Reproductive biology of Piper species (Piperaceae): a review to link the past to the future

Abstract Piper species are important components of the understory in tropical and subtropical forests, and some species are of economic and medicinal importance. This genus has been studied in different areas of science. However, some questions remain unanswered after the new circumscription as proposed by phylogenetic studies. Here, I review different aspects related to the reproductive biology of Piper species (phenology, sexual expression, floral biology, pollination and mating systems), comparing the results between Neotropical (representing most of the studies carried out so far) and Paleotropical regions. Seventy-five species were analyzed. Some patterns can be observed, mainly in Neotropical species, such as the predominance of annual flowering and fruiting, dichogamy (generally incomplete protogyny) with gradual and sequential exposure of stigmas, asynchronous pollen release, and pollination by insects. Some aspects have been poorly studied, such as variations in sexual expression, the efficiency of different pollinators, and the chemical composition of floral volatiles associated with their role in pollination. In addition, I discuss remaining gaps and further studies that are required, mainly on Paleotropical Piper species, to obtain basic information on the reproductive biology of these plants. From this, comparisons with Neotropical species can be made, allowing the verification of patterns in this genus.


Introduction
Piper has a predominantly Pantropical distribution, although it occurs in some subtropical regions (Jaramillo & Manos 2001).It has over 2,100 species and is the second-largest genus of angiosperms (Simmonds et al. 2021), with the highest diversity of species occurring in tropical America (700 species) and South Asia (300 species) Rodriguésia 74: e01012022.2023 (Jaramillo & Manos 2001).Species of this genus grow as shrubs, herbs, climbers, or treelets, and are important components of the understory in tropical forests (Hartshorn & Hammel 1994;Greig 2004).Some species are of economic importance, such as Piper nigrum L. (black pepper), which is used worldwide as a condiment (Greig 2004), and P. betle L. (betel vine) and P. methysticum G. Forst.(kava kava) which are used medicinally (Di Stasi et al. 2002;Dyer et al. 2004).Studies on the genus Piper have been conducted in several areas of science.In 2004, the book "Piper: a model genus for studies of phytochemistry, ecology, and evolution" was published, which addresses different ecological and evolutionary aspects, such as mutualistic interactions with ants (Letourneau 2004) and bats (Fleming 2004), in addition to aspects of pollination biology (Figueiredo & Sazima 2004).Despite the existing knowledge of this genus, some questions remain to be answered, especially after phylogenetic studies have redefined the circumscription of the genus (Jaramillo & Manos 2001;Jaramillo et al. 2008).
The systematics of Piperaceae has changed over time since Kunth's work in 1839, changing the number of genera within this family, in addition to the use of infrageneric groups or sections in some taxonomic treatises (see a summary of the taxonomic history of Piper by Jaramillo & Manos 2001;Sen & Rengaian 2022).After phylogenetic studies using sequences from the ITS region and chloroplast intron psbJ-petA (Jaramillo & Manos 2001;Jaramillo et al. 2008), the circumscription of Piper was altered to incorporate further genera.Three major clades are recognized, corresponding to the geographic distribution of the species.Neotropical species are divided into eight subclades, most of which are currently considered subgenera: Macrostachys, Radula, Peltobryon, Pothomorphe, Enckea, Ottonia, Schilleria, and the P. cinereum/P.sanctum complex, all of which are hermaphroditic (bisexual flowers) and are considered a basal group (Jaramillo et al. 2008;Sen & Rengaian 2022).By contrast, Paleotropical species are categorized into two geographic clades, i.e., tropical Asia (Piper s.s.) and the South Pacific (Macropiper), which produce unisexual flowers and are predominantly dioecious.Recent studies have evaluated the phylogeny of Asian pipers and found 17 infrageneric groups that constitute a sister group to the South Pacific group (Asmarayani 2018;Sen et al. 2019).
Here, I review the different aspects of reproductive biology of Piper, especially those related to pollination.I compare Neotropical with Paleotropical species and discuss knowledge gaps and potential topics for future studies.
