4 BULLETIN. 747, IT. S. DEPARTMENT OF AGRICULTURE. 
other words, the average of all plants, both good and bad, between 
these limits -shows a variation of about 87 pounds of coal per 1,000 
pounds of butter, while the poorer plants show a variation of about 
261 pounds. The better plants, however, show practically no varia- 
tion in the fuel used per 1,000* pounds between plants of that size. 
The same condition appears in similar curves plotted from data 
obtained on fuel consumption in other years ; hence there must be a 
definite reason for the variation in fuel cost in the creameries 
mentioned. 
A careful study of the reports was made to determine, if possible, 
the reasons for the differences, with the following conclusions : : 
1. "When the annual output was between 250,000 and 350,000 
pounds of butter the size of the boiler and engine was increased, as 
well as the length of time of firing the boiler, compared with a* 
creamery of smaller output. 
2. At about the annual production shown above, the size of the 
creamery building was .increased and the ratio of the coal required 
to heat the building to the quantity used for making butter is greater 
than in the smaller plants. 
3. Often when the- capacity of the creamery had reached an annual 
production of between 250.000 and 350,000 pounds of butter an 
additional helper was employed and the firing of the boiler was 
intrusted to him. It is probable, however, that he was not so eco- 
nomical in the use of fuel as the buttermaker. 
4. Frequently when the annual capacity reaches 250,000 pounds it 
is necessary to make two churnings daily, thus increasing the time 
of using power. 
While probably there are other reasons why the fuel consumption 
per 1,000 pounds of butter made does not decrease in proportion to 
the increase in the capacity of the plant, those mentioned are suffi- 
cient to account for much of the discrepancy. 
The curves of figure 2 show the average sizes of engines and boilers 
for plants of different capacity, and of figure 3 the average number 
of hours the boilers are fired. That portion of the curves between 
250,000 and 350,000 pounds of butter made annually should be noted 
especially. Between these points the average time of firing the boiler 
is increased from 7.3 to 8.5 hours, or 1.2 hours daily. There is prac- 
tically no difference in the size of the boiler or the engine, or in the 
length of time the boiler is fired, between plants where 150,000 and 
250,000 pounds of butter was made. In creameries below 150,000 
pounds' capacity, however, both the size of the equipment and the 
time of firing decrease. Above 350,000 pounds' capacity the size of 
the equipment and the total time of firing increases, but at a less rate 
than in the smaller plants. 
