10 Drs Christison and Turner on the Construction of 
■own phrase), the black objects remain equally dark, while the white objects 
become lighter and lighter, the contrast greater and greater, and the outlines 
of objects consequently more distinct. And hence it appears, that the me- 
thod of measuring light by the comparative intensity of shadows is an exact 
criterion of the relative fitness of different lights for such purposes as reading, 
writing, sewing, pencil-drawing, &c. Farther, the eye likewise judges of the 
forms of solid objects by the relative shades of their different surfaces. On this 
account, too, the method of Count Rumford measures correctly the value of va- 
rious lights. In fact, of the purposes served by artificial light, there are few in 
which the light does not act precisely according to the principle by which it af- 
fects his photometer. 
But there are some purposes served by lights generally, of their fitness for 
which it is no criterion. We cannot tell by it, for example, the relative fitness 
of several lights for distinguishing colours. Their fitness for that purpose will 
depend partly on their own colours, partly on those to be distinguished by 
them. It is certain, however, that colour generally is best appreciated with the 
whitest light. Now, an increase in the intensity of a light always tends to in- 
crease the purity of its whiteness, while an increase in quantity has no such 
effect ; and consequently it appears, that, for the special purpose of appre- 
ciating colour, variations in brilliancy or intensity are more important than va- 
riations in quantity. 
Some of the other more particular purposes of light, may perhaps be also 
similarly circumstanced. And hence the objection of Mr Ritchie, that the 
photometer is an incorrect measurer of light, because it takes into account the 
quantity only, is probably, to a certain degree, well founded. But still, as 
concerns the great uses of light, its indications are true. And it is worthy of 
notice, that, in regard to artificial light, the indications are true for all its pur- 
poses, so far at least as we have iioav examined them ; because artificial light 
never has been, and, whatever may be its intensity, never can be much used 
for distinguishing colours. 
The photometer of Count Rumford, it is well known, can only determine 
the relative illuminating power of two lights, not their actual quantity of 
light in reference to a fixed standard. It has no fixed scale of degrees, no zero 
to indicate total darkness, and no maximum to indicate the greatest illumina- 
tion. This is its great defect as a philosophical instrument, and its chief 
inconvenience in experiments like those about to be related. The experi- 
menters who have preceded us have endeavoured to remedy it, by using a wax 
or tallow candle of a given size for the standard light of comparison, and ex- 
pressing the results by corresponding numbers. A tallow-candle, so far as we 
can judge, is altogether inapplicable to the purpose ; for the colour of its flame 
is so dusky, that it cannot be easily compared even with a jet, much less with an 
Argand burner. A wax-candle, therefore, is the only standard of the kind 
which can be used at all. But, even then, it is impossible to compare together 
experiments made in that way with different candles and by different per- 
sons. Nay, it is exceedingly questionable whether much reliance can be put 
on such a standard, when the same candle is used by the same observer ; and, 
at all events, it is quite inapplicable to a train of experiments, each of which 
must be compared with all the rest. Such was the opinion of Count Rum- 
