REPORT ON THE ECHINOIDEA. 133 



to separate into distinct sub-genera on comparing such extreme forms as Pourtalesia 

 miranda, P. laguncula, and P. phiale with such forms as Pourtalesia ceratoijyga and P. 

 rosea. The former group is distinguished hj the extreme tenuity, almost transpar- 

 ency, of the test and its more or less bottle-shaped outline, while the latter group con- 

 tains species with a flattened test, a triangular outline from above and a comjjaratively 

 thickened test. 



On examining, however, the group of Pourtalesise to which Pourtalesia hispida, 

 P. carinata, and P. jeffreysi belong, the impossibility of maintaining this arbitrary 

 division becomes at once apparent. 



• Pourtalesia hispida, while having the general appearance of Pourtalesia miranda, 

 has a much thicker test, and in Pourtalesia carinata we have a stout test and a pro- 

 minent snout, with a posterior apex, and the closer tuberculation of the group to which 

 Pourtalesia rosea and P. ceratoj^yga belong. It is remarkable how great is the variation 

 in the extent of the separation of the bivium and trivium at the apical system in the 

 difi^erent species of the genus. 



In Pourtalesia rosea the genital plates join the ocular plates of the bivium (PI. XXII.* 

 fig. 6), in Pourtalesia laguncula, P. hispida, P. ceratopyga, P. cari7iata and P. Jeffrey si 

 the plates of the posterior lateral interambulacra extend entirely across between the bivium 

 and trivium, completely separating them (Pis. XXII. fig. 19; PI. XXII.* fig. 10; PI. 

 XXVIII. figs. 11, 12 ; PI. XXVIII." fig. 12). 



I have to add as one of the generic characters of Pourtalesia the existence of a 

 well-developed subanal fasciole, which in some of the species forms a broad band round 

 the anal snout. 



* Pourtalesia carinata (Pis. XXVII.% XXXVIII. figs. 30, 31 ; PL XLI. figs. 49-52; 

 PI. XLII. figs. 24, 25 ; PI. XLIII. figs. 20-23 ; PL XLV. figs. 46-52). 

 Pourtalesia carinata, A. Agassiz, Proc. Am. Acad., vol. xiv. p. 205. 



This is a large species with a comparatively stout test. It resembles in outline Pour- 

 talesia miranda, but is more gibbous, with a more regularly-sloping anterior extremity (PL 

 XXVIII." fig. 2), which brings the apex more posteriorly than in that species. The 

 greatest breadth of test seen from above is also more posterior (PL XXVIII." fig. 1). The 

 primary spines are more numerous on the median interambulacral line of the abactinal 

 side of the test on the anterior and on the odd interambulacra. The rest of the test is 

 quite thickly covered with small secondary spines increasing in size towards the ambitus, 

 and on the actinal side the keel, of the plastron carries still larger primary spines on 

 tubercles closely packed on the ridges of the line of the actinal keel (PL XXVIII." fig. 8), 

 and also crowded on the actinal part of the anterior interambulacra and on the inter- 

 ambulacral plate near the anal snout (PL XXXVIII." figs. 3, 4). The greatest height of the 

 test is posterior, the apex not corresponding with the apical system. The subanal fasciole 



