406 
Journal of Agricultural Research 
Vol. XI, No. 9 
The correlation coefficients obtained for the various crops are given 
in Table III. 
Tabus III .—Correlations between the differences in yield between the border rows of a 
variety and the differences in height between neighboring border rows of unlike varieties 
of cereals 
Cereal. 
Designation of test. 
Number 
of tests. 
Correlation coeffi¬ 
cient. 
Barley. 
Do. 
Oats. 
Do. 
Spring wheat. 
Do. 
Winter wheat. 
Do. 
Plant Breeding Nursery. 
Farm Crops variety test. 
Plant Breeding Nursery.... 
Farm Crops variety test. 
Plant Breeding Nursery.... 
Farm Cropse variety test. 
Plant Breeding Nursery. 
Farm Crops variety test. 
57 
48 
218 
121 
*93 
50 
183 
38 
-0. 5i9±o. 065 
— . 073 ± .097 
- . III± .045 
+ • * 55 ± 
+ . 060± . 048 
4 - . lOIzb . 094 
— . 092 ± . 049 
- • 337 ± -°97 
In the results in Table III as given for barley (Plant Breeding Nursery) 
each test is an average of three replications as the various varieties and 
strains are grown in the same order for each replication. In the winter- 
wheat results (Farm Crops Section) each test is an average of five plot 
results. With the exception of barley in the Plant Breeding Nursery 
and winter wheat in Farm Crops work, each test represents a single plot 
result. 
In a consideration of* these results certain environmental conditions 
are of interest. Until nearly the end of the crop season the barley and 
winter wheat in the Plant Breeding test stood up comparatively well. 
The oats and spring wheat stood up well until after heading and then 
nearly all oat plots lodged very badly, the spring wheat lodging con¬ 
siderably. A serious epidemic of black stemrust {Puccinia graminis 
tritici E. and H.), together with a very hot period shortly after heading, 
also caused serious reductions of yield in the Plant Breeding spring-wheat 
plots. In the Farm Crops work all cereals (with occasional exceptions) 
stood up well. 
The correlation coefficients for the Plant Breeding work show con¬ 
siderable effect, owing to the height of adjacent rows on the border rows 
of the same variety in the barley tests and much smaller effects in the 
oats and winter-wheat tests. 
In the Farm Crops test thfe correlation for the winter wheat is —0.337 
±0.097. The odds against an occurrence of a deviation as large as 3.5 
times the probable error is 1 to 82. There is only a slight indication of 
the effect of height in the barley tests, which is not great enough to be 
significant, and none for the oats or spring-wheat tests. 
CORRELATIONS FOR YIELD COMPETITION 
The effects of yield of adjacent rows on the border rows may be con¬ 
sidered in a somewhat like manner as in the effects of height. The 
