Alternation of Generations and Ontogeny. 209 
only that from the sporidium will penetrate the leaf. Obviously all 
these germ-cells contain all the determinants of the plant, for each 
is able to reproduce the whole life-cycle. Their difference in 
behaviour must be due to their different position in the lefrf-cycle,- 
to their having received a different “ tendency,” so that the pro- 
mycelial characters are latent in all but the teleutospore, the 
barberry-leaf-infecting characters latent in all but the sporidia, 
etc. 
That the principle of the relation of one stage in development 
to another must be extended beyond the simple limits of the 
development of parts of a single body-form is clearly shown by the 
metamorphosis of many insects. Here we have two very different 
body-forms, larva and imago, which may pass from one to 
another, without any change in external conditions. 1 That the 
relationship between the two body-types is of very special nature 
is clear from the fact that in many cases most of the larval 
structures are broken down during the pupa-state and entirely 
new ones formed for the imago ; and further one type of body 
may have organs not possessed by the other. 2 If a change of the 
type of development can take place, independently of a change in 
external conditions, in the groups of cells in insects, there would 
seem to be no difficulty in conceiving it as taking place in single 
cells, such as the spores and eggs of a fern. 
The metamorphosis of insects is, of course, only an extreme 
case of what is so common in animals, the development of larval 
stages very unlike the mature form. It is the usual absence of such 
larval stages in plants which makes the dimorphism of the 
Archegoniatae appear so striking. 
There is another line of argument which points to the fact that 
cells of the same organism freed from direct connexion with the 
somatic cells, may yet develop very differently under very similar 
or identical surroundings. The microspores and megaspores of 
Selaginella, for example, are both free cells developing under almost 
identical conditions of protection and nourishment, and yet how 
different are the results attained by the time they leave the 
1 Everyone who has reared silk-worms knows that they will pass 
from an egg to caterpillar, pupa and imago (moth) under the 
fairly constant conditions of an ordinary dwelling room. 
2 Within the single egg-cell of these insects there are the cha¬ 
racters of two “ specific body-forms,” one of which develops 
after the other similarly, and the egg and cell may carry the 
characters of gametophyte and sporopyte, one being latent in 
each. 
■1,'i 4 , 
