SEEDLING STRUCTURE IN THE ТЕССМІХОХ ДЕ, 93 
though epigeal, has cotyledons that do not expand and show only feeble 
development of chlorophyll *. 
We may conclude that the limits between hypogeal and epigeal ger- 
mination are easily transgressed in either direction ; that the question 
whether, in any particular case, the type of germination is primitive or 
derived must be judged upon its own merits ; and that the difference 
probably depends rather on proximate causes than on hereditary conformity 
to an ancestral type. 
Тнк EPICOTYL IN THE VICIEE. 
The structure of the epicotyl in the majority of the hypogeal Vicieæ 
(except Cicer arietinum) shows a number of peculiar features which have 
attracted the attention of many anatomists. These are as follows :— 
(i.) The presence of a cortical system of bundles running down the four 
ridges of the quadrangular stem. In the two ridges in the plane of the 
distichous leaves, i.e., in the intereotyledonary plane, run stout strands of 
thick-walled fibres. In each of the lateral ridges, in the plane of the stipules, 
runs a collateral fibrovascular bundle consisting of a small group of xylem 
and phloem in the normal orientation, often slightly increased by a cambium, 
and a group of fibres external to the phloem, each bundle being enclosed by 
an endodermis. The fibre bundles enter the leaf midrib, and the lateral 
fibrovascular bundles chiefly supply the large leafy stipules. The course of 
these bundles has received much attention and has been thoroughly 
elucidated f. 
(ii.) The presence, in the lowest 1-4 internodes, of a solid central core of 
primary xylem, and the absence of pith. In the higher internodes this 
central xylem strand is absent. This anomalous structure has received more 
than one interpretation. Mlle. Sophie Goldsmith (1876), who studied the 
seedling of Vicia sativa with great care, found that the central xylem is 
formed in a centripetal direction, differentiation commencing at the foci 
of the elliptical pro -ambium strand and proceeding in all directions ; thus 
forming also the centrifugal median xylem bundles of the leaves. She 
remarks (p. 43): “die Entwickelung der primordialen Vasalstriinge in den 
untern Stengelinternodien kann eine centripetal-centrifugale sein und dadureh 
* Tschermak (1904) showed that in Phaseolus hypogeal germination behaves as a 
Mendelian dominant to epigeal ; and as in most instances the phylogenetically older character 
is dominant to the younger, this may be taken as confirmatory of the primitiveness of 
hypogeal germination in this case. 
+ See especially Goldsmith, 1876; also Lestiboudois, 1848, p. 19; Nageli, 1858, p. 85; 
Van Tieghem, 1884, p. 133 (deals with fibro-vascular bundles only, and neglects the fibre 
bundles) ; Hérail, 1885, p. 218 (mentions that Cicer arietinum, an aberrant member of the 
tribe, is without cortical bundles). 
