Skip to main content
Annals of Botany logoLink to Annals of Botany
. 2010 May 11;106(1):29–36. doi: 10.1093/aob/mcq084

False paracarpy in Seemannaralia (Araliaceae): from bilocular ovary to unilocular fruit

Alexei A Oskolski 1,*, Dmitry D Sokoloff 2, Ben-Erik Van Wyk 3
PMCID: PMC2889795  PMID: 20462851

Abstract

Background and Aims

Seemannaralia appears to be fundamentally different from all other Araliaceae in the presence of a well-developed symplicate zone in its gynoecium, as well as in the ovule insertion in the symplicate zone (rather than in the cross-zone). The present investigation re-examined the floral structure of Seemannaralia with emphasis on the morphology and evolution of its gynoecium.

Methods

Flowers and fruits of Seemannaralia gerrardii at various developmental stages were examined using light microscopy and scanning electron microscopy.

Key Results

Ovaries in the flowers of Seemannaralia are bilocular. Each ovary locule corresponds to a carpel whose ascidiate part is distinctly longer than the plicate part. Each carpel contains one fertile ovule attached to the cross-zone, and one sterile ovule as well. The fruit is unilocular: its central cavity is occupied by a single large seed. In the course of fruit development, the growth of one ovule stops while another ovule develops into the mature seed. When this ovule outgrows the available space in the locule, the septum is ruptured, forming a united cavity of two carpels.

Conclusions

Despite literature data, the synascidiate zone is well developed in the gynoecium of Seemannaralia, and the ovules are attached to the cross-zone. Its preanthetic and anthetic gynoecium has nearly the same structure as gynoecia of most other Araliaceae. The Seemannaralia fruit resembles the paracarpous gynoecium but its ground plan is very different because the central cavity is formed by mechanical rupture of the septum. The term ‘pseudoparacarpy’ (‘false paracarpy’) is proposed to describe this condition, which has not been reported to date for indehiscent fruits in any taxa other than Seemannaralia. In this genus, the pseudoparacarpy has probably resulted from a decrease in seed number in the course of the transition from zoochory to anemochory.

Keywords: Anatomy, Apiales, Araliaceae, development, flower, fruit, gynoecium, paracarpy, Seemannaralia gerrardii, syncarpy

INTRODUCTION

Although the phylogeny and morphology of the Apiales have been investigated by many botanists, and significant progress has been made in this area of research in recent years (e.g. Plunkett et al., 2004a, b; Nicolas and Plunkett, 2009), some important aspects of floral evolution in the order remain enigmatic. In particular, the largest and obviously closely related families of Apiales, namely Araliaceae, Myodocarpaceae and Apiaceae, are similar to each other in the structure of their gynoecia whereas the Pittosporaceae is very distinct from them. The Araliaceae, Myodocarpaceae and Apiaceae possess mainly bi- or multilocular ovaries in a gynoecium with a long synascidiate zone and a very short symplicate zone, with the ovules inserted at the transition between the two zones, i.e. in a cross-zone (also referred to as a ‘Querzone’; e.g. Magin, 1977; Sokoloff et al., 2007). The few known cases of unilocular ovaries in members of Araliaceae and Apiaceae appear to result from carpel reduction, i.e. these gynoecia are pseudomonomerous (Baumann-Bodenheim, 1955; Magin, 1977, 1980). Unlike these three families, the symplicate zone in gynoecia of the Pittosporaceae is much longer than the synascidiate zone, and the ovules are arranged along this zone. As a result, members of this family have unilocular ovaries with a single cavity formed between adjacent carpels (e.g. Narayana and Radhakrishnaiah, 1982; Erbar and Leins, 1996).

As these two types of gynoecia are distinctive at the family level, a trend of evolutionary transition between them within Apiales could be of phylogenetic interest. Although no morphological constraints for such transitions have been suggested (Weberling, 1989; Erbar and Leins, 1996, 2004; Leins and Erbar, 2004), it is highly questionable whether the condition found in Pittosporaceae should be regarded as ancestral within the Apiales. As Stevens (2009) suggests, ‘floral morphology of Pittosporaceae is quite probably derived, rather than representing the plesiomorphic condition from which the distinctive flowers of [Apiaceae + Araliaceae] evolved’. Understanding evolutionary transformations of gynoecia in Apiales is in part complicated by the fact that precise relationships between Pittosporaceae and other monophyletic groups within Apiales have not yet been fully resolved. Some molecular data suggest a basal position for Pittosporaceae in relation to the Araliaceae, Myodocarpaceae and Apiacae (Andersson et al., 2006; Nicolas and Plunkett, 2009) whereas other phylogenetic studies have placed Pittosporaceae as embedded within the Araliaceae–Apiaceae alliance (Plunkett et al., 1996; Chandler and Plunkett, 2004).

