HARRIS: RELATIONSHIP OF OVULES TO SEEDS 253 
numbers of ovules per pod. The standard deviation may also 
be computed for the entire material. The slope of the regression 
line is | 
Z = .19833 — .04217 0. 
The slope of this line and the empirical mean deviations are shown 
in FIG. 3. 
It thus appears that while the relationship is an exceedingly 
slight one, in all the adequately large series of material from 
Meramec Highlands the capacity of the pods for maturing their 
seeds decreases as the number of ovules per pod increases. 
B. Analysis of The Data from the Vicinity of Lawrence, Kansas 
I have to thank my father, Mr. J. T. Harris, for the collection of 
a series of 100 pods each from 22 trees in the neighborhood of 
Lawrence, Kansas. All the trees grew in the same small field. 
The correlation between the number of ovules per pod and 
the number of seeds developing per pod for the total material 
from the 22 individuals is set forth in Table VI. The results are: 
A, = 4.916 = .013, A, = 4.116 = .OI5, 
Oo = .925 + .009, , os = 1.030 + .O10, 
Vo = 18.82, V, = 25.03, 
Tor =  .603.009, Yor = — .183 + .O14. 
For the lumped conclusions, where N = 2,200, I find: 
fos!L%ox, = 13.19. 
C. Analysis of the Data from the Vicinity of Sharpsburg, Ohio 
I am indebted to my grandfather, Mr. J. W. Harris, for the 
collection of 150 pods each from a series of 26 trees growing in the 
neighborhood of Sharpsburg, Athens Co., Ohio. 
Calculating from the grand total of nearly 4,000 pods summar- 
ized in Table VII, I find: 
A. = 5.493 =.011, As= 3.944 = .016, 
a = .983 += .008, Oo, = 1.453 = .O1l, 
V. = 17.89, V, = 36.83, 
‘os = 455 = .0090, Yor = — .034 = OIL. 
