SCIENCE 



[N. S. Vol. XXIX. No. 753 



cause in some ways the treatment is most ele- 

 mentary and in other chapters use is made 

 of the calculus. The field covered is satis- 

 factory on the whole, as is shown by search- 

 ing in the various chapters for discussions of 

 all the important general phenomena. It 

 would not be difficult to refer to certain sec- 

 tions which might be omitted with profit, but 

 on the whole there is little to criticize in this 

 respect, for the rights of the individual 

 teacher and author must be respected. 



The success of any text-book in physics 

 must be decided from its effect upon the mind 

 of the students who use it. If they are taught 

 by means of it to reason correctly, and if they 

 learn a consistent view of the great phenom- 

 ena of nature, it has accomplished its pur- 

 pose. A great deal naturally depends upon 

 the teacher, but certainly in the present case 

 the authors of the text-book referred to above 

 have done their full share. J. S. Ames 



The Johns Hopkins Univeesitt 



A Manual of North American Diptera. By 

 Samuel W. Williston. Third edition, illus- 

 trated. New Haven, Conn., James T. 

 Hathaway. 1908. Price, $4.00 postpaid. 

 Pp. 405, duodecimo, cloth. 

 The much-desired third edition of Dr. Wil- 

 liston's manual was actually published and 

 some copies distributed about August 28, 1908, 

 but on account of the absence of the author 

 on a fossil-hunting expedition in western 

 Texas only a few copies were sent out until 

 about the end of the year. It has, therefore, 

 received but little notice in reviews up to the 

 present time. 



The book, like the preceding editions, is 

 designed largely for beginners. It contains an 

 introduction, a treatise on the anatomy of 

 diptera, suggestions as to methods of collect- 

 ing, preserving and studying the insects, some 

 general remarks on the principles of classifica- 

 tion, a synoptic table of the families and a 

 series of chapters on the families. These last- 

 mentioned chapters each include one family, 

 giving in uniform style the following topics : 

 definition of the family, characters of the 

 larvse, habits of larvae and adults (often at 

 some length), and table of genera based on 



adult characters. In a few cases the larvse 

 are to some extent subdivided in a separate 

 table. Several of the chapters are written 

 partly or wholly by other entomologists. 



The illustrations form a new and conspicu- 

 ous feature of the work, numbering nearly a 

 thousand. While recent entomological litera- 

 ture has been drawn upon to some extent, a 

 large proportion of the figures are new and 

 drawn by Williston himself, representing an 

 immense amount of labor on his part. 



In the preface, after mentioning the suc- 

 cessive publications in which he had attempted 

 to outline the classification of North Amer- 

 ican diptera, the author states that he feels his 

 work in this line completed with the present 

 publication. Perhaps, for this reason, he has 

 allowed himself to express his views and even 

 his feelings to a greater extent than in former 

 editions. Many passages might well be quoted, 

 either as illustrating generalizations derived 

 from thirty years of strenuous scientific work, 

 or to illuminate points of disagreement be- 

 tween the author and certain younger dipter- 

 ists. A very few selections must suffice. 



Giantism in any group of animals is a special- 

 ization, and is, in general, an indication of ap- 

 proaching decadence. . . . No strong or dominant 

 group of flies, like the Tachinidse, Dolichopodidse, 

 Syrphidfe or Bombyliidae, has ever had in the past 

 a larger average bodily size than is found among 

 their living representatives. 



On the splitting of genera in the mosqui- 

 toes: 



I fear even Desvoidy's shade would turn pale 

 with envy in the contemplation of some of the 

 proposed genera of the modern culicidologists. 



On the " mere collector " : 



His labors are hardly more important than 

 tnose of the microtomist who cuts up frogs' eggs 

 and makes pictures of them. 



In the matter of wing nomenclature the 

 common system is wisely adhered to, while the 

 Comstock (here called the Comstoek-Need- 

 ham) is illustrated in a page of wings. Un- 

 fortunately, the tabular exhibit of homologous 

 terms is imperfect because Comstock's earlier 

 designations are used. The fact that there 

 are already three distinct forms of Comstock 

 nomenclature in existence is an excellent rea- 



