crazyben

What a fucking ridiculous year it turned out to be... Over the summer I had three bizarre incidents happen a) because of my bone condition several of my teeth became super weak for no good reason and I had to pay $2000 to have them removed. My mother is insane however and my thesis advisor paid instead! If she wasn't such an awesome thesis advisor/surrogate mother I don't know how I would've gotten things like this done; b) two weeks later I got some crazy infection and had to rush to the ER the first week of classes! If you're a grad. student you know how much it can mess you up to miss a single seminar hour. It was insane, I was on the same antibiotic that's used to treat Anthrax for two weeks AND was prohibited from both drinking alchohol and consuming dairy products! For a latte-swilling caffiene freak who also drinks liquor by the barrel it was tough times yo'; c) I finally found out why my knee always hurts when I hobble around like Dr. House, turns out my knee has become arthritic since I friggin' broke my kneecap. No wonder I still take pain killers. WTF. The doctor said "I would plan on not walking a mile again without substantial assistance." So that's that. I always knew that with my bone condition, which only affects like 1/200,000 people in worldwide, that a crazy freak injury that totally screwed me over for life COULD happen. Up till this one night at a cheap poolhall that served pitchers of Bluemoon for like $6, I hadn't actually ever broken a bone such that I just didn't recover. Anyhow, I'm like Dr. House now, hobbling around, saying crazy things to people (oh wait I've always done that...) and making sure I ace my graduate courses.
I just applied to six PhD programs in Ecology and Evolutionary Biology (spent like $1200 on the application fees and transcript fees etc all told) and will hopefully finish my M.S. this summer and move somewhere nice and cold for my PhD studies in August. The only good news is that a) I recently got to sleep with this hottie that I've been playing around with for like 6 months and b) more important to my scientific career, 2 of my papers are under peer review (1 was recently rejected with a 'revise and resubmit' option which is friggin' awesome, it hardly ever happens, usually they give you the acadamese equivalent of "fuck off, and die"). I've got 2 more to get churned-out in the meantime. Here's the couple that are under review. I didn't bother including the citation section because this is such a long blog but I will be happy to e-mail a full copy to anyone who so desires.

This first paper is at review for a journal called "Biology & Philosophy":

"Operationalizing the Selfish Meme" by Benjamin E. Hardisty-graduate student, Theoretical Evolutionary Biology
hardisty@mail.usf.edu

Introduction
In this paper, I assert that Dawkins' memes will be detectable in a larger sense than merely that they should lead people to build new types of buildings, believe in the existence of a higher power, or write blank verse poetry. I argue that memes should affect genotypes. Any memes that are successful at changing human behavior should be detectable first and foremost by the changes in fitness accrued to their bearers. In the first part, I describe the concept of memes as elucidated by Dawkins in his 1976 classic, The Selfish Gene. I also go over three fundamental properties that are hypothesized to make memes successful agents of cultural evolution: longevity, fecundity and copying fidelity (Dawkins 1989). I then outline H.C. Plotkin's work on memes and the evolution of culture (Plotkin 2003). Plotkin (2003), an evolutionary anthropologist, has done excellent work on the problems with Dawkins' conception of memes and I briefly describe his analysis and criticisms of the concept. I then outline the work of Plotkin (1988), Dunbar (1988), Dunbar et al (1999) and Hurford (1999) on the role learning is expected to play in the evolution of behavioral and culture-genetic coevolution. Evolutionary theory predicts that when it is beneficial to an organism, an organism will develop the ability to learn (Mayr 1963, 1974; Dunbar 1988; Dunbar et al 1999; Hurford 1999; Plotkin 1988; Plotkin et al 1979). This pressure has been hypothesized as one of the reasons why many species of social animals have the ability to learn in the first place (e.g., Mayr 1963, 1974; Dunbar 1988; Miller 2000). The work of Cavalli-Sforza & Feldman (1981) is notable as an example illustrating my contention that memes should be detectable due to their effects on fitness (and hence genotypes). Their work has shown that cultural change can drive biological evolution under some conditions (the ability to digest lactose being one example: Feldman & Cavalli-Sforza 1989; Durham 1991). Finally, I conclude the discussion with some relevant examples of memes that have affected biological evolution in our species.