Reproductive phenology studies require adult individuals already capable of producing inflorescences.Fleming (1985) suggested as an inclusion criterion the sampling of individuals with a height of ≥ 1 m.However, this criterion is not suitable for all Piper species (see Valentin-Silva & Vieira 2015).The latter authors proposed the mandatory presence of reproductive structures (inflorescences with flower buds, flowers, or fruits; Fig. 1a) in individuals or inflorescence scars on branches (Fig. 1b), which indicates previous reproductive episodes.Thus, sampling of juvenile individuals can be avoided, which would bias the results.
The number of sampled individuals varies among phenological studies on Piper species.However, to facilitate the comparison of results between studies, some standardizations must be applied.When possible, at least 15 adult individuals of each species should be sampled in biweekly observations, which has been suggested as suitable for phenological studies (Morellato et al. 2010a).Monthly observation is not appropriate, as some phenophases may not be recorded in some species (see Mariot et al. 2003).
Standardization of the analyses is also important.Phenological data can be analyzed using activity and intensity indices (Morellato et al. 2010a), which allow evaluation of the period of occurrence of phenophases and their duration, in addition to other phenological parameters.The frequency and duration of phenophases can be classified according to Newstrom et al. (1994).Phenophases can be described using circular statistics (Morellato et al. 2010b) and their seasonality can be assessed using the Rayleigh test (Zar 2010).To test the potential effects of abiotic factors (e.g., mean temperature, precipitation, and day length) on phenology, generalized linear models can be used for analyses (Crawley 2007).

Floral buds
Inflorescence production has been examined in a few studies, both on Neotropical (Marquis 1988;Mariot et al. 2003;Valentin-Silva & Vieira 2015;Valentin-Silva et al. 2018a) and Paleotropical (Devi et al. 2016a(Devi et al. , 2018) ) species.Some Neotropical Piper species show latent flower buds (Marquis 1988;Valentin-Silva & Vieira 2015;Valentin-Silva et al. 2018a), i.e., inflorescences are produced throughout the year, whereas anthesis of the flowers is restricted to a short period of the year, typically at the beginning of the rainy season.In addition, these species flower before those that do not have latent flower buds, and the duration of flowering is shorter, representing a possible escape mechanism from water stress (Valentin-Silva et al. 2018a).It is also possible that this temporal separation of the flowering period of these two species groups may reduce the competition for common pollinators.Due to the presence of latent flower buds, the separation of flowering into two phenophases (floral buds and flowers at anthesis) is important to consider avoiding errors in the calculation of phenological parameters (duration, concentration, start, and peak dates), in addition to not overestimating the period of availability of floral resources (pollen) to floral visitors.
Only two Paleotropical Piper species have been studied regarding phenology: P. mullesua Buch.-Ham.ex D.Don (Devi et al. 2016a) and P. sylvaticum Roxb.(Devi et al. 2018), both of which are dioecious.The respective authors recorded spike production throughout the year in both male and female P. mullesua and male P. sylvaticum plants.These results indicated that these species may have latent flower buds.However, this topic needs to be further explored in these and other Paleotropical Piper species, by observing whether they occur in both types of individuals (male and female) of dioecious species, as well as in inflorescences of monoecious species.
As some species with latent flower buds occur in different habitats (seasonal and aseasonal forests), it would be interesting to evaluate populations in these habitats to verify whether the production of latent inflorescences is an intrinsic characteristic of these species or a response to the environment.As an alternative option, greenhouse experiments could be conducted to evaluate the influence of water stress on the production of inflorescences and, consequently, on flowering (flowers at anthesis).According to Wright (1991), water stress can delay inflorescence production in Piper species; however, this study did not assess whether this effect also occurred in species with latent flower buds.
In Paleotropical species, an annual flowering pattern has also been observed at the species level (Devi et al. 2016a(Devi et al. , 2018)).However, the flowering peak can be simultaneous between male and female plants (Devi et al. 2016a) or sequential, with male plants flowering at a higher intensity before female plants (Devi et al. 2018).
Phenological patterns of species may differ between habitats.For example, Thies & Kalko (2004) observed that Piper species occurring in forest gaps tended to have longer flowering periods than species that occurred within the forest; in addition, the peak flowering period differed between habitats.Abiotic factors, such as climate, can also influence phenological patterns.Valentin-Silva et al. (2018a) showed that day length was positively correlated with the number of species flowering, as the flowering peak occurred during the rainy season when days are longer.This environmental variable is considered a trigger for flowering and is one of the most accurate predictors of this phenophase (Stevenson et al. 2008).This result corroborates the climatic hypothesis, which predicts that phenological activity is related to the seasonality of limiting environmental variables such as water and light availability (van Schaik et al. 1993;Fenner 1998).
