ECm.NUIDEA. II. 



the same as tlie /'.p/iia/r of \V\\'. Thomson from the Rockall Cliannel, thus loses its interest from a 

 zoogeographical point of view, since in an\- case this species really occnrs both in the Northern Atlan- 

 tic and in the Antarctic Sea. (Comp. Urecliiinis iiaresianus.) 



This species was taken 1)\ the -Ingolf at the following stations: 



St. II. (64° 34' La t. N. 31° 12' Long. W. 1300 fathoms i°6 C. Bottom temp.). 2 specimens. 

 — 40. (62° 00' — 21" 36' — 845 — i'l — — ) I - 



-83. (62° 25' - 28=30' — 912 - 3°5 - - ) I - 



The geographical distribution of the species is: Northern .Atlantic (S. of Iceland, Denmark 

 Strait! and .Antarctic Sea. It will doubtless be found to occur all over the .Vtlantic ( )cean. The bathy- 

 metrical range, as hitherto known, is 845 — 1975 fathoms. 



The ver\ interesting morphological relations of the bivium show that P. pliialc is really one 

 of the more primitive Pourtalesi;e, in spite of its modified form. The continuit\- of the ambulacra I 

 and V it has in common with Sternopatagits and Ponrfah:sin cariuata. which latter species through 

 its two pores in the plates I. a. i and V. b. i as well as b\- its large labrum, maintains the place as 

 the least modified of the Pourtalesia-s'^&cies, (\iz. among those species whose structure of the test is 

 thus far known)'. Otherwise important light is thrown on the structure of P. carinata by what has 

 here been made known of the structure of the actinal part of the test in P. p/iialr. A comparison of 

 the figure of the actinal side of P. pliialc (PI. VI. Fig. 7) with the PI. W. Fig. 42 of J^oven's On 

 Pourtalesias shows almost beyond doubt that the plates named by Loven 5. a. 2 b. 2 and \'. a. 2 b. 2 

 are wrongh" interpreted. The plate named \ . b. 2 is seen to agree very closely with the plate V. a. 2 

 in P. pliialc: but in case that plate is reall\- \'. a. 2, which can scarcely be doubted, the plate named 

 bv Loven 5. a. 2 really becomes the ambulacral plate V. b. 2. 

 To be sure, it is separated from the plate V. b. i, by the corner 

 of the labrum; but the connection between these two plates in 

 P. pliialc is already so ver>- narrow, that it is ver\- easily con- 

 ceivable how the total separation has been produced in P. cariuata 

 bv the great development of the labrum. The plate V. a. 2 in 

 Loven's Figure thus becomes a plate of interambulacrum 4. 

 I mav give here a copy of the figure from Loven with my 

 interpretation of the plates for the direct compari-son with P. pliialc 

 Figs. II and 12). I think it will be agreed that w\\ interpretation 

 thereof has all evidence of being the right one. But this leads 

 to the verv important conclusion that Poiirtalcsia caniiala is Fig. n. Part of 



5V Loven as the result of his, actinal plastron 



Fig. 12. Part of actinal 

 plastron of Pourlalesia 

 cai'inata after Loven. 



not amphisternous as thought b^ ^. , ^^ Pourlalesia 



evidently wrong, interpretation • of the plates in this figure, but /^/a/^-. 



I De Meijere (Siboga Echin. PI. XXI. 41S p. 16S1 represents Echinocrepis cuneata as having the same structure of 

 the bivial ambulacra, founding his opinion on PI. XXXV. a. lo of the - Challenger»-Ech. Echinocrepis seligera has its bivial 

 ambulacra separated bv the interambulacra i and 4 (Panamic Deep-Sea Echini. PI. 67. i, Fig. 167). Also the apical system is 

 very different in these'two species, compact in Eck. cuneata, disconnected in Ech. sctigera. It can then scarcely be doubted 

 that the latter species was unrightly referred to the genus Echinocrepis and will have to be made the tx-pe of a new genus. 

 (Comp. below p. 83— 84.1 



-' It is of course, the fragmentary- condition of his material of this species which has caused that interpretation. Not 

 knowing the real structure of P. phiale, Loven could scarcely interpret these plates in P. cariuata otherwise. 



