9 Bjipjnj 1 1 M 1 1 1 " 1 1 1 1 1 M 1 1 1 M O N o ;< 00 N N It'JoX 1 1 M M M 1 1 " 1 M 1 M 1 " 1 1 in M 9 Eijpjnj 1 1 1 1 1 1 1 1 1 1 " 1 1 M 1 M in M o r^ O M C MB 6o 00 b^ o b 6 w *^ M PJ comTh-^u-iinvOO 1 1 1 iTTTTTTTTTTi i i i lino "lO •no loO inO inO y^O inO inO i^X) X> b^ 'o b 6 " ►"■ M rl fnrn44"-. irivb h Ph < c HI 0 05 1 APPENDIX I 173 174 DISCOVERY REPORTS CO PIOX 1 1 1 1 1 1 1 ll-IMN'OT|-rOT)-MTt-^llHlMl IM 1 8 1 1 5- auaosajopv 1 1 1 1 1 1 1 1 1 1 ■-■ rooo 0-1-fO-l-"-l-^|'^l"l 1" 00 \n \n '^ 9 Bijpjnj 1 1 1 1 1 1 M 1 1 1 1 1 1 M 1 1 M 1 1 \ri \ri t> en leaox 1 1 1 1 1 1 1 1 1 |i-iMMh-«r)r^t^Mr^'-iN| 1 1 1 1 1 ^1 1 vb juaosajopY 1 ^ 1 1 1 1 1 1 1 1 1 |_«^KH-' 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 mmm"" iiiiiiiiiiiiii g 1 1 M 00 o luaosajopY |||||MmsO"0\ON"|l|l|ll|l||||| 9 Eniojnj w 1 N u-1 fO nvO °'^"||ll|l|ll||l||||| O 00 sO ^D CO w OmOioOu^OmOulOioOinpy^pioOioOy-iOioOioO a e - s s h cu < a a i c 0 e2 i'r'r'ri*'rf'j'"""i7iTTTTi iTi : i i i loOiriO^^OiriOwOiriO'npvnpiopyipy^Ou^OinOiri 00 b^b>6 b « " M N c-ifo^^>ninvb^KKobob b^c^o o « >-• APPENDIX I 175 CD «■ IKlox Ml 1 |>-ci/^Th-!)-cCu-iroir-OOr^^-*-N| 1 1 ^ 1 1 00 lO vO aU3DS3]0pY 1 1 1 1 1 N N ro 1/1 Tj- noo 00 to ■+ -^ N 1 1 I 00 t^^ 9 Biuo-md : " 1 M " '^ « " '^ " 1 " 1 1 1 1 1 1 1 1 ^ "^ ^ N I-* PIOJL i-'NC0MO^00"-'^>-Ht^0\00rt-\O 1 M 1 1 i-i " " " III 8 1 1 to 00 to juaDsajopv 1 1 1 1 M III 'O vO i, 9 Biipanj MMforj»oro^OTi-r^i-| mI i i i i i i t^ t^ 0 to to ^ en X PiojL 1 1 iMNrlii-iOa^-i-Or^iriri'i-Ml 1 1 ! Ill HH M W OJ •-• 1 1 1 1 8 1 1 M to M to to luaDsajopv 1 1 1 1 |'HN\OTt-(SJOfOmN'^M| 1 1 1 N 00 00 io 9 EijiDjnj 1 1 iNrlH-if^Tl-u-iMl 1 1 1 1 [ 1 1 1 1 ""? IBJOJ, 1 1 1 " 1 1 M "" M M 1 1 1 1 M to| 1 00 in to to luaosaiopY 1 1 M 1 1 1 1 1 " 1 1 1 1 1 1 1 1 1 1 ui M to r^ to w 9 Bi(pjnj 1 M " 1 1 1 1 " 1 1 1 1 1 M 1 1 1 1 vD to Sroo V^iOJL j lHl-lMlO'^Ot^O^O^OO^nmrldM | | 1 | ;:-l 1 00 JU33S3IOpV 1 1 1 |>-'f^\D00C0O00f^C0Nf^M| 1 1 1 ■^ K ^ 9 BlJTDjnj 1 " " ^ ^" 1 1 1 1 1 1 1 1 1 in '-' rl M FlOi 1 1 1 tOCOCOl-iNt^0O"l-tOO00-:)-u1-+Ni-i ° 1 1 O 1 1 CO io HI juaDsajopv 1 1 1 1 |"iHNinOOroroO00^inThMM 00 00 »o 9 Bijpjnj 1 1 1 to to c 1 1 .. 1 1 „ 1 1 1 1 1 1 1 1 M « 00 " " N Station Date bJD d c S inoy-iO y^9 ir»0 loO i^O mO »oo "^O WiO O y^O y^O u-)0 mo loO >oo vnO>^0 ii-»0>o o o '-' '-' N N mc')Tf-'^u-)iri^bvb i> i> oo 6o c^c^ •3 8^ D > d I) a 'a C h eu <; 176 DISCOVERY REPORTS l^ioj. 1 |MMMJincr)MNMMw| 1 1 1 1 2^ 1 1 00 CO CO o au3DS3]OpV 1 1 1 1 1 1 " " " ^ " " " 1 1 1 M in I-" 9 Bi[pjn_j II 1 "M II M 1 1 1 II 00 r p IBIOJ, "nM-ror^T|-ONi/lf^M| 1 1 1 1 1 1 ^ 1 1 00 N M O juaossjopv 1 1 1 |«\OioO>rONM| 1 1 1 1 1 1 9 Bi[i3jnj w M -^ f^vD 00u->MNm| 1 1 1 1 1 1 1 in w 2, M leioj. 1 „ „ N " 1 1 1 1 1 8 1 1 N 00 N o juaosajopv 1 1 1 1 ll-l-cmr^u^loN«| 1 | 1 1 OO 00 t^ ro CO f^ 9 Bi[pjnj InThMNOOwOOMNI 1 1 1 1 1 1 1 \0 S! S « CO en « IBWX « M M M 1 1 1 1 8 1 1 00 oo M N CO N luaosajopv 1 1 1 |«NNu-ir^Tj-inr)MM| 1 1 1 M M O n CO ;j. 9 Bijpjnj H -^^D ^ooo^mt^r^-*"] 1 1 1 1 1 1 OO 00 CO 2^ " en >< l^iox 1 M M M III {^1 1 CO CO M 00 00 1U30S9]OPY 1 1 1 1 |«nr^NI^^MMN«| 1 1 ^ CO M ^ CO Th ^ in n 9 Ei]i3jnj; |«MrIinioOO«N| 1 1 1 1 1 1 1 CO ^ 11 en >< PJOJ. 1 1 1 j M-^-^Ti-Tj-r^oo-ot^'l-'tl « o 1 1 o 1 1 00 vO M N luaosajopY 1 1 1 1 1 1 1 '^'^ « 2" ^"° t^ •*■ -t 1 M CO CO '-' t^ r^ Oi 9 Bi]iDjnj MM" '^^'-^ "" II II M 1 1. en Q 1 OmOinO'nOoOin 66""NMroco^T|-io irivb vO f^ KcO CO u-)0 loo ir»0 mo inO in O u^o in o loO £ c c 0 5 > i C o "a e2 APPENDIX I 177 0 ^ l^ioX 1 " 1 1 1 1 1 "" 1 1 1 1 1 1 1 1 1 1 '^l 1 M CO }u3Dsa[opv 1 1 1 1 1 II "" 1 II 1 1 II 1 11 0 PI t~ 10 vD to vO 1-1 9 '^ni-^nj 1 " M 1 1 1 M M M 1 M 1 1 1 \ " V,? IBIOJ, 1 1 1 1 1 "" 1 1 1 1 1 " 1 11 1 1 1 "11 00 CO juaosajopv IIIIMIIIMI-lllIM ""^ 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 I n »n 9 Biipjnj 1 0 1 1 1 1 1 "" 1 1 1 1 1 1 1 1 1 1 1 1 n t^ y^ 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 vb N ON > ° Flox 1 1 1 1 1 1 1 1 " 1 1 1 1 1 1 1 1 1 1 " 1 1 CO M 9 Bi]i3jnj 1 1 1 1 1 1 1 1 " 1 1 1 1 1 1 1 1 1 1 *J1 H 0 t^ 0 r^ 00 ^0 i^i°J. 1 1 1 1 1 1 M M M M 1 1 1 1 " 1 " M 10 ob M juaos^iopv 1 1 1 1 1 1 1 1 1 1 I 1 M 1 1 11 " m M 0 ^ 0 CO Ooo ?>« en X ^ 00 I^IOJL « M N r^rj-o^r^ThTi- t^-o n ri « 1 1 « 1 1 \0 1 1 CO CO r^ m luaDsaiopY 1 1 1 1 1 ^« ro^r.^ C..OM 1 1 " 1 1 ^ "^ " noo Th 9 Eiiiojnj » M C .. ^ ^^ - 1 1 1 1 1 1 , 1 1 1 1 00 n M 00 N ;^ ;, g>00 CD >< ^00 I^JOX 1 1 " 1 1 1 1 " 1 1 1 1 1 1 1 1 1 1 1 « 1 1 0 p N iu9osa,opv M 11 1 1 1 " 1 1 1 1 1 1 1 1 1 1 1 in WON 9 ^IlP-'nd II " 11 11 1 II 1 11 1 1 1 II 1 " S> f 1^ en ><■ ^00 1«10X 1 1 " 1 1 " " " '^" " 1 1 1 1 1 1 1 1 2 11 N 2 lU3DS3IOpY 1 1 1 1 1 " 1 " "" " 1 1 1 1 1 1 1 1 9 Eijiojnj 1 1 " 1 1 1 " 1 1 1 1 1 1 1 1 1 1 1 1 "^i Station Date OtoOinO»nOmo»noioomOioOtnO bb'HMrJNfnroThTt-in iovb sb t^ r^oo co 0 in 0 mo too inO too c>6 o •-' >-■ m n m MhHMMIHI-00 i^ s c c ca y . — 1 i d o E O i-« "-I N M fom-^Ti-inii-i\oo r^r^ooco ovoo o *-" i- n n MMMMI-ll-IMMMMlHIHMMMI-lMI-HWrsINNMNN APPENDIX I 179 IBJOX 1 " 1 1 1 1 1 M " 1 1 1 1 1 1 1 1 1 1 " 1 1 1 1 1 1 1 1 1 1 M 1 1 1 1 "1 1 in M CO juaosajopY 1 1 1 1 1 1 1 1 1 " 1 1 1 1 1 1 1 1 1 1 " 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 o ■h K 9 "'HP-'nd 1 " 1 M 1 M 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 M M 1 1 1 1 1 1 1 1 1 in ro 6 OS N o . N PIOJL 1 1 1 1 " M 1 1 1 1 1 M 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 M 1 1 1 " 1 1 N N M " 9 Bi]pjnj 1 1 1 1 " 1 1 M 1 1 1 M 1 M 1 1 M M M 1 1 1 1 1 1 1 1 1 1 1 1 MOM O N lEWJ, 1 1 1 1 n 1 " " 1 " M M 1 1 1 M M 1 1 1 1 1 1 1 1 M M 1 1 1 ^ 1 1 00 0 in juaosajopY 1 1 1 1 1 1 1 1 1 1 "" 1 1 1 M 1 1 M 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 o CO tn CO M 9 Bi[iDjnj 1 1 1 1 1 " 1 "" 1 1 M 1 1 1 1 1 M 11 1 M 1 1 1 1 1 1 1 1 1 1 1 1 00 -1- r^ CO \b CO vO 1-1 00 ? 00 "^ Fioj. !_,,_, (VI Mt^OOrorl 01 (MroM'-it-* lEloj, 1 I 1 iMHwMMMr^i>a^iMOt>r^]>i>Ti-u-)rii-.Mi-.i i i i i i i i i i i i 8 1 1 M }U30S3]OpY 1 1 1 1 1 1 M M 1 N f^\D ONw^ONt^r^r^-^ioNwi-iMi 1 1 1 1 1 1 1 1 1 1 1 en m^ CN O f^ 9 Eiipjnj; 1 1 1 1 " ^' 1 " " 1 1 " 1 - 1 M M 1 1 1 1 1 1 1 M M 1 1 1 i 1 1 1 00 <~' I^^OX li-tli-i|cO|COiOiOvOO^WMOOvOO'-iu~)'o m o mo u-joy^o y^o y^9 V^9 y^9 O o "-i iH M ri rof^'d-'^inioo^o r^r^obob 0(^0 0 *-> *- n n foro"+-1"ir)m\0\0 r^r^oo 2 an a c 2 a . 0 i3 > >' C o "a CO c 23-2 i8o DISCOVERY REPORTS O M I^JOX 1 M M 1 1 1 1 1 M 1 1 M 1 1 1 M M 1 1 1 1 1 1 M 1 1 1 M 1 1 1 1 M 1 1 1 1 CO 1 1 N 00 juaasgjopv M 1 1 1 1 1 M 1 1 1 1 M " 1 1 M " ^ 1 1 " ! 1 1 M 1 1 1 1 1 1 1 I M 1 1 1 1 1 1 I^^OX 1 |j|||ll||-'||"-r'l't "-.^ r^ u^ U-, r^ r^ o^^J^ u-,0 -i-mmr^-J-i-'N|MN|'-H-| j>--in-J- u^sO r* "^ lo m r^ O^O u^<; ■+r^r^m-t-i-'N| rlNj -h^i | i-.| 1 1 1 o O O ^" o o ^ '-' *^ r( 00 M CO M in . Piox I [ 1 1 1 j I 1 1 1 1 1 1 1 |'-«|>-<*h|'-.mNNN'-'N'+ u-<\0 'O'Ot^u". 0»d--1-r^f*>MNMN'H 1 ►- o 1 1 00 1 1 O O lO JU33S3[OpY 1 1 j 1 j t 1 I 1 1 1 1 1 1 JKHJHH-.l.-.H-NNN'-iri-h liOsD \OsOr^>/^0\-i--tr^roN«MN»H 1 m OOP 00 O Ov «> 2, o " ^2 PJOi 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 "1 1 1 1 1 1 M 1 1 1 1 1 1 1 1 1 1 1 1 1 " 1 1 " juaosajopY 1 1 1 1 1 1 1 1 1 M 1 M 1 1 1 1 1 1 M 1 n 1 1 M 1 1 1 1 1 1 1 1 M M 1 1 1 1 MOM O u^ CD — I^JOJL 1 1 1 1 1 1 M 1 1 M 1 1 " 1 1 M 1 1 1 1 M 1 M M M 1 1 1 1 1 1 1 1 1 1 1 1 1 1 " 1 1 N 6 - }U33S3JOpY 1 1 1 1 1 1 1 1 1 1 1 1 1 1 "1 M 1 1 1 1 1 M 1 M 1 M M 1 1 1 1 M 1 1 1 1 M 1 m M O N O o 1^ l^ioj. -|||||j|"lllll""ll"l"llll"llllllllllllllllllll ■^1 1 0-- try b N o luaasaiopY GO ^ r^ m m M CX3 M g Eijpjnj " 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 M 1 M 1 1 M 1 1 1 1 M 1 1 1 1 M 1 1 1 1 m PJOX 1 1 M 1 1 1 1 1 1 1 "1 1 M 1 1 1 "1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 M 1 M 1 1 M «| 1 b M juaosajopY 1 1 1 1 1 1 1 1 1 1 1 " M 1 1 1 1 1 " M 1 1 M 1 1 1 1 1 M 1 1 1 M M M M 1 1 en — lEJOX 1 1 1 1 1 1 |NN'-'f*^|'-'N-t-r^NNN"^r^m'-"'4-r^'-iNrO»-''-'M| 1 1 1 1 1 1 1 1 i i 1 | 1 j r-i 1 1 >o 1 1 -1- in auaosajopY 1 1 1 1 1 1 |MMi-.m|i-ipi-^mNciN^-r^mt-iThmMM(*iMH^M| [ i i i i t i i i i t i i i r-)0 i" ■"On M N PJOX ||[||i-i«lMi-4MMNNf^rn|Nnj||[|j[|[j||||j||||jll|||l| S 1 1 OS 6^ 5U30S310pY |||||[«-t|NiHNH.MNr^rn|NN||||||||||||||||||[l||j|[j| 9 Eiiiojnj 1 1 1 1 1 " 1 1 1 1 M 1 1 1 1 1 1 1 1 1 1 1 1 1 1 M 1 1 1 1 1 1 M 1 1 M 1 1 1 1 1 1 m M m^ c ■B 2 Is i'^ O u-i O i^i O un O >ri O lO O li-i O y-* O ]f> p '^' p Y^ O \r. p y-i 0 U-. 9 y> O y-- O *f.O u-. O u-. o i^ O u-1 O U-) O 1'^* 0 ro V V "-1 i^-O vb K i--6c do C>0>b b '<-< *-" M Mrnco^-^-io i/-*^b sb r^ KoC CC oob b ^ ^WMr^r^iih^iri ir,vb [[llllllllllllllllllllllllllllllllllllllllllll O mo "lO mo mo mo mO mO mo \n p mOmO mO V^OmO mo mp mp mo mO m, o mO mo m 'r^ rn '-i- '-+ in m^O 'O K f^C>bb^^-iNMr^ro^H-m m\b \0 t^ I^co cbc-bsb b ^ MNNr'irn^^-mm Total Percentage Av.length.mm. E E d o "a E K en C ca o h 1 APPENDIX I i8i o N . CO Pioj. MM MMMN Mill 8 1 1 IH to o juaossiopv 1 1 1 1 |lHnN00tHl>OOONt^'tC-«| | 1 ] 1 ^1 1 00 t^ ■* }U3DS3JOpV 1 1 1 1 1 1 •- N N r^o oo-^NcoNMwi 1 1 1 1 ^ in »n 00 " 9 EiIiMnj "II M " M 1 1 1 II 11 11 1 1 o m o M jb;o I 1 1 N"| rn'tO^ONOOiHCOvOvDThwNl | 1 -■ 8 1 1 p in CO o juaosajopY 1 1 1 1 1 mwOOvOOOmCOxOsD-I-mMI 1 1 " M On ON r^ 00 00 m M 9 Ei[iDjnj II ^ " 1 " - 1 "" II 1 1 11 11 1 1 1 1 1 o M M t^ " " M IH o n in . 00 jEjo I 1 1 1 |MT^N^-)t^^-l«ro^^-l■;^lnt^M■-■^Il-.| 1 8 1 1 M 7)- in 8 juaosajopY 1 1 1 1 1 1 |i-.-O0i-'N-l-"')-'nI^«"N>-'| 1 On t^ r^ r^ 00 00 m M 9 "HPJnj 1 1 1 1 N ^« « « « 1 « 1 1 1 1 1 1 1 1 1 1 1 o fO CO ?^ H. M p, Pioj, 1 1 1 IcoNlrj-roMmo^ ThGO in O 00 -^ fT) M 1 ) 1 1111 1 MMMM III 8 1 1 9 NO O CO juaasajopY 1 [ j j 1 1 |Nro»-t>nC>'^ooinOOO't-ro»H| l 1 in ro CO p 9 Eijiajnj 1 1 1 1 '"^ 1 ^ 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 '^'^l O leiojL 1 1 1 1 1 w ro CO n in\D c^^N00coco| 1 I 1 1 1 ai 1 vO N io i-t O 00 juaosajopY 1 1 1 1 1 lMNcnininc>'^MOOcoco| 1 1 1 1 1 CO ■^ O in 9 Eijpjnj 1 1 1 1 1 """ 1 1 " 1 1 1 1 1 1 1 1 1 1 1 1 «n ^00 « CO c .2 o 2 « O ""-O >nO i^-O u-jO inO inO inO mO »nO *~* N N rocO'^'^minv^^O r^i^ooco c^c>o o B B tffl-C c c > C o OJ "a B a .G "a I l82 DISCOVERY REPORTS o CO O . o I^JOX 1 « « - 1 N 1 C « 1 1 1 1 M « 1 , 1 1 1 , 1 1 1 1 1 1 ^1 1 0 CO N 00 1-1 }U33sajopY 1 1 1 1 M 1 1 M 1 1 1 - " 1 M 1 1 1 1 1 1 1 1 1 00 CO r 9 M M 9 Bi]i3jn^ 1 «"" 1 ^ 1 ^ " 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 PJOX I lOcOl^CJNOO^-^OONt^t^Niric^u-il Own 1 1 I w II MM MM 1 III 2 1 1 N oo o O CO Ov . ■^ . ON }U30S3IOpY 1 1 1 1 1 1 1 Nil-, -t-COI^Mt^r^NlOMinl MMNl 1 1 M ^ M p vO M 9 Bjipjnj 1 1 O CO.«N ^=° -> 1 " " " 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Tj- o» o ^ b M CO M O 00 M i TEJO T MCOCOt^OO^l^vOvOCOOOMCOTh^|N"i-M-t-| 1 l"| g;| | m CO M M o M aU3DS3]OpV 1 1 1 1 1 |MM^O.^OOMCOt^|«MM., -t| 1 |"| ^ %'P^ j ^ M 9 Biipjnj 1 "^ NCO00r--0vCco^rI|co|M| 1 1 1 1 1 1 1 1 1 1 1 1 1 1 °;^^ in M o to Pioj. 1 1 1 1 1 1 1 1 1 1 1 " 1 " i 1 1 " 1 1 1 1 M 1 1 1 co| 1 OO p M CO aU30S3[OpY 1 1 1 1 M 1 1 1 M " 1 " 1 1 1 " 1 1 1 1 1 1 1 1 1 00 CO 0 O 0 ^ o ro in . ^« vO N IBJOX j 1 1 M 1-1 N N Th t^ X^OO iri-^NNMNNwl | j"^! 1 1 1 S>1 1 00 CO M o lU3DS3]OpY 1 1 1 j 1 1 |wNTj-'OiO'4-Nn'HMNM| 1 l"^! 1 1 1 CO y^ N CO CO io VO M 9 Eiipinj 1 1 1 M M N 0. CO .o CO N 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 in CO M O CO ^ . ^ . TBJOT 1 1 1 1 "CONCOCOincoa>COrlON'tvD>«COCOI 1 "1 1 1 ?l 1 M in in M O 8 lU33S3|OpV 1 1 1 1 1 1 1 |MTi-Ti-o.r^cJONTi-sOir)cocol l"| 1 1 CO M o^ c^ 00 t^vij 9 BiiP-mj 1 1 1 1 M CO C, CO N M ^ 1 M 1 1 1 1 1 1 1 1 1 1 1 1 1 1 M N M CO 0 CO CO . IBJOJ. 1 |MMM|MnMM|OOt^'tvD>n'l-NN|Ml 1 1 1 1 1 %\ 1 M in M 00 m luaosaiopY 1 1 1 1 1 1 iMMMIODt^ Tl-vO m^NNlMl 1 1 1 1 1 00 CO 0 CP •^ 0^ in M 9 Eiipjnj 1 1 """ 1 "" M 1 1 1 1 1 1 1 M 1 1 1 1 1 1 1 1 in in o :*■ Station Date « 1 u^O iTi O ^^ 0 i^O uio inO u-)0 J^O »oO lOO >aO inOy^O V^ ONO 6 >•* *-> M ri for^V-h»nin\CiO t^f^coco O b^ O O ►-' *-* Jli ^ S g C o "E g ca c O h 1 1 1 i 1 I 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ' L i 0"^o »i^o u-)0 >oo mo loo «i^o u-)0 ino v^o y^o y^9 y^9 APPENDIX I 183 0 N • 4 I^Jox 1 1 1 " 1 1 1 1 1 1 " 1 1 1 1 1 1 1 1 - 1 1 1 1 1 CO, 1 CO p io t-l CO juaosaiopv 1 1 1 1 1 1 1 1 1 1 1 M 1 1 1 1 M " 1 1 1 M to M 9 Eiipjnj 1 1 1 " 1 1 1 1 1 1 " 1 1 1 1 1 M 1 1 1 1 1 1 1 0 PioX [rii-'MNc^f^-^oooooocoor^iniovo-i-nMlM<->'-i oil in CO o auaosaiopY 1 [ 1 1 1 ]'-lMfosOl>\Dt^c^^^u-lu-)0'+r^M|c^^-'M CO IT) lO '^ i^ i>sb 9 EIJIOJOJ |NMMNr^NmnNr^N»--i| 1 1 1 1 1 o 0 r i^ t^ 4 9 Eijpjnj M CON.O ^CO^ 1 1 M 1 1 1 1 M 1 1 1 1 1 M 1 o O 0 t^ CO CO i, 1.^ s 1 O \ri 0 »^0 inO m O inO toO lOO >oO tnO mO lOO m O 2 So c c ho,-*; i a o 1) i O O t-- M M M ror^-^h-^inu-jo-o r^r^oooo cncno O >-> m ri 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 i 1 1 1 1 1 1 1 ino y^o u-io »oo loo lop y^p ^o y^o y^p y^p y->o in 0*6 6 ^ *-* N ri forOTt-Thioio\bvb i> i>6o OO 0^0*6 b w ►^ 1 84 DISCOVERY REPORTS 00 .- '=2.x =0 >, CO ^0 =2. 3 00 '3 c n n c« Q IBJOJ, luaosajopv 9 Bljpjtljj PIOX jugosajopv 9 Et|i3jnj PIOX auaos3]opY 9 Biipjn_j^ 18JOX }u33sa]opv 9 BijiDjnj Piox luaosajopv 9 Eijioanj FJox JU33S9lOpY 9 Ei^pjnj r}-hf0in^f0^»or^'^'0\0ro| fDc*^'-'^N p-"NvOmt^'^\OvOc*^l cocOMtHN r) -^fommi-i u-jn Mf-500fO""100vOt^Or^C>mMT)-N-*-+| •-> M 1 00 u^vO inino^t^ONf^M -sj-N Th-^ cimOforol N'-'Nt-t M N " M in N N w N m r)roM-t-^ori^ncovDvO"l-vD"Ot^u-iu-, I nm n 't fo ■* N fOvO ^O ■*>£> "Or^'O'^l NNm f-I M M N n ro-^-^"-" t^OO O irif^)-i o^lO•-' iriM ^ M M foN t^r^vO ONi/^r^w O^u^M ION f^ N f M M •-■ M ■+ T)- N 00 ■+ r- n fO pr^ r*^ fo "^ in N ^ 00 O in f*^ M 0\ f^ \0 ^'^ (^ 00 o vb (:> m CO 00 00 vo c^ o "^ o lO o m O in o in o in o in O in 0 in O in o in o 0 o " M N CO CO -)- Tt- in in vO ^O t- r^oo 00 o^ o> o 1 1 o 1 in 1 1 O in i 1 in 1 O 1 m 1 O 1 m T T 1 O m O 1 in O in O in o in o 0 O hH l-( N M C*5 n Tl- m m^ \o t^ t~-00 00 o> 0^ m O >o O "O N N N N b 6 M w N N M N h PL, < o c o 00 o o O O 00 a S o APPENDIX I i8s o CO c> . T 00 o CO O a .2 „ 2 « l^ioj. }U33S3[OpY M I M M Tj- M 9 Eijpjnj lEJox JU33S3[OpY N -^cou^inoo ovcocoooo r^c>oo lopi m co rocor^co cocooGO r^OMX >on m m 9 Bjipanj lEloj. ^uaosajopY 9 Biipjn_j FJOJ. juaosajopv 9 Eijpjnj N CO CO N t^l »n p» •^\0 mo t^o tno^Nxo cot^-^r^m roiriinoo coco M-sO cor^-^r^irj ^n N -+0 N m N N N I^JOJ. juaosajopY 9 Eijpjnj CO ri N IT) rf- *-< 0^ ob tn iri CO CO i^ M ir^ IT) CO C\ O^ "^ M » O CO CO ;^ 00 CO >o "^ \j-, o ^ be C in o lO o >n o mj O in o "n O in o in o in o " N M CO CO '^h in ino ^ 1^00 OO Ov o> o CI o li^ o in O 1 »o 1 0 1 in 1 1 1 O m o 1 in 1 1 1 1 O m o in 1 o 1 in N-l M M CO CO '^h •t in in\o \0 t^ t^OO 00 0^ N M N N M in o in N CO CO -^ N N M I I in o I •H w N M 2 bo c c 2 S-^ h eu •< ■^ ^ O "a o 24 1 86 DISCOVERY REPORTS o FJOi 1 1 ■*• ^ t^ t^xO OvO00t~-t^-*-rot^"1MM 1 NfOwi-cMw 1 1 1 1 M 8 1 1 o in N lU3DS3JOpV 1 1 MMTj-invot^\Door^r^-tfor^"^ro" 1 mc^ihmnm i i i i m 00 00 ■+ oo 00 r^ 9 Eijpjrijj ||NMroN|fO|||||l|||||||||l|||||| N N ^ o ^ . N FloX 1 1 1 1 1 1 1 1 1 1 r' 1 1 1 1 1 1 1 1 1 M 1 1 1 1 1 1 1 N 1 1 in N aU33S3IOpV 1 1 M 1 1 1 1 1 1 1 " 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 N O t^ 2 r- O 00 • I«10JL " 1 1 1 " 1 1 1 1 1 1 1 " 1 " 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ^ 1 1 1 8 Th auaosajopY 1 1 1 1 1 1 1 M 1 1 1 " 1 " 1 M 1 1 1 1 1 1 M 1 1 1 1 IT) N O t^ •nob 9 Bijpjnj " 1 1 1 " 1 1 1 1 i 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 in MO" in ^ o CO I^WX M 1 " 1 " 1 *l " " M 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 CO , 1 Ov luaosajopv 1 1 M 1 1 1 ^1 "" 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 9 Eijpjnj 1 1 1 " 1 " 1 1 M 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 o * . M IBlox 1 1 1 M "> 1 1 1 " " 1 1 " " " 1 1 " 1 1 1 1 1 1 1 1 1 1 1 2 1 1 O 00 M au33S3iopv 1 1 1 M ^ 1 1 1 " " 1 1 " " " 1 1 " 1 1 1 1 1 1 1 1 1 1 1 fn Ov 0 P °"00 9 Biipjnj 1 1 1 1 1 " 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 « o r- o to pwx mNm 1 i-iij-j\0'Ou^'-'000Of^C0O>--'i-t| 1^1 1 1 00 00 a> 1-4 9 Eijpjnj "^11" "•"" '^ M 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 \0 \0 00 M M ^ .2 „ c S mo »nO u-)0 inO loO inO u^O mO toO mO mO >nO uio >nO »nO e E c c a C O "a CO Mi-ih-M»-.i-ii-.Mi-.i-ih-.i-(iHMi-iNMNNMOlMMrJNNNNMN 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 i 1 1 1 1 1 1 1 1 1 0»^0 lOO lOO u-»0 ii-)0 u->0 lOO u-)0»oO u-jO "-jO ijiO mO »00 "^ MiHi-ii-(Mi--H>-«| 1 1 1 1 1 009 •n 0 00 0 in >^ N N IE»ox ] 1 j |NNi-i(OI^coNNmroiON\ONsO>-'rnM| 1 it-ijwi-. t^ 1 1 ^ 1 1 00 JU3DS3(OpV 0 9 ''!ipjnj 1 1 1 1 1 ^ " 1 -> 1 " 1 1 M M M M M M 1 M 1 Ov CO^ 1. ad ►J p y^O u-)0 u-)0>00 100 u-)0 »00 >oO u-iO u-)Oi^Oii-iO inO 1^0 MM»-(hHi-.wi-.i-i.-.hHMMMHHi-.MNNNMMNMMNMNNM 1 i 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 irio mo in 0 loo »oO in 0 100 1^0 mO u^o in O ^n 0 m O yn O m '-' N N mrOThVinio^O^O t^Kdoob b- O O O ^ *-' N N c*icOTj-Thioin 1^" c c s s > c a CQ i88 DISCOVERY REPORTS 0 Flox 1 1 1 1 1 1 M " 1 1 1 1 " - 1 1 1 1 1 1 1 M 1 1 M 1 1 « 1 1 b N 'I- }u3DsajopY 1 1 1 1 1 1 1 1 " 1 1 1 1 "" 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 00 n O >« o 6 " N O CO 00 • N FJOJ. 1 1 1 1 1 1 t^ 1 1 00 3U30S3IOpY •HM j 1 1 Miou-^ior^c^N^orooiOHif^rjrrj-ron^HM i h-. | | 1 On r- ov :h N 9 Bijpjnj " 1 " 1 " 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 m Tj- r~- O vd FloX 1 |"|'Hl-l!^lCOU^rlvDMvO':^t^COOO^O"^Or-)lr, 1 j„-l-hH«M 0| 1 vO in On luaosajopY 1 t^ III « « II O O M " -' M o N Flox 1 1 1 1 1 1 " 1 1 — " " 1 1 " 1 " " " 1 1 1 1 1 1 " 1 1 n= 1 1 9 N m juaosajopv 1 1 1 M 1 " M " " 1 1 " 1 " " " 1 1 1! M " 1 1 :=8 t O FloX '"^ 1 " " 1 1 " M " 1 1 1 1 " 1 1 1 1 1 M M 1 1 2 1 1 N JU33S3|OpY "1 1 " " M " 1 1 " 1 1 1 1 " 1 1 M 1 1 1 1 M 1 9 Eiipjnj 1 1 1 1 " 1 1 1 1 1 M M 1 M 1 1 1 1 1 1 M M 1 M "=^f O FJOX 1 1 " M 1 1 1 " 1 1 M 1 1 M M 1 1 1 1 1 M 1 1 1 1 . 1 1 IT) N M }U33S3[OpY 1 1 " 1 1 M 1 " 1 M 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 O Flox 1 I.I 1 1 1 1 1 1 1 1 " 1 1 1 1 1 1 1 1 M M M 1 M M "II M lu3osa[opY 1 1 II 1 1 II II " II 1 II 1 II II M 1 1 1 II 1 1 " §1 Station Date J3 0 i/'fO i^O »oO w*0 'nO iTiO iTiO »00 U-, O »/^0 lOO u~iO »oO lOO »o E e c c h Oh <; £ > a C u o £ "3 Mi-.i-.i-ih-..-.i-(H-i-(i-iMrjrjMNrjriMrinr]NriMnfvir]nNN 1 1 1 1 1 1 1 1 1 i 1 1 1 1 1 i 1 1 1 1 1 1 1 1 1 1 1 1 1 1 u-)0 too u^O >00 tnO w^O »nO "OO "^O >oO i^O "^O mO »00 "^0 APPENDIX I 189 0 CO 6 N l^iojL 1 1 1 1 1 " 1 M 1 1 1 M M 1 " 1 M M 1 1 1 1 1 1 " M o m 8 N luaosgjopY 1 1 1 1 1 " 1 1 1 1 1 1 1 1 1 1 1 " 1 1 1 1 1 1 1 M 1 1 -8? 0 CO CO..: d i^'°J^ " 1 1 1 1 M 1 1 " 1 1 1 1 1 1 i 1 1 1 1 1 1 1 M 1 1 1 ^'11 o o w N JU33S3IOpV 1 1 1 1 1 1 1 1 1 - 1 1 1 M 1 1 1 1 1 1 1 1 1 1 1 1 1 1 "g>I 9 ByiiDjnj " 1 1 M 1 1 1 1 1 1 1 1 1 1 M 1 1 1 1 1 1 1 1 1 1 i 1 " S,^ 0 CO co-p JEJOJ^ 1 1 1 1 1 |'-'N|cOl-■^OCOOt^^I%OO^li^T)-^o|M|M| 1 1 1 1 8 1 1 o CO }U33S3IOpY 1 1 1 1 1 i-cN| cO"vDC^vO| Nl Ml | 1 1 0 o o o> o o ;, •H M ^ N 0 CO in '^ N lEJOX 1 1 1 " " " 1 1 1 " 1 " 1 1 1 " " 1 1 1 1 1 1 1 1 1 1 1 " °^ 1 1 Ov luaosajopv 1 1 1 ^ " " 11 1 " 1 " M 1 " " 1 1 1 1 1 1 1 1 1 1 1 " a> o o 2 ? 0 CO CO • M ]B}OJ^ ; 1 1 1 1 l'-'NC0^T|-Oioir)T|-C0ri-0-t0>-t-v0M^|i-.M| | 1 8 1 1 IT) GO N o o M juaosaiopv ■^ "^ M 0 CO f> .- N 3 m . 00 N FIOJL 1 M M " 1 1 " 1 11 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 M 1 1 0 O 00 N juaosajopY 11 1 1 1 " 1 1 " 1 1 1 1 1 1 1 1 1 1 1 1 11 1 1 1 1 11 ^gj 0 CO 00 M 1"!°^^ 1 1 " 1 1 1 1 1 1 1 1 1 1 11 M 1 i 1 1 1 1 1 1 1 1 1 1 " 1 1 in " 9>'!IP- C a; 0 mio^ovO t^r^oooo O^C^O 0 w '-' r> M rOflTh-^iomsOvO t^l>O0O0 0^ 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ino loO mo ino mo mO mO u^O mo irio loO mO mO inO m Tj-mmvb\b f^Kobdb 6sc>6 6 ^ --■ n m coco^^mmsbsb f^f^cbdb MMi-.hHi-.M«MH.MwMNnMNMMClNNNriNMrJNNN I go DISCOVERY REPORTS X! I— I Q w Ph Ph < o N ■-0 OO -a: & to -Si K <3 I in 00 i-i i o ro 00 in >? m o juaosajopY 1 1 1 1 1 |roNlOmi-<«nM| 1 1 1 1 1 1 1 1 o w 00 « i-i N 9 Eiipjnj^ III — ^ 1 " 1 1 1 1 1 II 1 1 1 1 1 1 1 c< vO N 'J- luaosajopv 1 1 1 1 1 1 1 1 - ^ " 1 1 " 1 1 1 1 1 1 1 II in in O ^ 2 j m 9 Eyipjnji II 1 M 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 O .-H VO N }U3DS3IOpY 1 1 1 I"! |«"rO'tNto|N| 1 1 1 1 1 1 1 CO O t^ "2 2" 00 9 BilP-inj 1 II 1 1 1 1 II 1 1 1 1 1 1 II M 1 1 1 1 1 1 1 Si juaosajopv 1 1 1 |rot-o ;^ 9 Ejipjnj 1 1 -"--^-^1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 HH CO rt - ? au33S3jopy j 1 1 1 1 1 (D m\0 lOsO lOsO M cj 1 i-t f 1 1 1 1 j -too ;^ o o CO N 9 Biipjnj[ 1 1 1 r' " ^ "" 1 1 1 1 1 1 1 1 1 1 1 1 1 1 O>00 ^ M I CO 0 }U33S3J0pY 1 1 1 1 1 " rovO t^ ro noo 1 m n « i i i « « i i 00 00 vO ^ CO t^ ;j. 9 Ei]i3jnj 1 1 " 1 '^'^^ 1 "" 1 1 1 1 1 1 1 1 1 1 1 1 1 0 " N N M < o "1 au33S3iopv 1 1 1 1 IMCOCOt^OOOfOI>OfO|i-|"| 1 1 1 1 CO in 0 "^ '*■ a ;j. Hi CO 9 Eiipjnj 1 " 1 " ^" 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 in in 0 7*" o CO lU33S3[OpV 1 1 1 1 |«U1WvOrOTj--+M-+| IMMWI j 1 1 M CO vO f^ covo ;j- CO 9 EinoJnj 1 1 " "^ " "1 ^ 1 1 1 1 1 1 1 1 1 1 1 1 1 1 M CO N o l-( t^ " 1 ^ o }U33S3I0pY 1 1 1 1 1 |rj-Mt^vOrOMvO<^M|"|«l |"l CO vO CO to r^ ;j. OS o 9 BflP-ind 1 1 ro M ro N « « 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 N ThOO 1-1 M ;, o <-< 00 JU30S3IOpY 1 1 1 M 1 N in n ^vD "'^|""1||||||| CO OT \0 00 N m ^ 00 1-1 9 Eijpjnjj 1 „ „ co^ r^ CO 1 » 1 1 1 1 1 1 1 1 1 1 1 1 1 1 in - ? aU33S3IOpY 1 1 1 1 1 |.i-ioin"vOcococoN| 1 1 I"! 1 1 in covO m Tj-co ;j. 1-1 o m »-i 9 Bi^pjnj 1 1 1 " 1 " ""^ 1 " 1 1 1 1 1 1 1 1 1 1 1 1 1 o t^ il-OO g s ■M 2;q 1) p Om0ino>n0>n0>n0>noino>n0>n0in0ino Total Percentage Av. length, mm. E O O M i-* N n cocoThTh>ninvOvO t^t^ccco OC^O O ^ 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 inO tno mO »nO ino mO ino tno ino tno ino in o^o O M M N N coco-^-^min^o^o r^r^oooo ooo o W«|-,l-IMI-.M«MI-IWMMWWlHMI-IIHMrJN APPENDIX II IQI o o ? M O o CO auaossiopv 1 1 1 1 1 "" 1 """"*" ^ 1 1 1 1 1 1 1 1 ^ s;|: 00 9 Ei]iojnj 1 1 1 1 1 1 " 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 o 1— 1 oo juaosaiopY 1 " 1 1 ['-'■^uir^oa'^NMMi-.fo| 1 1 1 1 1 00 m O "^ CO 9 Eijpjnj 1 1 1 1 "" "1 1 1 1 1 1 1 M 1 1 1 1 1 1 1 ir, o •* 1-1 00 as, juaossiopv 1 1 1 1 1 1 1 I'l'fi^^t^ot^'^coi 1 1 1 1 1 O OO vO =o •* <^ irt M CO N 9 Bijioanj 1 1 1 1 " 1 " 1 1 1 1 1 1 1 1 1 1 M 1 1 1 1 juaosajopv 1 1 1 IMNvOrONCOlNCOWi-'l 1 I"! 1 1 1 O vO M O N in ;j. 11 O O O 9 Bfipjnj 1 « N ^^^ * 1 « 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ThoO 00 ON in M N 1 vO CO M N ^uaosaiopv 1 1 1 1 '-ihHT^MrOh-<^oco-^N'-'»-''-'| ►^l 1 1 ] CO 9 Bijioinj " 1 " ^" 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 osoo 9^ o N > r ro luaossiopv 1 1 1 1 1 lvO'+-^0m0t^0>''*0u^0'o0m0''^0i^0''^0»'^0 Total Percentage Av. length, mm. Total in sample 00'-'^NMronTi--i-in u^so vo r^ r^ go go ON O^ O O •-' 1 1 1 1 1 1 Ill Ill IOC irio ino loo ir, o m o m o u^o tno ino »nO "r» OO o ^ *-" f^ M f^f^^'i-inioovo r^i^coco c^o^o o 192 DISCOVERY REPORTS 2 g T^^£> t^oo N\D o^Noo r^r^Ovoco mco -too l>«" " 1 i-iN^or^NOM"Ot-^r^ThTj-wM 1 Grand total examined = 1165 Grand total in all nets = 17,501 M 1 o N ok aU33S3IOpV 1 1 1 1 INinrt-lot^ionHMNI ImM"| I 1 M OvOO ^ « «^ Tj- •1- co in 9 Efipanj 1 1 1 "-^^ " 1 1 " 1 1 1 1 1 1 1 1 1 1 1 1 1 N M N 0 « M N T in N o " 00 lusDsappv ■l 1 1 1 1 " " " " 1 " 1 1 1 1 " 1 1 1 1 1 1 1 r^ CO 00 IT) ^ M m 9 Bijiojnj 1 1 1 1 ""^" 1 1 1 1 1 1 M 1 1 1 1 1 1 1 ^ N M O o l-l N > t }U3DS9IOpV 1 1 1 1 1 1 " 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 M 0 I^ 22 ^ 9 Bijpjnj 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 M 1 1 1 0 juaosaiopv 1 1 1 1 1 1 " 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 10 M 0 r^ 0 M N 9 Eijiojnj 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 > N >-' 1 juaosaiopv 1 1 1 1 1 1 1 1 " 1 ^ 1 1 " 1 1 1 1 1 1 1 1 1 00 ^ r^oo M r-- 9 Bi|pjnj 1 1 "" 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 rl c^ 0 o s r juaosajopY 1 1 1 1 1 1 1 1 1 " 1 """ 1 1 M 1 1 1 1 1 00 ^ 0 p 2:? ■i- 9 Ei[i3jnj[ 1 1 1 1 1 M 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 juaosajopY 1 1 1 1 i l"l |"NM|f^Tt-l |"| 1 1 1 1 1/1 9 B-nDanj M ^ "" 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1-1 S 8 CO ? i^ (^ N o o Q 8 }ua3sa]opY 1 1 1 1 1 1 " 1 " " 1 " 1 1 "" 1 1 " 1 1 1 1 Ov 0 00 0 9 Eijpjnj 1 1 1 1 1 " 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 w 0 M luaosappY 1 1 1 1 |MMOfO\Of^^NrlNM| 1 1 1 1 1 1 roOO ^ 5- 9 Eijpjnj " 1 1 1 " ^' 1 1 " 1 1 1 1 1 1 M 1 1 1 1 1 1 in " N S e 4-. & Oiri0io0iri0irj0u-)0i^0^00io0»o0in0>^0 Total Percentage Av. length, mm. "E g « 0 O O - — N N r^r^^^u-t u-i\0^ l^r^oOOO O^OO O "-• 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 I00io0»00m0"^0i'^0i^0>00in0"^0>n0io O^OOKMNNrOM't-^io trt\0 vO I^ t^ 00 00 0> O O 0 MWMWMMWMMMMMMMl-lMMMI-IIHNN [Discovery Reports. Vol. XIV, pp. 193-324, December, 1936] THE SOUTHERN SPECIES OF THE GENUS EUPHAUSIA By D. DILWYN JOHN, M.Sc Assistant Keeper, Department of Zoology, British Museum of Natural History -^v CONTENTS Introduction page 195 Material 197 The surface waters of the southern oceans 200 The species as aduhs 203 Euphausia lucens 205 Euphausia vallentini 211 Euphausia frigida 214 Euphausia superba 216 Euphausia crystallorophias 220 Euphausia hanseni 223 Euphausia spinifera 225 Euphausia longirostris 228 Euphausia triacantha 230 Euphausia similis 233 Euphausia similis var. arniata 236 Euphausia similis var. crassirostris 238 The relations of the species to one another 239 Group a 240 Group d 244 The southern group 246 The later larval development of five species 253 Larval development 253 Post-larval development 260 Development of Euphausia frigida 264 Euphausia vallentini 271 Euphausia triacantha 277 Euphausia longirostris 285 Euphausia spinifera 294 References 304 Appendix I 307 Appendix II 319 THE SOUTHERN SPECIES OF THE GENUS EUPHAUSIA By D. DILWYN JOHN, M.sc. (Text-figures 1-40) INTRODUCTION THIS report is an account of the adult characters and distribution of ten species of Eiiphausia from southern waters, and of the development from the second Calyptopis stage upwards of five of those species. The surface waters of the southern oceans are divided from south to north into three well-defined zones of waters of different origins : the Antarctic, sub-Antarctic and sub- tropical Zones. They are separated from one another by sharp boundaries, because, along the line where two meet, the heavier water of one zone does not mix with the lighter water of the other but sinks below the surface. Most of the species of Eiiphausia described here are found only in one or another of these zones. The characters of the species, more particularly those of the copulatory organs of the males, show that five of the ten make one group and four another group. I think they show too that, if it be supposed that the genus Euphausia arose in tropical waters, the individuals of each group represent, in the order that they occur from north to south, successive steps towards the colonization of Antarctic waters by the genus along two distinct lines. The ten species of Euphausia whose adult characters are described are : E. crystallorophias, Holt and Tattersall E. triacantha. Holt and Tattersall E. superba, Dana E. longirostiis, Hansen E. frigida, Hansen E. spinifera, Sars E. vallentini, Stebbing E. hanseni, Zimmer E. hicens, Hansen E. similis, Sars The first seven species occur only in the Antarctic or sub-Antarctic Zones, except that E. lucens is found too in the coldest water of the subtropical Zone. E. similis is always present in all but the coldest water of the sub-Antarctic Zone and may occur throughout it ; but it has a wider distribution than any of the other nine species : not only is it found like E. lucens in the coldest water of the subtropical Zone as well as in the sub- Antarctic, but it has been recorded from the neighbourhood of the Philippines and from off the south coast of Japan. To describe these eight species would be to describe a geographical unit, for they are the only members of the genus occurring in the Antarctic and sub-Antarctic Zones. E. spinifera from the subtropical Zone and E. hatiseni from tropical waters are included because they are closely allied to E. longirostris and E. triacantha, and together with them form one of the groups mentioned above. All the species have been described before, but the descriptions are in scattered 15,6 DISCOVERY REPORTS publications ; of some species a later worker has added to or corrected the description of an earlier without completely redescribing it ; three varieties of E. similis have been described in three separate papers ; some of the characters of E. longirostris have been described in one paper some in another; E. hanseni has been described from female specimens only, from an unknown locality ; no previous worker has had a collection, such as the present one, large enough in numbers and wide enough in range to enable the limits of the distribution of most of the species to be recognized. For these reasons illustrated descriptions of the ten species are given here ; the figures are new, the de- scriptions based upon new material. The five species whose development from the second Calyptopis stage upwards is described are : ^ /r/;?/Wa E. longirostris E. valktitini E. spinifera E. triacantha loo Fig. I. Chart showing positions of the stations made in the Falkland Sector in the summer of 193 1-2. The position of the edge of the pack-ice is shown by a scalloped line, the loose pack-ice along the course of the cruise in the Weddell Sea by irregular hatching. MATERIAL 197 In each of these species there are a few FurciUa that occur in much larger numbers than the others; they are what were previously known as dominant forms. In each they have the structural relationship to one another that Fraser (1936, p. 35) has shown to be the essential one of dominant forms, proving them to be the members of a series. They confirm Eraser's view that this part of Euphausian development has an order that hitherto was unsuspected, and that it is therefore shorter than was previously supposed. MATERIAL The collection upon which this report is based is a small part of that made by the R.R.S. 'Discovery II' during her second commission, 1931-3. The stations made during that time are shown in Eigs. 1-3. They include two sets of closely placed stations in the Ealkland sector of the Antarctic, one made in each of the summer seasons 193 1-2 and 1932-3 (Figs, i and 2). A series of lines of stations crossing and re-crossing 70° 70° 60° 50° 40° 30° 20° Fig. 2. Chart showing positions of the stations made in the Falkland Sector in the summer of 1932-3. The southernmost stations were along the edge of the pack-ice. 150° West 180 East Fig. 3. Chart showing positions of the stations made on the circumpolar cruises and of some in the Falkland Sector. Those from South Georgia to South Africa (833-843) were made in February and March 1932. Those east-about from South Africa to South America were made between April and October 1932; the turning point of each, except the last (974), was at the edge of the ice. Those between South Georgia and the Falkland Islands (1083-1120) were made between December 1932 and February 1933; those between South Georgia (1137) and the ice-edge (i 153-4) and to South Africa (1170) in March 1933. MATERIAL 199 the Antarctic and sub-Antarctic and the coldest parts of the subtropical Zones, encircling the Antarctic continent east-about from South Africa to South America, was made in the winter months (April-October) of 1932. A line of stations from South Georgia to the ice-edge off the Antarctic continent near the meridian of Greenwich and from there to the Cape was made in March 1933 (Fig. 3). 10 20 30 46 50 60 Antarctic Zone Souivt - 40 6d 80° 70 60° 50° 40° 30° 20° 10 10° 20 30" Fig. 4. Chart showing the probable average positions of the Antarctic, subtropical and tropical convergences in the South Atlantic (from Hart, 1934, Discovery Reports, viii, p. 5). The southern limit of the two sets of stations in the Falkland sector was the edge of the pack-ice — which had different positions in the two seasons — with the exception that the stations in the Weddell Sea during the first season were made in loose pack-ice itself. Each of the southern turning points of the circumpolar cruises was at the edge of the ice fringing the continent except that to the south-west of South America, where the amount of fuel remaining on the cruise from New Zealand was not enough for the ship to go on to the ice-edge. Among the eleven hauls of plankton nets made at most of these stations were two oblique hauls, one from approximately 250 to 100 m., the other from approximately 100 m. to the surface, taken with a conical net with a mouth i m. in diameter and the greater portion of its fishing part of stramin. This is the N 100 fully described by Kemp and Hardy (1929, p. 184). The specimens of Euphausia in almost all those net ,oo DISCOVERY REPORTS hauls taken east-about from South Africa to South America and those from about half the net hauls made in the Falkland sector in each of the two seasons have been examined. Some of the larvae described were sought for and found in the hauls of a finer and smaller net, 70 cm. in diameter, fished vertically at the same stations, or in other collections; adult E. crystallorophias and E. hanseni are described from material collected at another time.^ 30 I 40° I SUB-TROPICAL CONVERGENCE 50° ANTARCTIC CONVERGENCE SOUTH IOOOm- EOODm 3000 M 4000m SUB-ANTARCTIC ZONE < . SUB-TROPICAL WATER ANTARCTIC INTERMEDIATE CURRENT ^MIXED- WATER REGION ANTARCTIC SURFACE CURRENT WARM DEEP CURRENT / ANTARCTIC BOTTOM CURRENT / Fig. 5. Vertical diagrammatic section showing the vertical and meridional ranges and movements of the water masses of the subtropical, sub-Antarctic and Antarctic Zones. THE SURFACE WATERS OF THE SOUTHERN OCEANS^ The surface of the southern oceans is divided from south to north into three well- defined zones of different origins : the Antarctic, sub-Antarctic and subtropical Zones (Fig. 4). Where the waters of two zones meet the heavier sinks below the surface and sets up a hydrological boundary or convergence. The vertical and meridional movements of the waters of these zones and of the deep currents below them are shown diagram- matically in Fig. 5. 1 The positions of all the stations are shown in Figs. 1-3. In charts showing the distribution of individual species the only stations included are those from which the Euphausians where examined. 2 This section was written with the help of Mr G. E. R. Deacon, hydrologist, who will shortly be publishing in the Discovery Reports an account of the hydrology of the Southern Oceans. THE SOUTHERN OCEANS 201 The surface water of the southernmost, the Antarctic, zone is formed near the Ant- arctic continent and is a mixture of fresh water from melting ice and snow with up- welling deep water from the north (a part of the warm deep current to be described later). It is cold and poorly saline. It moves with the prevailing winds — towards the west near the continent, towards the east farther north — but it has as well a northerly flow which is probably small compared with the east or west movement. At a latitude which is determined by the deep-water movements (see Deacon, 1937, pp. 21-24) the Antarctic water sinks below the surface to form a deep current. Here, in this same latitude, it meets at the surface the lighter water of the next zone, the sub-Antarctic, and it is below this that it sinks sharply. The line along which this happens is known as the Antarctic convergence. ^ The depth of the Antarctic surface layer varies from about 80 m. in the far south between the east and west currents to 250 m. near the Antarctic convergence. Where the sub-Antarctic surface water meets the warmer water of the subtropical Zone it sinks below the surface along a line known as the subtropical convergence. There are reasons for believing that there is a less definite boundary separating subtropical and tropical water farther north. The Antarctic convergence runs uninterrupted by land around the southern hemisphere between the latitudes of 50 and 60° S. The course of the subtropical convergence between 37 and 47° S is broken by South America and New Zealand. Sub-Antarctic surface water is formed from the northward flowing Antarctic water which has sunk below the surface at the Antarctic convergence and from subtropical water which enters the sub-Antarctic Zone from the north as a subsurface current. Its depth increases from south to north and it is about five times as thick as the Antarctic surface layer. Its strongest movement is towards the east, but it generally has a slow movement to the north as well. The subsurface stratum, which flows towards the south, is a mixture of more saline water from the subtropical Zone and of sub-Antarctic surface water itself that has sunk below the surface at the subtropical convergence. In the most southerly part of the sub-Antarctic Zone, for about 100 miles north of the Antarctic convergence, there is a region of intense vertical mixing where this warmer, more saline, water from the north is mixed with water from the surface of the Antarctic Zone. From this region the current known as the Antarctic intermediate current sinks and flows towards the north. Below the northward flowing Antarctic intermediate current in the sub-Antarctic Zone is the southward flowing warm deep current. It climbs rapidly towards the surface at the Antarctic convergence and flows south beneath the surface layer in the Antarctic Zone. In the Atlantic and Indian Ocean sectors it comes from water which has sunk below the surface in tropical and subtropical regions ; it is present in the Pacific Ocean '■ The flow of Antarctic water to the north is strongest in summer when ice and snow are melting in the south. When new ice is being formed in winter it is considerably reduced. Our observations near the Antarctic convergence in winter suggest that the slowing up of the current leads to a southern movement of sub-Antarctic water at the surface ; north of the convergence was purely sub- Antarctic water, but the Antarctic water south of it was mixed with sub-Antarctic water. At such times sub-Antarctic animals (e.g. E. vallentini, see p. 214) may be found south of the convergence. 202 DISCOVERY REPORTS but its origin is uncertain. Part of it upwells far south and mixes with the fresh water from melting ice and snow to form the Antarctic surface water; part of it becomes cooled but only slightly diluted and gives rise to the heavy Antarctic bottom water. Both waters have a northward movement. There is then in the Antarctic Zone a move- ment of cold water away from the pole at the surface and bottom and a movement of warmer water towards the pole at an intermediate depth. Subtropical surface water is water which has spent some time circulating in sub- tropical or tropical regions. Near the subtropical convergence it appears to have either a component of movement towards the south or a smaller northward movement than the sub-Antarctic surface water. There are generally sudden increases of temperature, and sometimes of salinity, as one crosses the convergences between the zones from south to north. The increase in temperature across the Antarctic convergence is 2 or 3° C: where the convergence is far north (50° S) it is from about i to 3-5° C. in winter and from about 3-5 to 6° C. in summer ; where the convergence is far south (60° S) the figures are o to 2-5° C. in winter, 2-5 to 5° C. in summer. In the Pacific Ocean and in the western part of the Atlantic Ocean there is a small increase of salinity across the convergence from south to north : from 33-8 or 33-9 %o to 34-1 or 34-2 °l^^; the few observations in the eastern part of the Atlantic Ocean and the Indian Ocean make it appear that there is no increase there. The position of the convergence appears to be fixed by the deep-water movements, and so far as is known it varies only within very small limits. The subtropical convergence is well defined in the Atlantic Ocean and the western part of the Indian Ocean. It is usually marked by a sudden increase of temperature of about 4° C, and of salinity of about 0-5 %o. It is not so well defined to the south of Australia and in the Tasman Sea. There have been very few observations in the Pacific Ocean, but the convergence appears to be well marked there except in the east. There have not been enough observations to give a full account of the temperature and salinity changes across the subtropical convergence, but it can be said that its position is approximately that of the 11-5° C. isotherm in winter, of the 14-5° C. isotherm in summer, and of the 34-9 °j^^ isohaline. It does not appear to be stationary everywhere like that of the Antarctic convergence. It seems to vary most in the Atlantic Ocean and, so far, a definite seasonal variation has been found only to the south of the Brazil current: there its average position is 3-5° farther south in summer than in winter. It has other variations which are not seasonal changes. Briefly the surface temperature limits of the Antarctic, sub-Antarctic and sub- tropical Zones are as follows: In winter (August and September) In summer (February and March) Antarctic water: Where the convergence is Where the convergence is Sub-Antarctic water Subtropical water in about 50° S in about 60° S Less than 2° C. Less than 1° C. 2-11-5° C. Greater than 11-5° C. Less than 4-5° C. Less than 3-5° C. 4-5-i4-5°C. Greater than I4'S° C. THE SPECIES AS ADULTS 203 THE SPECIES AS ADULTS The adults of the ten species described in this report are easily distinguished from one another by their structural characters. Both sexes of each may be recognized by the struc- < o o > or u < CD q: LU Q. D < I H z < < a: I- z h z LJ _) _J < > ID z o 10 z UJ u 3 (T) m LU UJ SUB-TROPICAL CONVERGENCE-) ANTARCTIC CONVERGENCE EDGE OF PACK-ICE ANTARCTIC COAST- LINE u > o in N u (- u Fig. 6. Diagram showing the distribution of species of Euphausia in the surface waters of the Antarctic and sub-Antarctic Zones ; the blacked in portion of each column shows the normal range of that species, the entire column the possible range. ture of the antennular peduncle, the shape of the rostrum and front part of the carapace, and the presence or absence of abdominal spines or processes ; the males may be dis- tinguished by the structure of the copulatory organ on the first pleopod. They can also be separated when fresh by one who has seen and examined many of them by their general appearance — that sum of all the small characteristics of shape and colouring 2^^ DISCOVERY REPORTS which it is almost impossible to describe.