当前位置：文档库 › Response to three generations of selection for increased body weight in channel catfish, Ictalurus p

Response to three generations of selection for increased body weight in channel catfish, Ictalurus p

Response to three generations of selection for

increased body weight in channel catfish,

Ictalurus punctatus ,grown in earthen ponds

Mahmoud A.Rezk a,1,R.Oneal Smitherman a ,John C.Williams b ,

Amy Nichols a,c ,Huseyin Kucuktas a ,Rex A.Dunham a,*

a Department of Fisheries and Allied Aquacultures,Alabama Agricultural Experiment Station,Auburn University,

203Swingle Hall,Auburn,AL 36849,USA

b Research Data Analysis,Alabama Agricultural Experiment Station,Auburn University,Auburn,AL 36849,USA

c Department of Entomology,Clemson University,Clemson,SC 29634,USA

Received 12March 2003;accepted 18March 2003

Abstract

Selection response for body weight at marketable size was measured for channel catfish,Ictalurus punctatus ,grown in earthen ponds at 7500fish/ha.Three generations of mass selection for increased body weight in Kansas and Marion strains of channel catfish increased body weight from 453to 583g or 29%,and from 530to 642g or 21%,respectively,with cumulative realized heritabilities of 0.17F 0.016and 0.19F 0.012,respectively.Realized heritabilities for the third generation alone were 0.16F 0.016and 0.23F 0.015for Kansas and Marion strains,respectively.The results indicate that body weight can be significantly increased via mass selection in channel catfish,which should result in increased production and profitability in the catfish farming industry.

Keywords:Channel catfish;Ictalurus punctatus ;Selection;Body weight;Heritability;Selective breeding

1.Introduction

Channel catfish,Ictalurus punctatus ,is the main aquaculture species in the United States (Waters,2001).The projected production of edible catfish in 2002is estimated at

*Corresponding author.Tel.:+1-334-844-9121;fax:+1-334-844-9208.

E-mail address:rdunham@https://www.wendangku.net/doc/eb12894167.html, (R.A.Dunham).

1Current address:ICLARM Regional Research Center for Africa and West Asia,PO Box 1261,Maadi,Egypt.

https://www.wendangku.net/doc/eb12894167.html,/locate/aqua-online

Aquaculture 228(2003)69

–79

600million kg.Growth rate is one of the most important traits for genetic improvement in aquacultural species (Gjedrem,1983;Mahon,1983;Gjerde,1986).The amount of additive genetic variation for growth related traits in fish indicates potential for genetic improvement (Gjedrem,1983;Kinghorn,1983).Realized response to selection in fish is generally very high compared with what has been obtained in farm animals (Gjerde,1986;Dunham et al.,2001).Genetic improvement programs such as selection may allow additional aquaculture production to help meet the increasing demand for fish products.

Attempts to improve growth rate in fish through selection have been successful in several species (Dunham et al.,2001).Growth rate of coho salmon,Oncorhynchus kisutch ,was improved by more than 60%after four generations of mass and family selection for increased body weight (Hershberger et al.,1990).Hershberger et al.(1990)concluded that there was a significant effect of domestication and that a portion of the 60%increase resulted from domestication effects.In rainbow trout,Oncorhynchus mykiss ,an increase of 48%over the control was produced after three generations of family selection for increased 147-day weight (Kincaid et al.,1977).Potential for improvement in Nile tilapia,Oreochromis niloticus ,through selection was indicated by Jarimopas,unpublished data (Doyle and Talbot,1986),who reported a realized heritability of 0.19for 12-week weight.A combination of mass,family and index selection for seven generations increased body weight in Nile tilapia by 77%(Dunham et al.,2001).Heritability for increased body weight was 0.23F 0.05(Bondari et al.,1983)in O.aureus .Thien (1971)reported realized heritabilities of 0.1and 0.16for growth rate in male and female O.mossambicus ,respectively.In general,selection has been very successful for increasing body weight in fish.An average improvement of 6–7%per generation has been obtained for most species,but with some species such as Atlantic salmon,Salmo salar ,coho salmon,channel catfish,Labeo rohita and Nile tilapia exhibiting as much as 11–14%improvement per generation by utilizing mass selection,family selection or both (Dunham et al.,2001).

