[Date Prev][Date Next][Thread Prev][Thread Next]   [Date Index] [Thread Index] [Author Index

Mendel, Heredity, and Genetics



From page 258 of Life, Itself, Robert Rosen wrote:
 
"Evolution is concerned with "explaining" the characteristics of present generations retrospectively, in terms of the characteristics of temporally remote ones. Heredity is concerned with what passes between one generation and the next. Thus, if we view evolutionary processes as a kind of time integral of hereditary ones, in which a "generation" provides the individual time-step by which evolutionary chronicles are indexed, it is important to have a tangible bridge between today's generation and yesterday's, the immediately preceding one. This, roughly, is where genetics enters the picture.
 
Let us begin with a few historical comments.
 
Gregor Mendel was perhaps the first to concern himself with how the forms and behaviors of parents are related to those of their offspring. That is, he was comparing phenotypes, embodied in particular phenotypic characters, between one generation and the next. It seems to me, although as far as I know Mendel did not say so, that he was much influenced by the chemistry of his time. In [that] chemistry, a small number of different kinds of atoms, the chemical elements, could enter into an unlimited number of compounds. These compounds could be of utterly diverse character, or "phenotype"; they could be solid or fluid, crystalline or amorphous, red or yellow. In any case, the underlying atoms could be retrieved unchanged from any such combination into which they had entered, and recombined to yield new compounds, with new "phenotypes."  Chemical reaction, in its turn, could be regarded as a way of liberating the constituent atoms and recombining them, and thus turning given "phenotypes" into new ones. This clearly constitutes a powerful metaphor for what goes on between parents and the production of offspring. Note carefully that this viewpoint does not, in itself, constitute a reduction of heredity to chemistry; the relation between them is a metaphoric one, one of alternate realization. But it is powerful nevertheless.
 
Let us explore this metaphor a little further. In particular, let us ask whether the atomization of chemistry in any way "atomizes" the phenotypes manifested by individual chemical compounds in bulk. For instance: are there "atoms" for crystallinity, or solidity, or color, or any other such morphological characters? Even in chemistry, it is not nearly as simple as that; the atomization of a molecule does not, in that sense, at all entail a corresponding "atomization" of its phenotypic characters. Nor does it entail any kind of 1 to 1 mapping between these characters and those of the atoms, or even the molecules, which underlie them. Put in another language, which we developed earlier, the phenotype cannot be fractionated into "atoms" as a constituent molecule can; nor can the actual atoms be fractionated into separate phenotypic characters for which they are responsible. We will come back to this, or at least to its hereditary analogues, in a moment.
 
Returning to Mendel: he published his evidence for atom-like, particulate factors (as he called them) in heredity in 1866. As is well known, his papers sank without a trace. The fact is that no one then cared much about heredity per se; it was not until the connection between heredity and evolution, about which everybody cared, was slowly perceived, that Mendel's ideas were retrieved from oblivion. It was only then that the study of heredity, soon to be rechristened genetics, became a crucial part of biological main stream. By that time, Mendel himself was long dead.
 
Implicit in Mendel's ideas was a duality between the phenotypic characters, manifested in the forms and behaviors of individual organisms within a generation, and the particulate factors which pass between genera-- from parent to offspring. This is the genotype-phenotype dualism, which was first stated by August Weismann in his influential book Das Keimplasm, published in 1874, and to which I have already alluded. He proposed, in apparent ignorance of Mendel's work, a duality between what he called soma and what he called germplasm. Roughly speaking, the somatic part pertains to what is mortal in biology, with what we today identify with phenotype. Germplasm, on the other hand, pertains to what is immortal, flowing from generation to generation from the beginning of life on the planet. Thus the soma must be recreated anew in each generation, in such a way as to facilitate the flow of the precious germplasm. This was Weismann's conception of evolution itself.
 
We have already alluded to this dualism in the preceding section, where we identified organism with soma, and with life itself. Weismann, on the other hand, was the first to propose identifying life with germplasm, and with its flow. He thus, if indirectly, provided the crucial ingredient for the outlook embodied in contemporary biology.
 
But, as we have seen, Mendel himself was concerned primarily with this soma, with phenotype. he characterized his particulate factors in terms of their functional roles in this regard; i.e., in terms of how they manifested themselves in somatic forms and their generation. That was how his factors were recognized, and in turn, that is what the factors were for.
 
