SUMMARY 395 
It has been seen in the Lycopods that the root is constant neither in the 
time nor in the place of its appearance: it has been also seen that it 
originates in the epibasal region in Lycopodium and Jsoetes, but in Selaginella 
in the hypobasal. It need therefore be no cause for surprise, but rather of 
increased interest that the point of origin of the first root should fluctuate 
within the genus gwzsetum, Its indefinite position in different cases 
stamps upon it with special clearness the character of an accessory to 
the shoot itself, which its late appearance in certain Lycopods seems 
further to confirm. The whole embryo thus consists of a spindle-like axis 
with continued apical growth; its base is like that of /sve/es without any 
suspensor. The leaves and roots appear as appendages upon this spindle- 
like axis. 
Naturally, the embryogeny of the fossil Equisetales is not accessible 
for comparison. 
From the account of the Equisetales given in the above pages, it is 
possible to form some idea of a primitive general type for the phylum. 
They were probably, from the first, organisms with a prominent axis, 
while the leaves, of moderate size, were arranged in whorls, with 
elongated internodes between them. The root was an accessory addition 
to the shoot. Spore-production, which is so important an event in the 
antithetic alternation, does not figure in the early stages of life in 
any known Equisetal type, but appears only late in the individual life. 
There is little direct evidence among the Equisetales of any deferring 
of spore-production, by abortion of sporangia or of sporangiophores, com- 
parable with that which is so clearly indicated in the Lycopodiales. But 
comparative evidence shows that in the Equisetales spore-production is not 
restricted to branches of any definite rank, and transfers of the reproduc- 
tive function from branches of one rank to those of a higher rank may 
occur in nature, and are illustrated in various living species of Egutsetum. 
This, coupled with the fact that there is essential structural similarity 
between axes of all ranks in these plants, makes it seem probable that 
axes of lower rank, and finally even the primary axis itself, may have 
been fertile in a primitive Equisetoid type: that a deferring of spore- 
production by transfer from axes of lower to those of higher order 
occurred, and that thus the initial vegetative system was greatly extended. 
In the Calamarians a secondary development of tissues in the axis accom- 
panies the enlargement of the vegetative system, which thus attained 
dendroid characters, now only faintly reflected in the smaller living forms. 
It would appear from the elongated form of the lax cone in such types 
as Calamostachys, and especially from the usual intermixture of bract-leaves 
and sporangiophores in them, that among early Equisetal types a condition 
existed not unlike that of the undifferentiated Lycopod shoot of the Se/ago 
type: that is, a general-purposes shoot, in which the office of spore- 
production was not strictly differentiated from the function of nutrition, 
