1887.] NATURAL SCIENCES OF PHILADELPHIA. 89 



mid and the covering pieces along the ambulacra, at least near the 

 mouth, were intact and in place when the animal was deposited. 

 It is enveloped in a fine grained silicious mud, fine enough to pass 

 through the smallest opening, and to leave a cast of all cavities. 

 In this specimen there appears over the actinal center a small round- 

 ed knob, from which pass out radially, along the upper part of the 

 food grooves, delicate string-like impressions of the inner part of the 

 closed groove. From the inner angles of the spiracles, and passing 

 over the lip-like projections at those angles, are small elevated 

 rounded ridges connecting with the central knob, while the other 

 portion of the sj^iracle is depressed sufficiently to receive a good sized 

 set of plates. Considering that the parts comjDosed of this fine mud 

 are the counterparts of open spaces as they existed when the speci- 

 men was imbedded all plates and spines being removed by disinte- 

 gration after it weathered out of the matrix we may infer that there 

 was at the inner angle of the so-called spiracle a small channel or 

 opening, which probably served as the true spiracle, while the re- 

 mainder of the aperture which in this view of the case would repre- 

 sent a 'mere break in the test was all covered. At the posterior 

 opening the mud mould occupies a larger space, indicating a larger 

 opening ; otherwise we are not able from our specimens to give any 

 sj)ecial account of the anal opening ; neither can we obsei've any 

 special difference in the arrangement of the spines about the pos- 

 terior opening from that of the others. 



The shape and construction of the spiracles in Peiitremltidea is 

 very similar to that of Peiitremites, and we should not be surprised 

 to find its summit surmounted by a similar structure. We fully 

 agree Avith Etheridge- and Carpenter in placing these two genera in 

 the same family, but we are not so sure as to Mesoblastus, which we 

 think might be placed more appropriately with Sehizohlastm and 

 Cryptohlastus. 



The condition of the central opening in Pentremites cannot be 

 accurately determined from any of our specimens, but we have 

 distinctly seen that it is covered by several plates, independent of the 

 roofing by spines. 



The food grooves, which pass out between the spines at the re- 

 treating angles of the cone, are vaulted over by two rows of cover- 

 ing pieces which are alternately arranged. These pieces close the 

 central groove of the ambulacrum, whence they branch off" so as to 

 cover also the lateral grooves toward the pinnules. The plates cov- 

 7 



