TRANSACTIONS OF SECTION RP. 359 
1/100,000 of a gram molecule of solute per litre for his conductivity experiments, 
These solutions were therefore twice as strong as the gold solution with 1 gram 
per cubic metre, and twenty times as strong as the more dilute solution. This 
fact must be my excuse for placing before you the results of a few simple calcula~ 
tions as to the molecular distribution in these solutions, which have certainly 
given me an entirely new view of what constitutes a really dilute solution from 
the molecular point of view. 
In estimating the number of molecules in a given volume of solution the 
method adopted is to divide the space into minute cubical cells, each of which can 
exactly contain a sphere of the diameter of the molecule. In this way a form of 
piling for the molecules is assumed which, though not the closest possible, may 
quite probably represent the piling of water molecules. Taking the molecular 
diameter as 0:2 x 10~° millimetres—a figure which is possibly too small for the 
water molecules and too large for the gold—it is found that a cubic millimetre of 
solution contains 125 x 10'8 molecules, or 125 quadrillions. The head of an ordi- 
nary pin, if it were spherical, would have a volume of about 1 cubic millimetre. 
If these water molecules could be arranged in a single row, each molecule 
just touching its two nearest neighbours, the length of the row would be 
25,000,000 kilometres. A thread of these fairy beads, which contained the 
molecules of one very small drop of a volume of 6 cubic millimetres, would reach 
from the earth to the sun, a distance of about 150,000,000 kilometres, 
In a solution containing 1 grain of gold per ton, or 1 decigram per cubic 
metre, the ratio of gold molecules to water molecules is as 1 : 193,000,000. Each 
cubic millimetre of the solution, therefore, contains 6,500,000,000 gold molecules. 
If these are uniformly distributed throughout the solution each will be about 
400 micro-millimetres, or 1/60,000 of an inch, from its nearest neighbours. This 
is not really very wide spacing, for the point of the finest sewing-needle would 
cover about 1,500 gold molecules. 
If a cubic metre of solution could be spread out in a sheet one molecule in 
thickness it would cover an area of 1,680 square miles, and nowhere in this area 
would it be possible to put down the point of the needle without touching some 
hundreds of gold molecules simultaneously. 
According to Professor Liversidge, sea-water contains on the average about 
one grain of gold per ton. If this is the case, then the above figures for the dilute 
cyanide solution apply with only a slight modification to sea-water. No drop, 
however small it may be, can be removed from the ocean which will not contain 
many millions of gold molecules, and no point of its surface can be touched which 
is not thickly strewn with these. From this molecular point of view we must 
realise that our ships literally float on a gilded ocean! 
From time to time adventurers arise who attempt to launch upon this gilded 
ocean unseaworthy ships freighted with the savings of the trusting investor. In 
order that nothing which has been said here may tempt anyone to contribute to 
the freigh¢ing of these ships, let me hasten to point out that the weakest of the 
cyanide solutions here referred to is richer in gold than sea-water is reported to 
be. The practical conclusion from this comparisen is sufficiently obvious. If the 
eyaniding expert, whose business it is to extract gold from dilute solutions, finds 
that it does not pay to carry this extraction beyond a concentration of 2 or 3 grains 
per ton, even when the solution is already in his hand, and when, therefore, the 
costs of treatment are at their minimum, how can it possibly pay to begin the 
work of extraction on sea-water, a solution of one-half the richness, which would 
have to be impounded and treated by methods which could not fail to be more 
costly in labour and materials than the simple process of zinc-box precipitation ? 
It is generally unsafe to prophesy, but in this case I am rash enough to risk the 
prediction that if ever the gold mines of the Transvaal are shut up it will not be 
owing to the competition of the gold resources of the ocean. 
In these calculations with reference to the dilute cyanide solutions it is 
assumed that the gold molecules are uniformly distributed, that they are practi- 
cally equidistant from each other. There appears to me tu be considerable doubt 
whether we have any right to make this assumption. Leaving out of account for 
