1912] HARRIS—STAPHYLEA 403 
we find: 
7g=O0.3522+0.0131, 
psi=0. 3408+ 0.0131, 
Ta — psi = 0.00240.0185 
TABLE XVI, 1906 
SEEDS PER LOCULE 
Position ° I 2 3 4 « 6 Totals 
Dien a conta Oa TEz3 IT109 355 II5 33 II 3156 
Be attache 1076 133 190 60 16 4 2079 
Baek Soe 404 285 88 28 7 I 81 
| eh et epee 6 40 12 2 I 120 
Rete ewes 5 3 I 9 
SE OEAIS oars 3074 2170 646 212 58 16 4 6177 
TABLE XVII, 1906 
’ SEEDS PER LOCULE 
Fruits per infl. ° I 2 3 4 5 6 Totals 
Mepis oealics erik. 2 ee I oe 3 
Siete ants 299 238 82 29 14 4 666 
BS eR eS ri 761 556 170 54 15 4 1560 
Bee ol See 827 581 164 60 16 5 1053 
Beis see 591 309 120 37 7 I II55 
Ounce cee s 343 217 63 17 5 2 I 648 
ee gee Cena 115 85 25 6 231 
Bea oI 61 13 % I 168 
Boe ee ey 10 14 2 I 27 
| hs EE elena) 14 10 2 4 3° 
PE Os Oe ae ee ee wa 
3 ESN Rata Ba ie 21 9 4 2 36 
TOU Cs: 3074 2170 646 212 58 16 I 6177 
The reduction in correlation is only about one-eighth of the 
probable error of the determination! Further arithmetic or dis- 
cussion would be pedantic. I conclude that the correlation 
between the number of seeds developing and the length of the 
fruit is not merely a secondary result of the correlation of these 
two characters with position on the inflorescence or number of 
pods per inflorescence, but must be due to some force operating 
in the developing fruit itself. 
For Cercis I have no data for the relationship between the num- 
ber of pods per inflorescence and the fertility and length characters 
