58 On a Method of 



It might be inferred beforehand, that more plants would 

 grow (to take extreme cases) from a wall of rude stone ma- 

 sonry laid in clay, as is not an uncommon practice where lime 

 is scarce, and either pointed with lime or not, or from a wall 

 built by grouting, than from a compact and well-laid piece of 

 brick-work. Such is the fact ; and it is also true, that many 

 plants will grow on the surface or top of a wall, which will 

 not root in the interstices among the plaster, from the accu- 

 mulations of soil which so easily form in that part. But a 

 very great number, and many of the most desirable, will grow 

 in the interstices of even the most well-wrought wall, whether 

 of brick or stone, and even where the cement is of the firmest 

 quality, as I have put to the test of experience, times without 

 number. And this is the fact, of which I have found our 

 gardeners particularly incredulous; while I have never seen 

 one such attempt made in this country, if I except my own. 



This, however, is an essential point ; and it is needful, there- 

 fore, to bestow a few further words on it. The process may 

 commence with the very building of the wall, by laying the 

 roots in the mortar as the work proceeds ; roots of perennials, 

 of course, as it is not worth while thus to labour for annuals. 

 I need not point out all the plants which may thus be intro- 

 duced, as gardeners will easily supply what I omit ; but I 

 may mention, that this plan succeeds with the whole genus of 

 Dianthus for example, and that every pink or carnation that 

 I have ever seen tried, has thus rooted itself. And, with this 

 tribe, the effect is peculiarly pleasing and ornamental; as 

 their proliferous quality enables them to produce large cushions 

 from a single root, with which a wall can almost be covered, 

 were it deemed expedient. The same practice succeeds with 

 the Tussilago fragrans, with the Antirrhinums, Sedums, and 

 others ; but that which I omit may be easily conjectured, as 

 those which I have not thus tried will also leave room for the 

 endeavours of others. Of sowing seeds in the same manner 

 I have less experience, having never been able to return to 

 the only spot where I had the opportunity of fairly trying this 

 method, yet I see no reason why it should not equally succeed. 



For a wall already built, it is plain that another proceeding 

 is required. In this case, I have opened the pointing with a 



