My conclusion is that Mendel deliberately, though without any real falsification, tried to suggest to his audience and readers an unlikely and substantially wrong reconstruction of the first and most important phase of his research. In my book I offer many reasons for this strange and surprising behavior,53 but the main argument rests on the fact of linkage. Mendelian genetics cannot account for linkage because it was based on the idea of applying probability theory to the problem of species evolution. (...) Central to the theory is the law of probability according to which the chance occurrence of a combination of independent events is the product of their separate probabilities. This is the common basis of Mendel's first and second laws, but this is why Mendel's second law on independent assortment is enunciated in too general a way. From Morgan's work we now know that characters may not always be independent if their genes are located very close one to the other on the same chromosome. And this was also the basis of Mendel's personal drama: he surely observed the effects of linkage, but he had no theoretical tools with which to explain it. So he presented his results in a logical structure consistent with the central idea of his theory. Had he described the real course of his experiments he would have had to admit that his law worked for only a few of the hundreds of Pisum characters — and it would thus have been considered more of an exception than a rule. This is why he insisted on the necessity of testing the law on other plants, and this is why in his second letter to Carl Nägeli he admits that the publication of his data was untimely and dangerous.54.We can argue that already in 1866 Mendel was less confident that his so-called second law had the same general validity as the first — and that later he lost his confidence altogether. Contemporary testimony indicates that in the end he became as skeptical as all his contemporaries as to the scientific relevance of his theory.55 But he was wrong. His research is in no way the fruit of methodological mistakes or forgery, and it remains a landmark in the history of science. He was only the victim of a strange destiny in which the use of probability theory was responsible, at the same time, for the strength and for the weakness of his theory. We must still consider him the father and founder of genetics. (shrink)