In this context, the genus Seemannaralia, a member of Araliaceae that appears to be similar to the Pittosporaceae in the ground plan of its gynoecium, could be important for clarifying the evolutionary pathways of floral features in Apiales. This monospecific South African genus reportedly possesses a unilocular gynoecium that should be regarded as having a well-developed symplicate zone with apically attached ovules (Burtt and Dickison, 1975). As the embedded position of Seemannaralia within Araliaceae is well confirmed by molecular analyses, and direct relationships to Pittosporaceae are very unlikely (Wen et al., 2001; Plunkett et al., 2004a; Lowry et al., 2004), the structure of its gynoecium should certainly be regarded as derived from the condition which is typical for the majority of Apiales taxa. Therefore, Seemannaralia may be highly relevant in showing parallelism in the transformation of the gynoecia in different groups of Apiales. Because Seemannaralia remains a relatively poorly known genus and because of its relevance to general questions about gynoecium evolution in Apiales, a detailed study of flowers and fruits of this plant was conducted.

Seemannaralia gerrardii is a small to medium-sized deciduous tree with palmately lobed leaves. The species is restricted to a few localities in the eastern region of South Africa, in the provinces of KwaZulu-Natal and Mpumalanga (Burtt and Dickison, 1975). Originally, the species was described by Seemann (1866) as Cussonia gerrardii Seem. but was segregated by Viguier (1906) into the monotypic genus Seemannaralia R.Vig. Cussonia Thunb. is another African genus of Araliaceae. Although Seemannaralia obviously differs from Cussonia by its imbricate petal aestivation and dry laterally compressed fruit as opposed to the valvate petal aestivation and fleshy globose fruits in the latter genus, a close relationship between these taxa was recently supported on the basis of molecular data (Plunkett et al., 2004b). The placement of Seemannaralia within various other groups of Araliaceae as suggested by Viguier (1906) and Hutchinson (1967) is not supported by molecular data (Wen et al., 2001; Plunkett et al., 2004b; Lowry et al., 2004) nor by detailed morphological investigations (Burtt and Dickison, 1975).

The isolated taxonomic position of Seemannaralia was emphasized by Burtt and Dickison (1975), who found that the flowers have a unilocular ovary formed by two fused carpels, i.e. a paracarpous gynoecium. The description of Burtt and Dickison (1975) implies that the gynoecium of Seemannaralia either lacks a synascidiate zone completely or that the synascidiate zone is very short. Importantly, because the ovule attachment is apical in the ovary of Seemannaralia, they must be interpreted as being attached to the symplicate zone. As stated above, this remarkable floral feature is not reported elsewhere within Araliaceae, the majority of whose members have di- to pentamerous (sometimes polymerous) syncarpous gynoecia with separate ovarian locules (Philipson, 1970; Eyde and Tseng, 1971; Nuraliev et al., 2010). In particular, all Cussonia species share dimerous gynoecia. One-seeded fruits in Cussonia natalensis (Strey, 1981) are formed from bilocular ovaries (Sonder, 1862; Reyneke, 1981), and their septa between fertile and sterile locules of mature fruits remain intact (A. A. Oskolski, pers. obs.).

The occurrence of unilocular ovaries is reported in a few other members of Araliaceae, such as in Arthrophyllum and occasionally in some species of Osmoxylon, such as Osmoxylon micranthum, and Cuphocarpus. Unilocular ovaries of Arthrophyllum and Osmoxylon (Baumann-Bodenheim, 1955; Philipson, 1970, 1979) are formed by single carpels and it is very likely that the ovaries have a well-developed ascidiate zone. Therefore, these taxa may not be fundamentally different from other Araliaceae in the relative length of ascidiate and plicate zones. The phylogenetic placement of taxa with monocarpellate unilocular ovaries (Wen, 2001; Plunkett et al., 2004b) strongly suggests that the unilocular ovary is derived from bi- or multilocular ovaries in Araliaceae through a reduction in the number of carpels.

