September 6, 1912] 



SCIENCE 



319 



TABLE I 



Deviation of Yields of Varieties from the Yearly Means 



Rivet (Red) 



White Chaff (Red) . . . 

 Club Wheat (Red) . . . 

 Golden Drop (Red) 



Hallett's 



Bole's Prolific (Red) . . 

 Hardeastle (White) . . . 



Red Rostock 



Red Langham 



Bristol Red 



Red Wonder 



Red Chaff (White) . . . 



Browick (Red) 



Casey's White 



Red Nursery 



Wooly Ear (White) . . 

 BurweU (Old Red 



Lammas) 



Golden Rough Chaff 



(Red) 



Chubb Wheat (Red) . 

 Original Red (Hallett's) 

 Victoria White 



(HaUett's) 



White Chiddam 



Hunter's White 



(Hallett's) 



Number of Lots .... 



Averages 



+ 3.! 



+ 7.3 

 + 1.4 



+ 4.8 



- 1.4 



- 2.8 



- 1.0 

 + .6 

 + 3.1 



- 2.3 

 + 1.9 



- 1.0 



- 1.1 



+ .8 



- 3.8 



- 2.2 



+ 1.6 



- 5.3 



- 5.3 

 19 



+ 3.5 



+ 7.5 

 + .5 

 + 4.2 



+ 1.5 



+ 2.1 



+ 1.5 



- 5.3 



- 1.8 

 _ .2 

 + 3.0 

 + .5 



- 3.0 



- 2.3 



- 7.0 



+ 3.0 



- 3.5 



+ 9.3 

 + 1.8 

 + 8.7 



+ 5.4 



+ 6.4 



+ 3.2 



+ 7.5 



- 4.7 

 + .7 



- 1.7 



- 3.5 



- .3 



- 1.3 

 -11.7 



- 1.8 



- 3.7 



- .3 



- 3.0 



- 2.4 



- .5 



- 7.0 



- 2.5 - 



19 2 



+16.3 



+ 4.4 

 + 8.9 



+ 1.1 



- 2.6 



- 1.1 

 + 3.1 

 + 2.4 

 + 2.7 

 + 4.4 



- 1.9 

 + .4 

 + 1.4 



- 9.6 

 + .6 



- 3.4 



+ 1.4 



- .2 



- 7.1 



- 6.4 



- 8.7 



- 5.3 



22 



+11.6 

 + 3.4 

 + 9.8 



+ 1.3 



+ 7.5 



- 2.9 

 + .6 



- 1.9 



- 5.2 



- 3.6 



- 2.5 

 + 1.7 

 + 2.2 

 + 2.2 



- .7 



+ 1.7 



+ 2.0 

 + 1.5 

 -10.8 



- 3.0 



- 4.4 



-10.4 



22 



+ 7.0 

 + 5.1 



+ 5.9 



- 1.1 



+ 1.5 



2.5 



+ 6.7 



+ 5.5 



+ 6.6 



+ 6.6 



+ 1.9 



- .8 



+ 3.5 



- .1 

 + 1.7 

 + 1.3 



- 3.4 

 + 3.0 



- 5.0 

 + 4.1 



- 4.1 



- 4.1 



- 2.2 



- 2.4 



- 1.4 



- 5.0 



+ 1.0 



22 



+ 1.2 

 1.3 

 1.9 

 2.0 

 .1 

 2.3 

 5.3 



- 6.5 



- 1.4 

 + 1.5 



- .3 



- 5.3 



+14.3 



+ 7.2 



+ 9.2 



+ 1.0 



+ 1.0 



+ 2.2 



+ 5.2 



- 1.0 



+ .3 



+ .3 



4.5 



+ 2.3 



3.0 



+ .5 

 +10.5 

 + 1.0 

 -12.0 

 + 5.3 

 + 1.1 

 + 1.5 



1.7 

 + 4.0 



7.7 



+ 5.7 



+ 8.0 



3.1 



+ 4.3 



- 2.3 



- 4.0 



- 4.0 



- 3.5 



- 5.5 



- 5.0 

 + 3.3 



+ 3.5 



- 5.1 

 +10.4 



- .5 



+ 6.5 



- 6.1 

 + .3 



+ 2.9 



+ 7.2 

 - 9.2 



+ .8 



- 3.6 



- .5 



- 2.4 



- 1.7 



- 4.9 



- 2.9 - 9.5 

 22 20 



- 3.1 

 20 



- 8.3 

 + 3.3 



- 1.3 

 20 



- 2.5 

 + .6 



11 

 10 



7 



11 

 11 



10 

 226 



50.7 36.8 42.5 



42.9 1 51.8 1 20.5 



24.1 



46.5 



sted) in this manner. We note that the devia- 

 tions in the upper portion of the table are 

 generally positive, while those in the lower 

 half are generally negative. There are, how- 

 ever exceptions even in sign and the magni- 

 tude of the deviations varies greatly. 



How low the prediction value of a single 

 yield is may be seen at once by correlating be- 

 tween the relative yield of the same variety 

 in different years. Symmetrical intra-elass 

 tables^ or condensed tables may be formed or 

 the coefficient may be calculated from the 

 moments of the deviations of the individual 

 varieties by a convenient formula to be pub- 

 lished shortly. We find 



r = .266. 



It is most instructive to compare the corre- 

 lations between the relative yield of the dif- 

 ferent varieties in the same year. This fur- 

 nishes a measure of the influence of season. 



'Amer. Nat., Vol. 45, pp. 566-571, 1911; also a 

 second paper, Amer. Nat., ip press. 



Table II. shows the deviations of the yields 

 for each year from the mean yield of the va- 

 riety for all the years it has been grown. That 

 season has an immensely greater influence 

 than variety in determining yield is obvious 

 at once from a comparison of Table II. with 

 Table I. The coefiicient of correlation be- 

 tween the deviations of the different varieties 

 from their means in the same year is 

 r=.837. 



It is evident that with such a relatively low 

 value of the correlation between the yield of 

 varieties in different seasons, little importance 

 can be attached to such " variety tests " as 

 have been generally carried out in Agricul- 

 tural Experiment Stations. Yet' in the fu- 

 ture development of breeding, variety testing 

 must hold a most important place in station 

 work, for obviously it is idle to breed new 

 varieties unless they can be certainly demon- 

 strated to be superior to those already exist- 



