54 DR. D. H. SCOTT ON A NEW TYPE OF MEDULLOSE 
bearing much resemblance to that of the stele itself, on a small scale *. There is, so 
far, a marked agreement with the meristele of Suteliffia. In the Medwullosa, however, 
the main leaf-traces break up completely into the collateral bundles which enter the 
bases of the leaves; there is no evidence of fusion with neighbouring strands, nor any 
indication that a part of the strand remained behind in the stem after the foliar bundles 
were given off; the strands, in fact, constitute a leaf-trace system, and nothing more, 
They cannot, therefore, be directly compared to the meristeles of Suteliffia. 
On the other hand, the meristeles of the latter genus are evidently very different 
from the steles of JMedwilosa anglica, which are all of equal value, and constitute 
collectively the whole vascular system proper to the stem. The meristeles of Suécliffia 
can at most be regarded as subsidiary steles. 
The complex Permian Medullosee do not seem to afford any satisfactory analogies, 
for in them the principal steles are peripheral; the “star-rings” might be called 
subsidiary steles, but their position is central, and they have no direct relation to the 
leaf-traces. 
Sutcliffia is, in fact, an isolated type at present, and the interpretation of its structure 
correspondingly difficult. It is possible that we have in this genus the first beginning 
of dialystelic structure. ‘The Medullosez generally are characterized by the enormous 
development of their leaves, demanding an exceedingly complex and extensive 
vascular supply, which the simple protostele is evidently ill-adapted to provide. In all 
members of the family hitherto known, the megaphyllous requirements are met by the 
breaking up of the stele into a more or less elaborate system of distinct but anastomosing 
cylinders, some or all of which, as the case may be, are concerned in the provision of 
leaf-traces. In Suteliffia we have the unique ease of the protostele persisting as the 
main vascular axis of the stem, while giving off a peripheral system of subsidiary steles 
or meristeles, which form the points of departure for the actual leaf-traces. This seems 
the simplest explanation of the facts, and, if accepted, it probably involves our regarding 
Sutcliffia as the most primitive, anatomically, of the Medullosez. It might, no doubt, 
be possible to interpret the structure of Swtcliffia on an hypothesis of reduction; but 
considering the large dimensions of the plant and its high organization, such a suggestion 
seems to have no adequate basis. The structure of the petiole, on the other hand, 
fits in well with the idea of a relatively primitive position. There is one point, how- 
ever, which suggests eaution in our conclusions. Some of the detached petioles are 
very much larger than the leaf-bases attached to the one specimen of the stem which we 
possess. Hence it is evident that this specimen was not a full-sized one. It may have 
belonged to a young plant, or to the basal part of an older one, and it is quite possible 
that the larger stems may have had a more complex structure, just as we find in so 
many Ferns at the present day.  Swtcliffie, it is true, was not a Fern, and it may be 
that its vascular system acquired its mature form merely by secondary growth, only 
just beginning in our specimen. The possibility, however, of an increasing complexity of 
the primary structure in the later-formed regions of the stem must not be lost sight of. 
* Seott, 1599, pp. 92, 93; pl. 6. phots. 6-9; pl. 11. fig. 10. 
VT TE T ee ee UR Te RUPTA YE Ree pte e 
