During the flourishing era of modern molecular biology, the idea of the gene evolved away from chemical substance to non-material information that makes up the text of the “book of life” (Doyle 1997; Fox Keller 2000; Kay 2000; Rose 2007). This transition has started in the early 1960s and reached its (temporary) culmination at the beginning of the new millennium with the publication of the “human book of life”, i.e. the complete sequence of the human genome (Nerlich et al. 2002; International Human Genome Sequencing Consortium 2004). Around the year 2000 and during the following years, it was increasingly recognized that the metaphor of “information” was no longer appropriate and capable to cover how “life” and organisms act. Consequently, genetics and molecular biology turned their focus towards protein biochemistry, proteomics, and structural biology, as well as towards the analysis of the nature of metabolites (metabolomics) and their steady state concentrations and fluxes (fluxomics). Those disciplines emphasized the meaning of materiality of the substances involved as well as of its mutual interconversion, i.e., of three-dimensional structures, spatial relationships and molecular interactions. Concomitantly, researchers in these fields displayed less interest in information metaphors as manifested in linear sequences of letters and symbols for the characterization of DNA molecules (Jenner and Taithe 2000; Tanford and Reynolds 2001; Tyers and Mann 2003; Pappas 2006).

If one follows the historical outline of the information-driven conceptions of heritable phenotypic variation in genetics and molecular biology, it becomes obvious that a change in mindset is still underway. Changes often manifest themselves in the introduction of new metaphors. This is clearly established in genetics and molecular biology, already well developed in biochemistry, proteomics, and structural biology but only in its infancy in membrane and organelle biology. Now, times seem to be ripe for proteins, membranes, surfaces, organelles, in short “membrane landscapes” (MLs, see below for further explanations) to take over as the primary generators of metaphors and specific ways of thinking, researching, and writing about genes and DNA. Of course, this does not mean that genes and DNA will be abandoned as the “object” of scientific study. However, they may need to be examined from a new perspective, namely that of the proteome and – hopefully in near future – that of biological membranes and MLs. Under this new conceptual framework, genes and DNA will lose their meaning as the only representative and critical matter of inheritance.

It is a common place that changes in perspective and conception inevitably lead to the generation of new metaphors. In the following, we will try to show how the metaphors and narratives used in the biological inheritance discourse seem to be shifting from DNA to non-DNA matter. It is also important to note at the outcome that these shifts are not clear-cut breaks, radical transformations, or paradigmatic revolutions. But often, they emerge just as series of smaller steps that slowly undermine the previous foundation of certain metaphors, narratives and models, and pave new metaphorical and narrative paths that open up as a result of this erosion. Thus, DNA matter is not abandoned as “object” of study in the life sciences, just as “intention”, “purpose” and “meaning” have not been pushed aside in the historical sciences, but their status as the only representative and dominant generators of metaphors, narratives and models seems to be changing now.

The groundbreaking elucidation of the structure of the DNA double helix by James Watson and Francis Crick (Watson and Crick 1953) paved the path to the assumption that this macromolecule, which from a chemical point of view may be regarded as rather simple and boring, operated as the primordial substance of heredity in all living organisms. This assumption has become known among molecular biologists as the “central dogma of molecular biology” (Crick 1970). This dogma is essentially the belief that the genome of an organism, i.e., the entirety of its genes, fully explains the characteristic expression and specific combination of inherited traits, i.e., its unique “phenotype” (for comprehensive and outstanding discussions of the relationship between genotype and phenotype, see Chevin et al. 2022; de Vienne 2022; de Vienne and Capy 2022; Fisch 2022; Pontarotti et al. 2022; Robette et al. 2022; Shah 2022). It represents the foundation of a fundamental revolution that took place in molecular biology and genetics over the 20 years following the discovery of the DNA structure, which included the deciphering of the “genetic code” in 1961 (Nirenberg and Matthaei 1961). Based on this dogma, which is as simple as elegant and easy to understand, an attempt was initiated and seemingly completed to explain biological inheritance solely at the molecular level. Accordingly, the molecular substance of heredity is DNA, which is essentially a very long linear molecule consisting of only four different building blocks and is wound up in the nucleus of every cell in a strongly condensed fashion. Individual specifically “marked” sections of DNA form the genes that determine – independently or in combination with others – each of the inherited traits through a series of complex molecular processes, typically after the generation of RNA “transcripts” from the genes and then of proteins from the RNA “transcripts”.

