Excursion Chapter 3: The Origin Of Species

Introduction

The Origin of Species was intended to be an abstractof a larger manuscript that Darwin was working on when he received the celebrated letter from Alfred Russell Wallace. Darwin was rushed into publication of the Origin because of Wallace's independent discovery of natural selection and their joint presentation of a paper at the Linnaean Society in 1858. The first two chapters of the big book were eventually published under the title Variation of Animals and Plants under Domestication. The additional eight and a half chapters were only recently published under the title Natural Selection (Stauffer, 1975). This long version of the Origin included an extensive citation of sources.

The intelligent design (creationist) idea that life is like a manufactured object thoroughly refuted by Hume (see the critique of Excursion Chapter 1: The Origin of Life), Augros and Stanciu (1987, p. 21 fol.) and Arduini (1987).

The main difficulty with a single, unified modern theory of intelligent designis that no such thing exists! Who (or what) are the designer(s) and how do they design? And what did they design? the major groups of organisms? Which then evolved into their constituent species? That requires macroevolution. If latter doesn't exist (as Pandasseems to imply), then was each species created independently (special creation)? An alternate view is proposed by Ambrose (1982, pp. 143, 164) who says the designer took old species, and injected new information to make new ones. This is essentially theistic evolution. Most design proponents (creationists) claim that the designer (creator) works only by supernatural means yet Augros and Stanciu (1987, p. 229) that God made work through natural means. Thus there is potentially disagreement on every aspect of the "theory."

So what about diversity among the design proponents? On one hand, they appear to have uniform views simply because few authors have ever advanced any detailed answers to any of the important questions the "theory" is supposed to address, hence there is not much to disagree upon! When creationists do venture to put forth details (in the pages of the Creation Research Society Quarterly and at various conferences on Creationism), there is virtually no agreement among the various authors. In fact, they tend to ignore each others' ideas! Pandasadmits (p. 92) that intelligent design (creationism) embraces both "old earth" and "young earth" proponents. These are two such divergent world views as to render any differences between evolutionists trivial by comparison.

The Question of Kinds:The important question is not what unit of classification was originally designed, but how much can a designed entity change and evolve. This is not at all irrelevant. It determines how much evolution the design proponents (creationists) will admit to! The Pandasauthors have something in mind. They admit the evolution and common ancestor for the Hawaiian fruit flies discussed later in the chapter but not for chimps and starfishes (see Pandas, p. 127). What criterion is used to distinguish such cases? Speculation on the subject of the amount of change capable by a created kind runs rampant among creationists—there is no solid ground whatsoever upon which anything can be based! Groups as varied as man (a species), horses (a family), turtles (an order), sharks (a class), sponges (a phylum), algae (a group of phyla), bacteria and fungi (kingdoms) have been listed as examples of kinds (Morris, 1974, p. 87-88). Pandas' statement (p. 78, top of col. 2) that "design proponents accept the idea that species can change within limits. . ." implies that each species represents a kind designed and created de novo.

The limits to change: Pandas' "case" that there are limits to variation is very weak. Bumpus' study on house sparrows (which has questionable validity) only showed stabilizing selection in a population already adapted to a particular environment. The development of considerable geographic variation in the house sparrows introduced into North America illustrated the power of directional selection in creating adaptations to varied environments. Muller only did experiments with ionizing radiation to produce new mutations. No one has tried to produce unlimited change. Animal breeders, for example, have never tried to make their animals into something entirely new; they only wanted to improve or enhance certain useful properties. Yet, domestic dogs presents an enormous assemblage of forms, the extremes differing from each other far more strikingly than many natural species (Lovtrup, 1987, p. 368; Wells, Huxley and Wells, 1931, vol. 1, p. 374 fol.). If they were naturally occurring populations, they would probably be classified into a number of genera! Figure 3.1 illustrates the extent of variation exhibited by domestic dogs. Darwin says the same thing about the domestic races of pigeons (Darwin, 1968, pp. 82-83). Their variation, not only in plumage, but in size, beak shape, shape of the skull and variations in skeleton is such that, if they were wild birds, would be classified as different species, if not different genera. The same holds true for other domestic animals. Some of the more bizarre varieties of goldfish are pictured by Zahl (1973).