A basic transformation of the situation occurred roughly between the years 1900-1910. During these years, the identification came to be made between the hypothetical Mendelian "linkage groups" and the tangible, material, cytological structures called chromosomes. Accordingly, the Mendelian factors themselves, the genes, must be materially incarnated somehow in the structure of these bodies. So instead of resting content with characterizing these genes in functional terms, through their modulation of form and morphogenesis, we could rather imagine characterizing them intrinsically, in independent structural terms. That would, if accomplished leave only the problem of accounting for the genes' functional, morphogenetic activities in terms of this basic, intrinsic structure.
 
All this has a very contemporary ring; indeed, we are almost at the molecular biology of today. And of course, it is all fully in accord with the basic machine metaphor; Chromosomes could be fractionated from cells, and genes from chromosomes, purely as material structures, as independent material systems, without loss of information.
 
At the same time, the functional relation between genotype and phenotype was becoming progressively more complicated. Mendel thought that each of his functional factors could be associated with a fixed, demarcated phenotypic character; the one-gene-one-character hypothesis. As we have already suggested, it is not that simple; the dualism between genotype and phenotype does not extend to any kind of 1 to 1 mapping between them. Geneticists early discovered phenomena of pleiotropy, in which the "same gene" is involved in may phenotypic characters; and inversely, of polygeny, in which the same character is associated with many different genes. On top of this, there was growing evidence, soon to become an absolute necessity, for "cross-talk" among the genes themselves, directly or indirectly. All this was evidence of nonfractionability, but being uncongenial to a machine metaphor, it was regarded as a mere technical inconvenience, to be subsumed into a bigger machine.
 
All of this, it will be noted, pertains primarily to Weismannian soma. What about evolution, which was regarded as the primary issue? Here, the genotype-phenotype dualism manifests itself in another form. Namely, the operation of Darwinian selection pertains to phenotypes; to soma. On the other hand, evolution itself pertains to the flow of genotypes; to germplasm. Thus we must regard selection on the phenotypes as a modulator of this flow of genes from one generation to the next. This is where the notion of fitness comes in.
 
As presently viewed, fitness involves a decision made by natural selection about a particular soma, a particular phenotype. It is a decision that can be imputed to the associated genotype. A low fitness rating translates operationally into a disadvantage in populating the next generation; in leaving offspring. Thus a low somatic fitness serves as a filter, which prevents the associated genotype from reaching the next generation. Thus the gene flow is modulated, and the "gene pool" will manifest itself in somatically fitter individuals than we started with.
 
On the other hand, there are no "genes" for fitness either. Indeed, it cannot be regarded as a somatic feature in the usual sense at all. Fitness can only be defined operationally and retrospectively as far as individual organisms are concerned. In particular, it cannot be fractionated from anything. It is in fact a very mysterious concept; evolution in the Darwinian sense would be unthinkable without it, but it has always given evolutionists the greatest difficulty.
 
As we have seen, the introduction of the concept of genotype amounted to posing a new causal category for talking about somatic effects. That is, we could now answer a question of the form "why this somatic character?" with the answer "because these genotypic factors."
 
Biology today is furthermore completely committed to the idea that phenotype is a chemical concept. In other words, biological forms, and the morphogenetic processes which generate them, are ultimately chemical, or biochemical in nature. This, it will be seen, is a far cry from asserting an analogy of the kind sketched, above. This is a reduction.
 
On the face of it, it is an astonishing claim. It is not adduced on the grounds of great success in faithfully translating anatomical, embryological, or physiological processes into a syntactic chemical language. Quite the contrary, in fact, is true. It is adduced, rather, primarily on the grounds that otherwise we simply could not answer "why"? questions about these processes with a "because..." framed exclusively in terms of intrinsic, fractionable (i.e., chemical) structure. That is, unless we identify phenotype with biochemistry, we can no longer claim that the functional genetic factors originally posited by Mendel can be identified with fractionable pieces of chromosomal structure. Indeed; the viability of the entire machine metaphor in somatic biology currently rests precisely here; on the identification of phenotype with biochemistry. If it fails, then mechanism fails; but as we have stressed many times before, the alternative is not vitalism, it is complexity"