This “DNA-centric” conception was further strengthened and expanded by a large number of important discoveries in the 1970s to up to the beginning of the next millennium. These were, in particular, the advent of recombinant DNA technology (Linn and Arber 1968; Jackson et al. 1972), the launch of the Human Genome Project (Stephens et al. 1990; Watson 1990), and the publication of the complete draft of the human genome (International Human Genome Sequencing Consortium 2004). From the 1960s onwards, these scientific discoveries led to the formulation of what is called the “information model” of life, which applies from the bacterial cell to the human body. This model was specially mediated via the metaphor of the “Book of Life”, a metaphor particularly prevalent in the rhetoric surrounding the “Human Genome Project” (Nerlich et al. 2002; Lorimer 2005; Rose 2007). Genes were conceived as a kind of digital manual for the creation of all organisms, including humans. While DNA itself is matter, i.e., a physical substance of spatial structure and specific materiality, the ways in which DNA was assumed to be responsible mainly or even exclusively in generating the phenotype has been shrouded in metaphors and narratives of code, semiotics, text, letters, errors, writing, reading and erasing. This view of the genome as an information system, as a linguistic text written in the DNA code, has guided the theories and practices of molecular biologists since the 1950s (Brandt 2005). It culminated in deciphering the “Book of Life”, a narrative that has developed a life of its own with great impact (Kay 2000, p. 325). This view was also manifested in a speech by former US President Bill Clinton on June 26, 2000, at the press conference announcing the publication of the working draft of the human genome.

The information model promises people the opportunity to read the “Book of Life”, an “opus” that until now has only been accessible and understandable to its writer or creator. And accessible not only to read, but also to rewrite it, as Hans-Jörg Rheinberger (2000) pointed to the fundamental change that recombinant DNA technologies have brought about. According to this conception, life is something that is “written” and “read”. It is basically a linear text, and it is a text that scientists can not only read but can also edit and rewrite. Even without a further outline of how complicated this conception ultimately is, and despite the shift of the focus in genetics and molecular biology to biochemistry and proteomics, the importance of the information model in efforts to identify the basis of inheritance for all vital processes, from normal function, physiology, and behavior to dysfunction, aging, and disease, cannot be overestimated. The information model still has significant narrative and metaphorical impact, as well as economic, political, and cultural influence. And no doubt, as one of its major merits, it has provided a consolidated conception about nature, life, heredity and organisms in the public and published opinion (Haraway 1997; Weber 2003; Rose 2007), to up to the point that it has created a kind of “genetic fetishism” (Kay 2000, p. 342).

The “DNA- / environment-centric” conception of inheritance

The central concept of an epistemology of the information model of inheritance is the so-called “gene-for-thinking”, i.e., the search for the one and only gene that codes for the corresponding different manifestations of life or for the specific differences between organisms, being it specific functions, unique for certain but no other organisms, such as production of a venom, different phenotypic characteristics, such as eye color or body size, or different (patho)physiological configurations, such as a special mental or physical performance or a disease. This way of thinking with its oppressive narrowness becomes manifested in the countless studies on the search for the musicality gene, the intelligence gene, the obesity gene, the alcoholism gene, the homosexuality gene, etc. While some scholars warned from the beginning against believing this rhetoric, the public imagination and not only that held on to it. But as genetics and molecular biology has evolved, an increasingly influential group of scientists hinted to the limitations of the information model. It has been increasingly recognized that, on the one hand, many genes do not code for a single protein or a protein at all and, on the other hand, many (patho)physiological processes are determined by a multitude of interacting proteins, i.e., by a complex network of proteins, rather than a single protein.

