We also find series with breaks in them. Sometimes, in cells distributed at random, the egg has not developed and the provisions have remained untouched, but mildewed; sometimes, the larva has died before spinning its cocoon, or after spinning it. Lastly, there are parasites, such as the Unarmed Zonitis (Zonitis mutica, one of the Oil-beetles.—Translator's Note.) and the Spotted Sapyga (A Digger-wasp.—Translator's Note.), who interrupt the series by substituting themselves for the original occupant. All these disturbing factors make it necessary to examine a large number of nests of the Three-pronged Osmia, if we would obtain a definite result.

I have been studying the bramble-dwellers for seven or eight years and I could not say how many strings of cocoons have passed through my hands. During a recent winter, in view particularly of the distribution of the sexes, I collected some forty of this Osmia's nests, transferred their contents into glass tubes and made a careful summary of the sexes. I give some of my results. The figures start in their order from the bottom of the tunnel dug in the bramble and proceed upwards to the orifice. The figure 1 therefore denotes the first-born of the series, the oldest in date; the highest figure denotes the last-born. The letter M, placed under the corresponding figure, represents the male and the letter F the female sex.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 F F M F M F M M F F F F M F M

This is the longest series that I have ever been able to procure. It is also complete, inasmuch as it comprises the entire laying of the Osmia. My statement requires explaining, otherwise it would seem impossible to know whether a mother whose acts one has not watched, nay more, whom one has never seen, has or has not finished laying her eggs. The bramble-stump under consideration leaves a free space of nearly four inches above the continuous string of cocoons. Beyond it, at the actual orifice, is the terminal stopper, the thick plug which closes the entrance to the gallery. In this empty portion of the tunnel there is ample accommodation for numerous cocoons. The fact that the mother has not made use of it proves that her ovaries were exhausted; for it is exceedingly unlikely that she has abandoned first-rate lodgings to go laboriously digging a new gallery elsewhere and there continue her laying.

You may say that, if the unoccupied space marks the end of the laying, nothing tells us that the beginning is actually at the bottom of the cul-de-sac, at the other end of the tunnel. You may also say that the laying is done in shifts, separated by intervals of rest. The space left empty in the channel would mean that one of these shifts was finished and not that there were no more eggs ripe for hatching. In answer to these very plausible explanations, I will say that, the sum of my observations—and they have been extremely numerous—is that the total number of eggs laid not only by the Osmiae but by a host of other Bees fluctuates round about fifteen.

Besides, when we consider that the active life of these insects lasts hardly a month; when we remember that this period of activity is disturbed by dark, rainy or very windy days, during which all work is suspended; when lastly we ascertain, as I have done ad nauseam in the case of the Three-horned Osmia, the time required for building and victualling a cell, it becomes obvious that the total laying must be kept within narrow bounds and that the mother has no time to lose if she wishes to get fifteen cells satisfactorily built in three or four weeks interrupted by compulsory rests. I shall give some facts later which will dispel your doubts, if any remain.

I assume, therefore, that a number of eggs bordering on fifteen represents the entire family of an Osmia, as it does of many other Bees.

Let us consult some other complete series. Here are two:

1 2 3 4 5 6 7 8 9 10 11 12 13 F F M F M F M F F F F M F F M F F F M F F M F M

In both cases, the laying is taken as complete, for the same reasons as above.