ECHINOIDEA. II. 119 



show that the tridentate pedicellarise are richer developed. - - The triphyllotis pedicellarise are as usual, 

 like small tridentate ones. The spicules are long, spinulose rods (PI. XIV. Fig. 27 a-b), in striking contrast 

 to the very small spicules of canaliferus ; they lie transversely to the longitudinal axis of the tubefeet, 

 indistinctly arranged in two or three series. The plates of the rosette of the frontal tube-feet are well 

 developed, reaching to the point of the lobes. 



Schizaster Edwardsi Cotteau is nearly related to canaliferus and orbignycmus. Professor Joubin 

 has with the greatest liberality, for which I cannot thank him enough, sent me one of the type- 

 specimens for examination; I am thus able to give some additional information of characters which 

 are not mentioned in Cotteau's diagnosis of the species. The shape of the test is upon the whole 

 like that of canaliferus; only the anterior ambnlacral furrow is a little broader, its sides being almost 

 perpendicular, whereas in canaliferus they bend somewhat over the furrow. The pores are arranged in 

 a single regular series the most prominent difference from canaliferus. The labrum does not reach 

 the second ambulacral plate of the adjoining series; there are 5 6 large subanal tubefeet, the first of 

 these being on the 5th ambulacral plate. The lateral fascicle passes over the I3th ambulacral plate. 

 Only two genital pores, as pointed out by Cotteau. Of the pedicellarise I can give but very little 

 information, having found only a single small tridentate pedicellaria with simple, leafshaped valves, 

 and another small form (PI. XIV. Fig. 10) which is probably a small rostrate pedicellaria. The spi- 

 cules and rosette-plates as in canalifems. -- Though insufficiently known this species is easily disting- 

 uished from canalifertts by its single series of pores in the odd anterior ambulacrum and from orbigny- 

 anus (the northern form) by its spicules. But it is not possible for the present to say, if it is not per- 

 haps identical with the Caribbean form of orbignyanus, which might, from a zoogeographical point of 

 view, not be improbable. Also it has a very great likeness to Sch. lacunosus, and it is impossible for 

 the present to give other distinguishing characters between these two species than their geographical 

 distribution: one in the Indo-Pacific Ocean, the other at the Coast of Guinea; (S. lacunosus also has 

 a single series of pores in the anterior ambulacrum and quite small spicules). Before the Caribbean 

 form of S. orbignyanus has been closely examined and the pedicellarise of S. Edwardsi have likewise 

 been made sufficiently known, it is impossible to judge of the specific value of these two forms and 

 their mutual relations. 



Professor Doderlein (Op. cit. p. 255) has pointed out that among the (recent) species referred 

 to the genus Schizaster two groups may be distinguished, differing markedly by their globiferous 

 pedicellarise: in one group (S. fragilis, capensis, antarcticus and ventricosus) the valves of the globi- 

 ferous pedicellarise end in a single long, sharp tooth, in the other (S. philippii, canaliferus and japonicus) 

 they end in 4 6 short teeth. Though the number of genital pores is not in accordance with this 

 grouping, as might have been expected, Professor Doderlein thinks that nach Untersuchung auch 

 der anderen Arten von Schizaster die Aufteilung dieser Gattung in mindestens zwei Gattungen nach 

 den Merkmalen der globiferen Pedicellarien zu erwarten sei(n). In Revision of Echini Agassiz 

 says of Sch. ventricosus that it is intermediate between the species of the group of the genus to 

 which S.fragilis and S. Philippii belong and that formed by S. canaliferus and S. gibberulus*. It follows 

 from this that also Agassiz is inclined to divide the species into two groups, but he does not work 



