PHOTOCHEMICAL RESEARCHES. 67 



foreign gas. The retarding action of oxygen upon the mixture is the most 

 marked ; the addition of one per cent, of this gas to the chlorine and hydro- 

 gen mixture reduced the amount of action to -g^gth ; and the presence of a 

 mere trace of this gas (probably not more than loVoth per cent.) diminished 

 the action to one half of the normal amount. Excess of either chlorine or 

 hydrogen was found to act in the same manner, but not to such a remarkable 

 extent. This retarding action of oxygen accounts for the very great length 

 of time which it is necessary to lead the gas through the apparatus before 

 "the maximum action is attained. 



The diminution of the sensibility of the chlorine and hydrogen mixture 

 when foreign gases are present, gives a very accurate measure of the cata- 

 lytic action effected by such gases. 



The simple relations which exist between the amount of hydrochloric acid 

 formed by the action of the light and the time of exposure, and amount of 

 light, were first observed by Draper. We have confirmed his results in this 

 respect, and have proved that both laws hold good for diffuse solar as well 

 as for lamp-light. The relations are the following : — 



1. The amount of chemical action effected by a constant source of light 

 is directly proportional to the time of exposure. 



2. The amount of chemical action effected by the light in equal times, is 

 directly proportional to the amount of light. 



(These laws are of course only applicable when the phaenomena of induction 

 have been fully eliminated.) A third relation which we have established is, 

 that the amount of chemical action varies inversely as the square of the di- 

 stance between the source of light and the sensitive mixture. 



The experimental diflSculties which accompany the examination of the 

 relations existing between the amount of action and the mass of the gas, are 

 of so peculiar and considerable a nature, that although we have been occu- 

 pied for more than a month upon this branch of the subject, we have not as 

 yet succeeded in arriving at the law which regulates the action. We have, 

 however, already proved that after the light has passed through a certain 

 depth of the gas, it is no longer capable of causing a combination to take 

 place ; and we have further proved that the depth at which the light ceases 

 to act upon the mixture is very different for light from various sources. 

 Differences in this respect have not only been found in light from different 

 sources, but the diffuse solar light reflected from a perfectly cloudless sky is 

 found to differ, not only in the quantitj, but also in the quality of the chemi- 

 cal rays according to the sun's altitude. These interesting observations are 

 not complete, but the results as yet obtained give promise of further import- 

 ant I'elations being established between the nature and amount of the 

 chemical rays falling upon the earth's surface at various periods of the day. 



Reduction of the Chemical Action of Light to an Absolute Measure. 

 The difficulty of obtaining any constant terrestrial source of light threw 

 great obstacles in the way of reducing the chemical action of light to an 

 absolute measure. The normal source of light which we have chosen for 

 the calibration of our instrument (fig. 1 ), is a flame of pure carbonic oxide gas 

 streaming from a large (3 millims in diameter) platinum burner, and issuing 

 under a constant pressure of half a millimetre of water. By measuring the 

 volumes of gas burned by different-sized flames and observing the chemical 

 action produced, it was found that even with the homogenous flame of carbonic 

 oxide, the chemical action increases in a greater ratio than the volume of 

 gas burned. This relation between the action produced and the volume of 



f2 



