Staminodes: Their morphological and evolutionary significance.
For the majority of the angiosperms the functional stamen is differentiated into a basal supportive part, namely the filament, and upper microsporangia-bearing tissue, namely the anther (see, e.g., D'Arcy, 1996; Endress, 1994; Endress & Stumpf, 1990, 1991; Hufford & Endress, 1989; Weberling, 1989). Each anther consists of two equivalent halves, the thecae, joined together and with the filament by a connective. Each theca is built up of two pollen sacs (microsporangia), dehiscing in various ways. When stamens fail to develop into the above-mentioned sporogenous structures but retain the same characteristics of microsporophylls, they are usually referred to as sterile stamens or staminodes (e.g., Eames, 1961; Weberling, 1989).
Different definitions can be applied for staminodes. For Watson and Dallwitz (1992-), a staminode is a sterile stamen, or a modified structure identifiable as such, borne in the androecial region of the flower. It may be merely imperfect, vestigial, or specialized (e.g., petaloid or nectariferous). For Mione and Bogle (1990: 78), studying Hamamelidaceae, staminodes are "sterile floral appendages which are most certainly derived from stamens, i.e., appendages which are morphologically similar to stamens but are sterile."
The identification and description of a staminode often remains vague and arbitrary, and may overlap a whole range of different structures: It is an abstraction of something that is neither a stamen (except for those cases with clearly abortive anthers), nor a petal proper, nor any other clearly distinguishable organ. It is clear that, when a stamen aborts, the resulting structure should obviously be called a staminode. This is important for recognizing evolutionary trends in flowers, as the staminode represents a transitional phase from one category of organs to a totally different structure. Difficulties arise when there is absolutely no resemblance between the sterile structure and the fertile stamen. To interpret staminodes in an unequivocal way-like any other floral organ-necessitates a clear-cut approach of homology. For this purpose positional homology should be of major importance in the study of the morphology of staminodes. Staminodes may occur in the same whorl as fertile stamens, as a result of nu tritional limitations (e.g., Baillon, 1862b; Fukuoka et al., 1986) or as the result of a zygomorphic development of the flower (e.g., in many Scrophulariaceae: Endress, 1998, 1999; Reeves & Olmstead, 1998; Table II). An organ that shows no resemblance whatsoever to a stamen may be homotopic; that is, it takes the space in the flower usually reserved for members of the androecium but lacks all resemblance with stamens on a structural ground, even being restricted to vascular bundles, or is totally different in physionomy. The question is whether that organ can always be considered homologous with the stamen. This question cannot be answered positively in all cases, as the homology criteria of Remane (1952, in Sattler, 1994) remain arbitrary. The similarity criterion proposed by Patterson (1982), referring to a combination of topographic, ontogenetic, and compositional homology, was partly taken over by Albert et al. (1998), who distinguished between historical (having a single origin on a phylogenetic tree), p ositional (originating from the same organs), and process homology (having arisen by the same genetic process). The three definitions of homology used by Albert et al. 1998 (also called "orthology") apply to separate organismic levels (organisms, organ primordia, and genes) and may have different applications when discussed for the different levels separately. Ontogenetic homology, referring to a similar ontogeny of stamen and potential staminode (e.g., Kluge, 1988; Nelson, 1978) is another approach combining the historical, positional and process homology, where the staminode is a specialization appearing at one stage in the ontogeny of an organism. According to Sattler (1994) a 1:1 correspondence between structures that is the theoretical (static) criterion for homology is untenable and oversimplified, because of transformations of structures during development ("developmental hybridization") and the occurrence of homeosis, which may be partial or complete. Characters must be compared at all stages of devel opment, and because they eventually become transformed, partial correspondences and multiple relations must be taken into account. This leads to conflicts of homological interpretation, which are only resolved by a dynamic approach of morphology.
The definition of staminodes also implies the presence of heterotopic structures. A typical example of heterotopic staminodes are petals, if petals are considered a category different from the androecium. There is a broad literature covering the homologous nature of petals with stamens, as the subject has fascinated botanists since Goethe (see Weberling, 1989, for an overview). It is undeniable that petals often represent structures reminiscent of stamens and that there is a strong vascular and ontogenetic correlation between the petals and the stamens (see also Albert et al., 1998; Eames, 1931; Endress, 1994; Weberling, 1989). Staminodes can be seen as partially homeotic mutations. They develop from normal stamen primordia but have undergone altered developmental processes and patterns (Li & Johnston, 2000). The development of petals has gone a step farther by the onset of a novel developmental pathway. We could term this transformation from stamen to staminode, and to petal, "serial homeosis," but not in th e sense of Takahashi (1994). Takahashi (1994) proposed this term for the homeotic process occurring in the apetalous flower of Trillium apetalon (Trilliaceae), where there is a serial replacement of organ whorls from the center of the flower to the periphery.
Structures in flowers have often been described as staminodes either because of their superficial resemblance to stamens or because of their spatial association with the stamens. Indeed, it is sometimes very difficult to distinguish between structures that look like staminodes but are not homologous with stamens and those that are derived from stamens. As those structures have often been described as staminodes in the literature, the resulting misinterpretations can have far-reaching consequences for the definition of character states used in data matrices, and they can mislead hypothetical semophyleses of the androecium. It is clear that :he interpretation of staminodial structures meets the same difficulties as the definition of the nature of nectaries and demands a clear-cut characterization (e.g., Ronse Decraene & Smets, 1991c; Smets, 1986, 1988a, 1988b; Smets & Cresens, 1988).
Walker-Larsen and Harder (2000) recently presented a handsome survey of staminodial structures in the angiosperms. They discussed the possible origins of staminodial structures as the result of reductive processes in the androecium using the phylogenetic framework of angiosperm evolution presented by Chase et al. (1993). Patterns of staminode formation are intricately linked to patterns of evolution of whole floral structures. Therefore, staminodes will have different positions and functions in acyclic magnoliids, polysymmetric rosids, or zygomorphic asterids. The authors point to the functional integration of staminodes in the flower of many groups, as we will also discuss below. Shortcomings of their approach are caused by their reliance on literature citations about staminodes and also on certain shortcomings of the phylogenetic hypotheses they use to discuss staminode evolution.
In this article we present a survey of the occurrence of staminodial structures and their functionality in the flower and give an overview of possible misinterpretations of staminodes and their relevance in morphological studies. The difficulty of definition of a staminode may rest on uncertainty in interpreting the wide array of emergences on the floral receptacle. Therefore, a global morphological study, relying on floral anatomy, ontogeny, and external morphology, is needed to clarify this question. We consider staminodes only in hermaphroditic flowers, for the same reasons as given by Walker-Larsen and Harder (2000), because the origin and scope of these staminodes is different for unisexual flowers.
III. Possible Origins for Staminodes
Staminodes appear relatively early in the fossil record, and the same variations as in modern angiosperms seem to have been present since the Turonian. Apart from magnoliid fossils having inner and outer staminodes, there is an abundance of eudicots having one whorl of sterile stamens. They occur in Hamamelidae as a whorl alternating with antesepalous stamens, suggesting their homology with petals and as "a transitional stage between apetalous and petalous flowers" (Grepet & Nixon, 1996: 37). Also in Gapparales-like fossils, such as Dressiantha, five setiform staminodes alternate with five stamens (Gandolfo et al., 1998). Crepet & Nixon (1996) report the presence of antepetalous staminodial nectaries in flowers of Ericalean/Ebenalean affinity. They also suggest that staminodes are responsible for the derivation of nectaries and petals within the rosid-hamamelid complex by a "division of labor" in the stamens. In the ranunculids, staminodes and petals have been derived several times from stamens in separate li neages (see Drinnan et al., 1994).
Several functional explanations have been given for the origin of staminodial structures in flowers linked to evolutionary modifications of flowers (see Walker-Larsen & Harder, 2000). Staminodes may result from nutrient limitations, alterations in the construction of flowers, or adaptations to pollinators. However, different factors may contribute en masse to the elaboration of staminodes.
For obvious nutritional limitations, an entire whorl of stamens may become reduced or may completely disappear. This is illustrated by Rodriguez-Riano et al. (1999) in southwest European Fabaceae, where the incidence of reduced diadelphous androecia is correlated with an autogamous syndrome. In several taxa an inner stamen whorl (usually the antepetalous whorl) may be present, vestigial, or even absent within a single species or between different species of the same genus (Table I). Very often the upper flowers of a racemose inflorescence will not attain full development, leading to a partial sterilization of whorls. This process of reduction, once settled genetically, has affected several lineages of the angiosperms and has arisen several times independently (see Walker-Larsen & Harder, 2000). The process is consistent with the fossil record with the profusion of taxa that have apparent staminodial nectaries (Crepet & Nixon, 1996). The shape of these staminodes is characteristically stublike or sometimes not exceeding the stage of primordium (e.g., Figs. 2-3, 17-19, Myrsine africana, Samolus valerandi, Moringa, Linum). Walker-Larsen and Harder (2000) consider such nonfunctional staminodes temporary and doomed to be lost quickly. However, such structures may have a function in the flower that we do not grasp at this moment.
The trends in the reductive process of stamens (staminode origin) are understood as a semophyletic sequence that can still be traced in certain groups of plants. A reduction in size of stamens, correlated with a retardation of initiation of primordia, can be seen as a first obvious step in the process (Fig. 54). That one of two whorls is often retarded developmentally in diplostemonous flowers has been illustrated (see Ronse Decraene, 1992; Ronse Decraene & Smets, 1995a, 1998; Walker-Larsen & Harder, 2000). The occurrence of obdiplostemony with positional shifts of stamens is one of the mechanisms bringing about the reduction of one whorl in correlation with limitations in time and space for development (Ronse Decraene & Smets, 1995a). These reductive trends have phylogenetic implications as they are correlated with the configuration of the androecium in the eudicots: Diplostemony predominates, but there is a global trend to haplostemony or obhaplostemony. In some genera, species with staminodes coexist with species that have lost staminodes altogether (e.g., Linum, Hesperolinon: Narayana & Rao, 1976a, Samolus: Cans, 1998; Sattler, 1962).
In the monocots similar reductive trends are operating. The Zingiberales are a classic example of the semophyletic sequence in stamen reduction from an original dicyclic androecium running in a continuous sequence (the reductive process is represented with symbols used for floral formulas; A refers to the androecium, the numbers refer to the number of stamens in a whorl, and the raised circle refers to staminodes): Musaceae [A3+3 or A3+2(1[degrees]) / Heliconiaceae [A2(1[degrees])+3] - Lowiaceae / Strelitziaceae (A3+2) - Zingiberaceae [A2[degrees]+ 1(2[degrees])] / Marantacene [A1[degrees]/2[degrees]/0+1(2[degrees])] - Costaceae [A3[degrees]+1(2[degrees])] - Cannaceae [A2[degrees]+(2[degrees])] (see, e.g., Kirchoff, 1991; Kress, 1990). These reductions are correlated with a trend from small vestigial organs to specialized pollination mechanisms (pollinator attraction, trigger mechanisms, pollinator guidance: Endress, 1994, Walker-Larsen & Harder, 2000). The process of staminode formation must be seen as the o ngoing interaction of heterochrony and heterotopy. Heterochrony changes the developmental timing and rate of development of the organ, without changing the developmental direction; heterotopy changes the nature of the organs formed, not the timing and rate of morphogenesis (Li & Johnston, 2000).
In a first evolutionary step staminodes are incidental and must be seen as a response of the flower to a changing external or internal condition. For example, a trend to zygomorphy induces one side of the flower to become retarded in its development relative to the opposite side (e.g., Leguminosae: Tucker, 1984, 1996; Moringaceae: Ronse Decraene et al., 1998a). This leads to a retardation versus stagnation in inception of part of the androecium and finally to its sterilization or abortion. This shift to sterility can run from the adaxial side to the abaxial side; Emblingia has four adaxial stamens opposite the petals and four abaxial staminodes (Erdtman et al., 1969). The opposite occurs in Dactylaena and Euadenia (Gapparaceae), in which four adaxial staminodes fuse into a stalked appendage facing the single abaxial stamen (Figs. 4-5; Karrer, 1991). Different reductive trends may be correlated in one flower, as in Moringa (Ronse Decraene et al., 1998a). A generalized feature in the genus is that the antepetal ous stamen whorl is reduced to stublike staminodes with no obvious function. The flower also develops a strong oblique zygomorphy. As a result, one of the staminodes is much smaller than the others and is sometimes absent (Figs. 1-2; see Ronse Decraene et al., 1998a). The duality between two groups of stamens (heteranthery) is a frequently recurring pattern related to pollination and is one probable origin of staminodes in zygomorphic flowers. Pollen flowers with some feeding stamens and only a small number of larger pollinating stamens are characteristically arranged in two opposing groups in a monosymmetric pattern (see Endress, 1999; Vogel, 1978). The feeding stamens either produce either nonviable pollen or ultimately become completely sterile.
In a second step (probably simultaneously with the loss of pollen-producing activities), these retarded organs can become transformed (by a reversal of the original strictly reductive trend) and may gain another function in the flower. Statements of functionality versus nonfunctionality are sometimes dubious, as very little is known of the floral biology of flowers. Such alternations of trends are clearly very dynamic and are related to several internal (e.g., the degree of sterilization, occurrence of homeosis) or external factors (e.g., pollinator- flower relationships). It is essential that a genetic basis exist for a reprogramming of a moribund staminodial structure; otherwise, the staminode is doomed to disappear quickly. However, one can argue whether a stublike structure should still be present in flowers when it has no obvious function. At the same time it is an indication of an ongoing evolutionary process (He [beta], 1983). In the zygomorphic Scrophulariaceae the adaxial staminode can be variously d eveloped, may vanish completely, or be sometimes conspicuous (see Endress, 1998, 1999; Reeves & Olmstead, 1998; Walker-Larsen & Harder, 2000).
Table II illustrates the strong link between zygomorphy and the occurrence of adaxial staminodes spread over different genera Notable exceptions with anterior staminodes are Emblingia (Erdtman et al., 1969), Lopezia (Eyde & Morgan, 1973), and Pelargonium (Kumar, 1976; Sattler, 1973). The siting of staminode formation is linked to such factors as the orientation of the flower on the inflorescence and the type of visiting pollinator. The development of staminodes in zygomorphic flowers is independent of the number of stamen whorls, reflecting a different gene expression.
IV. A Redefinition of Staminodial Structures
A. SURVEY OF THE PROBLEM: SOME CASE STUDIES
Character research implies that characters are selected for their systematic value and their consistency (absence of homoplasy) in phylogenetic analyses. As staminodes are integral part of the androecium, they are involved in definitions, related to topologies of the androecium (see, e.g., Ronse Decraene & Smets, 1993, 1995a, 1998). Predominating ideas or theories of floral evolution have influenced the approach of the androecium and staminodes in particular. As staminodes are by definition vestigial structures, they have been linked to a reductive process in flowers, along the lines of the ranalean theory. This has repercussions on descriptions of "vascular stubs" that occur on the receptacle and that are related with preexisting staminal whorls. Also, reports on the presence versus absence of vascularization in staminodes are often contradictory. In this article we demonstrate that staminodes can be confounded with all sorts of structures. Therefore, a careful study of their nature is required, involving an atomical, ontogenetic, and--if possible--genetic investigations. It should be stressed here that the circumscription of floral characters in a hierarchical ordering of characters and character-states remains a necessity for all phylogenetic analyses and that staminodes must also be approached in this way (Table I). The difficulty in interpreting staminodial structures can be illustrated by following examples:
1. The flower of the monotypic genus Corynocarpus J. R. & G. Forst. (Corynocarpaceae) has a perianth differentiated as five sepals and petals and an androecium of five antepetalous stamens, alternating with antesepalous scales bearing a ventral nectary (Figs. 7-8). In the past, different interpretations have been given for the petaloid antesepalous scales. Krause (1960), among others, considered the scales staminodes and the nectaries opposite these scales a disk (namely, receptacular outgrowth). Narayana et al. (1986) interpreted the scale and nectary as part of a single staminodial structure; the nectary represents the modified anther part and the petal-like scale equals the transformed connective. Another interpretation, formulated by Philipson (1987), represented the petaloid scales as equivalent to petals and the nectaries as staminodes, not as a disk. However, he admitted that the common vasculature of scale and nectary could be an indication of a strong relation between "petal" and scale. Nonetheless, scarce ontogenetic evidence seemed to suggest that "it is not possible to distinguish between the primordia of petals and scales, nor between those of stamens and nectaries" (Philipson, 1987:13). In the absence of any clear evidence, Philipson opted for avoiding the use of the term "staminode" and preferred terms such as "petaloid scale" and "nectary" instead. The interpretation of Krause (1960) is supported by the fact that the nectary is inserted lower on the hypanthial slope and has no vascular connection (Ronse Decraene, unpubl. obs.). A thorough ontogenetic study should give more evidence of the nature of staminode and nectary, as the morphological evidence is still uncertain.
(2.) Harungana madagascariensis (Glusiaceae) and several other Clusiaceae possess antepetalous stamen fascicles alternating with five indented nectary scales (Fig. 9). Although the scales arise relatively late in ontogeny and have occasionally been interpreted as receptacular emergences (e.g., Leins, 1964), they are supplied by massive bundles alternating with the stamen traces and look superficially like anthers (Fig. 9). On this evidence Ronse Decraene & Smets (1991a) concluded that the nectaries represent a second staminodial whorl. In other taxa in which the presence of staminodes is beyond discussion (e.g., Samolus in Primulaceae) the staminodes also arise at a very late stage of development but are without vascular tissue (Caris, 1998; Ronse Decraene & Smets, 1995a).
(3.) Greyiaceae and Francoaceae have been shown to be closely related on the basis of rbcL data (e.g., Chase et al., 1993; Morgan & Soltis, 1993) and ontogenetic evidence (e.g., Ronse Decraene & Smets, 1999). Both families have interstaminal emergences that resemble staminodes and have been described repeatedly as such in the literature (e.g., Bensel & Palser, 1975b; Cronquist, 1981; Dahlgren & van Wyk, 1988; Morgan & Soltis, 1993; Takhtajan, 1997). Ontogenetic and anatomical studies have shown that these emergences represent nectaries with a receptacular origin and without vascular connection, and that they are not homologous with the stamens (Figs. 10-11; e.g., Klopfer, 1972; Payer, 1857; Ronse Decraene & Smets, 1999).
(4.) Staminodes in a similar antepetalous position either are vascularized (e.g., Erodium) or are not vascularized (e.g., Linum). For some taxa, reports are contradictory (e.g., Erodium: Figs. 12-13, Averrhoa: Fig. 14; Al-Nowaihi & Khalifa, 1971; Kumar, 1976; Narayana, 1966). For example, Al-Nowaihi & Khalifa (1973) consider the antepetalous "teeth" of Linaceae ligular appendages, not staminodes, because they lack vascularization. Indeed, other studies in Linaceae (e.g., Narayana, 1964; Narayana & Rao, 1976a, 1976b, 1977a, 1977c) report the absence of vascularization, except for Kumar (1976), mentioning short vascular stubs leading to the staminodes in Linum. The presence of antepetalous staminodes, together with antesepalous nectaries against the stamens in Erodium (Figs. 12-14), led to even more imaginative interpretations. Dawson (1936) and Kumar(1976) interpreted the androecium of Geraniales as originally triplostemonous with a progressive reduction series, leading to the transformation of outer antesepal ous stamens into glands and antepetalous stamens into scale-like staminodes. However, the staminodial nature of the nectary was put in doubt by Al-Nowaihi & Khalifa (1971) because of the absence of any vascularization. It is clear that the nectaries in Erodium have nothing to do with the staminodes and arise independently (Figs. 12-13). In Averrhoa, nectaries and staminodes occur as undistinguishable structures that are vascularized (Fig. 14). The presence of an intrastaminal disk, together with a whorl of vascularized staminodes (as in Toona, Cedrela of Meliaceae and Flindersia of Rutaceae: Sheela Lal, 1994; Sheela Lal & Narayana, 1994), makes the interpretation of staminodes more concordant among authors. In Cedrela the antepetalous stamens are suppressed, but their traces persist within the receptacle. The related genus Toona has persistent staminodes (Baillon, 1895; Harms, 1960).
