360 ROYAL SOCIETY OF CANADA 
Strange ” to groups of mucilage-bearing cells which alternate with them 
in the vascular bundle of Osmunda. This regular alternation of the two 
kinds of cells was first noticed by the writer in Todea barbara, and later 
in Osmunda regalis, and will be dealt with below. 
It is evident from the above account, that, although Terletzki dis- 
cussed the structures in question at some length, still he had no idea as 
to their mode of origination or of their function. Moreover, as will be 
seen below, he only saw the structure at one stage of its existence and 
therefore regarded it as a form of lacunar parenchyma peculiar to the 
xylem in bundles of certain shape. 
Thome (22) is the next author who makes mention of the struc- 
tures which form the subject of this article, and he is the first to give a 
fairly correct interpretation of some of the phenomena. In speaking of 
the vascular bundles in the stipes of ferns, he says (22, 125): “The 
scalariform tracheids form the main mass of the wood; next to them 
(daneben) there occur. everywhere several spiral, ring (annular), and 
reticulated tracheids which are much narrower than the majority of the 
scalariform tracheids, and which occur in small groups. They are the 
elements of the vessels which are first formed; the formation of the 
scalariform vessels occurs much later, proceeding from these in trans- 
verse section. Russow called them protoxylem cells, and the groups 
formed by them protoxylem groups. These groups lie everywhere either 
at the periphery of the xylem, or very near the periphery; in many cases 
they also lie isolated in the parenchyma.”........ “ On the formation 
of the scalariform tracheids they become functionless and tear away, 
mostly in the direction of the covering which they face (in Folge des 
Zuges, dem sie ausgesetzt sind)’ As will be shown below, this latter 
statement is by no means valid; for in many cases the protoxylem groups 
remain intact long after the secondary xylem is formed, and as a matter 
of fact never break up, whilst in the other cases they begin to disinte- 
grate long before there is any sign of the formation of secondary xylem. 
This point is especially well illustrated in certain preparations of Pteris 
aquilina, Todea barbara, Osmunda regalis, and other forms that have 
been examined by the writer. 
Continuing from the last passage, he says: “An air-conducting 
canal which contains the torn residue arises in their place. In most 
cases the walls of the parenchyma surrounding such a passage, extend 
into it in the form of protrusions like papille or tubes (wolben sich 
papillenartig oder schlauchformig vor), and fill it to a considerable ex- 
tent; however, they leave wide lacunæ between themselves. Where 
they come in contact, there arise, through reciprocal pressure, polygonal 
areas. Frequently the protoxylem appears crushed. Sometimes the 
nn 
