Optimization theory in plant evolution: an overview of long-term evolutionary prospects in the angiosperms.
I. Abstract II. Introduction III. Rationale and Methods IV. Historical Background: Outbreeding, the Prime Agent of Evolution in the Higher Plants V. Outbreeding Challenged: Autogamy and Apomixis, New Contenders in the Evolutionary Race VI. Apomixis in Tropical Trees VII. Apomixis, Dioecy and Entomophily VIII. Apomixis, Genetics and Embryology IX. Apomixis, Phylogeny and Cleistogamy X. The Current Record of Apomixis in the Higher Plants XI. Skepticism in the Way of Optimized Thinking XII. What If Phylogeny Is Second to Ontogeny? XIII. The Interface between Fitness and Long-Term Evolution XIV. Outlook XV. Concluding Remarks XVI. Literature Cited
There exists a consolidated optimized hypothesis for the flowering plants binding outbreeding, heterozygosity, genetic polymorphisms and successful long-term evolution. In other words, the rationale behind the theory of long-term evolution relies solely on the breeding system. The breeding system leads, in theory, to optimal adaptation and the plant to more successfully dealing with the hazards and stochasticism associated with natural selection. The hypothesis remains, however, largely, wishful thinking as substantial hard evidence vindicating the claim is conspicuously missing in reference texts. This critique aims at a suggestion that the assertion claiming the evolutionary advantages of outbreeding as the "chosen" mating system of plants is, to some extent, tinted with personal preferences. Bias against autogamy, and particularly against apomixis, is discernible in the literature.
III. Rationale and Methods
Hypotheses are legitimate constituents of scientific knowledge as long as verifications are carried out, however late this may take place (Popper, 1982). To some degree, this correlates with the term "value judgment" in Weberian sociology (Aron, 1967: 194); i.e., an explanation is held to be valid until it is replaced by a more suitable one--as long, of course, as there remained doubts regarding the truth of the former. In other words, the truth must prevail in the long run.
The hypothesis examined here is the deterministic association claimed to exist between outbreeding, heterozygosity and long-term evolution. A subliminal driving three in the above is whether stochastic or deterministic factors guide evolution in nature, the chronic row between the Darwinian selectionist school of evolution and the neutral theory of molecular evolution (Ayala, 1974, 1975; Dobzhansky, 1966; Dobzhansky et al., 1977; Mayr, 1983; Lewontin, 1985).
The enticing view that determinism guides the evolution of the flowering plants withstood the passage of time admirably. New admirers were seduced by the core of an argument which forecast that the adoption of outbreeding by plants, and its accruing load of heterozygosity, enhanced genetic fitness, as well as maximized the potential to deal with environmental chance.
Optimization theories are hardly testable in practice, and their role is not so much to demonstrate that organisms optimize as to foster the creation of testing methodologies (Maynard Smith, 1978). Examination of the subject suggests that there is more than just scientific interest in the polemic issue of optimization theory in the flowering plants, for personal bents can be inferred in texts.
A fairly comprehensive survey of the subject, including a limited search in the CAB abstracts, was carried out in the literature. The core of the argument is considered representative in the main body of discussion. For example, the literature on apomictic crop species is much too bulky to be treated here, but its absence did not prevent the main rationale from taking place. In addition, there is a substantial body of evidence of apomixis in terns, fungi and gymnosperms, which likewise fell outside the scope of the critique. Also, for reasons of space and because the subject would demand a treatment of its own, left out of the critique were the subjects of progressive evolution (Darwin, 1859) and the theory of punctuated equilibria (Eldredge, 1985), the latter including van Steenis's (1969) "therata theory." For those interested in this particular, many pertinent titles readily available.
In this treatment, self-pollination is selfing (carrying away of pollen to the stigma of an hermaphroditic flower--the same view as Percival's  but different from those of Faegri and van der Pijl  and Bawa et al. , for whom selling is the process of viable seed-set), self-fertilization is inbreeding and autogamy (fertilization from self-pollination), cross-pollination is outcrossing (pollen carried away to the stigma of another plant--the same definition that appears in the glossary of Bawa et al.  bat not in their text, where "outcrosser" rather than "outbreeder" is often used, as, for example, in this statement on page 16: "Tree species are largely outcrossed via self-incompatibility'), cross-fertilization is outbreeding, allogamy and xenogamy (fertilization from cross-pollination leading to amphimixis and seed-set), amphimixis is syngamy (sexual reproduction by a zygote through the fusion of male and female gametes), apomixis is agamospermy (asexual seed development without further discrimination of forms), polyembryony is the formation of more than one embryo per ovule and self-incompatibility is failure of an hermaphroditic flower to set seed with its own pollen.
IV. Historical Background: Outbreeding, the Prime Agent of Evolution in the Higher Plants
In the beginning, anthecology was restricted to finding out the purpose of floral parts and to discover how stamens and pistils interacted to produce a new plant (Baker, 1979). Soon afterward, researchers started guessing why some plants were self-sufficient reproductively while others needed the participation of a partner to produce offspring. In a prelude to what was to follow, observation and description of happenings in the external world did not suffice, as apparent deterministic trends were identified in the processes of plant reproduction. The dichotomy between appearance and reality in the plant kingdom defied thinking, and a teleological explanation was eagerly sought by investigators. Based on the observation of cultivated plants and domestic animals, Darwin (1878) reached the conclusion that, somehow, cross-fertilization must be favored in nature. The basis for this statement was his earlier announcement, based on studies of orchids, that "it is hardly an exaggeration to say that Nature tells us, in the most emphatic manner, that she abhors perpetual self-fertilization" (Darwin, 1882: 293). The statement went back to 1862, when the first edition of the work appeared. This declaration, together with the sentence "No plant self-fertilizes itself for a perpetuity of generations" (Thomas Knight, 1799, quoted in Baker, 1983), was destined to inspire an entire selectinnist school of thought bound to flourish in the second quarter of the 1900s, in the aftermath of the development of genetics. Right when Darwin disclosed his viewpoint, however, there appeared one unique communication that might have led him to pause for reconsideration. Had he noted the current literature, he would have become aware of the first record of apomixis in the angiosperms. In a nutshell, to his doorstep came the announcement that an Australian dioecious euphorbiaceous treelet called Alchornea ilicifolia had reproduced without pollen at the Royal Botanic Gardens, Kew, in 1839 (Smith, 1841; Baillon, 1865). The first official record of apomixis overlapped in time with the first declaration on optimization, passed unnoticed to it and vanished temporarily from the anthecological record.
In Darwin's footsteps, the case was made for a twofold evolutionary strategy; e.g., immediate fitness and self-fertilization associated primarily with short-term evolution, while successful long-term evolution associated itself inextricably with cross-fertilization, heterozygosity and genetic polymorphisms (Darlington, 1939, 1963; Stebbins, 1941, 1970, 1985; Huxley, 1942; Mather, 1943; Grant, 1958; van der Pijl. 1960; Percival, 1965). The possession of a heterozygotic gene pool and the manifold gene expression were regarded as the best assets with which to meet chance environmental hazards. This was particularly true for the assumed prevalence of genome heterozygosity in trees of the tropical wet forest, an endowment most needed to meet the stringent conditions imposed by unpredictable long-term evolutionary pathways (Stebbins, 1958, 1974: Proctor & Yeo, 1973; Bawa, 1974, 1980: Bawa & Opler. 1975: Levin, 1975; Arroyo, 1976; Ashton, 1976; Zapata & Arroyo, 1978; Beach & Bawa, 1980). A number of interpretations suggested how this could take place in the climax forest; e.g., pollen gene flow carried away by a diverse array of animal vectors, through recourse to mechanisms as diverse as zoophilous flowers, heterostyly, protogyny, dicliny, dioecy and self-incompatibility (Arroyo, 1976; Ashton, 1976; Zapata & Arroyo, 1978; Bawa et al., 1985a). Field studies suggested how long-distance zoophilous pollination could make up the genetic diversity of trees in tropical forests, through traplining euglossine bee pollination (Janzen, 1971; Frankie, 1975, 1976; Frankie et al., 1976), butterflies and hummingbirds (Stiles, 1978; Schemske, 1980; Webb & Bawa, 1983), moths (Appanah & Chan. 1981) and bats (Ayensu, 1974; Heithaus et al., 1974; Fleming & Heithaus. 1981). Sophisticated techniques using biochemical markers showed indisputably that some amount of outbreeding occurs in Central American forests. The work by Hamrick and Loveless (1989) demonstrated not only high genetic polymorphisms in shrubs and trees but also, through progeny tests, that outbreeding does occur in Panamanian forests. The role of nectarivoruus birds in pollination remained on hold as avian participation in the reproductive biology of tropical trees may relate more to gene dispersal through seed (Howe, 1977; Howe & Estabrook, 1977; Howe & Smallwood, 1982).
V. Outbreeding Challenged: Autogamy and Apomixis, New Contenders in the Evolutionary Race
Parallel developments in anthecology took issue with autogamy and apomixis. In principle, selfers could cope successfully with changing environments (Stebbins, 1957). However, selfing was not in high regard from the optimal evolutionary standpoint (Darlington, 1939; Stebbins, 1957). This view was best summed up through Darlington's (1963: 41) statement that "the reproductive history of the flowering plants is therefore largely an account of different ways in which they evade self-fertilization or escape from its consequences" (but see Cyril Darlington's reappraisal of the importance of apomixis in grasses in Chapman & Darlington, 1992).
A sense of disbelief arose among some of those who doubted that selfers might be at a selective disadvantage vis-a-vis outbreeders regarding long-term survival (Clausen & Hiesey, 1958; Jain, 1976). The disbelief went back to the primordia of pollination biology as some evolutionists were unconvinced of the evolutionary advantages of outbreeding (Herbert Baker, 1965, cited in Baker, 1983). So, in England, around 1876, the Reverend George Henslow diverged from Darwin after initially having sided with him. His revisionism was dictated, inter alia, by observation that some of the most successful species were autogamous weeds. Around 1891 he firmly established himself with "his championing of the sellers" (Baker, 1983: 12).
Autogamy historically enjoyed some amount of complacency as to its evolutionary potential because investigators could show indisputable evidence that it harbored genetic polymorphisms. Such a leniency did not apply to apomixis. Possibly because case studies were mainly confined to herbs, and mostly to weeds, apomixis was frequently equated with a second-class genetic system. Their members were evolutionarily doomed to extinction in the time scale of long-term evolution and the geological past since they became incapable of diversifying into novel phyletic lines. This line of thinking came to be known as "the blind alley of evolution" (Darlington, 1939; Stebbins, 1941; van der Pijl, 1960; Percival, 1965; Jain, 1976). The incapacity to form novel genetic recombinants further compounded the image of apomixis as an outcast among angiospermous breeding systems.
This framework remained unchangeable until the 1970s, when, overnight, one fact altered the prevailing long-term evolutionary historicism. The discovery of apomixis in arboreal species of the humid forests of the Far East (Kaur et al., 1978, 1986) dealt a blow to supporters of optimized plant evolution. The news caused concern among thinkers who were intent on explaining how evolution proceeds in the Tropics, the quintessential site of for biological diversity. If an announcement could become a watershed in the field of evolutionary biology of the higher plants, that certainly was it. The surprise meant that researchers were not prepared to discover agamospermy as a possible ordinary mating system occurring in that kind of environment since trees of the climax forest were thought to be prevalently outbred. Teleology was again summoned for an explanation of the significance of apomixis, "the solution to this fundamental problem, which concerns the structure of tropical forests as well as the mode of evolution of their components" (Ashton, 1988a: 627). The implications of the discovery of apomixis in tropical trees for evolutionary models fed on optimality were far-reaching. Experts intuitively knew it, and this translated into worried sentences: "The likely widespread occurrence of apomixis ... which is confirmed in a few taxa and inferred in a wide range of genera and families including important timber genera such as Shorea, and fruit trees which include Citrus, mangoes (Mangifera) and Garcinia has been a surprise, and contrary to theoretical expectations" (Bawa et al., 1990: 9). This tact led to the request that progenies be tested through allozyme studies to infer the true extent of outbreeding in the forest (Bawa et al., 1989).
