258 bulletin: museum of comparative zoology. 



there has been called /. maximus. As /. iowensis has certain char- 

 acters half-way between /. gigas and I. maximus, this misidentification 

 has served to throw the Cincinnati specimens into hopeless confusion. 

 The following key will show the more important characters which 

 separate the genera of asaphids occurring in the formations from the 

 Chazy to the Richmond. 



A. Depressed or concave border on both shields. 



a. Surface of shields ribbed. 



1. Free cheeks meeting in front Ogygites. 



2. Free cheeks separated in front Basilicus. 



b. Surface of shields not ribbed. 



1. Axial lobe narrow, prominent Isoteloides. 



2. Axial lobe wide, depressed Isotelus. 



B. Concave border on pygidium only. 



a. Eyes very large Nileus. 



b. Eyes small but very high Vogdesia. 



C. No concave border on either shield. 



a. Free cheeks meeting in front Onchometopus. 



b. Free cheeks separated in front Brachyaspis. 



The four more common species of Isotelus in the Middle and 

 Upper Ordo vician can be separated as follows : — 



A. Shields about three fourths as long as wide. 



a. Adult without genal spines. Shields subtriangular . ./. gigas. 



b. Adult with genal spines. Shields rounded /. iowensis. 



B. Shields less than three fourths as long as wide, regularly rounded. 



a. Adult without genal spines /. latus. 



b. Adult with genal spines I. maximus. 



This may at first sight seem to be an arbitrary arrangement of the 

 species, but these are not the only characteristics in which the above 

 species differ, the other points being given in the general discussion 

 of each species. The geographical distribution is also in agreement 

 with the above separation. For instance, the specimens which I have 

 described as Isotelus latus were considered by Billings the female of 

 /. gigas, but if this is true, why are all the hundreds of specimens found 

 at Trenton Falls, the narrow form? 



I have measured an extensive series of each species, and find the 

 ratio of length to the width of the shields to be an important clue to 

 their identification. From the above key, it might seem that it would 



