THE DANISH I X GOLF-EXPEL) III OX. VOL. Ill b. PUBLISHED AT THE COST OF THE GOVERNMENT THE DIRECTION OF THE ZOOLOGICAL MUSEUM OF THE I NIVERSITS ■ffi- COl'KXHAGKN. H. HAGERUP. PRINTED B-J BIANCO LUNOA/S 1916-1923. ts b v.5B ■ p1"- G-l Contents of Vol. Ill b. V. II. .1. Hansen: Crustacea Malacnsfraca, III: Isopoda, |>. 1 iii'>2 (16 plates), 1916. VI. H. J. Hansen: Crustacea Malacostraca, IV: C'umacea and Nebaliacea, \>. 1-86 (I plates), 1920. VII. H.J.Hansen: Crustacea Copepoda, II: Copepoda parasita ami hemiparasita, p. 1-92 (5 plates), 1923. ? THE DANISH INGOLF-EXPEDITION. VOLUME III. 5. CRUSTACEA MALACOSTRACA. III. BY H. J. HANSEN. WITH 16 PLATES AND A LIST OF THE STATIONS. ->4>ffi<$< \N COPENHAGEN. PRINTED BY BIANCO LUNO. 1916. Ready from the Press July the 5th 1916. CONTENTS. Crustacea Malacostraca. III. V. The Order hopoda. Page Introduction I On the Literature 2 Results and Questions 2 A. The Material 2 B. Classification and Morphology 3 C. Geographical and bathymetrical Distribution 4 Sub-Order Asellota 9 Group I. Ianirini 12 Iter a Leach 12 1. Itera albifrons Leach 12 lanira Leach 13 2. lanira maculosa Leach 14 3. — alta Stimpson 16 4. tricornis Kroyer 17 5. pukhra n. sp 19 6. — Vilhelmintz Stephensen 20 7. laciniata G. O. S a rs 20 8. — spinosa Harger 21 Acanthaspidea Stebbing 23 9. Acanthaspidea typhlops G. O. Sars 23 lanirella Bonnier 24 10. lanirella spongicola n. sp 25 11. — lie-vis n. sp 26 Katianira n. gen 26 12. Katianira chelifera u. sp 27 Group II. Haploniscini 2S Haploniscus Richardson 2S 13. Haploniscus bicuspis G. O. Sars 29 14. spinifer n. sp 31 15. arinadilloides n. sp 31 Hydroniscus n. gen 32 16. Hydroniscus abyssi n. sp 33 Group III. Munnini 33 Munna Kroyer 34 17. Munna Boeckii Kroyer 34 iS. — groenlandica n. sp 35 19. — Hanseni Stappers 36 20. — Kroyeri Goodsir 37 21. — Fabricii Kroyer 3^ 22. — minuta H. J. Hansen 39 23. — acanthifera n. sp 4° P.gc PUurogonium G. O. Sars • • 42 24. Pleurogonium 1 tier me G. O. Sars 43 25- 26. 27. — 28. — 29. latimanum n. sp 44 — intermedium n. sp 44 rubicundum G. O.Sars 45 spinosissimum G. O. Sars 46 pulehrum n. sp 46 Pseudomunna 11. gen 47 30. Pseudomunna hystrix n. sp 48 Dendrotion G. O. Sars 49 31. Dendrotion spinosum G. O. Sars 50 32. faradoxum n. sp 51 Group IV. Schistosomatini 53 Schistosoma n. gen 53 33. Schistosoma ramosum n. sp 54 Group V. Ischnoinesini 54 Ischnomesus Rich ardson 56 34. Ischnomesus profundus n. sp 56 35. — armatus n. sp 57 Haplomesus Richardson 59 36. Haplomesus quadrispinosus G. O. S a rs 59 37. angustus n. sp 61 3S. — insignis u. sp 63 39. tenuispinis n. sp 64 40. — modes/us n. sp 65 Heteromcsus Rich ardson 66 41. Heteromesus Jentatus n. sp 66 42. longirc in is a. Sp 68 43. — Sehmidtii II. sp 69 44. frigidus n. sp 71 Group VI. Pseudomesini 72 Pseudomesus n. gen 73 45. Pseudomesus brevicomis n. sp 74 Group VII. Macrostylini 74 Macrostylis G. O. S ars 75 46. Macrostylis spinifera G. O. Sars 76 47. 48. 49 5° 5i abyssicola n. sp 77 elongata n. sp 79 subinermis n. sp 80 longiremis M eincrt 81 longipcs n. sp 82 Page Group VIII. Nannoniscini S3 Nannoniscclla n. gen 84 52. Nannoniscella groenlandica n. sp S5 Nannoniscoidcs n. gen 86 53. Nannoniscoidcs angulatus n. sp 86 Nannoniscus G. O. S a rs 87 54. Nannoniscus simplex n. sp 91 55. oblongus G. O. S a r s 92 56. arcticus n. sp 94 57. ana/is 11. sp 95 58. laticeps 11. sp 96 59. rcticulatus n. sp 97 60. inermis n. sp 9S 61. cequiremis n. sp 99 62. plebejus 11. sp IOO 63. minutus n. sp 102 64. armatus n. sp 102 65. spinicornis 11. sp 104 66. affinis 11. sp 105 Group IX. Desinosoiiiatiui 105 Dcsmosoma G. O. S ars 106 67. Dcsmosoma globiceps Meinert 109 68. — latipcs n. sp HO 69. longispinum 11. sp 111 70. simile 11. sp 112 7 1 . gracilipes n. sp 113 72. politum n. sp 114 73. natator n. sp 115 74. laterale G. O. Sa rs 1 16 75. armatum G. O. Sars 118 76. — insigne 11. sp 118 77. plebcjum 11. sp 120 Group X. Ilyarachnini 120 Ilyarachna G. O. Sars 121 78. Uyarcu hna hirticeps G. O. S ars 122 79. bit amis 11. sp 125 80. dubia 11. sp 125 Si. /': rgendali Ohlin 126 82. — spinosissima 11. sp 127 Echinozone G. O. Sars 128 83. Echinozone coronata G. O. Sars 128 84. — arctica 11. sp 129 Group XI. Eurycopini 129 Syneurycope n. gen 130 85. Syneurycope parallcla 11. sp 131 Storthyngura VanhSffen. 132 86. Storthyngura magnispinis Richardson 132 Munnopsurus Richardson 134 87. Munnopsurus giganteus G. O. Sars 1 35 88. longipcs Tatters all 136 Eurycope G. O. Sars 137 89. Eurycope Murrayi Walker 137 Page 90. Eurycope nodifrons 11. sp 14° 91. — comuta G. O. Sars 141 92. iucrmis 11. sp 142 93. Hanscni Ohliu 144 94. — complanata Bonnier 145 95. — breviroslris n. sp 146 96. — producta G. O. Sars 147 97. mcgalura G. O. S ars 148 9S. — parva Bonnier 149 99. phallangium G. O. Sars 15° 100. — furcata G. O. Sars , 151 101. — mutica G. O. Sars 151 Group XII. Munnopsiui 152 Paramunnopsis n. gen 154 102. Paramunnopsis oceanica Tattersall 155 Mtmnopsis M. Sars 156 103. Munnopsis typica M. Sars 156 Munnopsoidcs Tattersall 158 104. Munnopsoides eximius n. sp 159 Pseudomunnopsis n. gen 160 105. Pseudomunnopsis Beddardi Tattersall 160 Sub-Order Flabellifera 162 Family CyinothoidEe 162 Sub-Family Cirolaniiue 164 Cirolana Leach 1 64 106. Cirolana borealis Lilljeborg 164 107. Sckmidtii H. J. H a n s e n 165 108. — Hanseni Bonnier 165 Eurydice Leach 165 109. Eurydice Grimaldii Dollf us 165 1 10. — caeca n. sp 166 Sub-Family -Egidre (uiisscript for ^gime) 167 sEga Leach 168 111. -cEga psora L 168 112. — tridens Leach 169 113. Stroemii Liitken 169 114. — cremdata Liitken 170 115. — monophthalma Johnston 171 116. — arctica L u t k e u 171 117. — ventrosa M. Sars 172 118. — gracilipes H. J. Hansen 173 Rocinela Leach 174 119. Rocinela danmofiiensis Leach 174 — maculata Sell. & Mein 174 Syscenus Harger 175 120. Syscenus in/clix Harger 175 Family Sphaeromidse 175 Limnoria Rathke 177 121. Limnoria lignorum Rathke 177 Bathycopca Tattersall 177 122. Bathycopca typhlops Tattersall 179 Family Anthuridre 180 r ( vat Intra Norman S: Stebbin g iSl 123. Cyathura truncata 11. sp 182 Anthelura Norman & St ebbing 1S2 124. Anthelura abyssorum Norm. & Stebb [N2 Calathura Norm. & Stebb 1S3 125. Calathura brachiata Stimpson I S3 Sub-Order Valvifera 185 Family Idotheidse 186 Mesidothea Richardson 1S6 126. Mesidothea Sabmi Kroyer 186 127. megalura G. O. Sars 1 S7 Idothea J. C. Fabricius 187 128. Idothea balthica Pallas 187 1 29. granulosa R a th k e [88 130. — emarginata J. C. Fabricius 189 131. — pelagica Leach 1 90 132. — metallica B os c 190 Synidothea H a r g e r 1 90 133. Synidothea nodulosa Kroycr 191 Family Areturida? 191 Arcturm Latreille 192 134. Arclurus Baffini Sabine 192 1 35. — scabrostts Norman 1 93 Pleuroprion zur Strassen 193 136. Pleuroprion hystrix G. O. Sars 194 137. Murdoch:' Benedict 195 138. frigidum n. sp 196 Astacilla Cordiner 197 139. Astacilla longicornis So werby 19S 140. .'arietina G. O. Sars 198 141. — intermedia Goodsir 199 142. — pusilta G. O. S a rs 199 143. — granulata G. O. Sars 200 Sub-Order Oniscoidea 200 Ligia J. C. Fabricius 201 144. Ligia oceanica Lin lie 201 Sub-Order Epicaridea 202 Familj Bopyridae 203 145. I hippolytei Kroyer 146. Pseudione Hyndmanni Bate & Westwood... 205 147. Pkryxus abdominalis Kroyer . 206 Family Dajidas.. 208 Dajus Kroyer 20S 14S. Dajus mysidis Kroyer 20S 149. — profundus n. sp .210 llolophry\it\ R i ch ardso 11 210 150. llolophryxus Richardt Kochler 210 151. acanthephyrce Stephensen 211 Aspidophryxus ('.. (). Sars 211 152. Aspidophryxus peltatu , ( ',. ( ). Sars 211 Family Cryploniscidaj 212 Clypconiscus Giard & Bonnier 213 153. Clypeoniscus Meinerti Giard & Bonnier .... 213 Areturoeheres n. gen 214 154. Areturoeheres pulchripes 11. sp. 214 Asia, illoechui n. ,l;iii 216 155. Astacilloechus Ingolfi n. sp 216 Parapodascon 11. gc-n 217 156. Parapodascon Stebbingii n. sp 217 ( 'umoechus 11. gen 219 157. Cumocchus msignis 11. sp 221 Sub-Order Gn.ithiidea 222 Gnathia Leach 222 158. Gnathia elongata Kroyer 224 159. — robusta G. O. Sa rs 225 160. — hirsuta G. O. S a rs 226 161. — abyssorum G. O. Sars 227 162. — albescens u. sp. 22S 163. — bicolor n. sp 229 164. — stygia G. O. Sars 23° Explanation of the Plates 233 Crustacea Malacostraca. III. By H. J. Hansen. V. The Order Isopoda. I ntrod ucti on. Before entering on the subject of the present paper I may refer to the "Introductory Remarks" in "Crustacea Malacostraca I" published in 1908, because they contain various statements that need not be repeated here. In that chapter I explained the limits of the area investigated by the "Ingolf" and other Danish expeditions to our northern dependencies; the principal sources (apart from the "Ingo for the material examined were enumerated, and the principles followed as to "occurrence" and "distribution" were laid down. Zoologists wishing to get some information on these and allied topics may find them in the paper mentioned. Our earlier knowledge of the Isopoda living at the coast of Greenland in depths down to nearly a hundred fathoms was rather good, but as to the fauna of the deeper tracts of the adjacent seas it was extremely poor. In 1913 K. Stepheusen enumerated 37 valid species of marine Isopoda, the deep-sea forms included, belonging to the fauna of Greenland. Of the Isopoda from Iceland and the Faeroes and the deep sea around these islands our knowledge was very poor; in various papers I have found a number of species also occurring at Greenland, and, besides, in all 9 species recorded which were unknown from the seas of the last-named country. The result is that according to the literature 46 valid species are known from the seas around our northern dependencies. In the present paper 164 species are enumerated, and all, excepting 3 species not seen by me, have been secured by Danish expeditions. The "Ingolf" has gathered 121 of these species, and no less than 74 have been taken only by that expedition; 61 of its species are established as new to science. The "Thor" (Dr. Joh. Schmidt) has secured no less than 46 species. 9 of which have been taken exclusively by this ship, and 4 among these are new. The Ryder-Expedition and the two Amdrup-Expeditions to I Greenland gathered several species, and 3 among them are new; many species, 2 among them new, are due to many other collectors in earlier or in more recent times. A comparison with the faunas of Norway and Great Britain may be of some interest, as the Malacostraca in the seas at these countries have been more thoroughly investigated than in any other area of our globe. In his "Account" Vol. II (1896— 1899) G. O. Sars described 84 valid species of marine Isopoda from Norway (the forms included in his work, but in reality belonging to the Danish fauna and not taken near Norway, I have omitted in this enumeration); of these 84 species, 33 (among them The Ingolf- Expedition. III. j. CRUSTACEA MALACOSTRACA. III. 14 Epicaridea) are unknown from the "In golf" area. In 1905 Tattersall enumerated 106 species belonging to the fauna of Great Britain and Ireland, but only 46 of these have been found in the "Ingolf' area, and the majority of the remaining 60 species belong to the so-called Lusitanian fauna, and will never be found at the Faeroes or Iceland. The numbers given may show that the seas around our northern dependencies have been proportionately very well explored, biit it is quite certain that numerous smaller forms living in the warm area in depths from 400 down to nearly 2000 fathoms have not yet been detected. On the Literature. The literature to be taken into consideration is very extensive, but only a little ma)- be said here. Only five works may be enumerated, as they were of special importance to the present task, and in my references their titles have been extremely abbreviated in order to save space. The main work is G. O. Sars: An Account of the Crustacea of Norway. Vol. II. I so pod a. 1899 (in reality 1896 — 1899). *-)n tne following pages it is quoted a hundred times, and, though some errors and several deficiencies are pointed out, the enormous value of this fine and highly instructive standard work must be emphasized. G. O. Sars: The Norwegian North-Atlantic Expedition. Crustacea, I. 1885, and II. 1886. — It contains elaborate descriptions of several forms, and a good deal about distribution. Harriet Richardson: A Monograph on the Isopods of Nor th America. 1905. (Bull. U. S. Nat. Museum, No. 54). The value of this book for the "Ingolf" paper was especially due to its enumerations of geographical distribution, and to its very complete and accurate lists of synonymy. The abbreviation used on the following pages is: Monograph. W. M. Tattersall: The Marine Fauna of the Coast of Ireland. Part V. Isopoda. 1905. (Fisheries, Ireland, Sci. Invest, 1904. II). This paper contains descriptions with figures of some interesting animals also captured by the "Ingolf" or the "Thor", and statements on the occurrence of numerous species. The abbreviation used by me is: Isopoda. K. Stephensen: Groulands Krebsdyr og Pycnogonider (Conspectus Crustaceo- rum et Pycnogonidor um Groenlandise), in "Meddelelser 0111 Gronland. XXII. 1913". This is a useful compilation of all earlier statements in the literature on the occurrence of the animals at Green- land, together with lists of synonymy and distribution outside Greenland. On the following pages it is quoted as: Conspectus. Results and Questions. A. The Material. A comparison of the Isopoda of the "Ingolf" area with the world's fauna of the same order may be of some interest. Caiman (1909) divided the Isopoda into six sub-orders; of these the Phreatoicidea have no representative, and of the rich sub-order Oniscoidea only a single species of Ligia is to be mentioned. But of the four remaining sub-orders some families, as Asellidse, Stenetriidte, Serolida;, Kntoniscidce and three small families of the sub-order Valvifera are to be discarded; of the big family Cymothoidae, comprising eight sub-families, only two sub-families, viz. Cirolaninse and yEghue occur in CRUSTACEA MALACOSTRACA. III. our area, and of the large family Sphceromidie only two genera, each with a single species, have been found. Most, and perhaps all, families or sub-families hitherto not met with within the "Ingolf area will never be found there, as they exclusively live in less cold areas, or belong to the sub-antarctic or colder temperate parts of the southern hemisphere. In the following pages no less than 70 new species are described. Of these 4 belong to the sub-order Flabellifera (sens. Caiman), 1 to the Valvifera and 4 to the Epicaridea, consequently 61 to the sub-order Asellota. Three of the new Epicaridea are rather interesting, but it can be said that, putting the sub-order Asellota aside, the material comprises only 59 species of the other four sub-orders together, and most of these animals are well-known forms; though, of course, our knowledge of the geographical and bathymetrical occurrence of the majority of these species is considerably expanded. But the material of the sub-order Asellota comprises 105 species, 61 of which are new. Tatter- sall had only 32 species of Asellota from Great Britain and Ireland, G. O. Sars 35 species from Norway; consequently the number of species from the "Ingolf" area is astonishing. The explanation is easily given. Most of the Asellota are, like the majority of the rich family Tanaidae, deep-sea animals, as most of them inhabit depths from 500 — 700 to 1870 fathoms. The majority of the deep-sea Isopoda hitherto known from all seas together in reality belong to the Asellota, and in spite of the rich gatherings ol the "Ingolf" it can safely be concluded that the warm area explored contains a good number of hitherto undiscovered Asellota. Two reasons may be given, viz. that one of the two richest places was Stat. 36, 1435 fathoms, where 17 species of Asellota (two pelagic species not included) were found in less than a liter of sifted bottom material, but the "Ingolf" has very few stations in the warm area with depths from 1000 to 1870 fathoms; furthermore each of no less than 38 of my new species of Asellota was found only at a single station. That the "Ingolf has gathered a material of deep-sea Asellota which, in proportion to that secured by other deep-sea explorations, is extremely rich, is mainly due to the method of collecting described in my report on the Tanaidacea, p. 3 (1913). B. Classification and Morphology. In his extremely valuable hand-book on the Crustacea (1909), Dr. W. T. Caiman wrote (p. - "The structure of the Isopoda is so diversified, and the number of forms included in the Order is so large, that their classification is a matter of some difficulty". This statement is true, but unfortunately a long time must pass before it will be possible to propose a realh satisfactory classification. Caiman adopted, with some modifications, that proposed by G. O. Sars, and I adopt Caiman's classification with a single additional alteration. But some remarks may be made here. As already stated, Caiman divided the order Isopoda into six sub-orders: Asellota, Phreatoicidea, Flabellifera, Valvifera, Oniscoidea and Epicaridea. The result is that these sub-orders stand as equivalent. but Caiman most correctly said that "they are of very unequal value", and pointed out some three instances. The Asellota stand in reality rather apart from the rest; not only does the structure of their pleopods isolate them, but they have preserved some primary features. The antennse have three joints in the sympod and, generally, a distinct exopod, the squama, which frequently is movable; but among other Isopoda an antennal praecoxa exists only in a few genera, and a squama is found only in Ligia (see later on). The coxs of the thoracic legs, except, generally, those of the first pair, are movable, and CRUSTACEA MAI.ACOSTRACA. III. besides, never developed as such plates as in Cymothoidse or Idotheidse, though sometimes they possess an epimeral process. Finally I have discovered the praecoxa of the maxillipeds well preserved in a couple of genera (see p. 9) but this joint has hitherto not been found in any member of the Malaco- straca, excepting Nebalia, while it is highly developed, f. inst., in Apas. The combination of these primitive features, together with the far from primary but very peculiar development of the two anterior pairs of pleopods, remove the Asellota rather far from the other Isopoda; probably features in inner anatomy or embryology may besides be found to show primary or aberrant characters. The quite small but very interesting sub-order Phreatoicidea cannot be discussed here. — The families hitherto referred to the Flabellifera are kept together only by the development of their abdominal limbs, but two of them, Gnathiidse and Anthuridse, differ much from the three other families: Cymothoidse (sens, lat), Serolidse and Sphseromidse, which constitute a natural group. In spite of the difficulties I have kept the Anthuridse in the sub-order, but found it necessary to remove the Gnathiidse, establishing them as a new sub-order, the Gnathiidea, which even is placed after the Epicaridea, because the family Bopyridse of the latter sub-order differs considerably less from the Cymothoinae among the Flabellifera than is the case with the family Gnathiidse, which in some features differs extremely from all other Isopoda, and, which is of special importance, the most aberrant features found in the structure, sexuel difference, and development of the Gnathiidse can scarcely be considered as modifications due to parasitism. The sub-order Valvifera is a most natural group, and the animals are easily separated from the Flabellifera in having the uropods developed as a folding-door covering the pleopods. But whether this single feature can be considered of a value sufficient for separating animals as the Idotheidse etc. from the Flabellifera is very doubtful and no other external character can be pointed out. Whether any internal character for the Valvifera can be found is questionable; if not, this sub-order may probably be- suppressed in a future ameliorated classification. The sub-order Oniscoidea cannot be discussed here; yet it may be pointed out that the family Tylidae differs considerably from the others. — The Epicaridea are a most natural sub-order, but the distance between it and some of the Flabellifera is, as already mentioned, not very great. It has been mentioned that I have now discovered the hitherto unknown praecoxa in maxilli- peds in two genera of Asellota. In two genera of the group Munnopsini I found an interesting fusion along the median Hue of the second pair of pleopods in the male. In Ligia I have found the preecoxa and the squama in the antennae; in a few types belonging to Cymothoidse and Valvifera three joints in the synipod of first pair of pleopods are pointed out. In Calathura brachiata Stimps. an unpaired, probably sensory, organ has been found in the telson; in species of the genus Macrostylis a pair of organs, probably sensory, were discovered in the abdomen. These morphological and anatomical features are only briefly enumerated; readers taking interest in them may find them described later on in the remarks on their sub-order, or in the descriptions of the genera or species in question. C. Geographical and bathymetrical Distribution. A restricted number of the forms are either decidedly, or most probably, pelagic, viz. Eurycopc Murrayi Walk., Paramunnopsis oceanica Tatt, Munnopsoides eximius a. sp., Pseudomunnopsis Beddardi CRUSTACEA MALACOSTRACA. III. Tatt, Eurydice Grimaldii Dollf. and /:. coeca 11. sp., Molina metallica Bosc, the larval stages oiSyscenus infelix Harg. and of Epicaridea. The species of the sub-family ^Eginse are either living on fishes or are found at the bottom, frequently in depths of several hundred fathoms, but probably they frequently swim about looking for a host, while their ovigerous females take no food and live only on the bottom. When speaking of bathymetrical distribution all these animals living either constantly pelagieally or during periods of their life as parasites may be put aside. Nearly all other Isopoda dealt with in the present paper are living during their whole life cither on or near the bottom. G. O. Sars has collected large numbers of specimens <>!' the genera Eurycope^ Ilyarachna and Munnopsis; he has described their methods of walking and swimming, but according to his statements one must conclude that they constantly live at or near the bottom. I do not know a single instance of any species of these genera, excepting Eur. Murrayi Walk., having been taken in any instrument not sunk to or very near to the bottom. It may be possible that some species of the genera Dcsmosoma G. O. Sars and Nannoniscus G. O. Sars sometimes may raise from the bottom and swim about, but judging from the fact that only a single specimen of these genera, viz. Desmosoma chelation Steph., has been recorded in the literature as taken pelagieally — and even at a station between Elba and Corsica, together with 3 species of Apseudes otherwise never before taken in this way _ this mode of living must be extremely exceptional, as both genera comprise numerous specie-. And many Isopoda, f. instance the species of the IscAnomesus-gTOUQ, are certainly quite unable to swim. A restricted number of Isopoda live either near the shore {lard) or in depths from very few t<> 50 or sometimes about a hundred fathoms {Ianira tricornis Kr., some species of Muinni, most sp. of Idothed). Ianira maculosa Leach is most remarkable, as it has been taken in most different depths, from a few down to more than a thousand fathoms, and the specimens inhabiting rather low water differ, as is shown later on, markedly from those captured in the deep sea. The great majority of the Isopoda from our area are bottom animals living in rather consider- able or great depths. The two richest places are the two "Ingolf" stations 78 and 36, both in the warm area. At Stat. 78, 799 fathoms, an enormous quantity of bottom material, especially sponges, was hauled up, and 22 species of Isopoda were found; at Stat. 36, 1435 fathoms, the bottom material did not fill a liter in a sifted condition, but it contained 18 bottom species, and two certainly baths - pelagic species were found in the meshes of the trawl. A number of bottom forms were also procured at two other of the deepest stations in the warm area, viz. at Stat. 24, 1199 fathoms, 15 species, and 8 species at Stat. 38, 1870 fathoms. With very few exceptions the forms taken at these four stations belong to the Asellota, and as to this sub-order we arrive at a similar result as that pointed out in my report on the "Ingolf" Tanaidacea on the rich family Tanaidae, that the richest harvest was yielded by some of the deepest stations south and west of southern Greenland. The only other rich pi are the above-named Stat. 78, 799 fathoms, far south-west of Iceland, and Stat. 25, 582 fathoms, in Davis Strait, where 16 species, 12 among them Asellota, were gathered. At the three stations i\ and 38, with depths from about 1200 to 1870 fathoms, together with Stat. 78, were taken in all 33 new- species of the Asellota. Of the total number of Isopoda only 17 species have been taken exclusively in temperatures below zero, and 15 of these in at least 293 fathoms, and generally in greater depths, thus exclusively CRUSTACEA MALACOSTRACA. III. iii the cold deep-sea area, while 2 species have besides been taken in lesser depths. As to the limit- ation of the warm deep-sea area, I proposed in 1913 that the stations with a depth between 300 and 600 fathoms and the temperature not below 30, and all stations with depths from 600 to 1870 fathoms and the temperature above zero, might be considered as belonging to the warm area. About 56 of the species (a few merely pelagical species not included) have been taken exclusively in this warm deep- sea area by our Danish expeditious, but of these 8 species are also known from some lesser depths in the Norwegian Fjords or Skager Rak. But a small number of species: Acanthaspidea typhlops G. O. Sars, Munna acanthi/era n. sp., Haplomesus quadrispinosus G. O. Sars and // angustus n. sp., Ilyarachna hirticeps G. O. Sars, Eurycopc inermis n. sp., Eur. brevirostris n. sp., E. producta G. O. Sars, Munnopsis typica M. Sars, Pleuroprion hystrix G. O. Sars, Astacilla granulata G. O. Sars and Calathura brachiata Stimps. have been taken both in the real cold deep-sea area and at rather deep or very deep stations in the warm area; three of these species, viz. Etirycope producta, Munnopsis typica and Calathura brachiata, have besides been taken in depths of less than 100 fathoms either near Iceland, Norway, East Greenland or in the Kara Sea, while a fourth species, Ilyarachna hirticcps, has been taken at 100 fathoms in Davis Strait and in 10 fathoms at Spitzbergen. Eurycopc inermis has been taken seven times in the warm area between 362 and 806 fathoms, and nine times in the cold area, in 293 to 1010 fathoms; Eur. brevirostris has been taken twice in the cold area and once in the typical warm area. As to the species mentioned of the genera Ilyarachna, Eurycopc and Munnopsis, the hypothesis that the animals sometimes live pelagically could at least to some degree explain their occurrence, and that the animals have three pairs of natatory legs, and can swim, is well known, but unfortunately no observation has ever been recorded that any species has been taken pelagically, and the three most common and widely distributed species, Ilyarachna hirticeps, Eurycopc inermis and especially Munnopsis typica, are so large, that they could scarcely escape or be overlooked. Munna acanthi/era has been captured nine times in the warm, and three times in the cold, area, and these stations are very far from the limits between the areas; Haplomesus quadrispinosus was taken by the "Ingolf" six times in the cold area in depths from 371 to 1309 fathoms, and three times at some of the deepest stations, 1199 to 1870 fathoms, in the warm area, and as to these two species even the possibility of swimming must be excluded. Pleuroprion hystrix G. O. Sars is known from stations decidedly belonging to both areas, but all the stations are not very far from the limits between the areas; as to several of the localities the same is the case with Astacilla granulata. In the report on the Tanaidacea I pointed out similar facts, viz. that 1 species of Apseudidae and 7 species of Tanaidse had been taken in both areas; the females of two of these species have no pleopods. In his Conspectus (1913), K. Stephensen attempted to give an explanation of this inter- esting distribution. He pronounced the submarine ridge between the Shetlands, the Fseroes, south- eastern Iceland, and between North-west Iceland and East Greenland to constitute an "absolute line of separation between two deep-sea faunas", so that no deep-sea form living in the warm area can be found in the cold area, and vice-versa; the only exception being some few Kchinodermata. Stephensen therefore assumed that the Tanaidacea in question had been taken pelagically when they bad been secured in the area, where they normally did not live. In order to support this hypothesis he put together the few records in the literature showing that animals of that order had been taken pelagi- CRUSTACEA MALACOSTRACA. III. cally, and referred besides to the gatherings of the "Thor" in the Mediterranean and the adjacent pa of the Atlantic. In his later report (1915) on these gatherings we find the following lines |p. 261: "H.J.Hansen mentions, among the Tanaidacea of the Ingolf Expedition ... certain species, the occur- rence of which is contrary to zoogeographical laws, inasmuch as the) appear to belong both to the boreal southern and northern arctic ocean deeps off the coast of Greenland". And later: "Dr. II. J. Hansen however, maintains that all Tanaidacea belong to the bottom fauna, despite the fact that the occurrence of the "doubtful" species is very easily explained if they are taken pelagically". Stephensen's expression: "contrary to zoogeographical laws" is a little too strong, as according to Th. Mortensen a few Echinodermata do not obey the "law"; the term "rule" would have been better, as rules generally are not without exceptions. He is quite right in saying that "the occurrei of the "doubtful" species is very easily explained if they are taken pelagically", but this explanation is only a hypothesis, and we must now examine the foundations for this assumption. All his statements on specimens of the order Tanaidacea shall be taken into account, while his utterances on Cumacea, on a couple of Gammaridea, etc. are discarded as being without the slightest value as to conclusi- on Tanaidacea and Isopoda. In his last-named paper Stephensen enumerated 16 specimens belonging to 3 species of Apsi as taken all at the same place in young-fish trawl with 25 m. wire out, while the depth of the sea ■ 600—620 m. The statement is valuable in showing that animals of the family Apseudidae in reality can swim about rather near the surface in a sea of considerable depth. Hut according to kind infor- mation from Dr. Joh. Schmidt the locality in question is situated between Corsica and Elba, about 18 sea-miles from the former, 16 sea-miles from the latter island, and only 5 — 6 sea-miles from the flat with the depth of 200 m. off Elba. The only other case recorded in the literature of a form of the Apseudidae having been taken pelagically is found in my report on the Isopoda etc. of the German Plankton-Expedition, where I said (p. 49) that a single very young specimen of Apseudes had been taken in the Gulf of Guinea at St. Thome. Now it must be emphasized that both females and males of the family have their pleopods, judging from their appearance, rather well developed for rowing, though not for vigorous movements, and the two above-named gatherings show that animals of the family have been taken, at least at night, swimming about at a proportionately moderate distance from two islands and not very far from a flat. Otherwise these two instances prove next to nothing. Dr. Joh. Schmidt has hundreds of times used by day, or at evening or night, the young-fish trawl in various depths without coming near the bottom; in this way he explored the seas at Iceland, at the Faeroes, and further south in the Atlantic to Morocco and the entire Mediterranean; he never caught a specimen of Apseudidse very far from any coast, and in reality only the single case mentioned, but animals of the genera Apseudes and Sphyrapus are so large that they must have been captured among the enormous quantities of Copepoda and other small Crustacea, ami are easily discovered in the samples. Furthermore a rather good number of species of Apseudes, taken in dredge or trawl at the bottom, are known from the seas explored by Dr. Joh. Schmidt. The explorations of Dr. Schmidt together with the fact that the German Plankton-Expedition did not take one single specimen of the whole order in its very numerous vertical hauls in most different depths show with sufficient clearness that animals of the family Apseudidae do not live pelagically in such a way CRUSTACEA MALACOSTRACA. III. that, f. hist, the distribution of Sphyrapiis anomalus can be explained. This species is known from so cold a locality as the Kara Sea, has been taken in large numbers in Lat. 690 24.6' N. off East Green- land in 9 — n fath., along the whole coast of Norway, at two places in the cold area north of Iceland, in large number in 582 fath., in the warm area in Davis Strait, and, besides, at numerous other places. Then the family Tanaidse. The whole literature contains two statements that animals of this rich family have been taken pelagically. The most interesting is that of Stappers (1911), who recorded two females of Pscudotanais forcipatus and one female of P. Lilljeborgii as taken in a vertical haul, 30—0 m., in Lat. 760 26' N., long. 60° 55' E. According to the map in his paper this place is only about 10 sea-miles from the coast of Novaya Zemlya, and the sea there, according to a map in the work on the "Vega" Expedition, is moderately shallow, the depth being between 100 and 180 m. As the females of the two small species of Pseudotanais have no pleopods, I venture to say that they cannot swim; they have most probably been clinging to floating seaweed or some other object. Thomas Scott obtained the small Leptognathia brcvircmis in a vertical haul in Loch Fyne, but whether females of this species, with their minute pleopods, are really able to swim is extremely doubtful, and the pelagic capture of the species in that Loch does not prove that it is able to swim or live pelagically for any time. And I think that nobody may suggest that the occurrence of this species of Leptognatliia at some of the deepest stations in the cold area far south of Jan Mayeu, at three of the deepest stations in the warm area in Davis Strait, in both cases a hundred sea-miles or far more from any coast and at one of these stations in each area in a considerable number of specimens, can be explained by the gathering in Loch Fyne, thus near the coast and in shallow water. In the report on the Tanaidacea I enumerated (p. 51 even 7 species of Tanaidae, among them Leptognathia breviremis, as "taken both in the real cold area and at rather deep or very deep stations in the warm area." The females of five of these species possess more or less developed pleopods, but with the single exception of A. breviremis in Loch Fyne they have never been taken in a vertical haul or a young-fish trawl. The females of two of the species have no pleopods, and it is of no importance for the distribution of the species of Tanaidse that their males have the pleopods very well developed, when these organs are feeble or quite absent in the females. The cases mentioned in my report on the Tanaidacea are in reality so numerous, the distances between the stations in the cold area so far from those in the warm area and vice versa, and, besides, the stations are frequently so extremely far from any coast or from shallow water, that the extremely few observations recorded in the literature on pelagic capture at moderate distance from, or near, the coast cannot be used to explain away the occurrence of the 8 species at the bottom in both areas. Finally the above-mentioned Isopoda found in both areas. I think it impossible to imagine that Calathura brachiata Stimps. and especially typical deep-sea animals as Munna acanthi/era 11. sp. or Haplomesus qiiadrispinosus G. O. Sars can swim only half an hour. Calathura brachiata lias been taken several times in the warm area down to 799 fathoms, three times in the cold area down to 371 fathoms, and, besides, in the Kara Sea. The two other forms are typical deep-sea animals, both taken three times in the one and several times in the other area; Munna acanthifera going down to 762 fathoms in the cold and to 1199 fathoms in the warm area, Haplomesus quadrispinosus to 1309 fathoms in the cold and to 1870 fathoms in the warm area. These Isopoda corroborate my results as to the CRUSTACEA MALACOSTRACA ill Tanaidacea; several species of each order live constantly in both areas, as is the case with a few of Echinodermata. The hypothesis on pelagic occurrence proposed by Stephensen .1- an explanation must be discarded as untenable. And the zoogeographical "law" that the submarine an absolute line of separation between two faunas must be reduced, recognized onlj as a generally \ good, most useful and interesting rule, but with a numbei ol exceptions. Sub-Order Asellota. In 1893 I proposed the theory that three joints in the sympod of all biramous appendages in the Crustacea ought to be considered the primary condition, and showed that this number of joints was still extant not only in various appendages in Phyllopoda and Copepoda, but besides in the antennae of the Mysidse and the Asellota, in the maxilluhe of nearly all and in maxillae of all orders of Malacostraca, in the thoracic limbs of the Leptostraca. But the 1 of the three joints in question, the prcrcoxa, has hitherto not beeii pointed out in the maxillipeds and thoracic legs af any species of Malacostraca, excepting the Nebaliickc. In examining closely the maxil- lipeds with the corresponding sternite of two large forms of Asellota, viz. lanira pulchra n. sp. and Munnopsurus gigantcus G. O. S., I discovered the hitherto missing praecoxa (PI. I, fig. (. e, ", and PL XII, fig. 5 a, pcx.) as a somewhat small but sharply limited and well chitinized plate between the sternite and the joint generally described as the first. In Asellus and in Munnopsis typica the praeo does not exist. But as it is found in the two genera lanira and Munnopsurus, so far distant within the very rich family Parasellida.-, it exists in all probability also in some other forms. In vain I have- looked for the praecoxa in representatives of the other sub-orders (excepting Phreatoicidea), but il stence in the Asellota mentioned is, in my opinion, a morphological feature of some importance. G. O. Sars has pointed out that in Asellus the basal part of the maxillipeds is in the female with marsupium produced in a lappet directed backwards, and on his figure of the maxilliped Munna Boeckii a much smaller lappet is seen. This peculiar structure, which is highly developed in the majority of Flabellifera, in the Bopyridse, etc., seems to be rare within the Asellota. A feature observed in several genera or groups of genera of the Asellota is partial or com; fusion of segments, and such fusion has, excepting in the abdomen, been overlooked in most cases by nearly all authors, who have figured articulations where only a feeble suture or even a transvi impression without suture exists. Some instances may lie mentioned here; in the description genera or groups they shall be treated more fully. Beddard figured his Munnopsis australis and Tattersall his Munnopsoides Beddardi as having the three posterior thoracic segments and the abdomen separated by three transverse articulate These do not exist; the three segments and abdomen are immovably fused, constituting a single piece, and the lines separating them are only impressions, and especially the two anterior of these impi sions are, besides, extremely curved, so that f. inst. the impression between the fourth and fifth segments has the part at the dorsal median line far in front of its lateral portions (PI. XIV, figs. 2 a and 3a). In Munnopsis typica we find a somewhat similar structure, as the three thoracic segments and the 2 The Ingolf-Expedition. 111. i- IO CRUSTACEA MALACOSTRACA. abdomen are immovably fused, but the lateral, very oblique lines between them are sutures, while on the median part of the dorsal surface only impressions are found. The aberrant group Ischnomesini (formerly the genus Ischnosoma) shows very interesting fusion of the segments. In all genera the head is immovably united with the first thoracic segment, and the very curved dorsal line between them is only an impression. While in the genus Ischnomesus the posterior thoracic segments and the abdomen are movable, with real articulations between them, we find in Heteromesus the two posterior segments and abdomen immovably fused, and limited only by dorsal impressions. And the genus Haplomesus goes still further, having the extremely long fifth segment fused with the next, so that almost two-thirds of the body is shaped nearly like a stiff stick, with thickenings and impressions on its posterior part. — In the Haplo// iseus-gvoup pronounced fusion of segments is also found. In 1905 my paper: On the Morphology and Classification of the Asellota-Group of Crustaceans .... (Proc. Zool. Soc. London, 1904, Vol. II) was published. Two main-points in that treatise must be taken up again. It is a well-known fact that the males of this sub-order have five pairs of pleopods, but the females only four pairs, and I pointed out that the three posterior pairs are homologous in the two sexes. Consequently it must be either the first or the second pair which are wanting in the females. As the first existing pair in the female form an undivided operculum in all genera of Asellota excepting Asellus (with Mancasellus Harg. and Coecidothea Pack.), and as in all these genera the first pair in the male have their sympods coalesced in the median line or, in Stenetrium, completely fused, I expressed the opinion that it might be concluded that the first pair were present, the second pair wanting in the female. But later I received from Dr. A. Vire three specimens of a most curious animal, Stenasellus Virei Dollf., in which the abdomen has its two anterior segments well developed, a feature not distinct in Asellus or any marine form of Asellota, but this peculiar structure made it possible to see that the first abdominal segment has in the female 110 pleopods, while such are found on second segment. Stenasellus is more allied to Asellus than to any marine genus of the sub-order. The posterior part of the body of Haphnuesus tenuispinis n. sp. seen from below (PI. V, fig. 4 f) shows clearly that the female operculum is not attached to the first abdominal segment. And it must now be considered as certain that the females of the Asellota have no pleopods <>n first abdominal segment, while their first pair, in most genera fused and constituting an operculum without even any vestige of a suture in the median line, are homologous with second pair in the male. It is seen that in 1904 I was led astray by the fact that in the males the first pair of pleopods are coalesced or fused in the median line (excepting in . Isellus and other fresh-water forms), while the two male pleopods of second pair are quite independent in all forms examined before 1905 by am author. But in Pseudomunnopsis Beddardi Tatt collected by the "Thor" I have now found pleopods of second pair in the male fused to such a degree that they constitute a large plate without any median suture, and only with a somewhat deep and moderately narrow posterior incision (PI. XIV, fig- 3^), and this plate is quite similar to a female operculum, excepting that it has the incision mentioned, in which the small copulatory organs are found. And in another form, Paramunnopsis oceanica Tatt., I found the same pair of male pleopods fused in about one-third of their length (PI. XIII, fig. 11 i), thus intermediate between Pseudomunnopsis and the great multitude of genera with the CRUSTACEA MALACOSTRACA III. n two pleopods of second pair independent to their sternitc. Tlu- Fusion of the second pair "I n pleopods in two genera is a very interesting morphological feature. As already mentioned, the fauna from the "Ingolf" area comprises 105 species, 61 of which new, and it was deemed necessary to establish 10 new genera. The sub-order is rich, and the general aspect of many of the genera extremely different. The necessity of dividing it into familii >ng been felt, and some attempts have been made. In his account, *',. 0. Sars divided the Norwegian forms of the sub-order (or tribe) into five families, while Vanhoffen in his work on the Isopoda (and Tanai- dacea) from the German South-Polar Expedition has eight families. In the above-named paper 11905) I critisized the classification of Sars, showing that four of his families are "far from distinct from 1 other"; based on a number of characters found in the pleopods I divided the sub-order into three families, Asellidse, Stenetriidse and Parasellidse, the two first-named quite small, the third comprising the great majority of forms, in reality uniting the four families of Sars. Especially is his family Desmo- somatida; badly defined, because it comprises forms like hchnomesus (Ischnosoma), Nannoniscus, Macrostylis and Desmosoma, which, according to many features, are very distant from each other. And such types as Schistosoma n. gen. and Pseudomesus n. gen. would be very difficult to place in his system. In vain I have attempted to find leading characters in the shape of any organ. As the best example may be taken the mandibles which in Ianira and allied genera have the molar process robust, subcylindrical with the end cut off, and passing through a number of types as Pleurogonium, Nanno- niscus, Macrostylis, Desmosoma, Ilyaracluio and Aspidonotus that process is gradually reduced, more slender, gradually conical and smaller, until it is very small in Aspidonotus and disappears in Mun- nopsis. But in Eurycope cornuta G. O. S., which by Sars is placed in the same family as the three last-named genera, the molar process is well developed, thick, and differing from that in Ianira only in being more obliquely cut off; in Munnopsurus giganteus G. O. S., which is closely related to Eury- cope and by Sars is referred to this genus, the molar process is only a quite low, broad and rounded protuberance. Every attempt to divide the very numerous genera — of course including those not found in our area — belonging to the four families of Sars into moderatelv well defined families will, in my opinion, be impossible. But as a kind of arrangement is very desirable I attempt here to subdivide the family Parasellidae H. J. H. into a somewhat larger number of smaller, but tolerably equivalent, groups. In this way it is possible to arrange genera showing somewhat close relationship into a kind of unit and point out its essential features, and the name "group" is much more neutral and somewhat less exacting as to sharp diagnoses than the name "subfamily" or "family". Of the three families into which I divided the Asellota, the Asellidse with its single spi Ascllus aquaticus L. known from Iceland and Greenland is omitted as being not marine. The Stene- triidse have no representative; all the forms dealt with belong to the family Paraselliche. 12 CRUSTACEA MALACOSTRACA. III. Group I. Ianirini. Body oblong, considerably depressed. Head free. Eyes frequently present, situated on the upper surface. Antennae with squama normally developed. Mandibles normal; molar process well developed, directed a little forwards, with the end cut off. Palp of the maxillipeds with second joint considerably or much expanded, and its two distal joints well developed. Thoracic segments freely movable. First pair of legs frequently a more or less prehensile organ; the other pairs of walking legs of moderate length, not increasing in length from second or third to seventh pair; accessory claw always discernible, frequently robust. Uropods always on the lateral or the terminal margin, frequently well developed, biramous, sometimes uniramous and, besides, much reduced. Remarks. The group comprises Icera Leach, Ianira Leach, Acanthaspidea Stebb., lanirclla Bonn., Katianira n. gen. and, besides, several genera not represented in our fauna, such as Ianiropsis G. O. S., Iccropsis Koehler, Carpias Richardson, etc. The genera differ somewhat from each other in the shape of the molar process, and considerably in the development of the first pair of legs, the length and thickness of seventh joint with claw and accessory claw in all legs, the shape of the female operculum and the development of the uropods. Ianira may be considered the typical and central genus of the group in all characters, while Tanirella to some degree, and especially in the abdomen, constitutes a transition to Pleurogonium of the group Munnini, and Katianira has a quite peculiar development of the first pair of legs. But in spite of differences the genera are on the whole allied, showing gradual differences in the features enumerated from genus to genus. laera Leach. Only a single species is known from our area. i. Iaera albifrons Leach. 1780. Oniscus marinus O. Fabricius, Fauna Groenl. p. 252 (not O. mar inns L.) 1814. Jcera albifrons Leach, Edinb. Encycl. VII, p. 434. 1838. /era nivalis Kroyer, Kgl. D. Vid. vSelsk. naturv. math. Afhandl. VII, p. 303; PI. IV, fig. 21. ! 1897. Jcera marina G. 0. Sars, Account, II. p. 104; PI. 43. 1905. — Richardson, Monograph, p. 450, figs. 503 — 504. ! 1910. Teem marina H. J. Hansen, Vid. Medd. Naturh. For. Kjobenhavn for 1909, p. 208, figs. 1—8. G. O. Sars and later authors have discarded the well-known name albifrons and applied marina O. Fabr. Fabricius undoubtedly described the present species, but referred it to Oniscus marinus L. It is impossible to interpret Linne's description, but Miers (1883) referred a collective species of Idothea to O. marinus L., saying that a specimen of Idothea preserved in Linne's collection in the Linnean Soc. London bears the name "marinus" "in what is undoubtedly Limueus's handwriting". For these reasons, and as the name /. albifrons is universally known, I prefer now to apply this name. In 1910 I pointed out that this species varies considerably in outline at Denmark. The spe- CRtSTACKA MALACOSTRACA. III. cimens from our northern dependencies all belong to the typical form shown in figs, i and 3 in m\ paper cpioted and, besides, drawn by Sars. Occurrence. Taken by the "Ingolf" at four places. West Greenland: Sukkertoppen, Lat 65°25' N., among alga.' at the beach; 9 spec. Godthaab, Lat. 64°n' N.; 1 spec. North-West Iceland: Isafjord, shallow water; main spec. Faeroes: Trangisvaag; 2 spec. This species has been taken at West Greenland from Lat. 69 '14' X. to Lat. 6i°5o,' N. at God- havn, Claushavn, Holstensborg, Sukkertoppeu, Godthaab, Fiskenaes and Frederikshaab. As to the last locality O. Fabricius wrote: "habitat sine numero ad littora inter fucos lapidesque," and it has been recorded as going down to 15 fathoms (H. J. Hansen). At Iceland it has been taken at the western coast far north in Adelvik (by Mag. W. Lundbeck) and far south at Reykjavik (various collectors), on the northern side at Akureyri in 0fjord (by Dr. A. C. Johanseu), on the east side in Bern Fjord (by Dr. A. C. Johanseu), and at Djupivogr (by cand. mag. H. Jonsson), finally south of Iceland at Vestman-0erne (by cand. mag. Ssemundsen); it was always taken in shallow water or at the beach. At the Faeroes it has also been taken at Thorshavn on the pier. Distribution. /. albifrons is common at Denmark (Meinert), and occurs everywhere in the Baltic, even to the end of the Gulf of Bothnia (test. Apstein). It occurs along the whole coast oi Norway, even to Vadso (G. O. Sars), and is stated to go further eastwards along the northern coast of Europe to Long. 55° E. (Apstein). It is distributed at Great Britain and Ireland, at the northern coasts of Holland, Belgium and France (various authors), and at the coast of la Vendee (Bonnier), but whether it occurs more southwards along the western coasts of Europe seems to be unknown; its existence at Naples in Italy (Carus) seems to me to be extremely doubtful. Finally taken at the east coast of North America from about Lat. 410 N. to 450 N. and at Labrador (Harger). lanira Leach. This genus has not been well understood by most authors. Henopomus Kr. is merely a syn- onym. Iolella Richardson with its synonyms lanthe Bovallius and Tole — a misprint for Iole - be separated from lanira in any natural way; Iolanthe (Tole) libbyi Oxtm. is only a synonym for lanira (Henopomus) tricornis Kr., and the other species referred to Iolanthe differ from the typical forms <.i lanira only in the degree of the development of the lateral expansions of the thoracic segments and abdomen. Furthermore Rhacura Rich, and Icerella Rich, may perhaps he cancelled as genera, and their forms referred to lanira. as I have been unable to find any generic difference worth mentioning between lanira and those two genera in the descriptions and figures published by Harriet Richardson: I am unable to ascribe generic value to characters derived only from the number and size of lateral pro- cesses and dorsal tubercles, when no other character can be found in the descriptions. lanira, as interpreted here, is a most natural genus, well distinguished from other allied genera by a set of characters. The molar process of the mandibles is strong, subcylindrical or a little broader 14 CRUSTACEA MALACOSTRACA. III. towards the end (PI. I, fig. 4 b), directed inwards and distinctly forwards, and with the end cut off transversely. The maxillipeds (PI. I, fig. 4 c) have the second joint of the "palp" about as broad as the lobe from the second joint — the praecoxa, a, not counted — ; the epipod is much longer than broad. — First pair of thoracic legs (PI. I, figs. 4 f — 4 g) somewhat shorter than the second, similar in both sexes and built as a kind of prehensile organ, as the fifth joint is considerably thickened, with a number of spines arranged along both sides near the lower margin, while the sixth joint is about two-thirds as long as the fifth, with a very close row of extremely small spines along the prehensile margin. All thoracic legs with the seventh joint short, the claw strong and moderately short, the accessory claw well developed, strong. — In the males the median lamella of the operculum is moderately broad at the base, nar- rowing considerably towards the middle and widening a little or moderately towards the posterior end. Female operculum not produced backwards, with the median part of the hind margin nearly transverse, or even a little concave. — Uropods from somewhat shorter to somewhat longer than the abdomen; peduncle and rami well developed, spiny. By these characters combined the genus is sharply defined from laniropsis G. O. S., Acantka- spidea Stebb., Janirella Bonn, and other genera. The most interesting characters for species and for sections of species are found in the development of epimeral processes or plates from the basal joint of the thoracic legs; such epimeral processes or plates are visible from above, and take part in shaping the outline of the thorax. It may be useful to give a tabular view of these characters in the seven species found in the "Ingolf" area. A. Epimeral plates developed at all thoracic segments. The plates are small, never produced into long, acute processes, but bifid at two or three of the segments. 1. /. maculosa Leach. 2. /. alta Stimp. 3. /. Iricornis Kr. B. Epimeral plates completely wanting 4. I. pulchra u. sp. C. Epimeral plates developed at the three posterior segments, but wanting at least at second and third segments. a. At first segment a long, narrow epimeral process; between the two lamellar processes of fourth segment a small, narrow epimeral process 5. /. laciniata G. O. S. b. No epimeral process at first segment; sometimes a minute epimeral tubercle between the two lamellar processes of fourth segment 6. /. VilhelmincB Steph. 7. /. spinosa Harg. 2. Ianira maculosa Keach. (PI. I, figs. 1 a— 1 f.) 1814. Ianira maculosa Eeach, Edinb. Encycl. VII, p. 434. ?i84&. Henopomus muticus Kroyer, in Gaimard, Voy. en Scand. Crust.; PI. 30, figs. 1 a— 1 u. 1847. Kroyer, Nat. Tidsskr., Ny Raekke, Vol. II, p. 366. ! 1897. Ianira maculosa G. O. Sars, Account, II, 11.99; PI. 40. This species, which at the Faeroes has been taken in 6 fathoms, and goes down not only to considerable depths but even to about 1100 fathoms (see later on), shows considerable variation according to the depth where the specimens lived. Specimens taken in depths to nearly 100 fathoms are mottled, CRUSTACEA MALACOSTRACA. 111. with reddish-brown dots, while in specimens from considerable depths such dots have disappeared. But much more important differences are found in the shape of the head and the relative size of the eyes. In specimens taken in comparatively shallow water, down to 70—100 fathoms, the eyes (fig. are proportionately very large, with black pigment, and .situated close to the lateral margins <>i the head; the head is not expanded laterally, and its surface is scarcely or slightly excavated along the anterior or the lateral margins. Fig-. 1 d exhibits the head of a specimen from my deepest station enumerated below; it is seen that the yellowish-brown eyes an- proportionately very much smaller than in the preceding specimen (fig. 1 a), and much removed from the lateral margins; the lateral parts of the head are much expanded and their surface is considerably excavated, so that the margin turns much upwards. Furthermore the surface is excavated trausvervely along the front margin, and espe- cially at its middle portion, which is turned much upwards. In all these particulars the two specimens from the deepest station, 1010 — 1140 fathoms, exceed all other specimens seen by me. Fvery possible transition between the two heads described and figured is found among my material. Fig. ic exhibits as to size of the eyes, their distance from the lateral margins, and the degree of dorsal excavation a specimen intermediate between those shown in fig. 1 a and fig. 1 d, and that specimen was taken in 295 fathoms. — Furthermore the surfaces of the three posterior thoracic segments have the elevations higher and more pronounced in specimens from deep water than in those from shallow water. Finally the shape of the abdomen varies a little, as is seen from a comparison of fig. 1 b, showing the abdomen of the specimen whose head is exhibited in fig. 1 a, with fig. 1 e, which belongs to the same specimen as fig. id; in fig. ie the postero-lateral parts of the abdomen are produced a little more backwards, art- more protruding, and less evenly rounded than in fig. 1 b, of the specimen with the eyes very large. Sars said that the female is 7111m., the male 10 mm. long, but such large males I have not seen. The ovigerous female shown in figs, ia — ib is only 3-9111111.; the female with marsupium from 295 fathoms (fig. 1 c) is 74 mm.; the female from the deepest station (figs, id — 1 e) has no marsupium and is 6-5 mm. long, while the male from the same station is 7 mm. My largest ovigerous female, from off Seydis Fjord, 135 fath., is 81 mm.; as to eyes and shape of the head this specimen is inter- mediate between the animals from 295 and from about rioo fathoms, which shows that the reduction in the size of the eyes, their distance from the lateral margins, and the excavations are not always quite proportional to the depths of their localities. Fig. 1 f shows the posterior part of the male first pleopods, and their posterior margin differs materially from that in Ianira alta (fig. 2 d). Occurrence. Taken by the "Ingolf" at thirteen places. West Greenland: Stat. 27: Lat. 64°54' X., Long. 55°io' W., 393 fath., temp. 3-8°; 1 spec. — — Mouth of Ameralik Fjord, Lat. 64°03' N., 5—70 fath., shells; 5 spec. Denmark Strait, off Angmagsalik: Stat. 94: Lat. 64°56' N., Long. 36°i9' W., 204 fath., temp. : 1 spec. Between Angmagsalik and West Iceland: Stat. 95: Lat. 650 14' X., Long. 7,0° yj \Y., 752 fath., temp. 2-i°; 4 spec. _ - - — Stat. 96: Lat. 65°24' X., Long. 29°oo' W., 735 fath.. temp. i-2°; 1 spec. 16 CRUSTACEA MALACOSTRACA. III. West Iceland: Dyre Fjord, 20 fath., mud; 7 spec. South-West of Iceland: Stat. 81: Lat. 6i°44' N., Long. 27°oo' W., 485 fath., temp. 6-i°; 5 spec. Stat. 78: Lat. 6o°37' N., Long. 27°52' W., 799 fath., temp. 4-5°; 3 spec. South of Iceland: Stat. 6: Lat. 63°43' N., Long. i4°34' W., 90 fath., temp. 7-0°; 1 spec. West of the Faeroes: Stat. 44: Lat. 6x°42' N., Long. 9°36' W., 545 fath., temp. 4-8°; 2 spec. North of the Faeroes: Stat. 143: Lat. 62°58' N., Long. 7°09' W., 388 fath., temp, h- 0-4; 2 spec. The Faeroes: Trangisvaag, between algae; 1 spec. It was known from two places off West Grenlaud, viz. Lat. 72°32' N., Long. 58°5' (not 51') W., 116 fath., stony bottom, and Lat. 66°32' N., Long. 55°34' W., 100 fath., stones with Hydroids and Balaui (H. J. Hansen). It has been taken by the l'Thor" south of Iceland at four stations: Lat. 63°i5' N., Long. 22°23' W., 116— 172 fath.; Lat. 63°i8' N., Long. 2i°3i' W., 89 fath.; Lat. 63°5' N., Long. 20°7' W., 290 fath., and Lat. 62°n' N., Long. i9°36' W., 1010 — 1140 fins., 2 spec. It is common at the Faeroes, where it has been taken about eleven times in depths from 6 fath. to 180 fath. (various zoologists). East of Iceland it has been taken (by Admiral Wandel) off Seydis Fjord, 135 fath., black ooze, 2 large spec, and 9 miles off the coast, 38 fath. (Mag. R. Horring). Finally it is known from East Greenland: Lat. 69°25' N., Long. 2O0i' W., 167 fath., large stones and clay; 1 spec. (Ryder Exp.). Distribution. At Denmark this species has been taken in the northern parts of 0resund, 15 — 22 fath. (W. Bjorck) and Storebelt, at various places in Kattegat, 7—15 fath., and in Skager Rak, 100 fath. (H. J. Hansen). It occurs along the whole coast of Norway, generally in 30 to 100 fath. (G. O. Sars); in the Barents Sea, 132 fath. (Max Weber); it is common in the North Sea (Zirwas; Copenhagen Museum), at Great Britain and Ireland, and is taken west of Ireland in depths from 26 to 388 fath. (Tattersall). Finally it is known from the northern coast of France and the south coast of Brittany (Bonnier). 