90 
Transactions Texas Academy of Science. — 1906. 
in that certain parts of a larva do not become incorporated into the 
later body, we should regard that larva as a first individual that by 
asexual means produces a later one. Different life cycles exhibit 
a great range of degrees as to how large the bud is in proportion to 
the larva, but no sharp line can be drawn between the various de- 
grees. Accordingly, there is no substantial difference whether the 
later individual arises as a small or as a large part of the earlier 
one. This shows how general a process metagenesis really is, how 
far from being the exception, and that life cycles are more compli- 
cated than generally thought. 
By larva we mean a motile immature stage of habitude different 
from the adult, and two kinds of larvae are to be distinguished. There 
is the more primitive kind found in indirect continuous development, 
where every part of the larva passes over into a later structure, 
and where asexual reproduction does not occur. And there is the 
secondary, metagenetic, larva of discontinuous life cycles, where 
certain organs of the larva take no part in forming adult structures, 
and where, consequently, asexual generation occurs. This together 
with other considerations suggests that asexual generation is always 
associated with immaturity of the organism, because it precedes the 
stage that generates by an egg cell: it is a secondary process inter- 
polated into the life cycle. This is in harmony with the contention of 
Claus 24 , further elaborated by Brooks , 25 that in the development of 
metagenetic medusae the polyp is to be considered a larva that has 
secondarily become sedentary. In any metagenetic cycle the begin- 
ning is an egg cell, the end is that individual that again produces by 
eggs. 
Metagenesis has originated from indirect continuous development 
in this way. The latter, it will be recalled, may be conceived of as 
a progression from the egg to the adult, in not a straight but an angu- 
lar line, the angle representing a larval stage. Now if such a larva 
ceases to pass wholly over into the adult, therefore buds off the lat- 
ter, the development becomes discontinuous at that point, another 
individual is added, there is a break in the development and metagen- 
esis is established. The known embryogenies affords us a practically 
complete series of intergradations from the indirect continuous de- 
velopment to the metagenetic discontinuous. 
Heterogenesis is the second kind of discontinuous life cycle. By 
it we mean a cycle composed of two or more successive individuals, 
some of which reproduce by unfertilized eggs, parthenogenetically, 
others by fertilized ones. There is alternation of the two forms of 
24 Lehrbuch der Zoologie, 1885, 5te Aufl., Leipzig. 
25 The Life-History of the Hydromedusae, 1886, Mem. Boston Soc. Nat. Hist., 3. 
