298 
AMERICAN JOURNAL OF BOTANY 
[Vol. 8 
variations in permeability to effects of electrolytes and metabolic products on 
interfacial soap films.^ 
Referring to the picture of the colloidal mechanism described above, 
it is obvious that, while the proteins or albumins and the pentosans must 
form separate strands or plates, the smaller aggregates in the more liquid 
portion or phase mingle with each other, and it might also be suggested that 
various unions might take place between molecular groups of pentosans 
with amino-compounds or salts, and that actual salts might be formed 
by the albumins when the colloid is immersed in hydroxides or salts such 
as those of potassium, sodium, or magnesium. Here then we have a crude 
statement of a theory of the colloidal condition of protoplasm upon 
which extensive experimentation as to the action of bases, salts of common 
metals, acids, amino-compounds, and vitamines has been carried on with 
the acquisition of results of positive value. These reactions are our real 
aim, and after we have crossed our bridge of hypothesis to the solid ground 
of facts, the fate of the bridge which may have served us well is of minor 
importance. 
The arrangement of living matter inferred has much to support it. 
It is one in which the separate components of the colloidal machine each 
present a separate and individualized capacity for hydration changes under 
any set of conditions or at any given temperature for example, and when 
immersed in water would move with differentiated speed to saturation or 
satisfaction. 
If the water in which the colloid is immersed holds substances in solution 
the ions of which may form combinations with the main components men- 
tioned above, their capacity for holding water in combination may be 
altered, and combinations or splittings in the metabolism of substances in 
protoplasmic colloids may also exert such action. Thus, a dissociation 
resulting in the freeing of hydrogen ions tends to increase the hydration 
capacity of the protein strands or aggregates of the mass might lessen 
that of the pentosans and soaps. Slightly acid amino-compounds as gly- 
cocoll would increase the hydration of the pentosans while exerting practi- 
cally no effect on the albumin or albumin derivatives. It is evident without 
further elaboration that in the albumin-pentosan-soap machine we have a 
mechanism capable of an almost endlessly diversified action in swelling 
and growth. 
With so much prelude we may now advantageously turn to a considera- 
tion of recently acquired results obtained by testing the action of salts, 
balanced solutions, amino-compounds, and vitamines on such colloidal 
mechanisms and on biocolloids and cell masses, living and dead. 
In an earlier paper^ I had advanced the idea that the common metals 
2 Clowes, G. H. A. On the action exerted by antagonistic electrolytes on the electrical 
resistance and permeability of emulsoid membranes. Proc. Soc. Exper. Biol, and Med. 
15: 108. 1918. 
^ MacDougal, D. T. Growth in organisms. Science, 49: 599-605. 1919. (Seep. 11 
of reprint). 
