CHAP, ii The Theory of Metamorphosis 73 



from each other, but are only to be distinguished by the 

 part they play in the further growth of the young leaf. 



The relationship between stem and leaf was examined 

 very closely by Bower in 1884, and a new view of the 

 morphology of the latter was advanced. At the outset 

 he drew attention to the undesirability of confining the 

 study of these forms to the higher plants, as Sachs had 

 done some years earlier, and insisted on comparing them 

 with the more lowly ones, to test the validity of the dis- 

 tinctions that had been drawn. Criticizing Eichler's theory, 

 he showed that the separation of the leaf into foliar base 

 and upper leaf divides the leaf into two parts that are not 

 co-ordinate, a plan which is unlikely to lead to true insight 

 into its morphology. He said, ' The most important point in 

 the morphological study of a shoot or branch system is 

 to ascertain the mode of origin and sequence of appearance 

 of the various parts, and their relations in these respects 

 to each other. In dealing with the leaf this principle is 

 not kept so clearly before the mind. Current morphology 

 still retains those obvious distinctions of sheath, petiole, 

 and blade, as co-ordinate structures, ignoring the fact that 

 the distinctive marks between them depend on processes 

 of growth in various directions '. Hence, he concluded 

 that Eichler's two categories are misleading. Bower 

 substituted the idea that the leaf is a podium, which may 

 or may not branch, and may develop in different ways 

 at different points. He proposed the term phyllopodium 

 to designate it, and showed how greatly it may vary. It 

 may be cylindrical as in Pilidaria, flattened, without wings 

 or midrib, as in WelwitscMa, winged throughout part or the 

 whole of its length as in Gnetum. It may branch, the 

 branches appearing in acropetal or basipetal order, some- 

 times merely as teeth at the margins of the wings, at others 

 as distinct pinnae which may branch in turn. Differences in 

 the distribution of its growth may lead to the appearance 



