ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 745 



A. Auricular arch not complete. Fam, 4. Arhaciadce. 



B. Auricular arch complete; rudimentary internal gill still 



retained. Fam. 5. Biadematidoe. 



C. Auricular arch complete ; no rudimentary gill. 



Fam. 6. EcJiinidm. 

 On the characters here detailed Professor Bell is led to place the 

 Echinometridee in the same family as the Echinidte. Though this is 

 contrary to the views of Gray and Agassiz, it appears to be supported 

 by histological observations ; and though his system refuses to make 

 use of the convenient separation of Desor, by means of which genera 

 were grouped under the Oligopori or Folypori, it seems to be sup- 

 ported by the history of the development and the variations in the 

 number of primary pore-plates which fuse to form a single secondary 

 plate. It is pointed out that there is no real difference in the " mode 

 of growth of the poriferous zone," as A. Agassiz believes, and it 

 is ui'ged that " whatever be the significance of the obliquity of the 

 morphological axis, there can be little doubt that it is of very great 

 importance." The Echinidse are, therefore, thus arranged : 



Group I. Body circular. EcnmiN^. 



(a) Secondary plates formed of three primary plates, e. g. 



Eddnus. 

 (/6) Secondary plate formed in adult of three or more than 

 three primary plates, e. g. Strongylocentrotus. 

 Group II. Morphological axis set obliquely to long axis of test. 



ECHINOMETEIN^. 



Group III. Morphological axis set at right angles to long axis 

 of test. Heteeocentein^. 



It is ujged that Heterocentrotus and Colobocentrotus stand further 

 from the Cidaridse than do Tripneustes and Tonopneustes, for in the 

 former there appear to have culminated " the elaborateness of the 

 ambulacral plate, the strength of the spines, and the size of the 

 buccal apparatus." 



The great body of the paper is made up of percentage measure- 

 ments and other observations on the groups united by Agassiz under 

 the head of " EcMnometradce," to which we have no space to refer. 



Crossaster.* — Professor Jeffrey Bell points out that no definite 

 diagnosis of this genus, as distinguished from Solaster, has yet been 

 given, and he directs attention to the fact that when we examine 

 the actinal surface of a ray of Solaster endeca we observe that, exter- 

 nally to the transverse set of spines which runs near the ambulacral 

 groove, there is a series of special plates bearing a comb of spines ; 

 though in a sense marginal, these plates are quite confined to the 

 actinal surface ; from the dorsal view one would not have the least 

 suspicion of their existence. It is not so with C. papposus, for there 

 the modified marginal plates are set on the upper part of the side 

 of the ray, and form a regular series of dorso-marginal plates. 

 Coming lastly to Lophaster, we find that the " differentiated marginal 



* Ann. and Mag. Nat. Hist,, viii. (1881) pp. 140-2. 



