690 ° CONCLUSION 
evolution of the race, the evidence is quite clear: it indicates that the 
large-leaved forms, in which solenosteli¢ or dictyostelic structure rules, 
originated from a smaller-leaved ancestry, with protostelic structure and a 
single strand of the leaf-trace. This is in full accord ‘with probability, 
according to the antithetic theory of origin of the leafy sporophyte ; for on 
that theory smaller-leaved would necessarily have preceded larger-leaved 
types.! 
Another mode of amplification of the stele, which often accompanies 
the first but is not necessarily associated with it, is by secondary thickening. 
The stem of Sphenophyllum (Fig. 217), and of Lepidodendron Petticurensis,® 
are examples of how a secondary development of vascular tissue may 
surround a solid protostele: this shows that medullation does not neces- 
sarily precede secondary thickening, but commonly the secondary thickening 
occurs where medullation is present: and indeed in some cases the two 
are in a sense complimentary, the secondary vascular tissue taking the 
place functionally of the primary tissue reduced by medullation; this is 
exemplified in the Calamarians (Fig. 225) and in Szgz//arza,? as also in 
some forms of Stigmaria,* and it is seen with special clearness in Lygino- 
dendron, Poroxylon, etc. In other types structurally more advanced, the 
secondary development may be held to have completely replaced the 
centripetal wood of the original stele. 
The distribution of secondary vascular development among the Pteri- 
dophyta indicates clearly that it is a phyletic afterthought, originated in 
relation to the increasing size of the vegetative system consequent upon 
continued apical growth, repeated branching, and leaf-enlargement, either 
separate or in combination. Enlargement of the primary stele, with or 
without attendant medullation, may meet the demand in some degree; 
but it is a fixed and limited scheme compared with that of secondary 
thickening, which is capablg of increasing the conducting tract in proportion 
to the demand. In some cases, however, it appears that a phyletic decrease 
of the secondary development has occurred, and it is probable that the 
feeble cambial activity in the nodes of Zyuisetum, and locally in the 
Psilotaceae may be vestigial remains of a more active increase in their 
predecessors, allied respectively to the ancient Calamarians and Sphenophylls. 
1This is, however, quite contrary to the opinions of Dr. Jeffrey, who holds that the 
large-leaved and small-leaved stocks were ‘‘separate back to the beginning of the period 
when the palaeontological record begins.” This view would recognise no transition from 
the structure characteristic of the smaller-leaved forms (cladosiphonic) to that characteristic 
of the larger-leaved (phyllosiphonic). But, as a matter of fact, this can be demonstrated 
to have occurred in the individual life of Ferns, and probably it has occurred also in other 
forms in the passage from small-leaved youth to large-leaved maturity. It has been 
pointed out repeatedly in Part II. how cladosiphony is the anatomical expression of the 
dominance of axis, phyllosiphony that of the leaf in the shoot: and the balance may 
be altered in the individual life. (See Jeffrey, Phz2. Trans., 1902, vol. 195, p. 144.) 
2Kidston, Proc. Koy. Soc., Edin., 1906-7, p. 208. 
3 Scott, Studies, Figs. 77-78. 47b., p. 234. 
