The Hybrid Hypothesis
8: Closing Thoughts
|
Whatever we see could be other than it is. Whatever we can describe could be other than it is. There is no a priori order of things.
—Ludwig Wittgenstein
Tractatus Logico-Philosophicus |
Continued from previous section
Now, let’s consider the case of a pig crossing with a panins from the standpoint of genetics (a panin is either of the two apes assigned to genus Pan). Some might suppose that in terms of their nucleotide sequences humans are too similar to panins, and too dissimilar to pigs, for them to be pig-panin hybrids. But anyone who does is either unfamiliar with the effects of meiosis at the molecular level, or does not see the ultimate implications of those effects for the genome of a hybrid. The camouflaging effect of gene conversion in backcross hybrids was explained in Section 3 (see the green sidebar there entitled “Why it may not be easy to evaluate this hypothesis with genetic data”) and in the diagram immediately below, which also appeared
| Genomic effects of backcrossing: |
|
previously. But an additional consequence of hybrid meiosis, typically, is a jumbling of the genome in later generations. In hybrids, meiosis rearranges chromosomes, and duplicates or deletes them, either in whole or in part. In the present context, then, the expectation that such rearrangements would have occurred makes it more difficult to determine whether genes being compared in pig, human and panin are truly equivalent. To make this clear, it may be worthwhile briefly to describe what happens to chromosomes during meiosis.
In mammals, chromosomes are paired and vary widely in number from one type of organism to another. Each human cell contains 23 pairs — 22 matched pairs (autosomes) and one mismatched pair (the X and Y chromosomes). Pairing is important during meiosis, the process that produces spermatozoa and eggs. Germ cells are produced by cell division. At the beginning of each such division, each chromosome unites with the other member of its pair, a configuration called a tetrad. With chromosomes linked in pairs, the machinery of the dividing cell will distribute one member of each such pair into each of the two “daughter cells” produced by the division. When a tetrad is formed, the two homologous chromosomes composing it actually exchange DNA in a process termed crossing-over. Meiosis is stable under ordinary circumstances when organisms having the same karyotype mate (read some basic information about karyotypes >>). There is little or no tendency for the number of chromosomes to change from one generation to the next. Chromosomes are not subjected to forces that rip them apart, rearrange them, and reassort them. But the picture changes in hybrids, where chromosomes become highly volatile.
When mating occurs between organisms belonging to different chromosets (as is almost invariably the case with organisms treated as separate species), some or all of the chromosomes of the resulting hybrid will be either unpaired or inexactly paired. During meiosis, chromosomes lacking a match do not join to form a tetrad. No cellular mechanism exists to deal with unpaired chromosomes, so they pass into daughter cells at random. In this case, different daughter cells end up with different chromosome complements. Partially matched chromosomes unite to form partial tetrads and exchange lengthy blocks of DNA so that large groups of genes are transposed to new chromosomes. The chromosomes are radically altered both with respect to their genetic content and their overall appearance. (MORE ABOUT CHROMOSOMAL MUTATIONS)
| Since some or all of the chromosomes are partially or completely unpaired during this transition process, the mode of inheritance is non-Mendelian. Pairing is an essential presupposition of Mendel’s laws. |
Obviously, the germ cells produced by such mechanisms would vary widely in genetic content. Depending on the cross in question, a larger or a smaller proportion of these gametes, and the later-generation hybrids produced by them, would be inviable. The process of exchange, breakage, repair, loss, and reassortment would continue — in each gamete, in each individual, in every generation, as hybrids mated among themselves or with pure individuals of either of the two parent types — until a generation was reached in which all chromosomes were paired and meiosis became stable once again.
| It is often asserted that a given type of organism has a particular form and particular habits because it is constrained by the demands of its “habitat.” Natural selection is seen as a force that shapes every aspect of an organism to fit its environment. But the stable form created by the chromosomal mechanisms just described would, in large part, only reflect the underlying meiotic (and, thus, reproductive) stability of its karyotype. (this idea is discussed at length elsewhere on this site) |
So the karyotype characterizing any such emergent, re-stabilized population would be unlike either of the karyotypes present in the parents that initially crossed to produce it. During the unstable period, chromosomes would have been broken up and reconnected in new configurations as well as reassorted into a new karyotype composed of chromosomes and genes previously present only in separate organisms. If a new type of organism — for example Homo sapiens — emerged from a chromosomal mishmash such as that just described, its chromosomes would differ in structure from those of either of its two parents. Repeated crossing-over, breakage, repair, and reassortment would have created chromosomes in which genes from both parents were mingled. And, as it happens, human chromosomes have in fact been extensively rearranged in comparison with those of panins, a fact that has never been satisfactorily explained in terms of conventional theory.
