59i 
Centripetal Xylem in Equisetum. 
through continuous lines of tracheides, into centrifugal, as the node is 
reached, and vice versa as the node is left in passing upward along the 
bundle. Such a condition is difficult to imagine, and is unknown elsewhere 
in the anatomy of stems. Assuming the alternative view, that these 
c centripetal ’ groups pass over the supra-nodal wood, an absurd condition 
is likewise presented, for we have centripetal wood, the suggested possible 
‘ remnants of a primitive central mass 5 of such wood, passing external to 
centrifugal wood, the remains of a mass to which it formerly lay internally. 
Such relation obviously cannot hold. Moreover, in the two species, 
E. hyemale and E. maximum , in which the statement is made that such 
a condition exists, there has been shown to be at the point in question 
some cambial growth. Hence the centripetal wood there passes outside 
secondary wood\ Yet, according to Gwynne-Vaughan (p. 779 of the above- 
mentioned paper), the view that these bundles are the vestiges of an ances- 
tral, solid, internal stelar mass is ‘ entirely in agreement with their apparent 
centripetal development and in particular with their cauline course ’. (The 
italics are the present writer’s.) Again, the metaxyiem-groups lie externally 
to the protoxylem, which is granted, apparently by all, to be endarch. Can 
the former then, if exarch , be the vestiges of tissues central in position, and 
thus necessarily internal to the carinal canals, as is the centripetal wood in 
C. pettycurensis ? The carinal canals of Equisetum and of the Calamites 
are indisputably homologous. 
It is further stated that the gradation in size and the sculpture of the 
walls of the tracheides of the lateral groups indicate a centripetal develop- 
ment. Size does not necessarily indicate the order of development ; and 
the sculpture of the tracheide-walls in metaxylem is, at least to a certain 
extent, independent of order of development. In his 4 Observations on the 
Anatomy of Solenostelic Ferns ’, Mr. Gwynne-Vaughan 1 mentions cases in 
the stems of which the first-formed tracheides ■ appear without order here 
and there throughout the whole xylem-mass, so that the differentiation is 
quite irregular’. Under these conditions, the sculpture is always scalari- 
form, and ‘ there is no difference whatever between the first-formed elements 
of the xylem and those formed later on’. Plate XLV, Fig. 1, shows 
a somewhat similar condition in Lycopodium. Here, though, a definite 
protoxylem has first formed ; lignification proceeds in general centripetally, 
but with much irregularity. Even in secondary wood completion of ligni- 
fication does not proceed in perfect centrifugal order. This was seen by the 
writer, for example, in the maturing wood of Betida. The loss of protoplasm, 
indicating the completion of lignification in tracheides, is decidedly irregular. 2 
1 Gwynne-Vaughan, D. T. : Observations on the Anatomy of Solenostelic Ferns. Ann. Bot., 
vol. xvii, p. 727 ; 1903. 
2 Special precautions were taken to assure the existence of this condition. Longitudinal sections 
of the wood were studied. The length of the tracheides and the shrinkage of their protoplasm in 
length were observed. These elements are formed in rather distinct tiers, and practically no 
