52 IOWA ACADEMY OF SCIENCES. 



from that its conductivity was determined. In all cases, 

 except for the initial part of the curve, the conductivity is 

 a linear function of the current. In the initial part of any 

 curve the scale readings of voltmeter and ammeter, begin- 

 ning at approximately zero, were small, and the probable 

 error in them and in the conductivity determined from them 

 is correspondingly large, and consequently the values for 

 this part cf the curve are somewhat indeterminate. This 

 remark applies equally to the curves determined when the 

 E.M.F. and current were of the order of magnitude meas- 

 ured by the galvanometers or by the voltmeters and 

 ammeters. I have considered in the same way the results 

 of Guthe and Trowbridge, Bose, and Eccles, and find the 

 same relation to exist between the conductivity and cur- 

 rent, except in the results of Eccles. In that case it is 

 not certain that the resistance obtained from his tabulated 

 values of E. M.F. and current is the resistance of the coherer 

 alone. The departure from the law is such as to indicate 

 that this is not the case. 



It is interesting to note that the conductivity of dielec- 

 trics (parofine and guta percha) as deduced from the meas- 

 urements of Leich* increases with the current flowing 

 through them in such a way as to be represented by a 

 straight line. Also that Ayrton and Perryf found the 

 potential difference between the two carbons of an arc 

 lamp was independent of the current strength provided the 

 distance between the carbons was kept constant, or in 

 other words, the apparent conductivity of the arc varies 

 directly as the current. Shaw:{; has shown that after the 

 electrical coherence or fall of resistance there is cohesion. 



The resistance does not increase again after the current 

 ceases to flow, unless the coherer is subjected to stress. 

 The resistance after lowering is not affected by a current 

 smaller than that which produced the lowering. 



All these results seem in harmony with the theory of 

 the coherer as advanced by Guthe and Trowbridge, that 

 "as the current flows, ions from the positive electrode 



* Wied. Ann. 66, p. UOi. 

 tPhil. Mag. May, 1883 

 tPhil. Mag. Mar., 1901. 



