INPUT AS RELATED TO OUTPUT 
It would seem from this table that there is a somewhat definite 
relation between the quantity of grain fed per day and the quan- 
tity of roughage fed. Only one of the droves receiving under 10 
pounds of grain per head per day received as much as 10 pounds 
of hay, and most of those receiving larger quantities of grain also 
received more roughage. However, the grain fed per day varied 
both with the weight of the cattle and the length of time they were 
kept on feed, as shown in Tables 7 and 8. Before the true relation 
of roughage input to grain input can be determined, the relation with 
these other factors must be taken into account. 
Table 7. — Droves classified by grain fed per day and weight of cattle at start of 
feeding period 
Rate of feeding grain (per head per day) 
Total 
droves 
Average weight at beginning of period 
5-9 
pounds 
10-14 
pounds 
15-19 
pounds 
20-24 
pounds 
25 pounds 
and over 
800 to 899 pounds - -- 
4 
5 
3 
7 
6 
9 
1 
7 
10 
4 
2 
23 
900 to 999 pounds 
3 
4 
2 
1 000 to 1 099 pounds 
1,100 to 1,199 pounds... . . ... ... . 
4 
8 
23 
23 
9 
Table 8. — Droves classified hy grain fed per day and length of feeding period 
Rate of feeding grain (per head per day) 
Total 
droves 
Number of days on feed 
5-9 
pounds 
10-14 
pounds 
15-19 
pounds 
20-24 
pounds 
25 pounds 
and over 
60to89 
2 
4 
7 
6 
3 
3 
7 
7 
3 
2 
1 
3 
3 
3 
8 
90 to 119 
14 
120tol49 
2 
2 
1 
1 
19 
150to 179 
11 
180 to 209 ... . .. . 
7 
210to239 
2 
4 
240to269 
1 
270to299 . 
1 
1 
300 to 329 
1 
1 
1 
330to359 . 
1 
4 
8 
23 
23 
9 
67 
The length of time on feed is highly correlated with the initial 
weight of the animal ; the lighter the animal, the longer it was kept 
on feed. If these intercorrelations are allowed for by the use of the 
net correlation analysis, the relations expressed in Table 9 appear. 
It is now evident that the rate of feeding grain is not correlated with 
the rate of feeding roughage (fig. 3). When the effect of the weight 
of the animals and the length of time on feed are eliminated, the 
correlation between the rates of feeding grain and roughage dis- 
appears. The multiple correlation between the three factors and 
the amount of grain fed per day was R=0.677±0.045, indicating 
that the input of corn is largely governed by the two factors, weight 
of animals and length of feeding period, and, as indicated by the net 
correlations, is more closely related to the weight of the animal than 
to the length of time on feed. 
96698°— 24 2 