The presence of arboreal apomixis in the wet forest took researchers off guard, but before long select examples from the same forest were available (e.g., Bernardo et al., 1961). In the past, examples of apomixis from Southeast Asia bad been restricted for the most part to cultivated plants like the mangosteen, especially to species of Citrus (Frost, 1926). Until the late 1970s most examples of agamospermous taxa other than weeds came from the North. In the subtropical flora of the Northern Hemisphere there had been many records of agamospecies and agamic complexes since the last quarter of the 1800s, mainly for herbs, weeds or shrubby Rosaceae (Gustaffson, 1947). In the late 1950s all that a fairly complete compilation of apomixis in angiosperms could find were 19 genera, mostly from the northern Europe (Love, 1960).
The comparatively late discovery of apomixis in the Tropics presumably caused so bewilderment because of the pervading and deeply ingrained thinking of the selectionist school of evolution; i.e., that asexual development of seeds is the pariah among breeding systems of angiosperms. However, a point often neglected by evolutionary biologists, that apomicts meet the first criterion for being victorious vis-a-vis selection and perpetuating their members through the geological ages; i.e., to leave large numbers of surviving fertile offspring. Another common charge, that apomicts are not prepared to meet the novel reaction norms demanded by a changing environment, came to be challenged by views that apparently not only the breeding system counts on adaptation and long-term persistence, as discussed later. The frequent use of two other disreputable expressions--apomixis regarded as a reproductive anomaly and an abnormal deviation from the sexual process--sounded dubious since the system, like amphimixis itself, is increasingly revealing strong embryological and genetic foundations, also mentioned later.
The dissent on apomixis recorded by Baker (1979, 1983) in the nineteenth century recurred in the twentieth century as others thought differently about the role of apomixis. A few were displeased with the way apomixis was looked at in the realm of evolutionary biology. In particular, there was dissatisfaction with theoretical assessments that denied any temporal significance to the phenomenon: "An examination of the real world suggests that agamospermy has proved to be a successful form of reproduction" (Richards, 1986: 456). The request for "examination of the real world" had been eloquently defended in a sister publication: "These opinions [on the deleterious effects of apomixis] are based not on facts but on generic conclusions and laws established in studies of organisms having sexual reproduction" (Khokhlov, 1976: 9). An unorthodox view by Khokhlov followed; namely, that genetic recombining phenomena and organic diversification may not, of necessity, be strictly bound to each other in the generation of variants: "It cannot be considered as confirmed that a recombination of genes during sexual reproduction is an essential pre-condition for progressive evolution." Khokhlov capitalized on the fact that investigators had long been familiar with ever-increasing case studies of apomixis in the angiospermous flora of European Russia (in European Russia alone between 25 and 70 percent of species belonging to "species complexes" are apomicts, according to Kupriyanov ). He used this evidence to back up three main assertions: that apomixis is not an abnormal chance deviation from sexual reproduction that necessarily culminates with extinction; that apomixis is one of the main evolutionary strategies of the flowering plants; and that apomixis is a regular mating system of the angiosperms.
VI. Apomixis in Tropical Trees
Reports of apomixis in tropical woody taxa are accumulating in the literature. Initial reports of agamospermy in arboreal genera of Southeast Asia included Lansium (Meliaceae; Bernardo et al., 1961) and Eugenia (Myrtaceae: Ashton, 1976). These were followed by reports of the possibility of higher incidence of agamospermy among rain forest trees of the Far East (Kaur et al., 1978, 1986). Subsequent studies suggested the occurrence of apomixis, much of it possibly pseudogamous in nature, in arboreal taxa as diverse as Shorea and Hopea (Dipterocarpaceae), Garcinia, Kayea and Calophyllum (Clusiaceae), Diospyros (Ebenaceae), Memecylon (Melastomataceae), Mangifera (Anacardiaceae), and Syzvgium (Myrtaceae) (Ashton, 1988b; Lughada et al., 1996). Independent field observations pointed to the existence of apomixis in the dipterocarps Shorea and Hopea (Ashton, 1977, 1979; Kaur et al., 1978, 1986; Chan, 1981).
Agamospermy has been reported for a number of neotropical woody taxa as well. The phenomenon has been documented in wild papaya (Badillo, 1971; Baker, 1976), two Clusia species in Guyana (Maguire, 1976), two Venezuelan forest trees (presumably melastomaceous; Arroyo, 1979) and Etythroxylum undulatum (Berry et al., 1991). Apomixis has been confirmed in shrubs and trees of the Brazilian Cerrado (Oliveira et al., 1992; Saraiva et al., 1996; Goldenberg & Shepherd, 1998) and in the rutaceous treelet Galipea jasminiflora from the Brazilian state of Sao Paulo (Piedade & Ranga, 1993). The prevalently neotropical malpighiaceous Peixotoa showed apomixis (Anderson, 1982). The few neotropical Bombacaceae investigated showed various mating systems: Pachyra oleaginea (Baker, 1960)--now Bombacopsis glabra (Tisserat et al., 1979) is apomict, one Eriotheca species is an outbreeder and another is a pseudogamous apomict (Oliveira et al., 1992); in Panama Quararibea asterolepis is an outbreeder and Cavanillesia platanifolia is an inbreeder as well as, possibly, a facultative apomict (Murawski et al., 1990).
VII. Apomixis, Dioecy and Entomophily
The literature records a strong association between dioecy and entomophily (Bawa, 1979; Bawa et al., 1985a, 1985b). However, cross-pollination, as a prezygotic event, is not tantamount to cross-fertilization. A plant may behave like an outcrosser without being an outbreeder, should syngamy fail to happen. Rather, the insect's deposition of pollen grains on a receptive stigmatic surface may be needed for endosperm development; i.e., a form of apomixis called pseudogamy (Gustafsson, 1947). Among others, pseudogamy has been reported for the dipterocarp tree Shorea ovalis of the humid forests of the Far East, the vector being the honey bee (Ashton, 1988a).
The observation that a considerable number of woody species in tropical forests are dioecious led authors initially to postulate an association between dioecy and outbreeding (Bawa, 1974; Bawa & Opler, 1975; Beach & Bawa, 1980). The evolution of dioecy is a source of sharp polemic between authors (Wyatt, 1983). Some authors believe that sexual selection may have directed evolution from hermaphroditism to dioecy (Willson, 1979; Casper & Chamov, 1982). Others think that dioecism may have evolved from distyly, hence leading to outbreeding, which, in this case, may not be deterministically generated (Wyatt, 1983). Almost simultaneously, the hypothesis that the evolution of dioecy meant the optimization of outbreeding met with skepticism in some circles (Bawa, 1980, 1982; Givnish, 1982; Willson, 1982). Disclaimers de-emphasizing the bond between dioecism and zoophilous outbreeding in neotropical forests (Arroyo, 1979; Bawa, 1980) followed in the wake of the discovery of apomixis in trees of the Malayan peninsula (Kaur et al., 1978). In addition, the same dioecious forest genus could show dioecious species in other biomes (Arroyo, 1979), and dioecious genera in oceanic floras showed dioecious species in mainland floras (Bawa, 1982; Baker & Cox, 1984; Cox, 1989). The above compilations run counter to rushing inductive ideas that dioecy may be disproportionately higher in tropical forests.
Overall, dioecious species account for approximately 7 percent of all flowering plants. However, the presence of dioecy in some groups, particularly among tropical trees, stands well above average. For example, Baker et al. (1983) quote that in the Barro Colorado Island flora in Panama dioecious tall trees averaged 21 percent; small trees, 7 percent; shrubs, 12 percent; and scandent plants, 8 percent. In addition, they quote that in a dry forest in Costa Rica 22 percent of the trees recorded were dioecious, in a rain forest in Nigeria 40 percent of the trees were dioecious, and in a wet Malayan forest dioecy was found in 26 percent of the arboreal local vegetation. In another instance, all 400 species of the evergreen trees of the clusiaceous forest genus Garcinia from the Far East are dioecious, while all ten species investigated so far have shown apomixis (Richards, 1990a).
The presumed association between dioecy and outbreeding is relatively short of examples. Rather, the reverse seems a possibility; e.g., evidence is mounting of a putative association between dioecy and agamospermy. The particular simultaneous evolution of sex and breeding system is a reality for increasing numbers of woody angiospermous taxa characterized by familial unrelatedness and from distinct latitudes, longitudes, biomes and selective norms. Representative examples of dioecy and agamospermy are Alchornea (Smith, 1841), Carica (Badillo, 1971), Clusia (Maguire, 1976), Viscum (putative, Barlow et al., 1978), Garcinia (Ha et al., 1988; Richards, 1990a, 1990b, 1990c; Thomas, 1997), Pandanus (Cox, 1990), Astronium (Allem, 1991), Genipa (Crestana, 1995), Ilex (Obeso, 1996), Commiphora (Gupta et al., 1996; Salomao & Allem, 2001), and Conceveiba (putative, Belem area, in the Brazilian state of Para, Ricardo de S. Secco, pers. comm., 1998). The request for more studies of the assumed correlation between apomixis and dioecism (Cox, 1989) remains.
VIII. Apomixis, Genetics and Embryology
The correlation between apomixis and polyploidy was established long ago (Mogie, 1986). More recently, a deterministic genetic character has been conferred to this association (Carman, 1997). The crossing of data suggested that apomixis may be under genetic control, as well as interacting regularly with dioecy, perennial life cycle, vegetative propagation and self-incompatibility (Asker, 1980). Much earlier on, there had been the suggestion that apomixis may be regulated by genetic and physiological phenomena (Clausen, 1954). In addition, the environment may influence the expression of apomixis; examples are known among grasses (Knox, 1967; Quarin, 1986). Recent developments stress the deterministic genetic and embryological foundations of apomixis in some of the higher plants (Mogie, 1992; Koltunow, 1993; Naumova, 1993; Carman, 1997). The assumed phylogenetic determinism of angiospermous apomixis in some angiospermous families and the discovery of apomixis in increasing numbers of arboreal taxa of the Tropics could turn out to be a breakthrough in evolutionary biology. By conferring a deterministic character on agamospermy, thus leveling it off to the category of an ordinary breeding system, cutting-edge technologies of the microdisciplines could be opening up hitherto unthinkable seminal thresholds in the biological sciences.
IX. Apomixis, Phylogeny and Cleistogamy
A relationship between apomixis and predetermined phylogenetic trend, for some groups of plants, may not be before its time. A fact suggestive of phylogenetic trend, Maguire's (1976) discovery of agamospermy in the dioecious Clusia in Guyana, was paralleled by the discovery of agamospermy in related dioecious clusiaeeous genera in the rain forests of Malaysia (Ashton, 1988b). In another instance, the discovery of apomixis in the melastomaceous Memecylon in the Malayan humid forest (Ashton, 1988b) was matched by that of agamospermy in the Indian melastomaceous Sonerila wallichii (Tisserat et al., 1979).
Additional suggestive evidence that parallelism may hold for select groups of disjunct floras recently became available with three striking discoveries in Brazilian Melastomataceae: 1) apomixis has been reported for nine shrubby melastomaceous species in the Cerrado savanna (Saraiva et al., 1996); 2) agamospermy has also been found in three melastomaceous arboreal genera, Clidemia and Miconia (Melo, 1995) and Henriettea succosa (Melo & Machado, 1996), the latter in a remnant of neotropical forest near the northeastern city of Recife; and 3) Goldenberg and Shepherd's (1998) study of eleven melastomaceous species of Cerrado resulted in the discovery of seven apomictic species, six of them belonging to the genus Miconia and one, to Leandra. all seven with viable seeds. Goldenberg and Shepherd's review lists ten melastomaceous genera with apomixis. The tribe Miconieae alone comprises 85% of all known apomicts in the family. On the basis of this fact, Goldenberg and Shepherd concluded that the trait may be phylogenetic and extrapolated the prospective existence of, 1300 agamospecies for the tribe.
