RESTITUTION-BODIES AND FREE TISSUE-CULTURE IN SYCON 807 
in their turn are produced in a totally abnormal, artificial way. 
The form produced, however, as also in the case of the choano- 
cyte spheres, though typical, is not in the least like anything 
occurring in the normal life of the species. Here, typical 
form and form-changes of an organic type are seen in artificial 
ageregates. Once more we see a series of organic forms very 
clearly as an equilibrium between external environment and 
inner constitution. Here, however, the inner constitution is 
simple, the changes are not running in grooves of heredity. 
It is probable that many adult forms of simple organisms, 
as well as simple developmental forms, are in this way deter- 
mined almost entirely by a direct relation of not highly- 
differentiated tissues with environment. The blastula, for 
instance, may or may not represent an ancestral adult form. 
It certainly is a primitive ancestral developmental form ; but 
it is this not for any adaptive or eventually ‘ organismal ’ 
reason, but because it is the simplest way in which 
a number of undifferentiated cells can arrange themselves in 
a fluid medium. See also Child (2) for an account of the way 
in which adult form may be largely determined thus in flat- 
worms, 
8. RESTITUTION-BODIES AND T'ISSUE-CULTURE. 
It is obvious that the spheres produced by isolation of sheets 
of collar-cells are ‘ tissue-cultures ’ in that they consist of one 
sort of cells only. Their history in ordinary sea-water is 
a history of gradual starvation, followed by involution, since 
the fluid does not contain sufficient nutriment. Numerous 
experiments were tried with a view to finding a suitable 
nutrient medium, but so far without success. 
(1) Pure culture of the Diatom Nitzchia, so successfully 
employed by Allen and others for feeding Echinoid plutei, 
were obtained and mixed with water containing preponderat- 
ingly choanocyte restitution-bodies. In a few cases, diatoms 
were seen in the collars of collar-cells, or partly embedded in 
the cell-body ; but they were apparently too long for con- 
venient ingestion. 
NO. 258 Y 
