A population investigation of Atlantic menhaden : a meristic, morphometric and biochemical approach

Citation
A population investigation of Atlantic menhaden : a meristic, morphometric and biochemical approach

Material Information

Title:
A population investigation of Atlantic menhaden : a meristic, morphometric and biochemical approach
Creator:
Epperly, Sheryan Patricia.
Place of Publication:
Tampa, Florida
Publisher:
University of South Florida
Publication Date:
Language:
English
Physical Description:
107p.

Subjects

Subjects / Keywords:
Atlantic menhaden
Dissertations, Academic -- Marine science -- Masters -- USF

Notes

Abstract:
Juvenile and mature adult Atlantic menhaden, Brevoortia tyrannus, were sampled 1978-1979 from estuarine and coastal waters of the Atlantic coast of the United States. Juveniles spawned in the North Atlantic in the spring had significantly lower number of vertebrae, trunk vertebrae, ventral scutes, anal ray supports, shorter, shallow heads, smaller eyes, shorter predorsal lengths and lower frequencies of the fastest allele observed at the transferrin locus than fall and winter spawned juveniles of more southern areas. Middle and South Atlantic juveniles were uniform with respect to all characters except vertebrae and dorsal ray supports. Latitudinal clines, with smaller or lower numbers of characters to the north, were observed for head depths, orbit diameters, predorsal lengths, vertebrae, trunk vertebrae, ventral scutes, anal ray supports and frequencies of the Tf(ll0) allele . Discriminant analysis successfully discerned the two populations. In decreasing order of power, the following variables were important in the discrimination: ln(HD/SL), ln (CPD/SL), ln (00/SL), SCUTE, TVERT, CVERT, and ln (PDL/SL). Characters of Middle and South Atlantic juveniles were homogeneous with respect to spawners of those areas. Adult samples collected in Rhode Island, May 1980, and proposed to represent the North Atlantic spring spawners were not homogenous with juveniles collected in that area. Based on meristics, morphometries and biochemical data, samples were not taken from a panmictic population. At least two populations are proposed. North Atlantic spring spawned juveniles are not from the same population as Middle and South Atlantic juveniles. The amount of gene flow between the populations is unknown. ( English )

Record Information

Source Institution:
University of South Florida
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
028608317 ( ALEPH )
F51-00037 ( USFLDC DOI )
f51.37 ( USFLDC Handle )
LD1801.F6m1981 E66 ( USF Library )

Postcard Information

Format:
Book

Downloads

This item has the following downloads:


Full Text

PAGE 1

A POPULATION INVESTIGATION OF ATLANTIC MENHADEN: A MERISTIC, MORPHOMETRIC AND BIOCHEMICAL APPROACH by Sheryan Patricia Epperly A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in the Department of Marine Science in the University of South Florida March, 1981 Major Professor: Dr. John C. Briggs

PAGE 2

Graduate Council Univer sity o f South Flor ida Tampa, Florida CERTIFICATE OF APPROVAL MASTER1S THESIS This is to certify that the Masters Thesis of Sheryan Patricia Epperly with a major in Marine Science has been approved by the Examining Committee on 25 November, 198 0 as satisfactory for the thesis requirement for the Master of Science degree. Thesis Committee: Major !7'6fessor: Dr. John7 C. Br1ggs __ Membe'f: Dr. J. Bi'll'Lnc( Sullivan Mt?mber: Dr. Ha ro 1 llumm l>tonbrafv Merrlt6er: Dr. Joseph J. Torres

PAGE 3

Sheryan Patricia Epperly 1981 All Rights Reserved

PAGE 4

ACKNOWLEDGEMENTS Many persons were involved in the successful completion of this research. Field collections were made possible by J. F Guthrie and aided by C. Krouse, D. Halliday, M. Judy, J. Terrill, D. Heinze, R. Gallo, K. MacPherson, A. Durbin, T. Durbin and J. DeVane. Many than k s are extended to the following commercial fishermen and their crews: Captain Abbott (Atlantic Breeze), Captain Nicastro (Rock away), Captain Davis (JohnS. Dempster, Jr.), Captain Loftes (Oceanstate), and T. Smith. Arrangements for these collections were made by A. Durbin, P. Doody, A. B. Crowthers, M. Covington, B . Humphries, B Jett, R. Cahoon, W. Nelson, R Chapoton, and D. Dudley. I am indebted to Zapata Haynie and Standard Products Corporations for their full cooperation and hospitality and to D. Ahrenholz, B. Lewis, W. Nicholson, H. Gordy, C. Krouse, and J. Hollingsworth for their help. I extend special thanks to my committee members and to S. W. Ross for help and encouragement, and to B. Harvey for her patience with my illegible scribbles. Two persons were instrumental throughout the course of the project. Enough appreciation cannot be expressed for the arrange ments, conversations, supervision and encouragement received from Drs. J Bolling Sullivan and Walter R. Nelson. i i

PAGE 5

TABLE OF CONTENTS Page LIST OF TABLES iv LIST OF FIGURES vi. ABSTRACT viii. INTRODUCTION 1 REVIEW OF THE LITERATURE 4 MATERIALS 7 METHODS 12 Meristics and Morphometries 12 Biochemistry 14 Data Analysts 14 RESULTS 16 Meristics and Morphometries 17 Biochemistry 58 DISCUSSION 79 CONCLUSIONS 92 RECOMMENDATIONS 94 LIST OF REFERENCES 96 i i;

PAGE 6

Table 1 Table 2 Table 3 Table 4 Table 5 Table 6 Table 7 Table 8 Table 9 Table 10 Table 11 Tabl e 12 Table 13 Table 14 LIST OF TABLES Collection data for each (a ) juvenile sampling location and (b.) Adult sampling location. Merist i c and morpho metric variab l es of Atlantic menhaden investigated in this study. Mean ovary indice s for adult Atlantic menhaden. Mean gonad indices for male adult Atlantic menhaden. Weight-length regressions of Atlantic menhaden. Intercepts and slopes of morphometric regressions for geographical areas. Intercepts and slopes of morphometric regressions for populations as designated .by June (1958, 1965}, Sutherland (1963) and Dahlberg (1970}. Probabtlities of significant t values by chance from independent t-tests of morphometric regression coefficients between geographical areas. Probabilities of significant t values by chance from independent t-tests of morphometric regression coefficients between populations as desig nated by June (1958, 1965), Sutherland (1963) and Da h 1 berg ( 1970}. Means of morphometric characters of juvenile Atlantic menhaden (a). By collection station, (b). By geographical area, and ( c) By population. Means of morphometric characters for adult Atlant i c menhaden. Juvenile Atlantic menhaden meristi cs summarized by areas and populations Means of meristic counts for adult Atlantic menhaden. Means of Atlantic menhaden meristics. iv page 10 13 18 18 19 22 22 23 23 24 27 32 33 47

PAGE 7

Table 15 Frequency of transferrin phenotypes of juvenile Atlantic menhaden. 65 Table 16 Transferrin gene frequencies of juvenile Atlantic menhaden. 66 Table 17 Hardy-Weinberg tests of transferrin phenotypes from (a). North Atlantic juveniles ( b). M iddle Atlantic juveniles (c). South Atlantic juveniles and (d). All juvenile samples. 69 Table 18 (a) Variances and (b). Tests for equality of variance s among geographica l areas for the transferrin alleles of juvenile Atlantic menhaden. 70 Table 19 Independent t-tests of Tf(llO) frequencies between geographical areas. 71 Table 20 Transferrin (a). Gene frequencies and (b). Hardy-Weinberg tests on transferrin phenotype frequencies of adult Atlantic menhaden. 74 Table 21 Transferrin (a). Phenotype and (b). Gene frequencies for adult and juvenile Atlantic menhaden. 75 Table 22 Results of enzyme specific stains on electrophoresed muscle homogenates of Atlantic menhaden. 76 v