The Meme Concept
In The Selfish Gene Richard Dawkins proposed the existence of an analogue of the gene he called the meme (Dawkins 1989). A meme is a unit of cultural inheritance, whereas the gene is a unit of biological inheritance. He stated that the meme is the only new and novel kind of replicator on our planet than the gene. Some examples that Dawkins gave of memes are: tunes, ideas, catch phrases, clothing fashions, tool designs and the construction of arches. Memes spread by parasitizing minds and using them to leap from mind to mind. One example: a scientist hears a good idea from another scientist and mentions it to his students and colleagues. If the 'good idea' catches on, it spreads from mind to mind indefinitely. As with genes, some memes may propagate for a long, long time and others may die out very quickly in the struggle between ideas. Dawkins goes on to claim that memes should be regarded as living structures, both metaphorically and literally (Dawkins 1989).
This isn't just a way of thinking about cultural evolution though: memes are actually supposed to be found in the brains of human beings (Dawkins 1989). Take the idea of God, for example: It is a very old idea. But how does it replicate itself? By spoken and written word, by becoming embodied in great art and architecture. What is its survival value? Or, what is it about the idea of a God that keeps it perpetuating itself? Well, Dawkins tells us that it probably makes people more comfortable with life. But Dawkins reveals something more here than a novel theory of how culture could be born and replicate some (or perhaps all) of its most interesting features. In his view, the fundamental property of genes is not that they code for the production of different proteins that influence an organism's ontogeny, but that they are self-replicating objects. Once an object can replicate, that's all it needs in his view. He states that, "We biologists have assimilated the idea of genetic evolution so deeply that we tend to forget that it is only one of many possible kinds of evolution (Dawkins 1989: 194).
Imitation is one way memes can cause themselves to be replicated (Dawkins 1989), but that is not all a meme can do. Some memes are better at replicating than others (Dawkins 1989), or getting others to take notice of them. There are three essential properties that all memes must have: longevity, fecundity and copying-fidelity. Longevity isn't as important as fecundity or copying-fidelity, as we might be more interested in just what makes some so long-lived vs. others that die out very rapidly. Fecundity is the rate of spread of a meme. The examples Dawkins gives are how many people one might hear whistling a certain tune and how many scientists cite somebody's idea after it's first published. Copying-fidelity, the third property of a successful meme, is a little harder to pin down because it doesn't seem to be quite the same thing as the copying-fidelity of genes, where a gene either copies properly or doesn't work. This does bring to mind the question of just what constitutes a copy of a meme. Dawkins answers that it is any discrete unit recognizable as embodying the whole idea: for example, the "Ode to Joy" in Beethoven's 9th Symphony may be considered as the meme that is Beethoven's 9th Symphony, but for others, it may be a whole movement of that work. Dawkins does suppose that memes compete with each other, but it is not with each other per se, rather they compete with each other for space in our brains. Dawkins further speculates that some memes may become cross-linked with other memes and their selection may be primary or secondary, like some gene complexes that control for a whole range of phenotypic attributes (Dawkins 1989).
Dawkins ends his discussion of memes by noting that although genes are halved in every generation, in the absence of effects like meiotic drive, memes can continue to be transmitted through the ages. In this sense, memes may be the immortality many seek in their genes. Memes may be the way we overcome our genetic predispositions (Dawkins 1989).

Learning and Evolution
If memes exist, and have any explanatory value as units of cultural inheritance, it will be because they act as selective forces that influence the survival or reproductive success of an organism. For example, the work of H.C. Plotkin on behavior and how it affects phenotypes (Plotkin 1988) and the work of Odling-Smee and others on the theory of "niche-selection" (see Day et al. 2003; Laland et al. 1999; Odling-Smee et al. 2003) all illustrate this idea with a profusion of examples. The work of Plotkin has been focused for some time on understanding gene-culture coevolution (Plotkin 1988, 2003). In Plotkin (2003), he discusses memes extensively. Plotkin says that part of the problem with memes is that they are seen primarily in terms of meanings and as being embodied in artifacts and their connection, if any, with biology is not a bother to memeticists.
