25 
much the form of a cone is lost sight of, it is still 
distinctly traceable. 
54. In bushes and plants of a similar growth the 
conical figui'e is too manifest to require demonstration; 
their branches, however much extended, diminish re¬ 
gularly to the point, not suddenly swelling and sud¬ 
denly diminishing, as we often sec it in the draw¬ 
ings in print shops, but forming a true conical figure, 
with the most graceful outline imaginable. 
55. Should a stem be cylindrical then it is truly 
cylindrical, and not deformed by bosses in one place 
and hollows in another. 
56. In branching, every species has its own peculiar 
arrangement, from which there is no departure. If 
branches are formed in one way in one part of a plant, 
they are formed in the same way in all the other 
parts, so that there can be no truth in figures which 
represent a plant with branches arranged one way 
on the right hand, and in some other on the left. In 
the Sycamore, for instance, the branches are con¬ 
stantly placed right and left, fore and aft, right and 
left, and so on, appearances to the contrary being 
owing to accidents — such as the destruction of the 
buds or rudimentary branches. 
57. In the course of its development a branch 
never forms an angle with the stem which produces 
it: it always forms a curve. It does not arise as if 
a hole had been bored into a trunk and a cone been 
driven in, but gentle curves of various kinds accom¬ 
pany the separation of the branch from its parent, so 
that nothing but a curve is allowable when repre¬ 
senting the junction of a branch and stem. 
58. Even the upper or re-entering angle (or axil 
as botanists term it), is also a curve of some kind, for 
B 
