
laying of fertilized eggs. Ornithological tables should, I think, 
also include data on the number of eggs that hatch, and in altricial 
birds the number of young that leave the nest. My own samples in 
this study were often so small that I avoided tabulating my results 
in terms of adjusted cohorts of 1000. To have done so would have 
magnified sampling errors and implied a far greater accuracy than is 
currently possible. 
(3) The number dying within the age interval in column x is usuall 
symolites ag ds The number of deaths is sometimes calied a sertaleey 
serlese 
(4) The rate of mortality is the number dying in a given age interval 
divided by the nunber alive at the start of that interval: tn avian 
life tables this result may be expressed as a percentage figure or a3 
the number dying per 1000. In vital statistics, mortality rates are 
customarily symbolized as q,; these rates are far more precisely cal- 
culated for humans than any mortality rates thus far computed for 
birds. While some other symbol (like m,) seems to me to be more ap-= 
propriate for the crude results I have often obtained, I have followed 
other biologists in using q, to identify mortality rates in this report. 
(5) The mean after lifetime is the average length of life remaining 
to each individual alive at the start of a given age interval. The 
term is synonymous with "life expectancy," “expectation of life," and 
"complete expectation of life." In human life tables this colum is 
often identified by the symbol e,. Pearl (191, pp. 513-515) has 
given a helpful table for quickly ascertaining elapsed time between 
any two dates in the direct computation of mean after lifetime, as in 
Farner's (1945) work on the American robin. This method is not in 
itself a precise one in many ornithological life tables, because the 
monthly mortality reported on banded birds is not necessarily represen-~ 
tative of unreported mortality in the same species. The dates on 
banding reports also contain other inaccuracies, which I will discuss 
later. Although Deevey (197) has described an excellent short cut 
in computations of mean after lifetime, I have avoided caluclating 
this statistic in the chapters that follow. 
(6) Survival rate is not ordinarily found in life tables but is 
mentioned frequently in quantitative studies of bird populations. 
Where I have occasionally used it, it is designated by the symbol s_. 
This of course is the same as 1 - Gye 
Basis of life tables 
Merrell (1946) pointed out at the AAAS meeting in Boston 
that life tables differ fundamentally in the way their primary data 
are gathered: 
Dynamic life tables summarize the survival and mortality of 
a given cohort (or age class) over a period of years. One could set 
up such a table for native Americans born in the year, say, 1850, and 
