We have presented the appearances of colour as exhibited by refraction, first, by means of subjective experiments; and we have so far arrived at a definite result, that we have been enabled to deduce the phenomena in question from the doctrine of semi-transparent mediums and double images. 300. In statements which have reference to nature, everything depends on ocular inspection, and these experiments are the more satisfactory as they may be easily and conveniently made. Every amateur can procure his apparatus without much trouble or cost, and if he is a tolerable adept in pasteboard contrivances, he may even prepare a great part of his machinery himself. A few plain surfaces, on which black, white, grey, and coloured objects may be exhibited alternately on a light and dark ground, are all that is necessary. The spectator fixes them before him, examines the appearances at the edge of the figures conveniently, and as long as he pleases; he retires to a greater distance, again approaches, and accurately observes the progressive states of the phenomena. 301. Besides this, the appearances may be observed with sufficient exactness through small prisms, which need not be of the purest glass. The other desirable requisites in these glass instruments will, however, be pointed out in the section which treats of the apparatus.[1] 302. A great advantage in these experiments, again, is, that they can be made at any hour of the day in any room, whatever aspect it may have. We have no need to wait for sunshine, which in general is not very propitious to northern observers. [1] This description of the apparatus was never given. OBJECTIVE EXPERIMENTS. 303. The objective experiments, on the contrary, necessarily require the sun-light which, even when it is to be had, may not always have the most desirable relation with the apparatus placed opposite to it. Sometimes the sun is too high, sometimes too low, and withal only a short time in the meridian of the best situated room. It changes its direction during the observation, the observer is forced to alter his own position and that of his apparatus, in consequence of which the experiments in many cases become uncertain. If the sun shines through the prism it exhibits all inequalities, lines, and bubbles in the glass, and thus the appearance is rendered confused, dim, and discoloured. 304. Yet both kinds of experiments must be investigated with equal accuracy. They appear to be opposed to each other, and yet are always parallel. What one order of experiments exhibits the other exhibits likewise, and yet each has its peculiar capabilities, by means of which certain effects of nature are made known to us in more than one way. 305.