522 Gwynne- Vaughan. — On the Real Nature of 
due to a natural tendency to cut as near to the apex as possible in order to 
avoid the inconvenience caused by the mature sclerenchyma. The tracheae 
in such regions would be too young for the complete disappearance of the 
primary wall. 
Vessels of the Pteris type seem to be very widely distributed among 
the Pteridophyta. They were observed in the Polypodiaceae, Cyatheaceae, 
Hymenophyllaceae, Gleicheniaceae, Schizeaceae, Marattiaceae, Ophioglos- 
saceae, Lycopodiaceae, and in the fossils Psaronius , Botryopteris , and 
Zygopteris. With regard to Zygopteris , Professor Weiss has pointed out 
to me that the presence of open passages in the side walls of the tracheae 
renders it easy to account for the thyloses he found far away from any 
living elements in the most central tracheae of the non-parenchymatous 
xylem mass. 1 
As a matter of fact, true and indisputably imperforate tracheides 
were not met, with in any one of the Pteridophyta examined. These were, 
however, far too few to permit any generalization. On the contrary, true 
tracheides must occur in the Calamites at any rate, for it appears that 
vertical striations can be observed on their pit-membranes. 
Even in the Gymnosperms the true tracheide is by no means so 
general a type as might be supposed. From some preliminary prepara- 
tions made by Miss S. Greves that I have been permitted to examine, it 
appears that true perforations occur in the longitudinal walls both in the 
primary and secondary xylem of some of the Cycads. This is probably 
the case also in the secondary wood of Araucaria ; at any rate, it has not 
been found possible to demonstrate a pit-membrane there. 
Considered from a mechanical point of view, elements of the Osmun- 
daceous type would appear to be very unsuitable for conferring strength and 
rigidity to the xylem as a whole. The fact that in the Ferns as a whole, 
and particularly in the Osmundaceae, the function of support is taken over 
almost entirely by masses of strong sclerenchyma in the ground tissue, has 
perhaps rendered the existence of this type of vessel possible. Even under 
these conditions it seems that mechanical exigencies prevent the width of 
the tracheal wall exceeding a certain limit. In the Osmundaceae the limit 
of width for a wall with a single series of pits is about 45 /x. Walls of 
greater width have the bars cemented together about half-way across, so 
that there are two series of pits, and may attain as much as 70 /x. With 
three series of pits they reach 90 j u, and with four up to 104 /x. In Nephro- 
dium Filix-mas , walls with a single series were found as much as 55 /x wide. 
Of course, when each pair of bars is cemented together along the whole 
width of the wall, as in Pteris aquilina , the mechanical efficiency is greatly 
increased and walls with a single series of pits may be as wide as 130 /x. 
1 F. E. Weiss, • On the Thyloses of Rachiopteris corrugata New Phyt., vol. iv, No. 4, 
p. 82, 1906. 
