28 
PACIFIC SCIENCE, VoL VI, January, 1952 
buff. The seed of Agathis rohusta is 11 to 12 
mm. long in the median line, and the pupal 
cell is 6 mm. long by 2.5 mm. wide and 
elongate-ovoid. Seeds from which Agathi- 
phaga queenslandensh had emerged were kindly 
sent by Brimblecombe. They were found to 
have subcircular emergence holes 1.5 mm. in 
diameter. 
It is doubtful if the tips of the long pupal 
mandibles can be apposed within the con- 
fines of the pupal chamber. The structure of 
the mandible makes it apparent that the 
emergence hole is made by a gouging or 
chiselling action of the teeth on the peri- 
phery of the circular, truncated, concave apex. 
Two holes are present in some seeds, indicat- 
ing that the two mandibles may not be able 
to work at the one point on the wall of the 
pupal cell. The movement of the pupal man- 
dibles must be produced by the muscles of 
the adult mandibles, in a manner similar to 
that described by Hinton (1946^: 284) in the 
pupa of the trichopteron Khyacophila dorsalis 
Curt. The alternative is that torsional move- 
ments of the whole body enable the man- 
dibles to make an emergence hole, and this 
seems very unlikely. It is evident from the 
material sent by Brimblecombe that the pupa 
emerges completely, or nearly so, from the 
seed before the emergence of the adult, since 
the pupal skins were free and not enclosed in 
the seed. The possession of tarsal claws by 
the pupa must greatly facilitate its emergence 
from the seed. 
Brimblecombe informs me that Agathi- 
phaga queenslandensis may destroy up to 20 
per cent of the fertile seed of the Queensland 
Kauri Pine, Agathis rohusta. Petrie (undated: 
9) mentions a loss of the order of 90 per cent 
of the seed. 
The seed of Agathis palmerstoni is also in- 
fested by Agathiphaga, but it is not clear 
whether this is A. queenslandensis or another 
species as I have not seen material which is 
definitely from Agathis palmerstoni seed. There 
was some doubt about the origin and identity 
of the seed recorded as from Como (.^), but 
adults from this material, except for minor 
variation in the venation, were identical with 
those from A. rohusta seed from Mary- 
borough. 
There is, of course, a possibility that 
Agathiphaga may infest the seeds of pines of 
the genus Araucaria which, in Queensland 
(but not in Fiji), occurs in the same areas as 
Agathis. 
Fig, 11. Hymenopterous parasite of A. queens- 
landensis. a. Facial aspect of head of larva; b, larval 
spiracle, 
A hymenopterous parasite attacks Agathi- 
phaga queenslandensis. Two infested seeds were 
found to contain only the head capsule of the 
moth larva and four or five dead larvae of the 
parasite. The head capsule and the pro- 
thoracic spiracle of this parasitic larva are 
shown in Figure 11 to facilitate future identi- 
fication. The parasitised larvae did not ex- 
ceed 1 or 2 per cent of the total number 
examined. I have no information on the dis- 
tribution of Agathiphaga vitiensis in Fiji or the 
extent to which it damages Agathis vitiensis 
seed there. Seeds containing living larvae were 
sent to me and 18 months later, in mid- 
September, pupae were present. No emer- 
gence of adults occurred even when the living 
pupae were removed from the seeds and kept 
in moist conditions, though the adults were 
fully developed inside the pupal case. The 
pupal head and forelegs were freely movable, 
but I did not observe any movement of the 
mandibles. 
