Previous Page Table of Contents Next Page



All-Union Pond Fishery Institute
Moscow, U.S.S.R.


Gynogenesis is a special form of sexual reproduction in which insemination is necessary but the head of the sperm penetrating into the ovum does not transform into male pronucleus; and the gynogenetic embryo develops at the expense of the ovum nucleus only. Consequently the gynogenetic offspring are all females identical to the mother. Reproduction of gynogenetic forms takes place when gynogenetic females mate with males of the bisexual form of the same and related species.

Gynogenesis cannot be regarded as usual fecundation which is characterized by amphimixis. It is not identical with parthenogenesis either, though very close to it. Brachet (1917) wrote that gynogenesis was a bridge built by nature connecting fecundation with natural parthenogenesis.

Gynogenesis was first found in some species of free-living nematodes. It is now known to exist also in some other worms, insects, fish and amphibians. Most cases of natural gynogenesis were discovered in the past ten years. Natural gynogenesis occurs more often than believed, though compared to parthenogenesis it is extremely rare.

Two cases of natural gynogenesis are known in fish, namely in Carassius auratus (Cyprinidae) and in live-bearing Mollienesia formosa (Cyprinodontidae); of which the gynogenesis of C. auratus has been studied more thoroughly because of its importance in the artificial rearing of this species.


The geographical distribution of C. auratus extends from Japan and China in the East to the countries of West Europe (Poland, Romania, Germany). The waters in Japan, China, Korea and the adjacent countries are populated by the typical form of this species, C. auratus L., and in areas farther west by its subspecies, C. auratus gibelio Bloch. Detailed information on the systematic status and biology of C. auratus can be found in the works of Berg, (1949) and Nikolsky (1954).

In nature C. auratus gibelio (and evidently C. auratus L.) occurs in two forms, the usual bisexual form having females and males and the unisexual gynogenetic form consisting of females only. There appears to be a certain regularity in the distribution of unisexual and bisexual populations within the limits of natural distribution. Populations in the eastern part of the area are bisexual with a predominance of females, the percentage of males varying considerably. Equal proportion of sexes has been registered only in a few cases. Evidently the numerical predominance of females in bisexual populations results from the mixing of the unisexual and bisexual forms in these populations. The factors determining the percentage of males in these gynogenetic populations are still unknown.

Further to the west, bisexual populations give way to purely female populations. Males are found in these populations as exceptions and do not comprise more than 1–3 percent. Such populations are common in West Siberia, the Urals and the European part of the U.S.S.R. In the countries of West Europe are found populations of both unisexual and bisexual types, bisexual being rarer.

In mixed populations of C. auratus gibelio, gynogenetic females mate with males of the bisexual form of their own species, and in unisexual populations with males of related species. In pond fisheries, offspring of gynogenetic females are usually obtained with the help of young male carp.

Unisexuality of C. auratus gibelio was observed by I.A. Anichenko in 1939–1940 (Golovinskaya and Romashov, 1947); but an explanation of this phenomenon was obtained only after the discovery of natural gynogenesis in this species by Romashov and Golovinskaya who obtained experimental evidence of this type of reproduction by detailed genetic analyses of females of the unisexual form (Krushinsky, 1946; Golovinskaya and Romashov 1947). These were immediately followed by cytological research into the processes of fertilization of the unisexual form of C. auratus gibelio, which confirmed the earlier findings (Golovinskaya, 1954).

2.1 Genetic analysis of the bisexual form of C. auratus gibelio.

Investigation of the bisexual form of C. auratus gibelio was carried out with the population of the Volma fisheries (in the Byelorussian Soviet Socialist Republic), which was brought from the Amur in 1948 (Golovinskaya, 1960; Golovinskaya, et al., 1965; Tcherfas 1966b). It was found that specimens of the bisexual form of C. auratus gibelio, when mated with each other, yielded equal numbers of female and male offspring.

The diploid number of chromosomes of the bisexual form was found to be 94 (by the Japanese scientist Makino in 1939). Maturing of bisexual females is accompanied by the usual meiosis. In the last prophase of the first meiotic division all chromosomes are represented by bivalents. The first division is reducing, and takes place shortly before ovulation ends with discharge of the polar body. As with all fishes studied, the ovum of the bisexual C. auratus gibelio is in the metaphase of the second meiotic division after ovulation and contains the haploid number of 47 chromosomes. Shortly after insemination the second (equational) division comes to an end and the second polar body detaches itself. The remaining female telophasic group and the sperm head turn into female and male pronuclei. When preparing for the first division of segmentation both pronuclei draw closer to each other. This process ends with the formation of the diploid metaphasic plate of the first division of segmentation.

2.2 Genetic analysis of unisexual forms of C. auratus gibelio

Genetic data testifying to natural gynogenesis of the unisexual form of Crucian carp were obtained by the analysis of progeny of females of unisexual populations and males of other species. Table I cites the relevant data of all experimental crossings known. In all the crossings tabulated the progeny consisted entirely of females of C. auratus gibelio.

The progeny obtained by crossing specimens under serial numbers 1–10 of Table I were subjected to a thorough biometric analysis which included the main systematic features; namely, the number of scales along the lateral line, the number of branchiostegal rays on the first gill arch, the number of rays in the dorsal and anal fins, the number of denticles on the last rigid ray of the dorsal fin, the formula of gular teeth and colouring of the belly. It was found that there is no difference between offspring (C. auratus gibelio) of different males. These offspring, while being raised in ponds, proved that they are not different in such characteristics as rate of growth, survival, and maturity cycle.

Males of leathery and linear carp bearing the dominant gene N were used in crossings 2 and 3 (Table I). It is worth recalling that the gene N in heterozygous state has a pleiotropic effect on a number of characters. It causes the reduction of scaly integument, dorsal and anal fins and gular teeth (Kirpichnikov, 1937; Golovinskaya, 1946). In crosses of leathery and linear carp with C. carassius, the reduction effect of the gene in hybrids of F1 is quite distinct (Kuzema). Thus gene N is a suitable genetic marker. In offspring obtained by crossings of unisexual females with leathery and linear males of Cyprinus carpio no signs of paternal heredity manifestation of N gene were observed. The most demonstrative evidence of the absence of paternal heredity was obtained in crossings 4 and 5 (Table I).