3. Ianira alta Stimpson. (PI. I, figs. 2 a — 2 d). 1853. Asellodes alia Stimpson, Smiths. Contrib. to Knowl. VI, p. 41 ; PI. Ill, fig. 30. ! 1880. /antra Harger, Rep. U. S. Comm. Fish and Fisheries, Pt. 6, p. 321; PI. II and III, figs. 9, 12 and 13. 1905. Richardson, Monograph, p. 475, figs. 531—532. Description. Intermediate between /. maculosa Leach, and /. Incur 11 is Kr. - The front margin of the head (fig. 2 ai conspicuously concave, with a somewhat short or moderately long median process, which is a little or considerably longer than broad; the antero-lateral part of the head is a little produced, with the corner subacute or obtuse, and the angle between the lateral and the anterior margin measuring distinctly less than 900. The eyes are small, and far removed from the lateral margins, as the lateral part of the head is considerably expanded. Thoracic segments at the sides somewhat feebly expanded; the lateral lobes of second to fourth segments with a broad, rather shallow incision dividing the lobe into an anterior, small, distally rounded lappet about as long as broad, and a posterior, broad, obliquely triangular portion (fig. 2 a); the lobe CRUSTACEA MALACOSTRACA in of first segment and of the three posterior segments is undivided and rounded. All segments ha a somewhat small epimeral protuberance very visible from above; the epimeral protub of first segment is longer than broad, those of second to fourth segments are bipartite, with their posterior lappet from a little to much shorter than their anterior lappet, and when the posterior lappet is small, it is covered by the plate of the segment. (In one of my two specimens, a female, the posterior lappet of these epimeral protuberances is quite small and invisible from above; in the other specimen, a male, the posterior lappet is narrower but only a little shorter than the anterior, and its tip is visible from above). The lateral plates of the three posterior segments are rounded and short; the epimeral protuber- ances of these segments are small and rounded. The posterior half of each lateral margin of the abdomen (figs. 2 b and 2 c) is adorned with about eight obliquely triangular saw-teeth, and the margin in each interval between two teeth is bent at the insertion of a strong seta. The median, flatly convex portion of the hind margin protrudes slightly more backwards than the convex postero-lateral part of the abdomen. The uropods (in the male) about as long as the abdomen. Length of the male 7 mm., of the female without inarsupium 67 mm. Remarks. Harger's figure of I. alia from above is on the whole good. The species is closely allied to /. maculosa, but the last-named species is separated from /. ulta in having no median frontal process, the antero-lateral angles of the head not produced forwards, rounded, and measuring more than 900, finally the two incisions on the hind margin of the median lamella of the male abdominal oper- culum are much deeper and the rounded minute protuberance in each incision considerably smaller in / maculosa (fig. 1 f) than in I. alia (fig. 2 d). Occurrence. Taken by the "Ingolf" at a single station. West of Iceland: Stat. 96: Lat. 65°24' N., Long. 29°oo' W., 735 fath., temp. V2°; 2 spec. Distribution. Off the east coast of North America at a number of places between Lat. 38 X. and 440 N., several of the localities not far from the U. States, others in the Bay of Fundy, and, besi it has been taken far south of Nova Scotia and east of Sable Island; depths 35 to 487 fathoms (Richardson). 4. Ianira tricornis Kroyer. (PI. I, figs. 3a-3b). ?i846. Hcuopomus tricornis Kroyer, in Gaimard, Yoy. en Scand., Crust., PI. 30, figs. 2 a— 11847. — — Kroyer, Nat. Tidsskr. Ny Rsekke, Vol. II, p. 372. 1901. Tole libbeyi Ortmaun, Proc. Acad. Nat. Sc. Philadelphia, 1901, p. 157, with fig. (ToU a misprint for Iol ! 1913. fan the libbeyi Stephensen, Meddelelser 0111 Gronland, Vol. 51, p. 70; PI. 3. Description. Stephensen's figures in the paper quoted convey a fairly accurate idea of this species. - The median frontal process is somewhat long; each antero-lateral angle is produced into a well developed process as long as, or a little longer than, broad, and terminating in a spine articulated to its end; each half of the anterior margin between the median and the lateral process is conspicuously convex. Eyes large and close to the lateral margins. 3 The Ingolf-Expedition. III. >■ 1 8 CRUSTACEA MALACOSTRACA. III. Thoracic segments on the whole somewhat similar to those in /. a//tr, yet presenting some sharp differences. First segment has its lateral lobe much narrower than in /. alta, obtuse, as broad as, or a little broader than, the epimeral process, which is longer than the lobe, a little curved, acute. Second and third segments have their lateral plates divided into two lappets by a broad, moderately deep incision with two angles; each lappet is somewhat small and distally rounded, and in each incision are seen two epimeral protuberances, the anterior small and rounded, the posterior much smaller. Fourth segment differs from the third in having its anterior lateral lappet distinctly larger, while the posterior is wanting; the epimeral plate has a well developed anterior protuberance, while the posterior is small or wanting. The lateral plates of the three posterior segments are small and rounded, excepting those of seventh segment, which are produced behind into an acute triangle; the epimeral protuberances at these segments are small but easily seen from above. The abdomen (fig. 3 a) has each postero-lateral part broadly rounded or subtriangular and pro- duced slightly beyond the flatly convex part of the hind margin between the uropods. The posterior half of each lateral margin (fig. 3 b) is adorned in a most peculiar way, showing some five or six broad and moderately low incisions with a small protuberance terminating in a more or less strong seta at the middle of each incision; each lappet between two incisions is broad or very broad and its rather short lateral margins concave, with the result that the lappet is narrower at the middle than at its nearly straight distal margin. — The uropods about as long as the abdomen. Length of the largest specimen, an ovigerous female (from Holstensborg) 8-6 mm. Remarks. /. tricornis is easily distinguished from the two preceding species by the antero- lateral processes of the head, each terminating in an articulated spine, by several points in the shape of the thoracic lateral lobes and the epimeral protuberances, and by the curious adornment of the lateral margins of the abdomen. — Kroyer's detailed description is good, but he had evidently examined a female which had the uropods anomalously small, while as a rule the uropods have the same length in both sexes. I have found very few specimens of various species of the genus Ianira with either both uropods small, or one uropod much smaller than the other. Though Orttnann's figure of his Tole (Iole) libbeyi is very poor, Stephensen's interpretation is certainly correct, but the latter author overlooked the fact that the animal belongs to the Kroyerian species. Occurrence. Taken by the "Ingolf" at three places. Davis Strait: Stat. 29: Lat. 65°34' N., Long. 54°3i' W., 68 fath., temp. 0-2°; 3 spec. Stat. 34: Lat. 65°i7' N., Long. 54°: 7' W., 55 fath.; 1 spec. Mouth of Ameralik Fjord, Lat. 64°c>3' N., 5 — 70 fath., shells; 5 spec. It lias been taken many times at West Greenland from Lat. 780 N. to Lat. 6o°4o' N., and several of the localities have been recorded in the literature (II. J. Hansen, Ortmann, Stepheusen). The localities may be briefly enumerated: Cape Alexander, 27 fath.; Disko Bay; Hunde-Eiland; Egedesminde, 25, 30 — 40 and 411 fath.; Northern Slromfjord, 7 — 20 fath.; Holstensborg, 40 fath.; Sukkertoppen, 30 fath.; Lat. 65°n' N., Long. 53°33' W., 50 fath.; Kangerdluarsuk, 5—15 fath., finally not far from Julianehaab, 22 fath. (many collectors). At East Greenland it has been taken by the 1st Amdrup Exped. at Tasiusak, Lat. 65°37' N., CRUSTACEA MALACOSTRACA. III. 15 — 20 fath. and 20—30 fath., among algae. Finally the Ilnd Amdrup Exped. secured it at Jan Mayen, 15 fath. and 50 — 60 fath. (Already recorded from Jan Mayen by Koelbel). Distribution. Taken south of Spitsbergen, 70 fath.. temp. 0*4 (G rs); at l.at. 7541/ X., Long. 24°25' E., 42 fath., temp. ^1-42°, and Lat. 78°5o' N., Long. 27°39' E, 10 fath., temp. 0-2° (Ohlin). The statement of ('.. < >. Sars in the 2nd Fram Exp. is too uncertain. 5. Ianira pulchra u. sp. (PI. I, figs. 4a-4h). Description. Body broad, only about twice as long as broad; the major part of its surface more or less conspicuously set with short, stiff hairs. - The head has the frontal margin considerably excavated (fig. 4 a), and at its middle a rather long rostral process much longer than broad, and with the end subacute or obtuse; no antero-lateral processes are found and the angles are about 90° Lateral parts of the head strongly expanded, so that the small, black eyes are very remote from the lateral margins. Thoracic segments without dorsal processes, but their lateral parts are strongly expanded out- wards; each lateral plate is cut off transversely, with the angles distinctly, or much, rounded; the plates of second to fourth segment are, besides, deeply bifid. When the animals are seen from above, no epimeral plate or process is observed; at most the first joint of the legs is just perceived at the bottom of the narrow intervals between the lateral plates of the three posterior segments. Abdomen about half as broad again as long; its lateral margins are nearly straight and very converging backwards; the postero-lateral part outside each uropod is somewhat produced backward-, forming a triangle about twice as broad as long. - The median lamella of the male operculum has its terminal part peculiarly shaped (fig. 4 h), as the inner half of each pleopod is produced in a rounded lobe about as long as broad and directed backwards, while the outer half is a subtriangular lobe directed mainly outwards, witli the distinctly concave hind margin nearly transverse, and this outer lobe reaches slightlv beyond the base of the inner lobe. - Uropods decidedly shorter than the abdomen. Length of a large female 9 mm. Remarks. This fine species is abundantly distinguished by several conspicuous characl from all other forms. The antennulas may be seen on fig. 4 a. Figs. 4b— 4 e represent the mouth-]' of the left side; these four figures may serve as types for the organs in question of the gen their morphological composition is easily understood by aid of the explanation of the plate. It need only be pointed out that the maxilliped (fig. 4e) possesses the pracoxa [a) described on p. 9. Occurrence. Taken by the "Ingolf" at four stations. Davis Strait: Stat. 25: Lat. 63°3o' N., Long. 54°25' W., 582 fath., temp. 3-3°, numerous spec. Between Angmagsalik and North-West Iceland: Stat. 95: Lat. 65°i4' N., Long. 30°39' W., 752 fath., teinp. 210; 15 spec. _ _ — - Stat. 96: Lat. 65°24'N. .Long. 29°oo'W., 735 fath., temp. i-2°; 5 spec. South-West of Iceland: Stat. 78: Lat. 6o°37' N., Long. 27°52' W., 799 fath., temp. 45 I * spec 2o CRUSTACEA MALACOSTRACA. III. 6. Ianira Vilhelminae Stephensen. 1913. Janira Vilhelmina K. Stephensen, Meddelelser om Gronland, Vol. 51, p. 68; Pis. I — II. This large and robust species is very characteristic. Stephensen's figure of the entire animal renders the species easily recognisable, but as he did not publish any real description, the more im- portant specific characters may be pointed out here. Description. Surface of the body without granulations or hairs. - The head has the rostral process long, while the sides are somewhat expanded and at each side produced in a broad, moderately long, triangular plate directed forwards and a little outwards, nearly as long as broad and with the end obtuse or subacute; the frontal margin between this plate and the rostrum convex. Eyes small and very distant from the lateral margins. The lateral part of the thoracic segments rather expanded and produced into lamellar lappets; second to fourth segments with two lappets at each side, and the four remaining segments with a single lappet; all lappets are subtriangular, with the end obtuse, and most of them nearly as long as broad, but those on the two posterior segments are a little longer than broad, distinctly longer than the preceding lappets and directed considerably backwards. Epimeral processes are completely wanting at the four anterior segments, while at the three posterior segments they are seen from above as small, obtuse or acute protuberances situated behind the base of each lappet. The postero-lateral parts of the abdomen produced at each side of the uropods into a large triangular plate about as long as broad. Remarks. The marsupium of one of the females is occupied by a species of Sphiclla. Occurrence. Not taken by the "Ingolf". Hitherto known only from Northern Stromfjord, West Greenland at L,at. 67°45' N., where Dr. V. Nordmann collected a good number of specimens; the depth was 213 — 218 fath., temp. -=-07°. 7. Ianira laciniata G. O. Sars. (PI. I, fig. 5 a). T872. [antra laciniata G. O. Sars, Forh. Vid. Selsk. Christiania for 1872, p. 92. ! 1897. Ianthe laciniata G. O. Sars, Account, II, p. 101 ; PI. 41. The figures and description published by Sars convey an excellent idea of this characteristic species, but a correction and an additional observation may be noted. Sars wrote: "Segments of meso- some with the lateral parts laminarly expanded, and each produced into two lanceolate lappets separ- ated by a deej) incision, those of the 4 anterior segments subequal, those of the 3 posterior ones rather unequal, the anterior lappet being much the larger". At a first view this seems to be correct, but on a closer examination it is seen that each lateral lamina has two lanceolate lappets only on second to fourth segments, while each of the four other segments have only a single lappet, because the apparently anterior lappet of first segment and the apparently posterior and somewhat small lappet of each of the three posterior segments are in reality marked off by a suture, being epimeral processes CRUSTACEA MALACOSTRACA III. 2I (fig. 5 a, cp, <••/) from the basal joint of the legs. Furthermore a small, very oblong epimeral proci (fig. 5 a, ep) is seen at the bottom of the cleft between the two lappets of fourth segment Most of the upper surface of the animals, especially on the thoracic segments, is finely granu- lated, and besides in most of the smaller and in some of the subadult or adult specimens clothed with fine hairs, generally short, but some long hairs may be interspersed. When hairs are scarce or wanting they have probably been lost. Among my rich material not a single female with the marsupiura fully developed is found. The largest female (from the "Ingolf" Stat. 35) measures 8-6 mm. from the tip of the rostrum to the middle of the hind margin of the abdomen; the largest male (from the same station) is 7-2 mm. Occurrence. Taken by the "Ingolf" at six stations. Davis Strait: Stat. 32: Lat. 66°35' N., Long. 56°38' \\\, 318 fath., temp. 3.90; n spec. Stat. 35: Lat. 65°i6' N., Long. 55°05' W., 362 fath., temp. 3-6°; n spec. — Stat. 28: Lat. 65°i4' N., Long. 55°42' W., 420 fath., temp. 3-5°; 2 spec. Stat. 27: Lat. 64°54' N., Long. 55°io' W., 393 fath., temp. 3-8°; 1 spec. Stat. 25: Lat. 63°3o' N., Long. 54°25' W., 582 fath., temp. 3-3°; 2 spec. West of North Iceland: Stat. 96: Lat. 65°24' N., Long. 29°oo' \\\, 735 fath., temp. i'2°; 1 spec. Besides it has been gathered three times in Davis Strait, viz.: Lat. 65°45' N., Long. 54 30' W., about 200 fath. (Stephensen); Lat. 65°36' N., Long. 56°24' W., 349 fath., temp. 32° (by Admiral Wandel), and Lat. 63°24' N., Long. 53° 10' W., 473 fath. (Stephensen). Distribution. Storeggen bank, off Molde, Norway, 400 fath.; besides off the coast of Nord- land and Finmark, 100 to 150 fath. (G. O. Sars). 8. Ianira spinosa Harger. 1879. Janira spinosa Harger, Proc. U. S. Nat. Mus., Vol. II, p. 158. ! 1880. — Harger, Rep. U. S. Comm. Fish and Fisheries, Vol. 6, p. 323: PI. II, fig. 10. ! 1881. Ianthe sfeciosa Bovallius, Bihang till K. Sv. Vet. Akad. Handl., Vol. 6, No. 4. p. 5; Pis. I III. This species is easily separated from the preceding forms by the conspicuous pair of dorsal submedian processes on each thoracic segment. The surface of the body generally more or les; but the hairs are much less numerous than in some specimens of /. laciniata. - The anterolateral plates of the head are lanceolate, acute, and from not fully twice to nearly three times as long as broad. Epimeral processes are completely wanting at the lateral lappets of the three anterior thoracic segments between the two lappets of fourth segment sometimes a minute, triangular epimeral tubercle may be seen. At the base of the hind margin of the lappets of the three posterior segments a small sub- triangular epimeral process is always visible, and these processes are seen on Bovallius' fig. 1 as teeth incorrectly not marked off by a suture, and at least those at fifth and sixth segments are distinctly a little longer than according to that figure. Harger established / spinosa on two small females, 8 mm. long, and published a single figure, viz., the animal from above. Bovallius established Ianthe speciosa on a very large male, measuring 21-5 mm. from the tip of the rostrum to the end of abdomen (the end of its processes, I suppose); he 22 CRUSTACEA MALACOSTRACA. III. had evidently not seen Harger's papers, but in a later treatise (1886) he keeps Harger's and his own form as "probably" two species. In 1887 I stated that Ianthe speciosa Bov. is only a synonym for Har- ger's species. Harriet Richardson, while writing the Monograph, had not seen any specimen. Noth- withstauding this she considered / speciosa to be specifically distinct from Harger's 1. spinosa, and copied Harger's description and an extract of that given by Bovallius; in a foot-note (p. 460) she added: "Since my manuscript went to press, the types of /. spinosa have been sent from Yale University, and I find it distinct from /. spinosa", but as she had not seen any specimen referred to /. speciosa and not any further material, the statement is of no value. And let us now look at the differences between the figures given by the two authors, and compare them with my material, 8 specimens. According to Harger's figure his specimens had the autero-lateral processes of the head less diverging, the thoracic lappets proportionately a little shorter and broader than on Bovallius' figure; besides, the abdomen has no subbasal dorsal tubercle on Harger's figure. My largest specimen (from the "Ingolf" Stat. 29) agrees as to these particulars well with Bovallius' fig. 1, but another male from Davis Strait differs more from that large specimen than from Harger's figure as to the direction of the frontal processes and the shape and size of the thoracic lappets. The only feature seemingly affording a specific character is the presence or absence of the dorsal abdominal tubercle. But in my specimens this tubercle varies much as to length and thickness. It is rarely shaped like one of the processes on the posterior thoracic segments; it is generally a little or much low-er and somewhat or much thicker than these processes, and in a small specimen, which has the lateral lappets as slender as my largest specimen, the abdominal tubercle is low and very broad. I am inclined to think that Harger's specimens possess the dorsal abdominal process as a low tubercle, and it is to be regretted that Richardson did not say anything on this topic in the foot-note quoted. As a result of my material and the literature I am forced to consider /. speciosa Bov. as only a synonym for Harger's species. It may be added that many of the numerous figures published by Bovallius, and especially those representing mouth-organs, are far from accurate. On the maxillipeds (his fig. 22) he overlooked the two coupling hooks, and the suture or articulation separating the long second joint from its lobe does not exist. On the figure showing the maxillula (fig. 16) and the maxilla (fig. 19) the proximal parts are partly omitted, partly wrong. On fig. 13, exhibiting the distal part of the left mandible, the movable lobe is not marked off, the row of strong seta; shows another aspect and ought not to be interrupted, the molar process is too short. But I found it unnecessary to draw a new set of figures of these appendages, as they did not exhibit differences worth mentioning from those of I. pulclira. My largest specimen, a male from the "Ingolf" Stat. 29, measures 15 mm. from the tip of the rostrum to the base of the uropods, 17 mm. to the end of the abdominal processes; the single female, that with Sphceronella, is i2"2 mm., or to the end of the abdominal processes, 13-2 mm. In a niarsupium of a specimen from Davis Strait, Lat. 66°32' N., I discovered the parasitic Copepod Spharonella curtipes H. J. H. described in my book on the Choniostomatidae. Occurrence. Taken by the "Ingolf" at three stations. Davis Strait: Stat. 31: Lat. 66°35' N., Long. 55°54' W., 88 fath., temp. i-6°; «/a spec. Stat. 29: Lat. 65°34' N., Long. 54°3i' W.. 68 fath., temp. 0-2°; 1 spec. South-East of Iceland: Stat. 4: 64°o7' N., Long. 11 "12' W., 237 fath., temp. 2'5°; 1 spec. CRUSTACEA MALACOSTRACA. III. The species was formerly known From Baffin Bay, Lat. 67°5q/ X., Long. 56 33' W., 98 lath. (Bovallius), and from two places in Davis Strait, viz.: Lat. 66°32' N., 55°34' W., 100 fath, and Lat. 65°35' N-> Long. 54°5o' W., 80 fatli. (H. J. Hansen). Besides, it has been secured by Admiral Wandel at a place north-west of Iceland: Lat. 66° 16' X., Long. 26 8' \Y., 330 fatli., temp, -f- o-i°, 1 spec. Distribution. Hitherto known only from Banquereau, Lat. 440— 45° X., about Long. 580 W. (Harger). Acanthaspidea Stebb. The name of the genus has been given by Stebbing in [893 instead of Acanthoniscu G. 0 which was preoccupied. Only the type species is to hand. The genus is allied to /antra, but differs in the following characters. The maxillipeds (PI. 1. Hg. 6 a) have the second joint of the palp very moderately expanded, and it is ver) much narrower than the broad lobe of the second joint. First pair of legs not prehensile, subsimilar to the other pairs; seventh joint of all pairs moderately long (fig. 6 c), the claw somewhat strong and rather long, the accessory claw slender. In the male the median lamella of the operculum (fig. 6di tapers from the moderately broad base to near the end, which is slightly widened; female operculum somewhat produced behind. Uropods moderately large, with the exopod quite small. Janthopsis Bedd. and probably Jolanthe Bedd. must be united with Acanthaspidea. A revision, based on the study of the appendages, of a good number of genera established in the literatim more or less allied to Ianira and Acanthaspidea, is much needed. 9. Acanthaspidea typhlops G. < >. Sars. (PL I, figs. 6 a— 6 e.) 1879. Acantkoniscus typhlops G. O. Sars, Arch. f. Math, og Naturv. Vol. 4, p. 434. ! 1885. — — G. O. vSars, North-Atl. Exp. p. 119; PL X, figs. 27—30. Sars has published a detailed description with four figures of the female, but he did not know the male. A few corrections and additions may be given here. Sars said of the antennae: "The 2 first joints of the peduncle jut forth on the outer side, as a strong, oblique, outward-directed, spiniform projection", and this description agrees with hi not correct. Sars has overlooked the first joint, which is very short and without any process from the real second joint is very long, while the "projection" from the third joint i>- somewhat shorter and not a process, but the nearly spine-shaped exopod or squama marked off by a distinct Of the thoracic segments Sars said in the diagnosis: "Epimera on 1st segment simple pointed, 011 the 3 succeeding segments two-lobed, on the 3 posterior three-lobed". But he did not distinguish between the different nature of some of the lobes. The lobes on the four anterior segments are only lateral expansions of the segments, and real epimeral plates or processes, projecting from joint of the legs, are quite wanting. The three posterior segments are laterally strongly expanded, and each expansion is bifid, so that the two anterior of the "lobes" originate from the segmen question, while the posterior, much shorter lobe (fig. 6 b, ,p) is an epimeral plate, thus a proo 24 CRUSTACEA MALACOSTRACA. III. first joint of the leg and, seen from above, well marked off by a transverse suture. It may be added that the lateral lobes of the segments are a little longer, more produced, in my older specimens than in the large female figured by Sars. The median lamella of the male operculum (figs. 6 d — 6 e) tapers, as said above, from the moderately broad base to near the slightly widened end. Each of its pleopods terminates in an inner, distally rounded lobe about as long as broad and an outer somewhat shorter, almost spiniform, acute process directed backwards and a little outwards. — The copulatory organ of each lateral part of the operculum is produced into a thin, long thread reaching nearly to the end of the peduncle of the uropods. The female operculum has been figured by Sars. According to Sars his large female was 12 mm. long; my largest specimen, a male, is 8-4 mm. - The specimens from the "Ingolf" Stat. 25 are very young, without seventh pair of legs, and they measure only 2— 21 mm. in length. They differ from developed specimens in having no dorsal tubercles on the thoracic segments, and besides their fifth and sixth segments have the anterior of the lateral lobes still undeveloped, being represented by a feeble angle. Occurrence. Taken by the "Ingolf at two stations. Davis Strait: Stat. 25: Lat. 63°3o' N., Long. 54" 25' W., 582 fath., temp. 3-3°; 27 young. Between Angmagsalik and North-West Iceland: Stat. 95: Lat. o5°i4' N., Long. 30°3c/ W., 752 fath., temp. 2-i°; 1 spec The species has been recorded from Davis Strait: Lat. 63°24' N., Long. 53°io' W., 473 fath. (Stephensen). And the "Thor" has captured it south-west of the Faeroes: Lat. 6i°i5' N., Long. 9°35' W. 463— 5J5 fath-i I spec- Distribution. Hitherto only recorded from a single station in the sea west of Lofoten, 457 fath., temp, -h 070 (G. O. Sars). Ianirella Bonnier. Description. As to the shape of the body intermediate between lanira and Pleurogonium, but somewhat more similar to the latter genus. Head with a rostral process and a pair of long lateral processes (PI. I, fig. 8 a). Eyes almost rudimentary or wanting. Antennulae with the peduncular joints well developed and a low number of joints in the flagellum. Antenna; in the main as in Zanira; squama distinct. - • Mandibles (fig. 8 c), maxillulse, and maxilla; almost as in Ianira\ maxillipeds (fig. 8d) somewhat shorter with second joint broader, fourth joint (second joint of the palp) at most as broad as the lobe from second joint. Thoracic segments produced into lateral processes or lobes. First pair of legs (fig. 8 e) prehensile, and especially third to fifth joints much thicker than in the following pairs; fifth joint with spines along its prehensile margin; sixth joint about half as long as the fifth and moderately strong; seventh joint conspicuously longer than in lanira, with the claw somewhat slender and the accessory claw very small. T1k- other legs somewhat slender and moderately long; claw and accessory claw nearly as in first pair. Abdomen with lateral processes, and behind the uropods considerably produced as a triangular CRUSTACEA MALACOSTRACA. Ill 25 part, which is longitudinally excavated below (fig. 8 g). - Operculum in both sexes | I consid- erably backwards; its median lamella in the male tapering from the base to the end; and the terminal part of each of its pleopods rather narrow, with the end rounded. — The nropods are verv small, consisting of two joints, and one of the rami, probably the exopod, is wanting; they are inserted much in advance of the end of the abdomen on its lateral margins. Remarks. EspecialK by the length of seventh joint of the thoracic legs, the length and shape of the claw and the minute accessory claw, the shape of abdomen and operculum and finally by the reduced nropods this genus is abundantly distinguished from [antra, and by differences especially in the mandibles, the first thoracic legs, the nropods and the male operculum from Pleurogonium. Five species have been established; the "Ingolf" has secured two new species. 10. Ianirella spongicola n. sp, (PI. I, figs. 7 a— 7 ci Description. Closely allied to /. Nanseni Bonn. Body rather convex, scarcely twice as l< as broad, measured to the end of the lateral processes. - The frontal process (fig. 7 a) much shorter than in I. Nanseni, only about half as long as the first joint of the antenuular peduncle; it terminates in a long spine, and has about two very small spines before the end. No eyes. - The first joint of the antennulse more than twice as long as broad; the flagellum 7-jointed. The lateral processes of the head and of the thoracic segments nearly as in /. Nanseni, each terminating in a strong, articulated spine and with some thin lateral spines; the only difference as to the processes is that the two processes at each of second to fourth segments are less divergent, and the anterior conspicuously more than half as long as the posterior (comp. fig. 7 b with Bonnier's Kg. 1 The surface of the body with the same number of processes as in /. Nanseni, viz. one on first thoracic segment, two on the head, on fifth to seventh segments and on the abdomen, three processes on second to fourth segments, but all these processes are somewhat or much shorter than in /. Nanseni. Fig. 7 b exhibits the major part af third thoracic segment; it is seen that the median process is only about as long as broad, and terminates in a long, articulated spine, while the submedian intermediate pair of spines are considerably smaller; Bonnier's fig. 1 1 of fourth segment shows the submedian pail times larger, even larger than the median process. The abdomen is more acutely produced (fig. 7 c) than in /. Nanseni and has the end acufc subacute; its four pairs of lateral processes are not, as in that species, subequal in size, but are conspicuously larger than the second pair, which are distinctly larger than the first or the four Length of the largest specimen, a male, 5-5 mm.; a female without marsupium is somewh; Remarks. The differences pointed out between my specimens and Bonnier's elaborate re] entation of/. Nanseni — especially those in the length of the rostrum and of the dorsal processes on second to fourth segments - are so strong, that I found it necessary to establish a new species on two somewhat mutilated specimens. Occurrence. Taken by the "Ingolf at a single station. South-West of Iceland: Stat. 81: Lat. 6i°44' N., Long. 27 00' W, 485 fath.. temp. 61 4 The Ingolf-Expedition. III. >. 26 CRUSTACEA MALACOSTRACA. III. ii. Ianirella laevis n. sp. (PI. I, figs. 8 a-8 g.) Description. The general outline of the body (fig. 8a) with its lateral processes is in the main as in /. spongicola and /. Nanseni, bnt the dorsal surface has no processes or spines, and several other differences are obvious. — The rostrum (fig. 8 b), which is scarcely as long as the sum of the two proximal antennular joints, is somewhat widened towards the middle and here armed at each side with a robust spine; its end is transverse or emarginate, and each angle has a strong spine. Eyes wanting. — First antennular joint (fig. 8 b) about twice as long as broad; the flagellum with 6 joints. The lateral processes of the head and of the thoracic segments have the terminal spine much smaller than in /. spongicola, and frequently minute; the processes are on the whole a little or some- what shorter than in the preceding species and without lateral spines; the anterior of the two processes at each side of second to fourth segments is much shorter than the posterior. The abdomen (figs. 8 a and 8 g) is distinctly less produced backwards than in /. spongicola, and has its end somewhat narrow and rounded. The four pairs of lateral processes differ extremely in size; third pair are distinctly larger than second, which are much larger than the very small first or fourth pair; each process terminates in an articulated spine. Length of the largest specimen, a female without marsupium, 4 mm. Remarks. I. glabra Richardson, I. abyssicola Richardson and /. Bonnieri Stephensen have no dorsal processes, but /. icevis differs from them in the shape of the rostrum, the abdominal processes, etc. — At the antero-lateral margins of first thoracic segment the epimera are visible as a low protub- erance with a spine; at the postero-lateral margins of the three posterior segments the epimera are generally visible from above as very low, but broad and rounded protuberances; in I. spongicola such epimera at the same four segments are also visible from above, but scarcely as conspicuous as in /. Icevis. Occurrence. Taken by the "Ingolf" at two stations. Davis Strait: Stat. 24: Lat. 63°o6' N., Long. 56°oo' W., 1199 fath., temp. 2-4°; 7 spec. Stat. 36: Lat. 6i°5o' N., Long. 56°2i' W., 1435 fath., temp. 1-5°; 4 spec. Katianira n. gen. Description. Body oblong, rather depressed, in general appearance somewhat similar to Tanira (PI. II, fig. 1 a). — No eyes. Antennulse with a small number of joints in the flagellum. Antennae shorter than the antennulse (PI. II, fig. ic); peduncle apparently 5-jointed, as the first joint has nearly or totally vanished, and the three following joints are short; flagellum with few joints. — Mandibles (PI. I, figs. 9 a — 9 b) differ from those in lanira especially in having the molar process more slender and, besides, tapering to the rather slender end which is obliquely truncate. Maxillipeds (fig. 9 c) have the second joint broad, its lobe broad though narrower than the joint, with two coupling hooks; third joint narrow; fourth joint formed by the complete fusion of two joints, subtriangular, considerably expanded, and a little narrower than the lobe from second joint; fifth joint -- answering to the sixth in other genera — long, distally produced inwards and forwards in an oblong lobe. Epipod of very moderate size, oblong. CRUSTACEA MALACOSTRACA. III. The thoracic segments laterally produced in plates (PI. II, figs. 1 a— i b); first, sixth and seventh segments each with a single lobe; in the four other segments each plate is divided into two lobes by a triangular incision. -— Thoracic legs of middle length; first pair in both sex< I d and i terminate in a fully developed, slender chela; the other pairs have the seventh joint robust md oi very moderate length (fig. i f), the claw somewhat strong and the accessory claw small and slender. The abdomen differs from [antra in having the part between the uropods ver) and protruding as a transverse plate. - The median lamella of the male operculum (fig. i i) is br< tapering to somewhat from the end, and then scarcely widening; the copulatory organ reaches about to the end of its pleopod (PI. I, fig. gd); the female operculum (fi.^. i hi rounded behind. -- The uropods (fig. i g) consist of two joints, the first quite short and hidden beneath the dorsal plate of the abdomen, the second somewhat large. Remarks. Katianira shows some resemblance to Tantra, but differs abundantly in having the antenna; short, the molar process of the mandibles tapering, only four joints in the "palp" of the maxillipeds, a real chela on first pair of legs, and the uropods without exopod. Only a single species is known. 12. Katianira chelifera n. sp. (PI. I, figs. 9 a— 9 d; PI. II, figs. 1 a— 1 i.i Description. Body about two and a half times as long as broad (fig. 1 a), with the sun smooth. — The lateral margins of the head (fig. 1 c) have a longer or shorter portion finely serrate; labrum is visible from above. — The antenuula; nearly reach the posterior margin of first thoracic segment; the first peduncular joint is slightly longer and thicker than the second; flagellum about half as long again as the peduncle, 5-jointed, with second joint about as long as the two following joints together. — Antenna; conspicuously shorter than the antennuhe (fig. re); flagellum shorter than the peduncle, 4-jointed. Thoracic segments peculiarly adorned, as all the free margins are serrate, being closely set with numerous small processes about equal in length (fig. 1 b); the lateral lobes of the segments are 1 acute, terminating in a conspicuous, articulated spine; on second to fourth segments the lateral lo of each segment differ much in size and shape, the anterior lobe being considerably shorter and seven times narrower than the posterior; on the fifth segment the anterior lobe is about as long a posterior. - The chela of first legs in the female (fig. 1 e) about four times, in the male (fig. id) more than four and a half times as long as broad; the fingers are considerably shorter than the hand, eqm in length, somewhat curved near the end, acute, and the movable finger is conspicuously thicker than the other. The abdomen (figs. 1 a, 1 g, 1 i) about as long as broad, on each lateral margin with fix six low protuberances, each terminating in a thick, conspicuous spine; besides, the intervals between the three anterior spines adorned with some minute saw-teeth. The protruding lobe between the uropods is less than twice as broad as long, and its moderately convex hind margin has 6- thin spines. - The operculum is placed somewhat backwards, in the female reaching nearly to, in the male even slightly overreaching, the posterior margin of the abdomen. In the male (fig. 1 i) each 4* 28 CRUSTACEA MALACOSTRACA. III. pleopod of the median lamella has the posterior margin concave and somewhat oblique, as the distal outer corner is produced into a triangular process directed backwards. The female operculum (fig. i h) is slightly longer than broad. — The terminal joint of the uropods a little less than half as long as the abdomen, with some stiff setae along both margins. Length of a male 2 mm. ; the largest female is without marsupium and nearly r8 mm. long. Occurrence. Taken by the "Ingolf" at a single station. South-West of Iceland: Stat. 78: Long. 6o°37' N., Lat. 27°52' W., 799 fath., temp. 4-5; 9 spec. Group II. Haploniscini. In general appearance somewhat similar to animals among the terrestrial Isopoda, excepting, of course, in having the abdomen not divided into segments. Head free. Eyes wanting. Antennae without normal squama, frequently with a long process on third joint. Mandibles normal; molar process sub- cylindrical or thickened towards the end, but otherwise as in Ianira. Maxillary palp with second joint not expanded, slender; the two distal joints well developed. The three, or at least two, posterior thor- acic segments coalesced at least on the dorsal side and immovably coalesced with the abdomen. All thoracic limbs are slender walking legs, with seventh joint at least moderately long, and on the posterior pairs rather long, the claw slender and no accessory claw. Uropoda minute and uniramous, or wanting. Remarks. The group comprises two genera, one of them new. The animals are very small, smooth, and live in considerable or great depths. HaploniSCUS Richardson. Description. Body more or less depressed. Front margin of the head distinctly sinuate or with a short process. — Antennulae (PI. II, figs. 2 a and 4 b) moderately short; first joint thick, twice to four times as thick as, but a little shorter than, the second; flagellum with a low number of joints. Antennae of very moderate length or rather short; first joint scarcely discernible; third joint with an acute, strong or long process, but whether this process may be a peculiarly developed squama is questionable; fifth joint considerably thicker than the sixth; flagellum shorter than the peduncle. — Mandibles (figured by Sars) nearly as in Ianira. Maxillipeds (fig. 2 e) with the two proximal joints moderately broad, the lobe from second joint broader than this joint and with few coupling hooks; the joints of the palp slender; epipod large, nearly triangular. Thoracic segments conspicuously expanded laterally, but their lateral margins are straight or slightly convex in their full length ; epimeral processes wanting. The three posterior segments either well marked off from each other and from abdomen towards their lateral margins, while their broad dorsal median parts are coalesced or even fused, or — in //. armadilloides — the articulation between fifth and sixth segments seems to be not fully undeveloped, while the two posterior segments and abdomen seem to be immovably coalesced, though moderately distinct sutures between them remain CRUSTACEA MALACOSTRACA. Ill on the broad median dorsal surface. The legs slender; Eirsl pan not prehensile and differing but little from the second pair; the posterior pairs with seventh joint and the claw (fig. 2 f) distinctly Ion) and thinner than on the anterior pairs; the claw is somewhat or very long (figs. 2 f and 4 c) and instead of an accessory claw a couple of hairs. Abdomen about as broad as, or considerably broader than, long, and posteriori) truncate, with a tooth or process at the end of each lateral margin. Operculum in the female (figs. 2I aiul.(di propor- tionately small, broadly rounded posteriorly. - Median lamella of the male operculum (it is unknown in H. armadilloides) broad a little from or at the base (figs. 2h and 31), and tapering considerably somewhat from the end, its distal part is somewhat widened and peculiar in structure; the copulat organ of second pleopods produced into an extremely long thread (fig. z h, > :; fig. 3g). — Uropods very small or nearly rudimentary, consisting of an oblong joint, and sometimes an exceedingly small proximal joint can be discerned (fig. 2 1). Remarks. In his Account (p. 119) Sars says that Nannoniscus bicuspis < . " > S. "is scarcely congeneric" with the type of the genus, N. oblongus G. O. Sars, and with N. casfiius Q. I >. S., "differing, as it does, considerably, in the structure of both the anteniue and the oral parts, and also in that ol the caudal appendages." In 1908 Harriet Richardson established the genus Haploniscus with A', bicuspis G. O. S. as the type, and described, besides, two new species. In 1914 Vanhoffen added two antarctic species. The "Ingolf" captured three species, two of them new. 13. Haploniscus bicuspis G. O. Sars. (PI. II, figs. 2 a— 2 1). 1877. Nannoniscus bicuspis G. O. Sars, Arch. f. Math, og Naturv. Vol. II, p. 352. ! 1885. — G. O. vSars, North-Atl. Exp. p. 122; PI. 10, figs. 31—45. Description. Body rather depressed, in the female from two and a half to two and two- third times as long as broad, in the male (fig. 2 a) about three times as long as broad. -- Front margin of the head nearly transverse with a minute protuberance at the middle (fig. 2 In. — The antennuke, especially their two proximal joints, thicker in the male (fig. 2 b) than in the female (fig sexes scarcely reaching the end of the penultimate joint in the antennal peduncles; flagellum in the male 5-jointed, in the female 4-joiiited. -- Antennae somewhat less than half as long as the b process on the third joint (figs. 2 b, 2 c, 2d) is robust, somewhat oblong-triangular, acute, proj upwards and a little or somewhat outwards from the middle of the upper surface of the joint; pen timate peduncular joint much thicker in the male (fig. 2b) than in the female (fig. 2d); last peduncu joint without any tooth or process at the end; flagellum with 13 joints. The three posterior thoracic segments and the abdomen fused on the broad median part of the dorsal surface (fig. 2 a); the limit between sixth and seventh segments generally perceptible, between fifth and sixth segments and between seventh segment and abdomen are only visible towards the lateral margins. Abdomen in the female (fig. 2 1) a little broader than long, with the postero-lateral processes only about one-third as long as the distance between them; uropods conspicuous, reaching to a little 30 CRUSTACEA MALACOSTRACA. Ill from the end (fig. 2 1), or nearly beyond the end (fig. 2 k) of the adjacent process; operculum slightly broader than long. — In the male the abdomen is a little longer than broad (figs. 2 a and 2h); the postero-lateral processes are much larger than in the female, more than half as long as the margin between them, which is incurved just above each uropod (fig. 2g); the uropods are conspicuous from below, reaching a little beyond the posterior margin (11 at figs. 2 g and 2 h). — The median lamella of the operculum is suddenly considerably widened a little from the base (figs. 2 h and 2 i), as each lateral margin is excavated near that base; the distal part of the lamella has two pairs of minute angles on the outer margin, respectively a little before and a little behind the middle; the thread of the copulatory organ is almost three times as long as second pleopod (fig. 2 h, c). Length of a female with marsupium 2-5 mm., of a male 2-6 mm. Remarks. Sars published a detailed description with figures of the female, but he overlooked the dorsal fusion of the segments mentioned, and he had no male. Occurrence. Taken by the "Ingolf" at fifteen stations, all in cold area. North of the Fseroes: Stat. 141: Lat. 63°22' N., Long. 6°58' W., 679 fath., temp. -4- cr6°; 8 spec. Stat. 138: Lat. 63°26' N., Long. 7°56' W., 471 fath., temp. -=- o-6°; 10 spec. Stat. 139: Lat. 63°26' N., Long. 7°3o' W., 702 fath., temp, -f- o-6°; ab. 32 spec. East of Iceland: Stat. 105: Lat. 65°34' N., Long. 7°3i' W., 762 fath., temp. ■— o-8°; 2 spec. Stat. 104: Lat. 66°23' N., Long. 7°25' W., 957 fath., temp. -f- rt°; 13 spec. Stat. 103: Lat. 66°23' N., Long. 8°52' W., 579 fath., temp. -=- o-6°; 4 spec. Stat. 102: Lat. 66°23' N., Long. io°26' W., 750 fath., temp, -f- 0-9°; 4 spec. North of Iceland: Stat. 124: Lat. 67°4o' N., Long. i5°4o' W., 495 fath., temp, -f- o-6°; 5 spec. Stat. 125: Lat. 68°o8' N., Long. i6°o2' W., 729 fath., temp. -^ o-8°; 3 spec. North-East of Iceland: Stat. 120: Lat. 67°29' N., Long. n°32'W., 885 fath., temp. -^ ro°; 4 spec. Stat. 119: Lat. 67°53' N., Long. io°i9' W., 1010 fath., temp, -f- ro°; 18 spec. South of Jan Mayeu: Stat. 118: Lat. 68°27' N., Long. 8°2o' W., 1060 fath., temp. -f- ro°; 2 spec. Stat. 117: Lat. 69°i3' N., Long. 8°23' W., 1003 fath., temp. -=- i-o°; ab. 33 spec. Stat. 113: Lat. 69°3i' N., Long. 7°o6' W., 1309 fath., temp. -^- ro°; 9 spec. Stat. 116: Lat. 70°05' N., Long. 8°26' W., 371 fath., temp. -\- 0-4°; 12 spec. Besides, it has ;been taken East of Iceland at Lat. 65°53' N., Long. 70 18' W., 1163 fath., temp. -=- ri° (G. O. Sars). Distribution. Taken by Sars in the cold area west of Norway at Lat. 63°5' N., 525 fath., temp. -~ i-i°, and at Lat. 69°4&' N., 649 fath., temp. -=- 07°. But when Sars records it from the warm area between Finmark and Beereu Eiland at Lat. 72°27' N., Long. 20°5i' E., 191 fath., temp. 3-5°, the statement seems to me a little dubious, because this species otherwise has been taken exclusively, and even at 18 stations, in the cold area in from 371 to 1309 fath.; perhaps the specimens from the warm area belong to the following species or to another closely allied form. CRUSTACEA MALACOSTRACA. [II. jj 14. Haploniscus spinifer 11. sp. (PI. II, Ggs. 3a-3h.) Description. This species is so closely allied to //. hintspis, that it is sufficient to point out the differences. — The antennnhe with only 4 joints in the flagellum (fig. 3 a). The antennae have the process on third joint nearer to its base (fig. 3 c) than in //. bicuspis\ in both I oi the pednncle has an oblong tooth or acnte process above at the distal end (figs. 3 a— 3 b), and the flagellum has only 11 joints. — The median lamella of the male operculum (fig. 31) is only about hall as long again as broad at the base, the proximal part of the outer margin somewhat convex with the incision found in //. bicttspts, and the terminal part of the lamella has its lateral margins 1 without any protruding angle. Two adult males from the same place (Stat. 22) differ considerably from one another; on shaped as the male figured of // bicztspis (fig. 2 a), but its postero-lateral processes are much la: (fig. 3 d), each being half as long again as the posterior margin between them. The other male lex almost like a female, as the lateral margins of the body are more convex, and the animal consequently broader in proportion to length, furthermore the antennulae and antennae are distinctly more slender. and the postero-lateral processes (fig. 3e) much shorter, being- even shorter than the margin between them. The largest male measures 2'8 mm. from the front margin to the end of the very long post lateral processes. No female with marsupium was secured. Remarks. The process at the end of the antenna! peduncle is the best character between //. spinifer and H. bicuspis, as the latter species has no trace of any process. A second allied species without any process at the end of the antennal peduncle is //. retrospinis Richardson, of which I have examined co-types from the U. S. Nat. Mus. Occurrence. Taken by the "Ingolf" at three stations, all in the warm area. Davis Strait: Stat. 36: Lat. 6i°5o' N.. Long. 56°2i' W., 1435 fath., temp. 1-5°; 4 spec. South of Davis Strait: Stat 22: Lat. 58°io' N., Long. 48°25' W., 1845 fath., temp. 1-4°; 4 spec. South-West of Iceland: Stat. 78: Lat 6o°37' N., Long. 27°52' W., 799 fath., temp. 4-5°; n but most of them very young. Haploniscus armadilloides n. sp. (PI. II, figs.4a-4d). Female. Body about two and a half times as long as broad, rather convex, and the major part nearly semi-cylindrical; the general appearance reminding one of a minute A rmadillo. the head somewhat expanded outwards and forwards, so that the anterior margin, which has part somewhat convex (figs. 4 a— 4 b), is conspicuously concave between the middle and rounded anterolateral angles. The frontal plate (fig. 4 b) is a rather broad triangle with the apex acute and each lateral margin sinuate. — Antennulae about as long as the head; flagellum the peduncle, 3-joiuted (fig. 4 b). — Antennae short, reaching about the posterior angle of first segment; the first joint could not be made out; the process on third joint is extrcmeh acute, projecting outwards and a little forwards from the anterior side of the joint, and reachii near the end of fifth joint; flagellum a little shorter than the peduncle, 7-jointed. 32 CRUSTACEA MALACOSTRACA. III. The articulation between fifth and sixth thoracic segments seems to be not quite undeveloped, while the two posterior segments and the abdomen seem to be immovably coalesced, and the limits between them are on the major part of the upper surface only moderately distinct sutures, but con- spicuous towards the lateral margins. On the thoracic legs the claw is longer than the seventh joint (fig. 4 c). Abdomen more than half as broad again as long (figs. 4 a and 4d); the postero-lateral processes are small, broader than long, acute; the posterior margin is more convex than in the two preceding species. — The operculum is nearly circular. — The uropods are very small, not reaching the end of the processes; a division into two joints could not be observed. Length of the largest specimen, a female without marsupium, 1-5 mm.; the other specimen juvenile. Remarks. In general aspect H. armadilloides differs considerably from the two preceding species, and it can roll itself nearly as many terrestrial Ouiscidse. It agrees with H. bicuspis in several features, thus in having a peculiar process on third antennal joint, in the postero-lateral abdominal processes, the shape of the uropods, the abdominal operculum, but shows a number of sharp differences, which, at least provisionally, are considered as not being of generic value. Occurrense. Taken by the "Ingolf" at a single station. South of Iceland: Stat. 54: Lat. 63°o8' N., Long. i5°4o' W., 691 fath., temp. 3-9°; 2 spec. Hydroniscus n. gen. Description. Of this curious genus only the female is known, and it is related to Haplomsats. - Body oblong, highly vaulted and contractile into globular form. Head anteriorly with a broad, rounded incision at each side of the moderately broad median part, which is produced as an obtuse process (PI. II, fig. 5 c). — Antennulae moderately short; second joint longer than in Haploniscus. Antennas short; third joint without process. — Mandibles (figs. 5 d— 5 e) somewhat similar to /antra, but the molar process is more thickened towards the end, and the movable lobe is very strong on the left mandible. Maxillipeds (fig. 5 f) with the two proximal joints and the epipod somewhat broader than in Haploniscus, while second and third joints of the palp are still more slender than in that genus. The three posterior thoracic segments and the abdomen are strongly vaulted (fig. 5 b) and, besides, so completely fused that only two rudiments of articulations between segments are visible on the sides; seen from above the abdomen has the lateral margins somewhat convex and converging to the narrowly rounded end (fig. 5 a); seen from the side (fig. 5 b), the lateral margin terminates in a small, triangular tooth not visible from above, and situated in advance of the somewhat protruding end of abdomen. — Thoracic legs slender, especially the posterior pairs; first pair (figs. 5 g and 5 h) with seventh joint moderately long and rather slender, on the posterior pairs (fig. 5 i) long and very slender; the claw rather long, shorter than seventh joint, and an accessory claw is wanting. Operculum (fig. 5 k, o) subangular before the middle. — Uropods completely wanting. Remarks. It is easily seen that this genus is closely related to Haplonisciis in spite of its peculiar aspect. It was impossible to find any vestige of uropods; if lost, their points of attachment must be discoverable. Only a single species is known. CRUSTACEA MALACOSTRACA. III. 16. Hydroniscus abyssi n. sp. (PI. II, figs. 5 a -5 k). Female. Body three times as long as broad, broadest considerabl) behind the middle surface extremely smooth. - • The frontal process of the head (fig. 5c) about as loi end broadly rounded; the antero-lateral angles are triangularly protruding. - Antennulae with tin 5-joiuted; the terminal joint very short. -- Antennae somewhat longer than tin- antennulae; the proximal joints of the peduncle are very difficult to discern, being placed in a deep excavation limited 1>\ tin- very high lateral plates of the head; last joint of tin- peduncle longer than tin.- preceding joint; 1 him nearly as long as the peduncle, g-jointed. The abdominal operculum a little longer than broad (fig. 5 k, 0); with a rounded media not reaching the base; the terminal margin somewhat short, almost straight Seen from the anal doors (a) are very conspicuous and reach the hind margin, and near the outer margin oi door the postero-lateral process (/>) is observed. Length of the largest specimen, which has no marsupium, 2-8 mm. Remarks. This interesting species is easily distinguished from all marine fsopoda hitherto known. Occurrence. Taken by the "Ingolf" at its deepest station South-West of Cape Farewell: Stat. 38: Lat 59°i2' N., Long. 51 05' W ith., temp. 1 3 7 spec., most of them very young, Group III. Munnini. Body very varying in aspect, but the four anterior thoracic segments conspicuously marked off from the three following segments, and second to fourth segments considerably to very much bro; than abdomen, which is less or more produced. Head free. Eyes, if present, situated on lateral pr< tuberances or processes of the head. Antenna; with the squama minute or wanting. the incisive part, the movable lobe, and at least a few setce well developed; molar process dii somewhat or even considerably forwards, either shaped nearly as in [antra or longer and thinner with the end oblique; palp in some genera reduced or wanting. Maxillary palp wil joint from rather broad to slender; the two distal joints well developed. - First pair of legs prehensile, their fifth joint being spiniferous and at hast robust, fr< thickened. The six following pairs "more or less rapidly increasing in length, simple, accessory claw generally discernible, sometimes long and strong (Mum on or above the lateral margins, and somewhat or considerably in fronl generally minute and subniarginal, but in a few forms strong, long, and with their insert Remarks. The group answers to the family Munnidae G. 0. Sars. Thi three of the four genera represented in the "Account" of Sars, and instead of the fourth mil una G. O. S., it has secured a new genus, Pseudomunna n. gen. In reality som< Munna and Dendrotion, differ much from each other in a number of features. And The Ingolf- Expedition. III. 5. 34 CRUSTACEA MALACOSTRACA. III. other genera have been discovered in other seas, especially at Kergnelen and in the Antarctic Ocean, so that the group, which answers to three families in Vanhoffen's work (1914), in the future may be divided in a satisfactory way, when those southern forms have been more closely investigated. Munna Kroyer. The best information on this difficult genus is found in Sars' "Account"; he pointed out the generic characters and described 5 Norwegian species. Some few corrections and additions to the northern fauna have since been published by myself (1910) and Stappers (191 1). In the present paper 7 species from the "IngolP area are mentioned, 3 of which are figured in Sars' work, and 2 are new to science. The species of Munna are far from easy to deal with, as most of them show considerable in- dividual variation, and many of the specimens are very mutilated. Variations in armature with spines, 111 the number of joints in the antennular flagella, etc., are pointed out in the descriptions of, or remarks on, several species on the following pages. An important specific character not mentioned by Sars is the shape of the median lamella of the male operculum ; in reality the shape of this lamella, especially its terminal part, affords, perhaps, the sharpest and most reliable character. Besides it may be pointed out that the coxae, first joint of the thoracic legs, are thick and developed as a kind of epimera which are attached to the lateral end of the segments (PI. Ill, figs. 7 b and 7 d) and, especially on the posterior pairs, not always easy to discern from the segments; these coxae or epimera — generally wanting at first segment ■ are frequently adorned with a tooth or with spines or processes, and sometimes the lateral margins of the segments are, besides, armed in a similar way. Sars' diagnosis of the genus must be altered a little. As M. acanthi/era n. sp. (and M. truncata Richardson) are completely without visual organs, the statement of Sars: "Eyes distinct ..." cannot be maintained. As to the first pair of legs a couple of lines in his diagnosis are not quite correct, as in most species there is no appreciable sexual difference in these legs. Finally it may be added that the female operculum has the posterior end rounded; in most species it is furnished with some spines on the proximal half of its ventral surface. 17. Munna Boeckii Kroyer. (PI. Ill, figs. 1 a- ib). 1839. Mil inui Boechii Kroyer, Naturh. Tidsskr. Vol. II, p. 612; PI. VI, figs. I— 9. ! 