Geology and Mineralogy. 397 



and it is evident from such a wholesale slaughter as he has been 

 forced to make that species should not be made on separate teeth 

 or a single bone, unless they represent an unusual discovery, just 

 such as that of Leidy 's first description of rhinoceroses from the 

 Great Basin. 



This review offers an opportunity to criticize the custom fol- 

 lowed here and by some other authors of attempting to discrim- 

 inate between the "generic" and "specific" characters. Such 

 characters are relative, and are generic or specific only in rela- 

 tion to some other definite type specimen. A safer way perhaps 

 is to use the term ' ' distinctive features, " as so many writers do. 



Finally, an important step is taken in this memoir in corre- 

 lating definitely the upper John Day beds with the so-called 

 Lower Miocene of the Great Plains. It seems that the paleon- 

 tological evidence for this is overwhelming, as has been suggested 

 by many writers — Merriam, Matthew, Osborn, and Peterson him- 

 self in earlier papers. There is a growing conviction, however, 

 that the dividing line between Oligocene and Miocene should not 

 be within the John Day formation, but at its top, and therefore 

 that the lower Rosebud comes at the level of the Upper Oligocene 

 of the Old World. e. l. troxell. 



4. Atlas der Krystallformen; by Victor Goldschmidt. Vol- 

 umes III-V, Heidelberg, 1916-18 (Carl Winters Universitats 

 buchhandlung). — Volumes I and II of the great work under- 

 taken by Professor Goldschmidt, were published in 1913 and 

 noticed in volumes 36 (p. 313) and 37 (p. 284) of this Journal. 

 All those who have had the opportunity to study these volumes 

 closely, must have appreciated keenly the care and thoroughness 

 with which the author had carried on his work. We have now 

 received volumes III, IV, and V of the same work each consist- 

 ing as heretofore, of 2 separate parts, one of the tables and the 

 other of text. Volume III bears the date of 1916 and includes 

 the species from danalite through the feldspar group. There 

 are 247 tables which are described in the accompanying text; 

 nearly 1,250 figures are devoted to the feldspar group alone. 

 Volume IV is dated 1918 and includes species from fergusonite 

 to ixionolite ; this embraces 133 tables. Volume V extends from 

 kainite to margarosanite with 123 tables. It is impossible not to 

 marvel at the perseverance and energy of the author in carrying 

 on so large a work to the extent noted in this brief summary, 

 and in accomplishing its publication in the same excellent form 

 as heretofore under conditions so unfavorable both on the per- 

 sonal and material side. All interested will now look forward 

 with confidence to the completion of the whole. The original 

 prospectus (see vol. 35, p. 553) mentioned "five or six volumes" 

 as the number looked forward to, but the material accumulated 

 by the author thus far seems to show that the original estimate 

 was much too small. In- any case, he is to be congratulated on 

 what he has accomplished by his arduous labors, and on his 



Am. Jour. Sci.— Fourth Series, Vol. L, No. 299.— November, 1920. 

 29 



