the Galvanic Current through Iron. 141 



netized iron, would be at once accessible to observation, while 

 the other, resistance of the unmagnetic, would have an im- 

 portant theoretic signification, inasmuch as it alone is compa- 

 rable with the quantities which in other metals we bluntly call 

 the resistance. 



Assuming, then, that these two quantities indeed differ (and 

 experience shows that this is the case), yet no way based on 

 special conclusions can be imagined in which we could decide 

 which is the greater. In such cases considerations having for 

 their starting-point the principle of the conservation of force have 

 recently been frequently applied with success. Taking the 

 same course, I place first of all a general principle which 

 results therefrom, and which hitherto, so far as is known to 

 me, has been expressed only for special cases. It is : — 



No force can of itself bring in conditions more favourable for 

 its own action than those which it meets with. 



This needs no explanation; even the expression " of itself " 

 has become universally familiar since it was introduced by 

 Clausius*. The proposition in which he first made use of it, 

 " Heat can never of itself pass over from a colder into a hotter 

 body," is itself a special case of the above principle. Another 

 is Lenz's law which determines the direction of the induced 

 current. Further, here belong the facts that in solids the 

 cubic coefficient of compression diminishes as the pressure in- 

 creases, that the specific heat and the galvanic resistance of 

 solid bodies increase with the temperature, &c. 



For the present case we may conclude from the above prin- 

 ciple that the resistance of iron as observed by us when a cur- 

 rent is conducted through it is greater than that ideal value. 

 Thence, in the next place, it follows that circularly magnetized 

 possesses a greater resistance than unmagnetic iron, at least 

 if the amount of the circular magnetization does not exceed 

 that which the current itself could produce. Evidently, how- 

 ever, we may drop this limitation ; for if the circular magne- 

 tization be greater than the current itself can generate, yet 

 there is always another intensity of current possessing this 

 property. For this latter, therefore, the above proposition 

 holds good. But the resistance of an iron wire possessing a 

 definite magnetic state] is independent of the intensity of the 

 current ; consequently that principle is valid for any amount 

 of circular magnetization. 



It follows, further, that the resistance must be lessened by 

 feeble longitudinal magnetizings. For, according to the above, 

 we may assume that circularly magnetic iron conducts the 



* Die mechanische Warintheorie, i. p. 81. 



t That this addition is necessary, will be shown in § 8. 



