1 42 Bower. — On the limits of the use 
Thus, there is another factor in the morphological problem 
beyond those above-mentioned, viz. the use of a comparison 
of closely allied forms, the results of which are accepted by 
some as overriding conclusions based on other grounds ; and 
whether or not, in the present state of our knowledge, we are 
justified in regarding such comparison as of first-rate import- 
ance, we must take into account this which may be called the 
‘ phylogenetic factor.’ An ideal system of morphology of the 
shoot, which should recognise the true homologies of all 
members, their origin, and metamorphoses, would be one 
based on a full knowledge of phylogeny, and what there is of 
arbitrariness in Sachs’ distinction is to be looked upon as a 
concession to the incompleteness of our knowledge on this 
point. How incomplete is our information and how uncertain 
our view, especially with regard to the descent of the Phane- 
rogams, all must be aware. But though our knowledge in this 
direction is at present far too scanty to form a general basis 
for an exclusively phylogenetic system of classification of 
members, there are certain points in the whole series of plants 
at which it is certainly sufficient for drawing a broad dis- 
tinction. We recognise that at various points in the series of 
plants c parallel developments 3 have taken place. If our 
morphology is ever to have a phylogenetic basis, we shall do 
well not only to admit the fact of these parallel developments 
having taken place, but, where such a course will conduce to 
clearness of conception, distinguish them from one another in 
our nomenclature. It will be well to begin upon what is 
certainly the most clearly ascertained, as it is also the most 
prominent example of parallel development in the vegetable 
kingdom, viz. the foliar differentiation of the shoot in the 
sporophore, as well as in the oophore generation 1 * . 
1 While the terms ‘ root ’ and ‘ shoot ’ may be accepted, as correlative terms, in 
the general sense proposed by Sachs, and as including the corresponding parts of 
oophore and sporophore generations, it must be clearly borne in mind that the 
differentiation of such parts must have arisen in the two generations in just as 
independent a manner as the further differentiation of the shoot into axis and leaf ; 
but there would be no sufficient advantage in marking this by a change of termino- 
"logy to justify disturbing terms which have met with general acceptance in their 
