LIFE -HISTORIES OF SAWFLIES. 107 



tendency to live in concealment common to both, make me 

 suppose that this little larva may represent a very early stage of 

 the larger one. 



When the larva is full grown it leaves its dwelling, and goes 

 into the ground, where it forms a blunt oval cocoon of particles 

 of sand and mould spun together; this takes place in the be- 

 ginning of June, and in this cocoon the larva remains without 

 change until the spring, somewhat bent, with the head downward ; 

 I found the larva in this state on opening one of the cocoons in 

 the winter. In consequence of the larva remaining so long 

 without change, and of my having so few examples, I had no 

 opportunity of observing the pupa, and so I have been prevented 

 from giving a figure of the insect in this state. I was afraid of 

 opening the cocoons one after the other, as by so doing I should 

 kill them, for it very seldom happens that the larva survives this 

 treatment. I cannot account for my not having noticed the 

 insect moving about in the glass, but the only imago which I 

 succeeded in rearing was found dead, lying on the mould ; it was 

 a female of Selandria candidata of Fallen, King's 8. repanda, 

 which seems to be a rather scarce insect everywhere. 



As my observations of this insect were so incomplete— I was 

 unacquainted with the pupa and the male — I resolved to wait 

 until I might have an opportunity of continuing my researches. 

 I have, however, not been fortunate enough to meet with the 

 larva again, and although for several years I made plans to go 

 over to Noordwijk in the beginning of June to examine the roses 

 in the botanical gardens at that place, I was only able to get there 

 once, and the result was nil. I also made enquiry, through an 

 acquaintance, of one of the large growers there, but with the same 

 result. Under these circumstaiices, I determined not to wait any 

 longer in the hope of becoming better acquainted with the life- 

 history of this species, but to publish what I knew on the subject. 

 Selandria candidata belongs to a small group which Thomson 

 separates as a genus, under the name of Poecilosoma, one of its 

 principal characteristics being the variegated colouring of the 

 abdomen, or, at all events, the occurrence of transverse marks or 

 bands on that part of the body. For myself, I am opposed to 

 founding divisions on such trivial characters, but I fear the 

 docility of entomologists is such that this little genus will be 

 accepted as so many others have been. 



