HYBRIDIZATION OF FISHES IN NORTH AMERICA
(FAMILY CENTRARCHIDAE)

by

W.F. CHILDERS
Illinois Natural History Survey
Urbana, Illinois
U.S.A.

1 INTRODUCTION

The Centrarchidae are commonly known as the sunfishes because some species are quite brilliantly coloured. They are endemic to North America, and many species are highly prized by the sport fishermen of that area. Their centre of origin was probably the Mississippi river basin, from where they have spread across the whole of North America. The Centrarchidae family contains 27 extant species. Natural hybridization occurs quite commonly between certain species within certain tribes; however, natural intertribal hybridization is quite rare. Hybridization occurs frequently between the two Pomoxis sp. in the tribe Centrarchini and between many of the species within the tribe Lepomini. Largemouth bass (Micropterus salmoides) and spotted bass (Micropterus punctulatus) natural hybrids have recently been reported.

Table I
Classification of the Centrarchidae ¹

¹ The scientific and common names are those suggested by the American Fisheries Society 1960 and the subfamily and tribal classification follows that of Branson and Moore (1962).

Although intertribal hybridization is rare in nature, laboratory experiments have revealed that viable hybrids can be produced from many intertribal crosses (West and Hester 1966). Intertribal hybrids have been successfully produced between the Centrarchini and the Lepomini and also between the Micropterini and the Lepomini. The white crappie, Pomoxis annularis, and the bluegill, Lepomis macrochirus, have been successfully hybridized. The warmouth, Chaenobryttus gulosus, and the largemouth bass, Micropterus salmoides, have been successfully hybridized both ways. The largemouth female has also been hybridized with the male bluegill, Lepomis macrochirus, and the F₁ hybrid has a body shape that resembles the largemouth bass more than the bluegill. The green sunfish, Lepomis cyanellus, has also been successfully hybridized with the female largemouth bass. In most if not all of these intertribal crosses, there are partial lethals expressed and the numbers of deformed individuals are usually fairly high. Because of the rather typical expression of partial lethals in the intertribal crosses, much research must be conducted before the value of intertribal hybrids to fisheries management can be determined.

2 EXPERIMENTAL SPECIES

Expression of partial lethals is much less frequent in intratribal crosses than in the intertribal crosses. In this paper is reported in some detail, the results of a study involving the hybridization of four species in the tribe Lepomini. The four species selected for study were the red-ear sunfish, Lepomis microlophus, the bluegill, Lepomis macrochirus, the green sunfish, Lepomis cyanellus, and the warmouth, Chaenobryttus gulosus. These species were studied because of local availability, taxonomic relationship, and similarities and differences in their morphology, habit selection, and reproductive behaviour. All four species are sympatric in Illinois and there is broad overlapping of their periods of reproduction. The males of all four species construct saucer-shaped nests which they guard with great vigour. The males also protect and care for the eggs and larvae. When the larvae develop into free swimming fry they leave the nest and the males show them no additional parental care. The three Lepomis species nest in colonies, and the warmouth is a solitary nester. The colony nesters sometimes nest in mixed colonies; consequently the functional life spans of gametes could be very important in controlling hybridization between these species. If gametes are capable of fertilizing and being fertilized over long periods of time, sperm driftage could result in the production of hybrid individuals. Experiments were conducted to determine the functional life spans of bluegill, green sunfish, and warmouth gametes. By stripping gametes and aging them for various periods of time prior to fertilization, it was determined that the average functional life span was approximately one minute for the spermatozoa and one hour for the ova. The brief functional life spans of the spermatozoa of these species are undoubtedly very important in reducing hybridization caused by sperm drifting from nest to nest.

3 HYBRIDIZATION EXPERIMENTS

Two types of experiments were used to produce hybrid sunfishes. In the first, referred to as stripping experiments, gametes were stripped from ripe adults and manually mixed. With this method it was possible to determine species isolation due to incompatibilities between sperm and eggs (primary genetic isolation). In the second type, designated isolation experiments, one or more pairs of fish composed of a male of one species and a female of another were isolated in small ponds to determine if they would hybridize when mates of their own species were absent.

