These breeds are always produced in the same way. The breeder selects a
pair, one or other, or both, of which present an indication of the
peculiarity he wishes to perpetuate, and then selects from the
offspring of them those which are most characteristic, rejecting the
others. From the selected offspring he breeds again, and, taking the
same precautions as before, repeats the process until he has obtained
the precise degree of divergence from the primitive type at which he
aimed.

If he now breeds from the variety thus established for some generations,
taking care always to keep the stock pure, the tendency to produce this
particular variety becomes more and more strongly hereditary; and it
does not appear that there is any limit to the persistency of the race
thus developed.

Men like Lamarck, apprehending these facts, and knowing that varieties
comparable to those produced by the breeder are abundantly found in
nature, and finding it impossible to discriminate in some cases between
varieties and true species, could hardly fail to divine the possibility
that species even the most distinct were, after all, only exceedingly
persistent varieties, and that they had arisen by the modification of
some common stock, just as it is with good reason believed that
turnspits and greyhounds, carrier and tumbler pigeons, have arisen.

But there was a link wanting to complete the parallel. Where in nature
was the analogue of the breeder to be found? How could that operation
of selection, which is his essential function, be carried out by mere
natural agencies? Lamarck did not value this problem; neither did he
admit his impotence to solve it; but he guessed a solution. Now,