As outlined above, Seemannaralia appears to be fundamentally different from all other investigated Araliaceae in the presence of a well-developed symplicate zone, with the synascidiate zone being either absent or very short, as well as in the ovule insertion in the symplicate zone (rather than in the cross-zone). The present study provides a re-examination of the floral structure of Seemannaralia with emphasis on the morphology and evolution of its gynoecium. This work is part of a general study of meristic trends in floral morphology throughout the Araliaceae (Sokoloff et al., 2007; Nuraliev et al., 2009, 2010).

MATERIAL AND METHODS

Inflorescences of Seemannaralia gerrardii (Seem.) R.Vig. were collected by the first and the third authors and B. J. de Villiers in April, 2007 in the KwaZulu-Natal Province of South Africa (collector's no. B.J. de Villiers 97 and 107; voucher specimens deposited in JRAU). Flowers and fruits at various developmental stages were fixed in FAA (formalin 5 %, acetic acid 5 %, ethyl alcohol 90 %) and stored in 70 % ethanol. For light microscopy, material was sectioned to 15–25 µm thickness using standard methods of paraffin embedding and serial sectioning. Sections were stained with safranin and alcian blue and mounted in euparal. For scanning electron microscopy, air-dried flowers and fruits were sputter-coated with gold using a JFC-1100 ion-coater, and were observed using a JSM-6380LA scanning electron microscope (JEOL, Tokyo, Japan). Premature flowers and fruits were photographed using a SteREO Lumar V12 (Carl Zeiss, Jena, Germany) photomacroscope fitted with a AxioCam MRc5 digital camera (Carl Zeiss).

RESULTS

Flowers of S. gerrardii are aggregated into 20- to 40-flowered open umbellules that are arranged into a terminal frondo-bracteose inflorescence (Fig. 1). The main axis (10–30 cm long) bears 5–14 second-order axes 10–25 cm long (Figs 1A and 2A) subtended by foliage leaves with trilobate to entire (sometimes five-lobate) blades 4–12 cm long on petioles 3–9 cm long. These subtending leaves fall by early anthesis; they are distinct from the foliage leaves situated below the inflorescence; the latter have larger five-lobate blades up to 25 cm long. The internodes in the upper portion of the main inflorescence axis (occasionally also in its middle portion) are shortened. The second-order axes (Figs 1D and 2C) bear terminal umbellules and 3–18 third-order axes (5–12 cm long) subtended by minute (4–8 mm long) triangular bracts, or rarely by small foliage leaves with entire lamina 1–3 cm long. Each third-order axis bears two prophylls inserted at different levels, and a terminal umbellule. Prophylls are caducous and fall off during anthesis.

Fig. 1.

Fig. 1.

Seemannaralia gerrardii: inflorescences at anthesis. Images taken on 4 April, 2007 at Roselands Farm near Richmond, KwaZulu-Natal, South Africa. (A) General view of a shoot with foliage leaves and a terminal inflorescence. (B, C) Umbellules with anthetic, preanthetic (arrowheads) and postanthetic (arrows) flowers. (D) Distal part of a second-order inflorescence axis with a terminal and several lateral (third-order) umbellules; the terminal umbellule bears immature fruits, while the lateral umbellules bear anthetic and just postanthetic flowers.

Fig. 2.

Fig. 2.

Seemannaralia gerrardii: inflorescences in the dry season, with mature fruits. Images taken on 21 September, 2009, 50 km east of Vryheid, KwaZulu-Natal, South Africa. (A) General view of an inflorescence. (B) General view of an entire plant with numerous fruiting inflorescences. (C) Second-order inflorescence axis with mature fruits (arrowhead = fruits of the terminal umbellule).

Each flower is situated in the axil of a narrowly lanceolate bract 0·5–1 mm long. The flowers are hermaphroditic, rather small (flower size at anthesis is approx. 3–5 mm), tetracyclic and have an inferior ovary. Elements of the adjacent whorls alternate with each other. The flowers have five small but conspicuous sepals approx. 0·5 mm long, five free, flat petals (Fig. 1B, C), five stamens and two carpels in a median position. Petal aestivation is imbricate or cochleate.

