experienced by Air in rushing through small Apertures. 485 



principal spiral ; and had there been no cooling effect due to 

 rushing through the stop-cock when it was nearly closed, would 

 have been more checked when the stop-cock was wide open 

 than when it was nearly closed, as the same number of strokes 

 of the pump must have sent considerably more air through the 

 apparatus in one case than in the other. A cooling effect on the 

 whole, due to the nishing through the nearly closed stop-cock, 

 is thus indicated, if not satisfactorily proved. 



Other calorimetric experiments were made with the stop-cock 

 immersed in water in one glass jar, and the air fi'om it, conducted 

 by a \Tilcanized india-rubber tube, to flow through a small spiral 

 of block-tin pipe immersed in a second glass jar of equal capacity ; 

 and it was found that the water in the jar round the stop-cock 

 was cooled, whde that in the other, containing the exit spu-al, was 

 heated, during the working of the pump, with the stop-cock 

 nearly closed, and a pressure of about three atmospheres in the 

 principal spiral. The explanation of this curious result is clearly, 

 that the water round the stop-cock supplied a little heat to the 

 air in the first part of the rapids, where it has been cooled by 

 expansion and has not yet received all the heat of the friction, 

 and that the heat so obtained, along with the heat produced by 

 friction throughout the rapids, raises the temperatm'e of the air a 

 little above what it would have had if no heat had been gained 

 from without ; so that about the end of the rapids the air has a 

 temperatm-e a little above that of the surrounding water, and is 

 led, under the protection of the india-rubber tube, to the exit 

 spiral with a slightly elevated temperature. This is what wovdd 

 necessarily happen in any case of an arrangement such as that 

 described, if Mayer's hypothesis were strictly true ; but then the 

 quantity of heat emitted to the water in the second glass jar, from 

 the air in passing through the exit spiral, would be exactly equal 

 to that taken by conduction through the stop-cock from the water 

 in the first. In reality, according to the discrepancy from Mayer's 

 hypothesis, which the other experiments described in this com- 

 munication appear to establish, there must have been somewhat 

 more heat taken in by conduction through the stop-cock than 

 was emitted by it in flowing through the exit spiral ; but the 

 experiments were not of sufficient accuracy, and were affected by 

 too many disturbing circumstances, to allow this difference to be 

 tested. 



To obtain a decisive test of the discrepancy from Mayer's 

 hypothesis, indicated by the experiments which have been 

 described, and to obtain cither comparative or absolute determi- 

 nations of its amount for different temperatures, some alterations 

 in the apparatus, especially with regard to the narrow passage 



