8o ECHINOIDEA. II. 



that its apical system is like that of P. laguncitla. evidently its nearest relation. (Loven. On Pour- 

 talesia. PI. VII. Fig. 52). 



I have found two kinds of pedicellariae' in P. rosra. viz. ophicephalons and tridentatc. The 

 ophicephalons jjedicellariae (PI. XI. Fig. 26) are rather large, with elongated, slender valves. The ter- 

 minal widening is smaller and has fewer teeth than in P. ccratopyga. The tridentate pedicellariae (only 

 one form foimd) have simply leafsliaped valves ; the endtooth is a little prominent, the apophysis con- 

 tinues into the edges of the blade (PI. XL Fig. 15). I have noticed especially that the ophicephalons 

 pedicellariae were found on the fragment of the posterior end ( — about the tridentate pedicellariae I 

 have forgot to notice that especially, so they may perhaps belong to the other fragments — ); they 

 are sufficientl)' characteristic for distinguishing this species from any other of the .species hitherto 

 known of this genus — and, evidently, it is the species represented by the anal snont-fragment which 

 must keep the name Pourtalrsia rosea, not that represented by the fragment with the apical system, 

 wliich is probabh' no T'oiirfalrsia at all. The affinities of Pourtalesia rosea must, of course, Ije left 

 undecided, .so long as we know almost nothing of its shape and structure of test". 



Pourtalesia laguncula A. Ag. and Taniieri A. Ag. have been treated above (p. 67). 



The question whether all the species referred to the genus Pourtalesia can rightly remain to- 

 gether in this single genus has repeatedly been treated. In the « Challenger '-Report (p. 132I Professor 

 Agassiz comes to the result that all the species must remain in one genus, though the character of 

 the test seems to indicate two natural groups (/-". ceratopyga and rosea forming one group, the rest of 

 the species another); in his last great work «The Panamic Deep-Sea Echini:> he is inclined to think 

 that the striking differences found in the various groups of species of Pourtalcsiae would seem to 

 warrant the splitting up of the genUs Pourtalesia into sirbsection.s. We might retain tiie name of the 

 genus, Pourtalesia, for the bottle-shaped types allied to P. iiiiraiida. such as P. Tainieri, P. laguncula. 

 P. Jeffreysi, and form a section of the genus for the elongate /-". phiale and another for the stout- 

 tested P. ceratopyga and P. rosea. P. hispida may yet be found to l)elong to a special genus». (Op. cit. 

 p. 141). Duncan |Ke\-isioii. p. 285) excludes from the genus P. mirauda and rosea on account of their 

 compact apical system and their postero-lateral interradia being separated dorsally. — Neither Agassiz 

 nor Duncan propose new generic names for the subdivisions. Pom el (Classification methodique (324) 

 p. 4(j) goes more radically to work. He divides the group into four genera. Pourtalesia is restricted to 

 the species miranda, hispida and (?) phiale ; a new genus, J'hvalopsis. is established for /-". laguncula, 

 another genus, Ceratophysa, for J\ rosea and ceratopyga, and a third genus, Phyale, for /'. Jeffreysi and 

 probably, P. cariuata. 



I cannot agree with any of these i)ro|)osed divisions of the genus; especial!)- those proposed 

 by Pfjuiel seem to me very unfortunate and ()uile in di.saccordance with the natural relations of the 

 species. Also Duncan's exclusion of J', luirai/dn from the genus Pourtalesia is very unfortunate, first 

 becau.se it is tlie type .species of the genus, and further because its apical .system i.s, in all i)rohal)iHl\-, 



' De Mcijcrc (Siboga-Ecli. ]>. 169) fimls the stHteineiit thai thf liivial ainbuhuM-a are in iinitual contact only on the 

 abactinal .side so da-ss das Sternum hochsten.s von den benachljarten Anibulacren unterVirochen .sein kann in Duncan's 

 remarks Op. cit. p. 2S1. As far a.s I can .see this is not the meaning of Duncan, on the contrary, he probably means to say 

 that in /-*. rosra and miranda there is no contact on the abactina! side between the two ]ioster<)-lateral interradia. In any 

 case no new inforiM.itioii on the structure of these two species is given there by Duncan. 



