PARALLELISMS DURING MAJOR EVOLUTIONARY TRANSITIONS
Alexander V. Markov
Borissiak Paleontological Institute, Profsoyuznaya 123, 117997 Moscow, Russia
2010, Third International conference dedicated to N. W. Timofeev-Ressovsky "Modern problems of genetics, radiobiology, radioecology and evolution" Alushta, 9-14 October 2010
Abstract. Parallel evolution is generally regarded as important evidence of directionality and predictability of macroevolutionary trends. Paleontological data imply that multiple parallelisms are typical during major evolutionary transitions (aromorphoses). This pattern reveals itself in numerous transitional fossils with mosaic distribution of ancestral and derived characters. The typical examples include: (i) ‘ornithization’ of theropods, a process during which different avian characters evolved independently in several clades of theropod dinosaurs; (ii) ‘arthropodization’, a major evolutionary transition from wormlike ancestors to typical arthropods; (iii) ‘hominization’, or parallel evolution of derived ‘human’ characters in different lineages of australopithecines; (iv) ‘mammalization’ of theriodonts, which involved parallel acquisition of various derived mammalian features in different lineages of theriodont reptiles; and many others. The same pattern of numerous homoplasies can be observed in smaller-scale evolutionary processes, e.g., during adaptive radiations of closely related founder populations on different islands or isolated lakes. This pattern tend to obscure phylogenetic relationships between taxa and impair the applicability of cladistic procedures based on parsimony analysis. The fact that parallelisms are an ubiquitous feature of major evolutionary transitions implies that evolution in general is largely constrained and canalized by multiple genetic, ecological, morpho-physiological and developmental limitations.
The ratio between randomness and directionality in evolution remains one of the most controversial questions in evolutionary theory. Parallel and convergent evolution is generally regarded as important evidence of directionality and partial predictability of macroevolutionary trends. Paleontological data imply that multiple parallelisms represent a typical feature of major evolutionary transitions (aromorphoses) and adaptive radiations (Romer, 1949; Krassilov, 1977; Tatarinov, 1987). Generally, the more we know about some particular transition, the more apparent it is that many important features of the derived taxon evolved independently in more than one subclade within the ancestral clade. This pattern reveals itself in numerous transitional fossils with mosaic distribution of ancestral and derived characters.
Different derived characters that comprise an aromorphosis sometimes develop almost simultaneously in several lineages, or they can appear at different times and even in different order. Derived characters gradually accumulate, until they all are combined together in one or few lineages. These lineages then will be considered as belonging to a new, ‘higher’ evolutionary grade (Tatarinov, 1987). The formation of a new evolutionary grade thus often have the appearance of a directed, regular process, which may even look ‘teleological’ (aiming for a goal).
Here, I discuss several examples from the fossil record showing that multiple homoplasies during the formation of a new major taxon are typical rather than exceptional. Further I consider several cases of recent and ongoing adaptive radiations and discuss possible explanations of the observed pattern.
The evolutionary transition from theriodonts to mammals (Late Permian – Triassic) took about 30 to 40 million years and was accompanied by successive acquisition of derived ‘mammalian’ characters (e.g., superior nasal conchae, molars with three tubercules, vibrissae, enlarged brain hemispheres, soft lips and lip muscles, jaw joint between dentale and squamosum, three bones in the middle ear) in several different lineages of theriodonts and early mammals. This pattern was called ‘mammalization’ of theriodonts (Tatarinov, 1976); the term highlights the parallel development of similar derived characters in different lineages. Later research confirmed multiple homoplasies at the base of mammalian phylogenetic tree, particularly in the evolution of the middle ear (e.g., Luo et al., 2007).
‘Ornithization’ was a process during which avian characters (e.g., pennaceous feathers) evolved independently in several clades of theropod dinosaurs (Kurzanov, 1987). Like ‘mammalization’, the ‘ornithization’ took several tens of million years (Late Jurassic to Late Cretaceous).
For many decades, the only known transitional fossil filling the gap between ‘reptiles’ and birds was Archaeopteryx, one of the most famous transitional animals, that lived about 150 million years ago. As long as Archaeopteryx remained the only fossil with combination of reptilian and avian traits, the origin of birds seemed to be a straightforward, linear process: dinosaurs gave rise to Archaeopteryx, which then gave rise to real birds (Ornithura).