1897. G. O. Sars, Account, II, p. 107; PL 44. According to Sars, the antennuke have 6 joints in the flagellum, and four of these joints are long; I have the same number in one of two Norwegian specimens, but only three long and two short joints in the other specimen, and the specimens from the Fseroes have only the last-named number. - The coxae of second to seventh pairs of thoracic legs are armed with spines not mentioned or figured by Sars; the greatest number of spines observed is two on second and seventh, three on third and fourth, and CRUSTACEA MALACOSTRACA. III. 35 four on fifth and sixth pairs of coxa', but frequently the number is lower, sinking to one or two on second to fourth pairs, and to two on each of the following pairs. The abdomen varies somewhat in breadth, but is always pyrifonn and never slender, being less than half as long again as broad, and generally rather broad; it has always three to five robust spines at each side, placed not in a line, but partly on and partly a little above the lateral margin. The median lamella of the male operculum (figs, ia — i b) has several small spines irregularly arranged on the proximal half of the ventral surface; the lamella is from almost three times to somewhat more than three times as long as broad, distinctly tapering from somewhat before to somewhat beyond the middle, while its distal part widens again conspicuously to the end, which is as broad as, or even a little broader than, the proximal half and distinctly emarginate, as the hind margin of each pleopod is a little oblique, while a minute tooth directed backwards is seen at its outer angle. Remarks. This species is easily recognized by the characters pointed out above together with the representation of Sars. Especially the shape of the median lamella of the male operculum is important, as it differs sharply from those in all the following forms excepting Al. Hanseni. The differences between M. Boeckii and M. groenlandica n. sp. are pointed out at the last-named species. Occurrence. Taken by the "Ingolf" at a single station. South-West of Iceland: Stat. 85: Lat. 63°2i' N., Long. 25°2i' W., 170 fath.; 1 spec. (poor). It has been taken by Dr. Th. Mortensen on three places at the Faeroes, viz. off Boro, 20 — 30 fath., 2 spec, (one spec, in Myxilla fimbriate!), north-west of Kalso, 60 fath., 7 spec., and off Nolso, 100 fath., many spec. Distribution. West coast of Norw*ay, 20—50 fath. (G. O. Sars); North Sea at Lat 57°I7' N., Long. 7°47' E., 27 fath. (Zirwas), and at some places in Scotland (T. Scott). The animals recorded by Meinert from Denmark belong to two other species (H. J. Hansen, 1910). 18. Munna groenlandica n. sp. (PI. Ill, figs. 2 a— 2 d). ?i846. Munna Fabricii Kroyer, in Gaimard, Voy. en Scand., Crust. PI. XXXI, figs. 1, a q (partim), 1847. — — Kroyer, Nat. Tidsskr., Ny Raskke, Vol. II, p. 380 (partim). Description. As to shape of the body and length of the legs this species is intermediate between M. Boeckii and M. limicola G. O. S. - The eyes (fig. 2 a) are as large as in J/. Boeckii. The antennulse (fig. 2 a) a little longer than in that species; the flagellum has in the adult female most frequently 4, but sometimes only 3. long joints, besides a proximal short and a terminal rudimentary joint (an antennula with the first of the long joints uncommonly long but not divided into two joints is shown in fig. 2 a). — The coxae of the thoracic legs somewhat less spiniferous than in M. Boeckii, generally with a single spine on the anterior pairs and two spines on the four posterior pairs. The abdomen (fig. 2 b) — the first segment not taken into consideration -- is generally a little narrower than in M. Boeckii, and is armed at each margin with two small spines, rarely one spine, and in a single female each side possesses four somewhat small spines visible from above. Uropods as in M. Boeckii. — The median lamella of the male operculum (figs. 2 c 2d) is characteristic: somewhat 36 CRUSTACKA MALACOSTRACA. III. from the base a single pair of large spines are found on the ventral surface, and in the largest male besides two or three spines; the lamella is about three times as long as broad, tapering from consider- ably before to a little beyond the middle and then with the margins nearly parallel to near the end, which is a little narrower; the terminal margin of each pleopod is deeply concave, as the pleopod terminates in a rounded setiferous lobe, while its outer part is produced into a rather large, triangular, acute process directed backwards. Length of the largest specimen, a female without marsupium from the Upernivik district, 3-4 mm.; a male from the same place is 3-2 mm. long. An ovigerous female from Egedesmiude is 27 mm. Remarks. In the list of the Danish Isopoda, etc. (1910) I had referred the specimens of this species to AI. Boeckii, but after the discovery of the value of the shape of the median lamella of the male operculum I saw that it is in reality a new species, though none of the characters found in other organs are very valuable, the best being the conspicuously feebler armature with spines on the abdomen. The opercular lamella differs sharply from those in the six other species by the shape of its end. Occurrence. Not taken by the "Ingolf". At West Greenland this species has been taken at three places. In the Upernivik district, Lat. ab. J2°4'j' N., 3 spec, were secured by Commodore Ryder; at Egedesmiude, Lat. 68°42' N., 1 spec, by Mag. Traustedt. In a bottle labelled: Godthaab (Lat. 64°n' N.) c. 50 fath., Holboll, I found 6 specimens of this species among 9 specimens of the real M. Fabricii, and they must be considered as determined by Kroyer. At East Greenland it has been secured at four places, viz.: at Angmagsalik, Lat. 65°3o' N., 2 spec, by Mag. Kruuse; at Tasiusak, Lat. 6j°3j' N., 3 — 5 fath., 5 spec, by the 1st Amdrup Exp.; at Lat. 67°4' N., at the beach, 2 spec, by the Ilnd Amdrup Exp.; finally at Danmarks-0, Lat. 70°27' N., 3 spec. by the Ryder Exp. 19. Munna Hanseni Stoppers. (PI. Ill, figs. 3a-3h.) 1911. Munna Hanseni Stappers, Crust. Malacost., in Due d'Orleans, Camp. Arctique de 1907, p. 91. Description. General aspect of body and limbs nearly as in M. Bocckii. — Eye-stalks and eyes together (fig. 3 a, 3 c, 3d, 3 e) conspicuously smaller than in M. Boeckii, generally forming a some- what conical or at the end broadly rounded protuberance which varies in shape, being from a little longer to conspicuously shorter than broad; the eyes are very reduced, containing only some few facets, or sometimes scarcely any facet can be perceived; the inner contents of the eyes are light brownish and much removed from the cornea. -■ The antenuulae (figs. 3 a, 3 c and 3 d) always consist of 8 joints, hut they vary considerably in length, as in specimens from the "Ingolf" Stat. 44 they are only some- what more than half as long as the distance between the eyes (fig. 3 a), while in the specimen from Stat. 81 they are even a little longer than that distance (fig. 3 d), and in the specimens from the two other stations they are almost as long; such long antennuke are, besides, more slender than in spec- imens from Stat. 44. First peduncular joint is moderately thickened, longer than broad; second joint somewhat or considerably longer than the first; the 5-jointed flagellum with the three intermediate joints more or less Long, but varying much as to their relative length; terminal joint very short. CRUSTACEA MALACOSTRACA. 111. 37 Thoracic segments and coxae arc most frequently scantily haired, but sometim lied with numerous hairs. The coxa; of second to fourth pairs of legs with a subcorneal or rounded protuberance at the middle of their outer margin, but in the small, hairy specimens from Stat 78 these tubercles could not be discovered; the coxae of the three posterior pairs of legs each with a single spine (fig. 3 b) or sometimes with two spines. Abdomen (the free basal segment not included) is oblong-ovate (fig. 3b), broadest much 1m the middle, not much varying as to breadth and generally about half as long again as broad. On the sides a single pair of spines are found, and in adult specimens the upper surface and the posterior half of the sides have in all about four pairs of long or very long spines, and the posterior margin has a pair of very long, thin spines; some or most of these spines are frequently broken oil, and at least sometimes they are partly or entirely wanting in not full-grown specimens. -- Uropods about as in J/. Boeckii. The median lamella of the male operculum (figs. 3 g — 3 h) has no ventral spines, and is some- what less than three times as long as broad, broadest somewhat from the base, then tapering to the end of the second third of the length, and then very feebly diverging to the end; the hind margin of each pleopod has at the outer margin a minute tooth directed backwards, and is slightly sinuate and oblique so that the margin of the whole lamella is feebly emarginate, especially at the middle; in the specimen from Stat. 81 the lamella is a little more widened towards the end, and its distal p is shaped as in M. Boeckii. Length of a large female with marsupium 3 mm., of a male 2-9 mm. Remarks. M. Hanseni is allied to M. Boeckii in most characters, but is instantly distinguished by the reduced eyes; the armature with spines on the abdomen of adult and many not-adult specimen- is also characteristic. Occurrence. Taken by the "Ingolf at four stations. South-West of Iceland: Stat. 78: Lat. 6o°37' N., Long. 2f& W., 799 Eath., temp. 4-5°: 9 small spec. _ . _ Stat 81: Lat. 6i°44' N., Long. 27°oo'W., 485 fath., temp. 6-i°; 1 spec. North of Iceland: Stat. 126: Lat. 670io.' N., Long. i5°32 W., 293 fath., temp. -7-0-5°; 3 spec. West of the Fseroes: Stat. 44: Lat. 6i°42' N., Long. 9°36' W., 545 fath., temp. 4-8°; 14 spec. Distribution. Only known from a place at the south coast of Novaya Zemlya, Lat. 70°2< Long. 56°35' E., 48 fath. (Stappers). 20. Munna Kroyeri Goodsir. (PI. Ill, figs. 4 a— 4 b). 1842. Munna Kroyeri Goodsir, Edinb. New Phil. Journ. Vol. XXXIII, p. 365; PI. VI, fig. 6 (tesl 1 ^97. — — G. O. Sars, Account, II, p. 109; PI. 46, fig. 1. This stout, short-legged species has been well figured and described by Sars; I have examined 6 co-types presented by him. -- Median lamella of the male operculum (figs. 4 a— 4b) without ventral spines, somewhat more than twice as long as broad, broadest somewhat from the base and tapering considerably somewhat from the end, where it widens rather rapidly to the end; the outer distal angle 38 CRUSTACEA MALACOSTRACA. III. is produced into a moderately small, oblong-triangular, acute process directed backwards and mainly outwards ; the hind margin of each pleopod somewhat sinuate, in the main vertical on the median line. Remarks. Sars in 1882 and in the "Account" says that a number of Kroyer's figures of his M. Fabricii belong to M. Fabricii Kr., G. O. Sars, and others to M. Kroyeri Goods., Sars. This statement led me astray in my list of the Malacostraca from West Greenland (1887), and in following Sars I made erroneous statements. Having now, assisted by "Account" and a rich material, studied the whole matter and examined again the specimens from Greenland determined by Kroyer, I have arrived at the result that some of these specimens belong to M. Fabricii Kr. but not M. Fabricii G. O. S. which in 1910 I gave the new name M. minuta, furthermore that the other specimens of Kroyer do not belong to M. Kroyeri Goods., Sars, but to M. groenlandica 11. sp. (see above). The statements on the species of Alumni in my list from 1887 and most of those in Stephensen's "Conspectus" ought in future to be left out of consideration. Occurrence. Not taken by the "Ingolf". Among the whole material of Munna from our area, I found only a single male, taken by cancl. mag. Saemundsen at Vestman-0erne, south of Iceland, in the littoral belt. Distribution. At the south and west coast of Norway, in comparatively shallow water (G. O. Sars); not known from Denmark, but a specimen has been recorded from the bay at Kiel (Apsteiu). Furthermore taken at a good number of places at Scotland, England and Ireland (several authors), besides at Jersey (teste Norman). Tattersall has recorded it as taken in 115 fath. west of Ireland. But I am inclined to think that several of the localities recorded ought to be referred to M. Fabricii Kr. When T. Scott recorded M. Kroyeri from Northbrook Island, Franz-Joseph Land, I am tolerably sure that his specimens belong to At. Fabricii Kr. 21. Munna Fabricii Kroyer (nee G. O. Sars). (PI. Ill, figs. 5 a-5 d.) ?i846. Milium Fabricii Kroyer, in Gaimard, Voy. en Scand., Crust. PI. XXXI, fig. 1, a — q (partim). 1847. Kroyer, Nat. Tidsskr., Ny Rrekke, Vol. II, p. 380 (partim). ! 1910. H.J. Hansen, Vid. Meddel. Naturh. Foren. Kjobenhavn for 1909, p. 211; PI. Ill, figs. 1 a — 1 e. The description and figures in my paper quoted may be sufficient for the recognition of this species. It is instantly separated from all northern forms by the regularly oval, not pyriform or ovate, shape of the abdomen, which, besides, has no lateral processes as in M. Kroyeri^ and at most a single siiblateral spine. The uropods are very thick and distally curved and produced into a strong, acute process directed backwards; seen from below (fig. 5 a) a strong, oblong process projects beyond the oblique terminal margin, and seen from above (fig. 5 b) the end of that process is perceived close by a slender process from the end of the upper wall. — The median lamella of the male operculum (figs. 5c — 5cl) is a little more than three times as long as broad, without ventral spines; in outline it is somewhat similar to that of M. Boeckii, excepting its most distal part, which is less widened and quite differently shaped; the outer distal angle of each pleopod is produced into a long, slender, acute pro- cess directed backwards and somewhat outwards, and the hind margin of the pleopod is somewhat convex. CRUSTACEA MALACOSTRACA. Ill It may be added that the legs are, as usual, distinctly varying as to length and thickness, being strongly or moderately strongly built, and the posterior pairs proportionately moderately short or somewhat long. Remarks. This characteristic species was certainly mistaken for, or confounded with, M. Kroycri or M. Boeckii by several authors until 1910. Occurrence. Not taken by the "Ingolf". At West Greenland it has been gathered at two places, viz. by Th. Holm at Upernivik (Lat 72°47' N.), 10 fath., and at Godthaab (Lat. 64°n' N.|, c. 50 lath., Holboll; as already stated. I foum specimens of M. Fabricii together with 6 specimens of M. groenlandica, all named .1/ Fabricii, in all probability determined by Kroyer, and originating from the last-named locality. It is not known from East Greenland. M. Fabricii is common at Iceland, where it has been taken by several zoologists at the following places. Off the west coast at Stykkishohnr in Brede Fjord, 30 fath.; off the north coast at Grim 15 fath.; off the east coast at Skalanes, 7— 8 fath., in Faskruds Fjord, 20—50 fath., at Djupivogr, 8 fatli., and 9 miles off the coast, 38 fath., finally south of Iceland: Lat. 64°i7' N, Long. I4°44' W., 44 fath. - Hitherto not known from the Faeroes. Distribution. At Denmark this species has been gathered in Langelandsbeltet, 15 fath. (H. J. Hansen). The Copenhagen Museum possesses it from the North Sea, Lat. 57°i6' N., Long. 5°3o' E, 30 fath. (taken by Capt. Orsted) and 1 old specimen from the west coast of Norway. It has been recorded from the south coast of Novaya Zemlya, Lat. 70-20' N., Long. 56 35' E., 48 fath. (Stappersi, and the specimens from Northbrook Island, Franz-Joseph Laud, referred by T. Scott to M. Kroyeri probably belong to M. Fabricii. Whether M. Brandti Zirwas (1910), recorded from sixteen places in the North Sea, belongs to M. Fabricii Kr. cannot be decided with certainty, but it is highly probable. More cannot be said on the distribution, as authors certainly have confounded it with other species. 22. Munna minuta H. J. Hansen. (PI. Ill, fig. 6 a.) 1897. Munna Fabricii G. O. Sars, Account, II, p. 108; PI. 45. fig. 1 mec Kroyer . 1910. — minuta H. J. Hansen, Vid. Medd. Nat. Foren. Kjobeuhavn f. 19/39, p. 213; PL III, figs. 2 a— 2 c. The description and figures published by Sars together with my notes and figures quoted may be quite sufficient for recognizing this small and thin-legged species. It may, however, be added that an examination of my large materiel has given the result, that in some specimens the abdomen free basal segment not counted — is about as narrow as figured by Sars and is furnished with a pair o\ small and slender lateral spines, while in the majority the abdomen has no lateral spines and i as broad as in fig. 2 c in my paper quoted or shows every shape intermediate between this figure and the figures of Sars. Consequently the abdominal shield varies from being slightly longer than broad to slightly more than half as long again as broad, but it is always ovate, broadest considerably before the middle. Finally an important specific character may be added. The median lamella of the male operculum 4o CRUSTACEA MALACOSTRACA. III. ifig. 6a) has several ventral spines on its proximal third; it is conspicuously broader than in any of the other species, only twice as long as broad, considerably constricted somewhat from the end, and then widening strongly to the end; each distal angle is produced into a somewhat long, spiniform, acute process directed mainly outwards; the long hind margin of each pleopod is a little sinuate and, in the main, transverve. Length of the largest females with marsupium (from West Greenland) 2-5 mm. Occurrence. Not taken by the -'Ingolf". It must be common at West Greenland, as it has been taken by several collectors off that coast at six places, from Lat. 72°43' N. to about Lat. 6o°43' N. The places are: Upernivik, 10 fath.; Egedes- minde; Lat. 66°3o' N., Long. 54°5o' W., 40 fath.; Lat. 64°52' N., Long. 53°io' W., 28 fath.; Godthaab, 4—10 and 25 fath.; finally not far from Julianehaab, 22 fath. -- In East Greenland it has been taken by Mag. Kruuse at Angmagsalik (Lat. 65°3o' N.) and Tasiusak (Lat. 65°37' N.); by the Ilnd Amdrup Exp. at Lat. 66° 1 5', low water. Near Iceland it has been taken by several zoologists at the following places. Off the west coast at Reykjavik (G. O. Sars) and in Talkua Fjord, 14 fath. Off the east coast in Yid Fjord, 8 — 12 fath.; in Bakka Fjord, 2 — 4 fath.; in Faskruds Fjord, 20 — 50 fath.; in Breiddals Vig, 6 fath.; at Djupi- vogr, 8 fath.; 9 miles of the coast, 38 fath.; finally south of Iceland at Lat. btfif N., Long. i4°44' W., 44 fath. — At the Fseroes it has been secured by Dr. Th. Mortensen twice: north-west of Kalso, 60 fath., and off Boro, 20 — 30 fath. Distribution. Taken a few times in Kattegat and Skager Rak, going down to 70 fath. (H. J. Hansen); off the west coast of Norway and "along the whole Finmark coast, in moderate depths" (G. O. Sars). Recorded from Cape Flora, Franz-Joseph Land, 30 fath. (T. Scott), from the south coast of Novaya Zemlya, 48 fath. (Stappers), and from Advent Bay, Spitzbergen (G. O. Sars). Besides from the North Sea, 34 fath. (Zirwas), from a few places at the west and south coasts of England (Norman), and the west coast of Ireland, 6—7 fath. (Tattersall). — According to Harger's description of the anten- nulse the animals from the east coast of New England referred by him to M. Fabricii cannot be this species, and as Richardson in the Monograph copied the text and figures of M. Fabricii G. O. Sars, but among the localities has most of those noted by Harger, the statements are of no value. 23. Munna acanthifera n. sp. (PI. Ill, figs. 7 a— 7 h.) Description. As to general shape of the body and length of the legs somewhat similar to M. limicola G. O. S. The upper surface of the body has always a number of hairs, and in some spec- imens the clothing is somewhat dense. — The front margin of the head armed with some very robust, horizontal spines, generally 4 (fig. 7 c), but sometimes 2, 3 (fig. 7 el or 5 (fig. 7 a). Eyes completely wanting, but the eye-stalks are very conspicuous and much varying as to length (figs. 7 a and 7 e), being from moderately short to extremely long, always conical, acute and curved a little or somewhat forwards. Antennulse (figs. 7 a and 7 f) nearly as in M. limicola; the basal joint somewhat large, but longer than broad; second joint somewhat long and the third short; flagellum 5-jointed, with 3 of the joints long. The thoracic segments vary much in armature. Bach segment has most frequently at the CRUSTACEA MALACOSTRACA 111. ^i dorsal median line either a pair of spines or three or four spines, either in a group (fig. ; a oi in a transverse line (fig. 7 b); these spines arc sometimes short or moderately short, sometimes long 01 very long. Besides, the three or four anterior segments have the lateral margins at the insertion of the coxae furnished with spines, from one to three or four at each margin (fig. ; .1 . Frequentl; some of the spines or most of them, especially those at the middle of the segments, tningly wanting, but they may have been broken off. In a male from the "Ingolf" Stat. 89 each segment has only one, but very long, dorsal spine, and a similar spine is found on the head. The coxae have also spines, from one to four or five (figs. 7 a, 7 b and 7 e); sometimes one of these coxal spines or one of the supercoxal spines is very long. — First pair of legs normal, and similar in both sexes. Abdomen (figs. 7 b and 7 d) oblong-ovate, about or somewhat less than half as long again as broad, with or without a single spine at each lateral margin, but somewhat more upwards a single pair, and besides two or three or, rarely, four pairs of dorsal spines; these spines vary much as to length and thickness, but the posterior dorsal pair are generally thick. The posterior margin of the abdomen is rounded; the uropods about as in M. Boeckii. - - The median lamella of the male operculum 7g — 7 h) without ventral spines, two and a half times as long as broad, broadest near the base and tapering to somewhat beyond the middle, while its distal part has the margins nearly parallel and the end in the main truncate; the hind margin of each pleopod is slightly sinuate, and at the outer margin with a minute tooth directed backwards. Length of a female 3-1 mm., of a male 2'8 mm. Remarks. It is evident from the description that this species is very variable as to length and curvature of the eye-stalks, and in number and length of the spines on the body. (The specimens figured have the spines moderately strongly developed.) The specimens vary from station to station, and, besides, sometimes considerably from the same station. But it must be emphasized that the vari- ation seems to be independent of the bottom temperature, and that it is quite impossible to separate the specimens from the cold area as another species. The antennuhe are uncommonly uniform in adult specimens, but younger specimens have only 2 long joints instead of 3 in the flagellum, and second peduncular joint is shorter than in the adults. Especiallv by the curious shape of the blind eye-stalks .)/. acanthi/era is instantly separated from all other arctic or European species, but it agrees in this feature with .)/. truncate Richardson (1908) from the North-West Atlantic. According to the description and figures published American authoress, M. truncata cannot be identical with M. acanthi/era, as it differs in three tea viz.: the surface of M. truncata is smooth, the end of abdomen uncommonly broad and truncate, and a higher number of joints is found in the antennuhe. The antennuhe of M. truncata contain according to description and figure 8 joints, but the minute terminal joint has certainly been overlooked, a eighth joint on the figure is as long as the seventh; the result is that the antennuhe contain 9 join Furthermore the authoress referred only 2 joints to the peduncle, but this always comprises joints; consequently the flagellum contains 6 joints, the first and the terminal short, while the remaining four joints are longer. But I never found more than 3 longer joints in the flagellum of .1/. acan In the marsupium of a female from Stat. 105 was found a new species of the genus Spharonella, belonging to the parasitic Copepoda. 6 The ingolt-Expedition. III. >. 42 CRUSTACEA MALACOSTRACA. III. Occurrence. Taken by the "Ingolf" at eight stations in the warm and three in the cold area. Davis Strait: Stat. 32: Lat 66°35' N., Long. 56°38' W., 318 fath., temp. 3-9°; 16 spec. Stat. 35: Lat. 65°i6' N., Long. 55°05° W., 362 fath., temp. 3-6°; numerous spec. Stat. 27: Lat. 64°54' N., Long. 55°io' W., 393 fath., temp. 3-8°; 2 spec. Stat. 25: Lat. 63°3o' N., Long. 54°25° W., 582 fath., temp. 3-3°; 10 spec. Stat. 24: Lat. 63°o6' N., Long. 56°oo' W., 1199 fath., temp. 2-4°; ab. 28 spec. West of Iceland: Stat. 89: Lat. 64°45' N., Long. 27°2o' W., 310 fath., temp. 840; 1 spec. South-West of Iceland: Stat. 81: Lat. 6i°44' N., Long. 27°oo' W., 485 fath., temp. 61°; 1 spec. Stat. 78: Lat. 6o°37' N., Long. 27°52' W., 799 fath., temp. 4-5°; 7 spec. East of Iceland: Stat. 105: Lat. 65°34' N., Long. 7°3i'W., 762 fath., temp. -f- o-8°; 2 spec. North of Iceland: Stat. 126: Lat. 67°i9' N., Long. I5°52' W., 293 fath., temp. -=- 0-5°; 8 spec. Stat. 124: Lat. 67°4o' N., Long. i5°4o' W., 495 fath., temp. -4- o-6°; 1 spec. Besides it has been taken by the "Thor" at a single locality. .South of Iceland: Lat. 62°n' N., Long. i9°36' W., 1010 — 1140 fath.; 3 spec. Pleurogonium G. o. Sars. Sars' account of this genus is excellent so far as it goes. He described three Norwegian species; in the present paper six species are mentioned, three of which are new. It is necessary to examine the origin of the lateral thoracic processes found in the majority of the species. In P. rubicundum G. O. S. and P. intermedium 11. sp. these lateral processes originate onlv from the coxse of the legs and are epimeral processes, while in P. spinosissimum G. O. S. some of them originate, as already observed by Sars in this species, from the segments and the others from the coxae, while in /'. pulchrum n. sp. only the segments have processes and teeth. The males are much smaller than the full-grown females, as second to fourth thoracic segments are much less expanded laterally, and these segments are, besides, conspicuously shorter in the males than in subadult or adult females; the general aspect of the thorax differs therefore considerably in the two sexes. Sars has not figured any male, but my fig. 9a of the male of P. latimanum may convey an idea of the general aspect of this sex. The first pair of thoracic legs are completely similar in both sexes. The abdomen seems frequently, but not always, to be proportionately a little broader in the male than in the female. The uropods are always dorsal, with the peduncle scarcely or not distinguishable, and the inner ramus at most about half as long as, and much thinner than, the other. The abdominal operculum in both sexes has been well figured by .Sars, and in both cases affords generic characters. Conspectus of the Species. A. No lateral processes on the thoracic segments. a. No epimeral processes, at most a low tubercular protuberance 011 the coxa: of the four anterior pairs. «. Hand of first pair of legs oblong-oval witli the lower margin arcuate; terminal margin of the preceding, the fifth, joint not longer than seventh joint with claw 1. /'. inerme G. O. S. CR1 STA< i \ M iLACOSTRACA Mi. 47 £. Hand of first pair of legs with the margins subparalle] to the truncate end; terminal margin of the preceding joint conspicuously longer than seventh joint with claw 2. /'. latimanum 11. sp. b. Coxae of at least second to fourth pairs of legs with conspicuous proces a. Processes on the coxae of second to fourth pairs of legs in the female somewhat short, acute; in both sexes no processes or tubercles on the three posterior pairs of coxae. Suit.. ahdomen not distinctly scaly 3. P. intermedium 11. sp. ft. Processes on the coxae of second to fourth pairs of legs in the female rather long, sub- cylindrical with the end obtuse; in both sexes distinct short processes or protuberanci at least fifth and sixth pairs of coxae. Surface of abdomen of a most conspicuously seals appearance 4. P. rubicundum G. 0 S. A. Conspicuous lateral processes on at least the four anterior segments. a. Well developed processes on all seven pairs of coxae; no lateral processes on the three posterioi segments 5. J\ spinosissimum < '.. < 1. S. h. No processes on the coxae; lateral processes on the three posterior segments. (Conspienon- processes in the dorsal median line of the thorax) 6. P. pule lint hi n. sp. 24. Pleurogonium inerme GO. Sars. (PL III, figs. 8 a— 8 b.) 1883. Pleurogonium inerme G. O. Sars, Forh. Vid. Selsk. i Christiania I". 1882, No. iS, p. 67; PI. II, tig. 5. ! 1897. — G. O. Sars, Account, II, p. 114; PI. 48, fig. 1. The males are smaller and especially much narrower than full-grown females. In both -< the coxae of the four anterior pairs of legs (fig. 8 a) have a low tubercular protuberance on the margin, while on the coxa; of the three posterior pairs at most vestiges of tubercles are found. The 3-jointed antennular flagella are scarcely as long as the sum of the two distal peduncular joints, hut the length ol the two distal flagellar joints varies considerably, each of them being either scarcely or considerably longer than third peduncular joint, and generally longer in the male than in the female, but not long as in the male of P. latimanum. First pair of legs (fig. 8 b) have their major distal part more slender than in P. latimam P. intermedium or P. rubicundum; fourth joint is somewhat longer than broad; sixth joint is oblong- ovate with the lower margin arcuate; seventh joint has two teeth, and the sum of this joint and claw is not shorter than the distal breadth of fifth joint. -- Surface of abdomen without distinct appearance. Length of a large female with marsupium 2 nun., of a male 15 mm. Occurrence. Not taken by the "Ingolf". It has been gathered in two places at the east (-oast of Iceland, viz. by Dr. A. C. Johansen in Bakka Fjord, 25—32 fath., 1 spec, and 43—52 fath., 1 spec, and by Mag. R. Honing in Faskruds Fjord, 20—50 fath., 3 spec — Furthermore taken by the Ilnd Amdrup Exped. at Jan Mayen, 50—60 tatli large number of spec. Distribution. Taken in the Sound off Laudskrona, 17-28 fath. (Bjorck); at Denmark in 44 CRUSTACEA MALACOSTRACA. III. the Kattegat, 17' a fath., and in Skager Rak, no fath. (H.J. Hansen); in several places on the west coast of Norway, 60 to 150 fath. (G. O. Sars); in the North Sea (Zirwas); recorded besides from some places at both sides of Scotland (Robertson; T. Scott), off Northumberland, 25 fath. (Norman) and off the west coast of Ireland, 199 fath. (Tattersall); T. Scott recorded it from Cape Flora, Franz-Joseph Land, 30 fath. 25. Pleurogonium latimanum n. sp. (PL III, figs. 9 a— 9 d.) Male. Closely allied to P. inerme. — The anteunular flagella considerably longer than the sum of the two distal peduncular joints (fig. 9 b); the two distal joints in the flagella longer than in P. inerme. The four anterior pairs of coxae partly with and partly without a vestige of a marginal tubercle. First pair of legs (fig. 9 c) differ considerably from those in P. inerme; fourth joint is much broader than long, the fifth extremely broad; sixth joint has the lower margin in the main parallel with the upper, somewhat sinuate and constituting an angle of about 100° with the nearly trans- verse terminal margin: seventh joint with a single process, and together with the claw somewhat shorter than the breadth of sixth joint. — Surface of abdomen not scaly. Length of the single apparently adult male ro mm. Remarks. P. latimanum is distinguished from P. inerme mainly by the first pair of legs, which distally are broader and especially have the sixth joint quite differently shaped; it may be added that in P. inerme I have not found any variation worth mentioning in the shape of fourth to seventh joint of these legs. I must therefore consider P. latimanum to be a valid species. Occurrence. Not taken by the "Ingolf", but by Th. Holm during the cruize of the "Fylla" in 1S84, at a single place. Davis Strait: Lat. 66°32' N., Long. 55°34' W., 100 fath., stones with Hydroids, 1 spec. 26. Pleurogonium intermedium n. sp. (PI. Ill, figs, ioa-iod.) Description. Intermediate between /J. inerme and /'. rubicundum. It differs from /'. inerme in having conspicuous processes on second to fourth pairs of coxse; these processes are in the female (fig. 10 a) oblong, conical and acute, much shorter than the differently shaped processes in the female P. rubicundum] in the male (fig. 10 b) these processes are at least sometimes distinctly shorter than in the female and less acute; the processes of first pair of coxse are in the female obtuse and much smaller than the three following pairs. The coxae of the three posterior pairs of legs are without any vestige of processes, thus about as in /'. inerme. — Anteunular flagella nearly as in P. inerme. — First pair of legs (fig. 10 c) with fourth joint conspicuously broader than long, thus a good deal thicker than in /'. inerme\ fifth joint is also a little larger, while sixth joint and seventh joint with claw are in the main as in /'. inerme. - Abdomen (fig. 10 d) more oblong than in the two species mentioned; its surface at most with vestiges of some scales. Length of an ovigerous female 1*3 mm., of a male 1 mm. Remarks. /'. intermedium is certainly a valid species, easily separated from allied forms by CRUSTACKA MAI,AC(>STKACA. III. 4c the characters pointed out. Furthermore it is smaller than P. inerme, and is taken only in water near to or below zero and in more considerable depths. Occurrence. The "Ingolf" lias captured this species at three stations, two among them in the cold area. North of Iceland: Stat. 128: Lat. 66°5o' N., Long. 20°02' W., 194 lath., temp. o*6°; 2 spec. — - — Stat. 126: Lat. 67°io,' N., Long. I5°52' W., 293 fath., temp, -h 0-5°; 3 spec. North-West of the Fseroes: Stat. 138: Lat. 63°26' X., Long, fcfi' W., 471 fath., temp, -f- 060; 40 spec 27. Pleurogonium rubicundum ( '.. (). Sars. (PI. Ill, figs, n a— 11 b.) 1864. Pleuracantha rubicunda G. O. Sars, Forh. Vid. Selsk. i Christiania f. 1863, p. 220. ! 1897. Pleurogonium rubicundum G. O. Sars, Account, II, p. 113; PI. 47, fig. 2. Sars has published an excellent figure and a good description of the female, but some notes on both sexes must be added. — In the female the processes on the coxae of second to fourth pairs are long, subcylindrical with the end rounded, sometimes (fig. 11 a) even longer than in the animal figured by Sars; the processes on first pair are conspicuously shorter and thicker than on the three following pairs; the coxae of fifth to seventh pairs are adorned with small, distally rounded, conspicuous pro- cesses, generally a little broader than long, and those on seventh coxae are smaller than tin- two other pairs. In the male the processes on second to fourth pairs of coxae are obtuse, not cylindrical, broader, but scarcely longer than in the male of P. intermedium (comp. fig. 10 hi: the processes on fifth and sixth pairs are smaller than in the female, and on seventh pair rudimentary. As to the antennulae and first pair of legs the figures of Sars are sufficient. - The abdomen is somewhat broad, especially in the male, and its surface, excepting on its most posterior part, has tin- appearance of being covered with a very large number of small, imbricate scales (fig. lib); in reality scales do not exist, but the scaly aspect is due to fine, sharp, impressed lines limiting small areas. Remarks. Both sexes are easily distinguished from the preceding forms by the seals aspect of the abdomen, and by possessing small processes on fifth and sixth pairs of coxae; the females are besides separated bv the long, subcylindrical and obtuse processes on second to fourth pairs of coxae. Occurrence. Not taken by the "Ingolf. But in 1899 it was captured by Dr. Th. Mortenseu at the Faeroes at Klaksvig, 10 — 15 fath., 3 spec. Distribution. It has been recorded from Kiel in the Baltic (Apstein), from off Landskrona and further northwards in the Sound, 12—28 fath. (Bjork), and in south-western Kattegat. 15 fath Hansen). It occurs along the whole coast of Norway from Christiania Fjord to Vadso, 6—30 fath. (G. O. Sars); finally it has been recorded from a few places at both sides of Scotland (Norman, T. Scott), in the North Sea off the northern coast of England (Normani, and at the east and west coas land, 6 to 37 fath. (Tattersall). 46 CRUSTACEA MALACOSTRACA. III. 28. Pleurogonium spinosissimum G. O. Sars. 1866. Pleuracantha spinosissima G. O. Sars, Beretning 0111 en Sommeren 1865 foretagen zool. Reise ved Kysterne af Christianias og Christianssands Stifter, p. 30. ! 1897. Phitrogonium spinosissimum G. O. Sars, Account, II, p. 115; PI. 48, fig. 2. Sars' representation of the female is very fine. The male, which he has not mentioned, is much smaller than the adult female, with second to fourth thoracic segments considerably shorter, the processes from the postero-lateral angles of second and third segments much shorter than in the female, and not as long as broad, while the fourth segment has its lateral margins without any anterior or posterior angle; all coxal processes nearly as in the female. Young females of the same size as the males agree with them as to the processes of the segments. Length of an ovigerous female 25 mm., of a male from the same locality (Jan Mayen) 1-5 mm. Sars has recorded 3 mm. as the length of the adult female. Remarks. In the marsupia of a good number of specimens from Iceland I found a species of the family Choniostomatidse. Occurrence. Taken by the "Ingolf" at two stations. North of Iceland: Stat. 128: Lat. 66°5o' N., Long. 20°02' W., 194 fath., temp. o-6°; 1 spec. Stat. 126: Lat. 67°ig' N., Long. i5°32' W., 293 fath., temp. -4- 0-5°; 1 spec. The species has never been found at Greenland, but is not uncommon at Iceland. At West Iceland it has been taken by Mag. W. Lundbeck in Gnundar Fjord, 11 — 12 fath., 22 spec, and Sars recorded it from Reykjavik. At the east coast of Iceland it has been gathered by Mag. R. Horring and Dr. A. C. Johansen in Bakka Fjord, 43—52 fath., and 13 — 15 fath., Seydis Fjord, 5—9 fath., Nord Fjord, 40 fath., Faskruds Fjord, 20—50 fath., and Breiddals Vik, 6 fath. The Ilnd Amdrup Exp. secured a good number of specimens at Jan Mayen, and at the Fseroes it has been collected by Dr. Th. Mor- tensen at Klaksvig, 10 — 15 fath. Distribution. Taken at Denmark in the northern half of the Sound (W. Bjorck) and Store Belt, in Kattegat and Skager Rak, 16 and no fath. (H. J. Hansen); at the coasts of Norway from Christiania Fjord to Vardo, 60 to 100 fath. (G. O. Sars), and at Cape Flora, Franz-Joseph Land, 30 fath. (T. Scott). Finally recorded from the Firth of Clyde (Robertson) and east of Northumberland (Norman). 29. Pleurogonium pulchrum n. sp. (PI. Ill, figs. 12 a — 12 e.) Female (without marsupium). Surface of the head with two pairs of minute denticles in a transverse row (fig. 12 b). - - First joint of the antennuhe with a small spine at the middle of the front margin and a somewhat long spiniform process a little from the end; flagellura 3-jointed, as in all preceding species. — Third joint of the antennae with a very conspicuous, spiniform process vertical on its outer margin a little beyond the middle, and a smaller process on the opposite margin. Bach oi the four anterior thoracic segments (fig. 12 a) has the lateral margins adorned witli a very long, slender and acute process, and besides with two to five very much smaller processes. ( >n first segment the large process projects from the autero lateral angle, and behind it two or three acute CRUSTACEA MALACOSTRACA. HI. denticles; on the three following segments the lateral margin is augularl) bent, and the long p projects from that angle; on third and fourth segments, but not on the second, a ver) small process is found in front of the large process, and behind the latter process the margin has three or four small processes on second and third segments, but only two on the fourth (fig. 12 c). Fifth and sixth ments each with a very long lateral process and a very small process in front of its base, but behind it the rounded coxa is seen; seventh segment only with a rather short lateral process. — First segment has a median dorsal process, which is extremely long, somewhat thick, denticulate and directed up- wards and a little backwards (a large portion of the process hast been lost in my specimen); each "i the other segments, excepting the fifth, has also a median process, hut the six dorsal processes deen very much in size backwards, so that the process on sixth segment is rather short; fifth segment has a minute median denticle. First pair of legs (fig. 12 d) robust, but deviating somewhat from those in the preceding forms: fifth joint is nearly longer than deep, with three teeth and two long articulated spines on its lower margin; sixth joint a little more than twice as long as deep; with two somewhat long spines on the prehensile margin; seventh joint with claw not much shorter than sixth joint. Abdomen — not including its basal segment — is not much longer than broad dig. 12 e). Each lateral margin from somewhat from the base to the uropods armed with 12 moderatelj l. spinosim besides, a number of a still more curious species, so that a somewhat more in! can be given. The Ingolf-Kxpedition. III. J. CRUSTACEA MALACOSTRACA. III. 31. Dendrotion spinosum G. O. Sars. (PI. IV, figs. 2 a — 2 c.l 1872. Dendrotion spinosum G. O. Sars, Forh. Vid. Selsk. Christiania for 1871, p. 273. ! 1897. G. O. Sars, Account, II, p. 116; PL 49. Four specimens are to hand, viz. three full-grown females, one of them with marsupium, and a probably adult male. The material, though rather mutilated, enables me to give some additions and corrections to Sars' description and figures, and, besides, to point out characters of specific value. The processes bearing the antennulse and antennae are in my specimens longer than shown by .Sars, being more than twice as long as broad (fig. 2 al. The antennulae are somewhat less elongate than in the next species; the basal joint is very long, but yet, according to Sars' figure, not half as long as the last peduncular joint of the antennae; the flagellum is 9-jointed in a full-grown female. — The antennae have the peduncle 6-jointed, but the first joint is short and not visible from above, being overlapped by the basal part of the antennula; according to Sars' figure second peduncular joint is not fully one-third as long as the fifth; fifth and sixth joints are extremely long, the sixth longer than the fifth and somewhat longer than the flagellum, which consists of 16 joints. The thoracic segments show features of interest. The first segment at each side feebly pro- duced into a low antero-lateral protuberance terminating in a very long, thin process (figs. 2 a — 2 b), and at the lateral part of the front margin of the segment is seen the coxa, which is produced in a rather short epimeral process [ep). Second to fourth segments each produced on the side in a somewhat thick protuberance somewhat less long than, or fully as long as, thick, and bearing the coxa, the anterior margin of which is visible from above at the end of the front margin of the process; just behind the coxa the thick protuberance terminates in a very long, slender process; the whole structure of the three segments is in the ovigerous female more similar to that found in the next species (shown in fig. 3 a) than to the figures of Sars, and in specimens without marsupium the structure is almost as on fig. 3 a, excepting that the lateral protuberances are a little shorter. The processes terminating the protuberances of the four anterior segments are all long, those on second and third segments longer than the others, but conspicuously shorter than the breadth of the head; furthermore they have no granules, are straight, slender, and distinctly tapering to near the end (fig. 2 b), which is slightly or a little thickened, cut off transversely, but with three or four minute or nearly obsolete teeth on the margin, and the end has a very long, strong seta (fig. 2 a); the first pair of processes have, besides, a similar seta on the upper side somewhat from the base, while the three other pairs possess the last- named seta and, besides, two or three setae placed more proximally on the process or at the end of the protuberance. The surface of these segments is furnished with a number of similar very long seta;, each inserted on a more or less elevated small foot; first segment has only one pair of such setae situated somewhat from the middle near the posterior margin, while second and third segments have four or five pairs placed in a row near the posterior margin and off the base of the protuberances; fourth segment has two or three pairs along the posterior margin and, besides, two pairs in a trans- verse line before the middle of the segment. The three posterior segments (fig. 2 c) differ much from the others. Bach is produced into a CR1 STACEA MALACOSTRACA. III. 5' lateral pair of nearly cylindrical protuberances about as long as, or a little longer than, the breadth of the corresponding segment, and bearing the coxa; on their ends; the coxa itself is nearly as broad the protuberance, but short and at the end produced in a slender process; the pro I fifth co are shorter than the breadth of the coxa.-, those of sixth pair are distinctly longer than, or even almost twice as long as, the preceding pair, and the processes of seventh pair of coxae are still a littl* and thinner; the processes have the end cut off and a long terminal seta, while the surface of the three segments has no long setse, but the protuberances from fifth and sixth segments have a single long dorsal seta. The processes are without granules. The abdomen of both sexes has been well figured by Sars. The four strong, curved spines on each lower margin and the shape of the opercula are characteristic. I found it impossible to point out a separate basal segment of the abdomen. Length of the female with marsupium 3-5 mm., of the female without marsupium 3*8 mm probably adult, very mutilated male has been about 2-8 mm. Remarks. Though my description of the thoracic segments with their protuberances and the processes differs materially from the representation given b\ Sars, I am sure that my specimens belong to his species, as the armature of the abdomen, the median plate of the male operculum, and an uropod preserved in a female agree completely with his figures. In the description of, and remarks 011, />. paradoxuni the differences between the two species are pointed out. Occurrence. Taken by the "Ingolf" at a single station. South-West of Iceland: Stat. 78: Lat. 6o°37' N., Long. zf& W., 799 fath., temp. 4-5 ; 1 spec. 1 Furthermore the "Thor" secured it in 1904 at the following place. South-West of the Faeroes: Lat. 6i°i5' N., Long, g^' W., 463— 515 fath.; 1 spec (?) Distribution. The Copenhagen Museum possesses two females gathered by the "Thor" Di Joh. Schmidt) in the North Sea at Lat. 58°32' N., Long. 4°i8' E.. 149 fath., thus not very far from South-Norway. Sars has recorded it from Hardanger Fjord, 150 fath. 32. Dendrotion paradoxum n. sp. (PI. IV, figs. 3 a-3 e). Description. The material comprises a number of not full-grown specimens thoracic segments consequently less expanded than in ovigerous females — and a single, perhaps adi male; all specimens have lost the major part of their appendages. This species is allied to /' rpi- iiosuiu, but it looks still more aberrant. The processes of the head are as long as, or shorter than, those in my specimens of I), spit First joint of the antennuke is extremely long; in a single specimen with an antennul well preserved (fig. 3 c) this joint is more than two-thirds as long as sixth peduncular joint of the antenna:; the flagellum is 8-jointed in this specimen. - Antennae with second joint longer spinosum, in the specimen mentioned half as long as the filth joint, which is a little shorter sixth; flagellum scarcely longer than sixth peduncular joint, 8-jointed. The four anterior thoracic segments (fig. 3 a) with the lateral protuberances a little longer than in D. spinosum, but their terminal processes are much longer than in that species, thos ond and 52 CRUSTACEA MALACOSTRACA. III. third segments distinctly longer than the others, and each considerably longer than the breadth of the head; each process is shaped almost as in D. spinosum, terminating in two to four small teeth on the truncate end (fig. 3 b), and a seta considerably shorter than in D. spinosum is inserted on or above near the end ; each process has furthermore a long dorsal seta before the middle, and those of second to fourth segments besides a seta beyond the middle and at least sometimes one or two proximal seta;. All these processes and a distal part of the protuberances are adorned with a number of sharp gra- nules (fig. 3 b). -■ These segments have very long dorsal setse, each seta inserted on an elevated foot, but each of the three anterior segments has only a single pair, placed near the posterior margin a little from the middle (fig. 3 a), while the fourth segment has an anterior and a posterior pair, in the large male even besides two pairs near the lateral margin. The three posterior segments have their lateral protuberances nearly longer than in D. spi- nosum] the processes on fifth pair of coxae are slightly shorter than those on fourth segments (fig. 3 a), but a little longer than those on sixth pair, and about three times as long as the processes on seventh pair of coxa;; all these processes are granulated like the anterior pairs. Each of these segments has a single pair of dorsal long hairs. The thoracic legs, as far as is known (fig. 3 a), seem to be somewhat longer than in D. spino- sum; the prehensile hand (fig. 3 d) is a little more slender than in that species. The abdomen shaped as in D. spinosum, generally with 3, in the large male with 4 pairs of marginal spines. The end of abdomen of one specimen has five minute teeth (fig. 3 e), but in other specimens the end has only a single tooth or two or three obsolete, irregular teeth. I have been unable to point out any difference in the shape of the opercula in either sex between this species and I >. spin t>s it in (uropods lost). Length of the large male 2'6 mm., while the other specimens are smaller, most of them 2 — 2-5 nun Remarks. D. paradox// in is easily separated from D. spinosum by the thoracic processes, which are conspicuously set with a good number of granules quite wanting in the other species. Furthermore the processes of the four anterior segments are longer than in D. spinosum, the processes on the coxse of fifth segment almost as long as the preceding pair of processes, while in D. spinosum they are quite short; in F>. paradoxum the coxal processes decrease much in length from fifth to seventh pair, while in />. spinosum they increase in length backwards. The existence of a pair of setiferous tubercles on the three posterior segments is also a valuable character, and the relative length of the joints in the antennular and antennal peduncles affords other characters, when these appendages have- been preserved. Occurrence. Taken by the "Ingolf" at a single station. South- West of Iceland: Stat. 78: Lat. 6o°37' N., Long. 27°52' W., 799 fath., temp. 4-5; 20 spec. It may be noted that enormous quantities of sponges were procured at that Stat. 78. Among the animals found were the .specimens of I >. paradoxum, a specimen of I >. spinosum, furthermore the two specimens of the above-described Pseudomunna hystrix, specimens of the extremely long-legged Pseudotanais longipes II. J. II., and of Typhlotonais eximius H.J. II., which possesses very elongate and slender chelae. I suppose that all these animals live on sponges, and that there may be some correla- tion between their armature or very elongate appendages and their mode of living. CRUSTACEA MALACOSTRACA. Ill y Group IV. Schistosomatini. It was deemed unavoidable to establish this group on a single genus (with a sir which seems to be intermediate between other groups, and rather distant from them all. In ordei avoid reiterations no characterization of the group is given, as the description of the genus together with the remarks may be sufficient. Schistosoma u. g« Description. Head somewhat similar to that of Desmosoma\ no eyes. Anteunulae moderateh long (PI. IV, fig. 4a); the basal joint somewhat slender and very oblong. The four proximal joints of the antennae moderately short, robust (the two distal, probably strong joints of the peduncle and the flagellum lost); squama very small. — Mandibles (fig. 4 c!) somewhat slender, in the main as in the In ////-(/-group; left mandible has the incisive margin somewhat long, a slender movable lobe and a few setae; the molar process is vertical on the corpus mandibular, slightly tapering towards the end, which is cut off, with the end irregularly shaped and adorned with a few short setae; palp well developed. - Maxillipeds (fig. 4 e) in the main as in /antra; second joint very broad; its lobe large, almost as long as the joint itself, much longer than broad, with the terminal margin very oblique and showing some triangular teeth; the palp has the first joint very broad, second joint moderateh broad and longer than broad, third joint only a little more than half as broad as the second and broader than long, the two distal joints well developed; epipod somewhat small. Thorax (fig. 4 a) divided into two sections, because the fifth segment is narrowed in from kind of short stalk. All segments movable. The three anterior segments laterally produced into obi protuberances directed outwards and forwards; the fourth segment has two pairs of lateral protuberant the first pair directed somewhat forwards, the second pair somewhat backwards; the three posterior segments each with a pair of lateral protuberances, projecting outwards on the fifth segment, backwards 011 the sixth, and much backwards on the seventh segment. All these thoracic protuberances have the terminal part moderately slender, the end cut off, and the coxae are inserted at the ba- the protuberances; the coxae have no processes. First pair of legs (figs. 4 a and 4f) moderately short and very slender, scarcely prehensile, without spines; fifth joint somewhat elongate; sixth joint not thicker than the fifth; seventh joint very oblong, as long as the very slender claw; especially the terminal part of sixth joint with a number of lonj hairs. The six other pairs of legs somewhat long, slender, subsimilar; fifth and sixth joints rathe long, with very few, nearly setiform, short spines; seventh joint (fig. 4 hi very oblong, termina a somewhat slender claw and a seta. Abdomen (in the male) shows two segments; first segment is slender and somewh: second is oblong-ovate, twice as broad as the first. The male operculum occupies lower surface of the main part of the abdomen; the median lamella tapers considerably to the end. — Uropods inserted on the lateral margins rather near the end; they are somewh; uniramous, two-jointed. 54 CRUSTACEA MALACOSTRACA. III. Remarks. This new genus occupies a very isolated position. The mouth-parts are in the main as in the Zantra-gtoup, the general outline of the body somewhat resembling that in Echinopleura (of the Desmosomatini), the first pair of legs feeble, not prehensile and without spines, the other legs nearly as in Plairogoiiittm, the male operculum uncommonly large. Only one species is known. 33. Schistosoma ramosum n. sp. (PI. IV, figs. 4a-4i). Male. Body a little more than three times as long as broad. — Anteunuke a little longer than the sum of the head and the two anterior segments; first joint (fig. 4 c) twice as long as broad and as long as second joint; flagellum a little longer than the peduncle, 8-jointed. The eight pairs of thoracic protuberances are all somewhat long, with a single small seta or two short setae at the end ; fourth pair considerably longer and thicker than the fifth. — The abdominal shield almost half as long again as broad; the posterior margin between the insertions of the uropods considerably convex. The uropods less than one-fourth as long as the last abdominal segment; first joint is short, second a little thicker and clavate, thicker beyond the middle than at the base. Each pleopod of first pair with an incision in the distal end. Length of the single specimen 1-4 mm. Remarks. As it was impossible as to a number of features to separate the generic from the specific characters, most of them have been mentioned in the description of the genus. Occurrence. Taken by the "Ingolf" at a single station. South of Iceland: Stat. 64: Lat. 62°o6' N., Long. io,°oo' W., 1041 fath., temp. 3-1°; 1 spec. Group V. Ischnomesini. Body somewhat or extremely narrow, generally most narrow at the middle, which is round. Head deeply immerged within first thoracic segment and so completely fused with it that only a more or less distinct, curved dorsal impression is seen as the limit, excepting frequently at the lateral margin, where a sharp suture is found. Eyes wanting. Antennuke with first joint thick, nearly globular, and second joint proportionately rather long or extremely long. Antennae without squama, long or very long. Mandibles without pal]), but otherwise in the main similar to those in the A/ww-group, having the molar process thick, subcylindrical, transversely truncate and directed a little forwards, or luing vertical on the corpus mandibular — Thorax very peculiarly shaped, as fourth segment is pro- duced somewhat or much backwards, fifth segment much to extremely much forwards, so that the distance between the insertions of the fourth and the fifth pair of legs is from very to exceedingly long; the produced parts of these segments are considerably, or very much, narrower than their leg- bearing parts. First pair of thoracic legs prehensile, similar in both sexes; fifth joint with two or three spines; the six other pairs slender, moderately or very long; fourth and fifth pairs subequal in length, CRUSTACEA MALACOSTRACA. MI. ,- and conspicuously longer than second and seventh pairs; the coxa without any process, seventh joint thin, claw short, frequently scarcely marked off. - Uropods inserted on Hit- margin, at most moderately long, frequently reduced, small. Remarks. This group is difficult to arrange among the other types of the family, but I insert it here as being the last of the groups with the molar process belonging to the fanira-type. The group is not only deviating from other Parasellidae, but its genera show must interesting differences. As all hitherto known species, excepting one, have been somewhat imperfectly studied, it may be useful to give a more detailed account. In 1866 G. O. Sars established the genus Ischnosoma on a very peculiar form, of which he I gave a full representation in the Account Later another species was established In Sars, a third species by Tattersall, and four species from the "Challenger" by Beddard. In 190S Harriet Richardson pointed out that the excellent generic name was preoccupied (in r820. and in 18321, and instead of it was proposed Ischnomesus Rich. Furthermore the authoress distributed the species hitherto referred to Ischnosoma into four genera, thus establishing three new genera, and described two new spei Having a comparatively enormous material of the present group, viz. Ischnomesus bispinosus « ".. 