We considered the most important difficulty in the way of instinct as hereditary habit, namely, the structure and instincts of neuter insects; these are very unlike those of their parents, and cannot apparently be transmitted to offspring by individuals of the previous generation, in whom such structure and instincts appeared, inasmuch as these creatures are sterile. I do not say that the difficulty is wholly removed, inasmuch as some obscurity must be admitted to remain as to the manner in which the structure of the larva is aborted; this obscurity is likely to remain till we know more of the early history of civilisation among bees than I can find that we know at present; but I believe the difficulty was reduced to such proportions as to make it little likely to be felt in comparison with that of attributing instinct to any other cause than inherited habit, or inherited habit modified by changed conditions. We then inquired what was the great principle underlying variation, and answered, with Lamarck, that it must be “sense of need;” and though not without being haunted by suspicion of a vicious circle, and also well aware that we were not much nearer the origin of life than when we started, we still concluded that here was the truest origin of species, and hence of genera; and that the accumulation of variations, which in time amounted to specific and generic differences, was due to intelligence and memory on the part of the creature varying, rather than to the operation of what Mr. Darwin has called “natural selection.” At the same time we admitted that the course of nature is very much as Mr. Darwin has represented it, in this respect, in so far as that there is a struggle for existence, and that the weaker must go to the wall. But we denied that this part of the course of nature would lead to much, if any, accumulation of variation, unless the variation was directed mainly by intelligent sense of need, with continued personality and memory. We conclude, therefore, that the small, structureless, impregnate ovum from which we have each one of us sprung, has a potential recollection of all that has happened to each one of its ancestors prior to the period at which any such ancestor has issued from the bodies of its progenitors—provided, that is to say, a sufficiently deep, or sufficiently often-repeated, impression has been made to admit of its being remembered at all. Each step of normal development will lead the impregnate ovum up to, and remind it of, its next ordinary course of action, in the same way as we, when we recite a well-known passage, are led up to each successive sentence by the sentence which has immediately preceded it.