22 BULLETIX 1300. U. S. DEPARTMENT OF AGRICULTURE 
ACREAGE 
Variations in acreage 'Table 3^ show a positive correlation in suc- 
cessive years ( + 0.36) persisting to a slight extent for two years 
( + 0.14). Aside from this tendency no important correlations with 
preceding condition are brought out in the present data. The acreage 
planted in a given year has no relation to the preceding crops ( + 0.04, 
+ 0.01). "What relation there is to corn price is negative ( — 0.22 the 
second preceding year. —0.10 the preceding year). Thus the acre- 
age is clearly not increased to an appreciable extent because of a 
scarcity of corn and because of high prices in the preceding year. 
Slight positive correlations are shown with pork production of the 
preceding year ( + 0.27 summer. —0.17 winter), as well as with that 
of the following year: and corresponding negative correlations are 
shown with pork prices ( — 0.29 summer, —0.11 winter). 
Apparently a large acreage tends to be planted because a large 
number of hogs is on hand which must be fed. and a small acreage 
because of a small number of hogs on hand. Altogether, however, 
acreage is determined only 5 to 10 per cent by the preceding hog 
situation. There is about 13 per cent determination by the acreage 
of the preceding year, leaving more than 75 per cent determination 
by factors other than the preceding corn and hog situation. 
YIELD 
The fluctuations in yield (Table 3) are undoubtedly due in the 
main to the weather, which is not dealt with here as a factor. The 
correlations between yields in successive years, however, throw light 
on the important question as to whether there are cycles in yield, 
determined presumably by cycles in the weather. The present data 
give no indication of the existence of any such cycle. There is ac- 
tually a negative correlation between successive yields ( — 0.15) and 
also between yields two years apart ( — 0.15). Any cycle of longer 
period than four years tends to produce a positive correlation be- 
tween successive years, unless this is eliminated in fitting a trend to 
the data. Inspection of Figure 2 shows that a cycle of 40 years may 
perhaps be eliminated in this way, but no shorter cycle can be affected. 
Thus the present data indicate that no cycle of less than 40 years 
can be of more than minor importance in determining yield and crop. 
While weather conditions are undoubtedly the most important 
factors, the average yield in the country as a whole may also be in- 
fluenced by the distribution of the crop. Thus a relatively large acre- 
age in the Corn Belt, associated with a small acreage elsewhere, will 
result in a high average yield for the country as a whole even under 
ordinary weather conditions. The slight positive correlation be- 
tween acreage and yield ( +0.19) possibly indicates that fluctuations 
in acreage are greater in the high-yielding sections of the country 
than in the low-yielding sections. Similarly the indications of a 
positive correlation between yield and pork production of the pre- 
ceding year ( + 0.20 summer/ - 0.22 winter), as well as in that fol- 
lowing, indicate that an excess of hogs in the Corn Belt determines 
the planting of an increased amount of corn in this high-yielding 
section. None of these factors are of much importance. The yield 
seems to be determined from 5 to 10 per cent by the preceding hog 