1 • Sars from Skager Rak and 11 species, 10 of which new, from the "Ingolf" area. I have attempted to find valid characters for the genera and new characters for the species. In reality the genera proposed by Richardson — with the possible exception of Rhabdomesus based on two very imperfectly known species from the "Challenger" - must be accepted as natural, but the majority of the generic char- acters pointed out by Richardson in the analytical key are wrong or of no value, while other charac- ters can be pointed out. To take a few instances. Harr. Richardson said in the key that the abdomen in hchnom bispinosus Sars consists of a single segment; the authoress had not seen this species but had probably taken the character from Sars' figures, which in this respect are not distinct, in reality not accurate. According to my examination of a good material of /. bispinosus the abdomen consists of two segments, the first short but well defined in front and behind by articulations, consequently both segments movable. - Furthermore the authoress said that the genus Haplomesus^ based on Ischnosoma qnadrispino. G. O. Sars, has the "abdomen composed of a single segment", while in Heteromesus, to which five species were referred, it consists "of two segments". These statements are misleading. Later I redescribe Haplomesus quadrispinosus and establish four new species in the same minis; of the five species referred by the American authoress to Heteromesus I have examined co-types of the two species esl in her paper, and describe in the sequel four new species. And based on such rich material it can be said, that both in Haplomesus and Heteromesus the abdomen is both above and below immovably fused with seventh thoracic segment, so that only a more or less pronounced transverse furrow or impression is seen between thorax and abdomen, while a furrow between an anterior short and a posterior large abdominal segment is most frequently wanting on the major part of the dorsal surface, but towards the lateral margins. But more can be said. In Haplomesus the three posterior thoracic ments and abdomen are immovably coalesced, forming a single piece; in Heteromesus the articulation between the fifth and the sixth thoracic segment is very distinct, while sixth and seventh segment; and abdomen are immovably fused; in Ischnomestis not only has the abdomen two movable seg- cj6 CRUSTACEA MALACOSTRACA. III. ments, but all thoracic segments -- excepting the first which is fused with the head — are free and movable. Some other good characters between the three genera can be pointed out, but they are found in the brief descriptions of the genera later on. But it may be useful to give an analytical key to the three genera; the fourth genus, Rhabdomesus Rich., is still somewhat doubtful, as its two "Chal- lenger" species were established on specimens without the anterior part of the body, and judging from the figures of their abdomen the genus must be cancelled and its species referred to Ischnomesus. A. The three posterior thoracic segments movable; the abdomen consists of two movable segments. Uropods two-jointed. Third joint of the autennal peduncles moderately short, not twice as long as the fourth Ischnomesus Rich. B. At least the two posterior thoracic segments and the abdomen immovably fused, constituting a single piece. Uropods one-jointed. Third joint of the autennal peduncles elongate, much more than twice as long as the fourth. a. The three posterior thoracic segments completely fused. Body slender to extremely slender. Antennuke with third peduncular joint and the 3-jointed flagellum well developed. Haplomestis Rich. b. Only the two posterior thoracic segments and the abdomen fused, as a real articulation is found between fifth and sixth segment. Body moderately slender or proportionately robust. Antennulse with the part beyond second joint reduced, consisting of two or three small or quite rudi- mentary joints Heteromesus Rich. IschnomeSUS Richardson. Body slender or very slender. Antennulse well developed, 6-jointed. Antenna; with third joint moderately short, not twice as long as the fourth. Maxillipeds (PI. IV, fig. 5 c) with second joint some- what large, its lobe at least as long as broad, and less than half as broad as the joint; the palp at least as long as second joint itself - the lobe not included — and its second and third joints are considerably expanded, conspicuously broader than the lobe mentioned. - The posterior thoracic seg- ments movable. First pair of thoracic legs with the fifth joint much deeper than the fourth, having its proximal half much expanded, and being broadest before the middle. Abdomen consists of two movable segments; the first is very short. — Uropods two-jointed, sometimes with a quite small exopod from the end of first joint. The "Ingolf" has secured two species, both new. 34. Ischnomesus profundus n. sp. (PI. IV, figs. 5a-5f). Male. Body about six times as long as the breadth of first thoracic segment. - Antennuke somewhat longer than the breadth of first segment; third joint half as long as second and as long as the flagellum (fig. 5 b). Third joint of the autennal peduncle about half as long again as the fourth (the following joints lost). CRUSTACEA MALACOSTRACA III. 57 First thoracic segment of a somewhat quadrangular aspect, as the major part of the lateral margins are subparallel; each anterolateral corner produced in a conspicuous, but somewhat small, process projecting horizontally forwards. The other thoracic segments without processes. Fourth seg- ment almost as long as second and third combined; fifth segment almost half as long again as the fourth, and nearly four times as long as the breadth near its front end. First pair of thoracic legs (fig. 5 d) robust; fifth joint somewhat curved, much thickened with a protruding, rounded part somewhat from the base, and this part set with some spines of very dif- ferent length, while the lower margin between the top of the protuberance and the distal end is some- what concave; sixth joint slightly shorter than the fifth, scarcely three times as long as deep, without spines; seventh joint with claw somewhat strong, about half as long as sixth joint. The other pairs lost excepting a few joints; second joint of fourth pair reaches slightly beyond the posterior end of its segment, while the corresponding joint of fifth pair, when directed forwards, is rather far from reaching the end of second joint of fourth pair. Abdomen (fig. 5 e) has its first segment very short but very distinct. Second segment about one-third as long again as broad, anteriorly considerably narrower than near the end; the lateral margins are somewhat convex, scarcely concave near the postero-lateral angles, which are well defined and sharp; the posterior margin with its major part considerably convex, and feebly concave towards the angles mentioned. The ventral excavation for the reception of the pleopods (fig. 51) with its posterior margin long, transverse and straight, leaving a considerable area between itself and the posterior margin; the median lamella of the operculum small, tapering from somewhat from the base to the narrow end, which does not reach the posterior margin of second pair of pleopods. (Uropods lost; Length of the specimen 4 mm. Remarks. I. profundus is allied to / bispinosus G. 0. Sars, but is considerably longer, and differs besides in several particulars. The processes of first segment project nearly forwards, and consequently slightly diverging, while in the male I. bispinosus they are longer and considerably diver- gent (comp. Sars' Account PI. 52). Fifth thoracic segment less elongate than in the male /. bispinosus; the prehensile organ of first legs is somewhat different in shape and in number, length and place of spines; the last abdominal segment differs in having sharply pronounced postero-lateral anj Occurrence. Taken by the "Ingolf" at its deepest station. South of Davis Strait: Stat. 38: Lat. 590i2' N., Long. 5i°o5' W., 1870 fath., temp. 1-3°; 35. Ischnomesus armatus n. sp. (PI. IV, figs. 6 a— 6 f). Male. The body nearly seven times as long as the breadth of the first thoracic segment, measured without taking its processes into account. - Antennulae (fig. 6 b) reaching 1 the posterior margin of second thoracic segment; second joint somewhat robust, slightly more than half as long as the entire antennula, with two or three spines along the interior margin: thin and only one-fourth as long as second joint; first joint of the flagellum much shorter than the second or the third. Third joint of the antennal peduncle not twice as long as fourth joint; 8 The Ingolf-Expedition. III. S- r8 CRUSTACEA MALACOSTRACA. III. distal joints of the peduncle with flagellum lost in my specimens). — Maxillipeds (fig. 6 d) in the main as in I. profundus, excepting that the epipod is considerably smaller, having more than half of the outer margin concave. Each of the four anterior thoracic segments (fig. 6 a) anteriorly at the sides produced into a pair of somewhat long, subacute processes directed forwards, considerably outwards and distinctly up- wards; the anterior margin of each process is about, or scarcely, half as long as the front margin of the corresponding segment between the origin of the processes (fig. 6 b). Fifth and sixth segments each posteriorly produced into a pair of similar processes directed backwards and somewhat outwards, but not upwards; the posterior pair of these processes are conspicuously longer than any of the other pairs. Fourth segment much produced backwards, being somewhat longer than second and third seg- ments combined; fifth segment distinctly less than half as long again as the fourth, moderately slender, slightly more than four times as long as its breadth near the front end. First pair of thoracic legs (fig. 6 e) more slender than in I. profundus; fifth joint more than twice as long as deep, with a protruding angle on the lower side somewhat from the base; this angle bears a long spine, and a still longer spine is somewhat more proximal, while the major, distal part of the lower margin is distinctly concave without spines; sixth joint is somewhat shorter than the fifth, more than three times as long as deep, without spines, and twice as long as seventh joint with claw. The other legs nearly all lost excepting the proximal joints; second joint of fourth pair reaches scarcely to the end of its segment, and is rather far from reaching the end of second joint of fifth leg. Abdomen with the first segment very short but very distinct (fig. 6 a). Second segment scarcely one-third as long again as broad, scarcely narrower at the anterior than at the posterior margin; the lateral margins are considerably convex excepting a little from the end, where they bend backwards and a little outwards (fig-. 6 f); the posterior margin is somewhat convex, with the uropods at its ends. The ventral excavation occupies nearly the whole surface excepting a moderately large posterior por- tion, and its hind margin is very convex. — The median lamella of the operculum is very large, distinctly narrowed near the base and then gradually a little expanded, with the lateral margins sub- parallel from before the middle to near the broadly rounded end of each pleopod; the lateral plates of the operculum unusually narrow. — The uropods scarcely half as long as the breadth of abdomen, two-jointed, and first joint not half as long as the second; no exopod. Length of the male 4-8 mm. Remarks. This fine species is abundantly distinguished from all other species of the group by the number and development of thoracic spines. In the outline of the abdomen and especially in the shape and relative size of the opercular plates it differs much from the two other species of the genus. But at least at the present state of our knowledge these differences are far from sufficient for establishing a new genus on / armatus. Occurrence. Taken by the "Ingolf" at a single station. Davis Strait: Stat. 36: Lat. 6i°5o' N., Long. 56°2i' W., 1435 fath., temp. 1-5°; 2 males. CRUSTACEA MALACOSTRACA. III. eg Haplomesus Richardson. Body slender or extremely slender. Antennulae well developed, 6-jointed. Antenna: with third joint elongate, two and a half times or more as long as the fourth joint Maxillipeds (fig. i It with second joint extremely large, the lobe more than half as broad as the joint and much shortei than broad; the palp not two-thirds as long as second joint itself -■ the lobe not includ< and it and third joints of very moderate size and narrower than the lobe of second joint, though the) much expanded, each conspicuously broader than long. -- The three posterior thoracic segments and abdomen immovably coalesced, constituting an extremely long, stiff part First pair of thoracic li with fifth joint a little less deep than the fourth, not expanded and of normal shape. — Uropods one- jointed, short. Remarks. This pretty genus has been established on /. quadrispinosus G. 0. Sars. The ma- terial contains this form and besides 4 new species, all deep-sea tonus. Their exoskeletoii is extremely brittle, and numerous specimens are therefore very mutilated. The colour of the animals generally whitish or pure white, but several of the specimens, especially females, of //. quadrispinosus are greyish. 36. Haplomesus quadrispinosus ( \. ( ). Sars. (PI. V, figs. 1 a— 1 p). 1879. Ischnosomd, quadrispinosum G. O. Sars, Arch. f. Math, og Naturv. Vol. IV, p. 435. ! 1885. — G. O. Sars, North-Atl. Exp. I, p. 126; PI. 11, figs. 26-29. Male. Body (fig. 1 a) extremely slender, from about eight to nearly eight and a half tim< long as broad across the first thoracic segment — the processes not included in the breadth. The integument of the whole body, excepting the thoracic processes, closely set with very fine, sharp granules, and at the lateral margins of the posterior part of the thorax and of the abdomen thi granules are a little larger (fig. 1 h), triangular. Antennulae, when directed backwards, reach beyond the middle of fourth thoracic segment: second joint (fig. 1 b) is extremely long, conspicuously more than half as long again as the four distal joints combined; third joint nearly as long as the flagellum. -- Antennre very long and slendi little shorter than the thorax; third peduncular joint long, but yet considerably shorter than second antennular joint; fifth peduncular joint subcylindrical; sixth joint very slender, but yet distinctly in- creasing in thickness from base to end, and a little longer than or about as long as the fifth ; flagellum very thin, somewhat or considerably longer than sixth peduncular joint, with about 18- Thorax on first segment with a pair of lateral processes, which are very long, from a 1 considerably longer than the breadth of the segment, moderately slender, tapering, a wards, much forwards and (fig. id) very considerably upwards; third segment has a pair of processes half or less than half as long as the first pair and frequently directed more outwards, but otherwise rather similar. (In the later "remarks" some variation in the processes segment a little longer than second and third segments combined. Fifth segment even slightly more than half of the entire animal, in adult specimens nine to nearly ten times as long as broad somewhat before its middle, subcylindrical to the beginning of the short leg-bea 6o CRUSTACEA MALACOSTRACA. III. which is twice as broad as most of the segment, and slightly or somewhat narrower than the anterior segments. First pair of legs (fig. i g) slender; fifth joint more than three times as long as deep, with two very long, nearly setiform spines and more distally a somewhat short, thin spine on the lower margin; sixth joint somewhat shorter and a little thinner than the fifth, with two somewhat short spines beyond the middle of the lower margin. The six other pairs of legs very long and very slender, without granulations, but with some very short and thin spines on fifth and sixth joints; second joint of fourth pair reaches, when stretched backwards, scarcely to the end of the corresponding joint of fifth pair. Abdomen (fig. i h) about as long as broad or slightly longer than broad, and about as broad as the leg-bearing part of fifth thoracic segment; the major part of each lateral margin is rather convex and, besides, serrate from triangular granules; posteriorly, where this serration terminates, a low emar- gination is found, aud the posterior margin between the uropods is moderately protruding and nearly semicircular. — The median lamella of the operculum, when seen in its natural position (fig. i i), tapers much from somewhat from the base to near the narrow end, but removed from the animal it is seen (fig. i k) that a rather broad part of more than the distal half of the lamella has been overlapped by the two lateral plates of the operculum. The copulatory process is short (fig. i 1), not reaching the end of its pleopod. — The uropods are triangular, very short, not reaching beyond the end of abdomen. Length of the three largest males (from Stat. 36, 38 and 116) 5 mm. Female. Bod}- (fig. 1 111) anteriorly much stouter than in the male, about five and a half times as long as broad across the first thoracic segment, the processes not counted. The integument as in the male. - ■ The antennulse (fig. 1 n) when directed backwards reaching about to the posterior margin of third segment; second joint only half as long again as the four distal joints combined. — Antenna; about as long as the thorax; sixth peduncular joint cylindrical, very slender and somewhat longer than the fifth; flagellum almost as long as the two distal peduncular joints combined, 17-jointed. The thorax with the same two pairs of processes as the male, but the first pair are much smaller (fig. 1 n), less than half as long as the breadth of its segment, while the second pair are more than half as long as first pair. The four anterior segments are proportionately considerably broader than in the male, while the fifth segment is less elongate, five or six times as long as its shortest breadth, and with the leg-bearing terminal part considerably narrower than the three anterior segments. — The legs about as in the male. The abdomen (fig. 1 o) conspicuously longer than broad, and narrower than the terminal part ' N., Long. 4°n' E., fath., temp, -f- [-4° (G. O. Sars). 37. Haplomesus angustus n. sp. (PI. V, figs. 2 a-2e). Youne- Male. Extremely slender, between ten and nine and a half times as long breadth of the anterior segments. The granulation on the body finer than in //. quadrispii Antenuula. (fig. 2b) much shorter than in the preceding species, when stretched backwards not reach- ing the posterior margin of second segment. Second joint scarcely as long as the distal joints con.' third joint only about as long as the fourth, but in this young specimen the flagellum conta two joints, as the distal joint has not yet been divided into two joints. Antenna, lost excepting th. four proximal joints of the peduncle; third joint about as in //. qnadrispinosus, nearly as long as thoracic segment with head. 62 CRUSTACEA MALACOSTRACA. III. Thorax with a pair of lateral processes on first segment and none on the others. These pro- cesses (figs. 2 a and 2 b) are very much shorter than in H. quadrispinosus, being of moderate length and thickness, each conspicuously less than half as long as the breadth of the segment, directed much upwards and outwards and somewhat forwards. The three following segments without any vestige of processes. Fourth segment about as in // qttadrtspinosus. Fifth segment extremely elongate, slightly more than half as long as the entire animal, and besides extremely slender, being about fourteen times as long as broad somewhat before the middle; its widened terminal part a little narrower than second segment. Seventh segment in my specimen unusually small (fig. 2d) and without legs (fig. 2e); these features show that the specimen must be rather far from adult. First pair of thoracic legs (fig. 2 c) in the main as in II. quadrispinosus, but fifth joint has a single very long, thin spine and two short spines, and sixth joint a single spine on the lower margin. The other legs, as far as preserved, nearly as in H. quadrispinosus) second joints of fourth and fifth pairs, when directed respectively backwards and forwards, do not quite reach one another. Abdomen (figs. 2 a, 2 d and 2 e) triangular in aspect, because, having the lateral margins con- spicuously sinuate, it increases somewhat irregularly in breadth from the base to the posterior margin ; each postero-lateral angle is produced in a conical process directed outwards and backwards, slightly longer than the uropods, conical with the end obtuse. The posterior margin between these processes is, taken as a whole, moderately convex, but a little more than its median third protrudes further backwards and has its margin semicircular; this protruding part is the terminal portion of the rather broad longitudinal median area of the surface (fig. 2 d). — The median lamella of the operculum (fig. 2 e) is towards the base distinctly narrower than in H. quadrispinosus, but otherwise of the same shape. — The uropods about twice as long as in that species. Length of the specimen described 4-8 mm. Remarks. This extremely slender species is instantly separated from the other forms by having only a moderately large pair of processes on first segment and none on the other thoracic segments; besides, the shape of abdomen affords excellent characters, as it differs much from the other forms excepting II. insignis. As the specimen described has seventh pair of thoracic legs totally wanting it must be very far from adult; I suppose that really full-grown males may measure 9 — 12 mm. in length. The type has been taken in the warm area. From a station in the cold area I have a male, which unfortunately is very mutilated, as two anterior segments with head are lost; this specimen is somewhat smaller, with fifth segment, as consequently might be expected, distinctly less slender, only ten times as long as its breadth before the middle, but in all other respects, especially the small seventh segment and the shape of the abdomen, this specimen agrees so completely with the type, that I must refer it to this species. As H. quadrispinosus has been taken some times in both areas, the occurrence of II. august us is less astonishing. Occurrence. Taken by the "Ingolf" at two stations. South of Denmark Strait: Stat. 18: Lat. 6i°44' N., Long. 3002o/ W., 1135 fath., temp. 3-0°; 1 spec. (o*, type). North of Iceland: Stat. 125: Lat. 68°o8' N., Long. i6°02' W., 729 fath., temp. -^o-8°; 1 spec, (mu- tilated o*). CRUSTACEA MALACOSTRACA. III. 63 38. Haplomesus insignis 11. sp. (PL W figs.3a-.3d). Male. Very slender, about seven times as long as the breadth of the anterior segments. Body very finely granulated, about as in //. angushis. Antennulae somewhat long (fig. 3 b), when stretched backwards reaching beyond the ; margin of third thoracic segment; second joint a little shorter than the head and first segment com- bined, and a little longer than the sum of the four distal joints; third joint conspicuously thicker and slightly longer than the flagellum. — Antennae (fig. 3 a) somewhat more than half as long as the bod) ; third peduncular joint conspicuously shorter than in the two preceding species, somewhat shorter than second joint of the antennulae; fifth joint increases conspicuously in thickness from the base to some- what from the end; sixth joint is somewhat longer than the fifth and increases somewhat in thicki; from the base to the middle; flagellum about as long as sixth peduncular joint, 18-jointed, increasing distinctly in thickness from the base to the middle, and tapering then to the thin end. The thorax has two pairs of very large, lanceolate processes (fig. 3 a) projecting respectively from first and fourth segment, while the sides of second segment (fig. 3 b) are produced in very short processes with the end obtuse. The processes of first segment are directed outwards, much forwai and rather feebly upwards (fig. 3 b), and their tips reach a little in advance of the front end of the head; they are proportionately broad, four times as long as broad, lanceolate, subacute and somewhat depressed. The processes of fourth segment are slightly shorter than in the first pair, besides they are directed considerably less forwards, and only a little upwards. Fifth segment very elongate, but yet considerably less than half as long as the entire animal, nine times as long as its breadth some- what before the middle; the terminal widened part is somewhat narrower than the anterior segments. First pair of legs (fig. 3 c) agreeing as to armature with spines with those in H. angiistus, but the three distal joints are distinctly more slender than in that species. The other six pairs mutilated or lost, but as far as can be seen they are more robust than in the two preceding species, but agi with them in the fact, that second joints of fourth and fifth pairs, when directed respectively back- wards and forwards, do not quite reach one another. Abdomen (fig. 3a and 3d) subtriangular in aspect, and on the whole somewhat similar to that in H. angusfus, especially as it has similar postero-lateral processes, but it differs in having the median part of the posterior margin subaugular, thus not broadly rounded, and the uropods are somewh; longer than in that species, somewhat more than twice as long as broad. - The median 1 the operculum distally less narrow than in H. angustus. Length of the specimen 4-5 111111. Remarks. H. insignis is easily separated from the two preceding species by the i cesses on fourth thoracic segment. In this character it agrees to some degree with //. / in the latter species this second pair of processes are bent strongly forwards, and the abdomen totally different. Occurrence. Taken by the "Ingolf" at a single deep station in the warm area. Davis vStrait: Stat. 36: Lat. 6i°5o' N, Long. 56°2i' W., 1435 f:lth- teniP- ''^ 1 sPec- 64 CRUSTACEA MALACOSTRACA. III. 39. Haplomesus tenuispinis n. sp. (PL V, figs. 4a-4f). Immature Female. The material consists of a specimen which has lost the three posterior thoracic segments and abdomen, and of another evidently younger specimen without head and the two anterior segments. After some hesitation I considered these two badly mutilated specimens as belonging to the same species, especially because they show considerable similarity in the shape of fourth seg- ment and its processes. Antennulae (figs. 4 a and 4 b) somewhat long, reaching the front end of fourth segment ; second joint half as long again as the four distal joints combined; third joint only a little longer and thicker than second joint of the flagellum, which is longer than first joint. — Antennas long and somewhat slender; third peduncular joint rather slender, and only half as long as second joint of the antennulae; fifth joint about two and a half times as long as the third, feebly increasing in thickness from base to end; sixth joint a little longer and distinctly more slender than the fifth; flagellum a little less than twice as long as sixth peduncular joint. 16-jointed. The thorax has a pair of moderately long processes on first segment (fig. 4 a) and a pair of distinctly longer processes on fourth segment. The processes on first segment are not quite half as long as the breadth of the segment, very slender, conical, slightly curved, acute, directed somewhat forwards, much outwards and feebly upwards. The processes on fourth segment are in the largest specimen (fig. 4 a) conspicuously longer, but not thicker, than first pair, directed horizontally a little outwards and very much forwards, reaching somewhat in advance of the front margin of third segment; in the small specimen (fig. 4 d) they are proportionately more slender and longer, reaching the front margin of second segment. Fifth segment — in the small specimen -- about eight times as long as broad somewhat before the middle; while its terminal leg-bearing part is unusually feebly widened, being much narrower than third segment. First pair of thoracic legs (fig. 4 c) in the main as in //. quadrispinosus, but fifth joint has only one very long, slender spine and one rather short spine, and sixth joint has a single moderately long spine. The other legs in the main as in //. quadrispinosus, but the distance between the ends of second joints of fourth and fifth pairs, when directed respectively backwards and forwards, is a little longer than in that species. The abdomen (figs. 4 c and 4 f) differs much from that in any other species of the genus, excepting to a certain degree H. modestus. In the three preceding species it is divided by a distinct lateral constriction into a moderately short proximal part answering, of course, to the first segment in Istlinom, sits, and a large distal part, but in //. tenuispinis the proximal part is not only elongate and rather little shorter than the distal part, but is even divided by a distinct constriction (fig. 4 e) into two portions, the anterior somewhat narrower and shorter than the posterior and besides, just in front of the transverse impression between them, armed with a pair of small, spiniform, sublateral processes (d) directed mainly backwards. Tin- constrictions and dorsal transverse impressions show that the ab- domen consists of two proportionately long anterior segments and the large posterior segment, all completely fused with each other and with the three posterior thoracic segments. The posterior seg- tnenl is somewhat longer than broad and broadest behind, with the lateral margins somewhat feebly CRUSTACEA M \l. iCOSTK \e"A. III. convex; each postero-lateral angle is produced in a small, quite low protuberance terminating in a very short spine marked off by an articulation; the posterior part of tin. abdomen between the uropods is produced into a rather large triangle much broader than long, and its terminal, subacute angli about ioo°. — The operculum (fig. 4 fi is slightly longer than broad, with the proximal margin v< convex, while its distal margin is long and vei \ feebly convex. The uropods, inserted between this plate and the lateral protuberances, are more than twice as long as broad, subacute, and feebl reaching the median triangle. Length of the younger specimen without head and the two anterior segments 2-5 mm.; as tin- lost parts have probably measured o-6 mm., the entire animal, which has the seventh pail ■ ' scarcely developed to full relative size, has consequently been about 3-1 mm. long. The specimen, ol which only the anterior segments with head remain, has certainly measured about 4-3 mm. Remarks. It seems to me very improbable, but yet not completely impossible, that .- future zoologists possessing a good material may find that the two mutilated specimens referred here to the same species in reality belong to two different species; in that case the name //. ten inspirits ought to be applied to the form, of which the head and the anterior thoracic segments are described here. The species is abundantly distinguished from the other forms. Occurrence. Taken by the "Ingolf" at two deep stations in the warm area. Davis Strait: Stat. 24: Lat. 63*06' N., Long. 56°oo' VY., 1 [99 lath., temp. 2-4°; anterior part of a - South of Davis Strait: Stat. 22: Lat. 58°io' N., Long. 48:2.s' \\'., 1S45 fath., temp. 1-4°; 1 without head, etc. 40. Haplomesus modestus n. sp. (PI. V, figs. 5 a-5 b). Female. Only an immature, extremely mutilated specimen, without head and first thoracic segment, is to hand. Third thoracic segment (fig. 5 a) with a pair of small, acute processes originating at ' margins and directed mainly forwards; second segment shows mutilated rudiments of probably processes; fourth segment with a pair of nearly similar processes placed dorsally margins. Fifth segment about six and a half times as long as broad somewhat before the : widened terminal part much narrower than the anterior segments. Abdomen with the proximal part about as long as broad, considerably shorter than the | part or segment, divided besides by a feeble lateral constriction into two parts or segments, and 1 off from last thoracic segment by a feeble constriction and dorsal impression, from the 1 segment by a feeble constriction. The posterior segment about as long as broad, increa: in breadth backwards, with the postero-lateral corners rectangular and each terminal thick, articulated spine (fig. 5 b). The posterior margin between the sublateral uropods circular. - The operculum slightly broader than long, but otherwise nearly as in the precedi. having the basal margin strongly convex and the posterior margin long and ven The uropods somewhat smaller than in //. tenuispinis, but showing about the same 1 Length of the fragment described r8 mm.; the animal has been slightly more than 2 mm. long. 9 The Ingolf-Expedition. III. >. 66 CRUSTACEA MALACOSTRACA. III. Remarks. Though the specimen described and figured is extremely mutilated, I think that the species is sufficiently characterized by the small processes on third and fourth segments and the shape of abdomen. Occurrence. Taken by the "Ingolf" at a single station. Davis Strait: Stat. 24: L-at. 63°o6' N., Long. 56°oo' W., 1199 fath., temp. 2-4°; 1 mutilated spec. HeterOlTieSUS Richardson. Body moderately slender or proportionately robust. Autennulse with the part beyond second joint considerably or extremely reduced, consisting of three or two very or extremely short joints. Antennae with third joint elongate, more than two and a half times as long as the fourth and even longer than the fifth or the sixth. Maxillipeds (PI. VI, fig. 4 o) with second joint extremely large, and its lobe more than half as broad as the joint and much shorter than broad; the joint itself is con- spicuously more than half as long again as the distinctly 5-jointed palp; second and third joints of the palp very small, the second slightly broader than long. — The articulation between fifth and sixth thoracic segment distinct, movable, while the two posterior segments and abdomen are immovably fused. First pair of thoracic legs (PI. V, fig. 6d; PI. VI, fig. 4 p) with fifth joint much expanded, con- siderably deeper than the fourth, broadest considerably beyond the middle, with a long, strong spine a little before the distal end of the lower margin, and a similar spine at the lower angle of the long, oblique terminal margin. — Uropods one-jointed. Remarks. The animals belonging to this genus are generally more robust than those of the preceding genera. They are greyish or brownish, frequently even somewhat dark. The species are evidently numerous, and some of them so closely allied that a careful investigation is necessary in order to separate them and make them recognizable. Five species have been established, viz. two "Challenger" species by Beddard, one species by Tattersall, and two species by Harriet Richardson. Our material contains four species, all new. 41. Heteromesus dentatus n. sp. (PI. V, figs. 6 a— 6 d; PI. VI, figs. 1 a-i c). Female. The material comprises the four anterior thoracic segments with head and appendages of a specimen with the marsupium rudimentary, and, besides, a much younger specimen. Bod)- moderately slender, densely set with minute, sharp granules, some among them a little larger, spiniform. -- Antennuhe about as long as the head (fig. 6b); second joint about as long as the interval between the antennuhe, and almost four and half times as long as the remaining distal portion, which consists of three joints (fig. 6 c), the two proximal oblong and slender, the third joint an exceedingly small nodule. - Antennae (fig. 6b) almost twice as long as the sum of the four anterior thoracic seg- ments with head; third joint slightly shorter than the breadth of first thoracic segment, and this joint has above near the distal end a conspicuous, oblong process directed upwards and forwards and ter- minating in a minute spine, and below, somewhat from the base of the joint, another somewhat longer, CRUSTACEA MALACOSTRACA. ill. 67 subcvlindrical process terminating in a small spine and directed downwards and considerabl) forwards; sixth joint of the peduncle is somewhat longer than the fifth; the- flagellum in the largest specimen even a little longer than the sum of the three distal peduncular joints. 21-jointed. - I.abruni without any trace of processes from the anterolateral angles. First thoracic segment (fig. 6b) above at the antero-lateral angles with a somewhat small, oblong, obtuse process having a couple of minute spines at the end; second segment with subsimilar pi and, besides, with a second pair of quite small processes above the middle of the lateral margins, and with some very small protuberances in a transverse row not far from the front margin; third segment with the antero-lateral processes short, and the second pair wanting, while some tubercles near the anterior margin are distinct; fourth segment without lateral processes, but with the dorsal tubercl Fifth segment in a small specimen somewhat less than four and a half times as long as broad a middle, a little longer than the remaining posterior part of the body, subcvlindrical to somewhat from its terminal widened part. First pair of thoracic legs (fig. 6 d) robust; fifth joint about half as long again as deep; eacli of the two long and strong distal spines has a minute seta somewhat from the end, and the long terminal margin has a short spine, a low minute plate with the end trispinose, and some exceedingly tiny spines; sixth joint is somewhat shorter than the fifth, about twice as long as broad, and its lowei margin has beyond the middle a single, short, thick spine with a seta on its middle; seventh joint with claw somewhat robust, and longer than the lower margin of sixth joint. The other six pairs of legs have the second joint scabrous, as it is set with numerous minute, sharp granules. The second joint of fourth pair reaches, when directed backwards, a little beyond the end of second joint of fifth pair, when stretched forward. The abdomen in the small specimen narrow, a little more than half as long again as broad; the part between the uropods moderately produced, with the hind margin nearly semicircular; ea lateral margin has somewhat in front of the uropods a conspicuous process (as in the male, PI. VI, fig. 1 a) directed backwards and outwards, and with the end truncate. — Operculum (fig. 1 c) conspicuous longer than broad, with the posterior margin considerably convex. - Uropods short, scarcely long as the breadth of the abdomen. Length of the small female 25 mm., of the large specimen, judging from the len preserved portion, about 3-9 mm. Male. Two specimens, one without the first segment and head, the other without head the three anterior segments, were found. The thoracic segments as to processes and sharp, more or less spiniform tubercles the female. Fifth segment (fig. 6 a) respectively four and a half and nearly five and a hall- long as broad before the middle, a little longer than the posterior part of the body, and i subcylindrical. Abdomen (PL V, fig. 6 a; PI. VI, figs. 1 a and 1 b) broader than in the female, a little less or a little more than half as long again as broad ; the processes on the lateral margins some* their ends well developed, directed backwards and somewhat outwards and downwards dorsal longitudinal area conspicuously broader than one of the lateral areas. - 68 CRUSTACEA MALACOSTRACA III. good deal longer than broad; the median lamella tapers moderately from somewhat from the base to a little from the end, and reaches beyond the end of the lateral, moderately broad plates. — Uropods only a little longer than in the female, scarcely half as long as the breadth of the abdomen. Length of the male, judging from the parts preserved, 3-6 mm. Remarks. II dentatus is instantly distinguished by the well developed, oblique, lateral pro- cesses on the abdomen from the six species seen by me, viz. //. spinescens Rich., //. granulatus Rich., //. Greeni Tatt. and the three species established later on, which have no vestige of these processes. It agrees with H. Green/ and differs from H. Schmidtii and H.frigidus in having a terminal and a proximal pair of processes on the third joint of the antennae; it agrees with H. spinescens in having the terminal process on third joint of the antennae, but differs in having, besides, the proximal process on that joint. It is somewhat allied to //. Thomsoni Bedd., but in this species the end of abdomen is bifurcate. Occurrence. Taken by the "Ingolf" at a single station in the warm area. South-West of Iceland: Stat. 78: Lat. 6o°37' N., Long. 27°52' W., 799 fath., temp. 4-5°; 4 muti- lated spec. 42. Heteromesus longiremis n. sp. (PI. VI, figs. 2 a— 2 f). Female. Only a very mutilated specimen, without head and first thoracic segment, is to hand. — Body finely granulated. Second segment (fig. 2 a) with a minute lateral tubercle and some tiny dorsal protuberances in a transverse row; third segment with similar protuberances but without lateral tubercles. Fifth segment somewhat long, four times as long as broad across its narrowest part, and conspicuously longer than the posterior part of the body; consequently the end of second joint of fourth pair of legs reaches, when directed backwards, scarcely to the end of the corresponding joint of fifth pair, when stretched forwards. — Thoracic legs somewhat long and slender, without sharp granules on second joint, but with the usual short spines on fifth and sixth joints. Abdomen (figs. 2 a and 2 b) about one-third as long again as broad; the part between the uropods moderately produced, with the hind margin forming a portion of a circle. — Operculum (fig. 2 c) about as long as broad, with the posterior margin transverse, very long and slightly convex, excepting at the rounded lateral angles. -- Uropods somewhat long, not fully half as long as the abdomen, straight, acute. Length of the preserved part of the body 3-5 mm. Male. The single specimen is without head and the four anterior segments. — Fifth thoracic segment (fig. 2 d) long, slightly more than six times as long as broad at the middle, and almost half as long again as the posterior part of the body. -- Abdomen, as usual, broader than in the female, only some- what longer than broad (fig. 2e); the produced posterior part with the hind margin forming a portion of a circle; the dorsal longitudinal median area, as usual, narrower than in the female, nearly as broad as one of the lateral areas. - - Operculum (fig. 2 f) distinctly broader than long, the median lamella with the proximal part broader, the distal part narrower than in //. ilnitnliis^ and scarcely leaching the end of the somewhat broad lateral plates; the distance between the operculum and the CRUSTACEA MAXACOSTRACA. III. 69 anal doors is considerably larger than in the males of //. dentahu or // — T't scarcely two-thirds as long as the abdomen. Length of the preserved part of the body 37 mm. Remarks. H. longiremis is easily distinguished* from the two following spec :des, from the two North-Atlantic species described by Richardson by having in b le fifth - ment longer both in proportion to thickness and to length of the coalesced part of the body, and having the female operculum proportionately broader with the hind margin long and slightly and finally by the long male uropods, etc. Occurrence. Taken by the "Ingolf" at a single very deep station in the warm . Davis Strait: Stat. 36: Lat. 6i°5o' N., Long. 56°2l'W., 1435 fath., temp. 1-5°; 2 very mutilal 43. Heteromesus Schmidtii a (PI. VI, figs. 3 a— 3 Female (without marsupium). Body somewhat robust, not fully five times as long as the breadth of first thoracic segment, finely granulate, but only on the lateral part of the last abdominal segment and the last thoracic segment can the granules be said to be real sharp minute tuben the head and the thoracic segments, excepting the major portion of fifth segment, adorned on the upper surface with irregular wavy markings or impressions. Antennulse (fig. 3b) nearly half as long as first thoracic segment with head; second joint distinctly shorter than the interval between the antennuke, and it has above near the distal end a small tubercle, which on closer examination is seen to consist of three extremely short joints ifig. 3 c). the first distinctly longer than, and more than twice as broad as, the second, while the third is somewhat smaller than the second, and has on the end a somewhat short and a long robust seta. — The anteniue have the third joint conspicuously longer than the sixth, as long as first segment with head, and it has no proximal process at the lower side and no subterminal process at the rounded end (the en the flagellum lost in my specimen!. — The labrum has each antero-lateral angle produced in but conspicuous tooth (fig. 3 b). First thoracic segment has each antero-lateral angle produced into a somewhat small protube- rance; some few tiny tubercles are more or less distinct on the upper surface of the four anter segments, and a similar pair are found on the lateral margins of second segment Fifth segmu sybcylindrical, with the terminal part not much expanded; it is not fully twice as long as broa. its anterior end, and considerably shorter than the posterior coalesced part of the body, but ously longer than the abdomen. First pair of thoracic legs robust; the leg preserved seems to be shaped and armed nea in H.frigidus (comp. fig. 4P), but the fifth joint has distally below a less sharp angle, and sixth joint is very robust with at least two strong spines. The other six pairs of legs differ 1 H. longire?nis in being distinctly shorter and less slender, and in having the second join with numerous minute, sharp teeth; second joint of fourth pair, when directed backwards, rear beyond the end of the corresponding joint of fifth pair stretched forward-. 7° CRUSTACEA MALACOSTRACA. III. Abdomen (fig. 3d) about one-third as long again as broad; the part between the uropods con- siderably produced, with the median part of its posterior margin very feebly rounded, almost transverse. ■ — Operculum (fig. 3 e) conspicuously longer than broad, with the posterior margin moderately long and slightly convex. — Uropods nearly straight, acute, a little less than half as long as the abdomen. Length of the single specimen 4.2 mm. Remarks. H. Schmidtii is easily separated from the female of H. longircmis by having fifth thoracic segment much shorter and thicker, and by a more oblong operculum ; from H. frigidus it is instantly distinguished by having the abdomen conspicuously broader and another shape of the oper- culum. According to my own examination of three specimens of //. Greeni Tatt. kindly lent me by Dr. Tattersall, this species differs from H. Schmidtii in several particulars: the three distal joints of the antennulae are several times longer than in H. Schmidtii, the first thoracic segment has at least frequently its antero-lateral angles produced into conspicuous conical processes directed mainly for- wards, the spiniform granules on the anterior segments and on second joint of the legs are consider- ably longer and consequently more conspicuous than in my species, finally the inner distal angle of third antennal joint is produced into a conspicuous small process or tubercle terminating in a minute spine - - a feature not observed or drawn by Tattersall — and, besides, this third joint has at the middle of the proximal half of its lower or inner margin an oblong, spiniform process directed for- wards and downwards, and figured by Dr. Tattersall. Besides, I have compared co-types of H. spines- cens Rich, and H. granulahis Rich, kindly lent me by the authorities of the U. S. National Museum. H. spinescens — the largest specimen, a female without marsupium, was 57 mm. long — differs from H. Schmidtii especially in having the end of third antennal joint produced in an acute process, while the fifth thoracic segment in the female is distinctly longer in proportion to its narrowest breadth than in H. Schmidtii, and nearly or fully as long as the posterior part of the body. (It may be inserted here that H. spinescens Rich, differs from H. Greeni Tatt. especially in having the distal part of the antennulse shorter and only two-jointed, in having no proximal process on third antennal joint and the surface of the anterior segments less scabrous). The female of //. granulaius — ovigerous females were 3-9 and 3-3 mm. long — is easily distinguished from //. Schmidtii in being more clumsy, with fifth thoracic segment conspicuously shorter, only a little more than half as long again as broad across its most narrow part, and only as long as the abdomen, furthermore in having the abdomen narrower than in H. Schmidtii and shaped nearly as in //. frigidus, the uropods less than one-third as long as the abdomen, etc. The species is named in honour of Dr. Joh. Schmidt, who, during his cruizes in the "Thor", collected a good number of the Isopoda described or enumerated in the present report. Occurrence. Not taken by the "Ingolf". Hut the "Thor" has secured it at a single place in the warm area. South of Iceland: Iyat. 62c57' N., Dong. 19°58' W., 508 fath.; 1 spec CRUSTACEA MALACOSTRACA. 111. 44. Heteromesus frigidus 11. sp. 1 PI. VI, figs. 4 a— 4 t). Female. Body still a little more robust than in //. Schmidtii. only about tour and a half times as long as the breadth of first thoracic segment. The surface of the body and It ,1111- lation and distribution of wavy markings or impressions (fig. 4 i) on the thoracic segments nearl; in the preceding form. Antennulae (figs. 4i, 4 k and 4 1) still more reduced than in //. Schmidtii. Second joint about half as long as head and first thoracic segment combined; at its upper distal angle is found a minute impression, in which is seen a tiny knot frequently difficult to examine; in reality this knot of two joints, the first (fig. 4 1, j) more than twice as broad as high, with a couple of fine setae on the margin, the second joint {4) exceedingly tiny, with a fine seta and a proportionately very loug and strong seta on the upper surface. — The antenna; nearly reach the posterior end of fifth thoracic segment; third joint is scabrous, closely set with minute, sharp granules, it is about as long as the head or as sixth joint, its end is rounded without any process, and it has no proximal process at the lower margin; flagellum about as long as the two distal peduncular joints combined, with about 14 joints. — The labrum has each autero-lateral angle produced into a small, but conspicuous tooth. First thoracic segment (fig. 41) with each antero-lateral angle produced into a more or les; conspicuous triangular tooth; sometimes a minute tubercle may also be found on the side of second segment, but the four anterior segments have in this sex uo transverse row of small tubercles. Fifth segment is from a little more to slightly less than twice as long as broad at its most narrow point, and somewhat shorter than the posterior part of the body, being about as long as the abdomen plus seventh thoracic segment. First pair of thoracic legs (figs. 4 p and 4 q) robust; fifth joint only one-third as long again i deep, with the usual long and very strong spine a little before the end of the lower margin and a similar spine on the lower end of the long, oblique terminal margin; the last-named margin besides, (fig. 4 q) two short, very thick spines, each with a seta near the middle, and a low plate with the margin serrate; sixth joint about as long as the fifth, almost twice as long as deep, with both margins considerably convex, and the lower margin has beyond the middle a couple of short, tlm spines, each with a seta at the middle; seventh joint with claw somewhat robust, as long 1 margin of sixth joint. The other six pairs of legs nearly as in H. Schmidtii, with second joint sea the end of second joint of fourth pair, when directed backwards, reaches somewhat beyond the corresponding joint of fifth pair stretched forwards. Abdomen (figs. 4I1 and 4 r) narrow, half as long again as broad; the longitudinal medis on the surface is conspicuously broader in proportion to the lateral areas than in any other species; the part between the uropods is much produced, with the terminal margin ver — Operculum (figs. 4 s and 4 t) somewhat longer than broad; its major part is circular, v anterior part is produced forwards, so that the lateral margins somewhat from the base are : concave or rather bent inwards. - Uropods generally feebly curved, somewhat less than half as Ion as the abdomen. j2 CRUSTACEA MALACOSTRACA. Ill Length of a female with inarsupium 3-9 mm., of a large female without marsupium 4-8 mm. Male. As usual, considerably more slender (fig. 4a) than the other sex, about five and a half times as long as the breadth of first thoracic segment. — Antennula: (fig. 4 b) with second joint conspicuously shorter than in the female; third joint (fig. 4c, j) in the main as in the male, but fourth joint {4) much longer than in the female, longer than thick. — The antenna: have their second joint somewhat longer and considerably thicker than in the female, a little longer than head and first thoracic segment combined; flagellum with about 13 joints. First thoracic segment has about three distinct tubercles in a transverse row (fig. 4 b), the three following segments each with a transverse row of distinct tubercles a little from the front mar- gin. Fifth thoracic segment far from cylindrical, as it tapers from both ends towards the rather narrow median part; it is about five and a half times as long as broad, and nearly longer than the posterior part of the body. Abdomen (figs. 4 a and 4 d) broader than in the female, scarcely one-third as long again as broad, with the posterior part distinctly less produced than in the female and evenly rounded, while the dorsal median area is scarcely as broad as one of the lateral areas. — Operculum (fig. 4 e) con- spicuously longer than broad; the median lamella in its natural position tapering very moderately from somewhat from the base to near the end, which overreaches the lateral plates; these are more oblong (fig. 4 e) than in H. longiremis and coarsely granulate (fig. 4g); in comparing fig. 4 f, represent- ing the median lamella removed from the animal, with fig. 4 e, it is perceived that its lateral parts are overlapped by the lateral plates, second pair of pleopods. — Uropods (figs. 4 a and 4d) longer than in the female, somewhat more than half as long as the abdomen. Length 37 mm. Remarks. The female of H. frigidus is easily separated from //. Schmidt! i by the narrow abdomen and other minor features; it is closely allied to H. granulatus Rich., but in the latter species both sexes have the fifth thoracic segment considerably shorter in proportion to breadth at or before the middle, and in proportion to the remaining posterior part of the body, than is the case in H. frigidus. Occurrence. Taken by the "Ingolf" at six stations, all in the cold area. North of the Faeroes: Stat. 141 : Lat. 63°22' N., Long. 6°58' W., 679 fath., temp. -f- o-6°; ab. 20 spec. Stat. 139: Lat. 63°36' N., Long. 7°3o' W., 702 fath., temp. -=- o-6°; ab. 20 spec. East of Iceland: Stat. 105: Lat. 65°34' N., Long. 7°3i' W., 762 fath., temp. -f- o-8°; 1 spec. Stat. 102: Lat. 66°23' N., Long. io°26' W., 750 fath., temp. -=- 0-9°; 2 spec. Stat. 101: Lat. 66°23' N., Long. I2°05' W., 537 fath., temp. -=- 07°; 1 spec. North of Iceland: Stat. 124: Lat. 67°4o' N., Long. i5°4o' W., 495 fath., temp. -7- o-6°; 11 spec. Group VI. Pseudomesini. This group is established on a single genus (with a single speciesi, which in general aspect is somewhat similar to the Ischnomesini, but differs from it in some important characters, and on the CRUSTACEA MALACOSTRACA. III. whole differs much from all genera of the familly. It maj he sufficient hert i<> refer t<. the follow iug description of, and remarks on, the genus. Pseudomesus n. gen. Description (the male unknown). Body (PI. VI, figs. 5a — 5b) elongate, slender, round, nar- rowed at the middle about as in Heteromesus; the integument is somewhat weak.— Head fi me- what ovate, narrowing forwards. Eyes absent. Antennulse short (PI. VI, figs. 5 a and 5 c); first joint thick, second joint longer than the first; flagellum very short. Antennae lost in the specimens excepting the three proximal joints, which are somewhat slender; third joint nearly as long as the two others combined; squama not discoverable. — Mandibles (fig. 5 d) in the main as in Macrostylis; left mandible with the few teeth on the incisive part thick, the movable lobe short and thick; the setae few in number, but the anterior setae thick and distally ramified; the molar process is short, nearly lamellar, distally very narrow, directed much back- wards and terminating in some setae; palp absent. Maxillipeds (fig. 5c) nearly as in Nannoniscus ; second joint long and moderately broad, twice as long as broad without counting the lobe, which i> of verv moderate size, longer than broad and at the inner margin terminating in a triangular pi the palp has its three proximal joints moderately broad, fourth joint well developed, about as long the third, fifth joint somewhat small; epipod long, reaching the middle of third joint of the palp, but almost three times as long as broad. Thoracic segments movable, without lateral protuberances. The three anterior segments some- what short, the second a little broader than the first or the third; fourth distinctly narrower than the third, posteriorly somewhat produced and narrowed as in Heteromesus ; fifth segment elongate, much produced and somewhat narrowed forwards; sixth segment much shorter and posteriorly slight!} broader than the fifth; seventh segment a little broader and somewhat shorter than the sixth. Thoracic legs slender and somewhat long, not differing much in length ; first pair are a little shorter than the second, but agreeing with them and differing from the othei pairs in having the fifth joint a little thicker than the third (fig. 5 f) and armed with a few spines on the lower margin, and their sixth joint is a little thicker than in the following pairs; in all pairs the seventh joint is moderateh long and slender, the claw almost setiform; the posterior pairs decrease perceptibly in leu fifth to seventh pair. Abdomen distinctly broader than seventh thoracic segment (fig. 5 b), nearly oval, posteriori; broadly rounded; a first free segment rudimentary. -■ The female operculum (fig. 5 g) leaves a m< ately broad lateral part and a rather broad posterior part of the lower surface uncovered. • small but thick, two-jointed, situated on the ventral side at the antero-lateral margins of the doors a little behind the operculum. Remarks. Pseudomesus shows considerable similarity to Heteromesus in fourth and thoracic segments, but it differs very much from the Ischnomesini in bavin- the head free, immerged within or fused with first thoracic segment, in the structure of the mandibles, whic much similarity to those in Macrostylis, Nannoniscus and Desmosoma, furthermore in ha\ The Ingolf-Expedition. III. ;. 74 CRUSTACEA MALACOSTRACA. III. anterior pairs of legs subsimilar, in the place of the nropods, etc. Pseudomestis is perhaps more allied to Nannoniscus than to any other genus, but yet differs in important features. Only a single species is known. 45. Pseudomesus brevicornis n. sp. (PI. IV, figs. 5a-5g). Female. Body scarcely six times as long as broad; second segment and abdomen subequal in breadth and broader than the other parts. — Head, seen from above (fig. 5 b), a little longer than broad; seen from the side very deep (fig. 5 a). — The antennuhc occupy a lateral position and consist of five joints (fig. 5 c); second joint considerably longer than the three distal joints combined, and with a couple of long terminal setae; third joint a little longer than the fourth, while the fifth is very small with two long setae. The thoracic legs show some particulars not mentioned above. First pair have three spines on the lower margin of fifth joint, each spine with a seta somewhat from the end, and in the intervals between the spines a long, low and thin lamella with numerous tiny hairs on the margin (fig. 5f, «); a similar lamella is also found on fourth joint; sixth joint without spines. Second pair have four spines and the same number of lamellae on fifth joint, two spines on the sixth joint. (Third pair want- ing). Fourth pair with a single spine on fifth joint, three spines and two lamellae on the sixth. The three posterior pairs have a peculiar, slender, triangular process on the posterior margin of third joint rather near the base, and a single conspicuous seta on the lower side of fifth joint a little from the end. Abdomen somewhat longer than the two posterior thoracic segments combined, somewhat longer than broad and rather highly vaulted. — Operculum (fig. 5 g) somewhat longer than broad, with a short proximal part much vaulted (fig. 5 a); the posterior half has the margin nearly semicircular. - The uropods as mentioned above; first joint thicker and much longer than the second. Length of two females with marsupium 3-2 and 2'8 mm. Remarks. As only a single species is known, some of the minor features mentioned in the description of the genus may be only of specific value. Unfortunately the male is unknown. — P. brevicornis differs very much from all other forms of Asellota hitherto known; it may be said to have some similarity in general aspect to Thambema amicorum Stebb., but differs in mouth-parts, thoracic segments, legs, etc. Occurrence. Taken by the "Ingolf" at a single station in the cold area. North of Iceland: ,Stat. 102: Lat. 66°23' N., Long. io°26' W., 750 fath., temp. ^0-9°; 2'/2 spec. Group VII. Macrostylini. Founded on a single very peculiar genus. In order to avoid reiteration it may suffice to refer to the following treatment of the yen us. CRUSTACEA MALACOSTRACA. III. 7- Macrostylis G. O. Sars. Sars has published a useful diagnosis of this most characteristic genus in his standard work; he had, however, only a single species, M. spinifera G. O. Sars, but in the appendix he add* ond species, .1/. \ Vana) longiremis Mein. A third species, M. latifrons Bedd., was secured by the Chal- lenger" in the North Pacific. But the genus is evidently rather rich, as I have six species, four among them new, from the "Ingolf" area, and, besides, fragments of a seventh form. The new form in most of the features pointed out by Sars as being of generic value with his .1/. spintfera, but a couple of characters, viz. his statement on the female operculum and on the uropods, must be drop] while some other features can be pointed out An examination of the structure with special refereni 1 to generic characters may be of some interest. The body is somewhat or considerably elongate, to sonic extent similar in all species, Head movable. Eyes wanting. Antennulae short, at most 5-jointed, in the males with mure sensory filaments and, besides, sometimes thicker than in the female. Antennae with 5 free joints in the peduncle, while the basal joint is very short and frequently scarcely distinguishable; squama seems to be wanting. The mouth-parts are well figured by Sars; the mandibles have the incisive part, the movable lacinia, and the setae well developed, the molar process strongly tapering to the narrow, obtuse, setiferous end and directed considerably backwards; palp wanting. Maxillipeds (PI. VII, fig. 4c, and Sars, PI. 51) with second joint long and more than twice as long as broad, with its lobe oblong; palp with second joint broad and long, third joint broad and short, the two distal joints short 01 nearly rudimentary; epipod very long. The three anterior thoracic segments constitute togethei a separate subquadrangular section, with each lateral margin forming a nearly uninterrupted, distinctly convex line; these segments dorsally well marked off, but yet scarcely, or at most slightly, movable; the four posterior segments are quite different, each very movable, but the articulation between seventh segment and abdomen seems to be slightly movable in some forms. The legs have been well figured and described by Sars; the basal joint of most pairs is distinct. but without any process; the three anterior pairs increase somewhat in length and very conspicuously in thickness and number of spines from first to third pair, the last-named pair having third and four joints somewhat or considerably expanded and seeming to be "fossorial in character"; fourth pai: siderably shorter and thinner than third or second pair, and adapted for walking; from four or seventh pair the legs increase much in length; second joint of seventh pair has a row of probably natatory setse along a part of the posterior margin of both margins (PL VII, fig. 3 e The abdomen is more or less oblong-quadrangular, with the posterior margin convex and the uropods inserted on the outer angles; a tergite of a basal segment was not found, but distinct in some species. When the abdomen is examined from above, and is not too opaque, a pe liar organ is seen shining through the integument (PL VII, figs. 1 c, 3 I and 4 I; PL VI, fig the lateral margin somewhat before its end; each organ is a cavity narrowing backwards which seems to open at the inward angle of the uropod; the central part of the cavit a lump of minute rounded or angular crystals. -- On the lower side of the abdomen the operculum yS CRUSTACEA MALACOSTRACA. III. terminates at least a little, and frequently far, from its end, but the excavation containing the pleopods is continued behind these to the end as a channel (PI. VI, fig. 6 a; PI. VII, figs. 2 g, 2 h, 6 gj, and on its bottom the two doors closing the anal aperture are generally seen. The female operculum is ob- long (though never so narrow as figured by Sars; its shape affords specific characters). — Uropods uniramous, long to extremely long, with two or many joints. Remarks. In the shape of the mouth-parts Macrostylis is allied to Nannoniscus G. O. S., but in other features, especially the thoracic segments and their legs, it occupies an isolated position. The above-mentioned organs in the abdomen are unknown in other Asellota; they are evidently a kind of statocysts to a certain extent analogous with the organs, or unpaired organ (see later on), in the telson of at least some forms of the family Anthuridae, and with the organs in the uropods of the family Mysidae. 