36 



NA TURE 



[March 1 1, 1909 



theory of evolution," somewhat like that of the 

 botanist Reinke. Man, the growing-point of progres- 

 sive life, is conscious of directive control. Is there 

 anything more real and certain to him, and is it 

 not the X factor in all life and evolution? " Tlie 

 master-word is nature's will to live," and as man 

 is not an outside observer of the universe, but an 

 organic part of it, the author goes on to show, in 

 very interesting chapters, that ethics is for life, and 

 that art is man's expression of life. J. A. T. 



A. Course of Pure Mathematics. By G. H. Hardy. 



Pp. xvi + 428. (Cambridge : University Press, 1908.) 



Price 12s. net. 

 The title of this book is rather a misnomer. As a 

 matter of fact, the most interesting part of it is in 

 the last two chapters, which contain an excellent 

 discussion of the logarithmic and exponential functions 

 based upon the definition of logs as an integral. The 

 preceding eight chapters deal with real and complex 

 variables, limits, convergence of series, and the funda- 

 mental theorems of the differential and integral cal- 

 culus. They are chiefly interesting as an illustration 

 of the fact that there is a growing number of uni- 

 versity teachers who arc resolved that, if they have to 

 teach elementary calculus, they will do it in the most 

 rigorous way that they can, exposing the fallacies 

 which used to be calmly ignored. There is a large 

 number of examples, many of which show how much 

 more attention has been given of late years in Cam- 

 bridge to the elements of general function-theory. 

 Mr. Hardy's book is more likely to be regarded as a 

 work on the calculus than anything else; as such, it 

 will be a useful companion to such treatises as those 

 of Lamb and Gibson. M. 



Clay Modelliug in Manual Training; jroni Plan, 



Elevation, and Section. By F. W. Farrington. 



With an Introduction by J. \V. T. Vinall. Pp. 47; 



plates xl. (London : Blackie and Son, Ltd., 190S.) 



Price 3s. net. 

 Clay Modelling in Manual Training. .'Scholars' 



Handbook. (Same publishers.) Intermediate and 



Senior, plates xl., price 4d. net. Junior, plates 



xvi., price 3d. net. 

 Any practical pursuit which leads to a scientific 

 training of the hands and eyes of young pupils should 

 receive encouragement in the schools; and modelling 

 in clay can, in the hands of a skilful teacher, become 

 a very useful aid in teaching several subjects. Mr. 

 Farrington indicates how clay modelling may assist 

 school teaching in arithmetic and geography, but 

 hardly develops sufficiently these and similar practical 

 applications of this form of manual work. The 

 books will serve to provide young teachers and pupils 

 with helpful guidance. 



Handbook to the Technical and Art Schools and 

 Colleges of the United Kingdom. Compiled from 

 Official Information. With an Index to Courses of 

 Instruction. Pp. xii+140. (London: Scott, 

 Greenwood and Son, 1909.) Price 35. 6d. net. 

 Tins useful directory of some of the most important 

 schools and colleges in the British Isles providing 

 instruction in science, technology, and art gives infor- 

 mation as to the governing authority, principal, and 

 secretary of each of the institutions dealt with, and 

 particulars as to the courses of instruction arranged 

 at each centre. Though comprehensive, the directory 

 is not complete, and it may be hoped that the request 

 made by the publishers for data of schools omitted 

 will be complied with by the respective authorities, so 

 that the omissions may be rectified in the second 

 edition. 



NO. 2054, ^'OL. 80] 



LETTERS TO THE EDITOR. 

 [The Editor does not hold himself responsible for opinions 

 expressed by his correspondents. Neitlier can he undertake 

 to return, or to correspond with the writers of, rejected 

 manuscripts intended for this or any other part of Nature. 

 No notice is talcen of anonymous communications.] 



lonisation in the Atmosphere. 



The apparatus designed by Ebert has been widely used 

 to determine the total charges per c.c. of the positive and 

 negative ions in the atmosphere. Except under unusual 

 conditions, the measurement of the positive charge exceeds 

 that of the negative charge by an amount very variable, 

 which averages perhaps about 20 per cent. Thus the ratio 

 of the charges has an average value not far different from 

 the ratio of the mobilities of the ions or from the ratio cJ 

 their coefficients of diffusion. 



The apparatus consists of a metal cylindrical testing 

 vessel with an insulated axial rod connected with the 

 central system of an electroscope. Air is drawn through 

 the testing vessel at a l^nown. speed by a small turbine 

 driven by clockwork. The quantity of electricity received 

 by the central charged rod is determined from a knowledge 

 of the electrical capacity and observations of the loss of 

 potential. 



The following simple experiments by Mr. F. W. Bates 

 and the writer led to unexpected results. .\ large hollow 

 cone of cardboard was placed so that the air entering the 

 testing vessel all passed through the cone, and the air 

 during its passage was strongly ionised by the ^ and y 

 rays of radium, or by the 7 rays alone. The instrument 

 itself was well screened from the rays, and the radium 

 bromide (14 mg.) was carefully sealed in a test-tube so 

 that no emanation escaped. The position of the radium 

 was varied, so that the number of ions detected in different 

 experiments covered a wide range. 



Assuming the value of the ionic charge to be 3-4X10-"* 

 E.S.U., and supposing that every ion carried unit charge, 

 then the values obtained, after necessary small corrections, 

 gave the following average number of ions per c.c. : — 



Without radium 1,280 



1,050 



The variation in the ratio may be due to changes in the 

 humidity or to the presence of dust. 



The main point is, however, strongly marked. Whilst 

 the y rays of radium produce equal quantities of positive 

 and negative electricity when they ionise gas in a closed 

 vessel, we find that on ionising air near Ebert's apparatus 

 there appears to be a large excess of positive electricity. 



Care has been taken in designing the apparatus to 

 avoid an external field. Since negative ions are under 

 almost all conditions more mobile than positive ions, we 

 should expect the negative ions to be captured more readily 

 than the positive in the testing vessel, unless, indeed, some 

 of the positive ions had a double charge. Again, it is 

 possible that a large number of the negative ions diffuse 

 to the top and sides of the testing vessel before entering 

 it. In that case the diffusion is unexpectedly rapid. More- 

 over, the ratio, positive to negative, remained unchanged 

 when the air was drawn through an earthed wide-meshed 

 wire cylinder, when the loss by diffusion of the negative 

 ions might be expected to show a relative large increase. 



The details require further investigation, but the main 

 and important result seems to be well established, namely, 

 that the Ebert apparatus, and others of like type, arc 

 misleading in indicating a large excess of positive over 

 negative electricity in the atmosphere. Thus when 

 observers have recorded the averagp ratio as 1-2 there may 

 really have existed equality, and the apnarcnt excess may 

 be due to the inequality of the rate of diffusion of the two 



