I.

II.

III.

IV.

v.

VI.

VII.

IX.

TABLE OF CONTENTS

Foreword ..................................Carlos M. Fetterolf, Jr., Great LakesFishery Commission

Solicitation of ideas and suggestions for astock concept symposium/workshop . . . . . . . . . . . . . . . . . .......A, H. Berst, Ontario Ministry ofNatural Resources

The stock concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .K. H. Loftus, Ontario Ministry of NaturalResources.1976.

An excerpt from Loftus, K. H.Science for Canada's fisheries re-

habilitation needs. J. Fish. Res. Boardcan. 33:1822-1857.

Fish genetics and selective breeding ...................K. H. Loftus, Ontario Ministry of NaturalResources. An excerpt from Loftus, K. H.1976. Science for Canada's fisheries re-habilitation needs. J. Fish. Res. Boardcan. 33:1822-1857.

History of fish genetics and cytogenetics:alpha, beta, gamma, and resolution .....................Henry E., Booke, Wisconsin CooperativeFishery Research Unit, U.S. Fish andWildlife Service, University of Wis-consin, Stevens Point, Wisconsin

Physiological and Behavioral Genetics and theStock Concept for Fisheries Management . . . . . . . . . . .Peter Ihssen, Fish and Wildlife ResearchBranch, Ministry of Natural Resources,Province of Ontario

.The LDH isozyme system of salmonids . . . . . . . . . . . . . . . . .

Edward Massaro, Medical School, Departmentof Biochemistry, State University of NewYork at Buffalo, Buffalo, New York

Genetic variations in populations of fish •-*~~*****-******Fred Allendorf, Department of Zoology,University of Montana, Missoula, Montana

Selective breeding: the ?Zipecialist" approachvs. cross-breeding: the %eneralist" approach ..........

Raymond Simon, U.S. Fish and WildlifeService, Fish Genetics Laboratory, Beulah,w-'iw

13

15

27

31

35

41

X. Summary and recommendations - roundtablediscussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . &?

XI. Species nuumgament . . . . ..*.......*..*.....................55D. A. Webster and W. A. Flick, Departmentof Natural Resources, Cornell university,Ithaca, New York. Reproduced from "Pro-ceedings of the wild Trout ManagementSymposium, Trout Unlimited, Inc., 1975.

FOREWORD

The 1976 Report of the Lake Ontario Committee (LOC) to theGreat Lakes Fishery Commission (GLFC) included the followingrecommendation:

Genetic Integrity of Fish Stocks:

A. There is concern that current stockingpractices in Lake Ontario will lead togenetic dilution and impairment ofestablished self-reproducing stocks.

It is recommended that a policy state-ment based on the "Stock Concept" bedeveloped by the GLFC and that theScientific Advisory Committee beinstructed to examine the problem,specify guidelines and references forthe selection of naturally-reproducingstocks, and where necessary, recommendappropriate research studies. It issuggested this subject could well war-rent a GLFC-sponsored symposium-workshop.

B.* The LOC requests the Commission contact each agency that has stocked fish in theGreat Lakes and request they provide allavailable information on the origin andknown characteristics of those stockedfish. The compiled information shouldbe in report form for the LOC 1977 annualmeeting.

In responding for the Commission at the Annual Meeting, June1976, Chairman Loftus-noted that similar recommendations had beensubmitted by the Upper Great Lakes Committees, acknowledged that the Commission is in full accord with the expressed concern, andstated-that the Colmnfssion is con&idering initiation of a large-scale binational workshop or symposium to address. the whole ques-tion of the stock concept end needs for research in fish genetics

%rt B of this recommendation is similar to a request fromthe Lake Michigan Lake Trout Technical Committee (LMLTTC) forwardedto the GLFC through the L&e Michigan Committee,. Cataloging suchinformation for lake trout has been an ongoing activity of the Com-mission for many years, but the requests of the IELTTC and the LCCresulted in renewed efforts to keep the information timely and inmore usable form.

1

2

and selection as applied to rehabilitation of Great Lakes fishatocka. The Commission then charged the Scientific AdvisoryCommittee (SAC) to investigate the feasibility of such a workshop.

As of November 1977 the genetic origins of lake trout stockedin Lake Michigan have been summarized and mapped in &aft form foruse by the LMLTTC, similar summarization has been initiated for LakeSuperior lake trout, and information gaps in our records for splakeand lake trout plants in lakes Huron, Erie, and Ontario are beingfilled. Plans are to summarize this information for all Great Lakeslake trout and splake plants. Similar efforts will be expended forother salmonids as appropriate.

Reporting at the Commission's Interim Meeting, December 1.976,the SAC proposed an exploratory session of invited papers supportedby the GLFC and co-sponsored with the American Fisheries Society atthe Annual Meeting of the International Association for Great LakesResearch, May 1977, with Dr. H. T. Booke, U.S. Fish and WildlifeCooperative Fishery Research Unit, University of Wisconsin--StevensPoint, as Convener. The Commission endorsed this proposal andappointed an ad hoc committee of Commissioner K. H. Loftus (OntarioMNR) end Alternate Commissioner J. Hemphill (USFWS) to work withDr. Booke to organize a session on the fundamentals of genetics asapplied to fish and their interaction with their environment.

This session was to be a prelude to the proposed binationalworkshop. The Commission agreed to support publication of thepapers in the Journal of Great Lakes Research if the participantsdesired, but the main thrust of the effort was to supply backgroundinformation and a resource base from which to consider further thedesign of the stock concept workshop.

The participants decided against formal publication, andagreed that the most practical way of dissemination of thematerials was through informal distribution by the Commission.The invited presentations should not be judged harshly from aneditorial viewpoint. They are author-edited talks for informa-tion-purposes. Considering the purpose of this document--provi-sion of a resource base--two other pieces have-been added:excerptsfrom Kenneth H. Loftus' paper,- "Science for Canada'sFisheries Rehabilitation Reeds!-:(J.. Nsh. Res. Board Canada 33:1822-1857); and "Species Management" by Dwight A. Websterand William A. Flick, originally Ijubliahed by Trout Unlimitedin their 1975 publication, "Proceedings of the Wild TroutManagement Sympo~siumfl;- The excerpts from the Loftus paper areplaced early in the document because they reflect the concern ofthe Commission in general terms.

3

The SAC will report their progress to the Commission at theInterim Meeting, December 1 and 2, 1977. At this time the SAChas asked A. H. Berst, Research Scientist, Fisheries Section,Fish and Wildlife Branch, Ontario Ministry of Natural Resources,to organize a meeting to discuss the proposed stock concept workshopand set in motion the preparatory stages. His correspondencesoliciting ideas and suggestions follows.

The Commission is particularly grateful to authors Dr. HenryE. Booke, University of Wisconsin at Stevens Point; Dr. PeterIhssen, Ontario Ministry of Natural Resources; Dr. Edward Massaro,University of New York at Buffalo; Dr. Fred Allendorf, Universityof Montana; and Dr. Raymond Simon, U.S. Fish and Wildlife Servicefor preparing their presentations on short notice; to Henry Bookefor his efforts to organize the session; to the InternationalAssociation for Great Lakes Research and their Program Chairman,Stan Bolzenga, for accommodating us; to authors Ken Loftus, DwightWebster, and Bill Flick, JFRB Canada Editor J. C. Stevenson andTrout Unlimited for permission to include the reprinted material;to Jane Herbert for recording and transcribing the session; andto Trudy Woods, Becky Andress, and Pat Lindvay for secretarialservices in putting the document together.

56DOntario

Ministry ofNatural

Resources

Box 50Maple, OntarioLOJ 1EO

Our file number

Your file number

1977 10 07

Mr. Carlos Fetterolf,Executive SecretaryGreat Lakes Fishery Commission1451 Green Rd.Ann Arbor, Michigan 48105

Dear Carlos:

Please find enclosed, a copy of a self explanatorymemorandum to scientists in the Fisheries Section ofthe Fish and Wildlife Research Branch. I would appre-ciate if you would forward copies of the memorandum toU.S. inlividuals and/or agencies whom you feel nightcontribute useful ideas and suggestions.

By copy of this letter, I am informing Dwight Websterand trust that he may see fit to send copies independ-ently to U.S. people.

Thank you.

-Yours very truly

Ontario

Ministry ofNaturalResources

1977 10 07

!?E:IOEA:lDUI~ TO:

Our file number

Your file number

All Research ScientistsFisheries SectionFish and Wildlife Research Branch

SUBJECT: Proposed Conference on Stock Concept.

I have been asked by the Scientific Advisory Committee ofthe Great Lakes Fishery Commission to organize a meetingin the near future on the above subject. The purpose ofthe meeting will be to set in motion the preparatory stagesfor a conference to be held some time *within the next coupleof years.

I believe that the paper "Science for Canada's fisheries re-habilitation needs" by Ken Loftus (J. Fish. Res. Bd. Canada33: 1822-1857) provides a basis of logic in support of sucha conference. The stock concept is obviously a component ofthe current management strategy for west coast salmon fish-eries. Many of us feel that it is also relevant to themanagement of freshwater fisheries.

Perhaps the objectives of such a conference would include(i) a review of evidence in support of the stock concept andits impact on marine fisheries, (ii) a review of documentedevidence and presentation of new evidence of the stock con-.cept in freshwater fisheries including salmonids, percids,coregonids, and centrarchids and (iii) a forecast of how theapplication of the stock concept in the strategy for manage-ment of freshwater fisheries would influence research and

management programs.

I would greatly appreciate your views on the above subject,and would especially appreciate knowing of any significantreferences (other than those cited in Lcf‘tus' _nac?cr) and/orthe nezes of any scientists either in research or management,who are directly or indirectly involved in this field.

8

Page 2All Research Scientists1977 10 07

Since there is a cocnitcent to hold the initial neetinqand report to the Scientific Advisory Concittee by 1977 12 01,I would a?Treciate your reply as soon as possible.

A.H. Berst, Research ScientistFisheries SectionFish and Wildlife Research Branch

C.C. K.H. LoftusJ.D. RoseboroughA.M. LawrieH.A. Regier

9

THE STOCK CONCEPT

K. H. Loftus, DirectorFisheries Branch

Ontario Ministry of Natural ResourcesParliament Buildings

Toronto, Ontario, Canada M7A 1W3

The follouing is excerpted frcm %cience for Canada's fisheriesrehabilitation needs," IL H. Loftus, 1976, Journdl of the FisheriesResearch Board of CsnMa 33:1822-1857 and reprodwed with the per-mission of the author and JFRB Canada Editor J. C. Stevenson.

The Stock ConceptThe reality of discrete spawning populations

has been recognized for several years as an im-portant consideration in the management of westcoast salmon. There has been an explicit attemptto harvest each recognized stock at the differentlevels appropriate to their continued productivity.In practice. however, this laudable and necessaryobjective has been difficult to achieve with morethan general precision because of the mixing ofstocks at sea, and because most of the harvestinggear dots not discriminate between stocks. Onlywhen the stocks are close inshore and near their“home” stream do they become adequately sepa-rated to allow an opportunity for fully selectiveharvesting. By the time this sort of separationoccurs there is little. if any. opportunity for har-vest by the traditional gear, and furthermore. thequality of the fish may have seriously deteri-orated.

In spite of the difficulties, the objective hasbeen partially achieved and, in the process, muchadditional information on the movements and

distribution of stocks in the ocean has beengenerated.

Ricker (1972) presented a comprehensive re- view of the knowledge of discrete stocks of west

coast salmon and trout. Impressive indeed is thelisting and description of discrete stocks within

species.. and parti&larty within single river SYS-terns. The question whcthcr genetic and/or en-vironmental infiucnces are involved in the emer-gence of discrete stocks is vital to managementand to rehabilitation. The cpiloguc of Ricker’sexhauLtivc work is, thcreforc, quoted here.

“In almost all cases where both genetic andenvironmcntai influtnces effe&ng natural

s t o c k dillcrenccs a m o n g P a c i f i c s a l m o n a n dsteelheads have been se:,rcheci for adequately,both have been found: though sometimes one.somet imes the o ther . i s rclativcly wcnk. o r i s

infrequently expressed.“Since season of return to the river is stronglyunder genetic control. and ability IO ‘home’ isapparen t ly under gcnctic con t ro l . a coro l l a ryis that mort or $1 of the identified stocks of

these fish diAer geneGcnll~ 10 come extent.“ S u c h conc!usions somctimcc cncountcr c r i t i -cism on the grounds that the)’ are too compii-cated. Nature (the ;irgmlent runs1 should notbe made more complex than it really is:species can exis t in one and the same r iversystem, a dozen or so even within a single lake- particularly when it is known that there issome interchange between them.“Certainly when t:\o or more hypotheses arcequally possible, it is customary to prefer thesimplest one. But how are we to know whichone is simplest? \Vhnt is the test of simplicit)?Wirh respect to any recurr ing d i f ference be-tween two salmon stocks, NC have three nltcr-natives:“1. The difference is completely environmen-tally determined.“2. The difference is completely hereditarilydetermined.“3. The difference is determined partly byheredity, partly by cnvironmcnt.“ S u p e r f i c i a l l y a t l e a s t . h y p o t h e s i s ( 1 ) a n dhypothesis (2) both seem simpler than hypoth-.esis (3); perhaps they really are so.“If we were to reject hypothesis (3) becauseit is too compiicntcd. then which of (1) and(2) is the sknplcr? Some writers have seemedto be l ieve tha t (1 ) i s s impler thnn (21. o r ntleas t ‘more corrservntive’ than (2 ) , nppnrcnflybecause it implies n more homogeneous geneticconstitution among the various popularions ofa species. For esnmple. if it has been denion-strafed that the number of sinle5 on fish of nc e r t a i n spel;ics i s nfTec~ed b y environmcnrnlt e m p e r a t u r e d u r i n g tie~~elopment. t h e y w i l lthen decide thnt ‘thcrc is no need 10 postulnte’any genet ic bas is icr nn observed dillrrence ins c a l e n u m b e r bct\\een t\\o loc;ll srocks o f ihntspecks.“But ohnt if UC\ were 10 2ypiy the snrne pro-ccdure i n revcr~-r? 5up;o52 \VC hn\e tkmon-strnted n hrrcilit.~rl!\ -dc:tiiniiicd <!il?Cr<nzi’ i nsc;~ls c0iir.t bcl\vc-n ~UU rroz;.s. 2s D r . SC.;\Xd i d f o r Str/~:o ::::ir,/r:~rr. chorlid \\e 11:c:: &cideth;lt t h e r e k n o ~CCII t o ro\r:~l:~!c :\II~ ~!!Lw (4the environnir~nl ‘\.‘I, CC.ilC 1l~llllbCl i n Ihal

s p e c i e s ? Tilnt nrviou~lv i\n‘: loric;ll; &VII rhcnt h e con\-e~s propozitlan c:tnnot b c logicnIeither.

10

“My strong opinion is that we should avoidany nppe:nl to simplicity or conservatism insuch questions. Time and again it has beendiscover4 that nature is more complex thananyone dreamed possible. Hence we shouldstick to whatever evidence is available, how-ever sketchy it may be.“In the matter at hand, the- evidence availableis now quite considcroblc. It indicates thatmost of the studied differences bc~.vccn~localstocks can and usually do have both a geneticand an environmental basis. Be it simple orcomplex, this should now be our normal ex-pectation in respect to any as-yet-unstudieddifference between stocks of salmon or trout.Only direct investigation can show which typeof influence predominates in a particular situa-tion.”

Larkin (1972) provides a view of the manage-ment implications of the stock concept. The dif-ficulties in achieving selective harvest of discretestocks are emphasized.

Available evidence strongly suggests that fullutilization of complex river systems by salmon,giving maximum production of young salmonfrom the system, and maximum returns to it, isachieved only when a complex of stocks is pres-ent. Salmon rehabilitation projects face the taskof optimizing reproduction on a stock-by-stockbasis and of learning how to recognize or tocreate stocks comparable to those which havealready been lost. Care may be necessary toavoid overloading parts of the river system to thedetriment of other stocks and projects will re-quire careful balance between stocks and carefulevaluation.

The significance of discrete spawning stocks isrecognized now as well in Atlantic salmon(Saunders 1967; Moller 1970; A. W. H. Needler,personal communication). Probably because ofthe rclntivcly small size of the total resourceand because it is comprised of small runs scat-tcrcd over so much coastline. less detail isavailable to describe these stocks. Less is knowntoo of the movements of separate stocks atsea, witness the recent surprise off Greenland

which precipitated harvest restrictions. It stemsprobnblc that many stocks have already been lostand that rchabilitntion will be the more difficultfor that r&son.