46. Macrostylis spinifera G. O. Sars. (PI. VI, figs. 6 a— 6 c; PI. VII, figs. 1 a— 1 c). 1864. Macrostylis spinifera G. O. Sars, Forh. Vid. Selsk. Christiania f. 1863, p. 219. ! 1897. — G. O. Sars, Account, II, p. 121; PI. 51. The specimen from the "Ingolf" Stat. 32 is normal and agrees with a number of specimens from Skager Kak, but the specimen from Stat. 80 differs in a couple of features. We may begin with re- marks on the typical form, thus a kind of supplement to the description and figures of Sars. The antennulae are 5-jointed, in the female slender with the fifth joint minute (fig. 1 a). Sars says that they are 4-jointed, which is incorrect; according to two of his figures the antennulae in the male are similar to those in the female excepting that they have three sensory filaments, while the female lias two. Hut in a male from Skager Rak the antennulae are considerablv longer and thicker than in the female, and fourth joint has four sensory filaments, while on the fifth, which is about as long as fourth and third joints combined, five such filaments were found; in another male from the last-named locality the antennulae are intermediate between those in the male mentioned and a normal female. — The antennae vary somewhat in length, in two ovigerous females the antennae do not respectively reach to the posterior margin of third thoracic segment and to the middle of fourth segment, in both cases the penultimate joint of the peduncle is only a little longer than the terminal: the long antennae have 7, the others only 5, joints in the flagellum. Fourth thoracic segment has each postero-lateral angle a little produced as a foot for a some- what small, robust spine; the three posterior segments have the postero-lateral parts somewhat pro- duced backwards and a little outwards, each bearing the leg below at the end, and on the angle a somewhat small spine; Sars' figures of male and female show the spines too thick and the majority not marked off, so that they afford the erroneous impression — not contradicted in his text - of being strong processes. — In the full-grown female without tnarsupium each thoracic segment has a ventral, spiniform process, which is long or very long on the two posterior segments, while the pro- cess on first sternite is moderately large, originating from the middle of the sternite and directed downwards and somewhat forwards; in females with marsupium the processes on the two posterior CRUSTACEA MALACOSTRACA. 111. 77 segments are as large as in females without marsupiurn, but the process on first sternite is small, and those on the four other steniites wanting. The sternites in the male armed about as in females with- out marsupiurn. — The dorsal spine on third joint of third pair of leg figured by Sars, stn and very curved. Abdomen conspicuously more than half as long again as broad, in the male slightly or not longer than the two preceding segments combined, in the female as long as these two segments and half of fifth segment combined. — The operculum reaches in the female (fig. 6b) to rather near the end of the abdomen, thus overlapping the anal doors; it is slightly more than twice as long as bn — thus conspicuously broader than drawn by Sars — and its distal end is broadly rounded. In the male the operculum (fig. 6 a) terminates far from the end, and even before 1 line between the insertion of the uropods, thus leaving a long ventral channel uncovered, and in the anterior half of this channel the large anal doors are seen. Length of adult females from Skager Rak 22 111111., of males 2 mm. The somewhat aberrant specimen from Stat. 80 is a scarcely full-grown female, 2 mm. long. The antenna; reach nearly the middle of fourth thoracic segment; the penultimate joint (fig. 1 a) "i the peduncle is one-third as long again as the terminal, thus proportionately longer than in the typical form; flagellum 7-jointed. The dorsal spine on third joint of third pair of legs is quite straight 1 and scarcely as strong as in the typical form. The abdomen (fig. 1 c) is distinctly less than half ; long again as broad, its lateral margins being more convex than in the typical form. But being un- able to find any further difference between this specimen and the Danish specimens I consider >■ be an individual or local variety. Remarks. M. spinifera differs sharply from the other species, excepting M.abyssicola and .1/. elongate in having the posterior part of the abdomen strongly produced as a long and distally broadh rounded lobe between the uropods. From M. abyssicola it differs in having the antennuke twic< long, the produced end of the abdomen much broader, the female operculum long, etc. .1/ elongate differs from M. spinifera in the extremely long, many jointed uropods, the antenna: only as long as the head, with the distal peduncular joints short, etc. Occurrence. Taken by the "Ingolf" at two stations. Davis Strait: Stat. 32: Lat. 66°35' N, Long. 56°3 than in M. longiremis, and appreciable differences are also found in third pair of legs and their si In a young female with seventh pair of legs only half developed, the uropods, especially their first joint, are shorter (fig. 4I1I than in the adults. — The single male found (from Stat. 139) is scarcely adult and has the operculum very mutilated; it differs Erom the females in having the ventral thorai processes conspicuously longer, the abdomen about half as long again as the uropods, and these uropods. which consequently are proportionately longer than in the females, have the first joint slightly m than twice as long as the second; in other respects this male does not show am differences worth mentioning from the female. Occurrence. Taken by the "Ingolf" at five stations, all in the cold area. North of the Faeroes: Stat. 139: Lat. 63°36' N., Long. 7°3o' W., 702 fath., temp. -i-ob°; 2 spei East of Iceland: Stat. 103: Lat. 66°23' N., Long. 8°52' W., 579 fath., temp. -^0-6°; 1 spec — - Stat. 102: Lat. 66°23' N.. Long. io°26' W., 750 fath., temp, -f-090; 10 spec. North-East of Iceland: Stat. 119: Lat. 67°53' N., Long. io°io/ W., toio fath.; temp. -5- io°; 2 s North of Iceland: Stat. 125: Lat. 68°o8' N., Long. i6°o2' W., 729 fath.; temp, -hoS3; 2 spec. Besides, it has been taken by the "Thor" at the following place in the cold area. East of Iceland: Lat. 66° 19' N., io°45' W., 765 fath- tetnP- ^0-92"; x sPec- (young ?|. 50. Macrostylis longiremis Meinert. 1 PI. VII, fig. 5 a). 1890. Vana longiremis Meinert, "Hauchs" Togter, p. 195; PI. II, figs. 63 ! 1899. Macrostylis longiremis G. O. Sars, Account, II, p. 250; Suppl. PI. H. Kg. 1. Sars has published good figures of the animal from above, of antennulae, antennae, maxilli] and several legs. As already stated in my remarks on M. subinermis, the processes on 11 The Ingolf-Expedition. III. ;. g2 CRUSTACEA MALACOSTRACA. III. and especially on the posterior sternites are long and slender in M. longiremis\ the female operculum, which is from somewhat less than to fully twice as long as broad, reaches to the insertion of the uropods, thus to rather near the end of the abdomen, and its posterior end (fig. 5 a) is somewhat broader than in allied forms. The setae and spines on the upper side of fourth joint of third pair of legs are characteristic as to length, and several among them are bifid at the end ('comp. the figure of Sars) and differ from those in M. subinermis. A single mutilated female from our area agrees completely in ail essential features with spec- imens from Skager Rak. Occurrence. Not taken by the "Ingolf". But it has been gathered by the "Thor" at a single locality. West of Iceland: Ivat. 03°46' N., Long. 22°56' W., 79 fath.; 1 spec. Distribution. Off the Skaw in Skager Rak, 125 fath. (Meinert). 51. Macrostylis longipes n. sp. (PI. VII, figs. 6 a— 6 g). Male. Slender, about five times as long as broad. -- Head of the usual shape. — Antennuke nearly as long as the head, somewhat robust, 5-jointed; first and second joint subcylindrical, subequal in length, together a little longer than the remainder of the appendage (fig. 6 b), and each about twice as long as deep; third joint short, thicker towards the end, scarcely as long as the fourth, which is thicker and has a transverse row of long sensory filaments a little from the end; fifth joint consider- ably longer than the fourth, somewhat clavate and with a number of long sensory filaments near the end. — (Antenna; lost, excepting the three small and slender proximal joints). — Maxillipeds with their distal joints as in .1/. subinermis. Thorax slender. The three anterior segments combined slightly longer than broad. Fourth segment a little narrower than the third, short, two and a half times as broad as long, with the postero- lateral angles broadly rounded. The three posterior segments subequal in length, equal in breadth, and slightly narrower than fourth segment; their postero-lateral part at each side (fig. 6 f) produced conspicuously backwards and a little outwards, distally rounded, and seen from above the posterior part of the coxte is perceived at their ends, but spines seem to be wanting; seventh segment scarcely one- third as broad again as long. First segment has a strong and somewhat long, acute ventral process directed mainly downwards (fig. 6a); sixth segment with a broad, low, acute process, while the process on seventh segment is slender, somewhat long, acute and directed downwards and a little back wauls. The three anterior pairs of legs of usual length and shape. Third pair (figs. 6 c) somewhat slender; the spine on third joint somewhat long and straight; most of the setae or spines on the two following joints of middle length and scarcely bifid at the end. Fourth pair, as usual, shorter than the third, not one-third as long as the animal; the four posterior pairs increase gradually in length backwards, but so much that sixth pair are a little longer than all thoracic segments combined, and seventh pair nearly half as long again as the entire animal, thus showing a quite anomalous devel- opment Sixth pair (fig. 6d) slender in proportion to their length; fifth joint is almost three times, CRUSTACKA MAI.ACOSTKACA. III. »3 and sixth joint twice, as long as the fourth; seventh joint is a little more than half as lonj^ as the fourth and shorter than the claw. Seventh pair (fig. 6e), and especially almost their distal half, ver) slender; second joint with a row of very fine and somewhat short setae along the proximal hall ••! the posterior margin, while the anterior margin has only about four set;e; third joint is slightl) Ion than the second; fifth joint is somewhat more than twice as long as the fourth but even slightly shorter than the sixth, which is extremely long; seventh joint nearly as long as tin- fourth, extremel) slender, and a little longer than the setiform claw. Abdomen conspicuously shorter than the two posterioi segments combined (fig. 6f), somewhat less than twice as long as broad, considerably tapering towards the base and again narrowing befori the postero-lateral parts, each of which is somewhat produced outwards as a triangle; the hind margin is unusually long, conspicuously longer than the basal breadth of the abdomen, and only moderatcK convex, slightly concave near the insertions of the uropods. The opercnlum ifig. 6 g) occupies about three-fourths of the length of the ventral excavation; its parts differ considerably from those in I/. spinifera or Af. abyssicola, but as they seem to be somewhat deranged in the specimen, I have figured them as well as possible without removing them from the animal, and omit a description. The anal doors are short, and situated considerably behind the operculum close at the end of abdomen. (The uropods are lost). Length 2-9 mm. Remarks. M. longipes is distinguished from the other species by the enormous length ol seventh pair of legs, as I suppose that the unknown female agrees with the male in this feature. In the antennulse and the shape of the three posterior thoracic segments this male differs considerabK from the same sex of the preceding forms. Occurrence. Taken by the "Ingolf" at a single station in the cold area. East of Iceland: Stat. 102: Lat. 66°23' N., Long. io°2o' \\\, 750 fath., temp. ^09°; 1 spec. Group VIII. Nannoniscini. Body oblong, considerably depressed, with the sides expanded. Head free; eyes wanting. An- tennulse dorsal, short, with very few joints in the flagellum. Antenna well developed; squama distinct. — Mandibles with the incisive part, the movable lacinia, and the setae well developed; process tapers strongly to the narrow, obtuse, setiferous end, and is directed somewhat backwards; palp 3-jointed. Palp of the maxillipeds with second joint much expanded and the two distal join small; epipod very long. — Thoracic segments movable, excepting that sixth and seventh segment- are generally fnsed at the middle. The thoracic legs inserted on the ventral side; first pair really prehensile, generally simple; the six other pairs without processes from first joint, a walking, and sometimes the three posterior pairs have, besides, natatorj setae on fifth and sixth joint-, furthermore they are moderately slender and not differing much in length; an accessory claw frequently distinct. — Abdomen at most with a vestige of a basal segment; uropods ventral, somewha biramous, with unjointed rami. 84 CRUSTACEA 1IALACOSTRACA. III. Remarks. This group comprises three genera from our area, two among them established as new (see the foot-note to Nannomscella). The animals are in general aspect somewhat similar to species of lai/ira, but they differ from all Ianirini in several important features, above all in the man- dibles. The antennulae are inserted on the upper surface of the head somewhat from the lateral margins and considerably or far behind the anterior margin of the head; between the antennulae the surface of the head shows a median broad area narrowing forwards, and both on the sides end in front con- spicuously marked off, on the sides generally by keels or sharp margins, while its anterior end generally is free, seen from above limited by a line; the shape of this part, which I name the front area, affords excellent characters. Nannoniscella n. gen. Description. Body considerably flattened. Head anteriorly irregularly rounded, with the front area well developed and reaching nearly beyond the anterior margin (PI. VIII, figs, i a— i b); the head is laterally expanded, with each lateral part produced forwards as a somewhat large triangular plate outside the proximal part of antennula and antenna. — Antennulae moderately short ifigs. i a — i b); first joint large and thick; second joint slender and shorter than the first; flagellum with few joints. Antenna; intermediate in shape and length between those in lanira and Nannoniscus\ squama well developed, triangular; fifth and sixtli peduncular joints rather long. — Mouths-parts (PI. VII. figs. 7 a — 7 d) in the main as in Nannimiscus; left mandible (fig. 7 a) has the incisive part somewhat long and slender, the movable lacinia thin, the setae moderately numerous, the molar process directed some- what backwards and very slenderly conical, with some short setae on the end. Maxillae (fig. 7 C) with the inner lobe somewhat narrow. Maxillipeds (fig. 7 dl long; second joint moderately large, somewhat less than twice as long as broad, its lobe long, not quite twice as long as broad ; third to fifth joints very broad, and the fourth a little broader than long; sixth joint small with an oblong lobe, seventh quite small; epipod very long and proportionately narrow, nearly four times as long as broad. Thoracic segments (fig. 1 a) rather crowded together and rather expanded; fifth and sixth seg- ments with more than half of their posterior margins very concave, while seventh segment has the major part of its anterior margin very convex, and the segment itself is more than twice as long at the middle than at the sides; the dorsal limit between sixth and seventh segment is distinct in its entire length, but whether the articulation is movable is somewhat dubious. In front of the lateral part of first segment is seen a somewhat small narrow plate (fig. 1 b. r/>) projecting outwards and for- wards, and this plate must be considered an epimeral process from the base of first pair of legs. - All legs of middle length, rather slender and adapted for walking (figs, id — ie); but first pair are yet somewhat shorter, and distally more robust than the other pairs; seventh joint with its short claw is '•11 first pair (fig. id) of very moderate length, on the other pairs (fig, 1 e) somewhat longer and slightly thinner; an accessory claw not developed. Abdomen broad, laterally much expanded (figs. 1 a and 1 f), posteriorly rounded. Operculum in the female oblong, and it occupies a very small part of the ventral surface (fig. 1 f). — Uropods ventral, moderately small, hiramous, with the rami unjoin ted, - (Male unknown). CRUSTACEA MALACOSTRACA. III. Remarks. This genus1 is on the whole allied to Vannoniscus, but differs in the following particulars: the antennulse are normal, not terminating in a vesicle; a pair of epimeral i mnd at first segment, sixth and seventh segments dorsally marked off from one another, and the female operculum is very small. Only a single species is known. 52. Nannoniscella groenlandica n. sp. (PI. VII, figs. 7 a— 7 d; PI. VIII, figs. 1 a 1 1 Female (with marsupium). Body about two and a half times as long as broad; lateral mar- gins of the major anterior part, viz. head and the four thoracic segments, are, taken as a whi somewhat convex, while the remaining thoracic segments and the anterior part of the abdomen ha the lateral margins nearly parallel, and this posterior section is conspicuously narrower than or third segment. The front area on the head is subtriangular, with the anterior end broadly obtuse and rounded; its lateral margins without keels. The protruding lateral plates reach about to the base of the anten- nal squama (fig. lb) and have the end acute. -- Antennulse a little shorter than the head; Qagellum ^-jointed, scarcely as long as the two distal peduncular joints combined. — Antennae with the peduncles almost half as long as the animal; sixth joint considerably longer than the fifth and almost . as the flagellnm, which is 8-jointed, with first joint about as long as the sum of the five following joints; squama nearly as long as third peduncular joint, much longer than broad, acute. The four anterior thoracic segments have their antero-lateral angles acute, but scarcely duced and without any terminal spine. The epimeral plates with the end very obliquely cut off, acute. Abdomen almost half as broad again as long; the major part of the lateral margins and the terminal margin constitute a" semicircular outline. - Operculum tig. 1 E) very small, longer than broad, increasing in breadth from the base to far bevond the middle, somewhat ovate, but the median half of the posterior margin is a little concave, and the whole posterior margin has a number of long setae. — Uropods with the peduncle narrow at the base and widening much to the very oblique end; endopod slightly longer than the peduncle, somewhat longer and thicker than the exopod. Length of the ovigerous female rz mm. Remarks. The female is easily distinguished from ever) other form of northern Asellota bj a number of features; unfortunately the male is not known. Occurrence. Taken by the "Ingolf" at a single place. West Greenland: Lat. 64°03' N., month of Ameralik Fjord, 5—70 fath., shells; ' After Pis. VII— VIII had been finished, I discovered that Vanhoffen in .914 had established the genus Austronix. with three antarctic species. It is not improbable that a more thorough examination of various features show that one or two - but scarcely the third - of his species are congeneric with my that be the case, the generic name given by me must be cancelled. Hut at our present state of knowledge 1 ; the name Nannoniscella until the male of my new species has been found and some points in the structure of the tho. segments, etc. of Vanhoffen's animals have been investigated. 86 CRUSTACEA MAI,ACOSTRACA. III. Nannoniscoides n. gen. Description (based on a mutilated male). Body considerably depressed. Head peculiarly shaped; the front area is limited by a pair of exceedingly high and thick keels, and at each side the head is produced into a very large, thick, oblong, subtriangular process directed somewhat outwards and then curved forwards, so that the proximal parts of the antennulse and antenna; are situated in a hollowing between the process and the keel. -- Antennulse inserted considerably from the lateral margins, short; first joint thick and somewhat short; second joint slender and rather short; flagellum with few joints. Antenute lost, excepting the four proximal joints; squama very short. -■ Mouth-parts nearly as in Nannoniscus ; right mandible (fig. 2 c) has the incisive part somewhat long, several seta', and the molar process moderately short, tapering to the moderately narrow, setiferous end, and directed considerably backwards; maxillipeds (fig. 2 e| long, second joint a little more than twice as long as broad, and its long lobe twice as long as broad; third and fourth joints very broad, and the fourth about as broad as long; fifth joint considerably narrower than the fourth; sixth joint short and some- what broad, without any real lobe; seventh joint small; epipod triangular, long, acute, slightly more than three times as long as broad. First thoracic segment considerably narrower than the head, with the sides rounded and no epimeral processes; second segment much broader than the first, at each side produced forwards into a large, triangular jjrocess directed forwards and a little outwards, while the other segments have no processes. Third segment is narrower than the second and broader than the fourth, and from the fourth the segments increase in breadth to abdomen. Seventh segment is short, and the median part of its dorsal surface so completely fused with sixth segment that no limit is seen. — (Legs mutilated). Abdomen broad, much broader than long. Male operculum somewhat large (fig. 2 d), with the median lamella very broad and near the end produced much outwards. — Uropods ventral, very small, biramous. Remarks. This genus is easily separated by the shape of the head and the anterior thoracic segments from all other forms. It is intermediate between Nannoniscclla and Nannoniscus, agreeing with the former and differing from the latter of these genera in having the antennulse normal, while it agrees with the latter and differs from the former in having the median dorsal part of sixth and seventh thoracic segments completely fused. Only a single species is known. 53. Nannoniscoides angulatus 11. sp. (PI. VIII, figs. 2 a— 2e.) Male (mutilated). Bod) not fully three times as long as broad, with the lateral parts some- what expanded. — Head (fig. 2 b) almost twice as broad as long; the thick and high upper keels con- verge somewhat forwards and protrude considerably beyond the anterior transverse and straight mar- gin of the front area. The lateral processes are deep, being triangularly keeled above, considerably longer than broad, with the end obtuse. - Antennulse somewhat shorter than the head; flagellum CRUSTACEA MALACOSTRACA [II. 87 3-joiiited, not much shorter than the peduncle, and its terminal joinl considerably longer than the second. — The antennal squama very small, much broader than long. Lateral processes of second thoracic segment reach as far forwards as the front m first segment; each process terminates in a small spine inserted on the obtuse end. The peculiar fun- on the four anterior segments are shown 011 Kg. 2 a. Abdomen distinctly broader than the posterior thoracic segments, about one-third a again as long. The lateral margins proximally somewhat convex and then slightly converging to the postero-lateral angles which protrude as somewhat small triangular corners; the posterior margin is very long and considerably convex, excepting a somewhat short concave part towards the lateral angles. The median part of the ventral side of the two posterior thoracic segments is vaulted, with a moderately large, oblong, obtuse process (fig. 2 d) somewhat in front of the operculum, which much vaulted; the very broad median lamella of the operculum is posteriorly emarginate, and eacli postero-lateral part produced into a long, somewhat curved and narrow, acute process directed out- wards. — The uropods are very small, with the endopod a little longer than the exopod. Length of the single specimen 2-2 mm. Remarks. The male of this species is abundantly distinguished from other Asellota by the keels and processes on the head, the two anterior thoracic segments and the shape of the abdomen Unfortunately the female is unknown. Occurrence. Taken by the "Ingolf in the cold area. North of the Faeroes: Stat. 139: Lat. 63°36' \\, Long. 7°3o' W., 702 lath., temp. : 01 Nannoniscus G. O. Sars. This genus was established in 1869 on females of a single species taken off the I. of,, ten In- lands. Several vears later Sars described another species, A', bicuspis G • > S., but in 1908 Hai Richardson with good reason removed it from Nannoniscus and made it the type for a new genu Haploniscus Rich. In 1897 Sars established X. caspius on a male from the Caspian S the state when Sars in his "Account" published description and figures of the type N. oblong GO thereby giving the first detailed account of a species of this genus. He described and fi« only the female but, besides, an animal which he believed to be the male of the same species, two specimens supposed to be males had the testes visible through the integuments com]., the f of this male from above on his PI. 50), while the operculum is an undivided round plate, shaped as in a female. According to his figure his male is in reality the female of a very diffe species, and the occurrence of testes in female specimens is certainly very interestinj that in the Cymothoin?e and some of the Epicaridea the specimens are first males and then fen furthermore, vestiges of hermaphroditism have been found in Sphc and i thing of this kind which has been found by Sars in his two specimens. cause I have both males and females of A', oblongus - and of four other >pecies of legs are always completely similar in both sexes, while in the specimen figured by 88 CRUSTACEA MALACOSTRACA. III. male of iV. oblongus first pair of legs are very thickened and, besides, armed with a few very strong spines, thus differing very much from the rather slender first pair in the females of N. oblongus; furthermore, not only the operculum but the outline of the body and the flagella of the antennce in the specimen figured by Sars as the male show exclusively female features. Consequently I consider the specimens described by Sars as being the males of N. oblongus to be in reality immature females with protandrous features in the internal sexual organs of a species with the first pair of legs very thickened; I propose the name N. crassipes for this peculiar form. Until 1914 G. O. Sars was the only author who had published anything based on personal observation on the genus Nannoniscits. In 1914 E. Vanhoffen established (in: Die Isopoden der Deut- schen Siidpolar-Exped. 1901 — 1903) two new species from the Antarctic Ocean, but descriptions and figures are rather poor. My material from the "IngolP area is extremely rich, comprising 13 species, 12 of which are new; of five species both males and females, of eight species only females, are to hand. Sars' figures of the female of N. oblongus and of the male of N. caspius convey together an excellent idea of the genus, but the rich material enables me to add some remarks to his description of the genus, and to point out some sexual differences. The outline of the body differs somewhat in the two sexes. In the females at least the three anterior thoracic segments are somewhat or considerably broader than the three posterior segments and abdomen, while in the males the anterior segments are a little narrower, but the posterior seg- ments and abdomen broader, than in the other sex, with the result that in the males the posterior half of the body is almost or completely as broad as the anterior portion (PI. YIII, fig. 3 a; PI. IX, fig. 2 a). The front area on the head is marked off laterally, and in most species by keels, each ter- minating anteriorly in an acute angle; the anterior margin of the front area is straight, concave or convex, and the length of this margin differs always in the sexes, being longer in the females than in the males, and it affords excellent specific characters. The antennulae are extremely characteristic (many figures on Pis. VIII — X), differing materially from those in any other Isopod, but the representations published by Sars are not quite correct. They are always rather short and consist of five joints. First joint is large, thick and more or less oblong; second joint is more or less slender at the base and increases in thickness to the end, which is cut off and, besides, produced in probably always three short protuberances or somewhat long processes; third joint is extremely short and thin, and not unfrequently difficult to perceive, while the fourth joint, which is short, is at the inner side of the end produced into a very conspicuous, oblong, oblique process. Fifth joint is always a vesicle, ovate (for inst. figs. 2d and 7c on PI. IX) or, rarely, very oblong (PI. IX, fig. 4 1)), at least frequently with a couple of fine hairs on the distal part of the lower side; it is attached to the end of fourth joint by an extremely narrow articulation. The antennae have the squama distinct, sometimes quite small, in Y. oblongus and especially iu N. caspius long; it is most frequently, but not always (as in A', oblongus, PI. VIII, fig. 4f) marked off by a suture from third joint Fifth and sixth joints moderately long. The flagellmn is slender and normal in the females; while in the males a considerable part (PI. YIII, fig. 6d| or almost the whole flagellmn (PI. IX, fig. 6a) is thickened, and in the thickened part the joints are fused and not discernible. — The mouth-parts have been well described by Sars. CRUSTACEA MALACOSTRACA. III. All thoracic segments movable excepting sixth and seventh segments, as these are immovabl) fused both above aiid below, the median half of the articulation between them having completely vanished, while a suture or articulation is distinct towards the lateral margin (PL \ III 3a and 6a; PI. IX, figs. 2a, 3 a, and 7a), a structure overlooked by Sars, who has drawn this articulation as if it were complete; the figures cited show, besides, that the articulation between fifth and sixth ■ ments is not straight, but peculiarly bent. The absence or presence of a high tubercle or proa of two processes, on the posterior part of the ventral surface of the thorax affords excellent specific, but scarcely sexual, characters. — First pair of legs are always a little and sometimes considerably much thicker than the following pairs; in most species this pair have few spines oi verj moderate length, but in V. inermis n. sp. and N. crassipes n. sp. they are more or less prehensile organs with long, strong spines on fifth joint. An accessory claw is frequently discernible and is sometimes strong. In the males known to me of three species and in the females of .V. inermis n. sp., the three postei pairs of legs have a number of very or extremely long natatorj seta along the upper margin of fifth and sixth joints (PI. VIII, fig. 3I1; PL IX, figs ia and 4 ei, while in the males of N. reticulata* n. sp. and A', plebejus n. sp. and in females of all species excepting N. inermh no such seta; were found. The shape of the posterior part of abdomen differs sometimes in the two sexes (PL VIII, I 3i and 3 k). In both sexes a considerable or large part of the ventral surface of the abdomen is not covered bv the operculum, which is nearly circular or oblong; in the male the median lamella is some- what broad, and its distal part affords specific characters. The uropods are ventral, moderately small, biramous; the eudopod considerably or, rarely, slightly longer than the exopod. Remarks. The best generic character is the very curious structure of the distal part of the antennulse; their terminal vesicle is, as far as I know, without parallel in any other Crustacean. None of the species ■ — excepting N. caspius (i. (). S. — is found in low water, and most of them are re; deep-sea forms. The males seem generally to be much more rare than the females, but that may be explained by the supposition that most of them, at least to some degree, have natatory habits. I think that future and really thorough deep-sea investigations will lead to the discovery of a great nuir of species in all oceans. The females of the thirteen species described here, together with N. crassipes 11. sp., ma) be pivided into two sections, and determined by the aid of the following kej S. Sect. I. Abdominal operculum with an acute process on the lower surface (sometimes may be partly broken off, but the irregular shape of the remaining protuberance reveals that it has existed). Sect. II. Abdominal operculum without any tubercle or process on the lower surface. Section I. A. Antero-lateral angles of first thoracic segment terminating in a spine; the ang econd ment with a short, fine seta B. Antero-lateral angles of first segment without any terminal spine, but with or withou angles of second segment with a real spine. a. Surface of the body without reticulation. The Ingolf.Expedition. III. 5. go CRUSTACEA MALACOSTRACA. III. a. First or second segment not half as broad again as sixth segment. i;. Abdomen with the margin just outside the base of each nropod convex or straight. f. Antennal squama long at least about as long as the diameter of third peduncular joint. Head at least six times as broad as the distance between the anterior ends of the keels limiting the front area 2. N. ob/oi/gus G. O. Sars. ff. Antennal squama conspicuously shorter than the diameter of third peduncular joint. Head about four and a half times as broad as the distance between the anterior ends of the keels limiting the front area 3. N. arcticus n. sp. iji;- Abdomen with the lateral margin just outside the base of each uropod considerably con- cave, as the posterior median part of the abdomen is produced backwards. 4. N. analis n. sp. j3. First and second segments conspicuously more than half as broad again as sixth segment. 5. JV. laticeps n. sp. b. vSurface of the body with a reticulate network 6. N. rcticulatiis n. sp. Section II. A. Second joint of the antennulae terminates in short or moderately long processes leaving nearly the entire fourth joint uncovered. Abdominal operculum posteriorly rounded or straight. a. Posterior ventral area of the thorax without any process. a. Antennular vesicle very long and narrow, between two and a half and three times as long as broad. Uropods with the endopod much longer than the exopod... 7. A', inermis n. sp. ft. Antenmilar vesicle pvriform, not half as long again as broad. Uropods with the endopod slightly longer than the exopod 8. A", cequiremis n. sp. b. Posterior ventral area of the thorax with a large, curved, acute process, but without anv se- cond process. a. Postero-lateral margin of the abdomen without any incision or tooth. S. First pair of legs moderately slender, without strong spines on the lower margin of fifth joint 9. A7, plebejus n. sp. §§. First pair of legs extremely thick, with two very long and robust spines on the lower margin of fifth joint 10. A^. crassipes u. sp.1 ft. I'ostero-lateral margin of the abdomen at each side with an incision and a very conspicuous triangular angle or tooth n. .V. minutus n. sp. C. Posterior ventral area of the thorax with an extremely large, curved, acute process, and in front of this in the median line a long, slender and straight process 12. A', a r mat us n. sp. B. Second joint of the antennulte terminating in long, oblong-triangular processes overlapping the fourth joint. Operculum posteriorly angular or even produced in an acute process. a. Anterior margin of the front area nearly half as long as the breadth of the head. Fourth joint of the six posterior pairs of legs terminating in a short seta 13. A', spinicornis n. sp. 1 This species, which has been briefly mentioned <>n p. S7--S8, is established 011 the so-called male of N.oblongui figured by Sars on PI. 50. His figures and description arc quite sufficient for recognizing this fine species, which has not been taken out of Norway. CRUSTACEA MALACOSTRACA III. q, b. Anterior margin of the front area considerably less than hall as long as the breadth of the head. Fourth joint of the six posterior pairs of legs terminating in a long and vei spiniform seta , , v# „,,,„,, n sp 54. Nannoniscus simplex n. sp. (PI. VIII, figs. 3 a -31). Female. Body as to general outline nearly as in .Y. oblongus G. O. S. (comp. tin figun ol Sars), from a little less to somewhat more than three times as long as broad, with second and third thoracic segments not much broader than the posterior segments, and the lateral margin- of fourth segment converging only moderately backwards. Head between four times and four and a half times as broad as the distance between the anterior ends of the keels limiting the front area (fig. 3 c), and the anterior margin of this area i straight. — Antennulae (fig. 3d) with first joint somewhat longer than broad; second joint slightly longer than the first and moderately slender, with the distal outer angle somewhat, the inner an slightly, produced; the process from fourth joint reaches at most the middle of the pyriform vesicle; the foot of each of the long setse on second joint transverse. Antennal squama well marked off, very oblong, acute, shorter than the diameter of third peduncular joint (fig. 3d). First thoracic segment with a well developed spine on each antero-lateral angle the male, fig. 3 b. — angles of second and third segments with a short, stiff seta. The two posterioi segments with a high, broad, rounded ventral protuberance (figs. 3 k and 3I) lying close at the front margin of the abdominal operculum, and without any process. — First pair of thoracic legs (fig. not much thicker than second pair; fifth joint on the lower margin with a seta beyond the middle, and at the end a long seta and a somewhat small spine; sixth joint with a spine before the middle and one at the end; accessory claw thin. .Second pair with about four short spines on the lower mar- gin of fifth joint, and four on the sixth joint. Seventh pair (fig. 3 f> has on the lower margin of fifth joint two moderately strong and two distal, long spines, three spines on sixth joint, but no 1 setse on these joints; accessor}- claw strong. Abdomen (fig. 3 k) with the lateral margins distinctly, but yet rather feebly, convex, converg to the somewhat narrowly rounded or subacute end. - - Operculum a little longer than broad, poster- iorly rounded, on the ventral surface with a high protuberance terminating in an acute pri vected backwards (figs. 3k and 3 1). — Uropods with the peduncle somewhat longer than broad; endopod distally thicker than at the base, and not fully twice as long as the exopod. Length of the largest adult specimen 2-6 mm. Male. Body a little more than three times as long as broad, with the lateral marc parallel (fig. 3 a). The head is about six and a half times as broad as the distance between the anterior ends of the keels limiting the front area (fig. 3 b), consequently the keels converge considerably than in the females, and besides they are higher, more protruding with their ends free; the margin between the keels is straight— Antennulae and proximal part of the antennae as in the fema (the flagella of the antennae lost). 92 CRUSTACEA MALACOSTRACA. III. The thorax, as already mentioned, differs somewhat from that of the female in outline, as first and second segments are only slightly broader than the sixth segment; the antero-lateral angles of first segment with spines, of second and third segments with setae as in the female; the rounded ven- tral posterior protuberance (fig. 3 i) as in the other sex. — First pair of legs (fig. 3 g) nearly as in the female; seventh pair (fig. 3 h) differs from those in the female in having the most distal spine on fifth joint short, while a long, stiff seta has replaced the long spine in the female ; furthermore the three posterior pairs have some long seta? on a part of the upper margin of fifth joint, and many extremely long setae on the upper margin of sixth joint. Abdomen (figs. 3 a and 3 i) differs in outline conspicuously from that in the female, as the lateral margins are scarcely or slightly convex, converge to a point somewhat beyond the insertions of the uropods, and then each margin bends abruptly much more inwards, thus constituting half of the posterior margin and being slightly concave; the end of abdomen is angular, but not acute. - Operculum slightly longer than broad; the median lamella is widened near the end, and each outer angle produced into a very conspicuous, slender process directed somewhat backwards and much out- wards; consequently the posterior margin of each pleopod is convex. — Uropods about as in the female. Length 1-9 mm. Remarks. As shown in the key (p. 89) A^ simplex is easily separated from the other forms of this section, and in reality from most of the other species of the genus seen by me, in having well developed spines on the lateral angles of first segment, and only stiff setae on the angles of second and third segments. In general appearance it is nearly similar to N. oblongus G. O. S., but the outline of the male abdomen is different, and the postero-lateral processes of the median opercular lamella afford a good character. In the marsupium of a female (from Stat. 90) a species of Sphmronella with ovisacs (family Choniostomatidaj) was found. Occurrence. Taken by the "Ingolf" at two stations in the warm area. West of Iceland: Stat. 90: Lat 64°45' X., Long. 29°o6' W., 568 fath., temp. 4-4°; 2 spec. ($). .South-West of Iceland: Stat. 78: Lat. 6o°37' N., Long. 27°52' W., 799 fath., temp. 4-5°; 10 spec. (6?, 4 c?). 55. Nannoniscus oblongus (i. O. Sars. (PI. VIII, figs. 4 a— 4 f). 1870. Nannoniscus oblongus ( ',. ( ). Sars, Forh. Vid. Selsk. Christiauia for 1869, p. 164. ' 1897. G. O. Sars, Account, II, p. 119; PI. 50. (partim; the animal described and figured as the male is the female of N. crassipes n. sp.). Female. Body from somewhat less to somewhat more than three times as long as broad, with second and third segments only one-fifth or one-sixth as broad again as the sixth segment, while the margins of fourth segment converge only moderately backwards. Head from six to eight, frequently about seven, times as broad as the distance between the CRUSTACEA MALACOSTRACA. III. anterior ends of the keels limiting the Eronl area (fig. 4a), and the short anterior margin of this ai is concave. — Antennulae (fig. 4 c) in the main as in A', simp/ex; first joint somewhat lunger than broad; second joint slender, a little longer than the first, with its terminal processes nearh broad; the process of fourth joint reaches beyond the middle of the pyriform vesicle; the i of the long setse on second joint oblong. -- Antennal squama not marked off at its than in any other species, excepting A', caspius G. 1 >. S.. from a little (fig. .) c\ to considerably or much longer than the diameter of third joint. Second thoracic segment (fig. 4a) with a moderately short spine on the anterolateral an. while the corresponding angles of first and third segments have a line or stiff, short seta. The ]>■ ventral median part of the thorax raised as a high and broad, rounded protuberance, which li' to the front end of the abdominal operculum, and has no process. The legs nearly as in N. Abdomen with the lateral margins somewhat convex and converging (fig. } e), and the • erior margin moderately broadly rounded at the end. -- Operculum about as long as broad, with the posterior margin nearly straight or even slightly concave, while the median proximal part of t!: surface is much vaulted and armed with a recurved, acute process. Uropods with the peduncle oblong, the endopod thicker and very considerably longer than the exopod. Length of full-grown specimens 2 — 26 mm. Male. Body about three and a half times as long as broad, with the anterior part of the thorax a little broader than the posterior. — The head is nearly nine times as broad as the distance between the anterior angles of the keels limiting the front area (fig. 4 b), as the keels conve; and protrude conspicuously with their acute ends. — Antennulae and proximal joints of the antenna as in the female (the antennal flagella lost). Thoracic leys as 111 the female, excepting that tin- three posterior pairs have natatory setse on fifth and sixth joints as in the male of \. very young male, ri mm. long, the natatory setae were wanting). Abdomen differs in outline from that of the female, as its median posterior part is prod backwards (fig. 4 c!) with the end somewhat narrowly rounded, and the margins somewhat end are distinctly concave. — Operculum a little longer than broad; the median lamella a little wid. towards the end, with each outer angle produced into a triangular tooth directed back little outwards, while the major terminal part of the lamella constitutes a kind of lobe wit margin semicircular. — Uropods with the peduncle distallv somewhat produced, otherwise nearly i in the female. Length of the largest male 2-2 mm. Remarks. N. ohlongus is easily separated from N. arcticus and N. analis in having 1 ably shorter distance between the anterior ends of the keels on the head, and the squama consider; longer; from N. simplex it is instantly distinguished in having spines on the anterolateral f second instead of first segment. As mentioned above, there is in the females some variation in the distance between tin of the keels of the front area and in the length of the squama. In most females from distance mentioned (fig. 4 a) is shorter than in specimens from Stat. 32, and it is still 1« specimens from Norway. As I was not quite sure that my reference of the "Ingolf" specimens to ' 94 CRUSTACEA MALACOSTRACA. III. was correct, because Sars has figured the female with the distance between the keels still shorter than in any of my specimens, and the operculum less cut off posteriorly, Prof. Sars, on my request, kindly sent me some fine females, but these have the frontal emarginatiou as broad as, or even a little broader than, in my specimens from Stat. 32, thus differing in this feature much from his figure, and the squama is in these Norwegian specimens much longer (fig. 4 f) than the diameter of third peduncular joint, longer than in any of the "Ingolf" specimens. After careful examination of the females to hand from various places I think that they all belong to A*, oblongus G. O. S. That the animal described and figured by Sars as the male is in reality a not adult female of a very different form, N. crassipes n. sp., is proved on p. 87 — 88. Occurrence. Taken >by the "Ingolf" at three stations in the warm area. Davis Strait: Stat. 32: Lat. 66°35' N., Long. 56^38' W., 318 fath., temp. 3-9°; 8 spec. (?). West of Iceland: Stat. 98: Lat. 65°38' N., Long. 26°27' W., 138 fath., temp. 5-9°; 1 spec, (young o*)- South-West of Iceland: Stat. 78: Lat. 6o°37' N., Long. 27°52' W., 799 fath., temp. 4-5°; 14 spec. (3 f the squama are easily observed and are certainly valid specific characters. While N. oblongus is known only from the warm area, .V a relic us is a cold water form. Occurrence. Taken bv the "Im'olf" at a sinele station in the cold area. CRUSTACEA MALACOSTRACA, III q. South of Jan Mayen : Stat 116: Lat. 70°05' X., Long. 8°26' W., 371 fath., temp. : 1 .-4°; 1 spi Besides, it has been taken h\ the Ilnd Amdrup Exped. at Greenland in Forsblad 1 Lat. 72° 1 7' N., 90 — 40 fath., 1 fine spec. 57. Nannoniscus analis n. sp. (PI. VIII, figs. 6a— 6k; PI. IX, figs. 1 a- . b). Female. Bod)' (fig. 6a) from three to nearly three and a half times as long as broad, with second and third thoracic segments considerably broader than, viz. even a little more than one-third as broad again as, the two posterior segments, while the lateral margins of fourth segment eonvei only moderately backwards, and the fifth segment is conspicuously broader than the sixth. Head (fig. 6 a) four times, or a little more than lour times, as broad as the distance between the anterior ends of the cephalic keels, and the anterior margin between these keels is feeblv com Antennulae (fig. 6c) nearly as in N. oblongus. -- Antennae nearly half as long as the body; Dagellum more than half as long again as fifth or sixth peduncular joint, 13-jointed, with first joint aboul long as the three following joints; squama marked off, conical and considerably shorter than the diameter of third joint. Second thoracic segment with a distinct, or even somewhat long, spine on the antero-lateral angles, while the angles of first and third segments have no spines, but frequently a short, fine The median ventral part of seventh segment raised as a high protuberance (figs. 6 i and 6k) lying close to the front margin of the abdominal operculum, and anteriorly stretching forward on the pi erior part of sixth segment, which, besides, has a tubercle of its own. — Legs somewhat similai those in N. simplex, but the accessory claw is more or less feeble on all pairs; the first pair (as in the male, fig. 6 e) are distinctly thicker than the second or third pair (fig. 6 fi, with two small spines on the lower margin of fifth joint, while fifth joint of third pair has five spines; the three posterior p (fig. 1 b) with still fewer spines than in X. simplex and no natatory sel Abdomen (figs. 6 a and 6 i) almost as long as broad, with the outline very characteris lateral margins outside the base of the uropods are conspicuously concave with an angular bend, because they turn somewhat suddenly inwards and then nearly abruptly backwards, e posterior part of abdomen is produced almost as a lobe rather broadly rounded at the end. - culum (figs. 6 i and 6 k) slightly longer than broad, posteriorly rounded; its proximal median part highly raised and produced into a long, strong, acute process directed backwards and consid< downwards. — Uropods (fig. 6 i) with the peduncle short, the endopod somewhat slender and not quite twice as long as the very slender exopod. Length of large specimens 27 mm. Male. Body not quite three and a half times as long as broad; the antei thorax only a little broader than the two posterior segments. - Head (fig. 6b) between and a half times as broad as the distance between the ends of the cephalic keels, and margin of the front area is a little concave. -- Antennulae as in the female. Antenn as in the female excepting the flagellum, which is extremely different; the Dagellum has 96 CRUSTACEA MALACOSTRACA. III. part extremely thickened, oblong-oval and as long as the four following slender joints combined (the end of flagellnm lost). Thoracic segments, excepting as to the breadth mentioned, nearly as in the female; the ventral posterior protuberances scarcely as high as in that sex. — The three posterior pairs of legs (fig. i a) with about four long setse on the upper margin of fifth joint, and a good number of very or extremely long setae on sixth joint. Abdomen posteriorly nearly as produced as in the female, but rather gradually (fig. 6 h), so that the distal third of the lateral margins is flatly and feebly concave, while the end is more narrowly rounded than in the female. — Operculum conspicuously longer than broad; the median lamella a little widened near the end, with the distal outer angle a little produced, triangular, while most of the terminal part of the lamella constitutes a kind of rounded lobe, divided of course in the median line. — Uropods about as in the female, excepting that their peduncles are a little longer. Length 2-i mm. Remarks. The female is instantly separated from all other species by the shape of the post- erior part of abdomen, while the male may be distinguished by some features taken together: the breadth of the frontal emargination, the somewhat short antennal squama, the shape of the abdomen and of the median lamella of the operculum. Occurrence. Taken by the "Ingolf" at a single deep station in the warm area. Davis Strait: Stat. 24: Lat. 63°o6' N., Long. 56°oo' W., 1199 fath., temp. 2-4°; numerous spec. (x6 o* and at least 50 5). 58. Nannoniscus laticeps n. sp. (PI. IX, figs. 3a-3di. Young Female (with seventh pair of legs not fully developed). -- Body (fig. 3 a) about three and a half times as long as broad. Head and the two anterior segments about two-thirds as broad again as sixth segment and abdomen; third segment not much narrower than the second, and some- what broader than the fourth, which has its lateral margins converging extremely backwards. Head (fig. 3 b) fives times as broad as the distance between the anterior ends of the cephalic keels, and the anterior margin of the front area slightly concave. — Antennulrt (fig. 3 c) with first joint a little longer than broad; second joint distinctly longer than the first, somewhat thick, with the terminal protuberances rather short; the process from fourth joint reaches conspicuously beyond the middle of the uncommonly small, oblong vesicle. — Antennal squama marked off, short, much shorter than the diameter of third peduncular joint. Antero-lateral angles of second segment terminate in distinct spines (fig. 3b), while the angles of first and third segments terminate in fine setae. The posterior median part of the ventral side considerably vaulted. Abdomen (figs. 3 a and 3d) conspicuously longer than broad, posteriori)' very broadly rounded, and the lateral margins nearly parallel. -- Operculum somewhat longer than broad, with the posterior margin considerably convex and, besides, almost angular, as a small median part is feebly produced; a spiniform, moderately small process projects on the ventral surface somewhat near the base, and at CRUSTACEA MALACOSTRACA. III. q? least in my single specimen it is directed nearly horizontally backwards. (Jropods with th< distinctly more than half as long as the endopod; that the rami are uncommonly robust maj probably at least to some degree, be dne to the young stage of the animal. Length of the young specimen 1-4 111111. Remarks. The extreme breadth of the head and two anterior segments as compared with the posterior segments renders it impossible to refer this specimen to any of the oth 1 this section (viz. forms with a ventral process on the female operculum), even if we suppose that a part of the difference may be due to age. Furthermore it differs materially in the shape oi the operculum. Occurrence. Taken by the "Ingolf" at a single station in the cold area. North of Iceland: Stat. 126: Lat. 67°ic/ NT., Long, ^a' W., 293 fath., temp. -^o-5c: 1 59. Nannoniscus reticularis n. sp (PI. IX, figs. 2 a— 2 1). Female (without marsupium). The body has the upper surface nearly everywhere adorned with a most conspicuous reticulated network (fig. 2 c); it is from somewhat less to somewhat more than three and a half times as long as broad, and the two anterior thoracic segments are somewhat, though not much, broader than the sixth or the seventh segment. Head (fig. 2 c) about four times as broad as the distance between the anterior ends of the cephalic keels, and the anterior margin of the front area is rather deeply concave - The antennulae (fig. 2d) differ from those in N. oblongus in having second joint somewhat thicker, and the proci from fourth joint shorter, scarcely reaching the middle of the pyriform vesicle. — Antennae lost, except- ing the proximal joints; squama oblong-triangular, much shorter than the diameter of third joint. Second thoracic segment (fig. 2 c) with the lateral parts produced considerably forwards, the end cut off and armed with a proportionately long spine, while first and third segments have their lateral parts much less produced, and on their angles at most short setae. Median ventral part of the three posterior segments very characteristic, especially when seen from the side (fig. 2 i); fifth segment raised in a thick, high, distally rounded protuberance; seventh segment vaulted as a very high pro- tuberance having a kind of incision in the posterior half, which is lower than the anterior half, and the latter is cut off and angular, with its posterior angle even sometimes produced a little backwards the protuberance of sixth segment is somewhat lower than that of the seventh, subcorneal, with end obtuse. — Legs normal; first pair (fig. 2 e) distinctly thicker than the following pairs, with I well developed spines on fifth and on sixth joints, and the accessory claw slender: the six other pai have only some few spines, and the three posterior pairs no natatory s< Abdomen (figs. 2I1 and 2 i) as long as, or a little shorter than broad, posteriorly fi scarcely produced, with the margins outside the uropods straight or slightly convex, and the pos end rounded. — Operculum slightly longer than broad (fig. 2 hi, posteriorly at the middle feebly distinctly, produced in a broad angle; its proximal part, seen from the side (fig. 2 i) much produced in a process of a very moderate size, curved horizontally backwards, acute, exopod distinctly more than half as long as the endopod. The Ingolf-Expedition. III. 5. 