192 CATALOGUE OF THE BLASTOIDEA. 



good reasons for removing this [P. laterniformis] and other subfusiform species as 

 Pentremites Reinwardth, P. lineatus, P. hip y rami dalis, P. Wortheni, and perhaps 

 P. Grosrenori from among the Pentremites, and grouping them together in a 

 separate subsection under another name. These and allied forms are remarkable 

 for their slender subfusiform shape, linear pseudambulacral fields, triangular base, 

 and simple summit-structure. These external differences would seem to imply 

 corresponding modifications in the internal economy of the animals of more than 

 specific importance. If from a more thorough study of such species it should be 

 deemed advisable to separate them from the genus Pcntremites, I would propose the 

 name Troosticrinus for the group." We have already expressed our concurrence in 

 this proposition by adopting Troostocrinus in 1 882 in the terms of Dr. Shumard's 

 remarks; and we had prepared a definition of the genus which should embrace all 

 the species named by Shumard (with the exception of P. laterniformis 1 , which is only 

 the internal cast of a Pentremites) and a few others as well. The morphological 

 portion of the present work was written according to this view ; and it was not until 

 the section on the deltoid plates was passing through the press that we saw any 

 reason to change it. But, as we have stated on pp. 36, 37, evidence was put before us 

 by Mr. Wachsmuth which seemed to indicate that the Silurian species Troostocrinus 

 Eeinwardti should be separated generically from its successors of Carboniferous age 

 on account of the anal deltoid being larger than its fellows, and appearing externally 

 above the radial limbs (see Fig. VII. on p. 193). But we hesitated to take this course 

 at first, though we have now decided to do so for two reasons — (1) Mr. Wachsmuth 

 has been good enough to examine for us all his available material of this species, 

 and he finds that the anal deltoid appears externally in all the specimens which are 

 sufficiently well preserved to show the structure of the summit. (2) Troostocrinus 

 Ilcimvardti has an anal spiracle; but in Shumard's Pentremites lineatus (PI. III. 

 figs. 14, 15), and in most of the other Carboniferous species, the anus opens separately 

 from the two posterior spiracles as in Tricoslocrinus (PI. XVIII. fig. 15; PI. XIX. 

 fi<*s. 15, 10). 



The combination of these two characters appears to us to be sufficient to warrant 

 the generic separation of Pentremites Eeinwardti and P. lineatus. The former, 

 having been originally described by Troost, was the first comprehensible type named 

 by Shumard in the group of species which he proposed to refer to Troostocrinus 1 . 

 There is therefore every reason for making it the type species of the genus as we now 

 define it ; while in accordance with a suggestion from Mr. Wachsmuth, we propose to 

 establish a new genus, which we have called Metablastus, for Pcntremites lineatus 



1 If /'. laUrniformis is really onh an interna] cast of a Pentremites, it might bo argued that Troostocrinus 

 is onlj a synonym of that genus, because /'. laUrniformis was the first named among the species referred by 

 Shumard to Troostocrinus. We have preferred, however, to regard the next species named by him as the 

 type of the genus ( /'. ReinwardH), and we believe that most palaeontologists will think that we have taken 

 the right course under the circumstances. 