PAGE 8

LIST OF FIGURES Page Figure 1. Collection sites for Atlantic menhaden. 8 Figure 2. Mean num ber total vertebrae of juvenile Atlantic menhaden samples plotted against latitude of collection station. 29 Figure 3. Mean number trunk vertebrae of juvenile Atlantic menhaden samples plotted against latitude of collection station. 35 Figure 4. Mean number caudal vertebrae of juvenile Atlantic menhaden samples plotted against latitude of collection station. 37 Figure 5. Mean number of ventral scutes of juvenile Atlantic menhaden plotted against latitude of collection station. 40 Figure 6. Mean number of interneural spines of juvenile Atlantic menhaden plotted against latitude of collection site. 43 Figure 7. Mean number of interhaemal spines of juvenile Atlantic menhaden samples plotted against of collection station. 45 Figure 8. Frequency distributions of the morphometric body depth and transformations of its proportion to standard length for Barnstable Harbor. 52 Figure 9. Cluster dendrogram of juvenile and adult Atlantic menhaden. 52 Figure 10. First and second canonical variables for juvenile Atlantic menhaden. 54 Figure 11 Electropherograms of red blood cell lysates of Atlantic menhaden. 59 Figure 12. Electropherogram of transferrin phenotypes of Atlantic menhaden. 63 vt

PAGE 9

Figure 13. Frequency of the Tf(llO) allele in juvenile Atlantic menhaden samples plotted against latitude of collection station. 67 Figure 14. Relative abundance o f North Atlantic juvenile menhaden line) and juvenile Atlantic menhaden from all other areas combined (broken line). 86 vii

PAGE 10

A POPULATION .INVESTIGATION OF ATLANTIC MENHADEN: A MERISTIC, MORPHOMETRIC AND BIOCHEMICAL APPROACH by Sheryan Patricia Epperly An Abstract Of a thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in the Department of Marine Science in the University of South Florida March, 1981 Major Professor : Dr. John C. Briggs viii

PAGE 11

Juvenile and mature adult Atlantic menhaden, Brevoortia were sampled 1978-1979 from estuarine and coa stal waters of the Atlantic coast of the United States. Juvenile s spawned in the North Atlantic in the spring had significantly lower number of vertebrae, trunk vertebrae, ventral scutes, anal ray supports, shorter, shallow heads, smaller eyes, shorter predorsal lengths and lower frequencies of the fastest allele observed at the transferrin locus than fall and winter spawned juveniles of more southern areas. Middle and South Atlantic juveniles were uniform with respect to all characters except vertebrae and dorsal ray supports. Latitudinal clines, with smaller or lower numbers of characters to the north, were observed for head depths, orbit diameters, predorsal lengths, vertebrae, trunk vertebrae, ventral scutes, anal ray supports and frequencies of the Tf(llO) allele. Discriminant analysis successfully discerned the two populations. In decreasing order of power, the following variables were important in the discrimination: ln(HD/SL), ln (CPD/SL), ln (00/SL), SCUTE, TVERT, CVERT, and ln (PDL/SL). Characters of Middle and South Atlantic juveniles were homogeneous with respect to spawners of those areas. Adult samples collected in Rhode Island, May 1980, and proposed to represent the North Atlantic spring spawners were not homogenous with juveniles collected in that area. Based on meristics, morphometries and biochemical data, samples were not taken from a panmictic population. At least two populations are proposed North Atlantic spring spawned juveniles are not from the same population as Middle and South ix

PAGE 12

A t l a ntic j uveniles The amount o f g e n e f low bet w ee n the popul ations is unknown. Abstract approved : or Prof e s sor Professor, Department of Marine Sci ence 25 November 1980 Date of A p p rova l X

PAGE 13

INTRODUCTION Atlantic menhaden, Brevoortia tyrannus (Latrobe), are important fish of the western North Atlantic, occurring in vast numbers in coastal waters. They play a major role in estuarine and coastal eco systems and support one of the largest commercial fisheries of the eastern United States. Hildebrand (1963) described the range of the species as Nova Scotia to Florida. Debates have arisen over the nature of the morphological variation in this species throughout its range. Information on certain meristics and morphometries, growth rates, migration patterns and the spatia-temporal nature of spawning has indicated the existence of two or more populations (June 1958, 1965; Sutherland 1963; June and Nichol son 1964; Higham and Nicholson 1964; and Dahlberg 1970). Decreases since the mid 1970's in both the number of older fish and in the number of juveniles in the North Atlantic may indicate the decline of a postulated northern population (June and Reintjes 1960; June 1961, Nicholson and Higham 1964a, b, 1965a, b; Roithmayr 1963; Nicholson 1975; Henry 1971; and personal communication by Ahrenholz, National Marine Fisheries Service, Beaufort Laboratory). However, tagging data showed regular north-south seasonal movements and substantial intermingling of individuals tagged at different localities (Costen 1971; Kroger et 1971; Kroger and Dryfoos 1972; Nicholson 1972, 1978; Dryfoos et 1973). Nicholson (1972) reported latitudinal

PAGE 14

in the relative density of each age group along the coast and stratification by size within each age group. The large fish moved farther north Atlantic menhaden, although appearing to spawn throughout the year, reach spawning peaks in s hore in the North Atl.antic in spring, nearshore in fall in the North Atlantic, and offshore in fall and winter in the Middle Atlantic and South Atlantic Bights (Higham and Nicholson 1964). Temporal segregation of spawners could effectively isolate spawning groups even though the groups otherwise intermingle. Nicholson (1972) argued that differences observed among fish from different regions may be attributable to environmental differences during early development The identification of populations within a species is essential to the understanding of fluctuations in distribution and abundance of the species. Each population may have characteristic natural mortality and growth rates, distribution, fecundity and/or other biological traits. The problem first involves identification of the populations then, secondly, identification of the specific biological characteristics necessary for good stock management. The desirable feature of biochemical data used in population identification is that it is possible to work with characters known to be genotypic. In morphological studies it is generally difficult or impossible to ascertain how much the genotype is disguised by the phenotype. The purpose of this research was to determine if Brevoortia tyrannus occurs as one population throughout its range without subdivision of the gene pool. Meristic and morphometric data were taken to compare this research with that of previous studies (June 1958, 1965; 2

