1909. ] LIFE-HISTORY OF THE AGRIONID DRAGONFLY. 267 
to the body. This circulation is rapid—as rapid as in the 
abdominal segments,—but that in the tibie and tarsi is extremely 
slow and irregular, the few visible corpuscles being constantly 
held up for quite long periods. 
In the antenne the circulation extends now to the apex of the 
5th segment. A few corpuscles can be seen travelling up the 
inner side very irregularly, there being constant delays and short 
rushes. At the apex of the 5th segment they quickly pass over 
to the outer side and travel back to the head in the same easy 
stages. 
Bowerbank (/. c.) mentions that the nymph he examined had 
antenne of 6 segments and that the circulation only extended 
to the extremity of the 4th segment—a condition probably 
reached about the 6th or 7th stage, but I have not paid sufficient 
attention to the circulation to be more precise. 
Stages 11, 12, de. 
If the nymph is not going to hatch out at the end of the 10th 
stage, the antenne do not change from the 6-segmented condition. 
It seems to be a rule in Agrion and /schnura that the 7-segmented 
condition only exists in the final stage, while in Pyrrhosoma the 
only two individuals which I reared through the last stages 
showed the 7-segmented condition in the penultimate stage. 
The labium may have either 6 or 7 hairs on each palp and 5 or 
6 pairs on the body, while occasionally no further development 
takes place after the 10th moult. If there are 7 hairs on each 
palp they are very crowded and the 7th appears at the extreme 
base of the palp. 
The majority of my nymphs hatched out into imagines at the 
end of the 10th, 11th, or 12th moults. Two Agrions passed 
through 13 stages, while the two Pyrrhosomas passed through 
11 and 12 respectively. A single Hnallagma passed through 
14 stages as did Hrythromma, while an Ischnura imago emerged 
at the 12th moult. I shall later on discuss the question of the 
variation in the number of moults in nymphs of the same species. 
Tue Movtts. 
In the case of the Agrion nymphs which hatched from the 
eggs at the end of August, the first moult took place in from 8 to 
12 days. Their average length, after hatching, was 1:3 mm., 
while after the first moult it was 1-6 mm. The second moult 
was a less regular event than the first, some nymphs undergoing 
it 14 days later, while others took more than three weeks to 
complete their second stage. The third moult again took place 
from a fortnight to five weeks after the second, while the fourth 
was even more irregular than the third. 
At the beginning of November the nymphs become very 
18% 