^ Some of the large species possess charac- teristic structures and rich colourings that make them easy to recognize with the naked eye- E superba may be recognized by its densely plumose thoracic legs and strong red colouring, E. triacantha by its brilliant red and orange and its three abdominal spines. But the smaller species without such obvious structures or such conspicuous markings may be as readily distinguished from one another by their general appearance when many have been seen and handled. The majority of the specimens that form the basis of a part of this report were identified at sea the day after they were taken. They were examined under a dissecting microscope and a large number of them were sexed and measured. The data so obtained will, it is hoped, be used for the preparation of another paper on the life histories of the '^ Kght of the ten species described here are the only Euphausia found in the Antarctic and sub-Antarctic Zones. Each of them, with the exception of E. triacantha, is normally confined not only to one of the two zones but to a part of one zone. But most of them may occur throughout the zone of which they normally occupy only a part, and some may be found even outside the zone to which they are usually confined. The more abnormal occurrences can mostly be explained by water movements that are local and peculiar for reasons of place or time, as the many examples described later in this report ' The normal and the possible range of latitudinal distribution of each of the species is shown diagrammatically in Fig. 6. , • • u The figure shows that the usual number of Euphausia to be found at one point in the sub-Antarctic is two, three or four. In the warmer sub-Antarctic E. shmlis, E. lucens and E lorigirostris occur. A little farther south E. vallentini is added to the number. Farther still E. lucens and E. similis cease to occur and E. triacantha appears. E longirostris is sometimes absent from the coldest water leaving only E. vallentim and E. triacantha. , ^ . ^, , In the ice-free part of the Antarctic Zone there are usually only E. triacantha and E frigida though in some large areas E. superba occurs too, widely distributed as far north as the convergence. Along and among the ice itself only E. superba is ound Farthest south of all is E. crystallorophias, a neritic species confined to the coast-line ot the Antarctic continent. The warmer sub-Antarctic species may be brought as far south as the Antarctic con- vergence and E. frigida and E. superba may be carried a short way north of the con- vergence'from the Antarctic. In this way it is possible that seven of the eight species may occur in one place, immediately north of the Antarctic convergence, and taken in one net This is indicated by the section AB in the figure. The largest number taken at any of the 200 stations shown on the distribution charts in this report was five; that number occurred at six stations all immediately north of the Antarctic convergence. 1 I have not seen E. hanseni nor many specimens of E. crystallorophias in a fresh condition. EUPHAUSIA LUC ENS 205 They were as follows: At five stations : At one station : E. vallentini E. vallentini E. triacantha E. triacantha E. longirostris E. similis E. similis E. lucens E. frigida E. frigida The complicated copulatory organs of the males of the different species show constant and unmistakable differences and form the most reliable criterion for separating them. In the accounts of the organs that follow under each species more details are given than are necessary for their recognition, and than would be seen under the low power of a binocular microscope — indeed the figures alone are sufficient for recognition. The details have been given because they are necessary for the comparative account of the copulatory organs which follows (pp. 244-252) ; in it the ten species are divided into two groups, and it is suggested that the male copulatory organs show that the members of each group are related to one another as the members of an evolutionary series. Hansen (1910, p. 79) says of the copulatory organs that "in all species (oi Euphausm) hitherto examined by me a spine-shaped process on the inner lobe and additional processes on the median lobe are totally wanting". But in a later paper (1912, p. 230) he records seeing the spine-shaped process in one specimen oi E. lucens and says that the additional process is present as a small spine in E. miicronata and as a mere rudiment in E.gibboides. Zimmer (1913, p. 117, pi. xiii, fig. 72, />.«.) says that he found in the speci- mens of E. siiperba he examined indications of the additional process on the median lobe in the form of a tiny spine. I have seen a very small spine-shaped process on the inner lobe in two or three specimens of E. siiperba, but not in E. lucens nor in any of the other species. I find a small additional process on the median lobe to be present in every one of the ten species except E. lucens ; in one specimen of E. superba there were two additional processes on the median lobe. Euphausia lucens, Hansen (Figs. 7-1 1 and 30 a) E. lucens, Hansen, 1905, p. 9; 1911, p. 26, fig. 8; Tattersall, 1913, p. 876; Hansen, 1915, p. 84; Zimmer, 1915, pp. 178-9; Colosi, 1917, p. 183, pi. xiv, figs. 6-8; Tattersall, 1924, p. 19; Illig. i930> P- 499; Tattersall, 1925, p. 6; Hardy and Gunther, 1935, p. 208. E. splendeiis (part), Sars, 1885, p. 80, pi. xiii, figs. 7-17. E. iincinata, Colosi, 1917, p. 186, pi. xiv, figs. 9-10. Description. The front margins of the carapace are very faintly convex behind each eye; the rostrum is a small broad triangle (Fig. 7). In the antennule the lobe from the end of the first segment of the peduncle is triangular and variable in size ; it may be very short, a small wide triangle, or it may be a long thin triangle one-third as long as the second segment of the peduncle. It is easily seen from the side. The dorsal keel of the third segment is strong ; as seen from the side it appears to rise a little less than half-way along the segment ; its upper margin is nearly straight or slightly curved. Fig. 7. E. liicens. a, front part of carapace and first segment of antennules from above, x 18. b, left antennular peduncle from the side, x 18. TERMINAL PROCESS LATERAL PROCESS MEDIAN LOBE INNER LOBE Fig. 8. E. Iticeus, inner and median lobe of the left copulatory organ of the male, from behind, x 82. EUPHAUSIA LUC ENS 207 The abdominal segments have no dorsal spines. The copulatory organ of the male is very similar to that of E. vallentini and E.frigida but easy to distinguish from either of them (Figs. 8 and 30 a). It differs from both in that the terminal process is considerably longer than the proximal : in E. vallentini and E. frigida it is shorter. The end of the terminal process is bifid, the inner of the two branches longer and stronger than the outer. The end of the proximal process carries on its hindermost and on its forward side a membranous expansion ; the latter is smaller than the former and is set a little farther back on the process (i.e. nearer its base) and does not extend so far forward ; it has a striated appearance which the hindermost ex- pansion has not. On the inner side of the distal end of the process, between the mem- branous expansions, there is a small tooth. Before the membranous expansions there springs from the outer side of the process a strong secondary process or spine which appears blade-like from some angles because one edge is more heavily chitinized than the other. The lateral process is strongly and sharply curved at the end and carries on that curve a strong tooth of a characteristic shape. Males may be sexually mature at the length of 11 mm., females at 12 mm. — that is, specimens of these lengths have been seen carrying ripe spermatophores. Both sexes may, exceptionally, reach the length of 18 mm. The vast majority of the specimens found were between 10 and 15 mm. long. E. hicens is the smallest of the species described in this paper. None of its larval stages has been described. Remarks. I cannot agree with Hansen (191 1, pp. 14-15) that the two specimens described by Illig (1908 a, pp. 54-5, figs, i, 2) as Thysanopoda megalops, sp.nov., cancelled by him later in the same year (Illig, 1908 b, p. 463), were possibly E. lucetis. Both were 20 mm. long, longer than any E. lucens I have seen ; they had, according to Illig, no denticles on the under edge of the carapace and no lappet on the basal segment of the antennular peduncle ; they had a median spine above the base of the eye-stalks ; and they differed in other ways from E. hicens. Colosi (1917, pp. 186-7, pl- xiv, figs. 9 and 10) described the new species E. uncinata from one male found among seventy specimens of E. lucens. It differed from E. hicens in only one way: the median lobe of the copulatory organ of the male had at the base of the lateral process an additional and smaller process, uncinate in shape with a strong secondary tooth near its apex. Tattersall (1924, p. 20) thought it reasonable to regard this as an abnormal male of E. lucens. I am satisfied that it is of this species and that what Colosi saw and drew as an additional process was not that but the lateral pro- cess of the succeeding moult, displaced perhaps in fixation. I have seen the same thing in E. lucens itself (Fig. 90) and in E. spinifera (Fig. 286 i), and there was no doubt that the additional process was within the tissues of the median lobe ; its apex lay below or within the base of the lateral process. The appearance does not, how- ever, seem to be a common one in the species of Euphausia described in this report. I have noticed it often in Thysanoessa vidua, in which it is not uncommon to see the rudiments of the three larger processes, the terminal, the proximal and the lateral, 2o8 DISCOVERY REPORTS protruding for a longer or a shorter way up into these processes ; one such is shown in Fig. 9 b. In his report on a collection of Schizopoda from the tropical Pacific, which contained no E. lucens, Hansen (1912, p. 230) in discussing the male copulatory organ of the genus Eiiphaiisio says: "The spine-shaped process is wanting (yet I found this process de- Fig. 9. a, E. liicem, median lobe of the left copulatory organ of a male from in front, x 67. b, Thysanoessa vicina, terminal proximal and lateral processes of the left copulatory organ of a male from behind, x 266. Both figures show the rudiments of the processes of the succeeding moult at the bases of the existing processes. veloped in the normal way in one of the specimens examined of E. lucens)." I have found no reference to this in any other work by Hansen, nor have I ever seen the spine- shaped process in E. lucens. Distribution. The stations at which E. lucens was taken are shown in Figs. 10 and 1 1 . It occurs constantly in all but the coldest water of the sub-Antarctic Zone, though it may occur in that too as far south as the Antarctic convergence. It has never been. taken south of the convergence. It occurs also in a narrow belt of the coldest water of the sub- tropical Zone just north of the subtropical convergence. Of the stations at which it was taken the three south-east of the Cape had the highest surface temperatures, that of the first of them, 20-1° C, being the highest. This is warmer water than they are usually found in, and their presence in it is accounted for by the shape of the subtropical convergence in this region. To the west of the Cape it lies in 37° S, but to the south of it the Agulhas Current causes it to bend sharply southwards. Its position varies more here than elsewhere (except perhaps south of the Brazil Current), and a mixing of sub-Antarctic and subtropical waters takes place throughout a large area. But the area would not include the northernmost two of the three stations at which E. lucens occurred. Although they were far from the convergence to the south they were EUPHAUSIA LUCENS 209 not so distant from it, and the waters of lower temperature in which E. lucens usually occurs, to the west. Apart from this exceptional area the highest surface temperatures of the stations at 70 60° 50'^ 4-0° 30° 60 0, "3 \ \ f-^ 1 1 A' /?"^ r9J r \ / 0 / . / « >&^/ — ^ — ~— -^ / * - " 1 ''^': / ^^^.°'^ '" " / * - / ^^^ »~ ' / r 1 "^ / ^\ / * ' ^v ^ " / ~ ^ / " ^ ' 1 -i^j" / / / • \ * ^ / / / ' * / / \ // \ ^ \ / \ ■ £ LUCENS . A -V if FRIGID^ D PACK 1 CE rrYTrnnrTVrY- PACK AND DRIFT ICE---;,y- / tr" — ~" ~*~"~^ 100° 90° 80° 70' 60° SO" 40° 30° 20" 10° O' 10 • Fig. 10. Chart showing the occurrence of £. lucetis and £". frigida at the stations made in the Falkland Sector in the summer of 1931-2. The thick Hne represents the Antarctic convergence. which it was found were from 12 to 14° C, except for the nearest of those to the south- west of Western Australia where it was i6-i2° C. The lowest surface temperature at which it was found, 5-02° C. at a station just north of the Antarctic convergence to the east of the Falkland Islands, is in another exceptional area; other warmer sub- Antarctic forms, e.g. E. similis, are found farther to the south here than elsewhere. Their presence is due to the unusually strong southward move- ment of sub-Antarctic water in the Falkland region. Elsewhere the lowest temperatures at which E. lucens was found were 6-5-7° C. DISCOVERY REPORTS oO° 150° West 180 East 150 Fig. II. Chart showing the occurrence of E. lucens and E.frigida at the stations of the circumpolar cruises made in February to October 1932, and in the Falkland Sector in the summer of 1932-3. The inner of the two thick lines represents the Antarctic convergence, the outer the sub-Antarctic convergence. EUPHAUSIA VALLENTINI Euphausia vallentini, Stebbing (Figs. 12-14, 30 i) E. vallentini, Stebbing, 1900, p. 545, pi. xxxvii; Holt and Tattersall, 1906, p. 3; Tattersall, 1908, p. 13, pi. iv, figs. 4-6; Hansen, 191 1, p. 30; 1913, p. 32, pi. v, figs, i a-f; Zimmer, 1914, p. 427; 1915, p. 178; Tattersall, 1924, p. 21 ; Illig, 1930, p. 499; Mackintosh, 1934, p. 76 et seqq.; Hardy and Gunther, 1935, pp. 217-19, fig. 94. E. splendens (part), Sars, 1885, p. 80, pi. xiii, figs. 7-17. E. patachonica, Colosi, 1917, p. 187, pi. xiv, figs. 11-14, P'- ^v- ^gs- 15-20. Description. The carapace is not pro- duced far forwards dorsally. The rostrum is a short sharp triangle about as long as or a little longer than broad; it is much more distinct than that of E. liicens. The anterior margin of the carapace is faintly convex behind each eye (Fig. 12 a). The gastric area of the carapace is moderately or sharply convex. The first segment of the antennular peduncle has a very characteristic lobe: it is large, broad and rounded and projects horizontally above the second segment (Fig. 12 a, b). In larger specimens, but not in smaller, there is a small rounded or tooth-like vertical projection on the upper surface of the second segment at the inner distal corner, and a smaller tooth-like pro- jection near the outer distal corner. There is a high dorsal keel on the third segment ; the part seen from the side is strongly rounded above and nearly vertical in front. There is a short thin mid-dorsal process from the posterior margin of the third abdominal segment (Fig. 12 f); rarely it is pig. 12. E. vallentini. a, front part of carapace and absent. first segment of antennules from above, x 12. b, left The COpulatory organ of the male is very antennular peduncle from the side, X15. c, median similar to that oiE. lucem and E.frigida but ^"'^""^ P™"'' °^ ''^'''^ abdominal segment, x 37. easy to distinguish from them (Fig. 30 h). The proximal process is considerably longer and heavier than the terminal ; its membranous expansions differ in appearance like those of E. hicens, the foremost having striations not possessed by the hindermost, but they are less unequal in size, and the foremost reaches more nearly as far forward as the hindermost; there is a small tooth on the outer side of the distal end of the process between the expansions as in E. lucens. The secondary process or spine on the proximal 212 DISCOVERY REPORTS process is wider, more blade-like, than in E. liicens. The terminal process is bifid at the end, the inner of the two branches stronger than the outer and either a little longer than it or of the same length ; the inner branch may have on its inner side a small blunt projection or tooth. The end of the lateral process is strongly curved but more widely Fig. 13. Chart showing the occurrence of E. valleiitini and E. superba at the stations made in the Falkland Sector in the summer of 1931-2. The thick line represents the Antarctic convergence. so than that of E. lucens ; it carries on the curve a strong tooth which points in the same direction as the end of the process. In the axil of that tooth on the hinder side there is always a tiny projection. Over 1800 specimens, apart from larval forms, of E. valleiitini have been measured. The largest male was 25 mm. long, the largest female 28 mm., but the vast majority of the specimens were less than 24 mm. in length. One male so small as 13 mm. and several females of 15 mm. were seen carrying spermatophores, i.e. were sexually mature at those lengths, but the usual size for sexually mature individuals is greater. EUPHAUSIA VALLENTINI 213 30° 90° 120 150° West 180 East ISO Fig. 14. Chart showing the occurrence of E. vallentini and E. superba at the stations of the circumpolar cruises made in February to October 1932, and in the Falkland Sector in the summer of 1932-3. The inner of the two thick hues represents the Antarctic convergence, the outer the sub-Antarctic convergence. 214 DISCOVERY REPORTS The earlier larval stages of E. vallentini are not known; those from the second Calyptopis upwards are described later in this paper. Distribution. The stations at which E. vallentini was taken are shown in Figs. 13 and 14. It occurs throughout the sub- Antarctic Zone except for a narrow belt of the warmest water to the north. The highest surface temperature of any of the stations at which it was taken was 14-2° C, at the station south-west of the Cape on March i, i.e. in summer. The stations at which it occurred in the Indian Ocean, Australian, and Pacific Ocean sectors were made in winter, and the surface temperature at the northern- most of them was lower, usually about 8° C. It was found a little way south of the Antarctic convergence in winter only. The flow of Antarctic surface water to the north, strongest in summer when snow and ice are melting in the south, is considerably reduced in winter when new ice is being formed near the Antarctic coasts. Our observations near the convergence in winter suggest that the slowing up of the current leads to a southern movement of sub-Antarctic water at the surface; north of the convergence we found purely sub-Antarctic water, but the Antarctic water south of it was mixed with sub-Antarctic water. In this way E. vallentini becomes carried into the Antarctic Zone. The numbers taken there were small compared with those from the nearest sub-Antarctic stations. The lowest surface temperature at which the species occurred was 0-74° C. at the southernmost station at which it was found on the line south-east of New Zealand in early September ; there was only one specimen. Euphausia frigida, Hansen (Figs. 10, 11, 15 and 30 c) E. frigida, Hansen, 191 1, p. 27, fig. 9; 1913, p. 30, pi. iii, figs. \a and l\ pi. iv, figs, i a-d\ Zimmer, 1914, p. 427; Tattersall, 1924, p. 20; Rustad, 1930, pp. 33, 43, 46-54, figs. 27-34; Ruud, 1932, pp. 52-4; Rustad, 1934, pp. 13-18, i^et seqq.\ Mackintosh, 1934, p. j6 etseqq.; Hardy and Gunther, 1935, many references. Euphausia sp., Tattersall, 1908, p. 14. E. splendens, Caiman, 1901, p. 23. E. crystallorophias (part), Illig, 1930, p. 500, fig. 182. Not E. frigida, Illig, 1930, p. 498. Description. The front margins of the carapace are faintly convex, the rostrum is very short and triangular (Fig. 15a). A small triangular or rounded lobe projects forwards from the inner distal corner of the first segment of the antennular peduncle ; it is difficult to see from the side. The third segment has a strong dorsal keel with, as seen from the side, straight upper and distal margins which meet nearly at a right-angle (Fig. 15 b). The abdominal segments have no dorsal spines. The male copulatory organ is very like that of E. vallentini and E. hicefis, but there are differences which make it easy to distinguish it (Fig. 30 c). It differs from that of E. lucens and resembles that of E. vallentijii in that the proximal process is longer than the terminal; it resembles both species in that the proximal process carries a blade-like secondary process or spine and in that it has two membranous expansions distally which EUPHAUSIA FRIGID A 215 differ from one another, the foremost having striations which are not present in the hindermost. The foremost expansion reaches a httle farther forward (i.e. distally) than the hindermost. Between them, on the outer side of the distal end of the proximal process, there is a conspicuous tooth-like projection easy to see with a low magnification ; it is in the position occupied by a very small tooth, which is often difficult to see under a high magnification if the membranous lobes are not so placed as to show it in profile, in E. lucens and E. valletitint. The terminal process is bifid at the extremity. The inner branch may be equal in length to or slightly longer or shorter than the outer; it is divided distally into two unequal lobes, the hinder shorter and smaller than the fore- most. The outer branch is narrower than the inner, triangular or lanceolate in shape. Fig. 15. E. frigida. a, front part of carapace and first segment of antennules from above, x 12. b, left antennular peduncle from the side, x 18. and its hinder margin is finely serrate (Fig. 30 c 2). The lateral process is sharply curved at the end and carries on the curve a strong tooth ; in the axil of that tooth on the hinder side there is a small or very small projection. Over 4200 specimens of E. frigida, apart from larval stages, have been measured. The largest males were 21 mm. long, the largest females 24 mm. In September and October some males and females only 1 1 mm. long, and many slightly larger, were found carrying ripe spermatophores, i.e. were sexually mature. The generation born the preceding season become sexually mature at this time before they are physically mature ; they grow to full size a month or two later. The larval stages of E. frigida from the second Calyptopis upwards are described later in this paper. Remarks. I am sure that Illig (1930, p. 498) is wrong in his only record of E. frigida, a female from subtropical water off the west coast of South Africa. He records none from the stations made by the ' Valdivia ' in Antarctic water where one would expect that the tow-nets could not fail to take some. E. frigida was in fact obtained there. I was doubtful of Illig's records of E. crystallorophias (1930, p. 500, fig. 182) from, among other localities, deep water far from land in the Antarctic. Through the kindness of 2i6 DISCOVERY REPORTS Dr A. Schellenberg of the Zoologisches Museum, Berlin, I was allowed to examine the animals identified as such from St. 135 east-north-east of Bouvet Island, and from St. 145 north of Enderby Land. They were not E. crystallorophias but E. frigida; Illig's figure of the copulatory organ is certainly not that of E. crystallorophias: it is something like the organ of E. frigida. Distribution. The stations at which E. frigida were taken are shown in Figs. 10 and II. It occurs throughout the Antarctic Zone from the convergence in the north to the edge of the pack-ice in the south. Our stations in the drift- and pack-ice of the Weddell Sea show that it is not found among and under the ice — except that it occurred at the two northernmost stations, east of the South Sandwich Islands. At these two stations the surface and deep waters were warmer than at the other stations made near them ; they were not so directly in the path of the Weddell Sea Current but were in eddies of warmer water.^ In such latitudes in summer the loosening pack-ice may be carried over water already containing E. frigida. The species occurred, but in much smaller numbers than in the Antarctic Zone, at a few sub-Antarctic stations just north of the convergence ; one of them was to the south- east of New Zealand, the others to the south-west of Cape Horn and to the east of the Falklands. They are carried over with the small quantities of Antarctic water that may cross the convergence anywhere at the surface, more particularly in summer; or in a deeper layer by the Antarctic water that sinks below the sub-Antarctic, from which they regain the surface. These water movements across and beneath the convergence are par- ticularly strong in that region, to the east of the Falklands, where E. frigida has been found most frequently to the north of its usual habitat. Euphausia superba, Dana (Figs. 13, 14, 16, 30 J) E. superba, Dana, 1852, p. 654, pi. Ixiii, figs, i a-o\ Sars, 18S5, p. 84, pi. xiv, figs. 5-9; Holt and Tattersall, 1906, p. 2; Coutiere, 1906, p. 8; Tattersall, 1908, p. 4, pi. i, figs. 1-12; Hansen, 1908, pp. 3, 7, pi. i, figs. 4a-;«; Zimmer, 1912, pp. 65-128, pis. viii-xiv, text-figs. 1-5 ; Hansen, 191 3, p. 27, pi. iv, figs. 2 «-^; Tattersall, 1913, p. 875; Zimmer, 1914, p. 424; Hansen, 1915, p. 79; Tattersall, 1918, p. 6; 1924, p. 18; lUig, 1930, p. 497; Rustad, 1930, p. 39, figs. 20-26; Ruud, 1932, pp. 20-51, figs. 5-17; Rustad, 1934, pp. n, 34 et seqq.; Mackintosh, 1934, p. 76 et seqq.; Hardy and Gunther, 1935, many references. E. murrayi, Sars, 1885, p. 82, pi. xiv, figs. 1-4. E. antarctica, Sars, 1885, p. 86, pi. xv, figs. 1-8. E. glacialis, Hodgson, 1902, p. 236, pi. xxx, figs. 1-8. E. australis, Hodgson, 1902, p. 238, pi. xxx, fig. 9. Description. This species is the giant of the genus and may be distinguished from all others by the long terminal segment of the mandibular palp which is at least seven times as long as broad, whereas in all others it is short and stout, about three times as long as broad. The setae of the thoracic limbs are much longer than in any other species of the -genus. The carapace has a distinct cervical groove, and the part in front of it is faintly convex 1 A single specimen of E. triacantha was found at each (p. 232). EUPHAUSIA SUPERB A 217 and keeled in the mid-dorsal line (Fig. 16 a). The front part is produced at the antero- lateral corners into strong projections one behind each eye. In old heavily chitinized males the projections are softened or disappear as such, because their margins have become turned upwards and backwards. The rostrum is short and triangular and may be a little shorter in the male than in the female. In old heavily chitinized males, and much more rarely in females, the lateral denticles of the carapace are reduced or wanting. b a Fig. 16. E. siiperbti. a, front part of carapace and antennular peduncles of female from above, x 10. 6, left antennular peduncle of male from above, xii. The antennular peduncles differ a little in the two sexes, but they are very variable in both, as a comparison of the accounts and drawings of Sars (1885), Tattersall (1908), Hansen (19 13) and Rustad (1930) shows. My description of them is based upon an examination of fresh material as well as upon these earlier accounts. In both sexes the characteristic structures are stronger in young specimens than in old. The antennular peduncle of the male (Fig. 16 b) is somewhat stouter than that of the female (Fig. 16 «). The lobe of the first segment is wide and strongly vaulted; it is usually a little narrower than the second segment, but it may be as wide ; its length varies from less than one-quarter to one-half the length of the second segment. Its shape is variable, but it is mostly longer on the inner side than the outer so that its distal margin is oblique ; it is usually emarginate too. There may or may not be a short strong spine or a blunter tooth-like projection at the outer distal corner of the first segment of the 2i8 DISCOVERY REPORTS peduncle. A wide lobe projects forwards from the upper surface of the second segment over the third. It may be as wide as the third segment and nearly half as long, its outer distal angle may be strongly produced forwards, outwards and downwards ; a keel which runs the length of the inner side of the upper surface of the segment turns outwards distally to end where the lobe bulges outwards above the median line of the third seg- ment. The lobe may in older specimens be much smaller and shorter, its distal margin straight, or it may be entirely wanting. The third segment has a high keel which may be wanting in old heavily chitinized specimens. In the female the lobe from the end of the first segment of the antennular peduncle is usually bigger than in the male; it may be wider or narrower than, and anything from over one-half to less than one-third as long as, the second segment. As in the male its shape is variable, but it is usually much longer on the inner side than on the outer, so that its distal margin is oblique, and it is usually concave or emarginate. There may or may not be a spine or projection at the outer distal corner of the first segment as in the male. The lobe from the end of the second segment is usually bigger than in the male, its outer distal corner more strongly produced; but it is variable. It varies from being (most often) wider than the third segment to being narrower, and from being more than one-half to one-quarter of its length. The outer distal corner is usually strongly produced forwards, outwards and downwards so that it sometimes reaches nearly as far forward, on the outer side, as the end of the third segment. There are keels on the second and third segments as in the male. The lobe of the second segment and the keel of the third do not appear to disappear in old heavily chitinized females as they do in males. The abdominal segments have no dorsal spines. The male copulatory organ is as distinct from that of E. crystallorophias and those of E. frigida, E. vallentini and E. lucens as the former is from the three latter, but it is nevertheless of the same pattern as the four of them (Fig. 30 d). The proximal part of the terminal process is bent nearly at a right angle to the distal part, which is about three times as long. The basal part, or foot, is bigger than in E. frigida, E. vallevtini and E. lucens, but the heel is not so sharp. The end of the process tapers and curls forwards; it is not bifid. The distal two-fifths of the proximal process is bent inwards and carries two mem- branous expansions, one distally, the greater part of it lying on the hinder side of the axis of the process, and one on the foremost side immediately near the bend. The distal end of the latter touches the former. The foremost expansion is striated, that which is distal and for the most part hindermost is not. There is at first sight no secondary process such as occurs in the three species already described ; none had previously been seen. But it is present in the majority of males, though not in all, as a spine bent for- wards over the foremost expansion. It may be as long as that shown in the figure or very much smaller, one-third as long or less, or it may be entirely wanting : of thirty heavily chitinized males examined it was present in the left petasmas of twenty, absent from those of ten. The lateral process is curved at the end; it carries no tooth or secondary process. EUPHAUSIA SUPERB A 219 On the inner and front margin of the median lobe, beyond the end of the lateral process, there is always a small additional process ; in one specimen there were two additional processes. I have seen a very small spine-shaped process on the inner lobe in one or two speci- mens of E. siiperba (cf. p. 205). Remarks. This species, the largest, most interesting and most important of the genus, has been studied more closely by other members of the Discovery staff than by me. Hardy and Gunther (1935, pp. 208-15) have written of its great importance in the ecology of Antarctic seas and more particularly of its economic importance as the only food of Blue and Fin whales in the south ; of its peculiar habit of swarming in dense shoals so that it becomes accessible to the whales in numbers large enough to satisfy them ; and of its occurrence as larvae and adults in different seasons on the South Georgia whaling grounds. Eraser (1936) has published a very full description of all the larval stages — more complete, and based upon far larger numbers of specimens, than any previous account of the development of a Euphausiid. Dr H. E. Bargmann will shortly be publishing a description of the internal anatomy of the species and an account, based upon the enormous collection of specimens now available, of its life history beyond the larval stages and of its distribution throughout the Antarctic Zone. Dr Bargmann has already examined a very large number and I am indebted to her for the following facts : females are usually a little larger than males but both may reach a length of 60 mm. ; the smallest female with spermatophores that she has seen was 32 mm. long, the smallest ripe male 41 mm. Distribution. I shall not attempt to give a full account of the distribution of E. siiperba here : Dr Bargmann will give a better one, based on far larger collections than I am considering, in her forthcoming report (see above). The stations at which it was taken in the commission of 193 1-3 are shown in Figs. 13 and 14. It was found on every one of our visits to the ice-edge, whether in summer or winter, and at the majority of the stations made far into the drift- and pack-ice of the Weddell Sea in January, including the southernmost of them. In the Falkland Sector, where our observations were for the most part made in summer, E. superba was found scattered throughout the Antarctic Zone nearly as far north as the convergence. It may be similarly distributed in other parts of the Antarctic in summer ; we did not find it to be so in winter on our circumpolar cruises east-about from S. Africa to S. America (Fig. 14). To the south-east of Africa, approaching Enderby Land, small numbers of larvae and adults were found at Sts. 851-4, far away from the ice-edge ; large numbers of adults and very large numbers of larvae were found at the ice-edge itself (St. 855). This was in April. At and near the ice-edge to the south of Australia, in May (Sts. 887-8), a few adults and many larvae were taken. At the ice- edge to the south of the sea between Australia and New Zealand, in June, many larvae but no adults were found (St. 912). The next catches were made in September at the ice-edge to the south-east of New Zealand ; there were larger numbers of larvae and adolescents but no adults (Sts. 952-5). 4-2 220 DISCOVERY REPORTS Fraser (1936, pp. 137-143) has examined and discussed the larval and adolescent forms taken at these successive visits to the ice-edge in winter, and writes of them: " In April Calyptopis 3 and early Furcilia stages are most abundant, in May Calyptopis stages have disappeared and Furcilia 2 predominates, in June it is Furcilia 6 (the last Furcilia), and in September adolescent forms, which have the chief place." Apart from the stations approaching Enderby Land it was at only one place far away from the ice-edge that E. superba was found: to the south-east of Kerguelen (Sts. 861-2). One adult and large numbers of larvae were taken there in late April. Our hydrological observations showed that at both of these stations there was colder Antarctic water de- flected from the south by the submarine ridge that connects Gaussberg on the mainland with Kerguelen. Other typical ice-frequenting planktonic animals were found at these two stations, and not at the stations before and after them. E. superba is sometimes found in small numbers to the north of the convergence east of the Falklands ; a strong tongue of Weddell Sea water flows north to this region. Fraser (1936, pp. 109-165) has discussed the distribution of is. superba of different ages, in terms of depth, latitude and nearness to ice, taking the records given above into account. In material collected at other times he has found the Metanauplius almost as far north as the convergence in the Falkland Sector and therefore considers that the adults spawn away from the ice. He shows that the Metanauplius to early Furcilia stages have a deep habit, spending most of their time in the southward-flowing warm deep current (see p. 201 of this paper), and he finds in this a reason for the concentration of young and adolescent E. superba near the ice-edge. " The habitat of the late Furcilia and early adolescents is at the surface and predominantly at the ice-edge. They will spread northwards in the northerly flowing surface water (see p. 201 of this paper) with the breaking up and drifting away of the field ice in the spring and summer" (p. 165). As they are carried northwards they will grow and become adult ; they will spawn and the deep-living younger larval stages will be carried south. So it is supposed that, during its life-history, E. superba makes a vertical cycle, along an axis that is approxi- mately longitudinal, in the Antarctic Zone. Euphausia crystallorophias. Holt and Tattersall (Figs. 17 and 30 e) E. crystallorophias. Holt and Tattersall, 1906, p. 3 ; Tattersall, 190S, pp. 9-12, pi. ii, figs, i-io, pi. iv, fig. 10; Hansen, 1908, p. 6, pi. i, figs. 3 a-c; Zimmer, 1914, p. 426, pi. xxvi, figs. 60-4; Tattersall, 1918, p. 6; 1924, p. 19; Mackintosh, 1934, p. -jbetseqq. ; Hardy and Gunther, 1935, p. 218. Not E. crystallorophias, Illig, 1930, p. 500, fig. 182. ? Euphausia sp. (calyptopis larvae), Rustad, 1934, p. 25. Description. The carapace has a right-angled expansion or projection behind the upper part of each eye (Fig. 17 a). The margins between these projections and the rostrum are concave; the rostrum is long and sharp and reaches as far forward as the front of the eyes. The carapace has a gastro-hepatic groove from which a nearly hori- zontal keel runs forward mid-dorsally to and along the rostrum. EUPHAUSIA CRYSTALLOROPHIAS 221 There is no dorsal lobe at the distal end of the first segment of the antennular peduncle as in the other species described; there is a short sharp spine, difficult to distinguish among the setae, on the outer distal corner (Fig. 17 a, b). The dorsal keel of the third segment is well developed and of the same shape as that in E. frigida but per- haps a little lower. The abdominal segments have no dorsal spines. The male copulatory organ shows many diff'erences from those of the four species already described, but it is on the same pattern (Fig. 30 e). In E. liicens, E. vallentmi and E. frigida the terminal process has a wide and somewhat foot-like base, the toe forming Fig. 17. E. crystallorophias. a, front part of carapace and first segments of antennules from above, x 12. b, left antennular peduncle from the side, x 14. the point of attachment, the heel being free on the outer side ; in E. siiperba the basal part of the process is bent nearly at a right angle to the distal part. The terminal process of E. crystallorophias is straighter and its base but little wider than the rest of it; but there is on the outer side, near the base and pointing back towards it, a blunt protuberance. The end of the process is divided into two very unequal parts. That which is the hinder- and outermost is by far the larger ; it is long and tongue-like and curled forwards, and its inner edge is serrate. The other part is finger-like, curved at the end, and lies within the larger as though protected in its curl. The proximal process has a secondary process as do those of the four other species (though it is reduced, and may be absent, in E. siiperba). The inner membranous part of it is expanded so that its edge is convex ; it is also serrate. As in the other four species there are at the end of the proximal process in E. crystallorophias two membranous expansions, one behind and one before, the latter striated the former not, but their proportions and shapes are very different. The foremost expansion is very small and set comparatively far back on the process, the middle of its base opposite the secondary 222 DISCOVERY REPORTS process. The hindermost and distal expansion is big. The proximal parts of its lower edges (which are uppermost in the figure) are serrate ; the serrate parts are not con- tinuous but are separated by a gap and lie in two different planes. The end of the lateral process is shaped somewhat like a cargo hook ; it has no tooth on the distal curve. I have not measured large numbers of this neritic species as I have of the others which are oceanic. Tattersall, who examined the big collections of the Discovery (National Antarctic) and Terra Nova Expeditions, says in the first of his two reports (1908, p. 11) that the largest adult specimens of both sexes measure 32 mm.; in the second report (1924, p. 19) he records a length of 33 mm. Tattersall (1908, p. 11, pi. ii, figs. 8-10) has given figures and a very brief account of a Calyptopis, a Furcilia and a Cyrtopia ; but nothing is known of the variety of Furcilia forms which occur and of which among them are dominant. Remarks. Zimmer's careful description and drawings of the copulatory organ (1914, p. 426, pi. xxvi, figs. 60-64) - Q. _l a: 3 E < u in a. > LJ ui LJ 1 l/l u 13 Ul SUB-TROPICAL CONVERGENCE ANTARCTIC CONVERGENCE - EDGE OF PACK-ICE ANTARCTIC COAST- LINE o OL , r> " .