The fact that most breeds of domestic animals can and do successfully interbreed and hence are considered the same species is beside the point. What is more important is to realize that they were developed in an artificial reproductive isolation enforced by the breeders. One could imagine selecting for a more "natural" reproductive isolation between two breeds of dogs using the selection protocol applied to Drosophilaby Koopman (1950, see below) thus making them true biological species instead of artificially sustained "pseudospecies". With any domestic animals, however, this would entail a prohibitive amount of time, work, money and animals, without any benefit to the breeds.

Even the experiments in artificial selection carried out by Drosophilageneticists had the modest goals of seeing how far a single, particular character (such as number of abdominal bristles) could be modified. When small populations were involved, the new phenotypes were formed entirely by genetic recombination of mutant alleles already existing in the population. Obviously a limit was reached when the genetic variance was exhausted. With larger populations, progressive changes in the character occurred at an irregular rate influenced by the occurrence of new mutations. The limit to change was brought about, not by exhausting the genetic variance, but by pleiotropic effects or by other genes closely linked with those regulating the character being selected causing sterility. This problem could be overcome by relaxing selection and waiting for recombination to unlink the sterility genes before resuming selection. Even so, in experiments selecting for an increase in abdominal bristle number in Drosophila, a fourfold increase—16 times the phenotypic standard deviation in the initial population—was attained. This is comparable in magnitude to many macroevolutionary changes (Futuyma, 1986, pp. 90-91, 207-210; Smith, 1989, pp. 113-117). The development of the most recent "Green Revolution" plant varieties was done with the aid of radiation-induced mutations (Sigurbjornsson, 1971).

It's nice that the "design proponents" agree that research should continue, but they are likely to let evolutionary biologists do the work of demonstrating the feasibility of evolutionary change. After all, who has the incentive to undertake an exhaustive study when he expects negative results? Also, it's too easy to get negative results by just not trying too hard or being persistent enough.

Allopatric Speciation

The Concept Explained

". . . divergence is usually a gradual process. It follows that situations must be expected, and they do occur, in which two groups of populations are too distinct to be regarded as races, but not distinct enough to be considered species. Whether they are named as races or as species is, then, arbitrary and decided solely on grounds of convenience. Biologically, however, these "difficult" situations are highly significant. They are borderline cases of uncompleted speciation, of speciation in the process, of speciation which happens to be unfinished on our time level but which may conceivably be completed in the future… It is no exaggeration to say that if no instances of uncompleted speciation were discovered the whole theory of evolution would be in doubt, we would have to conclude either that evolution did not occur or that the formation of new species is instantaneous. What is a difficulty to the cataloging systematist is a blessing to the evolutionist."

(Dobzhansky, 1958, p. 48; Pandas' footnote 1)

Such borderline cases are quite common, Mayr (1964, p. 165) reports that 94 (or 12.5%) of the 1,367 species and subspecies of native North American birds fall in this category. The species of European house sparrows are an example. These spread west from the Nile valley both north and south of the Mediterranean Sea during an interglacial period. Today the house sparrow and the related Spanish sparrow exist side by side as good species in Spain and the Balkans, yet they interbreed and produce a hybrid zone in Algeria. The Italian sparrow appears to be a stable hybrid of the other two, partially isolated from them by the Alps in the north and the Mediterranean Sea in the south. Yet there is a narrow hybrid zone with the house sparrow in the southern foothills of the Alps and one with the Spanish sparrow in Sicily (Summers-Smith, 1963, chapter 15). A great deal of research has been done on these so-called borderline cases to elucidate the speciation process. It is reviewed by Dobzhansky (1951), Mayr (1963, 1964, 1970) and Stebbins (1950). Shorter accounts are given by Futuyma (1986, chapter 8) and Avers (1989, chapter 8).