Efforts to genetically improve the aquacultural traits of channel catfish through selection are in the early generations.Realized heritability and heritability estimates for increased growth rate indicate potential for improving growth rate in channel catfish through selection (EI-Ibiary and Joyce,1978;Reagan,1979;Bondari,1983;Dunham and Smitherman,1983b,1985;Rezk,1993;Dunham and Brummett,1999).Heritability estimates for 18-month-old channel catfish ranged from 0.40to 0.87while realized heritability ranged from 0.24to 0.50.

Responses to one generation of selection were 63,73,and 54g (17%,18%,and 12%increase in body weight)in Rio Grande,Marion,and Kansas strains of channel catfish,respectively,when grown in earthen ponds at 7500fish/ha (Dunham and Smitherman,1983a,b).In channel catfish,an improvement of 12%was accomplished after two generations of mass selection for increased body weight (Brummett,1986;Dunham and Smitherman,1987;Dunham and Brummett,1999).Bondari (1980)was successful in selecting for increased size variability;but selection for increased size uniformity was not effective.Short-term selection experiments (Bondari,1983;Dunham and Smitherman,1983a,b)illustrate that body weight of channel catfish can be increased.However,longer-term selection experiments have not been conducted,and the nature of long-term selection response has not been evaluated in channel catfish.

M.A.Rezk et al./Aquaculture 228(2003)69–79

M.A.Rezk et al./Aquaculture228(2003)69–7971 The objectives of this study were to evaluate the response to selection for increased body weight and to determine realized heritability after three generations of mass selection in Marion and Kansas strains of channel catfish grown to marketable size in earthen ponds under commercial conditions common in the Southeastern United States.

2.Materials and methods

2.1.Experimental fish

The selection program was conducted with two randomly bred strains of channel catfish,Kansas and Marion,maintained at the Fisheries Research Unit,Alabama Agricultural Experiment Station,Auburn,AL.In each generation of selection,each group was reproduced and raised in a common environment to standardize environmental effects. Fish were harvested when they weighed approximately450g,enumerated,and individ-ually weighed(Chappell,1979;Dunham,1981;Brummett,1986;Rezk,1993;Dunham and Brummett,1999).

Kansas strain originated from a population of30–50fish from the Ninnescah River, Pratt,KS,in1991.The population maintained at Auburn was developed from six to eight pairings in1976(Dunham and Smitherman,1984).This strain possesses high growth rate, resistance to diseases,and late sexual maturity(Dunham and Smitherman,1984).

Marion strain was brought to the Fisheries Research Unit in Auburn from a federal hatchery in Marion,AL in1970.They were perpetuated in1976with six pairings (Dunham and Smitherman,1984).This strain is characterized by its high seinability, relatively large head,very poor disease resistance,relatively high percentage of albinism, moderate growth rate,a brassy color,and pre-feeding behavior in small ponds which includes schooling and rapid swimming causing a rippling effect on the pond surface (Dunham and Smitherman,1984).

2.2.Select lines

Kansas select and Marion select lines have been developed through three cycles of mass selection for increased body weight on the Kansas and Marion strains,respectively. Fish were harvested when they weighed approximately450g.The largest10%of the males and females were selected in each generation(Dunham and Smitherman,1983a,b, 1987).Kansas select grew to a final weight that was12%higher than its randomly bred population after two generations of selection(Brummett,1986;Dunham and Brummett, 1999).

2.3.Production of fingerlings

Fish were spawned in plastic-coated-wire pens250?150?50cm placed in0.1ha ponds.Each pen contained a wooden spawning box95?35?35cm that was open on the end facing the inside of the pond.Spawning boxes had a35?35cm removable cover above the end facing the dike.Adults from each of the Kansas and Marion strains and adults

that represented the largest 10%of males and largest 10%of females from the second generation of selection were spawned in the pens.One catfish pair was placed in each pen.

Spawning boxes were examined for eggs every 3or 4days.A total of 13,13,12,and 15spawns were obtained for Kansas control,Kansas select,Marion control,and Marion select,respectively.The egg masses were removed and incubated indoors in window-screen baskets (30?20?20cm)that were suspended in aerated hatching troughs.Water in the troughs was continuously flowing and agitated by electrically powered paddles.The eggs were given regular prophylactic treatments of formalin every 8h.Initially,treatments of 100ppm were administered.The concentration of formalin was gradually reduced to 30ppm for eggs within 1–2days of hatching.The frequency of treatment was also reduced gradually to once per day depending upon the extent of fungal growth in the eggs in the later stages.No formalin treatments were given to the sac-fry or while training fry to artificial feed,a 50%protein salmon starter.

