458 T. Mizuno on the Action of the Coherer. 



The chief results obtained from the above experiments may 

 now be summed up as follows : — 



(1) In platinum, lead, nickel, aluminium, cadmium, copper, 

 silver, steel, and potassium coherers, the action of 

 electric waves is to reduce their resistances at first to 

 a large extent, and then this reduction continues, 

 though with some intermediate rise and fall, until the 

 resistances assume certain final values. 



(2) In the case of iron, tin, bismuth, zinc, and antimony 

 coherers, their resistances are diminished at first, but 

 soon afterwards the changes become very irregular, the 

 diminution and increase occurring at random. It is 

 very interesting and also important to notice that with 

 the iron coherer the action is conspicuously irregular. 



(3) With one and also two small lead-ball coherers, the 

 resistance attains its final value very rapidly, so that 

 the successive stages of their histories present nothing 

 conspicuous. 



(4) In the alloys, namely german-silver, brass, electric 

 fuse, ordinary soldering fuse, Rose's and Wood's metals, 

 the general tendency is to the reduction of their resist- 

 ances to certain limiting values, though with more or 

 less irregularity. 



(5) In the case of the coherers with the mixed metals, 

 namely iron with silver, cadmium, Rose's and Wood's 

 metals respectively, and also silver with Wood's metal, 

 the mode of change of resistance seems to be chiefly 

 governed by the percentage ratios of the constituents. 

 In fact, the history of each coherer, as will be seen 

 from the figures, presents the character which would 

 belong to the predominating constituent. 



(6) With zinc, lead, potassium, and electric fuse, we find 



that the resistance suddenly assumes infinite value at a 

 certain stage during the experiment. 



Judging from the above results, it appears to me that the 

 action of the coherer might, in reality, be due to something 

 like welding which would take place on a small scale among 

 fine metallic particles in virtue of the induced microscopic 

 sparks. In fact I found that the lowering of resistance seems 

 to be greatly determined by the melting-point of each metal 

 and alloy ; that is to say, the lower the melting-point the 

 greater the reduction is. 



Of course it is here to be remembered that evidently this 

 reduction also depends upon the specific electric conductivity. 



For example, in the case of very easily fusible metals, 

 namely, Wood's and Rose's metals, the reduction comes out 



