254 



Mr. W. Brennand. 

 Fio. 1. 



laterally, a fresh portion of it is brought under the longitudinal 

 and the observation can be immediately repeated, several times if 

 desired. 



8. By comparing the darkening produced in the paper in 

 graphs 5 and 7, we easily show that we have to expose the paper fot 

 times as long to produce the effect caused by diminishing the distance 

 one half ; and that a light of intensity 4 acting for 1 second has the 

 same effect as a light of intensity 1 acting for 4 seconds. This 

 think might have been assumed ; Bnnsen and Roscoe, in their pa{ 

 (1863) above cited, have, however, taken great pains to prove it. 



9. My early experiments were designed to test the total effect 

 the sky and Sun for photographic purposes. I have always experi- 

 mented mainly by exposing the paper at right angles to the Sun's raj 

 Roscoe on the other hand, exposes his paper in a horizontal plane. 

 It will be seen below, that theoretic considerations have led me to 

 another method of observation, which gives directly the measure of 

 effect really desired, and does not require a clear heavens down to 

 horizon on all sides (the Octant Actinometer). I give as a firsi 

 example of my experiments the following table (A). The observatic 



