II 



but it t the copper lead firmly round the bottom of 



the alloy at b, instead of 46* being indicated at the galvano- 



. we cannot get inure than .ji-4-'- How is tl 



ucmbcring that the conduct! . >n is almost four 



x , and th he more we can bring the 



molecules of the two wctaU into friction the greater KM K 

 will be obtained, let us cast a couple in the following form. 



Take again a piece of 20 gauge sheet Iron, cut it to 



>.), instead of into the narrow strip ; roll this 



into a coil as before, and we have, after casting the /A upon 



couple in this form (Fig. 9). Lay the copper lead 



Fiy.Cf. 



(a) firmly on, and we have as a result about 40 galvr. not 

 equal to Fig. 4. Now to understand the reason of this, we 

 must seek an explanation in some other direction. 



rc turning to some older forms of Thermo-couples, 

 ution a few results from others, which we w ill call 

 Class 2. 



Take some Iron rings s inch diameter and p inch deep 



