Apomixis in Plants pp. 72-75 (1992)
By Sven Asker, Lenn Jerling

  Origin of Maternal Haploids from Seeds
of Sexual Plants
1. Spontaneous
2. From polyembryonic seeds (twin plants)
3. After exceptional types of pollination
     Intergeneric, interspecific, or intervarietal pollination
     Delayed pollination
     Use of irradiated pollen
     Use of pollen inactivated with compounds like toluidine blue
4. After treatments designed to give polyploids
     Colchicine treatment
     Temperature shock
From Asker. S.. Hereditas. 93. 277. 1980. With permission.

Parthenogenetic development of reduced female gametes (always, as far known, in connection with fertilization of polar nuclei, or "haploid pseudogamy") was earlier called "nonrecurrent apomixis" (Darlington. cited from Gustafsson40). However, the latter term must also include occasional cases of diploparthenogenesis, which are treated in the next section. We prefer not to use the term apomixis in connection with occasional haploid parthenogenesis — it is misleading, in any case. See also comments in Chapter 13: "Apomixis and Plant Breeding". In exceptional cases, instead male nuclei develop parthenogenetically (androgenesis, Section III.C).

Haploid formation is promoted by various circumstances (Table I). Haploids are known to occur at increased frequency among twin plants, resulting from polyembryonic seeds. In this case, haploid embryos may develop from the synergids. Special kinds of pollination can induce parthenogenesis: wide crosses, delayed pollination, or use of pollen treated for instance with X-rays, ultraviolet light, or toluidine blue. Haploids are also sometimes obtained after treatments designed to give polyploids, i.e., colchicine treatment or temperature shocks.4 Experimental initiation of parthenogenesis by treatment with plant hormones and other chemicals has also been attempted, although the results are still inconclusive (Chapter 6, Section III.D). A strongly increased frequency of haploid parthenogenesis is displayed by alloplasmatic wheat lines (with wheat chromosomes in Aegilops cytoplasm51,86).

Thus, tendencies towards parthenogenesis, at least at the haploid level, seem to be common in sexual plants, although their expression is strongly dependent on the choice of pollinator, as well as probably by various developmental and environmental factors.

Monoploids are generally weak and sterile, whereas polyhaploids sometimes are vigorous, and may play a role in evolution, for instance in the case of diploid-tetraploid-dihaploid cycles (cf. Asker3). Important for breeding and experimental purposes is the production, by chromosome doubling of monohaploids, of completely homozygous ("autodiploid") lines.

Some Examples of Diploid Parthenogenesis
from Sexual Plants
Genus   Author
Fragaria   Millardet. 1894
Solanum   Jorgensen, 928
Zea   Randolph, 1932
Primula   Emit, 1951
Triticum   Tschermak-Seysenegg, 1951
Petunia   Katayama and Adachi, 1969
Nicotiana   Todua et al., 1973
Brassica   Eenink (several papers)
Ficus   Arendt and Kazas, 1977
From Asker, S., Hereditas, 93, 277, 1980. With permission.

Little is known about the gene background of occasional parthenogenesis. In maize, genotypic influences of both female and male parents on the frequency of maternal haploidy have been confirmed," and lines with high frequencies of monoploid formation have been isolated." Here a polygenic inheritance is indicated. Very few gene mutations conditioning parthenogenesis are known. In barley, however, plants homozygous for the gene hap give rise to 30 to 40% haploids in their offspring by parthenogenetic development of reduced egg cells.42 It should be emphasized that parthenogenetic development of reduced and unreduced egg cells is not necessary under similar genetical and environmental control. The genetic regulation of parthenogenesis in established apomicts remains a controversial question (Chapter 6).

Summing up, parthenogeneric development is a must for natural apomicts, but it is very doubtful if occasional haploid formation by parthenogenetic egg cell development in sexual plants should be looked upon as a "tendency towards apomixis".

6. Matromorphy: Occasional Gametophytic Apomixis in Sexuals

Even parthenogenetic development of unreduced egg cells is sometimes realized in sexual angiosperms. Especially in offspring from distant crosses (interspecific, or intergeneric), so-called matromorphic plants are sometimes obtained, even where self-pollination after incomplete castration, or contamination with foreign pollen, can be excluded. (According to Nogler,63 matromorphy is a linguistically incorrect synonym.) The oldest report concerns Millardet's56 "faux hybrides" in Fragaria.

The records of such cases of "occasional apomixis" are so many that it is impossible to account for all of them here. Even if all reports are not quite reliable, the occurrence of matromorphic plants is established without doubt. Examples of matromorphy in crops are given in Table 2, and further examples from crops are given in Chapter 13. Matromorphy, or occasional diploparthenogenesis, should not be confused with matrocliny, meaning formation of true hybrids which resemble the mother much more than the father.

The origin, occurrence, and frequency of matromorphic plants are largely unknown. They are probably not uncommon, but their presence in offspring after open pollination is likely to be overlooked. Matromorphic plants could be detected by the use of pollen carrying dominant markers.

In Brassica, matromorphy has been extensively studied by Eenink.24-30 Plants of maternal type could be derived by asexual seed formation (I) from EMC's with twice the somatic number (premeiotic endomitosis has been observed in Brassica according to Eenink30), giving rise after meiosis to "unreduced" macrospores and embryo sacs; (2) by parthenogenetic development of egg cells in embryo sacs formed as a result of restitution after first or second meiotic decision; (3) by parthenogenetic development of reduced egg cells followed by reduplication or fusion. The last possibility could give rise directly to homozygous diploids and is thus interesting from the breeders' point of view (Chapter 13). As far as we know, there is still no genetical and embryological proof of such an origin of maternal "diploids".