. In the frcshwatcr arca littlc attention has beengiven to the significance of discrete spawningstocks. I am now-f the opinibn that our in+tten-

tion in this rcspcct may_ hive rendered some ofour mana~cmcnt inclfcctivc, and in 3ome cxcswhere rouiinc plantings h:lvc been invoivcd. cvcn

countciyroclllcti\c. A few csamplcs may Serveto cstahlish the possibility of parnllcls in somefrcshwntcr spccics to the stock co~cpt dcsc:ibcdfor Pacific salmon.

. Martin (1957; 1960) described very prccizc_. ._ homing behavior in lake trout at spawning time

.,

.

in 14SO-ncrc Louisa Lake whcrc the spawninggrounds 4crc scparatcd bv only a t’cw wrds. Hisconlirmin_c cvidcncc of &scrclc spnGin_r timesand within-lake locations for lnkc trout in a num-bcr of Algonquin Park IilkCS (personal conimuni-cation) is cstcnsive. 1.0itus ( 19%) &scribedpopulations of river-spn\vning I&c trout in LakeSuperior. The distribution of thcsc trout stocksovcrlappcd in the open lake but they scparatcdyear nftcr year and moved to the “home”(presumably nata!) rivets. He rccordcd some ad- -ditional spawning times and places for LakeSuperior lake trout as rcportcd by fishcrmcn andhatchery pcnonncl. Eschmeyer (1955) in dis-cussing lake trout reproduction in southern LakeSuperior suggested that take trout returned to thesame spawning shoals year after ycnr. J. B. Smith(1965) wc able to map more than 150 separateformer lake trout spawning grounds on the basisof rccollcctions by commcrcinl fishermen. Lawrieand Rahrer (1972) and Rcgicr and Loftus(1972) considered that the system of accumu-lating catch statistics in Lake Superior and inother lakes. a system which did not rcllcct stock-by-stock distribution, scrvcd to mask the realityof the fishing-up process to the cxtcnt that somestocks may have been decimated long beforetotal harvest ligurcy showed a dcclinc. In manyinstances, the trout rehabilitation program in theGreat Lakes implicitly. but not esplicitly. recog-nizes the stock concept. Successful rehabilitationmay be delayed by the lack of appropriate broodstocks. Olver and Lewis (1971) were able to findon!y a few csamplcs of planted lake trout whichsuccessfully reproduce thcmsclves.

Whitefish in Lake Huron are comprised of ap-parently discrctc stocks. Two are recognized byJ. J. Collins (personal communication) in theNorth Channel: two in South Bay (Budd 1957):two in Georgian Bay (Cucin and Regier 1965),and two in Lake Huron proper (Budd and Cucin1962; Spangler 197-l). Thcsc stocks may be thefew remaining from a larger original number ofstocks, Christie (1972) recognized two stocks ofwhitefish in Lake Ontario, one of which now ap-pears to be extinct.

Walleye literature also suggests strong parallelsto the stock concept. Lake Eric contains two ap-parently discrete stocks, one which reproduced inthe Western Basin and which has suffered asevere decline in recent years; another which rc-produces in the eastern basin and which hasremained rclativcly stable. Ferguson and Derk- .sen (1971) &scribed mixing of a Lake St. Clairwalleye population with those of southern LakeHuron and of western Lake Eric. In the Bay ofQuinte. Lake Onrario, Payne (1963) studied apopulation. then flourishing. now virtually cx-tinct. \Vithin fhc sync bny ccrtnin riser spn\vningpopulations persist at low lcvc’ls (W. J. Christic.pcrsonnl communication). In Lake Huron a

11

number of Jiscrctc rircr sp;lwiing wnllcyc stocksarc rccopnizcd in the Xloon. Shawanaga, French.Pickcrcl. hlississauga. and Echo rivers. Othersno dortbt.occttr. The same story is probably veri-fiablc for stocks in Lake Superior’s Goulais.Batchawana. and hlichipicotcn rivers, and Ryder(1965) recorded the demise of one such popula-tion near Red Rock in Nipigon Bay. The work ofUthc and Ryder (1970) presents further evi-dence that the stock concept may bc valid forwalleyes.

Rainbow trout. a species introduced to theGreat Lakes several decades ago have now be-come naturalized and appear to exhibit “stock”characteristics similar to those described forsteelhead on the west coast.

To date the significance of discrete spawningstocks. the stock concept of west coast salmon,appcars to have failed to attract the explicit at-tention it merits for management and rchabilita-tion throughout the freshwater area.

It is recognized that many stocks are seriouslydepressed and that some have already been lost.There are a number of causes for these losseswhich need to be more widclv recognized, betterunderstood, and guarded against. They include( 1 ) direct excessive exploitation, (2) incidentalcatch by gear directed at adjacent stocks orspecies which are in good supply, (3) geneticdrift imposed by the selectivity of gear fishingfor a particular size. This may tend to removeearly maturing and/or fast growing individualsfrom the stock, (4) hatchery plantings from un-selected stock origins may survive to maturityand mix with native stock during spawningthereby diluting the gent pool of a well-adapted

stock, (5) environmcntai change may force de-segregation at spalvning of formerly discrete butclosely related stocks causing intcrbrccding. Nodoubt a number of other mechanisms could con-tribute to loss of stocks essential to the full use ofan ecosystem.

Over several thousands of generations, mostspecies appear to have evolved discrete stocksadapted to similar yet discrete and specifichabitats. Fisheries science is not yet able to recog-nize either the habitat difTcrcnces or the specificcharacrcrisrics which dilicrentiate stocks. Man’sattempts to date to move stocks from one placeto another have been mostly unsuccessful at leastin terms of achieving natural reproduction by thetransplanted stock. Given that a naturally evolvedcomplex of stocks is apparently essential to fulluse of the productive capacity of a river system.and perhaps of lakes. it follows that it is impor-tant to learn to recognize and to preserve repre-sentative stocks. It is also important to learn howto reestablish stocks which have been decimatedor lost, anti to achieve a natural balance amongstocks. There is a strong implication that rehabili-tation requires the cuiture and planting of nu-merous small lots of the appropriate stocks untilnaturaf reproduction is reestablished. Such plant-ing requirements suggest a need for small. versa-tile, and mobile hatchery facilities rather than themajor “efficient” fish factory type of facility cur-rently in favor.

It may be necessary to establish a national genepool which would provide for the preservation ofrepresentative, particularly valued or endangeredstocks.

BUDD, J. C. 1957, Movements of tagged whitefish in northern LakeHuron and Georgian Bay. pans. h. Fish. Soc. 86: 128-l%.

BUDD, J. C., and D. CUCIN, 1962. Exploitation of Canadian LakeHuron whitefish. Trans. Am. Fish. Soc. 91: 223022h.

CHRXSTIE, W. J. 1972. Lake Ontario: effects of exploitation,introductions and eutrophication on the salmonid community.J. Msh, F&m, Board Can, 29: 913-919.

CTEIN, I)., &d H. A. REGIER. 1965. Dpamlca and exploitation oflake whitefish in southern Georgian Bay. J. Fish. Res. BoardCan. 23: 221-274~

ESCHMFPER, P. H. 1955. The reproduction of lake trout in southernLace Superior. Trans. h. Fish. Soc. 84: 47-74.

FERGUSON, R. G., and A. J. DERKSEN. 1971. Migrations of adult andjuvenile walleyes (Stizostedion vitreum vitreum) in southernLake Huron, Lake SL Clair, Lak~Zsnd-connect~~g waters.J. Fish. Res. Board Can. 28: 1132-l lb2.

12

LARKIN, P. A. 1972. The stock concept and management of Pacificsalmon. p. 11-15. In The stock concept in Pacific s&on,H. FL MacMillan LectEs in Fisheries. Univ. British Columbia,Vancouver, B.C.

LAWRIE, A. H., and-J.exploitation and

F. RAHRER. 1972. Lake Superior: Effects of

Fish, Res. Boardintroductions on the salmonid community, J.Can. 29: 765-776.

LOFTUS, K. H. 1958.trout in eastern259-277.

Studies of river-spawning populations of lakeLake Superior. Trans. Am. Fish. Soc. 87:

MARTIN, N. V. 1957.Ontario. Trans.

Reproduction of lake trout in Algonquin Park,Am. Fish. Soc. 86: 23102h.b.

MARTIN, I. V, 1960. Homing behaviour in spawning lake trout. Can.fish. Cult. 26: 34.

MOLLER, D. 1970. Transferring polymorphism in Atlantic salmon(Salmo salar). J. Msh. Res. Board Can. 27: 16174625.

OLVER, C. H., and C. A. LEWIS. 1971. Reproduction of planted laketrout in Gamitagama, a small precambrian lake in Ontario. Ont.Dep. Nat. Resour. 12 p.

PA&', N. R. 1963. The life history of the yellow walleye in theBay of Quinte. M.A, Thesis, Univ. Toronto, Toronto, Ont. &O p.

REGIER, H. A., and K. H. LOFTUS. 1972. Effects of fisheries ex-ploitation on salmonid communities in oligotrophic lakes. J.Fish. Res. Board Can. 29: 959-968.

RICKER, W. E. 1972. Hereditary and environmental factors affectingcertain salmonid populations. p. 19-160. In The stock conceptin Pacific salmon. H. B, MacMillan Lectures in Fisheries.&iv. British Columbia, Vancouver, B.C.

RYDER, R; A, 1968, I&a& cs and exploitation of mature walleyes(Stizostedion vitreum vitreum) in the Nipigon Bay region of

Lake Superior. J. Fish. Res. Board Can, 25: l&7-1376.

SAUNDERS, R. L. 1967. Seasonal pattern-of return of Atlantic salmon-- in the northwest Miramichi River, New Brunswick. J. Fish. Res.+imi f+n. j?b:.~l-3~. -

'SPAI@LER, 0. R. 1978. Mortality factors and dynamics of a Lake Huronlake whitefish population. Ph.D. Thesis. Univ. Toronto,Toronto, Ont. 87 p.

JJTHE, J. F,, and R. A._RYDEX. 1970, Regional variation in musclerqyogen~ polymorphism in walleye (Stizostedion vitreum vitreum)as -related to morphology. J. Fish. Res. Board Can. 27: 923-927.

13

FISH GENETICS ANDSEUCTNE BiUBZING

K. H. LoftusFisheries Branch

Ontario Ministry of Natural ResourcesParliament Buildings

Toronto, Ontario, Canada M7A 1W3

The following is excerpted from "Science for Canada's fisheriesrehabiUtation needs,' K. H. Loftus, 1976, JournkL of the FisheriesResearch Board of Canada 33: 1822-1857 and reproduced with the per-mission of the author and JFRB Canada Editor, J. C. Stevenson.

Fish gerretics and selective brectlinp - A num-ber of fish stocks on the west coast, cast coast,and in the freshwater area are depressed to criti-Cd levels or arc extinct. In the Great Lakes wheremultiple stresses have been evident. the discretestocks lost may number in the dozens and involvespecies such as lake trout, whitefish and othercoregonids, and walleye. In all areas. it seemsprudent to accept as fact the real possibility thatwe have lost many more stocks than can readilybe identified now. Most of these stocks have beenlost as a result of man’s intervention, directand/or indirect: fishing, industrializing. dam and/ot canal building, trying to reduce the nuisancefrom black flies. mosquitoes, or budworms, etc.We would be naive to assume that. having recog-nized the importance of discrete stocks. we willnow be able to organize the activities of competi-tive water users in such a way as to avoid the lossof additional stocks.

Former high abundance levels were supportedby communities of species, and by within-speciesstock complexes. Rehabilitation programs, there-fore. are directed toward the rebuilding of thesespecies and stock complexes to levels at whichthey are again able to reproduce eifrctiveiy. It isintended that management will be such as toallow them to maintain themselves through nat-ural reproduction. The implication here is thatrehabilitation of the Fraser or Saint John Riverecosystem, and of lakes Superior or Ontario re-quires the reestablishment of naturally spawningpopulations of each\of the species and stockswhich formerly made up the resource base inthose waters. It is recognized that habitat degra-dation by virtue of new species invasions, and bywater quality changes may have eliminated someof the precise habitats formerly occupied by somespecies or stocks. and that, therefore, we may beunable to fully regain the former levels of abun-dance. Aggressive habitat management programssuch as those undertaken by the Great LakesFishery Commission and the International JointCommission in the Great Lakes are needed tomake some of those habitats available again,while others must be considered lost. Havingaccepted that we are unlikely to fully regainformer abundance levels in our rehabilitation

programs, there are still major improvementsattainable.

In many instances where stocks persist. thoughat precariously low levels and where their habitatsremain healthy, the science ncccssary for theirrehabilitation appears to be availsblc. In caseswhere stocks are at extremely low levels. it maybc advisable. or nccessnry, to use sex productsfrom males only for cloning with females of aclosely related stock. as suggested by Calaprice(1969) and by Ihssen (1976) to obtain adequatenumbers of fish containing at least some of thegcncs necessary to effective rehabilitation. Furthercnrcful testing of this technique would be useful.

In many cases where stocks have become ex-tinct. and where they may go to extinction despiteour best efforts. and./or in those cases where thehabitats have been modified in some irreversibleway, the science required for rehabilitation is notyet available. A major new initiative of basicresearch in fish gcnctics and selective breeding isnecessary.

The necessary new research initiative may notbe difficult to mount because scattered existinginitiatives have already shown promise of success.In Ontario, for example. beginning in the late1950s under the guidance of F. E. J. Fry, anattempt has been made to create a new troutstock by hybridization and selective breeding,capable of occupying the vacant lake trout nichein Lake Huron. Hybrids were made between laketrout and brook trout, and the selection tech-niques to concentrate the deep-swimming laketrout character and the early maturing characterof brook trout (Tait 1970: Straight 1969) weredeveloped and applied through iofur or five gen-erations. The project has been successful in thelaboratory and large-scale field testing of theselected hybrid is currently under way in Lake .Huron. In this research an attempt was made tobuild a new trout to occupy the former lake trouthabitat which has been modified by the sealamprey invasion and the several fish communityconsequences of that event. It is suggested thatsimilar research projects may be necessary forother trout, salmon. whitefish. and walleye popu-lations tchel-e original stocks have vanished andwhere water management cannot be expected to

14

restore habitats of the quality required by theoriginal stocks.

More recently. work in selective breeding andhybridization has been initiated with salmon atSt. Andrews on the east coast, Nanaimo on thewest coast, and with rainbow trout at Winnipeg.Some of these studies reflect the anticipated needfor selection of special characters in aquacultureprograms rather than rehabilitation programs.

Whatever the specific objectives of the severalbeginnings that have been made, all have beencontributing to a better definition of the basicquestions that must be addressed by research.Some of these follow:

5.

1. What are the mechanisms or patterns of in-heritance for discrete characteristics? It will bcnecessary to select for certain characters (e.g.deep-swimming ability. or tolerance for lowoxygen levels) and at the same time avoid theincidental selection of an unwanted character(e.g. stream spawning).

of fry movement after emergcncc. crc’! it wasmentioned earlier that some stocks are thoughtto have been depressed through gcnctic dilutionfrom indiscriminant plantings of other stocksof the same species. Still the prospect of beingable to sclcct from elsewhere a stock appro-priate to the vacated niche of an extinct stockis intriguing. Further. if SCS products from .“closely-related” stocks can be helpful in re-habilitating a critically low. or lost stock, howcan the appropriate stock be identified?Is the approach suggested by Calaprice (1969) and by Ihssen (1976), that oi masimizingheterozygosity. the best direction to take inrehabilitating extinct stocks? How much varietyof parental material is required’?

2. What are the specific parameters in a changedenvironment which have become critical tosurvival of a stock?

The above are some of the questions for whichanswers will be needed in rehabilitation pro-grams. It is recommcndcd that a major researchinitiative in basic fish genetics and sclcctivc brccd-ing bc undertaken. Genetics. physiology. and bc-havior are three of the disciplines which need tobe part of such an initiative.

3. Having identified the parameters in (2), what arethe selection techniques appropriate to adapt-ing the new stock to the changed environment?

4. i-lzw are closely related stocks recognized? Bytime of spawning, spawning behavior. dir&on

While applauding the forward-looking researchin genetics and selective breeding that has alreadybeen initiated in at least four locations in Canada,it appears now to be appropriate to cramint thesein detail to determine whether some consolidationor further coordination might be appropriate tomore rapid progress on the basic questions beforeus.