PAGE 15

3 Sutherland 1963; Dahlbarg 1970) and u se d in conjunction with bioche mical data to attain the objecti ve.

PAGE 16

. REVIEW OF THE LITERATURE Numerous scientific investigations have been conducted on the biology of Atlantic menhaden. Annotated bibliographies published by Reintjes et (1960), Reintjes (1964), Reintjes and K eney (1975) and Reintjes and Hall (manuscript in preparation) index those publications pertinent to menhaden. Few biochemical studies have involved the Atlantic menhaden. Sindermann and Honey (1963) published the electrophoretic pattern of menhaden hemoglobin; Saffran and Gibson (1978) and Pokrywka and Gold (1980) investigated the kinetics of the hemoglobin components. Markert et (1975) characterized menhaden lactate dehydrogenase from different tissues relative to C subunit quantities. Reintjes (1966) referenced preliminary attempts by the National Marine Fisheries Service Laboratory in La Jolla, CA at gel electrophoresis of Atlantic, Gulf and yellowfin menhaden for the purpose of investigating intergradation between these species. Based on 6 fish they showed that patterns tyrannus x smithi hybrids were indistinguishable from those of fish identified as yellowfin menhaden. In contrast, Lewis (Eletrophoretic comparison of two species of menhaden and their hybrids in Indian River, Florida. Unpublished manuscript) tyrannus, smithi and their hybrids collected in Indian River, FL by the electrophoretic patterns of lactate dehydrogenase. Those identified as F 1 hybrids by their morphology appeared similar to B. tyrannus in 4

PAGE 17

their LDH patterns. Parya1bumins tyrannus have been characterized (Epper1y, S .P., Parvalbumins of Atlantic menhaden, Brevoortia tyrannus (CLUPEIDAE), Unpublished manuscript}. These calcium binding proteins, 13,900-15,300 daltons. were shown to consist of a major and minor component, both of which were monomorphic. DeLigny (1969) published an extensive review on the biochemical studies of fish populations and edited a volume of symposium papers on the subject (1971). Serological, biochemical and immunegenetical studies have identified subpopulations in Auxis (Taniguchi and Nakamura 1970), Catostoma (Koehn 1969, 1970), Centropristis (Chapman 1977), Chrysophrys auratus (Smith et 1978), Clupea harengus harengus (Sindermann 1962; Sindermann and Mairs 1959; Odense 5 et 1966; Ridgeway et 1970; Wilkins and Iles 1966; Zenkin 1966), Coregonus c1upeaformis (Clayton and Franzin 1970; Franzin and Clayton 1977; Kirkpatrick and Selander 1979), Cynoscion (Weinstein and Yerger 1976a, b), Cyprinus carpio (Scopes and Gosselin 1968), Esox lucius (Healey and Mulcahy 1980), Etheostoma (Echelle et 1976; Wiseman et 1978), Fundulus notatus (Tatum et 1979) Gadus morhua (Sick 1961; M0ller 1966; Cross and Payne 1978; Jamieson 1967; Jamieson and Jones 1967; Jamieson and Turner 1978), Ictalurus (Lodge 1974), Lepomis (Avise and Smith 1977), Menidia (Johnson 1973, 1974, 1975), Merluccius productus (Utter 1969), Mollienesia (Hewitt et 1963), Marone saxatilis (Morgan et 1973), Moxostoma macrolepidotum (Buth 1979), Notropis (Menzel 1970), Oncorhynchus nerka (Grant et 1980), Pagrus .(Manooch et 1976), Plecoglossus altivelis (Nishida and Takahashi 1978), Pleuronecte s platessa (deligny 1966; Purdom et al. 1976), Salmo

PAGE 18

clarki (Loude n s l ager and Kitc hin 1979), Salmo gairdneri (Northcote et 1970; Utter and H odgins 1 972), Sa11'10 trutta 1976), Sa1velinus fontin alis (Wright and Atherton 1970; McGlade and MacCrimmon 1979), Sard inop s oce11ata (Thompson a n d Mostert 1974), S aurida (Taniguchi 1 9 69), Sebastes ( O'Rourke 1960; Sindermann 1961; Wishard et. 1980), Sgua1us acanthias (Sindermann and Mairs 1961), Sprattus sprattus (Wilkins and Il es 1966), Theragra cha 1 eogramma (Grant and Utter 1980), Thunnus and Euthynnus (Cushing 1956; Suzuk i et 1958, 1959; Fujino 1970) and Zoarces (Frydenberg e t 1973; Frydenberg and Simonsen 1973; Hjorth and Simonsen 1975; Christiansen et al. 1976). 6

PAGE 19

MATERIALS Juvenile Atlantic menhaden were collected by haul seine, cast net and s urface trawl in estuaries along the eastern coast of 7 the United States (Figure 1 ). Adult menhaden were captured by commercial purse seines, pound nets and gill nets. Dates and locations of collections, methods of capture, and associated environmental data are given in Table 1. Adult menhaden were examined on board ship for sex and gonadal maturity. Indl'viduals exhibtting an elevated stage of maturity (_gonad index > 4.0} were assumed to spawn in the vicinity of capture (Higham and Nicholson 1964). Individuals so designated were used in this study to represent the spawning stock of a geographical area. They were assumed to be the parental stock of juveniles collected in the same geographical areas a few months later. Blood samples were obtained from freshly captured fish by caudal puncture with a syringe containing heparinized 1.75 % saline solution. Carcasses and blood samples were stored on ice in the field; the carcasses were frozen upon return to shore.

PAGE 20

Figure 1. Collection sites for Atlantic menhaden.

PAGE 21

. \ \\ .. GA \ / i ' .... \ .. -----PENN N.Y. ME. r\ ---_., '. ...... ) \ 1 VT. ( ', i ) . @.JUVENILES liJ ADULTS 9 \

PAGE 22

Table 1 Collection data for each (a). Juvenile sampling location and (b). Adult sampling location. Sample size at each juvenile collection station was generally 50 fish. Biochemical data available for those stations with an asterick. Water Water Secchi t 1 ethod a. Year of Month of temp salinity reading of Location collection collection Latitude NORTH ATLANTIC oc 0/oo em capture Barnstable Harbor, MA 78 9 41.9' 18.2 31.5 >100 haul seine Childs River, MA 78 9 41' 18.0 18.1 70 haul seine Winnepaug Pond, RI 78 9 41' 19.2 28.1 >100 hau. l seine Peconic River, NY 78 9 40.1' 19.0 1.1 >100 haul seine MIDDLE ATLANTIC Delaware Bay Leipsic River, DE 78 9 391 21.8 5 .2 25 sfc trawl White Creek, DE 78 9 38.9' 22.7 27.0 50 haul seine Chesapeake Bay Chester River, MD 78 9 39.5' 21.6 5.2 39 haul seine Onancock Creek, VA 78 9 37' 25.6 17.2 82 cast net Felgate Creek, VA 78 9 37.8' 24.3 12.6 53 cast net SOUTH ATLANTIC Bath Creek, NC 79 10 35.2' 27.0 5.0 46 haul seine Hancock Creek, NC 79 10 34.2' 25.0 5.0 48 haul seine Calabash Creek, NC 79 9 33.6' 29.0 3.0 32 haul seine Cooper River, SC 79 9 32.5' 25.0 9.0 53 haul seine C6oper River, SC 78 9 32.5' 29.8 17.7 71 cast net Lazaretto Creek, GA 78 9 32.5' 28.4 18.5 36 __. cas t net 0

PAGE 23

Table 1 (can't) Number of b Year of Month of Samples Method of Location Collection Collection Female Male Capture NORTH ATLANTIC SPRING SPAWNERS Narrangassett Bay 80 4 18 14 pound net NORTH ATLANTIC FALL SPAWNERS Narraganssett Bay 78 9 0 2 purse seine Weymouth Channel Boston, MA 78 9 3 5 purse seine Casco Bay, ME 79 9 4 4 purse seine MI DOLE ATLANTIC FALL SPAWNERS Port Monmouth, NJ 78 10 13 3 pound net Chesapeake Bay mouth 78 11 5 2 purse seine Seaside Park, NJ 79 11 30 15 purse seine Manasquan Inlet, NJ 79 11 11 4 purse seine SOUTH ATLANTIC FALL AND WINTER SPAWNERS Drum Inlet, NC 79 1 19 1 purse seine Cape Lookout, NC 79 1 1 0 purse seine Beaufort Inlet, NC 79 1 3 0 purse seine Walden Creek, Southport, NC 79 2 1 1 gi 11 net South River, NC 79 2 2 gi 11 net Drum Inlet, NC 79 12 12 2 purse seine __, __,