<3 XI C u u a, a. o J3 O o ■2^ "I 60"*- «J 1> C M w 1-, 1) -^ 250 DISCOVERY REPORTS The proximal process of E. crystallorophtas has a strong heel-hke angle on the outer side near the base. Its distal part is bent inwards. From near the bend there springs a strong secondary process very similar to that of E. lucens, E. vallentini and E.frigida, more heavily chitinized along its outer edge than its inner, which is membranous, convex and serrate. There are, as in the other four species, two membranous lobes on the end of the proximal process, a larger distal and hinder one, a very small striated one on the front side which has not been described before. The latter lies, as in E. superha, under the point from which the secondary process arises, not in front of it (i.e. not more distally than it) as in E. lucetis, E. vallentini and E.frigida. The terminal process of E. superba is not similar to that of E. lucens and the others; it does not have the same kind of foot-like base, and its distal end is curved and pointed, not cleft in two. The lateral process has no secondary tooth. The terminal process of E. crystallorophias has a narrow base, but there is a sharp projection near its outer side which may be homologous with the well-defined heel of the wide foot-like base of E. lucens, E. vallentini and E.frigida. Its distal end is divided into two very unequal parts, of which the hinder and outermost is the larger; its inner edge is, like the hinder edge of the outer lobe in E.frigida to which I suppose it to correspond, finely serrate. If these parts are indeed homologous, it is very interesting to find this closer relation between the neritic species of the Antarctic coasts and E.frigida from the Antarctic Zone than can be found between it and E. vallentini and E. lucens from the distant sub-Antarctic. The lateral process of E. crystallorophias, like that of E. superba, carries no secondary tooth. It is longer in proportion to the length of the terminal and proximal processes than in any of the other species, and it is differently shaped. It is perhaps significant that the cargo-hook shape of its end is one I have seen approached in one or more of the comparatively few petasmas of E. lucens, E. vallentini, E. frigida and E. superba that I have examined (Fig. 30 a 3, 6 3, c 4, >> r: c u :« h o n c n ^ CI) >, > 4J c> •n 4J u, «) V 4-) &. hJ > >> > > > > > > >>> e o t^ 'in "Fi 'J^ "^ '-I > K S" 1 a o > t! - sSSq <^ o + + N " S " , I- « m X- Is 3 o ° o « o ■" C ti ^ 2 " a. S ■" UI3.S •S bjO > ^ ■*-' tc O (/i '^ " 9, (11 r" m ■S t; u o X .3 " "2 Oral) ^ --3 c " " S " i- o S X -w x; S 3 Si 3 « r^ C X; O On CO eg lO m N o c o .o -a O a, .3 '0 Co .2 3T3 •a tin c U 1> " « l; tn o S •«, •' jj s s " ^ t- c/3 <4) 0) Cj 3 ■" S 3X2 3 "" « c S J P M 3 ^ D- «J .:.; OJ OJ S J- s O X CO 3 P — " 3-rt >< 3 u 3 X ^ > 4-J CO ^ L^ IJ CO c*.! a; 4J ,„ 4; rt u g CO -M (u tj -a 'o 2 3 U _n rt O. a, to C u -^ '^ '-5 u 3 -r « & o " 5x o is 3 -a a> ■-■ (L> Sf 3 -G +j O t— < CO CO n ^ ^^ ^ ., — 3 ■" CO E-H G ^ 'o, >> 05 JD ,p c« T3 U X 264 DISCOVERY REPORTS The different paths that development takes in the species for which it is known are shown in Table IV. Other forms of Furcilia than those shown as stages occur in most of the species ; it may be that in those I have interpreted here in terms of our Furcilia stages from the writings of others they occur in numbers large enough, if they were known, to make necessary some less simple interpretation of the path of development. It may be found that there are two instars where it appears that there is only one— as Eraser found that the Furcilia of Euphaiisia superba with five setose pleopods and seven terminal spines are composed of two stages. But the path of development shown for each species, even if it conceals a step by showing two steps as one, and even though some individuals develop along other paths, is the one followed by most of the individuals of the species. Eiiphausia triacantha, E. longirostris and E. spinifera are closely allied species forming with E. hanseni a natural group within the genus. E. spinifera lives in the sub-Tropical Zone, E. longirostris in the sub-Antarctic, and E. triacantha in the higher (the colder) sub-Antarctic and the Antarctic Zones. It is interesting to note that E. triacantha has a different path of Furcilia development to its allies. Descriptions of the larval development of five species of Euphatisia follow. All the measurements are of material preserved in formalin. I have followed English writers in calling the first thoracic limb by that name, and those that follow, the second to eighth. Scandinavian writers have called the first limb the maxilliped, and those following, the first to seventh thoracic limbs. Development of Euphausia frigida Rustad (1930) described most of the larval forms of E. frigida, none of which had been recognized before. He had large numbers (no) of the three Calyptopis; smaller num- bers (21) of the earlier Furcilia, my stages I-III; and small numbers (12 only) of the later Furcilia, his Cyrtopia— my FurciUa stages V and VI. Ruud (1932, p. 52, table 13) recorded a smaller collection of larvae, richer like Rustad's in Calyptopis than in later stages. Rustad's Furcilia, especially the later, were not numerous enough for him to recognize the steps by which development proceeds, but in a later paper (1934, p. 15) he showed that the seventy-five earlier Furcilia of a larger collection fell into three groups — my Furcilia stages I, II and III. The larval forms and the numbers of each that I found are as follows : Calyptopis: Second ... ■•• ••• ••• ••■ ■•• ■•• ^3 Third 50 Furcilia: With 3 pairs of non-setose pleopods ... ••• •■■ ••• i Stage I. With 4 pairs of non-setose pleopods 123 With 5 pairs of non-setose pleopods ... ... ... ••• ... ••• i Stage II. With 4 pairs of setose, I pair of non-setose pleopods 194 DEVELOPMENT OF E. FRIGIDA 265 Stage III. With 5 pairs of setose pleopods, antennal endopod unsegmented, 7 terminal spines on telson ... ... ... ... ... ... ... ... ... 220 With 5 pairs of setose pleopods, antennal endopod unsegmented, 5 terminal spines on telson ... ... ... ... ... ... ... ... ... i With antennal endopod segmented, 7 terminal spines on telson ... ... 4 With antennal endopod segmented, 6 terminal spines on telson ... ... 2 Stage IV. With antennal endopod segmented, 5 terminal spines on telson ... ... 117 With antennal endopod segmented, 4 terminal spines on telson ... ... 4 Stage V. With antennal endopod segmented, 3 terminal spines on telson ... ... 44 With antennal endopod segmented, 2 terminal spines on telson ... ... 10 Stage VI. With antennal endopod segmented, i terminal spine on telson ... ... 98 All of these are from the Falkland Sector of the Antarctic. Although Rustad (1930) has described the second and third Calyptopis stages, as well as the first of which I have none, I have thought it well to give my own descriptions of them in the terms and detail I use for describing later stages of this species and these and later stages of other species. The larval stages of E. frigida are very similar to those of E. vallentini. Second Calyptopis (Fig. 36 a) Description based on the examination of eight specimens, 1-7-1 -8 mm. long.^ The carapace is rounded anteriorly and is hood-like, coming down far in front and on either side of the head. It broadens out laterally from a point anterior to the middle point of the lateral margin. The upper hinder angle appears in side view to be faintly pointed because of the small and gently rounded projection in the middle of the pos- terior margin. There is no definite dorsal organ, but there may or may not be a small less characteristic projection in the region. The antennular peduncle is of two or of three segments ; the flagella are represented by small single segments. The other appendages, the antennae to the first thoracic limbs, are similar to those of other described Calyptopis. The limbless second to eighth thoracic segments are easily seen as narrow encircling folds of the integument. The abdomen is of six segments and equal in length to the carapace. The posterior margin of the telson is rounded and there are seven terminal spines. Third Calyptopis (Fig. 36 b) Description based on the examination of seventeen specimens. Twenty-three were measured and were 2-5-2-9 mm. long. The carapace seen from above or below, or better still at an oblique angle from above or below, is constricted immediately behind the eyes and then broadens out laterally and ventrally into faint postero-lateral wings. There is as in the preceding stage a low and gently rounded projection in the middle of the posterior margin. A denticle is present on each of the lateral margins near the posterior end. There may or may not be a small but definite projection in the region where the dorsal organ occurs in other species. ^ Rustad (1930) gives the length as about 1-5 mm. 266 DISCOVERY REPORTS The antennular peduncle is of three segments ; the spine from the outer distal corner of the first is as long as, or more often longer than, the sum of the second and third segments. Small sac-like buds of the second, or of the second and third, or even of the second, third and fourth thoracic limbs, are present, the second bigger than the third and so on ; they are enclosed in a common integument, not "free". The abdomen is of seven segments and is one and a half times as long as the carapace. The postero-lateral corners of the telson are rounded, the posterior margin is straight or feebly convex and has seven terminal spines which become progressively smaller from the outside to the centre. Furcilia, stage I (Figs. 36 c-e) Description based on the examination of over forty specimens. Fifty-three were measured and were 3-8-4-3 mm. long. In this stage there are four pairs of non-setose pleopods. The carapace may have no dorsal organ or a diminutive dorsal organ or a small projec- tion of a less definite and characteristic shape. The frontal plate is either rectangular, the anterior margin straight, the lateral margins parallel, the distal corners right angles, and the whole broader than long, or it is a little narrower distally than at the base so that the corners are not right angles and the whole is as long as broad. It is always broader and squarer than that of the same stage of E. vallentini. The antennular spine is longer than, or equal in length to, the sum of the second and third peduncular segments. The most usual condition of the thoracic limbs is that the buds of the second and third are free, the former beginning to bifurcate or bifurcated (Fig. 36 e) and a quarter to half the length of the first, the third about half the length of the second ; the buds of the fourth to eighth are small and enclosed in a common integument. The bud of the second limb may be the only one free and may or may not be bifurcated. There may or may not be a pair of small lateral projections from the tergum of the first abdominal somite, which are the beginnings of the "tergal wings" of the later stages and the adult. The abdomen is one and two-thirds as long as the carapace. The uropods reach as far as the lateral spines of the telson. Furcilia, stage II (Figs. T,6f-h) Description based on the examination of forty specimens. One hundred and forty-two were measured; one was 6-i mm. long but the others fell between 4-7 and 5-7 mm.^ In this stage there are four pairs of setose and one pair of non-setose pleopods. The carapace may have no dorsal organ or projection, a small or, more frequently, very small dorsal organ (never so high and abrupt as that of E. vallentini), or a low pro- jection. The frontal plate is usually wide with a rounded anterior margin sometimes 1 Rustad records a length so low as 4-2 mm. DEVELOPMENT OF E. FRIGIDA 267 Fig. 36. E. frigida. a, second Calyptopis from the side, x 33. b, third Calyptopis from the side, x 33. c, FurciHa, stage I, frontal plate and antennules, x 40. d, Furcilia, stage I, from the side, x 27. e, Furcilia, stage I, second thoracic limb, X53. /, Furcilia, stage II, second thoracic limb, X53. g, Furcilia, stage II, fourth thoracic limb, x 53. A, Furcilia, stage II, anterior part from above, x 20. i, Furcilia, stage III, second thoracic limb, x 53. k, Furcilia, stage III, fifth thoracic limb, x 53. /, tn and n, Furcilia, stage IV, frontal plate and antennules, x 27. 268 DISCOVERY REPORTS without, but mostly with a tiny spine at the centre ; in the smaller the lateral margins are still nearly parallel and the anterior margin may be straight, but in the larger the plate narrows a little distally (Fig. 36 h). It does not approach the " rounded triangular " shape as in E. valient ini. The antennular spine reaches as much as two-thirds or more of the way up the third peduncular segment ; the flagella are unsegmented. The development of the thoracic appendages varies from (i) Buds of second and third free, remainder low and in a common integument ; second nearly as long as first, un- segmented but bifurcated, endopod with terminal and other setae, a very small gill bud (Fig. 36/); third about half the length of second, less completely bifurcated and with no gill bud ; to (ii) Buds of second, third and fourth free ; second as long as first with its endopod of five segments, with a big gill bud ; third half as long as second and bifur- cated but not segmented, endopod with a terminal seta, with a gill bud ; fourth begin- ning to bifurcate and to bud off a gill (Fig. 36 »). The majority lie between these two extremes having the buds of the second and third appendages free and that of the fourth becoming free, the second unsegmented, strong gill buds on the second and third. The tergal wings of the first abdominal somite are well developed (Fig. 36 h), but they are not connected dorsally by a "collar" as in the same stage of E. vallentini (Fig. 37 e). Fiircilia, stage III (Fig. 36 i, k) Description of thoracic appendages based on examination of twelve specimens, of other characters, of sixty specimens. One hundred and forty-nine were measured. One taken unusually late in the year (May i) measured only 5-0 mm.; the remainder fell between 5-5 and 6-7 mm.'^ In this stage there are five pairs of setose pleopods and the antennal endopod is un- segmented. The carapace may have a crest too low to be called a dorsal organ or a small dorsal organ not so big nor so abrupt as that of the same stage of E. vallentini. The frontal plate is usually strongly rounded, though it does not narrow much distally, with a small spiniform process in the centre of its anterior margin ; it may be broadly rounded with no spiniform process. The antennular spine reaches about halfway along the third peduncular segment. The flagella are either unsegmented or beginning to segment and about equal in length to the sum of the second and third peduncular segments. The degree of development of the thoracic limbs is as follows : First limb. Endopod unsegmented as in the earlier larval stages. Second limb. Considerably longer than the first, endopod of five segments and feebly kneed (Fig. 36 2). Third limb. Endopod with five segments indicated or of five segments, much shorter than to knee of second or reaching to it. '^ Rustad records one so small as 4-44 mm. taken in February in inshore waters. DEVELOPMENT OF E. FRIGIDA 269 Fourth limb. Considerably shorter than third, bifurcated, with endopod unsegmented and without or with terminal setae, a strong gill bud. Fifth limb. Very short and not bifurcate, or bifurcate, a small or a strong gill bud (Fig. 36 k). Sixth limb. A low unbifurcate bud, no gill bud or a small gill bud. Seventh limb. Very rarely no gill bud, usually a small or a strong gill bud. The seventh gill bud appears before the sixth and is always bigger than it in this and the stages which immediately follow. Eighth limb. Usually not developed, very rarely a small gill bud. The tergal wings of the first abdominal somite are not connected dorsally by a collar as in this stage of E. vallentini. Of the 239 specimens examined all but two had seven terminal spines on the telson ; the two had five. Furcilia, stage IV (Fig. 36 /, m, n) Description based on the examination of twenty-three specimens. Eighty were measured and were 6-5-77 ™n^- long. In this stage there are five pairs of setose pleopods, the antennal endopod is segmented and there are five terminal spines on the telson. The carapace has a low and well-defined ridge, an ill-defined ridge, or more rarely no ridge, in the position of the dorsal organ. The frontal plate is rarely broad and evenly rounded with no central point, or triangular with strongly convex sides and a central point, or a sharply pointed triangle with feebly convex or straight sides (Fig. 36 /, m, n). The antennular spine reaches up to halfway along the third segment of the peduncle. The mandibular pulp is usually not segmented, rarely segmented. The degree of development of the thoracic limbs is as follows : First limb. Short. Endopod mostly with three, sometimes with four, rarely with five, segments. No gill bud. Second limb. Endopod long and kneed. Third limb. Endopod nearly as long as that of second, not kneed or feebly kneed. Fourth limb. Endopod considerably shorter than that of third, divided sometimes into three, more usually into five segments. Fifth limb. Bifurcate. Endopod rarely very short with only one or two terminal setae, most frequently longer with many setae but unsegmented ; rarely segmented. Sixth limb. Either a low bud beginning to bifurcate or short and bifurcate with no setae on endopod. Buds of the second to eighth pairs of gills are present, usually unbranched, the seventh slightly bigger than all the others, the eighth sometimes very small. The sixth to eighth pairs may be beginning to branch ; the seventh appears to be the first to show signs of branching. The tergal wings of the first abdominal somite are not connected dorsally by a collar. 270 DISCOVERY REPORTS Furcilia, stage V Description based on the examination of nine specimens. Thirty-six were measured and were 7-0-8-3 mm. long. In this stage there are three terminal spines on the telson. The carapace and frontal plate are as in the previous stage, the frontal plate being most often a triangle with straight sides. The antennular spine is as long as, or a little shorter than, the second peduncular segment. The mandibular palp may be either unsegmented or of two or three segments and setose. The degree of development of the thoracic limbs is as follows : First limb. Endopod of five segments and similar in shape to that of second but much shorter. No gill bud. Third limb. Endopod nearly as long as that of second and kneed. Fourth limb. Endopod considerably shorter than that of third but kneed. Fifth limb. Endopod much shorter than that of fourth, of five segments but not kneed. Sixth limb. From short bifurcate with endopod unsegmented but setose, to longer with endopod of three segments. Buds of the second to eighth pairs of gills are present. The second to fifth are single, the sixth to eighth beginning to branch ; or more often the second to fifth are single or beginning to branch, the sixth and eighth two-branched, the seventh three-branched. The new branches are in all much smaller than the buds from which they arise. The tergal wings are not connected dorsally by a collar. Furcilia, stage VI Description based on the examination of fifteen specimens. Sixty-two were measured and were 6-8-9-6 mm. long.^ In this stage there is one terminal spine on the telson. The carapace and frontal plate are similar to those in the two previous stages, the frontal plate being most often a triangle with straight sides. The antennular spine is as long as, or a little shorter than, the second peduncular segment. The mandibular palp is rarely not segmented nor setose, usually segmented and setose. The degree of development of the thoracic limbs is as follows : First limb. Endopod reaching well beyond knee of second. No gill bud or a very small gill bud. Fourth limb. Endopod nearly as long as that of second. ^ Rustad records one of only 6-o mm DEVELOPMENT OF E. VALLENTINI 271 Fifth limb. Endopod of five segments, not kneed or kneed. Sixth Hmb. Very rarely short, bifurcating and with endopod non-setose; usually longer or long with endopod setose and of two to five segments. The first pair of gills may or may not be present as very small buds. Most frequently the second to fifth pairs are not branched or just beginning to branch, the sixth two- branched, the seventh and eighth three-branched. They may be less well developed: the second to fifth single, the sixth to eighth just beginning to branch ; or the second to fifth single or just beginning to branch, the sixth and eighth two-branched, the seventh three-branched. They may be further developed: the second to fifth beginning to branch, the third more advanced than the second and so on, the sixth and seventh three- branched, the eighth four-branched ; or the second to fifth as in the previous clause, the sixth three-branched, the seventh and eighth four-branched. The tergal wings are not connected dorsally by a collar. Development of Euphausia vallentini No one has described any of the larval forms of E. vallentini. The following is a list of those that I have examined, with the numbers of each : Calyptopis: Second 25 50 Third Furcilia: Stage I. With 4 pairs of non-setose pleopods ... ... ... ... ... ... 62 With 3 pairs of setose, 2 pairs of non-setose pleopods ... ... ... ... 3 Stage n. With 4 pairs of setose, I pair of non-setose pleopods ... ... ... ... 47 Stage in. With 5 pairs of setose pleopods, antennal endopod unsegmented ... ... 135 Stage IV. With antennal endopod segmented, 5 terminal spines on telson ... ... 20 With antennal endopod segmented, 4 terminal spines on telson ... . 12 Stage V. With antennal endopod segmented, 3 terminal spines on telson ... ... 93 With antennal endopod segmented, 2 terminal spines on telson ... . . 15 Stage VI. With antennal endopod segmented, I terminal spine on telson ... ... iii All came from immediately north of the Antarctic convergence between the Falkland Islands and South Georgia, i.e. from one sector of the sub-Antarctic Zone and from near the southern limit of the habitat of the species. The larvae of E. vallentini are so similar in general appearance to those of E. frigida that they would be indistinguishable from them were it not for the dorsal organ of the carapace in all the stages described here, and the "tergal collar" of the Furcilia stages II-VI. Second Calyptopis (Fig. t,"] a,b) Description based on the examination of twelve specimens. Twenty-one were measured and were I-6-I-8 mm. long. The carapace is similar to that of the same stage of E. frigida except that there is mostly, but not invariably, a small but distinct dorsal organ. The appendages are as in the second Calyptopis stage of E. frigida, and the thoracic 272 DISCOVERY REPORTS segments without limbs, the second to eighth, can be recognized as narrow encircUng folds of the integument. The abdomen is of six segments and slightly longer than the carapace. The telson has seven terminal spines (Fig. 37 b). Fig. 37. E. vallentim. a, second Calyptopis from the side, x 47. b, second Calyptopis, telson from above, X 107. c, third Calyptopis from the side, x 30. d, Furcilia, stage I, frontal plate and antennules, x 57. e, Furcilia, stage II from the side, x 20. /, Furcilia, stage II, carapace and antennules from above, x 30. Third Calyptopis (Fig. 37 c) Description based on the examination of thirty-two specimens, 2-5-2-8 mm. long. The carapace is similar to that of the third Calyptopis of E.frigida except that there is invariably a small but distinct dorsal organ present. DEVELOPMENT OF E. VALLENTIXI 273 The antennular peduncle is of three segmentSj the spine from the first segment as long as the sum of the second and third. The second, the second and third, or the second, third and fourth thoracic limbs are represented by sac-like buds, the second bigger than the third, and so on, enclosed in a common integument. The second may show the beginnings of bifurcation ; the fourth appears to be more commonly represented by a bud than it does in the same stage of E. frig id a. The abdomen is of seven segments and one and a half times as long as the carapace. There are seven terminal spines on the telson which become progressively smaller from the outside to the centre. Furcilia, stage I (Fig. 37 ^) Description based on the examination of thirty specimens. Forty-five were measured and were 3 •2-4-0 mm. long. In this stage there are four pairs of non-setose pleopods. The carapace of all but a very few of those examined had a well-defined dorsal organ, an unmistakable abrupt protuberance. The frontal plate is rectangular with the distal corners nearly square in the smaller, more rounded in the larger. It is usually nearly as long as broad, rarely a little longer than broad (Fig. 37 d) — never so wide nor so square as in E. frigida. The spine from the outer distal corner of the first antennular segment is usually nearly equal in length to the sum of the second and third segments, sometimes equal to it. The most usual condition of the thoracic limbs is that the buds of the second and third are free (they are so in all thirty), and those of the fourth to eighth enclosed in a common integument ; the second bifurcate, from a quarter to more than half the length of the first ; the third not bifurcate. The third may be bifurcate and the bud of the fourth may be free ; very rarely the second limb buds show the beginnings of gills. Small lateral projections of the tergum of the first abdominal somite are present. The abdomen is nearly twice as long as the carapace. The uropods reach nearly to, or to the lateral spines of the telson. Furcilia, stage II (Fig. 37 e, f) Description based on the examination of twenty specimens. Forty-two were measured and were 4-5-5 -I mm. long. In this stage there are four pairs of setose and one pair of non-setose pleopods. The carapace has a conspicuous dorsal organ. The frontal plate narrows distally and is rounded so that it has the shape of a " rounded triangle". There is usually a diminutive spine in the centre of the anterior margin (Fig. 37/). In some the spine is wanting and the shape is less nearly triangular. The antennular spine reaches two-thirds of the way, or less frequently halfway, along the third peduncular segment; the flagella are short and unsegmented. 274 DISCOVERY REPORTS The buds of the second to fifth thoracic Umbs are free. The degree of their develop- ment varies from (i) second bifurcate and nearly as long as first, endopod with terminal setae, gill bud present ; third bifurcate and less than half length of second, with no setae, gill bud present; fourth not bifurcate, very short with no gill bud; fifth a very small simple bud ; to (ii) endopod of second of five segments and nearly as long as first ; third more than half length of second, endopod with terminal setae; fourth much shorter than third but bifurcate and beginning to bud off a gill ; fifth showing the beginnings of bifurcation and the budding of a gill. The tergal wings of the first abdominal somite are well developed and connected dorsally by an upstanding collar-like ridge. This tergal "collar" is very easily seen and it is the best character for distinguishing between this stage of E. vallentini and E. frigida (Fig. 37 e and/). Fur cilia, stage III Description based on the examination of thirty specimens. Of ninety-eight specimens measured two were 5-0 mm., one was 5-2 mm. and the remainder 5-4-6-I mm. long. In this stage there are five pairs of setose pleopods and the antennal endopod is un- segmented. The carapace has a prominent and unmistakable dorsal organ. The frontal plate is most frequently triangular with strongly or moderately convex sides; it is sometimes rounded with a central spine, more rarely rounded with no central spine. The antennular spine reaches about halfway along the third peduncular segment. The flagella are either unsegmented or beginning to segment and about equal in length to the sum of the second and third peduncular segments. The mandibular palp is unsegmented. The degree of development of the thoracic limbs is as follows : First limb. Endopod of two segments as in earlier stages, no gill bud. Second limb. Longer than first, endopod of five segments and feebly or strongly kneed. Third limb. Endopod of five segments, usually not kneed and as long as to knee of second, sometimes longer and kneed. Fourth Umb. Endopod not segmented, or beginning to segment, terminal and other setae, one-third to more than half as long as third, a strong gill bud. Fifth limb. Bifurcate, without or with setae, one-third to more than half as long as fourth, a gill bud. Sixth hmb. A low bud rarely bifurcate but usually with a small gill bud. Seventh limb. A gill bud usually, but not always, developed. Eighth limb. Usually not developed, rarely a small gill bud. The tergal wings of the first abdominal somite are joined dorsally by a collar as in the previous stage. There may be seven, six or five terminal spines on the telson. Of the specimens ex- amined forty-seven had seven terminal spines, seven had six and fifty-seven had five (see Table I, p. 256 and pp. 257-8). DEVELOPMENT OF E. VALLEXTINI 275 Fur cilia, stage IV Description based on the examination of ten specimens. Thirteen were measured and were 6-3-7-I mm. long. In this stage the antennal endopod is segmented and there are five terminal spines on the telson. The carapace has a prominent and unmistakable dorsal organ, quite unlike anything seen in this stage of E. frigida. The frontal plate is triangular with feebly convex or straight sides. The antennular spine varies from being a little longer to a little shorter than the second segment of the peduncle. The mandibular palp is unsegmented. The degree of development of the thoracic limbs is as follows : First limb. Endopod varies from having more than two segments indicated to being of five segments but not kneed; no gill bud. Second and third limbs. Endopods long and kneed. Fourth limb. Endopod of five segments, short and not kneed or longer and feebly or strongly kneed. Fifth limb. Varies from being short unsegmented but setose to having an endopod of five segments which is not kneed. Sixth limb. Usually very short and non-setose, rarely having an endopod with three segments indicated or completely marked off. There are seven pairs of gills, the second to the eighth, present. They may all be of single buds, the eighth pair small ; the second to fifth may be branching, the sixth and eighth single ; the second to fourth may each have a small branch, the fifth be branching, the sixth and eighth of two branches and the seventh of three. The collar joining the tergal wings of the first abdominal somite is not so high mid- dorsally as in the previous stage but it is distinct and characteristic enough to serve to distinguish this stage from the same stage of E. frigida. Furcilia, stage V Description based on the examination of thirty-five specimens for all but the development of the gills for which twenty were examined. Forty-five specimens were measured and were 6- 1-7-6 mm. long. In this stage there are three terminal spines on the telson. The carapace has a prominent and unmistakable dorsal organ which is, however, a little lower in the larger specimens than in the smaller. The frontal plate is triangular with feebly convex, or more often, straight sides. The antennular spine is usually as long as the second peduncular segment. The mandibular palp is either unsegmented or of three segments and setose. The degree of development of the thoracic limbs is as follows : First limb. Either short, the endopod having five segments indicated, or longer, the endopod having five segments ; usually there is no gill bud but there may be one. 276 DISCOVERY Rf:PORTS Fourth limb. Endopod nearly as long as that of third and strongly kneed. Fifth limb. Varies from having an endopod of three segments to having one of five which is feebly kneed but considerably shorter than that of the fourth. Sixth limb. Rarely not segmented and non-setose, most frequently not segmented but setose, often with endopod having three to five segments indicated. There are always seven pairs of gills, the second to the eighth ; the first pair may or may not be present as small buds. The following appears to be the most usual stage of development : the second to sixth and the eighth gills of buds of one small branch, the seventh of buds with two small branches arising from them. A variety of higher stages of development occurred which showed that the order in which the gills branch is not the same in all individuals. The highest development seen was the second to fifth gills of two, the sixth to eighth of three, strong branches. The tergal collar of the first abdominal somite is the same as in the previous stage. Fur cilia, stage VI Description of more general characters based on the examination of sixty-four specimens, of the mandibular palp on twenty specimens, of the limbs and gills on fourteen. Seventy specimens were measured and were 6-2-9-3 mm. long. In this stage there is one terminal spine on the telson. The carapace may have as strong and distinctive a dorsal organ as in earlier stages, or a smaller dorsal organ or only a crest ; even if it is only a crest it is higher and more definite in form than anything which occurs in this stage of E.frigida. The frontal plate is a triangle with straight margins or a lower triangle in which the lateral margins are distally slightly concave. The antennular spine is nearly as long as, or much shorter than, the second pe- duncular segment. The mandibular palp may be unsegmented but is usually segmented. The degree of development of the thoracic limbs is as follows : First limb. Either short with no gill bud, the endopod feebly kneed, or longer with a very small or small gill bud and the endopod strongly kneed. Second to fifth limbs. Endopods long and kneed. Sixth limb. Endopod of two to five segments and not kneed, or of five segments and kneed. The first pair of gills may or may not be present as small buds. The second to fifth may each consist of two unequal branches, the sixth of two, and the seventh and eighth of three, roughly equal branches ; or the second to fifth may consist of two nearly equal branches, the sixth to eighth of three unequal branches ; or the second to fifth may con- sist of two equal and one much smaller branch, the sixth of three, the seventh of four, the eighth of five, unequal branches. The tergal collar of the first abdominal somite is most frequently similar to that of the two previous stages; it is sometimes absent, so that no connection between the tergal wings can be seen. DEVELOPMENT OF E. TRIACANTHA 277 Post-Larval Stages Sixty-seven specimens of post-larval stages, all taken in one net haul in late February, were examined to discover at what size the tergal collar of the first somite of the ab- domen, which serves to distinguish certain of the Furcilia stages from those of E.frigida, is lost ; and at what sizes two structures characteristic of the adult, the broadly rounded antennular lappet and the spine of the third abdominal somite, appear. It was found that the tergal collar was absent from all and that the spine appeared before the lappet. An analysis of the specimens is shown below in lettered groups and the lengths of those of each group in Table V. Groups A, B and C were measured with greater accuracy than the larger individuals of groups D and E. A. Without the beginnings of the abdominal spine ... ... ... ... 18 specimens. B. With the beginnings of an abdominal spine : a small projection only ... 11 specimens. C. With a well-developed abdominal spine but with no lappet, or with the small beginnings of a lappet 13 specimens. D. With lappet less than, or about, half-formed 16 specimens. E. With lappet of shape and proportions of the adult 9 specimens. Table V. Showing lengths {in mm.) and numbers {in brackets) of post-larval E. vallentini, arranged according to the development of the abdominal spine and the antennular lappet A 6-.S (I) 6-8 (I) 7-3 (I) 7-6 (3) 77 (I) 8-0 (3) 8-1 (2) 8-2 (I) 8-3 (2) 8-5 (2) 8-6 (0 B 7-5 (I) 7-8 (i) 8-0 (I) 8-2 (2) 8-3 (I) 8-4 (i) 8-.S (I) 8-6 (I) 8-8 (I) 9-2 C 9-0 (i) D 9-5 (0 lo-o (2) lO-I (0 10-2 (I) io'4 (2) 10-5 (2) I0-6 (I) 107 (2) 12 (4) 13 (5) H (5) 15 (2) 14 (2) i.S (I) 16 (I) 17 (2) 18 (3) 278 DISCOVERY REPORTS The smallest post-larval stages, Group A above, having lost the tergal collar of the Furcilia stages and not yet having begun to develop the third abdominal spine of the adult, might be very difficult to distinguish with certainty from early post-larval stages of E. frigida if they w^ere taken from on or near the Antarctic convergence at a point where there had been some mixing of the surface waters of the Antarctic and sub- Antarctic Zones (see p. 204). Development of Euphausia triacantha The euphausian larvae described by Tattersall (1908, p. 21) as having a short blunt posterior spine on the carapace and a smooth anterior margin are undoubtedly of E. triaca?itha. Rustad (1930, pp. 54-7, figs. 35-7) described the first larval form of E. triacantha as such, a Furcilia with four pairs of setose and one pair of non-setose pleo- pods, from one specimen. In a later paper (1934, pp. 19 25, figs. 3-7) he described the first, second and third Calyptopis stages from five, eighteen and six specimens re- spectively, and a "Cyrtopia" with seven terminal spines on the telson (my Furcilia stage IV) from one specimen. The stages and the numbers of each that I found are as follows : Calyptopis: Second ... ... ... ••• ••• ••• ••• ■•■ ••• •■■ 25 Third 95 Furcilia: Stage I. With 4 pairs of non-setose pleopods 310 With 5 pairs of non-setose pleopods ... ... ... ... ... ... 5 With 3 pairs of setose, 2 pairs of non-setose pleopods ... ... i Stage II. With 4 pairs of setose, I pair of non-setose pleopods 343 Stage III. With 5 pairs of setose pleopods, antennal endopod unsegmented 247 With antennal endopod segmented, 8 terminal spines on telson ... ... 2 Stage IV. With antennal endopod segmented, 7 terminal spines on telson 136 With antennal endopod segmented, 6 terminal spines on telson ... ... 11 Stage V. With antennal endopod segmented, 5 terminal spines on telson 71 With antennal endopod segmented, 4 terminal spines on telson ... ... 1 Stage VI. With antennal endopod segmented, 3 terminal spines on telson 11 With antennal endopod segmented, 2 terminal spines on telson ... ... 3 Stage VII. With antennal endopod segmented, i terminal spine on telson 16 The vast majority came from either side of the Antarctic convergence between the longitudes of the South Sandwich Islands and Cape Horn. The larvae of E. triacantha can easily be distinguished from all other euphausian larvae occurring in the same waters. Second Calyptopis (Fig. 38 a) Description based on the examination of seven specimens. Fourteen were measured and were I •9-2-3 mm. long. The carapace is high and domed, the dome surmounted by a dorsal organ; the anterior margin is gently rounded ; the lateral margins broaden out strongly behind the DEVELOPMENT OF E. TRIACANTHA 279 level of the eyes to form strong postero-lateral wings. There is a short but stout posterior projection from the middle of the hinder edge of the carapace. The antennular peduncles are sometimes unsegmented, but are most often of two, sometimes of three, segments. The flagella are represented by small single segments. The antennae, mandibles, first and second maxillae and first thoracic limbs are present as in other species of Euphausia for which this stage is known and are of similar structure. The thoracic segments without limbs are clearly visible as encircling folds of the integument. The abdomen is of six segments. The telson has seven terminal spines decreasing in size from the outer to the inner, each having a row of spinules along each lateral margin. The three postero-lateral spines increase in size from the outer to the inner and each has a secondary spine, apart from spinules, arising from it. Third Calyptopis (Fig. 38 b) Description based on the examination of fifteen specimens. Thirty-three were measured and were 3 -0-3 -6 mm. long. The carapace rises dome-like mid-dorsally to a point surmounted by a large and con- spicuous dorsal organ. It is very broad anteriorly, with the margin gently rounded, usually with a tiny spine in the centre ; it is arched over each eye, constricted behind the eyes and then widened out into the postero-lateral wings. There is a strong median projection from the posterior margin which rises dorsally. A pair of lateral denticles is present. The antennular peduncle is of three segments ; the spine arising from the outer distal corner of the basal segment is considerably longer than the sum of the second and third segments. Buds of the second thoracic limbs are usually, though not invariably, present as large conical protuberances. The abdomen is of seven segments. The posterior lateral margins of the sixth seg- ment are produced backwards to form a pair of short spines, one running down either side of the telson ; they become longer in the Furcilia and are present in post-larval stages but not in adults. The telson is broader than in the second Calyptopis ; the ter- minal spines remain unchanged. The innermost of the postero-lateral spines is the longest, the outer the shortest; each of the three has a secondary spine arising from it; the inner two, but not the outer, have a row of spinules along the inner margin. Furcilia, stage I (Fig. 38 c) Description based on the examination of twenty specimens. One hundred and forty-six were measured and were 4-0-5-3 mm. long. In this stage there are four pairs of non-setose pleopods. The frontal plate is shield-like with the centre of its rounded anterior margin pro- duced into a short but strong spine, the rostral spine. The carapace has a very con- 28o DISCOVERY REPORTS spicuous dorsal organ. Its median posterior projection has a wide base and it reaches well beyond the hinder margins of the postero-lateral wings. The antennular spine is as long as in the third Calyptopis ; the flagella are unseg- mented. The degree of development of the thoracic limbs is as follows : Second limb. Long and bifurcate, the endopod with terminal setae ; a strong gill bud. Third limb. Usually short, about half as long as second, showing beginnings of bud- ding off of exopod and gill at once ; less often, longer and bifurcate, non-setose, with a small gill bud. Fourth limb. Always free. Usually simple and less than half as long as third, some- times beginning to show that exopod and gill will be budded off together. Fifth limb. Not free. There is a pair of very small lateral protuberances from the tergum of the first ab- dominal somite, the beginnings of the tergal wings of the adult (Fig. 38 c). There may be no posterior dorsal spines on the third to fifth abdominal somites, or very small spines on the fourth and fifth, or on all three. The telson is similar to that of the third Calyptopis. Fiircilia, stage II Description based on the examination of ten specimens. One hundred and twenty-nine were measured and were 5-0-6-5 mm. long. In this stage there are four pairs of setose and one pair of non-setose pleopods. The frontal plate is a little reduced in width and length compared with the previous stage, the rostral spine is stronger : this is a process which is continued in each of the larval stages which follow. The rostral spine reaches nearly as far forward as the end of the basal segment of the antennular peduncle. The dorsal organ is very strong. The median posterior projection of the carapace has gone, leaving only a low rounding of the margin in its place. The antennular spine is still longer than the sum of the second and third peduncular segments. The flagella are usually unsegmented, sometimes beginning to segment. The degree of development of the thoracic limbs is as follows : First limb. Endopod still of two segments as in earlier stages, usually only a little longer than exopod. No gill bud. Second limb. Endopod of five segments and feebly kneed. A two-branched gill bud. Third limb. Endopod with five or less segments indicated, or of five segments. Gill bud beginning to branch or of two branches. Fourth limb. Short, bifurcate, endopod non-setose or setose. Gill bud unbranched or branching. Fifth limb. Very short, non-setose, usually bifurcating and budding off a gill, or with a small gill bud. Fig. 38. E. triacantha. a, second Calyptopis from above, X33. b, third Calyptopis from the side, x 22. c, Furciha, stage I from above, x 30. d, Furciha, stage III, telson from above, x 22. e, Furciha, stage III, frontal plate and antennules from above, x 24. /, Furcilia, stage III, third to fifth abdominal segments from the side, X30. g, Furcilia, stage IV, telson from above, x 27. h, Furcilia, stage V, front part of carapace from the side, x 27. /, Furcilia, stage V, frontal plate and antennules from above, x 24. k, Furcilia, stage VII, frontal plate and antennules from above, x 25. 