Backcrossing
|
Evaluating the hypothesis with in vitro fertilization
I think it would be interesting to do in vitro fertilizations in order better to evaluate the chimp-pig hybrid theory of human origins. But I don’t know how many fertilizations you would have to do to actually reject the hypothesis, since some hybrid crosses take a lot of inseminations to produce a single hybrid. Consider, for example, the case of domestic hen (Gallus gallus) x capercaillie cock (Tetrao urogallus). The following is the information on that cross, cut and pasted from Handbook of Avian Hybrids of the World, my Oxford University Press book on bird hybrids: “Skjervold and Mjelstad (1992) artificially inseminated 25 hens with semen from 22 capercaillies. Of 1148 incubated eggs, eight embryos developed and three hatched. Surprisingly, seven of the eight showing development, and all of three that hatched and reached maturity, were sired by a single capercaillie. The hybrids had feathered legs down to the feet like capercaillies. When young, they sounded like chickens, when older, more like capercaillies. Sperm survival was drastically reduced for capercaillie semen in hen sperm storage glands (in comparison with that of roosters). Skjervold and Mjelstad say their results strongly indicate a genetic component in sperm survival and that treatment of recipient hens with immunosuppressive drugs would therefore be of interest. In an old report Schröder (1880) says several hybrids were reared. He also alleges that hybrids of both sexes were fertile in backcrosses to domestic fowl, to which, supposedly, the backcross hybrids were very similar.” There are several factors here that might complicate simple attempts at fertilization as a test of the chimp-pig hypothesis. One is that in some crosses many inseminations are required to produce just a few fertilizations. In the example given, the rate was only eight fertilizations resulting from 1148 inseminations, and, if anything, I would say a chicken and a capercaillie are more similar than a pig and a panin. So you might think the rate would be lower with a panin-pig cross, perhaps much lower. There is also apparently some unknown factor that allowed one male among the 22 capercaillie cocks to sire all offspring. So you would have to try lots of different pigs with lots of different panins before you could start to feel confident in your results. It might be that we are talking about a very, very rare hybrid that is very difficult to produce. And yet, if even one such was ever produced and it backcrossed, that single mating would be enough to get the whole process started. Then again, since no one has ever tried it, this might be a very easy cross to produce. So this method, I would say, would have greater prospects of confirming the hypothesis than of rejecting it. Because you would not be able to reach definitive conclusions by a simple failure to produce the hybrid. I know from the experience of looking at thousands of reports on hybridization that nothing is more common than someone failing to produce a hybrid and claiming it’s impossible, only to have someone else later report a success. |
Recall that a backcross is a mating of a hybrid with one of the parental types that originally crossed to produce it. If a hybrid backcrosses with an individual of one of those two types, then that parent's DNA will preponderate in the genome of the backcross offspring. If such offspring then backcross again, the progeny will be even more similar genetically to the backcross parent. And an even larger fraction of their DNA will be from the backcross parent. Moreover, as we have seen, even any DNA derived from the other, non-backcross parent becomes more and more like that of the backcross parent with each successive backcross due to the effects of gene conversion (if you need to refresh your memory on this point, please look at the diagram about backcrossing above, showing the genomic effects of backcrossing and read the green sidebar at right above, entitled “Why it may not be easy to evaluate this hypothesis with genetic data”).
In the case of early “humans,” if the initial hybridization were between a boar and a female panin (which, for reasons soon to be stated, seems the likelier possibility), the initial hybrid offspring would be born into a panin troop, which would increase the chances of the hybrid(s) backcrossing with panins. After an initial generation of backcrossing, the probability of additional backcrossing in later generations would be further increased because the backcross hybrids would be more panin-like and would again have grown up among panins. So they would be expected to have an even greater affinity for panins than did hybrids from the initial cross. And, as I have already emphasized, when a hybrid population repeatedly backcrosses to only one of its two parents, the DNA of that parent comes to predominate in the genomes of the hybrids. Long-term backcrossing over many generations can eliminate in the hybrids every genetic trace of the other parent. Of course, when the hybrid becomes genetically (genotypically) identical to one of its parents, it will also be indistinguishable in terms of its physical (phenotypic) traits.