Ovaries in the flowers of S. gerrardii are bilocular (Figs 3 and 4). Each ovary locule corresponds to a carpel whose ascidiate part is distinctly longer than the plicate part. The two laterally compressed carpels are united by their ascidiate parts. The septum between the two locules extends for the full length of the ovary. Each carpel contains one fertile ovule attached to the cross-zone (Fig. 4D–F) and one sterile ovule inserted just above the fertile ovule (Fig. 5A). Because the fertile ovule is pendent, it appears in cross-sections throughout the upper part of the synascidiate zone. The funicle of the fertile ovule is provided with a hairy obturator (Fig. 6A–C). Sterile ovules are much smaller than fertile ovules. The ventral bundles of adjacent carpels are fused in a heterocarpellary position (Fig. 4C–E). Just below the level of ovule insertion, two heterocarpellary ventral bundles unite with each other (Fig. 4F) and then immediately divide in a perpendicular plane, giving rise to two bundles each supplying a fertile ovule. Sterile ovules do not have any vascular supply.

Fig. 3.

Fig. 3.

Seemannaralia gerrardii: schematic drawing of a cross-section of just preanthetic flower in the upper part of synascidiate zone of the gynoecium. Grey areas, vascular bundles; ol, ovarian locules; fo, fertile ovules. Scale bar = 500 µm.

Fig. 4.

Fig. 4.

Seemannaralia gerrardii: ascending series of cross-sections of ovary in just preanthetic flower (light microscopy), continued in Fig. 5. (A) Section in lower part of synascidiate zone. (B) Lower part of synascidiate zone, just above (A). (C) Central part of the same section as in (B); note intact septum. (D, E) Central parts of sections in the upper part of synascidiate zone; note intact septum. (D) The same section as illustrated in Fig. 3; heterocarpellary ventral bundles are well separated. (E) Above (D); heterocarpellary ventral bundles are close to each other. (F) Above (E), nearly at the level of fertile ovule attachment (cross-zone); ventral bundles are united. (G) Symplicate zone: above, fertile ovules; below, sterile ovules. Abbreviations: fo, fertile ovule; ob, ovule bundle; ol, ovarian locule; st, septum; vb, ventral bundles; arrowheads, line of postgenital fusion in plicate zone. Scale bars = 100 µm.

Fig. 5.

Fig. 5.

Seemannaralia gerrardii: ascending series of cross-sections of ovary in just preanthetic flower (light microscopy), continued from Fig. 4. (A) Symplicate zone, at the level of sterile ovules. (B) Symplicate zone, above sterile ovules. (C) Plicate zone; carpels free. so, sterile ovule; pttt, pollen tube transmitting tissue; vs, ventral slit; arrowheads, line of postgenital fusion in plicate zone. Scale bars = 100 µm.

Fig. 6.

Fig. 6.

Seemannaralia gerrardii: dissected immature fruits (scanning electron microscopy). (A, B) Upper part of ovary with two fertile ovules visible. (C) Detail of (B) showing obturator. (D, E) Lower part of ovary; inner surface of locules in the area of ruptured septum. (F) Lower part of ovary; inner surface of a locule beside septum. fo, fertile ovule; arrowheads, ruptured septum. Scale bars (A, B, D) = 200 µm, (C) = 50 µm, (E, F) = 100 µm.

The very short symplicate zone (Figs 4G and 5A, B) forms a roof of the ovary and a massive intrastaminal nectary. Externally, the nectary is covered by relatively long unicellular papillae. The surface of each papilla is longitudinally striate. Carpel tips are extended into free plicate stylodia (Fig. 5C) bearing stigmas. Pollen tube transmitting tracts extend from each stylodium downward to the ovules. Just above the cross-zone, the transmitting tracts of adjacent carpels are united into a short internal compitum that divides beneath into separate transmitting tracts, which reach the ovules of each carpel.

Unlike the flower, the fruit of Seemannaralia is unilocular (Figs 6 and 7): its central cavity is occupied mostly by a single large seed but a remnant of a second fertile ovule may also be present. Initially, both fertile ovules of postanthetic flowers grow at an equal rate until they reach 2–3 mm in diameter (Figs 6A, B and 7A). Growth of one ovule then stops whereas the other ovule develops into the mature seed by expanding and elongating, reaching 6–8 mm in length and 4–6 mm in width (Fig. 7B, C). The ovule that develops into a seed can be localized in the abaxial or the adaxial carpel; both variants occur in nearly equal numbers of fruits. When the ovule outgrows the available space in the locule, the septum is ruptured, forming a united cavity of the two carpels. Vestiges of the ruptured septum can be seen as fine sutures on the cavity wall; their scar surface can be distinguished by scanning (Fig. 6D, E) and light microscopy (Fig. 7C). Mature fruits are indehiscent, laterally compressed, elliptical to rounded in outline (Fig. 7D), 9–12 mm in length and 7–10 mm in width. Their abscision is extended over the dry season (April–October) with the fruits remaining on the branches after the leaves have fallen (Fig. 2).

