408 MR. W. AND MISS A. BATESON ON FLORAL VARIATIONS 
garden-hybrids, which we may mention is also the opinion of 
Mr. Baker. 
It is therefore practically certain that flowers of the second 
type have come into existence in gandavensis as a variation 
occurring at, or since, its constitution as a hybrid, but whether 
such a variation ever occurs in its parents or in other wild 
Gladioli cannot be affirmed. 
In conclusion, we direct attention to the following important 
features of the case :— 
(1) The same spike bears flowers of two types of symmetry. 
(2) The second type of flower is of recent origin. 
(3) The symmetry of the second type of flower is nevertheless 
generally perfect, and forms intermediate between the two 
types are comparatively rare. 
(4) When the two forms occur on the same spike, the same 
colours occur in both, being distributed according to the 
symmetry of each, the distribution of the colours being as 
symmetrical in the one case as in the other. 
(5) Intermediate forms, though rare, occur; the symmetry 
both of form and colour is in them intermediate between 
those found in the two types; and the position taken up 
by the flower is intermediate between the positions assumed 
by the flowers of either type, and its approximation to 
either position is generally proportionate to the approxi- 
mation of its form to either type. 
IV. STREPTOCARPUS. 
We have been able to examine comparatively few flowers of 
this plant, and no exact record was kept of the total number of 
flowers seen. The observations given below were made chiefly 
on a batch of seedling plants of the species Resit, growing in pots 
in a stove-house, between the months of October and Decem- 
ber of this year. Among these plants we should say that about 
one in every twelve flowers was abnormal, but the proportion 
seems to vary greatly in different species. These abnormal 
flowers are of very various types, but we are unable to say at 
present whether one of these types can be considered more 
prevalent than others; for we have had too few specimens ot 
each to enable us to make any generalization. 