CALAPRICE, J. R. 1969. Production and genetic factors in managedsalmonid populations. p. 377-388.t rout in s t reams 1968.

& Symposium on salmon andH. R. MacMillan Lectures in Fisheries.

univ. British Coltmbia, Vancouver, B.C.

IHSSEN, P. 1976. Selec t ive breeding and hybridization in f i sher iesraanagement. J. Fish. Res. Board Can. 33: 316-321.

STRAIGHT, W. J. 1969. Depth distribution of splake of known abilityto retain swimbladder gas. &LSC. Thesis. York Univ., Douns-view, Ont. 4s p.

TAIT, i, S, 1970. A method of selecting trout hybrids (Salvelinusfontinalis x S. namaycush) for ability to retain swimbladder gas. J. Msh: Res. E?oard Can. 27: 39-t.5.

.

15

HISTORY OF FISH GENETICS AND CYTOGENETICS:ALPHA, BETA, GAMMA, AND RFSOLUTION

Henry E. BookeWisconsin Cooperative Fishery Research Unit

U.S. Fish and Wildlife ServiceUniversity of Wisconsin

Stevens Point, Wisconsin

I wish to thank the Great Lakes fishery Cdssion and theAmerican fishery Society for sponsorship. 1% Henry Booke, Assis-tant Unit Isader, Wisconsin Cooperative Fishery Research Unit. Iam supposed to give a talk on the history of fish genetics and fish

cytogenetics. Why did I divide the topic? PzdarUy because ofprejudice. I have an interest in cytogenetics. As I did morethinking about the subject, I realized that there's more to it thangenetics and cytogenetics. There are many other topics that shouldbe covured. In a sense my title is a bit of a misnomer. So I willbe covering fish culture, selective breeding, physiological genetics,cytogenetics, population genetics.

What is genetics? Essentially the science of heredity andvariation. Cytogenetics can be defined as the study of behavior ofchromosamss during mitosis and meiosis. A person working in thatarea is interested in the morphology of the chromosomes and theresulting descriptions of the chromosomes and this is known as thekaryotype. This person is working to define the specific morphologyof each chranosane and then obtain the diploid number. And then hefollows his interest by seeing what happens to the chromosomes duringmitotic and meiotic processes.

What I am going to do now is simply treat each area with a bitof history with the idea of getting some thought of where we standnow, and perhaps pose a few questions in terms of where we shouldgo in the future. Then each of our speakers wd.XL treat things abit more in detail, and then we hope that you will question us andstimulate good discussion.

1. History of Fish Culture

If anyone at all is worried about how animals change fromgeneration to generation or how we can maintain animals, we then canstart talking about fish genetics developing 2500 years ago in termsof a fish cultural activity. The earliest recordings of this ac-tivity appear to be around L75 E?C (The Classic of Fish Culture byFan Lai). This was in China. What they did was use wooden sticks,or faggots, and simply place these in ponds. The spawn would becollected around the sticks and was transferred to various other pondsand maintained. Carp, goldfish, and medaka were so maintained. 1found records of these activities from 475 BC and 12h3 AD (Kwei Sin

16

Chak Shik by Chow Hit). Later on, in 1639, there was a book publishedcalled, The Complete Book of AgrictiLture,fl again Chinese by Heu.Most fish cultural activities that were developed by the Chinesewere done because they were worrying about whether they were goingto have an adequate food supply on a steady basis. It appears thatindependently the Romans also developed a series of fish culturalactivities. Primarily cutting canals to the sea and then fencingoff the canals after the fish came in to hold them in these areasas a resource or as a food reserve.

There seems to be no great fish culture activity for almost500 years. From the fifth century through the 5th, 6th, 7th cen-turies you see some activity, and then, around the 13th century, inFrance, you have a record of a monk, Dom Pinchon, involved in fishculture. Apparently there was quite a bit of fish culture actititygoing on in the European monasteries. Pinchon mixed ova and sperm.He was able to artificially impregnate and maintain eggs in woodenboxes and move them about according to his needs. Further fishculture activity in Europe beyond monastery walls was not done until1763. -At that time in Germany, Stephan Ludwig Jacobi rediscoveredthe activities that went on during Pinchon's lifetime. In France,'JO years later (1863), J. Rem and A. Gehin started to get intoCub culture activities. All these activities seemed to be motivated3~ food needs.

,

Now I'd like to move on to North America and there you see notesin the literature with the indication of the terrible condition of theAvers. This is in the late 17th century. In the '19th century,the first fish hatchery in North Amer%sa was established 42 Mumford,New York and the name Seth Green appears in the literature as themanager of this fish hatchery. Hatcheries were thought to be neededTo restore our fish depleted waters. By 1857 there is a manuscript byThaddeus Garlick on artificial propagation of fish. LY 1868 ThaddeusMorris published the book, "American Fish Culture." By 1898, theU.S. Bureau of Fish and its first manual on fish culture. What'sinteresting about this manual? On page 181 of this manual it tellsof the mixing of whitefish eggs (Coregonus clupeaformis) times lakeherring (C. artedii) sperm. If Great Lakes hatchery workers ranout of sperm for the whitefish, they used lake herring sperm. Andyou wonder why you may have problems today in the Great Lakes.

The next accomplishment, in terms of a fish culture manual,comes during the 1950's. H. S. Datig's book on the culture and dis-

eases of game fishes. So briefly, this is our history of interestin fish culture. 'iat, as far as the 20th century goes, do we have.- in North Amerfca.for fish culture activities today, between Canadaand the United States? There are over 100 federal hatcheriesover 500 state and provencial hatcheries. What is my sublimeduction? We know how to grow fish. MO question about that.are going to leave fish culture and go into another branch ofgenetics, selective breeding.

andde-We

17

II. Selective Breeding

I'll define selective breeding in the qualitative sensefirst and later in the quantitative sense. In other words, we lookat particular fishes and we decide on an empirical basis we likethe color or we like this particular size or we like this growthrate and we select for them on an empirical basis. There is nostatistical basis to it. We first see this qualitative approach in1919 in Hackettstown, New Jersey. Charles Hayford was responsibleand he published his first paper on the subject in 1930. He workedwith the first North American pioneer in this field, Dr. GeorgeEmbody of Cornell University. They were able to breed and selectbrook trout for bacterial disease resistance, increased growth,size of fingerling, and increased number of eggs. Later in the1930'9, Dr. Lox-en Donaldson started his work. For the first timethat I've ever seen noted, George hbody was given credit for influ-encing Donaldson. This was done in a recent book some of you maywant to look at and is called "Fish of Hare Breeding" by NeilHines (Smithsonian Inst. Press). This book is a description ofDonaldson's fish selective breeding work for the better part of hislifetime. In 1931, H. 5. Davis was breeding brook trout for a spe-cific spawning regularity or spawning date. Louis Wolf, in NewYork during the 1930's was detecting strain differences in brooktrout in terms of specific disease resistance. This seems to be thebasic highlight history of North American selective breeding pro-gr-.

The quantitative approach, or the statistical approach, to measurethe effects of genetic and environmental contributions to the develop-ment of a particular trait is called, in genetics, simply heritabi-lity. Where is this being done to any extent? I would say in thelast 5-7 years there has been a definite sign of activity in theUnited States, in Washington, Wyoming (U.S.F.W.S., Fish GeneticsLaboratory), California, C&ads, perhaps at St. Andrews (for Atlanticsalmon,monids j

Israel (carp and Tilapia), Norway (for scxne of the sal-,,and England (flat?'ish). More on the subject of selective

breeding will be dealt with in terms of a quantitative approachby Ray Simon.

Now, I want to slip back a little bit in time to talk about a book that was published in 1926. This book was called nl'heBiology of Fishes" and was written by H. Kyle. It is one of thosetraditional books that shows up perhaps every generation and it sortof summarizes the field. In fact it is still available and TFHpublishes it. In 1926, what sort of appreciation did Kyle have ofgenetics as applied to fish? You might think th& his impressionscould be the general beliefs held by fishery workers at that time.He said what was done in genetics in 1926 did not fit MendelianLaws of Heredity. In other words, the expected ratios that you wouldexpect to appear as far as experin;ental. results are concerned didnot. But this man did not have appreciation of what was going on.All of you are probably aware of the name, Johannes Schmidt. Thiswas the man that worked with the eel; discovered their spawning

19

In the field of physiological genetics, it was demonstrated inthe 1920's, 1930'9, 1940's, that specific genes influenced the de-velopment of certain embryological stages. Pigmentation, for example,was demonstrated to be controlled by gene action. Who did much ofthis work? Mainly Hy~on Gordon, at the New York Zoological SocietyAquarium, that is now in Brooklyn, N.Y. where K. KaUman carrieson this work.

Meristic traits have been demonstrated to be under influenceof genetics or gene action. It has been shown that not all hybridswere necessarily intermediate in meristic characters. In otherwords, if you cross species A times species B you were supposed toexpect that all hybrids were to be intermediate in character. Thecharacter intermediacy idea was strongly influenced by Carl Hubbsin a 1955 publication. As far as species isolating mechanisms in-volving reproductive behavior are concerned, this type of stuc@ isstarting to come into the literature primarily with work on cyprino-dontid fishes. In 1957, the state of art in this field of fishphysiological genetics could be stmvned up in a statement by Gordon.Gordon indicated that, %ore highly inbred lines of fish should beused for experimental purposes." Why? Because you could get morevalid and verifiablevariability which isresults.

results because you would decrease geneticprobably being expressed in your experimental

IV. Cytogenetics

In the field of fish cytogenetics we are at a very beginninglevel or descriptive morphology stage for chranoscmes. The firstactivity that I can find recorded was in 1916. The diploid numbersof the cyprinodontid fishes that were recorded were incorrect, andthey were primarily incorrect for many other fish until the mid-1940's. Where do we get the prowess to say that they are incorrect?Fish chromosomes are very small, very hard to work with. Even withthe best of methods, we are dealing with materials that are probablyless than 10 microns in size. It wasn't until methods, spin-offmethods, that were developed in the human cytogenetics field, wereavailable that accuracy in fish chromosome studies increased. Theseinvestigators developed-methods for blocking the mitotic process in

cultured cells with a chemical called colchicine. Chromosome countsof blocked metaphase chromosomes could then be made. Once the meth-ods came on the scene, they permitted all sorts of studies, not onlyin fish (genetics), but other mammalian and vertebrate forms, wereable to be studied. However, you have got to have a laboratory setup to culture these cells in order to do this particular work. Prom1945 on, I would say we were getting good valid results because,even though these chromosomes are small, you could still get them ata stage where you could define their specific morphology, and thencount them. Now what can you do with this sort of information?

Initially, there was a great state of activity in terms ofspecies-specificity. You would think that each specific animal would

20

have a specific chromosomes number. Well for a while that was so.But it happens to be that fish tended to be a bit more conservativethan we thought and I’m basing this conclusion on approximately SO0species of fish that have been examined for chromosome number.Most fish tend to have diploid chromosome numbers around 48-50.However, in the salmonids, they are not conservative or boring.They vary from around 50 to 130. When you examine the cypriniformfishes, you find a variation too. A similar diploid number variationas in salmonids. From around 20 up to 104 (for carp), 100 for gold-fish depending on whose results you are interpreting. Most of theperciform fishes have a tendency to center their diploid number aroundSO. However, in certain cases you find that you can get highlyreasonable karyotype data which can be used with other charactersto determine species-specificity,

We have a case for usFng the karyotype or the formal number typeof chromosomes for a species as a systematic character. There issome indication, however, that you may find chromosomal polymor-phism. What is that? In terms of one species, there might be avaribility in chromosome number; perhaps by river system. Thereis somh evidence for this in Atlantic salmon. You might find achimaeric condition. Meaning, in the same species you can find amixfxre of chrcznosome numbers depending on what organ is examined.Tkis has been demonstrated in the rainbow trout on the West coast.

I

Where have we gone with fish chromosome work? We are at the verybeginning level. What I will call alpha level or the descriptivelevel. We are just looking at fish and describing the chromosomes,and we have found that, in general, the most FAimitive fishes, tSeearliest fishes, tend to have a high chromosome number. And in themore recent fishes, you have a reduction in chromosome number. Thech-xmosome number, in time, is reduced by a procee called fusion01 where two chromosomes become joined. The reverse of this canhappen but it hasn't been shown to any great extent in fish. Thislatter process is caUed fissicn or the splitting apart of chromo-somes. In the Great Lakes, what is the status of karyotype infor-mation? All the sa&nonids, all, the esocids, lampreys, and a moderatenumber of the minnows, suckers, catfishes, have been karyotyped.So we have a good level of information about chromosome morphologyand diploid ntrmber.

v. Population Genetics

About 1938, the biochemical method called moving-boundaryelectrophoresis was developed in Europe by Arne Tselius. Thiselectrophoresis method allows you to separate particular forms ofmolecules depending on their electric charge. in whatever form ofelectrophoresis $sed3- whether yon’re using starch, paper, or cellu-lose acetate as a medium for carrying proteins or other moleculeswith a charge. on them and then applying an electric field to themedium, we have found that the most infoArmzition, in terms of recog-

nition of specific forms of fishes, recognition of races, recognitionof populations; has come from the use of this particular tool. The

21

situation that held up its application in fish work was that we didnot make the association between a protein molecule with a chargeand the fact that this protein molecule was inherited.

In some genetic work with pigeons in 1941, it was demonstratedthat certain serum proteins showed up generation after generation.But it wasn't really appreciated that this association between aspecific molecule, a protein, and its inheritance existed. Rememberthis was before Watson & Crick. There was a great amount of litera-ture, perhaps a couple of hundred papers, published from about 1945to about 1961 where investigators were studying different moleculesusing electrophoresis and looking specifically for species-specificityin these molecules. Then in the early 1960'9, there was some workstarted in Clement Markert's lab, at John Hopkins University. Mar-kert was interested in serum proteins and how they acted in termsof the embryological development of an organism. And here we havethe beginning of the story of lactate dehydrogenase (LDH) and itsmany isozyme forms.

Markert's group could show, and other labs also demonstrated,that there was a certain specificity of this protein (LDH) as far asa species was concerned, But it wasn't until around 1966 in thiscountry end independently in England at Henry Harris' Laboratory,that Lewontin and Hubby discovered that there was a considerableamount of polymorphism at each gene locus for LDH and other enzymemolecules. What do I mean by gene polymorphism? It is simplywhere a gene that exists on a chromosome at a specific site, a locus,is expressed differently because that gene has been altered by muta-tion in a fish species. These genes are called alleles (genes existas pairs or alleles on sister chromosomes), and their expressed pro-ducts, allelic forms. If allelic forms are expressed, and we coulddemonstrate it by electrophoresis, we have a natural means of markinga fish population. From 1966 on, this field has developed in termsof application and papers published. The most immediate applicationis the ability to not only identify a species, but to identify a format the population level, subpopulation level, race level, or howeveryou want to treat a particular fish variant.

I'd like to pose a few questions in terms of m quick review offish genetics history. Alpha is a standard systematic, or systema-tist's term for the descriptive stage. But whatever walk of lifeyou are dealing with, whenever you are learning sanething new, thereis always a beginning descriptive stage and thus I've adopted theterm alpha from the systematist. The question being posed is Will

. . I be here tomorrow9 For 2500 ye&s man has been capable of cul-turing fish and having at least from an empirical approach some under-standing of how to do it: I would say now with the great number ofhatchery production units in this country, on this continent, yes,we can successfully culture fish. EM the question should be,Why should we just judge that a fish culturist should be given hisfull credit for doing a wonderful job because he produces so manypounds of flesh per unit food fedF The fish culturist is judgedon the amount of product he ships out of the hatchery and not on the

,

22

hrnctional rel&ionsfiip of that fish to being able to sustain itselfin an environment. We should be asking questions about whether thereis another way we should be-judging our products as they go out ofour shops, in other words, our hatcheries. If you want $0 be a bitmore mundane about it, I would like to ask what industry on the faceof this earth can produce a product, throw it away, and not judgeit? This industry is not going to be iti business very long. EMour government agencies do it every day in terms of hatchery products.If you disagree, please question me about it when we get into ourdiscussion.

.

23

The next level a systematist would study would be the arrangementor beta level. Have we approached this level at a33 in fish geneticstudies? In terms of selective breeding we are getting to that pointand Dr. Simon till talk about this later. I would say, in termsof use of internal markers, in other words, isozyme systems or pro-teins in general, we are getting to the point where we can under-stand something about specific populations and variation in thosepopulations.

