the Tracheae in the Ferns. 
52i 
according to the nature of the preservation, but in all cases they may be 
readily brought into agreement with the type of element described above, 
except the largest tracheae, which possess irregularly scattered more or less 
rounded pits. In these it is probable that each pit is an isolated perfora- 
tion whose cavity does not communicate with those of the others. 
It is no longer possible, therefore, to regard the xylem elements of the 
Osmundaceae as tracheides ; on the contrary, they are a special and very 
peculiar type of vessel. In each separate xylem mass of the stele there is 
a perfectly free passage for water in all directions from one element to 
another both vertically and horizontally ; to say nothing of the passages 
that occur in the very substance of the walls. 
In order to obtain some real genuine tracheides for the purpose of 
checking the observations made upon the xylem of the Osmundaceae, 
I next examined that of Nephrodium Filix-mas . In this plant the xylem 
elements are typically scalariform, with one series of pits only on each side 
of the wall. But here also it was found that the primary wall is com- 
pletely re-absorbed both in the region of the pits and between the bars 
of secondary thickening (Fig. 9). In fact, they have no right whatever to be 
called tracheides, but are vessels of the same type as those of the Osmun- 
daceae, the only difference being that they are typically scalariform. 
Pteris aquilina was next tried, and here a somewhat different type 
of element was met with, although still undoubtedly a vessel and not 
a tracheide. They are typically scalariform, and here again the pits, both 
on the end walls and on the side walls, are true perforations. They differ, 
however, from the vessels of the Osmundaceae and of Nephrodium , in that 
those parts of the primary wall that connect the opposite transverse bars 
of secondary thickening in the middle of the wall are here maintained 
intact even at maturity. The bars of each pair are, therefore, joined 
together by a cementing substance across the thickness of the wall. These 
persistent parts of the primary wall do not become lignified, but remain 
pectic in character and stain readily with ruthenium red. When a tracheal 
wall treated with this reagent is regarded in surface view, the pectic middle 
substance will appear as a red area shining through the lignified and 
unstained secondary layers and outlining the inner limits of the pit cavities 
as shown in Fig. 10. The shaded regions represent the pectic cementing 
substance connecting the bars in the middle of the wall, and it is seen that 
the widest extension of the pits greatly exceeds their apertures into the 
lumen of the vessel. 
Sachs in 1874 was already aware that the xylem elements of Pteris 
aquilina bore true perforations on their lateral walls. The figures he gives 
in his textbook (l.c. Fig. 27) illustrate most clearly and accurately the 
actual state of affairs. On the other hand, they escaped the notice of 
De Bary, Strasburger, and subsequent observers. Very probably this was 
