22 
32 
NATURE 
[Marcu 14, 1912 
those of igneous origin, many important rock- 
forming minerals, like the garnets, are not among 
those figured. 
that it is not intended to take the place of an 
ordinary text-book, but to supply the information 
which is not given in such works, being left for 
the laboratory-demonstrator to supply—and this | 
would seem to be the limit of the book’s useful- 
ness. 
Famous Chenusts. By E. Roberts. Pp. 247. 
(London: G. Allen and Co., Ltd., 1911.) Price 
2s. 6d. 
It seems a little difficult to understand the object 
of writing a series of disconnected short bio- 
graphies of distinguished chemists now that we 
have several readable histories of chemistry of 
moderate compass in which biography is woven 
into connected narrative. At the same time it 
must be stated, after conscientiously reading the 
book from cover to cover, that these biographical 
epitomes are well done. The chief contributions 
of each master are clearly indicated, and the human 
touches on the whole artistically added. Boyle 
hardly has his due proportion, and Berthelot and 
van’t Hoff are not included at all. 
Perhaps the chief thing to be said for this com- 
pact gallery of chemists is that it may help to 
stimulate an interest in history, and lead the reader 
to a more thorough study of the life-work of the 
great men who have made chemistry what it is. 
This is an educational and liberalising side of 
chemical study which in the past has been much 
neglected. It is a pure convention, and a mis- 
chievous one, that isolates the study of natural 
philosophy itself from the study of its history. 
AS: 
Earth and her Children. By Herbert M. Livens. 
Pp. 248. (London: T. Fisher Unwin, 1912.) 
Pricesss.emet: 
Mr. Livens practises the art of teaching nature 
study by means of pleasant little stories in which 
plants and animals speak autobiographically. The 
nature knowledge imparted in the twenty-four 
chapters into which the book is divided is much 
diluted by the conversational matter introduced, 
but the stories will please many children, and may 
lead a ow to observe nature > for themselve es. 
LETTERS RO” THE EDITOR: 
[The Editor does not hold himself responsible for 
opinions expressed by his correspondents. Neither 
can he undertake to return, or to correspond with 
the writers of, rejected manuscripts intended for 
this or any other part of Nature. No notice is 
taken of anonymous communications. ] 
Osmotic and Liquid Membranes. 
In an interesting letter in Nature of February 22 
Lord Berkeley has considered the question. of a 
possible osmotic cyc® in which a liquid such as ether 
is placed at the same time in connection through 
membranes with an aqucous solution of sugar and 
with water, while the latter is also in similar connec- 
tion with the solution. 
The water is kept at zero pressure, while the ether 
and the solution are at the osmotic pressure of the 
latter. 
NO. 2211, VOL. 89] 
The preface of the book suggests | 
As a consequence of a view of osmotic membranes 
which I have lately ventured to put forward in a 
paper in the Proc. Roy. Soc., he quite correctly 
deduces that equilibrium should exist for different 
strengths of the sugar solution provided the pressure 
is its osmotic pressure, and in addition that the ether 
should hold the same amount of water in solution at 
the different pressures. Or, to quote his words, ‘‘ the 
same solution of water in ether has two different 
osmotic pressures.” This he regards as impossible. 
Perhaps the best way to consider this question is to 
examine first the simple case of water placed on either 
side of a membrane permeable to water, but not to 
ether, and to suppose ether gradually added to one 
side. The osmotic pressure will rise, as shown by 
AB, to a maximum value at about 6 per cent. of 
ether. No solution of greater strength than this is 
possible until we reach 98'95 per cent. The osmotic 
pressure is now again zero, but rapidly rises, as shown 
PRESSURE. 
to) 6% 100 % ETHER IN 
WATER 
along DE, approaching the pressure axis P asym- 
ptotically; so that practically without any concentra- 
tion change we can have any osmotic pressure we 
please. 
The dotted curve has been added to show the normal 
character of the curve for substances which dissolve 
in all proportions, and probably if the pressure on 
both sides of our membrane were raised to a suffici- 
ently high value the two portions of the ether curve 
would unite somewhat after the fashion of the Van 
der Waals’s isothermals, and ultimately form the 
normal case, because experiments (not yet published): 
show that the portion DE moves to the left under 
pressure. 
If the above experiment be supposed repeated, but 
with sugar in solution on the other side of the mem- 
brane to that to which the ether is added, the first 
possible strong solution of ether is greater than 
98°95 per cent., or DE is moved to the right. As in 
the case of water, so in the case of sugar solution, 
DE can be moved to the left by raising the pressure 
on both sides of the membrane. For a 60 per cent. 
solution of sugar the pressure to bring it back to its 
old position is 80 atmospheres. 
Returning now to Lord Berkeley’s cycle, we are in 
a position to see how equilibrium can exist without 
an appreciable change in strength of the ether solu- 
tion when different strengths of the sugar solution 
are used. Referring to his figure, equilibrium exists 
across membrane bd at the osmotic pressure of the 
solution when the concentration of the ether is 
98°95 per cent., and practically the same concentration 
is required for equilibrium at the membrane bc, no 
matter what the pressure may be. 
Perhaps I may add that the view of osmotic mem- 
branes which I wished in my paper to emphasise is 
that a substance which acts as ether does in the con- 
dition at DE can, so long as the mechanical strength 
