tates that he was acquainted with only "one
questionable case of a contrary nature." Here, then, is a remarkable
and well-established instance, not only of a very distinct race being
established per saltum, but of that race breeding "true" at once, and
showing no mixed forms, even when crossed with another breed.

By taking care to select Ancons of both sexes, for breeding from, it
thus became easy to establish an extremely well-marked race; so peculiar
that, even when herded with other sheep, it was noted that the Ancons
kept together. And there is every reason to believe that the existence
of this breed might have been indefinitely protracted; but the
introduction of the Merino sheep, which were not only very superior to
the Ancons in wool and meat, but quite as quiet and orderly, led to the
complete neglect of the new breed, so that, in 1813, Colonel Humphreys
found it difficult to obtain the specimen, whose skeleton was presented
to Sir Joseph Banks. We believe that, for many years, no remnant of it
has existed in the United States.

Gratio Kelleia was not the progenitor of a race of six-fingered men, as
Seth Wright's ram became a nation of Ancon sheep, though the tendency of
the variety to perpetuate itself appears to have been fully as strong
in the one case as in the other. And the reason of the difference is
not far to seek. Seth Wright took care not to weaken the Ancon blood by
matching his Ancon ewes with any but males of the same variety, while
Gratio Kelleia's sons were too far removed from the patriarchal times
to intermarry with their sisters; and his grandchildren seem not to have
been attracted by their six-fingered cousins. In other words, in the one
example a race was produced, because, for several generations, care
was taken to 'select' both parents of the breeding stock from animals
exhibiting a tendency to vary in the same condition; while, in the
other, no race was evolved, because no such selection was exercised.
A race is a propagated variety; and as, by the laws of reproduction,
offspring tend to assume the parental forms, they will be more likely to
propagate a variation exhibited by both parents than that possessed by
only one.

There is no organ of the body of an animal which may not, and does not,
occasionally, vary more or less from the normal type; and there is
no variation which may not be transmitted and which, if selectively
transmitted, may not become the foundation of a race. This g