180 
OSTEOSTRACI. 
are less produced and less divergent than in the English heads.” A 
Glammis specimen mentioned below (No. P.3234) apparently rend¬ 
ering this conclusion unjustifiable, we venture to revert to Agassiz’s 
Eig. 21. 
Cepkalasjris lyelli, Ag.—Side view, restored by Lankester. [The opercular fold 
is too distinctly separated from the shield, having the appearance 01 a 
pectoral appendage.] 
original arrangement, and regard the differences as due to accident 
in preservation. 
20087. Type specimen, described and figured by Agassiz (p. 143, 
pi. i. a. fig. 2) and Lankester (p. 44, pi. viii. fig. 1); 
Glammis, Forfarshire. Since the original description of 
the specimen by Agassiz, the squamation of the caudal 
region has been more completely extricated from the 
matrix; and during this process the greater part of the 
opercular fold of each side was accidentally destroyed. 
Presented by Sir Charles Lyell, Bart., 1846. 
P. 3233. Portions of head and trunk of a similar specimen ; Glammis. 
o/ '7y/>£ . Enniskillen Coll . 
P. 3234. Small shield, much crushed and broken, labelled C. agassizi 
by Mr. William Davies, and agreeing in every respect with 
the definition of this supposed distinct species; Glammis. 
The portion of shield in advance of the anterior margin of 
the orbits measures 0*015 in length, that behind 0*02. 
Enniskillen Coll „ 
42140. Fragments and internal impression of shield ; Cradley, 
Herefordshire. Baugh Coll. 
P. 3235. Small shield, scarcely crushed, but wanting the external 
layers ; Cradley. Enniskillen Coll. 
36052. Portion of a similar specimen ; Cradley or Ludlow. 
Purchased , 1861. 
45945-^47, 47 a Four imperfect shields 
Whitbatch, near Ludlow 
Lightbody Bequest. 
