this view can be accepted as of general application. A beautiful illustration of the theory of warning colours is given by Belt in his “Naturalist in Nicaragua,” p. 321. All the frogs found in the woods round St. Domingo are, with one exception, protectively coloured; they are of nocturnal habits, and are devoured by snakes and birds. The exception was a species of bright red and blue colours, which hopped about by day and made no attempt at concealment. From these facts Mr. Belt concluded that this species was inedible, and on trying the experiment with ducks and fowls this was found to be the case. R.M.] [134] See the essay “Über den Einfluss der Isolirung auf die Artbilding.” Leipzig, 1872, p. 22. [135] [See also preceding note 133, p. 294. R.M.] [136] [Eng. ed. The habit of hiding by day occurs also in those caterpillars which resemble the bark of their food-trees. Thus _Catocala Sponsa_ and _Promissa_ conceal themselves by day in crevices of the bark, and are, under these circumstances, only found with difficulty. Dr. Fritz Müller also writes to me that in Brazil the caterpillars of _Papilio Evander_ rest in this manner in large numbers, crowded together into dense masses, on the trunks of the orange-trees, which they resemble in colour.] [137] “Über den Einfluss der Isolirung auf die Artbildung.” Leipzig, 1872, p. 21. [138] I am unfortunately not able to give exact numbers showing the relative proportions of the different forms, since I have never bred _S. Convolvuli_ from eggs, nor _C. Elpenor_ in sufficient numbers. [139] [With reference to _C. Porcellus_, see note 71, p. 188. R.M.] [140] [In the class of cases treated of in the foregoing portions of this essay, the external conditions remain unaltered during the lifetime of the caterpillar, but change of habit, and in some cases of colour, occurs when the insect has attained a size conceivable _à priori_, and are realized by observation, in which the environment itself may undergo change during the lifetime of the individual caterpillar. Thus, in the case of hibernating species, the colour which is adaptive to the autumnal colours of the foliage of their food-trees would not assimilate to that of the newly-opened leaves in the spring. I have already quoted (Proc. Zoo. Soc. 1873, p. 155) as instances of what may be called “seasonal adaptation,” the larvæ of _Geometra Papilionaria_, _Acidalia Degenararia_, and _Gnophos Obscurata_, and many more could be named. These species undergo a change of colour before or after hibernation, the change being always adaptive to the environment.