Fig. 7.

Fig. 7.

Seemannaralia gerrardii: fruits (light microscopy). (A) Dissected immature fruit with two ovules of the same size. (B, C) Dissected nearly mature fruits with one ovule developing into seed; arrowheads, ruptured septum. (D) Nearly mature fruit: general view. Scale bar common to all figures = 1 mm.

DISCUSSION

The present observations show that ovaries in the flowers of S. gerrardii are bilocular. The septum between the two locules extends for the full length of the ovary and not only at its apex ‘near the level of ovular insertion’, as Burtt and Dickison (1975, p. 457) noted. As far as we are aware ascertain, this is the first record of a phenomenon that could be termed ‘pseudoparacarpy’, not only in the Araliales but in the Angiospermae as a whole. The fruit superficially appears to be paracarpous (as is found in Pittosporaceae) but the single locule is formed by mechanical rupture of the septum by the developing seed.

The present data show beyond doubt that the flower of Seemannaralia has a bicarpellate syncarpous gynoecium; its ovary contains two locules separated from each other by a septum. Despite Burtt and Dickison's (1975) records, the synascidiate zone is well developed in this gynoecium and the ovules are attached to the cross-zone rather than to the reportedly long symplicate zone. In the present description, the preanthetic and anthetic gynoecium of Seemannaralia has nearly the same structure as gynoecia of most other genera of Araliaceae (including Cussonia), Myodocarpaceae and Apiaceae (e.g. Philipson, 1970; Magin, 1977; Reyneke, 1981; Lowry, 1986a, b). Therefore, Seemannaralia cannot throw any light on the evolution of the unilocular gynoecum with a well-developed symplicate zone, which is typical for the Pittosporaceae.

The discrepancy between Burtt and Dickison's (1975) results and ours is probably due to the different kind of materials that were used. Burtt and Dickison studied dry flowers taken from herbarium collections whereas we used samples fixed in FAA. The ovary septum in Seemannaralia (at least that part situated between the heterocarpellary ventral bundles) consists of thin-walled parenchyma that is easily damaged. Like Burtt and Dickison, we observed ‘unilocular ovaries’ (i.e. with split septa) in some transverse sections, but careful observations of serial sections revealed that these cases are localized and represent secondary ruptures of the septa.

The disappearance of the septum and the union of the locules into a common cavity take place in the course of fruit development. The fruit of Seemannaralia resembles the paracarpous gynoecium but its ground plan is very different because the central cavity is formed by mechanical rupture of the septum. The term ‘pseudoparacarpy’ is proposed here to describe this unusual case of united carpellary cavities.

Among other angiosperms, the destruction of the septa can serve as a mechanism to aid the dehiscence of the fruit in many taxa, for example in Impatiens (Caris et al., 2006). Besides, the septa are commonly partly or completely dissolved before or during anthesis in some Caryophyllales (Takhtajan, 1942, 2009; Rohweder, 1967; Bittrich, 1993), a phenomenon used to introduce the term ‘lysicarpous gynoecium’ (Takhtajan, 1942). Nevertheless, the mechanical rupture of the septum in the course of the development of indehiscent fruit has not been reported to date in any taxa other than Seemannaralia.

As Burtt and Dickison (1975, p. 460) suggested, the fruit of Seemannaralia is functionally equivalent to the typical anemochorous mericarp of Apioideae. In general, seed number in wind-dispersed diaspores tends to be reduced to one, which is the most common condition in anemochorous taxa (Ridley, 1930). Certain adaptive advantages of one-seeded anemochorous diaspores over di- or polyspermous ones have been experimentally demonstrated for Lonchocarpus pentaphyllus (Fabaceae), a tropical tree which has indehiscent wind-dispersed fruits containing 0–4 seeds (Augspurger and Hogan, 1983). In this species, the decrease in seed number per fruit is correlated with increases in both seed weight (i.e. improved nutrient storage for the embrio) and dispersal distance. The pseudoparacarpous structure of the fruit in Seemannaralia has probably resulted from such a decrease in seed number in the course of a transition from zoochory [as in Cussonia, where the fleshy fruits are dispersed by birds (Reyneke, 1981; our pers. obs.)] to anemochory. This suggestion may be tested by detailed studies of fruit dispersion in the two genera.

