7856 



Insects. 



chevaux de Jrise, consisting of ten or twelve little posts formed of dried 

 secretion : beginning at the stalk, it emits a little of the secretion 

 from the mouth, and raises its head to a certain height, whereupon 

 the fluid immediately dries (see fig. 1,/). Under the microscope the 

 palisades appear as if formed of dried bubbles or little bladders, and 

 are highly iridescent, as I have endeavoured to represent in fig. 2. 

 When the larva has covered the leaf-stalk, it turns round again, creeps 

 further on to the leaf, and raises a whole row of similar little posts, 

 placed quite close to each other, so that it is entirely surrounded by 

 them (fig. 1, c). In the open space inside this rampart, it bites a hole 

 in the leaf and begins to feed, as shown in fig. 1 a. When the hole 

 becomes very large, the larva erects its fortification on both sides of 

 the leaf. I have observed that in feeding it consumes its palisades, 

 and I have once seen a larva first eat up its old defences, and imme- 

 diately erect new ones enclosing a larger area. 



With regard to the object of this singular operation, it can only, I 

 think, be as a means of defence against the attack of apterous Hyme- 

 noptera, such as ants, or parasites of the genus Pezomachus. I have 

 never seen any other insect on a leaf which was occupied by a larva 

 of Nematus vallator. 



This larva is in no wise remarkable either for form or colour. Like 

 other Nemati, it has twenty legs, of which the anterior six are longer 

 than the rest and provided with claws. The fourth and eleventh seg- 

 ments are apodal. The head round, but depressed anteriorly, pale 

 green, with two brownish green stripes running obliquely across the 

 vertex ; trophi pale brown ; the eyes seated in round black spots. 

 The general colour of the body is pale green, clothed, as it were, with 

 brownish granules, more particularly on the sides above the legs and 

 round the spiracles. The legs are green, somewhat darker at the 

 joints. The caterpillar appears never to stretch itself at full length, 

 but is usually in the position shown at fig. 3. I have also never seen 

 it elevate the abdomen in the air, as is continually done by some 

 others of the genus, for example N. septentrionalis. The larva attains 

 a length of 6 lines. 



This species remains in the larval state about four weeks. Those 

 larva? which I had placed in glasses, for the purpose of rearing, hid 

 themselves in the earth on the 26th and 27th of June. I had, how- 

 ever, already missed other larva? from the tree, and these I must 

 regard as full-grown ; so that all the eggs were not laid in one week. 

 On the 29th of June I observed another individual spin up in the 

 folds of a withered leaf, from which it appears that they do not all 



