Another, between the 9th and 27th of May, stocked six Snail-shells with a family of thirteen, including ten males and three females.

A third, between the 2nd and 29th of May colonized eleven Snail-shells, a prodigious task. This industrious one was also exceedingly prolific. She supplied me with a family of twenty-six, the largest which I have ever obtained from one Osmia. Well, this abnormal progeny consisted of twenty-five males and one female.

There is no need to go on, after this magnificent example, especially as the other series would all, without exception, give us the same result. Two facts are immediately obvious: the Osmia is able to reverse the order of her laying and to start with a more or less long series of males before producing any females. There is something better still; and this is the proposition which I was particularly anxious to prove: the female sex can be permuted with the male sex and can be permuted to the point of disappearing altogether. We see this especially in the third case, where the presence of a solitary female in a family of twenty-six is due to the somewhat larger diameter of the corresponding Snail-shell.

There would still remain the inverse permutation: to obtain only females and no males, or very few. The first permutation makes the second seem very probable, although I cannot as yet conceive a means of realizing it. The only condition which I can regulate is the dimensions of the home. When the rooms are small, the males abound and the females tend to disappear. With generous quarters, the converse would not take place. I should obtain females and afterwards an equal number of males, confined in small cells which, in case of need, would be bounded by numerous partitions. The factor of space does not enter into the question here. What artifice can we then employ to provoke this second permutation? So far, I can think of nothing that is worth attempting.

It is time to conclude. Leading a retired life, in the solitude of a village, having quite enough to do with patiently and obscurely ploughing my humble furrow, I know little about modern scientific views. In my young days I had a passionate longing for books and found it difficult to procure them; to-day, when I could almost have them if I wanted, I am ceasing to wish for them. It is what usually happens as life goes on. I do not therefore know what may have been done in the direction whither this study of the sexes has led me. If I am stating propositions that are really new or at least more comprehensive than the propositions already known, my words will perhaps sound heretical. No matter: as a simple translator of facts, I do not hesitate to make my statement, being fully persuaded that time will turn my heresy into orthodoxy. I will therefore recapitulate my conclusions.

Bees lay their eggs in series of first females and then males, when the two sexes are of different sizes and demand an unequal quantity of nourishment. When the two sexes are alike in size, as in the case of Latreille's Osmia, the same sequence may occur, but less regularly.

This dual arrangement disappears when the place chosen for the nest is not large enough to contain the entire laying. We then see broken layings, beginning with females and ending with males.

The egg, as it issues from the ovary, has not yet a fixed sex. The final impress that produces the sex is given at the moment of laying, or a little before.

So as to be able to give each larva the amount of space and food that suits it according as it is male or female, the mother can choose the sex of the egg which she is about to lay. To meet the conditions of the building, which is often the work of another or else a natural retreat that admits of little or no alteration, she lays either a male egg or a female egg AS SHE PLEASES. The distribution of the sexes depends upon herself. Should circumstances require it, the order of the laying can be reversed and begin with males; lastly, the entire laying can contain only one sex.

The same privilege is possessed by the predatory Hymenoptera, the Wasps, at least by those in whom the two sexes are of a different size and consequently require an amount of nourishment that is larger in the one case than in the other. The mother must know the sex of the egg which she is going to lay; she must be able to choose the sex of that egg so that each larva may obtain its proper portion of food.

Generally speaking, when the sexes are of different sizes, every insect that collects food and prepares or selects a dwelling for its offspring must be able to choose the sex of the egg in order to satisfy without mistake the conditions imposed upon it.

The question remains how this optional assessment of the sexes is effected. I know absolutely nothing about it. If I should ever learn anything about this delicate point, I shall owe it to some happy chance for which I must wait, or rather watch, patiently.

Then what explanation shall I give of the wonderful facts which I have set forth? Why, none, absolutely none. I do not explain facts, I relate them. Growing daily more sceptical of the interpretations suggested to me and more hesitating as to those which I myself may have to suggest, the more I observe and experiment, the more clearly I see rising out of the black mists of possibility an enormous note of interrogation.

Dear insects, my study of you has sustained me and continues to sustain me in my heaviest trials; I must take leave of you for to-day. The ranks are thinning around me and the long hopes have fled. Shall I be able to speak of you again? (This forms the closing paragraph of Volume 3 of the "Souvenirs entomologiques," of which the author lived to publish seven more volumes, containing over 2,500 pages and nearly 850,000 words.--Translator's Note.)


CHAPTER 13. THE GLOW-WORM.