IN THB TROPICS. 
327 
between “ griine spate Erfnrter Folgererbse” with a 
colourless testa and “ purpur-violettschotigen Kneifel-erbse” 
with an orange-red to brown testa, Correns found, in the 
first generation and often in the same pod, seeds with an 
almost colourless testa and seeds with a fully orange-red 
testa. Violet spots also made their appearance. 
The appearance of violet spotB upon the testa in this way 
has also been noticed by Tschermak (59) ; but the former 
observation has not been confirmed and may rest upon a 
misunderstanding. Bateson (2) has pointed out that the 
result may have been due to conditions ; and I have myself 
found the testa of Sutton’s French sugar pea to change 
from gray to orange-brown upon exposure, so that these 
colours cannot be distinguished as separate plant characters. 
With regard to seed shape Tschermak (6)) obtained some 
remarkable results in the case of crosses between Pisum 
arvense (red flowers) and P. sativum (white flowers). In the 
first example described the former had “ slightly wrinkled” 
seeds, which may conveniently be called “ dimpled,” the latter 
had smooth seeds. No change was produced in the shape 
of the seeds as the direet result of crossing ; smooth “female” 
crossed by dimpled “ male ” and dimpled “ female ” crossed 
by smooth “ male” gave rise in each case to seeds identical in 
appearance with those of the maternal variety. The resulting 
plants (Fj) had all of them coloured flowers, and produced 
exclusively dimpled seeds with a coloured testa. Thus 
coloured flowers, coloured testa, and dimpled seeds were 
always associated ; and Tschermak regards this as indicating 
a correlation between a vegetative character and a sexual 
one, namely, between coloured flowers and testa on the one 
hand, and the character of the ovum from which dimpled 
cotyledons arose on the other. And upon this conception 
he bases important theoretical conclusions. 
But there is another possible explanation. £!orrens has 
found in many cases that the shape of the grains of Zea 
Mays depends solely upon that of the pericarp, by which 
