16 RATIO OF SPr:CIFIC HEATS. 



with the bluckeiied strip so obtained. Tlie high al)sor|)tive power of platinum 

 black led us to anticipate an increased teni[>erature difference between the bolome- 

 ter strip and the air, since this difference, due to the influx of radiant heat from the 

 walls of the receiver, would more ai)preciably affect the black strip than the bright 

 strip of the usual experiments. Indeed the temperature differences severally en- 

 countered should be in the ratio of the absorptions in the two cases. We made a 

 special measurement of the I'elative absorption of bright platinum and [ilatiniim cov- 

 ered ill rill' iiiaiiner in question, by exposing two surface bolometers, one of which 

 contained the blight platinum foil and the other the platinum-l)lacked foil, to the 

 radiations of warm water in a copper vessel.' These investigations showed that plat- 

 inum black absorbs the radiations from acopper surface for the temperatui'e inter- 

 val of 100° C. to 30° C, nearly fifteen times more powerfully than i)right platinum. 

 Hence as the change of «, due to radiation, is now a small quantity, it suflBces to 

 deduct the « value computed by equation (1) for a blackened bolometer strip from 

 the value obtained under othei'wisc like conditions but with the bright bolometer 

 strip, to divide this difference by 14, and eventually to add the correction so found 

 to all H values corresponding to uncovered bright j)latiiium foil. This eoirection is 

 in fact .so small that we deemed it sufficient to determine it foi- air onlv, and to 

 apply it without further change to the values for the other gases. 



IV. — Observations. 

 a. AIR. 



We will premiss the present section by communicating a few tables taken at 

 random from our JKiiriial of observation.s, in ordei' to afford an iiisii^dit into the 

 general charaeter aii<l the accuracy of our results. 



As an example showing the consistency of the results in relation to velocity 

 of efflux, we may cite the pressure measurements for air made on the 12th of June, 

 ISJ)."^. For an iiiteival of expansion ('VEinspringzeit") of 12 seconds for the case 

 of the smallest, and of 2 .seconds for the largest efflux opening, the data contained 

 in the tw(» follow ing tables were respectively obtained. The first column in each 

 shows the numbers read oft" on the scale of the sulj>huric-acid manometer' directly. 

 In the second column -\- shows that the pressure was too large, — that it was too 

 small, while ± denotes a correct pressure value. The third column contains data 

 for the position of the sliding contact on the Wheatstone bridge, and the fourth 



' U'c wish in this place again to thank Mr. Kurll).uim for his active participation in these 

 preliminary experiments. 



' '{'he numbers refer to an arbitrary scale, and do not show the height of the column of acid 

 at once. To find the pressure in centimetres of Hi;S04, they must be increased by about 26. 

 C/. data for comparison in table, pages 20, 22. 