Hybrids can maintain their existence as distinct entities only if at some point they can stop backcrossing and start mating among themselves. To do this, however, they would have to be more fertile than they do in the case of backcrossing. When two individuals of low fertility mate, the chances of producing progeny are much lower than when one such individual mates with a fully fertile parental-type individual. Moreover, in many crosses, hybrids will be of only one sex in the initial generation. In other crosses one sex among the hybrids is sterile. In either of these two cases, backcrossing is an absolute necessity. In some cases, the initial hybrids may be so rare that backcrossing is the only way to find a mate.
At the same time, however, fertility tends to increase in hybrids with each generation of backcrossing. When fertile, or partially fertile, backcross hybrids of both sexes begin to appear, the hybrid population has the option of breaking off and forming a separate (isolated) population. So long as they can produce in every generation a sufficient number of offspring to maintain the population, they can persist even though their fertility may remain poor for many generations thereafter.
In a pig-panin hybrid the chromosomes would be almost entirely unpaired. Meiosis would therefore be severely disrupted. There would be erratic segregation of the chromosomes into gametes, as well as extensive breakage, restructuring, loss, and reassortment of chromosomes — mass mutation. Genetic content would vary widely from one gamete to another. Under such circumstances, many gametes would not even have the genetic wherewithal to reach functional maturity (hybrids, especially ones derived from distant crosses, typically produce far fewer functional gametes than do nonhybrid individuals). Moreover, among those few gametes that did mature, even fewer would develop into mature organisms capable of producing offspring. So even if a rare pig-panin hybrid was actually able to find a mate of its own kind, such a pairing would be highly unlikely to produce viable offspring because, a mating between two individuals of low fertility has a small chance of success. The chances would be greater, if the initial hybrid backcrossed. And this supposition is consistent with observation since in a wide variety of crosses second generation hybrids are produced only in backcrosses, and not by matings among the F₁ hybrids.
The Direction of the Cross
| Rodolfo (470.15,16) notes “the extraordinary length of the genital tract of the sow and presence of obstructions, such as the sphincter and the rugate thick-walled cervix, call for special adaptations on the part of the boar for the deposition of the semen in a manner such that the spermatozoa may easily be carried into the Fallopian tubes where insemination probably takes place.” |
In the case of a pig-panin hybrid backcrossing would most likely have been with panins because the mother in the initial cross would, almost surely have been a panin. There are at least three reasons to reach this conclusion. The first is that a panin penis would probably be incapable of impregnating a sow, but a boar’s penis would be fully capable of carrying out the insemination process with sex roles reversed (Hill 1972, Rodolfo 1934). The second is that a humanlike hybrid would likely require a long period of nurture that a sow would not be able to provide. The third is that during estrus a pink sexual swelling appears on the rump of a female panin. Panin males do not attempt to engage in coitus, even with females of their own kind, unless this swelling is present (Goodall 1983, pp. 190-192; 1986). A boar, on the other hand, will mount any immobile object capable of supporting him, and will voluntarily ejaculate even into an inanimate tubular receptacle if it is of suitable diameter. “It does appear then as if, as far as the boar is concerned, coitus is largely a mechanical process” (Rodolfo 1934, p. 14, see also Sambraus 1990, Abb. 7; Walton 1952, p. 151). When threatened, panin females often attempt to appease the aggressor by crouching down and presenting their genitals.
A nubile little bonobo went exploring far from home,
It was the first occasion she’d gone seeking fruit alone,
When suddenly she heard a low reverberating snore,
That heralded a loud bellow and terrifying roar,
And as the startled simian was frozen still with fear,
A huge wild boar, aroused and mad, fell on her from the rear.
His mighty weight fair crushed her bones, his passion long frustrated,
So now he had his way with her, his ardour fully sated,
The traumatised young victim slowly climbed a jungle vine,
And soon was found and comforted by others of her kind,
Six months went by, in branches high, she bore a healthy daughter,
A special child, her shape and size was unlike any other.
The infant had her father’s eyes and long protruding nose,
With philtrum topping fleshy lips, both plump and red as rose,
And ruddy skin both damp and smooth, her head had soft brown hair,
Just like him she had trottered feet and full round derriere,
She had her mother’s fingers, although shorter and more blunt,
Her cry was an excited squeal, contentment drew a grunt,
As she grew the strangeling much intrigued an alpha male,
Until the amorous suitor’s firm intentions did prevail,
And so she bore a daughter who was very much like her,
Though walking round quite upright and with fair and sparser fur,
From whence a porcine heritage came down the female line,
The missing oink, our ancestor of pig and ape combined.