During the last three decades a great number of new transitional fossils were discovered, which, instead of further clarifying the situation, made it much more complicated and confused (this is a typical story in paleontology). Archaeopteryx is no longer alone. A variety of feathered creatures with different combinations of reptilian and avian characters was found in the Jurassic and Cretaceous rocks of China, Mongolia, South and North America and other regions. Some of them, like Confuciusornis and Enantiornis, are related to Archaeopteryx and usually classified as birds, although it is still disputed exactly how closely they are related to modern birds (Chiappe, 1995; Kurochkin, 1995; Fountaine et al., 2005; Walker et al., 2007).
Moreover, there are many transitional fossils that are classified as dinosaurs, but have a number of avian characters, including feathers. Apparently, many small coelurosaurian dinosaurs were acquiring different avian characters at different times. Feathers were quite common among coelurosaurs (Turner et al., 2007). Feathers probably evolved for thermoregulation; some species (e.g., Epidexipteryx hui) apparently used their large, brightly coloured feathers to attract mates in the way modern birds of paradise do (Zhang et al., 2008).
A very typical pattern during the ornithization was the mosaic distribution of morphological characters among clades. For instance, Epidexipteryx is “characterized by an unexpected combination of characters seen in several different theropod groups, particularly the Oviraptorosauria” (Zhang et al., 2008). Such a combination apparently implies multiple homoplasies and mosaic distribution of characters.
Feathers, that originally evolved for maintaining body temperature and attracting mates, eventually were used for gliding and flight. Recently paleontologists discovered several species of astonishing four-winged dinosaurs, e.g., Anchiornis who lived a little earlier than Archaeopteryx (Hu et al., 2009), and Microraptor, who lived some 30 million years later. Microraptor was probably a good flyer; it had large pennaceous feathers both on its hind limbs and forelimbs (Hone et al., 2010). Flying or gliding forms evolved independently in several different lineages of coelurosaurs, e.g., in Troodontidae, Dromaeosauridae, and Aviale (a clade containing Archaeopteryx, several other fossil genera, and Ornithura). Some dinosaur species even may have secondarily lost the ability to fly (Turner et al., 2007; Hu et al., 2009).
The similar pattern of mosaic distribution of characters and parallel acquisition of advanced traits revealed itself during the ‘arthropodization’, an evolutionary transition from wormlike ancestors to typical arthropods (Cisne, 1974; Tatarinov, 1987; Ponomarenko, 2004). Until recently, arthropods were thought to be descended from polychaetes. However, paleontological (as well as molecular) data provided no support to this hypotheses. Instead of predicted transitional forms between polychaetes and arthropods, numerous bizarre fossils were discovered with unexpected combinations of derived arthropod characters and ancestral characters of wormlike animals, probably not related to polychaetes.
Some fossils from the latest Precambrian (Ediacaran period) superficially resemble the expected transitional forms between polychaetes and arthropods. For instance, Spriggina has a crescent-shaped head similar to that of some trilobites, and long segmented polychaete-like body. However, its segmentation is very different from that of either arthropods or polychaetes: the segments of the body alternate along the midline, as if the left side of the body was shifted half a segment forwards or backwards relative to its right side (Fedonkin, 1998). Such ‘alternating bilateral symmetry’ is not found in polychaetes and arthropods, and is very difficult to explain in terms of comparative anatomy and embryology of extant invertebrates. This unusual symmetry is present in many other Ediacaran animals. The relationship between these forms and annelids or arthropods is not clear (Fedonkin, 1985, 1998).
Cambrian faunas of soft-bodied arthropod-like creatures also did not provide any simple picture. The most well-known finds of these animals came from the Middle Cambrian Burgess Shale (Canada) (Gould, 1989; Briggs et al., 1995). One of these forms, Hallucigenia, was especially difficult to understand. The first fossils of Hallucigenia seemed to have three appendages on each segment: two hard spines and one soft tentacle. There was no “alternating” symmetry typical of the Ediacaran animals: Hallucigenia has normal bilateral symmetry, but three appendages per segment are probably even more bizarre than alternating limbs. It was originally suggested that Hallucigenia walked on paired spines and had a single row of soft tentacles sticking from its back. Later it became clear that there were two tentacles per segment rather than one. This discovery resulted in a more credible reconstruction. The animal has been turned upside down and put on its paired tentacles, so that hard spines turned into protective appendages, like the spines of a sea urchin. In this position, Hallucigenia vaguely resembles Onychophorans (Gould, 1992). Other animals similar to Onychophorans were found in Burgess Shale (e.g., Aysheaia).