(5.) In the Hamamelidaceae it is difficult to distinguish among sterile phyllomes or appendages, staminodes, and nectaries. In Hamamelis, Mione and Bogle (1990) describe an antepetalous whorl as nectary primordia, while they describe the homotopic antepetalous whorl of Loropetalum as sterile phyllomes. In both taxa the sterile structures occasionally develop as staminodes, suggesting a staminodial origin (Baillon, 1871; Mione & Bogle, 1990). Sterile phyllomes similar to those of Loropetalum occur in other genera of Hamamelidaceae (Rhodoleia, Corylopsis: see Bogle, 1989; Endress, 1967; Figs. 15-16). Mione and Bogle (1990) argue that the sterile phyllomes of Loropetalum, nectaries of Hamamelis, and staminodes of Corylopsis are not derived from the same whorl of organs, because they do not arise at similar times in the development of the flower, because they have a different vascular connection, and because the least specialized genera, Maingaya and Dicoryphe, bear two whorls of both staminodes and sterile phyll omes with a different vascular supply. These facts indicate that progenitors of subfamily Hamamelidoideae of Hamamelidaceae probably possessed an androecium with at least three whorls with a functional divergence of nectariferous staminodes and sterile structures (cf. Bogle, 1989; Mione & Bogle, 1990).
From these examples it is clear that a topological definition for a staminode is sometimes sufficient but that in other cases it is not. In the example of Harungana, the topological criterion is supported by the vasculature and by the shape of the nectary. In Greyia and Francoa the stublike structures are associated with the androecium and alternate with the stamens, but they have nothing in common with staminodes. Indeed, the acceptance of the nectarial stubs as staminodes involves the acceptance of ancestral polyandry in Greyiaceae and Francoaceae, where the related rosid families are all basically diplostemonous.
The characterization of sterile stamens (i.e., staminodial structures) should be based on a combination of function and position in the flower rather than on their external morphology, as they are derived from a nonfunctioning stamen that must have preceded them in evolution (Figs. 52-54). Contrary to fertile stamens that are more or less restricted in their external morphology by their limited pollen-providing function, staminodes have evolved in a great variety of shapes, because of their varied functions, obscuring patterns of homology. The presence of vascular connections and the development of primordia are helpful tools for differentiating a staminodial structure. Therefore, we propose to characterize staminodes in two ways: a functional definition, with a major distinction between vestigial staminodes and functional staminodes (Fig. 52); and a topological definition, in which staminodes are approached on the basis of their relationship with the other organs in the flower (Fig. 53). The functional appro ach and topological definition can be combined in a time-related model, in which the evolution of the staminodes over time is stressed (Fig. 54).
Staminode evolution should be read as a progressive transformation series running from a fertile stamen into highly specialized forms. It runs from an imperfect, sterile stamen into a regressing or vestigial organ. Further evolution is biased between total loss and a conversion into a novel structure. Topology-based and function-based definitions are essentially hierarchical, in that a staminode either is homologous with a whole stamen or is part of an organ. As the androecium arises in a sequence in the flower, which may be spiral or whorled, a more detailed definition implies that the outer sphere (toward the perianth) has a staminodial nature (petalostaminodia) or that intervening (inner or middle) sterile whorls (antepetalous or antesepalous) are described as staminodial. Staminodes may also exist within a whorl or in a closed series. Staminodes as partial organs imply that they arise by the division of a common primordium.
B. EVOLUTION OF STAMINODIAL STRUCTURES: FUNCTION-BASED DEFINITION
1. Vestigial Staminodes
In its simplest form and as a primary step in stamen reduction, staminodes can persist as regressing or vestigial organs in the flower. Staminodes are an indication of a changing process, namely a reductive trend, either by the loss of a whole whorl of stamens in the transition from diplostemony to (ob)haplostemony (e.g., in Sterculiaceae, Geraniaceae, Primulaceae, Mytraceae; Ronse Decraene & Smets, 1995a; Figs. 12, 15; Table I) or by the partial reduction of stamens within a whorl (in relation to zygomorphy: e.g., in Scrophulariaceae, Verbenaceae: see Endress, 1999; Ronse Decraene & Smets, 1994, 1995a; Walker-Larsen & Harder, 2000; Figs. 20-21; Table II). Such structures can be defined as "vestigial staminodes." Although the extent of reduction differs, they are fundamentally little altered morphologically in comparison to fertile stamens. Such staminodes may possibly retain their vasculature (helping in their identification as staminodes), or the vasculature may fade out before reaching the organ or be lost completely (e.g., Euadenia: Arber, 1933; Raghavan, 1939). These staminodes may have a function in the flower, but this is not always clear.
Vestigial staminodes may be found at different stages of reduction, namely as a whorl of sterile stamens with small apical anthers (which are occasionally fertile) (e.g., Manilkara: Pennington, 1991; Anacardium: Figs. 20-21), as filaments without anthers (Paronychia: Figs. 17-19), as more or less small stubs (e.g., Anthirrhinum, Sagina, Moringa: Figs. 1-3), or as minute organs that appear in the early ontogeny of the flower but are no longer visible at maturity (e.g., Digitalis: Chatin, 1873a; Singh, 1979). Such staminodes may be initiated as a regular whorl of stamen primordia but abort at a certain stage of their development (e.g., Cedrela [Toona]: Baillon, 1895). However, the regular alternation of whorls may often become disturbed when the sterile structures arise after the fertile whorl, as is the case for centrifugal obdiplostemony (e.g., Theobroma in Sterculiaceae: Ronse Decraene & Smets, 1995a).
In other cases, staminode initiation is delayed until well after the initiation of the carpels (e.g., Paronychia decandra: Fig. 17; Samolus valerandi, Magodendron: Ronse Decraene & Smets, 1995a; Vink, 1995). Theophrastaceae, Sapotaceae, Myrsinaceae, and Primulaceae are examples of the derivation of the obhaplostemonous androecium from diplostemonous ancestors. Although all Theophrastaceae possess colored, attractive staminodes little different from petals, some genera of Myrsinaceae (e.g., Myrsine), and Primulaceae (e.g., Samolus) possess evidence of antesepalous staminodes: In all cases the staminodes arise after the initiation of the common stamen-petal primordia (Caris, 1998).
It is possible that a stamen or a whole stamen whorl has vanished externally but that evidence of residual traces persists internally. For example, in the Primulaceac the median sepal traces split tangentially and give off five internal traces, which alternate with the common stamen-petal traces. They appear in the petal tube as the fused petal marginals. In Steronema these bundles split twice, providing the petal marginal bundles and "staminodium" bundles, which come to lie in a ring with the stamens (Douglas, 1936). This induced certain authors to consider these marginal petal traces transformed stamen traces (e.g., Soldanella: Saunders, 1937-1939; Primula. Subrarnanyam & Narayana, 1976). Although this evidence appears to be a point for those who advocate vascular conservatism, it is absolutely not proof of a staminodial origin (see also Arber, 1933; Schmid, 1972).
In Mangifera indica L. or Anacardium occidentale L. (Anacardiaceae), reductions have affected the whole antepetalous whorl and four stamens of the antesepalous whorl (Figs.20-21). The antepetalous whorl may often be wholly suppressed, apart from the occasional presence of short vascular traces (Sharma, 1954); the single fertile antesepalous stamen receives a larger trace than do the sterile antesepalous stamens.
The abortion of stamens within a whorl can affect different halves of the flower, with intermediate half-fertile anthers in the genera Conospermum and Synaphea (Proteaceae). The configuration of the androecium is mirror imaged between the genera, with an adaxial (Synaphea) or abaxial (Conospermum) sterile anther and two lateral anthers with one half sterile (Douglas, 1997).
2. Functional Staminodes
In several instances staminodes have become adapted to fulfil novel biological requirements in the flower in response to a specific pollination syndrome. Petals (petalostaminodia or andropetals) also play that role, but most often in a more generalized way.
The different functions of staminodes can be summarized as follows (Fig. 53):
* Production of a food supply (nutrient bodies, sterile pollen, or nectar): nutritional function;
* Development of collecting structures in association with nectaries (as nectar recipients), triggering mechanisms for pollen dispersal, secondary pollen presenters, obstacles for selfing: structural function;
* Attraction of pollinators by display of colors, odors, or heat: attractive function.
Staminodes may fulfill several functions at the same time, namely producing nectar, collecting or holding it, and being optically attractive (e.g., Parnassia), or different sets of staminodes may have different functions in the same flower (e.g., inner versus outer staminodes in Himantandraceae: Endress, 1984, 1986). In the magnoliids the staminodes have multiple functions related to pollination, such as attracting and directing pollinators by their color, odor, food supply, and secretions, protecting the ovary against predation, effecting pollination or preventing selfing by their position or by movements, or providing shelter and warmth (Endress, 1984, 1994; Thien et al., 1999; Walker-Larsen & Harder, 2000).
The transition from nonfunctional sterile stamens to nectar-producing structures is apparently relatively easy, depending on a vascular connection, as in Azara (Flacourtiaceae), where the short stubs produce nectar through stomata (Figs. 36-37). In Loasaceae (subfamily Loasoideae) the antesepalous stamens have become differentiated into colored nectar recipients (Figs. 29-30; Hufford, 1990; Smets, 1988a, 1988b; Urban, 1892). For example, in Loasa (Loasaceae) the staminodes are bright yellow and red, contrasting with the white corolla. In Harungana madagascariensis (Choisy) Poir. (Clusiaceae) the antesepalous stamen whorl has become transformed into scale-like nectaries (Ronse Decraene & Smets, 1991 a; Fig. 9). In the genera Piciarium and Archytaea of Bonnetiaceae, nectaries are discrete, antepetalous structures alternating with the stamen clusters that are supplied by double bundles similar to stamens (Dickison & Weitzman, 1998). In certain families, such as Aizoaceae, part of the centrifugally developing sta mens grow into colored staminodes (Hofmann, 1993; Ihlenfeldt, 1960). Parnassia (Parnassiaceae) has a whorl of antesepalous stamens alternating with staminodial nectaries and resembling a fascicle of sterile stamens (He[beta], 1983). Floral anatomy and ontogeny demonstrate that the nectaries are equivalent to single reduced stamens (see Bensel & Palser, 1975a; Klopfer, 1972; Saxena, 1976).
These examples of staminodes are transformed structures, namely they are basically homologous to stamens, but they have been altered by their functional requirements. Because of their obvious role in the flower, contrary to vestigial staminodes, we prefer to describe this type of sterile structures as "functional staminodes."
C. STRUCTURAL SIGNIFICANCE OF STAMINODIAL STRUCTURES: TYPOLOGY-BASED DEFINITION
1. Acyclic Staminodes
Primitive taxa of the Magnoliidae often possess staminodial structures between stamens and tepals and between carpels and stamens (e.g., Bernhardt, 1996; Endress, 1984, 1986, 1990a, 1990b; Ronse Decraene & Smets, 1993; Walker-Larsen & Harder, 2000). These staminodes are typical of spiral flowers with little or no synorganization. They represent stepping stones between different organs (e.g., tepals-stamens-carpels) and have occasionally attained specific (overlapping) functions in the flower.
For Eames (1961), the first mode of attraction of the angiosperms consisted exclusively of these upper staminodes (Figs. 32-33). However, such cases are isolated and are not linked to the generalized condition with staminodes situated in the periphery of the flower. The initiation of floral organs in a close helical sequence leaves little space for differentiation between distinct groups of floral organs without disturbing the helix considerably. The transition between tepals, sepals, petals, and stamens can only be a gradual process in this case, in which staminodes play an important role as multifunctional transitional structures (e.g., protective structures versus nutrient bodies or showy attractive organs: see Endress, 1984, 1986, 1990a). In the eudicots the different functions often become separated in space and time.
2. Complete Staminodial Whorls
In many cases (see Table I) a whorl of stamens tends to become completely sterilized in the flower. In this way a diplostemonous androecium becomes transformed into an (ob)haplo-stemonous one. Evidence for a phylogenetic link between the two androecial configurations, running only in one direction, relies essentially on staminodial structures (see Ronse Decraene & Smets, 1995a). Staminodes evolved many times in the rosids (Walker-Larsen & Harder, 2000). In the Malvales and former Theales the occurrence of antesepalous staminodes is correlated with secondary multiplication of the other stamen whorl (Fig. 38). Staminodes tend to be the expression of a no-return reductive process, although they occasionally attain a new function in the flower (e.g., Bonnetiaceae, Clusiaceae, Malvaceae, Sterculiaceae, Parnassiaceae, Lepuropetalaceae: Figs. 2, 7, 9, 12, 15, 36, 38).
Whorls of staminodes related to a reductive trend also occur in the more primitive taxa with a polycyclic androecium. Monanthotaxis whytei (Stapf) Verdc. (Annonaceae) has two outer whorls of staminodes (Ronse Decraene & Smets, 1990a, 1993; Fig. 31). The more external whorl of six pairs appears in early development but is hardly visible at maturity; the next whorl of nine staminodes remains relatively large at maturity. Such cases probably represent stages in a stepwise reduction of a polycyclic androecium (see Ronse Decraene & Smets, 1993).
The corolla, or petal whorl, represents a special case of a complete staminodial whorl. Staminodes are sometimes petaloid, leaflike appendages that cannot be differentiated from the petals (e.g., in some Theophrastaceae, Corynocarpaceae: Figs. 7-8). They are evidence of a direct link between stamens and petals. As they are not different from petals or it is in some cases not possible to differentiate them (e.g., in some Caryophyllaceae), this kind of petaloid staminodes are best called "Petalostaminodia." Teratologica] cases of double flowers, as in Rosaceae or Malvaceae (e.g., Innes et al., 1989; Maclntyre & Lacroix, 1996) are a classic example of this transition. In other cases petaloid staminodes may be observed in the position that petals normally occupy (e.g., in Hamamelidaceae: Endress, 1967; Mione & Bogle, 1990, Caryophyllaceae: Ronse Decraene et al., 1998b; Fig. 25). The number of stamens can also be augmented at the cost of petals (Murbeck, 1918: "staminal pseudapetaly," quoted in Endress, 1967).
Petals represent a problematic case of staminodial origin, as it is generally assumed that the petals of a great many angiosperms have been derived from stamens and are homologous with them (e.g., Cronquist, 1988; Eames, 1961; Endress, 1986, 1994; Hiepko, 1965; Takhtajan, 1980, 1991; Weberling, 1989; Worsdell, 1903). In many cases it is difficult to determine when a petal ceases to be a staminode and when a staminode ceases to be a stamen (Figs. 24-25). At another extreme, petals can sometimes attain all structural and developmental attributes of sepals, concomitant with changing functions (Endress, 1994). Strictly speaking, petals must be seen as showy, flattened, and colored organs occupying the space between the sepals and the androecium. In comparison with staminodes within the androecium, the development of petals from stamens is an evolutionary step that has taken place repeatedly in angiosperm evolution.
Petals have probably arisen several times in the Ranunculales from outer (nectar-producing) staminodes (see, e.g., Drinnan et al., 1994; Endress, 1995; Hiepko, 1965; Kosuge, 1994; Figs. 22-23, 34-35). In Ranunculaceae there are transition series from inconspicuous staminodes to elaborate petaloid nectar leaves occurring among genera (Ronse Decraene & Smets, 1995b). The morphological homology between nectar leaves and stamens has been traced back ontogenetically in a number of species of Ranunculaceae by Erbar et al. (1998). A topological definition of staminodes is also in concordance with the nectary types proposed by Smets (1 988a), nectaria nectarophyllomina and nectaria staminodialia. The nectarophyllomina type of nectaries (or Helleborus type) correspond with the petalostaminodes characteristic of the Ranunculales (e.g., Ranunculaceae, Berberidaceae, Menispermaceae). The staminodialia type of nectaries (or Trigonia type) correspond with staminodes that are more strongly associated with the androecium.
Clear ontogenetic descriptions of homeotic shifts between petals and stamens are Sanguinaria, with an extra whorl of petals (Papaveraceae: Lehmann & Sattler, 1993) and Actaea (Ranunculaceae: Lehmann & Sattler, 1994), petals transformed into stamens in Macleaya (Papaveraceae: Ronse Decraene & Smets, 1990b), stamens occupying the position of petals in Saraca and Swartia of the Leguminosae (Tucker, 1988b), Dichapetalum (Dichapetalaceae: Breteler, 1973; Figs. 26-28, but see Table III), or double-flowered Hibiscus of Malvaceae (MacIntyre & Lacroix, 1996). Illustrations of Swartia in Tucker (1988c: 77) show that there are a single petal and three large stamens in one outer whorl, while the remaining stamens are crowded on a ring primordium. The transition of stamens into staminodes, and further into petals is best described by the term "serial homeosis."
The terms "andropetals" (related to and derived from stamens and similar to staminodes) and "bracteopetals" (related to and derived from bracts and sepals) distinguish between two kinds of petals in the angiosperms, even when shifts have occurred between petals and sepals (see Hiepko, 1965; Kosuge, 1993; Ronse Decraene & Smets, 1993, 1995b; Takhtajan, 1991). Important arguments for the presence of "andropetals" as opposed to "bracteopetals" (Hiepko, 1965; Takhtajan, 1991) are the vascular arrangement (one-trace organs; however, this distinction has little relevance because three-traced stamens may also occur), the ontogeny (similarity to stamen primordia in shape of primordia, retardation of growth of the petals), teratological cases, but most important the spatial relation between stamens and petals (existence of parastichies). Very often petals resemble stamens in having a stalk and a limited insertion area (clawed structures). Staminodes belonging to a staminal whorl may also become secondarily petaloid, a s in the Zingiberales (e.g., Kirchoff, 1991; Walker-Larsen & Harder, 2000).
Flowers are occasionally secondarily apetalous but may occasionally become secondarily petaliferous. In that case, outer staminodes may be differentiated as outer petaline structures, which confuses the limits between petals and staminodes as in Scytopetalaceae (Appel, 1996), or there is an outer receptacular corona without clear homology with staminodes (Passifloraceae: Bernhard, 1999a).
The strong link between petals and stamens has a genetic basis that has been extensively studied in the last ten years for the model genera Antirrhinum and Arabidopsis (e.g., Bowman et al., 1991; Coen & Meyerowitz, 1991: ABC model). At the same time, the petals are intermediate between stamens and sepals. We therefore assume that there are repeated evolutionary origins for petals, either from stamens (in the majority of eudicots) or from sepals.
4. Incomplete Staminodial Whorls
The presence of staminodes within a stamen whorl is often an indication of the monosymmetric development of the flower. A stamen whorl becomes partially sterile, as an adaptation to a "vectorized" pollinator visit. The reduced stamen usually occupies a position crossed by the symmetry line. Staminodial structures may be found within one or two whorls of stamens, depending on the androecial configuration that functions as the starting point.
In the Fabales stamens arise unidirectionally, and the abaxial part of the androecium is often "advanced" compared with the posterior part. Adaxial stamens are often smaller, as they lag in development (e.g., Chamaecrista: Tucker, 1996), are staminodial (as in Petalostylis with two antesepalous staminodes: Tucker, 1998; or in Cassia and Senna with three adaxial staminodes and a strong heteranthery: Tucker, 1996), or are missing. An extreme is Bauhinia divaricata, with a single stamen, nine staminodes, and a variable number of petals (Tucker, 1988b). Petals and all other stamen primordia are initiated but are arrested at a given stage of their development. Other species of Bauhinia have a variable number of staminodes, have sterile stamens, or none at all (Tucker, 1984, 1988b).
In many taxa of the asterids, zygomorphy is correlated with the occurrence of an adaxial staminode. Androecial initiation is unidirectional, with a delayed initiation of the adaxial staminode, which may not arise at all in some cases (e.g., Baillon, 1860b, 1862c; Bocquillon, 1861b; Chatin, 1873a; Endress, 1998, 1999; Payer, 1857; Singh & Jain, 1978). The posterior staminodes of many asterids either are small and reduced or can be secondarily increased in size, concomitant with a functional diversification (e.g., Kigelia: Neubauer, 1959; Penstemon, Scrophularia: Endress, 1994). Reductive trends of the posterior stamen in the Verbenaceae can be followed through several intermediates, ranging from the obvious presence of staminodes, to their initiation and consecutive loss and their total absence (Bocquillon, 1861b; Payer, 1857; Sattler, 1973).