282 DISCOVERY REPORTS Sixth limb. A low unbifurcate bud without or with a small gill bud. Seventh limb. A strong gill bud. The lateral protuberances of the first abdominal somite which appeared in the pre- vious stage have developed into strong outstanding wings. The base of each lies obliquely on the tergum in a postero-ventral direction. These structures persist to the adult stage when their anterior ends overlap the margins of the postero-lateral wings of the carapace. Each of the third to fifth abdominal somites has a posterior dorsal spine, the fifth the strongest, the third the weakest. The telson is similar to that of the previous stage. Furcilia, stage III (Fig. 38 d, e,f) Description based on the examination of ten specimens. Seventy-eight were measured and were b-^-l"] mm. long. In this stage there are five pairs of setose pleopods, but the antennal endopod is unsegmented. The frontal plate is a little more reduced, the rostral spine a little stronger than in the previous stage. The spine reaches nearly as far, or as far, forward as the end of the basal segment of the antennular peduncle. The carapace has a strong dorsal organ ; no trace of the median posterior projection of earlier stages remains : there is a deep depression in the margin in that region between the postero-lateral wings. The antennular spine is as long as the sum of the second and third peduncular seg- ments ; the flagella are segmenting. The mandibular palp is unsegmented. The degree of development of the thoracic limbs is as follows : First limb. Endopod as long as to knee of that of second limb, rarely of two segments, usually with three or more segments indicated. No gill bud. Second limb. Endopod long and kneed. Gill either of two nearly equal branches, or of two nearly equal and a third smaller branch. Third limb. Endopod shorter than that of second but kneed. Gill as in second limb. Fourth limb. Endopod of five segments, not kneed or feebly kneed. Gill usually of two branches, rarely with a third smaller branch. Fifth limb. Bifurcate. Endopod setose, not segmented or with segmentation in- dicated. Gill usually of two branches, rarely with a third smaller branch. Sixth limb. Bifurcating or bifurcate, non-setose. A large gill bud budding off one or two small branches. Seventh limb. A gill of two nearly equal branches and one or two smaller branches. Eighth limb. A small gill bud which may be branching. The tergal wings of the first abdominal somite may be unconnected, or they may be joined dorsally by a ridge of chitin or by an upstanding "collar" as in E. valleiitini. DEVELOPMENT OF E. TRIACANTHA 283 There is a strong posterior dorsal spine on each of the third to fifth abdominal so- mites as in the succeeding stages and the adult (Fig. 38 f). The telson is narrower distally than in the previous stage and the innermost pair of postero-lateral spines, still the longest, are beginning to thicken at the base (Fig. 38 d). Fiircilia, stage IV (Fig. 38_§') Description based on the examination of ten specimens. Eighty were measured and were 6-5-9-2 mm. long. In this stage the antennal endopod is segmented and there are seven terminal spines on the telson. The frontal plate is a little more reduced, the rostral spine a little stronger, than in the previous stage. The spine reaches as far forward as the end of the basal segment of the antennular peduncle. The carapace has a conspicuous dorsal organ. The antennular spine is as long as the sum of the second and third peduncular segments. The mandibular palp is elongated ; it may be non-setose and not segmented or be- ginning to segment, but it is more often segmented and setose. The degree of development of the thoracic limbs is as follows : First limb. Endopod of five segments and long, reaching beyond knee of second, not kneed or feebly kneed. No gill bud. Second to fourth limbs. Endopods long and kneed. Fifth limb. Endopod considerably shorter than fourth, of five segments, not kneed or feebly kneed. Sixth limb. Short. Endopod usually of three, sometimes of more segments, setose. The first limb has no gill bud. Those of the second to fifth are of two nearly equal branches and a third smaller or much smaller branch. The sixth gill bud is of three nearly equal branches with or without a fourth and smaller branch. The seventh is of three larger and one or two smaller branches, or of four larger branches with or without the buds of other branches ; the eighth of two nearly equal branches with or without buds of others. The end of the telson is narrower than in the previous stage and the innermost of the three pairs of postero-lateral spines are still wider and stronger (Fig. 38^). Furcilia, stage V (Fig. 38 h, i) Description based on the examination of ten specimens. Thirty-nine were measured and were 7-5-IO-2 mm. long. In this stage there are five terminal spines on the telson. The frontal plate is further reduced, the rostral spine so strong that it is more appropriately called a rostrum. It reaches nearly to or to beyond the end of the first segment of the antennular peduncle; a strong crest runs centrally down it from the dorsal organ (Fig. 38 h, i). 284 DISCOVERY REPORTS The antennular spine is a little shorter than the sum of the second and third segments of the peduncle. The mandibular palp is usually of three segments and setose. The degree of development of the thoracic limbs is as follows : First limb. Endopod of five segments, long and kneed, reaching well beyond knee of second. Fifth limb. Endopod shorter than those of anterior limbs but kneed. Sixth limb. Endopod of five segments but not kneed. The development of the gills is as follows : First gill. A small bud. Second to fifth gills. Of three equal branches without or with another smaller branch. Sixth gill. Of three larger branches and buds of others, or of four larger branches with or without the buds of others. Seventh gill. Of four or five larger branches with the buds of others. Eighth gill. Of two, three or more, usually four, larger branches and the buds of others. The end of the telson is narrower, the innermost of the postero-lateral spines thicker, than in the previous stage. The latter have lost the spinules of their inner margins. Fiircilia, stage VI Description based on the examination of nine specimens, 8-8-1 17 mm. long. In this stage there are three terminal spines on the telson. The frontal plate is a little reduced compared with that of the previous stage, the rostrum is as long. The dorsal organ may not be very distinct from the strong arched crest which runs forward from it to the rostrum. The antennular spine is nearly as long as, or much shorter than, the sum of the second and third segments of the peduncle. The mandibular palp is segmented and setose. The degree of development of the thoracic limbs is as follows : First limb. Endopod shorter than that of second. Fifth limb. Endopod kneed, much shorter than, to nearly as long as, that of fourth. Sixth limb. Endopod much shorter than that of fifth but kneed. The gills of five specimens were examined ; in four they were as follows : First gill. A small or large simple bud. Second to fifth gills. Of three equal branches and a fourth smaller branch. Sixth gill. Of four or five equal branches with buds of others. Seventh gill. Of six roughly equal branches with buds of others. Eighth gill. Of four, five or six roughly equal branches and buds of others. In the fifth specimen, the smallest one seen, only 8-8 mm. long, the gills were less developed, corresponding most nearly with those of stage IV. DEVELOPMENT OF E. LONGIROSTRIS 285 The telson is a little narrower, the innermost pair of postero-lateral spines a little longer and stronger, than in the previous stage. Furcilia, stage VII (Fig. 38 k) Description based on the examination of thirteen specimens, 8-5-1 1-4 mm. long.^ In this stage there is one terminal spine on the telson. The frontal plate is a little smaller than in the previous stage; the rostrum reaches from nearly to, to beyond, the end of the first segment of the antennular peduncle. The dorsal organ and crest are as in the previous stage. The antennular spine may be as long as or longer than the second segment of the peduncle. The beginnings of the lappet of the first segment may be present. The thoracic limbs of twelve specimens, from 9-0 to 11-4 mm. long, were as follows : First limb. Endopod nearly as long as that of second. Fifth limb. Endopod long and kneed. Sixth limb. Endopod as long as to knee of fifth or longer, feebly or strongly kneed. In the smallest specimen taken, only 8-5 mm. long, the fifth limb is considerably shorter than the fourth, the sixth only half the length of the fifth and not kneed. The gills of five specimens were examined. In four they were as in the five specimens of the previous stage ; in the fifth as in the sixth specimen of the previous stage. It appears then that specimens of Furcilia stage VII may have gills no farther developed than those of stage IV. Development of Euphausia longirostris Dohrn (1871, pi. xxx, fig. 54) figured the second Calyptopis of E. longirostris without recognizing it as the larva of a euphausian ; it is described as from the Indian Ocean. In the Challenger collection, Sars (1885, p. 170, pi. xxxi, figs. 30, 31) found and de- scribed a single specimen of a Furcilia with one pair of non-setose pleopods ; he con- sidered it to be a larva of a large species of Euphausia. Tattersall (1924, p. 22, pi. i, figs. 1-7; pi. ii, figs. 1-4), reporting on the Terra Nova collection, first recognized E. longirostris larvae for what they were. Describing his specimens in my terms he found : I specimen of Furcilia, stage I 4 specimens of Furcilia, stage V I specimen of FurciHa, stage II i specimen of Furcilia, stage VI 4 specimens of Furcilia, stage III 2 adolescents 9 specimens of Furcilia, stage IV He found in addition one larva {op. cit., pi. i, fig. i) with one pair of non-setose pleo- pods "so strikingly different from the later stages. . .that he hesitated for a long time before including it in the series". It is quite clear, now that the second and third 1 Since specimens of the previous stage were found of a length of 11-7 mm. it is probable that specimens of this stage are sometimes much bigger than any taken by me. 286 DISCOVERY REPORTS Calyptopis stages of E. lougirostris are known, that this larva is of another species. It came from south of New Zealand. I found one exactly similar to it at a station south- west of Western Australia ; and later stages of the same species— three specimens with one pair of setose and three pairs of non-setose pleopods, one specimen with four pairs of setose and one pair of non-setose pleopods — at a station south-west of the Cape. Both stations were in the warmer part of the sub-Antarctic zone ; adult and larval E. lougi- rostris, and large numbers of -E. similis were taken at them. I suspect the unknown larvae to belong to the latter species. The order of the appearance of the pleopods appears to be the same as in E. lougirostris and E. spinifera. Illig (1930, pp. 504-7, figs. 183-190) described a series of later larvae from off the west coast of South Africa and south of the Cape, all, as the surface temperatures show, from the sub-tropical zone, as E. lotigirostris. I was convinced, because of the locality from which they came, that they were E. spinifera, and have examined a number of them (p. 294) and found them to be so. The larval stages of E. lougirostris and the numbers of each that I found are as follows : Calyptopis: Second ... ... ••• ••• 9 Third 20 FURCILIA : Stage I. With I pair of non-setose pleopods 16 Stage II. With I pair of setose, 3 pairs of non-setose pleopods 6 Stage III. With 4 pairs of setose, I pair of non-setose pleopods 2 Stage IV. With 5 pairs of setose pleopods, antennal endopod unsegmented 5 With antennal endopod segmented, 6 terminal spines on telson i Stage V. With antennal endopod segmented, 5 terminal spines on telson 2 With antennal endopod segmented, 4 terminal spines on telson 3 Stage VI. With antennal endopod segmented, 3 terminal spines on telson 5 Stage VII. With antennal endopod segmented, I terminal spine on telson i These larvae were taken in the sub-Antarctic Zone south-west and south-east of the Cape and south-west of Western Australia, Second Calyptopis (Fig. 39 a, h) Description based on the examination of nine specimens, 2-o-2-i mm. long. The carapace seen from the side is dome-shaped and has a high and abrupt dorsal organ. Its lateral margins are sharply and deeply indented just behind the eyes and its anterior margin is emarginate in the centre, so that the carapace from above, dis- regarding its median posterior projection, has a cottage-loaf-like outline. The entire margin is beset with strong denticles. The median posterior projection is strong and usually curved upwards. The antennular peduncle is of two or three segments with no spine. The antennae, mandibles, first and second maxillae and the first pair of thoracic limbs are present. The DEVELOPMENT OF E. LONGIROSTRIS 287 Fig. 39. E. longirostris. a, second Calyptopis from the side, x 33. b, second Calyptopis, telson from above, X 66. c, third Calyptopis from above, x 25. d, Furcilia, stage I from the side, x 16. e, Furcilia, stage I, frontal plate from above, x 16./, Furcilia, stage III from the side, x 11. g, Furcilia, stage IV, front part of carapace and antennules, x 16. /;, Furcilia, stage V, front part of carapace and antennules, X23. /, Furcilia, stage V, telson from above, x 23. k, Furcilia, stage VII, carapace from the side, x 14. /, Furcilia, stage VII, third and fourth abdominal segments from the side, x 30. 288 DISCOVERY REPORTS thoracic segments without Umbs are clearly visible as narrow encircling folds of the integument. The abdomen is of six segments. The telson is short, wider distally than proximally ; its margins between the lateral and postero-lateral spines are concave ; the distal margin is emarginate with seven terminal spines, the innermost of which is much smaller than the others. Each of the postero-lateral spines has a secondary spine arising from it ; the middlemost of the three pairs is much the longest (Fig. 39 b). Third Calyptopis (Fig. 39 c) Description based on the examination of ten specimens. Twenty were measured and were 3'3~3'6 mm. long. The dorsal organ is high, abrupt and conical. The part of the carapace anterior to the constriction behind the eyes is larger in proportion to that behind than in the second Calyptopis. The entire margin of the carapace is closely set with strong denticles ; the single pair of lateral denticles do not appear in this stage of this species as they do in each of the others described in this paper. The median posterior projection of the cara- pace is of much the same proportions as in the second Calyptopis. The antennular peduncle is of three segments ; the spine from the outer distal corner of the first is longer, or much longer, than the sum of the second and third. The flagella are represented by small single segments. None of the second to eighth pairs of thoracic limbs is present as a free bud. The abdomen is of seven segments. The hinder lower edges of the pleura are denti- culate. The posterior lateral margins of the sixth somite are produced backwards to give a pair of short spines, one on either side of the telson as in the Furcilia stages of E. triacantho ; they persist to the last Furcilia stage. The telson is large, longer than the fourth to sixth somites, with its distal end expanded. Furcilia, stage I (Fig. 39 d, e) Description based on the close examination of ten specimens; sixteen were measured and were 4-2-4-7 mm. long. In this stage there is one pair of non-setose pleopods. The dorsal organ is a high cone rising abruptly from the carapace (Fig. 39 d). The frontal plate is very wide with an emarginate anterior margin and broadly rounded corners (Fig. 39 e). Its free edge is closely set with denticles, smaller on the lateral margins, stronger on the anterior. The margins of the carapace from the antero-lateral spines to the lateral denticles, which appear in this stage as short wide-based projections, are closely beset with denticles. There are no denticles along the margins of the con- cavities between the antero-lateral spines and the base of the frontal plate, and none posterior to the lateral denticles. The median posterior projection usually reaches as far back as the end of the second abdominal segment, but in one of the specimens it was very short and thick. DEVELOPMENT OF E. LONGIROSTRIS 289 The antennular spine is longer than the sum of the second and third peduncular segments ; the flagella are represented by single segments. Of the second to eighth thoracic limbs the second only may be free as large simple buds ; or the second and third may be free, the second large and beginning to bud off exopod and gill, the third small and simple. A pair of narrow lateral protuberances stand strongly out from the tergum of the first abdominal somite, their ends curved downwards and forwards : they are the beginnings of the "tergal wings". The hinder lower edges of the pleura of the abdominal somites are denticulate. There is a median posterior spine on the third abdominal somite, a quarter to half as long as the fourth somite. The telson is very similar to that of the third Calyptopis. Fur cilia, stage II Description based on the examination of six specimens, 5-0-5-4 mm. long. In this stage there is one pair of setose and three pairs of non-setose pleopods. The dorsal organ is a high and abrupt cone. The frontal plate is narrower than that of Furcilia stage I, but in five of the specimens it is otherwise similar to it ; in the sixth specimen there is a small spine, the beginning of the rostrum, shorter than the den- ticles on either side of it, in the centre of the anterior margin. The lateral margins of the carapace may, as in the previous stage, be denticulate from the antero-lateral spines to the lateral denticles, the more anterior denticulations being very small or absent. The posterior projection of the carapace is strong and of great length ; it may be as long as, or longer than, the first three abdominal somites. The antennular spine is longer than the sum of the second and third segments of the peduncle; the flagella are still unsegmented. The degree of development of the thoracic limbs is as follows : Second limb. Endopod of five segments and feebly kneed. A gill bud of two branches. Third limb. Short, bifurcate and setose but not segmented, a single gill bud; or endopod of three or four segments and gill bud budding off a second branch. Fourth limb. Free bud. Posterior limbs. Not free. The tergal wings of the first abdominal somite are stronger than in the previous stage and bent farther forward distally. The hinder lower edges of the pleura of the ab- dominal somites are denticulate. The spine of the third somite is stout and about half the length of the fourth. The telson is similar to those of the two previous stages. Furcilia, stage III (Fig. 39/) Description based on the examination of two specimens, 6-o and 6-3 mm. long. In this stage there are four pairs of setose and one pair of non-setose pleopods. D XIV 13 290 DISCOVERY REPORTS The dorsal organ is high and abrupt. The frontal plate is square, its lateral margins nearly parallel, its corners squarely rounded ; the anterior margin is weakly emarginate with a small projection, the rostral spine, arising from its centre ; the anterior and the lateral margins are strongly denticulate. The lateral margins of the carapace remain denticulate for a part of their length, from some way posterior to the antero-lateral spines to some way anterior to the lateral denticles. The posterior projection of the carapace is very strong and wide-based in both specimens ; in one it reaches to the end of the second, in the other to near the end of the third, abdominal somite. The antennular spine is longer, the flagella considerably longer, than the sum of the second and third segments of the peduncle. The flagella are not yet distinctly seg- mented. The thoracic limbs of the larger specimen are as follows : First limb. Endopod short, of same length as exopod. Second limb. Endopod of five segments and kneed. Gill of two unequal branches. Third limb. Endopod shorter than second and feebly kneed. Gill of two unequal branches. Fourth limb. Endopod much shorter than third, of five segments but not kneed. Gill of two unequal branches. Fifth limb. Very short, bifurcating, non-setose, with a single gill bud. Sixth limb. A small undifferentiated bud. Seventh limb. A small gill bud of two unequal branches. Eighth limb. A small undifferentiated bud. The hinder lower edges of the pleura of the abdominal somites are very finely denticulate. The spine of the third segment is about half the length of the fourth. The distal end of the telson is narrower than in the previous stage. The middlemost of the postero-lateral spines is still the longest, the innermost is greatly thickened at the base. Furcilia, stage IV (Fig. 39^) Description based on the examination of five specimens, 6-5-6-9 mm. long. In this stage the antennal endopod is unsegmented and there are five pairs of setose pleopods. The dorsal organ is high and abrupt. The frontal plate is wide and square with its corners rounded and its margins beset with denticles ; the centre of its anterior margin is produced into a strong rostral spine with a wide base which reaches as far as halfway along, or nearly to the end of, the second segment of the antennular peduncle. The same portions of the lateral margins of the carapace are denticulate as in the previous stage (Fig. 39^). The posterior projection of the carapace is strong and reaches as far as the end of the second, or halfway down the third, abdominal somite. The antennular spine is as long as or longer than the sum of the second and third segments of the peduncle. The flagella are long and segmented. DEVELOPMENT OF E. LONGIROSTRIS 291 The thoracic Hmbs of a specimen 6-5 mm. long are as follows: First limb. Endopod longer than exopod but still of two segments. Second and third limbs. Endopods long and kneed. Gills of two branches. Fourth limb. Endopod shorter but kneed. Gill of two branches. Fifth limb. Endopod short, of three segments. Gill of two branches. Sixth limb. A low bud, not bifurcating but budding off a gill. Seventh limb. A gill of two branches. Eighth limb. A small undifferentiated bud. There are no denticulations along the edges of the pleura of the abdominal somites. The tergal wings of the first somite are very strong. The spine of the third is about three-quarters the length of the fourth somite. The posterior margin of the telson is narrow and straight with seven terminal spines. The innermost of the postero-lateral spines is now the longest ; it has a thick base, and where it narrows distally on the inner margin there is a row of strong spines. Furcilia, stage V (Fig. 39 //, /) Description based on the examination of six specimens, y-o-yy mm. long. I have regarded the six specimens with six, five or four spines on the telson (see list on p. 286) as belonging to one group, and the description given below is based upon them. Their antennal endopods are segmented and they appear to be at roughly the same stage of development. Their lengths are as follows : With six terminal spines : i specimen, 7-0 mm. With five terminal spines: 2 specimens, both 7-5 mm. With four terminal spines: 3 specimens, 7-0, 7-4 and 7-7 mm. I have little doubt, by analogy with the other species described here, that if large numbers of this stage were taken the majority would be found to have five terminal spines on the telson. The dorsal organ is prominent. The frontal plate is as in the previous stage, but the rostral spine is longer, reaching nearly to the end or to the end of the second segment of the antennular peduncle. There are no denticulations along the lateral margins of the carapace (Fig. 39 //). The posterior projection of the carapace reaches from more than halfway down, to beyond the end of, the second abdominal somite. The antennular spine is as long as the sum of the second and third peduncular seg- ments ; the lappet of the first segment is not developed. The mandibular palp is small and unsegmented. The thoracic limbs of a specimen 7-5 mm. long are as follows: First limb. Endopod of five segments and over twice as long as exopod. Second limb. Endopod long and kneed. Gill of two larger and one smaller branch. Third limb. Endopod long and kneed. Gill of two large and one very small branch. Fourth limb. Endopod shorter but kneed. Gill as in third limb. 13-2 292 DISCOVERY REPORTS Fifth limb. Endopod of five segments and not kneed. Gill of two equal branches. Sixth limb. Very short, bifurcating, non-setose. Gill budding off a second branch. Seventh limb. A gill of two large and two small branches. Eighth limb. A very small gill bud, single or branching. The spine from the third abdominal somite varies from being over half the length of the fourth somite to being, more frequently, as long as it. The innermost pair of postero- lateral spines are thicker and longer than in the previous stage ; each has a row of strong spines along the distal half of the inner margin (Fig. 39 /). Furcilia, stage VI Description based on the examination of four specimens, 77-8-3 mm. long. In this stage there are three terminal spines on the telson. In three specimens the dorsal organ is high and abrupt, in one a low crest. The frontal plate and rostral spine are similar to those of the previous stage except that in one specimen the rostral spine is very short; it appears malformed. The posterior projection of the carapace varies from reaching nearly to the end of, to reaching less than halfway down, the second abdominal somite. The antennular spine is as long as, or shorter than, the sum of the second and third segments of the peduncle. The beginnings of the lappet of the first segment and of the carina of the third may be present. The mandibular palp is unsegmented and non-setose in one of the specimens, 7-9 mm. long, but of three segments and setose in the other three, 8-o, 8-2 and 8-3 mm. long. The thoracic limbs of a specimen 8-3 mm. long are as follows: First limb. Endopod considerably longer than to knee of second, weakly kneed. No gill bud. Second and third limbs. Gills of three roughly equal branches. Fourth limb. Endopod long and kneed. Gill of three roughly equal branches. Fifth limb. Endopod long and kneed. Gill of three branches, two large and one small. Sixth limb. Endopod short, of five segments and weakly kneed. Gill of three equal branches and a number of small buds. Seventh limb. Gill of four branches and a number of small buds. Eighth limb. Gill of two small and three smaller branches. The tergal wings of the first abdominal somite are changing their shape to approach that of the adult ; the free edge has a well-defined angle dorso-anteriorly, a poorly defined angle ventrally. The spine of the third somite is nearly as long, or as long, as the fourth somite. The telson is narrower than in the previous stage; an additional pair of lateral spines has appeared between the pair present in earlier stages and the postero-lateral spines. The middlemost pair of the latter is becoming reduced. DEVELOPMENT OF E. LONGIROSTRIS 293 Furcilia, stage VII (Fig. 39 k, I) Only one specimen was found, 8-5 mm. long. In this stage there is one terminal spine on the telson. The dorsal organ is a low crest. The frontal plate and rostral spine are much as in the previous stage ; the posterior projection of the carapace is shorter, reaching only to the end of the first abdominal somite. The antennular spine reaches some way up the third segment of the peduncle ; be- ginnings of the lappet of the first segment and of the carina of the third are present. The mandibular palp is long, of three segments and setose. The thoracic limbs are as follows: First limb. Endopod long and kneed. No gill bud. Second to fifth limbs. Endopods long and kneed. Gills of three branches. Sixth limb. Endopod very much shorter than that of the anterior limbs but kneed. A gill of three branches. Seventh limb. A gill of four branches and a number of small buds. Eighth limb. A gill of two small branches and smaller buds. The third abdominal somite is still the only one to have a median posterior spine ; it is shorter than the fourth segment (Fig. 39 /). The pair of postero-lateral spines of the sixth segment which appeared in the third Calyptopis stage are still present though very short. The telson is very narrow ; there are three pairs of lateral spines along its distal half Post-Larval Stages Twenty-six specimens were found of which ten were closely, and sixteen more superficially, examined. (i) With carapace retaining a long posterior projection. The posterior projection of the carapace is longer in all the larval stages of this species than in the corresponding stages of E. spinifera, and the earliest post-larval stages, unlike those oi E. spinifera, retain it. Three specimens were found, 97, 10 and 11 mm. long. The frontal plate is reduced compared with that of the later Furcilia stages, but its lateral margins are still nearly parallel at the base ; anteriorly they run obliquely forwards to the rostral spine giving the front half of the plate a triangular shape. The margins are denticulate. The rostral spine reaches from nearly to the end, to beyond the end, of the second segment of the antennular peduncle. The posterior projection of the carapace is as long as or longer than the first abdominal somite. The antennular spine reaches halfway or nearly halfway up the third segment of the peduncle ; the lappet of the first segment is strong but not yet bifid ; there is no spine from near the outer distal corner of the second segment ; the dorsal carina of the third is beginning to develop. The spine of the third abdominal somite is nearly as long as the fourth ; very small 294 DISCOVERY REPORTS spines are present on the fourth and fifth somites. The postero-lateral spines of the sixth are still present. (ii) With carapace having no posterior projection or, rarely, zvith a short stump remaining. Twenty-three specimens were found of which seven were closely examined. Their lengths are : three of 11-5 mm., two of 12-0 mm., and one each of 12-5 and 13-0 mm. Of the twenty-three two of the smaller, 11-5 and 12-0 mm. long, retain short blunt stumps of the posterior projection of the carapace. In the younger the lateral margins of the frontal plate are roughly parallel for only a short distance at the base, beyond which they turn by rounded angles obliquely inwards to make the distal part of the plate triangular. In the older the angles become at first sharp, and then a forwardly directed spine arises from each, forerunners of the post- ocular spines of the adult. The margins from the corners, or spines, to the base of the rostrum are still denticulate. The rostrum most frequently reaches nearly to the end, sometimes to or even beyond the end, of the second peduncular segment of the antennule. The antennular spine is as long as or much shorter than the second segment of the peduncle ; the lappet of the first segment is large and erect, and it is bifid in all but two of the specimens, both of them 11-5 mm. long; there may or may not be a small spine near the outer distal corner of the second segment. The postero-lateral spines of the sixth abdominal somite may or may not be present. The telson may have one to four pairs of lateral spines. The sixteen specimens examined more superficially range in size from 12 to 15 mm. Even in the largest of them the margins of the frontal plate are denticulate. Development of Euphausia spinifera Zimmer (1914, p. 429) recognized what he described as larval stages of E. spinifera, compared the younger with the older and showed that the former were the same as Euphausia schotti, Ortmann (Ortmann, 1893, p. 13, pi. vii, figs. 8, 8(7). Tattersall (1924, p. 26) recognized two Furcilia stages and a post-larval stage and compared them with those of £. longirostris. As already mentioned on p. 286 Illig (1930, pp. 504-7, text-figs. 183-190) described as E. longirostris a number of larvae which I was convinced must be E. spinifera because they were all from the sub-tropical zone, from stations where the surface temperatures varied from 15-9 to 17-1° C. I am indebted to Prof. A. Schellen- berg of the Zoologisches Museum, Berlin, for the opportunity to examine some of Illig's material. I saw those specimens from St. 90 described by Illig as ten females and seven larvae, 10-4 mm. long. They are undoubtedly the following stages of E. spinifera: Furcilia, stage III: 2 specimens. Furcilia, stage IV: 2 specimens. Furcilia, stage V: i specimen. Furcilia, stage VI: i specimen (and almost certainly a second with a damaged telson). Post-larval stages, 9 specimens. II I 2 3 6 DEVELOPMENT OF E. SPINIFERA 295 Illig described and figured stages IV-VI as well as stage VII which was not present at St. 90, but not stage III. One of the post-larval forms from St. 90 was smaller than, and differed in other ways from, any I had seen (p. 302). The larval stages and the number of each that I found are as follows : Calyptopis: Second ... ... ... ••• ••• ... ••- ••• ■•■ i Third 5 Furcilia: Stage I. With I pair of non-setose pleopods With I pair of setose, 2 pairs of non-setose pleopods ... Stage II. With I pair of setose, 3 pairs of non-setose pleopods Stage III. With 4 pairs of setose, i pair of non-setose pleopods Stage IV. With 5 pairs of setose pleopods, antennal endopod unsegmented Stage V. With antennal endopod segmented, 5 terminal spines on telson 3 Stage VI. With antennal endopod segmented, 3 terminal spines on telson 5 Stage VII. With antennal endopod segmented, i terminal spine on telson 9 Second Calyptopis (Fig. 40 a) Only one specimen was found, 1-9 mm. long. The carapace is expanded over each eye. The anterior margin is emarginate between the expansions; the lateral margins are constricted where the expansions end behind the eyes. The margin of the carapace anterior to these constrictions is denticulate. There is a very abrupt and high dorsal organ, conical in shape, and a median posterior pro- jection. The antennular peduncle is of three indistinct segments. The antennae, mandibles, first and second maxillae and the first pair of thoracic limbs are present as in this stage of other species. The abdomen is of six segments. The telson is slightly expanded distally ; it has seven terminal spines of which the innermost is the smallest and three pairs of postero-lateral spines of which the middlemost is the longest. Third Calyptopis (Fig. 40 b) Description based on the examination of four specimens, 3 -0-3 -3 mm. long. The dorsal organ is conical, abrupt and high. The carapace is constricted behind the eyes and broader in front, where there is a wide and rounded expansion over each eye, than behind. Between the expansions the anterior margin is emarginate. The margin of the front part of the carapace, anterior to the constriction behind the eyes, is beset with strong denticles, those on the lateral margin directed slightly forsvards, those on the anterior slightly inwards. There is a pair of lateral denticles and a strong median pos- terior projection. The antennular peduncle is of three segments, the spine of the first longer than the sum of the second and third. 296 DISCOVERY REPORTS The second to eighth thoracic segments are visible as narrow encircling folds of the integument ; on none of them are the buds of the limbs free. The abdomen is of seven segments, longer than the carapace. The posterior lateral margins of the sixth somite are produced backv^^ards into a pair of short spines ; as in E. triacontha and E. longirostris these spines persist throughout the later Furcilia stages but disappear before the adult form is reached. The telson is longer than the fourth to sixth segments of the abdomen and strongly expanded distally. Its posterior margin is emarginate ; the spines are as in the second Calyptopis. A fifth third Calyptopis occurred which I believe to be E. spinifera despite the strong differences between it and those described above (Fig. 40 c). It is much smaller than they are, only 2-5 mm. long, and differed from them in that (i) the dorsal organ is small, (ii) the front part of the carapace is much narrower than the posterior and not expanded widely over each eye, so that the narrow anterior margin is only faintly emarginate. I separate this specimen from the other four with greater confidence than these dif- ferences would give me because what appear to be two corresponding kinds of the first Furcilia stage occur (see below). It occurred in a net haul which included one of the four specimens of the third Calyptopis described above, one of each of the two kinds of the first Furcilia stage described below, later Furcilia stages and adults, of E. spinifera. Furcilia, stage I (Fig. 40 d) Description based on the examination of seven specimens, 4-0-4-2 mm. long. In this stage there is one pair of non-setose pleopods. The carapace has a high conical dorsal organ. The frontal plate is wide with broadly rounded corners and an emarginate anterior margin; the entire margin is beset with denticles, the lateral pointing slightly forwards, the anterior inwards. The only part of the lateral margin of the carapace which is denticulate is that along and immediately posterior to the antero-lateral spines. The posterior limit of the denticles is the point where the carapace broadens out into a postero-lateral wing, a point which corresponds with the constriction in the carapace in the third Calyptopis. The posterior projection of the carapace is longer than in the previous stage. The antennular spine is longer than the sum of the second and third peduncular seg- ments ; it is wide with a small number of strong spines along its inner margin (omitted from the figure). The flagella are unsegmented. Buds of the second thoracic limbs only are free ; they are large. Beginnings of the tergal wings of the first abdominal somite are present, standing strongly out with their distal ends curved forwards and downwards. There is a small median posterior spine on the third somite but not on the fourth or fifth. The hinder lower edges of the pleura are not denticulate. The telson is similar to that of the previous stage. DEVELOPMENT OF E. SPINIFERA 297 Fig. 40. E. spinifera. a, second Calyptopis, x 38. b, third Calyptopis, x 22. c, third Calyptopis, small kind, X22. d, Furcilia, stage I, x 18. e, Furcilia, stage I, small kind, x 18. /, Furcilia, stage IV, front part of carapace and antennules from above, x 22. g, Furcilia, stage V, front part of carapace and antennules from above, x 16. h, Furcilia, stage V, telson, from above, x 26. i, Furcilia, stage VI, carapace and antennule from the side, x 14. k, Furcilia, stage VI, third and fourth abdominal segments from the side, x 14. 14 298 DISCOVERY REPORTS Four Other first Furcilia occurred which I beheve to be E. spinifera (Fig. 40 e). They differ from those described above in the following ways : (i) They are smaller: 3-0, 3-1, 3-3 and 3-5 mm. long. (ii) The dorsal organ is inconspicuous or absent. (iii) The frontal plate is proportionately smaller and narrower than the carapace and its anterior margin is nearly straight. (iv) There are no denticulations on the lateral margins of the carapace by the antero- lateral spines. (v) There is no spine on the third abdominal somite. They occurred in three separate net hauls which contained, as well, other larval stages and adults of E. spinifera. They are certainly the same species as the third Calyptopis (Fig. 40 c) described from one specimen at the end of the last section : the frontal plate is very similar. If they are E. spinifera then there are in that species two kinds of the third Calyptopis and of the first Furcilia stage, the larger kind of one stage corresponding in structure with the larger of the other, the smaller with the smaller. I have not found two corresponding kinds of larvae in the succeeding stages, but the numbers of speci- mens found and examined were very small. Furcilia, stage II Description based on two specimens, 5-0 and 5-1 mm. long. In this stage there are one pair of setose and three pairs of non-setose pleopods. The dorsal organ is very high. The frontal plate is not so expanded over the eyes as in the first Furcilia: it is narrower and its lateral margins are nearly parallel. It has rounded corners, is emarginate anteriorly, and is denticulate as in the previous stage. A few denticles occur on the lateral margins of the carapace by the antero-lateral spines. The carapace has a long and strong posterior projection. The antennular spine is as long as the sum of the second and third peduncular seg- ments ; the flagella are unsegmented. The degree of development of the thoracic limbs is as follows : Second limb. Endopod of five segments and setose. A strong gill bud which may be branched. Third limb. Bifurcating, non-setose, with a strong gill bud which may be branching. Fourth limb. A free simple bud. The tergal wings are developed further, the spine of the third abdominal somite is stronger, than in the previous stage. The telson is the same as in the third Calyptopis and the first Furcilia. Furcilia, stage III Description based on three specimens, s-y, 5-8 and ca. 6-5 mm. long. In this stage there are four pairs of setose and one pair of non-setose pleopods. In one the dorsal organ is strong but not very high ; in the other two it is low. The DEVELOPMENT OF E. SPINIFERA 299 frontal plate is different in each of the three specimens : in one it is as in the previous stage ; in another it is similar except that the anterior margin is nearly straight ; in the third it is square with its anterior margin almost straight and with a tiny spine in the centre — the beginning of the rostrum.