464 BOTANICAL GAZETTE [JUNE 
PorrAvuLt (13) asserted that the metaxylem developed centripe- 
tally, and eleven years later GWYNNE-VAUGHAN was led by his 
study of the vascular system of E. gigantewm to adopt a similar 
view, although in the absence of an incompletely differentiated 
portion of the axis of this species he considered the centripetal 
development of the xylem as not established. As has been pointed 
out, a study of the young internode of E. giganteum confirms 
GWYNNE-VAUGHAN’S opinion, and it seems clear that in this 
species the direction of lignification, although subject to slight 
irregularities, is from without inward. On the other hand, QuEVA 
in 1907 showed clearly that in E. maximum the lateral metaxylem 
of the internode was differentiated centrifugally. Eames claimed 
that although there is a good deal of irregularity in the direction 
of its lignification, the internodal metaxylem was differentiated 
centrifugally in the majority of the bundles of E. maximum, 
E. arvense, and E. hiemale. This has recently been confirmed by 
Miss Barratt for the first two species. Serial sections of the 
internodes of young cone-bearing branches of E. arvense, E. limosum, 
and E. debile, in which the lateral metaxylem was incompletely 
differentiated, were examined for comparison with the young inter- 
node of E. giganteum. In all of them the differentiation of the 
metaxylem, although subject to occasional irregularities, as EAMES 
has pointed out, was in the great majority of cases clearly centrifugal. 
It seems difficult to doubt the essential homology of the char- 
acteristic lateral groups of metaxylem throughout the genus 
Equisetum. The question, therefore, arises whether the primitive 
order of development of the metaxylem was centripetal or cen- 
trifugal. It is undeniable that the metaxylem, both nodal and 
internodal, is better developed in E. gigantewm than in any of the 
other species the anatomy of which has been studied. Moreover, 
in a genus showing an obviously reduced stelar structure the 
species with the largest amount of xylem would naturally seem 
to be the most primitive. On the other hand, E. gigantewm is 
exceptional, so far as we know unique, in the centripetal develop- 
ment of its metaxylem. It is even possible that such centripetal 
development might be confined to the base of the internodes, 
although on general grounds and in view of the constantly much 
