Ficaria verna, and other Members of the Ranunculaceae. 57 
(c) Numerical relations } 
A study of the numerical relations between the different floral regions 
in the same flower set forth in Tables I-IV establishes the proposition 
that all the parts of the flower tend to vary at the same time and in the same 
direction . Thus we find that the larger number of nectaries are associated 
with more numerous perianth segments, and out of a total of forty-eight 
flowers with more than six perianth members only five had less than the 
normal number of nectaries, whilst twenty-nine showed an increase. Such 
evidence, then, gives no support to the assumption that supernumerary 
perianth segments represent modified nectaries. 
Still more striking is the numerical relation between carpels and 
stamens, or nectaries and stamens. From Tables II and III it is evident 
that when the number of stamens is in excess of the mean (viz. 30) the 
carpels and nectaries usually also exhibit an increase in number over the 
normal. The converse is, moreover, also true. A similar correlation is seen 
between the number of carpels and the number of ovules (cf. Table IV), 
though the number of cases examined (viz. 1 28) is too small to bring out 
this relation clearly. Hence, whatever the cause of increase or decrease, 
whether nutrition or some more subtle factor, it evidently tends to operate 
in a similar way upon the production of all the organs of the flower. The 
facile explanation sometimes resorted to, that increase of one type of organ 
takes place at the expense of another, is not borne out by the facts. On 
the contrary, it is clear that each type of structure is capable of exhibit- 
ing independent numerical increase or decrease without metamorphosis. 
Where metamorphosis does occur it is the exception rather than the rule. 
It is in conformity with this that the total number of parts in the flower 
as a whole exhibits a wide range of variation, viz. from thirty to 
seventy-one. 
It will also be seen from the data given that in general there is 
a marked tendency for all the whorls of the flower to exhibit trimery at the 
same time. 
(d) Transition. 
We have already noted the variability in form of the bracts, and that, 
though usually deeply divided, they always exhibit three main lobes which 
may be quite entire (cf. Fig. 9). Rarely, as noted by Irmisch (loc. cit.), an 
additional bract may be present accompanying a normal perianth of six 
coloured members — doubtless the result of bifurcation. In nearly all cases, 
however, where there is an additional bract it is found to consist of three 
entire lobes and to occupy the position of one of the outer perianth segments. 
An instructive example is represented in the diagram (Fig. 6, b). In this 
1 The author has been unable to obtain a copy of the paper by Mori (Firenze, 1910) on the 
correlative variation in this species. 
