90 THE SOLITARY WASPS. 



came up head first and pulled it within. As a ucual thing 

 unicolor walks straight in, carrying the bug in her mandibles. 



Three of our four individuals of unicolor excavated their 

 nests all at once instead of enlarging from day to day; but this 

 one, after making her nest on the morning of September fourth,, 

 was found early on the morning of the fifth excavating from 

 within. She w^orked in a very slow and dignified way, w^ithout 

 the least hurry or bluster, loosening the earth and then lying 

 quite flat and pushing it out with the end of her abdomen. 

 "Working backwards in this way she came out quite covered with 

 dust, and still lying flat, pushed the loose material away from 

 the entrance with her hind legs, but with a motion too gentle 

 to be called kicking. 



The nest was opened on September sixth and tliis time the 

 tunnel was successfully followed, and foiu* pockets, which led 

 off from it half way down, were discovered. (PI. XL, fig. 3.) 

 Three of these pockets contained partly eaten bugs, and one of 

 them had also a parasitic larva. The fourth pocket, which 

 seemed to have been most recently stored, contained three bugs,, 

 and on the venter of the last one taken in, near the base of 

 the first leg on the right side, and at right angles to the length 

 of the body, was an egg. We were surprised to find that such 

 a store of provisions had been taken into this nest, for if, as 

 we had taken for granted, it had been freshly made on Septem- 

 ber fourth by the wasp whose nest we had destroyed on the pre- 

 vious day, she had filled three pockets on the fifth and sixth, and 

 the bugs in two of these, as well as those carried in on the fourth, 

 had been partially destroyed,, not by the young of unicolor^ but 

 by parasites within their own bodies, before the sixth. "We 

 concluded, however, that this was the case, since our only alter- 

 native was to suppose that this nest had been made by another 

 wasp, and that it had been partly filled before we saw it on Sep- 

 tember fourth, and it seemed extremely unlikely that anything 

 could have been working there on the first three days of Sep- 

 tember without our seeing it. Moreover, if she was not the 

 wasp that we had been watching right along, why should she 



