THE WESTERN GRASS-STEM SAWFLY. 
13 
one stem. It is believed that the larva that finally reaches maturity 
has devoured its rivals. It is obvious that the remains of these 
superfluous individuals would naturally be counted when a census 
of exuviae was undertaken and would complicate the result. But 
from the best evidence obtainable it is almost certain that there are 
five instars in the larval life of this species. 
The length of the larval period is probably about 60 days, varying 
more or less with the warmth of the summer and the state of maturity 
of the host stems. The acceleration or retardation of the oviposi- 
tion period owing to an early or late spring has much to do with the 
date of maturity of the larvae, and possibly with the length of the 
larval period. August 29, 1911, at Kimballs, Utah, at an elevation 
of 7,000 feet, the writer found mature larvae in stems of Elymus 
condensatus. The next year, at the same place, 
oviposition was beginning freely during the first 
week of July. The determination of the larval 
period is wholly inferential, based upon the find- 
ings in a series of stems (figs. 7 and 8). 
The full-grown larvae vary greatly in size, their 
growth being governed, as is usual in the case of 
such borers, by the quality and quantity of food 
consumed. Those living in wheat stems are 
much smaller as a rule than those found in rank- 
growing grasses such as Elymus. Measurements 
of a series of individuals give variations of from 
8 to 14 mm. in length and from 1 to 2 mm. in 
diameter. 
When mature the larva always seeks the ex- 
treme base of the stem, where it soon begins its 
preparations for hibernation. Its first move is to 
cut a neat V-shaped groove entirely around and 
inside the stem, usually at or a little above ground 
level. This groove never severs the stem completely, but so weakens 
it that the upper stalk, swayed by the wind, will break off com- 
pletely when dr}-, leaving a stub that is very characteristic of the 
work of this insect (fig. 5). In this simple manner the larva provides 
for the eas}^ escape of the adult in the following summer. The length 
of the stub thus formed varies greatly. In Elymus condensatus the 
stub sometimes will project above the ground as much as 3 or 4 
inches, while in other grasses, and especially in wheat, stubs easily 
can be found less than an inch in length in all. 
Instances have been observed where two or more grooves had been 
cut inside the same stem, as if the larva had been uncertain as to the 
best place for severing the grass. After cutting its characteristic 
groove within the stem the larva forces a mass of the debris into the 
Fig. 8. — Western grass-stem 
sawfly: Mature larva. 
Enlarged 5 diameters. 
