October 8, 1903] 



NATURE 



549 



potential gradient of a volt per cm.) the same definite 

 legree of supersaturation, approximately fourfold, is 

 }uired to produce a cloud ; the phosphorus cloud, on 

 le other hand, does not require any sensible degree 

 F supersaturation for its production.' 

 There is evidence in these papers of strange mis- 

 )nceptions on the subject of ionisation. One is 

 arprised in a paper dealing with " ionised air " 

 > find such a statement as that on p. 53, 

 n„ = 3.6xio*, agreeing very well with J. J. Thom- 

 )n's 4x 10' as the number of ions in air ionised to 

 ituration by the X-rays." 



In measurements of the leakage of electricity 



irough air which has passed over phosphorus one 



rould expect the apparatus to be designed in such a 



/ay that there should be no danger of the leakage 



bserved being mainly due to the surface of the in- 



'sulating supports becoming conducting by contact 



with the phosphorus fumes. The failure to take such 



precautions detracts greatly from the value of the 



electrical observation described in these papers. 



Chapter ii. and the remaining chapters of the 

 volume on the structure of the nucleus contain an 

 account of experiments upon the clouds produced by 

 rapid expansion. There can be no doubt that such 

 experiments are easier of interpretation than those 

 made by steam-jet methods. Prof. Barus begins with 

 experiments on the colour phenomena attending the 

 rapid expansion of moist air containing nuclei, gener- 

 ally phosphorus and " punk " nuclei. It is only when 

 few nuclei are present, and the drops formed on ex- 

 pansion thus comparatively large, that normal coronas, 

 as Barus calls them, are seen surrounding a 

 luminous source viewed through the cloud. It is only 

 to such coronas that the ordinary theory of the corona 

 applies ; the gorgeous colour phenomena observed 

 when the drops are very small, numerous and uniform 

 in size are much more difficult to interpret. If it 

 were possible to deduce the size of the cloud particles 

 from the colour phenomena observed with a given ex- 

 pansion, a most convenient method of determining the 

 number of nuclei present would be available, for the 

 quantity of water separating out as a consequence of 

 a given expansion can be calculated, and hence the 

 number of drops could be determined if the size of 

 each were known. With this end in view Prof. Barus, 

 in the absence of an exact theon,' of the colours, 

 attempted to determine the size of the drops corre- 

 sponding to a given arrangement of colours by an 

 experimental method. On certain assumptions' the 

 relative numbers of the drops in a whole series of 

 successive expansions, giving a corresponding series 

 of colour phenomena, were known, the drops in the 

 final expansions being large enough to give normal 

 coronas, from which by comparison with Ivcopodium 

 coronas the radii of the drops, and hence their number, 

 could be determined ; thence could be deduced the 

 number and size of the drops in each of the previous 

 fwpansions. It is very doubtful if the method can be 

 made a trustworthy one. 



Expansion experiments made with other vapours 

 than that of water are next described, benzol, carbon 

 bisulphide, ethyl and methyl alcohol and other vapours 

 being used. Water vapour obviously differs from 

 most other vapours in one very important respect, i.e. 

 it is lighter than air. In the experiments made by 

 Prof. Barus the air was contained in a large vessel 

 with a pool of liquid at the bottom ; when the liquid 

 was water the moist air would rise to the top, and mix- 

 ing would thus take place automatically by convection 

 until the whole volume was saturated ; in the case of 

 liquids like benzol the heavy vapour-charged air would 

 lie at the bottom, the vapour only gradually diffusing 

 upwards. Uniform distribution of vapour, and hence 

 the production of circular coronas on expansion, are 

 NO. I 77 I, VOL. 68] 



to be expected with water, while with benzol, unless 

 artificial stirring has been employed or a long interval 

 has been allowed for diffusion, only the lowest strata 

 will be saturated with vapour, and the amount of 

 liquid available for each drop formed on expansion 

 will, if the nuclei are uniformly distributed, diminish 

 from below upwards ; distorted coronas, or in extreme 

 cases an arrangement of the colours in horizontal 

 strata, are to be expected. The upper part of the 

 vessel may remain free from cloud, the upper boundary 

 of the cloud marking the level at which just enough 

 vapour is present to give drops with the degree of 

 expansion used. Even when uniform distribution of 

 the vapour has been obtained, it will be destroyed by 

 the first expansion made and the subsequent entrance 

 of the dry air introduced to bring the pressure back 

 to that of the atmosphere. 



The phenomena observed by Prof. Barus are exactly 

 what one would expect from these considerations, but 

 he makes no reference to the above mentioned im- 

 portant difference in the conditions attending experi- 

 ments with water vapour and with other vapours. His 

 interpretation of the observed phenomena is, in fact, 

 quite different. " When sulphur or other nuclei are 

 put into the globe containing benzol vapour the result 

 is peculiar. Instead of distributing themselves homo- 

 geneously throughout the receiver they usually collect 

 in a heavy band near the bottom. This is invisible until 

 revealed by the first exhaustion, when a heavy sluggish 

 fog bank is seen, only a few centimetres high." 

 Again, " The most curious feature in connection with 

 benzol as well as the preceding liquids is the subsi- 

 dence of the invisible nucleated air immediately after 

 influx and without exhaustion." The "graded con- 

 densation " is interpreted as showing the nature of the 

 distribution in the vessel, not of the vapour, but of the 

 nuclei, and an elaborate series of experiments to deter- 

 mine the rates at which the nuclei travel in different 

 vapours is described ; that rate of diffusion of the 

 vapour rather than of the nuclei is involved is by far 

 the more natural interpretation. (In a short para- 

 graph, inserted apparently subsequently to the writing 

 of the paper, the possibility of this interpretation is 

 admitted.) 



The fifth chapter treats of the nuclei produced by 

 shaking liquids, particularly aqueous solutions. The 

 production of nuclei by shaking, bubbling and spray- 

 ing has been noticed by several observers, and the 

 effect of dissolved substances in the water upon the 

 persistence of the nuclei has been studied by Mr. H. A. 

 Wilson. Prof. Barus here gives an interesting series 

 of observations on a large number of solutions of vary- 

 ing degrees of concentration. These nuclei are re- 

 garded as minute drops of the solution employed, 

 which have evaporated until the concentration of the 

 dissolved substance becomes great enough to counter- 

 balance the effect of the curvature upon the vapour 

 pressure. The conditions of equilibrium of small 

 drops containing substances in solution are made clear 

 by a diagram. There can be little doubt that the 

 nuclei obtained by shaking solutions, and probably 

 also those produced from phosphorus and from mo.st 

 of the other sources used by Prof. Barus, are of this 

 nature. There is, indeed, nothing novel in the view 

 that nuclei of this kind exist. Barus, however, seems 

 to imply that all nuclei, including what other experi- 

 menters have taken to be the ions produced by X-rays 

 and similar agents, are of this type. 



An extraordinary interpretation is given (on p. 161) 

 of the experiments by which it was sought to 

 determine the difference in the action as condensation 

 nuclei of the positive and negative ions.- " If one 

 introduces nuclei or makes nuclei by aid of the X-rays, 

 in what is virtually the acid and alkaline side of a 

 battery, even if the ionised moist air is the electrolyte, 



