In _The Variation of Animals and Plants under Domestication,_ which appeared first in 1868 and in a revised edition in 1875, Darwin developed a theme to which he had accorded great rhetorical and evidentiary significance. He had begun the _ Origin_ with a description of artificial selection as practiced by farmers, stock breeders, and pet fanciers, thus using a reassuringly homely example—one recognizable by the general public as well as by members of the scientific community—to introduce the most innovative component of his evolutionary theory. In addition, domesticated animals and plants, because they were numerous and available for constant observation, provided a readily available body of evidence.
Reassuring as it was, the analogy between natural and artificial selection was far from perfect. The point of Darwin’s analogy was to make the idea of natural selection seem plausible by characterizing it as a grander version of a well-known process while emphasizing its efficiency and shaping power. He noted, for example, that some of the prize birds bred by London pigeon fanciers diverged so strikingly in size, plumage, beak shape, flying technique, vocalizations. bone structure, and many other attributes, that if they had been presented to an ornithologist as wild specimens, they would unquestionably have been considered to represent distinct species, perhaps even distinct genera. Darwin argued that if the relatively brief and constrained selective efforts of human breeders had produced such impressive results, it was likely that the more protracted and thorough-going efforts of nature would work still more efficaciously.
But as Darwin acknowledged, there were some fairly obvious reasons why the two processes might diverge. The superior power of natural selection—“Man can act only on external and visible characters: nature . . . can act on . . . the whole machinery of life. Man selects only for his own good; Nature only for that of the being which she tends” (_Origin,_ chap. 5)—might constitute a difference of kind rather than of degree, as might the much greater stretches of time available for natural selection. Further, although the mechanism of the two processes appeared superficially similar, their outcomes tended to be rather different. Natural selection produced a constantly increasing and diversifying variety of forms; it never reversed or exactly repeated itself. Anyone familiar with artificial selection would have realized that, although new breeds were constantly being developed and although neither improved wheat nor improved cattle showed any tendency to revert to the condition of their aboriginal wild ancestors, the strains produced by human selection were neither as prolific nor as durable as those produced by nature. Indeed, the animals and plants celebrated as the noblest achievements of the breeder’s art were especially liable to delicacy and infertility. Highly bred strains, long isolated from others of their species to preserve their genealogical purity, far from serving as a springboard for further variation, often had to be revivified with infusions of less-rarefied blood. Yet any relaxation of reproductive boundaries threatened subsidence into the common run of conspecifics.