Journal of Agricultural Research 



Vol. XII, No. 5 



of trees chosen there is a gradual variation of soil in this direction, and 

 these plots therefore include both high- and low-yielding trees in about 

 equal proportions, the difference from plot to plot being small, as shown 

 in Table X. 



TabIvE X. — Effect on variability of changing the shape of plots 



Kind of trees. 



Total navel-orange grove (Arlington) 



Do 



Do 



Portion of navel-orange grove (Arlington), 

 256 trees 



Do 



Do 



Do 



Do 



Do 



Walnuts, 1915 



Do 



Do 



Do 



Do 



Do 



Do 



Apples, 1914-15 



Do 



Do 



Do 



Do 



Do 



Do 



Number 

 of trees 

 per plot. 



16 

 16 

 16 

 16 

 16 



4 

 4 



9 

 9 

 9 



24 



24 



16 

 16 



24 

 24 



Shape of plots. 



3 by 3 



I by 9 N.-S. . 

 I by 9 E.-W. 



1 by I 



4 by 4 



2 by 8 N.-S. . 

 2 by 8 E.-W. 

 I by 16 N.-S. 



1 by 16 E.-W 



2 by 2 



I by 4 



3 by 3 



I by 9 N.-S. . 

 I by 9 E.-W . 



4 by 6 



1 by 24 



2 by 4 



I by 8 N.-S. . 



1 by 8 E.-W. 



4 by 4 



2 by 8 N.-S. . 



4 by 6 



I by 24 



Coefficient of 

 variability. 



28. i8±l. 48 

 28. 92 ± I. 46 

 28. 36±i. 43 



34- 47 ± I- 14 



16. 77 ±2. 05 



17. i7±2. 10 

 iS.39±i.88 

 16. 36 ±2. 00 

 io.09dbi. 31 



31. 09±i. 93 

 30.40±i. 89 



23. 94±2. 32 



26. 23±2. 57 

 22. 46±2. 16 



21. 62±3. 25 

 6. 56±i. 00 



19. 32±I. 81 



18. 7o±i. 74 

 21. 35±2. 18 



15- i3±i-97 

 16. 4i±2. 15 



i3-75±2.36 

 12. 8i±2. 20 



Changing the shape of plots of walnut trees had little effect upon the 

 coefficient of variability except in the case of the larger plots. The 

 coefficients vary only slightly from one another, regardless of shape, in 

 the case of the 4- and 9-tree plots. In the 24-tree plot, however, there 

 is a great difference between the 4- by 6- tree plot and the i- by 24- tree 

 plot in favor of the latter. It should be borne in mind, nevertheless, 

 that a plot of 24 walnut trees is an abnormally large plot, and in the 

 example just referred to it is a plot 250 feet by 350 feet, compared with 

 one 100 feet by 1,250 feet. This may take into consideration marked 

 variation in productivity, even on apparently uniform soil. 



In the case of the apple trees there is little difference between the 

 variability of plots of equal size but different shapes, even in the case 

 of the 24-tree plots. Regarding the question of the shape of the plot, 

 it therefore appears that in most cases there is no difference in the varia- 

 bility of a linear or a square plot. 



