PLANT MORPHOLOGY 65 



which are as a rule definite neither in number nor in arrangement, 

 and are subject to variation in both respects, according to the con- 

 ditions which may be imposed upon them by experiment : neverthe- 

 less, they accurately maintain their structural characters, and their 

 essential correspondence is thus established, but not their individ- 

 ual identity. It is clear that this is a comparison of a more lax order 

 than the recognition of their individual homogeny would be. 



But if room for doubt of the strictest homogeny be found in simple 

 cases such as these, what are we to expect from the comparisons of 

 less strictly similar parts of the plant, such as cotyledons, scale- 

 leaves, foliage-leaves, bracts, sepals, petals, stamens, carpels? How 

 far are these to be held to be homogenous, or in some less strict 

 sense homologous? Or, going still further, how are we to regard those 

 comparisons which deal with parts of different individuals, species, 

 genera, orders, or classes? What degree of homology is to be ac- 

 corded to them? In proportion as the systematic remoteness of 

 the plants compared increases, and the continuity of the connecting- 

 forms is less complete, so the comparisons become more and more 

 doubtful, and the use of the term "homology'' as applied to them 

 more and more lax, until w r e are finally landed in the region where 

 comparison is little better than surmise. It becomes ultimately a 

 question how far the term "homology" is to be held as covering 

 these more lax comparisons, which are certainly not examples of 

 "homogeny" in Lankester' sense, and are only doubtfully correlated 

 together on a basis of comparison of more or less allied forms. 



The progress of our science should be leading towards a refinement 

 of the use of the term "homology:" an approach must be made, 

 however distant it may yet be, to a classification of parts on a basis 

 of descent. But though this may be readily accepted in theory, it 

 is still far from being adopted in the general practice of plant 

 morphology. None the less, comparison is inevitably leading to the 

 disintegration, on a basis of descent, of the old-accepted categories 

 of parts: of these the most prominent, and at the same time the most 

 debatable, is the category of leaves, and they will lend themselves 

 best to the illustration of the matter in hand. 



To those who, like myself, hold the view that the two alternating 

 generations of the Archegoniatce have had a distinct phylogenetic 

 history, it will be clear that their parts cannot be truly comparable 

 by descent. The leaf of the vascular plant, accordingly, will not be 

 the correlative of the leaf of a moss. Even those who regard the 

 sporophyte as an unsexed gametophyte will still have to show 7 , on a 

 basis of comparison and development, that the leaves of the two 

 generations are of common descent. I am not aware that this has 

 yet been done by them. 



But the phylogenetic distinctness of the leaves in the sporophyte 