24

G a m m a.

Can weput it

together?

The next level of study for the systematist would be the syn-thesis level or gamma. We are far from that in tern of the kindsof products We are putting out in our environment and in our bowledgeof their ability to inherit particular characteristics.

25

Resolution: What can we resolve in terms of fish geneticsknowledge today? Definitely, we have the ability to define a pop-ulation in terms of certain genes, not every gene in the fish body,but genes that are representative of that particular animal. Asfar as selective breeding is concerned, we are just starting at thepoint where we are having an understanding of what trouble our houseis in. Again, I am going to leave that discussion to Dr. Simon andat this point I am going to stop and permit Dr. Peter Ihssen to speakabout fish physiological genetics.

. .

.

27

Physiological and Behavioral Genetics and theStock Concept for Fisheries Management

Peter IhssenFish and Wildlife Research BranchMinistry of Natural Resources

PrOrLnCb of Ontario

I believe that it is now well accepted that fish species aregenerally subdivided into more or less genetically isolated popula-tions or stocks. The stock has been defined as a population oforganisms that share a common gene pool and a common environment,Adaptive as well as non-adaptive genetic changes within speciesoccur at the stock level. Adaptation to the common environment oc-curs at the stock level as well as non-adaptive genetic changes dueto sampling variation of the alleles for small populations, calledgenetic drift. These processes, coupled with scbne genetic isolatingmechanisms, bring about the genetic differentiation among stocks.This leads to geographic clines or sharp distinctions among stocks,for example, due to discontinuities in environmental conditions.For some salmonids, their strong homing behaviour provides the me-chanism for genetic isolation. Migration counteracts these processesOf differentiation among stocks and leads tc greater genetic unifor-mity.

.

That a species can be composed of genetically distinct sym-pat&c stocks has been recognfsed for a long time. The Germanichthyologist Heincke has been credited with the pioneering work inthis area. He published a paper in 1898, In which he suggests thatthe Atlantic herring of the North Sea is represented by 30 smallgroups, which he called races, and defined as groups of individualswhich live in the same area, share the same habitat, and have aclose blood relationship. The distinctive features, in this case,were morphological features. This definition is very similar to theone I gave except Heincke used the term "close blood relationship",where we now say “share a common gene pool". Of course his workpreceded modern genetic theory and the recognition of MendelIswork. The stock concept has received renewed interest in the lasttwo decades due to the discovery of very considerable intraspecificgenetic variability for structural genes by electrophoretic tech-niques. I will be discussing genetic intraspecific physiological,and behavioral variability. Before I do, however, I would like tosay a few words about technique. Intraspecific variability has beenstudied traditionally by looking for differences for morphologicalcharacters. Body measurements, meristic counts, and morphologicalfeatures such as color variants have been observed for specimenscollected from the field. As soon as fish were reared in controlledenvironments (and even now of course this is very difficult for marinespecies) it was found that morphological characters are very muchinfluenced by environmental events, particularly during e=lY

28

embryonic development. Consequently, at least part of the varia-bility observed is not due to genetic differences, but environmentaleffects. Hence the differences observed among stocks may only par-tially, if at all, reflect the genetic relationship among stocks.Morphological characters, however, have received renewed interestwith the availability of high speed computers that can simultaneouslyanalyze a large number of characters using statistical techniquessuch as discridnant function analysis; but again, the differencesthat are found do not necessarily reflect genetic differences. Forexample, we looked at a number of lake trout populations from Ontariolakes and hatcheries for characters such as number of anal and dorsalrays, vertebrae, gill rakers, snd pyloric caeca. The difference innmber of gill rakers, anal and dorsal rays, and pyloric caeca thatcould be induced simply by rearing from fertilization in a differentenvironment (hatchery) were of the same magnitude as the differencesobserved among these populations. Hence one cannot draw conclusionsfrom such phenotypic characters about the genetic variability amongstocks.

Behavioral and physiological characters have not been studiedto the same extent for stock differences. However, usually in thesestudies more care was taken in controlling the environment, becauseit is very obvious that specimens collected from the field are notsuitable for comparative studies due to their different entiron-mental histories. The great advantage of the electrophoretictechnique is that the phenotypes observed are very representative ofthe genotypes and little influenced by the environmental effects.For behavioral and physiological characteristics of course this isnot the case since many processes intervene between the genotype andthe observed-phenotype and usually a number of genes control eachcharacter. However, many links between the electrophoretic geneticvariability and character of adaptative significance have not beencominclngly made to date.

Intraspecific variation for characters associated with migrationhave been studied for a few species of fish. Little is known aboutthe genetics of these characters, but it can be postulated fromwork on the behavior of other animals that these characters are DOZY-genetically controlled as has been found in Drosophila for example.Differences in spawning behavior have been observed for differentlake-trout stocks of the Great Lakes. For example, Loftus has re-ported on river-spawning of Lake Superior lake trout and Roycereported on differences in spawning time within season for New Yorkstate lake trout stocks. To what extent these differences reflectgenetic differencesis not known since no controlled experiments havebeen performed to my knowledge. Brannon showed that sockeye fry

.

reared from inlet and outlet stream populations of the Fraser Riversystem exhibited behavioral patterns under controlled laboratoryconditions corresponding to the up-stream and down-stream migrationbebavlorWin-the wild with hybrids intermediate in behavior. Johnsonand Groot studied the orientation behavior of smolts from differentBabine Lake populations and discovered differences in orientation inthe laboratory corresponding to the orientation required in the laketo find the outlet stream. In these experiments, gametes were

29

collected from the wild and reared in the hatchery without any of theenvironmental cues that were thought necessary for this behavior.

Another technique to demonstrate genetic differences are trans-plantation experiments. This work has shown, for example, thatdifferences in growth and size among stocks due to feeding behaviorwas environmentally induced, rather than genetically, at least insome instances. Recently a very interesting transplantation ex-periment was reported on by Bams on homing of pink salmon. Thiswork is particularly convincing since the maternal effects not ruledout in the studies I mentioned so far were ruled out by crossinglocal and donor males to identical donor females. Eggs were collectedfrom a stream, which is called the donor stream, and crossed withdOnOr and local males. %ocsln in this context means the environ-ment in which these crosses were tested. For a number of females,the eggs were divided in half, and fertilized by males from the twodifferent areas. Hence, differences found could not be due to maternaleffects associated with the eggs. However, maternal effects werenot ruled out in the other studies because eggs from different pop-ulations were used. Bams found that the local x donor strain returnedabout 10 times as frequently to the local stream as the pure donorstrain. It appears that some local genes are required to bring thesefish back to the home stream. Also, it was found that the migrationbehatior in the stream (of returning precisely to the planting site)was much more precise for the local x donor strain that had the localgenes. I think this is a very convincing piece of work indicatinginheritance of a behavioral character.

In our lab we have done some work on the inheritance of seasonalspawning time-and found that there is a strong correlation amongsisters for this character. Also, the mean spawning date of sistersis correlated with their mothers’ spawning date and hence their birthdate. The spawning time within season for individual females isvery precisely replicated in other seasons.influence is indicated.

Hence a strong maternal

With regard to physiological characters for a number of brookand lake trout populations of Ontario, no differences for high tem-perature tolerance were found. These were carefully controlledreplicated experiments for different acclimation temperatures.Similar comparisons were made for different naturalized Ontariorainbow trout populations and again no population differences forupper temperature tolerance were found. Also in our laboratorywe have been comparing two rainbow trout stocks for a number of phy-siological characters. One of these stocks is a hatchery straincommonly used by the Ontario Ministry of Natural Resources forplantings in lakes and streams. The other stock is a self-sustainingnaturalized stock from the Nottawasaga River which flows into LakeHuron. Males and females from these two stocks were crossed in adiallel experimental design (eggs from each female are split intotwo lots and each lot fertilized by a male from each strain and thesame males are then used in the same way on a female from the otherstock producing four crosses) which was repeated twelve times withabout 20 fish per family. This experimental design pemnits one to

30

separate male and female effects. It was found that most of thevariability in growth was accounted for by female effects with thefamilies derived from the hatchery stock growing faster than thefamilies derived from the naturalized strain. The rate of maturationof males on the other hand was almost entirely determined by themales with the families derived from hatchery males having a higherpercentage males in their second year. It is too early to assessmaturation of females since only a small percentage have matured

to date.

Hemoglobin concentrations for these fish were compared and itwas found that fish derived from the hatchery stock had highly sig-nificant lower total amounts of hemoglobin in whole blood than thefish derived from the naturalized stock. Also the amount of hemo-globin per red blood cell volume was lower for the hatchery stock.Also it was found that the amount of hemoglobin in whole blood ispositively correlated with the length of the fish within thesestrains.

In conclusion I want to make the point that usually man's ac-tivlties have the greatest impact at the stock rather than the specieslevel. And consequently it has been suggested by a number of fisheryscientists that without a clear idea of the specific features of thebiology of these stocks, a sound approach to fisheries managementis not possible. I thhk it is fair to say that until recentlythis concept has not been given the emphasis it deserves. The con-sequences have been not only the loss of some irreplaceable stocksof fish but also most probably the failure of some management pro-grams to re-establish self-sustaining stocks.

31

The LDH Isozyme System of Salmonids

Edward MassaroMedical School, Department of BiochemistryState University of New York at Buffalo

Buffalo, New York

When Henry Booke asked me to do this talk several months agohe said that my specific task would be to try to explain to theaudience of what value electrophoresis techniques might be in theidentification of stocks and environmental influences and things likethat. I have a question, therefore to ask the audience: Will Ibe wasting your time while I explain to you what electrophoresis isand how electrophoretic techniques are utilized in the identificationof fish species?

LDH isozyme patterns of the rainbow trout. Now, to some ofyou this just might be a lot of jibberish, However, what we havedone here is to separate the multiple molecular forms (isozymes)of the enzyme lactate dehydrogenase (LDH) frcun skeletal muscle, heart,intestine, and the eye by electrophoresis in a stabilizing medium.Now, how do we go about accomplishing this?

First of all, let me say something about lactate dehydrogenase.Lactate dehydrogenase is an enzyme, therefore it is a protein,therefore the infonnation for its biosynthesis is encoded-in thegenome of the organism.

Lactate dehydrogenase is a soluble enzyme and can be extractedfrom tissues simply by homogenizing a sample of tissue in distilledwater or in some type of buffer solution (at high protein concen-trations,, a buffer solution is not necessary). Usually, the homogenateis clarified by centrifugation and the supernatant solution, theextract, is applied to a stabilizing medium in preparation for elec-trophoresis. In our case, the stabilizing medium was starch gel.

The technique of electrophoresis (free boundary electrophoresis)was developed by Tiselius, Tiselius used a U-tube into which heplaced a solution containing protein molecules, keep in mind thatthese molecules are charged. Into one arm of the U-tube he placeda negative electrode and into the other, a positive electrode. Bypassing a D.C. electric current throughthe solution Tiselius dis-covered, using a UV optical system, that he could separate the pro-tefns according to charge differences. By having this apparatus con-strutted of movable sections, he was able to separate out certainproteins. This was a remarkable technique. However, it presentedproblems because the protein molecules were free in solution, andmany influences can affect the resolution of the protein moleculesin solution. For example, small changes In temperature can accel-lerate the rate of diffusion end, therefore, the resolution of themolecules. Thus; the apparatus has to be maintained at a constant

32

temperature, This is difficult and costly to achieve.

In more recent times, it was discovered that a more stablesystem could be achieved employing a strip of filter paper whichis AOthing more than a cellulose matting as a stabilizing medium.The paper is dipped into a buffer solution, the excess is blottedoff and a sample of a soluble protein mixture is spotted on the strip.The strip is subjected to electrophoresis; that is, a voltage gradient

is established across the paper and the component proteins of themixture are resolvedinto individual bands. NW, this is a verysimple technique, basically requiring nothing more than a powersupply, a strip of filter paper and a tissue sample. Very conven-ient. A few years after this technique was developed, somebody cameup with a substance called cellulose acetate. Electrophoresis oncellulose acetate was somewhat of an improvement over paper electro-phoresis because of the properties of the material. Filter paper hasbeen greatly improved over the last fifty years, but you still can'tget the type of resolution on filter paper that you can on celluloseacetate. Unfortunately, both of these methods suffer from severalinadequacies. One of these, which is, perhaps, not obvious, is thattheir capacity, the amount of material which can be applied to paperor cellulose acetate, is extremely limited. In some cases this mightnot be a problem, in other cases it is an extreme problem. A secondproblem, very similar to the problem encountered by Tiselius is thattheir resolving power is limited; fundamentally because what we are'doing is separating the protein molecules strictly according to chargeand diffusional movement of the proteins is not restricted.

Protein-molecules are charged molecules. The charge can benegative or positive. and one protein molecule can either be morecharged or less charged, negatively or positively, in relation tosome other protein molecule. Another characteristic of proteinmolecules is that of size. Size is a characteristic of a given typeof protein molecule, as is charge. Some types of protein moleculesare smaller than other types. Thus, we have two characteristics,charge and size, that we can exploit to separate protein molecules.Why not exploit both characteristics?

A while back someone came up with the brilliant idea to convertstarch into starch gel and use the gel as a medium for separatingproteins. A starch granule is nothing more than a gigantic curled-up

sugar polymer. Starch granules moistened with a buffer solution andpacked into a trough can be used as an electrophoretic stabilizingmedium. But, unfortunately, the interstices between packed starchgranules1ar-e so large that resolution is reduced by diffusion. For-

_ tunately, someone discovered that the starch granule can be unfolded-. and the linear-starch-polymers formed can be packed into a dense

matting of controllable pore size. The pore size of the gels formedis such that diffusional movement of protein molecules is greatlyinhibited which greatly improves resolution. All one has to do issuspend the starch in a buffer solution, boil it carefully (to un-fold the starch granules}, pour the hot suspension into a mold andallow it to cool. On cooling, a dense meshwork of starch polymers

33

is formed. By controlling the starch concentration (i.e. the linearpolymer concentration), the pore size of the meshwork can be controlled.Dy controlling the pore size, the rate of migration of proteins(which differ both in size and charge) can be manipulated which isof great importance in the study of the composition of complex mix-tures.

I would like to pass this technique on to you because it hasgreat potential in your field. Most certainly, it will find wideapplicability in fisheries management in that it provides a simplemeans for sampling the genetic composition of fish stocks. It re-quires only Simple equipment. All you need is a hot plate, stirringmotor and stirrer, starch, buffer solution, molds, and a D.C.

To prepare the starch gels I use in ngr research, the starch isboiled in a buffer solution, poured into a mold and allowed to cool.Gmple wells into which the tissue preparations are placed are pre-formed in the gel in the mold. For those of you who have never seen8 starch gel, it looks like a stiff block of gelatin.tissue samples,

I prepare wsuch as skeletal muscle (red or white muscle or a

mixture of both red and white muscle), heart, a piece of intestine,eye, etc. by homogenizing minced tissue in buffer solution (1 to 20 mlbuffer per gm tissue). The homogenstes are centrifuged and the clearsupernatant solution, containing the soluble proteins, is placed intothe sample wells of the gel. A voltage gradient is generated acrossthe gel and the proteins are separated according to size and chargeintensity. As is well known the enzyme lactate dehydrogenase can beseparated into a number of bands of activity. These multiple molecu-lar bands are called isozymes or isoenzymes. Now what does this allmean? Well, proteins like the enzyme lactate dehydrogenase are en-coded in the genome. Therefore, they are subject to alteration viamutation. Perhaps they are also subjected to change or variationin expressibility due to environmental factors. These alterations,variations, etc. are detectable by starch gel electrophoresis.

