Double r lowers and Flowerheads. 195 
(Cineraria), the transformation of the little yellow disc- 
florets into long white tubes (Pyrethrum, see Fig. 36) 
etc. We shall however leave such cases out of considera- 
tion ; they may be regarded provisionally as cases of 
spurious doubling. 
The genuine "doubling," on the other hand, as ex- 
hibited by the most diverse species, presents a very 
marked agreement with the conditions found in Chry- 
santhemum segetuni. Indications of a tendency to "doub- 
ling" occur both in forms of which a double variety is 
not offered by seedsmen and in those of which such 
are already on the market. For instance in 1892 I ob- 
served occasional tube-florets more or less completely 
transformed into ligulate florets in a culture of Bidens 
grandifiora in my garden. In other cases the variation 
is only seen when curves are plotted. For example I 
obtained the following very asymmetrical curve from 
a culture of the single variety of Chrysanthemum eoro- 
narium, a favorite garden plant whose double form 
has long been known (Fig. 35). 130 flowers on 25 
plants of a single crop were recorded, the flower at the 
top of the main stem and those on the primary branches 
alone being taken into account. I found : 
Ligulate florets 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 
Inflorescences 101 2 2 12 25 19 21 15 14 6 7 2 3 
That is to say, 18 on one side of the mean and 87 on the 
other, with a faint indication of a second maximum at 
the next figure in the BRAUN-SCHIMPER series, 21. It is 
clear that the double variety of this species could prob- 
ably be obtained from these plants, exactly in the same 
way as in C. segetuni. 
These considerations evidently lead to the hypothesis 
that the secondary maxima on the positive side of LUD- 
