Androdioecious Species

Androdioecious (and nearly or potentially androdioecious) species:

Androdioecy: The presence of separate male and hermaphroditic individuals in a breeding population. This dimorphism must be genetically determined.

Functional androdioecy: The male individuals must reproduce (gain fitness) only through male function, whereas hermaphrodites must gain fitness through both male and female function. If hermaphrodites gain some fitness through male function, male frequencies are expected to be less than 0.5 (Charlesworth 1984).

Gender diphasic populations (sex switchers) may be easily mistaken for androdioecious poulations. A dimorphism may exist in which individuals may be either male or be hermaphroditic in any given year, but this dimorphism is not genetically determined. Rather, individuals posess a plastic genotype which allows them to choose the phenotype that will maximize their fitness in a given year. Small or stressed individuals often choose the male phenotype while larger, healthier individuals choose the hermaphroditic morph, perhaps because successful seed production requires more resources than pollen production, such that small plants may be unable to successfully mature seeds. Some gender diphasic species with size-dependent male and hermaphroditic phases include green dragon (Arisaema dracontium, Araceae, Clay 1993), dwarf ginseng (Panax trifolium L., Araliaceae, Schlessman 1990), and Apodanthera undulata, Cucurbitaceae (Delesalle 1989).

Acer sp.; Aceraceae

- Several species of maple apparently have a complex form of androdioecy called heterodichogamous androdioecy by Gleiser and Verdu 2005: Males exist in low frequency along with two types of hermaphrodites, one which produces pistillate flowers first, then staminate, and another morph that first produces staminate flowers, then pistillate flowers.
- According to phylogenetic work by Gleiser and Verdu 2005, this androdioecious breeding system may be ancestral to dioecy in the genus Acer.

Anagellis monelli; Primulaceae

- Although this species is not androdioecious because there are no purely male individuals, it is clearly useful to the study of androdioecy, and subandrodioecy may be a good descriptor of the breeding system (as suggested by Gibbs and Talavera 2001).
- There appear to be two morphs in the population studied, one of which is "male dominant" (Gibbs and Talavera 2001). Five of 27 plants investigated had very low seed set in hand pollinations and natural pollinations. The seed set of "male dominant" plants was disjunct, and significantly lower (5 seeds/capsule) than that of the other plants (15-25 seeds/capsule). The "male dominant" plants were good pollen donors, so the low female fertility was not due to SI. It was also not due to insufficient pollen (plants were either hand-polinated or were within 2m of another plant).
- The species appears to be self-incompatible, determined by a single-locus, multi-allelic gametophytic SI (GSI) system (Talavera et al. 2001).
-SI seems to be common in the Primulaceae (heteromorphic SI in Primula sp., and non-heteromorphic SI in other members of the tribe Anagallideae
- Plants appear to be insect polinated, have a showy display, and plants with close neighbors are more likely to set seeds (Gibbs and Talavera 2001).
- Studies are from SW Spain
- Shortlived perennial
- Other species in the genus that were studied by Gibbs and Talavera 2001 appear to be autogamous selfers.
- Gibbs and Talavera 2001 suggest that the gender dimorphism may be widespread because of a historical paper (Willkomm 1870) which describes A. collina Schousb. and A. linifolia L. as having different fruiting pedicels. They say that these two species are now recognized as A. monelli, and that they may represent the two geneder morphs, however they do not say if the two gender morphs actually have different fruiting pedicels.
- This species needs more study. It would be nice to know:
Are there morphological differences between the gender morphs?
Do plants express the same morph year after year (or are they gender labile)?
Are the morphs genetically determined?
Are the morphs widespread, or is this just a rare deleterious mutation which reduces female fertility?
Do the "male dominant" plants produce more pollen or more-fertile pollen or more flowers?

