A considerable contribution to our knowledge of the genus Bembex has been made in the paper by Wesenberg (written in Danish) which has already been referred to. This paper deals with Bembex rostrata. It was translated for Mr. Ashmead by Mr. Martin Linell. [Illustration: A CORNER OF THE BEMBEX COLONY] It seems that rostrata makes its nest in solid sand, covering it up with loose sand, and usually, also, with a little flat stone, to prevent parasites from entering. The cell measures one cubic inch, the entrance tunnel being one and one half centimeters long, and arcuate. A cell contains four or five fresh flies (Lucilia, Eristalis, etc.), and torn–off wings, sucked–out thoraces, and in the middle of these, a big flat larva. When the larva is hatched the mother brings more and more flies, the flies being larger and larger as it grows. This adjustment of the size of the fly to the growth of the larva has also been noted by Fabre. Wesenberg says that fifty Bembecids will nest on a spot as big as a room during a period of three months. The time required for the development of the larva is two weeks, this giving five or six young ones for the season. He queries, “Does each female have more than one nest? and if so, how can she remember them?” To determine this point we marked six wasps by touching them with differently colored paints, putting near their nests pebbles painted to correspond with the owners, and then watched them closely for three hours. During this time the red wasp returned regularly to the red nest, the blue to the blue, and so on. They were watched for an hour and a half on the following day with the same result, so that it seems quite certain that spinolæ has only one nest at a time. To feed two larvæ at once, with interlopers thrown in, would be a heavier task than the most determined industry could accomplish. Chapter VII THE BURROWERS