The transition is still continuous; but it now takes place between yellow and green. We begin with, greenish yellows, and pass by the smallest perceptible transitions to yellowish greens, and so to pure green. After passing green there is another change of direction; we now have a greenblue series. There is still another turningpoint after passing blue; the series which follows is bluered, passing from blue through violet and purple to red. The change of colour in the spectrum is throughout so continuous that it is not possible to fix the exact point at which these changes of direction begin. All that can be said is that they begin somewhere in the region of red, yellow, green, and blue, respectively. Since the change of direction occurs, it must occur somewhere. At the precise point of its occurrence, there must be a simple colourtone, such as pure red, pure yellow, pure green, or pure blue. For instance, pure yellow is the point of transition between the redyellows and the greenyellows, and pure red is the point of transition between the purples and the redyellows.

It may be well to note here a question of some psychological interest which has been much discussed. Is it right to say that a bluegreen is a combination of blue and green, or a redyellow a combination of red and yellow? It has become the fashion of late to say that such a colour as a bluegreen merely resembles blue and green, but does not contain them as constituent elements. The colour itself, it is maintained, is perfectly simple. Now it is natural for common sense to distinguish one bluegreen from another, by saying that there is more or less of blue in it, or more or less of green in it. It does not appear to the present writer that any cogent arguments have been brought forward to show that this point of view is untenable. A bluegreen may approach so very near to pure green as to be barely distinguishable from it, so that the casual observer would regard it as a pure green. It seems strange to say that such a bluegreen contains no green at all. What is probably in the mind of those who deny the combination is that a bluegreen cannot be simply defined as blue + green. The components by entering into so intimate a combination are modified in a peculiar way. This modification is a new element which may be regarded as simple. The experience of the combination of blue and green is a simple experience, and seems to be identical in kind with the experience of the combination of yellow and red, and other such pairs. But the components abstractly regarded are not the less* discernible as partaking of the nature of blue and green. Because there is something new and simple in the experience, we have no right to infer that there is no complexity in it. It must, however, be admitted that the question is not an easy one; and the balance of authority seems to be against the view which I am inclined to favour. But in any case it is most convenient to speak of such a colour as bluegreen as a combination of blue and green. If the student is not inclined to believe that the colour actually is complex he may interpret the statement that bluegreen is a combination of blue and green as merely meaning that on the one hand it resembles blue, and on the other resembles green.

* Of course they are not separable, but they are under appropriate conditions distinguishable. The respect in which blue and bluegreen are seen to resemble each other when compared is different from the respect in which green and bluegreen resemble each other when compared. This appears to me a sufficient reason for inferring complexity in the bluegreen,

So far we have only considered difference in colourtone, apart from difference in intensity and saturation; but all the colours of the spectrum may vary in either of these respects so as to form a continuous series. Each of them may be made more or less pale by an admixture of white light. If the general intensity of the illumination be increased or diminished while the spectrum is being examined, and if the increase or diminution is not too great, the result is that all the colours in the spectrum vary in brightness while remaining the same in colourtone. But the change in brightness is in general accompanied by a change in saturation. Increased brightness makes a colour paler, and decreased brightness makes it darker—causes it to be mixed with black. When the increase or decrease is made sufficiently great, the colourtones tend to disappear in mere whiteness or blackness, respectively. They may be mixed with white light, and also lowered or increased in intensity, so that both changes are combined. All the colourdifferences recognised in ordinary life may be accounted for in these various ways. They are constituted by differences in primary colour tone, in intensity, and in saturation. Pink and rosecolour are whitish reds; maroon is a dark red, i.e., a red so diminished in intensity as to be strongly infused with black. Olive is a dark green. We usually call a pale green or blue a light green or blue. The series of colourmodifications obtained by making a colourtone, such as blue, paler or darker is psychologically quite analogous to to such a series as that of the bluegreens. Here, too, the question of simplicity or complexity arises. Those who maintain that no two distinguishable parts of the bluegreen series have, qua sensations, any common element, but that they are all simple and independent colourqualities, must maintain the same for the blueblack series. They must maintain that a black in which only the artist's eye detects a tinge of blue has no element in common with pure black, or with black that has a barely appreciable tinge of green.

Intensity is by no means independent of colour. In the spectrum, the physical light is most intense in the region of red. But for our experience the yellow is distinctly the brightest colour. The blue is less bright than the red, but the difference is by no means in proportion to the difference in the intensity of the illumination.

It should be noted that the red of the spectrum is not pure red, but, as Hering pointed out, is tinged with yellow.