The technique can be used for the identification of the multiplemolecular forms of enzymes,descriptive tool.

or other types of proteins as a purely

tool.However, it can also be used as an analytical

Lactate dehydrogenase is composed of four sub-units. These four sub-units form the functional molecule. So far as we know,the individual -sub-units, by and of themselves, are not active.The sub-units have to be put together in a tetrameric form beforethey will function, before the enzyme is active, so far as we know,Using starch gel electrophoresis, or other types of biochemicaltechniques, like chromophotography, it is possible to isolate individ-ual isozymes. There are five major bands of lactate dehydrogenase inthe cow, the horse, etc. If you isolate the slowest moving band,which is called LDH-5, and the fastest moving band, which is calledLDH-1, and combine them in a phosphate buffer solution in the presenceof a small amount of sodium chloride and freeze and thaw this solution,five isozymes are formed. LDH-1 and LDH-5 remain; but LDH-2, -3, and-&, migrating between LDIi-1 and LDH-5, are generated. What hashappened is that, on freezing in this solution, the tetramer diSSOCiRteS

34

and on thawing, the individual sub-units reassociate at random intoevery possible combination of four. Thus, if there are two differ-ent types of sub-units, as there are in mammalian LDH, five isozymeswill be fanned which can be separated by gel electrophoresis. Now,employing the freeze-thaw molecular hybridization technique in con-junction with gel electrophoresis we have an analytical. tool that canget some information on the actual molecu?ar composition of the groupsof LDH Isozymes of the rainbow trout. For example if we attempt tohybridize different groups of rainbow trout isozymes with equine orbovine LRH-5, we can get some interaction (certainly not a binomialrecombination by any means) between LDH-5 of the horse or cow andcertain groups of trout isozymes. By using this technique, thisintraspecific hybridization, we can obtain some information aboutthe structure of LDH molecules without resorting to techniques likeamino acid and peptide analysis, or other types of analyses. Usingthe hybridization technique, it is possible, for example, to showthat during the evolution of the salmonids (and other species offish) that, by the process of gene duplication, as in the case ofthe hemoglobins, new genes were established in the genome. Thesegenes are still undergoing variation and can be used for the identi-fication of specific strains and specific stocks of fishes.

35

Genetic Variation in Populations of Fish

Fred AllendorfDepartment of ZoologyUniversity of MontanaMissoula, Montana

Introduction

The era of modern genetics began at the beginning of the 20thcentury with the rediscovery of Mendel's principles. The field ofpopulation genetics quickly emerged in the following 30 ye=se By1930, a large proportion of our present total body of populationgenetic theory had been described by the three early giants of thisfield--R. A. Fisher, J. B. S. Haldane, and S. Wright. the theoreticalfoundation of population genetics, therefore, was established veryearly.

Unfortunately, however, the practice of population genetics didnot develop as rapidly. It is very difficult to examine geneticvariation in natural populations. Thus, although armed with thislarge body of theory, geneticists could not go out and look at geneticvariation in natural populations. They went out and looked at mor-phological differences; they looked at physiological differences;they looked at behavioral. differences. There was no way, however,they could interpret these differences in terms of $&n&lien genetics(i.e., allelic variation at single genetic loci).

The overwhelming majority of the experimsntal work performedwas with the fruit fly (Drosophila spp). This early work was almostexclusively accomplished with genetic markers associated with visiblephenotypic effects (e.g., white eyes or wingless). These low fre-quency genetic variants with a large, usually harmful, effect do notrepresent the type of variation in natural populations we are mostinterested in. We would like to be able to examine the geneticvariation which is present at high frequencies in natural populations

and can be used to characterize genetic differences between differentpopulations within a species. This, however, was impossible for along period of time following the origins of population geneticsin 1930.

In 1966, two independent publications described a technique whichfor the first time allowed the detailed examination of genetic var-iation in natural populations (Harris, Proc. Roy. E. Ser. g 16L:

.298; Lewontin and Hubby, Genetics 54: 5.m These two papers describedthe use of starch gel electrophoresis to examine genetic variationin populations of humans and Drosophila. One of the beauties of thistechnique is that it can be applied equally effectively to almostany organisms desired. Within the last year in my lab alone, thesame basic techniques have been used to examine genetic variation

in tent caterpillars, fish, frogs, mice, beers, and pine trees.

36

We now have a tool which allows us to examine in detail geneticvariation in fish populations. How do we use it?

Interpretation of Electrophoretic Data

Dr. Massaro has spent considerable time and effort outlining thebiochemical basis of electrophoresis. I would like to spend sametime outUning the genetic interpretation of these results.

Figure 1 shows the same enzyme (lactate dehydrogense - LDH)thcrt Dr. Massaro spoke about. The differences seen in this figurerepresent genetic differences between individual rainbow trout (Salmogairdneri). Each column (l-12) is a liver sample fran inditidti-fish. The observed differences represent two different geneticforms (A and A') of this enzyme in rainbow trout. These two genetictypes (i.e., alleles) are detected by their different rates of migra-tion when placed in a starch gel matrix which has an electric currentpassing through it.

Three different genetic types, or genotypes, can be seen in thisfigure--M, AA', and A'A'. Each individual carries two copies ofthis gene, one received from the mother and one from the father.Individuals csn therefore possess two copies of the fast migratingallele (genotype M), two copies of the slow migrating allele (A'A'),or one copy of each (genotype AA'). The five electrophoretic bands(or isozymes) seen in the AA' type results from the biochemicalstructure of LDH. As discussed by Dr. Massaro, four protein sub-unitsjoin togetherto form an active LDH molecule. If two geneticallydifferent sub-unit types (A and A') join together four at a time, theresult is the five different isozyme8 seen in the Ab' phenotypes.

Now, just so you don't go away with the impression that LDH isthe only enzyme biochemical geneticists work with, I should discussother enzymes which are also used in this work. There are presentlysome 40 enzymes being used in my lab to detect genetic variation inpopulations of salmonids (Allendorf, Mitchell, Ryman, and Stahl,Hereditas 86: (in press) ). The basic biochemical and genetic prin-ciples involved using these other enzymes are analogous to thosepresented for LDH here today. The more enzymes, and therefore geneticloci, we can examine the more detailed a picture we can develop of thegenetic structure of populations,-

A valid question at this point is--How do we know these dif-ferences I have presented represent s%nple genetic differences? Inthis regard, I am very fortunate to be working with salmonids. Itis relatively easy to perform experimental matings under controlledconditions with fish of known genetic types to test the inheritanceof these observed electrophoretic differences. These matings havebeen performed and confirm the genetic basis of this variation(Allendorf, Ph.D. thesis, Univ. of Washington, 1975).

Now that we have established the experimental basis of thistechuique, how do we use electrophoresis to characterize genetic

37

variation in fish populations? Let's return to Figure 1 showingdifferent genetic types of LDH in the rainbow trout, If we assumethese twelve individuals represent a sample from a population ofrainbow trout, we can CharaCteri that population for this geneticlocus by C&xd,ating the fr8qU8nCi8S Of these tW0 ti8liC types(A and A') in this sample. A total of 24 genes are represented inthis figure (two times the number of individuals). By examining thenumbers of different genetic types in our sample we can estimate fielefrequencies for the population from which this sample was taken.Each ti genotype individual has two copies of th8 A allele; similarly,each AA’ individual has one copy of the A allele. Therefore, thefrequency of the A allele in this sample is two times the number ofAb individuals plus the number of AA' individuals, divided by thetotal number of genes examined. Thus,

Likewise,

w8 have now estimated the tie18 fr8 uencies in the totalthrough this sample Of tWelV3 indiVi U&S.8 The more indiVP

populationduals we

examine, of course, the more accurate our al.1818 frequency’estimatesUsing these techniques we characterize differsnt populations

rihe frequencies of such genetic types for as many different geneticsystems (i.e., loci) as possible. If we carry out this procedure onthe 70 som8 lOCi We C8nnOw8xamin8 in salmonids,we can d8riV8 afairly detailed analysis of the genetic structure of these populations,

Applications

How can W8 apply this technique to the problems involved inmanaging populations of fish? The first question I would like to-approach is--How much genetic variation is there within individualpopulations? This is obviously a very important question. If we areconcerned with evolution, whether via natural selection or via ar-tificial selection in a hatchery, the amount of genetic variationpresent is the key factor. Lf there is no genetic variation betweenindividuals, then there will, not be any evolutionary change in thatpopulation. The rat8 of 8volutionaxy change is directly proportionalto the amount of genetic variation present. Therefore, if we wishto select a fish population for some particular characteristic, it isnecessary that there be genetic variation present for us to succeed.In addition, the loss of genetic variation following inbreeding, whichoften occurs in hatchery stocks, has been demonstrated to have haxm-ful effects (Kincaid, 1. m. Res. Bd. Canada 33:2L20). It is,

38

therefore, desirable to be able to measure the amount of geneticvariation within individual populations of fish.

Electrophoresis provides a simple objective technique formeasuring the amount of genetic variation within a population. Wecan analyze a sample of fish fram the population in question anddirectly quantify the amount of variatdon present.

How is this information directly important to managing fishpopulations? There has been much speculation that hatchery practiceshaoe reduced the amount of genetic variation in many-of our hatcherystocks of fish. These concerns were previously limited to specula-tion. NOW, however, we have a method for directly estimating howmuch genetic variation is actually present in these stocks. A corn--parison of stocks within a particular species will identify whichstocks do possess significantly lower amounts of genetic variation.We thus have a method for objective> estimating the loss of geneticvariation in hatchery stocks.

An extension of this technique is the creation of stocks withexceptional amounts of genetic variation. When selecting new stocksto be cultured, there ax-8 advantages for selecting stocks with asmuch genetic variation as possible. Such stocks can be created usingthese t8ChniqU8S by incorporating individuals from stocks which areinitially extremely genetically different, as measured with electro-phcresis. Thus, w8 ten both monitor the loss of any genetic variationand also artificially increase the amount of variation in a particularstock.

The next level of genetic variation of interest is the amount ofgenetic variation between different populations. Through the e-a-tion of allele frequencies at many loci W8 can 8Stimat8 the amount ofgenetic diVergem between populations. This gets us into the prob-lem of the stock concept. Having Separate genetic populations withina species is a critical factor in the management of such a species.A difficult problem has long been the identification of such stocks.Electrophoresis allows us to collect samples frcm a number of naturalpopulations within a species, and objectively define the populationstructur8 of that species. w8 can both OUtliII8 the major populationgroups and also estimate how genetically different local populationsare. We can, thus, for the first time describe the patterns of geneticdiversity within a particular species.

This genetic structure has very important implications for themanagement of a species. A good example of this on the local popula-tion-level-is a study I am involved in dealing with brown trout (2.triitta) $ojx&&ions present in Swedish Lakes (Allendorf, Ryman,Stahl, and Stennek, Hereditas 82, 19). We have found two completelygenetically isolated populations are equally frequent in this lake.However, they have different growth rates. We are currently attempt-ing to detail other environmental or physiological differencesbetween individuals in these two stocks.

39

The presence of these two separate stocks is an important factorin managing this lake, A harvesting system based on simple quotasand size restrictions could haV8 devasting effects. Proper manage-ment in this situation must treat these stocks separately. For allpractical purpOS8S these two stocks represent separate species endmust be managed as such.

On a larger scale, the genetic structure within a species tenb8 used as natural genetic tags. This is of critical importance whendealing with populations of salmonids which often travel long dis-tances from their waters of origin.

Populations of anadromous salmonids on the Pacific Coast aregenerally harvested before returning to their freshwater streams forreproduction. Harvesting quotas must take into account the stream oforigin of returning fish so that sufficient numbers ax-8 allowed returnto each local population in order to maintain that population. Geneticdifferences among local stocks, as measured by the techniques des-cribed here today, are currently being US8d to approach this problemwith populations of both Oncorhynchus and Salmo species along thePacific Coast.

We can go at least one step further in this us8 of genetic dif-ferences as natural tags in the management of fish populations. Thistechnique is dependent on finding genetic differences among stocks.We can, however, create stocks which are genetically distinct viathese techniques through systems of artificial selection.

An 8XCfthg example of this application is being applied to themanagement of steelhead (2. gairdneri) in the State of Wash43@on.Genetically marked hatchery fish are being planted in previouslyunplanted steelhead streams to monitor the effect of hatchery plantingon the natiVe stocks.us not only to

The biological nature of genetic tags allowidentify the planted hatchery fish but also their

PrOg8m. Thas, we can examine both the spawning success of hatcheryfish and the interaction between the progeny of the hatchery andnative fish.

Well, I'm already ten minutes over my alloted time SO 1 hadbetter stop right here. Thank you for your attention.

40

Figure 1. Diagram of electrophoretic patterns reflectinggenetic variation for LDH in the rainbow trout.

Genotypes

AA’

41

Selective Breeding: The "Specialist" Approachversus

Crossbreeding: T&?%&eralistR Approach

Raymond C. Simon*U.S. Fish and Wildlife Service

Fish Genetics LaboratoryBeulah, Wyoming

Trout hatchery personnel have repeatedly observed dif-ferences in growth and performance of different trout strains. Itis in fact axiomatic that if two or more strains of trout are retainedseparately, some differences, either great or small, are alwaysexpected. Strain differences are typically found in bioassay tests,nutritional studies and response to disease challenges. For example,some strains of trout are known to be highly tolerant of furunculosisand remain asymptomatic, while other strains are severely decimatedwithin the same hatchery environment. Furthermore, the difficultyin repeatability of bioassay results is understood to be due partlyto use of different strains at different localities.

Given that strain differences in trout are the rule rather thanthe exception, the idea of matching strains to particular managementobjectives or particular environments is clearly one way of derivingbenefit from these differences. Just such an activity was chargedas a major objective of the Fish Genetics Laboratory in 1973. Theproje&appro$Ching this issue was titled Genetics of Wild andHatchery bout Strains and has served to catalog a 7-w of strainattribute-h a final. aim of being able to specify which attributeswere correlated with superior performance. This activity then is oneof defining a SPECIALIST which is predicted to match certain environ-ments or management circumstances. Numerous differences in strainshave been identified by FGL studies. Several of these differencesare of obvious economic value, such as hatchery survival, food con-version efficiency and survival after release.

A stringent limitation on the specialist approach, however, isthe difficulty of testing performance under any appreciable fractionof the hundreds of different environments that exist. Difficultiesassociated with this limitation are several, if survival and perSor+ante (return to the-fisherman) are adopted as major criteria tomeasure success.

#&ymond C. Simon has since relocated as Staff Specialist, Divisionof Fishery Research, U.S. Fish and Wildlife Service, MatomicBuilding, Washington, D.C. 202LO.

.

42

For example, we can define several environmental factors whichwould be expected to discrirnlnate the superiority of certain strainsamong two or more strains being tested simultaneously:

1.2.3.

2.

4:

ii:9.

10.

abundance of foodtype of foodPHwater hardnesstemperaturepollution leveltype of pouutant if presentoxygen levelphysical factors (deep lake, farm pond, sluggish stream, etc.)competition and predation levels

Furthermore, seasonal fluctuations affect many of these factors,thus they are variables more often than they are constant. Given thenthat large numbers of different environments exist and that they aresubject to change, the search for ideal strains is apt to becomevery ccmplicated, and then perhaps for limited usefulness. Thenatural idea springs forth that the strain could use some help inthe form of selective breeding. Thus the few individuals which showthe most dramatic expression of what is desired are chosen out asparents for the neti generation and the specialist idea is advancedto yet further degree. The procedure is repeated for several genera-tions with the hope of attaining maximum levels of change. Thesekinds of selection have been performed in hatcheries rather often(few managers would use poor-looking fish for breeders unless theyhad no choice). Generally speaking, these activities present aproblem because they are uncontrolled experiments in the sense thatunselected controls are rarely retained for purposes of measuringthe changes anticipated.namely, that practitioners

A second problem may be more important,of artificial selection must assume they

know what is best. Almost inescapably, this leads to an attitude ofpaternity concerning the experiment. In other words, having oncedecided what is best, it is increasingly more difficult to believethat other alternatives might be better. There is yet a third prob-lem, caused by the fact that selective breeding, by definition,reduces the numbers of-breeding individuals to be used as parents.The price paid for gains which may have been made can be measuredfrom the effects of inbreeding, which inevitably has been accumulatedfrom the reduced numbers of parents. Some costs of inbreeding areshown in the following two tables, These data, where each number isbased on several hundred measurements, are thought to approximatethe inbreeding levels of some (possibly a majority) of trout hatch-eries. Eggs of each female were divided into two groups thus I x I(inbred matings) and I x 0 matings- from the same female are compar-isons of half-sibs.appreciable.

The adverse result of inbreeding is clearly

The alternative approach is to use crossbred individuals derivedfrom crosses between different strains for production purposes,, Thesecrosses have often, but not always, been shown to perform better thaneither of the parent strains. The main problem in this case is the

43

large number of different crosses that are possible. Twenty strains,for example, would yield 380 possible unique crosses. Results todate indicate that some strains do not do well in crossings whileothers are impressive. There are two clear advantages to the useof strain crosses. First, the idea of 'paternity" is minimized be-cause of the need to measure what is superior, rather than prog-nostfcating it. Secondly, the inbreeding which has accumulated canbe put to advantage because of the vigor displayed In many straincrosses.

