CXL I. TRÄGÅRDH. 
rows in the scales (fig. 22 e). The white cocoons are easily found on cutting 
a cone in two (fig. 21). The burrows in the base of the scales are plainly 
discernible from the upper or the lower side of the scale (fig. 22 b-d). The 
cocoons are flask-shaped, rounded at the bottom and taper towards the mouth, 
which corresponds to an irregular opening in the scale, made by the larva 
in order to facilitate the emerging of the pupa half-way through the opening, in 
which position the midge hatches (fig. 22 c). 
The biological import of this peculiar mode of pupation becomes evident 
if we take into consideration that the seeds fall to the ground, whereas the 
cones remain on the trees. If the larvae leave the seeds and pupate in the 
cones, they emerge in the crown of the spruce-tree close together and close 
to the place where to lay their eggs, being, moreover, during the pupal pe- 
riod well protected in the scales. 
If, on the other hand, they were to remain in the seeds and fall to the 
ground in these, they would have a considerable and — at least for such weak 
flyers as they are — dangerous distance to cover in order to find one another 
for mating purpose and to find suitable places for the eggs, and on the ground 
they would be exposed to numerous enemies. 
The fact that the pupation takes place in the scales is therefore doubtless 
an adaptation which greatly ensures the protection of the species. 
Perrisia strobi differs in the mode of pupation from Plemelrella abietina SEITN., 
which latter, according to SEITNER, pupates in the seeds. It seems, however, 
that SEITNER shelled the cones and kept the seeds apart, and it is just pos- 
sible that this accounts for the pupation having taken place in the seeds, and 
that if the seeds had been left in the cones the larvae of Plemeliella also 
would have pupated in the scales. 
Piatvgaster contorticornis RATzZ., the parasite of the spruce-seed 
gall-midge. 
'This species RATZEBURG bred from spruce-tree cones, and he suspected it 
of being the parasite of Cecidomyia strobi WINN. It appears not to have been 
found since RATZEBURG'S time until 1913, when it was bred from cones by 
KiIEFFER in Lorraine; it was not previously recorded from Sweden. For the 
description of both sexes the reader is referred to figs. 23-26. 
The larva of Platygaster contorticornis is an ento-parasite of the larva of 
Perrisia strobi, but does not kill the host in the seed, waiting until the latter 
has spun the white cocoon in the base of the scale, the inflated skin of the 
larva of Perrista serving as a cocoon of the parasite (fig. 27). 
The only conclusive evidence of Platygaster being the parasite of Perrisia, 
the breeding of it from the larva of Perrisia was still wanting, and it was 
not easy to establish the fact, as all tne cones at my disposal contained also 
other insects which might be the hosts of Platygaster. "The data accumulated 
during the breeding of the insects, however, showed that Platygaster accom- 
panied FPerrisia, emerging simultaneously and on examining cones very heavily 
infested with Perrisia a dead Platygaster was at last found in the larval skin 
Offa CFerrisid. 
The map fig. 28 shows, that Platygaster accompanies FPerrisia all over the 
country. It is true, that it was not found in 24 & of the localities, where 
