April 4, 1902.] 



SCIENCE. 



530 



ieal cliemistry, it is excellent because it con- 

 tains many things which every one ought to 

 know, and because one can justify many of 

 the mistakes and omissions on the ground 

 that the beginners should first get hold of the 

 general outlines of the subject, leaving the 

 troublesome exceptions until later. 



If one looks iipon the book as a work for 

 analytical chemists, for men who know the 

 practical details of their subject and who 

 would like to get a broader and more general 

 view of the theoretical side, the book is not 

 up to standard. Such men will be annoyed 

 by the quantitative application of the mass 

 law to the solubility of strong electrolytes, by 

 the tacit implication that nitrates are not 

 soluble in nitric acid, by the assumption that 

 continued addition of a salt with no common 

 ion will cause continued increase of solubil- 

 ity, by the statement that ion reactions are 

 necessarily more rapid than reactions where 

 ions are supposed not to take part. If they 

 have read the recent work of Kahlenberg on 

 the action of hydrochloric acid on oleates in 

 benzene solution, they may even ask them- 

 selves whether the electrolytic dissociation 

 theory is necessary in order to account for 

 results in aqueous solutions which are paral- 

 leled in solutions which do not conduct and 

 where the electrolytic dissociation theory 

 therefore does not apply. 



Wilder D. Bancroft. 



The Engineering Index; Five Years, 1896- 

 1900. Edited by Henry Harrison Suplee. 

 New York and London, The Engineering 

 Magazine. 1901. 8vo. Pp. 1030. Price, 

 $7.50. 



The first and second volumes of this index 

 to engineering literature, covering the years 

 1884-1895, were issued under the editorship 

 of Professor J. B. Johnson from notes pub- 

 lished monthly in the Journal of the Associa- 

 tion of Engineering Societies. Since January 

 1, 1896, this work has been done by the Engi- 

 neering Magazine, and the present third vol- 

 ume of the 'Index' is the gratifying result. 

 It contains about a hundred pages more than 

 the first and second volumes combined, while 

 the amount of matter is more than twice as 



much, owing to the arrangement of the page 

 in two columns and to the smaller type. The 

 number of periodicals indexed is about 350, 

 nearly six times as great as in the second vol- 

 ume. 



The index is a subject one, the titles of the 

 articles or papers being classified under head- 

 ings, each of which is subdivided into minor 

 ones. For example, under 'Education' there 

 are found twenty-one titles relating to engi- 

 neering education in general, these being 

 placed in alphabetic order according to the 

 first word of the title; then follow about 

 eighty special articles classified under sixty 

 subdivisions, beginning with Admission Ee- 

 quirements and ending with Yorkshire Col- 

 lege. Cross references are also given under 

 both the general headings and their subdivi- 

 sions, thus rendering it easy to follow special 

 lines of inquiry in different directions. The 

 styles of type used for the major and minor 

 headings are good ones, although perhaps a 

 little greater clearness might be secured with 

 styles somewhat lighter. 



The first volume of this series was called 

 by Professor Johnson 'The Descriptive Index 

 of Engineering Literature,' because there was 

 added to the title of each paper a brief note 

 giving an outline of its contents or an esti- 

 mate of its value. While the name has unfor- 

 tunately been changed, this excellent feature 

 of descriptive notes has been retained, and 

 these are of great assistance to the index 

 searcher, for they usually give a clearer idea 

 of the paper than can be obtained from its 

 title. For example, under the heading 'Gas 

 Engine' the title 'A Modern Motor' is some- 

 what vague, but the added note, 'The advan^ 

 tages of gas engines in points of economy, 

 efficiency, cleanliness and safety,' immediately 

 tells the reader whether or not the article is 

 likely to be of value to him. In this volume 

 the additional useful feature of noting the 

 approximate number of words in each article 

 has been introduced. The articles indexed 

 from periodicals in foreig-n languages appear 

 to be about ten or fifteen per cent, of the total 

 number; the titles of these are given in Eng- 

 lish translation, followed by the original in 

 parentheses. 



