EMBRYOGENY OF THE PTERIDOPHYTES 673 
of the greatest nutritive supply and take a strong curvature, as in Z. clavatum 
(Fig. 186). In others, again, it may provide the basal part of the embryo 
and root, without any swelling (Se/. spinulosa, Fig. 190), or an haustorial 
swelling may be formed, with convex curvature, on the side next to the 
food supply (Sed. Martensiz). In embryos without suspensor the hypobasal 
tier may maintain this same function, but it is usually only slightly enlarged 
(Eguisetum, Fig. 214; Fern, Fig. 15; Zsoedes, Fig. 191; Ophioglossum vulgatum, 
Fig. 260 4s; Botrychium Lunaria, Fig. 263). It would appear from the 
inconstancy of their development, and their position in relation to their 
obvious uses when present, that these haustorial growths are of the nature 
of relatively late and direct adaptations at or near to the basal region of the 
axis of the embryo, and it is significant that there is no special haustorial 
growth in Zyc. Selago or in Selag. spinulosa, both of them species believed to 
be primitive types of their respective genera. 
The extra-prothallial swellings, of the nature of protocorms, differ in 
origin and in function from the intra-prothallial haustoria (Figs. roz, 178, 
188): they spring from the epibasal tier, and do not serve as suckers. 
It has been argued at length above (p. 351, étc.) that there is good reason 
to believe them to be secondary in their origin: however greatly these 
gouty interludes may affect the form and appearance of the embryo, their 
effect is temporary, and the shoot ultimately. settles down into a normal 
Lycopodinous type. If this view of the protocorm as a special secondary 
development be accepted, then it may be put on one side as not directly 
affecting the bearings which embryogeny may have on the theory of 
origin of the shoot. 
The various types of embryogeny observed among Pteridophytes have 
now been reviewed, and it remains to attempt to separate the characters 
which are secondary, special, and fluctuating, from those which are primary 
and constant, with a view to some general estimate of the embryogeny 
as an aid to a morphological conception of the shoot. Following the 
reasoning contained in the preceding pages, the occasional swellings of 
the nature of a protocorm or of a haustorium, together with the curvatures 
and distortions which these often produce, may be set aside as secondary ; 
similarly, the precocious developments of root or of leaf, which sometimes 
upset the balance of parts in the embryo, may be set aside as special 
biological adaptations; for even where a cotyledon or a root appears 
early and anticipates apparently the other parts, still in all accurately 
observed cases the relation of the axis to the primary segmentations is 
found to be maintained. Further, the position of the first root is always 
lateral; its orientation and level of origin varies, as well as the time of 
its appearance: these facts point to its being an accessory part upon the 
embryonic body. The first foliar development is inconstant in position 
and time and number of the leaves, but it is constant in the fact that 
the protophylls are always lateral with respect to the point where the 
axis will appear, and orientated with regard to it, so that more or less 
2U 