Fry from each group were stocked in separate ponds,0.04–0.1ha each,at a rate of 148,000fry/ha.They were gradually shifted from salmon starter to 36%protein catfish crumbles.In August (3months of age),they were gradually shifted from catfish crumbles to 36%protein catfish floating feed.

2.4.Stocking and food-fish production

During the last week of March (10months of age),fingerlings were seined from the ponds and restocked separately at a rate of 7500individuals/ha in 0.04ha ponds.All fish were heat branded and stocked communally within six ponds.Fish were fed to satiation once every day with 36%protein floating catfish feed approximately at 1200–1500h.Dissolved oxygen level was maintained above 2.0mg/l using emergency aeration.Water was added to the ponds as needed to maintain an average water depth of 0.9–1.0m.

2.5.Harvesting and data collection

Fish were harvested during the last week of September (16months of age)by seining each pond twice and then removing the remaining fish by hand after draining most of the water from the pond.After harvest,fish were kept in aerated indoors-concrete tanks supplied with jets of running fresh water above the surface to improve aeration and water quality.Fish were sorted by seine haul,strain (Marion and Kansas),line (control and select),and sex,and were weighed individually.A split-plot arrangement (Steel and Torrie,1980)was implemented for analysis of body weight differences caused by strain,selection,sex,and their interactions.Strains were handled as main plots.Selection lines and genders were handled as subplots.The model that was implemented was

Y ijkl ?l tA i tP ij tB k tC l teAB Tik teAC Til teBC Tkl teABC Tikl

teBCP Tijkl tF ijklo

where l =the population mean;A =strain;B =selection line;C =gender;F =individual fish;P =pond;i =1,2;j =1,2,...,6;k =1,2;l =1,2;0=1,2,...,n where n is the number of individual fish in each group.

M.A.Rezk et al./Aquaculture 228(2003)69–79

M.A.Rezk et al./Aquaculture228(2003)69–7973 The Statistical Analysis System,SAS,release6.04(SAS Institute,1991)was used in the data analysis.Cumulative realized heritability was calculated(Falconer,1989)based on the equation

h2?R=eSD1tSD2t...tSD nT

where R=cumulative response to selection,SD1+SD2+...+SD n=selection differentials for generations1,2,...,n.Response to selection and selection differentials were adjusted for the second generation based on a standardized control weight as described next because controls reached different body weights in different generations because of environmental differences among generations.Mean weight for the Marion control line in the second generation was extrapolated using mean weights of both control lines in the first generation,and mean weight of Kansas control in the second generation.Adjusted mean weight of Kansas control and the extrapolated mean weight of Marion control in the second generation were generated based on the ratio of the mean weight of these two groups in generation1,and were then used to standardize both response to selection and selection differential for the second generation only(Dunham and Brummett,1999).

Standard error of cumulative realized heritability s h2was calculated as r R=eSD1tSD2tSD3T

where r R=standard error of the response to selection.

r R was calculated from the equation

r2?V p?eth2=N eTte1=mT R

where V p,=phenotypic variance,t=the number of generations,N e=the effective pop-ulation number,and m=the number of individuals measured(Falconer,1989).

Realized heritability from the second to the third generation(Warwick and Legates, 1979)was also calculated.Standard error of realized heritability was calculated using the same formula for the cumulative h2with the number of generations=1.

3.Results

Mean weight of the control lines was491g.Select lines averaged613g.Three generations of mass selection produced an average24.8%increase(P=0.0001)in body weight in both strains.On the average,there was an increase of8.3%in body weight per generation.The control line of the Kansas strain weighed453g while the select line weighed583g(P<0.01),a28.7%increase(Fig.1).The control line of the Marion strain weighed530g,while the select line weighed642g(P<0.01),a21.1%increase(Fig.1). There was an increase of9.8%and6.5%per generation in Kansas and Marion strains, respectively.There was no difference in weights between the two strains(P=0.95).The apparent difference in percent improvement is augmented by the difference(P=0.05)in average weight of the Kansas control and Marion control lines,453and530g, respectively,that were used to calculate the percentages(Table1).