98 CRUSTACEA MALACOSTRACA. III. Length of a specimen without marsupium 2 mm. Male. Surface reticulate as in the female. Body scarcely three and a half times as long as broad ; the difference in breadth between the anterior and the posterior segments (fig. 2 a) is smaller than in the female, and fourth segment is not only narrower than the first, but even a little narrower than the seventh. — Head between six and a half and six times as broad as the distance between the ends of the cephalic keels; the distal half of these keels is much more convex than in the female, and the anterior margin of the front area is extremely concave (fig. 2 b). — Antennulae and proximal joint of antennae as in the female. The thoracic segments, excepting as to the breadth already mentioned, nearly as in the female; even the ventral protuberances of the two posterior segments scarcely differ from those in the female, but the protuberance on fifth segment (fig. 2 k) is distinctly lower. — Legs as in the female; no natatory setae 011 the three posterior pairs. Abdomen (figs. 2 a and 2 1) distinctly broader than long and posteriorly scarcely produced. — Operculum a little longer than broad; the median lamella much broader before the middle than towards the end, with the outer distal corner angular, and the terminal margin of each pleopod oblique. — Uropods nearly as in the female. Length 18 mm. Remarks. N. reticulatus is instantly distinguished from all other species hitherto known by the very conspicuous network on nearly the whole dorsal surface. The frontal emargination is un- commonly deep, especially in the male. Occurrence. Taken by the "Ingolf" at a single station in the cold area. North of Iceland: Stat. 126: Lat. 67°ic/ N., Long. i5°32' W., 293 fath., temp. -=- o^0; 6 spec. (3 o*. 3 ?). 60. Nannoniscus inermis 11. sp. (PI. IX, figs. 4 a— 4 f). Female. Body about three and one-third times as long again as broad. Second and third thoracic segments nearly one-third as broad again as sixth or seventh segment, and the lateral margins of fourth segment converge only moderately backwards. Head a little more than four and a half times as broad as the distance between the anterior ends of the cephalic keels (fig. 4 a), and the anterior margin of the front area is slightly convex. — Antennulse (fig. 4 b) very characteristic; first joint longer than broad; second joint a little longer than the first, rather slender and slightly thicker towards the end, which has the angles, at most, feebly protruding; third joint is unusually long and totally visible; fourth joint with the process somewhat long, hut reaching slightly beyond the middle of the vesicle, which is uncommonly narrow and long, between two and a half and three times as long as broad. — Antennae slightly more than half as long as the body; fifth and sixth peduncular joints about equal in length, slender and somewhat thickened at the end; flagellum somewhat longer than sixth joint, with more than its proximal half unjointed, while the distal part is divided into 8 joints; squama small, triangular, slightly longer than broad, acute (fig. 4 b). Second thoracic segment with a rather slender spine on the antero-lateral angles, and a nearly CRUSTACEA M VLACOSTB VC V. III. ,,,, similar spine is found on each angle of third segment, while no spine or seta exists on the angles "i first segment. Median ventral part of the posterior segments (fig. 4 fi considerably vaulted, but without any protuberance or process. -- Legs of first pair (fig. 4 c) very considerably thicker and shorter than second pair (fig. 4d); especially fifth joint is thickened with six strong spines, the third and the fourth long; sixth joint somewhat thick with two spines of moderate size; accessory claw slender. Second (fig. 4d) to fourth pairs are slender; fifth joint with six or seven spines; sixth joint with about foul spines; seventh joint with claw slender, and the accessory claw very small. The three posterior pairs (fig. 4 e) with some spines on fifth joint; seventh joint with claw very slender, and the ace. law very short and thin; furthermore the fifth joint has alon<4 the half of the upper margin about seven very long setse, and above on the sixth joint a row of extremely long natatory setae. Abdomen (fig. 4 f) nearly as long as broad; the lateral margins converge feebly from before the middle to the posterior margin, which consequently is long; each half of this margin is scarcel) "i feebly concave, as the median part of abdomen is posteriori) slightly produced backwards with the terminal margin somewhat convex. — Operculum about as broad as long, without any ventral pr< or protuberance; the posterior margin is straight. — Uropods with the rami slender and somewhat long; the endopod almost twice as long as the exopod. Length of a large female with marsupium 3-3 mm. Remarks. N.inermis, which is the largest species known, is in general aspect, excepting to the posterior part of abdomen, somewhat similar to N. analis, but it is easily separated from this species by the shape of the posterior part of abdomen, and by having 110 process on the operculum. Besides, it is distinguished from all species of the genus by the elongate vesicle of the antennuhe, by the spines on first pair of legs, and finally by the fact that the female has natatory setse on fifth and sixth joints of the three posterior pairs of legs, a feature otherwise only found in males. Occurrence. Taken by the "Ingolf" at a single deep station in the warm area. Davis Strait: Stat. 24; Lat. 63°o6' N., Long. 56°oo' W., 1199 fath., temp. 240; 3 spec (?). 61. Nannoniscus aequiremis n. sp. (PI. IX, figs. 5a-5h). Female. Body about three and a half times as long as broad; second, third and fourth ments are rather a little broader than the posterior segments, consequently the animal i slender than N. oblongus, but otherwise similar in aspect. Head (fig. 5 a) nearly more than eight times as broad as the distance between the anterior ends of the cephalic keels, and the anterior margin of the front area is considerably concave, (fig. 5b) somewhat thick; first joint somewhat longer than broad; second joint scarcely longer the first, with the distal part rather thick, the process at the outer side somewhat large, nearly longer than thick and somewhat longer than the inner process; third joint scarcely discernible; fourth join thick, with the process short and thick, reaching about to middle of the vesicle, which is pyrifon and only a little longer than broad. -- Antennal squama somewhat small, triangul than broad, acute. IOO CRUSTACEA MALACOSTRACA. III. Second thoracic segment with small spines on the antero-lateral angles; the angles of first segment with a small seta, and somewhat similar seta; on the angles of third segment. Median ventral part of the two posterior segments (fig. 5 f) considerably vaulted; on seventh segment is found a some- what long and broad protuberance with a transverse impression (fig. 5 g) but without any acute process, and it is separated by a rather deep impression from the rounded and distinctly lower protuberance on sixth segment. — Legs moderately slender; first pair (fig. 5 c) only a little thicker than second pair, without any spine on fifth joint and with the accessory claw very distinct; the other pairs have few and nearly setiform spines, and the accessory claw strong (figs. 5d and 5e); the three posterior pairs without natatory seta;. Abdomen (figs. 5 f and 5 g) a little broader than long, posteriorly broadly rounded; the posterior half of the lateral margin and the entire hind margin everywhere convex, without any concave or straight portion. -- Operculum a little longer than broad, with the proximal half highly vaulted, but without any tubercle or process; the posterior margin straight. — Uropods with peduncle short and broad, the endopod considerably thicker, but only a little or slightly longer, than the exopod. Length of a female with marsupium 1-9 mm. Remarks. N. aequiremis is distinguished in having the exopod of the uropods conspicuously longer in proportion to the endopod than in any other species of the genus. Occurrence. It has not been taken by the "Ingoir, but by the Ryder Expedition at a single place. South of Jan May en : Lat. 70°32' N., Long. 8cio' W., 470 fath., clay with many small stones. 2 spec. ($). 62. Nannoniscus plebejus 11. sp. (PI. IX, figs. 6 a— 6 g). Female (without marsupium). Body a little more than three and a half times as long as broad; second thoracic segment a little more than half as broad again as the seventh, and the lateral margins of the fourth segment converge somewhat backwards. Head (fig. 6 b) a little more than four times as broad as the distance between the anterior ends of the cephalic keels, and the anterior margin of the front area scarcely concave. — Antennulae (fig. 6 c) with first joint considerably longer than thick; second joint about as long as the first, rather thick, with the outer terminal process moderately large, about as long as broad and scarcely larger than the inner process; third joint distinct; the fourth moderately thick, with its process reaching beyond the middle of the oblong-pyriform vesicle, which is somewhat or slightly less than twice as long as broad. Antennae, judging from the length of the moderately thick fifth joint — the sixth joint with flagelhun lost -- probably somewhat short; squama small, triangular, acute (fig. 6 c). Antero-lateral margins of the three anterior segments (fig. 6 b) without spines, each terminating in a somewhat small, stiff seta. The median ventral part of seventh segment (fig. 6f) is highly vaulted, forming an oblong, large protuberance, which posteriorly embraces the base of the operculum, is rounded anteriorly, and a little from the front end is produced into a strong, long, curved, acute process with its terminal part directed backwards. Thoracic legs simple; first pair (fig. 6d), though moderately CRUSTACEA MALACOSTRACA. III. slender, yet decidedly stronger than the other pairs (fig ill pairs with very few spines, while the accessory claw is moderately developed on first pair, but somewhat robust on the other pairs; natati seta; wanting. Abdomen (fig. 6 f ) as long as broad and posteriorly a little produced, with the result that the end is somewhat narrowly rounded, and the parts of the margins a little before the end or slightly concave. — Operculum a little longer than broad, with the proximal half <>f the ventral surface much vaulted, but without any tubercle or process; the posterior margin is convex. pods somewhat large; peduncle oblong; endopod somewhat robust and twice, or more than tv long as the slender exopod. Length of a female without marsupium, but having the ovaries swollen with 6 mm. Male. Bodv about four times as long as broad; first thoracic segment (fig. 6a) broad as the second, and almost or fully half as broad again as seventh segment, which i cuously broader than the front end of fifth segment. Head five to five and a half times as broad as the distance between the anterior ends cephalic keels, and the anterior margin of the front area is a little concave. Autennuke ; female. — Antennae (fig. 6a) not reaching the posterior angle of fourth thoracic segment; filth pedun- cular joint proportionately longer and conspicuously thicker than in the female, sixth joint longer than the fifth and gradually thicker towards the end; flagellum nearh as long as sixth joint, somewhat fusiform and thickest at the end of the proximal third, unjointed excepting the termii slender part, which is marked off as a joint; squama as in the female. Antero-lateral angles of the three anterior segments with seta.- as in the female; the ventr protuberance of seventh segment nearly as in the female, excepting that the process Legs as in the female; no natatory setce on the posterior pairs. Abdomen (fig. 6 g) shaped as in the female. — Operculum a little longer than broad; lamella broad, moderately tapering and slightly widened at the end; the posterior margin pleopod somewhat sinuate, and the postero-lateral margin not produced. - - Uropods as Length ra mm. Remarks. According to the specimens to hand A'. plebejtis must be a small species. It dii from all preceding forms and agrees with N. crassipes n. sp, A. minutus n. sp. and X. ar in possessing a large, curved, acute process on the vaulted ventral part of seventh thora< but from these three species it is easily separated by the characters in the key (p. 90). Th the body, no spines on the antero-lateral angles of any of the three anterior thoracic segments, a the uropods, afford other good characters. Occurrence. Taken by the "Ingolf" at a single station in the warm area South-West of Iceland: Stat. 78: Lat. 6o°37' N, Long, zfv' W, 799 &«!., temp. 45* (4 o*, 3 ?)• IQ2 CRUSTACEA MALACOSTRACA. III. 63. Nannoniscus minutus 11. sp. (PI. IX, figs. 7 a— 7 h). Female. Body (fig. 7a) about four times as long as broad; first segment slightly broader than the second and half as broad again as seventh segment, which is at least as broad as the front end of fifth segment; the major part of the lateral margins of fourth segment converge moderately backwards. Head (fig. 7 b) about three times as broad as the distance between the anterior ends of the cephalic keels, and the anterior margin of the front area a little convex. — Antennulae (fig. 7 c) with first and second joints nearly as in X. plebejits ; fourth joint with the process long, reaching conspicu- ously beyond the middle of the pyriform vesicle, which is about half as long again as broad. — An- tennae in a female with young in the marsupium reaching about the anterior angle of third thoracic segment; the two distal peduncular joints proportionately short and moderately thick; flagellum much longer than sixth peduncular joint, 8-jointed, but its first joint is nearly as long as the three following joints combined, and somewhat thickened; squama small, triangular, acute. First thoracic segment (fig. 7 b) with an uncommonly long and moderately strong spine on each antero-lateral angle; second segment with minute spines, and third segment with short, stiff seta; on the angles. Median ventral part of seventh segment vaulted and produced into a strong, long, very curved, acute, almost hook-shaped process (figs. 7g and 7 hi. — First pair of legs (fig. 7 d) some- what robust, while the other pairs (fig. 7 e) are slender; fifth joint of first pair with a feeble spine at the middle of its lower margin, and a more developed spine at the end; the accessory claw somewhat slender on all pairs; the three posterior pairs without natatory setae. Abdomen (figs. 7 f, 7 g and 7 h) somewhat longer than broad; the second third parts of the lateral margins converge a little to a point outside and slightly before the base of the uropods, where each margin suddenly is incised, forming a very conspicuous, sharp angle or even a low, triangular tooth; the posterior margin, which is moderately broadly rounded at the end, runs to the bottom of each incision. — Operculum (figs. 7 g and 7 h) a little longer than broad, posteriorly very broadly rounded, with the proximal half moderately vaulted and without any tubercle 01 process. — Uropods somewhat slender; endopod not quite twice as long as the exopod. Length of a female with young in the marsupium 15 mm. Remarks. N. minutus is a small species easily separated from all northern forms by the incisions in the margin of the abdomen. In this respect it agrees with the antarctic form N. bidcns Vanh., which has been insufficiently described and figured, but so much can be seen, that this species is much larger than N. minutus, and its antennae proportionately much longer than in our form. Occurrence. Taken by the "Ingolf" at a single station in the warm area. Davis Strait: Stat. 25: Lat. 63°3o' N., Long. 54°25' W., 582 fath., temp. 3-3°; 3 spec. (?). 64. Nannoniscus armatus 11 sp. (PL X, figs. 1 a— 1 fi. Young Female (last pair of legs only half developed). -- Body (fig. 1 a) somewhat less than four times as long as broad, with the anterior half and especially first and second thoracic segments CRUSTACHA MALACOSTRACA. Ill broad, while the major posterior half is narrow, as second segment is even slightly more than half as broad again as sixth segment, which is somewhat broader than the front end of fifth segment and a little broader than fourth segment; the major portion of the lateral margins of third segment con- verge exceedingly, those of fourth segment very considerably backwards, while the margin- of fifth segment converge a little forwards. Head (fig. i b) nearly four times as broad as the distance between the anterior ends of the cephalic keels, and the anterior margin of the front area is nearly straight Antennuke, could be ascertained, have the second joint longer than the first and rather thickened towards the end, which has the inner process well developed and the outer small; the process of fourth joint i long, the vesicle somewhat small, pyriform. - Anteunal squama small, triangular. Second thoracic segment seems to have only a stiff seta on the anterolateral an angles of first and third segments no setae or spines could be discovered. The ventral median part of seventh and the posterior part of sixth segment much vaulted and produced into a ver\ strong and extremely long process (figs, i e and i f), which is directed downwards and somewhat backw while its most distal, acute part is curved much backwards; in front of this process projects from the anterior part of sixth segment a long, slender, acute process (fig. I e) directed vertically downwards — First pair of legs (fig. i c) very moderately robust and slightly thicker than second pair; filth joint with two setse on the lower margin, sixth joint conspicuously longer than the fifth; accessory claw well developed. .Second (fig. id) to fourth pair very characteristic, as the fifth joint has on the lowei margin a moderately long spine at the middle, a very long, slender spine near the end, and s before the latter a long third spine; sixth joint has the distal half somewhat thick. The posterior pairs distinctly more slender than the others, without natatory setse. Abdomen (figs, ia and if) a little longer than broad, subtriangular, with the end some* narrowly rounded and the lateral margins rather feebly convex. -- Operculum distinctly long, with the posterior margin semicircular, and without any tubercle or process on the ven face. - Uropods with the exopod small, considerably less than half as long as the endopod, \ in the young specimen is somewhat thick. Length of the specimen, which has seventh pair of legs only half-grown, r6 mm. Remarks. JV. armatus is easily distinguished from all other species of the genu a long, slender, straight process on sixth segment in front of the very long, strong and cess of seventh segment. The four anterior pairs of legs afford further valuable characters the shape of the body in the adult female may differ somewhat from that of the young specii described - probably the posterior half as the bodv is proportionately somewhat less adult - the features pointed out will certainly prove to be sufficient for determination. Occurrence. Taken by the "Ingolf at its deepest station in the warm area. South of Davis Strait: Stat. 38: Lat. 59°"' N, Long. 51*05' W, 1870 fath., temp. r3' I04 CRUSTACEA MALACOSTRACA. III. 65. Nannoniscus spinicornis n. sp. (PI. X, figs. 2 a — 2 g). Female (not quite full-grown). Body even a little more than four times as long as broad, with the anterior segments more arched and more deeply telescoped (fig. 2 a) than in any of the pre- ceding species; second segment about one-third as broad again as the seventh, which is distinctly broader than the front end of the fifth. Head conspicuously longer than usual in proportion to its breadth; the cephalic keels very low, scarcely converging anteriorly (fig. 2 a), without real anterior angles, but the distance between their anterior parts is only a little less than half of the breadth of the head ; the anterior margin of the front area is a little convex. — Anteunulse (figs. 2 b and 2 c) very peculiar; first joint uncommonly narrow, more than half as long again as broad; second joint longer than the first, gradually increasing considerably in thickness from the base to the end, which is produced into three long, oblong-triangular, acute or subacute processes; the right antennula (fig. 2 b) drawn in its natural position shows only two of these processes, as the third process is covered by the vesicle, furthermore only the outer part of third joint is visible to the left, and the fourth joint itself is covered by the left process of second joint, while the process of fourth joint is uncovered and is long, more than half as long as the ovate vesicle. The left antennula had been turned in a different direction and, seen from above (fig. 2 c), the three terminal processes of second joint are visible from above, the fourth joint lies between two of these processes, and its major part is uncovered, but its process is completely covered. — Antennae lost, excepting its proximal joints; squama (fig. 2 a) small, broader than long, with the lateral margins subparallel, and the terminal margin long, transverse. Antero-lateral angles of the three anterior segments scarcely or slightly produced, acute, but spines or setae were not discoverable. Median ventral part of seventh segment (figs. 2 f and 2 g) some- what vaulted, and a little before its posterior margin produced into a small, acute process directed backwards. - Thoracic legs partly mutilated; judging from the second joint of first pair of legs these are at most a little thicker than second pair; the six posterior pairs (fig. 2d) are rather slender, and the outer angle of fourth joint has a short seta. Abdomen (figs. 2 e and 2 g) a little longer than broad, with the end rounded but yet subangular, and most of the margins regularly convex. — Operculum considerably longer than broad, subpentagonal, as the basal margin is long and very moderately convex, the lateral margins feebly convex and distally suddenly curved much inwards, the posterior part of the plate produced backwards, with the tip forming an angle of about 100°, and each half of the posterior margin a little concave. — Uropods (fig. 2 g) somewhat long; the peduncle very broad, the endopod rather robust and about twice as long as the slender exopod. Length of the specimen 1.5 mm. Remarks. .V. spinicornis is abundantly distinguished from all species described on the pre- ceding pages by the antennulse, the very long anterior margin of the front area and the shape of the operculum ; besides, the strong telescoping of the anterior segments is very characteristic. The differ- ences between N. spinicornis and the following species are pointed out below. CRUSTACEA MALACOSTRACA. III. IOr Occurrence. Captured by the "Ingolf" at a single .station in the cold area. South of Jan Mayen: Stat. 113: Lat. 69°3i' N., Long. 7°o6' W., [309 lath, temp : 66. Nannoniscus affinis n. sp. (PL X, figs. 3 a— 3 c). Female (without marsupium). Closely allied to .V. spinicornis and agreeing with it in most features, above all in the antennute, the shape of the antenna! squama, and the anterior thoracii ments, thus in important characters separating N. spinicornis and A', affinis from all other northern species. It is in reality sufficient to point out the three or four differences between V a/finis and X. spinicornis. The keels on the head as in .V. spinicornis without anteriorly protruding ends, and continued in the convex anterior margin of the front area (fig. 3 a), but the keels converge considerably forwards, so that the anterior margin of the area is much shorter than in N. spinicornis, being !•' one- third and one-fourth of the breadth of the head. — The ventral process on seventh thoracic segment (fig. 3 c) is much longer and stronger than in N. spinicornis. Thoracic legs - first pair lost — in tin main as in X. spinicornis, but the angle of fourth joint (fig. 3 b) terminates in a long, very strong, spini- form seta almost as long as the joint. — Operculum (fig. 3 c) almost half as long again as broad, ]> eriorlv conspicuously more produced than in iV. spinicornis. and terminating in a somewhat small, nearl spiniform median process. Length of the largest specimen scarcely 1-3 mm. Remarks. It was only after some hesitation that I established X. affinis as a species. I think, however, that the differences pointed out between it and the preceding form will prove to be valid specific characters; furthermore X. affinis was seemed at a pronounced warm-water station, while A spinicornis was gathered at one of the deepest and most typical cold-water stations. Occurrence. Taken by the "Ingolf at a single station in the warm area. South-West of Iceland: Stat. 78: Lat. 60 37' N., Long. 2f^2' W., 799 fath.; temp. 4-5°; (mutilated). Group IX. Desmosomatini. Hodv very oblong, moderately depressed. Head free; eyes wanting. Antennnhe with very few joints in the flagellum. Antennje moderately long, thicker in the males females; squama not perceptible. Mandibles with the incisive part, the movable lacinia, a well or very moderately developed; the molar process reduced, rather short or short, narrow, setiferous end, and directed somewhat backwards; palp 3-jointed or wan maxillipeds nearly as in Nannoniscini ; epipod somewhat long. Thoracic segments into two very different sections. The four anterior pairs of thoracic legs inserted on the : segments, and at least second to fourth pairs with well developed epimera; first pair the [ngolf-Expedition. III. ;. Io6 CRUSTACEA MALACOSTRACA. III. from the following pairs, and frequently strongly prehensile; the three following pairs of moderate length, adapted for walking and more or less fossorial, especially second pair. The three posterior pairs without epimera, decreasing conspicuously in length from fourth to seventh pair, and fourth pair longer than any other pair; these three pairs conspicuously natatory, having fifth and sixth joints flattened and somewhat or feebly expanded, with a moderate or low number of natatory setae, and seventh joint well developed. — Abdomen at most with a rudimentary basal segment; uropods ventral, the endopod unjointed, frequently moderately long, the exopod short or wanting. Remarks. This group comprises three genera hitherto established, viz. Desmosoma G. O. Sars, Etigerda Meinert and Echinopleura G. O. Sars, but for reasons given later on I cancel Eugcrda and refer its species to the large genus Desmosoma. The group occupies in certain features a position between the Nannoniscini and the Ilyrachnini, but differs materially from both in several characters, especially in the legs. The head has a more or less developed "front area" as described in the Nannoniscini. Desmosoma G. o. s ar s. This genus has, on the whole, been well characterized by Sars, but I must cancel the genus Eugerda Meinert adopted by .Sars. On Desmosoma, as limited in the present paper, some points may be mentioned. The antennulae have in the flagellum 2 joints, in the male of a single species 3 joints, when the peduncle is considered as 3-jointed. At the lateral margins of the four anterior thoracic segments epimera are always visible, and at least on second to fourth segment always marked off by a suture, while such a suture is sometimes distinct, sometimes not discernible, between first segment and the epimera; the shape of the epimera, especially of first and second pairs, affords excellent characters, but it must be added, that in some species if. inst. D. longispinum: PI. X, figs. 6a and 6e) the epimera are somewhat or considerably more produced in the male than in the female. At the margins of the three posterior segments no epimera are found; these segments are frequently proportionately broader in sometimes narrower in the male than in the female; besides, the shape of fourth segment may differ according to sex (PI. XI, figs. 5 b and 5 c). In 1) laterale G. O. S. the four anterior segments have on the ventral surface a long, spiniform process in the median line; in D. longispinum a small process protrudes obliquely forwards 011 first segment, and in some forms no ventral armature is found, but the feature has not been examined in several species. First and second pairs of thoracic legs differ so much in shape and adornment with setae or spines from species to species, that they afford some of the best characters, and must always lie care- fully studied. First pair are in some species considerably or extremely slender, and their use seems to be somewhat uncertain, while in other forms they are thick and, sometimes, even very strong pre- hensile organs. Second to fourth pairs arc to some degree similar in structure, but second pair are stronger and more peculiar than the third, and especially than the fourth pair. To the brief descrip- tion above of the three posterior pairs of legs may be added that in the male of D. natator (PI. XI, fig. 2 d) these legs are on the whole rather similar to the same legs in males of some species of Nan? CRUSTACEA MALACOSTRACA III nontscus. — The abdomen is generally broader in the male than in the Female, and in son* it shows further sexual differences. The female operculum seems to be rather uniformly shaped; the median lamella in the male operculum shows specific differences, but they are sometimes difficull investigate without dissection. The single character between Desmosoma and Eugerda is drawn from the number <>i rami the uropods. But this character cannot be of generic value, as, for instance, />. armatum G 0 which has no exopod on the uropods, is in the structure of the legs, etc., very closeh allied t<. /' (Eugerda) coarctatum G. O. Sars or D. (Eugerda) potitum n. sp., which both have biramous uropods; D. armatum is in reality more closely allied to these two species than to such forms with single-branche uropods as D. lineare G. O. Sars or D. instgne n. sp. Therefore the genus Eugerda must be suppi as not only valueless but misleading. From the North Atlantic with ramifications like Skager Rak, Kattegat, and the Mediterranean, 9 species were known, described by Sars, Meinert, Ronnier, and Stephensen. Of these onl} 3 are repre- sented in the material from the "Ingolf" area, but I here establish 8 new species. Thus no less than 17 species are known from the North Atlantic, including the cold area. I have, however, mutilated specimens of still 5, or more, new species, but, as the specimens have lost the main part <>f the two anterior pairs of legs, I do not think it advisable to describe them. These seventeen species may be divided into two sections, and determined by the aid of t following keys. The species not found in the "Ingolf" area are put in brackets and without numl Sect. I. First pair of legs with fifth and sixth joints slender; fifth joint has the lower margin either glabrous or with some few hairs or stiff setae. Sect. II. First pair of legs with fifth and sixth joints much thickened; especially fifth joint thick, with a very long and strong spine on the lower distal angle and in several sped. with a single spine or a few strong spines along the lower margin. Section I. A. Uropods biramous. Fifth joint of first pair of legs with the lower margin either naked 01 a couple of hairs. (First thoracic segment only about half, or less than half, as long i a. Fifth and sixth joints of first pair of legs extremely thin and elongate; fifth joint mai longer than thick ' /;- tenuimanu . b. Fifth and sixth joints of first pair of legs slender; fifth joint at most nearly five long as deep. a. Epimera of second thoracic segment triangular, with the anterior half acute or obt joint of second pair of legs distinctly more than twice as long as deep, wit 7 or 8, spines and setae along the lower margin • D. /?. Epimera of second thoracic segment with the anterior half rounded in front. 1 ,1 second pair of legs scarcely twice as long as deep, with 7 extremely stro along the lower margin B. Uropods without exopod. Fifth joint of first pair of legs with some setae on the lower most distal seta at least half as long as sixth joint. I0g CRUSTACEA MALACOSTRACA III. a. Abdomen posteriorly rounded, without postero-lateral angles. First thoracic segment consider- ably shorter than, or almost only half as long as, the second. a. Fiith joint of first pair of legs at the distal lower angle with a single seta, which is only half as long as sixth joint (£>. linear e G. O. S.) ft. Fifth joint of first pair of legs at the distal lower angle with two setse, which are almost as long as sixth joint (D. elongation Bonnier) b. Abdomen with produced and acute postero-lateral angles. First thoracic segment about as long as the second. a. Basal joint of the uropods much thickened, and these joints nearly touching one another at the median line of the animal, and covering the anal doors. Epimera at first and second segments in both sexes produced into very long processes 3. U. longispinum n. sp. ft. Basal joint of the uropods not much thicker than the endopod, and very remote from the median line. Epimera at first and second segments produced into moderately long or short processes. §. Lower margin of fifth joint of first pair of legs with four setae, of fifth joint of second pair with five seta;. Somewhat small species 4. /). simile n. sp. SS- Lower margin of fifth joint of first pair of legs with five setae, of fifth joint of second pair with about eight setae. Moderately large species 5. 1). gracilipes n. sp. Section II. A. Uropods biramous. (Abdomen with produced and acute postero-lateral angles). a. Outer angle of the head at the insertion of the antennula not produced into a free, outwards projecting process. Fifth joint of first pair of legs with three to five strong spines along the lower margin and at the end. «. Fifth joint of first pair of legs with four spines along the lower margin, and a fifth some- what short spine just above the very long spine projecting from the distal angle. Lateral margins of fifth segment scarcely concave, a little converging backwards. 6. D. politum n. sp. ft. Fifth joint of first pair of legs with three spines along the lower margin, and a fourth rather long spine just above the long spine projecting from the distal angle. Lateral margins of fifth segment (in the male) deeply concave, and each antero-lateral angle produced and ter- minating in a spine 7. />. natator n. sp. ;-. Fifth joint of first pair of legs with three spines along the lower margin, and no spine above the long spine projecting from the distal angle. Lateral margins of fifth segment in both sexes deeply concave, and the antero-lateral angles broadly rounded, without any spine. {D. coarctatum (Hansen, MS.) G. O. S.) b. Outer angle of the head at the insertion of the antennula produced into an oblong-triangular, acute process directed freely somewhat outwards. Fifth joint of first pair of legs with only two spines and an intermediate seta on the lower margin, and no spine above the very long spine projecting from the distal angle 8. I), laterale (Hansen, MS.) G. O. S. B. Uropods without exopod. CRUSTACEA MAJ.ACoSTK.ua. III. a. Abdomen with produced and acute postero-lateral angles. Fifth joint ol first pair ■ four spines along the lower margin, and a minute spine close above the very loir the distal angle .,. I >. armatun, b. Abdomen posteriorly rounded, without produced postero-lateral angles. Fifth j paii of legs with three spines or a single spine on the lower margin, and no spine long spine projecting from the distal angle. a. Fifth joint of first pair of legs with three spines on the lower margin, the terminal spine included (/>. angustum (Hansen, MS ft. Fifth joint of first pair of legs with a single spine, the terminal, on the lower in. S- Lower margin of fifth joint of first pair of legs more than twice as Ion the joint and slightly shorter than sixth joint, which is about lour times Fifth thoracic segment in the female much broader at the anterior than at the posterioi margin and more than twice as broad as the length of the median line, with the lati margins convex, and the surface strongly reticulate. Very large species helatum Stephei SS- Lower margin of fifth joint of first pair of legs a little less than twice as Ion- a- the depth of the joint, and slightly shorter than sixth joint, which is almost three times long as deep. Fifth thoracic segment in the female much broader at the anterior at the posterior margin, a little more than twice as broad as the length of its median line, with the lateral margins a little concave near the middle, and the surfaci reticulate. Moderately large species 10. I> SSS- Lower margin of fifth joint of first pair of legs scarcely half as long again as the depth of the joint, and considerably shorter than sixth joint, which is conspicuous! three times as long as deep. Fifth thoracic segment in the female narrower at the anterior than at the posterior margin, not one-third as broad again as the length of its u line, with the lateral margins scarcely concave at the middle, and the up] scarcely reticulate. Small species 67. Desmosoma globiceps Meinert. (PI. X, figs. 4 a -4 b). 1890. Etigrrda globiceps Meinert, "Hauch"s Togter, p. 194; PI. II. figs. 53-62- 11899. — G O- Sars, Account, II, p. 252; Suppl. PI. Ill, fig- 1. The figures and description published by Sars may be sufficient for the re species, but a few points may be mentioned here. First thoracic segment (fig. 4 a) is very short, only about half as long as the epimera somewhat produced and terminating in a spine. The anterior part of segment is distinctly produced, triangular, with the end acute or obtuse, terminating As figured by Sars, the legs of first pair are slender, and fifth joint without s. margin. Second pair of legs with fifth joint (fig. 4b) very characteristic, compres IIO CRUSTACEA MALACOSTRACA. III. expanded below and almost oblong-triangular, distinctly more than twice as long as deep, but scarcely ever more than two and a half times as long as deep (in the figure of Sars it is somewhat too narrow), with g — n - - rarely, and perhaps anomalously, with 7 or 8 — spines and setse along the lower margin, the three or four distal ones being very long and rather thick, spiniform with their terminal part setiform, the others gradually much shorter and thinner, so that the three or four proximal ones are real setse; besides, 6 or 7 long setse are found in a row on the outer side of the joint below the upper margin; sixth joint is conspicuously more than twice as long as deep, with 2 setse on the lower and 7 to 9 seta; on the upper margin, most of them long, and about the 3 distal ones very strong, nearly spiniform. — The abdomen, which has been well figured by Sars, is characteristic, a little longer than broad, ovate; the uropods are proportionately long, a little more than half as long as the abdomen, with the rami slender, and the exopod almost or not fully half as long as the endopod. Length of the largest specimen (from East Greenland) 26 mm. Remarks. D. globiccps is easily distinguished from D. tenuimanum G. O. S. in having the first pair of legs much stronger and shorter, a considerably lower number of spines and setse on fifth and sixth joints of second pair of legs, etc. The differences between D. globiceps and D. latipes are pointed out later on. Occurrence. Not taken by the "Ingolf". The Ilnd Amdrup Exped. captured 2 large females with marsupium at East Greenland near Cape Dalton, Lat. 69°3o' N., 9— 11 fath. At East Iceland Dr. A. C. Johansen secured 1 spec, in Breiddals Yik, 6 fath., mud and black sand; at the Fseroes 4 spec, have been gathered by Dr. Th. Mortensen in Klaksvig, 10—15 fath. Distribution. It has been taken at 10 places in the eastern and southern Kattegat, 12V2 — 44 fath. (H. J. Hansen), in the northern part of the Sound, 12 — 22 fath. (W. Bjork), and at 3 places in the Skager Rak, 6 to 15 miles from Jutland, 70 to 125 fath. (Meinert). It has not yet been discovered at Norway or Great Britain. 68. Desmosoma latipes 11. sp. (PL X, figs. 5 a— 5 f ). Female. Allied to D. globiceps. — First thoracic segment (fig. 5 a) not half as long as tin- second; the epimera scarcely produced, blunt or rounded. Second segment large; the epimera anteriorly rounded. Fourth segment a little longer than the first, but only a little more than half as long as the second. Fifth segment in the adult female somewhat narrower in front than behind, with the lateral margins gradually somewhat diverging backwards; in the younger female (fig. 5 a) the anterior part of the lateral margins converges extremely forwards. First pair of legs (fig. 5 b) with the distal half nearly more slender than in D. globiccps; fifth joint with two thin and somewhat short setse on the lower margin. -- Second pair (figs. 5 c and 3 d) extremely robust and conspicuously stronger than in E. globiceps; their fifth joint not quite twice as long as deep, with seven spines along the lower margin; the five distal spines are extremely thick excepting their short, setiform terminal part; furthermore, the two distal spines are extremely long, and the spines gradually decrease much in length from the penultimate to the most proximal; finally CRUSTACEA MALACOSTRACA. III. nl the basal spine is moderately slender and somewhat short, with the end setiform; cm the outer side of the joint somewhat below the upper margin eight strong, stiff setae arranged in a row. The sixth joint is twice as Ion"- as deep, with four seta- on the lower and nine along the upper margin; seventh joint not marked off from the short claw. (Seventh pail oi legs lo The abdomen (fig. 5 e) seems to be proportionately a little longer than in /' md, besides, less ovate. — Uropods a little more than half as long as the abdomen; the exopod slighth more than half as long as the endopod (fig. 5 f). Length of the single female with marsupimn 22 nun. Remarks. As already stated, D. latipes is allied to D. globiceps, but it is easilj distinguished by the second pair of legs, especially by its proportionately broader fifth joint with the very thick spines, and by the broader sixth joint with four setae along the lower margin. Occurrence. Taken by the "Ingolf" at a single station. Davis Strait: Stat. 25: Long. 63°3o' N, Lat. 54°25' W., 582 fath., temp. 3-3°; 5 spec (mutilated). 69. Desmosoma longispinum n. sp. (PI. X, figs. 6 a— 6 i). Female and Male. The whole upper surface of the body peculiarly sculptured, as each ment is adorned with numerous longitudinal, though somewhat irregular and partly interrupted, fine keels. -- First thoracic segment large, a little longer than the second, which is as long as, or a little longer than, the fourth; in the male (fig. 6 e) the epimera at all four segments are produced into extremely long processes, a little or somewhat longer than their segment and differing little in length, while in the female (fig. 6 a) the processes of the two first pairs are not mudTshorter than in the male, but the third pair are considerably and fourth pair much shorter than in that sex; the first pail processes are in both sexes directed forwards, the following pairs, besides, gradually somewhat outwards. Fifth segment a little broader near the anterior than at the posterior margin, with the lateral margins slighth' concave, and the antero-lateral angles broadly rounded; the segment is in the female souk what, in the male only a little, broader than long. First pair of legs (figs. 6 b and 6g) slender; fifth joint about five times as long as deep, a lower distal angle with two stiff seta; somewhat shorter than sixth joint, and two somewhat setae on the lower margin, the most proximal near its middle; sixth joint slender with a at the lower distal end. — Second pair (fig. 6 c) somewhat longer than, but almost as slender i pair; fifth joint more than five times as long as deep, increasing slightly in depth from base with five stiff seta; along the lower margin, and the distal seta inserted at the distal angle and ; as long as sixth joint; the last-named joint has two long setae at the distal part of its and a single short seta at the middle of the lower margin; seventh joint is uncommonly long, pair of legs (fig. 6 d) slender; fifth joint with about seven thin setae on the distal part of margin, and some of them (probably five) extremely long; sixth joint very slender, wit long setae and one shorter seta on the lower margin. Abdomen in the female (fig. 6 a) considerably longer than broad, subrectangular, wit II2 CRUSTACEA MALACOSTRACA. III. lateral angles rounded, and the postero-lateral angles produced into oblong processes directed some- what outwards, much backwards and considerably downwards; the posterior margin between the base of these processes a little convex, and the lateral margins slightly converging from rather near the base to somewhat from the end. — In the male the abdomen (fig. 6 e) is proportionately broader than in the female, with the major proximal part of the lateral margins somewhat convex, and the posterior margin more convex than in the female. — The operculum in the female posteriorly broadly rounded ; in the male (fig. 6 i) each of the pleopods of first pair tapers conspicuously in breadth from somewhat from the base to considerably beyond the middle, and then the outer margin suddenly turns consider- ably outwards, so that the distal part of the pleopod is a rather large triangle, with the posterior margin a little convex. — Uropods (fig. 6i) very peculiar, considerably less than half as long as the abdomen; basal joint very large, with the inner margin straight and lying close at the median line of the animal, so that the peduncles of the uropods completely overlap the anal doors when seen from below; endopod more than half as long again as the peduncle, slender; exopod wanting. Length of a female without marsupium 2'i mm., of the male r8 mm. Remarks. D. longispinum has the epimera at the four anterior thoracic segments produced into longer processes than in any other species hitherto known, though in this character it shows affinity to D. simile and the male of D. gracilipes. The sculpture on the surface of the animal is very char- acteristic of D. longispinum. The two anterior pairs of legs are somewhat similar to those in D. simile, though a little more slender and elongate. But in the very broad peduncles of the uropods and in their inner margins lying close at the median line D. longispinum differs sharply from all other species. Occurrence. Taken by the "Ingolf" at its deepest station in the warm area. South of Davis Strait: .Stat. 38: Lat. 59°i2' N., Long. 5i°05' W., 1870 fath., temp. 1-3°; 4 spec. 70. Desmosoma simile n. sp. (PI. X, figs. 7e-7e). Female. First thoracic segment about as long as the second or the third, and a little shorter than the fourth (fig. 7 a). The epimera at these segments anteriorly produced into very oblong-triangular, moderately long processes, each terminating in a distinct, small and short spine; the processes decrease a little in length from first to fourth pair, and the first pair are considerably shorter than their seg- ment. Fifth segment more than half as broad again as long; the anterior margin is straight, the antero-lateral angles produced into very conspicuous, triangular, acute processes directed outwards, and the anterior part of each lateral margin is somewhat concave. First pair of legs (fig. 7b) rather slender; fifth joint about three and a half times as long as deep, at the lower distal angle with two very strong, stiff setse somewhat shorter than sixth joint, while two somewhat shorter, stiff setce are found respectively before and beyond the middle of the lower margin; sixth joint with about three seta? on the lower margin. — Second pair of legs (fig. 7c) rather slender; fifth joint nearly three and a half times as long as deep, increasing slightly in depth from the base outwards; its lower margin has five stiff setse, the two distal, placed at the angle, are thick and very long, the longest even considerably longer than the sixth joint; furthermore the outer CRUSTACEA MALACOSTRACA. III. u, side of the joint has about five long setae placed in a row below the upper margin; sixth joint lias three long setae on the distal half of the upper margin, and on the lower margin a small seta at the middle and a couple at the end. — vSeventh pair of le^s (fig. 7 d) Inn- and very slender; filth joint with only three setae, two of which very long, on the distal part of the lower margin; sixth joint with four setae on the lower margin, and at least two of them very long. Abdomen (fig. 7 a) slightly longer than broad, but otherwise somewhat resembling that in D. longispinum, excepting that the posterior margin between the postero-lateral acute proi what more convex than in the female of that species. -- Uropods half as lout; as the abdomen, and inserted moderately far from one another (fig. 7e); the peduncle only somewhat thicker than the endopo exopod wanting. Length of an ovigerous female 2'2 mm. Remarks. The female of this species (the male is unknown) is separated from D. longispinum in having especially the two anterior pairs of epimeral processes much shorter, the anterior an fiftli segment produced into processes, setae above on the outer side of fifth joint of second pail "i legs, the uropods much removed from the median line, and their peduncles of normal size. Tin- dif- ferences between D. simile and D. gracilipcs are pointed out in the "remarks" on the last-named species. Occurrence. Taken by the "Ingolf" at a single deep station in the warm area. Davis Strait: Stat. 24: Lat. 63°o6' N., Long. 56°oo' W., 1199 fath., temp. 2-4°; 5 spec. 71. Desmosoma gracilipes n. sp. (PI. XI, figs. 1 a— r f). Female and Male. First thoracic segment slightly longer than the second, which is abou as long as the third or the fourth. In the female (fig. 1 a) the epimera at first segment are anti produced into a subtriangular process, about as long as broad and terminate in a minute spine epimera at the three following segments are similarly shaped, but from first to fourth pair gradually a little shorter. In the male (fig. 1 f) the epimera are anteriorly produced in processes more th; as long as in the female, and even longer than in the female of D. simile; each process termir. a small spine. Fifth segment in the female more than half as broad again as long, with all ang rounded, and the lateral margins subparallel, though slightly concave; in the male (fig ment is half as broad again as long, and differs especially from that in the female in ha anterior angles acute and a little produced outwards. First pair of legs (fig. 1 b) rather slender; fifth joint somewhat more than three times as deep, with five very strong setae along the lower margin, the two distal of thesi together at the angle and only somewhat shorter than sixth joint. -- Second pair (fig. 1 c) slender; fifth joint about three and a half times as long as deep and somewhat increasing i from base to the end, its lower margin with eight strong setae, the two most distal and somewhat longer than sixth joint, and, besides, a row with eight seta is seen on I of the joint below the upper margin; sixth joint has seven seta- on the upper and t margin. — Seventh pair (fig. id) slender; fifth joint with four long setae and one shorter seta on lower margin; sixth joint very slender, with about six setae on the lower margin The Ingolf-Expedition. HI. 5. ii4 CRUSTACEA MALACOSTRACA. III. Abdomen in the female (fig. i a) somewhat longer than broad, decreasing distinctly in breadth from a little from the base to near the produced acute angles, which are somewhat small processes directed mainly backwards; the posterior margin between these two teeth very convex. The male abdomen (fig. i f) is larger and especially distinctly broader than in the other sex, has the postero- lateral angles a little more produced, and the posterior margin between them a little less convex. — Uropods (fig. i e) not quite half as long as the abdomen, rather distant from one another; the basal joint only a little thicker than the endopod; exopod wanting. — Each pleopod of first pair in the male operculum has its lateral margins parallel from rather near the base to the suddenly rounded end. Length of a female with marsupium 32 mm., of a male 27 mm. Remarks. D. gracilipcs is allied to D. simile, but it is considerably larger; fifth joint of first pair of legs has five setae, that of second pair eight setoe along the lower margin, and the female differs, besides, from that of D. simile in having the epimera much less produced, and the anterior angles of fifth thoracic segment rounded. — In the male specimen the antennulse are 6-jointed, con- sequently the flagelltim 3-jointed. Occurrence. Taken by the "Ingolf" at two deep stations in the warm area. Davis Strait: Stat. 24: Lat. 63°o6' N., Long. 56°oo' W., 1199 fath., temp. 2-4°; 5 spec. (1 <$). Stat. 36: Lat. 6i°5o' N., Long. 56°2i' W., 1435 fath., temp. 1-5°; 1 spec. 72. Desmosoma politum u. sp. (PI. X, figs. 8a-8e). Female. First thoracic segment conspicuously broader and a little longer than the second, and as long as the fourth. Epimera at first segment constitute a pair of low, obtuse triangles; the three following pairs have the outer margin distinctly incised near the middle, and the front half a little protruding and rounded. Fifth segment a little more than half as broad again as long, with all angles broadly rounded, and the lateral margins a little converging backwards. First pair of legs (fig. 8 b) very robust; fifth joint scarcely twice as long as deep, with four moderately strong spines on the lower margin, the most distal placed on the angle and much shorter than sixth joint, while a moderately short spine — thus the fifth --is placed above the longest spine on the terminal margin of the joint; sixth joint a little shorter than the fifth, very oblong-oval, some- what less than three times as long as deep. — Second pair of legs (fig. 8 c) moderately strong; fifth joint long, between four and three and a half times as long as deep, increasing conspicuously in depth from the base to beyond the middle; its lower margin and end with seven spines, the five proximal of very moderate length, the sixth spine long and placed on the distal angle, while the seventh, inserted above the sixth on the terminal margin, is still longer, not much shorter than sixth joint; the upper margin with three strong setae. Sixth joint with three strong setae on the upper and a fine seta at the middle of the lower margin. — Seventh pair of legs (fig. 8 d) somewhat strong; fifth joint with a very long seta at the lower distal angle, and before this two much shorter setae; sixth joint longer and especially more robust than in the preceding forms, witli two somewhat short seta; on the lower margin, and three setae, one among them long and stiff, on the upper margin; seventh joint unusually short, terminating in a well developed claw somewhat shorter than the joint. CR1 STACEA MALACOSTRACA. Ill Abdomen (figs. 8 a and 8 e) a little longer than broad; the lateral margins omewhat backwards to the postero-lateral teeth, which are small, directed mainly backwards and plac der- ably before the end of abdomen, as the posterior margin is very convex - Uropods somewhat than half as long as abdomen, rather distant from the median line; peduncle somewhat thicker than the endopod; exopod minute, very thin and only about as long as the diameter of the endopod. Length of a large female without marsupium 22 mm. Remarks. D. poliftun is allied to D. coarctation (Hansen, MS.) G. O. Sars, but the latter separated from it by a very different shape of fifth thoracic segment, and in having only three spines on the lower margin and no spine on the terminal margin of fifth joint of first pair of legs. D.poli- tum is, besides, closely allied to D. armatum G. O. Sars, which, however, differs from it especially in having the first pair of epimera produced into somewhat long, oblong-triangular, acute processes, in having no exopod on the uropods, and in some further particulars. A very small specimen (from Stat. 90) with seventh pair of legs still wanting shows the curious feature that fifth joint of first pair of legs possesses only the long spine on the distal lower angle and no further spine at the lower margin, thereby being rather similar to the same joint in D. ins;. n. sp. and D. plcbcjum n. sp. But it is instantly separated from these species in having the distal lower angle of the joint mentioned broadly rounded, while in the two species named this angle is produi as a kind of foot for the very long spine. Occurrence. Taken by the "Ingolf" at two stations in the warm area. West of Iceland: Stat. 90: Lat. 64°45' N., Long. 29°o6' W., 568 fath., temp. y \ ; 9 ^pec. South-West of Iceland: Stat. 78: Lat. 6o°37' N., Long 27=52' W., 799 fath., temp. 4-5°; 3 very young). 73. Desmosoma natator n. sp. (PL XI, figs. 2 a— 2 e). Male. The upper surface of the body reticulate, nearly looking as if covered with very more or less rounded scales set close together but never overlapping each other. -- Head 1 and broad. First thoracic segment broader and a little longer than the second, but scarceh as the fourth. All epimera have their anterior section produced forwards and outwards spicuous, though moderately short, triangular processes, decreasing a little in length from fir: pair; the two anterior pairs are nearly as long as broad, and all pairs have the narrow end ten in a small but very distinct spine. Fifth segment anteriorly considerably broader than long, with front margin straight and long, the lateral margins considerably concave, the anterior augl- outwards, and their narrow end terminating in an outstanding spine. First pair of legs (fig. 2 b) very robust; fourth joint 011 the upper distal angle wit stiff setae or thin spines, and one of them nearly as long as the next joint Fifth joint as long as deep; its lower margin and end with four spines in all; the spine on the dis though a good deal shorter than sixth joint; the spine just above the angle and the two other proximal spines moderately long. Sixth joint scarcely shorter than the fifth, not quite two and times as long as deep. - Second pair of legs robust (fig. 2 c); fifth joint three times as Ion- as deep, JJ6 CRUSTACEA MAI.ACUSTRACA. III. increasing conspicuously in depth from the base to the end, with three spines along the upper and six spines on the lower margin and the end; the first of these latter spines is very short, while the two distal spines, and especially the most distal, are long and moderately thick. Sixth joint with three spines along the distal half of the lower margin. — Seventh pair (fig. 2 d) have the distal half peculiarly developed; fifth joint is somewhat long and rather strong, with about three stiff setae on the lower margin, while the median part of the upper margin has about six long, fine hairs; sixth joint is un- commonly strong and even a little longer than the fifth, with about four long, fine setse on the lower margin, and about thirteen extremely long, fine setse or hairs along the upper margin; seventh joint moderately long, slender, a little curved and terminating in a slender, somewhat long claw. — Fifth and sixth pairs rather similar to seventh pair, and especially their sixth joint possesses the same seta; along both margins. Abdomen somewhat broad (as generally in males of this genus), a little longer than broad (fig. 2 a), tapering a little in breadth from somewhat from the base to near the postero-lateral processes (fig. 2 e), which are moderately large, triangular, acute, directed backwards and somewhat outwards; the posterior margin moderately convex. — The median lamella of the operculum tapers conspicuously from rather near the base to a little from the end, where each outer margin turns suddenly consider- ably outwards, so that the short terminal part is a little expanded, and the posterior margin of each pleopod is oblique. — Uropods (fig. 2 e) rather distant from the median line, somewhat less than half as long as the abdomen; basal joint somewhat triangular and moderately small; the exopod tiny, very slender, about as long as the diameter of the endopod. Length of the single male 1-85 mm. Remarks. This animal is easily distinguished by the shape of the epimera and the fifth segment, and by the legs. First and second pairs of legs are stronger than in D. polihuu and some- what similar to those of D. coarctatum (Hansen MS.) G. O. S., but fifth joint of first pair has four spines in D. i/ti/a/or, only three in D. coarctatum. The development of the three posterior pairs of legs, above all the equipment of the upper margin of fifth and especially of sixth joint with very long, thin setse, is hitherto unique in the genus Desmosoma, and these legs show a close resemblance to the corres- ponding legs in the males of Naiuioniscus simplex n. sp., .V. oblongits G. O. S., N. analis n. sp. or the female of N. inermis. I think that this equipment makes the legs adapted for swimming — hence the name D. uatator -- but it is of course impossible to know at present whether these setse also occur in the female. Occurrence. Taken by the Ingolf at a deep station in the warm area. Davis Strait: Stat. 36: Lat. 6i°5o' N., Long. 56°2i' W., 1435 fath., temp. 1-5°; 1 spec. 74. Desmosoma laterale G. O. Sars. (PI. XI, figs. 3 a-3 e). 