4 BULLETIN 997, U. S. DEPARTMENT OF AGRICULTURE. 
The fuel consumption per day for the 2-plow tractors varied from 
about 18 gallons for fall plowing to about 11 gallons for drawing the 
hay loader. For the 3-plow tractors it varied from 23 gallons for 
plowing to 15 gallons for drawing the hay loader. The 2-plow trac- 
tors covered 6.6 acres per day in spring plowing and the 3-plow ma- 
chines 8.6 acres. The quantity of fuel required per acre was 2.7 
gallons for each size. 
The average cost per acre of power for the plowing done with 
2-plow tractors was about $2 and with the 3-plow about $2.20. The 
cost of power for the plowing done with horses on these farms was 
about $2.90 per acre. Based on the present prices of feed, fuel, and 
oil (September, 1921), the cost of power for plowing with horses 
would be about $1.60 per acre, and with tractors about $1.70. 
For most of the other operations the cost of power furnished by 
horses during the year of the investigation was slightly less than 
that furnished by tractors. The cost per acre of power for disking 
with tractors was $0.67; with horses, $0.64; for cutting grain with 
tractors, $0.67; with horses, $0.59. These figures represent the cost 
of power only, and do not include either the cost of man labor or 
that of the implements used. 
The average cost per day of 2-plow tractors for drawbar work on 
the home farm was about $12.67, and of 3-plow tractors about $17.73. 
The total cost of power furnished by the tractors for drawbar work 
at home during the year averaged $341. Based on the present price 
of fuel and oil (September, 1921), the cost would be about $280. 
This drawbar work on the home farm constituted 76 per cent of the 
total work done by the tractors, and only 76 per cent of the total 
annual charge for depreciation, repairs, and interest on investment 
is included in it. No charges for taxes, insurance, or shelter are 
included in the costs for either tractors or workstock. 
Nine of these men started farming with tractors; the others in- 
creased the size of their farms by an average of about 20 acres after 
the tractors were purchased. No change occurred in the size of 172 
of the farms, 81 were increased in size, and 24 were decreased. 
On the 172 farms where no change in acreage occurred the number 
of workstock was reduced by 2.2 head, an average reduction of 26 
per cent. Forty-four of these 172 men did not reduce the number 
of workstock, 62 disposed of 1 or 2 head, 43 disposed of 3 or 4 head, 
and 23 of more than 4 head. On these 172 farms 1 horse was kept 
for each 28.0 acres (total acres, not crop acres) before purchase of trac- 
tors, and at the time of the survey there was 1 horse for each 37.7 
acres. For all the farms an average of 1 horse was kept for each 27.6 
acres before the purchase of tractors, and there was 1 for each 37.9 
acres at the time of the investigation. 
With the tractors doing the bulk of the work of plowing and fitting 
the ground, the cultivation of corn was the operation which required 