Realized heritability from the second to the third generation was 0.16F 0.016and 0.23F 0.015for Kansas and Marion strains,respectively.Cumulative realized heritability,h 2,for increased body weight was 0.17F 0.016and 0.19F 0.012for three generations of selection in Kansas and Marion strains,respectively.Unadjusted cumulative selection response and differential were 130and 781g,respectively,in the Kansas strain and 112and 594g,respectively,in the Marion strain.After one generation of selection on the

same Fig.1.Percent improvement (response to selection)for Kansas select and Marion select channel catfish,I.punctatus ,relative to the controls after the first,second and third generations of selection for increased body weight.

Table 1

Mean body weight (BW),response to selection (R ),cumulative response to selection,selection differential (SD),all in grams,realized heritability (h 2)and cumulative realized h 2(Ch 2)for increased body weight in the first,second,and third generation of selection for Kansas strain (K)and Marion strain (M)channel catfish,I.punctatus ,in the parental generation,and Kansas control (KC),Kansas select (KS),Marion control (MC),and Marion select (MS)grown in earthen ponds at 7500fish/ha

Generation

Mean BW (g)Mean BW (g)KC KS R CR SD h 2Ch 2MC MS R CR SD h 2Ch 2Parent a

592572First b

45951354541640.330.3341348673731460.500.50Second c

4595130541420.000.17413d 461à2548170à0.150.15Third e

453583761304750.160.17530642641122780.230.19a

Chappell (1979).b Dunham (1981).

c Brummett (1986).

d Means weighted and adjusted based on th

e ratio o

f body weights found for Kansas and Marion controls in generation 1.MC not evaluated,but extrapolated from generation 1.

e This study.

M.A.Rezk et al./Aquaculture 228(2003)69–79

M.A.Rezk et al./Aquaculture228(2003)69–7975 Table2

Mean body weight for male and female Kansas control(KC),Kansas select(KS),Marion control(MC),and Marion select(MS)channel catfish,I.punctatus,after three generations of selection for increased body weight Genotype a Mean body weight(g)

KC KS MC MS Females b4232547249526132 n322281264430 Males b4831,461915651,46711 n380402445513 Line average4534,55833,55304,56423,5 n702683709943 Differences between selection lines l,2,between strains3,4,and between control and select genotypes5,6,were significant based on Student’s t-test as indicated by superscripts;odd numbers a=0.01,even numbers a=0.05.

a Males and females were compared within genotypes,control and select lines were compared within strain and sex,and strains were compared within sex and selection lines.

b No differences were found between males and females within genotypes.

populations,h2in Kansas and Marion strains was0.33and0.50(Dunham,1981;Dunham and Smitherman,1983a,b).

The gender of the fish affected body weight(P=0.001).Males averaged585g,while females averaged540g.This effect was not influenced by interactions of the selection process,the strain of the fish,or both(P=0.62,0.97,and0.52,respectively).Males were more numerous than females(Table2).

The statistical model that was implemented in this study accounted for effects of strain, selection and sex.These factors accounted for20%of the variation in body weight. 4.Discussion

One generation of selection for increased body weight in a previous study produced 12%and18%increase(Fig.1)in body weight in Kansas and Marion strains,respectively (Dunham and Smitherman,1983a,b).There was no additional improvement in the second generation;in fact,Marion select had decreased growth in the second generation in comparison to the first generation(Dunham and Brummett,1999).Several possible explanations for the non-linearity of the response to selection exist.These include(1) the effect of increased inbreeding in the select line;(2)the effect of domestication which might favor the control lines and has been shown to improve the performance of the domesticated strain by between2%and6%per generation(Dunham and Smitherman, 1987);(3)the inherent tendency of response to selection toward nonlinearity possibly because of changes in allele frequencies,chromosomal rearrangements,dominance,or epistatic effects;(4)approach of selection plateaus as additive genetic variation decreases, is hidden,or released from linkage groups;(5)and/or by genotype–environment interactions.

Sex differences in the current study were not influenced by interactions of the selection process,the strain of the fish,or both.Similar results were reported in other studies.While gender did not affect response to selection(Dunham and Smitherman,1983b),it had a

significant effect on body weight in channel catfish (Dunham et al.,1985).In the current study,this trend was observed in all genotypes,both control and select (Table 2).

Another interesting observation was that in each of the four lines,males were more numerous (Table 2).This could be due to either differential survival,inherently skewed sex ratios in these genetic lines,or both.Recent evidence indicates sex ratios in some species of fish can be affected by hatching temperatures (Dunham et al.,2001).

