July i, 1921 
Further Experiments in Field Technic in Plot Tests 493 
highest yielders under each method of test, and lines have been drawn 
at what may be considered the discard point for the season. Removal 
of one and three border rows, respectively, from either side of each 
plot lowers the discard point in the test. Due to the removal of the 
border rows, there are a number of changes in rank; but in general the 
varieties above the discard point with no border rows removed remain 
there under the other methods of test, except as has been previously 
noted. 
For the wheat variety test the least significant differences in bushels 
per acre for the various methods of test are as follows: No borders 
removed, 1.67; one removed, 1.71; two removed, 1.92; three removed, 
2.14. Lines have been drawn at what may be considered the discard 
point for the season. When no border rows were removed, Marquis 
gave a lower yield than Mindum or Preston; but with one border row 
removed from either side of each plot, Marquis equaled the other two in 
yield and maintained that rank with two and three border rows re¬ 
moved. 
In the barley variety test the least significant differences were found 
to be for no border rows removed, 3.94 bushels; one row removed, 3.62 
bushels; two rows removed, 3.58 bushels; and three rows removed, 3.97 
bushels. 
The removal of one border row from either side of each plot put the 
yield of one variety below the discard point adopted. The removal of 
three border rows put Svansota below the discard point. Removal of 
two border rows also raised Improved Manchuria from a rank of 6 to a 
rank of 4 in yield. 
