weighted change for a region (Table 3). ‘The changes in counts in 
each State and Province, in the two regions, and in all areas com- 
bined were tested to estimate the probability that the observed 
mean difference from zero change could have arisen by chance of 
sampling (using a t test, see Snedecor's Statistical Methods, 1956). 
‘The change in number of woodcock per route is more under- 
standable when expressed as a percentage of the average number per 
route in the region. The weighted mean woodcock per route was 
5.68 for the area west of the Appalachians, 6.43 for the area cast 
of the Appalachians, and 6.06 for all areas combined. The weighted 
changes in woodcock per route (Table 3) were divided by the weignted 
mean number of woodcock per route to obtain an estimate of the 
percentage change in woodcock population in the western and eastern 
regions and for all areas combined (Table 4). 
Two standard errors of the mean (twice the square root of 
the weighted variance from Table 2) indicate approximate 95 per cent 
confidence limits for the weighted mean change. These limits are 
expressed as percentages in Table }. 
RESULTS 
Tables 3 and 4 summarize the results of the woodcock 
singing ground counts. As pointed out previously, the results cited 
here relate to the routes actually covered and not necessarily to 
the total breeding population. A 7.3 per cent increase was recorded 
in the West and a 9.9 per cent reduction in the East, for an over- 
all net change in all areas of - 1.9 per cent. Confidence limits 
around these estimates extend in the eastern unit between - 1 and 
~ 19 per cent; in the western unit, between + 28 and - 13 per cent. 
For both areas combined, confidence limits ranged from + 9 to ~ 12 
per cent. Observed changes in the individual States and Provinces 
were tested to see whether any was large enough to be statistically 
significant at the 0.05 level of significance; none was. 
DISCUSSION 
A serious decline had been feared this year by some field 
observers who were aware of the severe weather conditions in March 
1960; so it is encouraging to find that the present counts indicate 
that population changes may not have been great. The changes in ~ 
woodcock on individual comparable routes (Tables 5 and 6), by States 
and Proyinces (Table 3), and by regions (Table 4) clearly suggest that 
at least this segment of the eastern population of the woodcock did 
not suffer a severe setback in the spring of 1960. 
T 