PAGE 24

METHODS Meristics and Morphometries Fi:sh were weighed to the nearest gram and x-rayed in the laboratory. Kodak Industrex R double coated film was exposed for 9-10 min at 2 rna and 50 kv and 65 kv respectively for juveniles and adults. Fork lengths were measured on adult gonads were removed, weighed and used to assess species fecundity. Scales for aging were removed from above the lateral line on the left side of the fish at a point below the dorsal fin origin. All measurements and counts were taken from radiographs with dial calipers and a 6X magnifier. Except as indicated below, measurements and counts were made as prescribed by Hubbs and Lagler (1958) Osteological terms follow Robertson (1959}. Table 2 lists the measurements and counts taken in the present study. Head length measurement follows Hildebrand (1948). Head depth measurement follows Hildebrand (1948). counts include the terminal vertebra supporting the hypural plate. This is consistent with Hildebrand (1963), Christmas and Gunter (1960) and Reintjes (A field guide to the North Atlantic menhadens, genus Brevoortia, Unpublished manuscript) but differs from June (1958, 1965), Sutherland (1963) and Dahlberg (1970). Interneural spine counts include anterior and posterior dorsal fin stays. Interhaemal spine counts include posterior anal fin stay. 12

PAGE 25

Table 2 Meristic and morphometric variables of Atlantic m enhaden investigated in this study Meristics Total vertebrae Trunk vertebrae Caudal Vertebrae Ventral Scutes Interneural Spines Interhaemal Spines Morphometries Standard Length (SL) Fork Length (adults) (FL) Weight (WT) Gonad Weight (adults) (GWT) Body Depth (BD) Head Length (HL) Head Depth (HD) Orbit Diameter (OD) Caudal Peduncle Depth (CPO) Predorsal Length (POL) 13

PAGE 26

14 Biochemistr y Sera were separated from red blood cells by centrifugation and stored frozen until analysis. Red blood cells were washed at least once with 1.75 % saline, lysed with deionized water and stored Small amounts of serum were incubated overnight (2:2 v/v) with 0.6 % rivanol in 0.52M Tris Gly cine buffer, pH 8.9, to precipitate all but the B 1 -metal-combining-globulins (Boettcher et 1958; Jamieson and Turner 1978). Red blood cell lysate samples were refrigerate centrifuged at 20,000 g for 20 minutes to precipitate cellular debris. The supernatant was incubated at 8 C 1:4 with 0.52M Tris-Glycine, 1 % mer capthoethanol buffer solution and bubbled with carbon monoxide. Samples were used immediately for electrophoresis. Muscle samples were homo-. genized 1 :1 with deionized water and centrifuged. The supernatants were used for electrophoresis. Disc gel electrophoresis of serum and red blood cell proteins followed Ornstein (1964) and Davis (1964); sodium dodecyl sulfate gel electrophoresis used the methods of Weber and Osborne (1969). Regular disc-gels (7 1/2 % acrylamide) were stained with coomassie blue according to McFarland (1977) and stored in 14% methanol-? % glacial acetic acid. Muscle homogenates were electrophoresed on cellulose acetate (Gauldie and Smith 1978) and on starch (Smithies 1959). Stains for specific enzymes on these latter two media are described with the results for these enzymes. Nomenclature for allelic designations follows Allendorf and Utter (1979) Data Analysis Collection stations were analysed individually, by geo-

PAGE 27

graphic areas and by populations as designated b y June (1958, 1965) and Sutherland ( 1963). Geographic are as were designated as f o l l ows: North Atlantic encompasses samples taken on Long Island and nor thward. This area is further d i v ided into those individual s w h ich are spring spawners or their offspring and fall spawners. Offspring of the fall spawners were not co 11 ected. Middle Atlantic encompasses samples taken between southern Long Island and the mouth of the Chesapeake Bay. This area includes samples from both the Chesapeak e and Delaware Bays. South Atlantic contains samples from Pamlico Sound, North Carolina and southward Meristic and morphometric data were analyzed with SAS-79.3 (Helwig and Council 1979), BMDP (Dixon 1977), SPSS (Nie et 1975), and Fortran IV utilizing simple statistics and descriptive and predictive multivariate statistical techniques. Results of cluster analysis were used as groupings for subsequent discriminant analysis. Biochemical data from each station and each geographical area were tested with chi-square for Hardy-Weinberg equilibrium. Samples were pooled in each geographical area and independent t-tests were used to test gene frequencies of binomial samples between the geographical areas (Sokal and Rohlf 1969) 15

PAGE 28

16 RESULTS Adult samples from each geographi c area were assumed to be representative of the parental population of j uveniles collected in estuaries of those same areas. The criterion for an adult females selection was the state of maturity of her ovaries. Higham and Nicholson (1964} calculated an ovary index for Atlantic menhaden from the formula : 0. I. = where W weight of both ovaries in grams, and L = fork length of the fish in millimeters. Their plot of median ovum diameter against ovary index showed an O.I. of 4.0 to represent the level at which most ova were maturing. They based their conclusions of geographical and seasonal distri bution of spawning on the assumption 11that females with an index of 4 0 or more would have actively spawned somewhere i n the vicinity of capture11 Only females with an O.I. of 4.0 or greater were used for this study Means of ovary indices for females are given in Table 3 Males were also used in this study. Because spermatogenesis has not been investigated in Atlantic menhaden, definitive levels of maturation cannot be assigned based on macroscopic inspection of the testes. However, G w. Link (University of North Carolina Institute of Marine Sciences, personal corrmunication) found that the number of tailed spermatozoa in testes of individuals of the genus

PAGE 29

17 Centropris t i s increased a s the calculated gonad index increased. Hayashi (1971) also concluded that the seasonal change of gonad index reflected the histological maturation process in the testes of Lateolajaponicus. It was assumed that the same correlation occurs in Atlantic menhaden. The change in gonad weight with heightened levels of maturation would be much greater for females than males as females incorporate lipids into the yolk of the ova. Gonad indicies {G.I.) were calculated for male Atlantic menhaden from the same formula as used for the female ovary index Only males with a G.I of 4.0 or greater were used in this study. The G.I. value of 4.0 represents a conservative estimate and should only include individuals with maturing spermatozoa even if the rate of gonad weight increase was as rapid as a femaless. Table 4 gives the means of male gonad indices. Meristics and Morphometries Weight-length regression of all adult and juvenile samples showed the growth of Atlantic menhaden to be allometric. The equation of the calculated r egr ession line is: weight= 1.75 x 105 (length) 3.0824 Regression s were also conducted for each geographic area. These results are given in Table 5. Regression coefficients were tested for equality between area pairs. Coefficients of area regressions were significantly different from each other (t-test, p=O.OOOl). Historically lengths of Atlantic menhaden have been expressed as fork length although sys tematists generally u se standard length. Body measurements were e x pressed as a proportion of standard

