I was right in my conjecture: the Bee's intention was to break open the door. Straight away, without heeding the raggedness of the orifice, she settles down in the cell which I have opened for her. Time after time, she fetches honey and pollen, though the larder is already fully stocked. Lastly, she lays her egg in this cell which already contains an egg that is not hers, having done which she closes the broken aperture to the best of her ability. So this purveyor had neither the knowledge nor the power to bow to the inevitable. I had made it impossible for her to go on with her purveying, unless she first completed the unfinished cell substituted for her own. But she did not retreat before that impossible task. She accomplished her work, but in the absurdest way: by injuriously trespassing upon another's property, by continuing to store provisions in a cupboard already full to overflowing, by laying her egg in a cell in which the real owner had already laid and lastly by hurriedly closing an orifice that called for serious repairs. What better proof could be wished of the irresistible propensity which the insect obeys? Lastly, there are certain swift and consecutive actions so closely interlinked that the performance of the second demands a previous repetition of the first, even when this action has become useless. I have already described how the Yellow-winged Sphex (Cf. "Insect Life": chapters 6 to 9.--Translator's Note.) persists in descending into her burrow alone, after depositing at its edge the Cricket whom I maliciously at once remove. Her repeated discomfitures do not make her abandon the preliminary inspection of the home, an inspection which becomes quite useless when renewed for the tenth or twentieth time. The Mason-bee of the Walls shows us, under another form, a similar repetition of an act which is useless in itself, but which is the compulsory preface to the act that follows. When arriving with her provisions, the Bee performs a twofold operation of storing. First, she dives head foremost into the cell, to disgorge the contents of her crop; next, she comes out and at once goes in again backwards, to brush her abdomen and rub off the load of pollen. At the moment when the insect is about to enter the cell tail first, I push her aside gently with a straw. The second act is thus prevented. The Bee now begins the whole performance over again, that is to say, she once more dives head first to the bottom of the cell, though she has nothing left to disgorge, as her crop has just been emptied. When this is done, it is the belly's turn. I instantly push her aside again. The insect repeats its proceedings, still entering head first; I also repeat my touch of the straw. And this can go on as long as the observer pleases. Pushed aside at the moment when she is about to insert her abdomen into the cell, the Bee goes back to the opening and persists in going down head first to begin with.