116 



Mr. H. Tomlinson on Thermal Conductivity [Feb. 21, 



boxes, that when some two hours or so had elapsed, the rise of tem- 

 perature of the water in each box seemed as measured by an ordinary 

 thermometer, marked off: in degrees centigrade, to be exactly the same, 

 namely, 10° C. 



But as the same experiment had shown that the pile in the box on 

 the left of the figure was very slightly more powerful than the other, 

 the water in the box on the right was diminished very carefully by 

 means of a small siphon, formed of india-rubber tubing, of very small 

 bore, which was kept closed by a pinch-cock at one end, the tube 

 passing through holes cut in the large box, through the sawdust, and 

 through a small hole cut in the lid of the small box, so as to dip 

 into the water. The little siphon was always kept charged, and, by 

 means of the pinch-cock, any desired small quantity of water could be 

 extracted from the box. 



By this means it was found easy to make one pile so neutralise the 

 other that a very slight use (if any is necessary) of the adjusting 

 magnet was required, even when the magnet was near its most sensitive 

 position with respect to the galvanometer needle. 



On sending a current through the coil M, it was in a minute or so 

 seen that the conductivity of the hard steel was diminished by the 

 longitudinal magnetisation, and the amount of diminution was roughly 

 determined by finding the amount of water necessary to be extracted 

 from the box to again bring the light to the same slow rate of motion 

 (about 2 divisions per minute) which it had before passing the 

 magnetising current, and comparing this with the original amount 

 of water in the box. The decrease of flow did not amount to 1 per 

 cent, of the whole, even with the strongest battery used (current 

 shown by the tangent galvanometer, 26°). 



This plan was found to be highly successful for determining whether 

 there was a diminution or not, but not equally so for determining the 

 amount of decrease, owing to the great difficulty in this method of 

 quickly making the light move at the same rate after magnetisation as 

 before. Another method was therefore tried, which promises to give 

 better results. 



The figure will perhaps best explain the plan now adopted for 

 making one pile completely neutralise the effect of the other. 

 Pi, P 2 are the two piles. 

 g the galvanometer. 



S the resistance of a shunt placed between the points AB, which 

 could be varied at pleasure. 



R the resistance of circuit AP X B, including the resistance of the 

 pile. 



The same quantity of water was now put into each box, and the 

 pile Pi being slightly more powerful than P 2 , the shunt S was adjusted 

 until there was little or no motion of the light on the scale. 



