In casting about for the means of exposing the pupæ he was about to experiment on, to a warm and equable temperature, Réaumur determined to conduct his first experiments in the Royal Conservatories, which were always carefully heated, and in which, as he with French _naïveté_ expresses it, "summer reigned in the depths of winter." In the month of January he carried thither a number of boxes containing pupæ of different species. The result was precisely what he had expected:--in the midst of a severe winter a number of butterflies appeared in his boxes, many of which would not naturally have made their appearance until the months of May, August, or even September; thus shortening the pupa state from four, seven, or eight months, to a fortnight, or to five, or to six weeks in different instances. Five or six days seemed to be equal to a month of the natural temperature. The butterflies thus developed were in no respect different from those which are brought into activity at the natural period. They were as active and perfect, as if their time and place of birth had been the green fields, instead of amidst the strange vegetation of these splendid Conservatories. Several of the mother insects deposited their eggs, accomplishing the last act of their existence as if summer had come, and died while the frosts and snow held all external nature yet in bondage. Not only, therefore, was the duration of the pupa state in these insects shortened, but their whole life was thus abridged by several months.
In November of the same year Réaumur recommenced his experiments, and again exposed a number of pupæ to the genial influence of these hot-houses. The result was the same. In the first week in December butterflies appeared, which would not, in natural circumstances, have been developed earlier than the May of the next year. There were some pupæ, in particular, whose development he watched with great interest. These pupæ belonged to a beautiful species of moth, which has two broods in a year; that is, it lays eggs in May which become butterflies in July, and then again lays eggs which become pupæ in August or September, but do not become butterflies until the following June. He was curious to see whether this second brood, instead of waiting for several months, would, like the first, disclose its butterflies in a considerably shorter time, now that it was exposed to the warmth of the Conservatory. Such actually proved to be the case; and thus two generations of these butterflies were obtained in one year. Alluding to the depredations of caterpillars, he quaintly remarks, "This certainly is not a secret which appears very profitable at present; but who can tell whether that which is useless to us to-day, may not possibly become of value to-morrow? Could we discover some new species of larvæ which would supply us with as good a silk as that of the silk-worm, and might be more easy to rear, but which only produced one generation in each year, and if it lived upon leaves which could be found all the year through, we might avail ourselves of this means of increasing the number of its broods." This remark deserves much consideration.