Biotic factors such as pollinators can also influence phenological patterns.Fleming (1985) observed a temporal partition among five Piper species observed in Costa Rica, which can be interpreted as a mechanism to avoid or reduce competition for common pollinators and increase pollination effectiveness (competition hypothesis;van Schaik et al. 1993;Fenner 1998).By contrast, Valentin-Silva et al. (2018a) examined 17 species of Piper co-occurring in Brazil and found an overlap in the flowering period between species groups (e.g., five of the eight species with latent flower buds flowered simultaneously at the beginning of the rainy season).As these species produce flowers with similar morphology, simultaneous flowering may attract more pollinators, thereby increasing pollination rates (facilitation hypothesis; van Schaik et al. 1993;Fenner 1998).Both competition and facilitation may be important in different areas, depending on pollinator availability.As these hypotheses are difficult to test in natural settings, the role of biotic vectors as modulators of plant phenology can be evaluated using the null model analysis (see Staggemeier et al. 2010).
The phylogenetic hypothesis predicting that closely related species tend to exhibit more similar phenological patterns than more distantly related species (Kochmer & Handel 1986) was tested on Piper species.However, Valentin-Silva et al. (2018a) found no association between phylogenetic relatedness and phenological patterns of the examined species; regardless, these authors suggested that the predominance of annual flowering frequency may be related to phylogenetic factors.Improving the phylogenetic resolution of species-rich clades, together with more phenological information, may thus provide new perspectives on how evolutionary history has affected the phenology of Piper species.

Fruiting
As with flowering, fruiting commonly follows an annual pattern at the species level (Opler et al. 1980;Fleming 1985;Marinho-Filho 1991;Thies & Kalko 2004;Valentin-Silva & Vieira 2015;Valentin-Silva et al. 2018a) and shows continuous pattern at the community level (Thies & Kalko 2004;Valentin-Silva et al. 2018a) in the Neotropics.In the Paleotropics, two Piper species showed a continuous pattern of fruiting (Devi et al. 2016a(Devi et al. , 2018)).Most of these studies did not separate data on immature and mature fruits due to the difficulty of recording ripe fruits in biweekly or even weekly observations (e.g., Valentin-Silva & Vieira 2015), as these fruits remain on the plant for one or two days (Marquis 1988;Thies & Kalko 2004).Therefore, the fruiting period is long, which may be related to the low availability of ripe fruits per plant and night (Fleming 1981;Thies & Kalko 2004).Fleming (1981) observed that in P. amalago L., a few ripe fruits (1 -3) were available per night over a period of 3-4 weeks.These analyses of ripe fruit availability and removal rate by dispersers were based on daily observations at night, as Neotropical Piper species are generally Rodriguésia 74: e01012022.2023 dispersed by frugivorous bats (Fleming 2004;Thies & Kalko 2004), mainly Carollia perspicillata (Linnaeus 1758) (Marinho-Filho 1991; Lima & Reis 2004).Further information on the dispersal ecology of Neotropical Piper species was compiled by Fleming (2004); however, the dispersal mechanisms of Paleotropical Piper species are still poorly understood.
Phenological patterns of fruiting can also differ between species, depending on the habitat in which they occur.Thies & Kalko (2004) observed fruiting in Piper species growing in a forest throughout the year, but with a staggered pattern among species.This pattern can be interpreted as a response to the limitation of dispersers, which helps reduce interspecific competition for dispersers in addition to reducing competition for germination sites.The authors also confirmed that Piper species occurring in forest gaps showed fruiting throughout the year, with each species showing more than one peak of fruit production and longer fruiting periods.The overlap of these long fruiting periods is associated with a broad spectrum of dispersers (bats, birds, and probably ants) and with the need for specific conditions for seedling establishment (gap formation).
Some of the hypotheses on factors influencing phenological patterns were also tested with respect to fruiting.Valentin-Silva et al. (2018a) showed that the number of species with fruits was positively correlated with mean temperature and day length of the months prior to the occurrence of the phenophase, and there was no phylogenetic structuring.Fruit production was constant throughout the year (continuous fruiting pattern), which is associated with the maintenance of frugivorous disperser populations (Snow 1965).These results indicate the importance of dispersers as modulators of the fruiting phenology of Piper species, as suggested by Thies & Kalko (2004).
