13 
Blue, yellow, and red, omitting white as neuter, we find, are liis 
three first colours ; and in the next range, we only want indigo to 
begin or terminate the series after the red and orange, between the 
yellow and red, to comj)lete tlie seven prismatic tints in tiie second 
series ; which, if we observe the joining of tlie red and IjIlic by being 
viewed at a certain distance, will seem very nearly to be the real 
case: indeed, a proper softening of one into the other would make 
it too nice for our discernment ; although, if the red is pure, it 
ought to be only violet, which in a bad light will ap])ear as dull as 
common indigo, as a deep purple is dulled by the yellower rays of 
candle-light and lilac, and all purples ajjpear brightest by day-light. 
The approach or nearer a})pearance of the two revolutions will also 
in this instance be found to form green, where the white took place 
before, as the yellow and blue join to form it ; and so it is with the 
other rings or revolutions which approach nearer and nearer, next 
losing the yellow as it were by mixing, and are both duller and paler 
from want of that opposition to stronger light or darkness, which 
probably towards the end became obscure, finishing in perfect white- 
ness, or light reflected on the face of the glass. The middle tint, if 
1 may so call it, at g, would not allow of a dark blue margin, as will 
be understood more perfect! \' hereafter. 
Having therefore produced the order of his tints, which without 
more instruction than his words we are enabled to do very perfectly, 
and as the tints come according to Nature, so we may conclude he 
was right, as may always be ])roved by the order of the colours ; for 
the prismatic tints, however irregular they may appear, or seem con- 
founded, are most perfectly and unerringly regular in their order: 
thus the rainbow, the colours in the flaws of stones, &c. follow the 
same rule, even to the coal called peacock coal, for the colours of 
which, see British Mineralogy, tab. 4Q. Thus if we look through the 
prism, at a window or a multitude of objects, or any varied number 
even of coloured things*, all will.be found to be reducible to certain 
♦ A print, with many objectSj coloured or uncoloured, may be convenient to examine. 