Initial spatial isolation of some kind that cuts off gene flow between two populations makes possible evolutionary divergence as the populations adapt to different environments. In general the characters that isolate species are genetically similar to characters that vary within a species. As two populations become more dissimilar genetically, it is less likely that their genotypes will be compatible in a hybrid. Recently it has been suggested that the mismatch repair system might be involved in this process (Rayssiguier et al, 1989; Rennie, 1990). Laboratory experiments have shown incipient reproductive isolation developing between isolated populations of Drosophilaand houseflies that have been exposed to divergent artificial selection for responses to various environmental factors (Futuyma, 1986, p. 226).

Natural selection may enhance reproductive isolation when two formerly allopatric populations come into contact and cross-mating produces less fit hybrids. This phenomenon has been produced in the laboratory. Chromosomal abnormalities were introduced into two mutant strains of Drosophila melanogasterto produce two artificial species. After two years in mixed culture where cross-matings were infertile, natural selection produced flies that preferred to mate with their own "kind" (Wallace, 1973). Similar results were obtained by Koopman (1950) who eliminated hybrids in cages with mixed populations of D. pseudoobscuraand D. persimilis. (In nature these two species are separated by ecoclimatic preferences but can breed freely in the lab.) When Paterniani (1969) eliminated hybrids between two populations of maize, they developed an isolating mechanism based on different timing of flowering. An analysis of data on 119 pairs of closely related Drosophilaspecies pairs by Coyne and Orr (1989) indicates that these same phenomena occur in nature: both pre- and post-zygotic isolation (see below) increase with time since species divergence and prezygotic isolation evolves more rapidly in sympatric species pairs.

Genetic Change In Small Populations

Pandastreatment here is a reasonably accurate but highly simplified account that can be very complex in all its details.

Reproductive Isolation

This section says little about the variety of isolating mechanisms specifically mentioning only differences in courtship behavior of Hawaiian fruit flies.

There are a variety of reasons why two species may remain genetically separate. These are usually referred to as isolating mechanisms and may be either premating or postmating (Futuyma, 1986, p. 112):

Premating mechanisms: Potential mates may not meet because they live in different geographic areas separated by a barrier, live in different habitats or have mating seasons at different times. If they do meet, they may not mate because inappropriate courtship rituals. If they do meet and mate, sperm transfer may not take place.

Postmating mechanism: sperm transfer may take place but the egg is not fertilized; or the zygote dies, or the hybrid organism has reduced viability or is partially or completely sterile.

A parallel but slightly different classification is into pre- and post-zygotic mechanisms. Among allopatric species, both pre- and postzygotic isolation evolves in conjunction with increasing genetic divergence while among sympatric species, where a degree of postzygotic isolation exists, prezygotic mechanisms are produced by natural selection (see preceding section on Allopatric Speciation: The Concept Explained).

The Hawaiian fruit flies: These were not discussed in the previous chapter of Pandas. In these flies, allopatric changes in courtship rituals appear to be the main reproductive isolating mechanism. See Carson et al(1970) for an overview of the Hawaiian drosophilids. Kaneshiro and Ohta (1982) and Hapgood (1984) give a popular introduction to these flies. The South American fruit flies mentioned by Pandas(p. 16) are discussed in more detail by Ayala (1978).

The Heliconiusbutterflies: During dry periods associated with the last Ice Age, the Amazonian rain forest was reduced to isolated "islands" surrounded by savannahs within which the various races of H. eratoevolved. When the climate became wetter, the races expanded their range. In some places their range borders occur at natural barriers, such as the Amazon river, strips of grassland along the crest of a range of hills, and the white sand forest in the Guianas. These butterflies has a bad taste and birds quickly learn not to eat them. Their brightly colored wings serve as warning coloration. Interestingly enough, a closely related species, H. melpomenehas a similar series of races whose color patterns mimic H. erato. Because both these species are bad tasting, this is as case of Mullerian mimicry where two (or more) species share the same warning color patterns, making it easier for the birds to learn them. The biology of these butterflies and their evolution by neo-Darwinian mechanisms are summarized by Turner (1981); a popular account with colored illustrations of the races is found in Turner (1975); the hybrids of H. melpomeneand their genetics are discussed in Turner (1971).

