Relation of Pteridosperm Anatomy to that of Cycads. 49 
of Cycadean descent, but Lyginopteris may represent a stock from 
which the Cycadoxylese arose as a side line, while the Cycads 
formed a more successful line of descent from the same stock. The 
so-called vascular anomalies appearing in some of the Cycad genera 
may be the expression of potentialities of development inherited 
from the ancestral stock, these potentialities manifesting themselves 
constantly in the other group, the Cycadoxylese. The parenchy¬ 
matous nature of the secondary wood, the gradual disappearance 
of the centripetal wood of the stem, and its retention in the leaves, 
characterise both lines. 
Prof. Weiss, in his Presidential Address to Section K at the 
British Association Meeting of 1911 (18, p. 5), draws attention to 
Chodat’s criticism of the view that the Lyginopterideae are the 
ancestral types of the Cycads (2). Chodat bases his criticism on 
the work of Bertrand and Cornaille (1) concerning the vascular 
system of living ferns, and is of the opinion that Scott’s interpre¬ 
tation of the primary vascular structure of Lyginopteris needs 
revision. He believes that the so-called “ mesarch ” structure of 
the primary wood and leaf-traces of this plant is of a different 
nature from that of the petoliar bundles of Cycads; in fact, 
according to Chodat, Lyginopteris is merely a specialised fern 
having no particular relation with the Cycads. In that form (and 
Heterangium) the primary bundles of the stem, though apparently 
mesarch, are really endarch, the whole of the metaxylem developing 
centrifugally. The metaxylem becomes extended laterally, and then 
curves inwards in the form of an inverted omega (a>). The lateral 
extensions unite on the inner side of the protoxylem, enclosing 
it, together with a few parenchymatous cells, thus forming a 
closed “ divergent ”—a typically filicinean unit of vascular 
structure. In Cycads, however, the vascular bundles of the 
petiole are truly mesarch, consisting chiefly of centripetal xylem 
(except near the base of the leaf-stalk); the protoxylem is in 
connection with the centripetal xylem, and is usually separated 
from the centrifugal xylem by a few parenchymatous cells. 
According to Le Goc (6), the centrifugal xylem is largely secondary 
in origin (particularly at the base of the petiole), while the 
centripetal xylem is wholly primary ; they are thus independent 
morphologically, though more or less secondarily connected for 
physiological reasons. Marsh (6a), however, while agreeing with 
Le Goc as to the nature of the centripetal and centrifugal xylem, 
maintains that there is a primary connection between them. The 