^ In one there are a few denticles on the lateral margin of the carapace by the antero- lateral spines, but in the other two there are none. The posterior projection of the carapace is strong, as long as the first and second abdominal somites. The antennular spine is as long as the sum of the second and third peduncular seg- ments ; the flagella are of the same length and beginning to segment. The thoracic limbs of the specimen ca. 6-5 mm. long are as follows: Second limb. Endopod of five segments and feebly kneed. A gill of two branches. Third limb. Endopod much shorter than that of second, of five segments and feebly kneed. A gill of two branches. Fourth limb. Short, the endopod of three indistinct segments, setose. A gill of two branches. Fifth limb. A small bud about to bifurcate and to bud off a gill. Sixth limb. No free bud. Seventh limb. A single gill bud. Eighth limb. No free bud. The spine of the third abdominal somite is very strong, half or more than half as long as the fourth somite. The posterior margin of the telson carrying the seven terminal spines is considerably narrower than in the previous stage. The middlemost of the three postero-lateral spines is still the longest ; the innermost is thickening or very thick. Furcilia, stage IV (Fig. 40/) Description based on the examination of six specimens. One is 6-o, two are 6-4, and one each 6-5, 6-7 and 6-8 ram. long. In this stage there are five pairs of setose pleopods and an unsegmented antennal endopod. The dorsal organ may be conspicuous but not high, or a low crest. In one specimen the frontal plate is large and square with rounded corners, nearly parallel lateral mar- gins, an emarginate anterior margin and its edges beset with denticles, but with no beginning of the rostral spine. In the other five it is as above except (i) that in two it is wider at the end than at the base so that the lateral margins are not parallel, (ii) that the anterior margin is only feebly emarginate or nearly straight and has at its centre a spine, the beginning of the rostral spine, standing well out beyond the denticles. In five of the specimens the posterior projection of the carapace reaches halfway or more down the second abdominal somite ; in the sixth it is only half the length of the first somite. The antennular spine may be longer or shorter than the sum of the second and third peduncular segments. The flagella are long and segmented. 1 Note that although this spine may be present in this stage it may be absent in the next. 300 DISCOVERY REPORTS The thoracic Hmbs of two specimens, {a) 6-o mm., (b) 6-7 mm. long, were examined. Their degree of development is as follows : First limb, a and b. Endopod of two segments, longer than exopod. No gill bud. Second and third limbs, a. Endopod long and kneed. A gill of two branches. b. Endopod long and kneed. A gill of two equal and a third smaller branch. Fourth limb. a. Endopod considerably shorter than third, kneed. A gill of two branches, b. Endopod long but not kneed. Gill as in second. Fifth limb. a. A short bifurcate bud, the endopod setose but not segmented. A gill of two branches, b. Endopod short but of five segments. A gill of two branches. Sixth limb. a. A very low bud. b. Bifurcate, non-setose. Gill of three unequal branches. Seventh limb. a. A gill of two branches, b. A gill of two larger and two smaller branches. Eighth limb. a. A very low bud. b. A small single gill bud. The spine of the third abdominal somite is very strong, nearly as long as the fourth somite. The posterior margin of the telson is narrower than in the previous stage, but it still carries seven spines. The middlemost of the postero-lateral spines is still the longest ; the innermost is greatly thickened with a group of spines on the inner margin where it narrows distally. Fur cilia, stage V (Fig. 40^, //) Description based on the examination of three specimens, 7'4-7'5 mm. long. In this stage the antennal endopod is segmented and there are five terminal spines on the telson. The dorsal organ is conspicuous but not very high. The frontal plate is not reduced compared with that of the previous stage : it is still wide and square with denticulate margins. But the rostral spine is much bigger, reaching nearly as far forward as the end of the first segment of the antennular peduncle. The posterior projection of the carapace reaches a half or a third way down the second abdominal somite. The antennular spine may be as long as the sum of the second and third peduncular segments. The mandibular palp is small and unsegmented. The thoracic limbs of a specimen 7-4 mm. long are as follows : First limb. Endopod of five segments, not kneed, shorter than to knee of second. No gill bud. Second to fourth limbs. Endopods long and kneed. Gills of one large and two small branches. Fifth limb. Endopod shorter but kneed. Gill as in second to fourth. Sixth limb. Bifurcate, endopod of two segments, setose. Gill of three branches. Seventh limb. Gill of three large and other smaller branches. Eighth limb. Gill of two small and other smaller branches. DEVELOPMENT OF E. SPINIFERA 301 The spine from the third abdominal somite is as long as or longer than the fourth somite. The telson is narrower than in the previous stage ; it still has, as in earlier stages, only one pair of lateral spines. The innermost of the three pairs of postero-lateral spines is very strong and long — now the longest of the three (Fig. 40 h). Furcilia, stage VI (Fig. 40 /, k) Description based on the examination of five specimens, 7-5-8-2 mm. long. In this stage there are three terminal spines on the telson. The dorsal organ may be high or it may be a low crest. The frontal plate is square with rounded corners and a denticulate margin, a little reduced compared with that of the previous stage ; the rostral spine is stronger and longer but reaches only to the end of the first segment of the antennular peduncle. In three of the specimens the posterior pro- jection of the carapace reaches beyond the end of the first abdominal somite ; in one it is shorter; in the fifth it is but a short stump reaching no farther posteriorly than the postero-lateral wings of the carapace (Fig. 40 i)} The antennular spine may still reach as far as the end of the third segment of the peduncle. The first segment may have a small simple lappet and the third the beginnings of a carina. The mandibular palp is unsegmented and non-setose in four specimens, but it is indistinctly segmented and setose in the fifth. The thoracic limbs of the specimen 7-5 mm. long are as follows: First limb. Endopod of five segments, as long as to knee of second, feebly kneed. No gill bud. Second to fifth limbs. Endopods long and kneed. Gills of three branches. Sixth limb. Very short. Endopod of three segments. Gill of three branches. Seventh limb. A gill of four large branches. Eighth limb. A gill of two small branches. The tergal wings of the first abdominal somite may be simple in shape as in previous stages or they may be becoming rectangular. The spine of the third somite is as long as or longer than the fourth somite (Fig. 40 k). The telson is narrow with one or two pairs of lateral spines. Furcilia, stage VII Description based on the examination of nine specimens. One was 7-8 mm. long and two each were 8-5, 8-6, 8-7 and 9-0 mm. long. In this stage there is one terminal spine on the telson. The dorsal organ is high in all the specimens (compare with the previous stage). The frontal plate may be somewhat reduced, but it is still square with rounded corners and strongly denticulate margins. The rostral spine is much stronger, rising with a wide triangular base from the frontal plate and reaching in all but two specimens to halfway up the second segment of the antennular peduncle ; in one specimen it is much shorter, 1 Compare with the next stage: the reduction of this projection appears to be rapid. 302 DISCOVERY REPORTS in the other completely wanting. A strong crest runs from the dorsal organ to the base of the rostral spine. The carapace has no posterior projection or, at most, a low blunt projection shorter than the postero-lateral wings. The antennular spine may still be as long as the sum of the second and third segments of the peduncle. The lappet of the basal segment is usually large and simple, its margin not incised ; the beginnings of the carina of the third segment may be present. In some of the specimens the mandibular palp is unsegmented and non-setose; in most it is of three segments and setose. The thoracic limbs of one of the specimens, 8-6 mm. long, are as follows: First limb. Endopod long and kneed. A very small gill bud. Second and third limbs. Endopods long and kneed. Gills of three roughly equal branches. Fourth and fifth limbs. Endopods long and kneed. Gills of two large and one smaller branch. Sixth limb. Endopod shorter, of five segments but not kneed. Gill of three large branches and small buds of others. Seventh limb. A gill of four large branches and small buds. Eighth limb. A gill of four medium-sized branches and one small bud. The tergal wings of the first abdominal somite are becoming rectangular. The spine of the third somite is very strong, longer or appreciably longer than the fourth somite. There are as yet no spines on the fourth and fifth somites. The postero-lateral spines of the sixth somite are still present but very short. The telson is narrow with one to three pairs of lateral spines. Of the postero-lateral spines those of the innermost pair are strong and wide, the middlemost are reduced, the outer very small. Post-Larval Stages In none of the specimens I have examined, nine from the present collection and nine from the material described by Illig (see p. 294), is there any trace of the posterior pro- jection of the carapace.! \^ jg retained by the earliest post-larval stages of E. longirostris. The nine specimens of this collection are of the following lengths : 9-5 mm. ... ... I 10-6 mm. ... ... i lo-o „ 2 ii-o „ I 10-2 „ I II-2 „ I 10-5 ,, I 13-0 „ I Tattersall's single specimen was only 8-5 mm. long (Tattersall, 1924, p. 26). I found one even smaller among Illig's material : it was only a little over 7 mm. and had a wide frontal plate with rounded antero-lateral corners and no trace of a rostral spine. This 1 Tattersall (1924, p. 26) says that his post-larval specimen "agrees closely with Ortmann's description and figure of E. schotti" (Ortmann, 1893, pi. vii, fig. 8). But Ortmann's figure shows a strong posterior projection on the carapace whereas Tattersall's post-larva, which I have been able to examine in the Natural History Museum, has none. Ortmann's figure appears to be of a late, but not the last, Furcilia. DEVELOPMENT OF E. SPINIFERA 303 must be an abnormal specimen, for the rostral spine is usually strong in the last three Furcilia stages. The shortest specimen of the present collection, 9-5 mm., differs from the others in the following ways : (i) the frontal plate is square as in the last Furcilia ; (ii) the anten- nular spine reaches more than halfway along the third segment of the peduncle ; (iii) the antennular lappet is simple with a rounded margin ; (iv) there is no spine from the outer distal corner of the second segment of the peduncle. The telson has two pairs of lateral spines. Otherwise it resembles the seven specimens described below. In the seven larger specimens, io-o-ii-2 mm. long, the frontal plate is no longer of the square larval form. For a short distance at the base of the plate the lateral margins are roughly parallel ; beyond that they bend obliquely inwards — to make the distal part of the plate triangular — by corners which are gently rounded in the five smaller speci- mens, but sharply angular in the larger two. It is from these corners that the post-ocular spines arise as in E. hngirostris. The margins of the frontal plate from the corners to the base of the rostrum are denticulate. The rostrum reaches halfway, or more often more than halfway, along the second peduncular segment of the antennule. The antennular spine varies from being as long to being half as long as the second peduncular segment. The lappet is wide upstanding and incised, with two, three or more points. There is a spine from the upper outer distal corner of the second segment of the peduncle. The hepatic spines of the carapace are not present, but the projections de- scribed in the largest specimen (below) can be seen to be developing. The free anterior margins of the tergal wings of the first abdominal somite may be concave as in the adult. The third abdominal spine is as long as or longer than the fourth somite ; very small spines are present on the fourth and fifth somites. The postero-lateral spines of the sixth somite are present in some, absent as in the adult from others. The telson has two to four pairs of lateral spines. In the largest specimen, 13 mm. long, the margins of the frontal plate are not denticulate ; the post-ocular spines are present at its basal angles. The rostrum reaches far along the second peduncular segment of the antennule. The lappet of the first, and the dorsal carina of the third, segment of the antennular peduncle appear fully formed. The hepatic spines of the carapace are not yet formed, but there is on either side of the carapace a large projection which is rounded off on all sides but the anterior where there is a vertical face; from that the hepatic spine will arise (see Fig. 19 o, p. 225). The free anterior margins of the tergal wings of the first abdominal somite are con- cave as in the adult. The third abdominal spine is nearly as long as the fourth somite ; the fourth and fifth are very small. The sixth somite has no postero-lateral spines. There are four pairs of lateral spines on the telson. 304 DISCOVERY REPORTS LIST OF REFERENCES Brook, G. and Hoyle, W. E., 1888. The Metamorphosis of British Euphausiidae. Proc. Roy. Soc. Edinb., XV, pp. 414-26, figs. I, 2. Calman, W. T., 1901. A catalogue of Crustacea and of Pyctwgonida contained in the Museum of University College, Dundee. 50 pp. CoLOSi, G., 1917. Eufausiacei. Raccolte planctoniche fatte dalla R. Nave "Liguria" nel viaggio di circon- navigazione del 1903-5.. . . 11, fasc. vii, Crostacei, Parte 11, pp. 165-205, 3 pis. CoUTlfeRE, H., 1906. Crustaces Schizopodes et Decapodes. Expedition Antarctique Fran9aise (1903-5), pp. 1-9, pis. i-ii. Dana, J. D., 1852-5. Crustacea. U.S. Exploring Expedition under the command of Charles Wilkes, U.S.N., vol. XIII, pt. I, 685 pp. (Folio Atlas of 96 plates dated 1855.) Deacon, G. E. R., 1937. The Hydrology of the Southern Ocean. Discover)' Reports, xv, pp. 1-122, pis. i-xliv. DoHRN, A., 1871. Untersuchungen iiber Bau und Entwicklung der Arthropoden. Zeit. wiss. Zool., xxi, pp. 356-79, Taf. xxvii-xxx. Eraser, F. C, 1936. On the development and distribution of the young stages of krill (Euphausia superba). Discovery Reports, xiv, pp. 1-192. Frost, W. E., 1934. The occurrence and development of Euphausia krohnii off the south-west coast of Ireland. Proc. R. Irish Acad., XLii B, pp. 17-46, 15 text-figs. 1935. Larval stages of the Euphausiids Nematoscelis megalops {G. O. Sars) and Stylocheiron longicorne (G. O. Sars) taken off the south-west coast of Ireland. Proc. R. Irish Acad., XLii B, pp. 443-58, pis. xii-xv. Hansen, H. J., 1905. Further notes on the Schizopoda. Bull. Mus. Oceanog. Monaco, No. 42, pp. 1-32. 1908. Schizopoda and Cumacea. Resultats du Voyage du S.Y. 'Belgica', Rapports Scientifiques, Zool., pp. 1-20, pis. i-iii. 1910. The Schizopoda of the Siboga Expedition. Siboga-Expeditie, xxxvii, pp. 1-123, pis. i-xvi. 19 II. The genera and species of the order Euphausiacea zvith account of remarkable variation. Bull. Inst. Oceanog. Monaco, No. 210, pp. 1-54, 18 figs. 1912. The Schizopoda. Reports on the Scientific Results of the Expedition to the Tropical Pacific. . . by the U.S. Fish Commission Steamer 'Albatross'. . .xvi ; Reports on the Scientific Results of the Expedition to the Eastern Tropical Pacific. . .by the U.S. Fish Commission Steamer 'Albatross' . . .xxvii. Mem. Mus. Comp. Zool. Harvard, pp. 175-296, pis. 1-12. 1913- Crustacea Schizopoda. Swedish Antarctic Expedition, 1901-1903, pp. 1-56, pis. i-vi. 1915. The Crustacea Euphausiacea of the United States National Museum. Proc. U.S. Nat. Mus., 48, No. 2065, pp. 59-114, pis. 1-4. 1916. The Euphausian Crustaceans of the Albatross Expedition to the Philippines. Proc. U.S. Nat. Mus., 49, No. 2129, pp. 635-54, Pl- 83- Hardy, A. C. and Gunther, E. R., 1935. The Plankton of the South Georgia Whaling Grounds and Adjacent Waters, 1926-1927. Discovery Reports, xi, pp. 1-456. Holt, E. W. L. and Tattersall, W. M., igo6. Preliminary notice of the Schizopoda collected by H.M.S. 'Discovery' in the Antarctic region. Ann. Mag. Nat. Hist. (7), xvii, pp. i-ii. Hodgson, T. V., 1902. Crustacea. 2. Schizopoda. Report on the collections of Natural History made in the Antarctic regions during the voyage of the Southern Cross, pp. 236-9, pl. xxx. Illig, G., 1908(7. Thysanopoda megalops, spec. nov. Erbeutet auf der deutschen Tiefsee-E.xpedition 1898-99. Zool. Anz., xxxiii, pp. 54-5, 2 figs. 19086. Berichtigung zu den Mitteilungen iiber die Thysanopoda- Arten der Deutsche Tiefsee-Expedition 1898-99. Zool. T^nz., XXXIII, p. 463. LIST OF REFERENCES 305 Illig, G., iQ'^o. Die Schizopoden der Deutschen Tief see-Expedition. Wiss. Ergebn. Deutsch. Tiefsee-Exped. 'Valdivia', xxii, 6, pp. 400-625. Kemp, S. and Hardy, A. C, 1929. The Discovery Investigations, Objects, Equipment and Methods. Discover)' Reports, I, pp. 141-232, pis. vii-xviii. Lebour, M. v., 1925. The Euphausiidae in the neighbourhood of Plymouth. II. Nyctiphanes couchii and Meganyctiphanes norvegica. J. Mar. Biol. Assoc, (n.s.), xiii, pp. 810-46, pis. i-ix. 1926^. The Euphausiidae in the neighbourhood of Plymouth. III. Thysanopoda inermis. J. Mar. Biol. Assoc, (n.s.), XIV, pp. 1-20, pis. i-v. 1926 b. A general survey of larval Euphausiids, with a scheme for their identification. J. Mar. Biol. Assoc. (n.s.), XIV, pp. 519-27, I fig. I926r. On some larval Euphausiids from the Mediterranean in the neighbourhood of Alexandria, Egypt. Proc. Zool. Soc. London, pp. 765-76, 4 figs. Macdonald, R., 1927. Irregular development in the larval history of Meganyctiphanes norvegica. J. Mar. Biol. Assoc, (n.s.), xiv, pp. 785-95, i fig., i pi. Mackintosh, N. A., 1934. Distribution of the Macroplankton in the Atlantic Sector of the Antarctic. Discovery Reports, IX, pp. 65-160, 48 figs. Ortmann, a., 1893. Decapodcn und Schizopoden. Ergebn. Plankton-Exped., 11, G. b., pp. 1-120, pis. i-x. RusTAD, D., 1930. Eitphausiacea with notes on their Biogeography and Development. Sci. Res. Norwegian Antarct. Expeds. 1927-28 and 1929-30 (Norske Vid.-Akad., Oslo), No. 5, pp. 1-83, text-figs. 1-5 1, pis. i-vii. 1934. On the Antarctic Euphausiids from the ' Norvegia' Expeditions, 1929-30 and 1930-31. Sci. Res. Norwegian Antarct. Expeds. 1927-28 et seq. (Norske Vid.-Akad., Oslo), No. 12, pp. 1-53, text- figs. 1-9. RuuD, J. T., 1932. On the Biology of Southern Euphausiidae. Hvalradets Skrifter (Norske Vid.-Akad., Oslo), No. 2, pp. 1-105, text-figs. 1-37. Sars, G. O., 1885. Report on the Schisopoda collected by H.M.S. 'Challenger' during the years 1873-76. The Voyage of H.M.S. 'Challenger'. Zool., xiii, pp. 1-228, text-figs. 1-4, pis. i-xxxviii. Stebbing, T. R. R., 1900. On some Crustaceans from the Falkland Islands collected by Mr Rupert Vallentin. Proc. Zool. Soc. London, pp. 517-68, pis. xxxvi-xxxix. Tattersall, W. M., 1908. Crustacea. VII. Schizopoda. National Antarct. Exped., Nat. Hist., iv, 42 pp., 8 pis. 1913- The Schizopoda, Stomatopoda, and non- Antarctic Isopoda of the Scottish National Expedition. Trans. R. Soc. Edinb., xlix, pp. 865-94, i pi. 1918. Euphaiisiacea and Mysidacea. Austral. Antarct. Exped. 1911-14. Sci. Reps. Ser. C, Zool. and Bot., V, pt. 5, pp. 1-15. 1924. Crustacea, Part VIII, Euphausiacea. Brit. Antarct. (Terra Nova) Exped., 1910. Nat. Hist. Rep., Zool., VIII, No. i, pp. 1-36, 2 pis. 1925. Mysidacea and Euphausiacea of Marine Survey, South Africa. Fish. Mar. Biol. Survey, Cape Town, No. 4, V, 12 pp., 2 pis. Zimmer, C, 1913. Untersuchungen liber den inneren Bau von Euphausia superba, Dana. Zoologica, lxvii, pp. 65-128, pis. viii-xiv, text-figs. 1-5. 1914. Die Schizopoden der Deutschen Siidpolar-Expcdition, 1901-3. Deutsch. Sudpol.-Exped., xv, pp. 377-445, pis. xxiii-xxvi. 1915. Schizopoden des Hamburger Naturhistorischen (Zoologischcn) Museums. Mitt. Naturh. Mus., Hamburg, xxxii, pp. 159-82, 41 figs. 15 APPENDIX I Table showing the positions at which the species of Euphausia occurred. Almost all the hauls were made with the i m. net, towed obliquely between the depths indicated ; but in surface hauls (0-5 m.) the net was towed horizontally. A few hauls made with the 70 cm. net are marked with an asterisk (*) ; while those which were examined only for E. superba are marked with a dagger (f). 3o8 DISCOVERY REPORTS E. similis var. a I- mat a 1 1 M M M 1 1 M M M 1 M 1 1 1 1 M M M 1 1 1 1 1 1 1 1 1 1 1 E. si)iiilis MMMIlxxlMllllllllMMMMIIIlxllllll E. tri- acantha 1 1 |x|x|x|xx|xx|x|x|x|x|xxxxxxxxxxxxx|x E. loiigi- rostris 1 1 1 1 1 1 1 1 X 1 1 1 1 1 1 1 1 1 1 1 1 1 1 M 1 1 1 ! X 1 1 XX 1 1 1 1 1 1 2 1 1 1 1 1 1 1 1 1 M 1 1 M M 1 1 M 1 1 M 1 1 1 1 1 1 hi 1 1 1 M M E. superbn IIMIIIIIIIII|x|x|xxl|xx||lM|xx|Mxxxxx E. frigida 1 1 1 1 1 1 1 |xxxxx|xxxxxxxxxxixlxxxx|xxxxxxx E. valleii- tiiii xxxxxxxxx||||||||||||||||lx|xxllxx|||l|| ^ kj 1 1 X M 1 1 1 M 1 1 1 1 M M M 1 1 1 1 1 1 M 1 1 M 1 1 1 1 1 1 1 1 1 Depth of net, m. o Tj- ^o O^-OOCO O»2-i-C000000v00 O-OO^hOO-O-O-O-O'-'OO^O0OO-O~O"O-O-O"O'-'O«Oi-.O'^- llllllllllllllllll Illlll Illlll "OO O O OCHf')i-.M Surface tem- perature, C. ^*f^'-' •- M I^ON - r^r] -00 rOO C^OCO N 1 i ^ ^^ ^ ^ ^ ^ ^^ ^ ^ O'^O0I>X^'i-r) OmN r^"+0 -^-lOOO 01 C>0 O lO'-HU-iOO O O O 01 M COO U-. U-, I-. 0 *-* *-•' iriio'^ ~i- -^ "lo 'id u-> in in "^ >^ C> '^ ■- »n in 'tj- "'-rt- o "c^ oo c«o oo 'sO'vor^r^r^f^r^l^r^O'sCO inininm^^-i-'^"^ c/f c/f^f c/fc/f c/f c/f c/f c/f (/} in x' c/f crT c/f x c/f c/f x' c/f c/f oo^ot^inr^r^^O r-^>nr^\o 'tfri^'"c^in^oViciO rn \0 in in 01 O I-" oj m in in -tj- 01 in oi oi oi o n O mo O mrocoin'Oco Ooi r^oi ►- oo^^o in^m>-« N moo O mminininin\D^DOvO^C' ininmrnininmininin 0) Q 1931 13. xi 14. xi 16. xi 18. xi 18. xi 19. xi 20. xi 21. xi 22. xi 24. xi 24. xi 25. xi 26. xi 28. xi 29. xi 30. xi 1. xii 2. xii 3. xii 5. xii 5. xii c N rlN NN N mc^rOfOfOThTl-Tl-Tfiriii-ji/^iy-jioio APPENDIX I 309 E. similis var. armata 1 1 1 1 1 1 M M 1 1 1 1 1 1 1 1 1 M 1 1 1 M 1 M 1 1 M 1 1 M 1 1 1 tq ^ 5 1 1 M 1 1 1 M 1 1 1 M 1 1 1 1 1 M M 1 1 1 M 1 1 1 1 1 1 1 1 M M E. tri- acaiitha lMIMMIIx|xM |MxxxMxx|x|x|xMIMIIM E. loiigi- rostris 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 M 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ■S, 1 1 1 1 1 1 1 1 1 1 1 1 M 1 1 1 1 1 1 1 1 1 1 M 1 1 M 1 1 1 M 1 1 1 1 1 .1 (0 xxxxxx] |xxx| |x| xxxxxl 1 1 1 |xxl |xx| 1 Ixxxxxx E. frigida x|xxxxxxxxxxxxx xxxxxx|xxxxxxxxx| 1 1 1 1 1 1 1 1 E. vallen- tini 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 M 1 1 1 1 M M 1 1 M M 2 1 1 1 1 1 1 1 1 1 M 1 1 M M 1 1 1 1 1 M 1 1 1 1 M 1 1 1 1 1 M M 1 Depth of net, ni. 0 OOOtJ-OOOOO •i-riOOOO-t-'tOO QCC j_-j.r^h-ii-.i-.Mu-i MMrnriroO'l'O'^ h-OOO 0*-0'-'0i-'0'-'0'-''-'0'-O-' O'-^O'-'Oi-Oi-'Oi-'Oi-'O'-iO'-'O'-'OO^-O'-OOO {lllllllllll Illlll'lllll'llll OCO'-'O'+O-l-ONOOOOO-J-rJ -t-OroO'-''+-^OOOOOOOOOOOTt-MinOO-^r^\D I^Ot^OMooOOOcnr^i-' '^'t'^- ri -^ u^uimr^O-^ riOf^O'-t 1- N mOO ^ r^-^-rfN- r^in Surface tem- perature. Oi>rlO'-'OOm ir.-oO'-'Or'^i-'rlsOw. OOOO 0 fX) 0 -l-roOOOO u-jt^f^, l^r-^ioyiM r^i/^Tf-ri 0 libi^b^obb 0 0 I* '<^ '^\ b b b b '-^ '>^ ''^ '<^ 1 1 1 1 Position >: ^ ^ ^ ^ ^^ ^ ^ ^ ^ ^ ^ ^ f ^^ ^ ^^f ^ '^ 0 "'o ^0 vb "^tJ- V^^iri ^ ^N ^M ^0 ^'-' 0 fO O^ '- M 00 ro 00 M ^f^Tf'-tii^Tj-row ONrOMwONNi-HH-.H-foo niD 000000 00 'D 00 000 C' 0000 00 ION i-.yO'«£)'«£)^0 M r^TJ-i-H N w '-' M f^r^f^iM M '^-■^■^''^-nr^mr^ r^N N N N M N M N NM N M c/f ctT c/f c/f c/f c/fc/f c/:" c/f t/f CO c/5" c/f c/f c/f t/} c/f coc/f t/T c/f M\D m>-> >-" f^f-. in li-iCO Tt-sjD ino r^\0 f -^^o r^o M ^f^„ mow -' -4-^-^N 01 f^n inro-^miow 00000 'TOO 0 i""i nOOOOOOCiOOO 0 000000 r-s£) m r^^^^f^\ri\r^o^>-' !1'^^'^ 0 inii-imioiotrju-) u-iii-jiomminmio^OOsOO^O Q Pj'x 'x 'x 'x "x 'x "x 'x !^-- .-H .- •- ■- ■- ■- •-- ■- ■- ■- c 0 0 M ro m so r^oo 0 inr^c-'- m-+\Doo 0 i-'nu-»o 'sOO'O^O^OsO^^ 0000 0 0 0 0 '-' —"-I *-* H- t-- r- r- r- r-. t-- r- 0 i>t^i>oooooocoGoooooooocGO 3IO DISCOVERY REPORTS •2 5 E. simil var arma 1 1 Mill 1 1 1 1 II II II II 1 1 i 1 1 1 1 1 X 1 X X X 1 X X 1 1 1 X 1 ^ 1 1 Mill 1 1 1 1 M X 1 1 M 1 1 1 1 1 1 1 X X X X X X X X X X X 1 1 X X E. tri- aiitlia 1 1 Mill 1 1 1 X xxxxxxxx 1 X X 1 1 X 1 X 1 1 1 1 1 1 1 1 1 1 1 1 ^j c ■ -a ^■|^ 1 1 II II 1 1 1 1 1 II XX M II 1 1 1 1 1 1 X 1 X 1 X X X 1 1 1 1 1 1 1 1 O C ^ 1 1 Mill 1 1 1 1 II 1 II 1 M 1 1 1 1 1 1 1 1 1 1 1 X X X 1 1 X 1 X X ^ « -o f^ 1, X 1 X X X 1 1 X 1 X X II X 1 X M X X X 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 t^ 1 1 II II 1 1 1 X X 1 X 1 X X X X XXX 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 < ^, fe3:S-S 1 1 Mill 1 1 1 1 X X X X 1 1 1 1 1 1 1 1 1 X X X X X 1 X X X 1 1 1 1 1 1 1 ^ c '- p [y 1 1 1 Mill 1 1 1 1 II II II II 1 1 1 1 1 1 1 1 X X X X X X 1 1 1 1 X 1 X O o ^ o o o o o 0 O o o o o o o X 4-r m o o O -t "t o „ „ ° „ „ ° o o o o ■^ „ -^ CO Dept of ne m. 0 ^ 1 1 ri O 0 O O O " 1 1 1 1 1 I^ in O O 'T V T r^ o u omoooo"0-' 1 1 1 1 1 1 1 1 "Ooor^ooo 0 O " 1 1 1 r^ v£3 O O O " 1 1 1 m N o 0 w in O V 1 t^ o tl n V 1 O 0 V r in O O O " 1 1 1 moo N ro vO t^ O - -t -t >y-. O ■^-u^Tj-r^HH in m \r t^ Tj- U-, KH o i^ N ir r*l ir\ en in O in tn N inoo in * ^ •-I N M -H W M •^ m ■- en M fM M rj hH r^ M N w HH r( w N M N •1 N " N HH r- f N ^. f^. 0-. rv. r^. urface tem- rature, ° C. ^- 00 to M c ro N O O Tl-OO in o 00 0 M O o •+ M Tt-^O 't ^ >-• CO M " r-2 •t O li-j 0 O r^ N CO M « vO „ M ^ M M o n r^ \0 in c^ n n •+ t^ o o n ■4- sO Ov O 00 1 1 1 1 1 "- i-i ^ »-t '-' M " « ^ ^ ^^^^ ^ ^ ^ ^ ^ ^ ^^ ^^ i? ^ ^ w W w t^ o r^ ri M ri O ^ a o V ^ ■I- V ^ 't- O sC o o lO r^ u-j ro »n O C) Tt CO N O OS O in r- o o r) o o o o o Q .> ro in o N CO o o o O fO rt- iTiOO C^ sD so in O ■+ O in M N "-* o 0 o G o fO OS O M M N CI CO en CO in in 'tJ- CO ro c^ rl -• o> ^ o '^ 00 '^ ^ N o crT cT) Oi tn C/} in m CT! C/^ (/2 C/} O! m ^ (Tj M M !n m C« CA) M C/2 Dh o i-H tJ- irj ■^ t^ i-. tJ- ro M OS cc t^ r^oO ^ sO •^ M sO m 0 CO »n 11 N -^ in >/"< n ri- T^- C^ HH in 1-1 -i rl -t in o M ■i- 0 ►H O o O 0 0 o o o 0 0 o o o o o o o o 00 r^ in CO OS vO M i-H N r^ ro M OS in ^ N OS r^ sO in 00 sD *0 sO O ^O IT) m m in m m m in Tj- rf ■* ^ -^ n c^ ro m ro o > _> ■- ■- .- -- .- .- •- •- ■- •- ■- •- Q ? <> 0 0 O r^ t^ 00 t^ 00 o 0 M ro in t-^ t^ 00 as M M m 03 OS M M N M M w rj M W h- N M M N N N M M c o ^ \n oo OS O ^ cn -^ in\D I^ 00 Cs 0 M N rj- in M N N rn f^ f^ n m m ^ ^ 'J- ^ * tn oo 00 00 00 00 00 00 00 OO 00 00 00 00 00 00 CO 00 00 CO CO 00 oo 00 M APPENDIX I 311 fcl g 5 5 liq bj §=§ lij s fc^ o X X X X X X X X X X X X xl |xxxxxx|xx XXXX XXXXXX I XXX X X X X X X X XXXXXX XXXX xxxxxxxx xxxxx 1 xxxxxxxxxxxxx XXXX XXX XX O ^O >£> I^ c^ o 0 M O " O " o 1 I I I I I I 00 O O O t^ O " w r^ M M *-. M o o « O " o III' O 00 0 M o o o 1 I I I I I I I. O O ^ ir, O C^ O C^O OCC mO O^^ I I I Tl ? r ? -i-MOOOO H c*lOGOO O Mt>0^OOO^OC>OMO c^ o 00 o ^ o ^ I I I I I I I ' O O vO O CO -t oor^oMO^o^c^C'CO PI — fo — r-l o N 6 6 o 00 W w w w w N w w w w w w w u u w a w w w w CD 00 0 M l-l 't ^ 0 °o 0 0 00 XI M Cfi CO c« c« CO V ^ PI N "0 CO 0 ^0 "0 0 IT) f-l ■+ o o 'J- CD m Cfi C/2 C/2 C/2 "0 "0 ^Os 0 IT) CO "0 '0 0 m 0 in CO 00 M > ? 6 > 4 On > 4 > vO t-- CO 0 't -^h ^ 't 00 CO CO CO 312 DISCOVERY REPORTS ^ S 5 S XXX xxxxxxxxxxxx fcl XXXXXXXXXX XXXXX XXX fe^ XXX XXXXXXXXXX b5 t^-? XXXXXXXXXX XXX X X X X XX XXXX XX xxxxxxxx xxxxxxxxx . s ■ XXXX XXXX cq XXX XXX X X X X X X O O O O vO o o mOMOO^OO 0"OmOi-.OOnO"0"OmO I- J. A i _L i I LI i J I I I I O 00 O Ov o o o O^000 OOO O u->0 O CO o o o M m m o ^ '^ Tj-oioo^^fM oco M o I O O O 1/-1 -i- O O O O N O ■-■ -I 1— -f— O MOOOOO^OOO^-^OMOwOinOOv ^ ^ .. I I I I I I I I I I I I I I I I I O^^HHOO'-'OO^MI^wOOroOOOrorl C>Tf rj M M HH n ^ rl n « r^ K^ N -H fO M ^ M i I I I •I- O o Cfi ■"I --' c ;. o o o o o o o u o . o o o o Or^"^t^00O-*-i^vOt^a>OM-:hir)r^O COOOOOO-""">H«NMMNNm C/3 CO C/f C/} M t/f C/}' (/f w" en' '/)" 7i' C/f r/f Ol uS '£ (A tn \f^ \0 N N O C> rl M r^ r^ o c^ -1- 00 "^>-' Vl ^M 01 inir^Tt-roO '-' ^ O u-)u", o u-jino in^ tJ- O O O O O G -J O O CI O O O C' n o o o OvvD fO»- Or^^moo O rol^ONsOOO w ro 1^ 00 Ov vO ^o -o 00 00 00 N r^ 1^ r^ r^ i^ X 00 O) 0 >H > > > > 1> > > 1 l-l M f^ ^ tr^ vO 0 N M 01 01 01 01 01 N o o n ro 'l- ■o l^ OC CO 00 00 CO oo oo 00 00 CO 00 CO CO 00 CO APPENDIX I 313 E. similis var. armatii ||l|||||||xxxxxxxxxxx|||lMIIIIIIIIIIII|x E. similis ||||||||xxxx|xxxMxxxx|IIIMIIMIllllllxx E. fri- acaiitha xxxxxxxxl 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Ixxxxxl lxx|xxxx| 1 E. longi- rostris ||||||xxxxxxxxM||x|xx||x||||||MI|||||xx MIMMIIMIIxxxxMlllllMIIIMIlllllMII -0 1 1 1 1 1 1 1 1 1 1 1 M 1 M 1 M 1 1 1 1 1 1 1 1 1 M 1 M 1 X 1 1 1 M 1 xxxxxxl 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 |xxxxx|xx|xxxx| 1 E. vallen- tini 1 1 1 Ixxxxxxl 1 1 1 1 1 1 1 1 |xxl |xxxx| 1 1 1 1 1 Ixxxxxx llMlllMlxxxxxxxxxllllllllMllllMMIMx Depth of net, m. 0 0 oo"-. 000 oooocxMO 000 0«0000-MM -t \r, -t n m -t " '^_0_" 0O^0MOO^p«O"O"0"0"0"000-0"0-0"0"0■-0" 0>hO--0'--- iiiiTtYiTiiiiiiiiiiiiiiiiiiiiiiiTimTiiiii oOooO"Or^OOO"i!noot^ 1^00 ot^oir>ooo-0-:hO-t-OOOvDO oooinO'tO Surface tem- perature, ° C. OOO^O>-OCOMCOMM'-M0O"OCO0OO5";J- 0 6 r-)iViKc>^ fofoo^b^ ^^ 0 ^ 0 -HH- (\i r^ 1 1 1 1 c _c C 00 =0 "0 "0 ""^ ^0 CO 0 Vi "■- CO "- "m 0 "" 0 £; '^ rp r? rl °r-> 'in "r- 00 °o °« 00 % 0 0 " °" ' M °^ •* "^ x "0 "m "jri c/T c/f t/T cr c/f c/f c/f c/f t/T M* t/J c/f m" c/T c/f crT m c/fw" c/5" w ~in V ^f^ 00 "K ^1- V, ^ ~in OD "K V Vl f^ M IT) 00 NO "2 Tj-0 0'^r^c*^'-*0>^^NNf^'-'^N'^0»/^-^r^ 0 0 0 0 0 0 0 0 0 0 0 0 0 ' 0 0 0 0 0 0 M O^vO N C>\0 n>- rot^O-N r-10 <>" " ^f^^" \0 ini/^u-i-t-+-h-|-T|--*--:t-min U-. in^OO^Ou-iuowl Q N p. ;> J, >■>•>■>■>"> ■> ■> ■> ■> ■? ■? > > >> > ^^ 2~ d- d o' ►-■ -■ N r^ lA lA 0 ri c- d 0 -■ -\ r^ 4 U-. vc r- " N N (^ f, „ „ „ „ M « rl rl N M N M M N c 0 '7} OVO " N C^Tt-int^OO C3^0 M ST'^'iC'^S " '^'^S M 00 o^o^o^c>o^oovco^o 0 0 0 0 0 " "^li" ooooooMoooooooococo o-o>ooo-. oc^c^o^o■c^ 16 314 DISCOVERY REPORTS tq •^ ■ ■*-» X « K a > £ X X X X X fci i X X X X XX X X X X lij X XXXXXXXX XX X X X X lii X X X X kl' XXX tj X X X X XX X X X X X X X X X X ^^. X X X X XXXXXXXXX fcl X X X X X X X X X X XXX M OO O O O O O O'-OwOC>O-'O'-'O'-'O'- cc NOoo^o N mo Or^O "^ N c^O -^O^N wOOOO OntnM mO li^O r^'-' '-' ►- O w O '- O ' i ' i ^ o r^ o M -I H- M en r^ I I ! I I . . o r^ o o o r^ o m ro o ^ o ^ ^ Oh.O'-O'-'Oi-'O I 1 I I I I I I I oo'-'or^ooNioo 0\0 O^ ^ C^ O O N O^ ? r ^ F Bc) o t- o cn CO i^ ^ ^ r^ o o o o -1- 6 I I I I I o a, w w ■i- W w w w w ^ ^ ^ ^ v£5 ^ ^ s:^^ ^^ ^ ^ ^ O N o rX X C/2 X X X "o ^M % 0 0 0 °^ °? o o 00 en O N CO X o ^ o o X XXX XX X M CO Q 00 ■ id .ii .ii OS d c- m 1 OS 00 Os OS OS .2 .s d d CO OS Os tj- inso r^oo Os m w~j »o ir^ m U-) Os OS Os Cs OS Os o OS so Os APPENDIX I 315 tj ■-: I- XX lij XXXXXX XXX xxxx XX feq xxxx xxxxxxx I Ixlxx fcj ^^ xxxxxxx X X XXX CiJ xxxxxxx fq xxxx xxxxx ki: xxxxxxxx XXX xxxx xxxx kl xxxx X X XXX OOOONOmOOOOCOO-^-t- Occ o OMO<^0^00"MMn"0-t'*-0 O" I I I I I I I I I I T I I I I I I I I I I I I I I I I I I I I I I I I I " I I I I -t-or^ooOMOMOoooooO'-'Or^oooooomTi-r^orooocor^ooi^'^o^oroo rir('-(N^ir)Nt-tO>ni-'Ooow*oomTi-c^000N0'-'O ro C) HH M r*) N N N M M N " - r^ N S2u 3 00 CO 0 \0 ^O 0 CN 0 0 " \0 0 0 0 M ^- rf iTi Ifi U^ \fi ^ o Oh ^^^^^^^^^ o o 'I- O o ^ ^ ^ ^^^^ ^ OS o M m M cfi !/3 "0 w "o 5 ^ CD !X) O Q m \0 0 0 OS M CO 0 c^ Tj- in 0 ^ M 'V OS OS oJ °0 00 0000 N t)-sO sO sO ^ sO t^ 0 OS M W M Cfi Vi'Simm C/2 ^0 CO 00 "0 "- ^1- "0 0 0 0 " a 0 OS 0 0 in 0000 M M i-< in m m U-) 10 10 3,.ii .2 .>< .>< .i; .a .a .•:• .-i .>:• .>< .S .2 .S .2 x x x i^ 1^ x x -"-HMMi-.h-ih-.rjMMMriMNr^ 1-.J-.1-1MN c 0 M m •^ in S.O ^^ sD sO sO so so o> OS OS OS OS X OS sD OS o OS O M ro in r^oo 00 00 CO OS Os O OS O' 16-2 3i6 DISCOVERY REPORTS fe) c « g ttl 5 !JJ CiJ C"* -S fe;-^ t^^ fcl o. c O I ^ n r- 3 . • 3 ;i U D ^' a. o X X X X X X X X X X X X X xlxl Ixl Ixlxxxlxxxxxxxxxxxx X XXX X X X X X X X X X X X X X X rl- O - Th O « o r^ o iH I I I I ' o 0 o "^ w 1 I I I o o N N o O "^ in o o o -^ in O O I I I O " cc 0 '^^ >n O 1 I I I U-) O O O o 0 O " o o 1 I I I I m ro tJ- -^00 o- 0 " o 1 I 1 o o o M rl n CO m o o cc M ro I I I O 00 •+ o I I o ^ ^ ^ ^ ^ ^ ^^ ^^ ^^ ^ ^ 00 O in o o CM o 0 cc o o CO M X ry2 'Sj cn "o o in "o ►H i ^ o in in 3> 0 so OS 00 OO in CO rX c/) cn M Cfi 0 ■+ in so io o sO vO o o vO vO vO vO o -*• ■^ ^ OO O O 01 0 o 01 " CO o 0 o CO SO in sC rosO so in o o so sO in in so in Xj Cfi cr. rji M m Tj C/2 o in rn o in sO sO o o en --• sO so o o « 00 sO m in so 01 1< X oi sO X X X CO X so CO o OI 't on 00 OS o> Os OS OS On OS OS so OS OS OS OS 00 OS o> o> 0> OS OI m on Tt in n O - o O O o O OS o APPENDIX I 317 E. similis var. armata 1 1 1 M 1 1 1 1 1 1 1 M M 1 1 1 1 1 1 M 1 M M 1 1 1 1 E. similis X 1 X 1 1 1 1 1 1 1 1 1 M M 1 1 M 1 1 1 M 1 1 1 M 1 1 1 E. tri- acantha xxxxxxxIxMxlMM lllllxM llllllxx E. longi- rostris X 1 1 1 1 1 1 M 1 1 1 1 M M 1 M 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 M 1 1 1 1 1 1 1 M 1 1 1 M 1 1 M 1 1 1 1 1 1 1 1 1 1 .1 1 1 1 |XXX|X|X|XX1X|XX1XXXX| |X|X| IXXXX XX |x|xxxxxxxxx|x||| x||xlx|x jxxxxxxx E. valleti- tini X 1 X 1 1 1 1 1 1 1 1 1 1 1 1 M M M 1 1 1 1 1 1 1 1 1 1 1 1 ^ 1 M X 1 1 1 1 1 1 1 1 1 1 M 1 1 1 M 1 1 1 1 1 M 1 1 1 1 1 1 Depth of net, m. oo-no-^o 00, 0 0 o#o-t-o, oI?o^o2on>o2o2oo"0-"^o2,no2,no"&oSofo2o2 y,-S L^iiiiiii^i^i.^iVi..^,oL4ii/.^iii>i-Vi^i^^^o ii :p,?^2?>2g,:rp;::!:r^2!:r;:'P;2=^;c »^ 2;?S=^??"?;2o?::;2; Surface tem- perature, °C. f»50o4-M"MMr->0^0 N -|-pC>vpOO in i^;nM"b6"-"6M " mmOOO" 1 1 1 1 1 1 1 1 1 1 1 c .0 S ^2 V ^^ ^S> \ \t \ t^^ ^S > ^S.^§^ ^S > >! V -„ °r^ V V V V W °o °r-i N =1- i'o 't- \ "rj t£ ifi rn rn en m" uS en uS muS t£ c/f in in in m en en "0 00 ^0 ^" ^ 00 00 Co "0-0 -- "" "" "g" ~^ :i t ° i? °o 0 °" V 0 °ov ■ 0 °- °" °" °o °o a> So °r- h "c^ N °cv 0 Q 1932 16. xi 17. xi 18. xi 19. xi 20. xi 21. xi 24. xi 24. xi 25. xi 26. xi 26. xi 27. xi 29. xi 29. xi 30. xi 30. xi 1. xii 2. xii 1933 12. iii d ,0 OOOOOOOOOOO 0 000000 M APPENDIX II Table showing the stations at which the larval and post-larval stages o/Euphausia frigida, E. vallentini, E. triacantha, E. longirostris and E. spinifera occurred. 320 DISCOVERY REPORTS .2 " c ^ " 5 3 s t5 I \ ggEJS uospj uo sauids (Buiuuai z A sSeJS uospi uo sauids |buiujj3; t" AI aSwS uosi-ij uo sauids (Euiujjsi 9 'p^iuaiu -3as podopUS {FUUOIUV uospi uo sauids [BUiuuaj C 'pajuniu -3os podopuo icuuoju^- uosi^i uo sauids iBUiuuaj S 'pa^uoiu -3asuTi podopua leuuajuy III 3^K1S II aSeig spodo3|d asojas-uou S I 33BIS spodoa[d asoj3S-uou £ rj O O '-' rj O O C> •1- M m --I i-ii-HNU^O'OrJN>-^sD-i- I N \D -i- r^ 1 i^. n, cd u pjjqj. puooag rrt 00 tr, o lO o O O o o N f^ rn O <) () O O O o o o o o o *-< o o o •-' o o f^. O O " 7 O ir^ O o in o o o r f in V in "-■ ^- M o o Tj- ^ o o »o 1 o o <^ O >n O 0 O 0 O o O o 1^ r^ O 0 O o o o O o o O r^ Tt- o n N H- ■too r^ ir-j o in o irj o •t o O 0 Tf- in in in o o o ^0 00 N m in in O M m o en fX ^ X X >< X ^ f*^ <^ Qs 6 i-^ '-' N X X X >6 r^H^od r^od d\ ID \0 r^ 00 On ^O O O O 'O i^ r^ r^ r>- r^ ^ ^ I-. fH u-iO lOvOOO O^ O <*! »0 'sO >-' N M N r^ccoooooo o o o O Ov O N o M N M O 0 en O (O o CO o in O APPENDIX II 321 2 ~ c « "1 SaSEJS ]BAJEl-lSOJ <3 t5 I A 3S«»S uospi uo sauids (EUIUU3J z ^i]\\ \ aSejs uospi uo ssuTds ]Buiiujaj +■ i^l}\\ •* Z O CO -^ M CO M Tt- t^ I " O O 10 -^ ^ f^l 01 rl m \I 38EJS III aSEJS a o a _>. U II aSEjg spodoaid PSO;3s-uou z '3S013S e I aSEJS ':h n O " M O 1^ O I O "^ fO N \0 I »rj CI t^ O ro '-' ^ PJ'MX puooai^ K^ lO M N O I 0 m r^ ro 0 7 '-^ 0 0 in 0 0 ^x> •i- 0 N " " n ^ in ^ 3 ?i t- o rn l' CD Q, o PL, " n o -t " n o ^ i^ to io u^ ^- "-< ^ in o^ o 10 xfirjimiTl rf Ti- C^ m Tj- m in i^ r^ in C^ 00 I>^D 00 rj moo — 00 00 O O ^^ ^ ^ ^ ^ ^ ^ N O O M ■* n o t^ ^ -J- ^ ■+ "^ Cfi m M en W M C/3 ro O N N o o O coco o o o 10 o \n -^Ti \r\ \n \n x X y. y. y. \0 t^oo o •*■ -, ►- - N rl rl M ro lo 00000 322 DISCOVERY REPORTS .53 t5 Fraction of catch examined 2 U-Jin _,|-(MI-lh-lMMM.-l)-( • 2 3 IIA 33EJS 111^ '-^ 1 1 11 1 " 1 11 1 1 1 1 1 1 1 11 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 i 11"'^ uosjai uo sauids iBuiuuai c 1 1 1 1 ^ 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 M 1 1 1 1 1 1 1 1 1 1 1 1 1 1 lA 33e;s III" -^ 1 1 1 1 " 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 11 1 1 II 1 1 i II"" uos{a) uo sauids [cuiiuj.->i t- 1 1 1 1 1 1 1 1 M 1 1 1 1 M M M 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 " 1 1 A 33EIS Mil:- ^ 1 1 " 1 1 1 1 1 1 1 1 1 1 II II 1 1 1 11 1 1 II 1 1 " 1 1 1 11 11-^" uos[ai uo sauids lEUiuuai 9 1 1 1 1 ^ 1 II 1 1 1 1 1 1 1 1 11 1 1 1 1 1 1 1 II 1 1 1 1 1 1 11 1 1 1 1 1 1 " 1 AI ^Se^s III?? ?;llll""IIIIIIIIIIIIIIIIIII^II^IIIM "1^1 uospj uo sauids IEUUIU91 s 'pajuoiu -33S podopua lEuuojuY II 1 1 " 1 1 1 1 1 1 1 1 1 1 M 1 1 1 1 1 II II 1 1 1 1 " 1 1 1 1 1 1 1 1 1 1 II III sS^lS Imlm i^lroNI 1 1 I I'-'l 1 1 |-^|'-'rM'-''^| 1 1 1 1 I'tNlr^l 1 1 |i-i 1 Iw^h 1 If-. w| lllll llll 1 fn llllll^ I"-"!!!! II II 39Bjy MM 1 MM MM 00 1 Mlr-'-H..^ 1 MM spodo3[d 3SOJ3S-UOU c *3soaas e MM 1 II II M II " II II II II II 1 II II II M II II II llll spodoajd asojas-uou S llll II 1 II M II II 11 II II " II M II 1 1 "" II II II 1 llll I sSejs 1 1 1 1 INI 1 1 t 1 1 lu-iroNwIN^OOt^lM^Nmin^'HhHrowlM^O'H-t--^ 1 1 1 1 llll 1 lllllll 1 On|" N C>nI "llll ■q. O 1 U PJ^MX Th|«| ||||||l«iM^inNMlT|-N||iol|coO""iriin||i-,||f-5 |||| pUOD3C^ II II II II II M' " 1 " '^ II II M 1" II " "^ II " ■" II " li II llll Depth of net m. 0 0 OnOO 0000 in 0 O-tOO " 0-l-*i--HO 00 Lr>OmOOrnOOOOrnOOO O'-'OO OO'-'OOO'-' 1 O'-'OOOOO'-'i-iO-i'n'-iO'-Oini-'OO^i-iQ— OOO-'O- OOOO llll iiiiiiAiiiiiiii (iiiiiiTiiiiii [III NO'^O 0'-'OOr^'Oi>ir^OO'-'000^0!?'NOO>nOOOOOOOC>00'-'OOOrnO<-nO O^OMr^ OOO-^O N^-int^'-inM^ONCOt^C^wNinNOOOin'-^mnovn^sOinNONOOinCvTi- -'OOon £ :: - : -„ - = r ^ :: :; :: : = = = :: " :: V ^ : V ' ?" - = V - ? - V ? ^ V ^ = " : = :::: Surface tem- perature °C. i- i-CNmor^ OO 0 -t- nin 0 n ^ r^r^OOO coincor^ r^oo n- oc^ o^oo^-oo^ '^ '^ T*" r^ n- go -+ m msO-Hro ^NOro rir^oin r^-+ lof^rnrnrh in-h m ^h on -^ •- w Tf-oinV c 'en O ^^^^ ^^ ^^ ^^^^ .^.^.^.^.^ .^.^ .^ ^ 'f'^ '^ ^ ^ ^^^^ ^-rh^N'O^N ^ov^r^N'-' OOO inOOt^M^ O'-' c/D >-' Nin i> 0 ro '-'Oo'^tn^H- OhHfHN hKin mm ^O o-^mmm N'l- m m mm -. ^ n m^Nm ,-, rjoO no 00 00 OOOOO 00 0 0 00 0 0 0 iDOOO rno^r^ Ntn O'^ wO soo^C'C^ln in^t- rt- *}- "^"^ r^ r^ r^ oo^oom I^-^mm Nin m-^ moo r^r^r^i>in u-itJ- ^t- tJ- -^m m m m \OinTi-Tt c/f c/f c/f c/f c/f c/f c/f c/f c/f c/:" c/f c/5" c/5" c/f co c/f" c/i c/f vf c/f t/T c/f c/f c/f c/5" t/T c/f" cc ^ ^N ""in 00 "in V "m ^- 00 00 "o "^O O ^O 't *^ 9 "^ ^7 \0 ""ov "m ^•-' "h^ ^ 00 "m "m mO-Tt- N'J- ^ ^ N^ w^mm^ NO m ^ mO O O rf rfo-N OOOO 00 00 00 OOOOO <-J 0 0 0 00 0 0 0 C'OOO ONinO mM mn inrn inC^O"-""^ •-'O n in \o >-^ 0^ m m r^mNN sOinmrn inm min ^in inm-o^Oin mm in in mm ■^ in in inmmm Date '-'^ONN^o'-^Ot^ 06 & t^ci m\Dr^i>dN moo Cn 6 dm 4- 6 '-^ ^t^ooo-^ ■"(^^w^^^lifHhHKHN NNNN ^H-. wN N ^ ^ ^ ^IN c o 'Si Mmc^in moo OO \O00 mooc>ooc ^-o co 0 "-"m in i-i n r^o^min min^r^ ON Nm mr^ooooooo-" ^ ^ Si '^ mm in \o sO •--f-NN r^r^t^i>oooo 0000 000 0 0 0^ 0 0 00 0 0 00 0 0 0 ^ ^ ^ ^ APPENDIX II 323 5<) t5 S3§E5S JEAJEJ-JSOJ 0 - l^ 1 1 1 CO M N u 3 IIA aS^S 1 1 1 1 1 1 " 1 1 I A aS^lS 1 "^ 1 1 " 1 " " 1 uospi uo sauids |euiluj3i t- 1 '^ 1 1 1 1 1 1 1 A aSEJS 1 " 1 "Mill uosiai uo ssutds iBUTiujaj 9 'pajuaui -33S podopua (Euuaiuv 1 " 1 1 1 1 1 1 1 AI a^ElS 1 -*■- 1 1 1 1 1 1 III sSi^lS 1 1 M "" 1 1 1 II sSeis •"" 1 1 1 1 1 1 1 I 33BIS ^ 1 1 1 1 1 1 1 1 '5, 0 a P-II1X ?" 1 M 1 1 1 1 pU039S °^l 1 1 1 1 1 M Depth of net m. 8 ? ? T t^ N 0 c^ ro U-) i-i h-. 01 0 N 0 0 w 0 m '-' lil i-H J> tJ- CN Ov 04 w o» S :;::;:.- ; Surface tem- perature °c. 0 00 6 N "i- io f^ 6 6 6^ 1 "■« V irj M 0 0 ^ 0 0 is V in 0 0 0 N M w w w ?^ ?" ? ». V V -^ ^O 0 0 »i^ tJ- N M ir^ <■) 0 N 0 0 0 000 f*l iri cs ^ I/-. 0 0 0 m N M M H M c/f c/f c/f c/f c/f c/f N 0 " in Th M oooooo O^ moo CO H. \0 f) ■*- ^ -^ ,j- Th D M hH tJ- IT) vo rj c 0 ■^ ■M CO c^ 00 00 0 r- 0 0 0 ^- CO 00 00 GO 00 00 "3 SaSBlS |BAJBI-lSO(J =° " 1 1 M 1 1 IIA sSws ^1 1 1 1 1 1 1 lA aSEis ^ 1 "Mill A a^^JS '^ 1 II 1 1 M AI aSEjs ^1 1 1 " - 1 1 III sSeJS 1 1 III II aS^lS 1 1 II " 1 " 1 spodo3[d 3SOJ9S-UOU Z '3S013S I 1 1 II 1 " M M UEUIS 1 1 "^"111 ggjEq 1 1 " 1 M ^ 1 CO 'a 0 a _>. 0 T3 IIELUS 1 1 "Mill 3§jBq 1 1 " M 1 ^ " puooag 1 1 II M " 1 Depth of net m. 0 tn 00 .^ 0 0 0"0"00 1 i i 1 1 1 1 1 0 0 ot^c^^<^c^ 0 0 CO^fMOOOOO •-( n >-( 01 I-" 00 a e g ; - - . . 1 1 , -. ^ - * ^ W 01 hH Surface tem- perature °C. u-1 GO NT^roOO^^ (^ 00 00\O>-'lSr^00 io 00 ^ocbinoo^K c .0 'co 0 0 0 v^CiOOuio 0 OS in Oo 0 0 0 0 *^oo^DT^N 0 >0 oO>nolo)N i-H tJ- OONNmww c/f c/f c/i" c/f oT c/f c/}" c/f V % \0 00 V ^^ ^0 vD w m TfOOOrooo 0 0 oooooo \0 "-I I> 00 ro i-i N in Date CO "^ c*^ "^ en'" ■"■ '"* "^ '^ "" M -, „ w 04 0 eg sO f^ >~tmf^r>-ooo -t i> OOOOOOONO^ON o I [Discovery Reports. Vol. XIV, pp. 325-350, plates I-V, June 1937.] THE REPRODUCTIVE SYSTEM OF EUPHAUSIA SUPERB A By HELENE E. BARGM ANN, Ph.D. CONTENTS Introduction pase 327 Material and methods 327 Acknowledgements 328 External Characters 328 Reproductive System 329 Previous work 329 Male reproductive system 330 Development of the copulatory organs 336 Female reproductive system 2^2 Fertilization -145 Summary ,^8 Bibliography . 