Can reproductive isolation occur without geographic isolation (i.e. can speciation be sympatric?) This is still a controversial question (Futuyma, 1986, p. 228; Mayr, 1963, p. 449 fol.). Rice (1985) succeeded in producing prezygotic reproductive isolation as a result of disruptive selection for habitat in laboratory populations of Drosophila melanogaster.

The Kaibab squirrel: Pandassays that "random mechanisms such as genetic drift and natural selection could possibly operate together to produce reproductive incompatibility." Natural selection is definitely not a random mechanism!! The Kaibab and Albert squirrels differ in the markings on their tails and bellies. Both live in ponderosa pine forests, eating the seeds and branch tips of the tree. They never meet for no ponderosas grow in the desertlike depths of the canyon (Colbert, 1976, p. 308).

Do Losses Lead To Gains?

Genetic drift, fixation, the founder effect and the bottleneck effect do not result in losses of genes (an entire locus- a part of a chromosome) which is usually very detrimental or fatal. At most only an allele of a gene is lost, or its frequency in the population changed. Each individual in the new species has the same number of genes and amount of information in its DNA as any individual of the old species. Only the variation in this information exhibited by the respective populations has been changed. And this variation can be made up by new mutations. Speciation is not a process involving a loss of information but a change in information.

An observational basis for the introduction of new information resides in the extensive evidence that new genes are formed from mutation in duplicate copies of old genes. This idea is supported by detailed molecular studies on a number of "new" genes (for example, the globin family) that arose in the course of evolution (Avers, 1989, pp. 83-88; Futuyma, 1986, pp. 451-455; Smith, 1989, chapter 11; Ohno, 1970). On the other hand there is no experimental work of any kind that sheds any light on how intelligent designers create new information. In fact, DNA, with all its nonsense sequences, etc. does not look like a product of any intelligence as we know it! Pandasmust explain why the DNA of all organisms studied is so full of nonfunctional garbage.

Does speciation fit with the theory that species were originally designed? Apparently that theory can accommodate any observations that might be made. It even includes evolution as a possibility (the Hawaiian Drosophilaspecies evolved)! Is there anything that the theory specifically says must occur or must not occur? Only then can it be tested scientifically. How can one distinguish between information supernaturally added to DNA and that produced by modification of previously existing DNA?

The theory claims to predict that "there may be species on the face of the earth that have undergone no substantial change since their beginning." What are the premises from which that conclusion is derived? It is admitted that species can change. How much is a "substantial" change? When did these species begin, 10,000 years ago or millions of years ago? And if we deal with fossil representatives of a species, we are limited to observing changes in the hard parts and cannot observe their DNA. Clearly this "prediction" cannot be satisfactorily tested. And it doesn't distinguish creation from evolution, because "living fossils" can be explained by stabilizing selection in stable environments.

Design proponents (creationists) have been around longer than evolutionists, yet they have never done any research or even suggested any concrete answers to questions concerning the limits of change, how to identify original species, and what the exact biological definition of species should be. In fact they appear to avoid making any definite statements on these matters.

The Tempo Of Evolution

Footnote 2 is to Darwin (1968, p. 292). There are demonstrable gaps in the fossil record itself. Yet, since Darwin, many examples of unambiguous intermediate forms linking groups have been found. These will be discussed in the critique of Excursion chapter 4 which deals with the fossil record.