Castilla elastica; Moraceae

- Pollinated by thrips (Sakai 2001).
- Hermaphroditic trees have separate male and female inflorescences (Sakai 2001).
- Some male inflorescences on male trees are much larger than those on hermaphroditic trees (Sakai 2001).
- There was no difference in diameter at breast height of male and hermaphroditic trees, so there is no size-dependent gender switching (Sakai 2001).
- Hermaphrodite and male pollen grains both stained red with methylene blue/phloxine-B solution, and no differences were observed under the microscope, suggesting that both morphs produce good pollen (Sakai 2001). Hermaphrodite pollen, however, was not tested in vivo.
- Sakai suggests that C. elastica is a colonist (Berg, 1972), so hermaphrodites may have invaded due a lack of pollination in newly founded populations.
- Androdioecy is derived from dioecy (Datwyler and Weiblen 2004). - Datwyler and Weiblen (2004) suggest that Helianthostylis could represent a second origin of androdioecy from dioecy within the Moraceae, however, I do not know of any data showing whether Helianthostylis is functionally androdioecious.

Caenorhabditis elegans; Nematoda: Rhabditida: Rhabditoidea

- Some nice pictures and videos.
- Hermaphrodites are XX, while males are X0.
- Hermaphrodites generally self-fertilize their ovules, and can only outcross by mating with a male.
- Fifty percent of outcrossed progeny are male, but outcrossing alone is not sufficient to maintain males.
- Males are produced by nondisjunction (an XX hermaphrodite produces a gamete containing no X, which produces an XO embryo after fertilization).
- Males are maintained in the lab at a frequency between .001 (Ward & Carrel 1979) and 0.001 (Hodgkin & Doniach 1997).
- Heat treatment (27C for 36h or 30C for 6h) increases male frequency to about 2% (Hodgkin 1983).
- Hermaphrodite embryos can also be converted to males by the loss of one X chromosome (Prahlad et al. 2003).
- Recurrent nondisjunction seems to be needed for the maintenance of males (Hedgecock 1976, Chasnov & Chow 2002, Stewart & Phillips 2002), similar to the maintenance of deleterious mutations by mutation-selection balance.
- The breeding system in C. elegans is derived from dioecy (Fitch and Emmons 1995 , Fitch et al. 1995, Figure from Fitch Lab).
- Hermaphrodites in C. elegans are essentially females that have the ability to produce sperm (Haag and Kimble 2000).
- Males will mate many times over as long as 6 days of adulthood (Hodgkin 1983).
- Males live longer than hermaphrodites in 3 of 4 androdioecious nematode species studied, and in 4 dioecious species. Males of the dioecious species lived longer than males of the androdioecious species (McCulloch and Gems, 2003).

Culcita macrocarpa;Culcitaceae

- A fern that grows in Spain.
- Gametophytes switch from male when young to hermaphrodite as they become older(Quintanilla et al. 2005 Flora 200:187-194).
- Quintanilla et al. 2005 describe this species as androdioecious because at any one time, some individuals are male and some are hermaphrodites. If the gender morphs were genetically determined, I would consider it to be androdioecious too. However, since gametophytes switch gender according to their age, I would consider this to be a gender diphasic system, where all individuals (gametophytes) are functional hermaphrodites which gain fitness through both male and female function.

Datisca glomerata; Datiscaceae

- (Liston et al. 1990)
- D. glomerata is a tall, wind pollinated, perennial herb, that grows along streams and occasionally on seeps in southern California, USA and Baja California, Mexico (Davidson 1973).
- Populations are generally small, ranging from just a few individuals, to 500 or more. Male frequencies range from 0 to 0.24 (Liston et al. 1990, Wolf et al. 2001b), and males produce three times as much pollen as hermaphrodites (Philbrick & Rieseberg 1994). - Androdioecy probably arose from dioecy (Rieseberg et al. 1992), but the phylogentic evidence is unclear (Swensen et al. 1998)

Eulimnadia texana; Conchostraca: Limnadiidae

- This androdioecious clam shrimp has been well studied by Steve Weeks, Naida Zucker and colleagues.

