This proves to demonstration not only the greater frequency of
anomalous skulls (and the same is true of physiognomical,
physiological, and psychological anomalies) amongst criminals, but
also that amongst these criminals between fifty and sixty per
cent. show very few anomalies, whilst about one-third of the whole
number present a remarkable combination, and one-tenth are normal
in this respect.


Amongst the statistical data exhibiting the primary
characteristics of the majority of criminals, the data connected
with relapsed criminals are especially conspicuous. Though
relapses, like first offences, are partly due to social
conditions, they also have a manifest biological cause, since,
under the operation of the same penal system, there are some
liberated prisoners who relapse and some who do not.

The statistics of relapse are unfortunately very difficult to
collect, on account of differences in the legislation of different
countries, and in the preparation of records, which, even under
the more general adoption of anthropometrical identification,
rarely succeed in preventing the use of fresh names by
professional criminals. So that we may still say, in the words of
one who is a very good judge in this matter, M. Yvernes, not
only that ``the Prisons Congress of London (1872) was compelled to
leave various problems undecided for lack of documentary evidence,
and especially the question of relapsed criminals,'' but also that
to this day (1879), ``we find varying results in different
countries, the exact significance of which is not apparent.''