376 Inquiries into the Laws of Dilatation of Solids, LiquidSf 



nient. We heated, in the first place, the tnh to a temperature 

 little inferior to that which we wished to obtain, and we after- 

 wards closed all t!ie apertures of the furnace. The e(|uilibrium 

 of heat tending tlien to be established throughout the wh.ole 

 heated masy, the temperature of the oil still rose some degrees, 

 an^, soon attained its maximum, where it became some time 

 stationary, and consequently easy to measure with precision. 

 It was then indicated by the horizontal thermometer, which was 

 sunk sufficiently in the oil, in order that the whole column of 

 mercury should be inserted in it: at the same instant we closed 

 by means of the blowpipe the fine point of the external part of 

 the air-tube, and we noted the barometrical height. This being 

 done, we withdrew the tube and carried it into a separate cham- 

 ber, the temperature of which was nearly invariable : we placed 

 it vertically, and in such a way that the point entered a mer- 

 curial bath perfectly dry. We broke this point ni the mercury, 

 and this liquid ascended until the equilibrium was estal)lished 

 with the external pressure : we left it in this situation a suffi- 

 cient time to give it precisely the temperature of the room, 

 which we knew with great precision by means of a very sensible 

 thermometer suspended beside the tube. When the ecpiilibiium 

 of temperature was perfectly estai)lished, we measured, by 

 means of a vertical scale fitted with an index, the height of the 

 column raised in the tube. We observed at the same time the 

 barometrical height, and the difference of these heights made us 

 acquainted with the elasticity of the cold air. We then with- 

 drew the tube, by taking all necessary precaution to retain the 

 mercury of which the column that had been raided was com- 

 posed. The tube and the mercury which it contained were 

 weighed: we afterwards weighed this same tube successively 

 empty and entirely full of mercury; by subtracting from the 

 result of this last weiglit those of the first two, we had the weight 

 of two equal volumes of mercury, the one with the volume of 

 hot air and the other with the volume of cold air ; and from 

 these v/eights we inferred the volumes themselves, which we 

 afterwards brought to what they would have been under the 

 same pressure, since we knew the elasticity of the cold air which 

 had been measured as we have indicated, and that of the warm 

 air which was equal to the pressure of the external air observed 

 at the instant of clrsing the tube. 



In order to judge more easily of the degree of confidence which 

 the results deserve to which we had been led, it will not be un- 

 availing to give some details relative to the precautions which we 

 took in every experiment. One of the greatest obstacles which 

 we meet with in this description of inquiries is owing to the dif- 

 ficulty of establishing a perfect uniformity of temperature jn a 



great 



