THE GENUS NOTOTRICHE. 213 
produced a chain of allied forms distributed over a fairly extended region. Others, 
again, once formed may have remained stable perhaps over a long period of time, and, 
by the intercalation of other unrelated species or perhaps by the dying out of those 
most nearly related to them, now appear as somewhat isolated and peculiar forms. 
It is of course also highly probable that some species, though they show a resemblance 
in their leaf-form may not really be very closely related, but may represent the outcome 
of variations in the same direction on the part of two widely distinct ancestors. 
From such a point of view it appears that the two contrasted leaf types, palmate and 
pinnate, may be regarded. 
Taking a 5-lobed “ palmate” leaf as the normal and original type of the lamina for 
the genus, it may be supposed that the evolution of the pinnate type has oceurred by 
successive mutations from the palmate type in the direction of the enlargement and 
elongation of the median lobe. Then with the further progress of variation along the 
lines of the two established and divergent types, a number of curious and elaborate 
forms have arisen which in the intricate variety of their leaves appear to be due rather 
to the indulgence of exuberant fancy than to the outcome of any definite response to 
particular conditions. 
Together with the variation in the leaves, floral and other variations would naturally 
have taken place, the expression of which can be seen in the presence or absence of 
a tube to the corolla, the arrangement of the stamens, and the shape and hairiness 
of the carpels. 
The history of the course of events in the evolution of the species of the genus 
Nototriche may, however, be regarded from another standpoint. It has already been 
pointed out that the species may be arranged in two groups on the character of the 
presence or absence of a corolla-tube (p. 206). In ten species the corolla-tube is 
practically absent, nine of them showing the stamens arranged in an elongated head 
(Pl. 30. fig. 9), three other species show a similar cylindrical arrangement of the stamens 
but possess a more or less definite corolla-tube, and in three others though the corolla- 
tube is short, the stamens are arranged in a globose head, as in almost all the species 
where the corolla-tube is well marked * (Pl. 30. fig. 8). 
* The species with the corolla-tube practically absent are tabulated below: where there is a short tube the 
length is stated :— 
Corolla-tube none. Corolla-tube none. Corolla-tube short. 
Stamens in cylindrical head. Stamens in globose head. Stamens in globose head. 
N. pichinchensis. N. flabellata. N. condensata (1-2 mm.). 
N. phyllanthos. N. Orbignyana (1-2 mm.). 
N. -Hartwegii. N. Lobbii (2 mm.). 
N. Macleanii. 
Corolla-tube present. 
N. pedicularifolia. 
pedicularifolia Stamens in cylindrical head. 
N. Lorentzii (?). 
N. Matthewsii. N. Jamesonii (3-4 mm.). 
N. stenopetala. N. Pearcei (2 mm.). 
N. epileuca, N. sulphurea (1°5-2'5 mm.). 
