414 MR. W. AND MISS A. BATESUN ON FLORAL VARIATIONS 
and not from a regular or actinomorphic flower. From such 
evidence, therefore, it would be wrong to draw conclusions as to 
the mode of origin of a zygomorphie flower from an aetino- 
morphie one. Therefore, though the facts warrant the statement 
that a new form of zygomorphie corolla may occur as a sudden 
variation, this may as yet be affirmed only in the ease of an 
irregular symmetry derived from another symmetry itself irre- 
eular. For example, the facts lead us to suspect that such an 
irregular, 4-petalled corolla as that of Veronica may have 
occurred as a sudden variation from the form of some 5-petalled 
Scrophulariaceous ancestor; but we are as far as ever from 
knowing how that irregular corolla of the ancestor was derived 
from a regular or actinomorphic form. At all events the facts 
now given have no direct bearing on this part of the problem. 
We do not now propose to attempt a discussion of the facts in 
their specific bearing, for the materialis not such as to entitle us 
to do so, still less can we aim at an analysis of the various forms 
of symmetry presented. 
It may, however, be pointed out that the examples taken show 
two methods by which a change of symmetry may be effected, 
and it is clear that these two processes are essentially distinct 
phenomena. For while, in the majority of instances given, the 
change of symmetry comes about by an alteration in the number 
of the parts, the case of the variation described in Gladiolus 
gandavensis and that of Linaria spuria No. 2 are alike in that 
they exhibit a change of symmetry attained not by an alteration 
in the number of parts, but by the selection of a different mor- 
phological plane about which the symmetry is developed. For 
inspection shows that in each of these cases the normal plane of 
symmetry of the flower has been forsaken and a new one 
substituted, so that the plane about which the symmetry of the 
flower is arranged falls through one of the segments adjacent to 
that through which it normally falls, and the parts of the 
flower are rearranged accordingly, but the number of the parts 
remains unchanged. Change of number of parts may also be 
associated with this alteration in the plane of symmetry, as 18 
seen in the flower of Linaria No. 3 (Pl. L. fig. 7). In other cases 
given, the change of symmetry is accomplished by a change in 
the number of parts. In Veronica for instance, though a zygo- 
morphie arrangement of 4 petals is normal, perfeetly zy gomorphic 
arrangements of 3 and 2 petals were shown to occur, in addition 
to two types of symmetrical flowers each with 5 petals. 