Thus, if an initial (F₁) hybrid was ever produced from this cross, its mother would almost certainly have been a panin — particularly given that a humanlike hybrid infant would require a mother that could hold it and nurture it during a prolonged period of development. The hybrid would therefore have grown up in a panin troop and, upon reaching sexual maturity, would have been in everyday contact with panins and would have thought of panins as its own kind. Many animals imprint on the animal that raises them and later prefer sex with animals of that type. The same might have been true of proto-humans raised by panins. I say “proto-humans” because any progeny of an initial cross between pigs and panins would have nearly half of their DNA from pigs and would thus be much more similar to pigs than are modern humans (the pig genome is about 10% smaller than that of a panin, so even in the F₁ hybrid, panin DNA would slightly preponderate). Only with successive generations of backcrossing would these piglike traits be reduced, as the backcross hybrids became genetically more similar to panins. An additional reason, then, to suppose that Homo sapiens is the result of backcrossing to panins is the observed preponderance of primate characteristics in humans. And the very fact that such backcrossing seems to have occurred suggests the mother in the initial cross was a panin: When the female in a hybrid cross raises her offspring among her own kind and in the absence of the father, the hybrid usually imprints on, and later seeks to mate with, animals of the same type as its mother.
The series of matings producing humans could perhaps have been completed in only a few generations, but could also have taken many, and have involved a complex mixture of backcrosses and matings of hybrid with hybrid. At this distance in time it’s probably impossible to reconstruct the exact series of events. However, for the reasons just stated, it does seem clear that extensive backcrossing to panins must have occurred. And yet, despite the homogenization of nucleotide sequences resulting from that backcrossing, the hybrids would not become physically indistinguishable from the backcross parent as quickly as their nucleotide sequences did. Under the hybrid hypothesis, this fact that humans have remained physically quite distinct from panins despite their extreme similarity in terms of protein and nucleotide sequences, can be easily explained.
Why are Humans Different from Panins?
Mclean et al. (2011) comment that “the genotypic basis of most human-specific traits remains unknown.” And such is still the case today (2022). Biologists say this because the degree of sequence similarity between humans and panins, which is greater than 98%, doesn’t seem big enough to account for the many obvious physical differences between the two. But under stabilization theory the development of an organism has little to do with point mutations at the nucleotide level. The crucial factor is instead the overall set of genes and regulatory sequences defined by its karyotype. Under that theory, differences between humans and panins would be due primarily to genes being present in different dosages, which would alter the mix of mRNAs and proteins produced during the various stages of development. The human genome, in comparison with the panin genome, would exhibit a shift in dosage toward the dosage seen in pigs, which would cause human morphology and behavior to become less panine and more porcine. Down syndrome is a well-known case in which a dosage difference produces a whole suite of altered morphological traits. No sequence differences exist in such individuals. The altered morphology is due solely to the presence of an extra copy of chromosome 21 (and therefore of an extra dose of the unaltered genes that that chromosome contains).
Scientists who think hybridization plays no important role in evolution often speak of “horizontal transfer,” which is the transfer, by means of viral infection, of genes from one type of organism to another. Some critics of the theory that humans are derived from hybridization between pigs and panins have suggested that horizontal transfer of genes from pigs to humans is a more plausible explanation of the many pig-like traits that distinguish us from other primates. The idea is that an initially ape-like animal was modified when infected with viruses of pig origin.
However, in my opinion, there are several problems with this view. One salient difficulty is that close association between humans and pigs is a relatively recent phenomenon, dating back only to the advent of agriculture, a few thousand years ago. But the existence of humans, as documented by paleontology, long predates that time. Of course, it might well have been the case that some animal similar to the domestic pig existed at a much earlier date, but pigs did not at that time live in regular association with humans or, for that matter, with human precursors. So if we somehow absorbed DNA from some other organism via horizontal transfer, why is it that the traits that distinguish us from primates don’t connect us instead with dogs? We’ve been living with dogs far longer than with pigs, and that contact has been more intimate, which would, presumably, make viral infection more likely. The close relationship between Homo sapiens and Canis familiaris dates back to a time when humans were living in caves. But the traits that distinguish us from other primates link us with pigs, not dogs.
In general, retroviruses are proposed as the agents of horizontal transfer, because viruses of this sort actually splice themselves into the DNA of the infected host, and therefore have the potential to be inherited. But for several reasons, retroviruses cannot be used to construct a satisfactory explanation of how humans might have obtained so many porcine traits.