Table I

Crossings that gave gynogenetic progeny of unisexual females of C. auratus gibelio (silver Crucian carp)

NumberMalesComposition of offspringInvestigators
   1Wild carp, Cyprinus carpio L. ()Silver Crucian carp, females onlyGolovinskaya, Romashov, 1947
   2Linear carp, C. carpio  L. ()""
   3Leather carp, C. carpio  L. ()
   4Carp, C. carpio L. radiation treatment 10kr"Golovinskaya, Romashov, Tcherfas, 1965
   5Carp C. carpio L. radiation treatment 100kr"
 16Silver Crucian carp bisexual form Carassius auratus gibelio"Golovinskaya, 1960
 17Silver Crucian carp unisexual form C. auratus gibelio Bl."Golovinskaya, 1954
   8Goldfish C. auratus L."Golovinskaya, Romashov, 1947
  9Golden Crucian carp, Carassius carassius L.
10Tench, Tinca tinca L.""
11Steed, Hemibarbus labeo Pall"Kryzhanovsky, 1947
12Roach, Rutilus rutilus L."Golovinskaya, Romashov, 1947
13Loach, Cobitidae: Misgurnus fossilis L.Golovinskaya, Romashov, Tcherfas, 1965
14Rainbow trout, Salmonidae: Salmo gairdneri Rich.Silver Crucian carp, females onlyTcherfas, 1968 

1 Other common names - Prussian carp, European goldfish.

In earlier experiments with C. carpio it was observed that 100 kr and 10 kr irradiation of sperm produces a most grievous effect on their nuclear apparatus while 100 kr irradiation results in complete inactivation of the nuclear apparatus (Romashov, et al., 1960). Therefore, when the eggs of female C. carpio were inseminated with sperm treated with 100 kr irradiation the whole progeny consisted of unviable gynogenetic haploids, not counting individual gynogenetic diploids resulting from spontaneous diploidization of the female chromosome complex. With 10 kr irradiation the effect was lower and the nuclear apparatus of the sperm took part in the development. However, as a result of the fact that the zygote included severely damaged male chromosomes, the whole progeny proved to be unviable, the possibility of individual normal specimens due to diploid gynogenesis, as in the case of 100 kr irradiation, being completely excluded. Similar results were obtained by inseminating the eggs of female C. auratus gibelio (bisexual form) with the irradiated sperm of C. carpio (Golovinskaya, et al., 1965). Inseminating the eggs of unisexual females with the irradiated sperm of C. carpio gave an absolutely normal progeny as in all other crossings (Table I). Consequently, irradiation of sperm, which is lethal in usual reproduction, produces no grievous effect in natural gynogenesis. Evidently, the state of the sperm nucleus (from the viewpoint of its genetic full-bodiedness) is of no consequence in the multiplication of females of the unisexual form.

The above is fully confirmed by the ability of unisexual females to mate with their own males from unisexual populations (crossing 7, Table I). As stated earlier, such males are very rare. Unlike normal males from bisexual populations they are characterized by a lower fecundity and genetic sterility. Evidence of the genetic deficiency of such a male (from the unisexual population of the Yakot fishery of the Moscow region) was obtained by crossing it with a female C. carpio. In spite of the fact that the eggs were of high quality (which was checked in control crossing of the same female with a normal male C. carpio) all hybrid offspring proved to be absolutely unviable, revealing the genetic deficiency of this male. However, for a few years this male gave quite normal gynogenetic progeny when crossed with females of its own kind in natural spawning in the pond and in artificial insemination. (Golovinskaya, et al., 1965).

2.3 Cytological analysis of natural gynogenesis of C. auratus gibelio

The exclusion of the male nucleus in the reproduction of the unisexual form of C. auratus gibelio, which was quite clearly proved by experimental crossings, was also confirmed by direct cytological observations of the sperm nucleus in the plasma of the ova. The first cytological analysis of the process of fecundation of the unisexual form of C. auratus gibelio was carried out by K.A. Golovinskaya in 1954. Later these results were confirmed in experiments with females of the unisexual form from many populations (Lieder, 1955, 1959; Statova, 1963). Observations showed that the centrosome brought in by the sperm doubles, and each of the two centrosomes forms a pole of the spindle of the first division of segmentation. However, during the whole period of preparation for the first division of segmentation the male nuclear apparatus remains as a dense chromatin formation and does not turn into male pronucleus as in usual fecundation. In the first minutes after insemination, the head of the sperm penetrates into the plasma and gradually comes closer to the female pronucleus. After that it can be detected in form in the zone of one of the poles of the spindle of the first division of segmentation, and later in one of the blastomeres. Lieder managed to trace the head of sperm up to the stage of eight blastomeres (Lieder, 1959). The further destiny of the male chromatin remains unknown. Evidently it is absorbed by the plasma of the ovum.

By analyzing the maturing process of the gynogenetic females in the population of the Volma fishery it was possible to ascertain the cytological mechanism securing the permanent number of chromosomes of the unisexual forms of C. auratus gibelio (Tcherfas, 1966 a and b).

Chromosomes in the somatic and sex cells of the Volma unisexual females were found to be 141, triploid set of the basic number, 47. Triploidy has also been confirmed by indirect methods, by determining ploidy by the number of nucleoli in the nuclei of the epithelial cells of the fin border of one-day larvae and by cytometric data (by the size of cells of some tissues). At present all these indices are widely used for determining ploidy without directly counting the number of chromosomes. Cytometrical investigations have shown that the size of erythrocytes and nuclei of triploid females is on the average 1.4 times greater than of diploid bisexual females.

The process of maturation of unisexual females differs considerably from the usual meiotic process, which is evidently closely connected with triploidy. In the late prophase of the first meiotic division all chromosomes are univalent and no conjugation of chromosomes is observed. In the last period of maturation, which ends with ovulation, it is possible to single out several main stages.

  1. Stage 1 - concentration of univalents in a limited area of the cytoplasm of the animal pole and formation of an achromatic, multi-pole figure in the zone of chromosomes. This stage is observed immediately after the nuclear capsule is broken and corresponds to prometaphase 1.

