Oct. 28. 1918 Experiments in Field Technic in Plot Tests 



257 



Table III. — Average yields of Ligowa oats {Minnesota 281); Haynes Bluestem wheat 

 {Minnesota 169), and Manchuria barley {Minnesota 105), grownin plots of the same 

 length, but varying in width and border rows removed 



From the data given it is evident that shape as well as size of plot is 

 of prime importance in considering increases in the yields of crops grown 

 in plots surrounded by alleys. The larger the plots, provided the ratio 

 of width to length remain approximately the same, and the nearer they 

 approach the form of a square the smaller the percentage of total area 

 exposed to alley effect and the less the increase in yield due to this cause. 



However, comparatively long and narrow plots can be more easily 

 sown with farm drills and harvested with binders than square plots of 

 the same size. The removal of a sufficient number of border rows from 

 the margins of long, narrow plots to obviate alley effect would appear to 

 be the proper procedure. Leaving no alleys between plots at seeding 

 time and removing a sufficient number of rows from each variety after 

 full heading to provide a pathway of the desired width would accom- 

 plish similar results. 



In Table IV are summarized the average yields, in bushels per acre, 

 together with the standard deviations for four :^-acre plots (ap- 

 proximate) of each of 11 varieties of oats, 5 varieties of wheat, and 4 

 varieties of barley (a) with no border rows removed, (b) with one border 

 row on either side of each plot removed, and (c) with two border rows 

 on either side of each plot removed. The yields included in this table are 

 for only one season and are not intended as a variety test report. They 

 are used only in so far as they supply data for the consideration of border 

 effect in variety trials for one season. Comparing the average yields 

 from the four ^V'^cre plots of each variety, it is significant that in 

 each instance the yields with no border rows removed are the highest, 

 with one border row removed the next highest, and with two border 

 rows removed, the lowest. Considering the average yields for all varie- 

 ties, the oats, with no border rows removed, yielded 9.14 bushels; the 

 wheat, 5.28 bushels; and the barley, 8.48 bushels per acre more than 

 where two border rows were removed from either side of each plot. When 

 only one border row was removed from either side of each plot, the oat 

 varieties yielded 2.20 bushels, the wheat 1.99 bushels, and the barley 

 2.86 bushels higher than when two border rows were removed. 



