474 OPHIOGLOSSALES 



only be held as the representatives, secondarily reduced, of leaves primitively 

 expanded above ground. 



Another feature for comparison is the balance between the root and 

 the shoot. Owing to the saprophytic mycorhizal habit of the prothallus 

 and in some of the species even of the sporophyte itself there is no 

 immediate need for leaf-expansion, though an effective root-system is 

 wanted, especially where it is itself mycorhizic. This finds its realisation in 

 the embryogeny ; for the root-development in the Ophioglossaceae is liable 

 to be hurried forward, and the development of the shoot to be postponed. 

 That is seen in O. vitlgatiim, where the first root may already have freely 

 emerged from the prothallus even before the shoot is clearly initiated. 

 B. Lunaria also shows the first root as predominant, and the shoot relatively 

 backward in development, with its succession of scale-leaves. Both these 

 familiar plants are thus relatively specialised types of their respective genera. 

 But the case of O. pendulum shows a still more extreme type; and it 

 seems not improbable that the precocious development of the root has 

 completely upset the balance of parts in the embryo, with the result that 

 the primary axis and cotyledon are difficult to locate, or may be even 

 entirely arrested. Comparing the embryos of the family as a whole, 

 it would seem probable that the primitive prothallus was above ground, 

 and that in the original state of the seedling even the first leaf was an 

 effective assimilating leaf, while those with one or more ineffective scales 

 show a more advanced adjustment to their underground habit. The 

 deferring of the period of functional activity of the shoot carries its 

 reflection back to the early steps of the embryogeny ; the relatively 

 late appearance of the axis with its appendages is thus explained, as weL 

 as the apparently precocious development of the root. The differences in 

 these respects shown by the various representatives of the family indicate 

 their unusual capacity for adjustment of such details. It is through con- 

 siderations of this nature that we may bring these embryos into relation 

 with those of other Pteridophytes where the embryo shows differentiation 

 at an earlier .stage. 



The late differentiation of the parts of the embryo in the Ophioglossaceae 

 brings with it a difficulty in their exact location relatively to the primary 

 segmentations of the zygote. There is no doubt that in the types under 

 consideration (excluding the type of B. oblhjituin}, the apex of the axis 

 arises in them all from the epibasal hemisphere, and allowing for distortions 

 due to unequal growth, it appears to be coincident with, or in near 

 proximity to, the intersection of the primary octant-walls. Thus as regards 

 the initial polarity the Ophioglossaceae resemble other types of Pterido- 

 phytes. The cotyledon appears in close relation to the apex of the axis, 

 both in time and space, and it usually lies between the apex of the axis and 

 the first root: but it will be remembered that in Isoetes, which offers some 

 other points of analogy, the root is on the opposite side of the axis to the 

 cotyledon. As to the exact point of origin of the first root there is some 



