. is nearer the
mark. This is also my own view, but for the purposes of this inquiry
we shall assume that it is 10 per cent.

How large a proportion of the criminal population is made up of the
middle and upper classes? An answer to this question would at once
show the exact relation between poverty and crime. If it could be
shown that the well-to-do classes, in proportion to their numbers, are
just as much addicted to the commission of criminal acts as the poorer
people, it would demonstrate that crime prevailed to an equal extent
among all sections of the community, and was not the work of one class
alone. Unfortunately, such statistics are not to be had. But, as the
facts are not to be got at directly, this does not mean to say that it
is impossible to catch a glimpse of what they are. This may be done in
the following manner:--According to the report of the Prison
Commissioners, between 5 and 6 per cent. of the persons committed to
gaol during the year ended March, 1890 (omitting court-martial cases),
were debtors and civil process cases. Now, it may be taken as certain
that in a very small proportion of these cases were the prisoners
working people. Nearly all these offenders are to be considered as
belonging to the well-to-do classes. Yet we see that they form 5 per
cent. of the criminal population, and it has to be remembered that the
fraudulent debtor is just as much a criminal, nay, even a worse
criminal in many instances than the thief who snatches a purse. In
addition to this 5 per cent. there is at least 3 per cent. of the
ordinary criminal population belonging to the higher ranks of life. At
the lowest estimate we have 6 per cent. of the criminal population
springing from the midst of the well-to-do, and if all cases of
drunkenness and assault were punished with imprisonment instead of a
fine, it would be found that the well-to-do showed just as badly in
the statistics of crime as their poorer neighbours.

In making this statement with respect to fines, I do not wish it to be
understood that all cases of drunkenness and assault should be
followed by imprisonment. On the contrary, it is a great mistake to
send anyone to gaol if it can possibly be avoided, and imprisonment
should never be resorted to so long as any other form of punishment
will serve the purpose. What is here stated is merely meant to bring
out the fact that the proportion of well-to-do among the prison
population does not accurately repres