32 



plant of leaves. A third plant was also defoliated. In a few instances 

 larvffi attacked the stems, in one case eating them off nearly to the 

 roots. 



Injury by the larger larvae is much worse, the younger ones appear- 

 ing to contine attack to cutting small holes here and there in the leaf. 



The punctures made by the females in oviposition and the holes 

 made by the larvte in escaping from the nidus in the leaf also cause 

 injury, particularly when the eggs arc deposited in a bunch, as shown 

 in the illustration (tig. 8, c/, h). The tip of the leaf withered and 

 nearly died as a result of this attack. 



The younger larv^ have the habit of resting during the daytime, 

 usually in a curved position like the letter J. When disturbed even 

 slightly, they promptly curl up spirally, after the manner of many Ten- 

 thredinids, and drop from the plants to the earth below. All of the 

 youngest larvi\3 that have come under observation rest thus on the 

 lower surface of the leaves, and are never curled spirally when upon 

 the leaf. The more mature larvpe may be found during the daytime 

 extended lengthwise on the leaves and more particularly along the 

 stems near the ground, but sometimes in other positions, with their 

 heads usually pointed upward but often downward. In this position 

 they are afforded sufficient concealment on fresh plants to readily 

 escape notice. On such plants, except when occurring in great num- 

 bers, the.y are eminently successful in eluding observation at nearly 

 any time. For example, only one or two of the nearly grown larvae 

 in a lot of upward of a dozen on a small plant were to be seen at one 

 time, the others being securely hidden from view. 



Of the first two broods reared — using the word not in the same sense 

 as generation, but as the progeny of one female — it was quite noticeable 

 that the larv* did not in any case desert the plant on which they had 

 hatched until they were mature or very nearly so. 



Injury is most apparent late in May and early in June, and almost 

 entirely to plants grown under glass, being particularly noticeable, 

 upon plants growing in shaded locations, such as are to be found in 

 greenhouses, under gutters, where the plants grow ranker. In some 

 cases injury appears to be confined to such locations. In the green- 

 house, according to Mr. Gallowa}^, "seldom more than two of the 

 worms are found at one time on the same plant. Two, however, are 

 enough, as they will riddle a half -grown plant in a week." 



Mr. Dorsett's experience with this species goes to show that it is 

 quite persistent in the greenhouse, and difiicult to dislodge after it 

 has obtained a foothold, in this respect ranking with several other 

 species, such as the "red spider" and the aphides, which are not, like 

 the sawfly, limited to any one genus of host plants. In spite of 

 frequent and S3^stematic fumigation and careful watchfulness in his 

 greenhouse, this species continues to be seen, although in small num- 



