1881'.] PROF. F. J. BELL ON THE ECHINOMETRID^. 419 



of three or more than three primary 



plates e- g- Strongylocentrotiis. 



Group II. Morphological axis set ob- 

 liquely to long axis of the test .... Echinometrin^. 



Group III. Morphological axis set at 

 right angles to long axis of the 



test ' HETEROCENTRINiE. 



On the present occasion the observations now to be recorded are 

 based on the classification of the ' Revision ;' the further details of 

 altered classification now proposed can only be worked out when 

 sufficient details as to the Triplechinidae have been presented to the 

 Society ; the Temnopleuridae have already '" been touched : but even 

 then Temnechinus and Trigonocidaris must have a place found 

 for them ; perhaps that will, after all, turn out to be not among the 

 Echinidse at all. 



I now proceed to the details of some of the genera of the so-called 

 family " Echinometradse." 



Heterocentrotus. 



If the student lets this paper follow in succession the third part 

 of these " Observations," he will, on examining the subjoined per- 

 centage values, be struck by the fact that there is not by any means 

 that marked diminution in the proportional values of the actinal and 

 abactinal systems to which attention could scarcely fail to have been 

 drawn in the study of the Temnopleuridee. The character of these 

 latter, though perhaps hardly so well marked, will be seen when the 

 species of the genus Echinometra come under inspection. 



1 have not been able to detect any very striking differences in the 

 characters of the buccal apparatus of H. mammillatus and H. trigo- 

 narius. The most important is, probably, their difference in size ; 

 for while a test of H. mammillatus, with a height of 26 millim., 

 gave as a measurement from the tip of the tooth to the top of the 

 epiphysis the almost paradoxical amount of 29 millim., two tests of 

 H. trigonarius, 21 and 2.5 millim. high respectively, gave for the 

 same distance 21 and 23 millim. in the two cases. 



In both cases there are ascending and descending processes, which 

 are perhaps a little better developed in H. mammillatus, as is also the 

 hammer-headed widening of the free end of the radius, and its division 

 by a median notch. 



* It is not yet time to forget the words of J. Miiller : — " Der Koi-per nur bei 

 querer Lage symmetriscb, welcher von der erstgenannten Gattung {Echino- 

 metra) bereita von Brandt erkannt, und diirch Corpus transversuni ausge- 

 druckt, von Agassiz aber uicbt benierkt worden, der diese Formen mit Echino- 

 metra fiir scbief aiigesehen hat " (Abb. Berl. Akad. Wiss. 1853, p. 128). 



2 P. Z. S. 1880. 