Strain,selection,and sex accounted for 20%of the variation in body weight in the current study.Noakes (1978)suggested that 70%of the growth rate variation in cultured fish may result from social interactions.That leaves 30%for all the other factors including those accounted for in the model.It appears that our data supports the hypothesis and data of Noakes (1978).

Mass selection for increased body weight has not always been successful in other fish species.Response to selection has been quite variable in O.niloticus and several studies obtained a lack of response to selection for increased body weight (Tave and Smitherman,1980;Hulata et al.,1986;Teichert-Coddington and Smitherman,1988;Huang and Liao,1990).Hulata et al.(1986)indicated that the lack of additive genetic variation in O.niloticus was possibly due to random genetic drift or inbreeding.The O.niloticus used by Tave and Smitherman (1980)and by Teichert-Coddington and Smitherman (1988)were shown to be highly homozygous (Brummett et al.,1988).A small founder population followed by two severe reductions in effective breeding number might explain the reduction in genetic variability,or O.niloticus may inherently have limited genetic variation for growth rate.Negative response to selection was observed in red tilapia and was attributed to epistatic reactions of the red color gene complex (Behrends et al.,1988).Moav and Wohlfarth (1976)also showed that mass selection did not produce improved growth rate in common carp,Cyprinus carpio .

The moderate cumulative realized heritability for increased body weight for three generations of selection and the moderate realized heritability for the third generation alone for increased body weight in channel catfish suggest that the continuation of the selection program will result in further improvement.Cumulative h 2in the Kansas and Marion strains was lower than h 2for the first generation (Dunham,1981;Dunham and Smitherman,1983b)and for the third generation because of the lack of or negative response to selection in the second generation.

The 25%increase in body weight gained through selection is comparable to the growth improvement obtained through interspecific hybridization between channel catfish females and blue catfish,I.furcatus ,males (Yant et al.,1975;Dunham and Smitherman,1987;Dunham and Liu,2002).The disadvantage of the hybrid is the difficulty in mating the two species;however,recent advancements in artificial spawning of catfish (Lambert et al.,1999;Dunham et al.,2000)should lead to commercialization of the channel-blue catfish hybrid.The body weight improvement gained through selection was greater than that reported for intraspecific crossbreeding of channel catfish (Dunham and Smitherman,1983a).

Further generations of selection for increased body weight in channel catfish should be evaluated since a selection plateau has not been reached.The rate of improvement appears less than that obtained in rainbow trout,O.mykiss (Kincaid et al.,1977),and that obtained in coho salmon (Hershberger et al.,1990),and more than that for common carp (Moav and

M.A.Rezk et al./Aquaculture 228(2003)69–79

M.A.Rezk et al./Aquaculture228(2003)69–7977 Wohlfarth,1976).Hershberger et al.(1990)accomplished more than60%improvement in body weight of coho salmon over the wild control through mass and family selection,but indicated that a significant amount of this improvement was due to a domestication effect. In general,selection has been very successful for increasing body weight in fish.An average improvement of6–7%per generation(7–10%in the current study)has been obtained for most species,but some species such as Atlantic salmon,coho salmon,channel catfish,rohu,L.rohita,and Nile tilapia exhibited as much as11–14%improvement per generation by utilizing mass selection,family selection or both(Dunham et al.,2001).

5.Conclusions

This study demonstrates that body weight of channel catfish can be improved by multiple generations of mass selection.Catfish production can be increased21–29%by three generations of mass selection of body weight.This improvement of growth would result in60–90million kg additional production if applied in the current catfish industry. Selection should also be combined with crossbreeding and hybridization to evaluate the possibility of additional improvement in growth rate from utilizing multiple genetic improvement programs since each of the three programs has been successful. Acknowledgements

This paper is Alabama Agricultural Experiment Station journal number8-985882. References

Behrends,L.L.,Kingsley,J.B.,Price III,A.H.,1988.Bidirectional-backcross selection for body weight in a red tilapia.In:Pullin,R.S.V.,Bhukaswan,T.,Tonguthai,K.,Maclean,J.L.(Eds.),The Second International Symposium on Tilapia in Aquaculture.ICLARM Conference Proceedings,vol.15,pp.125–133. Bondari,K.,1980.Cage performance and quality comparisons of tilapia and divergently selected channel catfish.