The possibility of a reversal of staminodes and the reappearance of fertility has been discussed by Walker-Larsen and Harder (2000) for the Scrophulariales. This reversion is correlated with a transition to radially symmetric flowers. We doubt that this process is possible, because reversals to radially symmetric flowers in asterids operate via the loss of the posterior staminode and the fusion of the posterior petals and the transition to tetramerous flowers (see, e.g., Ronse Decraene & Smets, 1994; Endress, 1999). Loss of stamens seems irreversible, certainly for whole stamen whorls and probably also for reductions within whorls, except for the occasional genetic mutation or monstrosity, unless one considers the event of peloric mutants (e.g., Antirhinum: Coen & Meyerowitz, 1991; Coen et al., 1995) as a leading factor in floral evolution. Although insights into molecular evolution of flower development rest mainly on homeotic mutants, their importance to floral evolution remain virtually unknown (cf. Li & J ohnston, 2000).
5. Secondary Staminodial Structures
In some families with a multistaminate, centrifugal androecium the outer stamen primordia are not developed beyond the stage of antherless structures (e.g., Dilleniaceae: Baillon, 1865, 1866; Endress, 1997; Fumana in Cistaceae: Nandi, 1998; Bixaceae: Ronse Decraene, 1989; Aizoaceae: Hofmann, 1993; Limnocharitaceae: Haynes et al., 1998). The existence of this kind of staminodes is probably linked to the secondary appearance of the centrifugal stamens and is induced by the rapid development of the flower (see Ronse Decraene & Smets, 1992). Centrifugal stamen development lags behind the development of other floral organs, and there is probably not enough time or nutrient allocation to attain a full development of the outermost stamens. Note that the presence of outer staminodes in a polyandrous androecium has often been interpreted as evidence for a reductive trend (see Ronse Decraene & Smets, 1992, 1993). In Paeonia, innermost stamens may be staminodial by pressures of the developing internal disk (Hiepko, 1966 ).
In some cases the outer staminodes of centrifugal androecia have become converted to new functions, linked with pollinator attraction. In Loasaceae subfamily Loasoideae a variable number of antesepalous staminodes develop into colored nectar collectors (Hufford, 1990; Smets, 1988a, 1988b; Figs. 29-30). In Dilleniaceae the outermost stamens may develop into a corona (Pachynema: Endress, 1997). The flowers of Scytopetalaceae are basically apetalous but have a showy corona (pseudocorolla) of staminodial origin (Appel, 1996). In the related Lecythidaceae, external staminodes have evolved in colored, complex structures (Endress, 1994). In Couroupita guianensis the abaxial part of the androecial ring primordium is detached as a broad flap of tissue with numerous staminodes covering the fertile stamens like a hood. This hood may contain fodder staminodes with pollen, or nectar may be produced at the base of the staminodes. Different pollination mechanisms and references hereto are abundantly discussed in Endress (19 94).
Secondary staminodial structures have the same characteristics as secondary stamens arising on common primordia. They may be vestigial or have evolved different functions related to pollination (Fig. 54).
V. Imaginary Staminodes
The difficulty in interpreting the homology of staminodes has often led to erroneous statements about structures surrounding the androecium. A striking similarity of intrastaminal emergences to filaments, prominent appendages of fused stamen bases, invaginations of the petals, or receptacular emergences, which are sometimes nectariferous, were often taken as evidence of a second aborted stamen whorl. Numerous examples exist in which sterile emergences have been interpreted as staminodes without supporting evidence (see Table III, Figs. 39-40). These appendages commonly arise very late in ontogeny and are not vascularized, or they are vascularized by various means. The following examples illustrate the difficulty in interpreting pseudostaminodial floral appendages:
1. Short-stalked glands occur at the base of the inner staminodes of Gomortegaceae (e.g., Brizicky, 1959) and Hernandiaceae (e.g., Kubitzki, 1969; Sastri, 1965) and on the outer, intermediate, or inner stamens of Lauraceae (e.g., Endress & Hufford, 1989; Kasapligil, 1951; Rohwer, 1994; Sastri, 1965; Vattimo, 1959; Fig. 41) and Monimiaceae (e.g., Endress, 1980; Sampson, 1969). Because the lateral appendages look superficially similar to reduced stamens, most authors have taken the basally inserted nectaries on the stamens of Laurales as evidence of reduced stamen fascicles and have interpreted the nectaries as lateral stamens in a clear state of reduction (e.g., Eames, 1961; Reece, 1939; Rohwer, 1994; Sampson, 1969; Sastri, 1952, 1965). Other evidence, especially a comparison with the lateral androecial lobes of Chloranthus (Chloranthaceae), has been used for arguing a derivation of lauralean stamens from primitively branched structures (Rohwer, 1994). Kasapligil (1951: 182), on the other hand, regarded the st aminal glands as "emergences produced de novo for a functional purpose." He observed that the staminal glands arise late in ontogeny from lateral meristematic regions of the stamens. Other recent observations of the floral ontogeny of the flower of Lauraceae support Kasapligil's view (Endress, 1980; Singh & Singh, 1985), because no difference is found between the early inception of stamens with nectaries and those without nectaries. Moreover, no fasciculate stamens are known in the Laurales, and the relative (vascular) independence of the nectarial appendages is due to their late appearance in ontogeny (Endress, 1980). However, Crane et al. (1994) interpret fossil evidence of a lauralean flower as having an outer whorl of six staminodes, apparently set in three pairs, each of which appears to be linked to a single stamen. Could the fusion of the staminode with a stamen lead to a tripartite structure? The question of paired staminodes is contradicted by a review study by Eklund (2000) of fossil Lauraceae flowe rs which demonstrates a basic and constant pattern of four trimerous stamen whorls, the innermost being staminodial and with only the third bearing paired glandular appendages. The question is clearly not settled, especially in comparison with the tripartite structure of Chloranthus (Chloranthaceae).
2. The intrastaminal appendages between the fused stamen bases of Amaranthaceae have been interpreted either as true staminodes representing a lost stamen whorl (e.g., Goldberg, 1986; Joshi, 1932; Joshi & Venkata Rao, 1934; Saunders, 1937-1939) or as emergences of the staminal tube without clear morphological identity (e.g., Eliasson, 1988; Payer, 1857; Schinz, 1934). Eliasson (1988) observed that broad filaments are correlated with an absence of interstaminal emergences and that small filaments share the presence of pseudostaminodes. The intrastaminal teeth arise late in ontogeny (Payer, 1857) and have no vascular connection (Schinz, 1934).
3. Pseudostaminodes and real staminodes may occur in a same flower, as in Sauvagesia (Ochnaceae), with an outer fringe of threadlike appendages and a whorl of five petaloid staminodes in the petal radii (e.g., Amaral, 1991; Eichler, 1875-1878; Goebel, 1933; Saunders, 1937-1939). Outer staminodes may co-occur with the five antepetalous staminodes (S. erecta), only the antepetalous staminodes may be found (S. glandulosa, S. guianensis), or only small appendages (Blastemanthus) may exist (Amaral, 1991). The outer appendages are best interpreted as a corona in colors that contrast with the real staminodes. However, these have also been described as staminodes (e.g., Amaral, 1991). The same interpretation holds for the corona of the Passifloraceae (Table III).
4. In the asterids, several families have "scales" on the corolla alternating with the stamens (e.g., in Apocynaceae, Boraginaceae, Cuscutaceae, Menyanthaceae, Hydrophyllaceae). These have occasionally been interpreted as stipular (e.g., Woodson & Moore, 1938) or staminodial in nature (Lindley, 1853, cited in Lawrence, 1937). However, other floral anatomical studies have shown that the scales are invaginations of the corolla tube, wit no relation to the androecium (e.g., Eichler, 1875-1 878; Lawrence, 1937; Rao & Arati Ganguli, 1963).
5. Intrastaminal appendages or lobes functioning as nectary have often been interpreted as evidence of staminodes or even carpellodes in the asterids, especially when the external morphology is reminiscent of these (Figs. 48-51; Eichler, 1875-1878; Sersic & Cocucci, 1999; Woodson & Moore, 1938). Other examples of incongruent interpretations are given in Table III.
B. RECEPTACULAR DISKS
Disklike nectaries (Figs. 42-47) also belong to the category of imaginary staminodes because they have often been taken for an aborted whorl of stamens (see Table III). This is illustrated by following examples;
1. In the Rhamnaceae there is no external (ontogenetic) evidence of a second staminodial whorl (Fig. 45); nor is there any link between androecium and disk (Bennek, 1958; Suessenguth, 1953a). However, the vascular supply of the disk, which can sometimes be similar to that of the antepetalous stamens, along with the disruption of the "alternance rule" (the stamens are antepetalous), has been used as support for the interpretation of the intrastaminal disk as modified stamens (Nair & Sarma, 1961; Prichard, 1955). Both interpretations--that is, the ontogenetic and the anatomical--can be supported to some extent, as another stamen whorl may have been present in an ancestral state but may be lost in extant Rhamnaceae. This interpretation is also linked to what family is considered the nearest sister group. The development of a disk can have "taken up" the vascular facilities provided for the now-missing antesepalous stamens. This demonstrates that a total rejection of the idea of a "lost" whorl, as well as the rec ognition of "evidence" of a lost whorl, are not to be considered too strictly. The derivation of an (ob)haplostemonous androecium from two stamen whorls could be described as a counterbalancing development within the flower, because the space occupied by stamens is invariably taken over by the developing disk. However, we occasionally observed an antesepalous staminode in Zizyphus lotus (Ronse Decraene, unpubl.), which does not support the interpretation of a staminodial disk.
2. In the Polygonaceae the position of lost stamens is often taken over by receptacular nectaries (Figs. 42-43). Emberger (1939) interpreted the nectaries of Fagopyrum as staminodes, because of the spatial and numerical correlation between stamens and nectaries. Indeed, there is a high correlation between the number of stamens and the presence of the glands. However, the internal variations of nectarial tissue, as well as anatomical evidence, firmly deny a staminodial nature (cf. Ronse Decraene & Akeroyd, 1988; Ronse Decraene & Smets, 1991c).
More examples of incongruent disks are given in Table III.
C. THE CONTEXT OF IMAGINATIVE THOUGHT
Interpretations of the homology of disk structures have varied according to the methods of investigation used. Floral anatomists attached greater importance to vascular elements and more often favored a phyllomatic (staminodial) nature; therefore, they interpreted floral disks more likely as transformed (reduced) organs. Scholars in floral ontogeny and systematists often ignored the vasculature and favored an interpretation of a receptacular nature for the disk, because of its late inception and the absence of a clear morphological resemblance to other floral organs. This has often resulted in contradictory interpretations in floral morphology. However, it is essential that both methods of investigations be given sufficient weight (see Arber, 1933; Gustafsson & Albert, 1999).
Floral anatomists, especially the American school of Eames (e.g., Berkeley, 1953; Blaser, 1954; Dawson, 1936; Eames, 1931, 1961; Heinig, 1951; Prichard, 1955; Tillson & Bamford, 1938) and the Indian school of Puri (e.g., Nair & Jain, 1956; Nair & Joshi, 1958; Nair & Sarma, 1961; Narayana & Rao, 1971; Puri, 1948) have been obsessive about describing staminodial structures on the basis of the presence of vascular traces and the current belief of primitive polyandry in angiosperms. Indeed, interpretations of staminodial structures (especially for disks) were often related with a hypothetical interpretation of ancestral polyandry and a given direction of evolution (e.g., Humiriaceae: Narayana & Rao, 1978; Geraniaceae: Dawson, 1936). This led to certain highly imaginative reconstructions of "ancestral" flowers.
Evidence of a staminodial nature of the disk was often sought in the presence of vestigial vascular stubs or vascular connections between the supply to the stamens or other organs and the disk (e.g., Nair & Joshi, 1958). There are indefinite possibilities for supplying the disk; the supply of the nectary is opportunistic as it becomes derived from the nearest source of vascular tissue, which is often the androecium. Trying to recognize staminodes or any other structures surrounding the ovary, if no structural evidence of their homology with stamens exists, is senseless. Smets (1986, 1988a, 1988b) restricted the term "disk nectary" to a secondary emergence of the receptacle (nectaria axialia) when there is no homology possible with staminodes and when it is not part of the gynoecium.
VI. Molecular Developmental Genetics and Staminodes
Recently, much emphasis has been laid on the study of expression of developmental genes in order to understand differences of morphological characters from an ontogenetic and phylogenetic perspective. Studies of the molecular controlling mechanisms of organ determination have led to the discovery of MADS box genes (see, e.g., Albert et al., 1998; Thei[beta]en et al., 1996; Yanofsky, 1995). These genes are partly responsible for floral organ determination, as demonstrated in the ABC model, with three distinct functions (Coen, 1991; Coen & Meyerowitz, 1991). In its simplest form the ABC model implies that A is responsible for sepal expression, A + B for petals, B + C for stamens, and C for carpels. Beyond the expression of this simple model, the overall expression of flower development is often more complex (see Albert et al., 1998; Kramer & Irish, 2000). Two gene activities have to be recognized, leading to a distinction between whorl identity and organ identity: one that influences the outcome or function of an organ (whether it be sepaloid or petaloid, etc.), or process orthology; and one that influences the position of organs, or positional orthology.
These two processes act independently, as a petaloid organ will occupy the same position as the original organ. For example, in double-flowered Begonia (Lehmann & Sattler, 1989; Ronse Decraene & Smets, 1990b), stamens have been replaced by petaloid structures that occupy the same position in the flower. On the other hand, in Macleaya stamens occur in the position of the petals in the other Papaveraceae (Ronse Decraene & Smets, 1990b) and in Calla (Araceae) in the position of tepals (Lehmann & Sattler, 1992).
Albert et al. (1998) interpreted the nature of organs mainly on the basis of gene activity. In a simple way AB gives petals, BC stamens, and ABC leads to staminodes. Staminodes thus appear as the result of an overlap of the genetic programs of the perianth members and stamens during floral development (cf. Erbar et al.  for the nectar leaves of Ranunculaceae).
Through examples of Lecythidaceae and Clusiaceae, Albert et al. (1998) and Gustafsson (2000) correlated the formation of staminodial structures with the expansion of the A function gene activity, which leads to the transference of petaloid characters to stamens.
However, this approach has certain shortcomings. The terminological distinction in zones of influence is not sufficiently detailed to recognize intrinsic variations of expressions of organs (there are different degrees of staminode development), it overlooks external environmental factors and pressures from pollinators, and it denies the historical dimension (what is derived from what), as process homology is equally influenced by time.
The explanation of a shift in gene activity is only a partial explanation for the existence of staminodes, as it is mainly a functional (teleological) explanation of gene activity. In the case of the nectar leaves of Ranunculaceae, Erbar et al. (1998) have demonstrated the homology with stamens in the presence of rudimentary adaxial pollen sacs in early developmental stages. The shift to an increasing A function may have been progressive or sudden, but little can be said about that, as the knowledge of the importance of genetic mutations to evolution is virtually nil.
A good case for the oversimplification of the molecular model is the example of sorrel or Rumex (Polygonaceae). In Rumex the perianth consists of two whorls of three sepaloid tepals. Ainsworth et al. (1995) and Albert et al. (1998) hypothesized that petals were ancestrally present in Rumex but that they were lost in evolution. They explain the present sepaloid perianth as the result of the loss of the B function and thus as the result of a secondary restriction of the "basal" petaloidy that is considered ancestral in the family. This explanation does not account for the shifts between trimery and pentamery operating in the family, the occurrence of outer stamen pairs, and the improbable distinction between petals and sepals that may not have been present in the ancestors of the Polygonaceae, as no extant Polygonaceae with both sepals and petals exist. The molecular explanation may refer to the process of development, but it is only partial evidence, as the hypothesized assumptions about the evolution of petal s have no morphological basis.
To restrict the explanation for stamen, staminode, and petal identity to an alteration in expression or function of B-class genes (e.g., Albert et al., 1998; Bowman et al., 1991; Weigel & Meyerowitz, 1994) is to oversimplify the development and identity of organs. The distinctions made between bracteopetals and andropetals by Hiepko (1965) and Takhtajan (1991), or the terms "homeosis"or "heterotopy," as the total or partial replacement of one part by another of the same organism (e.g., Sattler, 1988, 1994; Li & Johnston, 2000) explain the same as the molecular terminology, but they are based on a different point of view.
VII. Concluding Remarks
It is clear that the decision to recognize a lately arising primordium as a staminodium or merely as a secondary receptacular emergence is often a matter of subjective appreciation and is extremely difficult to assess. Therefore, reliance on indirect evidence can be helpful. For Harungana (Clusiaceae), Ronse Decraene and Smets (199la) hypothesized that the nectaries represent a staminodial whorl, and evidence was given in the vasculature and the external shape. In Proteaceae, Douglas and Tucker (1996) refuted a staminodial nature for the intervening nectary scales, although these are strictly speaking comparable to the nectary scales of the Clusiaceae. Proposed phylogenetic relationships (e.g., Chase et al., 1993; APG, 1998) can help in assigning the true nature of organs, although this evidence could be subject to circular reasoning. The association of Clusiaceae with clades having diplostemonous flowers (e.g., Linales, Ochnaceae, rosids I) supports the acceptance of a staminodial nature of the nectaries in Harungana. The association of Proteaceae with the Platanaceac at the base of the eudicots may be evidence against a staminodial nature.
Staminodial structures play an important role in floral evolution (see also Walker-Larsen & Harder, 2000). They are a reflection of the dynamism of the androecium (and flower) in response to changing conditions. Therefore, their importance should not be ignored, and a misinterpretation of structures that resemble staminodes must be recognized. The recognition of types of staminodial structures based on function (i.e., vestigial and functional staminodes) and position is only a partial characterization, but it is a necessary reflection of the complexity of floral forms.
We acknowledge the technical assistance of Anja Vandeperre in preparing the photographic material. This research was supported by a grant from the Research Council of the Katholieke Universiteit Leuven (0T/97/23).
IX. Literature Cited
Agarwal, S. 1963. Morphological and embryological studies in the family Olacaceac, I: Olax L. Phytomorphology 13: 185-196.
Ainsworth, C., S. Crossley, V. Buchanan-Wollaston & M. Thangavelu. 1995. Male and female flowers of the dioecious plant sorrel show different patterns of MADS box gene expression. Pl. Cell 7: 1583-1598.
Albert, V.A., M. H. G. Gustafsson & L. Di Laurenzio. 1998. Ontogenetic systematics, molecular developmental genetics, and the angiosperm petal. Pp. 349-374 in D. E. Soltis, P. S. Soltis & J. J. Doyle (eds.), Molecular systematics of plants, II: DNA sequencing. Kluwer Academic Publishers, Boston.
Alexander, I. 1952. Entwicklungsstudien an Bluten von Cruciferen und Papaveraceen. Planta 41: 125-144.
Al-Nowaihi, A.S. & S.F. Khalifa. 1971. Studies on some taxa of the Geraniales, I: Floral morphology of Averrhoa carambola L., Oxalis cernua Thunb. and O. corniculata L. with reference to the nature of the staminodes. Proc. Indian Natl. Sci. Acad. B 37:189-198.
----- & -----. 1973. Studies on some taxa of the Geraniales, II: Floral morphology of certain Linaceae, Rutaceae and Geraniaceae with a reference to the consistency of some characters. J. Indian Bot. Soc. 52: 198-206.
Amaral, M. C. 1991. Phylogenetische Systematik der Ochnaceae. Bot. Jahrb. Syst. 113: 105-196.
Andersson, L. 1998. Heliconiaceae, and Marantaceae. Pp. 226-230,278-293 in K. Kubitzki (ed.), The families and genera of vascular plants. Vol. 4, Flowering plants: Monocotyledons: Alismatanae and Commelinanae (except Gramineae). Springer-Verlag, Berlin.
APG (Angiosperm Phylogeny Group). 1998. An ordinal classification for the families of flowering plants. Ann. Missouri Bot. Gard. 85: 531-553.
