OF WASHINGTON. 75 



convinced that the principal cause of this failure is the work 

 of a single insect, the Lepidopterous gall-maker Ecdytolopha 

 insitiana. Dr. T. W. Harris gives a 'good account of what he 

 found as the most destructive enemy of the tree, viz., the well- 

 known Cyllene robinice, and attributes to it the complete disap 

 pearance of the tree in. some of the Western States. He is no 

 doubt correct in this opinion for the localities referred to by 

 him, but in the vicinity of Washington this Cerambycid yields 

 the first rank to the Ecdytolopha. This is an extremely abundant 

 species, and it is only to be wondered that its destructiveness 

 to the Black Locust did not attract the attention of Harris or 

 the other older authors. It oviposits in vigorous young 

 shoots, which usually die soon after the moth has issued from 

 the gall, or which at best remain sickly and never grow up to 

 a large tree. The wound rarely heals over ; on the contrary, 

 it is usually enlarged by the action of numerous sap-beetles or 

 other insects which invade the deserted gall. 



Cyllene robinics comes next in importance as an enemy of 

 this tree. Although a very common insect near Washington, 

 it is not so universally distributed as the gall-moth and seems 

 to live in large colonies, affecting all trees of small groves, 

 while long hillsides full of locust trees are not infested by it. 

 There are always numerous larvae in a single trunk, and a tree 

 once infested is usually occupied by two or three successive 

 generations of larvae until it dies. Xyleutes robinice would be 

 equally destructive but is much rarer here, and apparently 

 prefers other trees. Whether or not Agrilus egenus is to be 

 classed among the important enemies of the Black Locust 

 must be left undecided, but it certainly prefers to breed under 

 the bark of branches of small trunks which have already 

 greatly suffered from various other causes. 



The host of insects feeding on the leaves may be divided in 

 two classes. Those which feed externally on the leaves (nu 

 merous Lepidoptera, various Tenthredinidcz and Coleoptera) 

 do not seriously injure the tree. Total defoliation by them 

 hardly ever takes place, and the damage caused by all of them 

 combined is certainly much smaller than that done by the 

 second class, the leaf-mining species (various Tineidce and 

 Odontota dorsalis). These produce almost every year that 

 burnt appearance of the locust trees so noticeable in July or 

 later in the season, -and the premature loss of all or most 

 leaves naturally weakens the trees. The Elm trees in our 

 parks, when defoliated by the second generation of Galeruca 

 xanthomel<zna, produce a new growth of leaves and are green 

 in the autumn ; but the Black Locust, if once defoliated, re 

 mains so for the rest of the season. These leaf-mining species, 



