1 



546 TRANS. ST. LOUIS ACAD. SCIENCE. 



characters for this genus according to Mr. Carpenter), besides 

 other peculiarities in the ambulacral field. As we have seen, it 

 is mainly the deltoid piece which causes the differences in the 

 spiracle openings, and these pieces are subject to abnormal 

 developments almost as much as the hydrospiric sac* I see no 

 good reason to separate the first division from the second, be- 

 cause the number and relative position of these plates to each 

 other remains the same, and, as diflerences in the sizes of the ca- 

 lyx plates (as Mr. Carpenter remarks) are of no systematic value, 

 they should not be separated. 



All described Pentremites (except those which belong to the 

 genus Cadaster or Codonites) can easily be distributed in either 

 one or the other of these three divisions ; it is therefore impracti- 

 cable to divide the genus Pentremites into four or five new gen- 

 era, as has been proposed by Mr. Carpenter. For instance, the 

 diflerence of Troostocrinus clavatus, according to 

 Carpenter = Pentremites clavatus Hambach and 

 P. pyriformis Say, consists mainly in the different 

 length of the base and fork pieces, and there is cer- 

 tainly a closer relationship between these two spe- 

 cies than between P. clavatus and Reinwardtii or 

 Uneatus. P. Maia and Leda Hall resemble more 

 Pentremites conoideus Hall as P. Curtus and Roe- 

 meri Shumard or angulatus Phillips and granula- 

 tus Roemer. The difference between Pentremites 

 Pailleti. Reinwardtii^ clavatus Ham., Wortheni^ bipyrami- 

 dalis^ obliquatus^ and Woodmani, consists only in the different 

 development of the calyx pieces and illustrates the transition from 

 one species to another very well, but show no greater divergences 

 than there exist between Pentremites melo^ Sayi^ glaber^ Roe- 

 meri^ granulatus Roemer, neglectus^ etc. ; or between P. Jlore- 

 alis, pyriforjuis^ sulcatus^ elongatus^ hcmisphericus^ conoideus^ 

 and abbreviatus. But neither one of these divisions possesses 

 sufficient permanent characteristics to base a new classification 

 on, nor would they be any improvement on the one given by 



* To show the diversity in the development of the deltoid pieces, I give here (see Fig. 6) 

 an outline figure of a Pentremites Beimvardiii which has on one side the deltoid piece 

 laterally expanded. Besides this, I have a number of other anomalies in my collection, and 

 Carpenter himself reports of one. 



