ON VARIATIONS IN THE VALUE OF THE MONETARY STANDARD. 183 



prices, for these are the only prices of which there are records — of which 

 it may ahnost be said there can be records, retail prices being subject to 

 many difficulties, including the degree of ' retailing ' which may exist, 

 to become the basis of records capable of solving the practical problems 

 stated. 



In actual fact students and politicians have been still more limited. 

 There are prices and prices. Practically it is found that only the prices 

 of leading commodities, capable of being dealt with in large wholesale 

 markets, can be made use of, while the assumption must be made, so 

 numerous and varying are quaHties, that certain articles may be taken 

 as types. For instance, in dealing with the values of wheat it is con- 

 venient to take English wheat, as represented by the Gazette average, as 

 a type, and assume that all wheat sold in England varies in price as 

 English wheat does ; but the assumption may not be quite correct, while 

 the Gazette average itself, if we look at the question historically, only 

 goes back a hundred years, and previous to that the historical student 

 must fall back on the records of the value of wheat at a particular place, 

 such as Windsor. Even greater difficulties arise with regard to other 

 articles, while for some purposes the prices of large masses of articles, 

 which are obtained by such means as the division of the aggregate 

 quantities of the imports by aggregate values, are deficient, owing to 

 the changes in the qualities of the constituents of the group from year to 

 year or from period to period. In dealing with the question practically, 

 therefore, those concerned must always have an eye upon the data 

 and consider what is practically obtainable and what use may be made 

 thereof. 



Looking at the subject in this way, then, the most important con- 

 clusion seems to be that for such practical purposes as those above 

 mentioned the methods already followed may be and have, in fact, been 

 approximately successful. With improvements they may be made more 

 useful, although ideal perfection is unobtainable, and even inconceivable. 

 What has to be considered is that the prices of leading commodities are 

 likely to vary on the average as all commodities vary on the average ; 

 that, if such commodities are selected without bias, the result, as regards 

 the selected articles, may be accepted as representative of the average re- 

 sult ; that wholesale and retail prices will vary in much the same way, and 

 that consequently the mean of a number of prices taken almost at random, 

 if there is a sufficient number to get rid of chance errors, may be de- 

 pended on to measure variations in the monetary standard in lien of a 

 standard composed of all articles whatsoever, each receiving its proper 

 weight. Rough approximations only are possible, but according to the 

 logic of statistics the defectiveness of the statistical data, though it has 

 to be recognised, is not an insuperable barrier to the adoption of rough 

 approximations which are valuable. Let us see what has been accom- 

 plished. 



The ' Chronicon Preciosum,' dealing with the chief articles only, as to 

 which records of price happened to remain, and using a plan of means in 

 which each article was considered to be of equal importance, showed 

 approximately that the value of money had changed from century to 

 century; that, measured by the common articles of trade and consump- 

 tion, a sovereign did not go so far in the time of that book as it did 

 several centuries before — that it only commanded a sixth part of the 

 average of leading commodities which it had commanded several centuries 



