748 CHAPTERS ON PROPAGATING. [April 



protected from the exhausting effects of the atmosphere Ly means 

 of glass, and are shaded from the direct rays of the snn. But 

 cuttings of deciduous trees and shrubs as a rule, except they are 

 ripe and contain a full amount of fully elaborated true snp, will die 

 under the experiment, no matter what precautions may be tnkcn to 

 prevent that result. I have run the risk of being tedious over this 

 point, rather than be misunderstood by the inexperienced, for whom 

 these papers are intended, as it has the most important bearing on 

 the proper choice of cuttings and the time when production by this 

 means may be most successfully prosecuted. 



Eespecting the choice of cuttings, having in the preceding para- 

 graph endeavoured to impress on my readers the necessity that 

 they should be ripe, or at least in that stage of maturity in which 

 they contain in abundance the matter without which they cannot 

 develop roots, it is only further requisite on this point to indicate 

 several other important particulars, observation of which will 

 contribute to success. 



As a rule, side shoots are superior to leading ones, because they 

 are usually more solid or firm and short-jointed. Tlie best are 

 those which from their position have enjoyed the largest amount 

 of light throughout the season ; hence the recommendation 

 which we often see in books, that they sliould be selected from 

 the lowest branches, or those that sweep the ground, is not 

 always to be relied upon, because such branches are always more 

 or less overshadowed by those above them, resulting in flabby 

 foliage and weak, ill-conditioned wood. There can be no rule laid 

 down for the length of cuttings that would be applicable to those 

 of all kinds of trees and shrubs. But if there is no stem required 

 to stand above ground, as in the case of gooseberry and currant, 

 cuttings, any length from six to nine inches, may be deemed 

 generally convenient. Much stress is laid by some on the desir- 

 ability of having a small portion of the older M'ood attached to the 

 base of each cutting ; but provided the young wood is thoroughly 

 hard and ripe, I have never been able to convince myself tliat it 

 was of much importance, in so far as deciduous cuttings are con- 

 cerned. Hollow-stemmed subjects, such as the honeysuckles and 

 Lycesteria, are, however, exceptions, which may or should be 

 selected with a base of the two-year-old wood attached, because the 

 older wood is more solid, and consequently less liable to be injured 

 by accumulations of moisture in the tube. 



Evergreen cuttings, such as those of coniferte generally, hollies, 

 and arbutus, should always be selected with a heel or small section 

 of the old wood at the base. Laurels and other easily increased 

 evergreens may be selected with or without such a heel as is found 