All cleistogamous flowers are obligatory selfers (Uphof, 1938; Lord, 1981). However, at least one species of Araliaceae showed apomixis in cleistogamous blossoms (Elumeev, 1976). Apomixis is rampant in Rosaceae, Compositae, Ranunculaceae, Gramineae and Orchidaceae (Carman, 1997), while cleistogamy is very common in the latter two families (Uphof, 1938; Counor, 1979; Lord, 1981). The researcher Helene Ritzerow documented cleistogamy in four genera of Malpighiaceae in 1908. In cleistogams of Asticarpa she discovered that normal pollen grains are rarely produced but that fruit-set is normal; while studying embryos of A. hirtella, she discovered apomictic development (Uphof, 1938). Her findings of apomixis in cleistogams were refuted by Anderson (1980), who reported a rare modality of self-fertilization in cleistogamous flowers of Asticarpa and in three other malpighiaceous cleistogamic genera. It is peculiar that the few examples in which authors suspected the occurrence of apomixis in cleistogamic flowers were forcefully disputed by Lord (1981), who stated, authoritatively, that all known cleistogamous flowers are autogamous. It is unfortunate that although Lord compiled 56 angiospermous families and 287 species showing cleistogamy, the checklist included only 29 families.
The subject of apomixis and cleistogamy has recently experienced a revival. Long ago, in 1895, the German botanist Ernst Ule discovered cleistogams in the Brazilian melastomaceous Purpurella cleistoflora (Uphof, 1938). Percival (1965: 9) assumed that insects carried out autogamy in this melastomaceous species. She based her assumption on Ule's report that bumblebees, after the nectar, pierce the lower parts of the cleistogams and might carry out pollination. This polemic is anything but settled. For example, recent reports have presumed the formation of apomictic fruits in cleistogamous flowers of the melastomaceous genus Miconia in Brazil (Saraiva et al., 1996; Goldenberg & Shepherd, 1998).
X. The Current Record of Apomixis in the Higher Plants
As recently as 1979, examples of apomixis in the Tropics came principally from weeds or from well-documented case studies in Andropogoneae and Paniceae (Connor, 1979). As recently as 1992, Asker and Jerling's comprehensive book on apomixis included nothing on arboreal apomixis in the Tropics. This fact showed that compilative efforts were not being coordinated at all. Illustrative of this fact, Naumova (1993) recorded one anacardiaceous genus in her review; Tisserat et al. (1979), two; and Carman (1997), three.
The figures available for angiospermous apomixis differ considerably in the larger treatments. Khokhlov (1976) stated that 80 families and over 300 genera present apomixis. Tisserat et al. (1979) updated the 37 families reported by Nygren (1954) and ended with 59 families and 138 genera. Naumova (1993) tabulated only adventitious embryony and ended up with 57 families, 121 genera and 250 species. The latest comprehensive treatment (Carman, 1997) tabulated reproductive data for 348 angiospermous families and obtained the following results: apomixis exists in 33 angiospermous families; adventitious embryony exists in 53 families; polyembryony appears in 115 families and 255 genera; and anomalous reproduction was found in 163 families and 506 genera (Table I).
The possibility that the occurrence of polyembryony may regularly indicate the simultaneous presence of apomixis in the species (Hanna & Bashaw, 1987; Carman, 1997; Salomao & Allem, 2001) is a working hypothesis worth further investigation. For example, germinability tests carried out in Brasilia for 75 woody species of the Cerrado savanna and the Caatinga xerophilous vegetation from 1992 through 1999 showed polyembryony in 14 hermaphroditic and dioecious trees of the Neotropics, or about 20 percent of the total studied (Salomao & Allem, 2001). Field studies confirmed agamospermy for three of the above species, the anacardiaceous Astronium fraxinifolium (Allem, 1991), the bombacaceous Eriotheca pubescens (Oliveira et al., 1992) and the rubiaceous Genipa americana (Crestana, 1995).
Carman (1997) recorded 33 apomictic families. He regards apomixis as composed solely of two mechanisms: apospory and diplospory. Because he considers adventitious embryony a form of polyembryony, he listed separately 53 angiospermous families with this reproductive anomaly. However, a number of authors (e.g., Baker et al., 1983; Koltunow, 1993; Naumova, 1993) regard adventitious embryony as a third major process of apomixis. If this latter interpretation is followed, Carman's figures for apomixis rise to 86 angiospermous families, a figure remarkably close to that held by Khokhlov for the angiosperms: 80 families. If the reasoning advanced is true, it means that 86 out of 348 tabulated families of flowering plants contain apomictic genera and species. In addition, considering that 460 angiospermous families exist (Carman, 1997) and that reproductive data are missing for 112 of them, it is only realistic to await new announcements of anomalous angiospermous reproduction.
Forthcoming reviews will in all likelihood expand the above figures because none of the more recent examples of woody apomixis cited in this article appears in any of the discussed treatments. For instance, the recent record of apomixis in five Boehmeria species (Zang & Zhao, 1996) is significant since it amounts to 10 percent of the genus.
Future comprehensive reviews will disclose more apomictic angiospermous families, genera and species. For example, apomixis exists in the herb Phyllanthus odontadenius (Bancilhon, 1971), as well as in the herbaceous pantropical weedy species P. amarus. However, the euphorbiaceous genus Phyllanthus does not appear in any of the mentioned reviews. In another instance, Vochysia tucanorum, a tree typical of Cerrado, is a facultative apomict (Costa et al., 1992). This is the first record of apomixis in the family Vochysiaceae, which, likewise, does not appear in any treatment of anomalous angiospermous reproduction.
XI. Skepticism in the Way of Optimized Thinking
The assumed deterministic association between self-fertilization and immediate fitness lacked hard evidence (Jain, 1976). In much the same vein, the viewpoint that "when a species becomes apomictic, the maintenance of genetical variability by means of recombination comes to an end" (Percival, 1965: 198) came to epitomize an entire school of optimized thinking. This thinking, however, exposed some inconsistencies associated with optimization. Genetics, by highlighting the role of intraehromosomal phenomena, subdued the assumed prevalence of the breeding system in the release of variation, adaptation and survival. Although in principle favorable to cross-fertilization in matters of long-term evolution, Mather (1943) and Darlington (1963) were quick to point out that release of variation depends on recombination and, as such, is affected by linkage, a fact that extends even to apomicts that manage to pair chromosomes. They further argued that since gene storage may be homozygotic or heteruzygotic, release of variation depends on random segregation.
Others approached the problem with different perspectives. Thus, rather than going along unconditionally with outbreeding in the matter of plant evolution, a number of authors tried to suggest that the evolution of traits is dependent on heterozygosity. This view suited the situation of trees in climax rain forest, in which the high number of constituent species expressed a low pace of extinction, the result of action of the same selective forces over millions of years (Stebbins, 1974). The parallel view that biotic factors may prevail as selective agents inside the humid forest (Black et al., 1950) inspired ideas that the combined selective pressure put up by pests and diseases may have determined rain forest trees to value heterozygosity (Levin, 1975). This view was theoretically alluring, but an additional vexing question had surfaced in the meantime: genetic drift could prevail over selection in the tropical rain forest as a consequence of the extreme isolation of conspecific individuals. Elaboration of this point of view argued that demes in the wet forest are supposedly composed of selfers, while drift fosters genetic differentiation within, and particularly among, demes (Fedorov, 1966). If true, some amount of stochastic differentiation of local breeding populations could proceed in the wet Tropics irrespective of the effects of long-distance zoophilous gene flow and selection. The idea of drift as an agent of differentiation of small plant populations had been pioneered by Turesson (1922a), who regarded strict adaptation to the habitat as a factor enhancing leptokurtosis and culminating with ecotypic differentiation. Gene flow confined in dispersal has long been recognized as a mechanism behind the origin of races and neospecies formation (Wright, 1940, 1978; Runemark, 1970).
The extreme isolation of conspecific trees in the rain forest, often with less than one fellow individual per hectare (Black et al., 1950), led some researchers to regard such a low density as a stimulus for the evolution of selling (Baker, 1959, 1970; Whitmore, 1975). This view has been contested with antagonistic field observations, which suggests that strong-flying insects like the euglossines could cross-fertilize trees in the neotropical forest. However, euglossine bees could forage in much the same way as social bees do; that is, by staying around the same site for days and eventually causing leptokurtosis (Kroodsma, 1975).
Field research on pollination biology in Central American and Malayan forests gained momentum during the 1970s and 1980s and a consensus seemed just within sight: autogamy was on the way to limbo as outbreeding seemed prevalent in the local arboreal flora. The imbroglio led to a turnaround with the discovery of agamospermy in wet forests of the Neotropics and the Far East. These facts culminated with requests that embryological techniques be refined to properly distinguish true sellers from pseudogamous trees since seed formation may be deceptive in origin (Arroyo, 1979). A similar request expressed the worry that a major stumbling block stood in the way of reliable pollination studies; i.e., heterozygous adult trees may be using apomixis as one of their reproductive systems: "It is impossible to distinguish between self-compatibility and pseudogamy" (Bawa et al., 1990: 9).
The correlation between self-incompatibility and cross-fertilization was the next factor to be examined. Seasonal fluctuations of the environment and genetics may cause a breeding system to shift from one form to another during the same biological cycle (Fryxell, 1957: 150; Baker et al., 1983). In addition, the proclaimed association between self-incompatibility and outbreeding was not foolproof as it was known that in the same reproductive cycle, or as the season advances, self-incompatibility may give way to autogamy in some species (Stout, 1938: 282; Bateman, 1952; Fryxell, 1957; Baker et al., 1983; Wyatt, 1983). For example, the assumed prevalence and stability of self-incompatibility in trees of a Costa Rican humid forest broke down frequently for unknown reasons (Bawa, 1979). Continuing research in neotropical and paleotropical forests revealed a number of peculiarities, recorded in the following order (Baker et al., 1983): l) many tropical trees have incomplete self-incompatibility, hence self-and cross-fertilization can occur in the same tree at the same time; 2) the breeding system of some tree species may vary between individuals and populations; 3) in many tree species the incompatibility barriers are either incomplete or break down easily; 4) in a same species, some trees are self-incompatible while others are self-compatible; and 5) dioecism is more prevalent among large trees of the Tropics than in large trees of temperate regions.
XII. What If Phylogeny Is Second to Ontogeny?
The synthetic definition of evolution by Mayr (1963: 8)--"the production of variation and the sorting of the variants by natural selection"--joined views that evolution is an opportunistic phenomenon; i.e., stochasticism may ultimately prevail over determinism (Simpson, 1953; Stebbins, 1970). The selectionist view put forward by Grant (1980), following Theodosius Dobzhansky and Sewall Wright, that speciation is the fixation of new adaptive peak gene combinations, may not hold for all angiosperms, particularly if only allopatry is considered. For example, disruptive selection is assumed to lead to sympatric speciation in some groups in the animal and vegetal kingdoms (Mather, 1955; Maynard Smith, 1962, 1966; Thoday & Gibson, 1962; Thoday, 1972; Mascie-Taylor et al., 1986). If true, the rise of splitting polymorphic populations, forefathers of phenetic species, from disruptive selection, may not be such a rare phenomenon in cases of sympatry, despite Mayr's (1947) systematic denial that often what look like the same environment and the same habitat turn out, on closer examination, to be ecological niches that equate with allopatry. Sympatric speciation from disruptive selection assumes that a large number of shared morphological characters exist for some time between members of the populations. So, the view advanced is that kin polymorphisms in the same habitat may reflect key distinctive phenetic characters that do not express adaptive traits but mirror fixed collateral side effects of genetic or physiological phenomena linked to adaptedness and radiation (e.g., linkage, pleiotropism, polygenes).
In a parallel way, choice of the units of selection and the units of evolution have raised some of the most controversial debates in evolutionary biology (Brandon & Burian, 1984). Mathematical models pointed to the Mendelian population as the real workable unit of evolution (Wright, 1940), a line of reasoning espoused by some leading evolutionists (Dobzhansky, 1940, 1950; Simpson, 1953). Plant evolutionists like Ehrlich and Raven (1969) chose instead the local breeding population (deme) as the unit of evolution. In their conception, the plant species emerged composed mainly of small populations, often as small as a few square meters in size (Levin & Kerstner, 1974; Raven, 1976; Grant, 1980; Stace, 1980). Genecologists like Clausen (1951) regarded locally adapted populations of ecological races as the basic units of plant evolution. Mayr (1963) regarded the species as the unit of evolution, while Dobzhansky (1937: 419) believed that "the species is a dynamic rather than a static entity." Others followed his lead. Despite sharp discontinuities present in species, the cohesion of species could be traceable to similar selective forces operating along its entire range (Raven, 1977) or to express a primeval common ancestry (Grant, 1980). At the heart of the discussion of the unit of evolution there had always been the subject of the "species problem": the taxonomic species, the biological species and the phylogenetic (cladistic) species, a matter systematically pursued by some (e.g., Mayr, 1963, 1969, 1970, 1982; Grant, 1981, who recognized five types of species). The particular problem of the taxonomic species was best summarized by Turesson's (1922b: 344) seminal view that "the species problem is thus seen to be in a large measure an ecological problem." The sentence inspired proposals like the "polytypic species concept" for birds (Mayr, 1940, 1963, 1970) and the "multidimensional species concept" for plants (Wilkins, 1968). Other viewpoints considered the biotype ("morphs" sensu Turrill : individuals morphologically distinct within a population, with distinct genotypes) as the unit of evolution (Clausen, 1951; Stebbins, 1957). The ecotype (Turesson, 1922a, 1922b, 1925, 1930) and the ecological race (Clausen et al., 1945, 1947, 1948) were also considered units of evolution.
