64 BREEDING CROP PLANTS 



nating rows show a greater difference than the same varieties 

 grown in alternating blocks. Perhaps the extreme effect of com- 

 petition is shown by different varieties of corn grown in the same 

 hill. 



In the plots consisting of three rows each, grown at the Minne- 

 sota Experiment Station in 1916, a study of competition was 

 made. When varieties of different heights were grown in adja- 

 cent plots a considerable effect was obtained in the yields of border 

 rows in the barley and winter wheat nurseries. The effect of 

 competition has been observed at other experiment stations and 

 various ways of overcoming its possible vitiating influence have 

 been suggested. 



One of the easiest and most effective means of eliminating this 

 source of error is by the use of sufficiently wide borders which are 

 discarded at harvest. In the case of plots, consisting of a single 

 row, it is possible to make the planting plan in such a way as to 

 minimize effects of competition. The rows should be laid out 

 north and south and the varieties and strains most nearly alike 

 in habits of growth should appear side by side. At best this 

 method can do no more than decrease the error due to compe- 

 tition, while the elimination of effective borders overcomes com- 

 petition. The use of borders necessitates a larger experimental 

 area and is somewhat more expensive for a given number of 

 trials. 



CLIMATIC VARIATIONS 



One other disturbing factor to be considered in conducting plot 

 tests is variation induced by weather conditions. Its presence 

 is so obvious to any one who has worked with growing crops that 

 further comment is hardly necessary. In a year of deficient rain- 

 fall the varieties best qualified to subsist under a minimum water 

 supply will yield most. Some seasons are better for the growth 

 of early maturing varieties than for late ones. An epidemic of a 

 plant disease like rust may be fostered or hampered by weather 

 conditions. The question arises, how may errors due to this 

 source be overcome? Conducting an experiment over a period 

 of years is the only effective means at the disposal of the investi- 

 gator. The strain which fluctuates the least from year to year 

 and also gives a high average performance is most valuable for 

 the farmer. 