Fuchsia microphylla; Onagracae

-Arizmendi, Dominguez and Dirzo, Functional Ecology, 1996 suggest that this species is androdioecious. However the goal of the paper was to examine nectar robberby by humming birds, rather than to assess whether the species was functionally androdioecious, so insufficient information is available to determine if this is likely to be a good case of androdioecy or not.
- The species is hummingbird pollinated, and male flowers were robbed more often than hermaphrodie flowers (Arizmendi et al. 1996).
- They report that male flowers lack gynoecium and are longer and brighter than hermaphrodite flowers (Arizmendi et al. 1996).
- Gender remained constant over 3 years (Arizmendi et al. 1996).
- Self-compatible (Arizmendi et al. 1996).
- Arizmendi et al. 1996 do not report the sex ratio in the population they studied in Manantlan, Mexico.
- Arroyo and Raven, Evolution, 1975 reported that this species superficially appears to be *gynodioecious*, but is actually subdioecious. 90% of individuals were male.

Fraxinus lanuginosa; Oleaceae

-hermaphrodite inflorescences have 1/3 as many flowers as male inflorescences (Hiura & Ishida 1994).
-flowers have no nectar, and are polinated by beetles, syrphid flies and bees, all of which eat or collect pollen, and polinated by wind (Hiura & Ishida 1994).
-pollen germination on agar plates was really low, but about 3x higher for males than for hermaphrodites ( Ishida & Hirua 1998).
-hand pollinations with male pollen produced 2.6x as many seeds as pollinations with hermaphrodite pollen (Ishida & Hirua 1998).
-seed set for manual self-pollination treatments was very low, lower than hermaphrodite outcrossing, but don't know if caused by SI or inbreeding depression (Ishida & Hirua 1998).
- seed set for natural pollination was low for 3 of the 5 trees (Ishida & Hirua 1998), so there could be some pollen limitation.
-herms make 1/8 as much fertile pollen as males (Ishida & Hirua 1998).
- Ishida & Hirua 1998 suggest that pollen may be maintained in hermaphrodites to attract pollinators.
- male frequencies in populations analyzed by Ishida & Hiura 2002 were 0.59, 0.61, 0.47, 0.25, 0.11 and 0.35.
- outcrossing rates ranged from 0.66-0.99, and were correlated with male frequency, and with density (density was also correlated with male freq). In the populations with low male frequencies, outcrossing rates were significantly lower than 1, so apparently selfing occurrs in small populations with low male frequencies (Ishida & Hiura 2002).
- However, adult fixation index (F) was not different from 0, so selfed offspring may rarely survive to reproduce (Ishida & Hiura 2002).
- Over 5 years of study, 3 out of 354 individuals changed sex (0.8%). Since the trees live for about 100 years, it is possible that more individuals were sex changers (Ishida & Hiura 2002).
- Derived from hermaphroditism and in the same clade as F. ornus(Wallander 2001).

Fraxinus ornus; Oleaceae

- (Dommee et al. 1999)
- Sex ratio 1:1 or slightly male-biased, but only hermaphrodites set fruit during 2-4 yr of observation.
- Hermaphrodites produced viable pollen.
- Males are more likely to flower in consecutive years than hermaphrodites.
- No gender switching observed in 5 yrs of observation.
- Insect pollinated.
- No autogamy, but some plants are self-fertile.
- Lives in southern Europe, and is actively spreading.
- Derived from hermaphroditism (Wallander 2001).

Fraxinus apertisquamifer, F. bungeana, F. floribunda, F. paxiana, F. sieboldiana, F. trifoliolata, F. longicuspis and F. micrantha are all described as androdioecious in Wallander 2001. All of these species are in the section Ornus, and represent a single origin of morphological androdioecy (Wallander 2001). There are no studies on the functional status of androdioecy.

Lloydia serotina; Liliaceae

- Was suggested to be potentially androdioecious (Jones & Gliddon 1999), because male and hermaphrodite morphs were oserved. Individuals, however, switch gender gender from year to year (Manicacci & Despres 2001), so this species cannot be considered androdioecious or even nearly-androdioecious.

Mercurialis annua; Euphorbiaceae

- Appears to be derived from dioecy (Krahenbuhl et al. 2002)
- Populations apparently contain two genetic morphs: a canalized morph (hermaphrodites with separate male and female flowers), and a plastic morph, which can choose to be hermaphroditic or male, depending on the density of plants in the population (Pannell 1997a,b,c). Some authors consider this to be a form of androdioecy, others do not (Delph and Wolf 2005).

