138 
readily assumed, on the one hand, that the proposals of the 
advocates of a bi-metallic standard are opposed to what may 
be spoken of as the fundamental laws of economic science as 
taught by Adam Smith; while, on the other, the conditions 
which should be fulfilled by a theoretically perfect standard, 
itself subject to those laws and operating in harmony with 
them, have not been presented, the case having been made to 
rest mainly upon considerations of expediency. 
The primary purpose of a standard of value is clearly to 
express readily the mutual relations of all exchangeable 
commodities in common terms. Now, if all exchanges were 
direct and simultaneous, and we could conceive the standard 
as a thing apart, not itself exchanged, a variation in the 
value of the standard would not affect the exchanges. We 
can express the ratios of the temperatures of different bodies 
by the scale of Fahrenheit, Keaumur, or Celsius, and which- 
ever scale we use the relations between the temperatures 
remain the same. If it were bargained that cotton and 
woollen cloths should be directly exchanged in the ratio of 
two yards of the former to one of the latter, the exchange 
would be in no way affected were the standard yard to be 
elongated from 36 to 40 inches or contracted to 30 inches. 
Or, again, if a skilled labourer agreed to exchange his labour 
in the proportion of eight hours against the labour of three 
unskilled labourers each working eight hours, a variation in 
the length of the standard pendulum would not affect the 
bargain, provided that all worked simultaneously. With a 
standard fulfilling the function of a measure even in this 
simple sense, however, the problem might be complicated in 
two ways. In the first place, if the exchanges were not 
simultaneous, a variation in the standard would disturb the 
