296 Intelligence and Miscellaneous Articles. 



with a sensitive Thomson's galvanometer. By dipping a hot iron 

 bar in the mercury box and observing the deflections the masses of 

 mercury could be so equalised that their water values were equal ; 

 the water value of the boxes could also be determined. If i is the 

 intensity of the current, the quantities of heat in the two boxes are 

 ai 2 +b i, where 2 hi is twice the Thomson effect for the strength of 

 the current i. By inverting the bars their irregularities were also 

 eliminated. 



The deflection of the galvanometer was observed from 30 to 30 

 seconds, every 20 minutes before, during, and after passing the 

 current, and the intensity of the latter was observed every minute. 



It was thus found that when the temperatures of the ends pro- 

 jecting from the iron boxes were 63°*5 and 42°*5, and the strengths 

 of the currents were between 0*3 and 0*5 units (C.Gr.S.) the Thom- 

 son effect was proportional to the latter. The heat developed in a 

 second by unit current, at a place with the mean temperature 53°, 

 whose ends differ by a temperature of 1° C.,is about E= 0*000011215 

 thermal units. At higher temperatures a petroleum vapour bath 

 was used instead of a steam bath, in which the ends of the rods 

 near the iron boxes were at 124*1 and 92*7. In this case the 

 action for the mean temperature 108°*4 was E= 0*000013403. 



On the assumption of Tait that the Thomson effect is proportional 

 to the absolute temperature (273+53) and (273 + 108*4) the latter 

 number should be 0*000013121. 



On heating in boiling petroleum the places near the iron boxes 

 were at the temperatures 263°*5 and 220°, and for the mean tempe- 

 rature 243 c *25 the effect was 0*00001786, while it should be 

 0*00001897 on Tait's hypothesis. The observations agree therefore 

 with this hypothesis. — Atti delta B. Ace. di Torino, 22, p. 48, 1886. 

 Beiblatter der PJiysik, vol. xi. p. 463. 



ON HYGROMETRIC SUBSTANCES. BY HENRI DUFOUR. 



The author investigated the behaviour of various hygrometric 

 substances. He denotes by a the ratio between the aqueous vapour 

 absorbed and the weight of dry substance (moisture =100); and 

 by /3 the coefficient of hygrometric expansion, that is, the total 

 expansion which a bar of unit length undergoes when it has taken 

 up the maximum aqueous vapour. 



Horn 10 miUim. in thickness a=0*10 /3 = 0*061 



Gelatine a=0*34 /3 = 0*108 



Goldbeaters-skin a=0*43 /3=0*060 



The last bodv is what he most recommends. — Arch. d. Gen. [3] 

 xvi. pp. 197-199, 1886. Beiblatter der PhysiJc, No. 7, 1887. 



