435 ote 
title them to consideration as actiial though temporary residents, but 
as it is, we can only say that in our judgment somewhere between 50 and 
75 per cent. of our 166 recorded and counted species were found in 
numbers sufficient to make them worthy of note as effective agents, for 
good or evil according to their habits, at some time and in some part 
of the state. In this very general statement, however, we are making no sex 
application of the well established fact that birds in great variety of 6 
a widely different ordinary habit and habitat may concentrate locally and “oA 
for a considerable time in a habitat which through some exceptional 
development offers them unusual inducements.* 
On the whole it may be reasonably said that the general outcome 
of our survey warrants the statement that few kinds of birds are really ' 
insignificant everywhere and always and that it is a rational economic ; 
policy to preserve and protect all not known to be on the whole posi- 
- tively injurious and to decimate, and exterminate if possible, the few ae 
which can be definitely and positively so classed. | 
CONCERNING THE METHODS OF THE SURVEY 
We have frequently remarked upon the discretion to be used in 
applying the results of our survey, and we have ourselves frequently 
refrained from drawing seemingly warranted conclusions because of 
the obvious deficiencies of our data. Unable, of course, to make a com- 
.) plete census of the birds of the state at any time, we have resorted to 
# the method of random sampling; and unless this process is often re- Bi 
peated under every variety of circumstance and condition, the question ; 
may always be very properly raised whether the samples chosen ade- 
quately represent the whole. We have reason to regret the shortness A 
of the period over which the survey extended and especially the un- 
equal distribution of our data of observation in both time and space, 
largely due to our limitation to a single field party and the consequent = 
impossibility of making parallel observations in different places at the 
same time. As the phenomena we were studying were peculiarly sub- 
ject to seasonal changes, less rapid in summer and winter, but perplex- “4 
ingly so in spring and fall, it was logically necessary that we should 4 
have had at least three parties in the field at once, one for each section aoe 
of the state, all operating on similar programs especially as to times and 3 
areas. This being impracticable for us, profitable comparison of our =n 
data has been narrowly limited and considerable ingenuity has been 
called for in the adaptation and assimilation of data of different origit : 
in a way to make them fairly comparable. These differences were at- ‘ 
tributable to the conditions under which our methods of observation 
and record were used and not to the methods themselves, which seem to 
. . Oy 
us so well adapted to the end in view that we should not know how Stir 
— eee a 
*¥For an illustrative instance, see “The Regulative Action of Birds, upon eee 
Insect Oscillations’ by S. A. Forbes, Bul. Ill. State Laboratory of Natural History, a. 
Vol. I, Art. 6, pp. 1-32, May, 1883. i 
1: 
