Beview of liecent Geolo(ji(:al Liferature. 317 
lieen inferred on ii priori grounds." To the reviewer, this fact appears 
confirmatory of the suggestion recently advanced that the swimming 
powers of DinichtJiys were centered principally in the hinder part of 
the trunk and tail, and were in direct relation with the posterior process 
of the dorsal shield, a process which finds its most complete develop- 
ment in this genus. 
Of great interest is the author's description of the ventral armor, and 
his comparison of it with other forms. Wright's able interpretation of 
the system is supported in the main, and is at the same time completed 
by the discovery of an undoubted ventro-median plate. If Dr. Dean's 
view is correct that this element is single and undivided, the species 
presents a marked variation from the normal condition in the dinich- 
thyids, and in the Arthrodira generally; and yet, such a variation is 
readily explicable, when we consider the loo.se connection between the 
ventrals and their great tenuity, as a device for strengthening an other- 
wise fragile plastron. Nevertheless it must be observed that the evi- 
dence is not entirely decisive on this point, owing to the extremely 
weathered condition of the anterior portion of the ventro-median. What 
Dr. Dean considers to be the homologue of an antero-ventro-median may 
be in reality all that is left of that very element itself, which had be- 
come displaced downward so as to overlie the ventro-median proper, 
and afterwards had suffered almost total obliteration. In any case we 
know that two elements are potentially present, an anterior and a pos- 
terior, but whether the relation between them was one of fusion or of 
suture appears to be indeterminate from the specimen itself. It is un- 
fortunate, too, that the conditions of overlap among the several plates 
are not clearly decipherable from the present sijecimen, nor are they in- 
dicated in Prof, von Koenen's recent description of D. minor (?). The 
precise relations of the elements constituting the ventral armoring have 
yet to be ascertained from a specimen preserved in situ, such a one, for 
instance, as was l)riefly noticed in the last number of the Geologist, 
(p. 222). This much, however, is already certain: namely that the ven- 
tral plates of Dinicli-tJn/K are subject to greater variation, even within 
specific limits, than all the others of the derm covering. One class of 
variations may be alluded to here, which is known to affect different 
individuals of one and the same species, and that is the relative length 
of the anterior ventro-laterals as compared with the posterior pair. In 
one series, (A), the former ai-e longer than the latter, and the total 
length of the plastron is greater than in the corresponding series, (B), 
where the reverse condition obtains. Augmentation in length was only 
possible in a posterior direction, the anterior boundary being fixed; 
hence the first series must have had the abdomen protected by a plated 
covering over a greater area than in the second series. Although a 
matter of pure conjecture, it is not altogether unreasonable to suppose 
that these differences were correlated with sex. c. k. e. 
Fifteenth Annual Report of the Umted States Geological Survey to 
the Secretitrjj of the Interior, lsft:j-''94. By J. W. Powell, Director, 
(Pages xiv. 75."); with 48 plates, and 29 figures in the text: Washington, 
