IDEAS ABOUT LIVING UNITS, 1864-1909: A CHAPTER IN THE HISTORY OF GENETICS L. C. DUNN* The concept of the gene as a self-replicating unit ofheredity has been one ofthe most influential ideas oftwentieth-century biology. It was derived from Mendel's experimental proofofthe segregation ofalternative forms of a gene (now known as alleles) into different gametes in the formation of the germ cells of a hybrid. Although the principle was demonstrated in Mendel's paper of1866, it did not begin to influence biological theories until after its rediscovery, confirmation, and extension in the first years ofthe twentieth century. That period, say from 1900-1906, can now be seen as the chief break in the continuity of ideas about the transmission system ofheredity. What had happened prior to that time— except, of course, for the original publication of Mendel's pea experiments —had little genetic connection with the later course of development ofthese ideas. There was no such sharp break in studies ofevolution or ofdevelopment; in fact, many students ofthe latter field would say that it has not yet been affected in a major way by what happened in 1900. Views about the mechanisms ofevolution began to change more gradually in the 1920's and 1930's. It is, ofcourse, not surprising that the effects ofthe break should be felt first in the territory, so to speak, in which it occurred—that ofthe transmission mechanism ofheredity. It is interesting to reflect, from the vantage point of 1965, that that "territory" was not recognized as such until after the break occurred. Heredity, with the definiteness given to it by modern genetics, now seems to be a well-marked field, but in the nineteenth century this was not so. Then efforts to develop general theories of heredity werejudged primarily by their applicability to problems ofvaria- * Columbia University, Nevis Biological Station, Irvington-on-Hudson, New York. This will form a chapter in a forthcoming book, Genetics, 1864-10,30.: The Development ofSome of the Main Ideas, to be published by McGraw-Hill Book Co. 335 tion, evolution, and development, and we see signs today, after nearly one hundred years, ofa return ofthat attitude. Then it was due to the absence ofprecise questions about the transmission mechanism—what I referred to above as non-recognition ofthe territory. Now genetics threatens to disappear as a separate field because the solution ofsome ofits basic problems has led to its absorption into the general fabric of biological knowledge. The gene, whatever its fate as an "ultimate" unit in the structure and functioning ofliving material, has become a basic part of the description of that organization. The nature of the radical change that occurred around 1900 can be studied by examining some ofthe questions ofchiefinterest to biologists in the immediately preceding period. Three main groups of questions can be distinguished: those concerning the nature ofspecies differences and the effects ofhybridization on domesticatedplants; those having todo with morphology, development, and cytology, chiefly in animals; and those specifically concerned with theories ofheredity. The last took the form of ideas about living units as elements ofhereditary transmission and control of the development of form and function. Needless to say, all were directed toward general questions raised as the result ofthe intense interest in the mechanism of evolution evoked by the publication of Origin of Species in 1859. I shall deal first with the last set ofquestions and shall consider the history ofideas concerning living units in the nineteenth century and the state ofsuch theories today. Many ofthe documents describing the development oftheories ofheredity and ofliving units in the nineteenth century have been brought together, with commentary, by Alfred Barthelmess in his valuable book Vererbungswissenschaft [?]. The source of'ideas about living units could, ofcourse, be traced to much earlier times; Jean Rostand has made an attempt ofthis sort in his Esquisse d'une histoire de l'atomisme en Biologie [2]. But the particles imagined by Buffon, Diderot, and others in the period before experimental study began were purely speculative and did not lead to the discovery ofthe true atomisme which Mendel proved. The "elementary particles" ofMaupertuis (1751), however, were invoked to account for the segregation of a dominant gene for polydactyly...