111 
Sinnott.—Foliar Gaps in the Osmundaceae. 
where it is well out into the cortex a deep constriction appears opposite 
to it in the xylem ring. Thin serial sections were of course impossible 
to obtain, as in the living species. On the whole, it seems very probable 
that we have here to deal with a form somewhat resembling T. superba> 
where foliar gaps are always present, though sometimes so narrow that they 
are hard to make out. It would not be at all surprising if in the process of 
fossilization the structure of these rows of thin-walled cells should be 
entirely lost to view. The theory that such a form as O. Dunlopii has been 
derived from one with a stele of the type represented by Thamtiopteris 
Schlechtendalii , by the gradual alteration of the central tracheides into pith- 
parenchyma, is at present supported by too little evidence to warrant 
its acceptance. Most probably, if this were the course of evolution, the 
transitional forms between the two types would show leaf-traces departing 
without the formation of gaps in the xylem. These transitional forms are 
lacking, however, and, as we have shown above, wherever there is a ring 
of xylem around a pith of parenchyma, a departing leaf-trace, either at 
or soon after its separation from the stele, subtends a foliar gap. 
As far as it goes, therefore, the fossil evidence seems to support the 
view that foliar gaps are a primitive feature for the Osmundaceae. This 
testimony is not entirely sufficient, however, and we must turn to our other 
source of information, the structure of the conservative regions of the 
living members of the family. 
Of these portions of a plant which retain ancestral features, the seedling 
has long been recognized,, in connexion with the theory of recapitulation, to 
be one of the most important. The structure of the young plant in various 
members of the Osmundaceae has been investigated by Leclerc du Sablon, 
Faull, Seward and Ford, and Chandler. The first of these writers (8) 
worked out the structure and development of the vascular system in a young 
specimen of O. regalis. He found that there are originally two separate 
strands of xylem which fuse into a single protostele, surrounded by phloem. 
In the centre of this appears very soon a cluster of parenchyma-cells 
constituting a pith. Whenever a leaf-trace leaves the cylinder, it causes 
a break, which is soon repaired, in the continuity of the xylem ring. 
A little higher up in the stem, however, where the departing leaf-traces are 
numerous, gaps are formed more rapidly than they become closed, and the 
ring is consequently broken into a number of bundles, separated by ‘ rays ’ 
or foliar gaps. In no case, after the stele had assumed the tubular condition, 
was the absence of a foliar gap observed. 
Faull ( 9 ), who investigated young plants of O. cinnaniomea and 
O. Claytoniana , found a very similar state of affairs. The young stele 
possesses a continuous ring of xylem, which, however, is always broken at 
the departure of a leaf-trace. This break is subsequently closed, as in 
O. regalis . 
