28 ROYAL SOCIETY OF CANADA 
by decay until they eventually came to or slightly below the level of the 
surface of the marsh and so passed beyond direct observation, because 
of the concealment afforded by the grass. In many cases, however, the 
disintegration of the stumps continued much beyond this point with 
the result that with their more or less complete removal, their places 
became occupied by round holes of the same depth. Such holes may be 
observed from the railway track, and they are found to occur at intervals 
over the entire area of the marsh. One of the significant features of 
these holes is to be found in the evidence they afford that they were 
formed after the marsh turf had attained its present thickness or nearly 
so, and the disappearance of the trees must therefore be of very recent 
date. 
Finally, in this buried forest we have the most conclusive proof 
that the history of this marsh was in no way connected with tidal flow 
until a very late period in its development, and that this relation was 
established within very recent times cannot be doubted. A further 
discussion of this aspect of the problem will be deferred until we come 
to a consideration of the particular circumstances under which the very 
striking change here indicated was brought about. 
In the deeper portions of the trench there are vast quantities of 
fragments of twigs about two or four inches in length, and about the 
diameter of a lead pencil or less. This material presented the external 
appearance of perfect preservation, but it was really so completely 
disorganized as to have the consistency of putty, and upon drying out it 
contracted to a very small volume. It was, therefore, impossible to 
submit the fragments to microscopical examination, but in other ways 
it was possible to determine that they represent the residue of what must 
have been a very luxuriant growth of Cassandra calyculata. 
In very close association with these cassandra twigs, in fact forming 
an essential part of the same stratum, there were numerous small leaves 
which retained all their original structural features in a most remarkable 
manner, so long as they were kept moist, but desiccation brought about 
speedy disorganization. In many cases the green colour was preserved 
in a very remarkable manner. I had been informed of this fact by the 
foreman in charge of the excavation, but I felt very sceptical as to the 
correctness of his observation until verified by my own examination. 
It was a very simple matter to determine the ericaceous character of 
these leaves which were packed in closely in a layer of considerable thick- 
ness, precisely as we find at the present day in bogs of the cassandra 
stage of growth; and it was almost as easy to correlate them with 
Cassandra calyculata and Vaccinium macrocarpon. There were no 
doubt other species represented, but, if present, they were certainly in 
