﻿Notices respecting New Books. 1001 



resistance as resistance per unit area per unit length is surely open 

 to criticism, although worse definitions have been given. The 

 terms self and mutual inductance require recognition. On p. 10, 

 Benoit requires the accent. In the writer's opinion the critical 

 volume (p. 34) can have only one meaning, and it is not the 

 meaning for which the data are given here ; he latter are not of 

 much direct use ; and since they are exactly three times the value 

 of b given on the same page, they could easily be calculated by 

 those who may require them. On p. 35 ,Loschmidt is wrongly 

 spelt. The formula given on p. 50 for the boiling-point expresses 

 the change in boiling-point. 



On p. 39 the table of bursting strengths of glass tubing seems 

 very inadequate even allowing for the recommended factor two. 

 The numbers given no doubt correspond to the values obtained 

 . from a simple theory ; but in practice a very much greater factor 

 of safety than two is required. The table as it stands is only an 

 encouragement to disaster. In the value for log^> of mercury 

 vapour on p. 41 a decimal point is wanted. The approximate 

 formula of Callendar and Moss for the expansion coefficient of 

 mercury between 0° and 200° C. is stated on p. 55 to be correct 

 to 1 in "2000 ; at 0° itself it is in error by 5 parts in 1800. In 

 the formulas for the latent heat of water (p. 60) H. N. Davis' 

 value based on 37 available values should be given in a future 

 edition; this formula is L, = 92*93 (365-*) ' 3I50 . On p. 65, 

 Lummer is wrongly spelt. The word 'cathion' depends upon 

 a mistaken etymology ; we thought it was being purged from 

 scientific literature. The symbol N is used in two senses in the 

 formula near the top of page 98; in one case probably ~N m was 

 intended. On p. 106 the line beginning with the symbol ati 

 requires revising ; the numerical value given is that of a ; the 

 statement under the heading ' definition ' should be Energy of a 

 gas molecule at 0°G. = a0_=3p/2N or a = 2-02xl0~ 16 ergs per 

 deg. per gr. raol. E, is expressed in ergs per gr. mol. and in litres 

 atm./gr. mol., not per gram. 



The triviality of most of these corrections sufficiently indicates 

 the great care that has been expended in reading the proofs. The 

 authors are to be congratulated on their work. 



The mathematical tables include the values of e~ x , four-figure 

 logarithms from 100-1000 (with table of proportional parts), cor- 

 responding antilogarithms, five-figure logarithms from 100- 

 1000 and proportional parts, and four-figure tables of squares, 

 reciprocals, sines, cosines, tangents, and radians. We have used 

 some of these tables. The plan of putting a vertical line on each 

 side of the '5' column we have found to be a most excellent one. 

 The five-figure table fills two pages ; but they are arranged so 

 that no turning over is necessary. 



The index is an excellent one. 



The binding is a limp cloth which has a habit of curling up 

 when the book is lelt on the table. A stiff strong binding is 

 much more suitable for a book of this kind. 



