IN THE TROPICS. 
315 
It will be seen that there are seven cases, either of uncon- 
formability or of error, viz.:— 
1. 
02 
sm. y. 
sm. and wr. green 
* « 
© fe 
yellow 
2. 
a 
sm. y. 
o 
-+3 
wr. y. only 
smooth 
3. 
£ 
o 
sm. g. 
© 
02 
all four types 
© _ 
r) a 
green only 
4. 
GQ 
T5 
wr. y. 
do. 
© •£> 
wrinkled only 
5. 
6. 
© 
© 
02 
< 
wr. y. 
wr. y. 
© 
> 
CÖ 
sm. and wr. green 
wr. green 
© += 
çH © 
'S a? 
£ a- 
wr. y. (and g.) 
yellow 
7. 
wr. g. 
sm. green 
wrinkled only 
In cases 
2,4, 5, 
and 7 the most likely supposition is that 
round and wrinkled were not truly distinguished in sampling, 
whilst in cases 1, 5, and 6 we may suppose that a truly green 
seed had turned yellow on exposure after the manner of the 
Telephone parent."' 
Case 3 remains, and in this single case I am disposed to 
think that a mistake was made in taking up the plant, and 
that this was really the offspring of a yellow seed. 
(3) Tri-Poly hybrids. 
Other pairs of characters in peas are found to behave in 
the same way. Mendel worked with seven pairs, viz.:— 
(1) The shape of the ripe seed, whether smooth ; or 
angular and wrinkled. 
(2) The colour of the seed albumen (cotyledons), whether 
yellow ; or of a more or less intense green tint. 
(3) Coloured testa, flowers and leaf axils ; allelomorphic 
as a group to white in all three cases. 
(4) Pods, whether simply inflated; or soft and constricted 
between the seeds, the parchment layer being undeveloped 
in the latter case. 
(5) Colour of unripe pods, whether green ; or yellow. 
(6) Position of flowers, whether axial ; or terminal. 
(7) Length of axis, whether tall ; or short. 
Examples of cases 1 and 2 have already been described. I 
have also been able to examine cases 3, 4, and 7, and obtained 
results* agreeing closely with those of Mendel and other 
observers. Mendel found all possible combinations of the 
* In the case of crosses with other varieties, in which the green colour 
of the cotyledons is more constant, this form of irregularity is net obtained. 
