EASTMAN: THE DINICHTHYIDS, 27 



specimen is too imperfect to admit of a precise determination of t!ie 

 several elements, as the author has informed us by letter. 



The only other instance recorded where the plastron has been pre- 

 sei'ved in situ, is that made known by the writer at the Buffalo Meeting 

 of the American Association for the Advancement of Science. For the 

 discovery of this interesting fossil, science is indebted to Mr. F. K. 

 Mixer, Curator of the Buffalo Society of Natural Sciences, who found 

 the slab in place at the bottom of a small stream bed near Sturgeon 

 Point, on the lake shore, twenty miles west of Buffalo, N, Y. The 

 horizon at this point is the black Portage Shale, which has already 

 yielded a considerable number of fish remains.^ The plates were cor- 

 rectly determined by Mr. Mixer to be of Dinichthyid nature, and were 

 so labelled by him and placed on exhibition in the Museum of the 

 Buffalo Society. To this enthusiastic collector the writer is greatly in- 

 debted for the privilege of studying the specimen, and of presenting the 

 following description of it. 



Although the fossil has suffered considerably from aqueous and 

 atmospheric erosion, the salient features have been so far preserved as to 

 furnish points of control sufficient for reconstructing almost the entire 

 topography. The slight extent to which the diagram given in Plate 1 

 has been reconstructed may be seen from a comparison with a photo- 

 graph of the actual fossil, reproduced in Plate 4. In most cases the 

 sutural indications are so distinct, and continuous over such an area, 

 that we have only to produce them in the same general direction across 

 breaks in the surface until they meet, in order to complete the small 

 portions that are interrupted. Thus, among the prominent landmarks 

 that are left may be mentioned the terminal angles of the antero- 

 ventro-laterals, which overlie the postero-ventro-laterals in their natural 

 position. Half way between these points gives us the median line of the 

 body ; and as all the plates are arranged symmetrically with reference to 

 it, it is clear that the fossil has been in no wise distorted. A knowledge 

 of this fact permits us to supply the contours of one side from informa- 

 tion derived from the other, and fortunately the two sides supplement 

 each other to a remarkable degree. The only boundary lines that are 

 not tolerably distinct are the forwai'd portions of the antero-ventro- 

 laterals. We will consider the relations of the different plates in order. 



Ventro- Median Plates. — The first question that arises concerning the 

 median ventrals is whether they are represented by one element or by 



1 Mixer, F. K., Amer. Geol., Vol. XVIII. p. 223, October, 1896. Williams, 

 H. U., Bull. Buffalo Soc. Nat. Sci., Vol. V. pp. 81-84, 1886. 