  2. Stage 2 - formation of a three-pole spindle and distribution of chromosomes in three groups at the poles.

  3. Stage 3 - transformation of the three-pole spindle into a bipolar spindle. At this stage, conventionally corresponding to anaphase 1, chromosomes either distribute themselves into two equal groups or in proportion of 1:2, forming haploid and diploid groups.

The above process is abortive and results in no reduction. It ends in uniting all univalents into the triploid metaphasic plate of the only equational division of maturation. It is at this stage that the egg undergoes ovulation. Soon after insemination the division of the polar body takes place. The female triploid complex that remains in the ovum plasma gradually turns into a triploid pronucleus. Further, the process results in forming the triploid plate of the first division of segmentation.

Thus the stable number of chromosomes of the unisexual form of C. auratus gibelio of the Volma population is secured by the exclusion of the reduction division. Due to the exclusion of crossing-over and reduction which during the usual process of meiosis cause a genetic recombination, the progeny of each gynogenetic female from the Volma fishery is a genotypically uniform clone.

Females from the unisexual population of the Yakot fishery (the Moscow region) have proved to be triploid with ameiotic maturation. It may be presumed that the cytological peculiarities of Volma gynogenetic females are characteristic of other populations of the unisexual form of C. auratus gibelio.

Experiments to obtain unisexual C. auratus gibelio by parthenogenesis gave no positive results. Unfertilized eggs when submerged in water began irregular divisions and soon perished. In order to develop normally the eggs of unisexual females should be inseminated. No case of parthenogenesis of these fish has been observed. Suspected cases of parthenogenesis of some species of fish require reappraisal.

A comparison of the cytology of gynogenesis in C. auratus gibelio with data on parthenogenesis of other animals shows that natural gynogenesis of C. auratus gibelio is a typical case of triploid apomixis. It should be noted that polyploidy and, first of all, triploidy is quite a common phenomenon under natural parthenogenesis. As in other animals, triploidy of C. auratus gibelio is accompanied by the loss of ability for usual sexual reproduction (Baranov and Astaurov, 1956; Astaurov, 1965).

The origin of triploidy in gynogenetic females of C. auratus gibelio is still not clear. It can be presumed that initially the species had a diploid gynogenetic form, and triploidy came about later as a result of crossing diploid gynogenetic females which produce unreduced gametes with males of the bisexual form of their own species or related species. This assumption is based on the results of the well-known model experiments on the silkworm (Astaurov, 1955; 1956) and of some cases of natural triploidy (Daravsky, 1958; 1962). The assumption is borne out also by the features of the maturation process of unisexual females of C. auratus gibelio. The distribution pattern of univalents in haploid and diploid groups during the maturation process of the ovum may be regarded as a result of their cytological and genetic heterogeneity. Therefore, it is reasonable to presume that the genotype of gynogenetic females is of hybrid allopolyploid nature.

When studying the fecundation process of bisexual females of C. auratus gibelio, some instances of spontaneous diploidization of the female chromosome complex, caused by the return of the second polar body into the plasma of the ovum, were observed. This process resulted in spontaneous emergence of triploids in the progeny of bisexual females. Thus there is a second channel of triploidy of C. auratus gibelio which is not connected with hybridization, namely, autotriploidy.

It is comparatively easy to distinguish females of the unisexual and bisexual forms of C. auratus gibelio. This is of importance in investigations on mixed populations of the species. Table II cites generalized comparative results of some crossings of Volma females of the two forms. Distant crossings (1) and crossings with the use of irradiated sperm (2,3) were most suitable for distinguishing the two forms. In both cases the difference between the females is quite distinct during the process of embryogenesis; and, after hatching, by malformation in the progeny of the females of the bisexual form. Identification of females by the size of erythrocytes is also possible.


Another case of natural gynogenesis among fishes is that of the live-bearing fish, M. formosa, belonging to Poeciliidae (Hubbs and Hubbs, 1932). In nature this species is represented only by a gynogenetic form. In natural conditions, females of M. formosa live together with two related species M. sphenops and M. latipinna, and propagate with the help of males of these species. In morphological characters, females of M. formosa occupy an intermediate position between M. sphenops and M. latipinna. This led to the belief, in spite of evidence of gynogenesis, that M. formosa was a hybrid form.

Like the gynogenetic females of C. auratus gibelio, females of M. formosa also cannot propagate by parthenogenesis. When crossed with males of other species they give only female progeny of M. formosa.

During the entire period of investigations on natural populations of M. formosa, only three males were found. Phenotypically they resembled F1 males obtained by crossing M. sphenops and M. latipinna with females of M. formosa and masculinized by methyltestosteron. It was assumed that emergence of males may result from three types of development: phenotypical sex transformation of females; disturbances in their maturation; and introgression of paternal chromatin. No detailed investigation of these males was carried out (Hubbs et al., 1959; Haskins et al., 1960).

Experimental crossings over a period of 14 years from 1932 to 1946 yielded 8 000 gynogenetic specimens of 20 generations. Males of 50 different species, subspecies and races of Mollienesia and males of five other species belonging to Cyprinodontidae were used for crossing (Hubbs and Hubbs, 1946).

Much information on the nature of gynogenesis of M. formosa was obtained as a result of studies in the laboratory, and observations of natural population of the fish, using the method of transplantation of tissue (Kallman, 1962 a and b; Darnell et al., 1967). As known, compatibility of tissue is possible when the donor and the recipient are of related genetic constitution. In experiments with M. formosa the transplants were fins, spleen and heart. Transplantation was considered successful if no detachment of the transplants was observed during the whole period of the experiment. The degree of incompatibility of tissue, i.e. the degree of genetic relationship of the donor and the recipient, was determined by the time interval between transplantation and detachment.

When investigating laboratory populations, two types of transplantation were carried out: (1) transplantation of tissue between parents and their offspring, i.e. between P and F1 (la-P♀ and F1; lb-P♂ and F1); and (2) between the progeny of the first and following generations obtained from the initial parent pair. In crossings, males of M. sphenops were used. Transplantations of 1a and 2 were always successful. In transplantations of 1b, detachment of the transplants was observed. The results allowed the following conclusions:

  1. Gynogenetic progeny of M. formosa do not inherit any antigens from the father.

  2. Progeny obtained from a female is a clone genetically identical with the mother form and with one another.

The method of transplantation of tissue was used for investigating the clone structure of some natural populations. In these experiments the natural populations of M. formosa were found to include different clones. Kallman (1962b) assumed that different clones in gynogenetic populations may be regarded as the result of the gradual effect of mutation.

