720 Fatigue of Metals subjected to Rontgen Radiation. 



if the fatigue were due to absorption by gas. Also the 

 presence of these penetrating rays may be one reason why 

 ozone has a less influence. 



The other theory advanced to explain the phenomena is 

 that a change occurs in the metal itself. Ladenburg sug- 

 gested chemical corrosion of the surface to be the cause, but 

 the recent work of Hallwachs makes this hypothesis doubtful. 

 Then we have the theories of Allen and of Ramsay and 

 Spencer, that the radiation of metals is due to an actual dis- 

 integration of the substance. In some way, the ultra-violet light 

 releases the atomic energy and causes the atom to explode. 

 So if we begin with zinc (A) we soon get a disintegration 

 product zinc (B), then zinc (C), &c. It is rather unfortunate 

 that so fundamental an assumption as the transmutation of 

 the elements should be advanced except as a last resort 

 and with very convincing proofs. And certainly in the 

 present case such proof is lacking. In the first place, the 

 fatigue is not a permanent change and metals recover their 

 primitive efficiency. Then, too, Allen finds that the change 

 from zinc (A) to zinc (B) occurs in a few minutes. If such 

 a change takes place so quickly, we should expect to find some 

 evidence of this new product on the surface of metals which 

 have been exposed to sunlight for years. But the facts dis- 

 covered by him, that the fatigue progresses rapidly at first and 

 then at a slower rate and that the action may be expressed as the 

 sum of two exponentials, are important and interesting. But a 

 less radical assumption would satisfy his exponential equation. 

 Ramsay and Spencer go much further. They suppose the 

 ultra-violet light to produce a disintegration of the atom 

 accompanied by the ejection of a corpuscle, and that this 

 process is repeated until the number of corpuscles emitted is 

 equal to the valence of the element. The proof of this cir- 

 cumstantial disintegration rests solely on the fact that their 

 curves of radiation and time show breaks, or all their 

 measurements do not lie on a smooth curve. When we con- 

 sider the difficulty of keeping a source of ultra-violet light 

 constant for hours at a time and the errors liable to occur 

 when measuring the radiations from the plates, it is not sur- 

 prising there are irregularities in the curves. An inspection 

 of them does not carry conviction that successive disintegrations 

 of matter under normal conditions are involved. Nor do the 

 authors prove any extraordinary precautions to prevent errors 

 were taken. In fact they do not explain why other points 

 which lie at some considerable distance from the curves drawn 

 (see particularly curves III. and IV., plate vi., and curve 

 VIII., plate vii.) do not constitute points of disintegration. 

 After the exhaustive and rigorous treatment of the subject 



