December 4, 1908] 



SCIENCE 



803 



Some of the best evidence on the inheritance 

 of acquired characters has been too recent, 

 perhaps, for consideration in this edition. 

 Unless we are much mistaken in the next 

 edition this section will chronicle the greatest 

 advances in the theory of evolution. 



As for Mendelism the treatment given by 

 Plate is all too brief, though appreciative. 

 He entertains the view that the Mendelian 

 result follows between closely related indi- 

 viduals but not between distinct species; be- 

 tween the latter the characters blend. This 

 conclusion seems to the reviewer insufficiently 

 founded; the cross between a goldfinch and 

 a canary shows in the first generation no more 

 blending of characters than that between two 

 races of canaries. The method of inherit- 

 ance probably depends less on the degree of 

 relationship of the individuals crossed than 

 upon the nature of the characters concerned. 



Of the book as a whole one can speak only 

 in praise. Notwithstanding its conservative 

 attitude, it affords, in the reviewer's opinion, 

 the best general resume extant of modern 

 evolutionary data and theories ; there are other 

 excellent resumes but, for the most part, now 

 out of date. How quickly a book on this topic 

 may become out of date is indicated by the 

 growth of the bibliography in the successive 

 editions of Plate's book. In the first edition 

 there were about 210 titles ; in the second, 260 ; 

 and in the last 450. The new edition will be 

 widely read, but it deserves the greater ac- 

 cessibility that an English translation would 

 give. In any case it seems well adapted to 

 hold its ground, for some time to come, as the 

 standard treatise on Darwinism. 



Chas. B. Davenport 



The Fossil Turtles of North America. By 



Oliver Perry Hay. Carnegie Institution. 



Pp. 568, pis. 113. 1908. 



The appearance of Dr. Hay's extensive and 



richly illustrated memoir upon the fossil 



turtles of North America will mark a new 



departure in the study of this important and 



interesting order of reptiles. A bulky volume 



of nearly six hundred pages and over one 



hundred plates, aside from the numerous text 



illustrations, it will enable the student, for the 

 first time in many years, to understand and 

 appreciate the material at his command, for 

 there are few collections of extinct verte- 

 brates in America which do not have some 

 remains of turtles. We may now expect a 

 rapid increase in our still very defective 

 knowledge of these animals, since but very 

 few of the 276 species described in the present 

 work are completely known ; indeed, much the 

 larger part are yet imperfectly known. The 

 author, after the examination of nearly all 

 the types, as well as most of the known 

 material in America, has systematized and 

 correlated our present knowledge so that the 

 work wiU serve as the basis of the literature 

 for future studies. 



The reviewer has read attentively the ex- 

 tended and detailed introductory parts of the 

 volume on the structure, classification, geo- 

 graphical and geological distribution and evo- 

 lution of the Testudinata, and for the most 

 part has only commendation and approval. 

 But he can not approve the classification that 

 Dr. Hay adopts. The division of the order 

 into two chief groups or suborders, the Athecse 

 and Thecophora, first proposed by Dollo and 

 Cope, has been carried to an extreme by the 

 author, in that he would have the former a 

 primitive branch from the testudinate stem, 

 arising in early Trias or late Permian, and all 

 its aquatic adaptations and chelonid resem- 

 blances purely of parallel origin; an hy- 

 pothesis difficult to accept. He assumes that 

 the primitive turtles possessed two dermal 

 coverings, an inner represented by the cara- 

 pace and plastron of ordinary turtles, an outer 

 persisting in Dermochelys of the present time; 

 that, in aU other turtles, the outer has been 

 whoUy lost save a few ossicles in Toxochelys, 

 while in the latter only vestiges of the inner 

 covering have persisted. Such a development 

 of an outer dermal covering is not impossible, 

 as evidenced by osseous scutes overlying the 

 dermal clavicles in certain lizards, yet one can 

 hardly conceive of a condition in the early 

 reptiles which would bring about the concur- 

 rent development of two coverings; certainly 

 we have no warrant in calling the inner 