Cytological observations of natural gynogenesis of M. formosa show that gynogenetic females of M. formosa are diploid, the number of chromosomes being 45–46 (Haskins et al., 1960) which corresponds with the diploid number determined for related M. sphenops, M. latipinna and M. selifera. The diploidy of female M. formosa is also confirmed by the size of cells and their nuclei in fin regenerates and the amount of DNA in somatic cells of M. formosa, M. sphenops and M. latipinna (Meyer, 1938; Rasch et al., 1965).

The cytological pattern of maturation of ova of female M. formosa has not been studied. However, the results of transplantation experiments, that allowed the clone structure of gynogenetic populations of the species to be determined, show that the maturation process of M. formosa, as well as that of C. auratus gibelio, is of the ameiotic type.

Some crossings of females of M. formosa with males of M. sphenops and M. vittata (in these cases genetic markers were used) gave hybrids (Kallman, 1964; Haskins et al., 1960). F1 and F2 were constituted by females and males having characters of both parents. When mature they produced no progeny.

It was observed that hybrids accepted transplants from donors belonging to the mother clone, whereas graftings from the father to hybrids were not successful (Kallman, 1964). Kallman therefore observed that all tissue antigens of the hybrids were characteristic of the mother form. Regarding these hybrids as amphimictic diploids and trying to explain the mechanism of inheritance, Kallman came to the wrong conclusion, that females of M. formosa were entirely homozygous. Also, as a logical proceeding from these calculations was another wrong conclusion about restitution of the diploidy of gynogenetic females of M. formosa in the first division of segmentation. His conclusions contradicted the data on clone multiplication of female M. formosa. By cytophotometric analysis, other investigators (Rasch et al., 1965) have shown that these hybrids are triploids, with the diploid set of chromosomes from the mother and the haploid set from the father. Study of spermatogenesis of hybrid males from F1 has shown that their sterility is connected with the irregular distribution of chromosomes in meiosis. This explains the results of crossing.


It is necessary to emphasize that triploidy is characteristic of gynogenetic females of C. auratus gibelio and diploidy of M. formosa. This accounts for the difference in the nature of gynogenesis in the two species. Gynogenesis of M. formosa is evidently a result of mass hybridization between two related species (M. sphenops and M. latipinna) and the subsequent effect of natural selection. The origin of gynogenesis of C. auratus gibelio is not so clear. We assume that gynogenesis of C. auratus gibelio is of hybrid nature, and the emergence of triploidy may have been preceded by diploid gynogenetic forms.

In the case of autotriploidy, the possibility of spontaneous emergence of triploids in bisexual populations and their further destiny is of great interest. However, mention should be made of the well-known thesis that gynogenesis under any condition is determined by the presence in the genetic pool of populations of initial species of genes that control individual stages of gynogenetic propagation.

Studies of some general features of natural gynogenesis of M. formosa and C. auratus gibelio reveals the evolutionary importance of this method of reproduction of fish.

Change over from the usual bisexual multiplication to parthenogenesis has two main advantages: (i) Possibility of somatization during meiosis, resulting in preservation of the initial valuable genotype in a number of generations; (ii) rise in effectiveness of reproduction. Both these are true in natural gynogenesis of fish. The assumption that genetic advantages of unisexual lines of C. auratus gibelio may be connected with their hybrid constant heterozygous nature and heterosis (which are inherited in a number of generations due to gynogenesis) was made in the first study of natural gynogenesis of this species (Golovinskaya and Romashov 1947). Cytological studies have shown that there is no crossing-over and reduction during the process of maturation of ova of gynogenetic females of C. auratus gibelio; and this supports the above assumption. The above is true for females of M. formosa; clone reproduction of this species has been proved with the help of tissue transplantations.

Increasing effectiveness of reproduction under parthenogenesis is connected firstly with the doubling of the rate of reproduction due to wholly female structure of populations, and secondly with the removal of the loss of time necessary for meeting of two specimens (a male and a female) under usual reproduction. The latter is especially important for animals with low mobility (lizards, for instance) and when the imaginal stage is short (as in the case of some insects).

For gynogenesis fecundation is necessary, and therefore the main factor determining the numerical strength of gynogenetic populations is the number of males able to ensure the reproduction of gynogenetic forms.

Study of natural gynogenesis of different groups of animals has shown that usually reproduction of gynogenetic females depends greatly on males of the bisexual form of the same or related species. This factor limits their population and dispersion, thus depriving them of the main advantages of parthenogenesis. Natural gynogenesis of fish is peculiar in this respect. The results of numerous crossings of unisexual females of C. auratus gibelio and M. formosa have shown their ability to multiply with the participation of males of various species of fish. In experimental conditions, C. auratus gibelio gave progeny when crossed with males of ten different species representing three families. A large variety of species gave gynogenetic progeny of M. formosa It is therefore possible to conclude that in experimental conditions (under artificial fertilization) fish have no limitations in producing gynogenetic progeny by males of different species. Under natural reproduction it is necessary that the ecology of reproduction of gynogenetic females and of the bisexual species living with them should be adequate. It is also necessary that the numerical strength of the latter should be great. We have evidence of abrupt decrease in the numerical strength of gynogenetic populations of C. auratus gibelio when these conditions were not manifest.

In spite of the limiting effect of these factors, because of their ability to propagate with males of many species, natural gynogenesis of fish leads to increase in the numerical strength of the species, which is characteristic of natural parthenogenesis of other animals. The marked predominance of the unisexual form of C. auratus gibelio within the area of its natural distribution may be attributed to the above considerations.

Further investigations of the origin of triploidy of C. auratus gibelio and cytological analysis of natural populations connected with them, as well as research programmes of experimental triploidy would help clearer understanding of the phenomenon of natural gynogenesis of this fish. It is also necessary to study factors regulating the numerical strength of the unisexual and bisexual forms of C. auratus gibelio in mixed populations of this species and analyse the possibility of change over from one manner of reproduction to another. It is necessary to investigate the cytology of gynogenesis of female M. formosa.