Despite this, or eventually because of this, genetic sequence data are still growing exponentially across all organisms, from bacteria to humans. Google’s massive investment in the personalized genetic data service “23andme.com” is a perfect example of the scientific, economic, but also emotional significance of such data for people. Scientists repeatedly emphasized that these versions, or rather versions of genomes from different organisms, are based on a simplified and reductionist understanding of the gene. Such sequence data seem to contain more data than they actually do and promise existential meaningfulness and impact that they do not and cannot provide. The transformation of DNA sequences into products, i.e., transcription, translation, post-translational modification, subcellular targeting, is controlled by many intrinsic and extrinsic factors and is also extremely context-dependent, i.e., is decisively influenced by the neighboring sequences. This contradicts DNA sequencing’s belief in the explanatory power of genomic sequences, a belief based on a relatively simple match between genes, structures, and functions. The experimental evidence is overwhelming that a simple direct relationship between genotype and phenotype does exist only in exceptional cases. Over the past five decades, life has been transformed from the simple deciphering of a molecular (i.e., genetic) code to the mysterious transformation of the one-dimensional, linear, and comparatively simply structured genome into the three-dimensional networking and very complex system of proteins, membranes, subcellular structures, cells, organisms and bodies. Nevertheless, the following groundbreaking experimental findings and milestones in biochemistry, microbiology, genetics and molecular biology, starting in the 1950s and often awarded with the Nobel Prize, have initially been interpreted under a view which adhered to the information model:

(i)

In the early 1960s, the Danish biochemist Christian Anfinsen described the successful re-folding and restoration of the hydrolysis activity of the enzyme ribonuclease following its complete denaturation and unfolding and the resulting total loss of a defined three-dimensional structure and its enzymic activity (Anfinsen and Haber 1961).

(ii)

In the early 1960s, the American biochemist DLD Caspar and Aaron Klug (1962) together with the virologists Heinz Fraenkel-Conrat and Robley Williams (1955) demonstrated the reconstitution of infectious tobacco mosaic virus particles in vitro from their purified constituents, the capsid protein that constitutes the viral envelope and the single-stranded RNA trapped within the envelope. This process has been called “self-assembly” because no molecules are involved and required that are not themselves components of the newly assembled functional biological structure. Thus, the viral capsid protein and RNA are necessary and adequate (i.e. causally sufficient, see below) for assembly of the infectious tobacco mosaic virus particle.

(iii)

Subsequently, the principle of “self-assembly” was demonstrated for the “spontaneous” formation of a number of protein-nucleic acid complexes, including the heads of DNA bacteriophages (King et al. 1973; van Driel 1997; Rossmann et al. 2004), the subunits of ribosomes (Venema and Tollervey 1999), as well as specific protein-ribonucleic acid complexes, such as the signal sequence recognition particle (Massenet 2019). In some of these cases, however, the functional assembly was crucially dependent on the proportion of the individual constituents and / or their emergence along a defined temporal sequence and spatial arrangement and / or on the support of “scaffolding proteins”, such as chaperons, which are not contained in the final functional structure (Seth Horne and Grossmann 2020).

(iv)

In the 1950s, Robert Briggs and Thomas J. King (1960) succeeded in the first transfer of isolated nuclei (from skin fibroblasts) into embryonic cells (“enucleated” oocytes of the frog Xenopus laevis). Subsequently, in the 1960s, John Gurdon and coworkers (Gurdon et al. 1958) were able to demonstrate the complete reprogramming of the transplanted cell nuclei (from intestinal epithelial cells) upon transplantation into “enucleated” oocytes of Xenopus laevis. Thus, pluripotent stem cells (PSCs) were apparently obtained from differentiated cells which managed to differentiate into all cell types of an adult organism (Gurdon and Laskey 1970). During subsequent decades this pluripotency was demonstrated for mammalian organisms, including mice (Hochedlinger and Jaenisch 2002) and sheep (Wilmut et al. 1997).

Under the impression of the still overwhelming genetic information model, those data were interpreted in perfect agreement with it on the following basis: The information for (i) the three-dimensionality of proteins (i.e., their tertiary structure), as well as for (ii) and (iii) the assembly of (multi-subunit) protein complexes or protein-nucleic acid complexes exhibiting a defined quaternary structure, and for (iv) the ontogeny of organisms are all ultimately determined by the structure and function of polypeptide chains which is critically and exclusively determined by the linear amino acid sequences, with the “primary” information for the protein structure and function solely encoded in the corresponding genes. What else should be required for the development of all the structures and functions in cells, organisms and even the human body, with its huge variety? The consideration of the involvement of non-DNA non-informational factors seemed to be superfluous, even if one takes into account biological membranes, cell surfaces, and skin for the demarcation of cell and organismal boundaries as well as the separation of interior and exterior environments.

