Classificatory Position of Hemiaster elong-atus. 241 



to the same genus, one would saj that the specific difference 

 is in the position of the madreporic plate. 



With regard to Hemiaster zonatus^ A. Ag., the notice in 

 the ' Challenger ' Echini gives no structural details. 



Hemiaster gihhosus^ A. Ag., agrees with the Sindian forms 

 in every generic detail, excej)t in the structure of the apical 

 disk. 



Of course the whole question of the value of such generic 

 forms as Abatus, Tripi/lus, &c., depends upon the greater or 

 less importance of the presumed tixity of the position of the 

 generative plates. We cannot agree to the hard-and-fast 

 lines regarding Hemiaster laid down bj Loven, because we 

 believe that tlie position, as well as the number and perforation 

 of the costals of Loven (the basals of one of us) are variable, and 

 more so than the details of the actinal structures. With 

 some other naturalists we consider all these forms to be 

 simple groups of Hemiaster, some being possibly of subgencric 

 value. 



It is somewhat singular — and we believe it shows our 

 appreciation of Loven's generalization, that the madreporite 

 tends to pass backwards in Tertiary and recent species of 

 Mesozoic genera — that we should have been impressed with 

 the position of the madreporite in the Sindian forms as indi- 

 cating a variation that should be found in Tertiary Ilemi- 

 asters. 



The researches of M. P. de Loriol in our opinion confirm 

 the propriety of separating Palceostoma and the Hemiasters 

 with only two generative pore-openings and an encroaching 

 madreporite. M, de Loriol published in the ' Pala^ontogra- 

 pliica ' (xxx.) a work entitled " Eociine Echinoideen aus 

 Aegypten und der Libyscheu Wiiste " (1881), and in it are two 

 most interesting forms which bear considerably on the subject 

 before us. In the Libysche Stufe there is a Palceostoma, and 

 in the same deposit there is a Hemiaster Schweinfurthi, both 

 species of de Loriol's. Paheostoma Zitteli (p. 33, pi. viii. 

 ligs. 1-1 d) is closely allied to P. mirabile ; the shape of the 

 test and that of the peristome are the same, but the pores of 

 the ovarian plates " sind nicht deutlich zu erkennen." The 

 peristome has all the characters of Pakeostoma, but none of 

 Hemiaster elongatus—^'' Peristom nahe am vorderen Randc in 

 gleicher Ebene mit der iSchale, genau funfeckig, mit beinahe 

 gleichen fcjeiten. Der Umriss ist duich eiue deutliche Leiste 

 eingesiiumt." 



This interesting fossil form in no way resembles our Hemi- 

 asters. Hemiaster Schiveiiifurthi is described on page 34 and 

 figured on the plate viii. figs. 3, 4, y. 



Ann. (& Mag. N. Hist. Ser. 5. Vol. xiv. 19 