In this paper R refers to red-ear sunfish, B to bluegill, G to green sunfish, and W to warmouth. Matings between individuals of different species are designated to P₁ crosses and the resultant hybrids are designated as F₁ hybrids. F₂ hybrids are those produced by mating an F₁ male with an F₁ female. The P₁ cross of a male bluegill with a female green sunfish is designated B × G and the resultant hybrids are designated BG F₁ hybrids; GB F₁ designates the reciprocal hybrids.

3.1 Stripping Experiments

In the stripping experiments sperm and eggs stripped from the four parent species were paired in 16 different combinations to produce zygotes representing the four parent species and 12 hybrids. These experiments were designed to allow comparisons of rates of embryological development and the extent of viability of F₁ hybrids and their maternal parent species.

Eleven stripping experiments were conducted: Eggs from three red-ear sunfish, three bluegills, three green sunfish, and two warmouths were fertilized with sperm from males of all four species. The temperatures at which these experiments were conducted were well within the range of temperatures that embryos of the four species are subjected to under natural conditions. In Table II the viability of each kind of hybrid is compared to that of its maternal parent species.

Table II

The degree of viability of 16 different kinds of fishes produced by pairing gametes from red-ear sunfish, bluegills, green sunfish, and warmouths

¹ R = red-ear sunfish, B = bluegill, G = green sunfish, W = warmouth.
² Percentage based on number of eggs at the time sperm and eggs were mixed together and the number that hatched.
³ Percentage based on number of eggs at the time sperm and eggs were mixed together and the number of morphologically normal-appearing fry.
⁴ More than 90 percent of these larvae were morphologically deformed.
⁵ These fry appeared morphologically normal, but all behaved abnormally.

No hybrid type was significantly different from its maternal parent species in the percentage of zygotes that hatched; however, more than 90 percent of the WR and WB F₁ hybrids were morphologically abnormal.

Both WR and WB F₁ hybrids exhibited high mortality between the hatching and swim-up fry stages. At the time the experiments were terminated, only two percent of the WR hybrids and one percent of the WB hybrids appeared to be morphologically normal. All of these morphologically normal-appearing WR and WB F₁ hybrid fry were very sluggish. When petri dishes containing these hybrid fry were tapped with a pencil, the fry responded with weak swimming movements or not at all, and it is very doubtful that any of these fry would have become free swimming. Fifty-five percent of the WG hybrids and 75 percent of the pure green sunfish zygotes developed into normal-appearing swim-up fry (difference significant to 0.05 level). The WG hybrid swim-up fry appeared to be behaving normally. The remaining nine kinds of hybrids were not significantly different from their maternal parent species in the percentages that developed into normal swim-up fry.

The mean hatching time and standard deviation were calculated for each of the four parent species and each of the 12 kinds of hybrids. An analysis of variance revealed that WB F₁ hybrid zygotes hatched significantly sooner than pure bluegill zygotes when both kinds of zygotes were incubated at the same temperatures. Although WB zygotes hatched in less time, the newly emerged WB F₁ larvae were not as advanced in their development as the unhatched pure bluegill embryos. WR F₁ hybrids were not significantly different from pure red-ears in hatching time; however, the newly emerged WR larvae were not as advanced in their development as the pure red-ear larvae. There were no statistically significant differences in the time of hatching between the other 10 kinds of hybrids and their respective maternal parent species, and differences in the degree of development between the hybrids and their respective maternal parent species were not pronounced.

