IN THE TROPICS. 
413 
Summary. 
A somewhat complete summary of the more important of the facts 
dealt with in the present notice has already been given in a previous 
paper, see pages 308, 312, 328, 333 ; the account which follows 
deals chiefly with a few phenomena not previously referred to. 
The result of the crosses made with Telegraph and Telephone was 
on the whole fully to confirm Mendel’s account of the behaviour of 
cotyledon characters on crossing. In the crosses with Ringleader 
the agreement with expectation was complete ; in the case of those 
with the small smooth yellow native pea (No. 1) a single absolute 
exception appeared in F , in which green appeared instead of the 
expected yellow. In F 2 and F 3 a few minor irregularities were 
observed, also in respect of colour, which are probably to be accoun¬ 
ted for by imperfect ripening. In later generations the behaviour 
of these characters was more regular than in the earlier ones, and this 
is perhaps to be partly associated with the better health of the plants, 
but also, almost certainly, to an intensification of the characters 
following upon the cross. Nevertheless, even in F 2 , when the total 
number of seeds born upon individual plants was often very small, 
totals closely approximating to Mendelian expectation were obtained 
when the figures for several plants were combined. 
A summarized account of the crosses between French Sugar and 
No. 1 will be found on page 328 (example VI.). The colour charac¬ 
ters of the testa gave in F 2 the proportion 9 : 3 : 4,’ already observed 
by Tschermak in several instances, but not described by him in terms 
of gametes. Other definite characters such as presence and absence 
of the parchment layer in the pod, follow Mendel’s law, although 
dominance is not perfect in this case ; and certain other less definite 
characters very probably behave in the same way. 
The cross between No. 2 and Satisfaction, briefly described on page 
333 (example VII.), was only represented by a few individuals in the 
earlier generations, but on account of its rather peculiar interest and 
because many of its offspring still survive for further experiment, it is 
here described at considerable length. The colour characters of the 
testa yield in F 2 , almost certainly, the proportion 27 : 9 : 9 : 3 : 16, 
8(9)05 (8) 