"Learning" is a generic term that refers to the acquisition of information or knowledge of salient features of an animal's world and its relationship to that world by an individual animal, the storage of such information and its integration into already existing behavior patterns such that it may alter the future behavior of that animal (Plotkin 1988). It is no surprise that psychologists, behavioral ecologists and ethologists all study learning in one form or another. Plotkin believes that when learning is present it is a powerful device by which individuals exploit and adapt to their environments. The power and efficacy of learning stems from it's adding a feedback dynamic that supplements an animal's innate behavior. Plotkin himself is most interested in how the context of how such learned behavior plays a causal role in the evolution of the population to which the individual learners belong. If learning is a phenotypic trait that has evolved, then the assumption of its selection will be crucial to a full understanding of learning. Plotkin is thus interested in discovering in just what ways learning is a causal force in the process of natural selection (Plotkin 1988).
According to Plotkin, there are four types of learning that should be expected to have some effect on evolution (Plotkin 1988). The first one is what he calls "learned behavioral adaptation." As is known to ethologists, learning is adaptive. Plotkin calls this the most common view of the role of learning in evolution and believes that it is closely related to the generally accepted notion that learning is an evolved trait or set of traits. Learning can be seen as being ascribed a conservative, indirect role in evolution, no different from the roles of other adaptive phenotypic traits that increase fitness. In view of this, it is no surprise then that
Sommerhoff (see Sommerhoff 1950; Sommerhoff 1969) viewed learning as being "directional correlation." This means that, ideally, we should like to describe a flexible relationship between a phenotype and some particular aspect of its environment, with the conjunction of the two being the goal of any adaptation (Plotkin 1988). Sommerhoff realized that we could characterize adaptations on the basis of time-lag (Plotkin 1988). What Sommerhoff (1950, 1969) called long-term directive correlations are phylogenetic adaptations, and these are what furnish "unlearned" behavior (i.e., flexible learning behaviors). Short-term adaptations are things such as instincts that spatially orientate an organism (or its reflexes) (Plotkin 1988). Medium-term directive correlations are called "ontogenetic" adaptations. They involve adjustments to things that are stable relative to the organism's lifespan and ecological niche, that the organism must take account of (weather, for example) (Plotkin 1988).
The second type of learning is called "the exploitive system" (Plotkin 1988). This type of learning refers to the ability of some animals to select (choose) the environments in which they will live, these choices having an effect on the selection pressures to which the animals will be subjected. For example, a bee flees a noxious object. Or a forager encounters a new type of food item or a new type of predator. Such learning is concerned with how animals might use innate behavior that is supplemented with learned behavior. For example, one might study how a bird's memory allows it to map food sources so as to forage efficiently. Or one might study how an animal's innate use of biting or grasping objects is extended to use in novel situations. If this type of learning really exists, and affects evolution, we might expect to see reproductively successful individuals of the same species of learners being subjected to novel environmental pressures, and the potential for small groups of such individuals to become isolated breeders. As some evidence for this idea, Plotkin cites the research of Wyles (Wyles et al 1983) (Plotkin 1988).
The third category of learning is one of the best studied: mate-choice learning. In organisms that reproduce sexually, individuals that are reproductively successful contribute directly to gene frequencies of their gene pool (Plotkin 1988). Thus, of all behaviors mating is the most direct determinant of the constitution of the gene pool (Plotkin 1988). Mating behavior is thus the behavior of the phenotype that is most closely related to what could be called "the quintessence of evolution": gene-frequency changes in the gene pool of a breeding population (Plotkin 1988: 148). Mate choice is assumed by biologists to serve various functions: increasing access to resources, retaining adaptive attributes of the local environment, enhancing fitness and maintaining optimal inbreeding-to-outbreeding ratios. The idea behind studies of mate choice is that when mating behavior deviates from what would be expected on the basis of chance alone, and that deviation is due to processes within the animal, then mate choice can be assumed to be operating. Thus, if some form of learning influences mate choice then it may constitute the most direct causal role of learning in evolution. How mate choice might be expected to affect evolution is as follows: The constitution of the gene pool at time t1 results in orders for the construction of phenotype p1 having characteristics c1, including certain types of learning (in this example, mate choice learning) (Plotkin 1988: 152). If these forms of learning bias mate choice over a number of generations, then this will result in a constitution of the gene pool at some future time, t2 that will give rise to some phenotype, p2, with characteristics c2. Presumably c2 will be at least a partial result of mate choice learning in generation c1. Hence, we can conclude that mate choice learning should affect evolution.