Neobuxbaumia mezcalaensis; Cactaceae

- (Valiente-Banuet et al. 1997)
-A bat-pollinated columnar cactus.
-Self-incompatible.
- Habitat description and range (in Spanish).

Oxalis suksdorfii; Oxalidaceae

- (Ornduff 1972)
- Apparently functionally androdioecious, though morphologically tristilous.
- There are three flower morphs, called "long," "mid," and "short." The "mid" morph produces almost no seeds, and apparently functions as a male, while the other two morphs produce functional pollen and seeds.
- Self incompatible.
- Presumably derived from tristily with self-incompatibility.
- Insect pollinated.
- Grows in western USA.

Phillyrea angustifolia; Oleaceae

- Flowers [1] [2], Fruits
- A wind-pollinated shrub.
- Populations seem to range from functional dioecy (an established population in southern France, in which hermaphrodite pollen was unsuccessful in hand pollinations; Lepart and Dommee 1992) to functional androdioecy (a recently established population in southern France; Lepart and Dommee 1992 and a population in the Balearic Islands of Spain; Traveset 1994) to a system in which there is a continuum of gender expression among plants ranging from male to fully hermaphroditic (also in the Balearic Islands; Traveset 1994). - Hermaphrodite and male pollen are morphologically different (Traveset 1994), but there is variation in the ability of pollen to produce seeds in hand pollinations. In some populations, hermaphrodite and male pollen did not differ, whereas in others, hermaphrodite pollen was more successful than male pollen (Lepart and Dommee 1992, Traveset 1994). However, the validity of these results is questionable because self-incompatibility was not accounted for in these crosses, and only a small number of individuals was crossed. In vitro experiments did not show that hermaphrodite pollen was less likely to germinate than male pollen (Lepart and Dommee 1992). - Later crossing studies accounted for self-incompatibility (Vassiliadis et al 2000)
- Hermaphrodites produce functional pollen (crosses), but hermaphrodite pollen was less efficient than male pollen ( Vassiliadis et al. 2000).
- Paternity analysis of a small isolated population in southern France, with a 1:1 sex ratio showed that males and hermaphrodites had approximately equal sireing success (Vassiliadis et al. 2002). In other words, males did not show the siring advantage need for their maintenance with hermaphrodites.
- A minor caveat is that at least 47% of offspring were sired by pollen from outside the population, so the sex of the father could not be identified (Vassiliadis et al. 2002).
- If the self-incompatibility system is genetically linked to the sex-determination system, balancing selection on SI alleles could maintain the high frequency of males seen in this species (Vassiliadis et al. 2000). However, whether these two loci are indeed linked has not yet been studied.
- Sex ratios vary among populations, ranging from 1:1 or male biased in southern France (Depart & Dommee 1992), to hermaphrodite-biased in the Balearic Islands of Spain (Traveset 1994). A study of 13 sites in Spain and Portugal showed that the apparent frequency of males was greater than 0.5 at several sites (it ranged from 0.34-0.63), but the authors believe that the actual frequency of males is much lower (about 0.3 overall) because hermaphrodites were less likely to flower every year. The evidence is that when non-flowering individuals are included in the analysis, male frequencies were less variable among sites than hermaphrodite or non-flowering frequencies. Also, there were fewer hermaphrodites evident, and more non-flowering individuals at stressed sites (dry or recently burnt areas). The authors suggest that "reduced frequency of flowering by hermaphrodites relative to males may help to explain the maintenance of androdioecy in this long-lived, woody species" (Pannell & Ojeda 2000).
- The species has recently undergone extensive population growth in southern France (Vassiliadis PhD dissertation 1999, cited in Vassiliadis et al. 2002). Lepart and Dommee (1992) suggest that where P. angustifolia is presently an invader, populations are functionally androdioecious, because hermaphrodite pollen is functional, whereas in older, established populations, the species is functionally dioecious, because hermaphrodite pollen was not successful in hand pollinations (only two males tested).