1899. Eugerda lateralis (Hansen MS.) G. O. Sars, Account, II, p. 254; Suppl. PI. IV, fig. 1. Female. Head (fig. 3a— 3b) somewhat small; each lateral corner before the insertion of the antennula produced into a very conspicuous, oblong-triangular, acute process directed forwards and CR1 STACEA MALACOSTRACA. in- considerably outwards. — First thoracic segment narrower and somewhat longer than the second, which is scarcely as long as the fourth. Epimera well developed but not produced, having the margin of their anterior section evenly convex. Fifth segment considerably broader than long, subrectangular, with the lateral margins nearly parallel and all angles rounded. First pair of legs (fig. 3 c) very robust. Fifth joint only half as long again as deep, with a single very strong spine beyond the middle of the lower margin, a strung seta near the distal end, and on the angle a very robust spine, which is as long as the joint Sixth joint very lai ably longer than the fifth, somewhat less than three times as long as deep. - (fig. 3d) uncommonly robust; fourth joint on the distal lower angle with a very thick spine slightly longer than the joint; fifth joint about twice as long as deep, with five or four spines along the lower margin and the end; the spines increase gradually and strongly in length from the first to the I and if five are present the first is small, the others very strong; the two distal are distinctly curved. ami the most distal, which is placed just above the angle, is as Ion- as the sixth joint; the uppei margin with three strong setce. Sixth joint with three strong setae on the upper and two on the lower margin. — Seventh pair (fig. 3c) rather slender; fifth joint with five setae on the lower margin and three of them extremely long; sixth joint with two setse below. Abdomen somewhat longer than broad, decreasing conspicuously in breadth from somewhi from the base to near the postero-lateral teeth, which are somewhat slender and situated rather far in front of the end of abdomen, as the posterior margin is extremely convex. -- Uropods considerably less than half as long as the abdomen, rather remote from the median line; exopod between one-thin and half as long as the endopod. Length of the largest specimen, a female without marsupium, 2 mm.; a female with marsupiuin from Skager Rak is r8 mm. Remarks. Specimens of this species had been separated by me among the "Ingolf" and among material from Skager Rak, and I named them D. laterale. The three specimen last-named place were later handed over to G. O. Sars, who described and figured the species orrda lateralis. But as Sars has overlooked the processes from the sides of the head, and as his of the entire animal is not as correct in some few particulars as usually is the case, I have five figures and redescribed the species. It may be pointed out that D. lateral, is instantly distinguished from all other species of J), mosoma by the outstanding processes on the sides of the head, and by the armature of fir: legs. In females without marsupium each of the four anterior segments has a long, subvertica1 in the median line of the ventral surface. Occurrence. Taken by the "Ingolf at a single station in the warm area. Davis Strait: Stat. 25: Lat. 63*30' X, Long. 54°25' W, 582 fatb, temp. 3-3 ; 6 spec. Distribution. The only earlier known locality is in Skager Rak, 15 sea-miles lighthouse of the Skaw, 125 fath., ooze. s n8 CRUSTACEA MALACOSTRACA. III. 75. Desmosoma armatum G. O. Sars. (PI. XI, fig. 4 a). 1864. Desmosoma armatum G. O. Sars, Forh. Vid. Selsk. Christiania for 1863, p. 216. ! 1897. G. O. Sars, Account, II, p. 126; PL 54, fig. 2. The representation published by Sars is on tbe whole good, and quite sufficient for the recog- nition of the species. The shape of fifth and sixth joints in first and second pairs of legs and their equipment with spines and setae afford good characters. Only a few features may be mentioned as additions or corrections. The epimera of first thoracic segment are somewhat produced anteriorly, with the free inner or rather anterior margin concave, and the narrow end terminating in a robust, somewhat short spine. The epimera of second and third segments are slightly or scarcely produced forwards, but anteriorly angular with a minute hair. In my specimen and in a co-type of Sars I have found the seventh pair of legs (fig. 4 a) less slender than figured by that author; fifth joint has five setae on the lower margin, three or four of them very long, and this joint has, besides, near the upper margin, five long hairs; sixth joint has four seta; on the lower margin, three of them long, and some seta; at or near the upper margin. Length of the specimen from Greenland, a female with marsupium, 175 mm. Remarks. In having no exopod on the uropods D. armatum agrees with the two following species, but differs strongly from both in the shape and spines of fifth joint of first pair of legs. As to spines and setse on fifth and sixth joints of first and second pairs of legs, D. armatum is inter- mediate between D.politum and D. coarctation, but it differs from both these forms and from D. uafafor in having no exopod on the uropods. Occurrence. It has not been taken by the "Ingolf"; but in 1891 by the Ryder Expedition at the following place. Hast Greenland: Danmark O, Lat. 70°27' N.; 1 spec. ( with marsupium). 76. Desmosoma insigne n. sp. (PI. XI, figs. 5a-5g). Female and Male. In the female the major part of the upper surface of the body seems to be reticulated, in reality set with a large number of partly long, extremely fine, sharp keels; these keels are mainly transverse on the middle of the head, on second to fourth segments, and on the posterior median part of the other segments, longitudinal on the bead and on the more lateral parts of fifth (fig. 5 b) to seventh segment. In the male this adornment is much less developed, has vanished on large areas (fig. 5 c). The head is somewhat narrow (fig. 5a); the antennae in the female rather short, in the male proportionately longer and much thickened. — First thoracic segment longer and a little narrower than the second, but about as long as the fourth. Fifth .segment in the female (figs. 5 a and 5 b) broader than the preceding segments, slightly more than twice as broad as long in the median line, with the anterior margin somewhat concave, the lateral margins anteriorly convex, and near tbe middle I Rl STACEA MALACOSTR VCA. III. a little concave, taken as a whole considerably converging backwards, with the keels on most ol th( surface very conspicuous; in the male this segment (fig. 5c) is distinctly narrower thai third segment, scarcely twice as broad as long in the median line, with the anterioi I the lati margins feebly convex, their median part slightl) concave, and the keels on the surface develo] only towards the lateral margins. - Epimera subsimilar in both sexes; first to third pairs have their anterior section produced as triangles as long as or a little longer than broad, and terminating in spines, which are robust on first pair, thin on the two following pairs; fourth pair of epimera with the produced part quite small. First pair of legs (fig. 5 c!) with second to fourth joints robust, while the fifth joint is thick; this joint has the lower distal part somewhat produced as a foot for the very long and robust spine, and its lower margin, which is a little less than twice as long as the depth of the joint, slightly concave and without any spine, but with only a couple of short setae and a moderately Ion- seta just behind the base of the terminal spine. Sixth joint is slightly longer than the lower margin of fifth joint, and about three times as long as deep. -- Second pair of legs (fig. 5c) rather slender; fifth joint somewhat more than three times as long as deep, increasing a little in depth to beyond the middle, with about nine slender spines, gradually increasing in length, along the lower margin, and nine stiff setse in an oblique row on the outer side; sixth joint with slender spines, the two distal very long, on the lower margin, and about four setse on the outer side towards the upper margin. Seventh pair (fig. 5 f) somewhat slender or moderately strong; fifth joint with four long setae on the lower margin near its end; sixth joint with five or six long seta: along the distal half ol : lower margin. Abdomen (fig. 5 a) a little longer than broad, tapering considerably from near the base to the end, without postero-lateral teeth, and the posterior margin only a little convex. female a little emarginate behind; in the male the pleopods of first pair have the lateral margins nearly parallel to the rounded end. — Uropods (fig. 5 g) inserted only a little from the lateral mar they are more than half as long as the abdomen, with the peduncle twice as long as broad, consider- ably broader than, and a little more than half as long as, the slender endopod; exopod wanti Length of a female with marsupium 3-1 mm., of a male 1-85 mm. Remarks. D. insigne, D. plebejum n. sp., and D. chelation Stepheusen constitu within the genus Dcsmosonia, as the distal joints of the first pair of legs form fifth joint being the hand, the strong spine answering to the immovable finger, while- is the movable finger, and the hand has no spines on the lower margin. (A fourth specie is the antarctic Eugenia longimana Vanhoffen, as its first pair of legs have similar fers from the three other species in possessing an exopod on the uropods, an addition: the presence or absence of the exopod is less than valueless as a generic difference betwee and Engerda). It may be added that D. chelation Stepheusen, from off Elba in the M< considerably larger than D. insigne, as the type, a female with marsupium, i characters between the two species are pointed out in the key (p. 109), and, b< legs in D. chelation have fourteen spines on the lower margin and end of the I20 CRUSTACEA MALACOSTRACA. Ill spines on the corresponding margin of sixth joint. The most important differences between D. insigne and D.plebejum are pointed out in the key. Occurrence. Taken by the "Ingolf" at a deep station in the warm area. Davis Strait: Stat. 36: Lat. 6i°5o' N., Long. 56°2i' W., 1435 fath., temp. 1-5°; 6 spec. (3?, 30*). 77. Desmosoma plebejum n. sp. (PI. XI, figs. 6 a— 6 d). Female. The upper surface of the body without distinct lineal reticulation. — Head moder- ately small. — First thoracic segment shorter than the second, which is distinctly shorter than the fourth. Fifth segment conspicuously narrower at the anterior than at the posterior end, with the lateral margins somewhat diverging backwards and scarcely concave; the segment is not one-third as broad again as the length of its median line. — Epimera of first segment (fig. 6 a) anteriorly pro- duced as a small, subtriangular process terminating in a distinct spine; the three other pairs of epimera have their anterior halves rounded in front. First pair of legs (fig. 6 b) very robust. Second to fourth joints thicker than in D. insigne. Fifth joint very thick, with the distal lower part produced as a foot for the very long and robust spine, while its lower margin, which is scarcely half as long again as the depth of the joint, is straight and without spines, but with a seta at the terminal spine. Sixth joint proportionately considerably larger than in D. insigne, considerably longer than the lower margin of fifth joint, and conspicuously less than three times as long as deep. — (Second pair lost in my animals). — Seventh pair (fig. 6 c) somewhat strong; fifth joint rather broad, with five setae on the lower margin. Abdomen (fig. 6 a) somewhat longer than broad, ovate, without postero-lateral teeth. — Oper- culum small, scarcely half as long as the abdomen, about as broad as long, with the posterior margin rounded. — Uropods (fig. 6d) inserted far from the median line, small, not one-third as long as the abdomen; peduncle oblong, about half as long as, and little broader than, the endopod; exopod wanting. Length of a female with marsupium 1-65 mm. Remarks. This small species is allied to D. insigne, but easily distinguished by several cha- racters, some of them pointed out in the key (p. 109). It is in all probability confined to the cold area. Occurrence. Taken by the "Ingolf" at two stations in the cold area. East of Iceland: Stat. 102: Lat. 66°23' N., Long. io°26' W., 750 lath., temp. -=- 0-9°; 2 spec. North-East of Iceland: Stat. 120: Lat. 67°2c/ N„ Long. n°32' W., 885 fath., temp. ^ ro°; 1 spec. Group X. Ilyarachnini. Body moderately oblong or somewhat broad and thick. Head free and broad in proportion to length, without front area. Eyes wanting. Antennukc terminal; first joint plate-shaped; flagelluni frequently with several joints. Antennae longer than the body, as the two distal joints of the pe- duncle are very elongate, and the flagelluni has numerous joints; squama feebly developed or indi- CRUSTACHA MALACOSTRACA. III. 1 21 stinct. Mandibles very thick and peculiarly shaped; the incisive part short and obtuse, movable lacinia small or wanting, setse short or wanting, molar process either rather slender, tapering to the obtuse, setiferons end, or very reduced, being a very thin and somewhat or very short process with terminal setse. Palp of the maxillipeds witli second joint long and broad, third joint short and broad, the two distal joints well developed; epipod broad, moderately long. — Thoracic segments movable, divided into two sections, and the three posterior segments very different from the four anterior ones. The two anterior pairs of legs somewhat prehensile, with second joint very long, considerably or much longer than the third; third and fourth pairs moderately or very long, with second joint short, while the third is extremely long. The three posterior pairs, at least fifth and sixth pairs, natatory; fifth joint com- pressed and expanded especially on fifth pair, sixth joint somewhat or scarcely expanded, and both these joints most frequently, and at least on fifth pair, equipped with natatory setse along the upper margin or at both margins; seventh joint long. — Abdomen triangular, with the narrow end obtuse or acute. Uropods ventral, somewhat small; peduncle flattened, oblong-oval, with numerous marginal setae; endopod much shorter than the peduncle, exopod minute or wanting. Remarks. The family Munnopsidae as limited by Sars in his Account can be divided into three very natural groups. All agree in having the antennae elongate, at least third and fourth pairs of legs moderately to extremely long, and at least fifth and sixth pairs, generally also seventh pair, natatory, but they differ from each other in several characters, such as the position of the anteunuke, the shape and structure of the mandibles, the relative length of second and third joints in third and fourth pairs of legs, the posterior thoracic segments and the uropods. One feature is remarkable, viz. that while in the Ilvarachnini the third and fourth pairs of legs have the second joint short and the third extremely long, more than twice as long as the fourth, we find in the Eurycopini the second joint long, about as long as or much longer than the third, which is longer or shorter than the fourth, and in the Munnopsini both second, third and fourth joints in the same two pairs of legs short. The group comprises four genera established by Sars, viz. Ilyarachna, Echinozone, Aspidarachna and Pseudarachna, but only the two first-named genera are represented in the material from the "In- g;olf" area. Ilyarachna g. o. Sars. The generic characters have been well pointed out by Sars, excepting as to the antennulee. He wrote in the Account (p. 134—135): "Superior antennas originating close together from the front, basal joint large, sub-quadrangular in form, inner corner somewhat more projecting than the outer . . ." Sars figured his /. denticidata (PL 61) as having the inner distal corner of the basal joint of the anteu- nulae slightly produced and the outer corner strongly produced, which is correct, but opposed to his statement quoted. In /. longicornis G. O. S. and /. hirticeps G. O. S. he figured the outer corner of the same joint as slightly, the inner as strongly, produced, which is in accord with his generic diag- nosis, but is in reality wrong; in all species of Ilyarachna the antennular basal joint has the portion out- side the articulation of second joint produced into a very conspicuous, more or less triangular plate, while the distal inner corner is slightly or scarcely produced. The Ingolf-Expedition. III. >. l6 I22 CRUSTACEA MALACOSTRACA. III. The antennal squama is feebly developed (PL XI, fig. 7 a, so.; figs. 8 b and 9 b) or indistinct, it being a low protuberance on the outer side of third joint, and sometimes not marked off. Sars paid no attention to the existence of epimeral plates or processes at the four anterior thoracic segments. For instance, his figure of the female I. longicomis, seen from above, shows the anterolateral corners of the four segments a little produced and triangular, but according to my exa- mination of specimens from Skager Rak the structure is in reality rather different, agreeing more with Tattersall's figure of his Plunketti Tatt. (Isopoda, PI. VII). The first segment is at each side produced into a somewhat small, conical protuberance directed outwards, and with a single spine or two to four spines on the end ; below and a little in front of this protuberance is seen a process, which originates from the basal joint of the leg, is directed obliquely outwards and more or less forwards, and termi- nates in a spine or in two to four spines. The epimera of second segment are broad, bilobed, the posterior lobe rounded, the anterior conical, directed obliquely forwards and terminating in a spine; the anterolateral corner of the segment itself is produced into a kind of lobe which, seen from above, is narrow, directed forwards and outwards and, at least generally, without any spine. Third and fourth segments have their anterolateral corners produced and shaped as in second segment, but not marked off by any suture as figured by Tattersall, while the epimera are rounded, without any spine. As /. longicomis G. O. S. is not known from the "Ingolf area, it may perhaps be iuserted here, that by the remarks on the anterior segments the two best characters between /. longicomis G. O. S. and /. Plun- ketti Tatt. have disappeared; the third character (Tattersall, op. cit. p. 29), derived from the apparently strong difference in the shape of first joint of the antennula, must be cancelled, as the figure of Sars (see above) is quite wrong as to this feature; the fourth character, derived from the shape of the fe- male operculum, also disappears, as the upper (the basal) half of Sars' figure of the operculum is incor- rect. Consequently /. Plunketti Tatt. must be cancelled as a synonym for /. longicomis G. O. S. Of Ilyarachna as defined by Sars in the Account I have four species from our area, and three of them are new; a fiftli species not seen by me, but described by Ohlin from East Greenland (and Spitzbergen), is mentioned later on. But it is with considerable hesitation that I follow Sars in adopt- ing his genus Echinozone, which is so closely allied to Ilyarachna and based on two characters of so slight value (see later on) that it ought perhaps to have been withdrawn. My report here on Ilyarachna and Echinozone is unsatisfactory as to several points, because most of the specimens are mutilated, and of some species the material is very scanty. The single species, of which a rich material is to hand showed furthermore much variation according to age, locality and, apparently, also variation of purely individual nature. Perhaps the first pair of male pleopods may afford reliable specific characters, when a good material of several species is studied. 78. Ilyarachna hirticeps G. O. Sars. (PI. XI, fig. 7 a). 1870. Ilyarachna hirticeps G. ('). Sars, Forh. Vid. Selsk. Christiauia Eor 1869, p. 167. ! 1897. G. O. Sars, Account, II, p. 137; PI. 60. ! 1897. denticulata G. O. Sars, op. cit. p. 138; PI. 61, fig. 1. CRUSTACEA MALACOSTRACA. III. 123 According; to Sars1 figures of the animals from above, /. hirticeps and /. dm tint lata differ ex- tremely in the armature with spines on the anterior margins of the four anterior thoracic segments, and must therefore be easy to separate. But the examination of a large material from numerous places showed that /. denticulata must be cancelled, as already stated by Ohlin (1901); I found spe- cimens agreeing with /. hirticeps and others agreeing tolerably well with /. denticulata, but main- specimens showed intermediate features, and neither depth nor temperature at the bottom seems to be of real influence on the development and character of the spines mentioned. Some particulars on tin variation may be of interest. The largest specimen, a female from the cold area ("Ingolf" Stat. 126, 293 fath.) is 10-5 mm. long; it has nearly thirty spines, many among than moderately large, along the front margin of second segment. A large specimen from Stat. 103 (579 fath., temp. -^- o-6°) has about twenty-two spines on the same margin, and in a small specimen, 3-4 mm. long, from the same station the marginal spines are much less numerous, though very conspicuous. In a specimen 6-2 mm. long from Stat. 139 (702 fath., temp, -~- o-6°) the spines on the margin of second segment are somewhat small and only twelve or fourteen, while the head is almost smooth; in a smaller specimen 4-5 mm. long, from the same station, the head is quite smooth and the marginal thoracic spines few and feebly developed. — My largest specimen from the warm area, a female from Stat. 24 (1199 fath., temp. 2-4°), is 8 mm. long, has nearly thirty well developed spines on the anterior margin of second segment, and is completely similar to large specimens from the cold area; another female, 7 mm. long, from Stat. 24 has only sixteen or eighteen spines along the same margin and several among them large; a third specimen, 4 mm. long, from the same station has six large spines on the same margin. In specimens 6 — 7 mm. long from south of Iceland (Lat. 63°i5' N.) I find the anterior margin of the four segments extremely finely crenulate, the microscopic teeth being more numerous than according to Sars figure of /. hirticeps, and, besides, set with a small number of irregularly distributed, moderately strong spines, but probably some other spines have been broken off; in specimens 4-5 — 5 mm. long from the same place the fine crenulation is well developed, but no real spines are found, while the head has as usual numerous spines. In specimens 6—7 mm. long from another place in the warm area, viz. south-west of the Faroes in Lat. 6i°i5' X., the fine crenulation is very distinct, but scarcely any spine could be detected on the margins. — In very juve- nile specimens, from 2-5 to 3 mm. or rarely nearly 4 mm., the spines on the head and spines or crenu- lation on the margins of the four anterior segments are sometimes partly or totally undeveloped. All specimens, excepting the most juvenile, are distinguished from /. dubia n. sp. in having the head conspicuously broader than first segment, about as broad as, or frequently broader than, se- cond thoracic segment. / hirticeps is separated from I. longicornis G. 0. S. 1 = /. Plunketti Tatt) in having the basal joint of the antennula; (fig. 7 a) armed with a few or several spines on the outer margin, and, besides, this joint is proportionately longer than in /. longicornis. (Above, on p. 121, it has been mentioned that in Sars' main figure of /. denticulata the basal joint of the antennulce is correct, while it is wrong in his figures of /. longicornis and /. hirticeps.] Of /. longicornis one of the largest specimens is 47 mm., thus considerably larger than according to Sars, who said ''about 3 mm."; adult specimens of this species are rather similar to specimens of the same size of /. hirticeps. but can be distinguished bv first antennular joint and by having no spines on the head and no vestige of 16' 124 CRUSTACEA MALACOSTRACA. III. spines or crenulation on the front margins of the four anterior segments; small differences in the appendages may be derived from Sars' figures of both species. The shape of the proximal joints of the antennae is also of some significance ; their shape can be seen on fig. 7 a. Occurrence. Captured by the "Ingolf" at four stations in the warm area, ten stations in the cold area and one station with lesser depth. Davis vStrait: Stat. 35: Lat. 65°i6' N., Long. 55°05' W., 362 fath., temp. 3-6°; 3 spec. Stat. 27: Lat. 64°54' N., Long. 55°io' W., 393 fath., temp. 3-8°; 5 spec. Stat. 25: Lat. 63°3o' N., Long. 54°25' W., 582 fath., temp. 3-3°; 4 spec. Stat. 24: Lat. 63°o6' N., Long. 56°oo' W., 1199 fath., temp. 2-4°; ab. 16 spec. North of the Faeroes: Stat. 138: Lat. 63°26'N., Long. 7°56' W., 471 fath., temp.-^cr6°; 26spec. (young). — vStat. 139: Lat. 63°36' N, Long. 7°3o' W., 702 fath., temp. -=- o-6°; 2 spec. East of Iceland: Stat. 103: Lat. 66°23' N., Long. 8°52' W., 579 fath., temp. -f- o-6°; 6"2 spec. Stat. 102: Lat. 66°23' N., Long. io°26' W., 750 fath., temp. -4- 0-9°; 7 spec. — - Stat. 101: Lat. 66°23' N., Long. I2°05' W., 537 fath., temp, -f- 070; 1 spec. North of Iceland: Stat. 126: Lat. 67°i9' N., Long. i5°52' W., 293 fath., temp. -f- 0-5°; 3 spec. Stat. 128: Lat. 66°5o' N., Long. 20°02' W., T94 fath., temp. o-6°; 1 spec. North-East of Iceland: Stat. 120: Lat. 67°29' N., Long. n°32' W., 885 fath., temp. -=- ro°; 1 spec. Stat. 119: Lat. 67°53' N., Long. io°i9' W., 1010 fath., temp. -4- ro°; 4 spec. South of Jan Mayen: Stat. 117: Lat. 69°i3' N., Long. 8°23' W., 1003 fath., temp, -f- ro°; 1 spec. Stat. 113: Lat. 69°3i' N., Long. j°o& W., 1309 fath., temp, -f- It)0; 1 spec. Besides, /. liirticcps has been taken at nine places in our area. From Baffin Bay and Davis Strait it was recorded (by H. J. Hansen) from three localities, viz: Lat. 72°4i' N., Long. 59°5o' W., 227 fath.; Lat. 7i°io' N., Long. 58°56' W., 199 fath., and Lat. 66°32' N., Long. 55°34' W., 100 fath. — North of Iceland it was secured by Admiral Wandel in the Skagestrand Bay, 119 fath., temp. 2-9°. The "Thor' has captured it at the following five places. South of Iceland: Lat. 63°i5' N., Long. 22°23' W., 114— 172 fath.; many spec. — - — Lat. 63°5' N., Long. 20°7' W., 290 fath. ; 1 spec. — - Lat. 62°ii' N., Long. i9°36' W., 1010— 1142 fath.; 2 spec. South-West of the Faeroes: Lat. 6i°i5' N., Long. 9°35' W., 463—5151^11.; 8 spec. - - Lat. 6i°7' N, Long. 9°3o' W., 443 fath.; 9 spec. Distribution. The "Thor" captured some specimens west of most southern Norway: Lat. 58°32' N., Long. 4°i8' E., 148 fath. It has been taken along the coast of Norway from Stavanger Fjord to Vadso in Yarauger Fjord, 100 to 400 fath. (G. O. Sars). The Norwegian North-Atlantic Ex- pedition has gathered it at three stations off the coast of Norway in the cold area in depths from 417 to 649 fath.; furthermore north-east of the Faeroes in 1215 fath., temp, -f- 1-2°; between Norway and Beeren Eiland, 191 fath., temp. 3-5°; west of Beeren Eiland, 658 fath., temp. -=- 1-2°; filially at three stations near Spitzbergen, one with 1333 fath. and temp. -=- 1-4°, another only 146 fath. and temp. ri°, while the third, with 260 fath., lies even so far northwards as in Lat. 8o°3' N. Ohlin also recorded it from Spitzbergen; two of his stations are off King Charles Island, one of them even so far eastward as Lon.<;. 29°3y' E., and the depths of only 32—37 fath. and in lath. CRUSTACEA MALACOSTRACA. Ill 125 From these enumerations it is seen that this species goes down to 1199 lath- in the warm area, to 1333 fath. in the cold area, and that in high latitudes it has been taken in depths of 10 and of about 35 fath. 79. Ilyarachna bicornis n. sp. (PI. XI, figs. 8 a— 8 c). Female (without marsupium). Closely allied to / hirticeps, but distinguished by a few fea- tures. - - The head, which is about as broad as first thoracic segment (fig. 8 a), has its upper surface smooth, excepting that it possesses a single pair of conspicuous, robust, somewhat conical, acute pro- tuberances or teeth, placed at some distance from the hind margin and considerably apart from one another. — The basal joint of the antennulse (fig. 8 b) about as broad as the length of its inner margin, while the distal part outside the articulation of next joint is produced into a good-sized triangle terminating in a spine; the outer margin of the joint has no spines, but at most some short setae. - • The basal joint of the antennae (fig. 8 b) is broad, with the outer corner considerably produced and terminating in a thick spine. Second thoracic segment broader than the head and considerably broader than fifth segment (fig. 8 a). The four anterior segments with partly moderately small, partly very small spines along their anterior margins, about ten spines on the second and only a few on fourth segment; the shape of the sides of the four anterior segments and the epimera may be sufficiently seen on the figure. - The abdominal operculum has a high keel with many spines; the uropods without exopod. Length of the largest specimen 87 mm. Remarks. This species is certainly valid, not a variety of / hirticeps; the curious armature of the head and the absence of spines on the outer margin of the basal antennular joint must be good characters. Occurrence. Gathered by the "Ingolf" at a very deep station in the warm area. Davis Strait: Stat. 36: Lat. 6i°5o' N., Long. 56°2i' W., 1435 fath., temp. 1-5°; 2 spec. 80. Ilyarachna dubia n. sp. (PI. XI, figs. 9 a— 9 c). Female (without marsupium) and Male. Closely allied to I. hirticeps.-- Head distinctly nar- rower than the first and somewhat narrower than second segment; its surface is completely smooth, without any trace of spines. — Basal joint of the antennulse (fig. 9 b) scarcely as broad as the length of its inner margin, while the distal outer part is produced into a considerable or even rather long triangle terminating in a small spine and a feathered seta; the outer margin with at most a few thin, nearly setiform spines; in the female (fig. 9b) the flagellum consists of five joints, the four distal very slender and moderately long, while in the male the flagellum is considerably thicker and longer, with about nineteen short joints. — The anterior margin of the four anterior thoracic segments without spines or distinct crenulation. The shape of the epimera may be seen on fig. 9 a. - - The female oper- culum has a high keel with strong seta.'; the uropods without exopod. I26 CRUSTACEA MALACOSTRACA III. Length of the largest specimen, a female without marsupium, 3-8 mm., of a male 3 mm. Remarks. It is after prolonged hesitation that I establish this species on the rather small specimens to hand. I was inclined to think that they were only less than half-grown specimens of / hirticeps with the spines on the head and the spines or crennlation on the anterior thoracic segments still undeveloped. But the smooth and narrow head seems most probably to be a specific character, as most of my young specimens of /. hirticeps of the same size, or even a little smaller, have the head broad as in the adults, and in small specimens of /. hirticeps measuring 23—3 mm. from Stat. 138, and with the head nearly as narrow as in /. dubia, both the head itself and the thoracic segments have long spines. Furthermore, the fact that the male has the flagellum multiarticulate with well developed sensory filaments, seems to indicate that, though only 3 mm. long, it cannot be far from adult. But being, nevertheless, not quite sure, I choose to name the form /. dubia and must let the future decide its value. Occurrence. Taken by the "lngolf" at two stations in the cold area. Xorth-East of Iceland: Stat 120: Lat. 67°2c/ N., Long. n°32' W., 885 fath., temp. -f- ro°; 1 spec. (?) Stat. 119: Lat. 67°53' N., Long. io°ic/ W., 1010 fath., temp, -f- ro°; 2 spec. (o* and ?). 81. Hyarachna Bergendali Ohlin. 1901. Hyarachna Bergendali Ohlin, Bihang till K. Sv. Yet. Akad. Handl. Vol. 26, IV, No. 12, p. 37, figs. 8 a— 8 d. Of this species I have not seen any specimen. It has been established on three specimens, and wishing to examine some features I wrote to the Director of the department in question of the Riksmuseum in Stockholm, Dr. Hj. Theel, but he answered that the specimens could not be found, and after the death of Ohlin the)- were probably lost in one way or another. Consequently I can only make some remarks based on Ohlin's description and figures. /. Bergendali is certainly a valid and even an interesting species, as the antennae are much longer than in any other form from the North Atlantic; the penultimate joint of the peduncle is, ac- cording to Ohlin's fig 8 a, even a good deal longer than the whole body, and nearly half as long again as the following joint; if these joints are lost the species can be separated by having the proximal four-jointed part of the antenna: considerably larger than in the other forms of this genus. According to Ohlin's fig. 8 c the basal joint of the autennula has a somewhat aberrant shape; his fig. 8 d, repres- enting the fifth thoracic leg, shows that the third joint is considerably thicker, the fifth joint narrower in proportion to length, and the seventh joint longer in proportion to the sixth than in /. hirticeps. The anterior margins of the four anterior segments are said to be crenulate, and the fifth segment "in it at all emarginated behind", but the latter statement is somewhat astonishing. Ohlin wrote: "Oral parts in no respect differing from those in /. hirticeps" but unfortunately he did not say plainly whether a mandibular palp was found or not. Mis text contains nothing on the uropods, and I am not sure that they do not possess an exopod. For these and other reasons /. Ber- gendali is mentioned later on in the remarks on Echinozone arctica. CRUSTACEA MAIACOSTRACA. III. 127 Occurrence. Ohlin has mentioned a single specimen from East Greenland; it had been taken at Lat 7i°33' N., Long. 2i°3o' W., 106 fath. Distribution. East Spitzbergen: King Charles Island, Lat. 7>S°5o' N., Long. 27°39' L., 11 fath.; 2 spec. (Ohlin). 82. Ilyarachna spinosissima n. sp. (PI. XI, figs. 10 a — 10 e; PI. XII, figs. 1 a -1 c). Male and Female (without marsupiutn). — Head nearly as broad as second thoracic segment; in the male (fig. 10 a) above with several subcylindrical processes and minute tubercles, each terminat- ing in an articulated spine; four of the largest processes are placed in a row towards the posterior margin. and the third large pair a little from the antero-lateral margins; in the larger female the processes and tubercles are somewhat more numerous. — First joint of the antenuulae (fig. 10 a and 10 b) somewhat oblong, with a few strong spines on the outer margin; the outer distal part is produced somewhat forwards and terminates in a thick spine. - - The body formed by the four proximal antennal joints taken together is considerably larger than in /. hirticeps, and has a few tubercles of very different size. — The mandibles have the palp well developed; the left mandible (fig. 10 c) possesses a some- what small movable lacinia and some five short setae, while the moderately small molar process tapers from near the base to the narrow, setiferous end. The maxillipeds (fig. 1 a) with second joint propor- tionately longer and the epipod larger than in /. hirticeps. In the male (fig. 10 a) each of the four anterior thoracic segments has a pair of somewhat high processes a little from the front margin, one pair or two pairs of lateral processes, and the fourth seg- ment, besides, a pair of processes near the submedian part of the hind margin ; each of these processes is truncate with a terminal spine; fifth and sixth segments each with a larger number of processes and small tubercles on the surface and the lateral margins; the small seventh segment (fig. 10 a and 1 b) has two dorsal processes and a spine on each lateral angle. In the female, which is much larger than the male specimen, the same processes and tubercles are found on the segments and, besides, a good number of tubercles. — Fifth pair of legs (fig. 10 e) differ much from those in /. hirticeps; fifth joint is very moderately expanded, about two and a half times as long as broad, and with natatory setse only on the upper margin, while the opposite margin has some spines, three of which, on the proximal third, long and strong; sixth joint is not expanded and without real natatory seta. Abdomen (fig. 1 b) oblong-triangular as in the preceding forms, but the proximal half of its lateral margins has some protuberances, all probably terminating in spines; the upper surface has several moderately small or quite small tubercles almost regularly arranged on its median area and towards the lateral margins. — Operculum in the female without any conspicuous keel. - - Uropods (fig. 1 c) with the endopod somewhat long and no exopod. Length of the female without marsupium 6 mm., of the male 3-8 mm. Remarks. I. spinosissima is instantly separated from all other species of the genus by the numerous processes and tubercles on the surface of the body. The mandibular palps and the absence of the exopod on the uropods shows it to belong to this genus. 128 CRUSTACEA MALACOSTRACA. III. Occurrence. Taken by the "Ingolf" at two of its deepest stations in the warm area. Davis Strait: Stat. 36: Lat. 6i°5o' N., Long. 56°2i' W., 1435 fath., temp. 1-5°; 2 spec. ( temP- r4°; 1 spec As the specimen is unusually well preserved it has perhaps been taken from the meshes of the trawl, and may possibly be pelagic not too far from the bottom. Storthyngura Vanhoffen. Vanhoffen wrote (1914) that he established this new genus for the forms of Eurycopc "mit za- ckigem Abdomen, welche auf dem Korper mehr or weniger bedornt sind", while he left in the old genus all the species with the abdomen rounded. He enumerated eight species as belonging to Stor- thyngura, viz. .5". intermedia Bedd., J>". fragilis Bedd., S. atlantica Bedd., S. iiovw-zcalandia: Bedd., S. ca- ribbca Benedict, i". truncata Richardson, .V. magnispinis Richardson and S. elegans Vanhoffen. He over- looked my description and figures of Eurycopc pulclira H. J. H. (1897), as this species is well armed with processes and must be referred to Storthyngura. The existence of processes on thorax and abdomen is a very practical character, but if no other features could be found, the genus would seem to be of slight value. I have seen only two of the species enumerated, viz. 51 pulclira H. J. H. and S. magnispinis Rich., but these species show some characters which can be considered of generic value if they also are found in the other forms, which probably is the case. In both species the anteunal squama is not set off by an articulation. The three posterior thoracic segments, though well marked off in .V. pulclira, are immovably coalesced, while the articulation between thorax and abdomen allows in S. magnispinis a quite feeble and in S. pulclira no movement. The natatory legs have the fifth joint at least almost twice as long as broad, thus less expanded than in Eurycopc, and the median lamella of the male operculum has its central part con- spicuously narrower than the more distal portion. Only a single species is known from our area. 86. Storthyngura magnispinis Richardson. (PI. XII, figs. 3.1-311). 1908. Eurycopc magnispitlis Richardson, Proc. U. S. Nat. Mus. Vol. XXXY, p. 84, fig. 21. Description. Body rather oblong, moderately convex.- Head (fig. 3b) without dorsal keels or any limited front area; somewhat before the posterior margin a pair of broad and low, circular pro- tuberances. - Antennuhe considerably less than half as long as the body, in the main as in Eurycope, inserted on the dorsal surface of the head not far from one another; first joint depressed, about half as long again as broad, with the antero-interior angle produced into a conspicuous subcorneal process, while the distal outer, somewhat broad part is considerably produced and broadly rounded, so that the second joint, which is rather small and about as long as broad, is placed on the distal upper sur- face of first joint; third joint slender and about as long as the second; the flagellum with numerous CRUSTACEA MAI.ACnSTRACA. III. 133 joints. — The antenna; somewhat more than twice as long as the body. Third joint of the peduncle (fig. 3 b) large, with the inner distal corner produced into a rather long and robust process directed forwards and inwards; the squama is not marked off by any suture and is shaped as a subcorneal tubercle; fourth joint about as long as broad. -- Left mandible (fig. 3 c) witli the incisive part, the movable lacinia and the seta; well developed; the molar process is of moderate length, very thick at the base and tapering much to the end, which is cut off obliquely. Maxillipeds (fig. 3d) moderately broad; fifth joint very broad and slightly broader than the fourth; sixth joint with a somewhat small loin-; epipod more than twice as long as broad, with the distal half of the outer margin concave, and a protruding obtuse angle at the middle of that margin. First thoracic segment (fig. 3a) has a small median process at the front margin, while each anterolateral corner is produced in a robust and moderately long process directed outwards and much forwards. The three following segments each with a median process directed upwards and forwards; eacli antero-lateral corner is produced into a slender, moderately small process, and behind this pro- cess is seen the epimeron, which is bifid, with its anterior part produced into a long and a little curved process directed forwards and outwards, and its posterior part produced into a tooth. - The three posterior segments are, as already stated, dorsally fused, so that no suture but only an impression is seen on the broad median part; each segment has a pair of short dorsal processes, and the side is produced into a long, on fifth and sixth segments even very long, process. The dorsal processes 011 all thoracic segments are conspicuously longer in the male than in the female, while sexual difference is on the whole less developed in the lateral processes. — The articulation between thorax and abdo- men allows a feeble movement. The four anterior pairs of legs (figs. 3 e, 3 f and 3 g) in the main as in some smaller species of Eurycope; third (fig. 3 g) and fourth pairs subsimilar in length and scarcely as long as the body. Se- cond to fourth pair yet more robust than in Eurycope, with sixth joint increasing considerably in depth from base to end, and armed on the lower margin with a moderate number of strong spines; seventh joint uncommonly long, about two-thirds or more as long as the sixth. -- The three posterior pairs (figs. 3 h and 3 i) have the fifth joint about three times as long as broad, and almost the proxi- mal half of the lower margin without setae; sixth joint nearly as long as the fifth, but conspicuotish more narrow and closely set with setae along most of both margins; seventh joint -- with the claw not marked off — about three-fourths as long as sixth joint. Abdomen without any short basal segment; on the surface it has a median vertical tubercle near the base and a pair of submedian minute tubercles behind the middle, on each side a long basal process and a similar process behind the middle; the posterior part is a little produced but broad, with the hind margin straight or slightly concave, and each lateral margin of the produced part is convex and serrate (fig. 3 k). - The female operculum (fig. 3 k) nearly as long as broad, with the po- sterior half of the margin semicircular. In the male the operculum (fig. 3 1) is somewhat longer than broad; the median lamella with the major part of the lateral margins considerably concave, as each pleopod tapers from the moderately broad basal part to near the middle, then is gradually widened to somewhat from the end and finally tapers to the end, which has the outer corner produced as a tri- angular protuberance, and the hind margin of the lamella is consequently somewhat incised (fig. 31111; 134 CRUSTACEA MALACOSTRACA. III. the second pair of pleopods (fig. 3 n) have the distal part somewhat produced, moderately broad and rounded; the copulatory organ is somewhat slender, with the distal fourth filiform and reaching a little beyond the end of the pleopod. — The outer branch of the penultimate pair of pleopods is narrow. - Uropods (fig. 3 k) in both sexes of very moderate size; the peduncle is slender and about as long as the endopod, which is somewhat longer than the thinner exopod. Length of a female with marsupium 4-2 mm., of a large male 37 mm. Occurrence. Taken by the "Ingolf" at two deep stations in the warm area. Davis Strait: Stat. 24: Lat. 63°o6' N., Long. 56°oo' W., 1199 fath., temp. 2-4°; 6 spec. Stat. 36: Lat. 6i°5o' N., Long. 56°2i' W., 1435 fath., temp. 1-5°; 7 spec. Distribution. The type, hitherto the only specimen recorded, was taken by the "Albatross" at Lat. 39°49'N., Long. 68°28V2' W., 1467 fath., temp. 3-15° (Richardson). MunnopSUrUS Richardson. In 1912 Harriet Richardson published a small paper containing only the description of Mitn- nopsurus arcticus, n. gen., n. sp., from Novaya Zemlya. This form is Eurycopc gigantca G. O. Sars, but while Richardson's name for the species consequently must be cancelled, the genus may be accepted. To the same genus Eurycopc lotigipes Tatt. must be referred; that Eurycopc gigantea G. O. S. forma australis Vanhoffen from the antarctic deep-sea is a third species can be considered as certain, but Vanhoffen's statements and figures (1914) are too imperfect for closer comparison. In Eurycope the mandibles have the incisive part and the movable lacinia (on the left man- dible) well developed, and generally a row of setse, rarely a single seta; the molar process is consider- ably protruding, rather well defined and its end cut off (comp. figures in Sars' Account). In the two northern species of Munnopsurus the incisive part is simple with a single tooth or none, the movable lacinia and the row of sette completely wanting, while the molar process is shaped as a broad, large tubercle with the end rounded and bearing a small bundle of stiff seta: (PI. XII, fig. 6a, and, besides, figs. 3c and 3d on PL XX in "Dijmphna-Togtets zool.-bot. Udbytte", 1887). In no form hitherto stu- died mandibles intermediate between those in Eurycopc and Munnopsurus have been found, and there- fore the genus Munnopsurus must at least provisionally be considered as valid, though in all other characters it is closely allied to Eurycope, in that the antenna have a distinct squama well marked off. that all thoracic segments are movable, that second joint of second to fourth pair of legs are long, longer than the third joint, and that second pair is very similar to third pair. The name Munnopsurus is unfortunately badly chosen, as its species are related to Eurycope and not to Munnopsis, from which they differ much, not only in general aspect but in having second pair of legs nearly similar to third pair (in Munnopsis second pair is prehensile, similar to first and very different from third pair), in having second joint of third and fourth pairs of legs long, in the posterior thoracic segments, uropods, etc. The two above-named northern species are represented in the material. CRUSTACEA MALACOSTRACA. HI. 135 87. Munnopsurus giganteus G. < >. Sars. (PL XII, fig. 5 a). 1877. Eurycope gigantea G. O. Sars, Arch. f. Math, og Naturv. Vol. II, p. 353. 11885. G. O. Sars, North-Atl. Exp. p. 130; PI. XI, figs, r - 25. ! 1887. H. J. Hansen, Dijmphna-Togtets zool.-bot. Udbytte, p. 198; PI. XX, figs. 3—3 in. 1912. Munnopsurus arcticus Richardson, Bull. l'Inst. Ocean. Monaco, No. 227. The differences between this species and M. longipes are pointed out in the notes on the latter form. There is scarcely more than a single point to add to the accounts given by Sars and myself. Bnt that point is a morphological observation of interest. Fig. 5 a represents the proximal parts of the left maxilliped seen from below and attached to its sternite (sf). The small transverse plate, pcx, is the hitherto undiscovered prsecoxa of the maxillipeds, placed between the sternite (st) and the coxal joint, ex, generally considered as the first joint. The piece marked bus is the proximal part of the "basis", generally described as the second, long joint of the maxillipeds. So much may be said here; the whole morphological question has been dealt with on p. 9. — It may be noted here that the exo- pod of fourth pair of pleopods is rather broad, lamellar, with the end obtuse. Among the large material seen by me I have found only a single female with the marsupium fully developed; this female, from the Kara Sea, is 33 mm. long; the general size of the male is 24 — 25 mm. Occurrence. Taken by the "Ingolf" at seven stations, all in the cold area. North of the Faeroes: Stat. 140: Lat. 63°29' N., Long. 6°57' W., 780 fath., temp. -+- 0-9°; 2 spec. Stat. 138: Lat. 63°26' N., Long. f=,6' W., 471 fath., temp. H-o-6°; 4 spec. East of Iceland: Stat. 103: Lat. 66°23' N, Long. 8°52' W., 579 fath., temp, -i- 060; 1 spec. — Stat. 101: Lat. 66°23' N., Long. i2°o5' W., 537 fath., temp. -f- 070 ; 10 spec. North of Iceland: Stat. 126: Lat. 67° 19' N, Long. i5°52' W., 293 fath., temp. -=- 0-5°; 4 spec. — - vStat. 124: Lat. 67°4o' N, Long. i5°4o'W., 495 fath., temp, -f- o-6°; 2 spec. South of Jan Mayeu: Stat. 116: Lat. 70^5' N., Long. 8°26' W., 371 fath., temp. -=-04°; 1 spec. It has been taken in Denmark Strait off East Greenland at Lat. 66°42' N, Long. 26°4o' W., 292 fath., temp, -j-o-n0 (Grieg), and east of Iceland: Lat. 64°53' N., Long io°oo'W. (Grieg). Distribution. Taken by Admiral Waudel a little south-east of the Faeroes: Lat. 61 "23' X*.. Long. 4°2i' W., 505 fath., temp. -f- 0-4°, 1 spec. It was recorded from the Fierce Channel, 540 fath. (Nor- man) and from a place north-east of the Faeroes: Lat. 63°o6' X., Long. 2°46' W., 483 fath., temp. -4- 1-07° (Grieg). Sars recorded it from three stations in the cold area west of Norway, 350 to 634 fath., from two places in the cold area west and south-west of Beereu Eiland, 447 and 658 fath., finally from two places north-west of Spitzbergen, respectively 459 fath., temp, -f- ro°, and Lat. 80^3' N., 260 fath., temp. + n°. Furthermore recorded from King Charles Island, East Spitzbergen, 32 — 37 fath. (Ohlin), from near Beereu Eiland, 40 fath. (Hoek), from Novaya Zemlya (Richardson), and it is very common in the Kara Sea, 20 — 100 fath. (several authors). That I consider M. giganteus forma australis Vanhoffen from the antarctic deep-sea as a dif- ferent species is already mentioned. 136 CRUSTACEA MALACOSTRACA. III. 88. Munnopsurus longipes Tattersall. (PI. XII, figs. 6 a— 6 b). 1905. Eurycopr longipes Tattersall, Isopoda, p. 30 and 75; PI. X, figs. 1 — 8. Tattersall has published an elaborate deeription with figures of this species. He also pointed out differences between it and M. giganfeus, but some points may be added, and a view of the main differences may be useful. Of M. longipes I have only a single adult specimen, a male, but there is probably no sexual difference worth mentioning in any organ excepting the abdominal operculum. In M. gigantens the antennulae are less than half, in M. longipes considerably more than half, as long as the body. In M. gigantcits the first pair of legs, when stretched backwards, do not reach the middle of the abdominal operculum; their second joint is about as long as the sum of fifth and sixth thoracic segments and somewhat shorter than fifth joint, which is conspicuously thicker towards the base than at the end, and at least twice as long as the sixth; in M. longipes this pair are con- spicuously more slender and reach to the posterior end of the abdomen, while their second joint is very conspicuously longer than the sum of fifth and sixth thoracic segments, and only a little shorter than, or as long as, fifth joint, which is scarcely thicker at the base than near the distal end, and considerably less than twice as long as the sixth. — In M. gigantens the abdomen is broader than long, and as long as the three posterior thoracic segments combined; in M. longipes it is produced back- wards, somewhat longer than broad, and conspicuously longer than the three posterior thoracic seg- ments combined. -- In M. giganteus the median lamella of the male operculum tapers feebly near the end, and the end of each half is divided by a deep incision into two lobes, the outer almost as large and nearly as long as the inner; in M. longipes the median lamella tapers posteriorly much (fig. 6 b) to a little from the end; each half terminates in a short and broad, rounded lobe, and at the outer base of this lobe a somewhat small, triangular and acute tooth is seen. — Finally it may be mentioned that in M. gigunteus the molar tubercle on the mandibles is low, in M. longipes much higher (fig. 6 a); in M. giganteus the maxillipeds have second and third joints of the palp considerably broader in propor- tion to length than in M. longipes (comp. my fig. in "Dijmphna-Togtet" with Tattersall's fig. 5), while the difference in the shape of sixth joint mentioned by Tattersall must be merely accidental. As my specimens of M. longipes have lost most of the antenna; and thoracic legs excepting first pair, there may exist unknown differences between M.gignnleus and M. longipes in these appendages. Length of the best male 18-2 mm. — The other specimen is only 4-3 mm. long, and differs from the large specimen especially in the shape of the end of the median opercular lamella, but supposing that this difference and the somewhat more clumsiness of the preserved second joint of first legs art- due to age, I venture to refer it to the present species. Occurrence. Taken by the "Ingolf" at a deep station in the warm area. Davis Strait: Stat. 36: Lat. 6i°5o' N., Long. 56°2l' W., 1435 fath., temp. 1-5°; 1 spec, (large <$). The "Thor" captured the above-mentioned small male in the warm area. South-West of the Fseroes: Lat. 6i°i5' N., Long. 9°35' W., 463 — 515 fath.; 1 spec. Distribution. Some specimens taken in tow-net on liawl or dredge at two places west of Inland, 350 and 454 fath. (Tattersall). CRUSTACEA MALACOSTRACA. III. 137 Eurycope G. o. Sars. In his Account Sars has given a good description of the genus with eight Norwegian species. From Eurycope as limited by Sars a number of species not known at Norway have now been removed to Storthyngura Vanh. and Munnopsurus Rich. After their removal the remaining forms agree with each other in the following features. The body is oval and more or less oblong, without processes. Head generally with the front area sharply limited, at least on the sides. The antennulse are dorsal; first joint large, depressed, rather plate-shaped; second joint somewhat small and short, inserted on the upper surface of first joint before its front margin. Antennae with a distinct exopod marked off by a suture or articulation. Mandibles with the cutting edge divided into teeth, the movable laciuia on left mandible well developed, some setse or at least a single seta behind the laciuia, the molar process well developed, moderately to very thick with the end cut off. The articulation between thorax and abdomen and that between sixth and seventh segments well developed. The legs are mentioned in the diagnosis of the group; in this genus fifth joint of the natatory legs is very broad, considerably or much less than twice as long as broad. Median lamella of the male operculum at most very moderately narrowed at the middle. The material from our area comprises thirteen species, six among them described from Nor- way, four have been established on species from various other seas and three are new. As some of the forms unknown to Sars are more or less allied to one or another of his species, I found it useful to pay attention to some features not valued or totally omitted in his descriptions, and to correct a few points. Nearly all my specimens have lost most or all thoracic legs, but fortunately their shape or length are not necessary for the determination, as the peduncles of antennulce and antennae, the surface of the head, the maxillipeds, the uropods, and the abdominal operculum in both sexes afford a number of excellent characters. The thirteen species may be divided into two sections in the following way. Sect. A. Species with all thoracic segments freely movable Species 89 to 97. Sect. B. Species with fifth and sixth thoracic segments completely fused, without any dorsal suture between them excepting near the lateral margins Spec. 98 to 101. 89. Eurycope Murray i Walker. (PL XII, figs. 7 a— 7 b). 1903. Munnopsis? Murrayi Walker, Ann. Mag. Nat. Hist. Ser. 7, Vol. XII, p. 227; PL XVIII, figs. 1—6. 1905. Munnopsis Murrayi Tattersall, Isopoda, p. 24 and 73; PL V, fig. 8. 191 1. Tattersall, Nord. Plankton. VI. Die uord. Isopoden, p. 190, figs. 8—14. 1913. Munneurycope Tjalfiensis Stephensen, Vid. Medd. fra den naturh. Forening i Kjobenhavn, Vol. 64, p. 99, figs. 6—8. 11915. Stephensen, in Rep. Danish Oeeauogr. Exped. 1908—10, Vol. II, No. 3. p. 23, figs. 12—13. The rngolf-Expedition. III. 5. *" ^ CRUSTACEA MALACOSTRACA. III. The species has suffered from a peculiar fate. It is difficult to understand why Walker referred it, with a query, to Munnopsis, as a glance at his figure of the male shows that the animal stands far apart from that genus. Tattersall not only followed Walker, but even dropped the query; it is a new, interesting case that a wrong opinion not infrequently acts as a suggestion, so that a successor accepts it without critical consideration. Stephensen, who believed that the species established by him as Munneurycope Tjalfiensis was closely allied to but differing from M. Murrayi Walk., saw correctly that it could not remain in Munnopsis, and therefore he established a new genus for its reception, but neither in 1913 nor in 1915 was he aware that it would be difficult to point out any character worthy to be considered of generic value between his new genus and Eurycopc. In his first-named paper vStephensen referred the forms Munnopsis longicornis H.J.H. and M. oceanica Tatt. to his new genus Muuncurycope, but in 1915 he removed these two species, so that Munneurycope should only comprise M. Murrayi Walk, and M. Tjalfiensis Steph. Before entering on the question whether M. Tjalfiensis is a synonym to M. Murrayi or a sepa- rate species, we may examine some features in order to judge of the validity of the genus Munneu- rycope. In 1915 Stephensen published a good number of new figures of his M. Tjalfiensis. The first anteunular joint is somewhat longer than broad, while in most, but not in all, species of Eurycopc it is broader than long; furthermore, and what is of more interest, the joint in M. Tjalfiensis has no produced antero-lateral part, but in this respect it agrees with the joint in E. uodijrons u. sp. Conse- quently the outline of first anteunular joint does not afford any generic character. As to the shape of the body, the antennulte, antenna;, mandibles, maxillipeds, thoracic segments and legs — ■ as far as these are known — M. Tjalfiensis agrees in every point wrorth mentioning with Eurycopc. The uropods have the exopod extremely small as compared with the endopod, but in several species of Eurycopc the exopod is somewhat or considerably shorter than the exopod. The lateral plates of the male oper- culum show a difference, as the big coupling hook in M. Tjalfiensis (fig. 7 b) originates at the trun- cate, nearly transverse end of the plate, while, f. inst, in E. inermis (PI. XIII, fig. 2 h) the postero- interior angle of the plate is cut off in a way so that the margin where the hook is seen in reality is the somewhat oblique terminal part of the inner margin of the plate. Finally, the penultimate pair of pleopods have the exopod narrow with the end acute in several species of Eurycopc, while it is broader, obtuse, lamellar in M. Tjalfiensis, but unfortunately it is unknown in some species of Eury- t ope. All differences observed have been enumerated here, and I suppose that Carcinologists may agree that Munneurycope cannot be maintained. Hut still one particular may be mentioned. Stephensen pointed out an antennal scale in his species and added: "This last feature has not, as far as I am aware, hitherto been found in any Munnopstd" ; the antennal squama, however, had been figured and men- tioned by me in [887 ("Dijmphna-Togtet") as found in Eurycope gigantea. Then the question on the validity of /•.'