34^ Plates I-V following page 350 THE REPRODUCTIVE SYSTEM OF EUPHAUSIA SUPERB A By Helena E. Bargmann, Ph.d. (Plates I-V ; Text-figs. 1-26) INTRODUCTION THIS account of the reproductive system oi Euphausia superba forms the introduction to a further paper on the development and distribution of post-larval and adult krill. The composition of any specific population is estimated not merely by length measurements, but by the degree of maturity to which the individual members of each sex have attained, and before the life history and distribution of a species can be described, it is essential that these facts should be ascertained. In E. superba, external sexual characters appear first in half-grown specimens, but in a great number of such individuals these characters are not yet developed. In this investigation, therefore, to ensure correct sex determination, all the specimens both large and small were measured and dissected, and the degree of development of the external and internal genitalia was noted. Some new anatomical details resulting from this intensive investigation are set out in this short paper. MATERIAL AND METHODS John (1936), in his paper referring to the distribution of the genus Euphausia, places E. superba in his Southern Group, which is circumpolar in range. This species, which can occur in enormous numbers, is found in the colder water of the Antarctic Zone, and along the edge of and under the pack-ice. No detailed list of stations from which material was obtained is appended here, since it belongs more appropriately to the main paper, where it will be included. The post-larval and adult specimens in the Discovery Collections were preserved in alcohol, and in consequence the varying degrees of shrinkage due to the use of different preservatives, which Eraser (1936) encountered in his work on the larval forms, did not exist. Some specimens were, of course, more flexed than others, but each was straightened out as completely as possible on an opal glass scale, and its length from the anterior margin of the eyes to the end of the telson determined to the nearest millimetre. The anterior edge of the eyes, rather than the tip of the rostrum, was selected as one Hmit of measurement, being readily distinguished and less liable to damage. The carapace and body muscles were then dissected from the left side under a binocular microscope, and the reproductive system was examined. The sex and the approximate age of the specimen having been thus determined, external sexual cha- 328 DISCOVERY REPORTS racters, if developed, were also noted. In all, 5564 specimens ranging from 16 to 64 mm. were dissected, of which 2867 were males and 2697 females. Two series of sections, one longitudinal and the other transverse, were made of the reproductive region of two adult females, each measuring 50 mm. and having spermato- phores attached to the thelycum. The course of the oviducts was traced in these sections, the more important of which were photographed microscopically (cf. Plates II-V). Adult females were also used for a series of thick transverse hand-sections (about I cm. thick) through the fifth, sixth and seventh thoracic segments, from which, after further dissection, Fig. 23 was drawn. Finally, in order to determine the approximate number of eggs laid by one female, the ovaries of two gravid females were removed from the body, and the eggs were dissected out on to a large squared slide, separated from one another and counted. ACKNOWLEDGEMENTS I wish to acknowledge helpful criticism and advice from Dr Stanley Kemp, F.R.S. and Dr W. T. Caiman, F.R.S., and to thank my former colleagues, F. C. Fraser and D. D. John, for their invaluable assistance and for permission to quote from their papers and to reproduce certain diagrams. Dr N. A. Mackintosh has kindly allowed me to make use of the colour notes in an original sketch by him of a living specimen of E. superba. I was thus able to make the coloured plate of a male and female specimen. My colleague. Miss D. M. E.Wilson, has also helped me in very many ways, and I am very greatly indebted to her practical interest throughout the work. EXTERNAL CHARACTERS Euphausia superba is shrimp-like in appearance, attaining when fully grown a length of between 50 and 65 mm. The living specimens (PI. I) are remarkable for their transparency and their bright coral-red pigmentation, which often makes the occurrence of the large surface shoals of this animal very conspicuous. The body is covered with a thin, chitinous exoskeleton, which, although much thickened after the final moult before maturity, is always delicate enough to allow the internal organs and muscles to be seen. The adult is thus not very heavily armoured, spines being present only on the carapace and not on the abdomen, except for a well-marked pre-anal spine in both males and females. The carapace is not very extensive in either sex, although the whole body of the adult female is thicker, heavier and longer than that of the male. In a mature female 58 mm. in length, the depth of the carapace measured 13-5 mm. and the width 6-5 mm., whereas in a male 56 mm. long, the depth of the carapace was only 9 mm. and the width 4-5 mm. The ventral margin of the carapace does not overhang the bases of the legs, and the gills which project freely into the water are more exposed in the male than in the female. In the adults of both sexes there is a spine on each anterior, lateral corner of the carapace, this spine being sharp and long in the female, and blunt and short in the THE REPRODUCTIVE SYSTEM OF E. SUPERBA 329 male. The rostrum is bluntly pointed in the female and projects forwards to the anterior margin of the eye, whereas in the male it is rounded and extends only over two-thirds of the eye. The adult female has a pair of well-marked forwardly projecting lateral denticles on the ventral margin of the carapace at the level of the second pair of thoracic legs. There is only the slightest indication of such structures in the adult male, the denticles, which are present in young specimens, diminishing with each successive moult. The epimera of the abdominal segments are large, flat and blade-like. In the female they are produced posteriorly, in the male both posteriorly and anteriorly, except the first pair which alone project posteriorly. The last pair are triangular in both sexes, and are very large in the male. The eye-stalks are short and carry large well-developed eyes, which are slightly larger in the male than in the female. The antennular peduncles are three-jointed and exhibit small sexual differences. These have been described in detail by John (1936) in his paper on the genus Euphausia, so that nothing further need be said here. The basal part of the antenna is divided rather indistinctly into two parts, the first carrying the opening of the green gland and the second the antennal scale and flagellum. In the female, the antennal scale is truncate, extending about one-third along the third antennular joint. It carries a small sturdy denticle on its outer apex. In the male, the scale is more rounded, extending to the end of the second antennular joint, with a thinner, smaller denticle. The mouth-parts and thoracic legs are similar in both sexes, except the sixth pair of thoracic legs in the female, the coxopodites of which are modified to form the thelycum. This structure will be described in more detail later in the paper. The oviducts open between the thelycum and the ventral surface of the sixth thoracic segment. In the male the external genital pores are on the sternum of the eighth thoracic segment. The pleopods are all similar in the female, but in the male the first two pairs are much modified to form copulatory organs, an account of the development and structure of which will be given later. I do not propose to describe the external characters of E. siiperba in any more detail ; they have been fully dealt with by Sars (1885) and subsequent workers. My object has been to draw attention to the chief external differences between the adult males and females. REPRODUCTIVE SYSTEM PREVIOUS WORK Until the publication of two important papers, one by Zimmer in 191 3 and the other by Raab in 191 5, work on the reproductive system of the Euphausiacea was fragmentary. The first description of the system, though a short one, was given by Claus in 1868 of the genus Euphausia, together with a figure showing a spermatophore attached to the ventral surface of the female "unter zwei vorstehenden Flatten". The work of Boas 330 DISCOVERY REPORTS (1883) followed on a genus called by him Thysanopiis, but which according to the description of the thoracic appendages was evidently the genus Thysanopoda. A general account of the reproductive system of the genus Euphaiisia was given by G. O. Sars (1885) in his report of the Schizopoda of the Challenger Expedition, but no adult male or female specimens were obtained and the account is therefore incomplete. Chun (1896) reviewed the previous work on the group and described the system in Stylocheiron, giving an account of the histology of the ovary and testis. Zimmer (1913) worked on a considerable quantity of Eiiphausia superba Dana, and although the material was not sufficiently well preserved for histological examination, he was able to describe the general anatomy and to discuss the function of the various structures. His account was supplemented by Raab's work, which appeared in 1915, on Meganyctiphanes norvegica M. Sars and Euphaiisia krohnii Brandt. The material available was specially preserved and Raab was able to employ histological methods; consequently his descriptions are the most detailed of any published hitherto. None of the above workers, however, had access to such a large collection as that made during the course of the Discovery Investigations, and they were therefore unable to record a complete series of growth stages from the earliest appearance of the repro- ductive system to sexual maturity. MALE REPRODUCTIVE SYSTEM The general plan of the male reproductive system can be seen in post-larval specimens of 18-20 mm. in length, in which, however, there is as a rule no sign of modification in the pleopods. ca^ ala ha t da vd g pn vd *Fig. I. Dissection to show the development of the male reproductive system. :■: 15. ala, anterior lateral artery; ca, cephalic aorta; da, descending aorta; ^, gut; ha, hepatic artery; pa, posterior aorta; /, testis; vd, vas deferens. The testis / (Fig. i) lies partly below the heart and partly on the dorsal surface of the liver to which it is attached by a fold of mesentery. It consists of fifteen finger-like * Figs. 1-7 show progressive stages in the development of the male reproductive system. THE REPRODUCTIVE SYSTEM OF E. SUPERBA 331 follicles arranged round the outer side of a horseshoe-shaped band. The anterior follicles are short and broad, but the median and posterior ones become gradually more elongated. The number of follicles appears to be fairly constant, there being generally seven on each side and one occupying a median, anterior position. In the young specimen, each follicle is distinctly separate from its neighbours, but in the adult all the follicles become closely crowded together, although never actually fused. The vasa deferentia vd (Fig. i) run backwards from each end of the horseshoe-shaped band. They are paired, slender ducts lying below the heart and extending back as far t vd Fig. 2. Dissection to show the development of the male reproductive system. ■' 15. ega, external genital aperture; g, gut; pfi, posterior flexure; t, testis; vd, vas deferens. as its posterior margin. At this point, each duct bends outwards at right angles, runs between the musculature of the body wall and the carapace, and passes laterally and ventrally behind the luminous organ, to open on the sternum of the eighth thoracic segment near the middle line. The vasa deferentia are slightly dilated immediately behind the external genital apertures ega (Fig. 2). As can be seen from the diagrams, the heart and some of the main blood vessels^ are closely associated with the reproductive system. The hepatic artery ha runs into the lobes of the liver immediately behind the median anterior testicular follicle (Fig. i), while the cephalic aorta ca and the anterior lateral arteries ala branch off immediately in front. The descending aorta da passes between the vasa deferentia to the ventral body wall, and the posterior aorta pa leaves the heart just above the right angle bend. The intestine is ventral to the reproductive system. As growth proceeds, the vas deferens becomes slightly swollen in the region of the right angle bend, and a flexure pjl appears, so that there is a short anteriorly directed part from which the duct runs outwards and backwards to the external opening (Fig. 2). 1 The names of the blood vessels are taken from Zimmer's account of the Euphausiacea in Kiikenthal and Krumbach's Handbuch der Zoologie, vol. iii, pt. i, 1926-7. Fig. 3. Dissection to show the development of the male reproductive system. g, gut; pfl, posterior flexure; /, testis; vd, vas deferens. X 15. Fig. 4. Dissection to show the development of the male reproductive system, x 15. afl, anterior flexure; g, gut; pfl, posterior flexure; t, testis; vd, vas deferens. THE REPRODUCTIVE SYSTEM OF E. SUPERBA 333 The flexure increases in length and thickness, and in addition a swelling appears on It forming a lateral pocket, anterior to the point at which the duct bends outwards over the ventral musculature (Fig. 3). At the same time, the growth of the liver in a posterior direction alters the position of the testis, so that the latter comes to lie more closely and completely on the lobes of the liver (Figs. 1-7). A second flexure afl (Fig. 4) appears at a short distance behind the junction of the vasa deferentia with the testis, at the point where the vasa deferentia are most widely separated from one another, before they come to lie close together in a median position. Fig. 5. Dissection to show the development of the male reproductive system, x 15. afl, anterior flexure; ej, ejaculatory duct; g, gut; Ip, lateral pocket; pfl, posterior flexure; /, testis; sps, spermatophore sac; vd, vas deferens. This anterior flexure is bent first outwards and then inwards again. The vasa deferentia are also more swollen behind the external genital apertures. The follicles of the testis increase in size until they are closely crowded together, hiding the horseshoe-shaped band. Both the anterior and posterior flexures grow larger, the vasa deferentia begin to show coiling, while the lateral pocket Ip is more marked (Fig. 5). Coiling extends gradually over the whole length of the vasa deferentia between the anterior and posterior flexures. At the same time, the walls of the ducts become thickened, especially in the region of the lateral pockets Ip. At this stage there is often 334 DISCOVERY REPORTS a small mass of chitin in the enlarged part of the ducts ej behind the external opening. It is of the same shape as the full-sized spermatophore, but whether it contains active spermatozoa seems doubtful (Fig. 6). After some further growth, the mature condition is reached (Figs. 7 and 8). The follicles of the testis are often difficult to distinguish separately, because they are very tightly packed together. The horseshoe-shaped band, formerly regarded as part of the vasa deferentia, was shown by Raab (191 5) to be filled with germinal tissue even in quite young specimens, and is therefore more correctly considered to be part of the testis. t qf\ g vd [p pf\ sps Fig. 6. Dissection to show the development of the male reproductive system, y 15. afl, anterior flexure; ej, ejaculatory duct; ", gut; Ip, lateral pocket; pfl, posterior flexure; sps, spermatophore sac; t, testis; vd, vas deferens. Raab (191 5) determined the histology and function of the various parts of the repro- ductive system. He found that in the region of the anterior flexure afl, the lumen of the vasa deferentia is fairly wide and is lined with flattened cells of pavement epithelium. In mature males it is full of spermatozoa. The lumen is not so wide in the region of the posterior flexure pfl, being lined with columnar epithelium and containing, in addition to spermatozoa, a secretion which is the product of the epithelial cells. Behind the posterior flexure, where the lateral pocket Ip opens into the duct, the lumen widens again and the epithelium flattens. Chun (1896) called this part of the vas deferens the spermatophore sac sps (Figs. 7 and 8), and in mature specimens a spermatophore in process of formation is to be found in each sac. The spermatozoa, Fig. 7. Dissection to show the mature male reproductive system. Lateral view, x 15. afl, anterior flexure; ega, e.vternal genital aperture; ej, ejaculatory duct; c-, gut; Ip, lateral pocket; pfl, posterior flexure; sps, spermato- phore sac; t, testis; vd, vas deferens. Fig. 8. Dissection to show the mature male reproductive system. Dorsal view. X 15. afl, anterior flexure; ej, ejaculatory duct; g, gut; Ip, lateral pocket; pfl, posterior flexure; sps, spermatophore sac; t, testis; vd, vas deferens. 336 DISCOVERY REPORTS oval cells with round or oval nuclei, collect in the spermatophore sac together with a large part of the secretion from the preceding part of the duct. The epithelium of the sac secretes a chitinous cuticle, which encloses this mass in the spermatophore sheath. The spermatophore is flask-shaped and lies in the sac with its head towards the external opening. When fully formed it passes from the sac into the ejaculatory duct ej, the muscular part of the vas deferens immediately adjacent to the external genital aperture ega (Fig. 7), where it remains until transferred to the female. The lateral pocket Ip functions as a cement gland, and its secretion adheres to the neck of the spermatophore and serves to fix it on to the ventral surface of the female. In mature males, one spermatophore is to be found in each ejaculatory duct, and another one in each sperma- tophore sac, ready to take the place of the first after transference has occurred. The first two pleopods in the male are especially modified to effect transference. DEVELOPMENT OF THE COPULATORY ORGANS The endopods of the first pair of pleopods show secondary sexual modifications before any differentiation of the second pair has begun. The endopod is at first quite typical, consisting of a large blade-like setigerous lobe si and a smaller rod-like auxiliary lobe al (or appendix interna), which carries a number of small hooks or cin- cinnuli (Fig. 9). After a time, a blunt oval lobe p grows out below the auxiliary lobe: this is the first appearance of the chief copulatory organ, the petasma (Fig. 10). The petasma /> soon becomes cleft into two parts, the inner lobe il carrying two spines, and the median lobe ml carrying one spine (Fig. 11). Growth of these two lobes continues, the two spines on the inner lobe occupying terminal and proximal positions respectively, while the spine on the median lobe becomes lateral by reason of the extension of the tip of this lobe. A fold, the wing W, running parallel pig. 9. First pleopod, undifferentiated. to the margin of the setigerous lobe also auxiliary lobe; si, setigerous lobe, develops (Fig. 12) and which, as growth F'g- ^°- First pleopod. Development of the petasma. 1 ^1 ^ .A X -50. a/, auxiliary lobe; /), petasma; s/, setigerous lobe. contmues and the petasma extends ^ •' ^ *^ > > 5 further over the surface of the setigerous lobe, curves over the two processes of the 30- al. THE REPRODUCTIVE SYSTEM OF E. SUPERBA 337 inner lobe (Fig. 13). Growth of the median lobe proceeds more rapidly than that of the inner lobe, but the terminal process pt of the latter is now so long that it reaches almost to the top of the median lobe (Fig. 14). The proximal process pp now begins to broaden at its tip (Fig. 14) and finally shows a leaf-like expansion, while one or two small hooks /)a develop on the end of the median lobe (Fig. 15). Fig. II. Fig. 12. Fig. 13. Fig. 11-13. First pleopod. Development of the petasma. x 30. /'/, inner lobe; ml, middle lobe; pi, lateral process; pp, proximal process; pt, terminal process; w, wing. The adult condition has now been reached. This has been fully described by D. D. John (1936) in his paper on " The southern species of the genus Euphatisia ", and he has kindly allowed me to quote his description and to reproduce one of his figures. When the mature petasma is unrolled (Figs. 16 a and b) it can be seen that the "proximal part of the terminal process pt is bent nearly at a right angle to the distal part, which is about three times as long.. . .The end of the process tapers and curls forwards; it is not bifid. The distal two-fifths of the proximal process/"/) is bent inwards 338 DISCOVERY REPORTS and carries two membranous expansions, one distally a, the greater part of it lying on the hinder end of the axis of the process, one b on the foremost side immediately near the bend. The distal end of the former touches the latter. The foremost expansion b is striated, that which is distal and for the most part hindermost a is not [A secondary Fig. 14. Fig- 15- Fig. 14. First pleopod. Development of the petasma. ■: 30. //, inner lobe; ml, middle lobe; pi, lateral lobe; pp, proximal process; pt, terminal process; tv, wing. Fig. 15. First pleopod. Mature form. Left side, x 30. //, inner lobe; ml, middle lobe; pa, additional process; pi, lateral lobe; pp, proximal process; pt, terminal process; si, setigerous lobe; w, wing. process] is present in the majority of males, though not in all, as a spine c bent forwards over the foremost expansion b. It may be as long as that shown in Fig. 30, ^i, dz [my Fig. 16] or very much smaller, or it may be entirely wanting: of thirty heavily chitinised males examined, it was present in the left petasmas of twenty, absent from those of ten. The lateral process pi is curved at the end ; it carries no tooth or secondary process. On THE REPRODUCTIVE SYSTEM OF E. SUPERBA 339 the inner or front margin of the median lobe ml, beyond the end of the lateral process pi, there is always a small additional process pa; in one specimen there were two additional processes." A very small spine-shaped process d on the inner lobe il has been seen in one or two specimens of E. superba. There is nothing that I can add to this account of the adult petasma, except to say that I have not found the secondary process c on the membranous expansion b of the proximal process as frequently as John did. In fact, in most of the adult males examined by me it was completely absent. It may be developed only when certain conditions occur, or may even be a regional variation. Fig. i6«. Mature petasma, inner and middle lobes from behind (reproduced from John's paper). X 30. a, distal expansion; b, foremost expansion; w/, middle lobe; pa, additional process; pi, lateral lobe; pp, proximal process; pt, terminal process. Fig. idb. Terminal and proximal processes from in front, x 30. c, spine; d, spine-shaped process. In my observations, two additional processes />« occur more frequently on the median lobe than one, and I have also seen from time to time the small spine d on the inner lobe. On p. 207 of his report, John draws attention to the fact that processes of a future moult can be displaced, perhaps in fixation, and by penetrating the chitinous coat give a false impression of the presence of an additional process. I have frequently noticed this condition, while examining the developmental stages of the petasma. The modification of the second pair of pleopods (Fig. 17) proceeds rather differently, and does not begin until the petasma is well established in the first pair. The tip of the auxiliary lobe at exhibits a cleft on its outer side (Fig. 18) which grows rapidly and divides u o N 1> be O [in u •a -a a> D O o ^ J2 O ,; "^ S 3 J3 O 8"! ■a m o .. "^ OS 3 n "^ b o 3 M (an rz^ 1-1 Uh - 1 CJ ^ "a ^- to" o o o o o >. X X X r^ (U u U >< XI X3 -O 3 o _o _o rt _^- >, >> >-. <3 u Ut u c^ .2 .3 6 'x 'x 'x r^ 3 3 3 c« M rt X 1) 1) u JS _d ^ 00 " T3 tt-( uh tn U o o O M -S c e 3 ^ S Ji o .g O rt rt n fl) o o o jS « « « '-3 c 3 ^ ■-5 ■3 o O o e E e -a -t3 •o •a o o o o & C^ a. a, o O o o _u V Ji ° >t >Q Fig. 24a. Dorsal view of immature ovary, x 10. «, abdominal region; od, origin of oviduct; t, thoracic region. Fig. 246. Dorsal view of mature (not gravid) ovary. The oviducts can clearly be seen in young specimens, as straight tubes passing ven- trally from the outside of the ovary, below the posterior end of the heart into the coxopodites of the sixth pair of legs (Fig. 23), to open between the coxopodites and the sternite of the sixth thoracic segment by two large pores. The ovary is covered with thin connective tissue, by means of which it is suspended in the thorax. As the eggs develop, the connective covering grows too, and finally in mature females the eggs and their covering extend from the heart to the ventral body wall, forward over the liver, and backwards into the abdomen, causing considerable 3-2 344 DISCOVERY REPORTS expansion of the cephalothorax, the primitive saddle-shape of the ovary being lost (Fig. 246). The growth of the ovary is a continuous one, the eggs increasing in size and number until they reach an average diameter of o-66 mm. and a total number of between 11,000 and II 500. They become arranged in diagonal rows, extending from the dorsal to the ventral surface. Just before laying, the eggs are so tightly packed that they take on Fig 25 Dissection to show the female reproductive system in a mature (not gravid) specimen (anterior view), x 15. cm, coxal muscle; egp, external genital aperture; Jam, flexor abdommalis muscle; nc, nerve cord; od, oviduct; ov, ovary; sgl, shell gland; sp, spermatophore ; spm, sperm mass; th, thelycum. Deeper dissection shown on right side of diagram. a pentagonal or hexagonal shape towards the external surface and are roughly conical on the inner side. Raab (191 5) found that the eggs when first detached from the germinal layer had fine granular cytoplasm and a relatively large nucleus. As the eggs mature, the cytoplasm becomes coarsely granular, as more yolk material accumulates. The full-grown eggs are full of yolk. At the same time there develops round the oviducts a series of diffuse glands, con- sisting of actively secretory cells, which according to Raab produce the egg capsules. THE REPRODUCTIVE SYSTEM OF E. SUPERBA 345 These glands sgl (Plates II-V and Figs. 23 and 25) surround the oviducts od (Fig. 23) in their course through the thoracic cavity and extend into the coxopodites of the last four pairs of thoracic legs, including the last two vestigial pairs seven and eight (Fig. 25). They lie between the gills corresponding to these appendages, below the body wall and the large flexor abdominalis muscles /am, and extend under the ventral nerve-cord, in the region of the genital pores and posterior to them (Fig. 25). The oviducts can be traced through the gland masses both by dissection and by the examination of serial sections. Figs. 23 and 25 represent dissections to show their relation to the other parts of the reproductive system, and Plates II-V are photo- micrographs of transverse and longitudinal serial sections through the thorax of an adult female. At the point at which the oviducts leave the ovaries (Plate II, figs. 3 and 4), they are fairly thick-walled with a narrow lumen, but, on reaching the coxopodites of the legs, they widen to form a thin-walled atrium into which the secretion of the glands appears to be poured, to surround the eggs before they are laid (Plate III, figs. 2, 3 and 4). Fig. 26. Three views of the thelycuin. a, ventral; b, anterior; c, dorsal, a, anterior opening; c, coxopodite of sixth leg; egp, external genital aperture; p, posterior opening. It seems probable that the great flexor abdominalis muscles /aw, which run ventrally in the thorax and abdomen, and which increase markedly in size in the region of the last thoracic leg, assist in the expulsion of the eggs (Fig. 25). The coxal muscles cm may also operate in the same way (Fig. 25 and Plates II-V). The external genital pores egp (Fig. 25) are transverse slits surrounded by thick lips, and they open into the narrow space between the coxopodites and the ventral body wall (Plate IV, fig. i and Plate V, figs, i, 3 and 4). Sars (1885) and Chun (1896) described the genital pores in Euphmisia and Stylocheiron as being unpaired apertures on the ventral surface of the sixth thoracic segment. Zimmer (1913) found them on the basal j oint of the sixth pair of thoracic legs . Raab ( 1 9 1 5 ) , working on Meganyctiphanes and Etiphamia krohnii, states that they do not lie on the basal joints, but next to two plates (thelycum) connected to the legs close to the middle line. It will be seen from Fig. 25 and Plates II-V that in E. superba they lie on the coxopodites. Immediately between the genital pores egp is the posterior opening/) of the thelycum th (Fig. 26c), a pouch formed partly by two wing-like expansions of the coxopodites and partly by an outgrowth from the sternal wall (Plate V, fig. 3). These three plates are fused posteriorly and ventrally, but are wide open anteriorly forming the mouth of the 346 DISCOVERY REPORTS pouch (Fig. 26 a-c). Anteriorly, directly below the ventral body wall of the animal, the two coxal plates remain unfused in the middle line, forming an open groove, which runs from the wide mouth of the thelycum a to its posterior opening p between the genital pores, and which in Fig. 23 is closed by the sperm mass spm. Zimmer (191 3) described the thelycum as being derived entirely from outgrowths from the sternum, but the view of Raab (191 5), that both the sternum and the coxo- podites contribute to its formation, is confirmed in the present iresults, by examination of transverse and longitudinal serial sections. FERTILIZATION Fertilization must be external. The spermatozoa, which are very characteristic oval cells with large nuclei, have never been seen within the body of the female, although Zimmer (1913) described them as penetrating into the thorax. It is clear from his account, as Raab (191 5) pointed out, that he mistook the system of diffuse shell glands sgl for sperm masses. In all the fertilized females examined during the course of the present investigation, the sperm mass was always confined to the thelycum. The specimens selected for sectioning also carried spermatophores, the contents of which could be clearly seen in the same position (cf. Plate V, figs, i and 3) and were never found to have penetrated within the body. When transference of spermatophores is about to take place, the first two pairs of pleopods of the male, which are specially modified for the purpose, are extended for- wards under the thorax. The spermatophores are expelled from the ejaculatory duct and are caught by the pleopods, which transfer them to the thelycum. This operation must be a rapid one, for although male specimens have been obtained with spermatophores extruding from the genital pores, they have never been seen holding them in the pleopods. Zimmer gives a detailed account of this transference and Raab agrees with it. If the first pair of pleopods is bent forward under the thorax, the modified endopodites do not lie under the genital pores, and could not therefore take up the spermatophores when they are extruded. Further, all the special hooks etc. are on the under side of the endopodite, away from the ventral surface of the animal. The second pair of pleopods, however, if extended forward, coincides with the genital openings, the furrows on the endopodites being turned towards the ventral surface of the body. They are thus in a position to receive the spermatophores. The first pair of pleopods would be in the way, and therefore they are probably bent out laterally during the process. According to Zimmer, fertilization takes place with the animals lying abdomen to abdomen. In this position, however, the furrows on the second pair of pleopods are opposed to the ventral surface of the female, whereas the organ on the first pair is turned towards her. The spermatophores are evidently passed from the second to the first pair of pleopods, which then fix them into the thelycum of the female, who must be firmly held during this complicated proceeding. Zimmer states that the elasticity THE REPRODUCTIVE SYSTEM OF E. SUPERBA 347 of the parts may be sufficient to bring about transfer, but that the musculature alone cannot be effective. It is difficult to understand why he made this statement. The muscular system in E. superba is well developed, both in the body and in the appendages, and there is no anatomical evidence to show that the animal is not capable of the sustained muscular effort necessary to effect fertilization. The whole problem is one which can only be solved by observation of living material, but examination of well- preserved specimens confirms Zimmer's suggestions, in the main, as to the method of the transference of the spermatophores. The spermatophores are held in the thelycum by a chitinous cement substance adhering to their necks, and secreted by the lateral pockets of the spermatophore sacs in sufficient quantity to plug the mouth of the thelycum. The spermatozoa spm (Fig. 25) form a white mass in the wider end of the spermatophores when transference has just taken place, but very soon they penetrate up the narrow necks into the cavity of the thelycum itself, filling it completely and extending right up to the posterior opening, next to the female genital pores egp (Fig. 25 and Plate V, fig. 3). This penetration is effected by the action of a fluid contained in the spermatophores together with the spermatozoa, and which is secreted by the cell lining of the vasa deferentia in the region of the posterior flexure (see p. 334). This fluid expands, according to Raab, and pushes the sperm mass spm before it out of the spermatophore. Koltzoff (1906) described a similar process in some Decapoda, the spermatophores of which contain a fluid, which swells so prodigiously that it ruptures the chitinous sheath, and shoots the sperm through the egg membrane, thus bringing about fertiliza- tion. He called this substance "Explosionsstoff ". The passage of the spermatozoa into the cavity of the thelycum must be rapid, for it is much more usual to find the spermatophores empty and the thelycum full than vice versa, although specimens showing the latter condition occur sometimes in a big catch. Copulation generally occurs when the eggs have only reached a quarter or a third of their diameter when ripe. The spermatozoa must therefore retain their potency for some time. Normally, two spermatophores are inserted into the thelycum, but any number up to seven has been seen. In those specimens where copulation has occurred more than once, the spermatophores often show all stages between empty and full, indicating that they were affixed at successive intervals of time, and that possibly all of the spermatozoa are unable to penetrate at once into the thelycum, as its capacity is limited. As the eggs are laid, they come into contact with the sperm mass (Fig. 25 and Plate V, fig. 3), and fertilization is effected. Andrews (1904), in a detailed account of the breeding habits of the American crayfish, writes as follows: "As the eggs are laid they probably pass over the annulus [cf. thelycum in Euphausid] and a relatively small amount of sperm might fertilize all of them, if it came out of the sperm plug in time. Before laying the annulus is covered with glair and possibly this may act to bring the sperms out, as well as to protect them from the water. . . . Some osmotic factor may be here concerned 348 DISCOVERY REPORTS in bringing out the sperm. Other means of getting the sperm out of the waxy tube, that we have called the sperm plug, and from the interior of the annulus might be the pressure that the sternal plate between the fifth legs may exert on the annulus when the legs are forced forwards, as seemed to be the case about the time of laying." In E. superba, it seems likely that owing to the close proximity of the genital pores to the free end of the sperm mass (Fig. 25), the pressure due to the extrusion of the eggs will be sufficient to liberate the sperms at the same time from the posterior opening of the thelycum, and thus bring about fertilization. The shell membrane is probably still soft and glutinous, and will therefore allow the passage of the sperms through it. Meyer (1934) found that in Crangon vulgaris the shell membrane remained soft for some time after laying, the hardening process being a gradual one. In spite of the large number of eggs ( 1 1 ,230 and 1 1 ,407 were counted from the ovaries of two gravid females respectively), the process of laying is evidently a rapid one. On two occasions, females with greatly distended carapaces and carrying spermatophores were placed for observation in the aquarium on board the R.R.S. 'Discovery II'; these spawned overnight, and the following morning eggs were found in the water. One of the females was dissected and "many eggs were found in the ovary, floating in a milky fluid, which appeared to consist of an emulsion of tiny oil globules. The eggs were the same size as those found in the aquarium", that is o-6 mm. in diameter. Spawning in this instance was clearly incomplete, and this was probably due to the fact that, normally, gravid females occur at depths of 250 m. and more, where perhaps the increased pressure of the water assists in the liberation of the eggs. The eggs failed to develop in the aquarium, because the external conditions were not suitable. After the eggs have been laid, the ovary resumes its primitive saddle shape, and the shell glands are still much in evidence, extending up the oviduct for some distance, as well as remaining in the coxopodites of the legs. SUMMARY 1. Dissection shows that both male and female reproductive systems are developed before any external sexual characters appear. The sex of each specimen can be deter- mined in this way. 2. A series of stages in the development of the male and female systems was obtained. 3. Modification of the first pair of pleopods begins before the second pair shows any sign of diff^erentiation. 4. The course of the oviducts through the shell glands was traced by means of serial sections as well as by dissection. 5. Transference of spermatophores occurs in the first place before the female is gravid. 6. The spermatozoa retain their potency for some time. 7. Fertilization is external, occurring when the eggs are laid. 8. Spawning takes place rapidly, probably at depths of 200 m. and below. THE REPRODUCTIVE SYSTEM OF E. SUPERBA 349 After I had begun this investigation, a paper by Ruud (1932) appeared on the " Biology of Southern Euphausiidae ", in which Ruud drew attention to the fact that the "ex- ternal sexual characters do not tell us anything about the degree of maturity at which the animal has arrived. This must be ascertained by examining the internal organs, the ovary and the testis." A discussion of his results belongs more appropriately to a second paper now in preparation to which this is the introduction, but I have frequently referred to his work and wish to make full acknowledgement here. I was more fortunate than he in having access to larger collections of material, and consequently could carry my anatomical investigations farther. BIBLIOGRAPHY Andrews, E. A., 1904. Breeding habits of the Crayfish. Amer. Nat. xxxviii, pp. 165-206, 10 figs. Boas, J. E. V., 1883. Studien iiber die Verwandtschaftsbeziehungen der Malakostraken. Gegenbaurs morph. Jahrb. Leipzig, viii, pp. 485-579, pis. 21-24. Calman, W. T., 1909. A treatise on Zoology . Edited by E. Ray Lankester. Pt. vii. Appendiculata. 3rd fasc. Crustacea., pp. 244-52, 289-90. Chun, C, 1896. Atlantis. Biologische Studien iiber pelagische Organismen. Pt.v. Ueber pelagische Tiefsee- Schizopoden. Bibliotheca Zoologica, Stuttgart, vii. Heft 19, pp. 137-90, pis. 8-15. Claus, C, 1868. Ueber die Gattung Cynthia als Geschlechtsform der Mysideengattung Siriella. Zeit. wiss. Zool. XVIII, pp. 271-9, pi. I. Eraser, F. C, 1936. On the development and distribution of the young stages of krill (Euphausia superba). Discovery Reports, xiv, pp. 1-192. John, D. D., 1936. The southern species of the genus Euphausia. Discovery Reports, xiv, pp. 193-324. KOLTZOFF, N. K., 1906. Die Spermien der Decapoden. Arch. mikr. Anat. Lxxvii, pp. 364-571, 5 pis., 37 figs. KiJKENTHAL, W. and Krumbach, I., 1926-7. Handbuch der Zoologie. III. Pt. i, Euphausiacea, by C. Zimmer. Pp. 812-39, figs. 863-902. Meyer, P. F., 1934. Bin Beitrag zur Eiablage der Nordseekrabbe Crangon vulgaris, Fabr. Zool. Anz. cvi, PP- 145-57. 4 figs. Raab, F., 1915. Beitrag zur Anatomie und Histologie der Euphausiiden. Arb. zool. Inst. Univ. Wien, xx, pp. 125-58, pis. 1-2. Ruud, J. T., 1932. On the biology of southern Euphausiidae. Hvalradets Skrifter, Nr. 2. Oslo, pp. 5-105, 37 figs- Sars, G. O., 1885. Report on the Schizopoda collected by H.M.S. ' Challenger' during the years 1873-1876. Report of the scientific results of the voyage of H.M.S. 'Challenger' during the years 1873-1876, XIII, pt. 37, pp. 1-228, pis. 1-38. Zimmer, C, 1913. Untersuchungen iiber den inneren Bau von Euphausia superba Dana. Zoologica, Stuttgart, XXVI, pp. 65-128, pis. 8-14. DISCOVERY REPORTS VOL. Xrv .1 aTAjq lo 2naMiDaq2 tjuqa ho uDTayiz oaauojOD .a^ij MO^q ,Aa^aqu2 AiauAHqua DISCOVERY REPORTS VOL. XIV PLATE ] DISCOM.RY kLPORTS VOL. XIV I'; .AT if n oj i/> Mi: o u to 60 a 3 -o '> O -C bC C '$ o c o •a o g. Cr t3 to t-f u> < • a > g » <; w O C a. CU3 cv e o u > &, i~-( cv u C O > u t« kC to ^—4 0 I 1 r - O" Si, ?^ 3 q up rv^ CD -§ o G (J> 1-1^ O to Q 3 o w o a. DISCOVERY REPORTS VOL. XIV PLATE II !!*;& r ^v-vS #^!^> /■■■ rv^s«^ (M '^f ^k 'if- ^^^^ M(C ^!^ DISCOVERY REPORTS VOL XIV JJ u u u c -a 2 o & d t; o rt 1-1 . ,*- 3 ■« < 8l ad ed u "Sj '.1^! 0- ?pl 3 GO 1-5 « J3 Kfi C 3 p ■31 4> <1 a to >. Ph "5 5 D _a b w 3 j:: ^ J^ ^ H o u Cu d to a, U CQ g 4> Om w a T3 c« J= U3 a 4-1 o w 3 J3 -o *2 t> CA o c u o X \-l " u u •" )i u c g Um « s o 3 t« 0 a "^ o O u X ^—^ x; 15 Vi »-' Q -^ bC a- .E ci, Ui ^ Ji 60 c " h o u •a -5 ^ U U-t M 0 J3 — ci C^' rt X c c ^ " 3 « a -5 t£ '-' C 0 ^.s 4 0 VL, 3 c D- O . u > ri ^M (V3 a U3 — S" G CO > ^^ CV) _ a. G > Q o ^ ^f? vT! <• C» G a. c o " S n 3'^ D o o 3 a^i- 1' in ea h it OJ ts Sj w C Cr &j in. (W c o ..■. V4 y 3- c cr s cj o o 5' . a. g P ^ ,^ 'c 0 o g u . — ^ v— »-f /*• to e •-^ 5^ o O vj ; » 0 o o 5-- u c a. '^. -aa o a 5. u i- u Ol DISCOVERY REPORTS VOL XIV PLATE III V ^:-;^<;)»<:^. ' • V v^ ••if*-'. •■■.'. ■. k • id™ '">'■? •'/•■.•';. •'. 't%->H^ ^iC V. ■^ «.■* -•■ Ar "N ...rW M' ff •fv^ DISCOVERY REPORTS VOL. XIV I : .\ 4,' -s ■S n S=! " < ^ 2 t4 s s UJ s-g ;zi CO 1 ^* < O vS J3 CA) a do D O JJ << s^-s S •«\ o V y •s v; 60 -C 1 o tiO ^ O ui a ■?■ u 0.2 c ■a- ~ o ^B o 1-5 =5 c 0^ 00 n C V o a. H-) o ■* o c > > ^^ > G l-U ^5^ i_i-> ^^ n'S _, ?. ^. — r7 n «■ JD O cv M (J> ►a tr> tr a, e c J3 DISCOVERY REPORTS VOL. XIV PLATE IV -, &,. ^i ^^ , i' ■. ^ . ■•; ^ A ■'•'Ml :./-,/\\ 'Y-' r /?i i > 1 ' , » 4?- < . '. Ah' *r '^'^-^v/^^ I • ^vO DISCOVERY REPORTS VOL.XI\- < W D CO D X cu D <-l_C o c 1> PQ m G 5 D- to o c 0- ^■' <; o d Q CO o C > X o" ►fl 3 c 5 w mJ D O y J3 O ?i tr* b 3 ja »" a u Sj d Cr rt 5L Od C o l-l " a. c S « u a J3 o g El. ??* 06 /*. g /^ X 5 1. aa to b a •-1 o 3 q- ^ o u o u. !^ •JQ b Er S oo n P. o DISCOVERY REPORTS VOL.XIV PLATE V :-M f'^ ^' '.'»Xf\ ^<^' .-V.^./--- /*>'■■ -.-"v*- .^ > 'V ^^\€ M-/ [Discovery Reports. Vol. XIV, pp. 351-404, June 1937.] LARVAE OF DECAPOD CRUSTACEA PART IV. HIPPOLYTIDAE By ROBERT GURNEY, D.Sc. CONTENTS The larval genus Eretmocaris Bate page 353 Larvae from the Red Sea 354 Larvae from the Great Barrier Reef 363 Larvae from the Atlantic 366 Discussion 375 The genus Tozeuma Stimpson 3'7'7 Chorismus antarcticus Pfeffer 384 Discussion ^89 Savon marmoralus (Olivier) and allied forms 300 Latreutes mucronatus (Stimpson)? 