TKANSACTIOXS OF SECTIO^^ K. 820 



gation in tlie relationships of size and distribution of vasa both to tho othei- 

 structural elements of the stem and to the form of the plant in relation to its 

 environment. 80 far as I know the monocotylous tree-forms, there has been an 

 attempt in two different directions to provide an increased water-carrying- system 

 in them. There is the familiar one of the secondary cortical cambium in Dracrena 

 and other genera. In them the cambium merely repeats in its products the con- 

 •struction of the primary stem, and does not provide so copious an increase of carrying 

 area as does tbe system in dicotylous plants. And then in such plants as Barba- 

 cenia, many BromeliacCc'e, perhaps Kingia, we have an arrangement reminiscent of 

 the superficial root-system which is found in many polystelic arborescent Pterido- 

 phytes of the present day. There is a copious growth of adventitious roots from 

 the central vascular cylinder, and these pass down within the cortex, and from 

 its cells are no doubt able to draw water for the upper parts of the stem.^ Ulti- 

 mately many of these roots reach the soil. At best , hov/ever, neither of these 

 systems has been satisfactory. All that can be said for them is that they have 

 enabled the monocotylous trees in which they are found to hold their own in 

 xerophilous conditions. 



Of Phyla ivithin Dicofyledones and Monocotyledonefi. 



A brief reference only to the groups within the Dicotyledones and Monocotyle- 

 dones must conclude these remarks. AVhilst there is a wonderful concurrence in 

 the opinion of botanists as to the natural groups — real phyla, whether termed 

 cohorts, alliances, or series — into which many of the families of both Dicotyledones 

 and Monocotyledones fall, there is irreconcilable divergence of \\%vf as to their 

 genetic sequence or sequences. And this is not surprising when we remember that 

 we know nothing of the starting point or points of tlie classes themselves ; and 

 have, moreover, no critical mark by which to diagnose a primitive from a reduced 

 feature in many of the flower constructions to which, as characteristic of Angiosperms, 

 importance is attached. The desire to establish a monophyletic sequence of these 

 phyla is natural, and finds expression in pedigrees of Dicotyledones issuing from, 

 it may be, Ranales or Piperales, of Monocotyledones from, say, Apocarpse or Arales. 

 But all such attempts appear to me, in the present state of our knowledge, to be in 

 vain, "\Ve see in the phyla, as we know them, culminating series in our epoch in 

 lines of descent ; some, for instance Myrtales or Lamiales, progressive ; others, 

 like Primulales or Pandales, apparently not so. We also recognise that these 

 series group themselves in many cases as branches of broader lines of descent ; for 

 example, in the Bicarpellatoe of Gamopetalte, in the ITelobiete of Monocotyledones. 

 To a greater or less degree such relationships are traceable now, and as we obtain 

 more knowledge of the angiospermous plant-life of the world they will be widened. 

 But this is a different thing from the carrying back the pedigree of every phylum 

 of dicotylous and monocotylous plants to one or other of the existing ones, which 

 may possess what are taken to be elementary characters. We have, so far as I 

 Imow, no evidence to sanction the belief, or even the expectation, that there is 

 extant any family of Dicotyledones or Monocotyledones which represents, even 

 approximately, a primitive type in either class. The stem in each has gone. We 

 have the twigs upon a few broken branches. 



Amongst the phyla we cannot discern any one type that can be described as tho 

 dominant one. The multifarious adaptability of the angiospermous type has 

 given us diverse forms, suited, as far as we discern, no less well to the varied 

 environments of our epoch. Yet we are .able to difierentiate certain of them 

 which take precedence alike in point of number of species and in area of distribu- 

 tion. If we seek for some general character that marks these advanced groups 

 we find it in the tendency to greater investiture of the ovule, both in Dicotyledones 

 and Monocotyledones. This is brought about in different ways ; for instance, by 

 the sinking of the gynseceum in the torus as in Compositfe, by inclusion within a 



' I leave it to Palieophytologists to say whether this construction may sometimes 

 account for the profusion of roots alongside of stem-structure in fossil-sections. 



