622 THE AMERICAN NATURALIST. [Vor. XXXVI. 
ganglion. The seventh abdominal segment has two ganglia, 
smaller than those of the prothorax but larger than the remain- , 
ing eight, which are distributed — one in each of the inter- 
vening segments. The dorsal subsegment, interpolated between 
the prothorax and head, and which was previously mentioned, 
is shown here. Neither pair of the muscular bands which 
expand and contract the pharynx (m, Fig. 10), nor either of 
the sucking tubes below their point of juncture (5, Fig. 10), 
would be present in a section through the median plane, but I 
have figured them here diagrammatically for the sake of clearness. 
The most difficult, and, as well, the most interesting part of 
this study has been the tracing of the course of the alimentary 

Fic, 10. — G l di 

g f the larva. a, 2-5, c-c’, three silk glands attached at both ends ; 
d, e, two silk glands attached at one end; sr, silk receptacle ; sf, spinneret ; /, fat bodies ; 
, brain ; g, subeesophageal ganglion ; », band of regenerative cells of stomach; A, point 
of junction of sucking tubes; s, sucking pharynx; », muscle attachment of pharynx; 
o, esophagus 
canal posteriorly from the stomach. That others have expe 
rienced like difficulties in dealing with near relatives is shown 
by the contradictory statements made concerning Osmylus by 
students of its life history (Suckow, Ramdohr, Loew, Dufour, 
and Hagen). I have found the work particularly difficult, as 
the alcoholic specimens were too brittle for dissection, and I 
was obliged to depend entirely upon the study of sections. 
Hagen figures for Osmylus four Malpighian tubules attached 
at both ends to the intestine, the anterior attachment near 
the base of the stomach, the posterior one near the hinder 
end of the intestine. Three other tubules have only the ante 
rior attachment and have their posterior end free. All of the 
tubules are modified in their middle portion for the secretion 
of silk. 
