OF WASHINGTON. 



321 



stems would confirm the carnivorous habit of at least some of the 

 species of this genus in the larva state. 



It is an interesting fact, however, in this connection, that all the 

 larvae of Mordellistena have not the carnivorous habit nor the dorsal 

 tubercles so prominently developed, and I exhibit samples of four 

 other species by way of illustration. In Mordellistena unicolor 

 which I found July 13, 1874, in the stems of Ambrosia artemisi- 

 cefolia. and which was subsequently obtained by Mr. H. G. Hub- 

 bard from the stems of an undetermined plant at Detroit, Michi 

 gan, the dorsal tubercles are somewhat less developed than in 

 those of pustulata^ but still quite prominent. 



A single larva from the stems of the tall grass 7^riodia cuprea 

 was found April 8, 1888, the larva commencing work near the 

 flower and gradually working down to the roots. At from one 

 to two inches above the ground the stem is almost cut through so 

 that it is frequently broken off by the wind. August 9th the per 

 fect beetle issued and Chalcidid parasites were also bred from the 

 same stem. The species appears to be M. nubila Lee. ; but the 

 species of the genus are difficult to determine, and Prof. J. B. 

 Smith's paper has not helped to make the determinations easier. 

 This larva has the tubercles least developed, but they are faintly 

 discernible on the same joints as in the other species. This larva 

 also differs from the others in the more rudimentary thoracic legs 

 and comes much nearer in characteristics to the typical larva of 

 Mordella. 



A single larva, without number, from the stems of Amarantus 

 also shows the dorsal tubercles but faintly developed. 



In Mordellistena Jloridensis (Fig. 22), of which I exhibit good 



Fig. 22. Mordellistena Jloridensis: a, larva; 3, pupa; c, imago; d, hind 

 leg of same ; e, dorsal tubercles of larva seen from above (original). 