Punctuated Evolution

Pandas' description of punctuated equilibrium is somewhat misleading. The hypothesis states that most evolution takes place during speciation. The type of changes involved here are those typically distinguishing related species. They are not "major" changes, such as those distinguishing higher categories—genera, families, orders, etc. Such speciation is considered to take place over not 100's of years, but anywhere from 5,000 to 50,000 years (Lewin, 1981) or even 100,000 years (Ayala, 1983, p. 391). The mechanism involved is neo-Darwinian mutation and natural selection occurring in small peripheral populations and is the same as Mayr's peripatric speciation (Mayr, 1963, chapter 17). Extinction "due to factors other than competition" has no more to do with the punctuated equilibrium hypothesis of speciation than any other hypothesis of speciation. The hypothesis predicts gaps between species in the fossil record on the basis that one is unlikely to find fossil strata from the particular positions in space and time that document instances of such a local and fleeting phenomenon. But, in fact, several such examples of such species transitions do exist (Eldredge, 1985, pp. 78, 88). Thus, the hypothesis is not based exclusively on gaps in the fossil record.

The Failure Of Natural Selection

The fragmentary quote attributed to Mayr is really from Simpson and the Simpson quote is from Mayr! Pandasis not only confused here but guilty of quoting out of context in order to "prove" their spurious point that "evolutionists still have not solved the fundamental problem of how species originate." More extensive quotes from these two authors are given here:

"It is not so widely recognized that Darwin failed to solve the problem indicated by the title of his work. Although he demonstrated the modification of species in the time dimension, he never seriously attempted a rigorous analysis of the problem of the multiplication of species, of the splitting of one species into two."

(Mayr, 1963, p. 12—Pandas' footnote 3)

"There is another extremely important problem of evolution that has been only implicit up to this point. It was curiously neglected by Darwin, whose book called The Origin of Species is not really on that subject, but the neglect has been richly compensated in more recent years (emphasis added). The problem is how, in fact, do species originate; that is, not only how does one specific lineage evolved and adapt, but also how do multiple specific lineages arise and become divergently adapted."

(Simpson, 1964, p. 81, Pandas' footnote 4)

Mayr's work in particular more than makes up for the deficiency in the Origin. Thus, contrary to Pandas' conclusion, most evolutionists consider that allopatric speciation explains the origin of new species. Even Denton, who considers evolution a theory in crisis, says that speciation has been explained by neo-Darwinism (Denton, 1986, pp. 82-83, 344).

Furthermore, the intelligent design explanation, that species "were intelligently designed by some informative selection of the material for their genotypes" is totally empty of information content. We are given no explanation or understanding how this occurs! How does the designer (or designers) accomplish this? Furthermore, is each species made de novo (special creation), or is new information "injected" into a population that is speciating in an otherwise natural manner as suggested by Ambrose (1982, pp. 143, 164)? This latter suggestion is theistic evolution, a concept anathema to the special creationists (Morris, 1974, pp. 215-220). Evolutionists suggest the new information arises from mutations of existing genes or duplicated copies of those genes. How does one distinguish between such events and the "injection" of new information by a designer (creator)?

As we mentioned earlier, the amount of variation exhibited by domestic breeds is equivalent to that found within genera or even families of naturally occurring organisms. And this is remarkable, considering the short span of time (centuries) and the small population sizes involved. To proceed any farther requires waiting for the appropriate mutations to occur. In this context, remember that many of the advanced Green Revolution plant strains were derived from stocks irradiated to speed up the mutation rate.

The evolutionists' three problems:

Lines stable enough to be called species are probably brought about by stabilizing selection. Pandasis well aware of this process (see Excursion chapter 2). Why do they think this is a problem? Pandas' further statement that the world is filled with distinct and stable species is not entirely true. Speciation is a slow process taking thousands of years. There are many examples of species in the middle of this process which allows us the opportunity to study it. It has been mentioned earlier that 12.5% of the 1,367 species and subspecies of North American birds are only partially distinct and have known intermediate forms, indicating only partial isolation. This is the context of the quote from Dobzhansky (1958) on p. 79 of Pandas! Also Pandasapparently accepts the formation of the Hawaiian fruit fly species by evolutionary means (Pandas, p. 82). And instances of fossil intermediates between species have been found (Eldredge, 1985, pp. 78 fol., 88).

