I will cite from Barr's Buffon (Buffon's Natural History Containing a Theory of the Earth, a general history of man, of the brute creation, and of vegetables, minerals, &c. From the French. With notes by the translator. London: J. S. Barr) published in ten volumes between 1792 and 1807.
Although Buffon rejected the idea that species could be transformed into new species, he believed that human could produce new varieties and races of domestic animals and plants. Nevertheless, he regarded these varieties and races as being degenerated in comparison with the species in their original state, which he believed to be perfect creations.
Starting with an observation about the variation of dogs with external conditions, Buffon shows how dogs and other domestic species are productions of artificial selection by humans. Here's a relevant passage from page 310ff of volume 5 in Barr's Buffon (see here or here):
"From the same causes [climate, food etc.] arise that great variety so visible in the height, figure, length of the snout, form of the head, length and direction of the ears and tail, colour, quality and quantity of hair, &c. so that there seems to remain nothing constant in these animals but the conformity of their internal organisation, and the faculty of procreating together. And as those which differ most from each other can intermix and produce fertile individuals, it is evident that dogs, however greatly they may vary, nevertheless constitute but one species." (p. 310)
"Those animals which are independent and can chuse for themselves both their food and climate, are those which best preserve their original impressions, and we may believe the most ancient of their species are the most faithfully represented by their descendants. But those which mankind have subdued, transported from climate to climate, whose food, customs, and manners of living he has changed, may also be those which have changed most in their forms; and it is a fact that there are more varieties among domestic than wild animals; and as among domestic animals the dog is the most attached to man, lives also the most regularly, and who possesses sentiments to render him docile, obedient, susceptible of all impressions, and submissive to all restraints, it is not astonishing that he should be that in which we find the greatest variety not only in figure, height, and colour, but in every other quality. There are also other circumstances which contribute to this change. The life of the dog is short, his produce is frequent, and in pretty large numbers; he is perpetually beneath the eye of man, and whenever by an accident, which is very common in nature, there may have appeared an individual possessing singular characters, or apparent varieties, they have been perpetuated by uniting together those individuals, and not permitting them to intermix with any others; as is done in the present time, when we want to procure a new breed of dogs, or other animals." (p. 311)From artificial selection of dogs, Buffon takes a discourse into plants, where he believed the annual plants to be particularly likely to vary, because of their short generation time. Here, he contradicts his own doctrine, according to which varieties are degenerations of the original stock, because he admits that man can "improve" the species. He also states that the more domesticated or cultivated species vary the strongest:
"It is the same with annual plants (some of which may be said to be artificial or factitious), when compared with other vegetables. Wheat, for example, has been so greatly changed by man that it is not at present to be any where found in a state of nature; it certainly has some resemblance to darnel, dog-grass, and several other herbs of the field, but we are ignorant to which its origin ought to be referred; and as it is renewed every year, and serves for the common food of man, so it has experienced more cultivation than any other plant, and consequently undergone a greater variety of changes. Man can, therefore, not only make every individual in the universe useful to his wants, but, with the aid of time, he can change, modify, and improve their species; and this is the greatest power he has over Nature. To have transformed a barren herb into wheat is a kind of creation, on which, however, he has no reason to pride himself, since it is only by the sweat of his brow, and reiterated culture, that he is enabled to obtain from the bosom of the earth this, often bitter, subsistence. Thus those species, as well among vegetables as animals, which have been the most cultivated by man, are those which have undergone the greatest changes; and as we are sometimes, as in the example of wheat, unable to know their primitive form, it is not impossible that among the numerous varieties of dogs which exist at present there may not be one like the first animal of his species, although the whole of these breeds must have proceeded virtually from him." (p. 312f)