206 DISEASES OF THE ROSE. 



the act, because it doubtless does it in the night. On that occasion it lets itseli 

 slide to the earth upon a thread it had secreted for the purpose ; unfortunately, I 

 caught it. and put it under a glass with some rose-tree leaves, an undoubted mis- 

 take since, having attained its full growth, it would no longer eat, and utterly 

 refused the leaves. It should have been permitted to fall entirely to the ground, 

 where it would probably have enveloped itself in a casing of soft network, or a 

 shell, as it did under the glass, and have been changed into a nymph or chrysalis 

 to come forth a perfect insect the next spring. 



It is an experiment I recommend to amateurs, and one which requires great 

 care. It is the only means in our power of ascertaining the veritable insect which 

 produces the larva so destructive to roses. 



In my first work, I attributed this larva to a four-winged insect of the genus 

 Tcnthredo, and named it Tenthrcdo excavator, thinking it new. I have since 

 ascertained 1st, not only that it was not a new species, but moreover also, that 

 it is doubtful whether it is the parent of this larva 3d, that this red-fly, as the 

 gardeners call it (and under this name they confound many different species), is 

 the Tenthredo Roses of Panzius 1 3d, that it is not by digging in the peduncles 

 that it injures the Rose, but by boring the Rose when its bud is large, and thus 

 destroying the top of the blossom, which withers, and is partially ruined while its 

 peduncle yet remains firm since it destroys the Rose alone, which it pierces, and 

 sometimes only the upper part of that. Upon opening these altered blossoms, 

 worms are found larger than those which pierce the peduncles, and which are the 

 offspring of the eggs deposited there. Generally the larva of the saw-fly is of too 

 large a size by far, to allow of the supposition that it is the same insect, as is 

 evident from the figures of each given by Reaumur. 



As I am not mathematically certain upon this subject, 1 recommend the study 

 of the habits of the saw-fly, of which I have given a description in another place, 

 in order that it may be recognized. I will hereafter describe the white-footed 

 Tenthredo, supposed with more reason to be the parent of the larva, which exca- 

 vates the stem that bears the Rose. 



The red-fly of the gardeners, the Tenthredo Rosa of Panzius, and the Hylotoma 

 Rosce of Fabricius, is a four- winged insect, whose generic characters are, a single 

 radiated cellule and four cubitals at each upper or superior wing, and antennae 



1 More than sixty species of the ancient genus Tcnthredo live at the expense of dif- 

 ferent parts of the rose-tree, the leaves, blossoms, fruit, bark, etc. But the larvae of 

 the saw-fly, whose ravages I have described, cause more than three-fourths of the 

 depredations, and it is therefore important to distinguish them. 



Reaumur speaks in his memoir es, of a false caterpillar (tome v. p. 98,) which inocu- 

 lates the rose-tree, and gives a drawing of its larvae (pi. x.), also of the altered branches 

 (same pi, f . 1 & 2). It is not thus the ravages of my insect show themselves; it is 

 probable, as Saint-Fargeau thinks, that it is a different insect. 



Reaumur (mem. \. 102) speaks also of a false caterpillar which he calls Chenille 

 bizarre, on account of its raising the end of its tail like a serpent. It eats the leaves of 

 the rose-tree in June, July and August. He gives a drawing of this larva (pi. xiv. fig. 

 1, 2, 3,) and the fly or tenthredine which gives it birth, (same pi. fig. 10.) They are 

 seen in great numbers in the spring, especially in moist seasons. The rose-trees in 

 free soil are sometimes stripped of all their leaves, of which there remains only a sieve, 

 by this larva, which is rarely alone on a leaf, and eats its sides and upper part. 



Another larva, also distinguished by the same naturalist (pi. xii. f. 20, 21), eats the 

 lower part of the leaf. It is much smaller. Here are then four saw-flies, the two of 

 which I have here spoken and the two mentioned in the text, whose larvae commit 

 most havoc on rose-trees. The first, however, injure only the leaves. 



The worms found in fruit and vegetables are almost always larvae of the tenthre- 

 dines. 