The alpha temperature threshold of development (Shelford 1927) and the mean number of developmental units (degree-hours of effective temperature) necessary for 50 percent hatching were calculated for 12 kinds of zygotes. It was impossible to calculate alpha temperature thresholds for zygotes involving warmouth females because of similarities of incubation temperatures. The alpha temperature threshold of development is the theoretical temperature below which normal embryonic development does not occur. T test comparisons revealed that the alpha thresholds of development of the 12 kinds of fishes were not significantly different from one another. The alpha thresholds ranged from 17.8° to 18.6°C (64.0° – 65.5°F) and the mean alpha threshold of all 12 kinds of fishes was 18.3°C (64.9°F). Approximately 280 developmental units centigrade scale or 500 units Fahrenheit scale were necessary for 50 percent hatching.

3.2 Isolation experiments

Thirty-two isolation experiments were conducted. Different species were isolated in small earthen ponds. Each of the 12 possible hybrid producing combinations was tested in one or more ponds. The R × G, G × B, and W × G pairing successfully hybridized each time they were tested.

The R × B cross was attempted in four ponds. No hybrids were produced in three ponds although the ponds remained full and were uncontaminated by other fishes. Eleven small fish were found when the fourth pond was drained, and these fish were believed to have been RB F₁ hybrids although they were not positively identified as such (Childers & Bennett 1961). The water in this pond contained a high and constant clay turbidity that reduced the transparency of the water and caused the parent fish to be extremely pale in body colour. The normally scarlet portions of the opercular tabs of the red-ear males appeared as a faint rose colour. The results of all other isolation experiments were either negative or inconclusive.

Fish hybridization might result from sperm driftage or interspecific matings. Sperm driftage is an important cause of hybridization among certain species of fishes, particularly minnows and darters which live in flowing water habitats and simultaneously spawn in close proximity to one another (Hubbs 1955). Sperm driftage may also account for some hybridization between pond- or lake-dwelling centrarchids; however, since average functional life spans of sunfish spermatozoa are so brief and since there is such good synchronization in the release of sperm and eggs by a spawning pair, most hybrid sunfish are probably the result of interspecific pair formation.

The four experimental species are sexually dimorphic, closely allied, sympatric species. Signals that are in some way involved in reproductive isolation of such species are likely to be highly divergent and may involve specific differences in shape, colour, special movements, sounds, scents, etc. The precise signals which are operative in conspecific pair formation of the four experimental species are not known; however, specific differences in colour of opercular tabs, eyes, cheeks, and pelvic fins of nest-guarding males may be important in controlling the behaviour of ripe females. When a female ready to spawn approaches a nest-guarding male, she usually stops some distance from the nest and the male exhibits a courtship display (Miller 1963). Species recognition apparently occurs during this short time, and the female flees or remains in the vicinity of the nest and accepts the advances of the male.

Since in one isolation experiment there was an indication that the scarlet portions of the opercular tabs of male-red-ear sunfish might possibly prevent hybridization between male red-ears and female bluegills, experiments were conducted to test this hypothesis. Four small earthen ponds were each stocked with three ripe adult male red-ear sunfish and three adult female bluegills. The opercular tabs were clipped from all males stocked in two ponds and the tabs were left intact on the males stocked in the two other ponds.

The ponds were drained during early October, and several thousand small hybrid fry were collected from the ponds containing red-ear males whose opercular tabs had been removed. No small fish were found in the control ponds. An examination of the clipped males revealed that the blue portion of the opercular tabs had regenerated to almost normal size but the scarlet portions had not regenerated. One tab on each of these males had a small, narrow, yellowish-orange margin.

Two such tests cannot, of course, be considered conclusive proof that specific differences in the colour of the opercular tabs of male red-ears are highly functional in preventing hybridization with female bluegills; however, additional investigation of the importance of colour as a reproductive isolating mechanism in the sunfishes might prove rewarding.

According to Hubbs (1955), fish hybridization is controlled to a large extent by environmental factors. Sunfish hybrids appear to be more common in ponds which are choked with aquatic vegetation or have high turbidities than in clear-water ponds which have extensive spawning areas free from vegetation. In weed-choked ponds or ponds with high turbidities the range of visibility must be short, and under these conditions ripe females might occasionally spawn with males without observing preliminary courtship displays believed to be important in conspecific pair formation.