Phillyrea latifolia

-a Mediterranian tree. -may also be androdioecious, but has not been studied in detail.
-Lepart and Dommee 1992 state that they observed P. latifolia in one population, and there appear to be distinct male individuals, as well as hermaphrodites bearing both male and hermaphrodite flowers like in P. angustifolia.
-Aronne and Wilcock 1994 studied a population in southern Italy. 67% of plants were male, males and hermaphrodites produced equally viable pollen, but males produce three times as much pollen per flower and more flowers per node, as well as more nodes per branch. The species is wind pollinated.

Ricinocarpus pinifolius; Euphorbiaceae

- (Thompson et al. 1989)
- Insect pollinated.
- Lives in Australia.
- Completely self-compatible, with late abortion of some selfed seeds.
- Little geitonogamous selfing due to temporal separation of male and female functions.
- No pollen limitation observed.
- Individual flowers are either staminate or pistillate.
- The fraction of pistillate flowers per plant ranged continuously from 0 to 0.68 (Thompson et al. 1989).
- This species is unusual and is clearly useful in furthering the understanding of androdioecy and breeding system evolution in general. However, because this species has continuous variation in sexual phenotype, rather than two distinct sex morphs, it probably should not be considered to be androdioecious.
- The authors suggest that this unusual breeding system may be maintained because the population has two waves of flowering - a pistillate wave and a staminate wave. The predominately male plants produce staminate flowers during the pistillate wave, so are producing much more pollen than the other plants during the pistillate wave, so may sire more seeds than the hermaphroditic plants.

Sagittaria lancifolia Subsp. Lancifolia; Alismataceae

- An aquatic plant.
- Not strict androdioecy, but some populations may have a breeding system similar to functional androdioecy (called sub-androdioecy by Muenchow 1998).
- There appear to be two genetically determined sex morphs. One morph (called cosexual) has 35% pistillate buds, while the other morph (called predominant males) has 0-2% pistillate buds (Muenchow 1998).
- 84% of the population expressed the cosexual morph, while 16% was the predominantly male morph (Muenchow 1998).
- No gender switching was seen in the field or greenhouse (Muenchow 1998).
- Cosexual fathers produced all cosexual progeny, while male fathers produced 1:1 sex ratios, suggesting a single sex-determining locus at which maleness is dominant (Muenchow 1998).
- The staminate flowers of male plants were less likely to be eaten by weevils, which ate a large proportion of the staminate flowers on cosexual plants (Muenchow 1998).

Saxifraga cernua; Saxifragaceae

- A fly-pollinated herb, but mostly reproduces asexually via bulbils that have replaced flowers in the leaf axils.
- Circumpolar, arctic or alpine, throughout the Boreal Arctic, common.
- Efficient vegetative reproduction via bulbils, very little cpDNA variation ( Bronken et al 2001)
- Molau 1992 and Molau & Prentice 1992 suggest that the species self-incompatible is androdioecious.
-but later work shows that it is not androdioecious (Brochmann & Hapnes 2001)
- Brochmann and Hapnes 2001 found that S. cernua is strongly protandrous, with the male phase lasting for 6-10 days. All plants eventually developed full, papillolse stigmas, and most flowers had a phenotypically hermaphroditic phase (gynoecia reached matureity before all anthers dehissed). This is probably why the species was thought to be androdioecious, with only rare hermaphrodites.
- Brochmann and Hapnes 2001 also found S. cernua to be self-compatible. Seed production was low with either self or outcross pollination, and was increased with outcross pollination, but seeds were produced from self pollination.

