ECIIINOTIIURID^. 149 



phanus and that of an Astropyga or Micropyga is fully as great in these 

 particulars as that between Astropyga and such Echinothurids as Ara&- 

 osoma Owstoni or A. thetidis. In fact the resemblances between the two 

 families in the general character of the test are far more weighty than 

 are the differences. The crenulation of the primary tubercles is gener- 

 ally marked in the Diadematidae, but it is quite wanting in Micropyga 

 and Lissodiadema, in which genera the tubercles are like those of the 

 Echinothurids. 



The abactinal system of the Echinothurids represents merely an 

 extreme condition of the Diadematid form. There is really no difference 

 of importance between the arrangement of the plates in some species of 

 Arceosoma and that which we find in Astropyga and Choetodiadema. 

 When the abactinal system of Leptodiadema is compared with that of 

 some species of Echinosoma the differences are most striking, but when 

 we examine other genera we can trace every step of the transition from 

 one into the other. 



As regards the actinostome, it must be admitted that the fully plated 

 buccal membrane of the Echinothurids is quite unlike anything to be 

 found in the Diadematida3. As Doderlein has well pointed out, we may 

 regard the plating as a character developed in the Echinothurida) inde- 

 pendently, an interesting parallelism with what is found in the Cidaridie 

 and some Palaeozoic Echini, or we may look on it as a heritage, from 

 some ancestral form, which the Diadematidce have lost. The actinosto- 

 mal plates of the Diadematida3, aside from the customary buccal ten, are 

 usually numerous and often abundant. And it is interesting, if not, impor- 

 tant, to note that in nearly all Diadematids these plates are confined to, 

 or at least are much more abundant in, the ambulacra, and sometimes 

 form a double column in each ambulacrum. In Astropyga it is these 

 ambulacral plates which carry the pedicellarice, and in young specimens 

 what appear to be rudimentary tube-feet are sometimes present ; we have 

 never found any visible perforation of the plates, however. We incline to 

 the view, nevertheless, that the condition of the actinostome in the Dia- 

 dematids indicates the gradual loss of ambulacral plates similar to those 

 of the Echinothurids. 



The teeth, jaws, and perignathic girdle in the two families are so simi- 

 lar, we have not found any important constant difference. The jaws are 

 decidedly more inclined in the Echinothurids than in Diadema, but Astro- 



