The reeds invaded are laid horizontally, a condition on which the Bees likewise insist, if only to shelter from the rain the house-door, plugged with pervious materials, such as mud, cotton, or round, leafy disks. Their inner diameter attains an average of two-fifths of an inch. The length occupied by the cells varies greatly. Sometimes the Odynerus takes possession only of that fragment of the interval between two knots which the stroke of the pruning-knife has left free, a fragment longer or shorter according to the chances of the cutting. In that case, a small number of cells is enough to fill the available space. But generally, if the stump be too short and not worth the trouble of working, the insect bores through the partition at the end and thus adds a complete internode to the vestibule with the open entrance. In a lodging of this kind, some eight inches long, the number of chambers will amount to fourteen or fifteen. In thus enlarging the house by removing a floor, the Odynerus displays two separate talents, the plasterer’s and the carpenter’s. Her knack for wood-working, moreover, is extremely useful in another circumstance, as we shall see. The Three-horned Osmia, also an enthusiastic partitioner of reeds, does not employ this means of obtaining a spacious lodging at small cost. I find that she always leaves the first party-wall intact, building the row of cells against it, however short the section may be. To make an opening in a slight barrier is not one of her methods. She could do it if she wished; for to gnaw through the ceiling of the cell on hatching and then through the general door of the nest is a more difficult job. She possesses in her mandibles a tool powerful enough for the purpose; but she is not aware that a splendid gallery lies beyond the obstacle. How did the Odynerus learn, if she did not know from the beginning, what the Osmia, with her greater experience of the reed, does not know? Apart from the ingenious device of breaking down the party-wall in order to enlarge the premises, the Odynerus is the Osmia’s equal as a plasterer and partition-builder. The results of the two industries resemble each other so closely that we should easily confuse them if we merely examined the structure. We find in both cases, at irregular intervals, the same partitions, the same round disks of fine earth, of mud gathered wet on the brink of an irrigation-ditch or stream. Judging from the appearance of the materials, I imagine that the Odynerus has fetched her clay from the banks of the neighbouring torrent, the Aygues.