Another bizarre Middle Cambrian animal, Anomalocaris, was originally described as several different animals. Its oral disc was thought to be a strange jellyfish with an opening in the middle; sharp teeth were later discovered at the edge of the opening. Other fossils, originally described as shrimps, turned out to be paired appendages located near the mouth. Only when some better preserved specimens were found, it became possible to reconstruct this strange carnivore. Its segmented body had flat swimming lobes, located ventrally, but no segmented legs typical of arthropods. The only pair of large segmented appendages for grasping prey was near the mouth.
The soft-bodied fauna of Burgess Shale and other Cambrian localities contains a variety of animals with different combinations of ‘arthropod’ characters. For instance, there is Opabinia, apparently a relative of Anomalocaris, but with five eyes rather than two, a strange flexible proboscis, and no segmented limbs (Gould, 1989; Briggs et al., 1995). There are also Anomalocaris-like forms with biramous limbs (a typical arthropod feature) on the trunk segments, e.g., Schinderhannes from the Early Devonian of Germany (Kühl et al., 2009).
Paleontologists are trying to reconstruct the most plausible phylogenetic tree of the early arthropods based on the parsimony principle, which implies that the number of homoplasies should be minimized. But it appears to be not possible to remove all homoplasies from such trees. This is why in papers describing transitional fossils we often read either about “an unexpected combination of characters seen in several different groups” (Zhang et al., 2008), or about polyphyletic origin of some important morphological traits, e.g., biramous limbs of arthropods (Daley et al., 2009). Apparently, different arthropod characters evolved independently in different lineages, a process that resulted in mosaic distribution of derived traits among basal arthropods and their ancestors (Tatarinov, 1987; Ponomarenko, 2004).
Parallel evolution of derived ‘human’ characters in different lineages of australopithecines represents another example of multiple homoplasies during a major evolutionary transition. As long as the fossil finds were scarce, human evolution was thought to be a straightforward, linear process (exactly like the origin of birds, see above).
The discoveries made during the last two decades showed that human evolution was not linear. Hominids were a diverse group, and their evolution was accompanied by multiple homoplasies (McHenry, 1994; Lockwood and Fleagle, 1999, 2007). For example, it turned out to be extremely difficult to find out which of the australopithecine species gave rise to the genus Homo. Several species currently compete for the role of the human ancestor: Kenyanthropus platyops, Australopithecus afarensis, A. africanus, and recently discovered A. sediba, to name just a few. Each of these species has some advanced ‘human’ characters combined with some ancestral ‘ape-like’ ones. Once again we see multiple homoplasies and mosaic distribution of derived characters. The recent discoveries demonstrate that 2.5 – 1.8 million years ago in South and East Africa there were several populations of advanced australopithecines that probably experienced similar selection pressures and were evolving generally in the same direction, though with different speed. Some of these populations probably could interbreed with each other, while others could have been partially or completely isolated. The borderline between ‘apes’ (australopithecines) and ‘humans’ (Homo) currently looks more blurred and arbitrary than ever (Berger et al., 2010).
Irregular echinoids (Irregularia) are a subclass of Echinoidea adapted to sediment-feeding and burrowing lifestyle, which evolved secondary bilateral symmetry. In regular echinoids (Regularia), the periproct (flexible plated membrane carrying the anus in its center) is situated on top of the test and is surrounded by ten apical plates (apical system).
In the Jurassic period in several lineages of regular echinoids the periproct started to move backwards. This trend was triggered by the evolution of sediment feeding, because it is extremely maladaptive for a sediment feeder to have periproct on top of the test: such position of the anus increases the probability of repeated swallowing of the same particles of sediment.
Posterior shift of the periproct required a large-scale transformation of the apical system. This resulted in major changes of the test architecture, because the distal edges of apical plates function as growth zones where new plates of the test are produced during ontogeny. In different lineages of sediment-feeding echinoids these transformations were different. The ancient connection between the periproct and the apical system could not be abandoned at one stroke. The necessity to shift the periproct backwards resulted in origination of several aberrant forms with various additional plates; in one lineage (Spatangacea) the apical system was ruptured into two separate parts.
