[PENHALLOwW] CRETACEOUS AND TERTIARY PLANTS OF CANADA 38 
4. A narrow zone of phloem. 
5. A single row of large, scalariform tracheids. 
6. A narrow zone of protoxylem elements. 
?. Small celled parenchyma of the central region filled with pro- 
toplasm, but with no starch. 
In the xylem and other elements, the cell wall has been so far 
altered by decay that it is impossible to determine the specific structural 
features, and it is therefore impossible to establish the precise contact 
with Osmunda in this respect, but the features thus described indicate 
most clearly the general relation to the Osmundaceæ. Specifically the 
type differs from Osmunda and approaches Todea in the following 
respects :— 
1. The broad zone of thin-walled parenchyma internal to the sheath. 
2. The greater abundance of mucilage in large cells within the phloem 
and protoxylem regions. 
3. The zone of thin-walled parenchyma in the Donner region. 
It differs from the type of Todea and thereby approaches Osmunda 
in the following respects :— 
1. The special form of the stele. 
2. The occurrence of the secondary xylem in a single row. 
The great excess of mucilage in the phloem region. 
The great volume of the phloem parenchyma. 
5. The greatly diminished size of the cells in the central parenchyma. 
fe 
These relations are supported by the evidence afforded by the 
foliage and we cannot otherwise than conclude that the plant under 
consideration must have been a true Osmunda. 
In 1898 a rhizome of an Osmunda from Alliford Bay, Skidegate 
Inlet, Queen Charlotte Islands, was described by me under the name of 
Osmundites skidegatensis.1 There were also found certain fragments 
of foliage in all probability belonging to the same plant. Upon com- 
parison with existing species, it was found that the material might be 
referred to the type presented in Osmunda claytoniana. The results 
of these studies are now published in connection with this paper. It 
is worthy of note that the two specimens from the same horizon, and 
essentially from the same locality, are both to be referred to Osmunda 
of the type of O. claytoniana, and there can be little reason to doubt 

7 "Trans. Re soc, Cane) Valley ive, Se 
