366 ROYAL SOCIETY OF CANADA 
The object of the investigation was to determine the possible cor- 
relation of the entire structure with :— 
(a) The mucilage canals of the Marattiacee. 
(b) The resin canals of the Gymnosperms. 
(c) The parichnos of fossil and recent Lycopods. 
It resulted, however, in showing the correlation of the structure 
with :— 
(d) The carinal canals of the Equisetineæ, and the canals which 
are found adjacent to the protoxylem in forms like Zea Mays, Trades- 
cantia, and other phanerogams. 
The research was carried on through the various families of the 
Pteridophyta, as far as the material available would allow, and resulted 
in demonstrating the universal occurrence of the phenomenon in the 
Pteridophyta as a whole. 
In Pteris aquilina all the stages of development were observed, as 
material of all the parts at different periods of growth was available; 
in the other forms only those stages were observed that the material at 
hand would allow. 
In an adult plant of Pteris, as taken in the fall of the year (about 
Sept.) the following parts are found :— 
(1) A long rhizome divided off into nodes and internodes and 
bearing, 
(2) short branches at the nodes, which again bear 
(3) stipes, and 
(4) buds which give rise to fronds during the next year’s growth. 
(5) The growing point of the rhizome, which is free from 
branches, and varies greatly in length. For the purpose of the follow- 
ing investigation all these parts were thoroughly examined, and in the 
case of the stipe all the stages of growth—from the initiation of growth 
in the bud in spring up to the complete maturation of the frond—were 
obtained, and the differences at the various periods noted. 
It was thus possible to ascertain the distribution of the structure 
as between the various parts of the stipe and rhizome and with refer- 
ence to the age of the parts, as well as its origin. 
In the following account the bud will be dealt with first, as it re- 
presents the earliest appearance of the stipe, and thus shows the origin 
of the structure very clearly. The buds vary considerably in size and in 
the numerous specimens examined, ranged from about % in. in length 
to about 4 in. As a whole, the buds are regions of relatively slow 
growth, and this fact has an important bearing on the phenomenon 
under discussion, for, as will be shown further on, the structure in ques- 

