In the first place, the child is peculiarly susceptible to the appeals of change, novelty, chance, or happy suggestion; and often the failure to respond to a stimulus is due to distraction or to discomfort rather than to lack of intrinsic interest. Again, fatigue is a matter of considerable importance. In respect to fatigue, I should say that the first signs of restlessness, or arbitrary loss of interest, in a series of stimulations, is sufficient warning, and all attempts at further experimenting should cease. Often the child is in a state of indisposition, of trifling nervous irritability, etc.; this should be detected beforehand, and then nothing should be undertaken. No series longer than three trials should be attempted without changing the child's position, resting its attention with a song, or a game, etc., and thus leading it fresh to its task again. Furthermore, no single stimulus, as a colour, should be twice repeated without a change to some other, since the child's eagerness or alertness is somewhat satisfied by the first effort, and a new thing is necessary to bring him out to full exercise again. After each effort or two the child should be given the object reached for to hold or play with for a moment; otherwise he grows to apprehend that the whole affair is a case of "Tantalus." In all these matters very much depends upon the knowledge and care of the experimenter, and his ability to keep the child in a normal condition of pleasurable muscular exercise throughout. In performing colour experiments, several requirements would appear to be necessary for exact results. Should not the colours chosen be equal in purity, intensity, lustre, illumination, etc.? In reference to these differences, I think only that degree of care need be exercised which good comparative judgment provides. Colours of about equal objective intensity, of no gloss, of relatively evident spectral purity, under constant illumination--this is all that is required. The variations due to the grosser factors I have mentioned--such as condition of attention, physical unrest, disturbing noises, sights, etc.--are of greater influence than any of these more recondite variations in the stimulus. Intensity and lustre, however, are certainly important. It is possible, by carefully choosing a room of pretty constant daylight illumination, and setting the experiments at the same hour each day, to secure a regular degree of brightness if the colours themselves are equally bright; and lustre may be ruled out by using coloured wools or blotting-papers. The papers used in the experiments given above were coloured blotting-papers. The omission of yellow is due to the absence, in the neighbourhood, of a satisfactory yellow paper. The method now described may be further illustrated by the following experiments on the use of the hands by the young child.