LINNEAN SOCIEXr OF LONDON. 2^ 



we know of. Most of them were plants of great size, with rather 

 sturdy stems, bearing immense fern-like fronds ; the habit alto- 

 gether must have been something like that of a Tree-fern, but 

 their reproduction was by large seeds, borne on the fronds. To 

 adapt the vascular system of the stem to the supply of the large 

 and compound leaves, the type of structure was assumed which 

 {pace Mr. Tansley) it is still convenient to call polystelic, i. e. the 

 single vascular cylinder (which may be recognized in some of the 

 earlier members of the group) became broken up, in various 

 Avays, into a number of distinct cylinders, only connected at 

 intervals. So far the change was in the same general direction 

 as in the evolution of the higher Eerns ; the fossil famil}^ how- 

 ever, was not content with a complex primary vascular system, 

 but must have secondary growth as well. Now if you have a 

 number of vascular columns in the same stem, each growing 

 continuously in thickness on its own account, it is evident that 

 very special arrangements will be necessary to avoid overcrowding. 

 The difficulty was overcome, and the MedulloseaB for some time 

 flourished among tlie dominant families — the Permian formation 

 represents their Golden age. But one is tempted to think that 

 the system was too complicated to last ; at any rate it seems 

 not to have lasted, for thi-se elaborate stems have not been found 

 in any later rocks. Either, as Mr. Worsdell supposes, the 

 Medullosean stem became reduced and simplified to form the 

 Cycadean type of stem of later days, or, as I am more inclined to 

 believe, the family died out altogether. Even here, though we 

 seem to have an instance of a cumbrous mechanism, overreaching 

 itself in elaboration, yet it worked well enough for a time, and it 

 would be difficult to say exactly what the conditions were that 

 led to its being superseded. 



One of the most striking results of modern palaeobotanical 

 research, led by Williamson, has been to show how widely spread 

 among Palaeozoic plants was the power of secondary growth b}' 

 means of cambium ; probably quite as large a proportion of the 

 whole Flora possessed it then, as now. To a certain extent 

 indeed, secondary growth has " gone out '"' since then, for the 

 very flourishing modern class Monocotj^ledons liave dropped it, 

 and for the most part have done very well without it, though 

 some few have tried to retrace their steps. In spite of this 

 important defection, it is evident that for most land-plants 

 secondary thickening has been a highly successful system, and it 

 is an interesting question whether there was ever a time without 

 it. Was the power of cambial growth, at some period or other 

 however remote, a new acquisition, or is it as old as the vascular 

 tissues themselves? Jn Palaeozoic times eveiy class of land-plant 

 possessed secondary growth in a greater or less degree — G-ymno- 

 sperms, Pteridosperms, Lycopods, iSphenophylls, Horsetails, and 

 even Ferns, though among true Ferns it seems never to have 

 amounted to very much. Widely spread as it was, the evidence 

 on the whole points to cambial growth having been a secondary 

 acquisition in the history of the race, as in that of the individual 



