ROCK ROSE. 



271 



ROCK-WORK. 



escent species are not handsome 

 trees, from the liability of their 

 branches to he broken off by high 

 winds, which gives them a ragged 

 and untidy appearance ; and the 

 wood, notwithstanding all the very 

 extravagant praises of Cobbett, is 

 of very little value. The roots of 

 the Robinia also extend just under 

 the surface ; and thus a tree of this 

 genus occupies nearly twice the ex- 

 tent of ground which would be taken 

 up by a tree of the same size of 

 almost any other kind. The shrubby 

 kinds of Robinia are also liable to j 

 the same objections ; though the , 

 long racemes of rose-coloured flowers , 

 of the Rose Acacia {R. lii&pida) are i 

 so beautiful, that no shrubbery ; 

 should be without it. All the Ro- 

 binias are quite hardy, and they 

 may be grown in any soil, though 

 they thrive most in a rather rich 

 sandy loam ; care should be taken 

 always to plant them in a sheltered 

 situation, as no plants are more 

 injured by high winds. They are 

 propagated by seeds, layers, cut- 

 tings, and suckers, which are pro- 

 duced in great abundance. "When 

 Robinias are raised from seed, the 

 seeds should be steeped for twenty - 

 four hours in hot water before sow- 

 ing ; as, unless this is done, they 

 will sometimes remain in the soil 

 two years before they germinate. 

 When grown in strong clay, or in 

 any moist soil, old trees of Robinia 

 Pseud-Acacia are very often found, 

 when cut down, to be hollow at the 

 heart. Several plants, formerly 

 considered to belong to Robinia, are 

 now distributed through the genera 

 Caragana, Halimodendron, &c. Of 

 these, the most beautiful are the 

 Caraganas, and the Halimodendron, 

 or Salt Tree. 



Rocket. — See He'speris. 



Rock Cress. — Akindof A'rabis. 



Rock Rose.— See Ci'stus. 



Rock-work is a very common 

 ornament in gardens ; and, pro- 

 ducing a striking effect, it is intro- 

 duced more frequently than judi- 

 ciously. Rock- work may be divided 

 into two kinds ; that which is in- 

 tended to imitate natural rocks, and 

 that which is intended merely as a 

 nidus for rock-plants. Imitations 

 of nature should always consist of 

 large blocks of stone of the same 

 kind, and should, for the most part, 

 be disposed in imitation of some 

 kind of stratification. At the same 

 time, as in many parts of the coun- 

 try, large, round, or roundish, or 

 angular blocks of stone are found 

 distributed over the surface, it is 

 not objectionable to collect these to- 

 gether in groups so as to form a 

 feature in scenery, and to insert 

 plants among them. Rock- work, as 

 a mere nidus for plants, should 

 never be attempted on a large scale 

 without the introduction of large 

 blocks of stone, and some kind of 

 stratification being adopted ; and in 

 this case, as before mentioned, using 

 one kind of stone will produce an 

 effect in accordance with that of 

 nature. On a small scale, however, 

 different kinds of stone may be used, 

 more especially when these are well 

 covered with plants ; but even on a 

 small scale, one kind of stone has 

 always a better effect, and will be 

 felt more agreeable to the eye, than 

 a mixtiu-e of bricks, flints, pieces 

 of granite, freestones, and perhaps 

 marble, shells, fragments of carved 

 stone, and even roots, which are 

 not unfrequently seen in even the 

 best gardens. Rock-work should 

 always be an independent feature. 

 It rarely looks well when piled up 

 against a wall or around the roots 

 of a tree, or in any situation where 

 it is overshadowed by trees ; in 

 short, where it does not form the 

 prominent feature in the scene. It 



