82 OPHIOGLOSSACEAE [CH. 



followed by less regular divisions which disguise them in the resulting ellipsoid 

 body. The exact limits between the epibasal and hypobasal parts are lost, 

 and owing to the late origin of the several parts of the embryo Bruchmann 

 found it impossible to refer them to one source or to the other (Fig. 383). 

 The root, which is organised early, grows first in a horizontal direction, and 

 bursts laterally out from the prothallus, but the remainder of the embryo 

 rests within the prothallus, where a distended foot is formed. On this ovoid 

 cellular body, and opposite the neck of the archegonium, the apical cell of 

 the stem arises: it is immediately overarched by a small growth which 

 Bruchmann takes for a rudimentary cotyledon, and the embryo is now as 

 in Fig. 384, A. Even at this early stage it contains the endophytic fungus. 

 Successive roots may then follow, while the growth of the bud remains in 

 this species almost quiescent, though it forms a succession of small leaves 

 (Fig. 384, B) : of these about the eighth appears above ground, the rest serving 

 to protect the bud. A rudimentary fertile spike may appear on some of 

 these scale leaves. From this point onwards the development is as in the 

 adult plant. Comparing this development with that in B. virgmiann7ii the 

 relative position of the parts is essentially the same: the chief differences 

 are in their proportion. The root and foot are larger, and the axis later in 

 definition in B.Lujiaria: also in B. virghiianum the first leaf is itself expanded 

 above ground, while the same difficulty exists in defining whether the root 

 is epibasal or hypobasal in origin. It may be held that what is seen in 

 B. Limaria and virginianinn is a later and derivative state compared with 

 B. obliqujim, resulting from the elimination of the suspensor, which liberates 

 the embryo from the endoscopic orientation, and makes the awkward 

 curvature unnecessary (Jeffrey, Proc. Cimad. Inst. 1898). 



Helminthostachys shares with the '^ternatum" group the more primitive 

 endoscopic embryogeny with a suspensor. Lang {Ann. of Bot. 19 14, p. 21) has 

 shown how on the roughly cylindrical prothallus the axis of the archegonium 

 is oblique or even horizontal (Fig. 378). Consequently a like curvature to 

 that in B. obliqnuni will be necessary; it is seen to occur, and the relation 

 of the parts in the sporeling is closely similar also. At first the axis of the 

 young sporeling thus produced is upright (Fig. 385, rt:,^). It is only as the 

 plant passes to the adult state that its creeping habit is assumed. This is 

 the only member of the family that shows this prone position, and it may 

 be held to be secondarily acquired in relation to its heavy foliage. 



In Ophioglossnm the first division of the zygote is transverse to the axis 

 of the archegonium, and as in B. Luitaria there is no suspensor, and the 

 orientation is exoscopic. The first wall separates the epibasal from the 

 hypobasal hemisphere; but reference of the parts to any definite relation 

 with the initial cleavages is made specially uncertain by the fact that the 

 embryo attains considerable size before any differentiation occurs (Fig 386). 