In closing, one point requires heavy emphasis. Because environ-ments are changing, what does well today cannot be guaranteed to dosimilarly in the future. For this reason, it is essential to retainbroodstocks in unmixed state to permit construction of the fullpossibility of crosses for the future.

These remarks have been made in reference to fish which arereleased from a hatchery with subsequent requirement for fairly ex-tended, or greatly extended survival time to permit contribution toa fishery. !t”raditionally used hatchery stocks may not be quite soseverely compromised in uses where the fish are under total captivecontrol. Even here however, the advantages to be gained from cross-breeding deserve evaluation.

.

44

OutbredFemale

Body Weight (grams)

Summary of Means

Inbred OutbredMale Male

IX1 1 x 0

2.25 3.40

OXI 0 x .o

3.47 4.09

45

Mortality (Hatching to iLE7 days)

Summary of Means

Inbred OutbredMale Male

InbredFemale

OutbredFemale

I XI I x 0

34.9 13.6

0x1 0 x 0

10.1 12.0

47

Mark Noveck: Dr. Massaro, would you like to delineate some of thestaining methods you used for enzyme identification:

Massaro: A number of different types of enzymes can be detected bystarch-gel electrophoresis. However, I'll just confine my remarks tothe dehydrogenases because I don't wish to go into a lot of explanationconcerning substrates, etc., and I can use lactate dehydrogenase as anexample. The same basic system is used for the detection of dehydrogenases.The reaction, very simply, is one of a reducing series in which lactate isconverted to pyruvate. In effect, two hydrogens are removed from lactate.what do you do with these two hydrogens? This reaction, the conversion oflactate to pyruvic acid, requires a substance called a cofactor whichaccepts hydrogen atoms. This cofactor is called nicotanamide adeninedinucleotide (NAD). All YOU have to keep in mind is that NAD is reducedin the conversion of lactate to pyruvic acid. In effect, the two hydrogensare removed from the lactate molecule by the cofactor. Different dehydro-genases use different types of cofactors, but the same type of a reactionoccurs. years ago it was discovered that this dehydrogenation reaction couldbe made visible by using a tetrazolium dye (nitro blue tetrazolium) ~omm0d.yused in histochemistry. If, in a buffered solution, the substrate, ourenzyme, lactate dehydrogenase, plus its substrate, lactate, plus thecofactor, NAD, plus the tetrazolium are mixed together, a reaction willtake place in tihich the hydrogen is transferred from lactate to thecofactor and then to the tetrozolium. When the tetrazolium is reduced,it precipitates forming a blue or blue-purple precipitate. If a tissuehomogenate is prepared, for example, from skeletal muscle, a whole hostof soluble enzymes are going to be present in the supernatant of thishomogenate. How do we stain specifically for lactate dehydrogenase?Lactate is used as the substrate. Enzymes are highly specific and theywill do their job, usually only on a particular type or a particular groupof substrates, in this case lactate. Following electrophoresis, the starchgel is placed into a buffer solution containing lactate, the cofactor,and another substance which facilitates the reaction called PMS (phenazinemethosulfate). The hydrogens are reIlroved from the lactate by the cofactor.PMS removes the hydrogens from the cofactor and the tetrazolium removesthe hydrogens from PMS. Reduced tetrazolium is precipitated within thepores of the gel in the vicinity of the individual isozymes of LDH. Thereduced tetrazolium, which is called formazan, is precipitated in the poresof the gel and is very difficult to leach from the gel SO that, if treatedcorrectly, a rather permanent record is obtained. The gel can be preservedand you can keep it for a long period of time. Lactate, which contains thehydrogens, is converted to pyruvate and the cofactor is repeatedly reducedand reoxidized. -Thus, the cofactor is recycled. If one uses a largequantity of lactate one can, in effect, recycle this reaction for a longperiod of time. In simple terms, this means that you can start off witha very small amount of lactate dehydrogenase and by recycling, for a periodof time, the very small amount of lactate dehydrogenase can be detectedbecause formazan is continuously being precipitated in the vicinity of the

48

isozyme. A similar reaction, cofactor, PMS, nitro-blue tetrozolium,is used to detect all types of dehydrogenases. Even coupled reac-tions can be used in which the enzyme of initial interest is not adehydrogenase.

Dan Cobal: There are millions of lake trout stocked in the GreatLakes. It is my understanding that there seems to be minimal naturalreproduction and I would like to ask the panel what they wouldrecommend be done to rectify that situation.

.

Ray Simon: One of the things about biological problems is that theyare technically unique and to attempt to categorize an answer isprobably stupid. Presumably it would clarify the situation if thehistorical background of that situation were known. That is, is thefailure due to intervention through the adjunct of the hatchery?Habitat conditions during the year between the good reproduction andpoor--how have physical conditions of the habitat changed? It'sunanswerable unless you begin to glue together the description ofthat particular circumstance which I'm not able to do here becauseI don't know it. I presume you resolve it by digging further intothe question which puts it back on you. That's a poor answer.

Stanford Smith: The primary problem, Ray, is that in the areas whereyou don't have natural reproduction you have lost your native strains.Where you do have natural reproduction you still have them. Like inLake Michigan, Lake Huron we lost all of the lake trout, now we arecoming back with other strains and they aren't reproducing.

Ray Simon: It sounds like we'd better get Fred Allendorf to build onethat resembles the ones that do succeed in their natural habitat. I'mnot being facetious, I don't know how to solve the problem.

John Driver: You are mentioning the historical significance of thelake trout and I'm in charge of the Marquette Lake Trout Hatchery.I've gone back to the records as far as I can, 1946, 1943, when theywere first starting and they are very scanty and as you have indicatedthere has been much inbreeding within those stocks. But on LakeMichigan, three years ago at Charlevoix, eggs were taken from wildfish (planted fish which were mature), and the viability of thoseeggs was very good. So I wouldn't say that you are not getting repro-custion in Lake Michigan. You are getting reproduction but not neces-

sarilp natural from the standpoint of them reproducing on the reef.They have the capability of reproducing because those wild fish eggs,a couple million of them, have been taken and have done very well in

Y. a hatchery situation. So it lends me to believe maybe that they haven'tbeen planted correctly in the past, to some extent. I don't know.

~Maybe the habitat has been destroyed, but they still have the capabilityof reproducing well. At least in certain areas-off of the researchfacility at Charlevoix, for one.

49

.

I .

Henry Booke: One of the questions I was going to pose before thisaudience in terms of stimulating questions like we are getting rightnow is how do we match some of these markers that we have discoveredfor these particular fishes against successful forms? Successfulforms from particular hatcheries that we wish to maintain as a naturalresource. We have to wait awhile before we can do that but at leastattempts are being made. I think the best approach is to look at whatwe have now and try to characterize what we have and are defining assuccessful. Then see if there is some genetic basis for this successsuch that we can select for it. Maybe we are dealing with the samefish that are being affected by different environmental conditions.Is there a genetic basis for differences found among representativesof one species? I can recall a paper published in the early 60's,probably around 1964, by Eschmeyer and Phillips where they describedcrossings of lake trout, fat and lean forms of Lake Superior lake trout,and obtained intermediate forms. There was a strong implication thatthere was a genetic basis for the difference between the two forma.This work hasn't been appreciated as far as using it in the hatcherysystems around the Great Lakes.

Peter Ihssen: I have a comment to that same question. Indicationsare that the lake trout of a lake like Lake Superior was constitutedof a number of genetically isolated stocks. For example, there usedto be river spawning, deep shoal spawning and shallow shoal spawninglake trout. Now we are trying to rehabilitate the lake with essentiallyone or two strains which may not have the proper potential for thespawning behavior that is required in the lake under present environ-mental conditions. For example, it may be that because of the changein species composition in the lake that certain spawning areas are moresuitable than others as for example deep shoals compared to shallowshoals or river spawning areas. It may take some time for the stocksused in present plantings to select for individuals who have adaptedto this new set of conditions. I think such a process of adaptationis sometimes indicated. For example, for the initial rainbow troutplantings in the Great Lakes, at first only little reproduction wasnoticed and then slowly there was a buildup of the population. A similareffect is being noticed for pink salmon which have a shorter life cycleand consequently the process of adaptation is more noticeable. For laketrout, on the other hand, because of their long life cycle, it may takea long time before we notice an appreciable buildup of the stock due tonatural reproduction.

Ed Brown: Did you indicate earlier that, from the standpoint of rainbowtrout that very little of the natural, or genetic, variation had beenlost?

Fred Allendorf: Yes. In the great majority of populations that I haveexamined this is true.

50

Ed Brown: I guess, along this same line, we are interested in thelake trout situation and such questions are layman-type questions.Just how fast is natural selection in the hatchery in terms of asalmonid population.

Fred Allendorf: The loss of genetic variation that I was referring mto is the kind that is associated with having small numbers of individualsreproducing the population. But in regard to your question, yes,that's a real problem because you are not only losing importantgenetic variation but also you are selecting for characteristicswhich are good in the hatchery. And if you want to use your productout in the wild, then you are going against your ultimate goal. Ithink the important thing is to define what you want to do with yourhatchery product.

Mark Noveck: In the lakes, my particular area of interest is theeffects of certain chlorinated hydrocarbons on lake trout developmentand my research has shown that early developmental stages of the laketrout fry are probably more susceptible to toxicity of these compoundsthan later stages. This could possibly play a part in the difficultyin natural reproduction of the species. So I think something like thatshould also be considered in their natural habitat. Especially becauseof the large lipid area of the yolk sac, for instance, which acts likea sink for compounds like PCB's.

Neil Foster: I just wonder if our preoccupation with hatcheries mightbe missing the point, to some extent, when in fact one of the thingswe should be examining very carefully is the adaptive physfology of thelake trout egg on theactual-spawning reefs--if we are trying to establishself-sustaining populations in lake trout, shouldn't we know more aboutthe genetic variability in egg physiology, for example?

Henry Booke: Well I think one of the implications, at least I havepointed it out and I think others on the panel have enforced thisidea either directly or indirectly, is that you first of all must beable to characterize what you have there and then if that is the formthat you are saying is the successful form, then work from that point.But if you're working from the point of view where you are building ahusbandry and superimposing those fishes on the natural population, Ithink that is the long route. Unless of course you are dealing with aput-and-take fishery, then you don't care what you've got as long asthe-fisherman gets the hooked product, but this is also what I was

I am worried about the attitude thati

aluding to at the very beginning.hatchery people base the rates of success on how many pounds produced--and there is nothing wrong with that. But in the final analysis weare really interested in what is the functional status of that individualfish when it gets into the environment. How does it sustain itself? Ifit doesn't, then we are blaming it on the environment and not the hatchery.Maybe we're wrong and maybe we're right. But right now we don't have muchof a basis or much of an approach, except by these methods which arebeginning to be appreciated.

51

Russ Daly: I think historically we have had year classes, perhapssomething like 14 to 16 out there at one time. We only startedstocking lake trout in 1965, in realistic numbers anyway. Basically,I think it just takes a lot of lake trout to make a lot of lake trout.And I don't believe that on some of the historic reefs we can do thatyet. When we begin to analyze and think about it we took off 6 millionpounds, harvested over 17 inches, during the hay-day of Lake Michiganand I don't think we can approach that right now. So I think, as thegentleman pointed out, we are a little premature in expecting somethingto happen. Even with the original native stocks on Gull Island Shoals,in Lake Superior, it took quite a few years once that population wasbrought down to its knees to have it restore itself naturally. Andthen to have it not only restore itself, but to have it virtually wipedout two years ago by an intensive fishery. So we are dealing, I think,a numbers game too, and , getting back to the hatchery product, notwithstanding PCB's and the DDT found in sac fry, some of our LakeMichigan progeny now are quite healthy to the fingerling stage. Ithink we're playing a numbers game.

Andy Lawrie: I would like to make a comment and I'd like to perhapsask the panel for their judgement on something. Apropos of the rehabili-tation of lake trout, at least in Lake Superior, I think something thatis very often forgotten is that what we know about this process is theresult of a sampling program which is carried out in the largest freshwater body in the world. And it is dependent, at least on the Canadianside, to a very considerable degree, on the patterns of fishing thatare carried out by commercial fishermen. This tends to define theplaces from which the samples come and the time of year in which theyare taken. And it thereby determines the information which you getand which you use to judge the success of your program. Now I am notsure that this is equally true on the U.S. side but it is my understandingthat the U.S. government utilizes their commercial fishermen who wereselected for their faithful reporting in the days when that fishery wasfree and that they asked these people to continue fishing in the sameway and in the same places that they had traditionally done. so, tosome extent, I would expect that there is a sampling problem on thatside of the line as well. A more explicit comment, that I find interest-ing and which I must confess I have sat on for a long time, largelybecause I haven't managed to look at the data from that particular pointof view until recently, has to do with the widespread belief and practicethat planting large yearling lake trout is the route to success in thisgame. It has something to do with Dr. Simon's reference to the factthat larger fish at planting yield larger returns to the fisherman--and this is certainly true. I think in terms of everybody operatingon Lake Superior, at least. On the other hand, I have recently puttogether the data from the first four year classes which Ontarioplanted into Lake Superior--at the two ends of it. By chance thosethat went in at the east end were small, about 1/5 the size of thosethat went in at the west end. They didn't come from exactly the samestocks, but they came from stocks that originally were derived fromLake Superior waters and had been held for the most part in inland

.

-

.

52

lakes in Ontario. Generally speaking, those four year classessurvived in very small numbers to reach, despite lamprey predation,ages 11, 12 and 13. First of all, each cohort in the lake veryclearly shows increasing survival at a very rapid rate as the resultof lamprey control in 1961. That is, the later year classes havebeen suffering less predation and have done better, and quite sub- -stantially better. The data also show a strong bias in favor ofthe larger fish at the west end, if you look at the total productionthrough their entire period in the lake. But it shows a bias ofalmost 10:1 in favor of the small fish at the east end of the lakeif you look at the numbers that have been present after they havereached the age of maturity; at about 8 years. Being rapid growingand reaching a large size early, in a fishery which is prosecutedby a size selective gear and in the face of a predator which is alsosize selective for larger fish, is clearly not a very advantageousthing to do if what you are looking for is numbers of mature fish todeposit eggs for you eight years after you have planted them. Nowwhether or not the analysis will hold true with later year classes,I don't know. But I think that we have been party to an activitywhich has pleased us very much at the time because we've had visibleevidence of success early in the game and which has defeated, orpartially defeated, our purpose in the long run. The question thatI would like to pose to the panel is this: As a person with responsi-bility, or some responsibility, for management of fisheries in aprovince that has 800,000 lakes, I get a little bit up tight at thethought of trying to apply the traditional population dynamics solu-tions to management problems over bodies of water of that number; notto mention for the infinitude of stocks which you are now presentingme with to occupy those waters. And one thought which has occurredto us, and on which we had been operating and which I think is sharedby our colleagues in other jurisdictions, is that we may be able toreduce the magnitude of our problem by looking for so called type-setsof lakes. Lakes which have sufficient in common in terms of the com-munity that they support and the limnology which goes toward determiningthat community and the climate in which they happen to reside. And ifyou know one you can say at least with some degree of resolution thatyou know them all and that you therefore have some kind of handle onhow to deal with the populations and communities that reside in these

lakes. So the question for you -Dr. Simon's made the comment I thinkat one point in his reply to an earlier question, and I appreciate whyhe said it, that the biological problems unfortunately always have ahigh degree of specificity attached to them. How are we going to copewith this problem of multiple stocks with their different characteris-tics and their potential adaptation to rather specific parts of themajor environment that we are in, if we cannot find some mechanism forrecognizing and delineating generalities amongst these?

Henry Booke: I posed that same question before a Sea Grant group abouta week ago and we didn't arrive at an answer, even after we were ableto describe what we think are two populations of the lake whitefish inthe boundary waters of Wisconsin and Michigan. Ultimately when we resolve

53

fish groups where we can almost pinpoint individuals, certainlypopulations or subpopulations of fish, our problems will increase.How would you manage fish stocks especially when you have commercialfishing mortality as opposed to all other types of fish mortality?I can't answer that question now but it is something I would like tothink about for a long while because I think many of our problems,evolving either from our successes or our failures, are based on thefact that we haven't recognized that we do have genetic variation infish and it has to be maintained. How do we maintain it under dis-turbances created by man and perhaps exotics? I consider the lampreyan exotic in the Great Lakes. I wouldn't even attempt to try andanswer the question, but I think we should be aware of genetic varia-tion in view of the approaches that we have to study it. We have thestocks. We might just leave them alone and we would always have them.But that won't be the case. That's not realistic. Perhaps someoneelse on the panel would like to also answer Mr. Lawrie's question?