The second problem of long-term stability is also basically explained by stabilizing selection although the examples of it are exaggerated. For example there is not just one "shark". There are many species and genera. Some of the living genera go back to the Cretaceous. Even then, we can only say that the teeth and those parts of the cartilaginous skeleton that normally calcify secondarily and would form fossils, haven't changed very much. The quote from Thorpe (Pandasfootnote 5) comes from Taylor (1983, pp. 141-142).

As for the third problem, we have been mentioned that there is abundant evidence that new information (new genes) arise through mutation of duplicated copies of old genes. Again, it must be pointed out that the intelligent design "explanation"—that a "designer did it" is empty of any information content and provides no understanding of the phenomena at all.

Micro- And Macroevolution

Again, the two types of evolution ‘microevolution' and ‘macroevolution' (see Pandas, p. 61) grade into one another without any clear borderline. Some evolutionists (Rensch, 1960, p. 1; Futuyma, 1986, p. 397) consider micro-evolution to include those processes that occur within a species or lead to a new species while macro-evolution are those processes leading to new genera, families, etc. According to this view, Pandaschapters 2 and 3 really deal with microevolution! Macroevolution is usually studied through the fossil record. The neo-Darwinian position is that larger macroevolutionary differences arise through the successive accumulation of microevolutionary changes that occur within a species or during speciation (Futuyma, 1986, p. 397). Bock (1970) studied the adaptive radiation of the Hawaiian honeycreepers, specifically the macroevolutionary changes in the bill and feeding habits of the four genera of the subfamily Psittirostrinae and showed that these changes occurred by a series of steps involving micro-evolutionary modifications on the species level. See Futuyma (1986, pp. 32-33), Ralph (1982) and Lewin (1982, pp. 58-60) for an introduction to these birds. They are also mentioned briefly at the beginning of Pandas' Overview section 3. The magnitude of their morphological diversity is such that, if they were not connected by intermediate species, it would be enough to place them in different families or orders (Stebbins, 1982, pp. 68, 134). A comprehensive discussion of macroevolution and related topics is found in Levinton (1988).

References

Ambrose, E. J. 1982. The Nature and Origin of the Biological World. Ellis Horwood Ltd.

Arduini, F. J. 1987. Design, Created Kinds and Engineering. Creation/Evolution 7(1): 19-23 (Spring).

Augros, R. and G. Stanciu. 1987. The New Biology: Discovering the Wisdom in Nature. New Science Library of Shambala Publications Inc.

Avers, C. J. 1989. Process and Pattern in Evolution. Oxford University Press.

Ayala, F. J. 1978. The Mechanisms of Evolution. Scientific American 239(3): 14-27 (September).

Ayala, F. J. 1983 Microevolution and macroevolution. In: Bendall, D. S. (Editor). Evolution from Molecules to Men. Cambridge University Press. pp. 387-402.

Bock, W. J. 1970. Microevolutionary sequences as a fundamental concept in macroevolutionary models. Evolution 24: 704-722.

Carson, H. L., D. E. Hardy, H. T. Spieth, and W. S. Stone. 1970. The Evolutionary Biology of the Hawaiian Drosophilidae. In: Hecht, M. K. and W. C. Steere (Editors). Essays in Evolution and Genetics in Honor of Theodosius Dobzhansky. Appleton-Century-Crofts. pp. 437-544.

Colbert, E. H. (Editor). 1976. Our Continent: A Natural History of North America. National Geographic Society.

Coyne, J. A. and H. A. Orr. 1989. Patterns of speciation in Drosophila. Evolution 43(2): 362-381.

Darwin, C. R. 1968. The Origin of Species. Reprint of the 1^st(1859) Edition. Penguin Books.

Dobzhansky, T. 1951. Genetics and the origin of species. Columbia University Press.

Dobzhansky, T. 1958. Species after Darwin. In: Barnett, S. A. (Editor) A Century of Darwin. William Heinemann. pp. 19-55.

