130 ANNUAL OF SCIENTIFIC DISCOVERY. 



the hands of those who have no bias to do it otherwise than fairly, no 

 interest in. the result. 



Nor can I, while speaking of statistics, avoid referring to the Statistical 

 Congress which took place at Brussels about this time last year ; which 

 had mainly for its object, to produce uniformity among different nations, 

 in the selection of the facts which they should record, and in the manner 

 of recording them ; without which, indeed, no satisfactory comparisons 

 can be established, no results can safely be deduced. To bring about such 

 a uniformity absolutely, is, I am afraid, hopeless ; inasmuch as the grounds 

 of difference are, in many cases, so deeply embedded in the laws, the insti- 

 tutions, and the habits of the different countries, that no hammer of the 

 statist is likely to remove them. 



To understand, however, the points of difference, even if they are not 

 removed, is, in itself, one great step towards the object. It at least pre- 

 vents false conclusions, if it does not fully provide the means of establish- 

 ing the true ones. It gets rid of sources of error, even if it fails of giving 

 the full means of ascertaining truth. Take, for instance, the case of crim- 

 inal statistics. We wish to ascertain the comparative prevalence of differ- 

 ent crimes, either at different times or in different countries. For tin's 

 purpose, must we not know under what heads the jurists and statists of 

 the times or countries to be compared array the various offences which 

 are recorded ; with what amounts of penalty they were visited ; and with 

 what rigor, from time to time, the penalties were enforced ? 



That which is called manslaughter in one country, and assassination in 

 another, is called murder in a third. That which, in one country, is pun- 

 ished with death, in another is visited by imprisonment. The bankruptcy 

 which, in one country, is a crime, in another is a civil offence. The juve- 

 nile offences, which in one country are punished by imprisonment, and 

 swell the criminal calendar, in another are treated, as they should in many 

 cases be, only as a subject of compassion and correction, take 110 place in 

 the criminal calendar at all. 



Indeed, it is one of the difficulties which beset a large proportion of 

 these investigations, whether into morals, health, education, or legislation, 

 and which must always distinguish them from those which deal either 

 with matter or denned abstractions, that, in using the same terms, we are 

 often uncertain whether we mean the same thing ; whether, in fact, when 

 we are using the same denominations, the same weights and measures are 

 really employed. Such conferences, however, as those of Brussels, tend 

 much to limit the extent of error. 



With regard to the statistics of agriculture, the main object is, to pro- 

 cure such a knowledge of the facts as shall guide the operations of the 

 consumer and the merchant. I would suggest that they should be taken 

 and published at two periods of the year, once in the spring, recording 

 the extent of soil devoted to each kind of grain, a fact easily ascertained ; 

 the second time as soon as the harvest is concluded, announcing the 

 amount of the crop, as ascertained on several specimen fields, under differ- 