The final category of learning that might affect an organism's evolution is broadly called "social learning" (Plotkin 1988). Amongst biologists and evolutionary anthropologists there is widespread agreement not only that human culture is a product of evolution but also that human culture is itself an evolutionary process. Much like Dawkins, Plotkin sees that in both biological and cultural evolution, variants are generated and selection operates to winnow out unfit variants. The fit ones are then propagated by some kind of transmission process to descendent individuals (Plotkin 1988). R.I.M. Dunbar (1988) argues that social learning may be an important selective force. In his view, animals are social precisely because they can then solve some problems more efficiently than they could if they were solitary. Some functions of social living are to provide defense against predators (see Estes 1974; van Schaik 1983; Dunbar 1987), aid in the rearing of offspring and improving foraging efficiency (e.g., social carnivores can bring down very large prey that they otherwise could not) (see Kruuk 1972; Schaller 1972). Group living adds costs that animals need to offset, particularly lower ranking animals in hierarchical social groups (Dunbar 1988). His conclusion is that coalitions, for example those amongst gelada baboons and honeybees, are the solution to some of those problems, particularly the problem of reproduction. The formation of coalitions means that individuals must have some way of evaluating social standing, which Dunbar calls social learning. Animals that can manipulate the social terrain of their group more efficiently should have higher reproductive success relative to others within their group of conspecifics because they will be better able to procure access to resource (often attractive to females) and females (themselves a resource). Therefore, social learning should be selected for in many social organisms (Dunbar 1988), in accord with Plotkin's view (Plotkin 1988).
Coyne & Orr (2004) mention a speciation pressure called behavioral, or ethological, isolation, which refers to species differences that reduce the attraction between conspecifics during mating season. It is limited to animals and involves correlations between traits in different sexes. Typically, one sex (usually male) sends a signal for a preference for a certain trait to the other sex. Experiments to detect this type of isolation are usually testing for what is called "female-choice," "male-choice" or "multiple-choice." The old assumption was that it often involved only female-choice but it is now known that male-choice exists as well. Demonstrating behavioral isolation is often easy, but it can be very difficult to determine which specific traits are involved. Learning would be expected to play some role in the recognition of such traits, regardless of which type of choice is operating (Coyne & Orr 2004).
Dunbar et al (1999) have assembled some interesting articles on various aspects of culture and their evolution. The evolution of language, or the ability to learn to use a language (a neural and physiological ability) and their use as the dominant purveyors of information in human culture is discussed by Hurford (1999). Hurford notes that in any environment there will be scope for some variation in behavior that has little or no effect on fitness of itself. In relation to language ability, he states the view that it is possible for many small preadaptations to accumulate. He lists several proposed preadaptations that are supposed by experts in various subfields of linguistics and biology to have sped-up the evolution of language use (Hurford 1999). In lieu of Hurford's discussion, we should keep in mind that in some theories of mate choice (e.g., Miller 2000), initial selection for males that communicate more effectively, or even just fancifully, compared with other males, will contribute to "run away" selection that, if the preadaptations are in place, should lead to full-blown language use by all humans in just a few short generations (Miller 2000). Unsurprisingly, it is suspected that human language both evolved and spread very rapidly (Miller 2000). This should be another prime case with which to test the idea of culture affecting substantial genetic alteration.
Speaking of humans, Cavalli-Sforza (1994) says that the use of language greatly increases the efficiency of learning and is what forms the foundation of human culture. Perhaps it has been a more important force in the spread of humans and their survival because it has allowed humans to adapt to and master their surroundings in a very short time. Throughout our evolution, Cavalli-Sforza argues, it is language that has given modern humans much of their advantage over other species and made possible the complexity of our knowledge today. Language is an innovation that involves both biology and culture. It is clearly the result of selection acting on anatomy and physiology simultaneously. While language is a cultural creation, the ability to speak requires a precise anatomical and neurological foundation. What we call culture, or the ability to learn from the experience of others, relies on communication. The speed and precision of communication and our ability to memorize what we learn are factors that govern the efficiency of culture. However, it is not enough that culture merely exists for it to be useful from a biological viewpoint. Some examples demonstrate its usefulness however. Culture enables us to accumulate prior discoveries and helps us profit from experience transmitted by our ancestors- knowledge that we simply would not have access to on our own. Until the invention of writing, the accumulation of knowledge was limited by human memory alone but today that limit is gone (Cavalli-Sforza 1994).