4 HYBRIDS REARED IN PONDS

Large numbers of each of the 10 viable F₁ hybrid types were stocked in one or more ponds. The F₁ hybrids were reared to maturity in their respective ponds and the sex ratio, fecundity, and degree of heterosis of each F₁ hybrid population were studied.

4.1 Sex Ratios of F₁ hybrids

Sexually mature F₁ hybrids were collected from each population and sexed. Of the 10 kinds of viable F₁ hybrids, seven were predominately males (RB, BR, and BG were 97 percent males; WG were 84 percent males; and RG, GB, and BW were approximately 70 percent males), two were approximately 50 percent males (GR and RW), and one was predominately female (GW was 16 percent males). Ricker (1948) determined the sex of 428 BR F₁ hybrids in Indiana and found them to be 97.7 percent males.

Sex determination in sunfishes is very poorly understood. Bluegills, green sunfish, and their F₁ hybrids apparently have 24 pairs of chromosomes, and the sex chromosomes are indistinguishable from the autosomes (Bright 1937). Bright also reported that the chromosomes are so similar in shape and size that he was unable to detect specific differences. Roberts (1964) found that red-ear, bluegill, and warmouth sunfishes each have 24 pairs of chromosomes; green sunfish from North Carolina had 24 pairs; but green sunfish from West Virginia had only 23 pairs.

The unbalanced phenotypic tertiary sex ratios of the F₁ hybrid sunfish could result from unbalanced primary genetic sex ratios, specific differences in the strength of sex-determining factors, an overriding of the genetic sex by environmental factors, or differential mortality of the sexes.

Since the WG F₁ hybrids were 84 percent males and the reciprocal cross hybrids were 16 percent males, it is possible that the strength of sex-determining factors of warmouths are 5.25 times more powerful than those of green sunfish. Specific differences in the strength of sex-determining factors cannot alone explain the sex ratios of the remaining eight kinds of viable hybrids, since none of these were predominately females.

RB and BG F₁ hybrids were both 97 percent males. If differential mortality were the cause of these unbalanced sex ratios, much of the mortality would have had to occur after the swim-up fry stages, since in the stripping experiments total mortality between fertilization and the swim-up fry stages was only 14 percent for the RB and 27 percent for the BG F₁ hybrids.

It is not known which sex is the heterogametic condition for the sex chromosomes of the four experimental species; however, Haldane (1922) formulated a rule which furnishes a clue: “When in the F₁ offspring of a cross between two animal species or races, one sex is absent, rare, or sterile, that sex is always the heterozygous sex.” Using Haldane's rule, Krumholz (1950), in a study concerning BR F₁ hybrids, pointed out that the males of both bluegills and red-ear sunfish are probably homozygametic for sex and the females heterozygametic. The application of Haldane's rule to all possible F₁ hybrids produced from red-ear sunfish, bluegills, and green sunfish indicates that the female is the heterozygametic sex in these three species. Hybridization of male warmouths with females of the three Lepomis species resulted in partial or complete lethals, suggesting that in the warmouth the male is the heterogametic sex.

4.2 Reproductive success of hybrids

The reproductive success of each of the 10 kinds of viable F₁ hybrids was investigated in one or more ponds. The occurrence and abundance of F₂ hybrids were determined by seining, trapping, shocking, poisoning or draining the ponds after the F₁ hybrids were one or more years of age. RB, BR, and BG failed to produce abundant F₂ generations when in ponds which contained no other species of fishes. In contrast to these results, BR F₁ hybrids produced abundant F₂ generations in two ponds in Indiana (Ricker 1948). The other seven kinds of F₁ hybrids produced abundant F₂ populations when stocked in ponds containing no other fishes. Three of the seven kinds of F₁ hybrids which produced large F₂ populations when stocked in ponds containing no other fishes were also stocked in ponds with largemouth bass. RG F₁ hybrids and GB F₁ hybrids, when stocked with largemouth bass, produced only a few F₂ hybrids. No F₂ hybrids were found in the pond stocked with BW F₁ hybrids and largemouth bass. WG F₂ hybrids and GW F₂ hybrids were stocked in ponds containing no other fishes. Both of these F₂ hybrids produced large F₃ populations.