Eventually the complete separation of the periproct from the apical system occurred independently in at least five separate lineages. This evolutionary transition has been called ‘exocyclization’, because apical system with periproct inside is called ‘endocyclic’, while apical system that does not surround the periproct is called ‘exocyclic’ (Solovjev and Markov, 2004).
Other spectacular paleontological examples of parallel evolution include adaptive radiations of mammals on isolated continents (e.g., Simpson, 1980),
“angiospermization”, or parallel development of characters typical of angiosperms in different lineages of the Mesozoic gymnosperms (Krassilov, 1977; Ponomarenko, 1998), and Cambrian radiations of agnostids (Trilobita) and archaeocyaths. In the last two cases the homoplasies are so abundant that some paleontologists find it more practical to use ‘periodic tables’ of species instead of conventional hierarchical classifications or tree-like phylogenetic reconstructions (Rozanov, 1973; Naimark, 2011).
Parallel evolution takes place not only during major transitions. It appears to be a common feature of adaptive radiations of any scale. The same pattern of numerous homoplasies can be observed during adaptive radiations of closely related founder populations on different islands or isolated lakes.
One beautiful example concerns the parallel evolution of finches (Nesospiza) on two small islands in the Tristan da Cunha archipelago in the South Atlantic Ocean. Each of the two small islands (Inaccessible and Nightingale) has two varieties (incipient species) of finches: one with larger beak, specialized for feeding on the large seeds of the local tree, and another with smaller beak, specialized for feeding on the small seeds of the local grass. Ornithologists originally thought, based on morphology, that small-beaked finches from both islands are closely related to each other, and so are the two large-beaked finches. However, the genetic analysis demonstrated that each of the four varieties of finches is most closely related to the second variety from the same island, rather than to the variety with the same morphology from another island. In this case, parallel speciation of finches was obviously driven by: (i) similar ecological situation on the two islands (each island has a pair of distinct ecological niches); 2) similar ancestry: both islands were originally colonized by the same species of finches from the continent (Ryan et al., 2007).
Another example is presented by the threespine stickleback (Gasterosteus aculeatus) which has diverged into two distinct forms, benthic and pelagic, independently in each of the seven different lakes in British Columbia. This parallel adaptive radiation took place over the past 10,000 years (Rundle et al., 2000; Boughman et al., 2005).
Interesting example of parallel evolution was described recently in the Aral sea which is rapidly drying out. The sea has divided into two isolated basins: the Greater and Lesser Aral. Salinity has increased greatly in both basins, driving many species to extinction, but some bivalves survived and started to evolve rapidly. All suspension-feeding molluscs died out, whereas many sediment-feeders survived. They started to diverge and to occupy vacant niches. Burrowing sediment-feeders of the genus Cerastoderma started to move onto the surface of the sea bottom and turn into suspension feeders. The shell morphology changed accordingly. Interestingly, this process went on in a similar way in both isolated basins. Now this unique natural experiment came to an end because the salinity in the Greater Aral became to high for any molluscs to survive (Andreeva and Andreev, 2003).
The reasons underlying this ubiquitous pattern of multiple homoplasies during adaptive radiations are poorly understood, although some plausible hypotheses can be proposed. The fact that parallelisms represent a common feature of major evolutionary transitions and adaptive radiations implies that evolution in general is largely constrained and canalized by multiple genetic, ecological, morpho-physiological, and developmental limitations (Beldade et al., 2002; Brakefield and Roskam, 2006).
Mutations may be random, but selection is definitely not, so it is expected that genetic and phenotypic variation in closely related populations (species) under similar selection pressures should be limited, predictable, and similar between populations. This predictability and similarity of variation reveals itself in Vavilov’s law of homologous series in heritable variation.
This pattern may be related to some kind of general law of the development of complex systems. A complex system, such as an organism (or its developmental ‘program’), a system consisting of several interconnected blocks or elements, can exist in several stable states, but the number of such states is limited, and their nature is determined by the properties of the elements and their interactions (Meyen, 1975; Tatarinov, 1987).
