278 METAMORPHOSIS 



One might well conceive of these organs as derived phylogenetically from 

 short shoots, but in reality they exhibit many important points of difference 

 from them. The point which is of the greatest interest to us at present is 

 the difference in the mode of growth. The leaves have in general a very 

 limited capacity for growth, for in a short time they become fully developed, 

 their growing points as such disappear, that is to say, become transformed 

 into permanent tissue. Similarly growth in short shoots is, as we have learned, 

 very limited, but the growing points are often still retained, and can, if the 

 appropriate stimulus be applied, give origin once more to new members. A short 

 shoot may very often change into a long shoot, whilst the alteration of a leaf 

 into a shoot, whether long or short, is impossible. This is certainly the rule, 

 but the rule may have exceptions. Thus, as a matter of fact, we have been for 

 long acquainted with cases among ferns and other plants as well where apical 

 growth continues in the leaves for many years. We know again of many long 

 shoots in which growth is arrested owing to the complete transformation of 

 the growing point. As an example may be taken the case of a shoot ending 

 in a thorn, or flowers where the growing point is generally transformed into 

 an ovary. The difference quoted is not enough to serve as a distinction between 

 a short shoot and a leaf, but there are others, only one of which need be men- 

 tioned here, viz. their relative positions on the axis. 



There are quite a number of plants which form only one axis on which 

 no lateral members save leaves are produced, e. g. Isoetes, many ferns and 

 palms. In a second series may be placed the numerous Coniferae which develop 

 lateral branches only. The majority of the higher plants, however, form both 

 lateral buds and leaves, and between these members perfectly definite space 

 relations subsist. The bud is formed in the upper angle of the leaf insertion, in 

 the so-called leaf axil (Fig. 72, g), and is termed in consequence an axillary hud. 

 Its point of origin is sometimes more towards the base of the leaf, sometimes 

 more towards the side of the stem ; it appears sometimes at the same moment 

 as the leaf rudiment, sometimes considerably later. 



These recognized relations between the leaf and the axillary bud may 

 serve to distinguish these organs from each other, but they are applicable only 

 to radial (and bilateral) shoots. If the growing point be dorsiventral, on the 

 other hand, the leaves arise on the upper side, while the lateral branches arise 

 on the flanks and the roots on the under surface, at some distance, it is true, 

 from the growing point. 



The lateral branches in turn may be again branched. The shoots of secon- 

 dary, tertiary, &c., rank maintain the same relations we have described as 

 characteristic of those of primary rank. It is otherwise in the case of leaves. 

 The leaf-blade, the part of special interest to us, is very frequently flat and 

 dorsiventral, i.e. with clearly marked upper and under surfaces Such a surface 

 may be simple or it may be branched. Branching in many cases (Palmaceae, 

 many Aroidaceae) may result from rupture of an originally simple leaf, but 

 usually the lateral branches are predetermined from the beginning. Occa- 

 sionally dichotomous branching occurs (ferns and many Dicotyledons, e. g. 

 Utricularia, Ceratophyllum, and many species of Drosera) ; for the most part, 

 however, the branching is lateral and commences, in the simplest cases (the 

 only ones we are considering at present), either from the base towards the apex 

 (Umbelliferae, Leguminosae) or in the reverse direction from the apex towards 

 the base {Rosa, Potentilla), or, finally, in the middle, progressing towards either 

 end {Achillea millefolium). Each branch arising in this way may again branch 

 in a similar manner. All branches arise on the flanks of the mother axis and 

 the individual leaflets lie in this way more or less in one plane. The fact that 

 the leaves develop for the most part in a bud, that is to say, in a confined space 

 surrounded by older leaves, necessitates many deviations from the flattened 