Appel, O.1996. Morphology and systematics of the Scytopetalaceae. Bot. J. Linn. Soc. 121:207-227.
Arber, A. 1931. Studies in floral morphology, I: On some structural features of the cruciferous flower. New Phytol. 30: 11-41.
-----. 1933. Floral anatomy and its morphological interpretation. New Phytol. 32: 321.
Ayensu, E. S. 1972. Morphology and anatomy of Synsepalum dulcificum (Sapotaceae). Bot. J. Linn. Soc. 65: 179-187.
Baillon, H. 1860a. Recherches sur l'organisation et le developpement des Ericoidees. Adansonia 1:189-211.
-----. 1860b. Recherches sur l'organisation, le developpement et l'anatomie des Caprifoliacees. Adansonia 1: 353-380.
-----. 1861. Organogenie florale du sesame. Adansonia 2:1-4.
-----. 1862a. Sur la fleur des pivoines. Adansonia 3:45-49.
-----. 1862b. Sur le Mercurialis alternifolia Desr. et sur les limites du genre Mercurialis. Adansonia 3:167-176.
-----. 1862c. Organogenie florale des Martynia. Adansonia 3:341-348.
-----. 1865. Remarques sur les Dilleniacees. Adansonia 6:255-281.
-----. 1866. Sur l'organisation florale d'un Wormia des Seychelles. Adansonia 7: 343-347.
-----. 1870. Histoire des plantes, II: Elaeagnacces. Hachette, Paris.
-----. 1871. Nouvelles notes sur les Hamamelidees. Adansonia 10:120-137.
-----. 1872. Observations sur les Rutacees. Adansonia 10:299-333.
-----. 1873. Traite du developpement de la fleur et du fruit, IX: Chamaelauciees. Adansonia 11:361-3 65.
-----. 1874. Histoire des plantes, V: Euphorbiacees. Hachette, Paris.
-----. 1884. Les Xylolaena et la valeur de la famille des Chlenacees. Bull. Mens. Soc. Linn. Paris 1(52): 410-414.
-----. 1886. Remarques sur 1'organisation et les affinites des Podostemonacees. Bull. Mens. Soc. Linn. Paris 1(81): 644-648.
-----. 1892. Histoire des plantes, 11: Loranthacees. Hachette, Paris.
-----. 1895. Organogenie florale d'un Cedrela. Bull. Mens. Soc. Linn. Paris 2 (150): 1186-1187.
Beille, M. L. 1901. Recherches sur le developpement floral des Disciflores. Actes Soc. Linn. Bordeaux 56: 235-410.
Bennek, C. 1958. Die morphologische Beurteilung der Staub- und Blumenblatter der Rhamnaceen. Bot. Jahrb. Syst. 77:423-457.
Bensel, C.R. & B.F. Palser. 1975a. Floral anatomy in the Saxifragaceae sensu lato, I: Introduction, Parnassioideae and Brexioideae, Amer. J. Bot. 62: 176-185.
-----. 1975b. Floral anatomy in the Saxifragaceae sensu lato, II: Saxifragoideae and Iteoideae. Amer. J. Bot. 62: 661-675.
Berger, A. 1930. Crassulaceae. Pp. 352-483 in A. Engler & K. Prantl (eds.), Die naturlichen Pflanzenfamilien. Ed. 2. Vol. 18a. Engelmann, Leipzig.
Berkeley, E. 1953, Morphological studies in the Celastraceae. J. Elisha Mitchell Sci. Soc. 69:185-206.
Bernhard, A. 1999a. Flower structure, development, and systematics in Passifloraceae and in Abatia (Flacourtiaceae). Int. J. Pl. Sci. 160: 135-150.
-----. 1999b. Floral structure and development of Ceratiosicyos laevis (Achariaceae) and its systematic position. Bot. J. Linn. Soc. 131; 103-113.
----- & P. K. Endress. 1999. Androecial development and systematics in Flacourtiaceae s.1. P1. Syst. & Evol. 215: 141-155.
Bernhardt, P. 1996. Anther adaptation in animal pollination. Pp. 192-220 in W. G. D'Arcy & R. C. Keating (eds.), The anther: Form, function and phylogeny. Cambridge Univ. Press, Cambridge.
Beusekom, C. F. van. 1971. Revision of Meliosma (Sabiaceae), section Lorenzanea excepted, living and fossil, geography and phylogeny. Blumea 19: 355-529.
Blaser, J. L. 1954. The morphology of the flower and inflorescence of Mitchella repens. Amer. J. Bot. 41: 533-539.
Bocquillon, H.-T. 1861a. Observations sur le genre Oftia Adans. Adansonia 2: 5-12.
-----. 1861b. Revue du groupe des Verbenacees. Adansonia 2: 81-165.
-----. 1866. Memoire sur le groupe des Tiliacces. Adansonia 7:17-64.
Bogle, A. L. 1989. The floral morphology, vascular anatomy, and ontogeny of the Rbodoleioideae (Ha-mamelidaceae) and their significance in relation to the 'lower' hamamelids. Pp. 201-220 in P. R. Crane & S. Blackmore (eds.), Evolution, systematics, and fossil history of the Hamamelidac. Vol. 1, Introduction and 'lower' Hamamelidae. Clarendon Press, Oxford.
Bowman, J. L. & D. R. Smyth. 1998. Patterns of petal and stamen reduction in Australian species of Lepidium L. (Brassicaceac). Int. J. P1. Sci. 159: 65-74.
-----, ----- & E. M. Meyerowitz. 1991. Genetic interactions among floral homeotic genes of Arabidopsis. Development 112:1-20.
Breteler, F. J. 1973. The African Dichapetalaceae: A taxonomical revision. Meded. Landbouwhoogeschool Wageningen 73: 1-124.
Brizicky, G. K. 1959. Variability in the floral parts of Gomortega (Gomortegaceae). Willdenowia 2: 200-207.
Brough, P. 1933. The life-history of Grevillea robusta (Cunn.). Proc. Linn. Soc. N.S.W. 58: 33-73.
Buxbaum, F. 1961. Vorlaufigc Untersuchungen Uber Umfang, Systematischen Stellung und Gliederung der Caryophyllales (Centrospermae). Beitr. Biol. Pflanzen 36: 1-56.
Caris, P. 1998. Bloemontogenie en fylogenie van de Myrsinaceae en aanverwante taxa. Ph.D. diss., Katholieke Universiteit Leuven.
Chase, M. W., D. E. Soltis, R. G. Olmstead, D. Morgan, D. H. Les, B. D. Mishler, M. R. Duvall, R. A. Price, H. G. Hills, Y.-L. Qiu, K. A. Kron, J. H. Rettig, E. Conti, J. D. Palmer, J. R. Man hart, K. J. Sytsma, H. J. Michaels, W. J. Kress, K. G. Karol, W. D. Clark, M. Heden, B. S. Gaut, R. K. Jansen, K.-J. Kim, C. F. Wimpee, J. F. Smith, G. R. Furnier, S. H. Strauss, Q.-Y. Xiang, G.M. Plunkett, P.S. Soltis, S. M. Swensen, S. E. Williams, P. A. Gadek, C. J. Quinn, L. E. Eg-uiarte, E. Golenberg, G. H. Learn Jr., S. W. Graham, S.C. H. Barrett, S. Dayanandan & V. A. Albert. 1993. Phylogenetics of seed plants: An analysis of nucleotide sequences from the plastid gene rbcL. Ann. Missouri Bot. Gard. 80: 528-580.
Chatin, A. 1873a. Sur I'organogenie de I'androcee des Labiees, des Globulariees et des Scrofularinees. Bull. Soc. Bot. France 20: 41-45.
-----. 1873b. Organogenie comparee de I'androcee dans ses rapports avec les affinites naturelles. Bull. Soc. Bot. France 20: 327-330.
Clifford, H. T. & J. G. Conran. 1998. Johnsoniaceae. Pp. 336-340 in K. Kubitzki (ed.), The families and genera of vascular plants. Vol. 3, Flowering plants: Monocotyledons: Lilianae (except Orchi-daceae). Springer-Verlag, Berlin.
Cocucci, A. E. 1975. Estudios en el genero Prosopanche (Hydnoraceae), II: Organizacion de la flor. Kurtziana 8: 7-15.
Coen, E. 5. 1991. The role of homeotic genes in flower development and evolution. Annual Rev. P1. Physiol. P1. Molec. Biol. 42: 241-279.
----- & E. M. Meyerowitz. 1991. The war of the whorls: Genetic inteactions controlling flower development. Nature (London) 353: 31-37.
-----, J. M. Nugent, D. Luo, D. Bradley, P. Cubas, M. Chadwick, L. Copsey & R. Carpenter. 1995. Evolution of floral symmetry. Philos. Trans. Roy. Soc. London, B, 350: 35-38.
Coaran, J. G. 1998. Lomandraceae. Pp. 354-365 in K. Kubitzki (ed.), The families and genera of vascular plants. Vol. 3, Flowering plants: Monocotyledons: Lilianae (except Orchidaceae). Springer-Verlag, Berlin.
Cook, C. D. K. 1998. Hydrocharitaceae, and Pontederiaceae. Pp. 234-248, 395-403 in K. Kubitzki (ed.), The families and genera of vascular plants. Vol.4, Flowering plants: Monocotyledons: Alismatanae and Commelinanae (except Gramineae). Springer-Verlag, Berlin.
Copeland, H. F. 1961. Observations on the reproductive structures of Anacardium occidentale. Phytomorphology 11:315-325.
Crane, P. R., E. M. Frus & K. R. Pedersen. 1994. Palaeobotanical evidence on the early radiation of magnoliid angiosperms. P1. Syst. & Evol. (Suppl.) 8: 51-72.
Crepet, W. L. & K. C. Nixon. 1996. The fossil history of stamens. Pp. 25-57 in W. G. D'Arcy & R. C. Keating (eds.), The anther: Form, function and phylogeny. Cambridge Univ. Press, Cambridge.
Cronquist, A. 1981. An integrated system of classification of flowering plants. Columbia Univ. Press, New York.
-----. 1988. The evolution and classification of flowering plants. Ed. 2. New York Bot. Gard., Bronx.
Dahlgren, R. & A. E. van Wyk. 1988. Structures and relationships of families endemic to or centered in southern Africa. Monogr. Syst. Bot. Missouri. Bot. Gard. 25: 1-94.
D'Arcy, W. C. 1996. Anthers and stamens and what they do. Pp. 1-24 in W. G. D'Arcy & R. C. Keating (eds.), The anther: Form, function and phylogeny. Cambridge Univ. Press, Cambridge.
Daumann, E. 1931. Zur morphologische Wertigkeit der Blutennektarien von Laurus. Beih. Bot. Centralbl. 48: 183-208.
Dawson, M. L. 1936. The floral morphology of the Polemoniaceae. Amer. J. Bot. 23: 501-511.
De Wilde, W. J. J. O. 1974. The genera of tribe Passifloreae (Passifloraceae), with special reference to flower morphology. Blumea 22: 37-50.
Dickison, W. C. 1994. A re-examination of Sanango racemosum, 2: Vegetative and floral anatomy. Taxon 43: 601-618.
----- & A. L. Weitzman. 1998. Floral morphology and anatomy of Bonnetiaceae. J. Torrey Bot. Soc. 125: 268-286.
Dickson, J. 1936. Studies in floral anatomy, III: An interpretation of the gynoeceum in the Primulaceae. Amer. J. Bot. 23: 385-393.
Domke, W. 1934. Untersuchungen uber die systematische und geographische Gliederung der Thymelaeaceen nebst eines Neubeschreibung ihrer Gattungen. Biblioth. Bot. 27: 1-151.
Douglas, A. E. 1997. The developmental basis of morphological diversification and synorganization in flowers of Conospermeae (Stirlingia and Conosperminae: Proteaceae). Int. J. P1. Sci. 158 (6 Suppl.): S13-S48.
----- & S. C. Tucker. 1996. Comparative floral ontogenies among Persoonioideae including Bellendena (Proteaceae). Amer. J. Bot. 83: 1528-1555.
Douglas, G. E. 1936. Studies in the vascular anatomy of the Primulaceac. Amer. J. Bot. 23: 199-212.
Drinnan, A. E., P.R. Crane & S. B. Hoot. 1994. Patterns of floral evolution in the early diversification of non-magnoliid dicotyledons (eudicots). P1. Syst. & Evol. (Suppl.) 8: 93-122.
Eames, A. J. 1931. The vascular anatomy of the flower with refutation of the theory of carpel polymorphism. Amer. J. Bot. 18: 147-188.
-----. 1953. Floral anatomy as an aid in generic limitation. Chron. Bot. 14: 126-132.
-----. 1961. Morphology of the angiosperms. McGraw-Hill, New York.
Eckardt, T. 1959. Das Blutendiagramm von Batis P. Br. Ber. Deutsch. Bot. Ges. 72: 411-418.
Eichler, A. W. 1875-1878. Bluthendiagramme I & II. Engelmann, Leipzig.
Eklund, H. 2000. Lauraceous flowers from the late Cretaceous of North Carolina, U.S.A. Bot. J. Linn. Soc. 132: 397-428.
Eliasson, U. H. 1988. Floral morphology and taxonomic relations among the genera of Amaranthaceae in the New World and the Hawaiian Islands. Bot. J. Linn. Soc. 96: 235-283.
Emberger, M. L. 1939. Recherches sur la fleur des Polygonacees. Rev. Gen. Bot. 51: 581-599.
Endress, P. K. 1967. Systematische Studien uber die verwandtschaftlichen Beziehungen zwischen den Hamamelidaccen und Betulaceen. Bot. Jahrb. Syst. 87: 431-525.
-----. 1980. Floral structure and relationships of Hortonia (Monimiaceae). Pl. Syst. & Evol. 133:199-221.
-----. 1984. The role of inner staminodes in the floral display of some relic Magnoliales. Pl. Syst. & Evol. 146: 269-282.
-----. 1986. Reproductive structures and phylogenetic significance of extant primitive angiosperms. Pl. Syst. & Evol. 152: 1-28.
-----. 1990a. Patterns of floral construction in ontogeny and phylogeny. Biol. J. Linn. Soc. 39:153-175.
-----. 1990b. Evolution of reproductive structures and functions in primitive angiosperms (Magnoliidae). Mem. New York Bot. Gard. 55: 5-34.
-----. 1994. Diversity and evolutionary biology of tropical flowers. Cambridge Univ. Press, Cambridge.
-----. 1995. Floral structure and evolution in Ranunculaceae. P1. Syst. & Evol. (Suppl.) 9: 47-61.
-----. 1997. Relationships between floral organization, architecture, and pollination mode in Dillenia (Dilleniaceae). P1. Syst. & Evol. 206: 99-118.
-----. 1998. Antirrhinum and Asteridae: Evolutionary changes of floral symmetry. Symposia of the Society for Experimental Biology 51: 133-140.
-----. 1999. Symmetry in flowers: Diversity and evolution. Int. J. Pl. Sci. 160: S3-S23.
----- & L. Hufford. 1989. The diversity of stamen structures and dehiscence patterns among Magnoliidae. Bot. J. Linn. Soc. 100: 45-85.
----- & S. Stumpf. 1990. Non-tetrasporangiate stamens in the angiosperms: Structure, systematic distribution and evolutionary aspects. Bot. Jahrb. Syst. 112:193-240.
----- & -----. 1991. The diversity of stamen structures in 'lower' Rosidae (Rosales, Fabales, Proteales, Sapindales). Bot. J. Linn. Soc. 107: 217-293.
Engler, A. 1930a. Podostemonaceae. Pp. 3-68 in A. Engler & K. Prantl (eds.), Die naturlichen Pflanzenfamilien. Ed. 2. Vol. 18a. Engelmann, Leipzig.
-----. 1930b. Saxifragaceae. Pp. 74-226 in A. Engler & K. Prantl (eds.), Die naturlichen Pflanzenfamilien. Ed. 2. Vol. 18a. Engelmann, Leipzig.
-----. 1931a. Zygophyllaceae. Pp. 144-184 in A. Engler & K. Prantl (eds.), Die naturlichen Pflanzenfamilien. Ed. 2. Vol. 19a. Engelmann, Leipzig.
-----. 1931b. Rutaceae. Pp. 187-359 in A, Engler & K. Prantl (eds.), Die naturlichen Pflanzenfamilien. Ed. 2. Vol. 19a. Engelmann, Leipzig.
-----. 1931c. Simarubaceae. Pp. 359-405 in A. Engler & K. Prantl (eds.), Die naturlichen Pflanzenfamilien. Ed. 2. Vol. 19a. Engelmann, Leipzig.
-----. 1931d. Burseraceae. Pp. 405-456 in A. Engler & K. Prantl (eds.), Die naturlichen Pflanzenfamilien. Ed. 2. Vol. 19a. Engelmann, Leipzig.
Erbar, C., S. Kusma & P. Leins. 1998. Development and interpretation of nectary organs in Ranunculaceae. Flora 194: 317-332.
Erdtman, G., P. Leins, R. Melville & C. R. Metecalfe. 1969. On the relationships of Emblingia. Bot. J. Linn. Soc. 62: 169-186.
Eyde, R. E. & J. T. Morgan. 1973. Floral structure and evolution in Lopezieae (Onagraceae). Amer. J. Bot. 771-787.
Faden, R. B. 1998. Commelinaceae. Pp. 109-128 in K. Kubitzki (ed.), The families and genera of vascular plants. Vol. 4, Flowering plants: Monocotyledons: Alismatanae and Commelinanae (except Gramineae). Springer-Verlag, Berlin.
Friedrich, H.-C. 1956. Studien uber die naturliche Verwandtschaft der Plumbaginales und Centrospermae. Phyton (Horn, Austria) 6: 220-263.
Fukuoka, N., M. Ito & K. Iwatsuki. 1986. Floral anatomy of the mangrove genus Lumnizera (Combretaceae). Acts Phytotax. Geobot. 37: 69-81.
Gandhi, K. N. & R. D. Thomas. 1983. A note on the androecium of the genus Croton and flowers in general of the family Euphorbiaceae. Phytologia 54: 6-8.
Gandolfo, M. A., K. C. Nixon & W. L. Crepet. 1998. A new fossil flower from the Turonian of New Jersey: Dressiantha bicarpellata gen. et sp. nov. (Capparales). Amer. J. Bot. 85: 964-974.
Gerrath, J., C. R. Lacroix & U. Posluszny. 1990. The developmental morphology of Leea guineensis, II: Floral development. Bot. Gaz. 151: 210-220.
Gilg, E. 1894. Studien uber die Verwandtschaftsberhaltnisse der Thymelaeales und uber die anatomische Methode: Thymelaeaceae. Bot, Jahrb. Syst. 18: 488-574.
-----. 1925. Ochnaceae. Pp. 53-87 in A. Engler & K. Prantl (eds.), Die naturlichen Pflanzenfamilien. Ed. 2. Vol. 21. Engelmann, Leipzig.
Goebel, K. 1933. Organographie der Pflanzen insbesondere der Archegoniaten und Samenpflanzen. Part 3. Ed. 3. Gustav Fischer, Jena.
Goldberg, A. 1986. Classification, evolution and phylogeny of the families of dicotyledons. Smithsonian. Contrib. Bot. 58: 1-314.
Gustafsson, M. G. H. 2000. Floral morphology and relationships of Clusia gundlachii with a discussion of floral organ identity and diversity in the genus Clusia. Int. J. Pl. Sci. 161: 43-53.
----- & Albert, V. A. 1999. Inferior ovaries and angiosperm diversification. Pp. 403-431 in P. M. Hollingsworth, R. M. Bateman & R. J. Gomall (eds.), Molecular systematics and plant evolution. Taylor & Francis, London.
Gutzwiller, M.-A. 1961. Die phylogenetisehe Stellung von Suriana maritima L. Bot. Jahrb. Syst. 81: 1-49.
Haber, J. M. 1959. The comparative anatomy and morphology of the flowers and inflorescences of the Proteaceae, I: Some Australian taxa. Phytomorphology 9: 325-358.