Variations in the distribution of flowers on inflorescences were also reported by Valentin-Silva et al. (2018b).In P. arboreum, P. caldense, and P. cernuum, spikes with bisexual flowers and spikes with staminate flowers were observed.In addition to these types of inflorescences, P. chimonanthifolium showed spikes with bisexual and staminate flowers (mixed spikes), a characteristic described for the first time in a Neotropical Piper species.In all four species, the male spikes and the portion of mixed spikes with staminate flowers senesced after pollen release.Moreover, the distribution of inflorescence types varied between individuals of these four species, which directly influences their sexual expression (Valentin-Silva et al. 2018b).Piper cernuum was considered andromonoecious, whereas the populations of P. arboreum and P. chimonanthifolium appeared to be in transition to this type of sexual expression, as they are andromonoecious and hermaphrodite individuals.Piper caldense is probably androdioecious because individuals have hermaphrodite and male functions.Androdioecy is rare in angiosperms (Pannell 2002), and studies on other populations of this species may help answer this question.
A further interesting characteristic of P. arboreum, P. caldense, P. cernuum, and P. chimonanthifolium is variation in pistil length and in the size of spikes with bisexual and staminate flowers, indicating sexual dimorphism, as observed by Valentin-Silva et al. (2018b).The authors suggested that inflorescence length may be a diagnostic characteristic to identify the types of individuals in Neotropical Piper species that show sexual expression different from hermaphroditism.Because of the small size of flowers, analysis of floral morphology related to the functionality of flowers during anthesis is essential to determine the sexual expression of Neotropical Piper species.It is likely that studies of this type in other populations in the Neotropics will reveal further species with functionally unisexual flowers.
In the Paleotropics, Piper species with unisexual flowers and dioecy are predominant (Wanke et al. 2007;Jaramillo et al. 2008).In the Asian tropics, there are some species with bisexual flowers, such as Piper longum L. (Kanimozhi & Rodriguésia 74: e01012022.2023 Sujatha 2015), and in the South Pacific, some species are monoecious (Jaramillo et al. 2008).Despite this, Ollerton (1996) noted in P. novaehollandiae, an Australian species usually described as monoecious, individuals that only had spikes with staminate flowers.Female inflorescences tend to be shorter than those of males in some Paleotropical Piper species (Kanimozhi & Sujatha 2015;Devi et al. 2016b), indicating sexual dimorphism, similar to Neotropical species that have bisexual and staminate flowers.
Mixed inflorescences were observed in three dioecious species.Piper pedicellatum C.DC. has some inflorescences with bisexual and pistillate flowers in addition to female inflorescences (Lei & Liang 1998).Piper sylvaticum has some inflorescences with bisexual and staminate flowers in addition to male inflorescences (Devi et al. 2016b).In Piper mullesua, a few pistillate flowers have been observed along with staminate flowers in addition to male inflorescences (Devi et al. 2016b).However, these studies did not examine potential differences in length between these types of inflorescences.It would thus be interesting to analyze these variations in other populations, preferably under natural settings (Devi et al. 2016b studied cultivated individuals), or even in herbarium material (see Valentin-Silva et al. 2018b) to confirm such changes in the sexual expression in these species, considering that sexual expression can vary in time and space (Sakai & Weller 1999).

Floral biology
The minute flowers of Piper appear, at first glance, morphologically similar; both bisexual and unisexual flowers are perianthless and are protected by a bract, with 1-10 stamens and 3-4 carpels (Jaramillo & Manos 2001).However, there are variations regarding flower (relative size of filament and anther, orientation of the anther opening, presence or absence of pedicel) and inflorescence (packaging of flowers, position, length, presentation, color, and arrangement) (Jaramillo & Manos 2001).Studies involving aspects of floral biology were carried out with 29 species.
