Producible by Attrition and Contact 0/ Metals. 5 



informs us of the many errors into which he was led by the action of the rheo- 

 phores, with which the communication between the metals experimented on 

 and the galvanometer coil was established, heterogeneous as these metals and 

 the wires must often be. He had hoped that the specific action of the rheo- 

 phores " on the thermo-electric elements could be neglected, and that therefore 

 the observed deviation might be assumed to result only from the temperature, 

 or from the friction of the thermo-electric couple. A course of rather tedious 

 experiments have shown me that this supposition is most erroneous, and utterly 

 deceptive, when applied to refined investigations and highly susceptible instru- 

 ments. A multitude of contradictory and incoherent facts accumulated them- 

 selves like a chaos, before I arrived at this source of error." " A voluminous 

 journal of attempts was multiplied by the unexpected thermo-electric influence 

 of the rheophores destroying its value." 



For my own part, I can fully appreciate the difficulties and incoherent results 

 which obstructed Professor Erman's progress in this obscure investigation. The 

 same, along with many others, had bewildered me ; and to such an extent that 

 I more than once contemplated the abandonment of the subject. 



The adoption of new methods and the performance of multiplied experi- 

 ments have given me some confidence in the statements which here follow, 

 although they by no means correspond with results that have been published 

 by other investigators. 



It is convenient, in the first place, to state the conclusions to which my 

 experiments have led me, and which I have called Laws for want of a more 

 appropriate term. But I am aware that they should be the proved expression 

 of a far greater number of accordant facts than I have been able to elicit before 

 they can in strictness be admitted as the general laws which determine the 

 deflections of the magnetic needle when under the influence developed by the 

 attrition or therm o-contact of dissimilar metals. 



Law I. The agent which causes the deflection of the galvanometer needle 

 may be brought into action either by contact with each other at unequal tem- 

 peratures of certain metals, metallic ores, or some forms of carbon ; or by their 

 attrition against each other, whether at equal or unequal temperatures. 



Law II. When two different metals,* and sometimes two separated masses 



* To avoid perplexity I omit ores and carbon, although they are subject to many of the law;^. 



