284 Variation in Leaves and Flowers of Goldilocks. 
and the vast range of leaf-form in the foliage-leaves. Plants 
examined in other localities on previous occasions led to the 
belief that there was some connection between the character 
of the variation and the locality in which the variety grew, but 
no evidence was obtained either way on this occasion. Three 
hundred and seventy-six specimens were examined in detail. 
The sepals were five in number in 345 cases, while the six sepals 
in 24 cases and seven occurring six times plainly owed their 
larger number to an ordinary duplication of one sepal or of two 
sepals. In only one case was the number of sepals reduced 
to four. It is evident that the number of sepals is remarkably 
constant and as a rule they were arranged on a 2 spiral. 
But in the corolla there is a great amount of variation in 
number, shape and appearance of the petals. In gI cases 
there was no sign of petals at all, and in all of these the sepals 
had completely altered in general appearance; they were more 
papery in texture, somewhat larger, less leafy, and much 
yellower than in the typical sepals. These are shown in figure 
C from the front and D from the side. The edges were 
well-wrinkled and there was often a tendency to an obscure 
division into three parts as shown. No traces of nectaries 
were found, 
In 88 other cases one petal occurred. In nearly all this was 
well-developed and typical in shape, as at E. The nectary was 
well formed. The ends of the semi-circular mouth were con- 
tinued into the tissue of the lamina of the petal exactly at the 
veins next to and on either side of the middle one. As a rule 
the sepals on each side of the fully developed petal were green, 
like the type A, while those away from it were all more or less 
petaloid, some as in figure B, with a green centre and a yellow 
petaloid rim; and others more like C and D. Almost in- 
variably the petaloid sepals became thin, yellow and papery, 
but did not acquire the almost waxy texture and glossy surface 
of the true petals. In 74 flowers there were two petals, while 
42 had 3 petals each, and 32 possessed 4. In all these 
the tendency was for some of the sepals to be petaloid, and 
which of them were modified seemed to depend upon two 
factors. Frequently the alteration of the calyx took place in 
such of its members as made the whole flower more conspicuous. 
But in other cases the modification towards the petalised 
condition is much increased at the end of the spiral nearest to 
the petals or to the place where they would be under normal 
conditions. This is shown most plainly perhaps in the cases 
where there are two normal and three ‘ petaloid’ sepals. In 
this case Nos. I. and II. of the spiral, whether direct or reversed 
(see Figs. K, L), are sepaloid, four and five are petaloid and 
No. III. is either wholly petaloid or in the instances where it is 
half petaloid, the petaloid wing was always observed on the side 
Naturalist, 
