L] • ONTOGENETIC ADJUSTMENTS 289 



wards in the course of the individual development, while the lower indusium 

 is inconstant (Fig. 218). A very beautiful intermediate state is seen in 

 Histiopteris incisa (Fig. 219): here the orginal relation of the receptacle to 

 the margin is variable, but with a strong initial bias towards the lower sur- 

 face; and the lower indusium is abortive. In Pteris itself the lower indusium 

 is again absent, but here the flattened receptacle is clearl)- intra-marginal, 

 the first sporangia springing from the lower leaf-surface (Fig. 220). Thus the 

 "phyletic slide" is again complete; and it is carried out in a sequence 

 systematically independent of that of the Dennstaedtioid Series. These two 

 series are held as demonstrating real progressions, each within a distinct 

 natural circle of affinity. Traditional systematic grouping together of the 

 Ferns quoted in either sequence, as also their rich synonymy, show the near 

 kinship within either of these parallel sequences. Further, the species quoted 

 exhibit parallel progressive changes in their vegetative characters also, while 

 in either case there is a passage of their sori from the gradate to the mixed 

 state. These facts indicate that the sequences are natural and progressive, 

 not only in one but in many distinct characters. The identity of the sorus is 

 assumed throughout. The change in its position is held to be adaptive, and 

 that change has become hereditary. It involves readily intelligible steps in 

 homoplastic amendment. It is believed to illustrate the inheritance of pro- 

 gressive structural characters biologically advantageous, in plants akin, 

 developing without any restriction of time-limit, and carried out in a 

 plurality of series phyletically distinct. Only a few outstanding examples 

 have been quoted here: but there is evidence of a progressive slide of the 

 sorus from the margin to the surface of the blade having taken place in yet 

 other distinct phyla. The general conclusion zu/iich follows from this compara- 

 tive discussion is that a cJiange, first seen as an ontogenetic adjustment in more 

 primitive Ferns, is liable to become an accentuated and inherited feature in 

 more advanced members of the same natural affinity. In other zvords, that 

 ontogenetic adjustments — so-called fluctuating variations — continued without 

 any time-limit, may become hereditary. 



The incidence of a imiting factor may also be effective in establishing 

 heritable characters such as have been widely used in classification, as the 

 next instance will show. The initiative to develop, resulting in increasing 

 size of the individual, brings it up against such limiting factors. In Ferns this 

 becomes obvious in relation to the increasing size of the conducting tracts 

 (Chapter x). Since under suitable surface control these tracts supply material 

 to the tissues they traverse, and since all nutritive supply from them is a 

 unction of surface, the question of proportion of surface to bulk in the con- 

 ducting tracts is a critical one. As the size increases, supposing the originally 

 simple form of the conical stele to be maintained, the bulk of it will vary as 

 the cube of the linear dimensions, while the surface varies only as the square. 



BUI 19 



