64 



Vault moderately couvex, covered with Dumerous polygonal 

 plates aud bearin,<:j a small subcentral proboscis. 



This species does not seem, by its geueral form and structure, 

 to be nearly related to any other twenty-armed species, and it 

 cannot be mistaken for any of them. In its general form and 

 surface characters, it seems to be related to B. affinis, B. approxi- 

 matus, B. verneuilianus, B. variabilis and B. Coifnatus. Indeed, 

 were it not for the number of arms, it might be mistaken for 

 either one of them. The variable character of the regular inter- 

 radial areas reminds one of B. variabilis, B. verneuilanus and 

 B. cognatus, and if we had other specimens, we might expect 

 to find great differences, in this respect, in different specimens, 

 and, probably, also, in the azygous areas. The one thing, only, 

 that would stand to distinguish the species would be the number 

 of arms. That is all the distinguishing specific character we dis- 

 cover now, though we have only a single specimen of this species, 

 while we have hundieds of the others. We have not discovered 

 any eighteen-armed species in this line of evolution of species, but 

 we expect it will be found. We have a nineteen armed specimen 

 but it is abnormal and may have resulted from some unknown 

 injury to a fifteen, sixteen, or twenty-armed species or an un- 

 unkuown eighteen or niueteen-armed species. We figure it below. 



Found in the Burlington Group, at Burlington, Iowa, and now 

 in the collection of A. Albers. 



AN ABNORMAL SPECIMEN. 



Plate IV, Fig. 2/, right side view; Fig. 25, adjoining side view. 



The general form of the specimen here illustrated is that of the 

 last five above described species and their kindred, B. verneuili- 

 anus. It belongs to one of those six species or to an unknown 

 eighteen or niueteen-armed species. The basal plates appear to 

 be normal and so do four of the first primary radiala, ami the 

 first azjgous plate, followed by three plates, in the second range. 

 Otherwise the plates of the calyx are not in a normal position, 

 though the specimen appears to have been fully developed, has 

 niuotoen ambulacral openings to the vault and supports a large 

 proboscis. 



The first primary radial opposite the azygons area is a small 

 axillary plate, and may be seen on the left of figure 25. On one 

 of its upper sloping sides there are two plates in an interradial 

 position, and on the other sloping side there are three primary 

 radial plates, the last one being axillary, and supporting on each 