PAGE 30

18 T able 3 M e a n ovar y indicus for adult Atlantic menhaden. Area North North fall Atlantic Atlantic Middle South and Sta ti sti ca 1 spring fall At1 antic Atlantic spring winter parameter spawners spawners spawners spa w ners s p awners spa w ners Mean O.I. 8.3467 6.7348 9 .0903 12.3195 8.3467 10.0702 s/{fl 1.1904 0.5857 6:42 8 0 ti :1 5 .28. 1. 1904 0. 4025 n 20 8 60 38 20 l 0 6 Table 4. Mean gonad indicus for male adult Atlantic menhaden. Area PoEulation North North fall Atlantic Atlantic Middle South and Stati sti ca 1 spring fall Atlantic Atlantic spring winter parameter spawners spawners spawners spawners Mean G I. 6 .9499 4.2936 8.5657 10.7640 6.9500 7.9473 s/{fi 1 5606 0.2816 0.6790 2.2910 1.5606 0.6775 n 14 11 24 9 14 44

PAGE 31

Table 5. Weight-length regressions of Atlantic m enhaden. Standard error is given below each estimate. Note that there were no juvenile offspring collected from North Atlantic fall spawners. Hence, there i s not a regression for this area. log weight = a + b(log length) e e All samples Area North Atlantic spring spawners and offspring Middle Atlantic South Atlantic PoQulations Spring spawners Fall and winter spawners a -10.9537 0.0098 -11.2400 0.1154 -11.0373 0.0532 -10.5790 0.0892 -11.2400 0.1154 -10.8839 0.0467 o 3.0824 0.0098 3. 1453 0.0256 3.1027 0.0114 2 .9947 0.0204 3.1453 0.0256 3.0672 0. 0103 df 882 206 317 338 206 675 19

PAGE 32

20 length in this In order to compare these proportions to those calculated by previous researchers (except Dahlberg 1970 who used standard length) it was necessary to determine the relationship between fork and standard lengths. Regression of the two length on each other yielded the equations: FL = 10.2020 + 1 0745(SL), r = 0 .975 and SL = 2.7931 + 0.8842(FL), r = 0 .975 Allometric growth proved a major problem when analyzing menhaden morphometries. The magnitude of the body measurements coul d be expressed as a linear function of standard length. BD = 2.7685 + 0.3452 SL r = 0.995 HL = 4.2048 + 0.2855 SL r = 0.996 HO = 5.0000 + 0.2500 SL r = 0.996 00 = 3.6247 + 0.0442 SL r = 0.989 CPO= 2.3062 + 0.0840 SL r = 0.993 POL= 1.2296 + 0 .5493 SL r = 0.999 As a fish grew longer the proportion of body depth, head length, head depth, orbit diameter, and caudal peduncle depth to standard length grew smaller. Only the proportion of predorsal length to standard length remained constant. Similar regressions of morphometries on standard length for different geographic areas and populations as designated by June (1958, 1965), Sutherland (1963) and Dahlberg (1970) showed significant differences in slopes for body depth, head depth, and predorsal len gth between North Atlantic spring spawners and offspring and the Middle Atlantic. Slopes for head depth and predorsal length were signifi-

PAGE 33

21 cantly different between the North Atlantic spring group and the South Atlantic. Only the slope for h ead depth was significally different between the Middle and South Atlantic. This indicates significant growth differences between fish from each population The growth of fish from the Middle and South Atlantic was nearly identical. Regression results are given in Tables 6 and 7; t-test results are in Tables 8 and 9. Means of morphometri es are summarized in Tables 1 0 a n d 11. Based on means of morphometric proportions, juveniles of the Middle and South Atlantic were similar. However, proportions of head length, head depth, orbit diameter, and predorsal length of North Atlantic juveniles were significantly smaller Vertebrae were counted on 709 juveniles and 316 adults. Counts ranged from 46 to 49 vertebrae; many anomalous vertebral columns were observed. Means of total vertebral counts for each juvenile collection are shown in Figure 2. There is a definite cline with fish from higher latitudes having fewer vertebrae. Sutherland (1963) showed similar results with a step in vertebral counts occurring between latitudes 401N and 401N. June (1958) also found a distinct decrease in vertebral number north of Long Island Numbers of vertebrae were tested for homogeneity by analysis of variance. Heterogeneity was indicated between stations (p=O.OOOl). Grouping samples at 401N and northward into a northern population and samples southward of 40 latitude into a southern population still showed significant heterogeneity bet w een stations in the southern population (p=O.OOOl). Division of samples into geographical areas eliminated heterogenity between stations within each area. T-tests for comparison of .means between each of the areas showed significant differences between all area pairs

PAGE 34

Table 6. Intercepts and slopes of morphmetric regressi_ons for geographical area$, Note that there are no offspring from North Atlanti c fall spawners and hence no regression for tpis group . morphometric b + m (SL) estimate/standard error of estimate b BO m HL HO b llJ b m b 00 m b CPO m b PDL m North Atlantic spring spawners 2.1054 0.3530 3.1622 0.2873 2.5254 0.2554 3.2165 0.0452 2.4837 0.0834 -0.4207 0.5571 and offspring 0.2639 0.0022 0.1801 0.0015 0.1555 0.0013 0.0449 0.0004 0.0852 0.0007 0.0249 0 .0021 Middle Atlantic 3.9042 0.3396 4 .6336 0.2846 5.9132 0.2457 3.8547 0.0432 2.3501 0.0839 1.9867 0 .5456 0.3151 0.0023 0.1914 0.0014 0.1511 0.0011 0.0436 0.0031 0.0759 0.0006 0.2282 0.0017 South Atlantic 2.1797 0.3466 4.3347 0.2870 5.0435 0.2517 3.6155 0.0453 2.0246 0.0866 0.1542 0.5482 0.1876 0.0018 0 .0016 0.1176 0.0011 0.0459 0.0005 0.0584 0.0006 0.1505 0.0015 Table 7. Intercepts and slopes of morphometric regressions for populations as designated by June (1958, 1965), Sutherland (1963) and Dahlberg (1970) morphometric b + m (SL) estimate/standard error of estimate b BO m b HL m b HO m b 00 m b CPO m b POL m Spring spawners and offspring 2 .1054 0.3530 3.1623 0.2873 3.5254 0.2554 3.2165 0.0452 2.4837 0.0834 -0.4207 0.5571 0 .2639 0.0022 0.1801 0.0015 0.1555 0.0013 0.0449 0.0004 0.0852 0.0007 0.2491 0.0021 fall and winter spawners and 2 .9459 0.3432 4.5561 0.2847 5.4786 0.2481 3.7606 0.0437 2.2487 0.0843 1.7475 0.54 6 9 offspring 0 .1756 0.0014 0.1221 0.0010 0 .0952 0.0008 0.0306 0.0024 0.0475 0.0004 0.1374 0.0011 N N

PAGE 35

23 l able 8. Probabilities of a significa nt t valu e by chance from independ ent tests of morphometric r egression coefficients between geographical areas. Prob > t under H0 : m1 = m2 area comparisons North Atlantic spawners and offspring vs. Middle Atlantic North Atlantic spring spawners and offspring vs. South Atlantic Middle Atlantic vs South Atlantic BO HL 0. 0001 ns ns ns ns ns variables HO 00 CPO POL 0. 0001 ns ns 0. 0001 0.0006 ns ns 0. 0001 0. 0001 ns ns ns Table 9. Probabilities of significant t values by chance from independant t-test of morphometric regression coefficients between populations as designated by June (1958, 1965), Sutherland (1963) and Oalhberg (1970). Prob > t under H0 : m1 = m2 morphometric regression on BO HL HO 00 CPO POL prob >t SL 0.0006 ns 0. 0001 ns ns 0. 0001

