JiOCK-C.AKDKNS, PONDS. AND STJ{K.\ M LETS. 
71) 
parallel strata- if found in the natural roclis whence the stones were 
obtained should also occur throu^diout the artificial work; but wherever 
this idea is carried out the result is a failure from a picturesque as well 
as from a practical point of view, even if the rocks so constructed do 
have a resemblance to natural rocks. 
^Vhat we admire most in nature are not the re^nilar sedimentary 
layers, such as we find in a quarry pit, but rather such rocks as bear 
unmistakable evidence of having, at some period or other, passed 
through violent convulsions, and were, perhaps, cleft asunder and 
traversed by a deep chasm or ravine through which a streamlet found its 
way, now tumbling merrily over rocks and boulders, now meandering 
through green sward bedecked with flowers, and broken here and there 
by other rocks, which pierce the surface, and at one moment are high 
above the general level, and in another dip down far below the sward 
and are lost to view. Not only are such rocks infinitely more picturesque 
than a less broken and varied surface, but they also are far better 
adapted for choice alpines, owing to the fact that these natural projec- 
tions and recesses afford in most cases precisely the conditions the plants 
require. 
Whether a rock-garden should be large or small must ahvays depend 
on circumstances ; but wdiether it be of the most extensive or of the 
most modest proportions a natural appearance is absolutely indis- 
pensable. 
Unfortunately it is not always easy to effect this natural appearance, 
as work of this kind requires a special study of the rocks of nature, 
especially that of natural rocks in their simplest form cropping up from 
the ground. 
Having been asked to give a paper on this subject I have great 
pleasure in doing so. I do not by any means profess to have mastered the 
art of rock-building, but having been for nearly a quarter of a century 
landscape gardener to Messrs. Robert Veitch & Son, at Exeter, and 
having in that capacity designed and constructed a very large number of 
rock-gardens of all sizes and under the most varying conditions, I can 
at least lay claim to practical experience, an account of w^hich may 
possibly be of service to others, and prevent them from making mistakes, 
which I must confess to having frequently committed myself in my 
earlier works. 
With this object in view I will now^ venture to briefly explain my 
own method of arranging rock-gardens. 
Choosing a Site is not always quite as easy as it seems. The best 
is undoubtedly an uneven surface, away from large trees, and, in the 
ease of large gardens, not too near the house, because here the "svild 
character of rock scenery w^ould be incongruous and not in harmony with 
the necessarily geometrical surroundings. Sometimes, how^ever, it may 
happen that the house abuts against a hill or that there is an undulating 
slope or some other irregular feature, in which case it would be quite in 
keeping to have groups of rock apparently breaking through the sward 
here and there. 
It is still often supposed that any odd and useless corner even under 
trees is quite good enough for a rockwork. But unless the adornment of 