Peter Ihssen: We now have talked at some length about these geneticallydifferentiated stocks for these 800,000 lakes. Very little is knownabout the genetic variability among these stocks and I think one ofthe first things we have to do is get a better inventory of what isout there. We spend a lot of effort looking at the characteristicsof these lakes, but we have not spent much effort in looking at thecharacteristics of the fish in these lakes and I think we have to getsome feeling for the genetic resource we have and so I suggest thatgreater emphasis be placed on collecting this information.

Edward Massaro: I'd like to back up what Peter has said. We have alot of information about the physical-chemical characteristics oflakes, but in comparison to the-gross effort and quantity of moneyspent in this direction very little has been spent on understandingwhat the genetics ultimately mean, i.e., what are the physiologicaland biochemical characteristics of these individual strains, species,or whatever they might be. We don't know at this time. The only waywe will find out is by more study.

Fred Allendorf: In terms of 800,000 lakes I'd like to further statewhat Peter said. I think an important point is that sitting here, wecan't tell you how much you can generalize and how much you can't. Butthere are certain basic questions, which if you had some baseline data,you could answer. You want to know: How different are the differentstocks in these 800,000 lakes? You can look at populations from dif-ferent lakes and you can estimate how genetically similar they are.Maybe you don't have discrete stocks; maybe they are all very similar.The next level is genetic sub-division within lakes. These general

principles have to be found; but, they only can be found by going out'and looking at the stocks which are present and trying to come up with

some handles.

54

Andy Lawrie: I think a cofactor has been identified here as money.The discussion might be furthered little by another kind of cofactorat a later time and I would challenge you to start breeding biochemistswho live to be 99. We are currently in the process of trying to carryout an inventory on these same lakes in Ontario at an apparently sub-stantial rate per annum. Something like 15 or 20 lakes in each of 40some odd districts every year and our present reckoning is that byspending one day on these bloody lakes collecting the kind of informa-tion you can get from a single overnight gill net set, plus the limno-logical data that you can acquire in 99 years, we may have just begunto get an idea of the problem that we are talking about. So its aproblem of very substantial magnitude and of course I agree with allof you. We support genetics activity in our agency because what yousay is obviously correct--we don't have a handle on it and we have tohave.

Carlos Fetterolf: The Great Lakes Fishery Commission hopes to have abinational workshop on the stock concept at some time in the future.Through the Commission and a recommendation of the Scientific AdvisoryCommittee, Henry Booke put this special session together to serve as aprelude to the stock concept workshop and to also provide us with aresource document which the workshop participants may read and be broughtup to some sort of speed before entering the workshop. On behalf ofthe Commission I would like to thank Henry and his participants fortheir fine presentations.

.

55

Species Management

Dwight A. Webster and William A. FlickDepartment of Natural Resources

Cornell UniversityIthaca, New York

This paper, originally published by Trout UnHmited, Inc. (TU)in @Proceedings of the Wild Trout Management Symposium" (1975),sponsored by TU and U.S. Department of the Interior, is reproducedwith the permission of the authors and the publisher.

. The topic “Species Manarcment” may be developed in several ways. but because of the specialinterest of this group. it appears appropriate to focus on intraspecific diversification as it rcla:cs tomanagement or potential management The special case of wild YS. domestic strains will be notedin somewhat greater detail as a practical example of the benefits of how certain stocking require-ments can benefit from the attributes oi wild trout.

Historically, we have tended to regard a species of salmonid as a homogcnous organism, regard-kss of origin. True, certain liic history diffcrcnccs were strikingly apparent within anadromousspecies, and these were more or less take11 into account when fish culture was involved in manage-ment. But with more readily domesticated trout, it has been easier. indeed ;u;tomary; to deveiopbrood stocks as a convenient source of erps. Thus with the commonly cultured species. . brook,brown, rainbow and cutthroat trouts. . there has been the opportunity to cvolvc a different phil-osophy and management procedure based on the fact that the entire operation could bc conductedindependently of the realities of natural environments. Not so with the cultural programs forsalmon and such species as stcrlhead and lake trout, where either availability, economics or logis-tical constraints have worked against widespread establishment of brood stocks. The genetic con-

56

sequences of these two divergent proedures 3re potentially considerable. The management con-sequences are no less SO, but do seem to be well recognized or considered. Gcner3lJy, cultu~ts ofanadromous species and lake trout lack the brood stock option and are forced to more or lessmaintain the integrity of the gene pool as it relates to performance in the n3tural environment.Alternations in genetic make-up inevitably occur, though perhaps in3dvertently. within the cultur-al phases of the life cycle. It does not necessarily follow that these changes 3re favonble whenjudged by the ultimate objectives of the program or by standards set by native stocks in naturalhabitat.

Examples of species or intnspecific diversity are numerouSand only a wmpling can be includedin this review. We note in passing the subsmntial evidence for m&i or stock differences in cerurnanadromous species 3nd concentr3te on sometimes more subtle vsri3tions th3t may be of eqa31management importance to inland fisheries. One sport fishing example in the anadromous group,however, that demands attention, is the summer steelhead prognm developed in Washington(Millenbach 1972). Summer steelhead Sui~no gairdneri’’ are more desirabie from an 3ngling st3r.d.point than winter steelhead, since they enter freshwater from bI3y to late summer in prime condi-tion, with the spawning season upwards of 3 yesr away. Through selection of the earliest sp3wnersof the summer run, by advancing the spawning time another two months by light control and bythe development of improved fish cultural techniques, it h3s been possible to produce the smolt-sized yearlings in time to coincide with the natural spring outmigration. The spectacular successof this program is well known.

.~

_- -

Production of one-ye3r hatchery smolts has long been the goal in cultural programs for Atlanticsalmon &fmo r&r. But production of smoit-sized salmon does not necessarily mean productionof physiologic31 smolts, as 3 number of agencies have found out. A series of experiences currentlyunderway in Iceland may be cited to illustrate the need to meld manipulation of growth with phy-siological requirements. This country is especially favored to accelerate salmon growth in hatch-eries as abundant geothermal water gre3tly facilitates raising rearing water temperatures. But re-turns from releases of smolt-sized yerirling salmon. reared inside at elevaled temperatures, weremuch lower than 2-ye3r smolts reared under 3n ambient temperature regime (Isaakson 1973).When the former group was exposed to the norm31 natural light cycle and several weeks of wintertemperatures, survival response, measured by return to the stream, was immediate. Returns fromthe cohort of the 1972 year class reared under these conditional manipulations, released in spring1973 and returning as g&e in 1974, was the highest among several groups, including those of thetraditional two-year smolts (Iwakson. personal communication).

In Norway, Professor Skj3erwold is currently conducting 3n ehborste experiment to test fordifferences in growth of juveniles among a number of stocks of Atlantic salmon. Norwegi3n COI-leagues inform us that already there are indic3tions of substantial strain differences in this pan-meter that could have application in management (Kjell Jensen. personal communic3tion). Elrlier,Carlin (1969) showed striking differences occurred in survival, r3nging from one to seventeen pcr-cent, between different f3milies of salmon rcsrcd under comparable conditions.

A New York example illustr3tes different results from stocking two forms of indigenotis laketrout Salvefiws turma~cuslz. one from the Adirondack uplands. another from the L3kr Ontario -Fmger bkes b3sin. These forms represent two different stocks that invaded the afc3 followingthe retreat of the Laurentian ice sheet. A stock from an e3stern refugium gained sccess to themountain arcas vi3 high level g13ci31 lakes th3t prcdsted the opening of w3tcrways to the west.when Joke trout I’rom 3 Mississippi rcfugium inv3drd the Great Lsktls basin. Both Adirondlck 3ndFinger Lakes stocks of trou; wcrr used for stocking pcrpows, but since eggs wcrc much more rcJd-ily obt3inablc from the 13ttcr source. uL:e of the Finger L3kes stmin in rhc Adirondacks ~3s rncvil-able. Marked pktntings over the p3s.t 25 years h3ve shown virtu3lly no survival of the Finger Lakesstnin when pl3ntcd in Allirond3ck waters. 3hhough companion rtlcsses oi n3tive strains did showreascnahlc recoveri?% Reasons for this dil‘fercnc. h3ve not been irrevstigared, but the man3gcmcnrimplications 3re obvious.

fjomo of us hsve given thoug!lt to the nceJ for shemic3l lsmprcy Ferromvron nwirlus sontrolm Lake Onr3rio. L3mprcy war: thcrc for some time bciore t!lr: irtkc trout declined, 3nd I3mprc>and.substanti31 ulmonidpopul3tions cor’xlst in two of the Fiugcr L~IJXS trihut3ry to Lake Ont34o.Was there some signiticsncc :o the higil recovery of 3 token plsnticg ot‘ Fin:cr Lakes strain I&ctrout rn~ldc in L;lkc’ Or,tario in the 1950’s. r:iowrw that were made in commcrcul nc’ti riur <o:n-plctely JccimJtcC -I!IC i\.rrillcr): stock ;1!‘tcr two or three ycurs’! A goup oi Gdntists db~ubi;.g fortwo d3ys the 3pprucnt ;Inomniy oi Iamprcy-trout rciJtion in wntrcll Sew York bhc%, ~Y~-cr-vis tkupper Great Ltics. wmc ut> with the ruphcml~m ol“‘at~omnloil31lon”. In our simplistic appro4iilto solving XI immelliate problem have wc oscriookcd somcrhing in spccics m3nsgorncnt 11131 m3;have been already worked out by nature?

.

57

Acute environmental degradation through acid precipitation is developing regionally in parts ofUstCm Canada and the tJ~MK%res and the southern tip of Scandinavia, where bCx3=d acidityof preCipit3tiOn Originating in the Ruhr Valley of Germany and from Britain has renderlakes and streams fishiess (Oden and Ohl, 1970). In the past two dec;ldes, many iakcs taweti comer of New York’s Adirondack Mountains have experienced 10 to 100 fold increases inacidity, and, as in Scandinavia, entire fish populations of al1 native species have been decimated hsome of these waters. Solutions to the problem are long term, and, while local aikalization maybc prdctical on a limited basis, what about the existence of strains of fish more tolerant of low pH?Recent findings in Pennsylvania. where the problem is acid mine drainage. have indicated a widerange in acid io1er3nce of brook trout SufPelilrus fbrtlittalis, both by individuals in the same groupas weI1 as between sroups (Dunson and Martin 1973). One of the seveal strains involved in the=tests was a domesticated New York strain that proved the least resistant to this kind of environ-mental stress. A remarkable body of water exists in New York, holyever, Honnedaga Lake, wherebrook trout live at pH values fluctuating about 5.0 and with the additional burden of conentra-tions of zinc at 0.02 to 0.15 ppm. Nonacclimatized test fish live only 1 to 2 days under most con-ditions (Scholicld 1965). We do not know if there is any genetic basis for this adaptation.

Thermal degradation has claimed many former trout waters. There is 3 marginal range of habi-tat where a more thermally resistant strain of trout might be useful in management. Unfortunate-ly, the choice of “ideal” water for hatchery purposes, i.e., spring water with minimal seasonal var-iations in temperature, probably mitigates against retention of genes controlling extremes of tem-perature tolerance when brood stocks are maintained as closed systems. Studies on the hybridsplake fonrirualis x ~ma)~cush where temperature preference and lethal temperatures differ widelyin the parental stocks, demonstrated thermal inheritance of resistance to high temperatures in oneof the backcross hybrids involving the brook trout maternal parent (Ihssen 1973). Geographic31intraspecific variations in lethal temperatures have been demonstrated for two subspecies of Arcticcharr Mvelinus al~~inus (hiccauley 1958).

Splake are a new breed of salmonid filing a useful ecologiCa role in management. Experience:in Ontario (Martin and Baldwin 1960), New York and elsewhere indicated that the hybrid hadsatisfactory survival and excellent growth in lakes containing non-trout species such as suckers,minnows and sunfish. Like lake trout, splake turn readily to 3 piscivorous diet, but seem morereadily available seasonally for ~113llow water angling.

Ontario has also conducted 3 forthright selection program for splake retaining the deep swim-ming habit of lake trout. but earlier age at maturity of brook trout (Berst and Spangler.1 970; Tait1970; Ihssen and Tait 1974). The objective w3s to develop a fish occupying the niche vacated by

lake trout with the advent of lamprey, but ma:ure at a size below that most subject to lamprey at-tack. The breeding program was successful in retaining these characters by the F6 or F7 genera-tion, although it is not yet clear if management objectives have been achieved.

Food habits of salmonids, either intra- or interspecifically, provide some interesting biologicaldifferences and probably more management options than we now take advantage of. Scandinav-ians, especially in Sweden. have long been preoccupied with interactions of populations of charr&kclitrus alpinus and/or brown trout Saho trutta because of the intense program of river im-poundment ‘and lake level regulation. Nilssen and Pejler (1973) and Aass (1973). among others,have shown that when charr or brown trout occur allopatricslly, food habits are quite differentthan when they occur sympatrically. In the former case, both tend to feed on benthic organisms,wue in the latter, charr become plankivorous, leaving the benthic littonl fauna to trout. In NorthAmerican lakes it is common to manage for more than one species, frequently with rainbow trout

&lmo gairdneri as one of the combination. Since this species tends to utilize plankton more effi-ciently than either brook or brown trout, for example, differences in the feeding habits and otherRqkments provide for variations in season and methods of capture, a good example of manage-ment at the full species level

Sympatrjc populations of Arctic charr showing great divergencies in growth rates are most in-triguing and common throughout the range of this species. A stunted form, rarely exceeding 9 to10 inches in length, is largely a plankton feeder, while 3 larger or “normal” form has benthic oreven piscivorous food habits. Both forms also occur allopatrically. Recent studies making USC ofekarophorctic techniques have confumed that observed ecological and morphometric characten(& and number of gill rakers, position of mouth) sre indicative of real populXion differences(Lenart 1972). This diffcrcntiation is believed to have tzken place prior to thC last deglaciation(Behnke 1972 and others). Wh3t uw wn bc mr\dc of this kind of ecological diversity?

In some of the European lakes containing both dwarf charr and brown trout, trout may reach alarge size feeding on charr. Evidence is accumulating that fish eating proclivities are, at least to

some degree, under genetic Control. Aass (1973, and penonal communication) describes instances

58

of where strains of fat growing, chvrIating brown trout z&z&+sfiabit when introduced intoot?t- tugh-mountain Norwegian stun!?v ,i;~n i&es, &,3‘i were out-performed by the native strain

of&oNn trout when t~=~-c- .rea to lowland lakes. Aass points out that while the fish eating habitmay have evclved over thousands of years in some populations, there are examples of its development in reservoirs during the past 40 years, and that transfers from these populations also retainedthe bh eating habit.

In North America we have similar examples of predator-prey relationships that may be morethan casual or opportunistic: the giant Lahontan cutthroat of Pyramid Lake Sairno clarki henslluwi.the pure form of which may be extinct (Behnke, MS 1971); the Gerrard strain of Kamloops min-bow trout from Kootenay Lake (Hartman 1969). and Kamloops trout in Pend d’oreille, both populations utilizing kokanee salmon as forage; and the large brook trout fomerly inhabitsting theRangeiey Lakes of Maine that fed on the now extinct population of stunted charr or biurbacktrout Sdvelinus alpinus oquassa (Kendall 1918).

There is much evidence bearing on differences in vulnerability to angling. a reflection of varyingfood habits and/or behavioral characteristics. Many of these involve domesticated stocks and arenot considered here. Two strains of cutthroat living in the same body of water in Colorado showedup quite differently in angling catches over 3 two year period (Trojnar and Behnke 1974). One,the Snake River cutthroat, appeared three to four times as often as expected on the basis of theproportion present in the lake. This same phenomenon is commonly evident in Adirondack testwaters stocked with more than one strain of brook trout. One experiment involved a native Adi-rondack strain (Horn L3ke), 3 Canadian strain (Assinica Lake) and domestic fish, and the 3ngiingcatch expressed as a percentage oi the stock on hand at the beginning of the season was: Assinica70, Domestic 50, Horn Lake 20. Trout caught while surface feeding during the heat of mid-daywere invaria’biy of the Assinica strain.