PAGE 36

Table 10. Means of morphometric characters of juvenile Atlantic menhaden a. By collection station b. By geograph i ca l area and c. By population. Station a Morphometric Barnstable Childs Winnepaug Peconic L e ipsi c White Chester Onacock Felgate Standard length Harbor R iver Pon d River R i ver Creek River Creek Creek X 63.9129 73.1152 73.9250 80.8020 80.4568 76.3720 87.2060 60.3469 78.3688 s/{71 0.9078 0 7 079 0.78 1 2 0 .9527 1. 3579 0. 9775 0.8131 0.3267 0.9982 n 3 1 46 48 50 37 50 50 49 48 Height X 6.2818 9.3435 9.6229 1 5.2660 14.7541 11.0840 17. 9160 4.8041 13.5771 s/yn 0 .2654 0. 3136 0 .3663 0.5818 0.8653 0 .4339 0. 4840 0. 0769. 0.5 851 n 33 46 48 50 37 50 50 49 48 BD/SL 0.3862 0 .3844 0.3615 0.3941 0. 4112 0.3862 0.3962 0.3711 0.3917 s/vn 0.0046 0.0022 0 .0030 0.0024 0.0025 0.0022 0 .0028 0.0015 0.0023 n 24 46 47 50 37 50 50 49 48 HL/SL 0.3492 0.3353 0 3311 0 .3154 0.3370 0.3487 0.3385 0.3673 0 .3385 sl{n 0.0033 0 .0030 0.0019 0. 0015 0.00 1 3 0.0020 0 .0035 0 .0015 0.0015 n 31 46 48 50 37 50 50 49 48 HD/SL x 0.3 109 0. 3135 0 .294 1 0 .2994 0 .3244 0.4339 0. 3172 0.3382 0. 3194 s/{71 0.0025 0 0017 0. 0021 0.0016 0.0017 0 .001 9 0 .001 6 0 .0013 0 .0015 n 3 1 46 48 50 37 50 50 49 48 00/SL x 0 .0912 0.0916 0.0876 0 .0869 0 .0890 0.0960 0.0890 0.1031 0 .0940 s!{n 0. 0010 0.000 7 0 .0005 0.0005 0.0006 0.0007 0.0006 0 .0005 0.0007 n 3 1 46 48 50 37 49 50 48 48 CPD/SL x 0.1170 0.. 119 1 0.1154 0.1173 o 1132 0.1158 0.1163 0. 1152 0.1145 sf rn 0.0014 0 .0008 0 .0012 0. 0011 0 .0007 0.0008 0.0007 0.0006 0.000 8 n 3 1 46 48 50 37 50 50 49 48 PDL/SL X 0.5643 0.5545 0.5448 0 .5496 0.5663 0.5700 0.5713 0. 5801 0.5712 S/ {n 0 .0054 0 .0022 0.0030 0.0026 0.0020 0 .0036 0.0025 0 0016 0.0 0 1 9 N +=n 29 46 48 50 37 50 50 50 48

PAGE 37

Table lOa. (cont'd} Station Morphometric Bath Hancock Calabash Cooper Cooper Lazaretto length Creek Creek Creek River 1978 River 1 979 Creek X 58.7300 65.7776 59.6780 70.6245 64.6960 95.6940 S/ Vn 0.3007 0.4482 1.1208 0.6458 0.4364 1 .0503 n 50 49 50 49 50 5 0 vJeight X 4.6020 6.7531 6 .236 0 8.7673 7.3120 20.8360 s/{n 0 .0757 0.1386 0 .3752 0.2507 0.1496 0.2507 n 50 49 50 49 50 50 80/SL 0.3657 o. 3621 0.3934 0 .3803 0.3387 0.3819 s/{ri 0 0017 0 .0016 0 .002 1 0 .0018 0.0016 0 .0028 n 50 49 50 49 50 50 HL/SL x 0 .3726 0.3665 0.3412 0.3409 0.3432 0 .3419 s; rn 0.0010 0.0013 0.0015 0.0013 0.0012 0.0017 n 50 49 50 49 50 50 00/SL 0 .1049 0 .1005 0 1 011 0 .0973 0.0998 0.0894 s/{n 0.0005 0.0004 0.0008 0 .0005 0.0007 0.0005 n 50 49 50 49 50 I 50 CPO/SL x 0.1138 0 .1135 0.1196 0.1182 0.1179 0.1152 s/yn 0.0006 0.0006 0.0007 0.0006 0.0006 0.0007 n 50 49 50 50 50 50 POL/S L -X 0.5809 0 .5780 0.5708 0. 5605 0.574 8 0.5629 s/{11 0.0017 0 .0015 0 .0023 0.0016 0.0018 0.00 2 4 n 50 49 50 49 50 5 0 HO/SL x 0.3394 0.3344 0,3297 0 .3 160 0.3257 0.3137 sJ{n 0.0012 0.0010 0 .0015 0.0015 0. 0011 0.0014 N
PAGE 38

Table 10. (contd) b. Area c. Population Morphometric N. Atl. spring Middle Atl. South Atl. fall and winter spawned spring spawned Standard length X 73.9034 76.3906 69.2067 72.3741 73.9034 s; rn 0.5985 0.7158 0.7863 0.5624 0.598 5 n 175 235 298 533 .175 10.5215 12.3183 9 .0933 10.5152 s/ rn 0.3221 0.3699 0.3431 0.2609 0.3221 n 177 235 298 533 177 BD/SL x 0.3807 0 .3904 0,3787 0.3838 0.3807 s/yn 0. 0018 0.0013 0. 0010 0.0009 0.0018 n 168 235 298 533 168 HL/SL x 0.3309 0.3464 0.3510 0.3490 0 .3309 s/{il 0.0013 0.0012 0.0009 0.0008 0.0013 n 175 235 298 533 175 HD/SL x 0 .3037 0.3242 0.3265 0.3265 0 .3037 s/ rn 0.0011 0.0009 0 .0008 0 .0006 0. 0011 n 175 235 298 533 175 00/SL x 0 .0891 0.0944 0,0988 0.0969 0 .089 1 s/{n. 0 .0003 0.0004 0.0004 0.0003 0.000 3 n 175 233 298 531 175 CPD/SL x 0.1172 0.1151 0.1164 0.115 8 0 .1172 s/{n 0.0006 0 .0003 0.0003 0.0002 0.000 6 n 175 235 298 533 175 POL/SL x 0.5520 0.5720 0.5713 0 .5716 0 .5520 s/{ri 0.0016 0 0011 0 .0009 0.0007 0.001 6 n 173 235 298 533 173 N 0\

