With more plausibility it might be held that the universal discharge of colour in these cave-faunas is due, not to the presence, but to the absence of selection--i. e. to the cessation of selection, or panmixia. But against this--at all events as a full or general explanation--lie the following facts. First, in the case of Proteus--which has often been kept for the purposes of exhibition &c., in tanks--the skin becomes dark when the animal is removed from the cave and kept in the light. Secondly, deep-sea faunas, though as much exposed as the cave-faunas, to the condition of darkness, are not by any means invariably colourless. On the contrary, they frequently present brilliant colouration. Thus it is evident that if panmixia be suggested in explanation of the discharge of colouring in cave-faunas, the continuance of colour in deep-sea faunas appears to show the explanation insufficient. Thirdly, according to my view of the action of panmixia as previously explained, no _total_ discharge of colouration is likely to be caused by such action alone. At most the bleaching as a result of the mere withdrawal of selection would proceed only to some comparatively small extent. Fourthly, Mr. Packard in the elaborate _Memoir on Cave Fauna_, already alluded to, states that in some of the cases the phenomena of bleaching appear to have been induced within very recent times--if not, indeed, within the limits of a single generation. Should the evidence in support of this opinion prove trustworthy, of course in itself it disposes of any suggestion either of the presence or the absence of natural selection as concerned in the process. Nevertheless, I myself think it inevitable that to some extent the cessation of selection must have helped in discharging the colour of cave faunas; although for the reasons now given it appears to me that the main causes of change must have been of that direct order which we understand by the term climatic. As regards dogs, the Rev. E. Everest found it impossible to breed Scotch setters in India true to their type. Even in the second generation no single young dog resembled its parents either in form or shape. "Their nostrils were more contracted, their noses more pointed, their size inferior, and their limbs more slender[111]." Similarly on the coast of New Guinea, Bosman says that imported breeds of dogs "alter strangely; their ears grow long and stiff like those of foxes, to which colour they also incline ... and in three or four broods their barking turns into a howl[112]." [111] _Variation_, &c. vol. i. p. 40. [112] _Variation_, &c. vol. i. p. 40.