SOCIAL AND ECONOMIC THEORY. '@ 1/21 
ordinary increases in output (due to increasing popula- 
tion) at a constant price level. Similarly the drop in 
prices from 1880 to 1895 kept profits {vans 
ably below what would have resulted from the actual 
output at a constant price level, and in itself was 
instrumental in depressing that output. ~~ ~~ 
‘The “turnover” of foreign trade had become a 
relatively less important part of the whole trade of the 
country during the previous thirty years. 
‘ The annual trend of increase in trade freed from 
all fluctuation has to a great extent been made up of 
the larger output of existing businesses increasing 
continually in size, and to a relatively smaller extent 
of the output added by new businesses.’ 
The investigation had, perhaps, some value in indicating 
the kind of pitfalls that have to be avoided in dealing 
with this class of statistics, and the care that must be 
taken to make them ‘chemically clean’, so to speak, 
before the investigation. Our own Oxford Professor 
Edgeworth remarked on this paper that there was ever 
in the class known as the non-statistical reader a 
‘natural and not altogether unhealthy suspicion of 
any technical method, any organon which seemed 
intended to supersede the use of common sense. It 
was Locke, or some one who wrote, like Locke, against 
the Aristotelian syllogism, who protested that the 
Almighty had not dealt so very sparingly with the 
noblest of his creatures as to make them only bipeds, 
leaving it to Aristotle to make them rational. A 
similar prejudice on the part of common sense against 
correlation and other mathematical instruments is to 
be apprehended.’ 
VI. Examples. (b) Time series without the 
‘ growth’ element—Original inquiry. 
We can now look at an instance of a personal, as against 
a collective, inquiry. 
Mr. Edgar Smith recently made an investigation, the 
results of which are given in his book Common Stocks as