8 2 ECHINOIDEA. II. 



If we take these four groups to represent genera, or at least subgenera, which seems not at 

 all iinreasonable, the latter group must keep the name Pourtalesia. Of the names proposed by Pom el 

 two become synonyms only of Pourtalesia, viz. Phyalopsis (for laguncula} and Phyale (for Jeffrcysi}. 

 Only the name Ceratophysa may be retained; P.rosea is named as the first species of this genus, but 

 the diagnosis is made from ceratopyga. The latter species must then be taken as the genotype. For 

 the two other groups I may propose the names: Hclgocystis and Echinosigra. 



The old gemis Pourtalesia is thus divided into four genera (or subgenera), viz.: 



Helgocystis n. g. with the species carinata (A. Ag.). 



Echinosigra n. g. with the species phiale (W. Th.) (genotype) and paradoxa (Mrtsn.). 

 Ceratophysa Pomel with the species ceratopyga (A. Ag.). 



Pourtalesia A. Ag. with the species miranda A. Ag. (genotype), laguncula A. Ag., Tanneri A. Ag., 

 Jcffreysi W. Th., Wandeli Mrtsn. and hispida A. Ag. 



Perhaps the species Jeffreysi, Wandeli and hispida may yet prove to form a separate genus, 

 which would then get the name Phyale Pomel.; for the present, however, it seems not necessary to 

 separate these species from the genus Pourtalesia, though it must be conceded that they form a dis- 

 tinct group in that genus, differing from the other species in the shape of the test. P. Tanneri, how- 

 ever, is in some way intermediate between the two groups (by its narrow anal snout). That it should 

 be necessary to make P. hispida the type of a separate genus there is no reasen to suppose. 



Spatagocystis Challcngeri A. Ag. has been very carefully worked out, especially in the Panamic 

 Deep-Sea Echini (p. 141), as regards the structure of the test Three kinds of pedicellariae have been 

 figured in the Challenger >-Report (PI. XLJI. 10 12 and XLV. 39 43), though - - as is mostly the 

 case in that work not mentioned in the text I have found (on specimens examined in the British 

 Museum) two kinds of pedicellarise, viz. tridentate and rostrate. Further I find in my preparation a 

 single globiferous and an ophicephalous pedicellaria resembling exactly those of Urechinus Wyvillii. 

 As the specimens examined proceed from St. 147 from which station also Urec/i. Wyvillii is recorded, 

 I think these pedicellariae do really belong to that species, having only accidentally got between those 

 of Spatagocystis. The tridentate pedicellariae are richly developed, occurring in at least two different 

 forms, viz. one with simply leafshaped, more or less slender valves with the apophysis continuing into 

 the edge of the blade (PL X. Fig. 20 represents a small specimen of the slender form; larger specimens 

 are rather similar to those of Echinocrepis cuncata}, and another with rather short, broad valves, nar- 

 rowed in the lower part of the blade and terminating in a more or less prominent hook (PI. X. Fig. 10); 

 this is evidently the form figured in the Challengers-Report PI. XLJI. 10 and PI. XLV. 39 40 as a 

 large-headed pedicellaria. I have not found so much mesh work in this as figured iu the PI. XLV. 

 40 of the Challenger ; there is often nothing at all. The form figured in PI. XLII. 12, evidently an- 

 other form of tridentate pedicellarise, I have not seen. The rostrate pedicellarise, figured as short- 

 headed, toothed, cup-pronged* pedicellarise (PI. XLIL n and XLV. 41 and 43), are of a quite typical 

 form, with the outer edge of the rather short and broad blade provided with ca. 10 16 thick teeth 

 (PI. X. Fig. 18); the edge of the basal part is generally closely serrate, though not always so regularly 

 as in the specimen here figured. The stalk is more or less thorny (PI. X. Fig. 35). - There is a very 



