THERMO-ELECTRIC GENERATORS. 41 



and throwing the flame of a central fire on the different 

 junctions of the elements within the rings, he was able to 

 obtain with a battery of somewhat small dimensions an elec- 

 tric current equally powerful with that from two Bunsen's 

 cells. But in this arrangement the electro-motive force of 

 the couple was in proportion to the difference between the 

 temperatures at the ends of each bar, and therefore, in order 

 .to obtain a powerful effect, it was necessary that the bars 

 should be rather long, and this led to a great increase of 

 the resistance, for these heated alloys have much more re- 

 sistance than would at first be supposed. The apparatus 

 also required a useless consumption of the heat, which, by 

 radiation and by contact with the air, escaped from the lateral 

 surfaces of the prism without having traversed the whole 

 length of the bar, and therefore did not yield the maximum 

 of the useful effect of which it is capable as regards its trans- 

 mutation into electricity. Again, it often happened that the 

 apparatus received an excess of heat, and the metal fused 

 round the parts of the iron plates to which it was soldered.* 

 For these reasons Clamond was unable to use this kind of 

 pile for the great effects he was seeking to realize. He had, 

 therefore, to completely alter the original arrangement of his 

 pile, and after numerous attempts he arrived at the very re- 

 markable arrangement we are going to describe. The 

 original Clamond piles are, nevertheless, still used at the 

 present day in the workshops of Goupil and others, where 

 they continue to work with regularity. 



" In my new arrangement," says Clamond, " I have striven to 

 avoid all the defects previously enumerated, and for this object I 

 have formed the apparatus of three entirely distinct parts. 



* " The use of flames or of radiating surfaces," says Clamond, "renders 

 the heating of the couples very difficult to regulate, and it is seldom adopted 

 except for apparatus of small dimensions, and when gas is the combustible. 

 Besides, the surfaces to be heated (which are represented by the sections of 

 the couples or of the polar projections with which they are sometimes pro- 

 vided) being very small, but little of the heat produced by the flame is taken 

 up, and the products of combustion pass off at a very high temperature." 



