1884.] LAWS OF GROWTH. 287 



attending that method. I have heard gardeners speak of it in terms 

 of highest praise ; but one is not sure whether the inventor ov the 

 adopters have fully explained to us how such a method of training 

 becomes necessary. The fact that frost mny be .strongest near the 

 surface does not account for tlie phenomenon. There the growth is 

 sickly ; but the sickness comes not of damage. As weakly horizontal 

 light is prejudicial, so the absence of all horizontal light is fatal to 

 branch development. Seedling trees can be grown so closely 

 to^xether in the seed-bed as to exclude horizontal light from all the 

 plants except those at the edges of the bed ; in which case, only 

 those plants at the edges of the bed develop branches. If a plant in the 

 middle of such a bed develop branches, it will give irresistible proof 

 that the condition of horizontal darkness has been relaxed, and that, 

 after all, the^law of the 'survival of the fittest' is rather a law of ex- 

 ternal favour to the fortunate than of victory to the inherently strong. 

 Let us come to a case that will illustrate not one only but several of 

 the laws enumerated. We have such a case in the third form of tree 

 described, the form akin to that of a bell-jar. We have said this is 

 Nature's favourite form ; if the statement be true, the form ought to 

 vouch for it. In our treatment of this form of tree, attention has hither- 

 to been directed almost exclusively to its outline. The remarks we are 

 now about to make have reference to the laws that co-operate in pro- 

 ducing such a form, and must not be confined to external configura- 

 tion. Our inquiries here ought to be conducted when the tree is 

 naked as well as when it is clothed with leaves. To become 

 acquainted with their forms, we must assiduously study trees in 

 winter. Those who love trees for the freshness and colour of their 

 leaves only, may be in the right way to an acquaintance with them, 

 and to a proper appreciation of their beauty; but while their 

 sympathies remain so circumscribed, their progress toward these 

 things will be far from rapid. It is only an unique aristocrat like 

 Herr Teufulsdrokh that can shape overcoats from all nature, and it is 

 only in such as he that we can pardon a love for universal dress. Let 

 our present inquiry be made with a naked tree. At a distance, the 

 form selected is that of a bell-jar resting on its edges. Close 

 examination reveals a short thick trunk in the middle of the dome. 

 Between the trunk and the edges of the dome all round, there is a 

 commodious apartment, a vertical section of which, if taken at the 

 centre, would roughly resemble a pointed pediment. This apartment 

 is formed by a downward sweep of the lowest branches now on the 

 stem. Above where these branches are attached to the stem, other 

 branches, which take a less downward sweep, are given off. Higher 

 up the tree another series is given off at nearly right angles with the 

 stem. Still higher up, the branches begin to assume a more and 



