Heredity and its Bearings on Atavism. 148 
number of cells, it is conceivable that one of these may 
sradually lose its special character and become an initial 
cell, which will give rise to a new organism, or what is even 
more likely, that it will never acquire its special function. 
A process like that just suggested appears to occur in 
the Hydrozoa, in which a single ectoderm cell may give rise 
to a bud, or develop into sexual cells; but how the loss of 
specialisation in an ectoderm cell is brought about, cannot 
be stated definitely. It would appear to depend on the 
respective cell benefiting by the labour of the other cells, 
becoming, in fact, parasitic. As there are a number 
of cells in the ectoderm, the mesogloea, and the endoderm 
elaborating nourishment collectively, we can readily suppose 
a cell to feed on this elaborated material, if either its 
organisation be defective in the organs required for its special 
functions, or if its special function remain dormant, as the 
neighbouring cells are capable of satisfying all the require- 
ments of the individual. 
This occasional non-specialisation of cells amongst low 
forms of life, ¢.g.,in the Hydra, has developed amongst higher 
beings along definite lines, and the “non-functional” cells 
are aggregated into special organs, and we are wont to dis- 
tincuish between the “somatic” cells, which have undergone 
specialisation for maintaining the individual, and the 
“generative” or “sexual” cells, which bring about the 
perpetuation of the species. 
From my standpoint it is evident that the non-functional 
cells, which become the sexual cells, will depend for their 
existence on those food materials which are being elaborated 
by the differentiated, functional, or somatic cells, and 
therefore that sexual cells are influenced by the soma of the 
individual in which they live. 
Any change in the functions of the various organs con- 
cerned in the elaboration of food-materials must lead to an 
altered nutrition of the whole individual, the sex cells 
included; and provided the change is not sufficiently great 
to kill the soma of the individual, and provided the sex cells 
are not fully matured (ze., that they are still young enough 
to be affected by the change in nutrition), it is evident that 