As most Piper species are protogynous, anthesis begins with the exposure of stigmas, which occurs sequentially and gradually and in a basipetal direction (Valentin-Silva et al. 2015, 2018c;Vargas-Rojas & Vieira 2017).This mode of stigmatic papillae exposure is related to both incomplete protogyny and stigma longevity (4-16 days; Valentin-Silva et al. 2018c), which increases the likelihood for stigmas to receive sufficient pollen grains (Cruden 2000).Valentin-Silva et al. (2018c) observed this trait in Neotropical Piper species with three and four stigmas, indicating that the mode of stigmatic papillae exposure was independent of the number of carpels in the flower.
Pollen release, in turn, is asynchronous and sequential, as it occurs in one stamen at a time, typically between 10:00 am and 2:00 pm, showing clump formation on the anther (Valentin-Silva et al. 2015, 2018c;Vargas-Rojas & Vieira 2017).This pattern differs from that of androecium development, which is asynchronous and occurs in a specific sequence.The lateral pairs of stamens develop simultaneously, and the median pair develops sequentially, with the median-anterior stamen being formed before the median-posterior stamen (Tucker 1982).During anthesis, pairs of lateral stamens release pollen at different times on the same day or on subsequent days; the median pair of stamens releases pollen sequentially on subsequent days (Valentin-Silva et al. 2015, 2018c).This trait was observed in Neotropical Piper species with two, three, four, and six stamens.Therefore, it does not depend on the number of stamens in the flower, although in flowers with more stamens, the period of pollen supply tends to be longer, increasing the likelihood of cross-pollination (Valentin-Silva et al. 2018c).Asynchronous and sequential pollen release has also been recorded in the staminate flowers of Neotropical species (Vargas-Rojas & Vieira 2017;Valentin-Silva et al. 2018c).
Based on the similarities between bisexual and unisexual flowers regarding morphology, floral development (Tucker 1982;Lei & Liang 1998;Jaramillo et al. 2008), and sequence of pollen release (Vargas-Rojas & Vieira 2017;Valentin-Silva et al. 2018c), in addition to the long-lived stigmas in P. nigrum (Menon 1949;Martin & Gregory 1962), a genus-wide pattern in the dynamics of floral events may exist.However, further studies on Paleotropical Piper species are needed to test this hypothesis.Studies on the floral biology of Paleotropical species generally examined the timing of pollen release and viability, in addition to stigmatic receptivity (Kanimozhi & Sujatha 2015;Chen et al. 2018), but do not explain how this dynamic occurs.However, Chen et al. (2018) studied the stigmatic receptivity of ten cultivars of P. nigrum at three anthesis stages using Peroxtesmo KO (Machery-Nagel, Düren, Germany).From the images of this study, it is possible to observe that the stigmatic papillae become receptive sequentially, gradually, and in a basipetal direction, which reinforces the suggested similarity with Neotropical species.
Piper flowers have only pollen as a resource and generally have a sweet or citrusy odor, as reported for 13 Neotropical species (Figueiredo & Sazima 2000) and two Australian species (Ollerton 1996).These odors may be related to the attraction of pollinators, which is one of the key functions of volatile organic compounds (Farré-Armengol et al. 2013).However, studies to elucidate the chemical composition of these odors and their role in attracting pollinators are scarce.Machado et al. (2021) observed that linalool compounds were predominant in the flowers of P. mollicomum Kunth, and their emission was correlated with a higher frequency of bee visits.

Pollination and mating systems
Thirty-six Piper species had floral visitors and pollinators evaluated (Tab.S1, available on supplementary material <https://doi.org/10.6084/m9.figshare.22626970.v1>).Neotropical Piper species are entomophilous and are pollinated by different species of bees, flies, and beetles (Tab.1).Therefore, they are generalists in terms of ecology (number of pollinator species sensu Ollerton et al. 2007) and functionality (taxonomic diversity of pollinators sensu Ollerton et al. 2007).Other floral visitors have also been recorded, such as butterflies, wasps, hemipterans, and spiders (Fleming 1985;Figueiredo & Sazima 2000, 2004;Figueiredo 2003;Valentin-Silva et al. 2021).However, they were not considered pollinators because of their low abundance, their behavior and frequency of visits, and/or the absence of pollen adhering to the body.