ACKNOWLEDGEMENTS

We thank Bernard de Villiers for his assistance during field trips in South Africa, and Ludmila Kartseva for her help with electron microscopy facilities in the Komarov Botanical Institute. This research was supported by the Russian Foundation of Basic Research (projects 09-04-00618a and 09-04-01155a) and President of Russia grant No. MD-2644·2009·4.

LITERATURE CITED

  1. Andersson L, Kocsis M, Eriksson R. Relationships of the genus Azorella (Apiaceae) and other hydrocotyloids inferred from sequence variation in three plastid markers. Taxon. 2006;55:270–280. [Google Scholar]
  2. Augspurger CK, Hogan KP. Wind dispersal of fruits with variable seed number in a tropical tree (Lonchocarpus pentaphyllus: Leguminosae) American Journal of Botany. 1983;70:1031–1037. [Google Scholar]
  3. Baumann-Bodenheim MG. Ableitung und Bau bicarpellat-monospermer und pseudomonocarpellater Araliaceen- und Umbelliferen-Früchte. Berichte der Schweizerischen botanischen Gesellschaft. 1955;65:481–510. [Google Scholar]
  4. Bittrich V. Caryophyllaceae. In: Kubitzki K, Rohwer J, Bittrich V, editors. The families and genera of vascular plants. Vol. 2. Berlin: Springer-Verlag; 1993. pp. 206–236. [Google Scholar]
  5. Burtt BL, Dickison WC. The morphology and relationships of Seemannaralia (Araliaceae) Notes from the Royal Botanic Garden, Edinburgh. 1975;33:449–464. [Google Scholar]
  6. Caris PL, Geuten KP, Janssens SB, Smets EF. Floral development in three species of Impatiens (Balsaminaceae) American Journal of Botany. 2006;93:1–14. [Google Scholar]
  7. Chandler GT, Plunkett GM. Evolution in Apiales: nuclear and chloroplast markers together in (almost) perfect harmony. Botanical Journal of the Linnean Society. 2004;144:123–147. [Google Scholar]
  8. Erbar C, Leins P. An analysis of the early floral development of Pittosporum tobira (Thunb.) Aiton and some remarks on the systematic position of the family Pittosporaceae. Feddes Repertorium. 1996;106:463–473. [Google Scholar]
  9. Erbar C, Leins P. Sympetaly in Apiales (Apiaceae, Araliaceae, Pittosporaceae) South African Journal of Botany. 2004;70:458–467. [Google Scholar]
  10. Eyde RH, Tseng CC. What is the primitive floral structure of Araliaceae? Journal of the Arnold Arboretum. 1971;52:205–239. [Google Scholar]
  11. Hutchinson J. The genera of flowering plants. Vol. 2. Oxford: Clarendon Press; 1967. [Google Scholar]
  12. Leins P, Erbar C. Floral organ sequences in Apiales (Apiaceae, Araliaceae, Pittosporaceae) South African Journal of Botany. 2004;70:468–474. [Google Scholar]
  13. Lowry PP., II A systematic study of Delarbrea Vieill. (Araliaceae) Allertonia. 1986a;4:169–201. [Google Scholar]
  14. Lowry PP., II A systematic study of three genera of Araliaceae endemic to or centered in New Caledonia: Delarbrea, odocarpus and Pseudosciadium. 1986b PhD thesis, Washington University, St. Louis, MO, USA. [Google Scholar]
  15. Lowry PP, II, Plunkett GM, Wen J. Generic relationships in Araliaceae: looking into the crystal ball. South African Journal of Botany. 2004;70:382–392. [Google Scholar]
  16. Magin N. Das Gynoecium der Apiaceae – Modell und Ontogenie. Berichte der Deutschen Botanischen Gesellschaft. 1977;90:53–66. [Google Scholar]
  17. Magin N. Eine blutenmorphologische Analyse der Lagoecieae (Apiaceae) Plant Systematics and Evolution. 1980;133:239–259. [Google Scholar]
  18. Narayana LL, Radhakrishnaiah M. Floral anatomy of Pittosporaceae: five species of Pittosporum. Canadian Journal of Botany. 1982;60:1859–1867. [Google Scholar]
  19. Nicolas AN, Plunkett GM. The demise of subfamily Hydrocotyloideae (Apiaceae) and the re-alignment of its genera across the entire order Apiales. Molecular Phylogenetics and Evolution. 2009;53:134–151. doi: 10.1016/j.ympev.2009.06.010. [DOI] [PubMed] [Google Scholar]
