MUCILAGE-CANALS OF THE MARATTIACER. 449 
marked ring in the cortex. Here there is no appearance of 
an epithelium whatever, and the parenchymatous cells of this 
region secrete mucilage into the intercellular spaces, which 
thereby become much distended, ultimately breaking down the 
adjacent cells and giving rise to a great cavity filled with muci- 
lage. It is probable that there is a considerable amount of 
mucilaginous degeneration of the cell-walls, and it is even quite 
possible that the cavities are primarily due to this. These 
cavities filled with mucilage would be well described by Tschirsch’s 
term “schizo-lysigenous”’*. In portions in which the whole of 
the xylem had already become lignified, these peculiar mucilage- 
cavities seemed to be in many stages of development. 
In the aerial roots of Marattia attenuata, although the origin of 
the canals is obviously schizogenous, and a well-marked epithelium 
is early distinguishable, yet the initial meristematic groups are not 
clearly differentiated in the young parenchyma. Moreover, ata 
very early stage the schizogenous space is filled with a deeply- 
staining substance, probably mucilage, which does not seem to 
be materially added to as the cavity becomes larger, for it is not 
till the canals are approaching the adult condition that they are 
again filled with a clearly recognizable mucilage. It may be that 
the mucilage secreted at this early stage is correlated with the 
formation of the schizogenous space. In the roots of this species 
the young canal is often very irregular in shape, and bridge-cells 
are very frequent. When the canal becomes too wide for these 
cells to accommodate themselves to the increasing tension, they 
are torn away at certain parts, and in transverse section a number 
of them are frequently seen projecting into the cavity, beyond 
the other epithelial cells, much more prominently than, though 
similar to, d7.c’. in fig. 8. 
Although the object of the present investigation was simply 
to determine whether or not there were true mucilage-canals 
developed schizogenously in the Marattiacex, a word may not be 
out of place with regard to the question whether secretory glands 
and canals conform to one typical mode of origin or another, or 
whether there is likewise a mixed development. Althougha large 
amount of work has already been done in this direction, the matter 
hardly seems finally settled, even by Mlle. Leblois’ work t. She 
* « Angewandte Pflanzenanatomie,’ vol. i. p. 517 (1889). 
+ Leblois, Mlle. A., “ Recherches sur l’origine et le développement des canaux 
sécréteurs et des poches sécrétrices.” ‘ Annales des Sciences Naturelles,’ 7¢ série, 
Botanique, tom. v., vi. (1887). 