Neotropical Piper flowers are typically visited during the day, with a peak in the late morning, mainly between 10:00 am and 11:00 am (Kikuchi et al. 2007;Vargas-Rojas & Vieira 2017;Valentin-Silva et al. 2021).However, there may be some variations in the peak visiting periods between Piper species (Thomazini & Thomazini 2002;Figueiredo 2003).Visitations by these insects can be seasonal, as their proportions of visits vary throughout the year (Fleming 1985;Thomazini & Thomazini 2002;Wisniewski et al. 2019).Insects generally exhibit similar visiting behavior, crawling along a part or over the full length of the inflorescence, and they commonly visit more than one inflorescence per plant (Figueiredo & Sazima 2000;Kikuchi et al. 2007;Wisniewski et al. 2019;Valentin-Silva et al. 2021).However, the duration of visits varies between insects.Bees and flies tend to show shorter visitation times than beetles, which can remain on the inflorescences for up to an hour, sometimes with little movement (Figueiredo & Sazima 2000;Figueiredo 2003;Wisniewski et al. 2019;Valentin-Silva et al. 2021).
The visiting behavior of such insects (bees, flies, and beetles) can affect the rates of autogamy and geitonogamy (Lloyd & Schoen 1992) in self-compatible Piper species.In the Neotropics, both self-compatible (13 species; Marquis 1988;Figueiredo & Sazima 2000;Kikuchi et al. 2007;Valentin-Silva et al. 2021) and self-incompatible species (16 species; Figueiredo & Sazima 2000;Vargas-Rojas & Vieira 2017;Valentin-Silva et al. 2021) have been reported.Figueiredo & Sazima (2000) and Valentin-Silva et al. (2021) studied seven common species, three of which showed divergent incompatibility systems between the examined populations.Considering that interand intraspecific variation in outcrossing rates is common (Barrett 2003), progeny analyses are required to confirm whether self-compatible Piper species are exclusively autogamous or whether they can benefit from the presence of pollinators.
Despite pollination by different insects, no studies have examined whether pollinators differ in terms of effectiveness in pollinating Piper species.Such studies could be performed on self-incompatible Piper species because of the difficulty in emasculating the flowers and isolating the effect of self-fertilization.Thus, it is possible to assess whether pollinators differ in terms of their contribution to the sexual reproduction of these plants.Rodriguésia 74: e01012022.2023 In addition to biotic pollinators, wind pollination has been reported in conjunction with insects (ambophily; Figueiredo & Sazima 2000).The occurrence of anemophily in Piper species is still controversial.Culley et al. (2002) reviewed the origin of anemophily in angiosperms.Even considering that anemophilous species do not consistently possess all characteristics associated with this pollination syndrome, Piperaceae did not appear to be wind-pollinated.Some studies have ruled out this pollination mechanism due to the formation of pollen clumps on the anther, possibly indicating the presence of pollenkitt (Semple 1974;Valentin-Silva et al. 2021).According to these studies, such pollen clumps can remain on the anther from one day to the next, even after rainfall or wind.By contrast, Figueiredo & Sazima (2000) observed powdery pollen in Neotropical Piper species.Despite this divergence, other characteristics indicate better suitability of Piper flowers for entomophily: abundant insect visitation in most species and stigmas closely surrounded by stamens (Valentin-Silva et al. 2021).An experiment to simulate the removal of pollen from anthers by the wind showed that when pollen is transported by wind, the distance traveled (up to 30 cm) is insufficient for cross-pollination to occur (Valentin-Silva et al. 2021).A respective study on Piper species producing powdery pollen may elucidate the importance of wind as a pollen vector in the sexual reproduction of these plants.Valentin-Silva et al. (2021) evaluated the role of pollinators in the diversification of floral traits of Neotropical Piper species, and no correlation between the flower morphological characteristics and their pollinators was found, indicating phenotypic generalization.The authors suggest that dispersers which tend to be more specialized than pollinators of Piper may have influenced the diversification of the genus; however, further research is required.
