genealogists and others who would naturally be more
interested in the relationships than in pathological conditions. I
asked, however, that all such results be noted. Among 722 children of
first cousins I found 95 or 13 per cent who were defective in the
sense in which Bemiss used the term. This is much nearer the actual
percentage, but I have reason to believe, as will be seen hereafter,
that even this percentage is far too high. A good illustration of the
unconscious bias, which I tried to avoid is afforded by the reports on
the cause of death among children of first cousins. Only 58 replies
were given to this question, and of the 58 deaths 14 or one-fourth
were either accidental or otherwise violent, while only one person was
reported to have succumbed to pneumonia.

Many efforts have been made to investigate the occurrence of
degeneracy in the offspring of consanguineous marriages, by studying
communities in which such unions have been frequent, but the results
are untrustworthy. Huth[52] quotes a number of instances where
communities have lived for generations without crosses and with no
apparent degeneracy, while other writers tell of high percentages of
degeneracy. Smith's Island, Maryland, as has been said, seems
absolutely free from serious congenital abnormalities, in spite of the
great frequency of consanguineous marriages.

[Footnote 52: _Marriage of Near Kin_, chap. iv.]

The causes of degeneracy are so varied, complicated, and obscure that
even if consanguinity is a cause, there can be but few cases in which
it is not complicated by other factors. But for the same reason that
it is so difficult to prove any connection between consanguinity and
degeneracy, it is equally difficult to disprove such a connection. It
is very probable that from the mere operation of the law of heredity,
there must be a comparatively large percentage of degenerates among
the offspring of related parents, for defects which tend to be bred
out by crossing are accentuated by inbreeding. This may be the reason
for the disagreement among investigators of isolated communities. If
an island, for instance, were settled by a small group of families in
even one of which some hereditary defect was common, in the course of
a few generations that defect would be found in a relatively large
part of the population. While if the same island were settled by
perfectly sound families, there would only be a remote chance of any
particular defect