Backcrosses, outcrosses, a four-species cross, and a three-species cross involving F₁ hybrids are listed in Table III. The BW × B backcross was made by stocking adult male BW F₁ hybrids and adult female bluegills in a pond which contained no other fishes. The other 12 crosses listed in Table III were made by stripping gametes from ripe adults and rearing the young to the free-swimming fry stage in the laboratory.

R × RW, W × RW, B × RW, G × RW, R × GB, and RB × W young were killed after they developed into free-swimming fry because of the lack of ponds in which they could be stocked. All six kinds of fry appeared to be normal and probably would have developed into adults. Free-swimming fry of the remaining six crosses in the laboratory were stocked in ponds and did develop into adult fishes. BW × B, G × GW, and B × RG populations produced large numbers of young.

Table III

Successful backcrosses, outcrosses, four-way cross, and another cross involving F₁ hybrid sunfishes.¹

¹ R = red-ear sunfish, B = bluegill, G = green sunfish, W = warmouth.

Hubbs & Hubbs (1933) reported that in Michigen F₁ hybrids of bluegills, green sunfish, longear sunfish, pumpkinseeds, and orangespotted sunfish were unable to reproduce because males were sterile and ova stripped from the few adult females used in the experiments appeared distinctly abnormal. This study, often cited in the literature, has resulted in a rather widespread belief that all male hybrid sunfish are sterile. Results of my experiments conclusively establish that a number of different kinds of hybrid sunfishes produced in Illinois are not sterile, are fully capable of producing abundant F₂ and F₃ generations, and can be successfully backcrossed to parent species and even outcrossed to nonparental species.

4.3 Hybrid vigour

Heterosis has been defined (Manwell, Baker & Childers 1963) as “that condition where, with respect to one or more particular characteristics, the values for most, if not all, of the individual hybrids fall significantly outside the range formed from the means for both parent populations. In cases of positive heterosis - hybrid vigor - the hybrid shows a faster growth rate than either of the parents, or it possesses some other characteristic, often an economically significant one, at a ‘better’ level than the parents do.”

4.4 Rate of growth

In an attempt to determine whether certain F₁ hybrid sunfishes actually grow faster than their parent species, two experiments were conducted in which equal numbers of uniformly sized F₁ hybrids and parent species were stocked in ponds which contained no other fishes. Intraspecific competition is keener than interspecific competition because individuals of the same species are more nearly equal in their structural, functional, and behavioural adaptations. Consequently in experiments designed to compare rates of growth, it is imperative to use equal numbers of similarly sized fishes. In the first experiment the growth rate of BG F₁ hybrids was compared to that of green sunfish. In the second experiment the growth rates of GR F₁ hybrids, green sunfish, and red-ear sunfish were compared.

In both experiments the average increase in total length of the hybrids was not significantly different from the increases of the parental species. The population densities of the fishes in both ponds were much lower than would be found in most normal natural populations. In both experiments intraspecific and interspecific competition was undoubtedly quite light; consequently, the question of whether certain F₁ hybrid sunfishes are superior to their parent species in rate of growth cannot be answered until high density populations containing equal numbers of equal sized hybrids and parent species are studied.

