82 
Transactions Texas Academy of Science. — 1906. 
todes, Kotatoria and certain smaller taxonomic groups. Yet we will 
find that it is present in some species where it is not commonly recog- 
nized. Its teleological value is to increase the number of individuals 
before the egg cells are sufficiently matured for that purpose, and 
it is in general reproduction by an immature stage. Its relation to 
sexual generation is somewhat obscure, but we may try to approxi- 
mate it in the following way: 
Products of asexual reproduction either may become sexually 
generative or they may remain sterile. The latter condition is much 
the rarer; it is illustrated by various infertile buds of certain 
Cnidaria, such as nematocysts, nutritive polyps, etc., and in Ecto- 
procta by the aberrant vibraculariae and aviculariae. In the major- 
ity of cases metazoan buds or their division products generate by 
egg cells, which implies either that such buds must have contained 
germ cells at the start, or that they originate within themselves germ 
cells de novo, therefore from tissue cells. Now it is possible that 
a germ cell should become somatically specialized to a slight extent, 
and later lose such somatic differentiation; thus in the wall of an 
Echinoderm blastula all the cells bear cilia, yet in the case of a 
female blastula some of these cells must be early generations of 
egg cells, and therefore such egg cells must have temporarily as- 
sumed a somatic character, the development of cilia. But that germ 
cells can arise from somatic cells with any very much higher grade 
of specialization than in the case mentioned, would seem to be quite 
doubtful in the light of our present understanding of the continuity 
of the germ cells and of the nature of somatic differentiation . 8 Ac- 
cordingly, we will conclude that all buds that subsequently generate 
sexually must contain germ cells at the start; for is not the main 
criterion of a germ cell the maintenance of generalization, together 
with origin from a preceding germ cell? Now, buds may arise from 
parts of the organism that contain no gonads, no special centres of 
germ cell formation, which might seem strange and opposed to our 
conclusion that all buds of this kind must contain germ cells. This 
proves, on the contrary rather, that all germ cells of an organism 
need not be placed in specific germ glands, but that many of them 
may be diffusely scattered through the body. Such a wide dispersal 
is well known for the Sponges, and for other animals capable of 
budding we must conclude the presence, of germ cells wherever buds 
are formed. In the embryos of Pishes, Beard has shown the occur- 
rence of such cells even in the medullary tube and the intestine. 
To align the phenomena of a sexual reproduction and regeneration 
8 However, Child has recently shown that in eestodes muscular elements may 
become germ cells. 
