30 SCIENCE PROGRESS 



it traverses become positively charged also, though this is by 

 no means the only evidence of the particular sign of the charge. 

 1 will return to this question of ionisation later on and will 

 only say now that not only is the ionisation the result of the 

 interpenetration of the atoms but in all probability the ioni- 

 sation of radioactivity has no other cause ; when it occurs it 

 is a sign that the "domain" of the atom has been broken into 

 by some other atom or at least by an electron. 



And there is one other important result of the mutual 

 crossing. The a particle loses in energy in the act and there- 

 fore in range, as we have already seen. It is really not quite 

 clear whether it loses energy only where ionisation takes place 

 or whether it always loses energy and ionisation only sometimes 

 follows. Certainly the same expenditure of energy in different 

 gases does not always produce the same amount of ionisation. 

 The significance of this will only appear in full when we have 

 gone a little further with the experiment we are imagining to 

 be in progress ; I will call attention to it at the proper 

 time. 



Let us place screens of different metals in the path of the 

 a rays or allow them to penetrate as far as they will in different 

 gases : and let us compare the losses of energy (or of range) 

 which the a particle suffers in going through equal numbers 

 of atoms or molecules. A very remarkable law appears. The 

 loss of range in going through any atom is very nearly pro- 

 portional to the square root of the atomic weight. Thus an 

 atom of silver takes away twice as much energy from the 

 particle as an atom of aluminium, being four times as heavy. 

 There is no very obvious explanation of this result to be offered 

 at present. It is so striking and peculiar that it is sure to have 

 some very important meaning if we could but get at it. One 

 other law appears, also of great simplicity and much more easy 

 to make something of. Suppose the substance penetrated by 

 the a particle to consist of compound molecules, solid or gas : 

 we then find it possible to anticipate with exactness the loss of 

 range which the a particle will experience in passing through it. 

 We have only to allow for the separate atoms, without con- 

 sidering any molecular combinations they may have. The loss 

 of range due to a molecule of methyl bromide (CH 3 Br) is pro- 

 portional to n/i2 + 3v/i + \/8i : that due to benzene (C 6 H 6 ) to 

 6%/ 1 2 + 6\/i. Linkings and groupings and "rings" are of no 



