Secondary Xy/em in Psilotum . 5 1 1 
ordinary rhizome, since the nature of the branch, shown in section in Fig. 3, 
might be disputed. Examples were found, however, in which true rhizome 
branches attached to the stock of an aerial shoot contained a few secondary 
tracheides 1 . 
We must now discuss the reasons for regarding the outer tracheides as 
secondary. An important point is the late differentiation of the earliest of 
these elements. If one takes the case of the stock of an aerial shoot with 
diarch structure, the xylem differentiates centripetally until complete, and 
there appears to be a considerable pause before the outer tracheides arise. 
In this respect the fact is significant that Bertrand does not appear to have 
seen secondary tracheides in any part of the plant. This appeared so 
remarkable that the question suggested itself whether the production of 
these elements might not be exceptional and dependent on special conditions 
of growth, or perhaps shown only by one variety of the species. This 
however is improbable, because I learn from Miss S. O. Ford and from 
Miss E. N. Thomas that they also have found these elements. Hence it is 
probable that Bertrand did not examine material of suitable age. At what 
distance from the apex they first begin to develop could not be determined. 
Another fact suggestive of secondary growth is that successive 
differentiation of these tracheides is still proceeding in quite old parts of 
the subterranean stem, a long way back from the aerial branch. Thus at 
a in Fig. 44 in the text, differentiation is still going on, and the basipetal 
distance of this from the cut end of the aerial branch g (where ‘ primary ? 
development is complete) amounts to about 10 cm. 2 The stem at a is 
probably several months old. There is nothing to suggest that the 
apparently developing tracheides are permanently arrested immature 
elements, since the outer tracheides in the oldest stems examined were all 
mature, and true arrest in the metaxylem of one branch showed quite 
different characters. 
No definite cambium is present, but radial arrangement is often shown 
to a slight extent. In Fig. 6, PL XXXIII, r marks a radial row of five 
elements, all parenchymatous except the outermost one, which is a young 
tracheide containing protoplasm. In a few cases parenchymatous rays 
occur opposite the protoxylem-groups ; a rather good case is seen at r in 
Fig; 5. These last two features, though not particularly well marked here, 
are familiar characters of secondary xylem. 
Taking all these facts into consideration one is justified in assuming 
that the late-formed tracheides represent a comparatively slight amount of 
secondary xylem, and that this is probably a reduction from more normal 
secondary thickening is suggested by the fact that in Psilotum the leaves 
1 In other cases, where the branches were older and their rhizomic nature less certain, numerous 
secondary tracheides were present. 
2 Several centimeters of branched stem had been cut away above g. 