Evolutionary arms race between competing lineages that evolve in parallel can account for the similarity of selection pressures acting upon them, and it also may facilitate the emergence of a positive feedback in the process of parallel evolution (e.g., when a predator evolves faster gait, prey will evolve faster gait too, and so other predators will experience stronger selection pressure for faster gait) (Ponomarenko, 1998, 2004).
The mutations of regulatory genes may play an important role in major evolutionary transitions. Homeotic mutations can result in major morphological novelties. Is it absolutely inconceivable that such a mutation could instantly transform, for example, Anomalocaris, with only one pair of segmented limbs, which evolved specifically for grasping prey, into a form like Schinderhannes, with small segmented limbs on each segment of the body? And if so, could such mutation happen independently in several different lineages?
Such events probably happen sometimes. One possible example was found in the early evolution of mammals. Modern mammals have no lumbar ribs, while the earliest mammals had them. During the early evolution of mammals, lumbar ribs were lost and then sometimes acquired again, in several different lineages. It was shown experimentally that a loss-of-function mutation of the homeotic gene Hox10 restores lumbar ribs in modern mouse. It is suggested that “the loss or gain of Hox gene function to pattern the vertebral identities is a plausible mechanism for homoplasy of lumbar ribs in early mammals” (Luo et al., 2007).
There is no direct evidence of lateral gene transfer causing homoplasies during the adaptive radiations. But there is a related possibility which appears to be more plausible: episodic hybridization as means of genetic exchange between rapidly diverging lineages (Seehausen, 2004).
Studies of ongoing rapid adaptive radiations (e.g., lake Tana barbs, cichlids of the Great African lakes, etc.) showed that morphological and ecological divergence, facilitated by behavioural isolation mechanisms, may proceed much faster than the development of genetic incompatibility. For instance, lake Tana barbs started to diverge 10,000 – 25,000 years ago, and by now there are 15 distinct morphotypes, considered by some ichthyologists as 15 separate species (de Graaf et al., 2010). The morphological difference between the morphotypes is so great that, if the fossil skeletons of these fishes were found by paleontologists, they would probably describe several distinct genera or even families. However, these morphotypes still can interbreed and produce fertile hybrid offspring (although they probably seldom do it in nature) (Dzerzhinskii et al., 2007).
The same is true for the cichlids from the African Great Lakes. In some cases, even species that diverged 3 – 7 million years ago or more still can interbreed, despite of major morphological differences. Remote hybridization sometimes produces interesting morphological novelties; average amount of phenotypic novelty in F2 hybrids increases with genetic distance (Stelkens, 2009).
Was it also true for ancient adaptive radiations? Could Anomalocaris occasionally breed with Opabinia to produce a Schinderhannes-like hybrid? The idea may be less improbable than it seems at first glance.
Multiple parallelisms near the bases of many major and minor clades tend to obscure exact phylogenetic relationships between taxa and impair the applicability of classic cladistic methods based on parsimony analysis. These methods try to minimize the number of homoplasies, but if homoplasies are typical during adaptive radiations, then why should we consider a tree with minimal number of homoplasies as the most plausible of all trees?
Moreover, parsimony method can greatly exaggerate the “tree-likeness” of phylogenetic reconstructions of adaptive radiations. This point is illustrated by the recent study, in which 12 isolated experimental strains of viruses evolved under 2 distinct sets of conditions. The genomes of the evolved strains were sequenced and compared, and the common procedures of phylogenetic reconstruction unequivocally produced tree-like schemes (Paterson et al., 2010). However, in this case we know the real topology of their phylogeny: it was “star-like”, with 12 independent branches diverging from a single ancestral strain. The common stem in the tree-like reconstruction arose as a consequence of several homoplasies in the evolution of all 12 experimental lineages.
It seems perfectly plausible that some adaptive radiations (e.g., of lake Tana barbs, early arthropods during the Cambrian, and early irregular echinoids in the Jurassic) were “star-like” rather than “tree-like”. If this is so, then the attempts to obtain a ‘completely resolved’ dichotomous cladogram may be dramatically misleading.
Andreeva, S. I., and Andreev, N. I., 2003, Evolutsionnye Preobrazovaniya Dvustvorchatyh Molluskov Aralskogo Morya v Usloviyah Ekologicheskogo Krizisa [Evolutionary Changes in Bivalves of the Aral Sea Under Conditions of Ecological Crisis], Omsk State Pedagogical Univ. Press, Omsk, 382 p.
