1957 1 Parsons — Esophageal Valve in Hemiptera 143 
invagination in either of these bugs. The valves of Noton- 
ecta and Ranatra appear, to the author, to be equally simple, 
although Marks considered the latter to be somewhat more 
developed than the former. 
The form of the valve in Belostoma is more complex (see 
Text-fig. 1). Here a shallow invagination is present, which 
may offer resistance to the passage of material from the 
midgut into the foregut. Although an inner cell layer is 
present, there is nothing corresponding to the outer cell 
layer of Hesperocorixa or Pelocoris; instead, the annular 
cells occupy the posterior and outer surfaces of the invag- 
ination. 
The Pelocoris valve is more complicated than that of 
Belostoma , but less so than that of Hesperocorixa and 
Sigara. In its degree of complexity, therefore, it appears 
to be an intermediate between these two groups. The 
invagination is deeper than that of Belostoma but shallower 
than the invagination of corixids. The presence of an enton- 
noir indicates another advancement in complexity. In Pel- 
ocoris an outer cell layer is present, but the annular cell 
region appears to extend part way into the invaginated 
area, gradually merging with the tall cells of the outer cell 
layer (Text-fig. 2). 
The increased length of the invagination, the prominence 
of the entonnoir, and the placement of the annular cells 
in the corixid valve make it the most complex of the five. 
Sutton’s assertion that the entonnoir may extend as much 
as two-thirds of the length of the midgut was not borne 
out in the present study. The cuticula does extend farther 
into the midgut than in the other four bugs, but it was 
never observed to penetrate more than one-tenth the length 
of the mesenteron in the material used in the present study. 
In corixids, the annular cell region never extends into the 
invagination, as it does in Belostoma and Pelocoris. It is, 
instead, sharply set off from the outer cell layer, forming 
a distinct region which encircles the area between the ter- 
mination of the invagination and the beginning of the 
midgut (see Text-fig. 3). 
The conclusion of Sutton that the corixid esophageal 
invagination is long when the midgut contains small 