4.5 Electrophoretic patterns of haemoglobins

Red-ear sunfish, bluegills, green sunfish, and warmouths each have unique haemoglobin patterns in vertical starch gel electrophoresis. Most of the F₁ hybrids of these species yield haemoglobin patterns that are identical with those obtained by simply mixing haemoglobins of the two parental species; however, from 25 to 40 percent of the haemoglobin from BW, GW, and WG F₁ hybrids has electrophoretic properties different from the haemoglobins of the parental species. Oxygen equilibria for the haemoglobins from these three hybrids show greater heme-heme interactions than those for haemoglobin from any of their parental species. As a result of this haemoglobins from these three hybrids have better blood gas transport properties than those of their parental species, and in this respect each of these three hybrids is believed to exhibit hybrid vigour (Manwell, Baker & Childers 1963).

4.6 Vulnerability to hook-and-line capture

Although no controlled experiment has tested whether F₁ hybrid sunfishes are more vulnerable to angling than their parental species, certain F₁ hybrids are so easily caught that at several locations sport fishermen have almost completely eliminated substantial hybrid sunfish populations in a few days of angling, necessitating a daily creel limit on these hybrids.

5 HYBRIDS FOR SPORT FISHING

Over-population of sunfish is the single greatest problem encountered in the management of Illinois lakes and ponds containing largemouth bass and one or more of the Lepomis species. The Lepomis species have such high reproductive capacities and survival capabilities that they commonly become so abundant that they are unable to grow to sizes large enough to be of value to fishermen. Because certain kinds of F₁ hybrid sunfishes appear to be unable to produce sizable F₂ populations in ponds containing largemouth bass, a number of experiments are now in progress to test the usefulness of hybrids in combination with largemouths. Preliminary results indicate that several types of hybrids in combination with largemouth bass furnish fishing superior to that furnished by bass in combinations with the hybrids' parent species.

6 REFERENCES

American Fisheries Society, 1960 A list of common and scientific names of fishes from the United States and Canada. Spec.Publs.Am.Fish.Soc., (2):102 p.

Branson, A., and G.A.Moore, 1962 The lateralis components of the acustico-lateralis system in the sunfish family Centrarchidae. Copeia, (1):1–108

Bright, W.M., 1937 Spermatogenesis in sunfish. Thesis, Urbana, University of Illinois, 49 p.

Childers, W.F. and W. Bennett, 1961 Hybridization between three species of sunfish (Lepomis). Biol.Notes nat.Hist.Surv.Div.St.Ill., (46):15 p.

Haldane, J.B.S., 1922 Sex ratio and unisexual sterility in hybrid animals. J.Genet., 12(2):101–8

Hubbs, C.L., 1955 Hybridization between fish species in nature. Syst.Zool.,4(1):1–20

Hubbs, C.L. and L.C.Hubbs, 1933 The increased growth, predominant maleness, and apparent infertility of hybrid sunfishes. Pap.Mich.Acad.Sci., 17:613–41

Krumholz, A., 1950 Further observations on the use of hybrid sunfish in stocking small ponds. Trans.Am.Fish.Soc., 79:112–24

Manwell, C., C.M.Baker and W.F.Childers, 1963 The genetics of hemoglobin in hybrids. 1. A molecular basis for hybrid vigor. Comp.Biochem.Physiol., 10:103–20

Miller, H., 1963 The behavior of the pumpkinseed sunfish, Lepomis gibbosus (Linneaus), with notes on the behavior of other species of Lepomis and the pigmy sunfish, Ellassoma evergladei. Behaviour, 22(1–2):88–151

Ricker, E., 1948 Hybrid sunfish for stocking small ponds. Trans.Am.Fish.Soc., 75:84–96

Roberts, F.L., 1964 A chromosome study of twenty species of Centrarchidae. J.Morph., 115(3):401–17

Shelford, E., 1927 An experimental investigation of the relations of the codling moth to weather and climate. Bull.Ill.St.Nat.Hist.Surv., 16(5):311–440

West, J.L. and F.E.Hester, 1966 Intergeneric hybridization of centrarchids. Trans.Am. Fish.Soc., 95(3):280–8