Beldade, P., Koops, K., and Brakefield, P. M., 2002, Developmental constraints versus flexibility in morphological evolution, Nature. 416:844–847.
Berger, L. R., de Ruiter, D. J., Churchill, S. E., Schmid, P., Carlson, K. J., Dirks, P. H. G. M., and Kibii, J. M., 2010, Australopithecus sediba: A new species of homo-like australopith from South Africa, Science. 328:195–204.
Boughman, J. W., Rundle, H. D., and Schluter, D., 2005, Parallel evolution of sexual isolation in sticklebacks, Evolution. 59:361–373.
Brakefield, P. M., and Roskam, J. C., 2006, Exploring evolutionary constraints is a task for an integrative evolutionary biology, Am. Nat. 168 Suppl 6:S4–13.
Briggs, D. E. G., Erwin, D. H., Collier, F. J., 1995, The Fossils of the Burgess Shale, Smithsonian Institution Press, 238 p.
Cisne, J. L., 1974, Trilobites and the origin of arthropods, Science. 186:13–18.
Chiappe, L. M., 1995, The first 85 million years of avian evolution, Nature. 378: 349–355.
Daley, A. C., Budd, G. E., Caron, J.-B., Edgecombe, G. D., and Collins, D., 2009, The Burgess Shale anomalocaridid Hurdia and its significance for early euarthropod evolution, Science. 323:1597–1600.
Dzerzhinskii, K. F., Shkil, F. N., Abdissa, B., Zelalem, W., and Mina, M. V., 2007, Spawning of large Barbus (Barbus intermedins Complex) in a small river of the Lake Tana basin (Ethiopia) and relationships of some putative species, J. of Ichthyology, 47(8):639–646.
Fedonkin, M. A., 1985, Precambrian metazoans: the problem of preservation, systematics and evolution, Phil. Trans. Roy. Soc. London. Ser. B. 311:27–45.
Fedonkin, M. A., 1998, Metameric features in the Vendian metazoans, Ital. J. Zool. 68: 11–17.
Fountaine, T. M. R., Benton, M. J., Dyke, G. J. and Nudds, R. L., 2005, The quality of the fossil record of Mesozoic birds, Proc. R. Soc. London, Ser. B. 272:289–294.
Gould, S. J., 1989, Wonderful Life: The Burgess Shale and the Nature of History, W. W. Norton, N. Y., 347 p.
Gould, S. J., 1992, The Reversal of Hallucigenia, Natural History. 101:12–20.
de Graaf, M., Megens, H.-J., Samallo, J., and Sibbing, F., 2010, Preliminary insight into the age and origin of the Labeobarbus fish species flock from Lake Tana (Ethiopia) using the mtDNA cytochrome b gene, Molecular Phylogenetics and Evolution, 54:336–343.
Hone, D. W. E., Tischlinger, H., Xu, X., Zhang, F., 2010, The extent of the preserved feathers on the four-winged dinosaur Microraptor gui under ultraviolet light, PLoS ONE. 5(2):e9223.
Hu, D., Hou, L., Zhang, L., and Xu, X., 2009, A pre-Archaeopteryx troodontid theropod from China with long feathers on the metatarsus, Nature. 461:640–643.
Krassilov, V. A., 1977, The origin of angiosperms, The Botanical Review. 43(1):143–176.
Kühl, G., Briggs, D. E. G., and Rust, J., 2009, A great-appendage arthropod with a radial mouth from the Lower Devonian Hunsrück Slate, Germany, Science. 323:771–773.
Kurochkin, E. N., 1995, Synopsis of Mesozoic birds and early evolution of Class Aves, Archaeopteryx. 13:47–66.
Kurzanov, S. M., 1987, Avimimidy i Problema Proiskhozhdenia Ptits [Avimimids and the Problem of the Origin of Birds], Nauka, Moscow, 95 p. [In Russian]
Lockwood, C. A., and Fleagle, J. G., 1999, The recognition and evaluation of homoplasy in primate and human evolution, Amer. J. Phys. Antropology. 110:189–232.
Lockwood, C. A., and Fleagle, J. G., 2007, Homoplasy in primate and human evolution, J. Hum. Evol. 52:471–472.
