Reactionary fringe meets mutation-biased adaptation. 5.4. Taking neo-Darwinism seriously

This is part of a continuing series of posts by Arlin Stoltzfus on the role of mutation as a dispositional factor in evolution. In this post Arlin discusses his view of neo-Darwinism and why it is inconsistent with macromutations and lateral gene transfer. He equates neo-Darwinism with the Modern Synthesis (1959 version), a comparison that might be challenged. Click on the links in the box (below) to see the other posts in the series.

---

Reactionary fringe meets mutation-biased adaptation. 5.4. Taking neo-Darwinism seriously
by Arlin Stoltzfus
The Modern Synthesis is often described as the result of combining Darwinism and genetics. This description, in my opinion, is concise and historically accurate: the Modern Synthesis of 1959 is a sophisticated attempt to arrange the pieces of population genetics to justify a neo-Darwinian dichotomy in which variation merely supplies raw materials, and selection is the source of initiative, creativity and direction.

Reactionary fringe meets mutation-biased adaptation
Introduction
1. The empirical case
2. Some objections addressed
3. The causes and consequences of biases in the introduction process
4. What makes this new?
5. Beyond the "Synthesis" debate
    -Thinking about theories
    -Modern Synthesis of 1959
    -How history is distorted
    -Taking neo-Darwinism
      seriously
    -Synthesis apologetics
6. What "limits" adaptation?
7. Going forwardFor instance, to disallow the mutationist possibility that mutations initiate change, an abundant variation-maintaining gene pool is posited, and recombination replaces mutation as the proximate source of variation. To deprecate the idea of large changes, it is argued that the smallest changes are the most likely. The creativity of selection is justified by the way that selection brings together highly improbable combinations of alleles at many different loci. The opposing-pressures argument says that variation cannot be a cause of direction like selection.

Misguided Synthesis apologists have essentially re-written this theory to remove the neo-Darwinian ideology that was its original raison d'etre.

We will consider Synthesis apologetics later. My purpose here is to illustrate how to apply the Modern Synthesis of 1959, i.e., how to treat this theory like any other theory in science. The Modern Synthesis of 1959, and neo-Darwinism more generally, is a useful theory even when wrong.  As noted previously, we cannot carry out modus tollens reasoning about evolution without false models. Being able to describe the world by enumerating the laws that it does not follow is important for science.

Everyone understands this, I think, in regard to neutral models. Being able to specify a neutral hypothesis is an important skill.  We would be deeply disturbed if a student refused to develop and apply neutral models, arguing that the proper approach to theories in science is to change and expand them until they fit all the data, or arguing that neutral models have been rejected and good scientists should only use correct models.

Likewise, an important skill is to specify a neo-Darwinian hypothesis in which selection is a creative governing force and mutation is merely the supplier of raw materials. Neo-Darwinism has been influential precisely because it is a powerful explanatory theory, literally capable of explaining any adaptive feature as the accumulation of infinitesimals under the guidance of selection. Every evolved feature is like a sand-sculpture built by selection: each grain of sand may have a unique shape and history, but that hardly matters-- what matters is the force that brought them together.  

Let us consider, as an extreme case, lateral gene transfer.  For lateral gene transfer to occur in a manner conforming to the Modern Synthesis (1959) position on population genetics, the entire process of change must take place gradually by shifting allele frequencies, it must involve many loci at once, it must be fueled by recombination, and it must be governed by selection.

This is not difficult to conceive.  Consider the imaginary case of transferring a 5-bp segment denoted TRANS from species X to species Y.  If we can accomplish this, then we can scale it up from 5 bp to 5000 bp.  We begin with a species X genome with the sequence xxxxxxxxxTRANSxxxx and a species Y genome with the sequence yyyyyyyyyyyyyy.  The transfer process gives rise to an altered Y genome yyyyyyyyyyTRANSyyyy that did not exist previously.

The figure above shows how this could happen in a neo-Darwinian manner consistent with the Modern Synthesis of 1959.  The figure shows samples from the gene pool of species Y, at 3 successive points of time. Before the start of evolution (left figure), micro-mutational introgression from species X fills up the gene pool of species Y, so that there are many Y genomes with infinitesimal pieces of the TRANS segment, inserted into different points in the genome.  Evolution, in the Modern Synthesis, begins with multi-locus variation in the gene pool.  

From the beginning, critics of Darwin's thinking have objected that selection would not be able to distinguish what Mivart (1871) called "incipient stages of useful structures," e.g., the rudimentary beginnings of a wing, or in this case, the rudimentary beginnings of a gene.  The response of Darwin's followers has always been that one does not need to propose large initial steps (saltations), because selection is powerful and can leverage even the smallest effects. Perhaps the initial steps were selected for a different function, e.g., the rudimentary wings might have been used for temperature control, or for catching insects, rather than for flight. That is, the "early steps were not inadequate wings but well-adapted something-elses" (see Gould 1985).

Therefore, in our model of lateral gene transfer, we simply take neo-Darwinian thinking seriously: we assume that selection would recognize the beginnings of a new gene with some function, bringing together new combinations and increasing the frequencies of subsequences like TRA and NS-- indeed, an experimentally testable implication of our neo-Darwinian theory is that there exists a historical path of functional intermediates, beginning with these small fragments. Over time, recombination and selection would bring together a useful TRANS sequence in the Y genome. 

From this example of a 5-bp sequence, we can extrapolate to the lateral transfer of a sequence of any length.

Thus, neo-Darwinian lateral gene transfer is possible, following the Modern Synthesis of 1959.  In the scenario above, the process of lateral transfer takes place gradually by shifting allele frequencies, it involves many loci at once, it is fueled by recombination, and it is governed by selection.  Introgression of gene fragments from species X was vital for this process, but it was only a source of raw materials, not a source of creativity, direction or initiative.

Taking a disciplined approach to applying this theory sharpens our understanding of precisely why the natural process of lateral gene transfer is not consistent with neo-Darwinism. The fact that genetic material is transferred from one species to another is not the issue. Instead, what makes lateral transfer non-Darwinian is that it occurs by macromutations that transfer an entire gene, or a set of them, at once.  This mutational process is non-arbitrary and creative: unlike a micro-mutation with infinitesimal effects, the initial inter-genomic transfer mutation strongly shapes the subsequent evolutionary potential.

Likewise, the evolutionary emergence of a eukaryotic mitochondrion or chloroplast from an external bacterial lineage is not by itself contrary to neo-Darwinism: there is no conceptual difficulty in imagining a mitochondrion being built up gradually by thousands of micro-steps.  There is no conceptual difficulty in imagining a new gene being built up-- or laterally transferred, or transferred from organelle to nucleus-- one nucleotide at a time, via a gene-pool full of abundant variation. 

Neo-Darwinism is a fully general theory in the sense that it can generate a potential explanation for any feature considered as an end-state. However, neo-Darwinism does not accommodate every possible series of successor states, because it allows only infinitesimal changes driven by selection.  For the examples in the preceding paragraph, the neo-Darwinian theory fails because our historical reconstructions implicate saltations or discrete jumps. 

---