Schizopepon bryoniaefolius; Cucurbitaceae

- Grows at the foot of mountains in Japan (Akimoto et al. 1999)[abstract][html].
- Annual (Akimoto et al. 1999).
- Hermaphrodite flowers have three stamens and a pistil, while male flowers have only three stamens and lack a pistil (Akimoto et al. 1999 and Fukuhara & Akimoto 1999).
- Anthers and stigma in hermaphrodite flowers are very close to each other, and 8/16 bagged flowers set fruits autogamously with fully developed seeds (Akimoto et al. 1999).
- Six of 11 populations lacked males, while male frequencies were 5.5 to 28.3% in the other populations (Akimoto et al. 1999).
- The sex ratio did not depend on geography or population size (Akimoto et al. 1999).
- Hermaphrodite populations tended to have less genetic variation than the androdioecious ones because four of the hermaphroditic populations were entirely monomorphic (Akimoto et al. 1999).
- In androdioecious poulations, males were less homozygous than hermaphrodites (Akimoto et al. 1999).
- Populations with more males had lower inbreeding coefficients (F_IS) (Akimoto et al. 1999).
- Both male and hermaphroditic pollen stained blue with cotton-blue, and no morphological differences were observed under the electron microscope (Akimoto et al. 1999).
- The high frequency of monomorphic hermaphroditic populations may suggest a metapopulation structure with a high rate of colonization (and perhaps extinction).

Spinifex littoreus;Poaceae (=Gramineae)

- a wind-pollinated grass that lives on coastal sand dunes, with large vegetative colonies.
- this appears to be a case in which dioecy has been invaded by hermaphrodites (Connor 1996). - the species is dioecious in its continental distribution and on islands in its northern-most distribution, but hermaphrodites are present on oceanic islands. Connor (1996) interprets this as the breakdown of dioecy on oceanic islands (Connor 1996).
-Connor (1996) studied herbarium collections, but did not visit any populations in the wild, so it is difficult to know which sexes coexist. There were never more than 8 collections per island, and usually only one or two collections. These collections suggest that populations are either fully dioecious or fully hermaphroditic, with perhaps one trioecious population and no androdioecious populations. However, natural populations of this species should be investigated to determine if any populations are androdioecious.
- Even if no populations are androdioecious, studies of this species will help us to understand the invasion of hermaphrodites and breakdown of dioecy.
-Species is listed as androdioecious several floras, but the evidence used to reach the conclusion is unknown (Gilliland 1971 for Malaysia, Lazarides 1980 for tropical Southeast Asia, Monod de Froideville 1968 for Java and Telford 1993 for the Ashmore Reef, Timor Sea).
- According to Connor 1996:
- Pistillate flowers normally have sterile anthers.
- Male and female plants have structurally different inflorescences, a common character in dioecious peregrinating grasses.
- Hermaphrodite plants are structurally female. They have spikelets with pistillate florets as well as some spikelets with both perfect and staminate florets. Female spikelets have two florets, one neuter (with sterile anthers), and one pistillate (with sterile anthers). In the hermaphrodites, both florets produce stainable pollen.

Triops; Notostraca: Triopsidae

- (Sassaman 1991)

Ulmus minor; Ulmaceae

- studied in two Spanish populations by Lopez-Almansa et al. 2003.
- some individuals are primarily male, while others are hermaphroditic.
- phenotypic gender of plants was measured in two populations by estimating the number of seeds/flower and the number of stamen-bearing flowers per plant.
- the phenotypic gender of individuals was bimodally distributed, so even though most males were not pure males, they were distinct from hermaphrodites.
- there is no information about whether the dimorphism is genetic or if the plants are gender switchers. Phenotypic gender was constant across two years, but these are long-lived trees that don't flower until their 4th to 10th year of life (Lopez-Almansa 2002), so it would be necessary to observe individuals for much longer to determine if there is any gender switching. It would also be nice to know if there is any difference in the age, DBH or height of males and hermaphrodites.
- Sex ratios of natural populations are unknown because most natural populations have been destroyed by Dutch Elm Disease.
- There was no difference between males and hermaphrodites in number of flowers or total amount of pollen produced, and pollen was equally viable.
- The authors suggest that males may have a higher rate of clonal growth than hermaphrodites, and this may allow males to persist, and also allow them to have larger genets that will produce more pollen than hermaphrodite genets.
- The evidence that males have a higher clonal growth rate is that in a semi-natural population with a large amount of clonal growth, the male frequency is 0.75, whereas the experimental population without clonal growth has only 25% males. Both populations were presumably established without regard for the gender of trees planted.
- This is a long-lived, wind-pollinated, riparian species.
- The authors also suggest that the ancestral breeding system may be hermaphroditic (but provide no evidence, perhaps other elms are hermaphrodites).

Diana Wolf
dewolf@alaska.edu
907-474-5538