
61 
Inasmuch as the two censuses were taken three weeks apart (21 days) it was 
assured that on August 2-6 all Class IA, IB and IU broods had hatched since the 
July 11-17 count; or that Class IIIA, and IIIB broods from the July count were 
flying by the time the August count was taken. Either method of calculating total 
production gave approximately the same total but Method I is believed to be the 
most reliable in most instances. The 246 broods produced in 1953 is almost 
exactly the same as the 1952 production of 249 broods. 
From Tables II and VI we can now approximate the success of the breeding 
pairs or in other words the percent of successful females as shown in Table VII. 
It is extremely interesting to note that the over-all percent of successful 
females for 1953 was 42.8 compared to 42.4 for 1952. Mallard success dropped 
from 46.2 percent to 39.1 percent but broods of this species are the hardest of all 
to census due to their extreme wariness and their propensity for taking to cover at 
the slightest provocation and running for several hundred yards to other water areas. 
Increased cover and more abundant water in 1953 may have caused some broods to be 
missed so that the drop in success from 46 to 39 rercent may not be an actual one. 
A black labrador dog was used again in 1953 and it was agreed that in this habitat type 
a dog is essential if anywhere near a total count of broods is desired. The phenomenal 
success of the canvasback was evident from both nesting and brood data, and broods 
were much in evidence throughout the southeastern corner of Saskatchewan, 
Brood Averages 
Table VIII illustrates brood averages for 1952 and 1953, broods were recorded 
by 7 age classes in 1953 but for purposes of comparison, only the conventional age 
groups will be used. Brood averages were based on data collected either on the study 
area or in the general area surrounding. Class III broods were larger in 1953 for 
all species although there was no significant difference in Class O brood size nor in 
nest clutch sizes. Vastly improved cover conditions in 1953 could have reduced 
juvenile loss and caused the difference in size of Class III broods. 
Aerial VS Ground Voverage 
Several aerial censuses of the Redvers Study Area were flown to count both 
breeding pairs and broods. Flying was done in a Piper Super Cub 125 plane by 
Pilot-Biologist John Lynch and observer V. C. wonover. 
Data in Table 1X is taken from ground counts made on May 11-17 and June 1-6, 
a combination of which was used to arrive at total breeding population, The "adjusted" 
total for each species equals lone drakes and pairs actually counted plus a hen added 
for each lone drake. This was done for both aerial and ground coverage. Aerial 
counts gave adjusted totals on the following dates: 
May 12 May 16 May 17 May 21 May 24 
346 376 384 400 428 
An average of May 21 and 24 was used to compare with the calculated total breeding 
population. 