- The first is that retroviruses carry a very small amount of genetic information. When traveling between cells they must package themselves into a container, called a capsid. The typical capsid contains only about enough space for, say, 30 kilobases of RNA. This figure varies from one type of retrovirus to another, but by less tnan an order of magnitude. Compare this with the addition of an average human chromosome, which contains on average about 3 billion/46 = 65,000 kilobases of DNA. That is, about two thousand times as much information as the entire genome of a retrovirus. But a single hybridization event typically introduces not one, but many chromosomes into a genome. So all the genetic information required to produce the various pig-like traits of humans could be delivered to a panin egg by a single pig gamete, whereas, presumably, hundreds of thousands of retroviruses would be required to deliver the same information. Moreover, it is not at all clear why all those retroviruses infecting the precursors of humans would have come only from pigs, and not instead pigs and a variety of other types of animals.
-
Furthermore, much of the 30 kb of RNA that a retrovirus can carry across from one organism to another is its own genome. Any gene that it picks up from one organism to carry across can be no larger than about half of 30kb, at most. So that means about 15kb. So if you compare that to say, 10 human chromosomes, a typical quantity of DNA in a fairly average hybrid cross, the ratio of information would be (65,000 kb x 10):15kb = 43,333:1. In other words, the amount that can be carried across by a single retrovirus is truly minuscule in comparison with what can be transferred in a single hybrid mating.
-
Moreover, for a retrovirus to become heritable, it must enter the germ line. So it must initially infect a gamete, which is a far rarer event than the mere infection of somatic cells.
-
Further, I find horizontal transfer, as a theory of human origins, unsatisfactory from a philosophical standpoint. In science we are supposed to construct theories, to the greatest extent possible, with known phenomena. It’s well known that many new forms of life, treated as species, have been produced by hybridization and the various stabilization processes I describe in the general theory of evolution presented elsewhere on this site. But is there even one known organism treated as a species that has resulted from infection with retroviruses? It seems there is not even a single case. Certainly there are not many. So why frame an explanation of human origins in terms of this apparently unobserved phenomenon?
-
Another objection, of a general nature, is that change through horizontal transfer would take a long time. You don’t expect an organism to go out one afternoon and have one of its germ cells infected with 43,333 separate retroviruses. You would expect them to accumulate in the germ line over time, with the passage of many generations. But the fossil record shows that the typical new form of life arises abruptly and remains unchanged up to the time of its extinction. So where does that leave retroviruses, which presumably must be gradually accumulated in vast numbers in the germ line over many generations if they are to have an effect as large as those seen when chromosomes are introduced via sexual hybridization? If you were not trying to account for data of abrupt change followed by stasis, they might be satisfactory. But unless you are going to posit some unknown mechanism that might allow all those virally introduced genes to lie dormant, and then to magically be released to suddenly to have a large effect, I don’t see how anyone can expect them to produce the observed phenomenon, that is, the pattern seen in fossil record where new forms typically arise abruptly and then remain unchanged up to the time of their extinctions (i.e., Punctuated Equilibrium). This is not to say that they would have no effect whatsoever. I just can’t see why they would be expected to have in any way as large an effect as whole chromosomes introduced by hybridization.
-
One other phenomenon seems inconsistent with the retroviral thesis, and that’s the observation that hybrids typically have a reduced fertility in comparison with their parents. When you produce a new organism via hybridization and stabilize a new organism with a new karyotype, the simple fact that it has a new karyotype means that its offspring in hybrid crosses with preexisting forms will have a reduced fertility. This is because the chromosomes will not be paired in the hybrids. So hybrid sterility is accounted for by stabilization theory. On the other hand, merely sprinkling the genome with retroviruses is not expected to reduce the fertility of the affected organism. Such a hypothesis does not, then, account for the fertility problems actually observed in humans, whereas a presumed hybrid origin does.
The idea of horizontal transfer seems to be favored among people who are uncomfortable talking about sex or with those who suppose that hybrids are somehow always completely sterile, or that they don’t occur in a natural setting. Nothing could be further from the truth, though many people imagine otherwise. Anyone who doubts that hybrids occur naturally or that they are often fertile should take a look at my book on bird hybrids (Handbook of Avian Hybrids of the World, Oxford U.P., 2006) or on the mammalian hybrids page of this website, where I’m making public, page-by-page, a lengthy manuscript on hybridization in mammals. I can’t imagine that anyone who read the documented information offered there would still think that hybrids are sterile or rare, or that they do not occur in a natural setting. All that is simply an ancient wives’ tale that I’m working hard to dispel.
