Monozygotic twins arise from the division of a single fertilized ovum (zygote) in early embryonic life, whereas dizygotic twins arise from the implantation of two independently fertilized ova. The classical twin method compares the resemblance of monozygotic and dizygotic twins on the assumption that the former are genetically identical whereas the latter are similar to siblings. The twin method has been commonly credited1 to Francis Galton’s ‘The history of twins’ (1875),2 but this interpretation is wrong.3,4 What Galton did in this article was to track the life-history changes of twins to see whether twins who were similar at birth diverged in dissimilar environments or whether twins who were dissimilar at birth converged in similar environments. He found that on the whole they did not, from which he concluded that environmental factors (nurture) were much weaker than inherited ones (nature). He was unable to invent the classical twin method because of his inadequate understanding of the mechanism of heredity. (Remember that Mendel’s 1865 paper5 was not ‘discovered’ until 1900.6) In pursuing this analysis, it is also useful to consider ‘A theory of heredity’ (1875),7 which has further information about the twin study. (Galton’s papers are reproduced on the website galton.org.)
In Hereditary Genius (1869),8 Galton undertook a statistical investigation of the extent to which eminent men had eminent relatives and concluded that there was strong evidence of the inheritance of genius, understood as meaning ability. The book had a mixed reception, many of the critics suggesting that Galton had shown a hereditarian bias in interpreting his results and had underestimated the importance of common environmental factors within families.4 In response, he devised a questionnaire that he sent to Fellows of the Royal Society asking them among other things about their antecedents and about the origin of their taste for science. The results in English Men of Science (1874)9 are inconclusive, but the book has an introduction which coined the phrase ‘nature and nurture’ for the effects of heredity and the environment, and which suggested that identical twins could be used as evidence of the predominance of nature over nurture by relating anecdotes showing that twins who were identical at birth retained their near identity in adult life despite differences in nurture.
Galton then sent a questionnaire about their life histories to a large number of adult twins, reporting the results in ‘The history of twins’.2 He found that the pairs of twins could be classed into three groups by their degree of similarity: those who were very similar at birth, who were always of the same sex; those who were moderately similar, who could be of like or unlike sex; and those who were very dissimilar, who were all of the same sex. In the first group of twins very similar at birth, he found that differences in environment as adults had not caused them to diverge at all unless some form of illness had affected one of them. In the third group of twins who were very dissimilar at birth, an identity of nurture in childhood and youth had not tended to assimilate them at all. He did not consider the second group of moderately similar twins further. He concluded that ‘nature prevails over nurture when the differences of nurture do not exceed what is commonly to be found among persons of the same rank of society and in the same country’.
Galton knew the embryological difference between monozygotic and dizygotic twins from the work of Kleinwächter10 and Späeth.11 He also knew that monozygotic twins were always of the same sex whereas dizygotic twins were equally likely to be alike or unlike in sex. Since all 20 of the very dissimilar twins were of the same sex, he thought that they were probably monozygotic7: ‘Hence there is much probability that my cases of strong dissimilarity were usually, if not invariably, cases of true [monozygotic] twins. But I have no direct evidence one way or the other’. In the light of current knowledge, it seems likely that the 20 like-sexed dissimilar twins were dizygotic twins from the upper range of dissimilarity; that they were all like sexed may have been due to the greater ease of diagnosing dissimilarity in like- than in unlike-sexed twins. But this does not affect Galton’s argument that the similarity of their environment in childhood and youth did not tend to assimilate them.
Galton thought that the group of very dissimilar like-sexed twins might be monozygotic because of his understanding of the mechanism of heredity in 1875.4,7 He supposed that an individual’s inheritance from its parents consisted of a very large number of hereditary particles or gemmules, collectively called the stirp, of which a small proportion were patent, developing into the cells of the adult organism, whereas the rest remained latent and capable of transmission to the next generation. He also supposed that the stirps of siblings were almost identical, so that genetic differences between them depended on differences between the patent gemmules derived from them.
He then accounted for the differences between different types of twins as follows7:
What can be the reason that, out of identically the same primary stirp, either two absolutely dissimilar persons can be developed, or else two closely similar ones; while the intermediate cases are comparatively rare, so that they may be considered due to quite another and more common contingency – namely, that in which the twins are not produced out of the same ovum, but from separate ova? The answer I suggest is as follows: as regards the similarity of true [monozygotic] twins, there can be little difficulty; we should expect, on statistical grounds, that the two halves of any assemblage of germs would be much alike. The secondary stirps of the twins being alike, and the circumstances under which the bodily structure is developed out of them being almost identical, the results must be closely similar. On the other hand, as regards the dissimilarity, we might expect that if there had happened to be a sufficient delay before the division of the primary stirp, to allow its germs to arrange themselves according to their affinities, the twin halves of the primary stirp would be strongly contrasted.
In this passage, the primary stirp is the stirp of a zygote immediately after it has been fertilized and the secondary stirps of monozygotic twins are their stirps immediately after the zygote has divided into two. Galton explains the similarities of three classes of twins as follows. (i) Monozygotic twins who divided shortly after conception are very similar because they have very similar secondary stirps derived from the division of the same primary stirp, and these secondary stirps develop in very similar conditions. (ii) Dizygotic twins and siblings are moderately similar because they are derived from very similar primary stirps, which develop under somewhat similar conditions. No prediction is made about the degree of similarity. (iii) Monozygotic twins who divided a considerable time after conception are very dissimilar because the primary stirp, which divides into two, has been organized into contrasting halves.
Galton could not in 1875 have invented the classical twin method based on this non-Mendelian theory of heredity. What he did was to track the life-history changes of twins to see whether twins who were very similar at birth diverged in later life or whether twins who were very dissimilar at birth converged later, as a test of the relative importance of nature and nurture. The classical twin method of comparing the similarities of monozygotic and dizygotic twins on Mendelian assumptions was first used in the 1920s.3
Conflict of interest: None declared.