[ 24 ] 



II. The Sizes of the Water Particles producing the Coronal and 

 the Axial Colours of Cloudy Condensation. By C. Barus *. 



rf THE following table contains a summary of my data 

 JL for the sizes of water globules producing coloured 

 cloudy condensation either by the diffraction of diverging 

 rays, or by absorption of the strictly axial pencil. A brief 

 account of the general method of work will be found else- 

 where f. The nuclei used came from glowing charcoal, 

 though the corresponding result for sulphur nuclei will also 

 be instanced. In the table W denotes the relative number 

 of particles per cub. centim., N the absolute number, 1/N 1 ^ 3 

 the edge of a cube containing one particle, d the diameter of 

 the particle, all referred to centimetres. Experiment showed 

 that in successive adiabatic exhaustions between fixed pressures, 

 W = NolO n(1+bt)1 °zy, where n denotes the serial number of the 

 exhaustion, h is the coefficient of the time losses alone, y is 

 the ratio of the density of distribution of nuclei before and 

 after each exhaustion regarded as an adiabatic process, but 

 corrected for the heat evolved by the precipitation. 



In measuring the diameter, d, of the particles, the normal 

 (white-centred) coronas succeeding the 20th exhaustion were 

 used directly, by comparing their angular apertures, s, with 

 the coronas due to lycopodium particles under like conditions. 

 I assumed for the latter d = '0032 centim., while by trial 

 s =*45 centim., seeing that the radius of the goniometer falls 

 out. Hence numerically d= "00144/.9. In these experiments 

 the source of light and the special goniometer were each 

 250 centim. from the receiver of water globules. 



The diameters of the earlier coronal particles, i. e., those 

 corresponding to smaller values of n } were computed from 

 N', as this was the immediate result of the experiments. If 

 when r<=0, d = T>, then d = Y)W~^ and D = '00144 W^js. 

 In this way 1 found in case of the globe : — 



D = -000263 centim. for punk nuclei, D = '00035 centim. 

 for sulphur nuclei, and in the drum, D = '00036 centim. for 

 punk nuclei. 



These data give the number of nuclei arbitrarily put in 

 at the outset, but they seem to have an upper limit. The 

 datum for the drum was reduced from those of the globe by 

 the aid of the accompanying coronas. 



* Communicated by the Author. 



t American Journ. of Science, [4] xiii. pp. 81-94 (1902) ; Science, xv. 

 pp. 175-178 (1902). For allied experiments of. American Journ. of 

 Science, [4] xii. 1901 ; " Experiments with Ionized Air," Smithsonian 

 Contributions, pp. 1-95 (1901). 