No investigations have been carried out to determine the function of sperm under natural gynogenesis of fish or other animals. The assumption that the inactivated nucleus of sperm has no effect on the characteristics of the gynogenetic progeny is based only on morphological analysis. Experiments in transplantation of tissue were a considerable step forward in this respect. For C. auratus gibelio this problem is of great practical importance, as males of different species are used in commercial raising of the unisexual form of this fish. There is no information available on the nature of the biological inactivator causing inactivation of the male nucleus under gynogenesis. These and many other problems need further investigation.

Spontaneous emergence of gynogenetic offspring in natural conditions was observed under distant hybridization. Hybrid gynogenesis is based on the incompatibility of plasma and chromosomes of the crossing species. Chromosomes brought in by sperm get eliminated to some extent in the primary phase of development of hybrid embryos. Most of these hybrid offspring are haploids and mosaics with grave malformations. Their development is usually more regular when they resemble the mother type. Individual viable specimens emerging among hybrid offspring fully repeat the mother form. Their emergence may be accounted for by complete elimination of male chromosomes and diploidization of the female chromosome complex.

Hybrid gynogenesis of fish is described in the experiments in distant hybridization carried out by Nikoljukin (1952) and Kryzhanovskii (1947; 1956) and Kryzhanovskii et al., 1953.

Hybrid gynogenesis proves that in natural conditions mass hybridization facilitates the detection of specimens with hereditary tendency to gynogenesis.


Two unrelated phenomena form the basis of artificial diploid gynogenesis of fish. These are insemination of the ova with genetically inactivated sperm, and diploidization of the female chromosome complex.

Genetic inactivation of the male nucleus can be achieved by different means but ionizing radiation is most often used for this purpose. The quite distinct differential effect of this agent on the nuclear apparatus and cytoplasmatic components of the sex cell allows the sperm to be treated with very high doses of irradiation. When treated with high doses of irradiation, male chromosomes become genetically inactive, but this does not destroy the sperm's ability to fertilize the centrosome functioning normally to ensure the regularity of the process of the first divisions of segmentation. Progeny obtained in such crossings consist not only of unviable haploids but individual gynogenetic diploids, as the result of spontaneous diploidization of the female chromosome complex. This phenomenon has been observed in the natural gynogenesis of C. auratus gibelio as well.

The phenomenon of irradiative diploid gynogenesis is known as the Hertwig effect, named after the German scientist who first described it. The essence of the Hertwig effect is that when the normal ovum is inseminated with irradiated sperm, the affection of the embryo increases only when the dose is increased to a certain limit. When this limit is achieved, the affection decreases noticeably, and further increase in the dose of irradiation (up to the limit under which sperms still preserve their ability to fertilize) does not cause further affection of the embryo. Treating with high doses of irradiation produces malformed, unviable offspring with a very few normal individuals. This, paradoxical as it may seem, can be explained easily. The effect of irradiation on the embryo increases to a stage where the nucleus of the sperm still retains (if partly) its genetic functions and is able to take part in development. When the dose of irradiation is brought to the limit which causes complete genetic inactivity of the male nucleus, haploid and, in individual cases, diploid radiative gynogenesis takes place.

5.1 Experiments in radiative gynogenesis

Radiative gynogenesis of fish was observed for the first time in experiments with trout (Opperman, 1953) and much later with Misgurnus fossilis L. (Neifach, 1959). Special study of radiative diploid gynogenesis of fish was carried out recently in the Soviet Union. Species under investigation were: M. fossilis L., C. carpio L. and some species of Acipenseridae. Eggs and milt in all the experiments were obtained with the help of hypophysial injections. Sperm subject to irradiation was kept in test-tubes under low temperature (4–8°C). Irradiation of the sperm was carried out by means of usual X-ray units. The inseminated eggs were incubated in petri dishes in the laboratory.

When inseminating the eggs with sperm treated with irradiation doses varying from 100 kr to 600–800 kr, it was observed that complete genetic inactivity of the male nuclear apparatus occurs when the sperm is treated with 100–200 kr irradiation doses. These doses were therefore used in all the experiments. Evidence of genetic inactivity (when treating with 100–200 kr irradiation doses) was obtained through the changes in the curve of waste, general lessening of the affection of embryos and distinct haploid syndrome of the offspring, and in some cases, the emergence of individual normal larvae.

In experiments without any special inactivators, diploid gynogenetic offspring are very rare. On the average, they comprise 0.33 percent of the total fertilized eggs, for M. fossilis, about 2 percent for Acipenseridae, and 3 percent for C. carpio (Golovinskaya et al., 1963; Romashov and Belayeva, 1965). The small number of diploid gynogenetic offspring is evidence of the rarity of spontaneous diploidization of the female chromosome complex. Besides, it was found that individual females of C. carpio, and more especially of M. fossilis, vary greatly in their ability to produce gynogenetic diploids.

5.2 Genetic evidence of elimination of male inheritance

Elimination of male inheritance was proved by morphological analysis of gynogenetic offspring in the following crossings:

Acipenser ruthenus × Huso huso(100 kr)
H. huso × A. ruthenus"
M. fossilis × C. carassius"
M. fossilis × C. auratus"
C. carpio × C. carassius"

In a series of experiments conducted, a few normal offsprings of the maternal types were obtained in all cases. In control crossings, with insemination with non-irradiated sperm, the typical hybrid progeny that was obtained had morphological characters of both parents (Romashov et al., 1963; Golovinskaya, et al., 1963).

In experiments on radiative gynogenesis of C. carpio, marker genes were used as in the experiments with C. auratus, and the gynogenetic progeny were analyzed by the marker genes of the scale.

The type of the scaly integument of C. carpio is determined by two pairs of nonallelomorphic genes and (Kirpichnikov, 1937; Golovinskaya, 1946). The four known genotypes have the following formulae:

The gene N in homozygous state is lethal; therefore specimens and are unviable. Wild carp is a homozygous form with dominant gene S and recessive gene n:

Exclusion of father inheritance in gynogenesis of carp was proved by the emergence of scattered offspring in F1 when scattered female () was crossed with wild carp. In control crossing, all the progeny in F1 were scaly.

