and on the Species by which they were constructed. 253 



had the exuviee of the caterpillar compressed at the bottom, but no shroud of 

 the pupae. The inequality in the size of the cells may be attributed to the 

 smaller ones having been the residence of the males. After this examination 

 I divided the same portion of the nest transversely at the line k and /, fig. 5, to 

 exhibit the length of the cells, their longitudinal form and their position (fig. 6). 

 By this section it is also ascertained in what way the saw-flies escape from their 

 cells when they are hatched ; for at the end of each cell, sometimes at one 

 side, and as often on the other, is a circular lid (fig. m), formed of the same 

 leathery material as the entire comb, the outside being crossed with the hairs, 

 the inside smoothish, with the edge whitish and powdery, from the liquid pro- 

 bably secreted by the animal to soften the material during the operation of 

 cutting, which it performs with its two sharp mandibles (fig. h). Through this 

 aperture (fig. n) the saw-flies make their way into the woolly wall which sur- 

 rounds them, forcing themselves to the side next the bark, and then crawling 

 out into the open air. In two of the cells I found dead caterpillars (fig. 7), 

 which considerably resemble those of the genus Hylotoma : they are ochreous, 

 the head is large and horny, with a black eye-like dot on each side : the 

 body is covered with irregular transverse rows of black warts, the hinder 

 segments have short black spiny appendages on their sides, and the tail 

 is surrounded by ten shining black obtuse conical spines : they have six 

 pectoral, eight abdominal and two anal feet; at least I could discover no 

 more in the dried specimens ; but it is possible that others might be con- 

 tracted and concealed, or the lateral appendages may be employed instead 

 of them*. 



The more we reflect upon this singular nest, the more dissimilar will it 

 appear to be from anything of the kind hitherto discovered, whether we con- 

 sider its structure, or the period of the animal's existence when it is con- 

 structed ; for the compound nidus is generally the work of the parents, to 

 protect their eggs and feed their larvae, at once affbrding them an asylum 

 during three stages of their existence. Thus the bee forms its comb and the 

 wasp its nest ; the social ants also labour to form a dwelling ; but in this 

 Tenthredinous nest we find the larvce uniting and forming cells in which to 



* In Hylotoma the number of feet in the larvae is 20 ; those of Schizocerus are unknown, 



2 l2 