However, along this period some system theorists have expressed great reservations about this interpretation of genetic information (for instance see Gray 1992; Oyama 2000; Griffiths 2001). They argued that such a distinction between genetic informational reasons and non-DNA non-informational reasons was not empirically justified and was founded on incorrect metaphysical assumptions. Moreover, the use of the metaphor “genetic information” has been criticized by numerous geneticists as well as philosophers of science (Fox Keller 2000; Nelkin 2001; Venville et al. 2006; Walsh 2020). This critique of the concept of “genetic information” has in turn been characterized by others as – at least in part – exaggerated (Wheeler and Clark 1999; Maynard Smith 2000; Sterelny 2000; Kitcher 2001). A deeper discussion into this complex discourse is not necessary at this point. Simply, from the assumption that the genetic causes of development (of heritable traits and phenotypes) are informational, it cannot be concluded that genetic causes are explanatorily sufficient and that non-DNA causes are to be interpreted only as an explanatory background. If somebody wants to explain the characteristics of a roast, the reference to the recipe, i.e., the relevant information used is usually not sufficient. A better explanation would be to indicate which ingredients have been used and in the quality of the ingredients. In addition, it should also be stated which cooking and frying utensils and equipment have been applied and eventually how they have been operated. And of no minor importance would be mentioning who the cook has been and what training, qualifications, or competencies he or she had. A roast cooked from low-quality products or by incompetent people using inappropriate utensils will typically not be of high taste, regardless of the type of or adherence to a recipe that has been put together per se for a good roast. Recipes are usually not sufficient to explain the characteristics of a roast. So why should “genetic recipes” be sufficient to explain “phenotypic cooking”? The metaphor of “genetic information” is not sufficient per se to justify the view that all non-DNA non-informational factors (materials as well as environmental conditions) in the development of phenotypic traits are merely explanatory background, as Wheeler and Clark (1999) have claimed.

In this context, it may be instructive to point out that Gregory Bateson (1976) introduced the metaphor of the phenotype as a cake and development as cooking. He used these metaphors to argue that (i) phenotypes arise as a result of complex interactions between genetic and non-genetic “ingredients”, (ii) genes are not explanatorily privileged, and (iii) quantifications of the causal contributions of genetic and non-DNA factors are not possible or useless. Are there any other justifications for the view that non-DNA, non-informational factors are not necessary in terms of an explanation for inheritance and development? In fact, one is to consider that only the genetic, DNA, informational ones are subject to natural selection among all the factors responsible for development and inheritance. Numerous, sometimes divergent arguments have already been put forward against this view that only genetic factors become exposed to natural selection (Griffiths and Gray 2001; Sterelny 2001). But even if only genetic factors were the target of natural selection, it would not “automatically” follow that non-DNA factors should not be explicitly included in explanations of the development of inherited phenotypic traits. Rather, development may be the consequence of the causal interaction between genetically selected and non-DNA non-selected factors. The motivation of many biologists to treat non-DNA actors as an explanatory background in the case of “like-from-like” phenomena (Müller and Müller, manuscript submitted) may be due to the “claim” that the explanatory foreground has to be engaged in the generation of specificity during development. Although both genetic and non-DNA actors are generally considered to play a causal and necessary role in developmental processes, only the former are considered to contribute to specificity (Raff and Kaufmann 1991). However, it should be stated that the use of the term “specificity” in connection with developmental processes and genetic information by biologists is more likely to be assigned to the realm of intuition.

Regarding the inheritance of biological traits, the distinction between a sufficient explanation, presented in the foreground, and a sufficient causality of factors, only sometimes given in the foreground but usually hidden in the background, implies that the causal effect of some of these factors is simply self-evident, since it is generally accepted, especially by the “scientific community” as canonical knowledge. Thus, the causal involvement of those factors no longer appears to be worth mentioning, whereas that of others does. The latter apparently holds true for DNA and genes, which are then “consistently” classified as explanatorily (if not causally) sufficient. By contrast, the former applies to non-DNA matter, such as proteins and membranes, since those are typically evaluated as neither explanatorily nor causally sufficient, but as causally necessary or simply self-evident .

