J. W. Jenkinson 
153 
Plane of Symmetry and Sapittal Plane. 
Figs. 2 and 3. Schemes of Eegression constructed from Table II. 
In Fig. 2 tbe dots indicate the mean vahie of the angle between Plane of Symmetry and First Furrow 
for each class of angle between Plane of Symmetry and Sagittal Plane, in Fig. 3 the converse. 
left-hand side. As we shall see presently, and as indeed may be gathered from 
Table 11., this is due to the tendency of the first furrow to lie either in or at right 
angles to the plane of symmetry, and Professor Pearson has suggested to me that 
if the upper and lower arrays of the table were omitted the value of p would be 
still further reduced. This is, as a matter of fact, the case. 
Table III. is the correlation table constructed from the six middle arrays 
of Table II.; Fig. 4 the corresponding regression scheme. The line of regression 
is now practically horizontal, and the value of p — less than the probable error — 
practically nil. 
TABLE III. 
Correlation table constructed from the Middle 8t7-ip of Arrays of Table II. 
Plane of Symmetry and Sagittal Plane. 
T3 
C 
Sh is 
■£ o 
a g 
If 
c3 
go 
+ 
Totals 
45 
2 
2 
4 
6 
1 
1 
1 
45 
17 
1 
1 
7 
4 
5 
3 
3 
3 
27 
o 
3 
8 
24 
21 
14 
7 
4 
3 
o 
85 
1 
1 
•2 
.3 
8 
15 
15 
11 
3 
1 
1 
2 
63 
+ 
3 
6 
6 
4 
4 
1 
24 
3 
5 
5 
2 
3 
1 
2 
1 
22 
45 
45 
Totals 1 
2 
3 
12 
33 1 58 
55 
36 
18 
9 
Means + '5 
- -5 
+ •1 
+ •08 
+ -04 + -01 
-•21 
+ •13 
+ •11 
- -9 
4 
+ •7 
7 
+ •5 
238 
p=-009+ -044. 
Biometrilia v 
20 