Nor is there, under that theory, any reason to suppose that the human genome would contain genes sequentially identical to those of pigs. A given type of gene is very rarely present in only a single type of organism. Rather, the typical case for each kind of gene is for a wide variety of organisms to possess slightly altered versions of it. Also, a single type of organism will usually have multiple, differing copies of any given type of gene (these sets of similar but varying genes are called “gene families”). So in the present case, when pig-derived genes underwent crossing-over and gene conversion during hybrid meiosis, they would be expected in most cases to recombine with roughly equivalent panin-derived genes. As a result, successive rounds of backcrossing to panins would cause any such pig-derived genes to become more and more like their panin counterparts.
And yet, all of these converted genes would be expected still to code RNAs and proteins of the same general type that they did originally. The biochemical action of its protein product would not in most cases be greatly altered (for example, the various genes coding for actins, would still code for actins even after conversion). As has already been said, the main factor affecting development of an organism would be the overall set of genes defined by its karyotype, and the associated interactions of genes and regulatory sequences present in different dosages.
A karyotype in a zygote can be thought of as a set of initial conditions. The life cycle, of any stable organism is recursive because at each stage of its life cycle each of the cells in the organism applies the set of rules specified by the karyotype to produce the next stage. Development is the catch-all term for the complex series of events that convert a zygote, during the course of many successive cell divisions, into a mature organism.
Now suppose that a particular zygote contained a panin karyotype, and therefore the particular set of genes contained in such a karyotype, together with the regulatory sequences and all of the other molecules regularly present in that karyotype. Under such circumstances, the zygote begins dividing and its descendant cells go through the series of stages that ultimately produce an adult panin
However, suppose some of these same genes, regulatory sequences, and other molecules were deleted, and that other such genes, sequences and molecules were added, so that a different organism with a different set of gene dosages, to some degree more like the dosages seen in a pig was produced. Then the set of rules governing development would have changed. The interactions of genes and regulatory sequences at each stage of development would differ because those genes and regulatory sequences would be present in different dosages. At any given stage of development some proteins would be produced in greater quantities, others in lesser, so that within the developing organism each cell would respond differently from the way it would respond if it contained a panin karyotype. In this way, the differences at each stage along the path of development would cumulatively result in the production of a human being instead of a panin.
So, in brief, stabilization theory looks on development as a recursive process because, typically, a zygote with a particular karyotype uses the rules defined by the karyotype to give rise to other cells with a particular karyotype (usually the same one). The daughter cells then follow the rules defined by their karyotype to give rise to other cells with that karyotype, and so forth. The genes and regulatory sequences defined by the karyotype guide the process at every step. Alterations in the karyotype, such as the addition or deletion of genes and regulatory sequences, result in differences at each stage in development that accumulate to alter the ultimate developmental fate of the organism.
A test of the hypothesis
An earlier version of this webpage proposed that a survey of the human genome be carried out in an attempt to detect traces of pig in the human genome. During the years since that suggestion was made, I, together with two collaborators, did in fact carry out such a survey. But the results were mixed.
One fascinating finding was that scans with large numbers (hundreds of thousands) of randomly selected 40-mer nucleotide sequences from the pig genome produced matches in the human genome at a significantly higher rate than in panins. (A 40-mer is a sequence that's 40 nucleotides long.) We also found that the amount of the increase in the hit rate in humans versus panins exactly accounted for the amount of difference between humans and panins at the nucleotide level. For example, the hit rate with pig queries was 1.4 % higher in humans than in bonobos, which exactly corresponds with the previously reported amount of difference at the nucleotide level between humans and bonobos. The same was true for chimpanzees versus humans, where the difference was instead, in both cases, 1.8%. So we thought we had important results.
But then, just as a check, we conducted the same sort of scans but with randomly selected 40-mer queries from various nonprimates other than pig. The surprising result was that nearly every one of them produced higher hit rates in humans than in panins. So our initial results with pig queries could no longer be construed as having indicated a closer relationship between humans and pigs than between panins and pigs.
For months we were stymied by those results, but after considering various possible explanations we eventually homed in on the possibility that there was human DNA contamination in the publically available genome sequences for the various organisms we had used as queries. Over the last decade numerous papers have been published about the problem of biologists, attempting to sequence the genome of a given organism, inadvertently sequencing their own DNA instead. This can easily happen when the researchers, say, cough or sneeze into a test tube containing a tissue sample from the organism being sequenced. A flake of skin or an eyelash falling into the mix can do the same. Studies have shown that this kind of contamination is extremely widespread in the publically available genome databases, which are our only source of genomic sequences.
