1923] 
A Singular Habit of Saw fly Larvce 
11 
Perga or those of the subfamily Argiinse; that is, they would 
have a single, more or less impervious cocoon surrounded by a 
loosely woven, reticulate, outer cocoon which would be covered 
with long hair/’ 
Such literature as we have been able to consult on the habits 
of the Australian saw-flies of the genus Perga proves to be very 
interesting in connection with the South American larvae referred 
by Mr. Rohwer to the family Perryidae. Froggatt (1891, 1901 
and 1918) has published notes on the larval habits of several 
species of Perga. Concerning one of them, the ^‘steel-blue saw- 
fly”, P. dorsalis Leach, he says (1918): “The gregarious larvae 
feed at night, and rest during the day, clustered together in an 
oval mass, on the stem of the gum-tree upon which they are 
feeding. When disturbed, they exude a sticky yellow substance 
from the mouth, at the' same time raising the tip of the body, and 
tapping it down on the foliage. The leaves are devoured from 
the top of the young gum trees; and when the larvae are full fed, 
they crawl down the stem to pupate. I have found them fully 
developed in the middle of April; but when they descend from 
their resting place, they wander about over the grass for several 
days before they finally select a place in which to pupate, general- 
ly the softer soil against a tree-trunk. Into this they burrow to 
a depth of three or four inches, massing their large, oval cocoons 
in rows, one against the other. I watched several large swarms 
feeding upon the Peppermint-gums {Eucalyptus novce-anglicB) at 
our Experiment Station at Uralla, and afterwards in their 
erratic wanderings over the grass; and marked down their final 
resting place and dug up the cocoons. At Binalong, in April, I 
observed two large swarms marching in massed formation; the 
heads of the hind rows always rested upon those in front as they 
moved along steadily together. Every now and then, the front 
rank came to a dead stop, when they all rested for three or four 
minutes; then a number began raising up and tapping down 
the tip of the abdomen, whereupon the whole band took up the 
motion; the leading ranks made a fresh start, and all moved 
along again. In the largest band, I counted two hundred and 
fifty caterpillars.” 