One may ask at this point if there aren't some memes that create social customs that are deleterious to human health. Cavalli-Sforza says yes, but selection generally tends to create and maintain customs and institutions with social utility (Cavalli-Sforza 1994). Every day, we face choices that may be trivial or may affect us for years. We can consider these as "cultural selection." But unlike natural selection, which chooses between the best adapted individuals of a species, cultural selection proceeds through choices made by individuals. Ultimately, natural selection will still operate since it works on the cultural choices we make as well. If our choices help us reach maturity and reproduce, then our cultural decisions (as well as our biological predispositions) that generated these particular choices will be favored by natural selection. Therefore, we may deduce that every cultural decision will be favored by natural selection if it affects survival and reproduction, creating a positive correlation between these two forms of selection. Thus, memes face two levels of selection: the cultural and the biological. At the cultural level, a new idea must be accepted and then spread. At the biological level, however, that new idea will affect an individual's behavior, which selection will then operate on (Cavalli-Sforza 1994).

Cultural Change and Human Biological Evolution
Cavalli-Sforza & Feldman (1981) demonstrated that "cultural" traits can become genetic traits in their work on the diffusion of cultural innovations. In their view, it is extremely difficult to quantify the advantage conferred by the presence of a given behavior over its absence. For example, he notes that it often seems possible to assign to many of the enormous variety of behaviors in man and animals a precise adaptive meaning. Yet for others it may be difficult or even impossible to discover one. For example, shaking one's head from side to side means "yes" in much of India and "no" in Western societies. As another example, tail wagging has a different meaning in dogs and cats. Yet these facts in and of themselves do not contradict the notion that behaviors like tail wagging are acquired through natural selection, if we assume that the common adaptive feature in cases like this is the communication of emotions . Cavalli-Sforza and Feldman thus warn us that when considering the adaptiveness of a behavior, it is essential we distinguish between the general capacity to learn and any specific manifestation that might be the result of an interaction of genetic tendency and environmental factors (Cavalli-Sforza & Feldman 1981).
Cavalli-Sforza & Feldman (1981) arrived at their conclusions by studying the dynamics of changes within a population of the relative frequencies of forms of cultural traits under defined behavioral interactions. The studies had the goal of revealing the laws of transmission of cultural traits among individuals of the same or different generations and the resulting changes in frequencies over time that followed. Here the term "cultural" means any traits that are learned by any process of non-genetic transmission. For Cavalli-Sforza, "The most general conclusion to emerge from these studies is that, for animals that can learn, and especially for humans, it is difficult to partition the process of transmission into purely genetic and purely cultural components" (Cavalli-Sforza & Feldman 1981: 8). The important point here is that cultural transmission is capable of stimulating genetic transmission (Cavalli-Sforza & Feldman 1981). I argue below that how social learning affects evolution is how we will capture memes in action, or infer their past selection.
Cavalli-Sforza and Feldman refer explicitly to Dawkins' conception of meme when they state that the concept has its strongest support when dealing with traits that have some discontinuity between them (Cavalli-Sforza & Feldman 1981). They note some commonly studied discontinuities: atoms, elementary particles, genes, and DNA. Some traits are best discerned as perceptual thresholds (Cavalli-Sforza & Feldman 1981). As mentioned above, Dawkins also thought some memes would represent ideas or behaviors that would be similar to perceptual thresholds, for example Beethoven's 9th Symphony (Dawkins 1976).
All of the models referred to by Odling-Smee et al (2003) assume that cultural processes can modify both the selective environment to which humans are exposed and can affect which genetic variants that humans are exposed to. There is no assumption about causality and gene-culture interactions are viewed symmetrically (Odling-Smee et al 2003). In particular, these methods have often been an attempt to operationalize units of cultural transmission. The strength of the gene-culture coevolution approach is that it assumes that some human cultural activities cause feed-back that modifies selection pressures. This means that cultural transmission may affect the fate of some human genes (see Feldman & Cavalli-Sforza 1976; Cavalli-Sforza & Feldman 1981; Boyd & Richerson 1985; Durham 1991; Feldman & Laland 1996). Thus, the most relevant aspect of human selective environments are cultural in nature. In fact, some studies have shown that cultural transmission can cause extremely fast genetic responses in humans (see Dwyer et al 1990; Grant & Grant 1995a, b; Reznick et al 1997; Thompson 1998; Ehrlich 2000; Kingsolver et al 2001).
