I've been reading Ron Amundson's new history of biology book, The Changing Role of the Embryo in Evolutionary Thought.
I'll be posting a lot on this book over the next couple of weeks, as I have taken a lot of notes. In short, I find it to be a very interesting and thought-provoking revision of the history of evolutionary theory. I often find when I'm reading a history that things didn't happen the way I learned in school. Amundson gives example after example of the way that the history of biology was constructed to defend a specific set of beliefs -- those that constituted the Evolutionary Synthesis. The result is that in many particulars, the history of pre-Darwinian and post-Darwinian biology that you thought you knew just isn't the way things really happened. Amundson's particular interest is development, and especially the ways that developmental biology was jettisoned along the road to neodarwinism.
On page 144, August Weismann enters the story. If you know the history of biology, then you probably remember one key fact about Weismann: he innovated the distinction between germline and somatic cells. You probably remember this because the sequestration of the germline explains why Lamarckian inheritance is impossible --- changes to the somatic cells cannot affect the germline cells that ultimately produce gametes. That's an essential insight to population genetics: it underlies particulate inheritance. It also is usually taken to mean that the transmission of genes to offspring is independent of the processes of development; in other words, it underlies the distinction between transmission genetics and developmental genetics.
You may know a lot more than this about Weismann, but I certainly didn't, so I was intrigued by the real motivation for his theory.
Weismann enters Amundson's narrative as an example of pre-synthesis models of heredity. As it turns out, under Weismann's theory, the germline sequestration was necessary not to refute Lamarckian inheritance, but to allow embryonic development itself:
The central problem of the study of embryological development is explaining the increase in heterogeneity in the developing embryo. Seen in terms of the cell theory, increased heterogeneity could be conceived as cellular differentiation. How does the single cell of the zygote give rise through division to the specialized cells of the various parts of the body? The answer given by Weismann and [Wilhelm] Roux was the mosaic theory of development. Roux stated his version of the mosaic theory in 1885, the same year that Weismann proposed the germ-soma distinction. Mosaic or autonomous theories of development assert that the nature of body parts is determined in advance of their acutal development, and determined independently of the body parts around them. In contrast, regulative theories of development claim that body parts take their nature from their position within the embryo (Amundson 2005:145, emphasis in original).
How did this "mosaic theory" work? The cytoplasm was theorized to contain particles (ultimately called determinates) that comprised all the hereditary information. As the embryo differentiated, different cells received different subsets of the original determinates. Cells that received bone-determinates became bone cells, those that received heart-determinates became cardiac cells, and so on. Therefore the original heritable material was parcelled out to different somatic cells, meaning that no somatic cell contained all the determinates necessary to construct an entire body. In this theory, heredity and ontogeny were inextricably linked: indeed, the hereditary particles were directly responsible for the development of the organism.
It is easy to see that under this theory, the sequestration of the germline is essential, otherwise reproduction would be impossible. Only by setting aside a group of cells that would retain all the hereditary determinates could the ability to produce another individual be transmitted to the next generation. Weismann's mechanism of development dictated the germ-soma distinction.
As Amundson describes, the mosaic theory fell out of favor for two reasons. Embryologists didn't like it because it didn't really explain the formation of different tissues. Although the theory was intended to explain why the differentiation occurred (through the progressive assortment of determinates into different tissues), it didn't explain how those determinates actually got into their ultimate positions or how they determined the actual properties of different tissues.
And it was found to be inconsistent with the burgeoning field of genetics, under the influence of Thomas Hart Morgan. Cells were found to divide equally in all cases, and genetic material was found to assort equally into both daughter cells. This removed any empirical support for the "determinates", and left the mosaic theory without a credible mechanism.
Amundson's point in describing the theory is to point out that before Morgan, heredity was universally assumed to involve development: individuals inherited not an information-bearing particle, like DNA, but instead the ontogenetic program. This concept of heredity could not lead to population genetics, because it did not admit a mathematical analysis of inheritance along Mendelian lines. Instead, population genetics developed upon Morgan's assertion that the tranmission of traits from parents to offspring could be studied even in the absence of knowledge of how the traits develop.
Weismann's germ-soma distinction was thus resurrected as a key to Mendelian inheritance: it explained the transmission of genetic material to offspring as independent of the process of development. Amundson cites John Maynard Smith to this effect, and adds:
In 1927, Weismann was the emblem of the integration of development and heredity; in 1982, he was the hero who had divorced the two fields. Why the change? ... Embryology itself was almost forgotten, and Weismann's mosaic embryological theory was totally forgotten. This enabled a replacement of the historical Weismann ... with the modern pseudo-Weismann (whose germ line-soma distinction seemed to anticipate the genotype-phenotype distinction. References to Weismann were absent from formative Synthesis literature. Later in the century, Weismann was recalled to mind -- he was the person who had refuted Lamarckism with his germ line-soma distinction (ibid., 219).
Weismann has gained special importance in evolutionary theory for his argument against use-inheritance for a good reason: Darwin himself embarrassingly (at least to the modern sensibility) assumed that use-inheritance was an important source of heredity. Thus, Weismann's germ-soma distinction is placed as a corrective to Darwin, and thereby a major advance toward modern evolutionary theory.
In view of Weismann's actual ideas, this posturing is highly ironic.
UPDATE: I have intended to give Gould's Structure of Evolutionary Theory another look after finishing Amundson to compare their treatment of the embryology. But upon writing this, I felt like checking what Gould had to say about Weismann. Apparently he didn't get Amundson's point at all --- perhaps the genotype-phenotype distinction was drilled into Gould once too often, also. Instead, Gould focuses on Weismann's later theory of germinal selection as a precursor to his own multilevel selection ideas. Thus he gives Weismann a good 20 pages, but doesn't outline the importance of his mechanism of heredity to the germ-soma distinction.