-----. 1961. The comparative anatomy and morphology of the flowers and inflorescences of the Proteaceae, II: Some American taxa. Phytomorphology 11: 1-16.
-----. 1966. The comparative anatomy and morphology of the flowers and inflorescences of the Proteaceae, III: Some African taxa. Phytomorphology 16: 490-527.
Hakki, M. I. 1998. On floral morphology and embryology of Usteria guineensis Willd. (Loganiaceae). Bot. Jahrb. Syst. 120: 275-293.
Harms, H. 1925a. Passifloraceae. Pp. 470-507 in A. Engler & K. Prantl (eds.), Die naturlichen Pflanzenfamilien. Ed. 2. Vol. 21. Engelmann, Leipzig.
-----. 1925b. Achariaceae, Pp. 507-510 in A. Engler & K. Prantl (eds.), Die naturlichen Pflanzenfamilien. Ed. 2. Vol. 21. Engelmann, Leipzig.
-----. 1960. Meliaceae. Pp. 1-172 in A. Engler & K. Prantl (eds,), Die naturlichen Pflanzenfamilien. Ed. 2. Vol. 19b1. Duncker & Humblot, Berlin.
Hartog, M. M. 1878. On the floral structure and affinities of Sapotaceae. J. Bot. 16: 65-72.
Haynes, R. R., D. H. Les & L. B. Holm-Nielsen. 1998. Limnocharitaceae. Pp. 271-275 in K. Kubitzki (ed.), The families and genera of vascular plants. Vol. 4, Flowering plants: Monocotyledons: Alismatanae and Commelinanae (except Gramineae). Springer-Verlag, Berlin.
Heel, W. A. van. 1966. Morphology of the androecium in Malvales. Blumea 13: 177-394.
-----. 1967. Anatomical and ontogenetic investigations on the morphology of the flowers and the fruit of Scyphostegia borneensis Stapf (Scyphostegiaceae). Blumea 15: 107-125.
Heinig, K. 1951. Studies in the floral morphology of the Thymelaeaceae. Amer. J. Bot. 38: 113-132.
He[beta], D. 1983. Die Blute. Einfuhtrung in Struktur und Funktion, Okologie und Evolution der Bluten. UI-mer, Stuttgart.
Hiepko, P. 1965. Vergleichend-morphologische und entwicklungsgeschichtliche Untersuchungen uber das Perianth bei den Polycarpicae. Bot. Jahrb. Syst. 84: 359-508.
-----. 1966. Zur Morphologie, Anatomie und Funktion des Discus der Paeoniaceae. Ber. Deutsch. Bot. Ges. 79: 233-245.
Hofmann, U. 1993. Flower morphology and ontogeny. Pp. 123-166 in H.-D. Behnke & T. J. Mabry (eds.), Caryophyllales: Evolution and systematics. Springer-Verlag, Berlin.
----- & J. Gottmann. 1990. Marina L. und Triplostegia Wall. ex DC. im Vergleich mit Valerianaceae und Dipsacaceae. Bot. Jahrb. Syst. 111: 499-553.
Huber, H. 1998. Dioscoreaceac. Pp. 216-235 in K. kubitzki (ed.), The families and genera of vascular plants. Vol. 3, Flowering plants: Monocotyledons: Lilianae (except Orchidaceae). SpringerVerlag, Berlin.
Hufford, L. 1990. Androecial development and the problem of monophyly of Loasaceae. Canad. J. Bot. 68: 402-419.
----- & P. K. Endress. 1989. The diversity of anther structures and dehiscence patterns among Hamamelididae. Bot. J. Linn. Soc. 99: 301-346.
Hutchinson, J. 1959. The families of flowering plants. Vol. 1, Dicotyledons. Ed. 2. Clarendon Press, Oxford.
Ihlenfeldt, H. D. 1960. Entwicklungsgeschichtliche, morphologische und systematische Untersuchungen an Mesembryanthemen. Feddes Repert. 63: 1-104.
Innes, R. L., W. R. Remphrey & L. M. Lenz. 1989. An analysis of the development of single and double flowers in Potentilla fruticosa. Canad. J. Bot. 67: 1067-1079.
Johnson, D. S. 1935. The development of the shoot, male flower and seedling of Batis maritima L. Bull. Torrey Bot. Club 62: 19-32.
Joshi, A. C. 1932. Dedoublement of stamens in Achyranthes aspera, Linn. J. Indian Bot. Soc. 11: 335-339.
-----. 1936. Anatomy of the flowers of Stellera chamaejasme. J. Indian Bot. Soc. 15: 77-85.
----- & C. Venkata Rao. 1934. A contribution to the anatomy, morphology and cytology of the flower of Digera arvensis Forsk. J. Indian Bot. Soc. 13: 201-228.
Juncosa, A. M. 1988. Floral development and character evolution in Rhizophoraceae. Pp. 83-101 in P. Leins, S. C. Tucker & P. K. Endress (eds.), Aspects of floral development. Cramer, Berlin.
----- & P. B. Tomlinson. 1987. Floral development in mangrove Rhizophoraceac. Amer. J. Bot. 74: 1263-1279.
Kallunki, J. A. 1998. Andreadoxa flava (Rutaceac, Cuspariinae): A new genus and species from Bahia, Brazil. Brittonia 50: 59-62.
Karrer, A. B. 1991. Blutenentwicklung und systematische Stellung der Papaveraceae und Capparaceae. Ph.D. diss, Univ. of Zurich.
Kasapligil, B. 1951. Morphological and ontogenetic studies of Umbellularia californica Nutt. and Laurus nobilis L. Univ. Calif. Publ. Bot. 25: 115-240.
Khosla, C. & H. V. Mohan Ram. 1993. Morphology of flower, fruit and seed in Polypleurum stylosum. Aquatic Bot. 46: 255-262.
Kirchoff, B. K. 1991. Homeosis in the flowers of the Zingiberales. Amer. J. Bot. 78: 833-837.
Klopfer, K. 1972. Beitrage zur floralen Morphogenese und Histogenese der Saxifragaceae, VII: Parnassia palustris und Francoa sonchifolia. Flora 161: 320-332.
Kluge, A. G. 1988. The characteristics of ontogeny. Pp. 57-81 in C. J. Humphries (ed.), Ontogeny and systematics. British Museum (National History), London.
Kopka, S. & F. Weberling. 1983. Zur Morphologie und Morphogenese der Blute von Vochysia acuminata Bong. subsp. laurifolia (Warm.) Stafleu (Vochysiaceae). Beitr. Biol. Pflanzen 59: 273-302.
Kosuge, K. 1993. Petal evolution of the Polycarpicae. Abstr. XV Int. Bot. Congr., Yokohama, p. 92.
-----. 1994. Petal evolution in Ranunculaceae. Pl. Syst. & Evol. (Suppl.) 8: 185-191.
Kral, R. 1998. Xyridaceae. Pp. 461-469 in K. Kubitzki (ed.), The families and genera of vascular plants. Vol. 4, Flowering plants: Monocotyledons: Alismatanae and Commelinanae (except Gramineae). Springer-Verlag, Berlin.
Kramer, E. M. & V. F. Irish. 2000. Evolution of the petal and stamen developmental programs: Evidence from comparative studies of the lower eudicots and basal Angiosperms. Int. J. Pl. Sci. 161 (6 Suppl.): 529-540.
Krause, J. 1960. Corynocarpaceae. Pp.22-35 in A. Engler & K. Prantl (eds.), Die naturlichen Pflanzen-familien. Ed. 2. Vol. 20b. Duneker & Humblot, Berlin.
Kress, W. J. 1990. Phylogeny and classification of Zingiberales. Ann. Missouri Bot. Gard. 77: 698-721.
Kshetrapal, S. 1970. A contribution to the vascular anatomy of the flower of certain species of the Salvadoraceae. J. Indian Bot. Soc. 49: 92-99.
-----. 1973. Vascular anatomy of the node and flower of Happea dichotoma Willd. Bot. Gaz. 134: 1-4.
Kubitzki, K. 1969. Monographie der Hernandiaceen. Bot. Jahrb, Syst. 89: 78-209.
Kumar, A. 1976. Studies in Geraniales, II: The floral anatomy. J. Indian Bot. Soc. 55: 233-253.
Larsen, K. 1998. Costaceae. Pp. 128-132 in K. Kubitzki (ed.), The families and genera of vascular plants. Vol. 4, Flowering plants: Monocotyledons: Alismatanae and Commelinanae (except Gramineae). Springer-Verlag, Berlin.
J. M. Lock, H. Maas & P. J. M. Maas. 1998. Zingiberaceae. Pp. 474-495 in K. Kubitzki (ed.), The families and genera of vascular plants. Vol. 4, Flowering plants: Monocotyledons: Alismatanae and Commelinanae (except Gramineae). Springer-Verlag, Berlin.
Lawrence, J. R. 1937. A correlation of the taxonomy and the floral anatomy of certain species of the Boraginaceae. Amer. J. Bot. 24: 433-444.
Lehmann, N. L. & R. Sattler. 1989. Floral development and homeosis in Begonia semperflorens: Cultorum "Cinderella." Amer. J. Bot. (Suppl.) 76: 43 (abstract).
----- & -----. 1992. Irregular floral development in Calla palustris (Araceae) and the concept of homeosis. Amer. J. Bot. 79: 1145-1157.
----- & -----. 1993. Homeosis in floral development of Sanguinaria canadensis and S. canadensis "Multiplex" (Papaveraceae). Amer. J. Bot. 80: 1323-1335.
----- & -----. 1994. Floral development and homeosis in Actaea rubra (Ranunculaceae). Int. J. Pl. Sci. 155: 658-671.
Leins, P. 1964. Die fruhe Blutenentwicklung von Hypericum hookerianum Wight et Arn. und H. aegypticum L. Ber. Deutsch. Bot. Ges. 77: 112-123.
Li, P. & M. 0. Johnston. 2000. Heterochrony in plant evolutionary studies through the twentieth century. Bot. Rev. (Lancaster) 66: 57-88.
Link, D. A. 1991. The floral nectaries of Geraniales, III: Lepidobotryaceae J. Leonard. Bull. Jard. Bot. Belg. 61: 347-354.
-----. 1992. The floral nectaries of the Geraniales and their systematic implications, IV: Ctenolophonaceae Badre. Flora 187: 103-107.
Litt, A. 1997. The single fertile stamen of the Vochysiaceae: Derived four times or once? Amer. J. Bot. (Suppl.) 84: 48 (abstract).
Maas-van der Kamer, H. & T. Weustenfeld. 1998. Triuridaceae. Pp. 452-458 in K. Kubitzki (ed.), The families and genera of vascular plants. Vol. 3, Flowering plants: Monocotyledons: Lilianae (except Orchidaceae). Springer-Verlag, Berlin.
MacIntyre, J. P. & C. R. Lacroix. 1996. Comparative development of perianth and androecial primordia of the single flower and the homeotic double-flowered mutant in Hibiscus rosa-sinensis (Malvaceae). Canad. J, Bot. 74: 1871-1882.
Mair, O. 1977. Zur Entwicklungsgeschichte monosymmetrischer Dicotylen-Bluten. Diss. Bot. 38: 1-88.
Mattfeld, J. 1960a. Salvadoraceae. Pp. 232-239 in A. Engler & K. Prantl (eds.), Die naturlichen Pflanzenfamilien. Ed. 2. Vol. 20b. Duncker & Humblot, Berlin.
-----. 1960b. Stackhousiaceae. Pp. 240-254 in A. Engler & K. Prantl (eds.), Die naturlichen Pflanzenfamilien. Ed. 2. Vol. 20b. Duncker & Humblot, Berlin.
McConchie, C. A. & J. W. Kadereit. 1987. Floral structure of Vallisneria Caulescens Bailey & F. Mueller. Aquatic Bot. 29: 101-110.
Melville, R. 1984. The affinity of Paeonia and a second genus of Paeoniaceae. Kew Bull. 38: 87-105.
Michaelis, P. 1924. Blutenmorphologische Untersuchungen an den Euphorbiaceen. Bot. Abh. 3:1-150.
Mione, T. & A. L. Bogle. 1990. Comparative ontogeny of the inflorescence and flower of Hamamelis virginiana and Loropetalum chinense (Hamamelidaceae). Amer. J. Bot. 77: 77-91.
Moncur, M. W. 1988. Floral development of tropical and subtropical fruit and nut species: An atlas of scanning electron micrographs. Natural Resources Series, 8. CSIRO, Melbourne.
Moore, J. A. 1936a. The vascular anatomy of the flower in the Papilionaceous Leguminosae, I. Amer. J. Bot. 23: 279-290.
-----. 1936b. The vascular anatomy of the flower in the Papilionaceous, Leguminosae, II. Amer. J. Bot. 23: 349-355.
Morgan, D. R. & D. E. Soltis. 1993. Phylogenetic relationships among members of Saxifragaceae sensu lato based on rbcL sequence data. Ann. Missouri Bot. Gard. 80: 631-660.
Murray, M. A. 1945. Carpellary and placental structure in the Solanaceae. Bot. Gaz. 107: 243-260.
Murty, Y. S. 1954. Studies in the order Parietales. J. Indian Bot. Soc. 98: 226-238.
Nair, N. C. & R. K. Jain. 1956. Floral morphology and embryology of Balanites roxburghii Planch. Lloydia 19: 269-279.
----- & T. C. Joseph. 1957. Floral morphology and embryology of Samadera indica. Bot. Gaz. 119: 104-115.
----- & R. K. Joshi. 1958. Floral morphology of some members of the Simaroubaceae. Bot. Gaz. 120: 88-99.
----- & P. N. N. Nambisan. 1957. Contribution to the floral morphology and embryology of Leea sambucina Willd. Bot. Not. 110: 160-172.
----- & V. S. Sarma. 1961. Organography and floral anatomy of some members of the Rhamnaceae. J. Indian Bot. Soc. 40: 47-55.
Nandi, O. 1998. Floral development and systematics of Cistaceae. P1. Syst. & Evol. 212: 107-134.
Narayana, L. L. 1960. Studies in Burseraceae, II. J. Indian Bot. Soc. 39: 204-209.
-----. 1964. A contribution to the floral anatomy and embryology of Linaceae. J. Indian Bot. Soc. 43: 344-357.
-----. 1966. A contribution to the floral anatomy of Oxalidaceae. J. Jap. Bot. 41: 321-328.
----- & D. Rao. 1966. Floral morphology of Linaceae. J. Jap. Bot. 41: 1-10.
----- & -----. 1969. Contributions to the floral anatomy of Humiriaceae, 1. J. Jap. Bot. 44: 328-335.
----- & -----. 1971. Contributions to the floral anatomy of Linaceae, II. Phytomorphology 21: 64-67.
----- & -----. 1974. Contributions to the floral anatomy of Linaceae, IV. Curr. Sci. (Bangalore) 43: 391-393.
----- & -----. 1976a. Contributions to the floral anatomy of Linaceae, 6. J. Jap. Bot, 51: 92-96.
----- & -----. 1976b. Contributions to the floral anatomy of Linaceae, 7. J. Jap. Bot. 51: 349-352.
----- & -----. 1977a. Contributions to the floral anatomy of Linaceae, 8. J. Jap. Bot. 52: 56-59.
----- & -----. 1977b. Contributions to the floral anatomy of Humiriaceae, 6. J. Jap. Bot. 52: 145-153.
----- & -----. 1977c. Contributions to the floral anatomy of Linaceae, 9. J. Jap. Bot. 52: 231-234.
----- & -----. 1978. Systematic position of Humiriaceae, Linaceae and erythroxylaceae in the light of their comparative floral morphology and embryology: A discussion. J. Indian Bot. Soc. 57: 258-266.
-----, P. Satyanarayana & P. S. Prakasa Rao. 1986. The floral anatomy of Corynocarpus laevigatus (Corynocarpaceae). Phytomorphology 36: 325-329.
Nelson, G. 1978. Ontogeny, phylogeny, paleontology, and the biogenetic law. Syst. Zool. 27: 324-345.
Neubauer, H. F. 1959. Uber das Staminodium von Kigelia aethiopica Decne. Oesterr. Bot. Z. 106: 546-550.
Norris, T. 1941. Torus anatomy and nectary characteristics as phylogenetic criteria in the Rhocadales. Amer. J. Bot. 28: 101-113.
Palser, B. F. 1962. Studies of floral morphology in the Ericales, VI: The Diapensiaceae. Bot. Gaz. 124: 200-219.
Patterson, C. 1982. Morphological characters and homology. Pp.21-74 in K. A. Joysey & A. E. Friday (eds.), Problems of phylogenetic reconstruction. Systematics Association Special Volume No. 21. Academic Press, London.
Payer, J. B. 1857. Traite d'organogenie comparee de la fleur. Masson, Paris.
Pax, F. & K. Hoffmann. 1936. Capparidaceae. Pp. 146-223 in A. Engler & K. Prantl (eds.), Die naturlichen Pflanzenfamilien. Ed. 2. Vol. 17b. Engelmann, Leipzig.
Pennington, T. D. 1991. Morphology. Pp. 75-85 in T. D. Pennington, The genera of Sapotaceae. Roy. Bot. Gard., Kew, England, & New York Bot. Gard., Bronx.
Philipson, W. R. 1987. Corynocarpus J. R. & G. Forst.: An isolated genus. Bot. J. Linn. Soc. 95: 9-18.
Prichard, E. C. 1955. Morphological studies in Rhamnaceae. J. Elisha Mitchell Sci. Soc. 71: 82-106.
Puri, V. 1948. Studies in floral anatomy, V: On the structure and nature of the corona in certain species of the Passifloraceae. J. Indian Bot. Soc. 27: 130-149.
-----. 1951. Role of floral anatomy in the solution of morphological problems. Bat. Rev. (Lancaster) 17: 472-534.
Radlkofer, I. 1896. Sapindaceac. Pp. 277-366 in A. Engler & K. Prantl (eds.), Die naturlichen Pflanzenfamilien. Vol. III, 5. Engelmann, Leipzig.
Raghavan, T. S. 1939. Studies in the Capparidaceae, II: Floral anatomy and some structural features of the Capparidaceous flower. J. Linn. Soc. Bot. 52: 239-257.
Rahn, K. 1998. Alliaceae, and Themidaceae. Pp.70-78, 436-440 in K. Kubitzki (ed.), The families and genera of vascular plants. Vol. 3, Flowering plants: Monocotyledons: Lilianae (except Orchidaceae). Springer-Verlag, Berlin.
Rao, V. S. & Arati Ganguli. 1963. Studies in the floral anatomy of the Apocynaceae. J. Indian Bot. Soc. 42: 419-435.
Reece, P. C. 1939. The floral anatomy of the avocado. Amer. J. Bot. 26: 429-432.
Reed, C. F. 1955. The comparative morphology of the Olacaceae, Opiliaceae and Octoknemaceae. Mem. Soc. Brot. 10: 29-79.
Reeves, P. A. & R. G. Olmstead. 1998. Evolution of novel morphological and reproductive traits in a clade containing Antirrhinum majus (Scrophulariaceae). Amer. J. Bot. 85: 1047-1056.
Ridsdale, C. E. 1974. A revision of the family Leeaceae. Blumea 22: 57-100.
Rodriguez-Riano, T., A. Ortega-Olivencia & J. A. Devesa. 1999. Types of androecium in the Fabaceae of SW Europe. Ann. Bot. (London) 83: 109-116.
Rohweder, O. & K. Huber. 1974. Centrospermen-Studien, 7: Beobachtungen und Anmerkungen zur Morphologie und Entwicklungsgeschichte einiger Nyctaginaceen. Bot. Jahrb. Syst. 94: 327-359.
Rohwer, J. 1994. A note on the evolution of the stamens in the Laurales, with emphasis on the Lauraceae. Bot. Acta 107: 103-110.
Ronse Decraene, L. P. 1985. Een bijdrage tot de klassifikatie van Dilleniidae en Rosidae sensu Cronquist met behulp van meeldraadkenmerken. Ph.D. diss., Katholieke Universiteit Leuven.
-----. 1989. Floral development of Cochlospermum tinctorium and Bixa orellana with special emphasis on the androecium. Amer. J. Bot. 76: 1344-1359.