. Tjalfiensis as separate from E. Murrayi. In 1915 Stephensen enumerated four differences between his own animals and the descriptions and figures oi Walker and Tattersall. His animals had "a blackish brown or dark colour", which he thinks is a specific character, because this unusual colour was not mentioned by the two English authors. It is true that Walker said nothing on the colour, but the description had been published more than four and a half years after the capture of the specimens, ami I suppose that they had been preserved al- CRUSTACEA MALAO KSTRACA. III. 139 most four years before his description was written, and thai therefore the colour had faded consider- ably; Tattersall said in 1911: "Farbe schwarz". — Stephensen wrote: "The exopodite is evenly rounded, not indented; "exopodite" is decidedly an error instead of "epipodite", and when Walker figured the distal part of the inner margin of the epipod of the maxillipeds as deeply and irregularly serrate it is certainly erroneous; such serration is not found in any species of the group Eurycopini. — I do not quite understand Stephensen's statement on the palp of the maxillipeds: "the median edge of the 4. joint being somewhat concave" in .17. Tjalfiensis ; judging from his figure he ought to have said "third joint", as this joint has on his figure the inner margin somewhat concave; unfortunately his figure is as to this point either incorrect or shows an individual variation, as I have examined the maxillipeds of some specimens, two among them of his material, and found the inner margin of third joint convex, as shown on my drawing (fig. 7 a), consequently as to this point agreeing with Walker's figure of /:'. Murrayi. That, as Stephensen pointed out, there is a difference in the small terminal joint of the maxillipeds between his species and Walker's figure is true, but it is easily seen that Walker's figures as to such minor points are not accurate, and his text contains nothing on that feature. - Finally Stephensen wrote: "Furthermore, the metasome of M. Murrayi is stated as having a blunt carina; no trace of any such is visible in the specimens here concerned." The upper surface of the abdomen is so thin-skinned that it generally is irregularly impressed or somewhat folded, and no keel is visible, but on a well preserved abdomen I observed a feeble, rounded median keel. For the rest, Stephensen and Walker use the term "metasome" for very different sections of the body, as Walker wrote (p. 228) : "The first of the last three segments (metasome) with a large oval tubercle on each side of a central cleft." On well preserved specimens I find a longitudinal median cleft not only on the first segment but on the anterior part of the two other segments of the so-called "metasome" — on sixth and seventh thoracic segments the cleft widens posteriorly to be a triangular depressed area — and the surface at each side of the cleft is raised like a badly defined blunt keel or oblong tubercle. It is seen that Walker wrongly applies the name "metasome" used by Sars in his Account, and Stephensen has followed Sars. — I have now analyzed all the differences pointed out by Stephensen as distinctive characters for his Munneurycope Tjalfiensis, and it is absolutely certain that it must be cancelled as only a sy- nonym. But it may be added that most of Stephensen's figures in 1915 are fairly accurate, and quite sufficient not only for recognizing the species, but for judging of its peculiarities and affinities. Occurrence. E. Murrayi has not been gathered by the "Ingolf", but at a single locality by the "Tjalfe" and at five places by the "Thor". West of Cape Farewell: Lat. 6o°07' N., Long. 48°26' W. 2000 m. wire out, 1 spec. (Stephensen). South of Iceland: Lat. 6i°34' N., Long. i9°05' W., Young-fish trawl, 1N00 m. wire out; 4 spec. — - Lat. 6i°3o' N., Long. i7°o8' W., 1800 m. 14 spec. — - Lat. 62°47' N., Long. i5°03' W., — 1500 m. 2 spec. South-West of the Faeroes : Lat. 6o°oo' N., Long. io°35' \V., Young-fish trawl, 1000 m. wire out; 3 spec. - - — Lat.59°52'N., Long. 9°53'W., 1500 m. - 5' _,spec. Distribution. Taken at several places west of Ireland in towuet or young-fish trawl in depths between 350 and 1710 fath. (Walker and Tattersall.). Stephensen recorded it in 1915 from seven 140 CRUSTACEA MALACOSTRACA. III. places in the East Atlantic between Lat. 50°25' N. and Lat. 35°53' N., in every case taken in the young- fish trawl, with the length of the wire out varying from 1500 to 2800 m. According to all gatherings E. Alitrrayi is pelagic in the deeper layers of the North-Atlantic. 90. Eurycope nodifrons n. sp. (PI. XIII, figs. 1 a— 1 d). Male. Body (fig. 1 a) a little more than two and a half times as long as broad. — The head (fig. 1 b) has no sharply defined front area, and consequently no rostral projection from the vertex be- tween the first joints of the antennuke, but the area in question is raised longitudinally, with its posterior half forming a broad, rounded keel marked off by a transverse impression from the anterior half, which is a much broader, considerably vaulted and rounded protuberance. Of the antennulse only the left is present. Basal joint rather plate-shaped, slightly broader than long; the lateral margins converge much forwards to the anterior margin which is transverse, and consequently the joint has no antero-interior lobe; second joint, which is inserted considerably behind the front margin of the first, is medium-sized and slightly longer than the third. — Antenme (fig. 1 a) with the squama well set off, moderately small and somewhat broad at the base, triangular, scarcelv one-third as long as the breadth of the third joint. — Mandibles nearly as in E. cornuta. Maxillipeds (fig. 1 c) in the main as in E. Murray/', but a little more narrow, and the epipod is about three and a half times as long as broad, with the end obtuse and the outer margin not concave. First thoracic segment (fig. 1 a) uncommonly long, much longer than the second, while the fourth is very short at the middle; the epimera considerably produced, seen from above triangular and acute. The three posterior segments with their anterior margins extremely convex, and the segments have a longitudinal median impression. Abdomen longer than in the following forms, nearly as long as broad. — The pleopods of first pair in the male (fig. 1 d) have the proximal two-thirds of their outer margin somewhat concave; at the beginning of the distal third the)- are as broad as at the base, and then each tapers posteriorly; the terminal part of each pleopod is divided by a rounded incision into two lobes, the inner a little longer than broad, distally broadly rounded and much longer than the outer, which is much narrower and shorter and directed more outwards; the broad end of the inner lobe has several setae, while the narrow end of the outer is naked. The lateral plates of the operculum rather broad, with the terminal part feebly produced, and the thick coupling hook originating before the rounded end. — (Uropods lost). Length 5 mm. Remarks. At first sight E. nodifrons is somewhat similar to E.inermis n. sp. or E. cor- intta G. O. S., but on a closer examination it is seen to be intermediate between E. Murrayi and /:'. inermis in having no front area marked off by keels and no rostral projection, in the shape of the anterior half of first antennular joint, the shape of the three posterior segments and the abdomen, as the segments have the anterior margin very convex, and the abdomen is somewhat long. One of the most interesting differences between E. Murrayi and E. nodifrons is, that in the former species the basal joint of the anterior pairs of legs is not produced as an epimeral triangular plate, while in B. nodifrons epimeral plates are well developed as in /.. cornuta and several other species. CRUSTACEA MALACOSTRACA. III. 141 Occurrence. Taken by the "Ingolf" at a single deep station in the warm area. Davis Strait: Stat. 36: Lat. 6i°5o' N., Long. s6°2i' W., 1435 fath., 1-5°; 1 spec. 91. Eurycope cornuta G. O. Sars. (PL XII, figs. 8a-8k). 1864. Eurycope cornuta G. O. Sars, Forh. Yid. Selsk. Christiania for 1863, p. 209. 1880. robusta Harger, Rep. U. S. Com 111. Fish and Fisheries for 1878, Ft. VI, p. 332; PL III, fig. 15. ! 1897. cornuta G. O. Sars, Account, II, p. 145; PL 64. The richly illustrated account of this species published by Sars is good as to most particulars, but in order to distinguish E. cornuta easily from the two or three next species it is found necessary to redescribe several features. Sars wrote: "... front produced to an acute, rostriform projection", but this is not correct. The projection or rather the anteriorly protruding part of the front area (fig. 8 a) is not acute; even in co- types of Sars the end of the projection is emarginate and its upper surface somewhat concave long- itudinally, as the lateral margins of the front area are raised; furthermore the projection, when seen essentially from the side, has its upper margin horizontal, with the anterior part not curved down- wards, and the end of the projection free. — The basal joint of the antennulse (figs. 8 b, 8 c and 8d) varies somewhat in shape in individuals even from the same haul; the inner margin is straight, some- what or considerably shorter than the breadth of the joint, and from a little to a good deal shorter than the distance between the antero-exterior point of the front margin and an obtuse or rounded angle on the hind margin; the anterior margin is irregularly but considerably concave, as the antero- interior part of the joint is produced into a somewhat short lobe; second joint, which originates on the surface of the first considerably behind its front margin, is middle-sized, conspicuously thicker in the male (figs. 8 c and 8 d) than in the female (fig. 8 b); the third joint is much more slender, as long as or somewhat longer than the second. — Antenual squama (fig. 8 e, so) well set off, oblong-triangular, conspicuously less than half as long as the breadth of third antenual joint, and armed at the end ami laterally with a few small spines. Abdominal operculum in the female seen from below (fig. 8 g) much broader than long, in the main transversely oval, with somewhat more than the posterior half of its whole margin almost semi- circular; on the lower surface it has a median keel, which is rather far from reaching the anterior or the posterior margin. Seen from the side (fig. 8 f) the female operculum is highly vaulted, with its lower margin angular at the middle, and the posterior angle is rounded. — The median lamella (fig. 8 h) of the male operculum about two and a half times as long as broad, at the middle feebly narrowed, and the distal half of each outer margin flatly convex excepting at the end, as the terminal part of each pleopod is produced backwards as a small plate (fig. 8 i), which is broader than long, with its hind margin somewhat convex and the postero-lateral corner produced into a small, triangular tooth directed outwards. The lateral plates (fig. 8 k) of the male operculum are very broad, though consider- ably longer than broad, without any broad, hairy lobe on the outer side near the base; the outer 142 CRUSTACEA MALACOSTRACA. III. margin is slightly concave somewhat before the end, which is feebly produced as a short and moder- ately broad lobe; the copulatory organ has its second joint long, a good deal longer than the plate, especially because the part containing vesicle and duct is two and a half times as long as the distance from base to vesicle; the thick hook is placed somewhat before the end of the plate, and the distal part of the inner margin of the plate inside the articulation of the hook is a little concave. — The uropods have the basal joint thick and the rami are, as stated by Sars, subequal in length, but the outer ramus is more slender than the inner. Remarks. The differences between E. cornuta and the two next species are mentioned with these forms. Occurrence. Taken by the "Iugolf" at a single station. North of Iceland: Stat. 128: Lat. 66°5o' N., Long. 20°02' W., 194 fath., temp. c/6°; 1 spec. Furthermore it is known from three places in Baffin Bay, viz. Lat. yi'io' N., Long. 58°56' W., 199 fath. (H. J. Hansen); Umanak Fjord, ab. Lat. 710 N. (H. J. Hansen) and Lille Karajak Fjord, ab. Lat. 70°3o' N. (Vanhoffen). — But when Grieg (1907) recorded it as having been captured by the Duke of Orleans off East Greenland at Lat. 75°58I ,' N., Long. i4°o8' W., 300 m., temp. 0-4°, and, besides, in Denmark Strait at Lat. 66°42' N., Long. 26°4o' W., 550 m., temp. ^o-ii°, I consider these statements as possible but not certain, because I do not feel convinced that in the one or in both cases the animals in question have not belonged to E. incrniis 11. sp., formerly mixed up with E. cornuta by Sars (see below). Distribution. It occurs "along the whole coast of Norway, from the Christiania Fjord to Vadso, in depths ranging from 50 to 400 fathoms" (G. O. Sars). In Skager Rak it has been taken 15 sea-miles from the lighthouse of the Skaw, 125 fath. (Meinert), and at a number of places between Jutland and Norway, in 130 to 340 fath. (H. J. Hansen); the "Thor" has taken it in the North Sea off southern Norway: Lat. .58°32' N., Long. 4°i8' E., 148 fath. As E. robusta Harger is synonymous with E. cornuta, this species is known from the Gulf of St. Lawrence, 220 fath. (Harger), but when Harriet Richardson (1905) said: "Atlantic coast of North America" I do not know any observation on which this too loose statement has been based, and the authoress had evidently not seen any specimen in the U. S. National Museum, as she only reprinted the text of Sars. In 1886 Sars wrote that the Norwegian North-Atlantic Exp. had taken E. cornuta at nine stations in the cold area between Norway — Spitzbergen and Iceland — the Faeroes in depths from 350 to 12 15 fath., and that the animals were in part unusually large. I am sure that all these animals 111 reality belong to E.inermis 11. sp. and /.. Hanseni Ohlin, both larger than E. cornuta, similar in aspect and inhabiting the cold area. Stuxberg's statement, copied by Sars and Stephensen, that E. cornuta has been taken in the Kara Sea, is also certainly wrong. 92. Eurycope inermis u. sp. (PI. XIII, figs. 2 a— 2 1). Description. .Similar to I'., cornuta, but larger and a little more slender (fig. 2a), as a rule a little more than two and a hall times as long as broad. CRUSTACEA MALACOSTRACA III 143 Tlie front area between the basal joints of the antennulse is rather narrow, longitudinally eon- cave with the lateral margins raised (fig. 2 b| and it tapers strongly forwards, but instead of being horizontal as in E. cornuta or E. Hanseni its anterior part, seen somewhat from the side, slopes down- wards and terminates below without being well marked off from the front just above the vaulted clypeus. -- First joint of the antennulae (fig. 2 c) shaped as in /■'.. cornuta, especially as the autero-interior lobe is somewhat short, considerably broader than long; second joint inserted and shaped as in K cor- nuta, considerably thicker in the male than in the female, a little or somewhat shorter than the more slender third joint. — Antennal squama (fig. 2 d) oblong-triangular, terminating in a spine, and slightly more than half as long as the breadth of third joint. - - Maxillipeds (fig. 2 e) have the second joint, counted to the articulation of third joint, half as long again as broad (in this respect it agrees with E. cornuta, as Sars has figured the second joint too short); the distal terminal corner of the epipod more rounded than in E. cornuta. Thoracic segments (fig. 2 a) nearly as in E. coruuta\ the epimera a little more produced than in that species. The female abdominal operculum very different from that in /:. cornuta. Seen from below (fig. 2 k) it is broadest much before the middle, and then the most lateral part at each side constitutes a rounded rather short lobe directed outwards; the great portion of the operculum behind these lobes is triangular with each lateral margin a little sinuate and the end narrowly rounded; the surface has a high keel from somewhat from the base to near the posterior end. Seen from the side (fig. 2 i) the operculum is highly vaulted; its lower margin is bent, forming an angle of about ioo°, which is placed so far backwards as the insertions of the uropods; the lower margin in front of that angle is long and straight, and behind the angle it is somewhat short and almost vertical on the lower surface of the abdomen. — The male abdominal operculum also differs much from that in E. cornuta. Its median lamella (fig. 2 f) is narrower, about three times as long as broad and more constricted near the middle; the terminal part of each pleopod forms a lobe (fig. 2 g), which is conspicuously longer than broad, with setse along its somewhat convex outer margin and the rather narrowly rounded end; at the outer base of the lobe the pleopod is suddenly a little broader, so that a small and low, rectangular and protruding part is observed. Each lateral plate (fig. 2 h) has a kind of low but very broad, rounded lobe on the basal part at the outer margin, and this lobe has some marginal hairs; the distal half of the outer margin is considerably concave, and the terminal part of the plate is produced into a lobe as long as broad, with both margins concave and the end narrow with a few setse; the copulatory organ has its second joint rather short, only a little more than half as long as the plate, and the part containing vesicle and duct is only somewhat longer than the part from base to vesicle; the distal part of the inner margin, which runs across the coupling hook, is more concave than in E. cornuta. - Uropods (fig. 2 1) with the peduncle thick, the endopod considerably longer and much thicker than the endopod. Iyength of large males from the cold area 9—93 mm., of females 9—11-5 mm., but the two largest females, both. from the "Iugolf" .Stat. 103, have no marsupium and are 11-2 and 11-5 mm. long. The length of adult specimens from the warm area 5-5 to 7 mm. Remarks. E. inermis is abundantly distinguished from E. cornuta by having the anterior part of the front area curved downwards and without any freely protruding end, by the unequal length 144 CRUSTACEA MALACOSTRACA. III. of the rami of the uropods, and by the totally different shape of the abdominal operculum both in male and female. But I have been unable to find any difference excepting in size between specimens from the warm and from the cold area. — The differences between E. inermis and E. Hanseni are pointed out later on. Occurrence. Taken by the "Ingolf" at five stations in the warm and nine in the cold area. Davis Strait: Stat. 35: Lat. 65°i6' N., Long. 55°05' W., 362 fath., temp. 3-6°; 9 spec. Stat. 28: Lat. 65°i4' N., Long. 55°42' W., 420 fath., temp. 3-5°; 5 spec. Stat. 27: Lat. 64°54' N., Long. 55°io' W., 393 fath., temp. 3-8°; 4 spec. West of Iceland: Stat. 96: Lat. 65°24' N., Long. 29°oo' W., 735 fath., temp. 1-2°; 1 spec. South-West of Iceland: Stat. 76: Lat. 6o°5o' N., Long. 26°5o' W., 806 fath., temp. 4-1°; 1 spec. North of the Faeroes: Stat. 139: Lat. 63°36' N., Long. 7°3o' W., 702 fath., temp. -=-06°; 2 spec. East of Iceland: Stat. 103: Lat. 66°23' N., Long. 8°52' W-, 579 fath., temp. -^cr6°; 12 spec. Stat. 102: Lat. 66°23' N., Long. io°26' W., 750 fath., temp, 4-0-9°; 3 spec. North of Iceland: Stat. 126: Lat. 67°i9' N., Long. i5°52f W., 293 fath., temp. -=-0-5°; 5 spec. Stat. 124: Lat. 67°4o' N., Long. i5°4o' W., 495 fath., temp. ^o-6°; i'/2 spec. North-East of Iceland: Stat. 120: Lat. 6702g' N., Long. n°32' W., 885 fath., temp. 4- ro°; 8 spec. Stat. 119: Lat. 67°53' N., Long. ioci9' W.. 1010 fath., temp. — ro°; 2 spec. South of Jan Mayen: Stat. 117: Lat. 69°i3' N, Long. 8°23' W., 1003 fath., temp. -~- ix>°; ab. 18 spec. — - — Stat. 116: Lat. 70°05' N., Long. 8°26' W., 371 fath., temp. 4-0-4°; 10 spec. Besides, this species has been taken by the "Thor" at two places in the warm area. South-West of the Faeroes: Lat. 6i°i7' N., Long. 9°3o' W., 443 fath.; 5 spec. Lat. 6i°i5' N., Long. 9°35' W., 463— 515 fath.; ab. 15 spec. Distribution. E. inermis has not been separated from E. cornuta by Sars, and I am sure that his animals from at least several of the nine stations in the cold area enumerated by him in 1886 as belonging to E. cornuta in reality are E. inermis; some of his specimens from the deepest stations may belong to E. Hanseni. It is probable that the specimens from the Kara Sea referred by Stux- berg to E. cornuta belong to E. inermis; whether some or all of the above-mentioned specimens referred by Grieg to E. cornuta belong to this species or to E. inermis cannot be guessed. E. inermis certainly lives at the bottom, as, in spite of its considerable size, it has not been taken by the "Thor" in the young-fish trawl when this instrument did not touch the bottom. The list shows it to have been found from 293 to 1010 fathoms and to be widely distributed both in the warm and the cold area. 93. Eurycope Hanseni Ohliu. (PL XIII, figs. 3 a— 3 e). 1901. Eurycope Hanseni Ohlin, Bihang till K. Sv. Vet. Akad. Handl. Vol. 26, IV, No. 12, p. 34, figs. 7 a — 7!". Description. Similar and closely allied to B. inermis in size, general appearance, and most features. — The front area (fig. 3 a) between the basal joints of the antennulss is, seen somewhat from the side, horizontal to the end as in B. cornuta, but longer than in this species, consequently very CRUSTACEA MALACOSTRACA. III. H5 different from that in /•'. inermis; the anterior end is emargiuate and protrudes freely and considerably above the base of the elypeus. The basal joint of the anteuuuke (fig. 3 b) differs much in shape from that in E. cornuta or E. inermis\ the inner margin is even slightly longer than the breadth of the joint, as the anterior inner part is produced into a large, triangular lobe which is a little longer than broad and twice as long as in the last-named species; second and third joints mainly as in A. inermis. — The antennal squama (fig. 3 cl is longer and more narrow than in /:'. inn mis, being some- what more than half as long as the breadth of third joint. The thoracic epimera, the uropods (fig. 3 e), the female operculum and the lateral plates of the male operculum do not afford any valuable specific difference from E. inermis. Judging from my single male the median lamella of its operculum has the terminal lobes shorter than broad (fig. 3 d!, thus considerably shorter than in E. inermis. Length of the largest female, without marsupium (from Stat 113), 10 mm., of the single male 67 mm. Remarks. E. Hanseni is easily distinguished from E. inermis l>v the shape of the anterior part of the front area, from both E. cornuta and E. inermis by the very long lobe from first antennular joint, from E. cornuta besides by the widely different shape of the abdominal operculum in both sexes. Occurrence. Taken by the "Ingolf" at four stations of the deeper part of the cold area. East of Iceland: Stat. 105: Lat. 65°34' N., Long. 7°3i' W., 762 fath., temp. -f-o-8°; 3 spec, (small). vStat. 102: Lat. 66°23' N., Long. io°26' W., 750 fath.; temp. -4-0-9°; r spec. South of Jan Mayen: Stat. 118: Lat. 68°27' N., Long. 8°2o' W., 1060 fath., temp. -4- ro°; 1 spec. — • - — Stat. 113: Lat. 69°3i' N., Long. 7°o6' W., 1309 fath., temp. -4- ro°; 4 spec. (large). Distribution. Ohlin recorded it from a place nearly midway between East Greenland and southern Spitzbergen, viz. at Lat. 77°52' N., Long. 3°5' W., 1460 fath., temp. -:- 1-4°, and from west of Horn Sound, West Spitzbergen, Lat. 76°36' N., Long. 12° 10' E., 929 fath. -- Certainly some of the deep- sea specimens of Sars' specimens of E. cornuta from the Norw. North-Atlantic Exped. (comp. p. 1 \j, belong to E. Hanseni, which evidently is a cold water form inhabiting from considerable to very great depths. 94. Eurycope complanata Bonnier. (PI. XIII, figs. 4 a— 4 e). 1896. Eurycope complanata Bonnier, Campagne du "Candan": Ann. de l'Univ. Lyon, Vol. XXVI, p. 6 PI. XXXIV, figs. ia-ir. The species has been elaborately described and on the whole sufficiently figured by Bonnier; some features may yet be mentioned and a few figures added. The front area (fig. 4 a and 4 b) differs from that in E. cornuta in being distinctly broader witli the end very deeply incised; the incision is rounded at the bottom and the processes limiting it are about as long as broad, triangular, very acute. -- - The first joint of the antenuukc at least frequently broader than long, and it differs from that in E. cornuta especially in having the inner margin distinctly convex (fig. 4 a) and the inner distal lobe not triangular but obtuse and somewhat broadly rounded, The Ingolf-Expedition. III. 5. 1 9 I46 CRUSTACEA MALACOSTRACA. III. conspicuously broader than long. — The antennal squama (fig. 4 a) oblong-triangular, large or moderately large, conspicuously or even considerably more than half as long as the diameter of third joint. — The maxillipeds have the second joint uncommonly narrow (figured by Bonnier) towards the base, while the outer corner of the epipod is very broadly rounded, and the terminal corner acute, as shown by Bonnier. The abdominal operculum in the female (fig. 4e) somewhat reminds one of that in E. inermis\ its greatest breadth is somewhat before its middle, the most lateral part broadly rounded and setiferous, and from these seta; the lateral margins converge strongly backwards to the median line, but the distal part of each margin is distinctly concave, with the result that the terminal part of the operculum is somewhat produced, and the end is narrowly rounded; the lower surface of the operculum is vaulted and has a median, obtuse keel, which is low at the base and gradually higher backwards to near the end of the plate. — In the male operculum the median lamella (fig. 4 c) is rather similar to that in /;'. cornuta, but the terminal, produced, transverse part of each pleopod is shorter, without any real outer tooth. But the lateral plates are very characteristic (fig. 4d); each plate is considerably narrower than in E. cornuta, with the distal part of the inner margin deeply concave, with the result that the produced distal part of the plate is very narrow, much narrower than in any other species; second joint of the copulatory organ is not much shorter than the plate, and the part from its base to the vesicle is uncommonly long, only a little shorter than the portion containing vesicle and duct, and the distal part is unusually thick and terminates in a tiny spine. — The uropods, which have been well figured by Bonnier, are peculiar, as the peduncle is transverse, being widened obliquely towards the median line of the animal; the endopod projects from the middle of the posterior margin, and the exopod, which is much shorter and thinner, at the outer end of that margin. Length of the largest specimen, a female without marsupium (from Stat. 24), is 6 mm.; the male (from Stat. 36) is only 3 mm., but Bonnier had a male 5 mm. long. Remarks. Especially Bounier's figures of the uropods and of the lateral plates of the male operculum make the determination of my specimens sure. Occurrence. Taken by the "Ingolf" at two deep stations in the warm area. Davis Strait: Stat. 24: Lat. 63°o6' N., Long. 56°oo' W., 1199 fath., temp. 2-4°; 4 spec, (and 4 frag- ments). — Stat. 36: Lat. 6i°5o' N., Long. 56°2i' W., 1435 fath., temp. 1-5°; 2 spec. Distribution. Hitherto known only from the Bay of Biscay: Lat 44°I7' N., Long. 4°38' YV., 504 fath. (Bonnier). 95. Eurycope brevirostris n. sp. (PI. XIII, figs. 5 a-5 i). Description. Somewhat more than twice as long as broad, and in general aspect (fig. 5 a) some- what similar to E. con/it/n, but very much smaller, and the middle portion of the body, especially fifth thoracic segment, is broader as compared with head or abdomen than in E. conntta, inermis, ffansem, or complanata. - The epimera considerably produced, with a minute spine on the subacute end. CRUSTACEA MALACOSTRACA. III. 147 The front area (figs. 5 b and 5c) is well developed but somewhat short; its end is flatly rounded or straight, but at each angle a spiniform acute tooth projects forwards. — First joint of the autennulse (figs. 5 b) in the main as in E. complanata, a little broader than long, with the inner margin somewhat convex, the inner distal lobe broad and shorter than broad; second joint inserted somewhat behind the concave front margin of the first, middle-sized, and shorter than third joint. - Antenna] squama (fig. 5 d, ex) small, only about one-third as long as the diameter of third antennal joint; its end is cut off and bears a spine. — Maxillipeds (fig. 5 e) differ from those in E. inermis or /:'. cornuta in Inning the epipod broader with the distal subtriangular part less produced. Abdominal operculum of the female (fig. 5 h) somewhat similar to that in E. cornuta in the feature that the median keel terminates considerably before both margins, but this highly vaulted operculum has the carinate part of the lower margin, when seen from the side, nearly straight; seen from below the operculum differs considerably from that of E. cornuta, as it is only somewhat broader than long, broadest far before the middle, and from the narrowly rounded lateral angles the margins converge backwards as partly moderately convex lines to the rather broadly rounded end. - In the male operculum the median lamella (fig. 5 f) differs mainly from that in E. cornuta in having the pro- duced terminal part shorter and broader, without any outer tooth; the lateral plates (fig. 5 g) are about half as long again as broad, thus conspicuously longer than in E. cornuta, and somewhat triangular in outline, with the inner basal angle and the outer angle very broadly rounded, while its distal part is shaped nearly as in E. cornuta, and the distal part of the inner margin crossing the coupling hook is a little concave; second joint of the copulatory organ is a little shorter than the plate, with the pro- portion between the distance to the base of the vesicle and the distance between this point and the end nearly as in E. cornuta, but the distal fourth of the organ is much thicker than in this species. — Uropods (fig. 51) short and thick; the peduncle without any lateral expansion; the endopod at most a little longer than the peduncle, not three times as long as thick, nearly twice as thick as, but only a little longer than, the exopod. Length of the largest female with marsupium (from Stat. 138) 27 mm., of the largest male only 2'i mm. Remarks. E. brevirostris is easily separated from the preceding forms by a number of features; especially the outline of the small body, the female operculum and the uropods afford good characters. Occurrence. Taken by the "Ingolf" at two stations in the cold and one station in the warm area. South-West of Iceland: Stat. 78: Lat. 6o°37' N., Long. 27°52' W., 799 fath., temp. 4-5°; 2 spec. (poor). North-West of the Faroes: Stat. 138: Lat. 63°26' N., Long. 7°56' W., 471 fath., temp. -f- ev6°; 18 spec. East of Iceland: Stat. 103: Lat. 66°23' N., Long. 8°52' W., 579 fath., temp, -f- o-6°; 1 spec. 96. Eurycope producta G. O. Sars. (PL XIII, fig. 6 a). r866. Eurycope producta G. O. Sars, Nyt Magazin for Naturvid. Vol. 15, I, p. 113. (The whole volume dated 1868). ! 1898. G. O. Sars, Account, II, p. 147; PL 65. i9* 148 CRUSTACEA MALACOSTRACA. 111. This species, which is instantly distinguished from all other forms of this section (comp. p. 137) by the extremely conspicuous, long and rather broad, linguiform front area slightly bilobed at the end, has been well described and figured by Sars. Only some remarks may be made. As to outline E. producta differs from E. brevirostris in having the head and the abdomen broader in proportion to fifth thoracic segment than in that species, and thus is more similar to E. cormtta. Sars said that the front area, the "linguiform projection", is "narrow", but this term is misleading, as the "projection", according to his figures and in co-types presented by him, is long and somewhat broad to the laterally rounded, at the middle emarginate end; furthermore it is longitudinally very concave and, as stated by Sars, the most anterior part of its outer margins is conspicuously denticulate. — The epipod of the maxillipeds varies considerably in shape; sometimes the distal half of its outer margin is, as figured by Sars, scarcely or slightly concave, sometimes distinctly more concave, as the outer angle is a little produced, and in a specimen from the "Ingolf" Stat. 78 the shape shown in fig. 6 a was found, where the margin in question is rather concave, the outer corner being considerably produced. E. producta is allied to E. brevirostris in most features excepting the front area. The antennal squama is scarcely or at most a little larger than in E. brevirostris. The female operculum is carinate as in this species, but yet, as figured by Sars, somewhat broader and posteriorly more broadly rounded. The male operculum differs mainly in having the most distal part of each pleopod of first pair pro- duced and triangular (figured by Sars). Uropods with the exopod very much thinner but slightly shorter than the endopod. Occurrence. Taken by the "Ingolf" at three stations, two in the warm and the third in the cold area. Davis Strait: Stat. 25: Lat. 63°3o' N., Long. 54°25' W., 582 fath., temp. 7,-t,0; 2 spec. South-West of Iceland: Stat. 78: Lat. 6o°37' N., Long. 27°52 W., 799 fath., temp. 4-5°; 3 spec. North-West of the Faeroes: Stat. 138: Lat. 63°26' N., Long. f$6' W., 471 fath., temp. -^o-6°; 1 .spec. Furthermore it has been taken by Admiral Wandel in Davis Strait: Lat. f>5°49' N., Long. 56°28' W., 235 fath., temp. 4-4°, 1 spec; further south it has been taken by Cand. mag. Stephensen at West ( Greenland in Brede Fjord, ab. Lat. 6i° N., 164 fath., 2 spec. The Ryder Exped. secured it off East Greenland: Lat. 69°28' N., Long. 20V W., 167 fath., 1 spec. Finally it has been taken twice by the "Thor" at the following places: .South of Iceland: Lat. 62°io' N., Long. I9°36' W., 1010— 1142 fath.; 2 spec. South-West of the Faeroes: Lat. 6i°i5' N., Long. 9°35' W., 463—515 fath.; 4 spec. Distribution. Taken at several places along the coast, of Norway from Christiania Fjord to Yadso, in 60 to 400 fath. (G. O. .Sars). Later recorded from a locality west of Ireland, 382 fath. (Tattersall). 97. Eurycope megalura ('.. < >. Sars. 1872. Eurycope iitcgalunt G. O. Sars, Forh. Vid. Selsk. Christiania for 1871, p. 27). ! 1898. G. O. Sars, Account, II, p. 151 ; PI. 69. A couple of damaged specimens are referred to this species, as they agree witli Sars' figures in the size of the posterior thoracic segments, in the shape of the maxillipeds and the uropods, while CRUSTACEA MALACOSTRACA. III. r.49 the abdomen is shaped as in Sars' figure, but seems to be still a little broader and longer in propor- tion to the three posterior thoracic segments. The front cephalic area seems to differ from the figure of Sars in having no angles at the sides of the terminal emargination, but according to experience in other forms no stress can be laid on this difference. Occurrence. Taken by the "Ingolf" at a single station in the warm area. South-West of Iceland: Stat. 78: Lat 6o°37' N., Long. 27°52' W., 799 fath., temp. 4-5°; 2 spec. Distribution. Hitherto known from Stavanger Fjord and Hardanger Fjord, South-West Norway, 150 to 200 fath. (G. O. Sars) and from a locality west of Ireland, 191) fath. (Tattersall). 98. Eurycope parva Bonnier. (PI. XIII, figs. 7 a— 7 e). 1S96. Eurycope parva Bonnier, Campagne du "Caudan", Ann. l'Univ. Lyon, Vol. XXVI, p. 600; PL XXXIII, figs. 4 a— 4 e. The male of this fine species has been figured and described by Bonnier, but the female is unknown. The following description is based on the female, but excepting the operculum it may hold good for both sexes. Description. Body uncommonly oblong, from slightly less to considerably more than two and a half times as long as broad. — The front cephalic area is extremely large, only a little longer than broad, as its lateral margins converge feebly forwards (comp. fig. 4 a of Bonnier), while the long- anterior margin is a little convex, and the surface is feebly arched without keels at the lateral margins. — The first joint of the antennulse (figs. 7 a and 7 b) is a little longer than broad, with the lateral margins somewhat convex, the outer margin longer than the inner, and the front margin a little oblique; second joint originates somewhat before the front margin of the first, is longer than thick and longer than third joint. — Antennae with third and fourth peduncular joints (fig. 7 a) somewhat slender and subcylindrical; fourth joint with the outer margin at least as long as the diameter of the joint; squama very small, about one-third as long as the diameter of the joint, rather slender, sub- conical, with the obtuse end terminating in a small seta. -- Maxillipeds (fig. 7 c) moderately slender; fifth joint produced forwards and inwards, so that the distal inner angle is acute, and the inner margin straight excepting towards the base; sixth joint with the lobe short; epipod almost three times as long as broad, with the end obtuse, and the outer margin conspicuously concave somewhat beyond the middle, but without any trace of an outer corner. The four anterior thoracic segments in the median line together extremely short (figs. 4 a and 4b in Bonnier) as compared with the sum of the three posterior segments; the epimera at first seg- ment are considerably produced, while those at the three other segments are mainly or totally invisible from above, as they are overlapped by the rather produced corners of the segments. Fifth segment, is at the lateral margins as long as the sum of the two posterior segments, and its anterior margin is extremely convex; this segment is dorsally fused with sixth segment, so that a suture between them is seen only near the lateral margins. Abdomen somewhat small, not much broader than long. The female operculum (figs. ;d and !5Q CRUSTACEA MALACOSTRACA. III. 7 e) is extremely vaulted and the central part of its median line strongly keeled; seen from the side (fig. 7 d) the operculum is raised so much that it is shaped like a broad triangle, with the angle on the lower margin rounded and situated a little behind the middle; seen from below (fig. 7 e) the operculum is considerably broader than long, with the broadest part rather near the base, the anterior margin with each lateral third considerably concave, the 'lateral angles broadly rounded, and the margins behind these angles a little convex and strongly converging to the posterior margin, which is trans- verse, with the median part somewhat convex. — As to the male operculum the reader is referred to Bonnier. — Uropods with the peduncle rather small and short, considerably or much shorter than the endopod, while the slender exopod is scarcely one-third as long as the endopod (fig. 4 d in Bonnier). Length of an ovigerous female 27 mm.; the male according to Bonnier 3 mm. Remarks. E. parva differs strongly in general shape and in several characters from the other forms in this section, in which fifth and sixth thoracic segments are dorsally fused; the female oper- culum is very peculiar. Occurrence. Taken by the "Ingolf" at a deep station in the warm area. Davis Strait: Stat. 36: Lat. 6i°5o' N., Long. 56°2i' W., 1435 fath., temp. 1-5°; 2 spec. Besides, it has been taken by the "Thor" at the following place. South-West of the Faeroes: Lat. 6i°i5' N.. Long. o,°35' W., 463—515 fath.; 1 spec. ($ ovig.). Distribution. Known from the Bay of Biscay: Lat. 44°i7' N., Long. 4°38' W., 504 fath. (Bonnier). 99. Eurycope phallangium G. O. Sars. (PL XIII, fig. 8 a). 1864. Eurycope Phalangium G. O. Sars, Forh. Vid. Selsk. Christiania for 1863, p. 210. ! 1898. phallangium G. O. Sars, Account, II, p. 147; PI. 66. This species has been well represented by Sars as to most particulars, so that a few notes are sufficient. -- Sars' figure of the front area of the head, his "frontal projection", agrees badly with that found in co-types presented by him, while this area in his co-types is similar to those in my "Ingolf" specimens. The area (fig. 8 a) tapers anteriorly less than figured by Sars; its lateral margins are raised like keels, each of which projects as a minute, triangular tooth at the end, while the anterior margin of the area between these teeth is transverse, straight, moderately short and generally showing two or three tiny teeth. — The antennal squama (fig. 8 a) is small, much less than half as long as the diameter of third joint, slender, and tapering feebly to the truncate end, which has a single seta. Occurrence. Taken by the "Ingolf" at two stations. Davis Strita: Stat. 32: Lat. 66°35' N., Long. 56°38' W., 318 fath., temp. 3-9°; 1 spec. Stat. 25: Lat. 63°3o' N., Long. 54°25' W., 582 fath., temp. 3-3°; 6 spec. Distribution. Common at the south and west coasts of Norway, going northwards to Hasvig in Finniark, occurring in depths from 50 to 300 fath. (G. O. Sars, 1898), in Christiania Fjord taken 111 40—60 and even in 30 fath. (Sars 1864). A specimen was taken in Skager Rak, 15 sea-miles from the lighthouse of the Skaw, 125 fath. (H. J. Hansen). Finally gathered on the- "Porcupine Bank", west of Ireland, 293 fath. (Tattersall), and recorded with a query from two places respectively east and wist at Southern Scotland (T. Scott). CRUSTACEA MALACOSTRACA. III. ,Si ioo. Eurycope furcata G. O. Sars. (PI. XIII, figs. 9a— 9b). 1870. Eurycope furcata G. O. Sars, Forh. Vid. Selsk. Christiania for 1869, p. E65. ! 1898. G. O. Sars, Account, II, p. 148; PI. 67, fig. 2. To the elaborate representation published by Sars only a couple of remarks may be made. - Fig. 9 b exhibits the outline of the front cephalic area in a co-type presented by Sars; it shows the area broader in proportion to length with the end more broadly incised than in Sars' fig. 2; the spec- imens presented by him agree with each other in the shape of the area. Fig. 9 a, which represents the head of an "Ingolf" specimen (from Stat. 78), shows a somewhat different shape of the area; it has the lateral margins straight and converging to near the end. which is broadly emarginate with the lateral corners rounded. — The antennal squama (fig. 9a) is very peculiar in this species; it is moder- ately slender, long, a little to much more than half as long as the diameter of the third joint, and it reaches considerably beyond the end of fourth joint, and terminates in a couple of setae on the obtuse or truncate end. — It may be emphasized that, excepting the difference mentioned in the shape of the front cephalic area, my specimens agree well with the figures drawn by Sars. Occurrence. Taken by the "Ingolf at two somewhat deep stations in the warm area. Davis Strait: Stat. 24: Lat. 63°o6' N., Long. 56°oo' W., 1199 fath., temp. 2-4°; 1 spec. South-West of Iceland: Stat. 78: Lat. 6o°37' N., Long. 27°52' W., 799 fath., temp. 4-5°; 2 spec. Distribution. Sars has recorded E. furcata from off Lofoten, 100 — 120 fath., and from Sta- vauger Fjord. — I have not found it mentioned by any other author. 101. Eurycope mutica G. O. Sars. (PI. XIII, figs. 10 a — 10 c). 1864. Eurycope mutica G. O. Sars, Forh. Vid. Selsk. Christiania for 1863, p. 210. ! 1898. — G. O. Sars, Account, II, p. 149; PI. 68, fig. 1. To the representation of Sars three new figures are added, and a few remarks must be made. The antennal squama (fig. 10 a, ex) is very characteristic; it is shaped as a small tubercle scarcely as long as broad, but distinctly marked off, and it terminates in a very conspicuous, thick, generally straight, stick-shaped seta terminating in a hair. — The abdominal operculum in the male has been incorrectly figured by Sars; the more distal part of both the median lamella and the lateral plates are curved upwards — seen from below (fig. 10 b) downwards — but the lateral plates reach scarcely to the end of the lamella, while in Sars' figure they reach considerably beyond the lamella; furthermore each pleopod of the lamella terminates in two lobes (fig. 10 c), the inner broader and obliquely rounded, the outer somewhat or very narrow. — As to other features E. mutica, which is a very char- acteristic species, has been well figured and described by Sars. Occurrence. Taken by the "Ingolf" at a single station. West of Iceland: Stat. 87: Lat. 65°02-3' N., Long. 23°56-2' W., no fath.; 1 spec. In the Waigat Strait, ab. Lat. 700 X., at West Greenland Mag. Traustedt secured a single spec- 1.52 CRUSTACEA MALACOSTRACA III. imen; Mag. R. Horriug gathered 5 spec, in Faskruds Fjord, East Iceland, 50— 20 fath., and Dr. Th. Mor- tensen captured 3 spec, off Nolsd, the Fseroes, in ab. 100 fath. Distribution. According to Sars B. mutica has been taken in the inner part of the Christiania Fjord and at many places along the coast of Norway, as far north as Bodd (about Lat. 6772° N.). Sars wrote: "It is not strictly a deep-water species, being often found in only a few fathoms' depth among algae." At Denmark it has been taken in the northern part of the Sound, 14 fath.. in the southern Kattegat, 15 fath., and in Skager Rak off the Skaw, 70 fath. (H. J. Hansen). Finally recorded from the North Sea off Northumberland, 59 fath. (Norman) and from the Irish Sea near Isle of Man, 40 fath. (Tattersall). — The species is so small that it has probably been overlooked very frequently. Group XII. Munnopsini. Body rather oblong; anterior section of the thoracic segments, and especially third and fourth segments, somewhat or most frequently very much broader than the posterior segments and abdomen. — Head free; eyes wanting. Autennuke dorsal with first joint plate-shaped; the flagellum with many joints. Antennas some or several times longer than the body, as the two distal joints of the peduncle are extremely elongate; squama somewhat small, more or less distinct. Mandibles somewhat slender, with the distal parts well developed or much reduced, the molar process either slender conical or wanting; palp well developed or wanting. Maxillipeds normal; epipod rather oblong. — Thoracic seg- ments divided into two sections; the four anterior segments movable, the three posterior either mov- able or fused with each other and with the abdomen. First pair of legs very slender or moderately strong, more or less prehensile, with second joint long; second pair differ from first pair only in being thicker, much longer and with seventh joint long. Third and fourth pairs differ extremely from second pair in having second and third joints short, and the enormous length of these legs is due to fifth and sixth joints, which are extremely elongate. The three posterior pairs natatory, fifth and sixth joints a little or much expanded with marginal setse; seventh joint wanting. — Uropods terminal or a little ventral, uniramous, slender. Remarks. This group differs from the Eurycopini in having uniramous uropods, no seventh joint on the three pairs of natatory legs, and above all in second to fourth pairs of legs, as the second pair are similar to the first and very different from third pair, which have not only the third but above all the second joint short. Our knowledge of most of the forms of this group has hitherto been very poor, and it may be useful to give a brief review. The only fully known species is Munnopsis typica M. Sars. Munnop- v/.v Murrayi Walk. ( Munneurycope Tjalfiensis Steph.) has been wrongly referred by Walker and Tattersall to Munnopsis, but it is, as shown above, a species of linrycope. In 1895 I described Mun- nopsis longicornis H.J. II. on a single mutilated male captured by the German Plankton-Expedition. Stephensen (1913) quoted the following statement from Walker: "As G. O. Sars has pointed out, Mkii- i/opsis longicornis Hansen differs in the structure of the mandibles from the generic description, as CRUSTACEA MALACOSTRACA. III. x53 does the present species." [M. Murrayi); but this quotation shows that at least neithei Walker nor Stephensen had read my text, as I wrote (1895) on M. longicornis: "in detn Bau der Schwimmfusse und Uropoden (so weit sie vorhanden sind) stimmt es mehr mit Munnopsis typica M. Sars als mit Eurycopc gigantea G. O. Sars ubereiii, aber es weicht in dem Bau der Mandibeln griindlich von beiden ab". These words may prove sufficiently that it was very easy for Sars to point out that difference. M longicornis H. J. H. is allied to M. oceanica Tatt. and M. spinifer Vanhoffen (1914). Of the remaining species referred to Munnopsis, M. australis Redd, will be mentioned presently; .1/. latifrom Bedd. and M. longiremis Richardson are closely allied but, like M. gracilis Bedd., too imperfectly known for discus- sion on affinities. The material from our area comprises four species, viz. Munnopsis typica M. Sars, M. oceanica Tatt., a new species closely allied to M. australis Bedd., and the form described by Tattersall as Mun- nopsoides Beddardi Tatt. But a study of these forms showed that it was necessary to divide the genus Munnopsis into three genera. M. typica has the anterior thoracic segments and especially third segment much broader than the rather narrow three posterior segments, and these three are immovably coalesced with each other and with the abdomen; the mandibles have no molar process, well developed palp, and on the left mandible only rudiments of seta; near the lacinia; the natatory legs have the two distal joints moderately broad; in the male the second pair of pleopods are free to the base with their copulatory organ extremely elongate, and the uropods are moderately long; this species must of course remain as the type for the genus Munnopsis M. Sars. But M. oceanica Tatt. must be removed; it agrees with M. typica in having well developed mandibular palps, but the mandibles have a number of well developed seta; before the incisive dentate part and a slender conical molar process; the difference in breadth between the anterior and the posterior thoracic segments is moderate; the three posterior seg- ments and the abdomen are movable, the fifth joint of the natatory legs is much broader than the sixth; the male pleopods of second pair have their proximal part completely fused in the median line, and the copulatory organs moderately elongate; the uropods are long. For the reception of M. oceanica Tatt. the generic name Paramunnopsis is proposed. Munnopsis australis Beddard has by Tattersall been taken as type for his genus Munnopsoides, to which he also referred his new species M. Beddardi Tatt. .Supposing that M. australis in the main features agrees with my M. eximius n. sp. (described later on) these three species show the following features. They agree with Munnopsis typica and differ from Paramunnopsis n. gen. in having no molar process and the setse rudimentary on the mandibles, and, besides, in the fact that the three posterior thoracic segments and the abdomen are immovably fused ; they differ from both in some features, viz. the mandibles have 110 palp, the anterior thoracic segments are very broad and the posterior very narrow, the fifth segment even extremely narrow, the natatory legs have the two distal joints very narrow, and the uropods are very short. But in one most interesting feature the two species M. Beddardi Tatt. and M. eximius n. sp. are very different. In M. eximius and therefore, according to the above-named supposition, in M. australis Bedd., the male pleopods of second pair are quite independent, and the free part of their copulatory organ more than twice as long as the pleopod (PL XIV, fig. 2 111; in the male of M. Beddardi Tatt. the same pleopods are fused to such a degree that they constitute a single large plate (PL XIV, fig. 3 k) with a posterior oblong incision, in which are found the two copulatory organs, The [ngolf-Expedition. III. >, 20 J54 CRUSTACEA MALACOSTRACA. III. wliich are small and quite short. This structure, which has been mentioned on p. 10 as being of great morphological significance, I am not prepared to consider as being only of specific value. It seems to me, that this feature makes it necessary to establish a new genus; as Tattersall has taken Munnopsis australis Bedd. as the type for his genus Miinnopsoidcs, I keep this generic name, and, according to the above-named supjiosition, refer my new species, eximius, to it, but the result is that I must estab- lish a new genus for the reception of Munnopsoides Beddardi Tatt, and the name Pseudomunnopsis n. gen. is proposed. Paramunnopsis n. gen. Description. Body rather oblong; third and fourth thoracic segments somewhat broader than the fifth segment, which is conspicuously broader than the seventh. — First antennular joint is an oblong plate without any protruding distal angle. Antennae extremely long; the squama (PI. XIII, fig. nb, ex) somewhat small but protruding. Mandibles (figs, iic-ne) with the incisive part produ- ced and its edge divided into some triangular teeth, the movable lacinia on left mandible somewhat long with teeth on its terminal margin, the seta; numerous and well developed, the molar process slender and tapering to the acute or subacute end; palp long and robust. Maxillipeds (fig. nf) nearly as in Munnopsis. The three posterior thoracic segments and abdomen movable; fifth segment produced much forwards, overlapping a more or less considerable part of the fourth (PI. XIV, fig. i a). — First pair of legs moderately slender, prehensile (second pair in P. spinifer Vanh. considerably longer and thinner than first pair, in the main as in Munnopsis). Third and fourth pairs extremely long, nearly as in Munnopsis. The three posterior pairs differ from those in Munnopsis in having the fifth joint not only much broader than the sixth but differently shaped, as the anterior margin is nearly straight, tin- major part of the posterior margin extremely convex. The two anterior pairs of male pleopods in P. uccatiica (they are unknown in the two other species) show characters probably of generic value. First pair (fig. n g) increase much in breadth from somewhat from the base to beyond the middle, and then tapers to the rather narrow end, wliich has two pairs of small lobes. Second pair (fig. n i) have about their proximal third fused, so that no vestige of a suture is seen in the median line; first joint of the copulatorv organ long and directed forwards, while the second joint is produced in a long, nearly filiform part reaching considerably beyond the pleopods, and the vesicle is rather far removed from the base of the joint. — Uropods uniramous, two-jointed, long or extremely long. Remarks. As the type I take Munnopsis oceanica Tatt, but M. longicornis H.J.H. 11895) and .1/. spinifer Vanh. (1914) belong to the same genus. In the description of the genus features in all three species have been taken into account; excepting the male pleopods most characters can be seen in the text or in the descriptions of the two last-named species. — It may be observed that all three species are really pelagic animals, but only one is represented in our material. CRUSTACEA MALACOSTRACA. III. 155 102. Paramunnopsis oceanica Tattersall. (PL XIII, figs. 11 a— 11 i; PI. XIV, figs. 1 a— 1 b). ! 1905. Munnopsis oceanica Tattersall, Isopoda, p. 23 and 72; PI. V, figs. 1—7. 1911. Tattersall, Die nordischen Isopoden, in Nordisches Plankton, VI, p. 187, with figures. 1914. Vanhoffen, Deutsche Siidpolar-Exp. 1901 — 1903. Vol. XV, Zoo!. VII, p. 581, figs. 11 a — b (? and 11 c — d). Description. Body of a female with niarsupium about three times, of the males (fig. 1 a) three and half times as long as broad, without processes or teeth on the dorsal surface or on the lateral margins of thorax and abdomen. — Antennal peduncles nearly twice as long as the body. Molar process on left mandible (figs. 11 c and 11 e) slender, tapering from the base to the acute end, with five or six oblong acute teeth on the distal half of the posterior margin. - Maxillipeds (fig. 11 f) with fourth and fifth joints conspicuously broader than second joint, which tapers towards the base; sixth joint produced in a lobe, which is much longer than broad and much longer than the joint; distal part of the epipod somewhat produced, but the narrow end obtuse. Fifth joint of the natatory legs (Tattersall : Isopoda, fig. 6) twice or a little less than twice as long as broad. — Abdomen large, about as long as the sum of the three posterior thoracic segments, oblong-ovate. The median lamella of the male operculum (fig. 11 g) about three times as long as broad; the inner pair of terminal lobes (fig. 11 h) much longer but much narrower than the outer pair. Se- cond pair of pleopods described in the diagnosis of the genus. - - Female operculum (PI. XIV, fig. 1 b) conspicuously broader than long, much shorter than the abdomen, somewhat convex but not carinate, and posteriorly emargiuate at the middle. — Uropods in the female as long as, in the male somewhat shorter than, the abdomen (fig. 1 a), and second joint nearly two and a half times as long as the first. Length of a male 6 mm. (Tattersall recorded 7 mm.), of a female with niarsupium 5 mm., of a female without niarsupium 6 mm. Remarks. In this description the differences between P.oceanica and the two other less well- known species are indirectly taken into account. The first thoracic leg was figured by Tattersall. Occurrence. Taken by the "Ingolf" at a station in the warm area. Davis Strait: Stat. 36: Lat. 6i°5o' N., Long. 56°2i' W., 1435 fath., temp. 15°; 1 spec, (mutilated o*)- Besides, it has been taken by the "Thor" in young-fish trawl far south of Iceland, at Lat. 6i°3o' N., Long. i7°o8' W., 1800 m. wire out, 3 spec. Distribution. Tattersall recorded it as taken twice west of Ireland, respectively in townet from 730 to o fath. and in young-fish trawl, 1150 fath. Vanhoffen recorded it from the South Atlantic, Lat. 350 10' S., Long. 2°33' E., vertical net from 3000 rn. to surface. A badly preserved specimen from the Sub-Antarctic Ocean: Lat. 6i°58' S., Long. 95V E., vertical net from 2000 m. to surface, referred by Vanhoffen to this species seems to me a little less certain, as he described and figured the molar process of its mandibles as being somewhat different from the usual shape. 156 CRUSTACEA MALACOSTRACA. III. Munnopsis M. Sars. Diagnosis. Body with the anterior section of thoracic segments and especially third segment much broader than the three posterior segments and abdomen.— First antennular joint with the distal inner part considerably protruding beyond the insertion of next joint as a broad triangle with the end obtuse. Antennae mainly as in Para munnopsis ; squama low. moderately distinct. Mandibles reduced; the incisive part terminates in a couple of obtuse teeth, the setae are few and quite short, the molar process wanting, the palp well developed. -- The three posterior thoracic segments immovably fused with each other and with abdomen; fifth segment not overlapping the fourth. Natatory legs with fifth joint about three times as long as broad, not broader than the sixth, and the posterior margin very moderately convex. — The median lamella of the male pleopods moderately broad, feebly broader beyond the middle and then tapering considerably to the end, which terminates in two pairs of small lobes; second pair of pleopods free to the base, and each copulatory organ has first joint rather short and thick, second joint produced in a filament of excessive length. — Uropods moderately long. Remarks. In this diagnosis some characters not found in earlier descriptions are pointed out mainly in order to separate Munnopsis from the other genera of the group. Of the species hitherto referred to Munnopsis only M. typica seems to remain in it. 103. Munnopsis typica M. Sars. 1861. Munnopsis typica M. Sars, Forh. Vid. Selsk. Christiauia for i860, p. 84. 1 JS68. — — M. Sars, Bidrag til Kundskab 0111 Christiauiafjordeus Fauna, in Nyt Magaz. for Naturv. Vol. XV, p. 310; Pis. VI— VII. 1887. — H. J. Hansen, Dijmphna-Togtets zool.-bot. Udbytte, p. 196; PI. XX, figs. 2— 2 e. ! 1897. — G. O. vSars, Account, II, p. 133; Pis. 57—58. 1905. — — Richardson, Monograph, p. 486; figs. 544—546. To the fine paper of M. Sars (1868) I added in 1887 (1886) some particulars, chiefly of morpho- logical significance, and these two treatises together with the representation of G. O. Sars and the diagnosis above of the genus give a full account of this interesting form. — As to size it may be said that according to G. O. Sars the length of the adult male is 11 mm.; H. Richardson recorded 13 mm., but Ohlin had from East Greenland a female 15-5 mm. and a male measuring 18 mm., while one of my largest specimens, a female without marsupium from the "Iugolf" Stat. 138, is 177 mm. long. Females from the Kara Sea I found infested with Spharonella Munnopsidis H.J.H. (1897), and on a couple of females taken by the Ilnd Amdrup-Fxped. at East Greenland I found in the marsupium a Sphmronella, probably the same species, but not yet investigated. Occurrence. Taken by the "Ingolf" at eight stations. Davis Strait: Stat. 32: Lai. 66°35' N., Long. 56°38' W., 318 fath., temp. 3-9°; ab. 20 spec. Stat. 31: Lat. 66°54' N., Long. 55°34' W., 88 fath., temp. 16°; 1 spec. Stat. 35: Lat. 65^16' N., Long. 55^5' W., 362 fath., temp. 3-6°; ab. 10 spec. Stat. 28: Lat. 65°i4' N., Long. 55°42' W., 420 fath., temp. 3-5°; 1 spec. Stat. 27: Lat. 64°54' N., Long. 55°io' YV., 30,3 fath., temp. 3-8°; 5 spec. CRUSTACEA MALACOSTRACA. III. 157 Between Iceland and the Faeroes: Stat. 4: Lat. 6y°oy' N., Long. iicI2' W., 237 fath.; temp. 2-5°; 1 spec. Stat. 2: Lat. 63°04' N., Long. o.°22' W., 262 fath.; temp. 5-3°; 2 spec. North of the Faeroes: Stat. 138: Lat. 63°26'N., Long. 7°56' W., 471 fath.. temp, -r- cv60; 12 spec. It has been recorded from two places at Grinnell Land, the most northern locality being at Lat. 79°44' N. (Miers), and was taken at several places off West Greenland and in the eastern part of the Baffin Bay between Lat. 78°i8' N. and Lat. 69°3i' N. (several authors); in all six places north of Lat. 770 N. the depths were from 15 — 20 to 50 fath., but in Baffin Bay they were from 100 to 199 fath. Furthermore it has been secured by the "Thor" south of Iceland: Lat. 63°i5' X., Long. 22°23' W., 119 — 172 fath., 1 spec; and by Admiral Wandel at four places, viz.: North of Iceland in the Skagestrand Bay, 119 fath., temp. 2-9°, 1 spec; East of Iceland: Lat. 63°i5' N., Long. 9°35' W., 270 fath., 6 spec; nearer to Iceland off Seydis Fjord, 135 fath., 2 spec; finally East of the Fseroes: Lat. 6i°23' N., Long. 5°04' W., 255 fath., temp. o°, 3 spec. Mag. R. Horring gathered it in Nord Fjord, East Iceland, 35 — 55 fath., 1 spec. At East Greenland M. typica has been taken by Danish and foreign expeditions at ten places between Lat. b