Proceedings Annual Conference Southeastern Association Fish and Wildlife Agencies34,88–98. Bondari,K.,1983.Response to bidirectional selection for body weight in channel catfish.Aquaculture33,73–81. Bondari,K.,Dunham,R.A.,Smitherman,R.O.,Joyce,J.A.,Castillo,S.,1983.Response to bidirectional selec-tion for body weight in blue tilapia.In:Fishelson,L.,Yaron,Z.(Compilers),Proceedings of the International Symposium on Tilapia in Aquaculture.Tel Aviv University,Tel Aviv,Israel,pp.300–310.

Brummett,R.E.,1986.Effects of genotype?environment interactions on growth,variability and survival of improved catfish.Doctoral dissertation,Auburn University,AL.

Brummett,R.E.,Halstrom,M.L.,Dunham,R.A.,Smitherman,R.O.,1988.Development of biochemical dichot-omous keys for identification of American populations of Oreochromis aureus,O.mossambicus,O.niloticus, O.urolepis hornorum and red tilapia.In:Pullin,R.S.V.,Bhukaswan,T.,Tonguthai,K.,Maclean,J.L.(Eds.), The Second International Symposium on Tilapia in Aquaculture.ICLARM Conference Proceedings,vol.15, pp.135–141.

Chappell,J.A.,1979.An evaluation of twelve genetic groups of catfish for suitability in commercial production.

Doctoral Dissertation,Auburn University,AL.

Doyle,R.W.,Talbot,A.J.,1986.Artificial selection on growth and correlated selection on competitive behaviour in fish.Canadian Journal of Fisheries and Aquatic Sciences43,1059–1064.

Dunham,R.A.,1981.Response to selection and realized heritability for body weight in three strains of channel catfish grown in earthen ponds.Doctoral dissertation,Auburn University,AL.

Dunham,R.A.,Brummett,R.E.,1999.Response of two generations of selection to increased body weight in channel catfish,Ictalurus punctatus compared to hybridization with blue catfish,I.furcatus ,males.Journal of Applied Aquaculture 9,37–45.

Dunham,R.A.,Liu,Z.,2002.Gene mapping,isolation and genetic improvement in Catfish.In:Shimizu,N.,Aoki,T.,Hirono,I.,Takashima,F.(Eds.),Aquatic Genomics:Steps Toward a Great Future.Springer-Verlag,New York,pp.45–60.

Dunham,R.A.,Smitherman,R.O.,1983a.Crossbreeding channel catfish for improvement of body weight in earthen ponds.Growth 47,97–103.

Dunham,R.A.,Smitherman,R.O.,1983b.Response to selection and realized heritability for body weight in three strains of channel catfish,Ictalurus punctatus ,grown in earthen ponds.Aquaculture 33,89–96.

Dunham,R.A.,Smitherman,R.O.,1984.Ancestry and breeding of channel catfish in the United States.Circular,vol.273.Alabama Agricultural Experiment 349Station,Auburn University,AL.

Dunham,R.A.,Smitherman,R.O.,1985.Improved growth rate,reproductive performance and disease resistance of crossbred and selected catfish from AU-M and AU-K lines.Circular,vol.279.Alabama Agricultural Experiment Station,Auburn University,AL.

Dunham,R.A.,Smitherman,R.O.,1987.Genetics and breeding of catfish.Southern Cooperative Series Bulletin 325,Southern Regional Research Project S-168.Alabama Agricultural Experiment Station,Auburn Univer-sity,AL.

Dunham,R.A.,Joyce,J.A.,Bondari,K.,Malvestuto,S.P.,1985.Evaluation of body conformation,composition and density as traits for indirect selection for dressout percentage of channel catfish.Progressive Fish-Culturist 47,169–175.

Dunham,R.A.,Lambert,D.M.,Argue,B.J.,Ligeon,C.,Yant,D.R.,Liu,Z.,https://www.wendangku.net/doc/eb12894167.html,parison of manual stripping and pen spawning for production of channel catfish ?blue catfish hybrids and aquarium spawning of channel catfish.North American Journal of Aquaculture 62,260–265.

Dunham,R.A.,Majumdar,K.,Hallerman,E.,Hulata,G.,Mair,G.,Bartley,D.,Gupta,M.,Liu,Z.,Pongthana,N.,Rothlisberg,P.,Horstweeg-Schwark,G.,2001.Status of aquaculture genetics and prospects for the third millenium.Proceedings Conference on Aquaculture in the Third Millenium,Bangkok,pp.129–157.

EI-Ibiary,H.M.,Joyce,J.A.,1978.Heritability of body size traits,dressing weight,and lipid content in channel catfish.Journal of Animal Science 47,82–88.

