24 SMITHSONIAN INSTITUTION 



sort of crenulation ; toward the center they are frequently somewhat 

 effaced. This plate sculpture is all a modification of a single plan; the 

 different varieties depend upon the degree to which the union and enlarge- 

 ment of the radiating costse proceed, and they may all be more or less 

 present in the same specimen. Some of the specimens have the brachials 

 beyond the lower secundibrachs perfectly smooth, others with a strong 

 transverse ridge, and still others with a girdle of detached granules. 



Upon the predominance of the various kinds of plate sculpture in the 

 interbrachial pavement Waagen and Jahn have proposed the following varie- 

 ties, which they admit are not constant: 



a. Plates of pavement rather irregularly convex in lower part; passing somewhat 



abruptly into broad ridges forming several interrupted, transverse bands, 

 or rows, resembling chains, more or less parallel, arched downward. 



var. schlotheimi. 



b. Plates of pavement in upper part with more or less deep pits, forming a net- 



work of rhombs ; in lower part stelliform. var. typica. 



c. In lower part a complicated network into which root-like prolongations from 



the sides of the lower arm-brachials enter, and from which the arms appear 

 to rise suddenly, as if they emanated from roots. var. schroeteri. 



Comparing the figures, it is impossible to identify these three varieties 



with any certainty, as they all run into one another, the characters often being 



all combined in one specimen. 



Species 2. 6". decoratus Waagen and Jahn. 



Like elegans, but the radiating costas are formed of separate granules. 

 Calyx very large. 



These characters are found intermingled in my specimens (see PI. II). 



II. Group subornatus: Calyx plates not sculptured. 

 Species 3. 5". cenonis Waagen and Jahn. 



Plates smooth or granular, convex, with crenulated margins. Pave- 

 ment as in var. schlotheimi. 

 Species 4. S. subornatus Barrande (MS.). 



Plates smooth or granular, flat or concave in the middle, straight or 

 irregularly plicated ; interbrachial pavement of smooth, flat plates ; arm 

 plates also smooth, and but slightly raised. 



To these might be added another variety under 5\ elegans from a fine 

 specimen in my collection in which the plates of the iBr pavement are smooth 

 except for a pustulose center. 



All of these species and varieties are derived from the transition beds 

 between etages E 1 and E 2 of Barrande's section in central Bohemia, hitherto 

 called Silurian, but which Dr. Ulrich is now inclined to consider as more nearly 

 equivalent to the lowermost Devonian (Helderbergian) formation of this 

 country. In view of the recent discoveries in Tennessee, whereby the range 



