470 
Bulletin Xo. 1G2.—1915. 
and are of little or no consequence so far as the general results are 
concerned. 
It has been found inexpedient to attempt any classification on the 
basis of the cause of death, since in so many cases this could not be 
determined and has not been uniformly recorded. Death may result 
from a great variety of causes, among which perhaps the most 
important are lack of constitutional vigor, neglect by the parents 
(especially during cold weather), specific infections, and accidents, 
especially those resulting from the fighting and interference of adult 
birds. 
Figure 1 shows a high incidence of deaths at time of hatching 
which rapidly declines to a minimum on the fifth day, after which 
there is a correspondingly rapid rise to a marked mode from the tenth 
to fifteenth days. The first two or three days and the tenth to 
fifteenth days after hatching appear, therefore, to be the most 
critical periods in the life of the young squabs. The minor irregular¬ 
ities probably have little meaning, and in most cases no explanation 
is apparent, as for example, the much greater number of deaths on 
the eleventh day than on the twelfth. On the other hand the greater 
height of column 1 over the 0 column is probably due to the fact that 
many squabs living only a few hours after hatching are really not 
recorded as dead until the following day. There can be no doubt, 
however, that the marked depression in the death rate at the fifth 
and sixth days is of significance, and this point has therefore been 
chosen as the limit between the deaths at hatching and immediately 
following, and those grouped around the age of about two weeks. 
From this time the number of deaths drops off again rapidly, and the 
curve then swings out gradually to about 50 days, after which there 
is a steady but slight decrease in the number of deaths per month for 
the full period of nearly six and one-half years since the first squabs 
were produced in our experimental work. It must be borne in mind, 
however, that this gradual decrease in the number of deaths recorded 
does not necessarily imply a decreasing death rate, in fact probably 
the reverse is true. The death rate at any age should be based on the 
