420 MR. W. AND MISS A. BATESON ON FLORAL VARIATIONS 
of the nature of specific changes, and to point out that it may 
help us to measure the size of the integral steps of Variation. 
An objector may say that there is no evidence that the variations 
we have described are such as lead to the formation of new species 
and new forms. That is perfectly true; but nevertheless it is clear 
that they may be of this nature, and that new forms might thus 
originate ; and while, on the other hand, there is no evidence of 
the occurrence of variations other than these, by which new 
forms of symmetry may be produced, this class of Variation is 
entitled to be very carefully considered. For our own part, we 
think, further, that evidence can be adduced to show that this 
class of Variation does lead to the formation of distinct forms ; 
but this is a much wider subject and must needs be postponed 
for the present. 
There is, however, another class of Variation which is known to 
occur, and space will not be wasted if we point out very briefly 
the relation of this other kind of Variation to that described by 
us. By the elaborate researches of Galton *, it has been shown 
that the frequency of the occurrence of certain variations obeys 
the Law of Error; that is to say, that, speaking generally, the 
greater the departure from the normal form, the rarer will be the 
variation. Galton has shown that this is true of several variations 
in size &c. of Man; and Weldont has further established the 
same for variations in proportional measurements of the Shrimp 
(Crangon vulgaris), &c. Though these are the only variations 
which have been properly investigated by a statistical method, it 
may be seen by inspection that be resulting proposition cannot be 
true of the variations which we have been considering. For in these 
cases of symmetrical variations, as we have shown, the variation 
is frequently complete and seldom ineomplete, and the per- 
fection of the variation is out of all proportion to the frequency 
of its occurrence. If we suppose in the case of Linaria spuria 
that the flowers having the normal form or the form No. 2, or 
some intermediate ion could be arranged in a series, it would 
then be found that there were a great number of normals, a few 
intermediates, and a considerable number of flowers with the form 
No.2. The form No. 2 is thus, as it were, another normal. It is 
uo doubt true that flowers having substantially the form No. 2 
* F. Galton, * Natural Inheritances, &c.” 
t W. F. R. Weldon, Proc. Roy. Soc. xlvii. 1890, p. 445. 