2g8 Discussion and summary of generic characters 300 Literature <02 T LARVAE OF DECAPOD CRUSTACEA PART IV. HIPPOLYTIDAE By Robert Gurney, D.Sc. (Text-figures 1-137) THE LARVAL GENUS ERETMOCARIS BATE HE genus was founded by Bate (1888, p. 895) to include four species of larvae taken in the Pacific near Japan and in the Atlantic near the Cape Verde Islands, all characterized by the great length of the eyestalks. Ortmann (1893, p. 78) described a fifth species, but neither was able to figure the legs, which were all lost. Brooks and Herrick had, however, given excellent figures (1891, pi. ix, x) of a similar larva, probably identical with Eretmocaris stylorostris Bate, as the larva of Stenoptis, and Chun (1888) had described another form under the name of Miersia clavigera. Ortmann recognized that both these larvae belonged to Bate's genus. Lo Bianco (1901, p. 439 and 1909, p. 609) reared from M. clavigera a post-larval stage which he identified as Ligiir edwardsi, but Caroli (19 18) was able to prove that Chun's larva is actually a stage in the develop- ment of Lysmata seticaudata. The larva of L. seticaudata is characterized by the great elongation of the eyestalks and the precocious development of leg 5, before the appearance of legs 3 and 4, into an enormously long appendage with a paddle-like enlargement of the propodus. The coxa and basis are correspondingly enlarged, so that even when the rest of the appendage is lost, as it so generally is, it is possible to determine its size in comparison with leg 4. These characters are found in all the four species described by Bate, but Erettnocaris dolichops Ortmann, apparently had leg 5 not larger than leg 4, and no doubt does not belong to the same adult genus as typical Eretmocaris. Whether Bate's four species can be regarded as all belonging to Lystnata is by no means certain, and it is more probable that they represent two or more genera. I have had the opportunity of examining a large number and variety of these larval forms from the following sources : (i) At Ghardaqa on the Red Sea I obtained a series of stages including one specimen which moulted to the post-larval stage and permits the identification of the series to which it belonged with Lysfnata. Although these larvae agree very closely with those of L. seticaudata I consider their description of value as tending to establish the characters of the genus and to furnish evidence that Eretmocaris is a composite genus. (2) From the Great Barrier Reef I have a large material for the use of which I am indebted to Mr F. S. Russell. This material enables me to illustrate certain additional specific types which, so far as can be said at present, seem also to fall within the genus Lysmata. 354 DISCOVERY REPORTS (3) Dr Wheeler has been kind enough to give me specimens from Bermuda of two distinct forms, one of which differs strikingly from all the others I have seen in having the propod of leg 4 expanded as well as that of leg 5. (4) The Discovery material which has been entrusted to me by Dr Kemp contains a few specimens of Eretmocaris larvae which are separable into five types, two of which have already been described by Bate. LARVAE FROM THE RED SEA Larvae of the Eretmocaris type were taken in very small numbers in plankton from the deeper water outside the reef at Ghardaqa, and occasional specimens of stage I appeared at night near the laboratory. Two forms can be distinguished, the most obvious distin- guishing character being the presence or absence of dorsal spines on abdominal somite 5 ; but there can be little doubt that the specimens referred to as "species 2" actually represent two species, since some specimens have supra-orbital spines while others have not. Apart from this difference the larvae, having regard to the scantiness of the material, cannot be separated. Species R.S. I, Hippolysmata? Stage I. Length 2-3 mm. (Figs. 1-7). Rostrum long and slender, reaching end of antennular peduncle. Carapace with anterior and posterior median dorsal tubercles ; anterior ventral margin with four teeth. Abdominal somite 5 with a pair of small dorso-lateral teeth. Telson deeply concave behind. Antennule. Exopod with inner feathered seta and four aesthetes ; of these the inner one is short and stout, but bears at its end a spoon-shaped membrane with thickened midrib and distal margin (Fig. 3). It seems to be a general rule in Caridea that there is a feathered seta and four aesthetes, one of which differs from the others in most cases by tapering to a fine point. In Saron and a species allied to it I have detected a very delicate membrane on either side of it, but narrower and much more difficult to see than in Lysmata. Antennal scale with four distinct segments; two outer setae and eleven inner and terminal ; endopod a slender rod bearing a long feathered seta. Maxillule. Palp unsegmented, with five setae. Maxilla, exopod with five setae; endopod with three small inner lobes bearing setae, unsegmented; four inner laciniae distinct. Maxillipede i. Endopod of four segments; exopod with three terminal setae, and one on outer margin ; coxa large, with long setae. Maxillipedes 2 and 3 with long exopods bearing three terminal and six lateral setae. One very small leg rudiment. Chromatophores of legs, mouth region and telson dark olive brown; dark red in thorax between maxillipedes, and rosy red in antennules. THE GENUS ERETMOCARIS 3SS The long exopods have a characteristic slow waving motion. There is no definite proof that this larva belongs to the following series, but the general appearance in life and mode of swimming left little doubt in my mind of the relationship. The resemblance Eretmocaris. Species R.S. I, Hippolysmata't Fig. I. Stage I, ventral. Fig. 3. Stage I, end of antennule. Fig. 5. Stage I, mandible. Fig. 7. Stage I, maxilla. Fig. 2. Stage I, lateral. Fig. 4. Stage I, antenna. Fig. 6. Stage I, maxillule. to stage I of Lysmata seticaudata is very close. It should be noted that the figure of the ventral view does not give a correct idea of the length of the appendages which were seen foreshortened in this position. Erettnocaris. Species R.S. I, Hippolysmata> Fig. 8. Stage II, dorsal. Fig. 9. Stage II, end of antennule. Fig. 10. Stage II, antenna. Fig. II. Stage III, dorsal. THE GENUS ERETMOCARIS .357 Stage II. Length 3 mm. (Figs. 8-10). Rostrum slender, without dorsal teeth, reaching to base of antenna! scale. Carapace with small supra-orbital spines ; ventral margin with three teeth. Abdominal somite 5 with a pair of dorsal spines. Telson less deeply concave, with 8+8 spines. Eyes on long stalks about half length of eye ; eye with stalk about one-quarter length of body. Peduncle of antennule one-quarter length of body, unsegmented, with base a little enlarged ; exopod apparently lacking the feathered seta, but with five aesthetes of which one, as before, has a membranous border. Antennal scale with ten inner and terminal setae, but only three distal segments. Outer distal seta much longer than proximal one ; endopod shorter and more slender, with seta as before. Exopods of maxillipedes 1-3 with four terminal setae; maxillipede 2 with six and maxillipede 3 with eight lateral setae. Leg i a large biramous rudiment. Stage III. Length 3-5 mm. (Fig. 11). Rostrum short, not reaching base of antennal scale. Telson separated from somite 6, slightly broader than long, with 8+8 spines, of which one is lateral. Eyestalk longer and more slender than in stage II; eye with stalk to body as i : 3-5. Peduncle of antennule as long as eye, of two segments, and with small endopod. Antennal scale unchanged ; flagellum reduced to a very small knob without a seta. Leg I developed, with exopod. Pleopods absent. Uropods developed, the endopod small, without setae. Of this form I have no older stages. Species R.S. II, Lysmata sp. Stage II. Length z-^-z-'] mm. (Fig. 12). Rostrum very small, not extending beyond frontal lobe. Carapace with anterior and posterior papillae, but no median spine; supra-orbital and antennal spines absent; pterygostomial spine present, and two small marginal teeth. Abdominal somite 5 without dorsal spines. Telson not deeply concave, with 8+8 spines (in one specimen 7+7 only). Eye with stalk about one-third length of body. Peduncle of antennule nearly one- third of body length. Antennal scale very long and slender, with eleven inner and terminal setae ; two distal segments distinct ; one outer seta near end ; endopod a very small process one-tenth length of scale. Leg I fully developed, endopod very long and slender. Leg 5 fully developed, the propod very long and slightly expanded; lengths of segments (in mm.): Ischio-merus Carpus Propodus Dactylus i'48 1-14 i'3 "06 Between legs i and 5 is a single large biramous rudiment. Exopods of maxillipedes and leg I with four terminal setae; total setae 5, 10, 12, 10. Pleopods absent. General colour red. Eretmocaris. Species R.S. II, Lysmata sp. Fig. 12. Stage II, dorsal. Fig. 14. Stage IV, part of telson. Fig. 16. Stage IX, carapace, dorsal. Fig. 13. Stage IV, lateral. Fig. 15. Stage IV, part of antennal scale. Fig. 17. Stage IX, part of telson. THE GENUS ERETMOCARIS 359 It will be seen that this stage differs very much from the corresponding stage in species I in having legs i and 5 fully developed. In this respect it agrees with L. seti- cmidata as described by Caroli. It is very doubtful if species I really belongs to this genus. I have seen no specimens of stage III. Stage IV.? Length 4-3 mm. (Figs. 13-15). Rostrum very small. Carapace with minute tooth in front of anterior tubercle; supra-orbital spine absent, but antennal and pterygostomial spines present. Telson widest at end ; greatest width equal to half length ; one pair of lateral spines and two small spines at each angle, 8 + 8 in all. Anal spine absent. Eye with stalk to body as i : 3-5. Antennular peduncle as long as eye. Antennal scale slender, with outer distal spine and no segmentation ; flagellum very small. Legs 1 , 2 and 5 fully developed, but leg 5 lost. Legs 3 and 4 rudimentary. Pleopods absent. Endopod of uropod with setae. Stage V.-" Length 5-45 mm. Rostrum small, without teeth. Carapace with median dorsal tooth. There is a well- marked groove in the gastric region from which lines can be faintly traced forwards and downwards. These lines seem to be characteristic of this type of larva and will be shown in figures of other forms in which they are clearer. Telson nearly three times as long as wide, with 6+6 spines, of which one pair is lateral and one at the angle ; apparently one of the small spines at the angle and the innermost spine are lost at this stage. Peduncle of antennule 1-4 mm., with rudiment of stylocerite; flagella about two-thirds length of peduncle. Antennal scale nine times as long as wide; fiagellum one-sixth of its length. Mouth-parts differing little from stage I. Exopod of maxilla with fifteen setae, broad in front. Maxillipede i with endopod of four segments ; coxa well developed with long setae ; exopod with five setae. All legs fully developed, but leg 5 lost. Pleopods absent. Between stage V and the last stage I have only a single specimen. Judging by the increase in length of the carapace, it seems there must be three, or perhaps four, intermediate stages, making eight or nine in all. Caroli describes eight stages in L. seti- caudata, but admits that there may be a stage intermediate between his sixth and "penultimate", making a total of nine. Stage VIII? Length 7 mm. Rostrum with one large dorsal spine. Carapace with supra-orbital, antennal and pterygostomial spines, and with large median dorsal spine; surface with lines well marked. Telson nearly three times as long as wide, with two pairs of small lateral spines and 5+5 distal "spines. Eye with stalk about one-third length of body. Peduncle of antennule 1-9 mm.; 36o DISCOVERY REPORTS endopod 2-55 mm.; exopod faintly segmented, with three groups of aesthetes; stylo- cerite large. Flagellum of antenna as long as scale. Legs I and 2 not chelate. Pleopods large. This specimen is included in this series, ahhough the presence of a supra-orbital spine shows that it must belong to a distinct species. Stage IX? (Figs. 16-19). One specimen in the last stage was taken in plankton and moulted the same night to the post-larval stage. As the animal was not examined alive I have only the moulted skin for description. Rostrum i mm. long, with four dorsal teeth. Carapace 1-9 mm. long, without supra-orbital spines, but with dorsal, antennal and pterygostomial spines. A series of lines are clearly marked (Fig. 16). These lines, which do not seem to bear any relation to the carapace lines described by Boas and others, are repeated in almost exactly the same arrangement in specimens from the Barrier Reef and Bermuda. In a specimen from Bermuda, which was examined very soon after preservation, they were very clearly picked out in red. In Caridion gordo7ii there are two transverse red lines, the one immediately in front of the anterior tubercle, and the other running across from the gastric groove. They correspond to the two transverse lines in Lysmata, but the posterior longitudinal lines cannot be traced. It is possible that they may none the less exist and be visible in the moulted skin. Abdominal somite 5 without dorsal spines; somite 63-6 times as long as deep. Telson 1-3 x 0-45 mm. with one pair of lateral spines and 5+5 terminal spines, of which the second is very large. Anal spine absent. Peduncle of antennule i -95 mm., with small stylocerite ; outer flagellum 4-9 mm. long, segmented, with aesthetes on segments 2, 3, 4. Antennal scale 1-85 x 0-35 mm., flagellum very long. Mouth-parts of the same form as in earlier stages. Exopod of maxilla very large, somewhat truncated in front. Maxillipede i with exopod expanded at base and bearing here ten setae ; distal part with six setae. Legs i and 2 subchelate. The following table gives measurements (in mm.) of the legs. Leg 4 is lost. Propod of leg 5 expanded, the length about 4^ times the width. Pleopods large, with rudimentary setae. Dactylus Propodus Carpus Ischio-merus Total I 2 3 5 o-iy 0-14 0-25 0-2 0-56 0-25 1-55 3-0 073 1-67 II 1-85 1-3 2-0 372 2-76 4-06 5-65 8-77 THE GENUS ERETMOCARIS 361 Eretmocaris. Species R.S. II, Lysmata. Fig. 18. Stage IX, mandible. Fig. 20. Post-larval stage I, lateral. Fig. 22. Post-larval stage I, maxillule. Fig. 24. Post-larval stage I, maxillipede i. Fig. 26. Post-larval stage I, peduncle of antennule. Fig. 28. Post-larval stage I, dactyl of leg 4. Fig. 19. Stage IX, end of leg 5. Fig. 21. Post-larval stage I, telson and uropods. Fig. 23. Post-larval stage I, maxilla. Fig. 25. Post-larval stage I, maxillipede 2. Fig. 27. Post-larval stage I, part of outer flagellum. 362 DISCOVERY REPORTS Post-larval Stage I (Figs. 20-28). Length 9-8 mm. Moulted from last larval stage. Rostrum as long as eyes, with three dorsal and one ventral tooth. Carapace with median dorsal tooth and small antennal and pterygostomial spines, but no supra- orbital. Pleura of abdominal somites 4 and 5 with small spine; somite 6 twice as long as deep. Telson shorter than uropods, nearly 3^ times as long as wide, with two pairs of dorsal spines ; sides fringed with setae ; posterior margin with small spine at angle and a pair of long spines, between which is a pair of feathered setae. No anal spine. Basal segment of antennule with pointed stylocerite ; outer flagellum with five groups of aesthetes on segments 4-6; sixth segment divided, with rudimentary accessory branch. Mandible without palp or incisor process. Maxillule with proximal lacinia pointed; palp small, with one apical seta. Maxilla with three inner laciniae, the proximal one reduced ; palp small, with one seta. Maxillipede i with large bilobed epipod; exopod expanded at base and with eight terminal setae; endopod unsegmented with two small apical setae. Maxillipede 2, epipod with small lobe representing rudiment of podobranch; exopod with twelve setae; endopod bent downwards at carpus, dactylus fused with propod. Maxillipede 3 with rudimentary epipod ; exopod reduced and without setae ; dactyl and propod fused, this segment more than twice length of carpus, and a little shorter than ischio-merus. Leg I with small chela, propod shorter than carpus. Leg 2 carpus very long, faintly divided into twenty-eight segments. Legs 3-5 about equally long, dactylus of 3 and 4 with four inner spines. Legs 1-4 with large vestigial exopods. Epipods absent. Pleopod I endopod less than half as long as exopod, without appendix interna. Pleopod 2 endopod shorter than exopod, without appendix masculina. Exopod of uropods with outer spine near end, and partial joint at this point. There can be no doubt that this specimen belongs to the genus Lysmata, though the mouth-parts are far from having reached the form characteristic of the adult, and the epipods have not yet appeared. At the same time the general form of these appendages fully agree with those of Lysmata, the outer flagellum of the antennule shows a distinct rudiment of the accessory flagellum, and the carpus of leg 2 is divided into many seg- ments. In addition the absence of a supra-orbital spine is regarded as a feature of the genus. The identification of the species at this early stage is out of the question, though the only species recorded from this region is L. trisetacea Heller. It is of interest to note the reduction of the exopod of maxillipede 3 . Reduction at this transitional stage seems to be the rule in Caridea, and its significance is unknown. THE GENUS ERETMOCARIS 363 LARVAE FROM THE GREAT BARRIER REEF The Barrier Reef material examined includes 157 specimens of Eretmocaris, which appears to be very common in these waters. No stages earlier than stages 4 or 5 have been seen, although I have searched a number of the plankton samples for younger specimens. The material is, on the whole, very well preserved, but leg 5 is generally lost, only eighteen specimens having retained one or both of these legs. Three distinct species can be separated by easily observed characters, and the following numbers show the relative frequency : Species i Species 2 Species 3 132 15 10 It may well be that two species are included in "species i ", separable by small details in rostrum, length of eyes and antennule ; but I have not attempted a revision of the material to determine this point. As none of the species can be named it seems sufficient to establish the more marked types which can be easily recognized. All seem to belong to one genus, and that probably is Lysmata. Two species, Eretmocaris remipes Bate and E. longicaulis Bate have been described from the neighbourhood of Japan, but neither of these can be identified with any of the Barrier Reef species. I give below short descriptions of the oldest specimens of each form. Species B.R. I (Figs. 29, 30) Length less than 8 mm. Rostrum very short, with one dorsal spine. Carapace with median dorsal spine, and antennal and pterygostomial spines; supra- orbital spines small or absent. There is a deep transverse groove in the gastric region and lines as described above. Eye about as long as stalk, together less than one-third length of body. Telson about three times as long as wide, with two pairs of lateral spines and 5+5 terminal spines of which the second is the longest. No anal spine. Peduncle of antennule nearly one-third length of body. Leg 5 ischium and merus when distinguishable about equal, together twice length of carpus; propod about i\ times length of carpus and 5! times longer than wide. Species B.R. II (Figs. 31-35) Length up to 7 mm. Rostrum long and slender, with three dorsal spines. Carapace with median dorsal spine large and forming a procurved hook ; antennal and pterygostomial spines present, but no supra-orbital. Carapace not deeply grooved, and lines not traceable. Telson a little more than three times as long as wide, without lateral spines and with 5+5 terminal spines of which 2 and 4 are the longest. 364 DISCOVERY REPORTS Eyestalk much shorter than eye, together less than one-third length of body. Peduncle of antennule one-quarter length of body. Maxilla, exopod broad and truncated in front: endopod with three distinct seta- bearing lobes. Maxillipede i, endopod of four segments: exopod expanded at base and bearing here ten setae : coxa large, with long setae : epipod large, bilobed. Maxillipede 2 with epipod. Leg 5 ischio-merus a little more than twice length of carpus; propod less than ij times length of carpus, its width a little more than one-third of the length. Eretmocaris. Species B.R. I. Fig. 29. Species B.R. I. Fig. 30. Species B.R. I, part of telson. Species B.R. Ill Length up to 4-6 mm. Rostrum very small, without dorsal spine. Carapace with dorsal and pterygostomial spines, but without supra-orbital and antennal spines ; lines not clearly traceable. Telson three times as long as wide, with two pairs of lateral spines and 5+5 terminal, of which the second is very long, and the fourth about half its length. Eyestalk more than twice as long as eye, together more than half length of body (body : eye as i-22-i-5 : i). Antennular peduncle nearly half length of body, with rudiment of stylocerite. THE GENUS ERETMOCARIS 36s Mouth-parts as in species B.R. I. Maxilla with exopod broad and truncate in front. Leg 5 absent in all specimens. In all three species the youngest specimens have leg 5 fully developed while leg 4 or legs 3 and 4 are rudimentary. Fig. 31. Eretmocaris. Species B.R. II. Fig. 33. Species B.R. II, base of antennule. Fig. 35. Species B.R. II, part of telson. Fig. 32. Species B.R. II, mandible. Fig. 34. Species B.R. II, maxillipede i. 366 DISCOVERY REPORTS LARVAE FROM THE ATLANTIC The Discovery material includes a small number of larvae which, by reason of their long-stalked eyes, may be assigned to the genus Eretmocaris. Two of them can be certainly identified with E. corniger Bate and E. dolichops Ortmann, while a third almost certainly belongs to Lysmata seticaiidata} Of the other two, one is characterized by the possession of a spine on the eyestalk, and in this respect resembles a larva described Eretmocaris. Species A. I, A. II. Fig. 36. Species A I. Fig. 38. Species A II. Fig. 37. Species A I, part of telson. Fig. 39. Species All, part of telson. by Kemp (191 6) as probably belonging to Hippolysmata ensirostris. The other two represent types not previously described. In addition to these Atlantic forms from the ' Discover}'' I have a few specimens from Bermuda for which I am indebted to Dr Wheeler. Neither of these seem to be represented in the Discovery collection. There are, therefore, apart from Lysmata seticaudata, at 1 From Station 708, 10° 20' S, 34° 54' W. THE GENUS ERETMOCARIS 367 least seven distinct types of Eretmocaris larvae from the Atlantic. One of the Bermuda forms is perhaps the same as that described by Brooks and Herrick, and may be identical with E. stylorostris Bate. Species A. I (Figs. 36, 37) See Kemp, 1916, Rec. Ind. Mus. xii, p. 403, fig. 5. Hippolysmata ensirostris. St. 701. 14° 39' N, 25° 51' W. Two specimens. St. 709. 14^ 01' S, 36" 30' W. One specimen. Length 6-14 mm. Rostrum reaching end of eyestalk, with three dorsal teeth near end. Carapace with large dorsal tooth and supra-orbital, antennal and pterygostomial spines. Abdominal somites without spines, pleura rounded; somite 6 nearly three times as long as deep. Telson shorter than uropods, without lateral spines and with 5+5 terminal spines, of which spine 2 is the longest. Eyestalk shorter than eye, with large dorsal spine ; eye and stalk to length of body as I : 37- Peduncle of antennule reaching nearly to end of eye ; rami long, but broken. Flagellum of antenna about as long as scale. Legs all lost, the stumps increasing in size backwards and arranged in a semicircle, leg 1 widely separated, leg 5 closely apposed. Leg 4 with exopod. Pleopods present, small. These three larvae represent a younger stage than that figured by Kemp, which is no doubt the last stage and has the rostrum very much longer and with numerous dorsal teeth. Specimens of precisely the same form were taken by the ' Discovery ' at six stations off the east coast of Africa. Species A. II (Figs. 38, 39) Eretmocaris stylorostris Bate, 1888, pi. xcv, fig. 3?. Stenopus hispidus Brooks and Herrick, 1892, pis. ix, x?. St. 709. 14° 01' S, 36° 30' W. One specimen. Length 6-i mm. Rostrum small, with one dorsal tooth. Carapace with small dorsal tooth and antennal and pterygostomial spines; supra-orbital spine absent. Abdominal somites without spines, pleura rounded ; somite 6 not much more than twice as long as deep. Telson three times as long as wide, without lateral spines: terminal spines 5 + 5, of which 2 and 4 are longest. Eyestalk longer than eye, together nearly half length of body. Peduncle of antennule shorter than eye, about one-third length of body. Legs all broken off, but stump of leg 5 larger than that of leg 4, which has an exopod. Pleopods large. Bate's Eretmocaris stylorostris may be an earlier stage of this species. The species 368 DISCOVERY REPORTS figured by Brooks and Herrick differs in having the eyes considerably shorter, but may well be the same species. Fig. 40. Eretmocaris. Species A. III. Fig. 41. Species A. Ill, part of telson. Species A. Ill (Figs. 40, 41) Bermuda. 28. v. 35. Several specimens of same stage. Length 6-o8 mm. Rostrum short and straight, with one dorsal tooth. Carapace with small dorsal tooth and supra-orbital, antennal and pterygostomial spines. Lines on carapace well marked. Abdominal somite 6 three times as long as deep. Telson shorter than uropods, with two pairs of very small lateral spines and 5 + 5 terminal spines, of which the second and fourth are much the longest. THE GENUS ERETMOCARIS 369 Eye and stalk about equal, together about one-third length of body. Peduncle of antennule a little shorter than eye. Legs I and 2 not chelate, but leg 2 with very long carpus and very short propod. Leg 2 longer than leg i and leg 3 than leg 2; leg 4 shorter than leg 3, slender, with exopod. Leg 5 enormously developed; ischium and merus not distinct; merus and carpus each with terminal spine ; propod flattened and expanded, the distal margin on both sides serrated ; length five times the greatest width. Pleopods small. An older specimen diflPers little from description given, but legs i and 2 have rudi- mentary chelae, and the telson has lost the lateral spines, while spine 4 is very much smaller than spine 2. Species A. IV (Figs. 42-49) Bermuda. Jan. 1935. Length 9-05 mm. Rostrum nearly as long as segment i of antennule, with one dorsal tooth. Carapace with dorsal tooth, and supra-orbital, antennal and pterygostomial spines, the two latter very small. In this specimen, which was examined soon after preservation, the lines shown on the carapace were distinctly marked in red. Telson 3^ times as long as wide, without lateral spines; apex with 5+5 spines, of which the second is much the longest ; posterior margin protuberant between the two large spines. Eyestalk much shorter than eye, together less than one-third length of body. Peduncle of antennule rather short and stout, one-fifth length of body, with pointed stylocerite ; flagella about zl times as long as peduncle, the outer one thickened at the base. Antennal scale five times as long as wide, with large terminal spine. Palp of maxillule with three setae, unsegmented. Exopod of maxilla very large and broad anteriorly; endopod unsegmented and without distinct inner lobes; endites well developed. Maxillipede i exopod with eight pairs of distal setae and enlargement at base bearing nine setae; endopod of four segments; coxa large, with long setae; epipod large. Maxillipede 2 ischium and merus, and propod and dactyl fused ; epipod large. Legs i and 2 chelate, carpus of leg 2 long and slender. Leg 3 with propod slightly expanded. Leg 4 propod very much expanded, about twice as long as wide ; exopod present. Leg 5 very long; ischium and merus fused and together i| times as long as carpus; propod i| times as long as carpus, and about twice as long as wide ; pleopods large. Measurements of legs {in mm.) Leg Propodus Carpus Ischio-merus I 2 3 4 S 0-55 0-3 1-4 X 0-34 1-5 X 0-75 4x2 07 1-6 i-i6 1-26 3-15 I -06 1-8 2-25 2-5 4-89 3-2 Eretmocaris. Species A. IV. Fig. 42. Species A. IV. Fig- 43- Species A. IV, telson. Fig. 44. Species A. IV, mandible. Fig. 45. Species A. IV, maxillule. Fig. 46. Species A. IV, maxilla. Fig. 47. Species A. IV, maxillipede i. Fig. 48. Species A. IV, part of leg i . Fig. 49. Species A. IV, leg 4 propodus. THE GENUS ERETMOCARIS 371 A single specimen of this remarkable form was taken by Dr Wheeler just before my visit to Bermuda. A number of red chromatophores were still visible, as shown in Fig. 42. Fig. 50. Eretmocaris. Species A. V. Fig. 52. Species A. V, mandible. Fig. 54. Species A. V, maxillipede i. Fig. 51. Species A. V, end of telson. Fig. 53. Species A. V, maxilla. Species A. V (Figs. 50-54) Eretmocaris corniger Bate, 1888, p. 900, pi. 145. E. corniger Ortmann, 1893, P- 79- St. 690. 03° 17' S, 29° 57' W. One specimen. St. 701. 14° 39' N, 25° 51' W. One specimen. Length of rostrum 3-1 mm. ; of body 67 mm. Rostrum very long and slender, with six dorsal teeth. Carapace with large dorsal 372 DISCOVERY REPORTS tooth and supra-orbital and pterygostomial spines ; antennal spine very small. Abdominal somite 3 with large, slightly procurved, dorsal spine ; pleura rounded ; somite 6 twice as long as deep. Telson three times as long as wide, without lateral spines, and 5+5 terminal spmes of which spine 2 is very large and the inner six quite small. No anal spine. Eyestalk shorter than eye, together less than one-third of body length. Peduncle of antennule shorter than eye, less than one-quarter body length; flagella extendmg beyond rostrum. Antennal scale longer than eye; flagellum a little longer than scale. Exopod of maxilla very broad and square in front ; endopod with three distinct inner lobes, unsegmented; four endites well developed. Maxillipede i endopod of four segments ; coxa large, bearing long setae ; exopod with five terminal setae, and basal enlargement bearing seven setae; epipod large, bilobed. Legs I and 2 not chelate. Leg 4 with exopod, endopod lost. Leg 5 lost, but with enlarged base. Pleopods present, small. Bate's specimen had the rostrum broken, but Ortmann describes his as having it as long as the antennular flagellum, and with eight dorsal teeth, including the tooth on the carapace. Both specimens came from the neighbourhood of Cape Verde. The Discovery specimens are from much farther south and west, and far out in the ocean. Species A. VI (Figs. 55-59) Eretmocaris doUchops Ortmann, 1893, p. 79, pi. v. St. 689. 05° 59' S, 29° 49' W. One specimen. Length 11 -6 mm. Rostrum small and slender, without teeth. Carapace with dorsal papilla ; no supra- orbital spine ; anterior margin produced into a pointed process above antenna, and lower margin with four teeth. Abdomen without spines, but pleura of somites 3-5 narrowing and rather acute ; somite 6 about three times as long as deep. Telson 3^ times as long as wide, without lateral spines; apex with 4+4 spines. Eyestalks extremely long, divided by a constriction into two segments, of which the first is the shorter; eye not distinctly marked off from stalk, together nearly as long as the body. Peduncle of antennule about one-quarter length of body ; flagella longer than peduncle. Antennal scale shorter than peduncle of antennule ; flagellum more than three times as long as scale. Maxillule, palp unsegmented, with five setae. Exopod of maxilla very large, extending far beyond endopod, rather narrow in front. Maxillipede i endopod of four segments ; exopod with eight terminal setae, and very slight basal enlargement bearing two setae ; epipod large, bilobed. Maxillipede 2, endopod of four segments ; epipod with rudiment of podobranch. Legs I and 2 chelate ; carpus of leg 2 very long and slender. No trace of epipods or THE GENUS ERETMOCARIS 373 arthrobranchs on legs. Legs 3-5 lost, but leg 4 apparently with exopod. Base of leg 5 not so large as that of leg 4, so that the leg is probably not enlarged. Pleopods slender. This must be the same form as that described by Ortmann, but it differs in having the first segment of the eyestalk only about half the length of the second, whereas Ortmann describes it as about equal. E. longicaulis Bate is a very similar form, with the eyestalk enormously long, and with pointed abdominal pleura; but the eyestalk is not jointed, and it appears that leg 5 must be much enlarged. Fig. 55. Eretmocaris. Species A. VI. Fig. 57. Species A. VI, maxillipede 2. Fig. 59. Species A. VI, part of telson. Fig. 56. Species A. VI, maxilla, setae of exopod omitted. Fig. 58. Species A. VI, maxillule. Species A. VII (Figs. 60-62) St. 97. 33° 11' S, 16° 55' E. One specimen. Length 10 mm. Rostrum very small, upturned, without dorsal tooth. Carapace without dorsal tooth, and with supra-orbital, antennal and pterygostomial spines. Abdominal somites 1-4 with dorsal spines, pleura rounded. 374 DISCOVERY REPORTS Telson as long as uropods, a little more than twice as long as wide, without lateral spines ; apex with a pair of stout spines and two pairs of setae between, the outer pair longer than the spines. Eyestalk very long and slender, together with eye more than half length of body. Peduncle of antennule much shorter than eyestalk, about one-quarter length of body. Mandible with two series of spines filling the gap between molar and incisor portions. Maxillule, palp unsegmented with three setae. Exopod of maxilla not extending much beyond endopod, not very broad; endopod with three small inner lobes bearing setae. Legs I and 2 chelate, carpus of leg 2 nearly equal to carpus and propod of leg i. Leg 4 Fig. 60. Eretmocaris. Species A. VII. Fig. 61. Species A. VII, part of telson. Fig. 62. Species A. VII, mandible. Measurements of species of Eretmocaris /row Barrier Reef and Atlantic Leg 5 Species Length mm. Rostral teeth Supra- orbital Eye : body Propodus width : Propodus: Antennule peduncle: body spine length carpus B.R.I 7-35 2 + 2-94 4-4 1-57 37 B.R. II 6-84 2 — 3-37 3-2 I 44 4-0 B.R. Ill 4-6 — — 1-26 — 2-o8 A.I 6-14 3 + 37 — 47 A. II 6-1 I + 2-1 5-2 1-6 3-4 A. Ill 6-24 I -1- 2-9 5-0 2-1 3-4 A. IV 9-05 I -1- 3-4 19 1-25 5-3 A. V 9-8 6 -1- 3-45 — — 4-8 A. VI II-6 — — 1-42 — 4-0 A. VII lO-O — ■ -1- 1-8 ■ — • 4-3 Total length is given inclusive of rostrum and lengths of eye and antennule are compared with this total length except in A. V where the rostrum is unusually long. In this specimen eye and antennule are compared with length of body without rostrum. THE GENUS ERETMOCAR IS 375 without exopod, long and slender. Leg 5 slender, a little shorter than leg 4. Pleopods long and slender. I have also two specimens from St. 1575 (18° 32' S, 4i°35'E) which agree in all respects with this description, but in which the antennal flagellum is retained. This flagellum is an enormously long, unsegmented rod, profusely covered with spicules. In one of the specimens the body is 9 mm. long and the flagellum 31 mm., though a little broken at the end. Key to the larvae described 1. Abdominal somite 5 with pair of dorsal spines ... ... R.S.I Without these spines 2. Leg 5 not larger than leg 4 Leg 5 distinctly larger ... 3. Eyestalk of two segments Eyestalk not divided 4. Abdominal somite 3 with dorsal spine Without this spine 5. Carapace with dorsal spine strongly procurved This spine not procurved 6. Eyestalk with spine Without spine 7. Eye and stalk more than half length of body ... Much less than half 8. Eye and stalk less than one-third length of body ; leg 4 propod dilated Eye and stalk more than one-third body 9. Supra-orbital spines present Absent 10. Eyestalk longer than eye, together nearly half body ... Eye and stalk about equal; together about one-third length of body 3 4 A. VI {E. dolicJiops Ortmann) A. VII A. V (£■. corniger Bate) 5 B.R. II 6 A.I 7 B.R. Ill A. IV ... 9 10 R.S. II A. II A. Ill DISCUSSION I have described here twelve forms of " Eretmocaris" larvae, and an endeavour must be made to determine whether any of them represent distinct generic types. One of them (R.S. II) can be quite certainly identified as belonging to Lysmata, and this larva agrees so closely with that of L. seticaiidata that the presumption is justified that any considerable departure from that type must represent a distinct genus. The characters of Lysmata, as founded on these two forms, are: (i) Rostrum short, with few dorsal teeth. (2) Carapace with antennal and pterygostomial spines, and dorsal tooth. (With or without supra-orbital spine?) (3) Abdominal somites without spines, and with rounded pleura. (4) Eyes carried on long, but not excessively long, stalks. (5) Endopod of antenna in stages I and II a slender rod, with one long seta; reduced to a short stump in stages III and IV. (6) Leg 4 with exopod; propod not dilated. (7) Legs 5 fully developed in stage II when legs 3 and 4 are rudimentary. (8) Leg 5 enormously large, with dilated propod. 376 DISCOVERY REPORTS What is perhaps the most important character, namely, the precocious appearance of leg 5, cannot, of course, be used when only isolated specimens of older stages are available, but it may prove that in all cases in which leg 5 has the same predominance over leg 4 it appears before it. Applying these characters, so far as is possible, to the forms here described it seems that the following may with some probability be assigned to Lysmata; Species R.S. II; B.R. I, II, III; A. I, II, III. Species A. I is evidently congeneric with the form referred to Hippolysmata by Kemp. In his form the rostrum had a series of small teeth at the base, and resembled that of some specimens in post- larval stage I which could be determined as H. ensirostris, and were taken at the same time off the Orissa coast. It may be accepted, then, that the larva of Hippolysmata does not differ strikingly from Lysmata in late stages ; but we do not know the early stages, and it may be that there is a difference in the time of appearance of leg 5. It is un- fortunate that I have no late stages of my form R.S. I which so closely resembles Lysmata, but differs remarkably in the late appearance of leg 5. It is not impossible that this may be the larva of Hippolysmata multiscissa, and that the genera may be distinguished by the order of appearance of the legs. There cannot be any doubt that R.S. I is at least closely related to Lysmata, and I refer it provisionally to Hippolysmata. Although the spine on the eyestalk is perhaps the most striking feature of Kemp's larva, and of A. I, I do not think it is such a character as must necessarily be found in all species of the genus. Kemp's figure shows that legs 1-4 are quite normal, and do not have the propod dilated. Of the remaining four species from the Atlantic A. IV from Bermuda differs from the Lysmata type only in having the propodus of legs 3 and 4 dilated, and that of leg 5 is of unusually broad form; there is still the same discrepancy in actual length between legs 4 and 5. How far such a difference can be regarded as generic it is impossible to say. In the closely allied species Caridio?i gordoni and C. steveni (Lebour, 1930) the larvae differ in the presence or absence of large procurved spines on the carapace, and also in the greater or less expansion of the propod in certain appendages, and we do not know enough to exclude the possibility of large differences between species in Lysmata and Hippolysmata. It seems best to regard this form also provisionally as within one of these two genera. The other three larvae differ so much from Lysmata that they cannot possibly be included in it or in Hippolysmata. Species A. V has still the same general form, with long eyestalk and, presumably, enlarged leg 5, and its genus must be sought within those allied to Lysmata. The fol- lowing genera are generally regarded as forming the Latreutid group of Hippolytidae: Latreutes Stimpson. Paralatreutes Kemp. Toseuma Stimpson. Lysmatella Borradaile. Trachycaris Caiman. Lysmata Risso. Gelastocaris Kemp. Hippolysmata Stimpson. Bythocaris Sars. Merguia Kemp. Mimocaris Nobili. THE GENUS TOZEUMA 377 The larvae of Latreutes and Tozeuma are known, and they have no resemblance whatever to Eretmocaris. Bythocaris need not be considered, as it is said to have no free larva. Gelastocaris, Lysmatella, Merguia are not known from the Atlantic; so that if the " Latreutid group" is a reality we are confined in speculation to Trachycaris and Mimocaris. That seems to be all that can be said. The remaining two species, A. VI and A. VII, cannot be even approximately placed. The only character which A. VI has in common with Eretmocaris is the enormously long eyestalk, and this is also peculiar in being jointed. Leg 5 is probably not enlarged. Until something is known about its legs it is useless to speculate about its identity. Larva A. VII, again, has extreme elongation of the eyestalk, but here it is known that leg 5 is not modified, and leg 4 has no exopod. Having regard to the characters of such Hippolytid larvae as are known, I regard it as most improbable that either of these larvae can be placed among the Hippolytidae at all, and the only suggestion that can be made is that they may be Pandalids. In Chlorotocella^ there is great elongation of the eye, though without distinction between eye and stalk, and I have referred to Pandalidae (1924, p. 114) a larva which agrees with Pafidalus platyceros in having the abdominal and carapace margins serrated, while the eyestalks are as much elongated as in some Eretmocaris. This character alone cannot therefore be regarded as confined to the Latreutid group of Hippolytidae, and the genus Eretmocaris is an ill-founded composite of unrelated larvae. THE GENUS TOZEUMA STIMPSON I am indebted to Dr J. F. G. Wheeler for specimens of Tozeuma carolinense Kingsley in stage I which were hatched at Bermuda Biological Station on 30 May 1936. For these and other interesting specimens I would like to express my thanks to him. Stage I (Figs. 63-70). Length of rostrum 0-9 mm. ; body 3-0 mm.; total length 3-9 mm. Rostrum nearly twice as long as antennule, without teeth, rather deep, parallel-sided in side view as far as distal third where the lower margin slopes sharply upwards to end in an acute point. Carapace without supra-orbital spines; margin smooth, with small pterygostomial spines. Abdominal somite 3 with large pointed dorsal process, curving slightly forwards; somite 5 elongated, with a pair of large lateral spines. Telson concave behind, the spines all of about the same length. Antennule with very short feathered seta and four aesthetes on exopod. Of the aesthetes three are long and blunt-ended, one much stouter than the other two, and one is much shorter and pointed, apparently without the membrane seen in Eretmocaris R.S. I. ^ Material from Ghardaqa, not yet published. 