-----. 1992. The androecium of the Magnoliophytina: Characterisation and systematic importance. Ph.D. diss., Katholieke Universiteit Leuven.
----- & J. R. Akeroyd. 1988. Generic limits in Polygonum and related genera (Polygonaceac) on the basis of floral characters. Bot. J. Linn. Soc. 98: 321-371.
----- & E. Smets. 1990a. The floral development of Popowia whitei (Annonaceae). Nord. J. Bot. 10: 411-420. (Corrected in 1991 in Nord. J, Bot. 11: 420.)
----- & -----. 1990b. The systematic relationship between Begoniaceae and Papaveraceae: A comparative study of their floral development. Bull. Jard. Bot. Belg. 60: 229-273.
----- & -----. 1991a. Androecium and floral nectaries of Harungana madagascariensis (Clusiaceae). Pl. Syst. & Evol. 178: 179-194.
----- & -----. 199lb. The impact of receptacular growth on polyandry in the Myrtales. Bot. J. Linn. Soc. 105: 257-269.
----- & -----. 1991c. The floral nectaries of Polygonum s.1. and related genera (Persicarieae and Polygoneae): Position, morphological nature and semophylesis. Flora 185: 165-185.
----- & -----. 1992. Complex polyandry in the Magnoliatae: Definition, distribution and systematic value. Nord. J. Bot. 12: 621-649.
----- & -----. 1993. The distribution and systematic relevance of the androecial character polymery. Bot. J. Linn. Soc. 113: 285-350.
----- & -----. 1994. Merosity in flowers: Definition, origin, and taxonomic significance. Pl. Syst. & Evol. 191: 83-104.
----- & -----. 1995a. The distribution and systematic relevance of the androecial character oligomery. Bot. J. Linn. Soc. 118:193-247.
----- & -----. 1995b. Evolution of the androecium in the Ranunculiflorae. Pl, Syst. & Evol. (Suppl.) 9: 63-70.
----- & -----. 1998. Notes on the evolution of androecial organisation in the Magnoliophytina (Angiosperms). Bot. Acta 111: 77-86.
----- & -----. 1999. Similarities in floral ontogeny and anatomy between the genera Francoa (Francoaceae) and Greyia (Greyiaceae). Int. J. Pl. Sci. 160: 377-393.
-----, E. Smets & D. Clinckemaillie. 1995. The floral development and floral anatomy of Coris monspeliensis. Canad. J. Bot. 73: 1687-1698.
-----, J. De Laet & E. Smets. 1998a. Floral development and anatomy of Moringa oleifera (Moringaceae): What is the evidence for a caparalean or sapindalean affinity? Ann. Bot. (London) 82: 273-284.
-----, E. Smets & P. Vanvinckenroye. 1998b. Pseudodiplostemony, and its implications for the evolution of the androecium in the Caryophyllaceae. J. Pl. Res. 111: 25-43.
-----, H. P. Linder, T. Diamini & E. Smets. 2001. Evolution and development of floral diversity of Melianthaceae, an enigmatic Southern African family. Int. J. Pl. Sci. 162: 59-82.
Rutishauser, R. 1997. Structural and developmental diversity in Podostemaceae (river-weeds). Aquatic Bot. 57: 29-70.
Sampson, F. B. 1969. Studies on the Monimiaceae, II: Floral morphology of Laurelia novae-zelandiae A. Cunn. (subfamily Atherospermoideae). New Zealand J. Bot. 7: 214-240.
Sastri, R. L. N. 1952. Studies in Lauraceae, I: Floral anatomy of Cinnamomum iners Reinw. and Cassytha filiformis Linn. J. Indian Bot. Soc. 31: 240-246.
-----. 1965. Studies in the Lauraceae, V: Comparative morphology of the flower. Ann. Bot. (London), n.s., 29: 39-44.
Sattler, R. 1962. Zur fruhen Infloreszenz und Blutenentwicklung der Primulales sensu lato mit besonderer Berucksichtigung der Stamen-Petalum-Entwicklung. Bot. Jahrb. Syst. 81: 358-396.
-----. 1973. Organogenesis of flowers: A photographic text-atlas. Univ. of Toronto Press, Toronto and Buffalo.
-----. 1988. Homeosis in plants. Amer. J. Bot. 75: 1606-1617.
-----. 1994. Homology, homeosis, and process morphology in plants. Pp. 424-475 in B. K. Hall (ed.), Homology: The hierarchical basis of comparative biology. Academic Press, San Diego.
Saunders, E. R. 1937-1939. Floral morphology; a new outlook with special reference to the interpretation of the gynoecium. Vols. 1-2. W. Heffer & Sons, Cambridge.
Saxena, N. P. 1976. Studies in the family Saxifragaceae, X: Floral morphology and systematic position of Parnassia. J. Indian Bot. Soc. 55: 282-288.
Schinz, H. 1934. Amaranthaceae. Pp. 77-85 in A. Engler & K. Prantl (eds.), Die naturlichen Pflanzenfamilien. Ed. 2. Vol. 16c. Engelmann, Leipzig.
Schmid, R. 1972. Floral bundle fusion and vascular conservatism. Taxon 21: 429-446.
Sersic, A. N. & A. A. Cocucci. 1999. An unusual kind of nectary in the oil flowers of Monttea: Its structure and function. Flora 194: 393-404.
Setoguchi, H., H. Ohba & H. Tobe. 1996. Floral morphology and phylogenetic analysis in Crossostylis (Rhizophoraceae). J. Pl. Res. 109: 7-19.
Sharma, M. R. 1954. Studies in the family Anacardiaceae, I: Vascular anatomy of the flower of Mangifera indica L. Phytomorphology 4: 201-208.
Sheela, Lal. 1994. A contribution to the floral anatomy of Cedreleae (Meliaceae). Feddes Repert. 105: 449-455.
----- & L. L. Narayana. 1994. Floral anatomy and systematic position of Flindersia R. Br. Feddes Repert. 105: 31-36.
Simpson, M. G. 1998. Haemodoraceae. Pp. 212-222 in K. Kubitzki (ed.), The families and genera of vascular plants. Vol. 4, Flowering plants: Monocotyledons: Alismatanac and Commelinanae (except Gramineae). Springer-Verlag, Berlin.
----- & P. J. Rudall. 1998. Tecophilaeaceae. Pp. 429-436 in K. Kubitzki (ed.), The families and genera of vascular plants. Vol. 3, Flowering plants: Monocotyledons: Lilianae (except Orchidaceae). Springer-Verlag, Berlin.
Singh, V. 1979. Early floral development in Digitalis purpurea. Phytomorphology 29: 239-245.
----- & D. K. Jain. 1978. Floral organogenesis in Adenocalymna alliaceum (Bignoniaceac). Beitr. Biol. Pflanzen 54: 207-215.
----- & A. Singh. 1985. Floral organogenesis in Cinnamomum camphora. Phytomorphology 35: 61-67.
Sleumer, H. 1935. Olacaceae. Pp. 5-32 in A. Engler & K. Prantl (eds.), Die naturlichen Pflanzenfamilien. Ed. 2. Vol. 16b. Engelmann, Leipzig.
Smets, E. 1986. Localisation and systematic importance of the floral nectaries in the Magnoliatae (Dicotyledons). Bull. Jard. Bot. Belg. 56: 51-76.
-----. 1988a. Florale nektarien van de Magnoliophytina: karakterizering en systematisehe betekenis. Ph.D. diss., Katholieke Universiteit Leuven.
-----. 1988b. La Presence des "nectaria persistentia" chez les Magnoliophytina (Angiospermes). Candollea 43: 709-716.
----- & E. Cresens. 1988. Types of floral nectaries and the concepts "character" and "characterstate": A reconsideration. Acta Bot. Neerl. 37: 121-128.
Speta, F. 1998. Hyacinthaceae. Pp.261-285 in K. Kubitzki (ed.), The families and genera of vascular plants. Vol. 3, Flowering plants: Monocotyledons: Lilianae (except Orchidaceae). Springer-Verlag, Berlin.
Steyn, E. M. A., P.J. Robbertse & A. E. van Wyk. 1987. Floral development in Greyia flanaganii with notes on inflorescence initiation and sympodial branching. S. African J. Bot. 53: 194-201.
Stoudt, H. N. 1941. The floral morphology of some of the Capparidaceae. Amer. J. Bot. 28: 664-675.
Subramanyam, K. & L. L. Narayana. 1976. A contribution to the floral anatomy and embryology in certain members of Primulaceae. J. Indian Bot. Soc. 55: 274-282.
Suessenguth, K. 1953a. Rhamnaceae. Pp.7-173 in A. Engler & K. Prantl (eds.), Die naturlichen Pflanzenfamilien. Ed. 2. Vol. 20d. Duncker & Humblot, Berlin.
-----. 1953b. Leeaceae. Pp. 372-390 in A. Engler & K. Prantl (eds.), Die naturlichen Pflanzenfamilien. Ed. 2. Vol. 20d. Duncker & Humblot, Berlin.
Takahashi, H. 1994. A comparative study of floral development in Trillium apetalon and T. kamtschaticum (Liliaceae). J. Pl. Res. 107: 237-243.
Takhtajan, A. 1980. Outline of the classification of flowering plants (Magnoliophyta). Bot. Rev. (Lancaster) 46: 225-359.
-----. 1991. Evolutionary trends in flowering plants. Columbia Univ. Press, New York.
-----. 1997. Diversity and classification of flowering plants. Columbia Univ. Press, New York.
Thei[beta]en, G., J. T. Kim & H. Saedler. 1996. Classification and phylogeny of the MADS-box multigene family suggests defined roles of MADS-box gene subfamilies in the morphological evolution of Eukaryotes. J. Molec. Evol. 43: 484-516.
Thien, L. B., H. Azuma & S. Kawano. 1999. New perspectives in the pollination biology of basal angiosperms. Abstr. XV Int. Bot. Congr., Yokohama, p. 90.
Tillson, A. H. & R. Bamford. 1938. The floral anatomy of the Aurantioideae. Amer. J. Bot. 25: 780-793.
Tschunko, A. H. & N. H. Nickerson. 1976. The androecium of Suriana maritima. Rhodora 78: 160-164.
Tucker, S. C. 1984. Origin of symmetry in flowers. Pp.351-395 in R. A. White & W. C. Dickison (eds.), Contemporary problems in plant anatomy. Academic Press, New York.
-----. 1988a. Heteromorphic flower development in Neptunia pubescens, a mimosoid legume. Amer. J. Bot. 75: 205-224.
-----. 1988b. Dioecy in Bauhinia resulting from organ suppression. Amer. J. Bot. 75: 1584-1597.
-----. 1988c. Loss versus suppression of floral organs. Pp. 69-82 in P. Leins, S.C. Tucker & P. K. Endress (eds.): Aspects of floral development. Cramer, Berlin.
----. 1996. Trends in evolution of floral ontogeny in Cassia sensu stricto, Senna, and Chamaecrista (Leguminosae: Caesalpinoideae: Cassieae: Cassiinae): A study in convergence. Amer. J. Bot. 83: 687-711.
-----. 1997. Floral evolution, development, and convergence: The hierarchical-significance hypothesis. Int. J. Pl. Sci. 158 (Suppl.): S143-S161.
-----. 1998. Floral ontogeny in legume genera Petalostylis, Labichea, and Dialium (Caesalpinioideae: Cassicae), a series in floral reduction. Amer. J. Bot. 85: 184-208.
Urban, I. 1892. Bluthen- und Fruchtbau der Loasaceen. Ber. Deutsch. Bot. Ges. 10: 259-265.
Vanvinckenroye, P., E. Cresens, L. P. Ronse Decraene & E. Smets. 1993. A comparative floral developmental study in Pisonia, Bougainvillea and Mirabilis (Nyctaginaceae) with special emphasis on the gynoecium and floral nectaries. Bull. Jard. Bot. Belg. 62: 69-96.
Vattimo, I. de. 1959. Notas sobre o Androceu de Aniba Aubl. (Lauraceae). Rodriguesia 21/22:339-345.
Venkata Rao, C. 1952. Floral anatomy of some Malvales and its bearing on the affinities of families included in the order. J. Indian Bot. Soc. 21: 171-203.
----- & Ramalakshmi, T. 1968. Floral anatomy of Euphorbiaceae, I: Some non-cyathium taxa. J. Indian Bot. Soc. 47: 278-298.
Vink, W. 1995. Revision of Magodendron (Sapotaceae) with observations on floral development and morphology. Blumea 40: 91-107.
Vogel, S. 1978. Evolutionary shifts from reward to deception in pollen flowers. Pp.89-96 in A. J. Richards (ed.), The pollination of flowers by insects. Academic Press, London.
Wagner, W. L. & E. M. Harris. 2000. A unique Hawaiian Schiedea (Caryophyllaceae: Alsinoidese) with only five fertile stamens. Amer. J. Bot. 87: 153-160.
Walker-Larsen, J. & L. D. Harder. 2000. The evolution of staminodes in angiosperms: Patterns of stamen reduction, loss, and functional re-invention. Amer. J. Bot. 87: 1367-1384.
Watson, L. & M. J. Dallwitz. 1992-. The families of flowering plants: Descriptions, illustrations, identification, and information retrieval. August 1999 version. ttp://biodiversity.uno.edu/delta/
Weberling, F. 1989. Morphology of flowers and inflorescences. Cambridge Univ. Press, Cambridge.
----- & H. Uhlarz. 1983. Zur Morphologie und Morphogenese der Blute von Cadaba juncea (Sparm.) Harv. (Capparidaceae). Beitr. Biol. Pflanzen 58: 267-281.
Weigel, D. & E. M. Meyerowitz. 1994. The ABCs of floral homeotic genes. Cell 78: 203-209.
Winkler H. 1931. Linaceae. Pp. 82-130 in A. Engler & K. Prantl (eds.), Die naturlichen Pflanzenfamilien. Ed. 2. Vol. 19a. Engelmann, Leipzig.
Woodson, R. E. & J. A. Moore. 1938. The vascular anatomy and comparative morphology of Apocynaceous flowers. Bull. Torrey Bot. Club 65: 135-165.
Woon, C. & H. Keng. 1979. Stamens of the Dipterocarpaceae. Gard. Bull., Singapore 32: 1-51.
Worsdell, W. C. 1903. The origin of the perianth of flowers with special reference to the Ranunculaceae. New Phytol. 2: 42-48.
Wydler, H. 1863. Ueber die Bluthe von Melianthus. Flora 1863: 145-151.
Yanofsky, M. F. 1995. Floral meristems to floral organs: Genes controlling early events in Arabidopsis flower development. Annual Rev. Pl. Physiol. Pl. Molec. Biol. 46: 167-188.
Zohary, M. & B. Baum. 1965. On the androecium of Tamarix flower and its evolutionary trends. Israel J. Bot. 14: 101-111.