Others advised researchers to shy away from the issue of the unit of evolution while stressing that investigators should coordinate their efforts in order to understand the genetic structure of the deme (Carson, 1987a). A useful innovation in a much-troubled field was the proposal that the term "ecotype" should hold for "parallel ecological races with convergence of habit," while "ecological race" should be associated with geography (Bocher, 1977). This latter view had been defended earlier (Clausen et al., 1945, 1947, 1948; Grant, 1977, 1981). Ecological races were regarded by some as forerunners of new species (Clausen et al., 1945, 1947, 1948; Grant, 1963; Ehrendorfer, 1968).
Other investigators did not concern themselves with the physical individualization of units but concentrated instead on the theoretical angles of the problem. Views denying recognition to the breeding system, heterozygosity and introgression as major phenomena of any predetermined, long-term evolutionary strategy (Williams, 1966, 1975, 1979; Carman, 1997) joined forces with views denying to evolution a deterministic character (Williams, 1975, 1979; Gould & Lewontin, 1979). Critical reviews of the adaptationist program rebuffed in particular two tenets of the selectionist school of thought: that morphological characters reflect adaptedness and that speciation is only guided by natural selection (Williams, 1966; Gould & Lewontin, 1979; Carson, 1987a). Unlike what Dobzhansky and Grant defended, the alternative view stressed that recombination may produce products without apparent selective advantage; that is, without increasing the genetic fitness of the organism (Turesson, 1922a; Khokhlov, 1976; Lewontin, 1985).
As Mayr (1983: 331) put it, "the target of selection is always a whole individual, and an individual is a developmentally integrated whole." His thinking on individual selection was expanded by others, who accepted that, because selection acts on the phenotype, long-term evolution may not properly be taken into account in the daily interplay involving selective forces and the local breeding population (Solbrig, 1976a, 1976b, 1979; Dobzhansky et al., 1977; Lloyd, 1979b, 1979c; Mayr, 1983). Dissenters denied recognition to the phenotype as the unit of selection by arguing that the true repositories of the heredity material were "replicators" and "vehicles" (Dawkins, 1989); the unit of selection was the gene, although the possibility existed that the same genes could be around for millions of years, lodged either in "vehicles" or in distinct evolved phenotypes that carried the same genomes of ancestors. This perennial row between the selectionist school of thought and the neutralist theory of evolution had convinced some revisionists that much of the polemic stems from one's state of mind when he or she approaches the problem (Lewontin, 1970).
In his 1970 communication, Richard Lewontin debated an intractable issue in evolutionary biology, rival theories group selection versus individual selection. Among others, he denounced that most theoretical basis for the subject had been built up based on examples from the animal kingdom. To counter the effects of stochasticism on the demography of populations intent on evolutionary change and persistency, a shift of focus became compelling. This came with the proposal that an adequate way to counter stochasticism was for long-term evolution to depend on group selection; e.g., "evolution by differential extinction of groups" or "the differential extinction (or reproduction) of groups of individuals" (Alexander & Borgia, 1978). Of necessity, a beneficial character fixed by drift had to be present in the populations, and this trait was held directly accountable for the perpetuation of the populations through geological times (Maynard Smith, 1964; Williams, 1966; Alexander & Borgia, 1978). Hence, group selection has evolutionary consequences, since one essential feature of this type of selection is the obligatory extinction of some constituent populations over space and time (Maynard Smith, 1976). Part of the above was a natural outgrowth of views (Stebbins, 1970) that allopatric populations are not affected by the action of modal evolutionary rates since populations may be evolving at quite different evolutionary speeds. Related views (Lloyd, 1979b) argued that outbreeding has evolutionary value because it works in association with group selection. Lloyd was apparently oblivious that group selection demands spatial separation of groups and that zoophilous gene flow is mainly leptokurtic. He partly corrected this with the afterthought that "individual selection rather than group selection is primarily responsible for the evolution of the characters of populations" (Lloyd, 1979b: 604).
Another viewpoint implied that outbreeding accounts for the large number of extant flowering plants, while simultaneously asking why there were so many angiospermous species (Stebbins, 1981). However, Stebbins elaborated a chart in which he wrote: "I have listed these factors (four) in order of their importance, as I see it, on the basis of current data: 1. greater flexibility with respect to seed production. 2. seed dispersal. 3. seedling establishment. 4. pollination biology." Thus he placed the mating system in fourth place in importance in a system measuring long-term survival potentials of species. With this, a subtle but significative shift of focus took place almost unobtrusively for plant survival and persistency; i.e., plant population biology joined the breeding system and associated phenomena of reproductive biology in matters of long-term evolution. An evidence backing up Stebbins's (1981) view, up to 95 percent of all deaths happen during seed germination and seedling establishment (Hickman, 1979). Stebbins associated substantial speciation with outbreeding. However, as formerly discussed, disruptive selection may generate new taxa. Hence, in a point of view antagonistic to cladism, Cronquist (1987) argued that mother species may persist long enough in the evolutionary race after radiating into new taxa, rather than becoming extinct, as cladists claimed. If true, a stock could generate distinct stocks with their own adaptive norms while remaining alive and thus increasing organic diversity. In central Brazil, support for Cronquist's (1987) argument has been found through a notable example of sympatric speciation in the eupborbiaceous genus Manihot, in which the mother species (M. violacea) and the derived species (M. irwinii) still live in the same habitat, the Cerrado of the sierra of Pireneus in the state of Goias (Allem, unpubl, data). In closing, the pivotal role edaphics may play in causing edaphic determinism and hence differentiation has to be taken into consideration. For example, the catalytic importance of soil chemistry in the promotion of phenetic radiation is known for a number of situations (Snaydon & Davies, 1982; Kruckeberg, 1986; Bradshaw & Hardwick, 1989).
XIII. The Interface between Fitness and Long-Term Evolution
Allelic series, recombination, linkage, epistasis, pleiotropism--in short, a plethora of genetic phenomena were all potentially capable of codifying different expressions of physiological or morphological characters in distinct environments (Darlington, 1939, 1958; Clausen & Hiesey, 1958; Grant, 1963, 1964, 1975, 1977; Allard, 1975). The end result could be developmental and polymorphic differentiation of neutral traits, due to pleiotropy (Mayr, 1970: 93, 422), or could culminate with speciation (Davis & Gilmartin, 1985). More, however, was demanded if the stalemate was to be resolved; i.e., what supposedly are the characters that confer selective advantage on populations intent on outmatching and then outlasting competitors in an evolutionary time scale? Which traits more forcefully related to long-term evolution? Did everything concerned with adaptation and survival revolve around the breeding system? Those unnamed beneficial traits were adaptive traits and ultimately gave their lucky bearers the upper hand in matters of descendancy: Their offspring were bound to outdo and outlast those of their competitors. But how elastic did the transcending mold designed to evade the evolutionary hazards awaiting the plant species have to he? The crucial question defied think tanks, and a timid reply came in the form of the rationale offered for "genetic assimilation" (Waddington, 1959; Stearns, 1989) or the synonymous "threshold selection" (Mayr, 1970:110). Both ways of thinking assigned some maneuverability to the phenotype in the face of environmental stresses.
Missing from the set, however, were much-needed simulation models capable of plotting the genome's malleable capability to codify features against the selective forces represented by the hazards of changing environments and to see how the genome coped. So, part of the ongoing evolutionary struggle was fairly predictive if the view were accepted that the processes of evolution which operated in the past operate nowadays, albeit on distinct phenotypes (Stebbins's  "principle of genetic uniformitarianism"). Such processes purportedly brought evolutionary change, but there remained uncertainty as to how much of all this evolutionary change resulted from selection (Lewontin, 1974). Adaptation had always been at the heart of the evolutionary thinking, and a general-purpose model of apt fitness equated the victorious individual with successful reproduction in the habitat (Dobzhansky, 1956, 1968; Lewontin, 1957). In much the same vein, the best evolutionary strategy was that which maximized the odds of persistence vis-a-vis the environment, to the extent that individuals or populations showing wide ecological tolerance were evolutionarily dubbed "progressive" (Dobzhansky, 1968). The profile of the plant qualified to outlast present-day competitors had been enunciated and, surprisingly, did not differ significantly from the profile advanced earlier in the twentieth century by Gote Turesson in Sweden and Jens Clausen's teammates in California, on the evolutionary chances of the ecological race.
As time went by it became increasingly clear that adaptation needed more clarification if the message it meant to convey were to be meaningful at all; i.e., Dobzhansky's (1968) wording, inter alia, sounded like a concept rather than a definition. Explanations like the cohesion of the genotype (Mayr, 1983)--the individual as a whole--addressed the issue but were relatively economic on the more embracing answer being sought. In marked contrast, the genecological conception of the reaction type (Turesson, 1922a: 111) fit naturally in the context by filling a most-felt gap. The reaction type disputed space with the term "phenotype." Turesson's emphasis on the reaction type, at the expense of the term "phenotype," was not merely semantic. It carried in its scope a dynamic picture of genetic variability and genetic plasticity missed in the everyday use of genotype-phenotype. Turesson contended that a morphological detail of the plant reflects an interaction between genotype and environment, best referred to as the plant's "reaction type." For him, not all distinct phenotypic expressions of a morphological trait were necessarily adaptive. It followed that morphological characters, in marked contrast to the buffered genome background, could be considerably out of the reach of selection, particularly if dispensable and nonessential parts of the organism, and not the plant as a whole, were subject to the action of selection. If true, this could give the organism a chance to recover from diverse attacks, an inspiring viewpoint if it is remembered that the definition of "phenotype" stresses observable characteristics of an organism, thus facilitating the visualization of the effects of phenotype selection. In light of the sense-impression theory, the expression of characters of the phenotype might mimic the reaction type at work; i.e., part of the external characters observed associated with neutral plasticity and polymorphism and part reflected selectionist adaptedness.
To an extant, Turesson was ahead of his time by shifting the focus of the action of natural selection wholly to internal selection, to genomic selection, hence associating the debacle of the organism to genetic and physiological background failures, processes vital to the successes of adaptedness and the establishment of the reaction type. With this view in mind, it now became easier to conjecture about how selection might operate on workable, but not healthy, phenotypes under conditions of environmental stress; i.e., phenotype failure and death resulting from genomic failure to deliver in unfavorable circumstances. The possibility that selection mostly operates internally in a background composed of molecules, genetic material, cells, tissues and organs, all targets that a machinery composed of physiology and genetics tries to keep out of the reach of disaggregating selection by preserving the integrity of molecular relationships taking place to secure the success of adaptedness, sounds genecologically seminal. Unlike Stebbins (1968), who assigned to internal selection the prime task of eliminating deleterious mutations, Turesson regarded adaptation (adaptedness sensu Dobzhansky) as concerning solely the plant's genotypic constitution--its biochemistry and the surrounding abiotic environment. His discourse on the reaction type concentrated on internal selection, but in a fashion much different from that taken up, in passing, by George Stebbins in 1968. Turesson's view, to some extent, also intersected with those that regarded evolution as the work of integrated sets of polygenes (Mayr, 1970; Carson, 1985, 1987b). Turesson's pioneering view was that the whole genetic structure of the plant, not mere modifications in the morphology of some of its organs, enabled it to live in a given habitat. To him, the genetic makeup determined the geographical and ecological range of the species.