One less well-known example is that of the link between yam cultivation and the mutation of the sickle-cell allele in West Africa. In the case of the Kwa-speaking yam cultivators in West Africa, the frequency of a hemoglobin allele that causes sickle-cell anemia is believed to have increased as a result of the indirect effects of yam cultivation (Odling-Smee et al 2003). These people traditionally cut clearings in the rain forest to grow their crops, creating standing water and increasing the breeding grounds for malaria-carrying mosquitoes. This, in turn, intensified selection for the sickle-cell allele, because of the protection it offered against malaria in the heterozygous condition. Simply plotting the cultural practice of yam cultivation against the frequency of occurrence of the sickle-cell allele is not sufficient to demonstrate the relationship between the cultural practice and allele frequencies but the more sophisticated methods yield the correct result. A cultural behavior, or meme, caused Kwa-speakers to plant lots of yam, thus contributing to an increase in mosquitoes' breeding as a result and the incidence of malaria increased. This selected for a genetic mutation that confers immunity against malaria for most, but a horrific death occurs for a few. This is not a complex illustration of the idea of niche construction as a selective agent, nor even of cultural behavior as a selection pressure. The relevant meme may be as simple as "yams are yummy when roasted" or "yams are easy to plant," but it is a clear example of a cultural behavior, a certain style of agricultural production, and genetic evolution. Those who aren't resistant to malaria either develop a resistance or die. Perhaps a cultural response, such as developing some effective anti-malarial remedy was possible, but human genes responded instead to the increased selective pressure that enlarged mosquito breeding grounds introduced (Odling-Smee et al 2003).
Cavalli-Sforza & Feldman's modeling methods, as well as others (including Boyd & Richerson 1985; Durham 1991), have allowed investigations of the evolution of: lactose tolerance (see Feldman & Cavalli-Sforza 1989; Durham 1991), sickle-cell alleles (see Jackson 1996; Durham 1991), language (see Aoki & Feldman 1987; Aoki & Feldman 1989), handedness (see Laland et al 1995), incest taboos (Aoki & Feldman 1997), sex-ratio modifications (see Kumm et al 1994), hereditary deafness and sign language (see Aoki & Feldman 1991) and sexual selection with culturally selected mating preferences (see Laland 1994, Odling-Smee et al 2003). All of these may be construed, along with the examples of lactose digestion genes and sickle-cell alleles, to be examples of memes acting as selective pressures on human genes. Thus, we see that human cultural inheritance can influence human genetic evolution in two ways: 1) it can directly influence our differential survival and reproduction and 2) it can contribute to human niche construction, thereby enlarging our realized niche and increasing the fitness of our descendants by making our physical environment a better "fit" to our descendents than it is to us (Odling-Smee et al 2003).

Conclusion
We have seen that Dawkins' concept of the meme may be a useful tool for students of culture (Dawkins 1989), but also that it appears to lack the analytical rigor (e.g., Plotkin 2003) necessary to put it to any good use beyond ad hoc hypotheses as to the nature of the very meme asserted to be responsible for a given behavior or trait. Perhaps, as Plotkin (1988) argues, it is the crucial element of time lag that causes the directional correlations (e.g., Sommerhoff, 1950; Sommerhoff 1969; Plotkin 1988) that are so important in social learning and mate-choice learning. If so, then these areas would be prime grounds for testing different hypotheses about memes. As I have argued here, memes, though lacking proper operationalization, may yet provide a clue as to how cultural evolution proceeds by furnishing examples of phenotypically "plastic" traits that then contribute to genetical evolution by causing increases in an organism's fitness. As Cavalli-Sforza & Feldman's work (1981) has shown, as well as Endler's (1984 CITE), frequency trait distribution changes can be discerned through careful analysis. These changes may be instances whereby cultural evolution drives biological evolution, particularly in the case of animals that learn by imitation or other means (e.g., Cavalli-Sforza 1981, Cavalli-Sforza 1994).
The work of Odling-Smee et al (2003) on "niche-construction" may also provide further evidence that the biological underpinnings of behavior can give rise to phenotypic plasticity, which then gives rise to further biological evolution due to organisms modifying their own environments such that their fitness is increased. Further research will no doubt shed more light on the mysterious nature of how biology can drive cultural evolution and how culture can drive biological evolution. The small list of cases of cultural behaviors suspected of giving rise to genetic evolution attests to the strength of my conclusion that the best way to test how useful memes are will be when we can test their effects on our genome. If memes are to have any importance amongst biologists, let it be in the old proving grounds of the laboratory or the field.