PAGE 39

Table 11. Morphorretric Fork !ength X s/{n. n Standard length sf {Tl n Weight X S/ {fl n BD/SL X s/ rn n X s/{0 n Means of morphometric characters for adult Atlantic menhaden Area Population North Atlantic North Atlantic Middle South Spring Fall spring fall Atlantic Atlantic spawners spawners spawners spawners 276.4412 276.3684 264.5000 246.5957 276.4412 260.3933 2.8190 3.1699 1 .8943 3. 3643 2 .8 190 l. 7460 34 19 84 47 34 150 247.6565 250.7595 236.1928 223.3415 247.656 5 234.4 4 7 6 2.3931 2.8218 l. 8067 3.1793 2. 3931 1. 5987 32 19 83 41 32 143 439.8233 441.3158 374.0353 286.4375 439.8233 354.7830 12.0714 13.3966 8.4222 12.5789 12.0714 7.5916 34 19 85 48 34 152 0.3615 0.3519 0.3558 0.3546 0.3615 0.3549 0.0021 0.0024 0.0023 0.0024 0.0021 0.0015 32 19 83 41 32 143 0.3007 0.3000 0 .3048 0. 3060 0.3007 0.3045 0.0017 0.0024 0.0014 0.0024 0 0017 0. 0011 32 19 83 41 32 143 N '-J

PAGE 40

Table 11 (cont'd) Area Morphome-North Atlantic North Atlantic tric spring spawners fall spawners HD/SL X 0.2695 0 .2691 s/{n 0.0013 0.0024 n 32 19 OD/SL x 0.0581 0 .0579 s/{n 0. 0003 0.0005 n 32 19 CPD/SL X 0.0932 0.0904 s/yn 0.0007 0.0010 n 32 19 PD!JSL X 0.5552 0.5554 s/{fi 0.0023 0.0044 n 32 19 Middle South Atlanttc Atlantic 0.2710 0.2736 0.0012 0.0016 84 41 0.0596 0. 0611 0.0003 0,0006 83 41 0.0936 0.0949 0.0005 0.0006 83 41 0.5546 0,5556 0,0017 0.0021 83 41 Population Spring spawner s 0.2695 0.0013 32 0. 0581 0.0003 32 0 0 932 0.0 007 32 0.555 2 0.0023 32 Fall spawners 0.2715 0 .00 0 9 143 0.0598 0.00 0 3 143 0.09 3 6 0.00 0 4 143 0.5550 0.0013 143 N (X)

PAGE 41

Figure 2 Mean number total vertebrae of juvenile Atlantic menhaden samples plotted again s t latitude of collection station. One s tandard error is given about each mean.

PAGE 42

' 1\f 101 0 <.0 1'-..q-30 w 0 ::::> 11...J

PAGE 43

31 (pNorthM iddle=O.OO Ol, PNor th-South=O.OO Ol; P M iddl e-South=0.0002 ) Tests of means of the two previou s ly desi g nated populations were also significantly different (p=O.OOOl). Analys i s of variance of adult s amples did not indicate heterogeneity. Based on vertebral counts June (1965) concluded that sprin g spawne r s in water s n orth of Long I s l and constituted a single homogeneous population which was different from nonspawners in that a rea in tha t t ime peri od. H e found t h e s e northern spring nonspawners, northern fall spawners and autumn and winter spawners in the South Atlantic constituted another homogeneous group Means of juvenile and adult samples from each geographical area and population are given in Tables 12 and 13. Vertebral counts were further divided into trunk and caudal vertebrae. Counts ranged from 15 to 20 and 28 to 32, respectively, for trunk and caudal vertebrae. Figures 3 and 4 show means of these meristics for juveniles plotted against latitude. The clinal structure is present only in the diagram for trunk vertebrae. Standard errors about caudal vertebrae means were large. Analysis of variance of all juvenile caudal vertebral counts was not significant. The same test for trunk vertebrae was significant between stations (p=O.OOOl). This indic ates that the clinal variability shown in total vertebral counts was due to change s in the number of trunk vertebrae. Division into north vs south population s did not eliminate the between station heterogeneity in the southern population (p=0.0026). Further analysis for homogeneity in e a c h g e o g raphi c al are a was not sign ificant. T-te sts of trunk vertebrae means showe d significant differences between the North and Middle Atlantic comparison (p=0.0002) and the North and

PAGE 44

32 Table 12. Juvenile Atlantic m enhaden meristics summarized by areas and populations Area Population Meristic N. Atlantic Middle South Spring Fa 11 and Hinter Spring Atlantic Atlantic Spawned Spawned Vertebrae x 47.6836 47.9277 48.1040 47.6836 48.0262 s/yn 0.0412 0.0330 0.0322 0.0412 0.0234 n 177 235 298 177 533 Truok Vertebrae X 17.4598 17.7949 17.9799 17.4598 17.8985 s/vn 0.0458 0 .0374 0 .0322 0.0458 0.0247 n 174 234 298 1 7 4 532 Caudal Vertebrae :\ 30.2184 30. 1282 30.1242 30.2184 30.1259 S/Vn 0. 0521 0.0420 0 .0395 0 .0521 0.0288 n 174 234 298 174 532 Scutes x 31.9253 32.6340 32.7215 31.9253 32.6330 0.0815 0.0619 0.0592 0.0815 0.0429 n 174 235 298 174 533 Dorsal Ray Sueports X 20.8741 20.8142 21.1027 20.8741 20.9592 0.0701 0.0507 0.0460 0. 0701 0.0349 n 143 183 185 143 368 Anal Ray Sypports X 21.1429 21.8295 21.5282 21 1429 21.6587 0.0821 0.0614 0.054 7 0.0821 0 .0414 n 161 217 284 161 501

PAGE 45

Table 13. Mean of meristic counts for adult Atlantic menhaden Area Meristic North Atlantic North AtlMtic Middle South spring spawners fall spawners Atlantic Atlantic Vertebrae X 47.9375 47.7895 48.0000 48.0976 s/yn 0.0998 0.1447 0.0521 0.0911 n 32 19 82 41 Trunk_ Vertebrae X 18.0000 18.0000 17.8780 17.7805 s / -.Jn 0.1100 0 1325 0.0532 0.0742 n 32 1 9 82 41 Caud al Vertebrae x 29.9375 29.7895 30.1220 30.3171 s/yn 0.1094 0. 1636 0.0658 0.0952 n 32 1 9 82 41 Scutes X 33.1290 32.4211 32.4634 33.2308 s/vn 0. 1655 0.2572 0.1379 0.1621 n 3 1 1 9 82 39 Dorsa 1 Ray Sup(2orts X 20.9355 21.0526 2 1 .0366 21.1250 s/yn 0.0920 0.1617 0.0636 0.1025 n 31 19 82 40 Populat ion Spring spawners 47.9375 0.0998 32 18.0000 0.1100 32 29.9375 0. 1094 32 33.1290 0.1655 3 1 20.9355 0 .0920 31 Fall spa11mers 48.0000 0.0047 142 17.8662 0.0415 142 30.1338 0.0531 142 32.6714 0.1026 140 21.0638 0.0 515 141 w w

PAGE 46

Table 13 (cont'd) Area Meristic North Atlantic North Atlantic spring spawners fall spawners Anal ray X 21.5625 22.1053 s/yn 0.1954 0.2008 n 32 19 Age X 4.1250 3.7368 s/Vn 0.1665 0.1499 n 32 19 Middle South Atlantic Atlantic 21.6341 21.5854 0 ,1091 0.1258 82 41 3.2471 2.6596 0.0747 0.0926 85 47 Population Spring Fa 11 spa\'lners spawners 21.5625 21 6831 0.1954 0.0783 32 142 4 1250 3.1258 0 .1665 0. 0611 32 151 w +>-

PAGE 47

Figure 3. Mean number trunk vertebrae of juvenile Atlantic menhaden samples plotted against latitude of collection station. One standard error is given about each mean.