The closest humans came to simulating changing environments was through the realization of extensive genecological studies in Sweden and California (Turesson, 1922b; Clausen et al., 1940). Through these genecological studies, Turesson realized that only total mobilization of the genome could respond adequately to critical threatening selective external factors; i.e., the successful species was that which was capable of forming myriads of ecological races, physiological, morphological, edaphic and so forth. A species thus constituted had more chances to diversify and to become more competitive to spread and colonize distinct types of environments. Ecological races were not only raw material for selection but also, typically, forerunners of evolution (Clausen, 1951; Simpson, 1953). This meant that the need to count on eclectic genomes capable of dealing satisfactorily with usual and unusual selective scenarios, a sort of an all-out effort that would require of the plant genome such a volume of capabilities that only a few select species (and their descending evolutionary variants), would be in condition to respond adequately.
The discovery of arboreal apomixis created a new scenario in tropical anthecological research. The real possibility that tropical trees use various breeding systems is enhanced by the fact that pseudogamy and autonomous apomixis do not ordinarily occur together (Asker, 1980) and that obligatory apomixis is an event somewhat rare in plants (Asker & Jerling, 1992). Mixed breeding systems may turn out to be the best evolutionary strategy for the plant species. The capacity to manifest reversible mating systems according to chance, thus enabling different breeding systems to operate over space and time, is generally regarded as a prized bonus from the evolutionary standpoint (Darlington & Mather, 1949; Fryxell, 1957; Imam & Allard, 1965; Heslop-Harrison, 1966; Jain & Allard, 1966; Mather, 1966; Jain, 1975; Richards, 1996). Mixed breeding systems in herbaceous and arboreal species have long been known (Stout, 1938; Cuatrecasas, 1964; Randell, 1970; Carpenter, 1976; Crisp, 1976; Solbrig, 1976b; Chan, 1981; Chantaranothai & Parnell, 1994; Murawsky & Bawa, 1994; Crestana, 1995).
The early view that apomixis may aim at freezing the accumulated heterozygosity inherited from outbred ancestors (Clausen, 1954) was accepted more recently as explanation for the origin of part of the heterozygosity stored in trees of climax vegetation (Baker, 1979; Baker et al., 1983; Bawa et al., 1990). Baker et al (1983: 189) suggested that apomixis in woody taxa of tropical forests can be traced to scarcity of pollinators and to freeze heterozygosity from ancient outcrossers. The Far Eastern finds, and the almost simultaneous verification of the same phenomenon in trees of neotropical forests, led to the acknowledgment that apomixis may be a regular breeding system in the climax vegetation of tropical forests (Arroyo, 1979; Baker, 1979).
Apomixis is expected to be discovered in increasing numbers in flowering plants and to eventually become a regular companion to amphimixis. The assumption rests on the fact that the processes that determine the differentiation of embryonal structures are very similar for both systems and may lead either way (Mogie, 1992; Koltunow, 1993; Naumova, 1993). The occurrence of apomixis in trees from areas so far apart and with distinct ecologies seems not to express chance deviation from sexual reproduction but to observe developmental determinism with supraspecific phyletic implications and a bearing on long-term evolution. If true, the next locations for discoveries of arboreal angiospermous apomixis, in addition to those already reported for the bombacaceous Bombacopsis (Baker, 1960; Baker et al., 1983) and the euphorbiaceous Mallotus (Webster, 1967), are Africa's paleotropical savannas and humid forests.
The inflexibility of dioecism, traditionally associated with outbreeding, was regarded as counterproductive to the reproducing species (Lewis, 1942; Grant, 1975). At that time arboreal apomixis was unheard of. Renewed explanation of why agamospermy occurs in dioecious trees of tropical vegetation argued that recourse to this breeding system may aim at overcoming pollination hardships caused by pollinator scarcity and the large spatial isolation often existing among conspecific individuals (Maguire, 1976; Baker et al., 1983). Two examples support this view. Deterministic apomixis seems to be a fact for the Malayan dipterocarp tree Garcinia scortechinii (Thomas, 1997), known only from female specimens. Spatial isolation is a fact for female members of the anacardiaceous Brazilian Astronium graveolens, which often cluster by the tens in stands where no male is found, and may represent another instance of deterministic apomixis. Female trees of Astronium are rarely visited by flying vectors in the Cerrado; the insects prefer the mellifluous flowers of the male specimens (Allem, 1991).
Apomixis often shows up in a state of equilibrium with sexual reproduction in many species, thus making for very flexible specific breeding systems (Clausen, 1954; Nygren, 1954; Fryxell, 1957; Asker & Jerling, 1992; Richards, 1996), while most if not all apomicts are facultative sexuals (Asker & Jerling, 1992; Carman, 1997).
A link between apomixis and fitness is favored by some authors (Ornduff, 1970; Lloyd, 1979b; Ellstrand & Levin, 1980), but there is no credible evidence that this view automatically handicaps apomicts in long-term survival versus extinction. If more evidence becomes available to support the interpretation that scarcity of pollinator service fosters the adoption of autogamy (Levin, 1972) and that as population density declines autogamy increases, regardless of hitherto prevailing mating systems (Solbrig, 1979; Antonovics & Levin, 1980), then perhaps agamospermy has been overlooked by investigators as an additional device in this type of situation. If true, apomictic facultative sexuals and dioecious agamospermic species could turn out to be, from the points of view of both long-term evolution and fitness, disputing space and competing with selfers. If this holds, longevity prospects for many apomicts may not be as bleak as formerly anticipated, for they may be on a par with other outcasts.
Reversibility from obligatory inbreeding back to outbreeding is considered an evolutionarily unlikely phenomenon, hence enabling forecasts. In spite of this, however, discussion of the fortune of inbreeders vis-a-vis long-term evolution has been carefully avoided by, among others, Jain (1976). This, seemingly, represented an evasive movement that could mean a break with the conventional past; i.e., inbreeding on the brink of no longer being regarded as a second-class genetic system. In much the same tone, the adoption of apomixis by flowering plants is considered irreversible (M. A. Rozanova, cited in Khokhlov, 1976). If true, this may not, of necessity, become instrumental in their demise. Quite the contrary, on two counts: obligatory apomixis seems to be virtually absent in the angiosperms; and multiple breeding systems seem to be the norm for a number of angiospermous species.
A more radical departure from conventional thinking is the view that apomictic speciation is suspected to exist for a number of angiospermous groups. Urbanska-Worytkiewicz's (1974) thought has been strengthened by Carman's (1997) double conclusion that an apomictic evolutionary trend seems to be discernible in part of the angiosperms, while apomixis may eventually act as the springboard toward the formation of new angiospermous genera and species. Apomixis has been reported in 80 (Khokhlov, 1976) and 86 (Carman, 1997) angiospermous families, which means that the breeding system occurs in 25 percent of the 348 families tabulated so far.
XV. Concluding Remarks
Skepticism and belief about optimality punctuate the subject of long-term evolution of the flowering plants. Dicliny, for instance, is regarded as fostering cross-fertilization in Moraceae and Orchidaceae (van der Pijl, 1969; Stebbins, 1974). In the monoclinous flowers of Palmae and Araceae (van der Pijl, 1978), as well as in the Caricaceae family (Baker, 1976), however, the trait is believed to foster geitonogamy, a form of autogamy. Heterostyly could, in turn, aim at the attraction of different pollinators to regulate excessive resource competition (Lloyd, 1979a). A fact inadequately explored, inbreeders can also be rich in heterozygosity (Lewis, 1963; Allard & Kannenberg, 1968; Allard et al., 1968; Hamrick, 1979; Grant, 1981) or even outdo outbreeders in this regard (Ellstrand & Levin, 1980). Similar examples carrying dubious meanings exist in evolutionary biology.
The true meaning of heterozygosity in wild plants remains an open subject, particularly if heterozygosity is assumed as the major evolutionary agent of long-term evolution. More evidence is needed if outbreeding is to strictly relate to optimization of evolution, through the enhancement of physiological processes, the enhancement of reproductive biology, and the enhancement of increased fitness, to cite some optimized possibilities. This essay aimed at suggesting that the possession of heterozygosity per se and its association with the history of long-term evolutionary success, as advocated in the form of truism by a number of plant evolutionary biologists, is unconvincing. The sense-impression dialectic has not been adequately explored by optimization theory, and humans are impressed by what they see. Extrapolations from phenomena like inbreeding depression as known for cultivated plants or from deleterious consanguinity and endogamy as recorded for mammals are by no means certain to hold good for wild plants.
A consensus on the prospects of long-term evolution depends on satisfactory answers to critical questions. A number of pertinent questions for more convincing evidence on optimization theory in plant evolution have been already addressed, and now it is the turn of researchers to investigate them.
For example, an imaginative point of view suggests that it should be discovered first under what conditions heterozygotes prove selectively superior to homozygotes (Mayr, 1955; Dobzhansky, 1959; Brown, 1983; Scharloo, 1989). Outbreeding may, on occasion, be a handicap to the population. For example, strictly outbred maritime Danish races of Viola tricolor vanished in a few generations owing to inadequate pollinator service (Clausen, 1951). Also--a major challenge to evolutionary theory--one-third of all known flowering plants are selfers (Allard, 1975), a case to be reckoned with before hypotheses like the optimality of outbreeding can be accepted in toto. Another challenging instance, the view that "a species which reproduces apomictically or by self-fertilization is unlikely to have the same amount of variability as an outbreeding species" (Bradshaw & McNeilly, 1991: 10) was completely overshadowed by an ensuing daring viewpoint advanced by the same authors that migration, rather than progressive evolutionary change, may be the strategic plant reaction to climatic change (Bradshaw & McNeilly, 1991). This put genecology once more at the foreground of thinking on evolutionary biology.
Other antithetic instances exist. For instance, heterodox finds came to be recorded as "the heterozygosity paradox" (Wyatt, 1983, commenting on Anthony Brown's research); i.e., the discovery of considerable heterozygosity in allozyme surveys of autogamous natural populations. Similar points of view were raised by others, but this time with a bearing on the evolutionary implications of allogamy vis-a-vis reaction norms and features (Scharloo, 1989; Schlichting, 1989; Stearns, 1989). Genetic variability per se became a meaningless expression if environmental constraints to with it should interplay and conflict were not simultaneously defined (Brown, 1983); i.e., if the needed genes were not present in the heterozygous genotype. In addition, "the presence of allozyme or similar variation is not necessarily a clue to the evolutionary potential of a species, since it is only appropriate variation which can play a part, and allozyme variation does not necessarily have any influence on climatic adaptation" (Bradshaw & McNeilly, 1991: 11). This declaration was capped with the view that "in the dark age of electrophoresis, researchers did not study the functional effects of the differences between the enzyme variants found to be so abundantly present. To close the gap between genome variation and genetic variation in fitness, biologists must go upward from the molecular differences via physiological and functional effects to fitness" (Scharloo, 1989: 471).
To sum up, current knowledge lacks decisive evidence that a particular breeding system enjoys the upper hand over other systems competing for evolutionary persistence. To date, if the breeding system participates alone ill the long-term successful evolution of species, three breeding systems stand out historically in the evolutionary story of the angiosperms: outbreeding, inbreeding and apomixis. The likable plant theory binding long-term evolutionary success, outbreeding, and heterozygosity remains in the offing, hence it does not seem impudent at present to assume that all contestants seem relatively equally positioned at the starting line of the evolutionary race.
Table I Apomixis in angiosperms Families General Source 80 300 Khokhlov, 1976 59 138 Tisserat et al., 1979 36 200 Hanna & Bashaw, 1987 57 121 Naumova, 1993 33 126 Carman, 1997
XVI. Literature Cited
Alexander, R. D. & G. Borgia. 1978. Group selection, altruism, and the levels of organization of life. Ann. Rev. Ecol. Syst. 9: 449-474.
Allard, R. W. 1975. The mating system and microevolution. Genetics 79: 115-126.
--& L. W. Kannenberg. 1968. Population studies in predominantly self-pollinated species, XI. Genetic divergence among the members of the Festuca microstachys complex. Evolution 22: 51-528.
--, S. K. Jain & P. L. Workman. 1968. The genetics of inbreeding populations. Adv. Genetics 14: 55-131.
Allem, A. C. 1991. Estudo da biologia reprodutiva de duas especies florestais Aroeira e Goncalo-Alves da Regiao do Cerrado. Pesquisa em Andamento, Cenargen 2: 1-5.