Falconer,D.S.,1989.Introduction to Quantitative Genetics.Longman,Harlow,England,UK.

Gjedrem,T.,1983.Genetic variation in quantitative traits and selective breeding in fish and shellfish.Aqua-culture 33,51–72.

Gjerde,B.,1986.Growth and reproduction in fish and shellfish.Aquaculture 57,37–55.

Hershberger,W.K.,Myers,J.M.,Iwamoto,R.N.,McAuley,W.C.,Saxton,A.M.,1990.Genetic changes in the growth of coho salmon (Oncorhynchus kisutch )in marine net-pens,produced by ten years of selection.Aquaculture 85,187–197.

Huang,C.-M.,Liao,I.-C.,1990.Response to mass selection for growth rate in Oreochromis niloticus .Aqua-culture 85,199–205.

Hulata,G.,Wohlfarth,G.W.,Halevy,A.,1986.Mass selection for growth rate in Nile tilapia (Oreochromis niloticus ).Aquaculture 57,177–184.

Kincaid,H.L.,Bridges,W.R.,V on Limbach,B.,1977.Three generations of selection for growth rate in fall spawning rainbow trout.Transactions of the American Fisheries Society 106,621–625.

Kinghorn,B.,1983.A review of quantitative genetics in fish breeding.Aquaculture 31,283–304.

Lambert,D.M.,Argue,B.J.,Liu,Z.,Dunham,R.A.,1999.Effects of seasonal variations,thyroid and steroid hormones and carp pituitary extract on the artificial production of channel catfish (Ictalurus puntatus )?blue catfish (I.furcatus )hybrid embryos.Journal of the World Aquaculture Society 30,80–89.

Mahon,G.A.T.,1983.Selection goals in oyster breeding.Aquaculture 33,141–148.

Moav,R.,Wohlfarth,G.,1976.Two-way selection for growth rate in the common carp (Cyprinus carpio L.).Genetics 82,83–101.

Noakes,D.L.G.,1978.Social behavior as it influences fish production.In:Gerking,S.D.(Ed.),Ecology of Freshwater Fish Production.Blackwell,Oxford,pp.360–382.

M.A.Rezk et al./Aquaculture 228(2003)69–79

M.A.Rezk et al./Aquaculture228(2003)69–7979 Reagan,R.E.,1979.Heritabilities and genetic correlations of desirable commercial traits in channel catfish.

Mississippi Agriculture and Forestry Experiment Station Resource Report.Stoneville,MS.

Rezk,M.A.,1993.Response and correlated responses to three generations of selection for increased body weight in channel catfish,Ictalurus punctatus,grown in earthen ponds.Doctoral dissertation,Auburn University,AL. SAS Institute,1991.SAS/Stat Users Guide Release6.06Edition.SAS Institute,Raleigh,NC.

Steel,R.G.,Torrie,J.H.,1980.Principles and Procedures of Statistics:A Biometrical Approach.McGraw-Hill, New York.

Tave,D.,Smitherman,R.O.,1980.Predicted response to selection for early growth in Tilapia nilotica.Trans-actions of the American Fisheries Society109,439–445.

Teichert-Coddington,D.R.,Smitherman,R.O.,https://www.wendangku.net/doc/eb12894167.html,ck of response by Tilapia nilotica to mass selection for rapid early growth.Transactions American Fisheries Society117,297–300.

Thien,C.M.,1971.Estimation of realized weight heritability in tilapia(Tilapia mossambica Peters).Genetika7, 53–59(in Russian with English abstract).Cited from:Siitonen,L.,1986.Factors affecting growth in rainbow trout(Salmo gairdneri)stocks.Aquaculture57,185–191.

Warwick,E.W.,Legates,J.E.,1979.Breeding and Improvement of Farm Animals.McGraw-Hill,New York. Waters Jr.,P.J.,2001.Response and correlated response to selection of channel catfish,Ictalurus punctatus,for resistance to Edwardsiella ictaluri,Flexibacter columnaris,low levels of dissolved oxygen and toxic levels of nitrite.Master of Science thesis.Auburn University,AL.

Yant,R.,Smitherman,R.O.,Green,O.L.,1975.Production of hybrid(blue?channel)and channel catfish in ponds.Proceedings Annual Conference of the Southeastern Association of Game and Fish Commissioners29, 82–86.