THE SHAPES OF BRANCHES. 135 



of form result, are in many cases largely due to tlie inequali- 

 ties in tlie circimistances of tlie parts while in the bud (which, 

 are however representative of inequalities in ancestral cir- 

 cumstances) ; yet these are clearly not the sole causes of the 

 unlikenesses that eventually arise. For the leaf-buds whence 

 the larger leaves in Fig. 205 were developed, instead of being 

 at first more favourably circumstanced than the others, were 

 less favourably circumstanced. So that this bilateralness 

 that results from the unequal sizes of the leaves, must be con- 

 sidered as wholly due to the differential actions that come into 

 play after the leaves have assumed their tj^ical structures. 



§ 225. How in the arrangement of their twigs and leaves^ 

 branches tend to lapse from forms that are approximately 

 symmetrical to forms that are quite asymmetrical, need not 

 be demonstrated : it is sufficiently conspicuous. But it may 

 be well to point out how the tendency to do this further 

 enforces our argument. The comparatively regular budding- 

 out of secondary axes and tertiary axes, does not usually 

 produce an aggregate which maintains its regularity, for 

 the simple reason that many of the axes abort. Terminal 

 buds are some of them destroyed by birds ; others are bur- 

 rowed into by insects ; others are nipped by frost ; others 

 are broken off or injured during gales of wind. The envi- 

 ronment of each branch and its branchlets is thus ever 

 being varied on all sides : here, space being left vacant by 

 the death of some shoot that would ordinarily have occupied 

 it ; and there^ space being trenched on by the lateral growth 

 of some adjacent branch that has had its main axis broken. 

 Hence the asymmetry or heterogeneity of form which the 

 branch assumes, is. caused by the asjonmetrical distribution 

 of incident forces — a result and a cause that go on ever com- 

 plicating. 



§ 226. One consjDicuous trait in the shapes of branches 

 has still to be namecl. Their proximal or attached ends 

 differ from their distal or free ends, in the same way that 



