1909.) LIFE-HISTORY OF THE AGRIONID DRAGONFLY. 283 
Such a moult as that described by Edwards is apparently a 
“ srowth-moult ” pure and simple. Unfortunately, however, it 
is not always so easy to determine the nature of a moult 
Development involves a certain amount of local growth which 
presumably proceeds steadily between the moults. Ifa “ growth- 
moult ” is interpolated, the ectodermal structures will noes remain 
as they were before the “ growth-moult,’ but will show some 
advance as the result of that moult, so that, if in the ideal 
individual there are a definite number of developmental stages 
each showing a definite advance, the interpolated ‘“ growth- 
moult” upsets this state of things. But the developmental 
stages must get right again by the end of the larval period, 
so that there must be some power of adjustment; in other 
words, the interpolation of a single “ growth-moult” will not 
necessarily add merely one moult to the life-history, but may 
necessitate one or more additional developmental moults for 
adjustment. The more normal moults in the life-history, the 
more probable it will be that the developmental stages will be 
complicated by stages for adjustment. The lengthening of the 
later stages of larval life, which Packard (J. c. p. 615) regards as 
chiefly growth-stages, seems to me to be a lengthening of the 
developmental stages owing to the more radical changes which 
they involve. The last stage of the nymph, for instance, in which 
the most marked internal changes take place is always a long one 
and corresponds with the pupal stage of the Holometabola. 
The fact that certain moults have nothing whatever to do with 
growth (e. g., of the pronymph of the dragonfly; of the sub- 
imago of Ephemerida ; moult to pupa and imago, &e.) seems fatal 
to a theory that growth is the cause of the moult, while the fact 
that certain moults have apparently nothing to do with develop- 
ment (e. g., the moult of larve after hibernation) seems fatal to 
a theory which makes development the cause of the moult. 
Since, however, both causes seem to be at work, it is possible 
that both theories, combined and readjusted, may give us a true 
explanation of the significance of the moult. 
BIBLIOGRAPHY, 
Baraitron, HE, 1892.—“Sur le déterminisme physiologique de la 
métamorphose chez le ver a sole.” C. R. Acad. Sci. exv.. 
pp. 61-64. 
BATAILLON, “La Métamorphose du Ver a soie et@le. 
déterminisme évolutif.” Bull. Sci. France et Belgique, 
xxv. pp. 18-55. 
Benn, W. 1835-—‘ Decouverte d’une circulation de fluide 
nutritif dans les pattes de plusieurs insectes hémiptéres.” 
Ann. Sci. Nat., Zool. ser. 2, iv. pp. 1-12. 
BowErRBANE, J. 8. 1833.—‘‘ Observations on the Circulation of 
the Blood of Insects.” Ent. Mag. 1. pp. 259-244. 
19* 
