People have always known that heredity is a complicated thing — that it is more than a simple "mixing" model would account for. But this misses something of a key point, which is that until around the time that Mendel's work was "rediscovered" (1900, but there is a lot that could be said about how much of a "rediscovery" it was), neither scientific nor folk views of heredity were as "hard" as we now think of them. That is, even someone like Darwin believed that there had to be a lot of role for "development," and did not accept that heredity was a locked-in "code" as we think of it today. It is not until the 20th century that this "hard" hereditarian view really becomes dominant, though it did have a few proponents in the late 19th century (e.g. Francis Galton, August Weisman).

("Hard" heredity assumes that the development of the organism is mostly due to inborn characteristics, e.g. a set of instructions or code, and that only these instructions are passed on to kin, not abilities or traits gained through later development or environmental influences. "Soft" heredity, which even Darwin believed in, says that some acquired characteristics are potentially passed on through breeding. In the late 19th century, developmentalism and "soft" heredity was still seen as essential to understanding organismic development. By the early 20th century the view of heredity had gotten very "hard." Today we know it to be a very complex situation indeed, with cytogenetics and epigenetics and genetic regulation and whatnot not really fitting into either of those categories.)

As for folk understandings — the presence of unusual, unexpected traits is common enough that people would have already seen it in families, and certainly in lifestock. In breeding terms it is known as a "sport" — a "surprise" trait that shows up occasionally and defies a simple "blending" expectation. While I would not be surprised if it at times led to suspicions of adultery (who would be surprised at the human capacity for jealousy?), note that such occasional surprises would in theory be noticed in family histories, because these things do not come out of nowhere (there were probably great-grandparents or cousins with the same trait expressed).

On the history of heredity, Peter Bowler's The Mendelian Revolution is a valuable little read.

Two folk beliefs can come into play here. One is the very ancient idea of "influences," and the other - why can't I think of the term? - that a female's first mate continues to influence her offspring by later mates.

Influences are like, if a pregnant woman is startled by a rabbit, the child will have a hare-lip, quite literally. If she sees a grotsque person, the child will be deformed. So besides sensibly saying, oh, look, Baby has Grandma's hair, it could be blamed on her having a yellow and creme boudoir.

Two folk beliefs can come into play here. One is the very ancient idea of "influences," and the other - why can't I think of the term? - that a female's first mate continues to influence her offspring by later mates.

Influences are like, if a pregnant woman is startled by a rabbit, the child will have a hare-lip, quite literally. If she sees a grotsque person, the child will be deformed. So besides sensibly saying, oh, look, Baby has Grandma's hair, it could be blamed on her having a yellow and creme boudoir.