Eldredge, N. 1985. Time Frames. Simon and Schuster. N. Y.

Futuyma, D. J. 1986. Evolutionary Biology. 2^ndEdition. Sinauer Associates, Inc.

Hapgood, F. 1984. Fruit Fly Fandango. Science 84 5(7): 68-74. (September).

Kaneshiro, K. and A. T. Ohta. 1982. The Flies Fan Out. Natural History 91(12): 54-59 (December).

Koopman, K. F. 1950. Natural selection for reproductive isolation between Drosophilapseudoobscura and Drosophilapersimilis. Evolution 4: 135-148.

Levinton, J. 1988. Genetics, Palaeontology and Macroevolution. Cambridge University Press.

Lewin, R. 1981. Evolutionary Theory Under Fire. Science 210: 883-886. (21 November).

Lewin, R. 1982. Thread of Life. Smithsonian Books.

Lovtrup, S. 1987 Darwinism: The Refutation of a Myth. Croom Helm.

Mayr, E. 1963. Animal Species and Evolution. The Belknap Press of Harvard University Press.

Mayr, E. 1964. Systematics and the Origin of Species from the viewpoint of a zoologist. (reprint of a 1942 book). Dover Publications.

Mayr, E. 1970. Populations, Species, and Evolution: An Abridgment of Animal Species and Evolution. The Belknap Press of Harvard University Press.

Morris, H. M. (Editor). 1974. Scientific Creationism. Creation-Life Publishers.

Ohno, S. 1970. Evolution by Gene Duplication. Springer-Verlag.

Paterniani, E. 1969. Selection for reproductive isolation between two populations of maize, Zea mays. Evolution 23: 534-5447.

Ralph, C. J. 1982. Birds of the Forest. Natural History 91(12): 40-45 (December).

Rayssiguier, C., D. S. Thaler and M. Radman. 1989. The barrier to recombination between Escherichia coliand Salmonella typhimuriumis disrupted in mismatch-repair mutants. Nature 342: 396-401 (23 November).

Rennie, J. 1990. Kissing Cousins: A DNA repair system stops species for interbreeding. Scientific American 262(2): 22-23 (February).

Rensch, B. 1960. Evolution Above the Species Level. Columbia University Press.

Rice, W. R. 1985. Disruptive selection on habitat preference and the evolution of reproductive isolation: an exploratory experiment. Evolution 39: 645-656.

Sigurbjornsson, B. 1971. Induced mutations in plants. Scientific American 224(1): 86-95. (January).

Simpson, G. G. 1964. This View of Life. The World of an Evolutionist. Harcourt, Brace and World.

Smith, J. M. 1989. Evolutionary Genetics. Oxford University Press.

Stauffer, R. C. 1975. Charles Darwin's Natural Selection: Being the second part of his big species book written from 1856 to 1858. Cambridge University Press.

Stebbins, G. L. 1950. Variation and Evolution in Plants. Columbia University Press.

Stebbins, G. L. 1982. Darwin to DNA, Molecules to Humanity. W. H. Freeman and Company.

Summers-Smith, D. 1963. The House Sparrow. Collins.

Taylor, G. R. 1983. The Great Evolution Mystery. Harper and Row.

Turner, J. R. G. 1971. Two thousand generation of hybridisation in a Heliconiusbutterfly. Evolution 25: 471-482.

Turner, J. R. G. 1975. A Tale of Two Butterflies. Natural History 84(2): 29-36.

Turner, J. R. G. 1981. Adaptation and evolution in Heliconius: A Defense of NeoDarwinism. Annual Review of Ecology and Systematics 12: 99-121.

Wallace, B. 1973. Man's Humanity. Saturday Review of the Sciences. February: 48-49.

Wells, H. G., J. S. Huxley and G. P. Wells. 1931. The Science of Life. Doubleday, Doran and Co.

Zahl, P. A. 1973. Those Outlandish Goldfish. National Geographic 143(4): 514-533 (April).

(from Frank Sonleitner's critique of Of Pandas and People)