THE STRUCTURE OF PROTOPLASM 
trum with their plastids, mouth-piece, blepharoplasts, cilia, etc., are visibly 
highly polarized structures, and it will be of great interest to determine 
whether this visible polarity coincides with the major axes of the adult 
cell of Pediastrum and the apparent affinities which determine the orien- 
tation of each cell with reference to its fellows and to the form of the colony 
as a whole. This problem is by no means insoluble with the modern culture 
methods available for keeping such algae in large numbers under continuous 
observation. 
That in Pediastrum the swarmspores show polarity in at least two axes 
and that this polarity determines their mutual interactions and final position 
in the colony is shown conclusively, as I have pointed out elsewhere, by the 
symmetrical organization of the colony and the method of its formation 
from a group of freely swimming cell units. Polarity exists in probably 
all the tissue cells of metaphytes and coenobic plants and in many proto- 
phytes. It is apparently independent of the uni- or multinucleated 
condition of the cell, which shows that it is in some cases at least a more 
generalized characteristic of the cell as a whole rather than a mere expression 
of the space relations of the nucleus and cytoplasm in a diagram like that of 
Rabl. Bilateral and radial symmetry are shown also in the cells of desmids, 
diatoms, and other protophytes, these latter showing in Rhumbler's term 
the anomogeneity of the protoplasm as a liquid. It is sufficiently clear, 
it seems to me, that in the presence of polarity and the various symmetry 
relations we have a fundamental distinction between cell organization 
and that of polyphase colloidal systems as they are commonly produced 
in vitro. 
Furthermore, the outstanding fact not sufficiently recognized in the 
theory of colloids as at present developed is that we may have at least more 
or less differentiated colloidal systems within a colloidal system. The cell 
must at least be conceived as a complex of such colloidal systems, some 
possibly simple two-phase systems as perhaps a vacuolated nucleole, some 
polyphase as the nucleus taken as a whole within the cytoplasm. In such 
cases the interior system as a whole will show a tendency to form convex 
surfaces toward the enveloping system or systems. Perhaps the nearest 
approach to an experimental demonstration of such an organization is 
found in Hardy's development of the evidence for secondary interior phases. 
If a mixture of gelatine and water is cooled to a certain temperature, we 
obtain a watery interior phase in a continuum of denser gelatin ; if now the 
cooling is continued small particles appear in the watery disperse phase 
which may make chains which anastomose to form reticula constituting a 
secondary interior gelatine-rich phase. In this case we should still have no 
polarity in the mass as a whole. The interior phases would be repeated 
equally in all directions throughout the whole system. None the less the 
experiments as described are highly suggestive as to the physical relations 
under which the internal structures of such a complex system as the nucleus 
