498 
Journal of Agricultural Research 
Vol. XXIX, No. 1 (P 
weight. April, 1917, at Rounton, Eng¬ 
land was too cold, as were January, Feb¬ 
ruary, and March. There was a difficult 
lambing season with a small lamb crop, 
but the later months were better. 
August was too wet, with over 6 inches 
of rainfall, but the humidity was not 
high. In 1921 at Croyden the lambing 
season was favorable (fig. 9). The 
summer was particularly dry, accom¬ 
panied by humidity a little above the 
limits. Water had to be carted for 
livestock for six months. 
A comparison was made between the 
good and bad years at Urbana, Ill., 
from 1906 to 1922 {20). (The informa¬ 
tion obtainable was not so complete as 
that for England.) The rate of gains 
made by the lambs was taken as an 
indication of the effect of the summer 
conditions upon them. The mean 
weights of each breed—Southdowns, 
Shropshires, and Rambouillets—were 
plotted for the years 1907, 1911, 1913, 
1915. The weights were taken each 
month, and twice a month during 
some years. The mean temperatures 
and humidity for two-week periods 
were also plotted. Hammond has 
shown the normal growth of lambs to 
be regular. Figures 15 and 16 show a 
comparison of the growth or increase 
in weight with the mean temperature 
and humidity. The irregularities indi¬ 
cate that a high mean temperature 
accompanied by high humidity is corre¬ 
lated with the retarded growth of the 
lambs. In some cases there was a loss 
in weight for a two to four week period. 
The graphs presented here (together 
with others on file at the University of 
Illinois) show that retarded growth 
follows high temperature and humidity 
regardless of the d^ate or the age of the 
lambs. Curves drawn for 1913 (on file 
at the University of Illinois) show com¬ 
paratively uniform growth of lambs 
(fig. 11). The difficulty in making use 
of the weights of the university flock 
grew out of the different dates of lamb- 
ing ; evidently influenced to some ex¬ 
tent by the weather condition of the 
preceding mating season. Mean 
growth was rendered insignificant by 
this irregularity, and retardation in 
growth was noted at critical periods 
regardless of the age of the lambs. 
A mean relative humidity higher 
than 80 per cent and a mean tempera¬ 
ture higher than 70° F. seem to be 
detrimental. The sheep can stand a 
rather high mean temperature if the 
mean humidity is not over 60 or 65 per 
cent. The rate of growth was more 
uniform in 1913 than for the other 
three years plotted. The study has 
not been carried far enough to give 
definite limits of temperature and 
humidity. The Rambouillets, which 
come from a country where the tem¬ 
perature runs higher than it does in the 
section from which the Shropshire and 
Southdowns come, seem to be less 
affected by unfavorable conditions. 
The late-born lambs are more seriously 
affected by unfavorable conditions 
than the early-born lambs. 
A good sheep year at Urbana, Ill., 
closely approximates the limits of a 
good sheep year in England, although 
it has a wider range, being warmer in 
summer and colder in winter. Fig¬ 
ures 10, 11, 12, and 13 are hyther- 
graphs and climographs of the years 
1912, 1913, 1915, and from June, 1920, 
to June, 1921, respectively, at Urbana, 
compared with the limits of a good 
sheep year at Urbana. It was found 
that summers unfavorable for the 
growth of the lamb are rather common. 
The high humidity during the summer 
of 1912 was thus detrimental (fig. 10). 
During the hot summer of 1913, which 
was accompanied by a comparatively 
low humidity, the lambs made a regular 
and satisfactory growth (fig. 11). 
Although the summer of 1915 was cool 
and wet, the comparatively high 
humidity retarded the growth of the 
lambs, particularly of those born late 
in the spring (fig. 12). The period 
from June, 1920, to June, 1921, was a 
favorable year for sheep, and falls, 
almost entirely within the linlits of a 
good sheep year (fig. 13). 
HOUSING OR SHELTER FOR SHEEP 
The climatic limits under which 
sheep can be successfully kept may be 
extended by providing shelter and 
extra food. In the important sheep 
countries the sheep are seldom housed, 
and depend largely upon grazing for 
their food supply, although in some 
sections some additional food is pro¬ 
vided during the winter. Shelter or 
housing makes it possible to raise sheep 
successfully in sections which have 
low temperatures, heavy snowfall, or 
excessive cold winter rains. The types 
of shelter may vary from sheds, con¬ 
structed to break the wind and protect 
them from the cold rains, to warm well- 
constructed barns. In comparing 
graphs of these sections with the com¬ 
posite graph for good sheep countries, 
one should make some corrections for 
housing and various kinds of shelter. 
But no data have been found on how 
much correction is advisable. 
The sheep farm at the University of 
Illinois has two types of building— 
one a shed which is open to the south,. 
