198 PROCEEDINGS OF THE NATIONAL MUSEUM Vou. 113 
this ‘‘step.”” However, further consideration of this possibility would 
be difficult without more information on the biology of the rest of 
the species in the subgenus Smicronyz, particularly such species as 
S. halophilus Blatchley, S. lutulentus Dietz, S. lepidus Dietz, and 
S. vestitus LeConte, which have been found associated with plants 
other than Cuscuta, but resemble the species which do breed in 
Cuscuta. 
1. SUBGENERA Pachyphanes, Pseudosmicronyz, AND Desmoris: 
These subgenera could have developed through further modification 
of species or groups of species which developed in step II (see above). 
All three groups share the basic characteristics of the genus Smicronyz 
with the subgenus Smicronyz, but exhibit what are probably modifica- 
tions of the characteristics of the latter group. 
Species in these three subgenera have been found breeding in plants 
of a number of genera, particularly genera of Compositae, but are not 
known to breed in Cuscuta. 
Discussion: The morphological changes which characterize the three 
American subgenera involve (a) the rostrum of the female, it being 
longer, smoother, and more tapered before the antennal insertions than 
in the male; (b) the genitalia (Gn Pseudosmicronyz and Desmoris), 1.e., 
the ovipositor and associated parts being elongate and the male geni- 
talia being usually elongate; (c) the second segment of the antennal 
funicle, which is now noticeably longer than the third segment. These 
changes all reach their extremes in subgenus Desmoris (figs. 34, 94, 
140). However, in all three subgenera there are species in which one 
or more of the three main types of change has not gone very far. In 
the scapalis group of Desmoris, the median lobe of the male genitalia 
is not elongate (fig. 93). In subgenus Pseudosmicronyz, the female 
genitalia and second funicular segment of the griseus group show little 
elongation; in most species of the ovipennis and abnormis groups, the 
rostrum of the female is not very smooth or tapered before the antennal 
insertions. In subgenus Pachyphanes, the genitalia are much as in the 
typical subgenus (compare figs. 48, 50, 54, 73). Thus, while the sub- 
genera show certain general tendencies away from subgenus Smicronyz, 
there are a number of points of interdigitation with that subgenus. 
Because of the overlap between those groups, they are treated here as 
subgenera rather than genera. 
The three subgenera show the most modification in the body parts 
which are closely associated with oviposition. If there was a change 
in hosts, from species of Cuscuta to such a genus as Helianthus, it would 
have indicated a radical change in the accessibility of the oviposition 
site to the female weevil. The seed capsule of Cuscuta is rather ex- 
posed, but the achenes of tubuliflorous composites are usually beneath 
the long, closely spaced corollae of the disk flowers. It seems likely that 
