June 23, 1916] 



SCIENCE 



897 



The definition of mathematical probability 

 from which are developed the elementary 

 theorems of probability is quoted from Czuber, 

 and is about the usual definition of a priori 

 probability. The author is rather emphatic in 

 his criticism of the idea of replacing the 

 a priori probabilities of Laplace by the empir- 

 ical ratios of Mill, Venn and Chrystal. He 

 believes the distrust of a priori probabilities is 

 due to a misapprehension of the true nature of 

 Bernoulli's theorem, which is the cornerstone 

 of the theory of statistics. The chapter on 

 probability a posteriori deals with the criti- 

 cisms of Bayes's rule in a rather constructive 

 manner, by indicating the limitations under 

 which Bayes's rule will give correct results in 

 practise. The author makes the connection 

 between a priori probabilities and statistical 

 series by the use of the well-known theorem of 

 Tchebycheff. In this connection he offers a 

 proof that the limit of a relative frequency 

 a/s when s becomes infinite is the postulated 

 a priori probability p. It seems to the re- 

 viewer that the notion of limit here employed 

 is not quite the rigorous notion ; for, the state- 

 ment that the probability that I a/s — p I <C 8 

 approached 1 as a limit, is not the same as 

 the usual statement that I a/s — p I becomes 

 and remains leas than 8. The author does not 

 seem to discriminate in this connection be- 

 tween a point of condensation and a limit 

 point. 



One of the most interesting and important 

 parts of this book is its neat and striking ap- 

 plications of Bernoulli, Poisson and Lexis 

 series to the characterization of actual data. 

 Purthermore, the application of the Lexian 

 ratio and of the Charlier coefficient of disturb- 

 ancy is clearly shown. Taken as a whole, this 

 book will be found of much value to students 

 of the mathematical methods in statistics. 



H. L. EiETZ 



Gould's Practitioner's Medical Dictionary. 

 Third edition, revised and edited by R. J. E. 

 Scott, M.A., B.C.L., M.D., of New York. 

 Pp. XX -|- 962. Flexible cloth, round corners, 

 marbled edges. P. Blakiston's Son & Co., 

 Philadelphia. Price $2.75. 



The history of medical dictionaries begins 

 with the fifteenth century. The first works of 

 the kind are the " Synonyma " (Venice, 1473) 

 of Simone de Cordo or Simon of Genoa and the 

 contemporary Pandects of Mathceus Silvaticus. 

 Both these works are alphabetical lists of 

 medicinal simples, but a goodly number of real 

 medical dictionaries were published during the 

 Renaissance period, in particular those of 

 Lorenz Fries or Phryesen (1519), Henri 

 Estienne or Stephanus (1564) and Jean de 

 Gorris (1564). 



In the seventeenth century appeared the 

 famous " Lexica " of Bartholommeo Castelli 

 (1607) and Steven Blancard (1679) which passed 

 through many editions. After these the num- 

 ber of medical dictionaries is legion. Among 

 the best known of more recent times are those 

 of Robert James (London, 1743) and P. H. 

 ISTysten (Paris, 1810), which, in 1855, was en- 

 tirely rewritten by Emile Littre and Charles 

 Robin and is still a standard source of refer- 

 ence. In England, the dictionary of the New 

 Sydenham Society (1878-99), in America that 

 of Frank P. Foster (1888-93), and in France, 

 Galtier-Boissiere's " Larousse Medical illus- 

 tre " (1912), are monmnents of scholarship. 

 Gould's large illustrated medical dictionary 

 (1894), frequently revised and reedited, has 

 been of great practical use to the medical pro- 

 fession. Of late years the tendency has been 

 towards handy volumes of reasonable thick- 

 ness, printed on thin paper, with flexible 

 covers, and of these the new edition of Gould's 

 Practitioner's Dictionary is an excellent ex- 

 ample. 



This new edition is unsurpassed as to com- 

 prehensiveness, clearness and size. It contains 

 over 70,000 words. To reduce the size of the 

 book and to make it a handy volume a small 

 type had to be selected, but the type is very 

 clear and legible and is even a little larger than 

 that used in "Webster's Unabridged Dictionary. 

 Each word is accompanied by its pronuncia- 

 tion and followed by its etymology. The defi- 

 nitions are clear and concise. 



The book contains all the numerous and 

 latest eponyms in their proper alphabetical 

 order, such as Abderhalden's test, Alzheimer's 

 disease. Lane's kink, Meltzer's method, 



