BOROTARTRATE OF POTASH. 
225 
these preparations would most readily pass through the diaphragm of the 
dialyser, and thus pass into the circulation. In the first instance, he had 
looked forward to very important results in this respect, and expected to find 
that whilst some preparations of iron, such as the sulphate and chloride and 
other preparations of a crystalline character, would be most readily diffusible, 
others, such as citrates and preparations of that description, would be found 
to belong to the class of colloid bodies. He very soon, however, found that 
the apparently at any rate uncrystallizable preparations.—the scaling prepa¬ 
rations,—were all of them capable of diffusing very well,—rapidly, and to a 
great extent. The borotartrate of potash being one of those substances that 
had been always looked upon as absolutely amorphous, he had operated upon 
it several times, and had found, in opposition to what he had anticipated, 
that it was capable of diffusion, and diffused very well, in fact quite as 
rapidly and to as great an extent, within a specified period, as substances 
which were looked upon as good diffusers. 
He referred to these results at present only incidentally, as he had not 
carried the investigation as far as he intended. On a subsequent occasion he 
would give the details of the experiments, but for the present he might state 
with reference to the uncrystallizable or scaling preparations of iron, that 
they were all diffusible in a high degree, as also were other similar prepara" 
tions, such as borotartrate of potash. With reference to the preparations of 
iron, he had not found any marked difference in the quantity of iron that 
passed out of the dialyser from solutions of equal strength, whether the iron 
existed as protosulphate or persulphate, as citrate or ammonio-citrate, or ci¬ 
trate of iron and quinine. He had not found in these preparations the 
difference in diffusibility which was anticipated ; but there were, as was w r ell 
known, other preparations that were not diffusible. Thus, for instance, per- 
cliloride of iron would diffuse readily, but if it w r ere made to dissolve as 
much moist hydrated peroxide of iron as it was capable of taking up, so as 
to convert it into a highly basic salt, on putting this into a dialyser, none of 
the iron would pass through. There were other preparations which resem¬ 
bled this in regard to diffusibility, and he had been anxious to indicate classes 
of preparations that were and also that were not diffusible, so that it should 
be known, or might be inferred, to what extent preparations used in medicine 
M ould be absorbed by diffusion or retained in the alimentary canal. All this, 
however, had yet to be much elaborated, but in connection with it he had 
one further remark to make, which M r as that, with reference to the molecular 
condition of solid bodies, there appeared, in addition to the crystalline state, 
to be at least two amorphous states, the vitreous and the granular, from one 
to the other of M r hich a body may pass with marked alteration of property, 
such as solubility, and that there was probably another or fourth state, the 
colloidal, the transition to vhich involved another change of property, 
namely, that of diffusibility. Although the colloidal state had been found 
to exist in the highest degree among amorphous bodies, it would perhaps be 
too much to say that the same molecular condition necessarily represented 
both the amorphous and colloidal states. He was not aware that any very 
decided opinion had been formed as to the cause of the colloidal state of 
bodies. Possibly it might depend upon some such molecular condition as 
Professor Tyndall had assumed to influence the diathermacy of gases, the 
more complex molecules appearing to absorb heat to a greater extent than 
the less complex ones.* But m ith reference to this part of the subject, a far 
* Ml-. Graham, in his paper “ On Liquid Diffusion,” says,—“The equivalent of a colloid 
appears to be always high, although the ratio between the elements of the substance may be 
simple. Gumrnic acid, for instance, may be represented by C 12 H n O u , but judging from the 
VOL. IX. Q 