Table I The occurrence of complete staminodial whorls in the Magnoliatae as derived from (ob)diplostemonous or dicyclic androccia (a) Family Tribe, genus, or species Alliaceae Allium, Trichlora, Gilliesia Anacardiaceae Pentaspadon Bombacaceae Chorisia Bonnetiaceae Ploiarium, Archytaea Brexiaceae Brexia, Ixerba Burseraceae Santiria Caesalpiniaceae Dimorphandra Caryophyllaceae Paronychia, Herniaria, Habrosia, Schiedea, etc. Celastraceae Lophopyxis (Lophopyxidaceae) Combretaceae Thiloa Commelinaceae Murdania, Anthericopsis (1), Palisota (2) Connaraceae Connarus Corynocarpaceae Corynocarpus Crassulaceae Sempervivum Diapensiaceae All Dioscoreaceae Dioscorea sects. Macrocarpaea Asterotricha Dipterocarpaceae Dipterocarpus oblongifolius Fabaceae Teramnus, Biserrula sp., Vicia sp., etc. Geraniaceae Erodium Hamamelidaceae (b) Hamamelis, Corylopsis, Loropetalum, Hyacinthaceae Albuca Hydnoraceae Prosopanche Hydrocharitaceae Lagarosiphon Johnsoniaceae Hodgsoniola Lepuropetalaceae Lepuropetalon Loasaceae Mentzelia sect. Bartonia Lomandraceae Sowerbaea Linaceae Linum, Reinwardtia, etc. Medusandraceae Medusandra Melastomataceae Poteranthera, Anplectrum Meliaceae Toona, Cedrela Mimosaceae Pentaclethra Moringaceae Moringa Myrsinaceae Myrsine Myrtaceae Darwinia, Chamaelaucium Ochnaceae Sauvagesia, Leitgebia Olacaceae Olax (c), Liriosma Onagraceae Clarkia sp. Oxalidaceae Averrhoa, Oxalis corniculata Parnassiaceae Parnassia Primulaceae Samolus Pterostemonaceae Pterostemon Ranunculaceae Clematis sect. Atragene Rutaceae Diosmeae, Flindersia Sapotaceae Magodendron, Mimusops, Synsepalum, Achras, etc. Simaroubaceae Alvaradoa Sterculiaceae Buettneria, Theobroma, Abroma, etc. Suriancaceae Suriana Themidaceae Brodiaea, Dichelostemma Theophrastaceae All Thymelaeaceae Gnidia, Craspedostoma Tiliaceae Brownlowia, Pentace Triuridaceae Seychellaria Xyridaceae Xyris Zygophyllaceae Tribulus (occasionally) Position of staminodial Family whorl Alliaceae Opposite inner tepals Anacardiaceae Antepetalous Bombacaceae Antesepalous Bonnetiaceae Antepetalous Brexiaceae Antepetalous Burseraceae Antepetalous Caesalpiniaceae Antesepalous Caryophyllaceae Antepetalous Celastraceae Antepetalous (Lophopyxidaceae) Combretaceae Antepetalous Commelinaceae Opposite inner (1) or outer tepals (2) Connaraceae Antepetalous Corynocarpaceae Antesepalous Crassulaceae Antepetalous Diapensiaceae Antepetalous Dioscoreaceae Opposite inner tepals Dipterocarpaceae Antesepalous Fabaceae Antepetalous Geraniaceae Antepetalous Hamamelidaceae (b) Antepetalous Hyacinthaceae Opposite outer tepals Hydnoraceae Alternitepalous Hydrocharitaceae Opposite outer tepals Johnsoniaceae Opposite outer tepals Lepuropetalaceae Antepetalous Loasaceae Antesepalous Lomandraceae Opposite outer tepals Linaceae Antepetalous Medusandraceae Antesepalous Melastomataceae Antepetalous Meliaceae Antepetalous Mimosaceae Antepetalous Moringaceae Antesepalous Myrsinaceae Antesepalous Myrtaceae Antepetalous Ochnaceae Antepetalous Olacaceae Antepetalous Onagraceae Antepetalous Oxalidaceae Antepetalous Parnassiaceae Antepetalous Primulaceae Antesepalous Pterostemonaceae Antepetalous Ranunculaceae Alternitepalous Rutaceae Antepetalous Sapotaceae Antesepalous Simaroubaceae Antepetalous Sterculiaceae Antesepalous, Suriancaceae Antepetalous Themidaceae Opposite outer tepals Theophrastaceae Antesepalous Thymelaeaceae Antesepalous Tiliaceae Antepetalous Triuridaceae Opposite inner tepals Xyridaceae Opposite outer tepals Zygophyllaceae Antesepalous Family Authority Alliaceae Rahn, 1998 Anacardiaceae Eichler, 1878 Bombacaceae Eichler, 1878 Bonnetiaceae Dickison & Weitzman, 1998 Brexiaceae Bensel & Palser, 1975a Burseraceae Engler, 1931d Caesalpiniaceae Eichler, 1878 Caryophyllaceae Ronse Decraene et al., 1998b; Wagner & Harris, 2000; Figs. 17-18 Celastraceae Cronquist, 1981 (Lophopyxidaceae) Combretaceae Eichler, 1878 Commelinaceae Faden, 1998 Connaraceae Saunders, 1939 Corynocarpaceae Narayana et al., 1986; Philipson, 1987; Figs. 7-8 Crassulaceae Berger, 1930 Diapensiaceae Palser, 1962 Dioscoreaceae Huber, 1998 Dipterocarpaceae Woon & Keng, 1979 Fabaceae Eichler, 1878; Rodriguez-Riano et al., 1999 Geraniaceae Kumar, 1976; Payer, 1857; Figs. 12-13 Hamamelidaceae (b) Endress, 1967; Mione & Bogle, 1990; Figs. 15-16 Hyacinthaceae Speta, 1998 Hydnoraceae Cocucci, 1975 Hydrocharitaceae Cook, 1998 Johnsoniaceae Clifford & Conran, 1998 Lepuropetalaceae Engler, 1930b Loasaceae Urban, 1892 Lomandraceae Conran, 1998 Linaceae Kumar, 1976; Narayana, 1964; Narayana & Rao 1971, 1976a, 1976b, 1977a, 1977b Medusandraceae Cronquist, 1981 Melastomataceae Eichler, 1878 Meliaceae Harms, 1960, Sheela Lal, 1994 Mimosaceae Eichler, 1878 Moringaceae Ronse Decraene et al., 1998a; Figs. 1-2 Myrsinaceae Caris, 1998 Myrtaceae Baillon, 1873; Ronse Decraene & Smets, 1991b Ochnaceae Eichler, 1878; Goebel, 1933 Olacaceae Agarwal, 1963; Baillon, 1892; Sleumer, 1935 Onagraceae Eichler, 1878 Oxalidaceae Al-Nowaihi & Khalifa, 1971; Eichler, 1878; Kumar, 1976; Moncur, 1988; Fig. 14 Parnassiaceae Bensel & Palser, 1975a; Engler, 1930b; Saxena, 1976 Primulaceae Caris, 1998; Ronse Decraene & Smets, 1995a; Sattler, 1962 Pterostemonaceae Engler, 1930b Ranunculaceae Eichler, 1878 Rutaceae Engler, 1931b; Sheela Lal & Narayana, 1994 Sapotaceae Ayensu, 1972; Eichler, 1878; Hartog, 1878, Moncur, 1988; Pennington, 1991; Vink, 1995 Simaroubaceae Engler, 1931c Sterculiaceae Venkata Rao, 1952; Heel, 1966; Fig. 38 Suriancaceae Tschunko & Nickerson, 1976 Themidaceae Rahn, 1998 Theophrastaceae Cronquist, 1981; Eichler, 1878 Thymelaeaceae Domke, 1934; Gilg, 1894 Tiliaceae Bocquillon, 1866; Eichler, 1878; Heel, 1966 Triuridaceae Maas-van de Kamer & Weustenfeld, 1998 Xyridaceae Kral, 1998 Zygophyllaceae Engler, 1931a (a)No information is available for Centroplacus (Pandaceae), Daphniphyllaceae. (b)The nectaries of Hamamelis represent an inner staminodial whorl, whereas in Corylopsis one or two whorls of supplementary staminodial nectaries are said to arise next to the staminodial whorl (Endress, 1967). We observed that an antepetalous staminodial whorl is initiated following antesepalous stamen initiation before gynoecium initiation. Two inner protuberances bearing stomata are initiated much later and represent--we believe--receptacular nectaries (Figs. 15-16). (c)Not exactly antepetalous in Olax due to reductions of stamen number. Table II Presence of staminodes within a whorl of fertile stamens Key: 1 = antesepalous whorl; 2 = antepetalous whorl (alternisepalous whorl); 3 = antesepalous and atenpetalous whorl; 4 = position of staminodes; 5 = number of staminodes; 6 = presence of zygomorphy in the flower Family Genus Anacardiaceae (Figs. 20-21) Mangifera, Anacardium Amaranthaceae Lithosperma Bignoniaceae Catalpa, Kigelia Caesalpiniaceae Tamarindus, Bauhinia, Cassia, Amherstia, etc. Cannaceae Canna Capparaceae (Figs. 4-6) Euadenia, Dactylaena Caryophyllaceae Microphyes (1), Ortegia (2) Chrysobalanaceae Hirtella, Parinarium Combretaceae Lumnitzera Commelinaceae Cochliostema (1) Palisota (2), Murdannia (3), Polyspatha etc. (4) Costaceae All Dichapetalaceae Tapura Gentianaceae Hoppea Geraniaceae Pelargonium Gesneriaceae Sanango Globulariaceae (incl. Selaginaceae) Globularia Haemodoraceae Schiekia (1), Pyrrorhiza (2) Heliconiaceae Heliconia Hydrocharitaceae Nechamandra, Maidenia, Vallisneria Krameriaceae Krameria Lamiaceae (a) Lavendula, Bystropogon, Salvia Linnacaceae Linnaea, Abelia Loasaceae Petalonyx crenatus Loganiaceae Usteria guineensis Malpighiaceae Stigmaphyllon, Gaudichaudia, Camarea Marantaceae All Mimosaceae Neptunia (Staminate flowers) Morinaceae Morina Musaceae Musa Myoporaceae Oftia, etc. Olacaceae Ptychopetalum Onagraceae Lopezia Orchidaceae All Pedaliaceae (incl. Martynia, Sesamen Martyniaceae, Trapellaceae) Podostemonaceae Podostemon Pontederiaceae Hydrothrix Proteaceae Conospermum, Protea (1), Synaphea (2), Placospermum (3) Rutaceae Cusparieae Sabiaceae Meliosma Solanaceae Salpiglossis, Schizanthus, Anthocercis Scrophulariaceae Bonnaya, Gratiola (1), Ixianthus (2), Digitalis (3), etc. Surianaceae Suriana Tecophyllaeaceae Tecophilaea (1), Zephyra (2) Trigoniaceae Trigonia (1), Lightia (2) Verbenaceae Duranta, Stachytarpheta, Lantana, etc. Vochysiaceae Salvertia, Vochysia, Qualea Zingiberaceae All Family 1 2 3 Anacardiaceae (Figs. 20-21) - - + Amaranthaceae + - - Bignoniaceae + - - Caesalpiniaceae + + - Cannaceae - - + Capparaceae (Figs. 4-6) + - - Caryophyllaceae + - - Chrysobalanaceae - - + Combretaceae - + - Commelinaceae + + + Costaceae - - + Dichapetalaceae + - - Gentianaceae + - - Geraniaceae - + - Gesneriaceae + - - Globulariaceae (incl. Selaginaceae) + - - Haemodoraceae + + - Heliconiaceae + - - Hydrocharitaceae - + + Krameriaceae + - - Lamiaceae (a) + - - Linnacaceae + - - Loasaceae 1 - - Loganiaceae + - - Malpighiaceae + - - Marantaceae - - + Mimosaceae - - + Morinaceae + - - Musaceae - + - Myoporaceae + - - Olacaceae + - - Onagraceae + - - Orchidaceae + + - Pedaliaceae (incl. + - - Martyniaceae, Trapellaceae) Podostemonaceae - - - Pontederiaceae + - - Proteaceae + - - Rutaceae + - + Sabiaceae - + - Solanaceae + - - Scrophulariaceae + - - Surianaceae - + - Tecophyllaeaceae + - + Trigoniaceae + - + Verbenaceae + - - Vochysiaceae - + - Zingiberaceae - - + Family 4 Anacardiaceae (Figs. 20-21) One antesepalous fertile Amaranthaceae ? Bignoniaceae Adaxial upper, or three adaxial Caesalpiniaceae S2, 5/P post, Spost+P Cannaceae Only Ppost fertile Capparaceae (Figs. 4-6) Slat & Spost Caryophyllaceae S1, S2 (1), S4, S5 (2) Chrysobalanaceae Spost Ppost Combretaceae Antepetalous (variable number of stamens) Commelinaceae Pant+Sant-lat (1), Plat-post (2), Spost (3), Plat-post+Spost (4) Costaceae S, Plat-post? Dichapetalaceae S2,S4 Gentianaceae ? Geraniaceae Pant Gesneriaceae Spost Globulariaceae (incl. Selaginaceae) Spost (vascular bundle) Haemodoraceae Slat-ant (1) or Plat-post (2) Heliconiaceae Sant Hydrocharitaceae Alternitepalous (and antetepalous if petal considered as staminode) Krameriaceae Sant (S1) Lamiaceae (a) Spost, S4 and S5 (Slat-post) Linnacaceae S2 Loasaceae Spost and Slat-post Loganiaceae Spost (fertile or absent) Malpighiaceae S1, 5 Marantaceae Plat-ant or 0, Slat-post (only Ppost fertile) Mimosaceae Abaxial stamens Morinaceae Slat-ant (S1,S3?) Musaceae Ppost Myoporaceae Spost Olacaceae Alternipetalous Onagraceae Sant Orchidaceae Spost, Plat-post Pedaliaceae (incl. Spost, Slat-ant Martyniaceae, Trapellaceae) Podostemonaceae Incomplete inner whorl Pontederiaceae ? Proteaceae Sab (1), Sad (2), Slat and Sab (3) Rutaceae S2, 3, 5(4), Ppost Sabiaceae Opp. Pant and Ppost Solanaceae Sant (opp. sepal 1), S1,2,5 Scrophulariaceae Slat-ant (1), Slat-post (2), Spost (3) Surianaceae Antepetalous, variable (combination of fertile and sterile stamens Tecophyllaeaceae Ppost+Slat-post (1), Slat-post (2) Trigoniaceae In adaxial part of flower (1), S3 (2), partly transformed as stami- nodes Verbenaceae Sab, Slat-ant Vochysiaceae Plat & ant Zingiberaceae Plat-ant, Slat-post Family 5 6 Anacardiaceae (Figs. 20-21) 9 + Amaranthaceae 3 - Bignoniaceae 1, 3 + Caesalpiniaceae 1-2, 6 + Cannaceae 1-4 (-5) + Capparaceae (Figs. 4-6) 3 + Caryophyllaceae 2 - Chrysobalanaceae ? + Combretaceae 1 + Commelinaceae 1, 2-3 + Costaceae 5, fused + into labellum Dichapetalaceae 2 + Gentianaceae 2-3 ? Geraniaceae 3 + Gesneriaceae 1 or absent + Globulariaceae (incl. Selaginaceae) 1 + Haemodoraceae 2 + Heliconiaceae 1 + Hydrocharitaceae 1-2? (if + petal considered as staminode) Krameriaceae 1 + Lamiaceae (a) 1-2 + Linnacaceae 1 + Loasaceae 3 - Loganiaceae 1 + Malpighiaceae 2-3 + Marantaceae 1-2(0)+2 + Mimosaceae 3 + Morinaceae 2 + Musaceae 1 + Myoporaceae 1 or absent + Olacaceae 3 - Onagraceae 1 + Orchidaceae 1, 2 + Pedaliaceae (incl. 1, 2 + Martyniaceae, Trapellaceae) Podostemonaceae 1 - Pontederiaceae 2 - Proteaceae 1 + Rutaceae 3-5 + Sabiaceae 3 - Solanaceae 1, 3 + Scrophulariaceae 1,2 + Surianaceae 1-5 - Tecophyllaeaceae 2-3 + Trigoniaceae 2-6 (1), + 1 (2) Verbenaceae 1, 2, 3 + Vochysiaceae 2, 4 + Zingiberaceae 2+2 + Family Authority Anacardiaceae (Figs. 20-21) Copeland, 1961; Sharma, 1954 Amaranthaceae Eliasson, 1988 Bignoniaceae Eichler, 1875; Neubauer, 1959 Caesalpiniaceae Eichler, 1878; Tucker, 1988b, 1988c, 1996, 1997, 1998 Cannaceae Kress, 1990 Capparaceae (Figs. 4-6) Karrer, 1991; Pax & Hoffmann, 1936; Ronse Decraene, unpubl. Caryophyllaceae Ronse Decraene et al., 1998b Chrysobalanaceae Eichler, 1878 Combretaceae Fukuoka et al., 1998 Commelinaceae Faden, 1998 Costaceae Larsen, 1998 Dichapetalaceae Baillon, 1874 Gentianaceae Kshetrapal, 1973 Geraniaceae Kumar, 1976; Sattler, 1973 Gesneriaceae Dickison, 1994 Globulariaceae (incl. Selaginaceae) Saunders, 1937 Haemodoraceae Simpson, 1998 Heliconiaceae Andersson, 1998 Hydrocharitaceae Cook, 1998; McConchie & Kadereit, 1987 Krameriaceae Cronquist, 1981 Lamiaceae (a) Eichler, 1875; Payer 1857 Linnacaceae Eichler, 1875 Loasaceae Urban, 1892 Loganiaceae Hakki, 1998 Malpighiaceae Eichler, 1878 Marantaceae Anderson, 1998; Eichler, 1878; Kress, 1990 Mimosaceae Tucker, 1988a Morinaceae Hofmann & Gottmann, 1990 Musaceae Kress, 1990 Myoporaceae Bocquillon, 1861a; Cronquist, 1981 Olacaceae Sleumer, 1935: fig. 1H Onagraceae Eyde & Morgan, 1973 Orchidaceae Endress, 1994 Pedaliaceae (incl. Baillon, 1861, 1862c; Cronquist, Martyniaceae, 1981 Trapellaceae) Podostemonaceae Baillon, 1886; Rutishauser, 1997 Pontederiaceae Cook, 1998 Proteaceae Haber, 1959, 1966; Douglas, 1997 (1,2); Douglas & Tucker, 1996 (3) Rutaceae Baillon, 1872; Eichler, 1878; Kallunki, 1998 Sabiaceae Beusckom, 1971 Solanaceae Eichler, 1875; Mair, 1977; Murray, 1945 Scrophulariaceae Chatin, 1873a; Eichler, 1875; Payer, 1857; Singh, 1979; Walker- Larsen & Harder, 2000 Surianaceae Gutzwiller, 1961; Tschunko & Nickerson, 1976 Tecophyllaeaceae Simpson & Rudall, 1998 Trigoniaceae Eichler, 1878; Kopka & Weberling, 1983; Smets, 1988a Verbenaceae Bocquillon, 1861b; Eichler, 1875; Payer, 1857; Sattler, 1973 Vochysiaceae Eichler, 1878; Kopka & Weberling, 1983; Litt, 1997 Zingiberaceae Larsen et al., 1998 (a)Payer (1857) described the initiation and later abortion of the posterior staminode in the Lamiaceae. This is rejected by Chatin (1873a), who states that no trace of a fifth stamen exists in the Labiates, unlike the Scrophulariaceae. Eichler (1875) mentions the presence of a rudimentary stamen in Bystropogan. Table III Various misinterpretations of receptacular emergences as staminodes (pseudostamiondes), carpellodes, or structures with unknown or debatable homologies Family Genus or species Achariaceae Ceratiosicyos, Acharia, Guthriea Aextoxicaceae Aextoxicon Amaranthaceae Achyranthes, etc. Apocybaceae (Figs. 48-49) Vinca, Allamanda, etc. Bataceac Batis Brassicaceae None Burseraceae Balsamodendron Capparaceae Cadaba, Capparis Celastraceae Celastrus Clusiaceae Hypericum, Harungana, etc. Crassulaceae None Ctenolophaceae Ctenolophon Dichapetalaceae (Figs. 26-28) Dichapetalum Dipentodonta- Dipentodon ceae Epacridaceae None Euphorbiaceae Croton, Cluytia, Mercurialis, etc. Fabaceae Phascoleae Flacourtiaceae Casearia, Azara, (Figs. 36- etc. 37, 39-40) Francoaceae Francoa Geraniaceae Geranium, Pelar- gonium Greyiaceae (Figs. 10-11) Greyia Humiriaccae Sacoglottis, Vantanea, Humiria, etc. Hydrophyllaceae Phacelia glaberrima Ixonanthaceae Ixonanthes, Ochthocosmus Lauraceae None (Fig. 41) Lepidobotrya- Lepidobotrys ceae Melianthaceae Melianthus Nyctaginaceae Mirabilis Ochnaceae Sauvagesia Olacaceae (b) Aptandra, Strom- bosia, Octok- nema, etc. Opiliaceae Opilia Oxalidaceae Oxalis Paeoniaceae Paeonia Passifloraceae (c) Passiflora, Adenia, Crossostema, etc. Peridiscaceae Peridiscus Podostemonaceae Polypleurum, etc. Polemoniaceae Cantua, Cobaea, Phlox (Fig. 47) Polygonaceae Fagopyrum, Polygonum (Figs. 42-43) Primulaceae Primula, Soldanella Proteacceae None Rhamnaceae Zizyphus, Helinus, etc. (Fig. 45) Rhizophoraceae Crossostylis Rubiaceae Mitchella Rutaceae None Salvadoraceae Salvadora, Dobera Sapindaceae Xanthoceras Sarcolaenaceae Xyloolaena Scyphostegiaceae Scyphostegia Simaroubaceac Picrasma, Brucea (1); Samadera (2); Picramnw, Eurycoma, etc. Stackhousiaceae Stackhousia, Tripterococcus Tamaricaceae Tamarix Thymelaeaceac None Turneracceae Turnera, etc. Vitaceac (including Leeaceae) Leca, Vitis (Fig. 44) Zygophyllaceae Balanites Protagonist Family evidence Achariaceae External Aextoxicaceae External Amaranthaceae External Apocybaceae (Figs. 48-49) Anatomical link with the gynoecium Bataceac External, no anatomical Brassicaceae External Burseraceae Anatomical Capparaceae Anatomical, external Celastraceae Anatomical Clusiaceae Anatomical, ontogenetic Crassulaceae External, anatomical Ctenolophaceae Anatomical Dichapetalaceae (Figs. 26-28) External Dipentodonta- External ceae Epacridaceae External Euphorbiaceae External Fabaceae Anatomical Flacourtiaceae External (Figs. 36- 37, 39-40) Francoaceae External Geraniaceae Anatomical Greyiaceae (Figs. 10-11) External Humiriaccae External Hydrophyllaceae External Ixonanthaceae Anatomical Lauraceae Anatomical, external (Fig. 41) Lepidobotrya- Anatomical ceae Melianthaceae External Nyctaginaceae External Ochnaceae External, ana- tomical Olacaceae (b) External Opiliaceae External Oxalidaceae External, anatomical Paeoniaceae Anatomical Passifloraceae (c) External, anatomical, developmental Peridiscaceae External Podostemonaceae External Polemoniaceae Anatomical (Fig. 47) Polygonaceae External (Figs. 42-43) Primulaceae Anatomical Proteacceae Anatomical, external Rhamnaceae Anatomical (Fig. 45) Rhizophoraceae External Rubiaceae Anatomical Rutaceae Anatomical Salvadoraceae External Sapindaceae External Sarcolaenaceae External Scyphostegiaceae External Simaroubaceac Anatomical, external Stackhousiaceae External Tamaricaceae Anatomical, External Thymelaeaceac Anatomical, External Turneracceae External Vitaceac (including Leeaceae) External (Fig. 44) Zygophyllaceae Anatomical Family Description Achariaceae "Glieder eines zweiten staminalkreises"; staminodes Aextoxicaceae Fleshy, bilobed glands alternating with the stamens Amaranthaceae Interstaminal