256 Mr. W. Breimaml. 



Tables are added of half-hourly readings at Manchester, givi 

 al actinic effects for different seasons of the year, &c. 



e. Roscoe and Baxendell, " On the Relative Chemical Intensities of 

 direct Sunlight and diffuse Daylight at different Altitudes of the Sun." 

 in ' Boy. Soc. Proc.,' vol. 15, 1866-67, pp. 20-24. 



By "total daylight" is meant the whole resultant action of the 

 Sun and sky on paper exposed horizontally. 



By "diffuse daylight" is meant the same action when the Sun w 

 stopped out. 



The "direct sunlight" was taken as the difference between I 

 two ; it does not appear to have been observed directly. 



d. Roscoe, " On the Chemical Intensity of Total Daylight at Kew 

 and Para," in ' Phil. Trans.,' 1867, pp. 555-570. 



e. Roscoe and Thorpe, "On the Relation between the Sun's Alti- 

 tude and the Chemical Intensity of Total Daylight in a Cloud 

 Sky," in ' Phil. Trans.', 1870, pp. 309-316. 



2. My observations made at Dacca, in 1861-1866 (repeated 

 Milverton, in Somersetshire, during the last year), were made in 

 entire ignorance of the work of Sir H. Roscoe ; his results, therefore, 

 so far as they agree with mine, afford an independent support to mj 

 theory. My experiments have been directed largely to ascertaining 

 the laws of the distribution of the actinic power in the sky, and thus 

 the work of Sir H. Roscoe overlaps mine at particular points only. 

 So also Roscoe has taken numerous observations of the sky more 

 less clouded ; I take no observation except when the sky is clear, as 

 find even a very slight haze to produce large differences in the 

 measurements, and to bring into the numerical results complications 

 that 1 have not at present attempted to deal with. 



3. The method of measurement I adopted, is the darkening pro- 

 duced in sensitised photographic paper ; for this effect I accept 

 Roscoe's term of " the chemical action." My method of measurement 

 differs from that of Roscoe in one important point : I use strips cut 

 from one uniform sheet of ordinary photographic paper; all my 

 measurements are so far relative, and I obtain the same numerical 

 results (ratios) with any paper. I compare ultimately the effect of 

 the Sun and of a candle on this same paper. Roscoe, by preparing 

 special paper with definite proportions of nitrate of silver, Ac., 

 depends on thus reproducing paper of exactly the same sensitiveness. 

 I make each measurement numerically (as did Roscoe) by comparing 

 the shade produced with some standard blackness. 



4. I assume that in the burning of a stearine candle, the " chemical 

 action " is proportional to the material consumed. I have taken as 

 my unit (i) of measure of chemical action, the darkening produced 

 at a distance of 1 inch from the wick of the candle, when 100 grains 

 were consumed, which, in the candle I used in India, occupied about 



