GENERAL LAWS. 



departures from this either way is usually limited to 5 or 7. If in 

 one year there are 7 such marriages, it is sure to be preceded or 

 followed immediately or mediately by another year in which the 

 number is only 5. Again, the number of men between 30 and 45 

 contracting marriage with women above 60 is annually 18, subject 

 to a small occasional variation one way or the other. In the same 

 way, it appears that the number of men from 45 to 60 marrying 

 women above 60 is annually 27. 



The same regularity is found to characterise the number of mar- 

 riages between couples within any other prescribed limits of com- 

 parative age. 



84. The number of children resulting from each marriage cannot 

 be considered as depending on will. But assuredly the calling into 

 the world of illegitimate children must be admitted to have the 

 character of a voluntary act ; yet it is found, that in each country, 

 the annual* number of illegitimate children bears a fixed ratio to 

 the number of marriage-born. In France and Belgium this ratio 

 is 1 to 1 3. In England the proportion is found to be exactly the 

 same, and this appears to occur, from year to year, in both 

 countries with all the regularity of physical law. 



85. The statistics of crime being especially susceptible of exact 

 record, have been submitted to the same careful examination by 

 M. Quetelet, from whose researches it appears, that in the same 

 country the same number of crimes of the same description are 

 committed annually; and this curious result is found equally to 

 attend those classes of crimes which it would seem most impos- 

 sible to foresee. But, connected with these criminal statistics, 

 there is a circumstance still more curious and remarkable. It 

 necessarily happens, in the administration of criminal justice, 

 that, through the want of sagacity in the examining magistrates, 

 and a multitude of fortuitous circumstances surrounding the 

 accused, a considerable number of guiltless persons are brought 

 to trial. Now, will it be believed, that such is the prevalence of 

 general laws, that even in this class of moral phenomena, founded 

 on the results of fallible judgment, a rigorous law prevails, and 

 we find that, in each country, the proportion of persons charged 

 with offences who are acquitted, is from year to year the same ? 

 Thus, for instance, in France, 39 in every 100 of those accused 

 are as regularly acquitted as if the decisions of the juries were 

 made by putting 61 black balls and 39 white ones into a box, 

 and deciding the fate of the criminals by ballot. 



86. It is not only, however, voluntary acts which are subject 

 to this numerical regularity. Collectively speaking, persons re- 

 member and forget certain things with as much regularity as if 

 memory and attention were the result of wheel-work. A very 



95 