Asexually developing embryos were formed especially after so-called "prickle pollination". There were differences between lines concerning parthenogenetic ability (female) and parthenogenesis inducing ability (on the male side). Delayed (prickle) pollination and application of gibberellic acid increased the number of matromorphic seeds; obviously temperature also influences the seed set. Various evidence indicated that unreduced embryo sacs occurred.

The production of plants by matromorphy is under genetic control, as the frequency of such plants depends on the genetic constitution of both the female and the male parent.

Matromorphy is a mode of reproduction that seems to mimic natural gametophytic apomixis, but some differences remain. First, matromorphy seems to be a response to extreme conditions such as distant pollination, whereas during "normal" conditions probably only sexual reproduction occurs. Second, pollination seems always to be necessary for seed formation; if parthenogenetic egg cell development takes place, it is not accompanied by a similar development by the central nucleus, or polar nuclei. Third, the most important natural types of gametophytic apomixis, mitotic diplospory, and apospory seem not to be involved in matromorphic offspring production.

Probably the most common pathway, in Brassica as well as in other genera where matromorphy has been demonstrated, would be production of unreduced female gametophytes and gametes by formation of restitution nuclei after first or second meiotic division. Probable exceptions are Fragaria61 and Raphanobrassica,32 where occasional apospory seems to play a role.


Automixis, or automictic parthenogenesis, occurs in several parthenogenetic animal groups. It implies that normal female meiosis takes place. Later, however, the unreduced chromosome number is restored by fusion of two reduced nuclei, formation of a restitution nucleus, or by endomitosis.84

In higher plants, a similar mechanism is known to occur regularly only in one species. Rubus caesius,22,38  which is tetraploid and has earlier been judged to reproduce by gametophytic apomixis.

Only reduced embryo sacs and egg cells are formed. Both sexual and "parthenogenetic" reproduction occur. Pollination is necessary for endosperm formation. Sexual reproduction takes place by fusion between one egg nucleus and one or sometimes two sperm nuclei. Before parthenogenetic development, two nuclei fuse, which are produced by division of a haploid egg nucleus. Before this event, a pollen tube has entered a synergid.

Sometimes the chromosome number seems to be increased by additional fertilization or another chromosome doubling following fusion. Even "haploid-diploid" chimaeras may be formed. Thus, Rubus caesius seems to have a mode of reproduction quite different from the related blackberries.

Further investigations, especially genetical ones, are badly needed here. Similar conditions are liable to exist even in other plant species judged to be apomictic.

In maize, a kind of automixis has been described depending on homozygosity for a gene el, which causes elimination of the second meiotic division, followed by parthenogenetic egg cell development.90

Mogie57 defines automixis as a process whereby a new individual is formed from a product or products of a single meiotically dividing cell. Among these are both sexual processes, involving nuclear fusion, and asexual processes. The latter ones are the same as discussed by us under matromorphy, with the exception that first division restitution is excluded.


In androgenesis, or 'male parthenogenesis", the offspring has the genotype of the male gamete nucleus. After the male nucleus has entered the egg cell, the nucleus of the latter degenerates. The pollen nucleus divides in the cytoplasm of the egg cell, giving rise to a haploid embryo.

Androgenesis was first observed after a cross Nicotiana digluta X N. tabacum.18 By experimental androgenesis (in vitro proliferation of haploid pollen grains), haploid plants were produced from various Nicotiana species.12 A screening method has been worked out for detection of androgenetic haploids in this genus.65

After crosses between Begonia socotrana and B. X semperflorens-cultorum, Preil and Lorenz67 obtained "patromorphic" offspring which must have originated either from androgenesis or by elimination of the maternal chromosomes.

In maize, androgenetic haploid formation occurs spontaneously at a rate of about 1 X 10-5. Plants with the mutation "indeterminate gametophyte" produce androgenetic haploids at a high frequency.49 Androgenesis has also been studied in Hordeum.35

In Crepis, Gerassimova37 demonstrated androgenesis by X-ray treatment of plants homozygous for a dominant marker and pollination from untreated plants homozygous for the corresponding recessive genes. Ehrensberger31 obtained paternal haploids in Antirrhinum from crosses between irradiated egg cells X normal pollen. The reverse combination, normal egg cells X irradiated pollen, gave rise to maternal haploids. Haploids have been produced by androgenesis in apple." Such haploids are also produced from a Gossypium strain with hemigamy (next section).

The above-mentioned examples concern sexual plants. But androgenesis may also appear in apomictic plants, even here as an anomaly. Rao reported "male parthenogenesis" in "tetraploid Job's tears", Coix aquatica, which later turned out to have hemigamy. In diploid, facultatively apomictic Potentilla argentea, isozyme studies of population material indicated that certain plants could have arisen by androgenesis followed by chromosome doubling (Nordborg, unpublished).


This peculiar form of pseudogamy was first described in Rudbeckia.7 Here, the sperm nuclei sometimes do not fuse with the egg cell, and they begin to divide independently and simultaneously. As the embryo sac nuclei are unreduced, chimaeric embryos with both the somatic and half the somatic chromosome number arise. In Zephyranthes,79 male nuclei have been observed to give rise to diploid tissue after restitutional division. Chimeral plants are also formed by hemigamy in Coix aquatica.70 Hemigamy may also occur in Amelanchier.14

In a hemigamous strain of Gossypium barbadense, about 40% haploids were formed in the progeny.16 When female plants of this strain were crossed with nonhemigamous male parents, about 1% androgenetic hapboids appeared in the offspring. This technique allowed the deliberate production of haploids of different strains of cotton.

Gerlach-Cruse39 induced hemigamy in Arabidopsis by X-ray irradiation of floral parts, followed by pollination with untreated pollen. The occurrence of pseudogamy was indicated by genetic studies, using suitable markers.