6 INTRODUCTION. 
It was observed to me by a very good friend, " that as the seven 
tints were permanent, that was an evidence of tlieir sufficiency as 
primitives and their original derivation, and that water was perfect 
in itself, although it was found to consist of hydrogen and oxygen." 
The latter was once thought a pure element, and led to wrong con- 
clusions. The former being thought perfect, it will be evident, has 
also caused wrong conclusions; and as water is not necessary to be 
formed to produce one of its original parts, hydrogen or oxygen, so, 
it will be less necessary to mix yellow and red and yellow and blue* 
to make a yellow ! when yellow is originally and necessarily so, to 
form the very ingredients so unhappily combined, and leading into 
continued and self-proving errors. Thus water does not form hy- 
drogen and oxygen, but hydrogen and oxygen form water. Yellow, 
red, and blue, I presume, therefore, will more properly form the re- 
mainder of the seven prismatic tints, than either two or more of the 
seven will form a single primary colour. I would not have insisted 
so much on this subject, but that the present improving state of 
natural science seems to demand a concordance of the primitives of 
prismatic tints and substantial colours, which appear so much to 
depend on each other ; and it has been understood by many, from 
the time of sir Isaac Newton, that the prismatic tints may be imi- 
tated by what I would consider as simple primitives, viz. yellow, 
red, and blue. 
Now to me it seems that, if these are allowed to be the proper 
primitives of the prismatic tints or colours, there will be a perfect 
concordance on the most permanent footing. 
I should not have set about this work without a confidence of its 
utilit)', although I had not intended or expected to carry it quite so 
far. I therefore may not have arranged it so regularly as I wish, 
* It has been, by some, a long mistaken notion, that as yellow is placed between 
«range and green among the seven tints, it was a natural result from them, when the 
■contrary is so naturally and truly the case, as is proved by the very first hypothesis, 
seeing that yellow and red make orange, &c. 
