SOME WEST INDIAN ECHINOIDS. 19 



est doubt), one can not feel sure that the specimens examined by him are really 

 the same species— on the contrary, it is beyond dovibt that both hystrix and fene- 

 strafum are among them, perhaps also riolaceuin — and tlius ins statement loses its 

 weight. In fact, 1 find, after having examined a rather considerable number of these 

 forms (especially hystrix), that the shape of the plates and the extent of the mem- 

 branous interstices between the plates is not so very variable. Thus in C. hystrix 

 I never find the wide membranous spaces between the plates so characteristic of 

 A. fenestratum; A. belli in this respect comes near to A. fene stratum, though the 

 membranous spaces are not quite so conspicuous; the same holds good of 

 A. violaceum. Instead of giving long descriptions I think it is sufficient to refer 

 to the figures. Plates 11-13. 



The most conspicuous difTerence is found in the arrangement and size of the 

 primary tubercles of the actinal interambulacra. (Plate 11 , fig. 2 and Plates 12-13.) 

 In A. helli there are near the middle line two arched series of large tubercles, one 

 tubercle to each plate." In the other species there is certainly something cor- 

 responding to this, but the series are less curved, and there is generally a tubercle 

 only on every second plate. Further, in A. belli these tubercles are larger than in 

 the other species; in A. fenestratum they are not much smaller, but in A. violaceum 

 and in C. hystrix they are considerably smaller. In the larger specimens some large 

 tubercles appear between the marginal and the inner series, and here again the same 

 fact holds good: Larger and on each plate in A. belli, smaller and not on each plate 

 in the other species. 



In the abactinal side of the test, as well as in the apical and oral systems, I 

 do not find any reliable differences, in size or structure, between the four species. 

 In the two young specimens of A. belli the apical plates are still in contact, and the 

 genital openings have not yet appeared. 



The spines do not seem to afford any reliable differences either. In the speci- 

 mens of A. belli before me all the primary actinal spines are broken; in the specimen 

 of 25 mm. horizontal diameter only a single spine has kept the hoof; judging from 

 this the hoof is not especially large in this species, as is also the case in the other 

 species." The miliary spines, those on the actinal as well as those on tiie abacti- 

 nal side, have a small (poison) gland at the point. The tubefeet are as in the 



other species. 



The pedicellariaj afford the most conspicuous differences as shown m the 

 original description.'^ Tetradactylous pedicellari:e I have been unable to find in any 

 of The specimens of A. belli at hand ; that they will prove to occur also in this species 

 I can scarcely doubt. The tridentate pedicellariae, so very characteristic that the 

 species mav be at once distinguished thereby (especially the small form), have been 

 sufficiently described and figureil in the Inyolf Kchinoidea. (Pt. 1, p. 55, pi. 13, 

 figs. 10, 11, 22.) The large form (fig. 10) I have not found in any of the specimens 

 at hand. 'l have nothing to add concerning the triphyllous pediceUariae. The 



a There may, however, be some inconsistency in this character; eome of the plates may want the 

 large tubercle, but still in such case? this series of tubercles is generally more prominent and regular than 



in A. fenestratum. ,^..1.1.1 



b I may note that in none of the European specimens of A . Jmcstralum at my disposal has the hoof 



been preserved, all the primary actinal spines being broken, 

 c /(i(7o(/' Eehinoidea, i)t. 1, p. 55. 



