Mr R. Adie on Thermo- Electrical Experiments. 353 



joint, together with a portion of the bismuth bar, was placed betwixt 

 the chops of a strong vice ; on pincing the vice, the bismuth was 

 crushed together, emitting a crackling sound like tin, when bent back- 

 wards and forwards ; there was no decided action on the galvanome- 

 ter. Repeating this experiment with another similar in the vice- 

 joint placed between two cards in the vice, in order to prevent the 

 mass of iron in the vice from quickly withdrawing any heat which 

 the violent crushing developed, at the moment of compression the 

 galvanometer made a swing of 10° in the same direction, as if the 

 joint had been very slightly heated. Had molecular change been 

 alone able to develop an electrical current, the galvanometer should 

 have in both cases moved, and, if I apprehend rightly, the extent of 

 the action should have been very great. 



In forming mercurial amalgams, abundant molecular change can 

 be produced without any perceptible variation in temperature. To 

 use mercury like a bar of metal in a thermo couple, a glass-tube, 

 open at one end, and fused at the other around a piece of platina wire, 

 was filled with mercury ; the platina wire was connected to one end of 

 the galvanometer wire, and a bar of bismuth, connected to the other 

 extremity, was made to dip into the mercury at the open end of the 

 tube. In this arrangement, the bismuth was slowly combining 

 with the mercury, yet there was no effect on the galvanometer so 

 long as care was taken to preserve equality of temperature through- 

 out the couple. The same was found for lead and tin. When 

 unequal temperatures existed, then the well-known thermo-electric 

 effects were noticed, where bismuth is the generating metal to mer- 

 cury, and mercury the generating element to lead or tin. From 

 some unseen cause, most probably connected with the circumstance 

 of one of the elements in these couples being a fluid, the density 

 of the amalorams form inor has no influence on the direction of the 

 electrical current developed ; for, when one equivalent of tin and 

 mercury are amalgamated, the specific gravity by experiment is 

 ir05, by calculation 12*15 ; and for one equivalent of bismuth and 

 mercury, the specific gravity by experiment is 12' 08, by calculation 

 12*55 :* the calculated specific gravity is on the supposition that the 

 metals unite without change of volume. Lead and mercury give 

 exactly the same density as calculation shews, but from the above 

 numbers it is seen, that the amalgam of tin undergoes a greater 

 change in volume than that of bismuth, yet the thermo-electric re- 

 lation of tin to mercury is nearly like that of lead, and quite difierent 

 from bismuth. 



I have formerly mentioned cases of other metals where the influ- 

 ence of the molecular change is distinctly seen on the thermo- electri- 

 cal current generated. The two best examples of this kind may be 



* The proportion of mercury was tried for luilf an equivalent and for two 

 equivalents, which only varied in degree the changes noticed. 