Luo, Z.-X., Chen, P., Li, G., and Chen, M., 2007, A new eutriconodont mammal and evolutionary development in early mammals, Nature. 446:288–293.
McHenry, H. M., 1994, Tempo and mode in human evolution, Proc. Nat. Acad. Sci. USA. 91: 6780–6786.
Мeyen, S. V., 1975; Problema napravlennosti evolutsii [The problem of directionality of evolution], in: Vertebrate Zoology, V. 7, VINITI, Moscow, pp. 66–117. [In Russian]
Naimark, E. B., 2011, One hundred species of the genus Peronopsis, Paleontol. J., in press.
Paterson, S., Vogwill, T., Buckling, A., Benmayor, R., Spiers, A. J., Thomson, N. R., Quail, M., Smith, F., Walker, D., Libberton, B., Fenton, A., Hall, N., and Brockhurst, M. A., 2010, Antagonistic coevolution accelerates molecular evolution, Nature. 464:275–278.
Ponomarenko, A. G., 1998, Paleobiology of angiospermization, Paleont. J. 32(4):325–331.
Ponomarenko, A. G., 2004, Artropodizatsia i eyo ecologicheskie posledstviya [Arthropodization and its ecological consequences], Ecosystemnye perestroiki i evolutsia biosfery. 6:7–22 [In Russian]
Romer, A. Sh., 1949, Time series and trends in animal evolution, in: Genetics, Paleontology and Evolution, G. L. Jepsen et al., eds., Princeton Univ. Press, Princeton, pp. 120–130.
Rozanov, A. Yu., 1973, Zakonomernosty Morphologicheskoi Evolutsii Arheotsiat i Voprosy Yarusnogo Raschleneniya Nizhnego Kembriya [Regularities of Morphological Evolution of Archaeocyaths and the Questions of Stage Partition of the Lower Cambrian], Nauka, Moscow, 164 p. [In Russian]
Rundle, H. D., Nagel, L., Boughman, J. W., and Schluter, D., 2000, Natural Selection and Parallel Speciation in Sympatric Sticklebacks, Science. 287:306–308.
Ryan, P. G., Bloomer, P., Moloney, C. L., Grant, T. J., and Delport, W., 2007, Ecological speciation in South Atlantic island finches, Science. 315: 1420–1423.
Seehausen, O., 2004, Hybridization and adaptive radiation, Trends Ecol. Evol. 19(4):198–207.
Simpson, G. G., 1980, Splendid Isolation: The Curious History of South American Mammals, Yale Univ. Press, New Haven, 266 p.
Solovjev, A. N., and Markov, A. V., 2004, The early evolution of irregular echinoids, in: Proceedings of the XI International Echinoderm conference, Heinzeller, T., and Nebelsick, J. H., eds., Taylor & Francis, Munich, pp. 551–556.
Stelkens, R. B., Schmid, C., Selz, O., and Seehausen, O., 2009, Phenotypic novelty in experimental hybrids is predicted by the genetic distance between species of cichlid fish, BMC Evol. biol. 9:283.
Tatarinov, L. P., 1976, Morphologicheskaya evolutsia teriodontov i obschie voprosy filogenetiki [Morphological Evolution of Theriodonts and General Questions of Phylogenetics], Nauka, Moscow, 257 p. [In Russian]
Tatarinov, L. P., 1987, Parallelizmy i napravlennost evolutsii [Parallelisms and directionality of evolution], in: Evolutsia i Biotsenoticheskie Krizisy [Evolution and Biocenotic Crises], L. P. Tatarinov and A. P. Rasnitsyn, eds., Nauka, Moscow, pp. 124–144. [In Russian]
Turner, A. H., Makovicky, P. J., and Norell, M. A., 2007, Feather quill knobs in the dinosaur Velociraptor, Science. 317:1721–1723.
Walker, C. A., Buffetaut, E., and Dyke, G. J., 2007, Large euenantiornithine birds from the Cretaceous of southern France, North America and Argentina, Geol. Mag. 144(6):977–986.
Zhang, F., Zhou, Z., Xu, X., Wang, X., and Sullivan, S., 2008, A bizarre Jurassic maniraptoran from China with elongate ribbon-like feathers, Nature. 455:1105–1108.