Fig. 1

“DNA- / environment-centric” conception. During early development of the zygote (Z), produced by the parents, the somatic (S) and germ cell (G) lineages become totally separated. The entirety of S, which are copied by cell division and then become differentiated, creates the phenotype (P) of the organism. G which is also copied by cell division during adult life may ultimately lead to Z of the next generation upon cell fusion which will develop in the absence of any direct effect exerted by S. According to the “DNA-centric” conception, during the course of both cell division and fusion, DNA is the only matter to be transferred from the mother to daughter S and G, as well as from the parental G to the offspring Z, respectively. The “DNA-centric” conception has been supplemented with the impact of environmental factors (E) which induce specific changes on S, such as mutations, leading to differently differentiated and functional S* and creating an altered P*. DNA is thought to be the only matter to be transferred from S to S*, mother to daughter G and parental G to offspring Z along cell division and cell fusion, respectively, and to remain unaffected by E

It is noteworthy that over the past five decades – in addition to genes and depending on the inherited phenotypic trait – environmental conditions have successfully “struggled” for acknowledgment as factors of explanatory (and causal) adequacy for the development of phenotypes by both scientists and laymen. Accordingly, the majority – but not the entirety – of the heritable phenotypic traits and the differences between organisms are thought to be determined by the interaction of genes / DNA and environmental factors. Strikingly, this canonical DNA- / environment conception of inheritance (Fig. 1) continues to systematically exclude other factors, namely non-DNA materials, among them proteins and membranes and other subcellular structures and assemblies.

On the one hand, the “DNA- / environment-centric” canonical conception presents genes and environmental factors as explanatorily sufficient and causally necessary for the inheritance of phenotypic traits. On the other hand, proteins, membranes and others non-DNA cellular elements are presented only as explanatory background. How did this unjustified differentiation between explanatory foreground and background, causal sufficiency and necessity of DNA, environment and non-DNA elements come about? The description of the forces and motivations at the various societal, cultural, and economic levels underlying this obvious trajectory of the “DNA- / environment conception” of biological inheritance should be at the center of future Science and Technology Studies.

The “DNA- / epigenetics-centric” conception of inheritance

In epigenetic studies, in general, and epigenetic-environmental studies, in particular, dualisms and binary differentiations, such as organisms and environment or nature and nurture, are essentially dealt with. At the same time entanglements, indeterminacies and simultaneities become apparent when, for example, epigenetics is conceived as a “mediator” and as the “intermediary” or “in-between” genes and the environment (Leuzinger-Bohleber and Fischmann 2014; Subramanaiam 2014). The question therefore arises as to what the implications are when binary boundaries dissolve and shift. Does this possibly lead to something other than just a binary understanding of nature-culture, body-mind, gene-environment, body-surroundings, as materialist and feminist philosophers of science have repeatedly suggested? (Weber 2003). The problem with those binaries is that it typically remains unclear where the boundary between the apparently antagonistic entities is positioned. Regarding the gene-environment dichotomy, where does the action of genes, e.g., responsible for melanin production, become initiated and where does the effect of UV-light end, and vice versa. Putative answers are “at the level of the gene”, induction of melanin expression in melanocytes, synthesis of melanin protein, its transport to the cell surface, distribution to keratinocytes, protection against UV-light due to conversion into heat, and “at the level of UV-light”, its generation by the sun, propagation via the atmosphere, partial absorption by sun crème and clothes, stimulation of the synthesis of melanocyte-stimulating hormone, and induction of melanin expression. Thus, it is simply not feasible to exactly define the terms “gene” and “environment” and to simply delineate the boundary between them, as it is true for most binary systems.

Conrad Hal Waddington is usually credited as the one who introduced epigenetics into the areas of embryology and genetics in the 1940s (Squire 2017). The term is considered a neologism from “genetics” and “epigenesis” (Müller and Olsson 2003). It resulted from debates about whether preformation or epigenesis, respectively, provides the correct view for the development of organisms. From the perspective of preformation, developmental processes could be explained from what was already there before. New things, such as molecules, structures, organs, would develop from existing ones. This was based on the early idea that the complete organism is already contained in the egg or sperm in miniature form and the embryological development is based solely on its growth (Müller-Wille 2014). While preformation is associated with a more deterministic and reductionistic explanatory model, “epigenesis” is considered holistic because it focuses on the adaptability of organisms (Schuol 2016). “Epigenesis” thus focuses less on the pre-existing than on the ever-new emergencies in the embryological course of development as well as the interaction with environmental factors. Waddington drew on both perspectives. He developed an epigenetic understanding of development in concert with an explicitly non-preformistic genetics. Consequently, he emphasized the importance of genes as well as described environmental factors as one aspect among many factors influencing developmental trajectories (Waddington 1957).