5.3 Cytological analysis of radiative diploid gynogenesis

As stated before, the emergence of externally normal and fully viable offspring, when sperm are treated with irradiation, is possible only if they are diploid. Cytologically, diploidy was confirmed by the number of nucleoli in the nuclei of epithelial cells of the fin border of one-day old larvae, which had the number of nucleoli equal to that of the maternal form and approximately twice as high as that of the unviable haploids. Similar analysis was carried out on loach, carp and sterlet (Golovinskaya et al., 1963; Romashov et al., 1963; Romashov and Belayeva, 1965).

The investigation of the fertilization process under radiative gynogenesis carried out on loach has explained the mechanics of diploidization of the female chromosome complex and revealed some peculiarities in the behaviour of the irradiated male nucleus (Romashov and Belayeva, 1964).

After ovulation, the ovum of fish is in metaphase 2 and contains the haploid number of chromosomes. During the first minutes after insemination, anaphase 2 occurs; and a little later the second polar body emerges. Cytological analysis has shown that irradiating the sperm does not affect the regularity of the stages of the second meiotic division and subsequent transformation of the female haploid complex into female pronucleus. In the case of loach, about an hour after insemination (at a temperature of 18–19°C) female chromosomes form the metaphasic plate of the first division of segmentation, and the embryo begins its haploid development.

In experiments where the percentage of gynogenetic diploids was high, it was possible to investigate the mechanics of diploidization of the female chromosome complex. It was observed that in diploid radiative gynogenesis, the second meiotic division also proceeds regularly and ends in the formation of the second polar body. However, without leaving the surface of the ovum, the polar body is gradually absorbed by the ovoplasma and forms the second female pronucleus. Both pronuclei take part in the formation of the zygote and thus secure the diploidy of the embryo.

Observations of the cytological behaviour of the head of the irradiated sperm during the first few minutes after insemination have proved the exclusion of the male heredity material in radiative gynogenesis. At the same time it was observed that the behaviour of the inactivated male nucleus is different under artificial and natural gynogenesis. Cytological analysis has shown that irradiated sperm preserves the ability to turn into male pronucleus in the first few minutes after insemination. Therefore in radiative diploid gynogenesis, at a certain stage of development, the ovum contains three pronuclei (two female and one male); and in haploid gynogenesis, two pronuclei (one male and one female). Pycnotization of the male nucleus takes place later at the end of the prophase of the first division of segmentation. The pycnotic nucleus of the sperm gets into the central zone of the spindle, in the way of anaphasic groups of female chromosomes diverging to the poles, affecting the regularity of the procedure of the first divisions of segmentation.

In natural gynogenesis of C. auratus the formation of male pronucleus has never been observed, and during the segmentation divisions the head of the sperm is always in the zone of one of the poles, apart from the female chromosomes.

The analysis of the mechanics of diploidization under radiative gynogenesis has proved that, if the female is heterozygous, segregation among diploid gynogenetic offspring is inevitable. It is connected with random disjunction of homologues in the first meiotic division. With such a mechanism if diploidization complete homozygosity of gynogenetic diploids may be expected as a result of unification of two identical sets of chromosomes.

Experiments on carp have absolutely confirmed segregation in the gynogenetic progeny. Gynogenetic progeny was obtained from female carp of three genotypes: scattered ; scaly heterozygous () and linear heterozygous by two pairs of genes (). As a result of random disjunction of homologues in the first meiotic division, haploid sets of ova could be represented after the second meiotic division by the following combinations of genes:

female scattered-s n
female scaly-S n, s n
female linear-S n, s n, SN, s N

Evidently, as a result of diploidization, the following compositions of the progeny could be expected:

from the scattered female: - only scattered
from the scaly female - scattered
and scaly
from the linear females - scattered
and scaly
Doubling haploid sets with the gene N gives lethal combinations and

In experiments conducted four scattered females produced 39 scattered offspring. Five scaly, heterozygous females gave 30 scaly and 11 scattered fingerlings. Thus segregation of gynogenetic offspring was ascertained with accuracy. As gynogenetic progeny proved to be extremely small in number it may be assumed that the numerical ratio of scaly and scattered specimens in these experiments was accidental.

The composition of the gynogenetic progeny of the linear female proved to be unexpected. Three linear females gave 20 gynogenetic diploids: 13 linear and 7 leathery, all of which were sure to be heterozygous through the gene N (and consequently through many other genes).

The question is: how could the unification of the second polar body and the female telophasic group, that parted in the second equational mitotic division of meiosis, result in the emergence of a heterozygote? A few possible explanations of this phenomenon are given by Golovinskaya and Romashov (1966). The most probable cause of this phenomenon is crossing-over of two non-sister chromatids of the bivalent, which took place in the first meiotic division. In this case, in the locus in which the exchange takes place, each of the homologues becomes heterozygous; and instead of the initial NN and nn, both homologues become N n. In the case of divergence of homologues in the first division of meiosis, the same groups of N n and N n are segregated; in the second, the chromatids with the loci N and n are segregated. In the case of diploid radiative gynogenesis, loci N and n unite again and the restoration of the heterozygous state N n occurs. It is quite obvious that such a recombination concerns all loci under crossing-over. The heterozygosity of gene S was proved after crossing linear females of gynogenetic origin with a scattered male. If the gynogenetic linear specimens were homozygous as to gene S, the progeny obtained by crossing with the scattered male would have been scaly and linear. However, this crossing gave carp of four genotypes, which is evidence of gene S heterozygosity in gynogenetic females.

Data obtained have disproved the earlier assumption that gynogenetic progeny obtained by radiative gynogenesis were fully homozygous. It has been observed that homozygosis is confined to crossing-over that occurs in the first meiotic division.

5.4 Characteristics of gynogenetic progeny

Most complete data on the characteristics of gynogenetic progeny were obtained from carp reared up to the maturation period. All the specimens reared turned out to be females. They were normal in their external appearance but slow in growth. Six out of eight gynogenetic females when crossed with male carp gave normal progeny.

Gynogenetic progeny of Acipenseridae were few, and it was possible to rear the larvae up to the period when they began active feeding. In this period their viability was lower than in the control experiment.

