Anatomy of the Osmundaceae % 531 
elements necessary for its physiologically effective construction. Below this 
limit the preceding stages will continue to exist undisturbed. If now a still 
later modification should arise requiring no more elements for its expression 
than the one immediately preceding it, it will eventually catch the latter up 
at its lower limit of penetration. There the two modifications will be 
superposed and may coexist, or, if their coexistence at the same level in the 
stem is impossible, the later modification will replace the earlier and 
eliminate it from the ontogeny altogether. 
A series of modifications will appear in orderly succession in the 
sporeling stem only when the lapse of time has not yet been great enough 
for the successive stages to have caught each other up, or when each stage 
requires a successively increasing number of elements for its effective 
expression. Otherwise, in course of time the ontogeny will become tele¬ 
scoped and the sporeling will begin with the latest modification. In the 
case of a new modification that has caught up and overlapped its pre¬ 
decessor, the structure of the sporeling stele in this region will depend upon 
the effect of the superposition of the new modification upon the older one. 
This will vary according to the nature of the two modifications. They may 
be able to coexist without serious interference with each other at the same 
level in the stem, or the later one may so affect the earlier that it is no 
longer recognizable as such. Again, it is conceivable that a new modifica¬ 
tion should require even fewer elements for its expression than its pre¬ 
decessor. In this case it might pass by its predecessor and appear at 
a lower level in the sporeling than the really more primitive stage that 
phylogenetically preceded it, provided the latter is not deleted by the over¬ 
lapping. All these and other factors still more obscure related to the 
special conditions of existence of the sporeling itself must, I believe, be 
taken into account when considering the phylogeny of the Osmundaceae in 
the light of the vascular structure of the sporeling. On these grounds the 
early appearance of the xylem-sheath pockets in the sporeling of Osmunda 
regalis is in no way incompatible with the true intrastelar nature of the pith. 
It seems clear that they are two entirely independent phenomena. 
From this point of view also the criticism of the reduction theory put 
forward by Dr. Kidston and myself on the grounds of lack of sporeling 
evidence is not altogether valid. The term £ reduction ’ is taken to include 
simplification in structure consequent upon decrease in size, and it now 
seems to me that a sporeling could only be expected to provide evidence 
of such a reduction upon the supposition that it had an effect upon every 
stage in the plant’s life. It is conceivable, however, that the reduction 
should only affect the later more complex stages in the development of 
a plant in such a manner that they are no longer formed. Such a reduction 
would be without influence upon the sporeling, which would still repeat the 
ascending series of changes up to the latest stage of reduction. 
