
Before presenting a set of ornithological life tables in 
Part II of this report, it seems necessary that we further examine 
the mathematical weaknesses of both the dynamic and time-specific 
methods of analyzing mortality reports of banded birds. Although 
this examination is tedious in that it requires mental concentra- 
tion on minute details, it will reward us by leading to a better 
understanding of analytical methods encountered in this type of 
research. The two important conditions to be noticed here are (1) 
constant and (2) fluctuating annual mortality rates. 
If only one mortality factor is represented in the mortalit 
reports, Will this yaens accurate data on the total annual mortalit 
rate? This circumstance is closely approached in recovery reports of 
banded ducks and geese, where as many as 97 per cent of the recoveries 
may come from hunters. The actual percentage of a population taken 
by duck hunters is still unknowm. Without attempting to guess even 
the approximate dynamics of such populations, let us assume that 
60 per cent of the individuals alive on September 1 in species A are 
shot each fall, and that after the hunting season 25 per cent of the 
survivors are annually wiped out by accidents, diseases, and preda- 
tion. The turnover in such a theoretical population is shown in 
table 7. In connection with this table, the following things should 
be noticed: 
(a) By definition, mortality rate equals the number dying divided 
by the number exposed to the risk of dying. This latter value of 
course decreases with each death. In actuarial practice, the annual 
mortality rate is conventionally based on the number alive at the 
start of each year or age level. This compromise is crude but 
nevertheless inescapable. In the table under study, the 25 per cent 
mortality for disease, etc. was computed on the basis of the number 
exposed to this risk at the end of the hunting season; it actually 
represcote only 10 per cent of the number alive at the start of the 
year. 
(bo) In game species, the number of birds crippled and lost by hunters 
can reasonably be regarded as a more or less fixed percentage of the 
number shot. The inability of hunters to report the band number on 

When mortality factors operate at different but consecutive parts 
of a year, they affect the overall mortality rate as follows: 
mM. = 1 = (lem')(1-m'!)(l-m''!) etc. 
where is the overall mortality rate for the year, m', m'', and m''' 
the mortality rates for specific factors operating at consecutive 
periods. If 3 factors (m, M5, and za) operate contemporaneously, then 
m 2M, +m +m C 
31 
