EMBRYOGENESIS IN LEPTOSPOR ANGI ATE FERNS 165 



metabolism, they are probably not identical. A basipetal auxin 

 gradient within the prothallial tissue has been demonstrated (Albaum, 

 1938), and there may be acropetal gradients of carbohydrates and other 

 metabolites having their source in the older region of the prothallus. 

 Thus there may be physiological difTerences on opposite sides of the 

 ovum. These may be small but they may be none the less important. 

 They may account for the fact that the embryonic axis is typically 

 aligned on the axis of the prothallus, the epibasal segment being 

 invariably towards the prothallus apex. A conception of this kind may 

 provide a working hypothesis relating to the cytoplasmic organisation 

 of the ovum and zygote and to the subsequent cleavage pattern and 

 organogenic development. A hypothesis along these lines would also 

 be in general accord with contemporary biochemical conceptions of 

 morphogenesis, i.e. that different specific substances, or different con- 

 centrations of them, may be involved in the inception of the different 

 organs. The need for concepts which will relate embryonic develop- 

 ment and metabolism becomes evident when we contemplate the 

 organisation of the post-octant embryo, in particular the positions of 

 the primary organs relative to the archegonium and the prothallus. 



Our knowledge of the ontogenetic growth pattern is almost entirely 

 based on the 'stages of development' as ascertained in anatomical 

 studies. These 'stages' are distinctive for the species and they occur 

 with great fidelity in the normal development. In the ferns, each octant 

 undergoes several divisions and the embryo becomes a spherical mass 

 of histologically similar meristematic cells. Yet, at this early stage, the 

 main pattern of development, or organisation, has already been 

 established: the shoot apex, the first leaf, the first root, and the foot 

 have been determined, and the orderly development of the sporophyte 

 has begun. In short, it would appear that in the post-octant embryo 

 there is a patternised distribution of metabolites, different morpho- 

 genetic substances, or different concentrations of the same substances, 

 being present in each of the quadrants. As an over-simplification, we 

 might perhaps say that a 'leaf-forming substance' has accumulated in 

 the leaf quadrant, a 'root-forming substance' in the root quadrant, and 

 so on. But however we may describe these events, the ultimate problem 

 is to explain how the characteristic patternised distribution of metabolites 

 is brought about in an initially homogeneous system {see Chapter III). 



In a fern embryo entering on the post-octant phase of development, 

 the basic pattern might be modified in various ways, e.g. by changes in 

 the gene-controlled metabohsm, the cells both of the embryo and of the 

 prothallus being affected, or by environmental factors. Thus mycor- 

 rhizic nutrition of the prothallus may not only modify the nutrition of 

 the embryo in a general way: it may affect one quadrant more than 



