1911-12.] Effects of Seasonal Changes on Body Temperature. 115 
In October 1909, when the work was commenced, the birds were pullets, 
having been hatched in the previous April or May. They were leg-branded 
and numbered, and twenty-five were picked from each pen for observation, 
except in the case of IV., where fifteen was the number started with. In 
the course of the first year several of these had disappeared, and by the end 
of the second year not more than forty-one out of the original 140 remained. 
A few had died, and the rest had been sent to market or disposed of in 
other ways. 
In the accompanying tables (I. to VI.) the actual figures are given for 
each individual in degrees centigrade, with the date and the temperature 
of the air in the pen at the top of each column. The number second from 
the bottom in each column represents the mean temperature for all the 
individuals in the pen observed on that day ; while the last number, printed 
in italics, shows the mean for those on which the observations were con- 
tinued for two years. 
In the Charts I. to VI., the upper curve represents the mean temperature 
of all the hens in each pen from which monthly records were obtained for 
one year ; the second curve the same for those which were under observa- 
tions for two years. Underneath these are plotted out curves for the air 
temperature in the pen and the barometric pressure as it stood at the time 
when the observations were made. The curve of egg-production * at the 
bottom will require a word of explanation. Suppose that one has a flock 
of one hundred hens, and that by this flock one hundred eggs are laid 
daily, then the egg-production is said to be 100 per cent. ; if sixty eggs are 
laid per day by the one hundred hens, the egg production is 60 per cent., 
and so on. In Charts VII. and VIII. the curves are plotted from the mean 
figures for all the hens that were examined on the same day, that is, when 
the temperature and other conditions were approximately the same, for in 
order to minimise errors due to accidental causes, it is desirable that the 
averages be taken from as large a number of individuals as possible. 
It is conceivable that atmospheric conditions other than temperature, 
such as humidity, sunshine, strength and direction of wind, etc., may 
affect body temperature, and in order to ascertain to what extent this 
is the case, I have appended a fairly complete weather table for the days 
on which the temperature records were made. This, by the kindness of 
* This curve of egg-production is taken from a paper by Rice, Nixon, and Rogers 
(Bulletin 258, College of Agriculture Publications , Cornell University, September 1908). 
It was obtained from trap-nested hens living at the same station and under the same 
conditions as those on which the present temperature observations were made, and indeed 
Professor Rice informs me that this curve of egg-production is practically correct for hens 
anywhere in New York State. 
[For continuation see p. 130. 