  20. Nuraliev MS, Beer AS, Oskolski AA. Vascular anatomy of flowers in Tupidanthus calyptratus and related species of Schefflera in view of the origin of floral polymery in Araliaceae. Botanicheskiy Zhurnal. 2009;94:625–642. (in Russian) [Google Scholar]
  21. Nuraliev MS, Oskolski AA, Sokoloff DD, Remizova MS. Flowers of Araliaceae: structural diversity, developmental and evolutionary aspects. Plant Diversity and Evolution. 2010;128 (in press) [Google Scholar]
  22. Philipson WR. Constant and variable features of the Araliaceae. Botanical Journal of the Linnean Society. 1970;63:87–100. [Google Scholar]
  23. Philipson WR. Araliaceae. In: Van Steenis CGGJ, editor. Flora Malesiana. Vol. 9. The Hague: Martinus Nijhoff; 1979. pp. 1–105. [Google Scholar]
  24. Plunkett GM, Soltis DE, Soltis PS. Higher level relationships of Apiales (Apiaceae and Araliaceae) based on phylogenetic analysis of rbcL sequences. American Journal of Botany. 1996;83:499–515. [Google Scholar]
  25. Plunkett GM, Chandler GT, Lowry PP, II, Pinney SM, Sprenkle TS. Recent advances in understanding Apiales and a revised classification. South African Journal of Botany. 2004a;70:371–381. [Google Scholar]
  26. Plunkett GM, Wen J, Lowry PP., II Infrafamilial classifications and characters in Araliaceae: insights from the phylogenetic analysis of nuclear (ITS) and plastid (trnL-trnF) sequence data. Plant Systematics and Evolution. 2004b;245:1–39. [Google Scholar]
  27. Reyneke WF. ‘n Morfologies-taksonomiese studie van die familie Araliaceae in suidelike Afrika. 1981 DSc thesis, University of Pretoria, South Africa. [Google Scholar]
  28. Ridley HN. The dispersal of plants throughout the world. Ashford, UK: L. Reeve; 1930. [Google Scholar]
  29. Rohweder O. Centrospermen-Studien. 3. Blutenentwicklung und Blutenbau bei Silenoideen (Caryophyllaceae) Botanische Jahrbücher. 1967;86:130–185. [Google Scholar]
  30. Seemann B. Revision of the natural order Hederaceae. Journal of Botany. 1866;4:293–299. [Google Scholar]
  31. Sokoloff DD, Oskolski AA, Remizowa MV, Nuraliev MS. Flower structure and development in Tupidanthus calyptratus (Araliaceae): an extreme case of polymery among asterids. Plant Systematics and Evolution. 2007;268:210–234. [Google Scholar]
  32. Sonder OW. Araliaceae. In: Harvey WH, Sonder OW, editors. Flora Capensis. Vol. 2. Dublin: Hodges Smith; 1862. pp. 568–570. [Google Scholar]
  33. Stevens PF. Angiosperm Phylogeny Website. 2009 http://www.mobot.org/MOBOT/research/APweb/ . Accessed 2 November 2009. [Google Scholar]
  34. Strey RG. Observations on the morphology of the Araliaceae in Southern Africa. Journal of Dendrology (South Africa) 1981;1:66–83. [Google Scholar]
  35. Takhtajan A. The structural types of gynoecium and placentation. Bulletin of Armenian Branch of the Academy of Sciences of the USSR. 1942;3–4:91–112. (in Russian with English summary) [Google Scholar]
  36. Takhtajan A. Flowering plants. 2nd edn. New York: Springer; 2009. [Google Scholar]
  37. Viguier R. Recherches botaniques sur la classification des Araliacées. Annales des sciences naturelles. Série 9. Botanique. 1906;4:1–210. [Google Scholar]
  38. Weberling F. Morphology of flowers and inflorescences. Cambridge: Cambridge University Press; 1989. [Google Scholar]
  39. Wen J, Plunkett GM, Mitchell AD, Wagstaff SJ. The evolution of Araliaceae: a phylogenetic analysis based on ITS sequences of nuclear ribosomal DNA. Systematic Botany. 2001;26:144–167. [Google Scholar]

Articles from Annals of Botany are provided here courtesy of Oxford University Press

RESOURCES