Internodal Vascular Strands of Equisetum. 28 
have been caused by the obsolescence of the elements of the meta- 
xylem abutting on the protoxylem-group, any of the elements of 
the metaxylem may cease to develop as tracheides. This is only 
what we should expect in view of the fact that sometimes there is 
no lateral metaxylem. 
If we accept the view that the lateral groups of metaxylem are 
parts separated from the protoxylem relatively recently in the 
phylogeny, the question as to whether these lateral groups con¬ 
tribute to the formation of the trace seems of less importance. It 
is possible to reconcile the late Mr. Gwynne-Vaughan’s denial of 
the existence of this contribution (9) with Mr. Eames’ assertion of 
such a contribution (7, p. 593) by supposing that the metaxylem 
may contribute to the trace if the latter is unusually large or the 
carinal group of protoxylem relatively less well-developed, while in 
other cases the carinal group of tracheides may itself suffice for 
the emission of traces. 
We have not so far discussed the vexed question of the direction 
of lignification of the lateral groups of metaxylem. The fact that, 
more often than is usually believed to be the case, they are not separate 
from the carinal canal is not, of course, incompatible with their 
centripetal development. Still, very possibly, but for their supposed 
independence from the protoxylem, the late Mr. Gwynne-Vaughan 
might never have asserted their probable homology with centripetal 
xylem. This author did not mention, possibly he overlooked, an 
earlier assertion than his of the centripetal development of the 
lateral strands of wood. Poirault, in a paper read as early as 1887, 
and published in 1890, stated that the lateral xylem of the bundle of 
Equisetum developed centripetally (14, p. 13). It is true that he 
called this lateral wood “ secondaire ”—a misleading expression, 
for to judge from his description he does not regard the lateral 
tracheides as being formed by a cambium, but merely uses the 
word “ secondaire ” to bring out the contrast between the meta¬ 
xylem and the protoxylem, the elements of which he describes as 
“ primary vessels.” Since the first publication of Mr. Gwynne- 
Vaughan’s views, however, have appeared Queva’s excellent figures 
of developmental stages of the metaxylem. The latter’s very 
careful descriptions, based on material giving remarkably complete 
series of stages, seem to show that the lateral metaxylem develops 
centrifugally (15), though probably, as pointed out by Mr. Eames 
(8, p. 592), there is some irregularity in development. 
The nature of the ring of short reticulate tracheides known as 
the nodal, or more accurately the supra-nodal xylem, is one which 
