
Table 8.--Effect of Combining Mortality and Survival Data in a 
Single Life Table Involving Constant Mortality Factors 
This table continues the study of the theoretical population in 
table 7. It is assumed that 10 per cent of the birds alive at 
the start of each year are recaptured and released, that one-half. 
of the number shot are reported, that 5 per cent of the number left 
are retrapped, and that 1 per cent of the diseased birds are dis~ 
covered each year. All recoveries are then grouped to form a life 
curve for a time~specific table and are used as a mortality curve 
in a dynamic table. 



"Frequency by Age Intervals ~~ 
Sample O-I”_ i-2 2-3, ~— 3-4 C-5~—COS-6~=C«~rTTtco htaad’ 
Recaptured (10% of 1) 100 30 9 2.70 0.81 0.24 12.75 
Shot (50% of 2) 300 90 27 8.10 2.43 0.73 428.26 
Retrapped (5% of 3) 20 6 1.8 0.54 0.16 0.05 28.55 
Found dead (10% of i) 10 3 0.9 0.27 0.08 0.02: 14.27 
“Total 430 4129 30.7 11.61 3.40 1.0, °° 
TIME-SPECIFIC LIFE TABLE 
Alive at start 1,30 129 38.7 11.61 3.48 1.0; 
Deaths 301 90.3 27.09 8.13 2. 1.04 
Mortality rate 70 70 70 70 70 §=©100 
DYNAMIC LIFE TABLE 
Deaths 4,30 129 38.7 11.61 3.48 1.04 
Alive at start 613.83 183.83 5).8316.13 4.52 1.04 
Mortality rate 70 70 71 = ~=72 77 100 

When birds are retrapped by a single bander or a research station, how- 
ever, the survival reports thus obtained do not necessarily involve a 
random sampling of the age distribution in a bird population. Time- 
specific analyses of common terns retrapped on Cape Cod and reported 
by the Austins (1932, 1938, 1942) contain an additional variable which 
is examined in a later section. 
Statistical treatment of fluctuating mortality rates 
Are dynamic and time-specific analyses reliable when the over- 
all mortality rate is fluctuating from one year to another? That a 
Fiuctuation does exist between the first- and second-year mortality 
rates of many birds has long been surmised by ornithologists. As early 
as 1926, Whittle and Whittle attempted to measure the difference for 
purple finches, but the homing responses of young and adult birds have 
usually frustrated attempts to analyze retrapping data along these lines. 
In more recent years, Lack (193a, 193b, 19h6a, 1946b) has successfully 
used mortality reports in centralized banding files to demonstrate the 
degree of this fluctuation in a number of British birds; Buss (196) 
has reported a similar condition in winter-trapped ring-necked pheasants; 
and more extensive fluctuations have been described for European cormorants 