Anderson, C. 1982. A monograph of the genus Peixotoa (Malpighiaceae): Taxonomic history, anatomy and morphology, chromosomes, distribution, apomixis, new taxa. Contr. Univ. Michigan Herb. 15: 1-92.
Anderson, W. R. 1980. Cryptic self-fertilization in the Malpighiaceae. Science 207: 892-893.
Antonovics, J. & D. A. Levin. 1980. The ecological and genetic consequences of density-dependent regulation in plants. Ann. Rev. Ecol. Syst. 11: 411-452.
Appanah, S. & H. T. Chart. 1981. Thrips: The pollinators of some Dipterocarps. Malaysian Forester 44: 234-252.
Aron, R. 1967. Main currents in sociological thought, 2. Pareto, Weber, Durkheim. Penguin, New York.
Arroyo, M. T. K. 1976. Geitonogamy in animal pollinated tropical angiosperms: A stimulus for the evolution of self-incompatibility. Taxon 25: 543-548.
--. 1979. Comments on breeding systems in neotropical forests. Pp. 371-380 in K. Larsen & L. B. Holm-Nielsen (eds.), Tropical botany. Academic Press, London.
Ashton, P. S. 1976. An approach to the study of breeding systems, population structure and taxonomy of tropical trees. Pp. 35-42 in J. Burley & B. T. Styles (eds.), Tropical trees: Variation, breeding and conservation. Linnaean Society Symposium Series, 2. Academic Press, London.
--. 1977. A contribution of rain forest research to evolutionary theory. Ann. Missouri Bot. Gard. 64: 694-705.
--. 1979. Some geographic trends in morphological variation in the Asian Tropics and their possible significance. Pp. 35-48 in K. Larsen & L. B. Holm-Nielsen (eds.), Tropical botany. Academic Press, London.
--. 1988a. Systematics and ecology of rain forest trees. Taxon 37: 622-629.
--. 1988b. Dipterocarp biology as a window to the understanding of tropical forest structure. Ann. Rev. Ecol. Syst. 19:347-370.
Asker, S. 1980. Gametophytic apomixis: Elements and genetic regulation. Hereditas 93: 277-393.
Asker, S. E. & L. Jerling. 1992. Apomixis in plants. CRC Press, Boca Raton, FL.
Ayala, F. J. 1974. Biological evolution: Natural selection or random walk? Am. Sei. 62: 692-701.
--1975. Genetic differentiation during the speciation process. Evolut. Biol. 8: 1-78.
Ayensu, E. S. 1974. Plant and bat interactions in West Africa. Ann. Missouri Bot. Gard. 61: 702-727.
Badillo, V. M. 1971. Monografia de la familia Caricaceae. Asociacion de Profesores, Universidad Central de Venezuela, Maracay.
Baillon, H. 1865. Sur la parthenogenese etla suppression du genre Caelebogvne. Adansonia 6: 368-379.
Baker, H. G. 1959. Reproductive methods as factors in speciation in flowering plants. Cold Spring Harbor Symposium on Quantitative Biology 24: 177-191.
--. 1960. Apomixis and polyembryony in Pachira oleaginea (Bombacaceae). Amer. J. Bot. 47: 296-302.
--. 1970. Evolution in the Tropics. Biotropica 2: 101-111.
--. 1976. "Mistake" pollination as a reproductive system with special reference to the Caricaceae. Pp. 161 169 in J. Burley & B. T. Styles (eds.), Tropical trees: Variation, breeding and conservation. Linnaean Society Symposium Series, 2. Academic Press, London.
--. 1979. Anthecology: Old testament, new testament, apocrypha. New Zeal. J. Bot. 17: 431-440.
--. 1983. An outline of the history of anthecology, or pollination biology. Pp. 7-28 in L. Real (ed.), Pollination biology. Academic Press, Orlando, FL.
--& P. A. Cox. 1984. Further thoughts on islands and dioecism. Ann. Missouri Bot. Gard. 71: 230-239.
--, K. S. Bawa, G. W. Frankie & P. A. Opler. 1983. Reproductive biology of plants in tropical forests. Pp. 183-215 in F. B. Golley (ed.), Tropical rain forest ecosystems: Structure and function. Elsevier, Amsterdam.
Bancilhon, L. 1971. Contribution a l'etude taxonomique du genre Phyllanthus (Euphorbiaceae). Boissiera 18: 1-81, pls. 1-22.
Barlow, B. A., D. Wiens, C. Wiens, W. H. Busby & C. Brighton. 1978. Permanent translocation heterozygosity in Viscum album and V. cruciatum, sex association, balanced lethals, sex ratios. Heredity 40: 33-38.
Bateman, A. J. 1952. Self-incompatibility systems in angiosperms. Heredity 6: 285-310.
Bawa, K. S. 1974. Breeding systems of tree species of a lowland tropical community. Evolution 28: 85-92.
--. 1979. Breeding systems of trees in a tropical wet forest. New Zeal. J. Bot. 17:521-524.
--. 1980. Evolution of dioecy in flowering plants. Ann. Rev. Ecol. Syst. 11: 15-39.
--. 1982. Outcrossing and the incidence of dioecism in island floras. Am. Nat. 119: 866-871.
--& P. A. Opler. 1975. Dioecism in tropical forest trees. Evolution 29: 167-179.
--, D. R. Perry & J. H. Beach. 1985a. Reproductive biology of tropical lowland rain forest trees, I. Sexual systems and incompatibility mechanisms. Amer. J. Bot. 72:331-345.
--, S. H. Bullock, D. R. Perry, R. E. Coville & M. H. Grayum. 1985b. Reproductive biology of tropical lowland rain forest trees, II. Pollination systems. Amer. J. Bot. 72: 346-356.
--, P. S. Ashton, R. B. Primack, J. Terborgh, S. M. Nor, F. S. P. Ng & M. Hadley. 1989. Reproductive ecology of tropical forest plants: Research insights and management implications. Biology International, Special Issue, 21: 1-56.
--, --& S. M. Nor. 1990. Reproductive ecology of tropical forest plants: Management issues. Pp. 3-13 in K. S. Bawa & M. Hadley (eds.), Reproductive ecology of tropical forest plants. UNESCO and the Parthenon Publishing Group, Paris and Carnforth.
Beach, J. H. & K. S. Bawa. 1980. Role of pollinators in the evolution of dioecy from distyly. Evolution 34: 1138-1142.
Bernardo, F. A., C. C. Jesena & C. C. Ramirez. 1961. Parthenocarpy and apomixis in Lansium domesticum Corr. Philippines Agricultural Review 44: 415-421.
Berry, P. E.,H. Tobe & J.A. Gomez. 1991. Agamospermy and loss of distyly in Erythroxylum undulatum from northern Venezuela. Amer. J. Bot. 78: 595-600.
Black, G. A., T. Dobzhansky & C. Pavan. 1950. Some attempts to estimate species diversity and population density of trees in Amazonian forests. Bot. Gaz. 111: 413-425.
Bocher, T. W. 1977. Convergence as an evolutionary process. Bot. J. Linn. Soc. 75: 1-19.
Bradshaw, A. D. & K. Hardwick. 1989. Evolution and stress-genotypic and pbenotypic components. Biol. J. Linn. Soc. 37: 137-155.
--& T. McNeilly. 1991. Evolutionary response to global climatic change. Ann. Bot. 67 (Suppl. 1): 5-14.
Brandon, R. N. & R. M. Burian (eds.). 1984. Genes, organisms, populations: controversies over the units of selection. MIT Press, Cambridge.
Brown, W. L. 1983. Genetic diversity and genetic vulnerability--An appraisal. Econ. Bot. 37: 4-12.
Carman, J. G. 1997. Asynchronous expression of duplicate genes in angiosperms may cause apomixis, bispory, tetraspory, and polyembryony. Biol. J. Linn. Soc. 61:51-94.
Carpenter, F. L. 1976. Plant-pollinator interactions in Hawaii: Pollination energetics of Metrosideros collina (Myrtaceae). Ecology 57: 1125-1144.
Carson, H. L. 1985. Unification of speciation theory in plants and animals. Syst. Bot. 10: 380-390.
--. 1987a. The genetic system, the deme, and the origin of species. Ann. Rev. Genet. 21: 405-423
--. 1987b. The process whereby species originate. Bioscience 37:715-720.
Casper, B. B. & E. L. Charnov. 1982. Sex allocation in heterostylousplants. J. Theor. Biol. 96:143-149.
Chan, H. T. 1981. Reproductive biology of some Malaysian dipterocarps, III. Breeding systems. Malaysian Forester 44: 28-36.
Chantaranothai, P. & 3. A. N. Parnell. 1994. The breeding biology of some Thai Syzygium species. Trop. Ecol. 35:199-208.
Chapman, G. P. & C. D. Darlington. 1992. Apomixis and evolution. Pp. 138-155 in G. P. Chapman (ed.), Grass evolution and domestication. Cambridge University Press, Cambridge.
Clausen, J. 1951. Stages in the evolution of plant species. Cornell University Press, Ithaca.
--. 1954. Partial apomixis as an equilibrium system in evolution. Caryologia (Vol. Suppl.): 469-479.
--& W. M. Hiesey. 1958. Experimental studies on the nature of species, IV. Genetic structure of ecological races. Carnegie Institute of Washington Publications 615: 1-312.
--, D. D. Keck & W. M. Hiesey. 1940. Experimental studies on the nature of species, I. Effect of varied environments on western North American plants. Carnegie Institute of Washington Publications 520: 1-452.
--, --&--. 1945. Experimental studies on the nature of species, 11. Plant evolution through amphiploidy and autoploidy, with examples from the Madiinae. Carnegie Institute of Washington Publications 564:1-174.
--, --&--. 1947. Heredity of geographically and ecologically isolated races. Am. Nat. 81: 114-133.
-- , --&--. 1948. Experimental studies on the nature of species, III. Environmental responses of climatic races of Achillea. Carnegie Institute of Washington Publications 581:1-129.
Connor, H. E. 1979. Breeding systems in the grasses: A survey. New Zeal. J. Bot. 17: 547-574.
Costa, R. B., P. Y. Kageyama & G. Mariano. 1992. Estudo do sistema de cruzamento de Anadenanthera falcata Benth. Vochysia tucanorum Mart. e Xylopia aromatica Baill. e area de Cerrado. Rev. Brasil. Sementes 14: 93-96.
Cox, P. A. 1989. Baker's Law: Plant breeding systems and island colonization. Pp. 209-224 in J. H. Bock & Y. B. Linhart (eds.), The evolutionary ecology of plants. Westview Press: Boulder, CO.
--. 1990. Pollination and the evolution of breeding systems in Pandanaceae. Ann. Missouri Bot. Gard. 77: 816-840.
Crestana, C. M. S. 1995. Ecologia da polinizacao de Genipa americana L. (Rubiaceae) na Estacao Ecologica de Moji-Guacu, estado de Sao Paulo. Revista do Instituto Florestal, Sao Paulo, 7: 169-195.
Crisp, P. 1976. Trends in the breeding and cultivation of cruciferous crops. Pp. 69-118 in J. G. Vaughan, A. J. Macleod & B. M. (3. Jones (eds.), The biology and chemistry of the Cruciferae. Academic Press, London.
Cronquist, A. 1987. A botanical critique of cladism. Bot. Rev. 53:1-52
Cuatrecasas, J. 1964. Cacao and its allies: A taxonomic revision of the genus Theobroma. Contr. U.S. Natl. Herb. 35: 379-614.
Darlington, C. D. 1939. The evolution of genetic systems. Cambridge University Press, Cambridge.
--. 1958. Evolution of genetic systems. Ed. 2. Oliver and Boy& Edinburgh.
--. 1963. Chromosome botany and the origins of cultivated plants. Ed. 2. Allen and Unwin, London.
--& K. Mather. 1949. The elements of genetics. Allen and Unwin, London.
Darwin, C. 1981 . The origin of species. Harmondsworth, England, Penguin.
--. 1878. The effects of cross and self fertilization in the vegetable kingdom. Ed. 2. John Murray, London.
--. 1882. The various contrivances by which orchids are fertilised by insects. Ed. 2. John Murray, London.
