4 Jo Experiments to determine tie Denftty of the Earth. 



^weights were much warmer than the cafe, they would warm that fide which was next to 

 them, and produce an afcending current of air, which v/ould be accompanied by a defcend- 

 ing current towards the oppofite fide, and two horizontal currents near the top and bot- 

 tom ; the latter of which would impel the ball towards the weight. Though it feemed 

 fcarcely probable that the weights fhould happen to be wai^mer than the cafe, and that in 

 a fufficicnt degree to afford any perceptible confequence, our author neverthelefs deter- 

 mined to fubmit the matter to examination. He therefore placed two lamps under the 

 weights while in the mid-way pofition, and placed a thermometer with its bail clofe to the 

 outfide of the cafe, near that part which one of the weights would approach when brought 

 into the pofition to exert its attraftion upon the ball. The door of the room was then 

 fliut, and fome time afterwards the weights were moved to this lad- mentioned pofition. 

 At firft the arm was drawn afide only in its ufual manner; but in half an hour the effedt 

 was fo much increafed that the arm was drawn fourteen divifions afide inftead of about 

 three, as it would ocherwife have been ; and the thermometer was raifed near i|- °, namely 

 from 61° to 624 °. On opening the door, the weights were found to be no more heated 

 than juft to prevent their feehng cool to the hand. 



From the great effeft of a difference of temperature it became dcfirable to afcertain the 

 temperature of the weight itfelf, and of the air near the cafe, in the fubfequent experiments. 

 A fmall thermometer was therefore inferted in one of the balls, and another placed clofe to 

 the cafe. A number of experiments were then made, in which the differences of tem- 

 perature between the weights and the external air were obferved ; the weights being in 

 fome inRances heated by a lamp, and in others cooled by means of ice. It was found 

 that a higher temperature.in the weight occafioned a greater deviation or apparent attrac- 

 tion of the ball j and, on the contrary, that a lower temperature caufed the ball to be 

 driven by a current in the oppofite dire£l:ion. 



After this minute and fcientific inveftigation of the caufes which are capable of pro- 

 ducing error in the refults, Mr. Cavendifli proceeded to make his fundamental experiments. 

 Thefe are tabulated at length in the memoir itfelf, and followed by a detail of the method 

 of computing the denfity of the earth. The firft thing to be done is to determine from 

 the time of a vibration, what force is required to draw the arm afide ; and the next, to 

 find the proportion which the attraftion of the weight upon the ball bears to that of the 

 earth upon the fame ball, fuppofing the ball to be placed in the middle of the cafe : and from 

 thefe refults (fince the diameter and fpecific gravity of the weights and balls compared with 

 water, and the diameter of the earth, are alfo known ; together with the fads that the 

 quantities of matter are as the attraiftive forces at like diftances, and vary in the inverfc 

 duplicate ratio of the diftance) the mean denfity of the earth itfelf may alfo be found. 

 Several corrections requifite to be made in this denfity are then ftated ; namely, on ac- 

 count of the refiftance of the arm to motion, the attraftion of the weights on the arm, 

 and on the remote ball; the attra£lion of the copper rods, and of the cafe itfelf; and for 

 the alteration of the attraction of the weights upon the balls according to the pofition of 

 the arm ; which laft is of more Cgnification than all the reft. For all which I mull, for 

 the fake of brevity, refer to the TranfaCtions. 



57.'* 