It is possible that the low rate of growth of gynogenetic carp and the low viability of sturgeons are the result of their high homozygosity. Data on the normal fecundity of the gynogenetic progeny have proved to be of great value for selection.

5.5 Methods of increasing the numerical strength of diploid gynogenetic progeny

As stated earlier, the level of spontaneous diploidization of the female chromosome complex is very low and diploid gynogenetic offspring are very rare. To increase the frequency of diploidization of the female chromosome complex, the method of temperature shock is used. At present this method is used in experiments in ploidy of animals that represent quite different systematic groups, most often of fish and amphibians. Treating the eggs with low and high temperatures in the period of meiotic divisions results in different disturbances in the meiotic process. Diploidization of the female chromosome complex may be caused by the disintegration of the spindle, due to which none of the chromosome sets can form the polar body, or by the return of the polar body into the plasma of the ovum (Rott, 1965; Betina and Rott, 1966; Romashov and Belayeva, 1965). When fertilized in the usual way, such unreduced ova develop into triploid specimens; and when inseminated with genetically inactivated sperm, diploid gynogenesis takes place.

Investigations have shown that the frequency of cases of induced diploidization of the female chromosome complex greatly depends on the individual characteristics of the female and on the experimental conditions. These most essential conditions are: (1) temperature shock, (2) the stage at which temperature treatment is applied and (3) duration of treatment.

The output of gynogenetic diploids as a rule varies greatly, being very high under a most favourable combination of the parameters.

Temperature treatment gave good results in experiments on M. fossilis. As has been mentioned above, the average output of diploid gynogenetic offspring of M. fossilis comprises 0.33 percent. When the eggs inseminated with irradiated sperm were cooled at a temperature of 1–3°C for 10 minutes after insemination (the stage of anaphase 2) the number of diploid gynogenetic offspring was increased up to 14.9 percent. In the most successful experiments, diploid gynogenetic offspring comprised 50–62 percent of the total fertilized eggs.

In thermal shock experiments when the eggs were warmed up to 34°C for 8 minutes after insemination (at the stage of early anaphase 2), the investigators managed to obtain about 17 percent of gynogenetic diploids (Romashov and Belayeva, 1965).

At present, methods of temperature shock are being sought by experiments on carp. Mastering these methods is of interest for the study of experimental triploidy of fish, particularly for the study of prospects of using polyploidy in fish farming.


Experiments carried out on C. carpio and M. fossilis and some Acipenseridae have proved that it is possible to obtain gynogenetic progeny of fish. Investigations have revealed the cytological mechanics and genetic characteristics under radiative diploid gynogenesis. The aim of future experiments is to work out methods of raising the numerical strength of diploid gynogenetic offspring.

It is possible to visualize the prospects of using artificial diploid gynogenesis (Golovinskaya, 1965, 1968). This method holds great prospects for intensifying homozygotization in some line of breeding. Within one generation it is possible to obtain a result that otherwise requires a long period of inbreeding, which, due to late maturation of carp, is very impeded. This method undoubtedly will be useful in selection of heterosis combinations. Such heterosis will evidently be possible in crossings of the topcross type.

The practical importance of artificial diploid gynogenesis is that it makes it possible to obtain entirely female progeny, due probably to the homogametic nature of the female sex. This conjecture is confirmed by the fact that all carp of gynogenetic origin were females.

Experiments on carp have shown that artificial diploid gynogenesis can be used as a special method, opening wide prospects for in-depth analyses of the cytological nature of the fish or other organism under investigation.

In conclusion it is necessary to note an important cytological difference between natural and artificial gynogenesis of fish. In natural gynogenesis the initial number of chromosomes remains the same due to non-exclusion of the first meiotic division. It leads to reproduction of the maternal genotype in an unlimited succession of generations and increases the heterozygosis of offspring under the influence of mutation.

In artificial gynogenesis the initial number of chromosomes remains the same due to the exclusion of the second meiotic division which results in segregation, i.e. the progeny is not genetically identical with the maternal form and is characterized by high homozygosity.

Along with the method of irradiative diploid gynogenesis, other mechanisms of artificial gynogenesis modelling natural gynogenesis are of great interest for future investigations.


Astaurov, B.L., 1955 Polyploidy and parthenogenesis of silkworm. I. Triploid artificial parthenogenesis. Biull.mosk.Obshch.Ispyt.Prir.(biol.), 60(2):34–64

Astaurov, B.L., 1956 Polyploidy and parthenogenesis of silkworm. II. Variance and selection in obtaining triploid clones in connection with evolutional odd-polyploid parthenogenesis in nature. Biull.mosk.Obshch.Ispyt.Prir.(biol.), 61(2):45–66

Astaurov, B.L., 1965 Experimental polyploidy of animals. Sborn.“Poliploidiya i selektiva”, izd. “Nauka”, M-L:43–56

Baranov, P.A. and B.L. Astaurov, 1956 Introduction. Sborn.“Poliploidiya”. izd.“Inostrannaya literature”, M:5–22

Berg, L.S., 1949 Fishes of fresh waters of the USSR and adjacent countries. Izd.Akad. Nauk SSSR, M-L, II:826–31

Betina, M.I. and N.N.Rott, 1966 Cytomorphological analysis of maturation and fertilization of axolotl eggs after thermal treatment resulting in heteroploid embryos. Citologiya, 8(4):475–84

Brachet, A., 1917 L'oeuf et les facteurs de l'ontogénése. O. Doin., Paris, 349 p.

Darevsky, I.S., 1958 Natural parthenogenesis of some subspecies of rock lizard, Lacerta sascicola Eversmann. Dokl.Akad.Nauk SSSR, 112(4):730–2

Darevsky, I.S., 1962 Origin and biological role of natural parthenogenesis in the polymorphic group of rock lizards of the Caucasus. Zool.Zh., 41(3):397–408

Darnell, M., E.Lamb and P.Abramoff, 1967 Matroclinous inheritance and clonal structure of Mexican population of the gynogenetic fish Poecilia formosa. Evolution, 21 (1):168–73

Golovinskaya, K.A., 1946 Linear form of carp. Dokl.Akad.Nauk SSSR, 54(7):637–40

Golovinskaya, K.A., 1954 Reproduction and heredity of the Crucian carp. Trud.VNIORKh, 7:34–57

Golovinskaya, K.A., 1960 Males of Crucian carp and their crossings with carp Rybovod.i Rybolov., 6:16–7

Golovinskaya, K.A., 1965 Natural and artificial gynogenesis of fish. Sborn. “Teoretitcheskije osnovy rybovodstva”, izd. “Nauka”, M:106–8

Golovinskaya, K.A., 1968 Artificial gynogenesis of fish and prospects of creating heterosis combinations. Sborn. “Heterosis v zivotnovodstve”, (in press).