Davis, J. I. & A. J. Gilmartin. 1985. Morphological variation and speciation. Syst. But. 10: 417-425.
Dawkins, R. 1989. The selfish gene. Ed. 2. Oxford University Press, Oxford.
Dobzhansky, T. 1937. Genetic nature of species differences. Am. Nat. 71: 404-420.
--. 1940. Speciation as a stage in evolutionary divergence. Am. Nat. 74:312-321.
--. 1950. Mendelian populations and their evolution. Am. Nat. 84: 401-418.
--. 1956. What is an adaptive trait? Am. Nat. 90: 337-347.
--. 1959. Variation and evolution. Proc. Amer. Phil. Soc. 103: 252-263.
--. 1966. Are naturalists old-fashioned? Am. Nat. 100: 541-550.
--. 1968. Adaptedness and fitness. Pp. 109-121 in R. C. Lewontin (ed.). Population biology and evolution. Syracuse University Press, Syracuse, NY.
--, F. J. Ayala, G. L. Stebbins & J. W. Valentine. 1977. Evolution. W. H. Freeman, San Francisco.
Ehrendurfer, F. 1968. Geographical and ecological aspects of infraspecific differentiation. Pp. 261-296 in V. H. Heywood (ed.), Modern methods in plant taxonomy. Academic Press, London.
Ehrlich, P. R. & P. H. Raven. 1969. Differentiation of populations. Science 165: 1228-1232.
Eldredge, N. 1985. Time frames: The rethinking of Darwinian evolution and the theory of punctuated equilibria. Simon & Schuster, New York.
Ellstrand, N. C. & D. A. Levin. 1980. Recombination system and population structure in Oenothera. Evolution 34: 923-933.
Elumeev, E.A. 1976. Peculiarities of flower structure, flowering and fruiting in Eleutherococcus senticosus. Pp. 248-252 in S. S. Khokhlov (ed.), Apomixis and breeding. B. R. Sharma, transl. Amerind, New Delhi.
Faegri, K. & L. van tier Pijl. 1979. The principles of pollination ecology. Ed. 3. Oxford University Press, Oxford.
Fedorov, A. A. 1966. The structure of the tropical rain forest and speciation in the Humid Tropics. J. Ecol. 54: 1-11.
Fleming, T. H. & E. R. Heithaus. 1981. Frugivorous bats, seed shadows, and the structure of tropical forests. Biotropica 13:45-53.
Frankie, G. W. 1975. Tropical forest phenology and pollinator plant coevolution. Pp. 192-209 in L. E. Gilbert & P. H. Raven (eds.), Coevolution of animals and plants. University of Texas, Austin.
--. 1976. Pollination of widely dispersed trees by animals in Central America, with an emphasis on bee pollination systems. Pp. 151-159 in J. Burley & B. T. Styles (eds.), Tropical trees: Variation, breeding and conservation. Linnaean Society Symposium Series, 2. Academic Press, London.
--, P. A. Opler & K. S. Bawa. 1976. Foraging behaviour of solitary bees: Implications for outcrossing of a neotropical forest tree species. J. Ecol. 64: 1049-1057.
Frost, H. B. 1926. Polyembryony, heterozygosis and vhimeras in Citrus. Hilgardia 1: 365-402.
Fryxelk P. A. 1957. Mode of reproduction of higher plants. Bot. Rev. 23: 135-233.
Givnish, T. J. 1982. Outarossing Versus ecological constraints in the evolution of dioecy. Am. Nat. 119: 849-865.
Goldenberg, R. & G. J. Shepherd. 1998. Studies on the reproductive biology of Melastomataceae in "Cerrado" vegetation. P1. Syst. Evol. 211:13-29.
Gould, S. J. & R. C. Lewuntin. 1979. The spandrels of San Marco and the Panglossian paradigm: A critique of the adaptationist programme. Proceedings of the Royal Society of London, Series B. 205: 581-598.
Grant, V. 1958. The regulation of recombination in plants. Cold Spring Harbor Symposium on Quantitative Biology 23: 337-363.
--. 1963. The origin of adaptations. Columbia University Press, New York.
--. 1964. The architecture of the germplasm. John Wiley and Sons, New York.
--. 1975. Genetics of flowering plants. Columbia University Press, New York.
--. 1977. Organismic evolution. W. H. Freeman, San Francisco.
--. 1980. Gene flow and the homogeneity of species populations. Biologisches Zentralblatt 99: 157-169.
--. 1981. Plant speciation. Ed. 2. Columbia University Press, New York.
Gupta, P., K. R. Shivanna, H. Y. M. Ram & P. Gupta. 1996. Apomixis and polyembryony in the guggul plant, Commiphora wightii. Ann. Bot. 78: 67-72.
Gustafsson, A. 1947. Apomixis in higher plants, II. The causal aspect of apomixis. Lunds Universitat Arsskriet 43: 69-178, 181-370.
Ha, C. O., V. E. Sands, E. Soepadmo & K. Jong. 1988. Reproductive patterns of selected understorey trees in the Malaysian rain forest: The apomictic species. Bot. J. Linn. Soc. 97: 317-331.
Hamrick, J. L. 1979. Genetic variation and longevity. Pp. 84-113 In O. T. Solbrig, S. Jain, G. B. Johnson & P. H. Raven (eds.), Topics in plant population biology. Columbia University Press, New York.
--& M. D. Loveless. 1989. The genetic structure of tropical tree populations: Associations with reproductive biology. Pp. 129-146 in J. H. Bock & Y. B. Linhart (eds.), The evolutionary ecology of plants. Westview Press, Boulder, CO.
Hanna, W. W. & E. C. Bashaw. 1987. Apomixis: Its identification and use in plant breeding. Crop Science 27: 1136-1139.
Heithaus, E. R., P. A. Opler & H. G. Baker. 1974. Bat activity and pollination of Bauhinia pauletia: Plant-pollinator coevolution. Ecology 55: 412-419.
Heslop-Harrison, J. 1966. Reflections on the role of environmentally-governed reproductive versatility in the adaptation of plant populations. Transactions Proceedings of the Botanical Society of Edinburgh 40: 159-168.
Hickman, J. C. 1979. The basic biology of plant numbers. Pp. 232-263 in O. T. Solbrig, S. Jain, G. B. Johnson & P. H. Raven (eds.), Topics in plant population biology. Columbia University Press, New York.
Howe, H. F. 1977. Bird activity and seed dispersal of a tropical wet forest tree. Ecology 58: 539-550.
--& G. F. Estabrook. 1977. On intraspecific competition for avian dispersers in tropical trees. Am. Nat. 111: 817-832.
--& J. Smallwood. 1982. Ecology of seed dispersal. Ann. Rev. Ecol. Syst. 13: 201-228.
Huxley, J. 1942. Evolution: The modern synthesis. Allen and Unwin, London.
Imam, A. G. & R. W. Allard. 1965. Population studies in predominantly self-pollinated species, VI. Genetic variability between and within natural populations of wild oats from differing habitats in California. Genetics 51: 49-62.
Jain, S. K. 1975. Population structure and the effects of breeding system. Pp. 15-36 in O. H. Frankel & J. G. Hawkes (eds.), Crop genetic resources fbr today and tomorrow. Cambridge University Press, Cambridge.
--. 1976. The evolution of inbreeding in plants. Ann. Rev. Ecol. Syst. 7: 469-495.
--& R. W. Allard. 1966. The effects of linkage, epistasis and inbreeding on population changes under selection. Genetics 53: 633-659.
Jauzen, D. H. 1971. Euglossine bees as long-distance pollinators of tropical plants. Science 171: 203-205.
Kaur, A., C. O. Ha, K. Jong, V. E. Sands, H. T. Chart, E. Soepadmo & P. S. Ashton. 1978. Apomixis may be widespread among trees of the climax rain forest. Nature 271: 440-442.
--, K. Jong, V. E. Sands & E. Soepadmo. 1986. Cytoembryology of some Malaysian dipterocarps, with some evidence of apomixis. Bot. J. Linn. Soc. 92: 75-88.
Khokhlov, S. S. 1976. Evolutionary-genetic problems of apomixis in angiosperms. Pp. 3-17 in S. S. Khokhlov (ed.), Apomixis and breeding. B. R. Sharma, transl. Amerind, New Delhi.
Knox, P. B. 1967. Apomixis: Seasonal and population differences in a grass. Science 157: 325-326.
Koltunow, A. M. 1993. Apomixis: Embryo sacs and embryos formed without meiosis or fertilization in ovules. Plant Cell 5:1425-1437.
Kroodsma, D. E. 1975. Flight distances of male euglosssine bees in orchid pollination. Biotropica 7: 71-72.
Kruckeberg, A. R. 1986. An essay: The stimulus of unusual geologies for plant speciation. Syst. Bot. 11: 455-463.
Kupriyanov, P. G. 1986. The role of apomictic genetic complexes in the flora and the evolutionary significance of apomixis in flowering plants. Istochniki lnformatsii V Filogeneticheskoi Sistematike Rastenie. V. N. Tikhomirov, ed. Vllth Moscow Conference on Plant Phylogeny, 23-25 December, 1986. Vol. 2. Moscow: Nauka, 38-40 (in Russian).
Levin, D.A. 1972. Competition for pollinator service: A stimulus for the evolution of autogamy. Evolution 26: 668-669.
--. 1975. Pest pressure and recombination systems in plants. Am. Nat. 109:437-451.
--& H. W. Kerster. 1974. Gene flow in seed plants. Evolut. Biol. 7: 139-220.
Lewis, D. 1942. The evolution of sex in flowering plants. Biol. Rev. 17: 46-67.
Lewis, H. 1963. The taxonomic problems of inbreeders or how to solve any taxonomic problem. Regnum Vegetabile 27: 37-44.
Lewoutin, R. C. 1957. The adaptations of populations to varying environments. Cold Spring Harbor Symposium on Quantitative Biology 22: 395-408.
--. 1970. The units of selection. Ann. Rev. Ecol. Syst.1: 1-18.
--. 1974. The genetic basis of evolutionary change. Columbia University Press, New York.
--. 1985. Population genetics. Ann. Rev. Genet. 19:81 102.
Lloyd, D. G. 1979a. Evolution towards dioecy in heterostylous populations. Pl. Syst. Evol. 131: 71-80.
--. 1979b. Parental strategies of angiosperms. New Zeal. J. Bot. 17: 595-606.
--. 1979c. Some reproductive factors affecting the selection of sell-fertilization in plants. Am. Nat. 113: 67-79.
Lord, E. M. 1981. Cleistogamy: A tool for the study of floral morphogenesis, function and evolution. Bot. Rev. 47: 421-449.
Love, A. 1960. Biosysternatics and classification of apomicts. Feddes Repertorium 63: 136-148.
Lugada, E. N. & C. Proenea. 1996. A survey of the reproductive biology of the Myrtoideae (Myrtaceae). Ann. Missouri Bot. Gard. 83(4): 480-503.
Maguire, B. 1976. Apomixis in the genus Clusia (Clusiaceae): A preliminary report. Taxon 25: 241-244.
Mascie-Taylor, C. G. N., J. B. Gibson & J. M. Thoday. 1986. Effects of disruptive selection, XI. Gene flow and divergence. Heredity 57: 407-413.
Mather, K. 1943. Polygenic inheritance and natural selection. Biol. Rev. 18: 32-64.
--. 1955. Polymorphism as an outcome of disruptive selection. Evolution 9: 52-61.
--. 1966. Breeding systems and response to selection. Pp. 13 19 in J. G. Hawkes (ed.), Reproductive biology and taxonomy of vascular plants. Pergamon Press, Oxford.
Maynard Smith, J. 1962. Disruptive selection, polymorphism and sympatric speciation. Nature 195: 60-62.
--. 1964. Group selection and kin selection. Nature 201: 1145-1147.
--. 1966. Sympatric speciation. Am. Nat. 100: 637-650.
--. 1976. Group selection. Quart. Rev. Biol. 51: 277-283.
--. 1978. Optimization theory in evolution. Ann. Rev. Ecol. Syst. 9: 31-56.
Mayr, E. 1940. Speciating phenomena in birds. Am. Nat. 74: 249-278.
--. 1947. Ecological factors in speciation. Evolution 1: 263-288.
--. 1955. Integration ofgenotypes: Synthesis. Cold Spring Harbor Symposium on Quantitative Biology 20: 327-333.
