JUNE 14, 1901.] 
vapor-tension of water to only a slight ex- 
tent. 
These facts can be accounted for in two 
ways, either by assuming that the colloids 
in solution have very large molecular 
weights or that colloidal solutions are not 
true solutions at all, but mechanical mix- 
tures, in which the colloid is in a very fine 
state of subdivision. The latter view is 
now generally held for a large number of 
reasons, and for the metals, at least, seems 
to be made very probable by the method 
employed in the preparation of such solu- 
tions. 
The work of Bredig and his pupils has 
to do with colloidal solutions of the metals, 
and the method employed in preparing such 
solutions is very interesting. Two bars of 
the metal in question are thrown into the 
circuit of a suitable electric current, and 
the lower ends of these poles are immersed 
in pure water, as shown in the figure. The 
ends of the bars of metals, g, g, are brought 
sufficiently near that an arc is established 
between them beneath the water. The 
metal is torn off from the bars in such a 
fine state of division that it forms in the 
water a colloidal solution. 
In such solutions the metal is so finely 
dividea that it cannot be seen under the 
most powerful microscope, the solution ap- 
pearing homogeneous under such condi- 
tions. 
By the above method colloidal solutions 
of platinum, palladium, iridium, silver, 
gold and cadmium were prepared. These 
SCIENCE. 
941 
solutions had the properties of colloidal so- 
lutions in general, and it is quite evident 
that such solutions are but mechanical mix- 
tures of the very finely divided metals with 
the solvent, water. 
Such solutions of the metals have some 
quite remarkable properties, and one of 
these has been studied extensively by Bre- 
dig and his pupils. 
It was found that colloidal solutions of 
platinum have the property of decomposing 
hydrogen dioxide catalytically. A catalyzer 
is a substance which, in relatively small 
quantity, can effect a large transformation. 
The solution of metallic platinum fulfilled 
this condition, a small amount of the solu- 
tion decomposing a large amount of hydro- 
gen dioxide. Bredig states that a gram, 
atomic weight of platinum in 70,000,000 
litres of water can appreciably accelerate 
the velocity of the decomposition of hydro- 
gen dioxide. 
Further, in order that a substance may 
act catalytically, it must apparently not 
take part in the reaction which it produces, 
and must remain in an unaltered condition 
after the reaction is over. The colloidal 
solution of platinum also fulfilled this con- 
dition. Bredig has shown from a study of 
the velocity of the reaction, by a well- 
known method, that the decomposition of 
hydrogen dioxide by the finely divided 
platinum is a monomolecular reaction ; 
i. €., only one substance—the hydrogen di- 
oxide—takes part in the reaction. The 
metallic platinum, therefore, does not enter 
into the reaction at all, but remains un- 
changed in the solution. 
We now come to the most important 
part of the paper. The author sees an 
analogy between the catalytic action of the col- 
loidal platinum and the action of organic fer- 
ments. They point out that recent work 
has shown that there are many reactions 
which are effected by both unorganized and 
organized ferments, and also by the contact 