Golovinskaya, K.A. and D.D. Romashov, 1947 Investigations in gynogenesis of the Crucian carp. Trud. VNIORKh, 4:73–113

Golovinskaya, K.A., 1966 Segration in scaly coverings characteristics in diploid radiogynogenesis of the carp. Trud.VNIORKh, 14:227–35

Golovinskaya, K.A., D.D. Romashov, N.B. Tcherfas, 1963 Radiogynogenesis of the carp. Trud. VNIORKh, 12:149–67

Golovinskaya, K.A., D.D. Romashov and N.B.Tcherfas, 1965 Unisexual and bisexual forms of the Crucian carp. Vop.Ikhtiol. 54(37):614–29

Haskins, C., E. Haskins and R. Hewitt, 1960 Pseudogamy as an evolutionary factor in the poecilid fish Mollienesia formosa. Evolution, 4:473–83

Hubbs, C., G. Drewry and B. Warburton, 1959 Occurrence and morphology of phenotypic male of a gynogenetic fish. Science, 129: 1227–9

Hubbs, C. and L. Hubbs, 1932 Apparent parthenogenesis in nature in a form of fish of hybrid origin. Science, 76:628–30

Hubbs, C. and L. Hubbs, 1946 Breeding experiments with the invariably female, strictly matroclinous fish, Mollienesia formosa. Genetics, 31:218

Kallman, K., 1962a Gynogenesis in the teleost, Mollienesia formosa (Girard), with a discussion of the detection of parthenogenesis in vertebrates by tissue transplantation. J. Genet., 58:7–21

Kallman, K., 1962b Population genetics of the gynogenetic teleost, Mollienesia formosa. Evolution, 16(4):497–504

Kallman, K., 1964 Homozygosity in a gynogenetic fish Poecilia formosa. Genetics, 50:260–1

Kirpichnikov, V.S., 1937 Main genes of the scales of carp. Biol.Zh., 6(3):601–29

Krushinsky, L.V., 1946 Sixth annual conference on dynamics of organism development. Usp.Sovetsk.biol. 26(2):289–304

Kryzhanovsky, S. G., 1947 The system of carps. Zool.Zh., 26(1):53–64

Kryzhanovsky, S. G., 1956 Data on herring dynamics. Trud.Inst.Morf.Zhiv.Akad.Nauk.SSSR, 17:254

Kryzhanovsky, S.G., N.N.Disler and E.A.Smirnova, 1953 Ecologo-morphological regularities in the development of perch. Trud.Inst.Morf.Zhiv.Akad.Nauk SSSR, 10:3–138

Lieder, U., 1955 Männchenmangel und naturliche parthenogenes bei Silberkarausche Carassius auratus gibelio (Vertebrata, Pisces). Naturvissenschaften, 42(21):590–4

Lieder, U., 1959 Uber die Eientwicklung bei Mänchenlosen Stammen der Silberkarausche Carassius auratus gibelio (Bloch)(Vertebrata, Pisces). Biol.Zbl. 78(2):284–91

Makino, S., 1939 The chromosomes of the carp, Cyprinus carpio, including those of some related species of Cypinidae for comparison. Cytologia, 9(4):430–40

Meyer, H., 1938 Investigations concerning the reproductive behaviour of Mollienesia formosa. J.Genet., 36(3):329–36

Neifach, A.A., 1959 Radiative ionization effect on early development of fish. Trud.Inst. Morf.Zhiv.Akad.Nauk SSSR, 24:135–59

Nikoljukin, N.I., 1952 Interspecies hybridization of fish. Oblizd., Saratov, 312 p.

Nikolsky, G.V., 1954 Special Ichthyology. Izd.“Sovetskaya Nauka” Moskva, 456 p.

Opperman, K., 1953 Die Entwicklung von Forelleneiern nach Befruchtung mit radiumbestrahlten Samenfäden. Arch.mikroskop.Anat., 83, abt. II (H. Iu4):141–89

Rasch,E., R.Darnell and K.Abramoff, 1965 Cytophotometric evidence for triploidy in hybrids of the gynogenetic fish Poecilia formosa. J.exp.Zool., 160(2):155–70

Romashov, D.D. and V.N. Belyaeva, 1964 Cytology of radiative gynogenesis and androgenesis of loach. Dokl.Akad.Nauk SSSR, 157(4):964–7

Romashov, D.D. and V.N. Belyaeva, 1965 Increase of diploid gynogenetic larvae of the loach (Misgurnus fossilis L.) by applying temperature shock. Biull. mosk.Obshch.Ispyt.Prir.(biol.), 60(5):93–109

Romashov, D.D., et al., 1960 Diploid radiogynogenesis of fish. Biofizika, S,4:461–8

Romashov, D.D., 1963 The possibility of diploid radiogynogenesis of the Acipenseridae. Radiobiologiya, 3(I):104–10

Rott, N.N., 1965 Changes in reaction of eggs to thermal effect in the period of maturation, fertilization and early segmentation of axolotl. Cytology, 7(2):205–11

Statova, M.P., 1963 Data on reproduction of Crucian carp in ponds of Moldavia. Izv. Akad.Nauk Moldav.SSR, 5:49–54

Tcherfas, N.B., 1966a Natural triploidy of females of the unisexual form of Crucian carp. Genetika, 5:16–24

Tcherfas, N.B., 1966b Meiotic analysis of unisexual and bisexual forms of Crucian carp. Trud. VNIORKh, 14:63–82

Tcherfas, N.B., 1968 Main results of cytogenetical analysis of unisexual and bisexual forms of Crucian carp. Trud.nauch.Soveshch.Vop.Genet.Selekt.Gibridiz. Ryb.(in press).

Previous Page Top of Page Next Page