THE WESTERN GRASS-STEM SAWFLY. 
15 
If the same amount of moisture had entered the stem where the 
larva was hibernating it probably would have caused its death. 
This experiment, taken in connection with others that were not so 
directly conclusive, seems to prove that the porous plug in the stub 
must prevent in some way the admission of an undue amount of 
moisture into the chamber below, although water readily penetrates it. 
The gallery below the plug is always entirely free from debris, 
forming a hibernation chamber and later a pupation cell. Within 
this chamber the larva lies with its head up and usually pressed 
against the barrier at the top, always on the alert to retreat down- 
ward at any sign of disturbance. It descends by alternately flexing 
and straightening the body, bracing itself first by the jaws, then by 
the caudal horn as it hitches its way down In ascending, the caudal 
horn is thrust against 
the side of the gallery 
or the cocoon, the 
body is straightened, 
the j aws obtain a pur- 
chase to hold the dis- 
tance gained, when 
the body is again 
drawn up until the 
caudal horn is applied 
to the side wall for 
another push. 
Late in the summer 
or during the autumn 
the larva spins for it- 
self within the hiber- 
nation chamber an 
almost transparent 
tube of filnry silken 
tissue. This silk tube is sometimes several times the length of 
the larva, is closed at both ends, and is free from the sides of the 
chamber, so that often it can be readily withdrawn entire. When 
first constructed this fabric is comparatively strong and pliant but 
after some months it grows more brittle and is easily ruptured. As 
a rule it remains intact until the emergence of the adult. Even the 
presence of half a score of parasitic larvae often fails to wreck the deli- 
cate structure during the winter. 
The longevity of the sawfly larvae is remarkable and is worthy of 
mention. September 8, 1911, a number of stubs of Elymus conden- 
satus containing Cephus larvae were gathered and set upright in sand 
within doors. From time to time this sand was moistened but 
finally was allowed to stand perfectly dry. During October, 1912, 
Fig. 10.— Western grass-stem sawfly: Infested wheat stubs, enlarged 
3 diameters, the two left-hand ones opened to show hibernating 
larvae in situ. 