Of particular importance is the concept of “epigenetic landscape” introduced by Waddington (1940), which refers to the process of ”decision-making” of cells or tissues for a developmental pathway. This represents a surface embedded in a multidimensional space of cellular metabolism (Slack 2002). There are also illustrations that show the underside of the epigenetic landscape. There, the supporting pegs of the surface represent genes that together form and shape the landscape. Depending on the nature of the landscape, a sphere moves on it and can take different paths until it reaches its destination, the fully differentiated tissue (Waddington 1940).

For Waddington, the ability of cells and tissues to respond to an impulse or signal from the environment, as well as the “canalization” of development through numerous branched decisions are still under the control of genes (Slack 2002). Waddington was less interested in the specific triggering signals (‘inducer’), but above all in the fact that cells react or can respond differently to signals, making different developmental pathways possible (Waddington 1968). The epigenetic landscape stands then as a dynamic system in which cells and tissues react differently to signals, taking multiple developmental paths in different regions. At the same time, there is no arbitrary development but a limited number of possibilities, and everything remains genetically controlled. Depending on the level of development of the organism and action of environmental stimuli, genes shape the epigenetic landscape. Thus, what happens is not genetically predetermined. Consequently, a hard gene determinism or “DNA-centric” conception seems to soften. It becomes clear that Waddington opened up to the complex of environmental influences. At the same time, he adhered to the idea of a program according to which development takes place and does not completely neglect the conception of the gene as determinant. With the concept of canalization, Waddington made it clear that the course of development depends on numerous factors and that evolution selects canalized developmental trajectories, i.e., epigenetic trajectories that show some resistance to being changed, which are also genetically controlled. However, since development is constantly being redefined, it is nevertheless not predetermined. In our view, although Waddington acknowledged influencing factors and environmental influences, he remained within a “DNA-centric” conception and his model of ’canalization is therefore a “DNA-centric” one. In addition, Waddington’s ideas expanded the concept of heredity, as he stated: We need a heredity system which does not merely contain information, but which acts as algorithms or programmes and thus leads to the production of a phenotype which takes its place between the genotype and the environment. It is the phenotype which acts on the environment (for example, in metabolism) and it is on phenotypes that the environment exerts its natural selective forces. (Waddington 1968).

Waddington also used the term “program”, which assumes a fixed sequence of developmental steps (not to be mixed up with “open programs”, which may result in different phenotypes under different environments). At the same time, he emphasized the interactions between genotype and environment, which he described as nonlinear, but ongoing in different directions. A famous example of this interplay is the formation of skin calluses on the chest of ostriches. Waddington assumed that calluses had developed on the chest of the ostrich ancestors in response to environmental factors in the developmental system. Since it proved to be an advantage to sit on hot and rough ground, these have been preserved (Waddington 1942). This implies something that is still characteristic of epigenetics today: Environmental factors have led to the formation of skin calluses during life, and the descendants were born with these, although they have not longer been exposed to those environmental factors. It is common knowledge that most, if not all, organisms share the capability to modify their specific characteristics in direct response to altered environmental cues. This phenomenon has been termed phenotypic plasticity. Waddington’s explanation for this was what he called “genetic assimilation”, the process “by which a phenotypic character, which initially is produced only in response to some environmental influence, becomes, through a process of selection, taken over by the genotype, so that it is formed even in the absence of the environmental influence” (Waddington 1961). Waddington recognized that a specific phenotype which had been triggered by a given environmental cue may acquire a constitutive nature, and be maintained during normal environmental conditions, even after termination of the initial environmental cue, provided selection for organisms had continued for several generations which display highest susceptibility for expressing this phenotype upon stimulation. He correctly argued for an increase in allele frequency as consequence of the selection leading to improved, i.e., more reliable and consistent expression of the altered phenotype (Waddington 1953). This hypothesis has further been supported in the course of experiments preformed with inbred stocks missing genetic variation which were resistant toward genetic assimilation in response to selection (Scharloo 1991).

In this sense, phenotypic plasticity is lost, i.e., development is canalized, through natural selection, which drives the fixation of genetic factors that were either already present, but cryptically and in low frequency in the original population, or produced by de-novo mutations. In other words, driven by the process of “genetic assimilation” selection apparently manages to abrogate this environmental sensitivity, leading to fixation of a pr