The pollination biology of Paleotropical Piper species has been poorly studied (see Sen & Rengaian 2022), even though these species have predominantly unisexual flowers and probably depend on pollen vectors.The few existing studies were carried out mainly with P. nigrum, species originating in India that is widely cultivated throughout the humid tropics (Greig 2004).Spring tails (Collembola) were observed to visit the flowers in populations of four varieties of P. nigrum cultivated in Puerto Rico (Martin & Gregory 1962).However, as these insects are wingless, they likely contribute little to pollen flow between plants.Some authors proposed that water and wind may be important pollen vectors for sexual reproduction of black pepper (Menon 1949;Gentry 1955a,b), however, there is no experimental evidence (see Martin & Gregory 1962;Sasikumar et al. 1992).This species, which comprises several varieties that can be hermaphroditic, dioecious, or monoecious (Menon 1949), is self-pollinating or even apomictic (Gentry 1955b;Sasikumar et al. 1992; but see Pooja et al. 2022), both characteristics can be adaptations to cultivation (Martin & Gregory 1962).However, the occurrence of apomixis needs to be confirmed through an embryological study, which could also be performed on Neotropical Piper species with a high fruiting rate and low visitation frequency, such as P. umbellatum L. (Figueiredo & Sazima 2000;Valentin-Silva et al. 2021).
In addition to studies on P. nigrum, there are only two other studies on the aspects of pollination with Paleotropical Piper species.Ollerton (1996) observed that in Australia, flies (gall midges; Diptera: Cecidomyiidae) visit and lay eggs on male inflorescences of P. novae-hollandiae Miq., which produces sweet-smelling flowers.However, as no female inflorescences were observed, whether these flies contribute to pollen flow between plants is not yet known.An interesting floral visitor was observed by Okuyama (2014) in the Philippines, a crab species was found on male inflorescences of an unidentified Piper species.Although the author recorded the presence of anthers attached to the body of this crustacean, the absence of visits to female inflorescences and the record of this crab in only one individual of Piper indicates that this animal probably contributes little to pollen flow between plants.

Conclusion
The reproductive biology of Piper has advanced in recent years, mostly in Neotropical species.Some patterns can be observed, such as the predominance of annual flowering and fruiting, dichogamy (generally incomplete protogyny) with gradual and sequential exposure of stigmas, asynchronous pollen release, and pollination by insects.However, these aspects have not been extensively studied in the Paleotropical Piper species, and further studies are required to obtain basic information on the reproductive biology of these plants.From this, comparisons with Neotropical species can be made, allowing the verification of patterns in this genus.It Rodriguésia 74: e01012022.2023 will also be possible to apply these results to evolutionary studies, for example, regarding floral diversification.
From existing knowledge, it is possible to point out some gaps that deserve attention.Phenological data can be associated with germination ecology to assess their influence on reproductive success, especially for seasonal vegetation.Phenological responses can also be studied in the context of climate change.For example, the production of flower buds (latent and non-latent) in a scenario of lower precipitation and the consequences for subsequent phenophases and population dynamics.Another interesting point to be evaluated is the frequency of occurrence of cryptic unisexuality in morphologically bisexual flowers, considering that hermaphroditism is a typical characteristic of the Neotropical clade, and the selective pressures that can culminate in changes in the sexual expression of these plants.Regarding floral biology, nothing is known about the dynamics of floral odor release or the plant tissues that produce it.Chemical ecology studies can assess when the release of volatile organic compounds occurs, considering the predominance of dichogamy in Neotropical species and unisexual flowers in Paleotropical species.Moreover, fractions of these floral odors can be used in bioassays that evaluate their role in attracting different pollinators in a generalist pollination system such as Piper.Genetic analyses in self-compatible Piper species can contribute to studies on the evolution of mixed mating systems.Due to the organization of inflorescences and the visiting behavior of pollinators, the occurrence of this type of mating system may not be related to the unpredictability of pollinators, as commonly reported in the literature.
Piperaceae is a mega-diverse (approximately 3,500 species) and early diverging lineage of angiosperms.Besides Piper, there are four other genera in the family (Manekia, Peperomia, Verhuellia, and Zippelia) that have been little studied in relation to their reproductive biology.Information on these plants can be useful for understanding the diversification of angiosperms.Thus, the study of different aspects of the reproductive biology of Piper remains a promising field of research.

Figure 1
Figure 1 -a-b.Criteria for sampling adult Piper individuals -a.presence of reproductive structures; note spikes with floral buds (light green color; black arrow), flowers at anthesis (cream color; white arrow), and fruits (green color; arrowhead); b. presence of inflorescence scar on the branch (circle).

Table 1 -
Studies carried out on the pollinators of Neotropical Piper species.= Halictinae species were not identified, as well as those of Syrphidae. *