--. 1963. Animal species and evolution. Oxford University Press, London.
--. 1969. The biological meaning of species. Biol. J. Linn. Soc. 1: 311-320.
--. 1970. Populations, species and evolution. Belknap Press, Cambridge.
--. 1982. The growth of biological thought: Diversity, evolution, and inheritance. Belknap Press of Harvard University Press, Cambridge.
--. 1983. How to carry out the adaptationist program? Am. Nat. 121: 324-334.
Melo, G. F. A. 1995. Biologia floral e sistema reprodutivo de cinco especies de Melastomataceae na mata de Dois Irmaos, Recife, Pernambuco. M.S. thesis, Universidade Federal de Pernambuco, Recife, Brazil.
--& I. C. Machado. 1996. Biologia da reproducao de Henriettea succosa DC. (Melastomataceae). Rev. Brasil. Biol. 56: 383-389.
Mogie, M. 1986. On the relationship between asexual reproduction and polyploidy. J. Theor. Biol. 122: 493-498.
--. 1992. The evolution of asexual reproduction in plants. Chapman and Hall, London.
Murawski, D.A. & K. S. Bawa. 1994. Genetic structure and mating system of Stemonoporus oblongifolius (Dipterocarpaceae) in Sri Lanka. Amer. J. Bot. 81: 155-160.
--J. L. Hamrich, S. P. Hubbell & R. B. Foster. 1990. Mating systems of two bombacaceous trees of a neotropical moist forest. Oecologia 82: 501-506.
Noumova, T. N. 1993. Apomixis in angiosperms: Nucellar and integumentary embryony. CRC Press, Boca Raton, FL.
Nygren, A. 1954. Apomixis in the angiosperms, II. Bot. Rev. 20: 577-649.
Obeso, J. R. 1996. Produccion de frutos y semillas en Ilex aquifolium L. (Aquifoliaceae). Anales del Jardin Botanico de Madrid 54:533-539.
Oliveira, P. E., P. E. Gibbs, A. A. Barhosa & S. Talavera. 1992. Contrasting breeding systems in two Eriotheca (Bombacaceae) species of the Brazilian Cerrados. Pl. Syst. Evol. 179: 207-219.
Ornduff, R. 1970. Pathways and patterns of evolution: A discussion. Taxon 19: 202-204.
Percival, M. S. 1965. Floral biology. Pergamon Press, Oxford.
Piedade, L. H. & N. T. Ranga. 1993. Ecologia da polinizacao de Galipea jasminiflora Engler (Rutaceae). Rev. Brasil. Bot. 16: 151-158.
Popper, K. 1982. Unended quest: An intellectual autobiography. Fontana/Collins, Glasgow.
Proctor, M. & P. Yen. 1973. The pollination of flowers. Collins, London.
Quarin, C. L. 1986. Seasonal changes in the incidence of apomixis of diploid, triploid, and tetraploid plants of Paspalum cromyorrhizon. Euphytica 35: 515-522.
Randell, B. R. 1970. Adaptations in the genetic system of Australian arid zone Cassia species (Leguminosae, Caesalpinioideae). Austral. Bot. 18: 77-97.
Raven, P. H. 1976. Systematics and plant population biology. Syst. Bot. 1: 284-316.
--. 1977. The systematics and evolution of higher plants. Pp. 59-83 in C. E. Goulden (ed.), Changing scenes in natural sciences, 1776-1976. Special Publication 12. Academy of Natural Sciences, Philadelphia.
Richards, A. J. 1986. Plant breeding systems. Allen and Unwin, London.
--. 1990a. Studies in Garcinia, dioecious tropical forest trees: Agamospermy. Bot. J. Linn. Soc. 103: 233-250.
--. 1990b. Studies in Garcinia, dioecious tropical fruit trees: The origin of the mangosteen (G mangostana L.). Bot. J. Linn. Soc. 103: 301-308.
--. 1990c. Studies in Garcinia, dioecious tropical forest trees: The phenology, pollination biology and fertilization of G hombroniana Pierre. Bot. J. Linn. Soc. 103: 251-261.
--. 1996. Breeding systems in flowering plants and the control of variability. Folio Geobotanica et Phytotaxonomica, Praha, 31: 283-293.
Runemark, H. 1970. The role of small populations for the differentiation in plants. Taxon 19: 196-201.
Salomao, A. N. & A. C. Ailem. 2001. Polyembryony in angiospermous trees of the Brazilian Cerrado and Caatinga vegetation. Acta Bot. Bros. 15:369-378.
Saraiva, L. C., O. Cesar & R. Monteiro. 1996. Breeding systems of shrubs and trees of a Brazilian savanna. Arquivos de Biologia e Tecnologia, Sao Paulo, 39: 751-763.
Scharloo, W. 1989. Developmental and physiological aspects of reaction norms. Bioscience 39: 464-471.
Schemske, D. W. 1980. Floral ecology and hummingbird pollination of Combretum farinosum in Costa Rico. Biotropica 12: 169-181.
Schlichting, C. D. 1989. Phenotypic integration and environmental change. Bioscience 39: 460-464.
Simpson, G. G. 1953. The major features of evolution. Columbia University Press, New York.
Smith, J. 1841. Notice of a plant which produces perfect seeds without any apparent action of pollen. Transactions of the Linnaean Society of London 18: 509-512, table 36.
Snaydon, R. W. & T. M. Davies. 1982. Rapid divergence of plant populations in response to recent changes in soil conditions. Evolution 36: 289-297.
Solbrig, O. T. 1976a. On the relative advantages of cross and self-fertilization. Ann. Missouri Bot. Gard. 63: 262-276.
--. 1976b. Plant population biology: An overview. Syst. Bot. 1: 202-208.
--. 1979. A cost-benefit analysis of recombination in plants. Pp. 114-130 in O. T. Solbrig, S. Jain, G. B. Johnson & P. H. Raven (eds.), Topics in plant population biology. Columbia University Press, New York.
Stace, C. A. 1980. Plant taxonomy and biosystematics. Edward Arnold, London.
Stearns, S. C. 1989. The evolutionary significance of phenotypic plasticity. Bioscience 39: 436-445.
Stebbins, G. L. 1941. Apomixis in the angiosperms. Bot. Rev. 7: 507-542.
--. 1957. Self-fertilisation and population variability in the higher plants. Am. Nat. 91: 337-354.
--. 1958. Longevity, habitat, and release of genetic variability in the higher plants. Cold Spring Harbor Symposium on Quantitative Biology 23:365-378.
--. 1968. Integration of development and evolutionary progress. Pp. 17-36 in R. C. Lewontin (ed.), Population biology and evolution. Syracuse University Press, Syracuse.
--. 1970. Variation and evolution in plants: Progress during the past twenty years. Pp. 173-208 in M. K. Hecht & W. C. Steere (eds.), Essays in evolution and genetics in honor of Theodosius Dobzhansky: A supplement to evolutionary biology. Appleton-Century-Crofts, New York.
--. 1972. Ecological distribution of centers of major adaptive radiation in angiosperms. Pp. 7-34 in D. H. Valentine (ed.), Taxonomy, phytogeography and evolution. Academic Press, New York.
--. 1974. Flowering plants: Evolution above the species level. Edward Arnold, London.
--. 1981. Why are there so many species of flowering plants? Bioscience 31: 573-577.
--. 1985. Polyploidy, hybridization, and the invasion of new habitats. Ann. Missouri Bot. Gard. 72: 824-832.
Stiles, F. G. 1978. Temporal organization of flowering among the hummingbird food-plants of a tropical wet forest. Biotropica 10: 194-210.
Stout, A. B. 1938. The genetics of incompatibilities in homomorphic flowering plants. Bot. Rev. 4: 275-369.
Thoday, J. M. 1972. Review lecture: Disruptive selection. Proceedings of the Royal Society of London, Series B, 182: 109-143.
--& J. B. Gibson. 1962. Isolation by disruptive selection. Nature 193: 1164-1166.
Thomas, S. C. 1997. Geographic parthenogenesis in a tropical forest tree. Amer. J. Bot. 84: 1012-1015.
Tisserat, B., E. B. Esan & T. Murashige. 1979. Somatic embryogenesis in angiosperms. Horticultural Review 1: 1-78.
Turesson, G. 1922a. The species and the variety as ecological units. Hereditas 3: 100-113.
--. 1922b. The genotypical response of the plant species to the habitat. Hereditas 3: 211-350.
--. 1925. The plant species in relation to habitat and climate. Hereditas 6: 147-236.
--. 1930. The selective effect of climate upon the plant species. Hereditas 14: 99-152.
Turrill, W. B. 1964. Plant taxonomy, phytogeography, and plant ecology. Pp. 187-224 in J. G. W. B. Turrill (ed.), Vistas in botany. Macmillan, New York.
Uphof, J. C. T. 1938. Cleistogamic flowers. Bot. Rev. 4: 21-49.
Urbanska-Worytkiewiez, K. 1974. L'Agamospermie, systeme de reproduction important dans la speciation des angiospermes. Bulletin de la Societe Botanique de Prance 121: 329-346.
Van der Pijl, L. 1960. Ecological aspects of flower evolution, 1. Phyletic evolution. Evolution 14: 403-416.
--. 1969. Evolutionary action of tropical animals on the reproduction of plants. Biol. J. Linn. Soc. 1: 85-96.
--. 1978. Reproductive integration and sexual disharmony in floral functions. Pp. 79-88 in A. J. Richards (ed.), The pollination of flowers by insects. Linnaean Society Symposium Series, 6. Academic Press, London.
Van Steenis, C.G.G.J. 1969. Plant speciation in Malesia, with special reference to the theory of non-adaptive saltatory evolution. Biol. J. Linn. Soc. 1: 97-133.
Waddington, C. H. 1959. Canalization of development and genetic assimilation of acquired characters. Nature 183: 1654-1655.
Webb, C. J. & K. S. Bawa. 1983. Pollen dispersal by hummingbirds and butterflies: A comparative study of two lowland tropical plants. Evolution 37: 1258-1270.
Webster, G. L. 1967. The genera of Euphorbiaceae in the southeastarn United States. J. Arnold Arb. 48: 303-361, 363-430.
Whitmore, T. C. 1975. Tropical rain forests of the Far East. Clarendon Press, Oxford.
Wilkins, D. A. 1968. The scale of genecological differentiation. Pp. 227 239 in V. H. Heywood (ed.), Modern methods in plant taxonomy. Academic Press, London.
Williams, G. C. 1966. Adaptation and natural selection: A critique of some current evolutionary thought. Princeton University Press, Princeton, NJ.
--. 1975. Sex and evolution. Princeton University Press, Princeton, NJ.
--. 1979. The question of adaptive sex ration in outcrossed vertebrates. Proceedings of the Royal Society of London, Series B, 205: 567-580.
Willson, M. F. 1979. Sexual selection in plants. Am. Nat. 113: 777-790.
--. 1982. Sexual selection and dicliny in angiosperms. Am. Nat. 119: 579-583.
Wright, S. 1940. Breeding structure of populations in relation to speciation. Am. Nat. 74: 232-248.
--. 1978. Evolution and the genetics of populations, 1. Variability within and among natural populations. University of Chicago Press, Chicago.
Wyatt, R. 1983. Pollinator-plant interactions and the evolution of breeding systems. Pp. 51-95 in L. Real (ed.), Pollination biology. Academic Press, Orlando, FL.
Zang, G. G. & T. N. Zhao. 1996. Preliminary report of agamic complex in Boehmeria. China's Fibre Crops 1:19 (in Chinese).
Zapata, T. R. & M. T. K. Arroyo. 1978. Plant reproductive ecology of a secondary deciduous tropical forest in Venezuela. Biotropica 10: 221-230.
ANTONIO C. ALLEM,
Embrapa Recursos Geneticos e Biotecnologia
70849-970 Brasilia, DE Brasil
|Printer friendly Cite/link Email Feedback|
|Author:||Allem, Antonio C.|
|Publication:||The Botanical Review|
|Date:||Jul 1, 2003|
|Previous Article:||How Paleozoic vines and lianas got off the ground: on scrambling and climbing Carboniferous--early Permian pteridosperms.|
|Next Article:||Vegetation ecology and conservation on Tuscan ultramafic soils.|