80 
the whole of the interior of the tree being more or leas filled 
up with this material.” Now, this is typical of the subsidiary 
nests made in live trees where Termes gestroi is feeding, and it is 
probably such nests that Robinson explored and described. Robinson 
found one nest out of many investigated which contained eggs — 
that is, a parent colony. Personally, I have never found a parent 
nest in live rubber, although such do occasionally occur in old rubber 
where surface timber has gradually rotted away. Always the 
parent nest has been in a log or stump. One might search long in 
attacked young rubber trees for a parent colony without success, but 
if such were found it would only serve further to obscure the general 
truth. Cai^y’s clue of the logs and stumps was not followed up, and 
this otherwise valuable report is marred by the failure to provide 
that most desirable of all information, the location of the main nest. 
In April, 1906, E. P. Stebbing,- Forest Zoologist in India, 
entered the discussion because of specimens sent from Mergui Estate. 
Stebbing, who, of course, had no first hand knowledge of living Termes 
gestroi , larged the need for procuring full information upon its life 
history. Among points which he indicated as requiring elucidation, 
there are two of vital importance — namely, “ Where do the termites 
found in the nest in the crown of the root come from in the first 
instance?” and “Are there any galleries radiating from the nest to 
other parts of the plantation ? If so, where do they go ? ” Now this 
is the first consideration of the problem by a trained Economic 
Entomologist, and it is interesting to note that at once the true 
line of investigation was suggested. 
In May, 1908, Pratt supplied the material for answering 
Stebbing’s questions in an interim report upon bis termite work. 
Tracing back runways from 15 or 20 eight-year old rubber trees, all 
of them terminated in a nest, about two feet in diameter, constructed 
round a buried stump. In this nest was discovered the first 
authentic queen. Runways ramified from the main nests, broaden- 
ing out iu places to form large open spaces from which branch 
burrows diverged, the larger of these leading to subsidiary nests 
inside stumps, logs, and fallen trees. This was indeed an important 
verification of Carey’s statement, and laid the foundation of a 
rational method of treatment. Pratt pointed out that the age of the 
rubber, and the nature of the. soil had little, if any, effect upon the 
incidence of the attack, but what really mattered was the, presence of 
timber. The supposed deadly effect of lalang upon Termes gestroi, 
he rightly ascribed to the general absence of timber on old lalang 
areas. Further promise of being on the right track at last was given 
in the following passages : “ The most important point in connection 
with the present abundance of Termes gestroi ... is that it 
.attacks dead as well as living wood . . . The fact that it is among 
the dead wood, and rapidly multiplying lias been ignored. It is only 
when the rubber trees are being attacked that any measures are 
