144 PENETRATION PHENOMENA IN LIQUID WATER 



three ''discrepancies" is impressive. (One would not expect them to be 

 identical, even if the experimental errors were negligible, for the meas- 

 urements yield three closely related but essentially different quantities.*) 



It merits mention here that the reality of the discrepancies found by 

 Michl and by Philipp has never been generally accepted by workers in 

 the field. Thus Rutherford (9), in 1930, dismisses the effect as ''small 

 and difficult to account for." Gray (10), in an authoritative review pub- 

 lished in 1944, also discounts the direct measurements on liquids, imply- 

 ing that they are in error in some unknown way (s) ; generalizing from an 

 analysis of the stopping powers of a number of compounds, mostly of 

 C, H, and and almost all in the vapor phase, he concludes that pos- 

 sible effects of chemical binding and state of aggregation on the stopping 

 power for fast particles are at most of magnitude ±1 per cent. As far 

 as applications are concerned, it has always been the custom, in any con- 

 siderations into which the stopping power or range of energetic charged 

 particles in liquid water enters, to presume the water to have behavior 

 identical with that of a mixture of hydrogen and oxygen gas having the 

 same composition and density as the liquid. 



We now have two new and different measurements, performed with 

 modern techniques. The work of Appleyard, particularly in its semi- 

 quantitative agreement with the earlier work, undermines the com- 

 placency sketched above, although a great difference between liquid and 

 vapor is so remarkable in the light of theory, as will be discussed below, 

 that still another, independent confirmation would appear to be de- 

 manded, and more accurate data are necessary in any event. However, 

 the conclusion of de Carvalho and Yagoda that Sr is the same for liquid 

 and vapor (and solid) throws a cloud of uncertainty over the entire situ- 

 ation. It is clearly imperative to have a vigorous attack on the experi- 

 mental problem, preferably from several different directions, not onty 

 to establish the correct results, but also to identify the nature of the 

 various errors. 



In view of this situation, it does not seem propitious to investigate 

 extensively the theoretical aspects of the problem. Only a qualitative 



* Some doubt may also be raised that the medium in any of the experiments was 

 truly "water." Thus Appleyard used rather concentrated electrolytic solutions, in 

 which the ions were separated, on the average, by only about 3-4 water molecules. 

 And in all the experiments the distance traversed by the alpha particle in the liquid 

 was small — of the order of magnitude of 10^ molecular diameters of water in the 

 experiments of Michl, of Philipp, and of de Carvalho-Yagoda, and even less in that 

 of Appleyard — and the water layer might have had structural abnormality (of sur- 

 face in the experiments of Philipp and of Appleyard, of intersurface in those of Michl 

 and de Carvalho-Yagoda). It has not been possible, however, to invoke a plausible 

 reason why such peculiarities might account for the observations. 