To check whether the genome sequences for the query organisms that we had been using might be contaminated we scanned them for the presence of Alu sequences. Biochemical studies have shown that Alu sequences are present in primates only, so they should not show up in the genome sequences of nonprimates, so long as those sequences are not contaminated with human DNA. But if they are contaminated, they should show up, because Alus make up about 11% of the human genome.
As it turned out, we did detect Alus in the genome sequences published for the various nonprimate query organisms that we used in our study. The presence of such contamination explained why our scans produced higher hit rates in humans than in panins with queries, not just from the published pig genome, but also from a wide variety of other published nonprimate genomes. The reason was that the published genome sequences do not accurately represent the actual genome sequences of the various query organisms we were using. Instead, they are composed of sequences from those organisms with an added fraction of human DNA sequences. But, obviously, for the published genome of any given nonprimate query organism, no panin sequences will be present in the contaminating fraction because panins do not work in laboratories. Only humans contribute to the contamination.
So, to sum up, we think we understand our results now, but we can no longer infer the stupendous conclusions that we once did. We will, however, continue to seek genetic methods of testing the hybrid hypothesis.
Concluding Remarks
It seems to me that the information thus far presented is consistent with the idea that humans and gorillas originated by hybridization. For humans, the case appears strong, because the hypothesis accounts for such a large number of observations. I consider the gorilla guilty by association, even though far less empirical data is available — both for this animal and, in particular, for one of its two posited parents. I reach this conclusion because 1) the case for human hybridity is persuasive, 2) humans and gorillas both exhibit a pattern of infertility that is otherwise unexplained, and 3) the modicum of genetic and morphological information available for this animal is consistent with the posited hypothesis.
The tentative scenario that I picture is that human beings came into being via hybridization between a pig, whose best modern representative is Sus scrofa, and an ape, best represented today by Pan paniscus. I assume, as a working hypothesis, that before this hybridization event a population of Pan paniscus-like panins was distributed throughout the geographic region now occupied by panins, not just south of the Congo-Zaire-Lualaba river barrier where such animals are found today, but also north, in those areas where only P. troglodytes is now found. Judging from what is known of the African climate in prehistoric times, I think the range of this proto-panin would probably have extended farther north than it does today, particularly in the Nile Valley. It would seem that sometime during the Pliocene, or more probably the Pleistocene, pigs came into contact with panins and hybridization occurred.
Now, it could be that such hybridization occurred only once in very ancient times (perhaps 5,000,000 or more years ago), producing the earliest hominids (australopithecines), and that various early human types hybridized to produce subsequent human types. It may be, too, that various hominid types each arose via a separate cross between a pig and a panin. One possible scenario is that a hybridization event occurred just prior to the time that modern humans first appeared (estimates for this date range from 100,000 to 140,000 years ago), perhaps somewhere in the Nile Valley, followed by an indeterminate number of generations of backcrossing to panins. During this time the hybrids would have improved in fertility, eventually breaking off from the panin population to breed strictly among themselves. During this backcrossing period many pig-derived DNA sequences would become more and more like those of panins. The resulting high level of similarity to panins, together with the fact that primate traits would predominate in hybrids resulting from multiple generations of backcrossing to panins, would explain why humans have invariably been grouped with primates, and would also account for the fact that a connection between pigs and human beings has been overlooked.
It also seems likely that panins crossed with Hylochoerus meinertzhageni, the giant forest hog, to produce the gorilla. This event way well have happened very recently (in fact, it may still be happening); it appears that no fossil remains have been found for the gorilla. The common chimpanzee population (P. troglodytes), also, appears to have been affected by this backcrossing, with some genetic influence carried across from Hylochoerus. This influence seems to be reflected today in such distinctive traits of the common chimpanzee as large body size, heavy jaws and canines, the occasional sagittal crest, and higher levels of genetic and morphological variability (as compared with P. paniscus, which here is considered a genetically isolated residual population). Hybridization with pigs may also have contributed to the increased variability and size of P. troglodytes populations (in comparison with P. paniscus). Thus, as a working hypothesis, P. troglodytes itself can be viewed as a kind of hybrid, but only in the sense that some degree of genetic leakage seems to have seeped through from the human and, especially, the gorilla populations. The human genetic influence seems to be minimal, because backcrossing probably stopped long ago, but in the influence of gorilla heredity on P. troglodytes — since hybridization between gorillas and P. troglodytes appears to continue even today.