We have reviewed a sampling of options that have been employed in species management, othersthat might be explored to a greater extent. in either case, the innovation in some way usually ex-pioits diversification combined with an awareness of physiological or ecological needs. These ex-

.

amples refute the notion of 3 homogeneous equally useful strain for multipurpose management.The examples have focused on natuni populstions or those resulting from plantings of fish from

nokdomestic stocks. Of particular relevance and interest in this Symposium is the relative perfoorm-ana of wild and domestiured strains of trout. A substzniai body of irliormation of wide-spreadorigin has accumulated on this subject, but at the risk of being provincial. we hsve to review theessence of this ph3se of species management on the basis of our New York expe;iments and exper-iences (Vincent 1966: Flick and Webster 1962, 1964: Flick 1971; and unpublished data\. Currentawareness and contributions in this 3rca in C3iiforni3, however. should be noted in psssing (for ex-ample, Cordona 3nd Nicob 1970,3nd W. D. Weidlein, personal communication).

New York fish managers, IS well 3s others. have noted th3t generally low angling or test nettingrecoveries followed plantinp of domestic3ted stnins of brook trout in the Adirondack hlountains.New York (Ziliiox snd Pfeiffer 1960). These obxrvations included the mitigated environment ofreclaimed waters devoid of competitive fish, where few survivors were found after age two. Thelongevity of natuni popuktions of wild brook trout. in contrast commonly extended to three orfour years, and age five fish \RI ‘re not r3rc (lot tit snd person31 obsem3tions of 3uthors1. An earlyexperiment by Greene (1952) suggested that wild hatchery reared trout offered m!mapement petentiai. This and other consider3tions prompted 3 program initiated in 195s to quantify more pre-cisely relative performance bctwcen domestic and F1 wild 1 h3tchery reared strains. The investiga-tion was funded from priv3te sources and conducted on private lands, thus greatly enh3ncing tlexi-bility of opentions and control over angling.

-,

Data on survival within the first year of life ~3s obtained from a small (0.5 acre) drain3ble pondfed by a small cold brook. Spring fingerlings. rck3sed in spring. were inventoried in autumn. 3ndsometimes rrturned to the pond 3nd inventoried a final time the following sprinK. Data on survivalthrough the life span of any given cohort ~3s obt3incd by planting spring or fall fingerlings ir. n3:tural ponds containing no fis!i or only brook trout, and estimating st3ndinp crop 3t scmi3nnu3l in-tervals. All of the waters used (Long. Bear and Bay Ponds) were loc3ted within distances 01‘ fivemiIes of one another in-tile tteadwaters of the St. Regis River in the northern Adirondack Moun-t & S .

Six wild stnins of trout were involved in thcw studies. four n3tivc to the Adirond3cks 3nJ tworecently n3tumlizcd in New York w3tcrs I’rom the j3mcs B3y 3rr\3 of QUCIXC: pcrformuncc ~l’cnl>four will bc cited in this pqxr. The New York locations wcrc 3 mountain-top pond containingonly n3tivt: trout (ilorn L&z). a l3rpr. clecp, asidolropic body of water (Honned3ga L3k). 3nd n3sm311 uphnd brook (Long Pond Outlet). Longevity in 311 popul3tions was a minimum oi five ye3rs.but the stream population was stunted, 3versging 6 to 7 inches in length in tr3p and angling

59

Yn~Ples* and natural mortality was excessive after age two. Canadian stmins of trout wcre from theAssinica or Temiscamie area southeast of James Bay, where they attained a sue of over five poundsand longevity extcndcd to nine years (confirmed by known age naturalized trout).

Domestic Steins Of trout were from CWO sources. J so-called “Berlin” strain cultured at the ~a-tional Fish Hatchery at Cortland, Sew York, and a “New York” strain. generally uwd for stockingin that State. Both had been propagated from brood stocks with a Ions Ilistory o(J cloScd genepool. Under cultuml conditions. both showed rapid growth, attaining lengths of 4 to 6 inches asfall fmgerhgs and an early age at n’l3tUrity (0+ In larger males and I+ for females). Compared withwild hatchery reared groups, they were robust in appearance and exhibIted a wide range of behav-ion1 differences (Vincent, lot tit).

Results of a series of experiments involving five year classes comparing two to seven differenth’ew York brook trout strains in each experiment were consistent In showing substantially higher

SUtYiVal Of wild groups between spring and fall of the fit year of life (Flick and \\‘ebster 1964and unpublished). Among the twelve paired comparisons, survivaI in wild strains avcn~ed 15 per-cent higher, with a range of 12 to 43 percent. Two experiments evaluated the effect of parentalhistory on survival, but it made no difference whether all groups were reared to maturity In ahatchery environment on a standard hatchery diet or in 3 natural envuonment on natural food;survival oi wild stCihs was always higher. One experiment included two interstrain hybrids be-

tween wild and domestic stocks, and both hybnds also proved superior in survival to that exhibitedby the domestic parent.

Survival and production data over the life span of the several strain cohorts obtained from semj-annual estimates of population size and growth in natural ponds were no less convincing on thepositive attributes of wild strains in a pond management program for brook trout. All live experi-ments with two or more strain-cohorts led to esscntialiy the same conclusions. The 1960 yearclass in Long Pond provides an example of the results obtained and data on planting is given inTable 1. The larger size of domestic fingerlings reflects the adaptation of this strain to cultunlpractices and would be regarded of positive survival value. To eliminate possible effects of finclipping, one of the ventrcll iiis was used to identify the domestic group and one of the wildstrains.

Population size of the three cohorts in the 1960 year class is shown in Figure 1. Domestic strainfish were essentially extinct at age three. while both wild strains exlsted in substantrai numbers.Wild fish at all ages dominated the population. Since angling took place from age two onwards,the curves reflect losses due to this source as well as natural causes. Domestic strain fish were m-itially larger in weight, but 311 groups resched a climax size of 3bout 0.8 pounds (Figure 2). Tlussize is not a definitive parameter, since it merely relfects response to conditions of stock densityduring the course of the experiment.

The combined effects of growth and survival are depicted by the biomass (population site x meanweight) present through the life span (Figure 3). Also shown are the pounds of fish in each strainremoved by angling. After age one, it is clear [hat a substantially bugh btomass is on hand In the twowild stnins, also reflected in angling catch. Here 3 total of 40, 73 and 68 pound5 MS h3rwstedfrom Domestic, Long Pond Outlet and Honncdaga sframs, respectively. Reduced utches duringages 4 and 5 were due lo decreases in fishing effort.

Comparison of the three groups is facilitated by computation of gross production (Table 2); this eliminates the effects of angling and includes an estimate of all biomass elaborarcd. including losses

to natural ~3~95 during each semi-annual Interval (Ricker 1958). Gross prod~rlon is n3dlly rr3ns-formed (after subtracting the weight at stocking 01’ the strain-cohort) to 3 r.1110 showinf the numberof pounds produced in the pond per pound of trout stocked. For rhr Domchtlc str31n. this amount-ed to 7 pounds, for the two wild strains, 80 and 51 pounds or 3 proportlonai r3tmS of 1: I?:?. Asimilar, but independent rating. can be calculated from biomass harvcstrd by 3nrlmS diwkd bybiomass stocked, or 1:13:8. These relative ratings were calculated for five expcrlments. 3nd with-out exwption, wad strains gave substantially higher recoveries when judged by thcsc parameters

< (Table 3).. Perforrnane of the interstrain hybrid of Assinica x Domestic in Bay Pond ~3s esp~~~3lly notable.

_ q-his cohort exhjbitrd faster initial growth and higher survival than elthcr of the pux PJrcntalstrains, resulting b lifetime gross productIon estimates of 3b0ut 1 .bOO i)ounds I’9r the hYbrld% comepared Wj(h 460 and 265 pounds oi the wild 3nd dOmcStlC parcnt.ii SIOCkS. In fiW JnKl:% ‘(CJSons,432 pounds of hybrids wex removed. 3vera~ing 1.4 pounds. In scvcrJI ofhcr x3t~rs. lkcsc Inter-str;ljn }ly&,rlds have consjs[cntJy ouI&fown ciomesttc slnins stocked dt the Same LMC, sukWjtinL3hybrid vigor in this character.

A&je from intrinsic v3iues i.n d3t3 showing wide dispxity cf perform3ncr under nJtur31 fieldconditions, the ratios m Table 3 have direct besrmg on &ne!it-cost Judgments uhcn h3ccheV rc3wd

60

ftih are used in the kind of management programs under considention. They form a viable altema-tive for judging the effectiveness of a hatchery product compared with traditional methods based onhatchery performance alone. The several experiments exhibited a range of values in favor of wildstrains, but an average of about five times higher weight recovery can be taken as the basis for aworking estimate (Table 3). Thus, for any given unit cost per pound of fish in the hatchery, wildstrains produced five times as much poundage in natute as domestic strains. This assumes no diifer-en&i in rearing costs between the two gmups, but there is considemble cushion to absorb any likelyadditional cost associated with raising wild strains. Furthermore, most hatchery techniques anddiets have been developed with domesticated strains in mind so that changes favoring wild strainscould modify potential added production costs.

The use of wild strains of trout for stocking appropriate waters in rehabilitation or maintenanceprograms may constitute only 3 small part of the total propagation program of governmental agen-cies. But economic payoffs seem assured, and the development of high quality angling experiencesthrough appropriate regulations and use of special strains has a definite spot in the repertoire offishery managers.

Finally, we would like to echo the plea, so earnestly expressed on numerous occasions by Dr.Robert Behnke of Colorado State University, that we preserve the genetic integrity of such of ourheritage of salmonid fishes as we will have left, and that as managers, we recognize the genetic di-versity and plasticity of salmonid fishes in the context of species management and use it when appropriate for innovative improvements in the more traditional approach to management (Behnke1971).

lfhe term Fl wild is used here to clarify the semantics connected with the term wild as loosely applied in the litera-ture, and signifies the first pnention of hatchery reared stock from original native populations unaltered (pre-sumably) by previous cultured introductions This convention was suggest by hloylc (1966 MS).

Table 1 . S tak ing deu o n t h e 1 9 6 0 Y e a r Clau nlcrsed in Long P o n d . O c t o b e r 1980.

strainA-ram L-W Total W4ight

NUttlbN hched tPOUtttlSl Pin Clp

DblWSiC so 4 . 3 1 7 . 8 R V

Long Pond o u t l e t 360 2 7 2 4 A 6

HomQgs 3 6 0 2 9 2 8 LV

Table 2 Conputrtion of rehtion of the biotneo of gross production to biomass of trout staked. 1960 ysrr c*u - Long Pond.

Swin

Domrstic

Long P o n d o u t l e tHonn4d4s4

GIOP Rodueion Eionwsv Stocked Groo P r o d .(@oundsJ boundst Biorrwm Stocksd Rati-

116 1 7 . 5 7 1

1 9 3 2 . 4 8 0 12 .1 9 4 2 9 6 6 1 0

Strain B u r ‘59l Lot!4 ‘6of Long ‘6x1 Brr ’61s 8ev ‘68f

Dolultk 5 11) 1 (11 1 0)’ 1 ( 1 ) 1 ( 1 )

Long P o n d Out*1 3 (41 1 2 (131

xoinrdspa 4 IS) 1O (101

Horn 7 (6) 4 (7)

Aninicr 3 131

Ash I D o m 7 112)

Tomisomi 2 (4)

62

REFERENCES CITED

Aass, Per. 1973. Some effects of lake impoundment on salmonid in Norwegian hydroelectric reservoira, Abstractsof Uppsula Dksertatlons from the Faculty of Science. Number 234, 13 p.

Behnke. Robert. 1971 MS. The toogeognphy, rystematics and management of cutthroat trout Presented atAm&an Fkhenes Society Exhibit- Salt Lake City. Mimeo 8 p.

Behnke. Robert. 1972. The systcmatics of nlmorud fishes of reantly glaciated lakes 1. Fish Bd. Can. 29(6):639.k67.

- -

Bent. A. H. and G. R Spanger. 1970. Population dynamics of Fl splake Salvclinur fontimlir x S. numuycwh inLake Huron J. Fish Bd. Gn 27(6): 1017-1032

Cartin, Barge. 1969. Salmon tagging rxperimentr Swed Sal- Rer Inst. Report LFI MEDD. 2-4:8-13.Cordone, A J. and S. 1. Nicola 1970. Hy~est of four strains of rainbow trout, Salmo pirdnrrii, from Bexdslcy

Reservoir, Californiz Calif. Fish and Game. 56(4):271-287.Damson. W. A. and R R Martin. 1973, Sur*ival of brook trout in a boderived acidity gradient Ecology. Sq6):

137&1376.Flick, W. A and D. A Webster. 1962. Problems in sampling wild and domestic stocks of brook trout Suhclimu

fontinufis. Trans. Amer. Fish Soc 91(2): 146144.Flick, W. A. and D. A Webster. 1964. Comparative fist year survival and production in wild and domestic strains

of brook trout Salvelinw fontinalis. Tram Amer. Fish Soc- 93:58-69.Flick, W. A- 1971. New trout for old waters N. Y. State Consen. 25(6):18-21.Green, C. Willard. 1952 Results from stocking brook trout of wild and hatchery strains at Stillwater Pond. Traru.

Amer. Fish Soc 81:4X2Hartman, G. F. 1969. Reproductive biology of the Gerrard stock rainbow trout In T. C. Northcote (ed.) Sym-

posium on salmon and trout in streams, K R .MachfiJlan Lectures in Fisheries Univ. British Columbia, Vmcower. 8. C pp 53-67.

Ihssen, Peter. 1973. Inheritance of thermal resistance in hybrids of Sulvefinw fontthtZir and S mrmoycurh. 1. Fish.Rer Bd. Can. 30(3):401-408.

Jhssen, P. and J. S. Tait. 1974. Genetic differraces in retention of swim bladder gas between two populations oflake trout Sulvtfinur numaycu~h. 1. Fish- Res. Bd. Can. 31(8): 135 l-l 354.

isaksson, Arni. 1973. The result of tagging experiments at Kollafjordur Experimental Fish Farm from 1970through 1972. ICtS C.!& 1973/m:26,7 p.

Kendall, W. C 1918. The Rangefey Lakes, Maine; with special reference to the habits of the fishes, fuh cultureand angling BuL Bur. Fish 35:487-594.

McCauley, R. W. 195.8. Thermal relations of geographic raas of Ploelinur. Can J. Zool’ 36:655-662.Martin, N. V. and N. S. Baldwin. 1360. Obscrvatlons on the life history of the hybrid between eastern brook trout

and lake trout in Algonquin Park. Ontario. J. Fish Res. Bd. Can 17(4):541-550.Millenbach, C. 1972 Genetic selection of steelhead trout for management purpose& Int. Atlantic Salm. Found.

4(1):25%257.Moyic, P. B. 1966 MS. Comparative behavior of different strains of hatchery reared brook trout Solvclinitr~onli-

~fit Mitchill. blasters Dsrtation, Cornell Univ. 107 pp. l

Niin, N&Avid and Birger Prllrr. 1973. On the relation between fish fauna and zooplankton composition innorth Swedish lakes. Insti of Freshwater Res Rep. 53:51-77.

Nyman, Lennart. 1972. A new approach to the taxonomy of the~“Solvelinuroipinw species complex.” Inst ofFreshwater Rer Drottnmgbolm. Reporr 52: 103-131.

Oden, S. and T. Ah!. 1970. The acidification of Scandinavian waters Ymer, Anbok: 103-122. (Translation Ser.No. 2535, Fish Rcr Bd. Can.)

Ricker, N. L 1958. Handbook of computations for biological statistics of fish population% Bull- 119. Fish RerBd. Can.. 300 p.

Tait, J. 1970. A method of selecting trout hybrids Salvclinw/onrinulir MirchiN. Tranr Amer. Fish. Soc 89(1):3s.50.

Trojnar, J. R. and R. 1. Behnke. 1974. Management implications of ecological vgregation bet&een two introducedpopulations of cutthroat trout in a small Colorado lake. Trans. Amer. Fish- Soc- 103(3):423-430.

Vincent, R. E. 1960. Some infIuenccs of domestication upon three stocks of brook trout Sof~elin~s JonridisNirchill. Tronr Amer. Fish Soc. 89(1):35-50.

ZiJliox, R G. and Martin Pfelffer. 1960. The use of rotenone for management of New York trout waters. Can.Fish Cult 28:3-12