appendages, staminodes Apocybaceae (Figs. 48-49) Nectaries derived from carpellodes Bataceac Whitish spatulate, slender-stalked, denticulate "appendages" or staminodia; "Petalen die durchaus als staminodien gelten konnen" Brassicaceae Transformed median stamens Burseraceae A disk staminal in nature Capparaceae Remnant of a former staminal supply, episepalous glands, or unilateral appendage Celastraceae A disk of staminal nature (evidence of small traces) Clusiaceae Transformed staminodes Crassulaceae Staminodes or carpellodes Ctenolophaceae Disk as modified stamens Dichapetalaceae (Figs. 26-28) Petals and disk lobes are staminodes Dipentodonta- Staminodial(?) ceae nectary glands Epacridaceae Staminodes repre- sented by a cluster of ante- petalous glands (1); or hair bun- dles on corolla tube (2) Euphorbiaceae "Ecailles ou glan- des de nature staminodiale avec loges d'an- theres steriles" (1); staminodial origin of inner or outer whorl (3 stamen whorls) (2) Fabaceae Disk as sterilized branches of the androecium Flacourtiaceae Antepetalous (Figs. 36- staminode-like 37, 39-40) disk append- ages: "stami- nodienartigen Diskus- forsatzen" Francoaceae Staminodes Geraniaceae Original triplo- stemony with transformation of outer stamen whorl Greyiaceae (Figs. 10-11) 10 small staminodes Humiriaccae "Des languettes etroites et subulees qui sont des filets depourvus d'antheres" (1); the disk can be interpreted as the inner sterilized part of the staminal tube (2); inner staminal whorls staminodial (3) Hydrophyllaceae "The nectary appears to be the morphologically homologous to an inner whorl of stamens" Ixonanthaceae Disk of staminal origin Lauraceae Splitting of organ in three parts (Fig. 41) and sterilization of lateral parts (1); result of association of three stamens and transforma- tion of lateral parts in glands (2); telomic structure (3) Lepidobotrya- Staminodial disk ceae Melianthaceae Posterior stamen as part of abax- ial nectary; "dass das hin- tere mediane Stamen fehlt, ... und zum Nectarhalter geworden ist" Nyctaginaceae Interstaminal appendages interpreted as aborted stamens Ochnaceae Staminodial disk, outer stami- nodes (corona) Olacaceae (b) Staminodes Opiliaceae Scales alternating with the sta- mens Oxalidaceae Tongue-like structures separated from the back of the alternipetalous stamens Paeoniaceae "The disk is largely androecial in nature" (1); "the disk represents a sterlised park of the androccium" (2) Passifloraceae (c) Staminodes as "Spitzchen... die sich als Staminod. deuten lassen (1); corona partly staminodial (limen) (2); 5 alternating ridges in Basananthe; two whorls of 5 protrusions in Crossostema interpreted as original triplostemony (3) Peridiscaceae Multilobed disk of staminodial origin Podostemonaceae The sepal-like staminodes arise at the base of the filament Polemoniaceae 5 Vestigial antepetalous (Fig. 47) stamen traces split up in numerous small branches and supply disk Polygonaceae "Les nectaires isloles sont (Figs. 42-43) des etamines reduites" Primulaceae A third whorl of vestigial traces (1); as petal marginal traces derived from the dorsal sepal trace and incorporated in the corolla (2) Proteacceae Intrastaminal scales (e) Rhamnaceae Disk from modified stamens (Fig. 45) (evidence of obdiplostemony) Rhizophoraceae Inner whorl of staminodes alternating with the stamens in 3 of the 10 species (f) Rubiaceae "The disk may represent an ex- pansion of the receptacle, or a second whorl of carpels" Rutaceae Sterilized branches of the androe- cium, branches from staminal straces (modi- fied stamens) Salvadoraceae Antepetalous "Zahnchen oder Diskus-Drusen" Sapindaceae 5 alternipetalous staminodes Sarcolaenaceae Disk with "cinq ecailles al- ternisepales" (1); nectary disk of probable sta- minodial origin (2) Scyphostegiaceae Three stubs in front of the in- ner perianth and opposite the sta- mens (related to petals) (1); or stamens (2) Simaroubaceac Variously receptacular derived from antesepalous stammens, or mixed antepetalous and carpellary traces; outer whorl of sterile carpels Stackhousiaceae "Die Drusen selbst entsprechen moglicherweise einem zweiten Staminalkreis" Tamaricaceae "The disc is staminal in nature being formed by the staminal bases and their stripules" (1); inner antipetalous staminal whorl (2) Thymelaeaceac Disk is the inner part of androecium Turneracceae 5 glands or protuberances between the stamens and petals Vitaceac (including Leeaceae) Staminodial tube, (Fig. 44) staminodial scales Zygophyllaceae Disk of staminodial nature (vascular supply derived from stamen traces) Protagonist Family authority Achariaceae Goldberg, 1986; Hooker & Masters 1871, cited in Harms, 1925b Aextoxicaceae Ronse Decraene, 1985; Smets, 1988a Amaranthaceae Joshi, 1932; Joshi & Venkata Rao, 1934; Saunders, 1939 Apocybaceae (Figs. 48-49) Woodson and Moore, 1938 Bataceac Eckardt, 1959: 416, Johnson, 1935: 23 Brassicaceae Alexander, 1952; Bernhardi, 1838, cited in Eichler, 1878; Goebel, 1933 Burseraceae Shukla, 1995, cited in Narayana, 1960 Capparaceae Stoudt, 1941; Weberling & Uhlarz, 1983 Celastraceae Berkeley, 1953 Clusiaceae Eicher, 1878; Payer, 1857; Ronse Decraene & Smets, 1991a Crassulaceae Eichler, 1878 Ctenolophaceae Narayana & Rao, 1971 Dichapetalaceae (Figs. 26-28) Breteler, 1973 Dipentodonta- Cronquist, 1981 ceae Epacridaceae Chatin, 1873b (1); Cronquist, 1981; Eichler, 1875 (2) Euphorbiaceae Baillon, 1862b (1); Eichler, 1878; Gandhi & Thomas, 1983; Goebel, 1933; Michaelis, 1924 (a) (2) Fabaceae Moore, 1936a, 1936b Flacourtiaceae Eichler, 1878; (Figs. 36- Gilg, 1925; 37, 39-40) Ronse De- craene, unpubl. Francoaceae Bensel & Palser, 1975b; Takhta- jan, 1997 Geraniaceae Dawson, 1936; Kumar, 1976 Greyiaceae (Figs. 10-11) Cronquist, 1981; Dahlgren & Van Wyk, 1988; Steyn et al. 1987 Humiriaccae Baillon, 1860a: 208 (1); Narayana & Rao, 1969, 1977b: 150, 1978 (2); Winkler, 1931 (3) Hydrophyllaceae Cosa, 1995, cited in Sersic & Cocucci, 1999: 402 Ixonanthaceae Narayana & Rao, 1966 Lauraceae Daumann, 1931 (1); Eames, 1961, (Fig. 41) etc. (2); Rohwer, 1994 (3) Lepidobotrya- Narayana & Rao, ceae 1974 Melianthaceae Eichler, 1878; Wydler, 1863: 149 Nyctaginaceae Buxbaum, 1961; Friedrich, 1956 Ochnaceae Cronquist, 1981; Gilg, 1925; Goebel, 1933; Saunders, 1939 Olacaceae (b) Sleumer, 1935 Opiliaceae None Oxalidaceae Kumar, 1976 Paeoniaceae Eames, 1953, 1961 (1); Goebel, 1933; Melville, 1984 (2) Passifloraceae (c) Harms, 1925a: 480 (1); puri, 1948, 1951 (2); De Wilde, 1974 (3) Peridiscaceae Cronquist, 1981; Hutchinson, 1959 Podostemonaceae Khosla & Mohan Ram, 1993:257 Polemoniaceae Dawson, 1936 (Fig. 47) Polygonaceae Emberger, 1936: 591 (Figs. 42-43) Primulaceae Dickson, 1936 (1); Saunders, 1939 (2); Subramamyam & Narayana, 1976 Proteacceae Ronse Decraene, 1985 Rhamnaceae Nair & Sarma, 1961; (Fig. 45) Prichard, 1955 Rhizophoraceae Setoguchi et al., 1996 Rubiaceae Blaser, 1954: 538 Rutaceae Tillson & Bam- ford, 1938 Salvadoraceae Mattfeld, 1960a: 232; Ronse De- craene, 1985 Sapindaceae Bonnier, 1879, cited in Smets, 1988a Sarcolaenaceae Baillon, 1884 (1); Cronquist, 1981 (2) Scyphostegiaceae Baehni, 1937, 1938 (1); Swamy, 1953, all cited in Heel, 1967 (2) Simaroubaceac Ejehler, 1878; Engler, 1931c; Nair & Joseph, 1957; Nair & Joshi, 1958 Stackhousiaceae Mattfeld, 1960b: 243 Tamaricaceae Murty, 1954: 235 (1) (1); Zohary & Baum, 1965 (2) Thymelaeaceac Domke, 1934; Heinig, 1951; Meisner, cited in Gilg, 1984 Turneracceae Cronquist, 1981: 409; Ronse Decraene, 1985 Vitaceac (including Leeaceae) Nair & Nambisan, (Fig. 44) 1957; Ridsdale, 1974; Suessenguth, 1953b Zygophyllaceae Nair & Jain, 1956 Detractive Family evidence Achariaceae Ontogenetic, but vascularized Aextoxicaceae External Amaranthaceae No evidence for extra whorl of stamens Apocybaceae (Figs. 48-49) Anatomical, derivation of disk traces well below ovary Bataceac Lacking; ontogeny needed Brassicaceae Ontogeny, external, anatomical Burseraceae None Capparaceae External, ontogenetic Celastraceae None Clusiaceae Ontogeny Crassulaceae External, ontogeny Ctenolophaceae External, anatomical Dichapetalaceae (Figs. 26-28) External Dipentodonta- None ceae Epacridaceae Ontogenetic Euphorbiaceae External, anatomi- cal Fabaceae External, ontoge- netic Flacourtiaceae Ontogenetic (Figs. 36- 37, 39-40) Francoaceae Ontogenetic, ana- tomical Geraniaceae Ontogenetic, ana- tomical Greyiaceae (Figs. 10-11) Ontogenetic, anatomical Humiriaccae Anatomical; ontogeny lacking Hydrophyllaceae None Ixonanthaceae Ontogeny lacking Lauraceae Ontogenetic (Fig. 41) Lepidobotrya- External ceae Melianthaceae Ontogenetic, ana- tomical Nyctaginaceae Ontogenetic Ochnaceae External Olacaceae (b) External Opiliaceae External Oxalidaceae Anatomical, developmental Paeoniaceae Developmental Passifloraceae (c) Developmental Peridiscaceae No ontogenetic evidence Podostemonaceae External Polemoniaceae None (Fig. 47) Polygonaceae External, anatomical (Figs. 42-43) Primulaceae Ontogenetic, anatomical (Coris) Proteacceae Developmental Rhamnaceae External, developmental (Fig. 45) Rhizophoraceae Developmental Rubiaceae No ontogenetic or anatomical evi- dence Rutaceae External Salvadoraceae External Sapindaceae External Sarcolaenaceae None Scyphostegiaceae Anatomical Simaroubaceac Anatomical Stackhousiaceae No ontogenetic or anatomical evidence Tamaricaceae External, developmental Thymelaeaceac External Turneracceae External Vitaceac (including Leeaceae) Developmental (Fig. 44) Zygophyllaceae Developmental Family Description Achariaceae Nectary-like bodies Aextoxicaceae 5 pairs of coalescent glands derived from 10 initial structures Amaranthaceae "Nebenblatter" (1); pseudostaminodia (interstaminal appendages, part of staminal tube) (2) Apocybaceae (Figs. 48-49) "Proliferation of receptacular tissue between the androecium and the gynoecium" Bataceac No conclusive evidence for staminodes or petals Brassicaceae Receptacular nectaries with variable development and vascular connections Burseraceae None Capparaceae Late appearance in ontogeny, outer morphology Celastraceae None Clusiaceae Receptacular emergences Crassulaceae Receptacular appendages, dorsal appendages of carpels Ctenolophaceae Extrastaminal, receptacular Dichapetalaceae (Figs. 26-28) Nectary glands, disk lobes Dipentodonta- None ceae Epacridaceae "Le disque n'est que le gonfle- ment de la par- tie du receptacle qui supporte l'ovaire" Euphorbiaceae Variably episepa- lous or epipeta- lous disk lobes with vascular supply from dif- ferent sources Fabaceae Very late initiation of disk, diplo- stemonous flowers Flacourtiaceae Scales are of same (Figs. 36- number as sta- 37, 39-40) mens and ap- pear much later Francoaceae No vascular con- nection, late ini- tiation, extrastaminal Geraniaceae Basically diplo- stemonous, nec- taries receptacular Greyiaceae (Figs. 10-11) No vascular connection, late initiation, extrastaminal Humiriaccae Disk variously supplied by bundless from the stamens, or without vascular connections Hydrophyllaceae None Ixonanthaceae None Lauraceae De novo emergences, staminal (Fig. 41) appendages Lepidobotrya- Receptacular disk ceae Melianthaceae No evidence for staminodial nature Nyctaginaceae Extensions of the staminal tube Ochnaceae Paracorolla as in Passifloraceae (no transition with stamens), next to true an- tepetalous sta- minodial whorl Olacaceae (b) Extrastaminal or intrastaminal disk Opiliaceae "Glandes volumin- euses" (1); intrastaminal nectary disk (2); "five broad disk lobes alternate with the sta- mens" (3) Oxalidaceae Extrastaminal appendage or gland, ligular appendage Paeoniaceae Receptacular disk Passifloraceae (c) Extrastaminal nectary receptacular in nature; also the 5 antesepalous nectarics of Adenia (no positional and time relation with the androccium); 5 alternating ridges dubiously staminodial Peridiscaceae None Podostemonaceae Petaloid, spathulatetepals Polemoniaceae None (Fig. 47) Polygonaceae Receptacular mamillae (Figs. 42-43) Primulaceae No external evidence Proteacceae Nectar scales as secondary organs Rhamnaceae Variable intrastaminal (Fig. 45) disk, intrastaminal thickening of the receptacle Rhizophoraceae Instrastaminal appendages Rubiaceae None Rutaceae Enlargement of the floral axis be- tween the sta- mens and the base of the ovary Salvadoraceae Considered as fused stipules of the stamens (1); receptacular na- ture (no vascu- lar supply) (2) Sapindaceae Disk with 5 long, fleshy append- ages Sarcolaenaceae Evidence is lack- ing to assign a staminodial na- ture to the disk Scyphostegiaceae Extrastaminal disk glands Simaroubaceac Supply of disk highly variable in the family (g) Stackhousiaceae None Tamaricaceae Stipular appendages ("Stipularzahnche- n"), staminal tube with teeth Thymelaeaceac "Receptaculareffig ura-tionen" (no evidence of transitions) Turneracceae Nectar secreted by broadened abaxial parts of filaments (nectarotheca) Vitaceac (including Leeaceae) Disk arising from (Fig. 44) the base of the gynoecium Zygophyllaceae Receptacular emergence Detractive Family authority Achariaceae Bernhard, 1999b Aextoxicaceae Baillon, 1870 Amaranthaceae Eichler, 1878 (1); Eliasson, 1988; Payer, 1857; Schinz, 1934 (2) Apocybaceae (Figs. 48-49) Rao & Arati Ganguli, 1963: 433 Bataceac None Brassicaceae Arber, 1931; Bowman & Smyth, 1998; Eichler, 1878; Norris, 1941, etc. Burseraceae None Capparaceae Pax & Hoffmann, 1936; Payer, 1857; Weberling & Uhlarz, 1983 Celastraceae None Clusiaceae Leins, 1964 Crassulaceae Eichler, 1878; Payer, 1857; Smets, 1988a Ctenolophaceae Link, 1992 Dichapetalaceae (Figs. 26-28) Cronquist, 1981; Leenhouts, 1956, cited in Breteler, 1973 Dipentodonta- None ceae Epacridaceae Payer, 1857: 578; Smets, 1988a Euphorbiaceae Baillon, 1874; Beille, 1901; Venkatao Rao & Ramalak- shmi, 1968 Fabaceae Smets, 1988a Flacourtiaceae Bernhard & En- (Figs. 36- dress, 1999 37, 39-40) Francoaceae Ronse Decraene & Smets, 1999 Geraniaceae Payer, 1857; Sat- tler, 1973; Smets, 1988a Greyiaceae (Figs. 10-11) Ronse Decraene & Smets, 1999 Humiriaccae Smets, 1988a, this study (based on descriptions in Narayana & Rao, 1977b) Hydrophyllaceae None Ixonanthaceae None Lauraceae Endress, 1980; Kasapligil, 1951; (Fig. 41) Payer, 1857; Singh & Singh, 1985 Lepidobotrya- Link, 1991 ceae Melianthaceae Payer, 1857; Ronse Decraene et al., 2001 Nyctaginaceae Rohweder & Huber, 1974; Vanvincken- roye et al., 1993 Ochnaceae Eichler, 1878; this study Olacaceae (b) Cronquist, 1981; Reed, 1955; Smets, 1988a Opiliaceae Baillon, 1892: 412 (1); Cronquist, 1981 (2); Reed, 1955: 41 (3) Oxalidaceae A1-Nowaihi & Khalifa, 1971; Eichler, 1878 Paeoniaceae Baillon,, 1862; Eichler, 1878; Hiepko, 1966 Passifloraceae (c) Bernhard, 1999a Peridiscaceae None Podostemonaceae Baillon, 1886; Engler, 1930a; Rutishauser, 1997 (d) Polemoniaceae None (Fig. 47) Polygonaceae Ronse Decraene & Akeroyd, (Figs. 42-43) 1988; Ronse Decraene & Smets, 1991c Primulaceae Payer, 1857; Ronse Decraene et al., 1995 Proteacceae Brough, 1933; Douglas & Tucker, 1996 Rhamnaceae Bennek, 1958; Payer, (Fig. 45) 1857; Suessenguth, 1953a Rhizophoraceae Juncosa, 1988; Juncosa & Tomlinson, 1987 Rubiaceae None Rutaceae Penzig, 1887, cited in Tillson & Bamford, 1938 Salvadoraceae Gluck, 1919, cited in Mattfeld, 1960a (1); Kshetrapal, 1970 (2) Sapindaceae Radlkofer, 1896 Sarcolaenaceae Smets, 1988a Scyphostegiaceae Van Heel, 1967 Simaroubaceac Smets, 1988a Stackhousiaceae None Tamaricaceae Eiehler, 1878; Payer, 1857 Thymelaeaceac Gilg, 1894 Turneracceae Smets, 1988a Vitaceac (including Leeaceae) Gerrath Ct al., 1990 (Fig. 44) Zygophyllaceae Ronse Decraene, unpubl. obs. (a)Michaelis (1924) gives several arguments for a staminodial nature of the nectaries, including the position of the glands and external shape. The late initiation and variable position of the nectaries (e.g., Beille, 1901) are arguments against this. (b)True staminodes occur occasionally in the Olacaceae (e.g., Agarwal, 1963; Baitlon, 1892; Sleumer, 1935: Olax, Liriosma). In some genera of tribes Anacoloseae and Aptandreae there are disklike appendages with a possible homology to stamens (Sleumer, 1935:7: "Es konnte sich bei diesn lappigen Drusen oder dem gekerbten Drusenring auch um umgebildete aussere oder inner Stam. handeln"). (c)In Crossostemma and Basananthe a second series of small protrusions alternates with the stamens (Bernhard, 1999a). (d)Podostemoideae posses an additional envelope, homologous to prophylls or leaves (spathella), whereas Tristichoideac have a cuplike cover (Cupule) (Rutishauser, 1997.) (e)Haber (1959, 1961, 1966) interprets the scales as a petal whorl, because their vasculature is connected with the sepal lateral bundles, their tetramerous plan, and the alternisepalous position. Joshi (1936) interprets the disk scales of Stellera (Thymclacaceae) in the same way. (f)The "staminodes" in Crossostylis alternate with the stamens and are the same in number. They have no vascular connection and no obvious function. They are situated between the stamens and a nectary disk that is present in all species (Setoguchi et al., 1996). Juncosa (1988: 86) states that the intrastaminal appendages are initiated long after the stamens have developed and are therefore "clearl not staminodes." The common diplostemonous androecium in the family (symplesiomorphy) supports the latter interpretation. (g)The disk of Brucea receives its vascular supply from the three sources: branches from the antesepalous staminal traces, a first whorl of disk traces is found alternating with the stamens, followed by a second whorl opposite the stamens, but as a part of the carpellary tissue (Nair & Joshi, 1958). Engler (1931c) described Alvaradoa as having five sterile antepetalous stamens. In other cases (e.g., Eurycoma) the description of antepetalous ("wahrscheinlich Staminodien": 1.c.: 381).
|Printer friendly Cite/link Email Feedback|
|Author:||Decraene, L.P. Ronse; Smets, E.F.|
|Publication:||The Botanical Review|
|Date:||Jul 1, 2001|
|Previous Article:||Remains of palms (palmae) at archaeological sites in the new world: A review.|
|Next Article:||Screening methods for the evaluation of crop allelopathic potential.|