566 STATE BOARD OF AGRICULTURE. 



which are dependent on the cjiitacilv and liealing oflecls in the resistances, 

 and in Ihe former the limitations arc only tliose that can be overcome. If 

 if were not necessary, for exami)le, to load a fjalvanomelcr system with a 

 mirror for the i)nri)ose of readinj; a defied ion, and if again the moment of 

 inertia of the movable coils were as small as possible and if still fnrther 

 the coils were wonnd non-indnctively and for the annnlment of capacity, 

 then the galvanometer would have a large figure of merit. Progress in 

 galvanometer construction is being rapidly made, and it only rests with 

 the future to produce an extremly sensitive apparatus. The use of a 

 galvanometer of the dynamometer type as a detector for liquid resist- 

 ances was suggested more than fifteen years ago by the late professor 

 Kowland (84) of Johns Hopkins University. It is surprising to note that 

 apparently little attention has been given to this suggestion. . The follow- 

 ing quotation is taken from his Physical Papers, page 294: 



"I also give a few methods of resistance measurement more accurate 

 than usually given by means of the telephones or electrodynamometers 

 as usually used and especially suitable for resistances of electrolytic 

 liquids'' — "I have introduced here many new methods, dej)ending upon 

 adjusting two currents to a phase-difference of 90° which I believe to be 

 a new principle. This I do by passing one current through the fixed coil 

 and the other through the suspended coil of an electrodynamometer. By 

 this means a heavy current can be passed through the fixed coils and a 

 minute current through the movable coil, thus multiplying the sensitive- 

 ness possibly a thousand times over the zero current method." ''Using 

 non-inductive resistances, methods 10 and 14 are especially good and I 

 have no doubt, will be of great value for liquid resistances. The liquid 

 resistances must, however, be properlj'^ designed to avoid polarization 

 errors." "Altogether the results of experiment justify me in claiming 

 that these methods will take a prominent place in electrical measurements 

 especially where fluid resistances, * * * are to be measured. The new 

 method of measuring liquid resistances with alternating currents allows a 

 tube of quite pure water a meter long and G mm. in diameter having a re- 

 sistance of 10,000,000 ohms to be determined to one in a thousand or even 

 one in ten thousand. The current passing through the water is very small, 

 being at least 500 times less than that required when the bridge is used 

 in the ordinary way. Hence polarization scarcely enters at all." We 

 reproduce two of his figures below (Fig. 5). In these two methods the 

 concentric circles are the coils of the electrodynamometer. Either one 

 is the fixed coil and the other is the hanging coil. Oblong figures are the 

 inductances and when near each other are mutual inductances. A pair 

 of cross lines is a condenser. This quotation is given in full because 

 of its bearing on the present work and to aid in counteracting the general 

 notion that the telephone is good enough. 



There have been several objections to the use of the dynamometer but 

 none of these appear to have any foundation. It has been claimed 

 that the most serious objection arises from the fact that the two cur- 

 rents might sometimes be in quadrature. In this case, though the bridge 

 is unbalanced and a current is passing through the coils, there will be no 

 deflection of the galvanometer scale. Should such a possible condition 

 arise it could be eliminated by the introduction of a condenser to produce 

 a "lead" in the current. In all our experiments we have never experi- 

 enced this trouble and yet if it should arise we have a suitable means of 

 controlling it. 