4-2 Fig. 63. Stage I. Lateral. Fig. 66. Antenna. Fig. 69. Maxillipede 2. Tozeuma carolinense . Stage I. Fig. 64. Mandibles. Fig. 67. Maxillipede i. Fig. 70. Maxilla. Fig. 65. Maxillule. Fig. 68. Telson. THE GENUS TOZEUMA 379 Antennal scale with two small outer setae and nine inner and terminal setae ; with two distal segments traceable. Inner margin with slight protuberance before first seta. Endopod spine-like. Maxillule, endopod unsegmented with five spines. Maxilla, exopod with five setae ; endopod unsegmented, with three inner lobes bearing setae; apex with three setae. Maxillipedes 1-3 each with endopod of four segments, and with apical setae of segment 4 long and slender ; exopods with three apical setae ; exopod of maxillipede i with one outer marginal seta, the other two with a pair of setae. No later stages of this species are known ; but I am indebted to Mr F. S. Russell for permission to describe later larvae belonging to two distinct species of Tozeuma taken in plankton from the Great Barrier Reef. One of these (species B.R. II) is represented by a single specimen, and is evidently the same as that described by Kemp (1916, p. 400) as the larva of T. armatiim. In view of the fact that several species are known from Australian waters I do not feel justified in accepting so precise an identification. Tozeuma species B.R. I Of this form there are eight specimens apparently representing five stages, of which the youngest is certainly stage III, and the oldest probably stage VIII. Stage VII is not represented, and it is probable that there is also a ninth stage. Stage III (Figs. 71, 72). Length of rostrum 2 mm., body 5-6 mm. ; total length 7-6 mm. Rostrum with two small ventral spines, broadening at base over eyes, with a spine on either side representing the supra-orbital spines. Abdominal somites very elongated ; somite i with anterior projection overlapping carapace, but not produced, as in later stages, into a spine. Somite 3 with an enormous posterior process ending in a pro- curved hook; somites 5 and 6 with lateral spines. Anal spine absent. Telson large, deeply hollowed, with small lateral spine ; spine 2 fused with angle of telson ; spine 3 reduced to a small seta. Peduncle of antennule segmented ; endopod represented by a small tubercle bearing a seta. Antennal scale slender, unsegmented, with outer apical spine and thirteen setae; endopod spine-like, about three-quarters length of scale. Exopods of maxillipedes 2 and 3 with eight and ten setae. Legs I and 2 rudimentary, leg i fairly large and biramous. Uropods with endopod rudimentary; exopod with nine setae. Stage IV (Figs. 73, 74). Length of rostrum 4-0 mm., body 7-4 mm. ; total length 11-4 mm. Rostrum with three small ventral spines. Carapace with large supra-orbital spines and pterygostomial spines. Abdomen unchanged, but anterior overlapping part of somite i with sharp ventral angle, and somite 5 with very small posterior dorsal spine. Telson more deeply forked, with 8+8 small spines, of which three are on outer margin and the fourth forms the point of the fork. Fig. 71- Stage III. Fig. 74. Stage IV, telson. Tozeuma B.R. I. Fig. 72. Stage III, part of antennal scale. Fig. 75. Stage V, telson. Fig. 73. Stage IV. THE GENUS TOZEUMA 381 Endopod of antenna unchanged. Leg I developed, its exopod with eight setae. Legs 3-5 rudimentary. Pleopods absent. Endopod of uropods nearly as long as exopod, with several setae. Stage V (Fig. 75). Length of rostrum 6-o, 3-6 mm. Body 9-3, 7-56 mm. Total length 15-3, n-i6mm. The great difference in length between these two specimens suggests that two species may be represented, but they agree in all other respects, and differ very little from the preceding stage. The telson is narrower, the limbs of the fork closing inwards (Fig. 75). The anterior overlap of abdominal somite i has now a definite pointed process. The antenna is unchanged. Legs I and 2 are developed, with exopods, and there are three pleurobranchs, the first, on leg I, large and foliated, the third very small and rudimentary. Pleopods absent. Stage VL Length of rostrum 8-2 mm., body 11-05 mm.; total length 19-25 mm. Abdominal somites 4 and 5 with small posterior dorsal spine. Antenna unchanged. Legs 1-3 developed, with exopods. Five pleurobranchs, decreasing in size backwards to a small rudiment on leg 5. No pleopods. Stage VIII? (Figs. 76-84). Length of rostrum io-6 mm., body 14-7 mm.; total length 24-7 mm. Rostrum with about sixteen small ventral spines. Abdominal somite 2 with small anterior spine on pleura. Telson very long and narrow (width : length about i : 6), deeply forked at end, with three pairs of large outer spines and four pairs on posterior margin. Anal spine present. Antennule, exopod faintly segmented, with sbc groups of aesthetes; endopod shorter than exopod; segment i of peduncle with long pointed stylocerite and small ventral spine. Antennal scale very long and slender, narrowed at end ; flagellum very much longer than scale, with basal segment distinct. Maxillule, palp unsegmented, with five spines. Maxilla, endopod with small basal segment marked off, distal part with six setae; exopod very large, extending beyond endopod, and broad and square at end. Legs I and 2 chelate, leg i shorter than leg 2, dactyl shorter than palm. Carpus of leg 2 nearly as long as chela. Legs 4 and 5 without exopods. Pleopods large, with small setae. Fig. 76. Stage VIII?. Fig. yg. Stage VIII?, mandibles. Fig. 82. Stage VIII?, antenna (setae omitted). Tozeuma B.R. I. Fig. 77. Stage VIII?, telson. Fig. 80. Stage VIII?, maxillule. Fig. 83. Stage VIII?, leg 2. Fig. 78. Stage VIII?, maxilla. Fig. 81. Stage VIII?, antennule. Fig. 84. Stage VIII?, leg i. THE GENUS TOZEUMA Tozeuma species B.R. II 383 Stage VI? (Figs. 85 88). Length of rostrum 15-3 mm., body io-8 mm. ; total length 26-1 mm. Rostrum with about twenty-five ventral spines ; towards the end larger spines alternate with very minute spinules. General form the same as in species I except for the immensely long rostrum and the dorsal process of abdominal somite 3, which has a large posterior spine at its base. Tozeuma B.R. II. Fig. 85. Stage VI?. Fig. 87. Stage VI?, maxillule. Fig. 86. Stage VI?, telson. Fig. 88. Stage, VI? maxilla. The specimen is somewhat more developed than stage VI of species I, though it has no pleopods, for the antennule has a small rudiment of the stylocerite and the flagellum of the antenna is a rod nearly half the length of the scale. The telson is five times as long (median length) as it is wide anteriorly, deeply forked, with three pairs of lateral spines and four pairs within the fork. The mouth-parts are as in species I (Figs. 87, 88). There can be little doubt that this is the same species as that figured by Kemp, and 384 DISCOVERY REPORTS it may well be that it is the larva of T. armatum, since the posterior spine on the abdominal process may perhaps persist as the spine present on somite 3 in the adult. Coutiere (1905, p. 21) has described a larva, Caricyplnis aciihts Cout., which he gives reasons for identifying with Tozeiima. This supposition can now be dismissed, since his larva is totally unlike that here described, the identification of which cannot be doubted. CHORISMUS ANTARCTICUS PFEFFER The plankton collected by the 'Discovery' and the 'William Scoresby' in the neighbourhood of South Georgia contains larvae of three species of Caridea. Of these one cannot be identified ; but the other two belong to Crangon antarcticus Pfeffer and Chorismtis mUarcticiis Pfeffer, the former being the commoner of the two. In Chorismiis ontarcticiis the eggs are large, 1 7 x 1-3 mm. I have not seen any eggs at all near hatching, but in one specimen they were sufficiently far advanced for the structure of the embryo to be made out without much difficulty. Rudiments of the full number of appendages were present, and the telson could be seen to be deeply cleft and apparently with 7+7 spines. The telson of the first free larva diff^ers both in shape and in number of spines, so that a more primitive form of telson seems to be retained in the embryo. It is usual in the Caridea for the embryonic telson to bear only six spines, and I am not certain that there is an exception in this case. More advanced material should be examined. Stage I. Length 6-6-7-5 mm. (Figs. 89-95). Rostrum reaching end of antennular peduncle. Carapace with dorsal papilla and pterygostomial spine. Abdominal somites without spines. Telson slightly concave behind, with 10+ 10 spines. Peduncle of antennule unsegmented ; endopod absent. Scale of antenna constricted at end, with one outer seta near end and eight inner and terminal setae crowded together; four distal segments indistinctly marked; flagellum longer than scale, with basal segment marked oflt; constricted at end into a short denticulate spine. Palp of maxillule with five setae, unsegmented. Endopod of maxilla rather large, with three inner lobes ; exopod large, with numerous setae, the seta at proximal end very large; four endites well marked. Maxillipede i, endopod of four segments; coxa large; epipod present, bilobed. Exopods of maxillipedes with three apical setae, those of maxillipedes 2 and 3 with twelve and fourteen lateral setae respectively. All legs present, large, legs 1-3 with exopods without setae. Legs i and 2 with rudimentary chelae. Five pairs of pleuro- branchs present. Pleopods absent, or present as very small papillae. THE GENUS CHORISMUS 38s Chorismus antarclicus Fig. 89. Stage I. Lateral. Fig. 91. Antennal scale. Fig. 93. Maxilla. Fig. 95. Telson. Fig. 90. Mandible. Fig. 92. Maxillule. Fig. 94. Maxillipede I. 386 DISCOVERY REPORTS Stage II. Length 8-9 mm. (Figs. 96, 97). Rostrum stouter, in some cases with three or four very minute dorsal teeth, but generally smooth. Carapace with supra-orbital spine. Telson marked off from somite 6 but with the same shape and number of spines. Anal spine present. Fig. 96. Stage II. Chorismus antarcticus Fig. 97. Stage II, mandible. Fig. 98. Stage III. Antennule with peduncle segmented. Antennal scale unsegmented and with outer apical spine ; flagellum longer than in stage I and without apical spine. Legs much more developed, legs i and 2 with chelae; exopods of legs 1-3 with setae. Pleopods present, small. THE GENUS CHORISMUS 387 This stage differs very little from stage I, but can be most easily separated by the presence of a spine on the antennal scale. Stage III. Length 11 mm. (Fig. 98). Rostrum a trifle longer than antennular peduncle ; with five dorsal teeth and either smooth or with one small spine below. Telson with posterior margin straight; width at end more than half length (58 : 99), with 10 + 10 spines. Antennule with stylocerite, and with inner branch as long as outer, both unsegmented. Flagellum of antenna about 4 mm. Legs almost fully developed, legs i and 2 with large chelae. Carpus of leg 2 divided into four segments. Exopod of maxillipede i not widened at base and without setae in this position. Epipods of maxillipede 3 and legs i and 2 absent in one of two specimens examined, but traceable in the other. No trace of arthrobranchs. Pleopods large, without setae. Uropods fully developed. Of sixty-nine seen only six were in this stage, which is no doubt the last larva. Post-Larval Stage L Length about 16 mm. (Figs. 99-105). Rostrum slender, not widening towards end, with 6-8 dorsal teeth and 3-4 ventral. Carapace without supra-orbital spine; with antennal and pterygostomial spines. Ab- dominal somites 3 and 4 slightly gibbous ; pleura of somite 5 slightly pointed. Telson parallel-sided, about three times as long as wide, with two pairs of lateral spines and 12-16 slender spines on the slightly convex posterior margin. Anal spine large. The appendages are all nearly of adult form. Mandible with small rudiment of palp. Carpus of leg 2 of eight segments. Exopod of maxillipede 3 large, without setae. Legs 1-3 with large vestigial exopods. Gill formula and epipods as in adult, but no arthrobranch seen on maxillipede 3. Some specimens differ in having no trace of an exopod on leg 3 and those of legs i and 2 reduced to small papillae, while somites 3 and 4 are not gibbous; but they are either of the same size, or but little larger, and do not seem to represent a second stage (Fig. 99). Measurements of leg 2 (in mm.): Dactylus Propodus Carpus Merus Ischium Last larva Post-larval i Post-larval 2? 0-23 0-27 0-3 0-39 0-43 0-45 071 1-2 1-5 0-49 i-o 1-05 0-45 0-9 0-85 Coutiere (1905, p. 26; 1907, p. 15) has suggested that the larva described by him as Hippocaricyphiis bigibbosiis may belong to Chorismus. The possibility that it might be C. antarcticus is now excluded, and it is most unlikely that the development of C. tuber- ciilatiis could differ so profoundly as Coutiere 's supposition would require. Coutiere 's larvae were taken at two stations in the central Atlantic,^ a fact which, in itself, would seem to exclude their identification with a genus confined to the Antarctic. 1 St. 1851, 36^^ 17' N, 28° 53' W; St. 2187, 38° 04' N, 26° 07' W. Fig. 99. Post-larval stage I?. Fig. 102. Post-larval I, maxillule. Fig. 104. Post-larval I, maxillipede i. Chorismiis antarcticus Fig. 100. Post-larval I, telson Fig. loi. Post-larval I, mandible. Fig. 103. Post-larval I, maxilla, setae of exopod omitted. Fig. 105. Post-larval I, maxillipede 2. THE GENUS CHORISMUS 389 DISCUSSION This species provides an interesting example of abbreviated development. In Bythocaris and Cryptocheles the free larva is entirely suppressed (Sars, 1885), the young hatching in the form of the adult. In Spirontocaris polaris the first free larva is much more developed than in Chorismus antarcticus and has lost all the characters distinctive of stage I (Stephensen, 1935). In Chorismus the first larva, though it has the full number of appendages, retains certain characters of the normal first larva which one would think might well have been lost. For example the antennule is unsegmented and has no trace of the inner branch; the antennal scale has no spine, and retains traces of segmentation and few setae, while the flagellum, though so far developed, terminates in a denticulate spine, showing that it has developed from the spine-like form found in Hippolyte and Spirontocaris but not in Caridion. The exopods, although furnished with far more setae than is normal in stage I, have the apical setae arranged asymmetrically, as is usual in stage I in Hippo- lytidae and certain genera of other families of Caridea. When development is shortened it is commonly the case that the telson has more than the normal number of spines (e.g. Sabinea, Palaemon boreUii). Sollaud (1923) has already noted the retention of some of the characters of the normal stage I in Palaemo- nidae in which larval development is greatly shortened or suppressed, and particularly that the telson retains the wide triangular form until transformation to the adult. Distribution of larvae of Chorismus antarcticus Station no. Position Larvae Post-larval 125 53° 28' S, 36° 20' W 3 — 127 53° 48' S, 37° 08' W I — 130 54° 06' S, 36° 23' W 2 — 134 54° 22' S, 35° 56' W 49 — 135 54° 22' S, 35° 39' W — 54° 62° 06' S, 55° 08' W — 541 62° 22' S, 55° 23' W — 544 62° 26' S, 57° 15' W — 546 62° 46' S, 57° 11' W — WS 18 54° 07' S, 36° 23'-W — WS 19 54° 00' S, 36° 20' W — WS 20 53° 52' S, 36° 00' w — WS33 54° 59' S, 35° 24' W — WS42 54° 41' S, 36° 47' W I WS47 54° 22' S, 37° 50' W 2 — WS48 54° 24' S, 38° 09' W 2 1 WS243 51° 06' 8,64° 30' w — I MS 67 S. Georgia, E. Cumberland Bay — 2 MS 71 )» — 13 MS 74 ») — 2 Total 69 19 390 DISCOVERY REPORTS In Spirontocaris polaris the first larva has the antennule fully segmented, with small inner branch ; the antennal scale has the apical spine, and no segmentation ; supra-orbital spines are present ; the exopods of the maxillipedes (they are absent from all legs) have four symmetrically placed apical setae (Stephensen). The most remarkable feature of shortened development is the apparent difficulty of eliminating the three first obligatory stages of the normal larva. In normal development the uropods appear almost without exception in stage III, and, however fully the free larva is provided with pereiopods, or even pleopods, when hatched, the uropods are never present. In the few cases in which the course of abbreviated development is satisfactorily known the uropods appear, as in the normal larva, in stage III even if the adult form is to be attained at the next moult — for example, Palaemonetes varians lacustris, Palaemon potiiina (SoUaud, 1923). Even more striking is the fact that in Astacusfluviatilis, although the young on hatching have all the appendages more or less of aduh form, still the uropods do not appear till stage III. In ScJerocrangon, although the young hatch in the adult form, and attach themselves to the parent, the embryo just before hatching has no uropods (Wollebaek, 1906). SARON MARMORATUS (OLIVIER) AND ALLIED FORMS Saron marmorattis is not uncommon on the coral reefs at Ghardaqa, and two females hatched their eggs in the laboratory. A close similarity was at once noticed between these larvae and others which were at the time common in the plankton, and it was assumed at first that all belonged to this species. They were all characterized by their rich colour, generally red, and compact form, with small posterior tooth on the carapace. It was not possible to spare the time necessary for detailed examination, but it became evident later that there must be at least two species present, though they could not be easily distinguished without risk of injury to those which it was desired to keep alive. I was also singularly unfortunate in getting these larvae to moult to post-larval. Exami- nation of the preserved material shows that there are actually at least three distinct forms, and that the material does not admit of separating them into series. It is also impossible to determine with sufficient certainty that any of the later larvae belong to Saron. The close resemblance of these larvae to Latreutes (Gurney, 1936) is evident, but they all lack the dorsal spines present on abdominal somites 4 and 5 in L.fticoruni, and also the spine present in that species in the middle of the margin of the carapace. They agree in general form, in shape of telson and in the presence of a small tooth at the posterior angle of the carapace. Further, the fact that in all these forms there are four apical setae on the exopods is an important point when we take into account the fact that there are three in all other Hippolytidae known. I have given below a description of a larva from the Barrier Reef which I attribute with some doubt to Latreutes, for the sake of comparison with the Red Sea larvae. It is quite certain that these larvae all belong to closely related species of Hippolytidae, THE GENUS SARON 391 though it is impossible to identify them, and it may well be that Latreiites is among the forms from the Red Sea. In view of the doubtful relations between the various genera of the Hippolytidae further study of this well-defined group of larval forms would be useful, and the problem of their identity would be easily solved at Ghardaqa if it could be attacked at a rather later season than that of my own visit, and with better opportunities for getting plankton. Saron marmoratus (Olivier) Stage I. Length 2-5 mm. (Figs. 106-112). Carapace with small pterygostomial spine and two anterior marginal teeth; anterior and posterior papillae present, small; ventral margin sinuate, with small tooth at posterior angle; rostrum absent. Pleura of abdominal somite 2 pointed, the rest rounded ; without dorsal or dorso-lateral spines. Telson rather narrow, with 7+7 spines. Exopod of antennule with inner feathered seta long, and four aesthetes, of which one is slender and pointed, with a narrow hyaline margin. Antennal scale short and broad, with two outer setae and ten inner and terminal setae ; four terminal segments distinct ; endopod spine-like. Maxillule without outer seta on basis ; endopod unsegmented, with five setae. Endopod of maxilla segmented, with basal lobe very well marked ; lacinia 2 large; exopod with five setae. Exopods of maxillipedes 2 and 3 with six setae, the four apical setae symmetrically placed. Rudiments of legs i and 2 biramous. General colour very red, with many deeply placed red chromatophores. When these are contracted a pale yellow tinge remains general. As I supposed at the time that the Saron larvae were hatched that the larvae found in the plankton belonged to the same species, no attempt was made to keep the larvae through a moult. I find it now impossible to be sure which, if any, free larvae belong to Saron. There is a species which has the pleura of the abdomen of the shape characteristic of S. marmoratus, but there is also another form of stage I which is like Saron, but has a small downturned rostrum, which Saron has not. There are therefore apparently three species. One which is described below is quite distinct by reason of its large rostrum and the shape of the abdominal pleura, but the others cannot be satisfactorily dealt with. One specimen in the last stage moulted to post-larval and should have given the opportunity of identifying its genus, but this specimen was lost by accident before the essential characters had been made out. I give, therefore, only a description of a stage II larva which I believe to be Saron, and leave the rest undescribed. Stage II. Length 3-3 mm. Rostrum shorter than antennule, triangular in dorsal view, and very slender in side view. Carapace without supra-orbital spine ; with large pterygostomial spine and an anterior process extending under the eye. Abdominal somite 2 with pleura large and pointed, the rest rounded. 392 DISCOVERY REPORTS Antennule with base widened, and rudiment of stylocerite. Antenna, basis with small ventral spine; endopod short, with small apical spine; scale constricted at end, with traces of segmentation. Exopod of maxilla with large proximal setae ; eleven setae in all. Saron marmoratiis Fig. 107. Stage I, lateral. Fig. no. Stage I, mandible. Fig. 108. Stage I, antennule, part. Fig. III. Stage I, maxillule. Fig. 106. Stage I, dorsal. Fig. 109. Stage I, antenna. Fig. 112. Stage I, maxilla. Exopod of maxillipede i with outer seta at base and five apical setae; coxa large; endopod of four segments without outer setae; epipod large, not bilobed. Exopods of maxillipede 3 and legs i and 2 with six setae. Legs 3-5 rudimentary ; leg 3 large and biramous. Three rudimentary pleurobranchs present. Pleopods and uropods absent. FORMS ALLIED TO SARON 393 Hippolytidae, Species R.S. I. Stage I. Length 3-9 mm. (Figs. 113, 114). Rostrum large, extending beyond antennules. Carapace as in Saron, with small posterior tooth and anterior spine and marginal teeth. Pleura of abdominal somites 2-5 pointed. Telson narrow, as in Saron. Antennal scale a little broader than in S. marmoratus; endopod stout at base and constricted at end into a denticulate spine. Exopods of maxillipedes 2 and 3 with four apical setae. Rudiments of legs 1-5 traceable. Stage II. Length 4-3-4-8 mm. (Figs. 115-117). Rostrum large, broadening at base over eyes. Carapace without supra-orbital spine ; with large, bluntly pointed, process under eye. Telson unchanged except for additional inner pair of spines. No anal spine. Antennule stout, peduncle segmented, with ventral spine, and slightly enlarged at base with trace of stylocerite. Antennal scale with segmentation still traceable, and 17 inner and terminal setae; endopod stout with basal segment marked, and small terminal spine; basis without spine. Legs I and 2 with incipient chelae; legs 1-3 with exopods bearing 8.6.6 setae; legs 4 and 5 rudimentary, leg 4 with rudiment of exopod. Pleopods present as small buds. Rudiments of pleurobranchs on legs 1-4. The description is taken from a specimen moulted from stage I. Stage III. Length 5-4 mm. (Figs. 118-121). Telson nearly rectangular, slightly wider at end, the width about half the length; no lateral spines; terminal spines 7+7. Anal spine absent. Antennule with small stylocerite; exopod with feathered seta and aesthetes in three groups, but not segmented. Antennal scale unsegmented, without outer setae and generally without outer apical spine. A very small spine seen in one specimen (Fig. 120). Endopod nearly as long as scale. Legs I and 2 subchelate, leg 4 with exopod. Pleurobranchs large; rudiments of epipods on maxillipede 2 and legs 1-4. Pleopods bilobed. Uropod, exopod with 14 setae; endopod small, not jointed to basis, with two small apical setae. To show how very much alike these different species of red carid larvae are I give a figure (Fig. 122) of a second species, in stage IV, which can be distinguished by the form of the rostrum and shape of the abdominal pleura. Stage IV. Length 5-3-6-7 mm. (Figs. 123-128). Rostrum in some cases with two small ventral teeth and one dorsal; carapace un- changed, but with very small tooth in front of anterior dorsal papilla. Telson nearly or quite three times as long as wide, narrowed at end, without lateral spines, and with 5+5 terminal spines. Anal spine absent. Antennule with stylocerite but no trace of Fig. 113. Stage I, telson. Fig. 116. Stage II, antenna. Hippolytid R.S. I. Fig. 114. Stage I, antenna. Fig. 117. Stage II, maxilla. Fig. 115. Stage II, lateral. FORMS ALLIED TO SARON 395 otocyst. Antennal scale without outer distal spine in all specimens seen ; endopod as long as exopod, unsegmented. Exopod of maxilla very narrow proximally, with one large seta at end. Maxillipede i with large epipod having a small anterior lobe which seems to be a rudiment of a podo- branch ; endopod segments i and 2 with outer seta ; coxa large. Legs I and 2 with large chelae ; carpus of leg 2 undivided. Leg 4 with exopod bearing 120 118 Fig. 118. Stage III, dorsal. Fig. 120. Stage III, antennae. Hippolytid R.S. I. Fig. 119. Stage III, carapace. Fig. 121. Stage III, mandible, left side. eight setae. Epipods present on legs 1-4. Pleopods large, without setae, but with rudiment of appendix interna. This stage was observed to arise by moult from stage III and to moult itself to post- larval. Unfortunately this specimen died before the moult was completed, and the legs were not fully freed. Post-larval Stage I (Figs. 129-133). Rostrum with three large ventral teeth and six dorsal, of which three are on the carapace ; carapace with antennal spine and a group of three small spines at anterior angle, but without supra-orbital. Telson long and narrow, its greatest width a little less 130 132 133 Hippolytid R.S. II. Fig. 122. Stage IV, dorsal (of a species distinct from R.S. II). Figs. 124, 125. Stage IV, mandible, right and left. Fig. 127. Stage IV, maxillipede i. Fig. 129. Post-larval stage I, rostrum. Fig. 131. Post-larval stage I, maxillule. Fig. 133. Post-larval stage I, maxillipede 2. Fig. 123. Stage IV, carapace. Fig. 126. Stage 1V, maxilla. Fig. 128. Stage IV, end of telson. Fig. 130. Post-larval stage I, telson. Fig. 132. Post-larval stage I, maxilla. FORMS ALLIED TO SARON 397 than one-third the length ; two pairs of dorso-lateral spines ; posterior margin with two pairs of spines, the inner pair very large, and four pairs of long setae, of which the outer three are feathered. Antennule with narrow pointed stylocerite. Antennal scale broad, with outer apical spine. Mandible with incisor process and rudimentary unsegmented palp. Maxillule, proximal endite narrow; endopod small, curved, with two setae. Maxilla, endite i very much shorter than endites 3 and 4, with a small lobe representing endite 2 ; proximal setae of exopod not very long ; endopod small and slender. Maxillipede i, exopod with five setae on basal part, and six distal; coxa and basis large ; epipod large, with small anterior lobe like a rudimentary podobranch. Epipod of maxillipede 2 as that of maxillipede i, with anterior lobe representing a podobranch. Exopod of maxillipede 3 large, with six setae. Small epipods present on legs 1-4, but no trace of arthrobranchs. Legs i and 2 not free from moult, but carpus of leg 2 distinctly divided into three segments, segments i and 3 very much smaller than 2. Although this specimen is not perfect most of the characters of systematic importance can be established ; but it cannot be definitely stated that the carpus of leg 2 shows the final segmentation ; or that the arthrobranchs may not appear in later stages. Judging by the development of Spirontocoris cranchii and S. occulta (Lebour, 1936) one would expect the carpus to be divided from the first to the full, or nearly the full, number of segments found in the adult, and it is improbable that there would be no trace of arthro- branchs, if the adult possesses them. According to the keys to the genera of Hippolytidae given by Caiman (1906) and Kemp (1914) the combination of characters found in this species is not found in any known genus. The larvae are so exactly like those of Saron that there cannot be a doubt that they belong to a genus related to it ; but no such genus is known which lacks arthrobranchs. The resemblance to the larvae of Latreiites is rather close, but Latreutes and its allies have a mandible without palp or incisor process. The lack of a palp is not a character of much importance, since in the two very closely allied species Spirontocaris cranchii and S. occulta (Lebour, 1936) the latter has a palp while the former has not; but the absence or presence of an incisor process seems to be a character of primary value. One point is of rather special interest in this developmental series, namely the small number of stages. In most Hippolytidae known there are as many as nine stages (Lebour, 1936, p. 99), but these may be greatly abbreviated as shown above for Chorismus which has three only. In the present form there are only four, but stage I is a perfectly normal larva with quite small rudiments of the legs. In Hippolyte proteus there appear to be four stages (Gurney, 1927) instead of the nine which Miss Lebour has found in H. varians. A total number of nine stages is found commonly among Caridea, and is not known to be exceeded. It is interesting to note that both Eraser and John in their admirable accounts of the development of Euphausidae (1936) agree in finding three Calyptopis and six (or rarely seven) Furcilia stages, the term Furcilia being used to include all stages between Calyptopis and post-larval. Accepting, then, the interpretation of 398 DISCOVERY REPORTS Stages 1-3 of Caridea as equivalent to the Protozoea of Penaeidea and the Calyptopis of Euphausidae (Gurney, 1926), the total number of possible stages in Caridea is precisely the same as it is in Euphausidae. LATREUTES MUCRONATUS (STIMPSON)? Stage IV. Length 3-9 mm. (Figs. 134-137). Rostrum longer than antennule, with three dorsal teeth ; carapace with tooth in front of anterior dorsal papilla; ventral margin with a series of small anterior teeth and Fig. 134. Stage IV. Latreutes mucronatiis} Fig. 135. Telson. Fig. 136. Antenna. Fig. 137. Maxilla. conspicuous posterior tooth. Pleura of abdominal somite 2 very large, pointed behind; somite 3 with small median dorsal spine; somite 5 with large lateral spines. Telson four times as long as wide, with two pairs of lateral spines and 6 + 6 terminal spines. Anal spine present. GENERIC CHARACTERS OF HIPPOLYTIDAE 399 Antennal scale narrowing towards end, with large apical spine ; endopod short and stout. Maxilla, exopod very broad in front, with very large proximal seta. Maxillipede i coxa large; epipod large; exopod with five setae. Maxillipede 3 with ten setae on exopod. Leg I endopod short, not distinctly segmented, with incipient chela; exopod with ten setae. Leg 2 endopod rudimentary, bifid at end. Legs 3-5 rudimentary. No rudiments of exopods seen on legs 2-5. Four large rudiments of pleurobranchs on legs 1-4 and very small one on leg 5 ; no epipods. Pleopods present as small buds. Uropods with numerous setae on both branches. One specimen of this form was found in plankton from Barrier Reef St. i . A young specimen, perhaps in post-larval stage 2, St. 65 is recognizable as belonging to L. miicronaUis. The larva almost certainly belongs to the genus Latretites, and it is a fair speculation to refer it a species which was also taken in plankton in much the same locality. DISCUSSION AND SUMMARY OF GENERIC CHARACTERS The general characters of the Hippolytid larvae then known were summarized by Miss Lebour (1932), and I give an extended summary below. It must be confessed that they do not throw any clear light upon the systematic grouping of the Hippolytidae, though they suggest that the separation of a Latreutid group is not justified. As Miss Lebour has said, the characters of the larvae point to a division into new families or subfamilies, but our knowledge is not sufficiently complete to carry the suggestion farther. It is peculiarly disappointing that I am unable to determine the parentage of the remarkable group of larvae from the Red Sea, since the characters of Saroti and Latreutes are left uncertain. Before any definite conclusions can be put forward it is necessary not only to know more about these two genera, but also the development of Thor, which, on adult structure, seems to be inseparable from Spirontocaris (Lebour, 1936)- The larval characters of the Hippolytidae, so far as they are known, may be sum- marized thus: Hippolyte Rostrum present in stage I, remaining simple, broad at base. Carapace without supra-orbital spines, or with very small spines in late stages ; margin denticulate. Abdominal somite 5 with dorso-lateral spines. Telson broad in early stages, and later with two pairs of lateral spines and twelve terminal. Anal spine present in late stages. Antennal scale with tendency to lose segmentation; endopod spine-like in stages I and II. 400 DISCOVERY REPORTS Maxillule with outer seta. Exopods of maxillipede 3 with three terminal setae in stage I. Exopods on legs i and 2 only; legs 3-5 remaining apparently non-functional. From four to nine larval stages. Chorismus Rostrum long and rather stout. Carapace with supra-orbital spines ; margin smooth. Abdominal somites without spines. Telson broad, with twenty spines in stage I. Anal spine present. Antennal scale segmented, with one outer seta; endopod a long flagellum in stage I. Maxillule without outer seta. Exopod of maxillipede 3 with three apical setae in stage I. Exopods on legs 1-3. Development greatly abbreviated. Spirontocaris Rostrum absent in stage I (except in S. sptnus), remaining very short and broad. Carapace with small supra-orbital spines and denticulate margin. Abdominal somite 5 with or without dorso-lateral spines. Telson rather deeply indented, in late stages with two pairs of lateral spines and twelve terminal. Anal spine present from stage I. Antennal scale segmented, endopod spine-like. Maxillule without outer seta. Exopod of maxillipede 3 with three terminal setae in stage i . Exopods on 2, 3 or 4 pairs of legs. Legs 3-5 long, slender, functional. Development in four species greatly abbreviated. Caridion Rostrum present in stage I ; becoming long and slender. Supra-orbital spines present ; margin of carapace denticulate. Abdominal somite 5 with dorso-lateral spines. Telson very deeply indented; with two pairs of lateral spines and twelve terminal in late stages. Anal spine present in late stages. Antennal scale segmented; endopod a rod with two terminal setae in stages I and II. Exopod of maxillipede 3 with three terminal setae in stage I. Exopods on legs 1-4. Legs 3-5 very long and slender, propodus sometimes expanded slightly. Nine larval stages. GENERIC CHARACTERS OF HIPPOLYTIDAE 401 Lysmata and Hippolysmata Rostrum long and slender in stage I. Carapace with supra-orbital spines and denticulate margin. Abdominal somite 5 with or without dorso-lateral spines. Telson deeply indented; in late stages with two pairs of lateral spines and ten terminal. Anal spine absent. Eyes borne upon long stalks. Antenna! scale segmented ; endopod a rod with one seta. Maxillule without outer seta. Exopod of maxillipede 3 with three terminal setae in stage I. Exopods on legs 1-4. Leg 5 appearing before leg 4 and of enormous size; with propod greatly expanded. Probably nine larval stages. Tozeuma Rostrum of extreme length. Supra-orbital spines very small ; margin of carapace smooth. Abdominal somite 3 with large dorsal spine ; somite 5 with lateral spines. Telson deeply indented, becoming forked, with three outer spines and ten terminal of which one forms the point of the fork. Anal spine absent, or present in late stages. Antenna! scale slightly segmented; endopod spine-like. Maxillule without outer seta. Exopod of maxillipede 3 with three terminal setae in stage I. Exopods on legs 1-3. Probably nine larval stages. Saron (and allied forms) Body stout, very highly coloured. Rostrum present or absent in stage I ; later stout, and broad at base. Carapace without supra-orbital spines; anterior margin denticulate; with posterior tooth. Abdominal somites without spines. Telson narrow in stages I and II, scarcely indented; later without lateral spines and with ten terminal spines. Anal spine absent. Antennal scale segmented ; endopod spine-like. Exopod of maxillipede 3 with four apical setae. Exopods on legs 1-4. Number of larval stages uncertain. Reduced to four in one species. 7-2 402 DISCOVERY REPORTS Latreutes (stage I only) Rostrum small. Carapace with strong pterygostomial and posterior teeth, without anterior marginal teeth. A strong tooth in middle of margin in L. fucorum. Abdominal somite 5 with large dorso-lateral spines; spines also on somite 4 in L. fucorum. Telson very narrow. Antennal scale with segmentation almost lost; endopod spine-like, but with a seta also. Maxillule without outer seta. Exopod of maxillipede 3 with four apical setae. (No later stages known with certainty.) LITERATURE Bate, C. S., 1888. Crustacea Macrura. Challenger Reports, Zool., xxiv. Brooks and Herrick, F. H., 1891. The embryology and meiamorphosis of the Macrura. Mem. Nat. Acad. Wash. V, pp. 321-576. Calman, W. T., 1906. Notes on some genera of the Crustacean family Hippolytidae . Ann. Mag. Nat. Hist. (7) xvn, pp. 29-34. Caroli, E., 1918. Miersia clavigera Chun, stadio misidiforme di Lysmata seticaudata. Pubb. Staz. zool. Napoli, II, pp. 177-89. Chun, C, 1888. Die pelagische Tienvelt in grosseren Meerestiefen. Bibl. Zool. Kassel, I. CouTiERE, H., 1905. Note priliminaire sur les Eucyphotes recueillis par S. A. S. le Prince de Monaco a I'aide du filet a grande ouverture. Bull. Mus. Oceanog. Monaco, No. 48. 1907- Sur quelques formes larvaires enigmatiques d'Eucyphotes, provenatit dcs collections de S. A. S. le Prince de Monaco. Bull. Mus. Oceanog. Monaco, No. 104. Eraser, F. C, 1936. On the development and distribution of the young stages of Krill (Euphausia superba). Discovery Reports, xiv, pp. 1-192. GURNEY, R., 1924. Decapod Larvae. Terra Nova Expedition Zoology, viii. No. 2, Crustacea, part ix. 1926. The protozoeal stage in Decapod development. Ann. Mag. Nat. Hist. (9) xviii, pp. 19-27. 1927. Larvae of the Crustacea Decapoda. Trans. Zool. Soc. Load, xxii, pp. 231-86. 1936. Notes on sotne Decapod Crustacea of Bermuda. L The larvae of Leptocheh and Latreutes. Proc. Zool. Soc. Lond. 1935, pp. 785-93. John, D. D., 1936. The southern species of the genus Euphausia. Discovery Reports, xiv, pp. 193-324. Kemp, S. W., 1914. Notes on Crustacea Decapoda in the Indian Museum. V. Hippolytidae. Rec. Ind. Mus. x, pp. 81-129. 1916. Notes on Crustacea in the Indian Museum. VII. Further tiotes on Hippolytidae. Rec. Ind. Mus. xii, pp. 385-405- Lebour, M. v., 1930. The larval stages o/Caridion, with a description of a new species, C. steveni. Proc. Zool. Soc. Lond. 1930, pp. 181-94. 1932. The larval stages of the Plymouth Caridea. III. The larval stages of Spirontocaris cranchii {Leach). Proc. Zool. Soc. Lond. 1932, pp. 13 1-7. 1936. Notes on the Plymouth species of Spirontocaris. Proc. Zool. Soc. Lond. 1936, pp. 89-104. Lo Bianco, S., 1901 . Le pesche pelagiche abissali eseguite dal " Maia" nelle vicinanze di Capri. Mitth. zool. Stat. Neapel, xv, pp. 413-82. 1909. Notizie biologiche riguardanti specialmente il periodo di maturitd sessuale degli animali del Golfo di Napoli. Mitth. zool. St. Neapel, xix, pp. 601-18. LITERATURE 403 Ortmann, a., 1893. Decapoden mid Schisopoden. Plankton Expedition, Bd. n, G. b. Sars, G. O., 1885. Norwegian North Atlantic Expedition. Zoology. Crustacea, I. SOLLAUD, E., 1923. Le diveloppement larvaire des Palaemoninae. Bull. Biol. France Belg. lvii, p. 510. Stephensen, K., 1935. The Godthaab Expedition 1928. Crustacea Decapoda. Medd. om Gronland, lxxx, no. I. WoLLEBAEK, A., 1906. Le diveloppement du genre Sclerocrangon G. O. Sars. Bergens Mus. Aarb. 1906, No. II. DISCOVERY REPORTS Issued by the Discovery Committee, Colonial Office, London on behalf of the Qovemment of the Dependencies of the Falkland Islands Vol. XIV, pp. i-vi TITLE-PAGE AND LIST OF CONTENTS CAMBRIDGE AT THE UNIVERSITY PRESS 1937 Price ninepence net Cambridge University Press Fetter Lane, London Nm York Bombay, Calcutta, Madras Toronto Macmillan Tokyo Maruzen Company, Ltd All rights reserved PRINTED IN GREAT BRITAIN BY WALTER LEWIS MA AT THE CAMBRIDGE UNIVERSITY PRESS DISCOVERY REPORTS Vol. XIV, pp. 1-192 Issued bji the Discovery Committee, Colonial Ofjice, London on behalf of the Qovernment of the Dependencies of the Falkland Islands ON THE DEVELOPMENT AND DISTRIBUTION OF THE YOUNG STAGES OF KRILL (EVPHAVSIA SUPERB A) F, C. Fraser, B.Sc. CAMBRIDGE AT THE UNIVERSITY PRESS 1936 Price twenty'Seven shillings net Cambridge University Press Fetter Lane, London New York Bombay, Calcutta, Madras Toronto Macmillan Tokyo Maruzen Company, Ltd All rights reserved PRINTED I N G REAT BRITAIN BY LEW IS MA AT THE CAMBRIDGE UNIVERSITY PRESS DISCOVERY REPORTS Vol. XIV, pp. 193-324 Issued by the Discovery Committee, Colonial Office, London on behalf of the Qovemment of the Dependencies of the Falkland Islands THE SOUTHERN SPECIES OF THE GENUS EUPHAUSIA D. Dilwyn John, M,Sc. CAMBRIDGE AT THE UNIVERSITY PRESS 1936 Price eighteen shillings net Cjmbridge University Press Fetter Lane, London A^** York Bombay^ Calcutta, Madras Toronto Macmtllan Tokyo Maruzen Company, Ltd All rights reserved PRI NTED I N GREAT BRITAIN By LEW IS MA AT THE CAMBRIDGE UNIVERSITY PRESS DISCOVERY REPORTS Vol XIV, pp. 325-350, plates I-V Issued by the Discovery Committee, Colonial Office, London on behalf of the Qovemment of the Dependencies of the Falkland Islands THE REPRODUCTIVE SYSTEM OF EVPHAVSIA SVPERBA by Helene E. Bargmann, Ph.D. CAMBRIDGE AT THE UNIVERSITY PRESS 1937 Price six shillings and sixpence net LONDON Cambridge University Press FETTER LANE NEW YORK • TORONTO BOMBAY • CALCUTTA • MADRAS Macmillan TOKYO Maruzen Company Ltd All rights reserved PRINTED I N G REAT BRITAIN BY LEWIS MA AT TH E CAMBRI DGE UNIVERSITY PRESS DISCOVERY REPORTS Vol. XIV, pp. 351-404 Issued by the Discovery Committee, Colonial Oifice, London on behalf of the Qovemment of the Dependencies of the Falkland Islands LARVAE OF DECAPOD CRUSTACEA ^7 PART IV. HIPPOLYTIDAE by Robert Gurney, D.Sc. CAMBRIDGE AT THE UNIVERSITY PRESS 1937 Price seven shillings and sixpence net LONDON Cambridge University Press FETTER LANE NEW YORK • TORONTO BOMBAY ■ CALCUTTA • MADRAS Macmillan TOKYO Maruzen Company Ltd All rights reserved LEWI S M A AT THE CAMBRIDGE UNIVERSITY PRESS ■ m/f .*?V"**#'*^^* ■:J^~^^,^M ^LJ'^-^: v> . -^ f^h*%' ■**^t'*' ■ 1 ■^^ . - ■ . Vi.>K :$^-/ 1, *V' .--^ 4 >■<: 7Vt^ ^* c VX' y^.../.