106 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY 
Nat. Sci., Phila., 60: 487, 1908), and though they live from nine 
months to three years as aquatic larvae, flying for only a few 
weeks, their presence in a given region is more the result of their 
imaginal life (flight wanderings or migrations of females) than of 
their larval existence. At least recent changes would seem to de- 
pend on the imaginal states. 
The present paper is an attempt to trace, and to some extent to 
explain, the New England distribution of the Odonata. In this 
study, it has been necessary to examine the continental distribu- 
tion of species. The local elements that have been investigated 
have included topography, geological influence including glacial 
phenomena, temperature, moisture, and halophilous control, water 
temperatures, variability in seasonal occurrence, migration, and 
wind carriage. These will be discussed in their turn. 
Dr. C. H. Kennedy in his very kind and generous review (Ent. 
News, 31 : 206-208, 1920) of the author's Manual of New Eng- 
land Odonata (Mem. Thoreau Mus. Nat. Hist., 2 : 1-102, 1917- 
20) suggested that the tables "showing the New England states 
from which each species has been recorded . . . have little value 
as these are political regions while it is the physiographic and cli- 
matic areas that control distribution." The reason for the tables 
was not that the writer was ignorant of such a fact, — but be- 
cause, first as stated, he wanted the Manual to conform in plan to 
Dr. Calvert's List of Odonata (1905), one of the Occasional Papers 
on the Fauna of New England published by the Boston Society of 
Natural History; and secondly, because until many "political" 
stations are listed it is quite impossible to discuss the distributional 
characteristics of species intelligently. Even now there are many 
corners of these much investigated States where collecting is 
needed. Dr. Kennedy will agree ^ that only when it is possible 
to piece together township- into state-, and state- into continental- 
lists, can zoogeography be properly understood and studied. As a 
proof of the value of the publication of these tables two state-lists 
have been suggested, which when published, will be the result of 
intensive field work, valuable in the extreme, as units in a patch- 
work zoogeographical whole. 
J For he later says, "All odonate records in New England [are] of great 
value, for which reason Dr. Howe's Manual is especially opportune." 
