H. S. Williams — Ventral Plates of the Holonema. 287 



tions of the plates of Holonema, as will be seen by comparing 

 figures 1 and 2.* 



The corresponding plates are indicated by the same letters 

 in the two figures : i. I., interlateral, c, cornu or lateral spine, 

 a. 7ii. v., anterior median ventral, a. v. I., anterior ventro-lateral, 

 m. v., median ventral, p. v. L, posterior ventrolateral. The 

 continuous lines indicate outlines that are preserved in the 

 specimens, and the dotted lines, the probable outlines of the 

 perfect plate. 



On the specimen from which figure 2 is made there are also 

 traces of impressions of heavy spine-like bones, but their shape 

 is not determined ; their presence however is sufficient to indi- 

 cate that the same animal, of which the plates figured are 

 preserved, possessed also either pectoral limbs or, at least, some 

 more massive bones than the thin, flat plates of which the 

 form is distinctly preserved. 



The "pectoral limb," or "pectoral spine" as it is called in 

 the description by Copef (fig. 5), is probably the terminal 



Ventral carapace of Pterichthys {Both- 

 riolep'is) canadensis Whiteaves, after 

 Whiteaves. -k 



Terminal finger 

 of pectoral limb 

 of B. canadensis, 

 nat. size, after 

 Whiteaves. 



" Pectoral spine 

 of Holonema sp." 

 nat. size, after 

 Cope. 



* After handing this paper to the editor I have noticed that Mr. A. Smith 

 Woodward had already pointed out Claypole's error of interpretation, saying 

 '•The post-dorso-median " plate of Claypole is obviously the anterior median 

 ventral, while the •• post-dorso-lateral" and " dorsolateral" of the same author 

 are the anterior and "posterior ventro-lateral plates respectively." Catalogue of 

 Fossil Fishes, Brit. Museum, part II, p. 315. 



f On the pectoral limb of the genus Holonema of Newberry ; Proc. TJ. S. N. 

 Museum, vol. xiv. 



