26 bulletin: museum of comparative zoology. 



frout ; so that an interlateral element, such as is present in Cocco- 

 steus, cannot be said to exist. We are therefore limited to assigning a 

 strictly lateral (external) position for the so called " claviculars " or 

 coracoids. 



A second specimen of the ventro-median preserved in the Cambridge 

 collection (Catalogue No. 1299) shows the longitudinal ridge on the vis- 

 ceral surface more strongly developed than the first, and is both thicker 

 and wider towards its posterior extremity. 



There is yet a third specimen, which is smaller and of somewhat 

 different configuration from the preceding ; this is shown in Plate 3, 

 Fig. 1. The visceral surface is embedded in the matrix, so that its 

 character cannot be made out. In form it is somewhat suggestive of 

 the paraspbenoid bone of Ctenodus, but its structure as seen under the 

 lens proves it to be Dinichthyid. The plate was obtained by Dr. Clark 

 in the Cleveland Shale, near Lindale, Ohio. Owing to its smaller size, 

 it may be referred with some reservation to D. intermedius. 



From the same locality as the preceding. Dr. Clark has also obtained 

 two unique plates, one of which is preserved in counterpart, and is 

 shown in Plate 5, Fig. 1. Lanceolate in outline, and perfectly sym- 

 metrical, it presents a very graceful appearance ; its length is 29 cm., 

 and its maximum width 12.5 cm. Only the visceral aspect is exposed, 

 and this is marked by two slightly oblique ridges, such as occur also in 

 the corresponding position of D. terrelH. The plate is abruptly trun- 

 cated in front, and bears indications of overlap by the antero-ventro- 

 ruedian. We shall find that additional light is thrown upon these 

 relationships when we consider the plastron immediately to be described. 

 The specimen is somewhat thinner than other ventral plates that have 

 been noticed thus far, and it differs also in form. For the present, it 

 must be regarded as representing an unknown Dinichthyid. 



Overlapping Median Yentrals preserved in Situ. 



So far, but two instances have been reported where the ventral plates 

 were retained in their natural relations with respect to one another. 

 The less perfect of these was described very briefly by von Koenen,^ by 

 whom it is doubtfully referred to D. minor. Only the left half of the 

 plastron is preserved in this case ; its entire length is assumed to have 

 been about 16 cm., and its width 6 or 7 cm. The condition of the 



1 Koenen, A. von, Ueber einige Fischreste, etc. (Abhandl. Gesellsch. Wis- 

 sen. Gottingen, Vol. XL. p. 18), 1895. 



