84 
later than normal was anticipated, especially in divers. and teal. 
Production 
Road Count Observations - Table VI gives a comparison of the species and 
number of broods observed on the 12 transects from 1950 to 1954. As of July 25 
(the data comparable to previous years listed), the 1954 figures showed a reduction 
in mallard, gadwall, baldpate, redhead, and lesser scaup broods. Canvasback and 
pintail remained approximately the same with slight increases in blue-winged teal and 
ruddy ducks. The total number of broods observed at this time was 47 percent below 
the 1952 total with the greatest loss in mallards which showed a reduction of 76 
percent from the 1952 figure and 42 percent from 1953. Though high water may have 
allowed broods more cover in the flooded emergent vegetation and thus bias the counts, 
we feel that it could not account entirely for the discrepancy. The beatout data for 
the 120 study potholes (Table X) closely paralleled the above reductions. A check of 
the worksheets for the l2 transects revealed 58 pairs and 81 lone females still 
present on July 25. This could indicate a potential of 139 additional broods. 
On the July 25 to August 12 rerun of the 12 transects, the total number of 
broods observed rose from 153 to 350, Table IX. Mallards jumped to 50 percent 
below the 1952 base and equal to the 1953 population. However, these figures cannot 
be compared as this count was completed nearly three weeks later than those of 
previous years, even though it was estimated that this season was two weeks later 
than normal, phenologically. The July 25 to August 12 data as compared to that up 
until July 25 reveals a 50 percent increase in mallard broods, 175 percent in baldpate, 
300 in green-winged teal, 186 in blue-winged teal, 166 in shoveler, 325 in redhead, 
31 in canvasback, 186 in scaup and 445 in ruddy. In other words, it appears that the 
survey had been completed before the peak of the hatch in most species probably due 
to the lateness of the season for divers and loss of earlier nestings in some puddlers. 
Table IV shows the classes of broods observed by species, road counts, and shows 
45 percent of-the total broods in Class I and 74 percent of the total broods in Classes 
I and II on August 12. This total is heavily weighed by normal late hatchers such as 
redhead, scaup and ruddy though mallards and blue-winged teal are well represented. 
Table V shows the distribution of the broods. Transects VI to XII are the northernmost 
and are generally in what is considered better diving-duck bahbitet. Table X reflects 
in part the data of Table VII. It shows the change in percent species composition of 
the late hatching divers for the two 1954 runs. Mallards still composed a lower per- 
centage of the total hatch as compared to that of other years but again, the percentages 
at such a late point in the season were heavily weighed by the late hatching divers. 
One conclusion seems to have shown itself above others from this late run of 
the transects. It appears that in order to gain a truer index of the actual results of 
the hatch, instead of hypotheses of the potential hatch as gained from early runs and 
from breeding population counts, a later closing date, as late as is possible, is 
desirable. On July 25, the picture portrayed by data gathered up until that time was 
not too rosy. It changed materially in the period from July 25 to August l2 due to late 
hatching divers and late dabblers. It is apparent that the data gathered during this 
period is not comparable to any previous data on hand but it may serve as a base for 
future studies on this area if they are to be continued. 
