The soil not occupied by those clumps of woody vegetation is almost bare and consists of fine, dry, very loose sand, which the wind heaps into little dunes wherever the stems and roots of the holm-oak interfere with its dissemination. The sides of these sand-dunes are generally very smooth, because of the extreme lightness of the materials, which slide down into the smallest depression and of their own accord restore the evenness of the surface. You need but push your finger into the sand and take it out again to bring about an immediate landslip which fills up the hole and restores things to their original condition without leaving a visible trace. But, at a certain depth, which varies according to the more or less recent date of the last rains, the sand retains a lingering dampness which keeps it in its place and gives it a consistency that enables it to have small excavations made in it without a subsequent collapse of walls and roof. A blazing sun, a gloriously blue sky, sandy slopes that yield without the least difficulty to the strokes of the Wasp’s rake, game galore for the grub’s food, a peaceful site hardly ever disturbed by the foot of man: all the good things are combined in this Bembex paradise. Let us watch the industrious insect at work. If the reader will sit with me under the umbrella, or consent to share my Rabbit-burrow, this is the sight which he is invited to behold, at the end of July: a Bembex (B. rostrata) arrives suddenly, I know not whence, and alights, without preliminary investigations or the least hesitation, at a spot which to my eyes differs in no respect from the rest of the sandy surface. With her fore-tarsi, which are armed with rows of stiff hairs and suggest at the same time a broom, a brush and a rake, she works at clearing her subterranean dwelling. The insect stands on its four hind-legs, holding the two at the back a little wide apart, while the front ones alternately scratch and sweep the shifting sand. The precision and quickness of the performance could not be greater if the circular movement of the tarsi were worked by a spring. The sand, shot backwards under the abdomen, passes through the arch of the hind-legs, gushes like a fluid in a continuous stream, describes its parabola and falls to the ground some seven or eight inches away. This spray of dust, kept up evenly for five or ten minutes at a time, is enough to show the dazzling rapidity of the tools employed. I know no other example of this swiftness, which nevertheless in no way detracts from the easy grace and the free movement of the insect, as it advances and retires first on this side, then on that, without discontinuing its parabolic streams of sand.