1826.] ofPkotometrt/, with Remarki. 379 



fior would it follow on my principle, that any invariable or neces- 

 sary proportion subsists between the heating and illuminating 

 power ; and in some cases the contrary is decisively known to 

 be the case : hence when the term photometer is applied to 

 instruments or apparatus on the principle of comparing the illu- 

 minating intensities ; in point of fact, a totally different meaning 

 is attached to it. In other words, there is a want of definition 

 as to what it is that we intend to measure. 



The quantity of light, or its density, may, in certain cases, be 

 calculated upon optical principles ; but it does not necessarily 

 follow that this bears a proportion to the heating effect. Still 

 less that it is proportional to the illuminating effect on our eyes. 



If this last be the point to be measured by a photometer, it is 

 evident that the standard ultimately referred to can be no other 

 than the human eye. 



When two hghls of different intensity are simultaneously pre- 

 sented to the eye, we are unable, by the greatest degree of expe- 

 rience, to determine with any sort of numerical accuracy the 

 ratio of the illuminating effects ; but there are cases in which, as 

 we have already seen, by making the lights undergo certain 

 modifications, an estimate may be formed. Such are the methods 

 proposed for deducing the ratio from the means employed to 

 equatize the lights ; and of this equalization (at least within 

 certain limits), the eye can very sufficiently judge. Another 

 such principle may be that of diminishing the light till it actually 

 vanishes, or ceases to produce any impression. The different 

 degrees of attenuation necessary may give the ratio of the inten- 

 sities in different cases. The degree of accuracy to which any 

 such determinations may be carried will depend entirely on the 

 accuracy with which the eye can judge of the equalization of 

 the lights, or the point of evanescence ; but the object to be 

 measured is an impression made upon the same organ. It is 

 consequently to be remembered in all such researches that the 

 same uncertainty attaches to the thing to be measured, as to the 

 standard by which the measure is attempted. 



Upon the consideration of the inherent difficulties and uncer- 

 tainties of the thermometrical method, even in its most unobjec- 

 tionable form, it seems now to be admitted that we can only 

 look for accurate comparisons on some modification of the prin- 

 ciple of equalization of light ; but as yet no method combinino* 

 simplicity and accuracy of application with soundness of prin- 

 ciple has been proposed. 



With respect to the general principle of equalization, many 

 methods by which it may be effected will obviously occur. The 

 methods hitherto alluded to all go upon the plan of altering the 

 distance of one of the lights, but in some cases other principles 

 may be resorted to ; for example, with gas lights it is easy to 

 alter the intensity of one by regulating the supply. This method 