I think the question of the gorilla’s hybridity will take longer to resolve than the human case. An immediate, obvious, hindrance is the paucity of information available for the gorilla and, especially, for the forest hog. In addition, several factors seem to indicate that the gorilla may be more highly backbred to panins (particularly to P. troglodytes): 1) Ongoing hybridization seems to be occurring in the case of the gorilla; 2) Gorilla and panin chromosome counts are identical (2n=48), while the human count is lower (2n=46); 3) Fewer morphological differences seem to exist between the gorilla and the two panins than between either of panin and Homo sapiens.
I must admit that I initially felt a certain amount of repugnance at the idea of being a hybrid. The image of a pig mating with an ape is not a pretty one, nor is that of a horde of monstrous half-humans breeding in a hybrid swarm. But the way we came to be is not so important as the fact that we now exist. As every Machiavellian knows, good things can emerge from ugly processes, and I think the human race is a very good thing. Moreover, there is something to be said for the idea of having the pig as a relative. My opinion of this animal has much improved during the course of my research. Where once I thought of filth and greed, I now think of intelligence, affection, loyalty, and adaptability, with an added touch of joyous sensuality — qualities without which humans would not be human.
When it comes to topics like human origins, where the opinions are rigid and the evidence thin, reservation of judgment is best. It is my hope that the arguments presented here will serve as an intellectual springboard allowing the mind to rise above the inflexible creeds of traditional evolutionary thought. Even if the hybrid hypothesis is wrong, any satisfactory theory of human evolution will have to address the facts touched upon in the foregoing discussion. Wrong or right, I believe a final answer is at hand. The obstacles to the acquisition of such knowledge are by no means insurmountable. Scientists around the world are gathering more data every day. If this rising tide of information indicates that the ideas that we have always had about our origins are wrong, we should not hesitate to correct our errors. Time after time, science has dispelled dogma and brought us things that were once beyond imagination. From tiny bacteria to vast galaxies, from telephones to rocket ships, our knowledge has continued to expand. Perhaps we will even at last be able to rend the veil that has so long obscured our own origins. If the hybrid hypothesis is correct, we will be able to find out where we came from. One simple thing is essential to that discovery: In the immortal words of Professor Bernhardt, “It isn’t faith that makes good science, Mr. Klatu. It’s curiosity!”
You can help:
|
Pause you who read this, and think for a moment of the long chain of iron or gold, of thorns or flowers, that would never have bound you, but for the formation of the first link on one memorable day.
—Charles Dickens Great Expectations |
What’s Big is Pig
They say we ape the apes alone,That their form is like our own,
That molecules suggest the same:
They’re clear links we can’t disclaim.
With DNA wise dons decide:
The gap between is slim, not wide.
They judge what’s true and dole it out.
And what they tout, we dare not flout.
Yes, bigwig prigs will dance their jig.
But, truth to speak, what’s big is pig.
— Gene McCarthy
A speculative email from a reader:
Hi Eugene,
You might recall the PBS documentary, “Meet the Coywolf.” In that documentary it was suggested that the hybrids have always existed in the territory where the ranges of coyote and wolf overlapped, but the hybrids were never as successful as either parent in that environment. But then the environment changed drastically, with the arrival of humans, such that the hybrids were far better adapted to it than either parent.
I’m thinking that something similar could have happened with chimps, pigs and humans: Humans (or human-like monsters) were always present in the overlapping ranges of chimps and pigs, all through the ages that northern Africa was forested, but they were a marginal population. But then, say five MYA, the forest started to retreat as the climate dried out. Suddenly, in geological terms, those hybrids were at a big advantage compared to either parent, because, unlike chimps, they could forage out on the savanna due to their upgraded cooling systems, and they were smarter than pigs and furthermore they could use found objects as rudimentary weapons. So the chimps disappeared from nearly all of their former range while humans were left behind by the retreating tide, so to speak, still living in the same territory as pigs (as shown by the Laetoli footprints), hunting them more and more efficiently and perhaps still occasionally mating with them.
That speculative scenario suggests to me that “humans” are still being produced in the overlapping ranges of chimps and pigs; they just haven’t been noticed because no one has been looking for them. Primatologists tend to study small groups intensively rather than trying to suck up all the data about thousands of animals.
Cheers,
Robert
Most shared on Macroevolution.net:
Human Origins: Are we hybrids?
On the Origins of New Forms of Life
Mammalian Hybrids
Cat-rabbit Hybrids: Fact or fiction?
Famous Biologists
Dog-cow Hybrids
Prothero: A Rebuttal
Branches of Biology
Dog-fox Hybrids