PAGE 48

@ II 3VC:l831H3/\ >1Nn8 1 0 'o:;:t 36 w 0 :::> tt-<( ...J

PAGE 49

Figure 4. Mean number caudal vertebrae of juvenile Atlantic menhaden samples plotted against latitude of collection station. One standard error is given about each mean.

PAGE 50

L 0 -q. 0 C't) I 0 ('I') 0 C't) I 0 C\1 0 C't) I 0 C't) lVOn\IQ NV3V'J -,_ ----0 C\1 C't) 0 co C't) 0 c:o C't) 0 0 3 8 w 0 ::> t:: 1-..J

PAGE 51

South A tlantic comparison (p=O.OOOl). There was no significant difference between the means of Middle and South Atlantic areas. In contrast to the juveniles, adult Atlantic menhaden e 'xhi bi ted heterogeneit y in' cauda 1 vertebra 1 counts but not in trunk vertebrae (p=0.0083). The southern population was not homogeneous with respect to caudal vertebrae (p=0.0029). Further separation by area eliminated all heterogeneity. Independent t-tests found no significant differences between means of all possible area and population comparisons. Ventral scutes were the most variable of the meristics studied with counts ranging from 29 to 36 scutes. Heterogeneity occurred in juvenile counts between stations (p=O.OOOl), between stations within populations (Pnorthern=0 0317, Psouthern=0.0005) and between stations within the North Atlantic (p = 0.0317) and within the South Atlantic areas (p=0.0004). Significant heterogeneity in the 39 North Atlantic was due to the low mean count of the Barnstable Harbor, Massachusetts station. Means of northern latitude stations were significantly lower than Middle Atlantic (p=O.OOOl) and South Atlantic areas (p=O.OOOl). Means of scutes between the Middle and South Atlantic stations were not significantly different. Means of each station are plotted against latitude in Figure 5 June (1958) calculated a mean number of ventral scutes for his stations north of Long Island as 31.58 scutes. The mean of stations Long Island and southward was 32.78. These values compare favorably with means calculated for this study of 31.92 and 32.68 scutes for the northern and southern areas, respectively. Adults, too, exhibited heterogeneity of variances between areas within the s outhern population (p=O.OOOB). Separation into geo-

PAGE 52

Figure 5. Mean number of ventral scutes of juvenile Atlantic menhaden plotted against latitude of collection station. One stand ard error is given about each mean.

PAGE 53

0 0 C\J (") l'v'cUN3/\ 0 LO ,..... 41 w 0 :::> II
PAGE 54

graphica l areas eliminate d the heterogeneity. Interneural spines (dorsal ray supports) of 172 adult and 662 juvenile individuals ranged from 18 to 23 spines. This meristic was the most difficult. to count accurately . Radiographs of smaller fish were the most illegible and consequently many individuals were not used in the analysis of this character. Unfortunately, most of those eliminated were North Atlantic juvenile samples. There is a 42 1:1 relationship between interneural spines and dorsal rays which facilitates comparison of these results with those of June (1958). Means of each juvenile collection station are illustrated in Figure 6. There appears to be no relationship between numbers of interneural spines and latitude. Analysis of variance showed significant hetero geneity between the stations (p=O.OOOl}, between stations in each population (pNorth=0 .0003, Psouth=O.OOOl) and between stations within the North Atlantic area (p=0.0003). Means were significantly different between the Middle and South Atlantic areas (p=O.OO. Ol). June (1958) calculated the means of the northern and southern populations, respectively, as 20.91 and 21.77 dorsal rays Means for adults from this study for the same populations were 20.94 and 21.06, respectively. Homogeneity within areas and within populations was demonstrated by analysis of variance on adult samples. Juveniles in the more northern latitudes had fewer interhaemal spines (Figure 7). Counts ranged between 18 and 24 spines. Tests for.homogeneity of variances were significant between stations, (p=O.OOOl) between stations in each of the two populations (pNorth=O.OOOl, Psouth=O.OOOl) and between stations in the North Atlantic area (p=O.OOOl). The source of significance in the North

PAGE 55

Figure 6. Mean number of interneural spines of juvenile Atlantic menhaden plotted against latitude of collection site. One standard error is given about each mean.

PAGE 56

(/) l l I l.U z ij I CL (/) _J <"'.( I c r l l l :::> I w z n : U 1-z 20.50 z < U J 20.30 20.101j 42 40 38 36 34 32 LATITUDE

PAGE 57

Figure 7. Mean number of interhaemal spines of juvenile Atlantic menhaden samples plotteed again s t latitude of collection station. One standard error is given about each mean.

PAGE 58

I (/) 21 .90 l1J z 0.. (/) ...J <{ 2 lU <{ I 21.30 cc j 1 j w 1z z < w ::?! 42 40 I I I 38 36 LATITUDE l 34 I 32 .p. CJ)

PAGE 59

47 Table 14 Means of Atlantic menhade n meristics. North Atlantic spring fall and winter spawners and offspring spawners and of-Fspring Vertebrae x 47.7225 48.0207 s/-vn 0. 0377 0.0207 n 209 675 Trunk_ Vertebrae X 17.5437 17.8917 s/.y-rt 0.0444 0.0214 n 206 674 Caudal Vertebrae X 30.1748 30.1276 s/vn 0.0476 0.0253 n 206 673 Scutes X 32.1073 32.6805 s/yn 0.0794 0. 0401 n 205 673 Dorsal Ray Support X 20.8851 20.9882 S/vfn 0.0598 0.0290 n 174 509 Anal Bay Supports X 21.2124 21.6641 s/yn 0.0764 0.0365 n 193 643

PAGE 60

48 Atlantic was due to the low count s for individuals from Winnepaug Pond, Rhode Island M eans w e r e signifi c antl y differe n t betw ee n the North and S outh Atlantic are a s (p= O.OO O l). Adult s amples showed n o significant heter o g e n eity or diffe r ences b e tween means for th e area s -and Biologists often use a scaling variable to standardize body measurements. Atchle y et. (1976) presented re sults of empirical analyses ''indicating large and systema tic change s in both the s t ructure and the underlying distributions of data when ratio proportion s were compounded bet\'Jeen continuous vari a b l e s They warned again s t using proportions as ra w variables for multivariate analyse s suggesting the use of residuals of a covariance analysis as input variables. Traditionally, arcsine transformations of data have been used to rectify the leptokurtosis (Sakal and Rohlf 1969}. However, this type of transformation does not correct for skewness introduced by scaling variables (Sakal and Rohlf 1969; Atchley et 1976). Hills (1978) noted that the statistical difficulties associated with ratios arise because the ratio is not a linear function of the numerator and denominator. He pointed out that the log transformation of the ratio is a linear function of the log of the numerator and log of the denominator. This type of transformation rectifies kurtosis and decreases skewness. Atchley et. noted an increased spuriou s correlation between variables which increa sed as a function of the d e n o minator coefficient of variation. S purious correlation is i mportant a s multivariate statistics as principle component, factor and di s crimninant anal ys e s are sensitive to this They did conce de, however, that mos t morpholog ical data i s positively correlated not z ero correlated, the extreme which they e xpound ed. In length

PAGE 61

49 is often the scaling vector used to standardize body measurements. Unf ortunately, this measurement u s u ally has the hig hest coefficient of variation. Since the intent was to use the data in mulitivariate analyses the effect of ratio formation on the structure of the data had to be investigated. Figure 8 illustrates the effect of ratios and their transformation on the distributions. Clearly the logarithmic transformation of proportion s changed the data base the least. Hence, morphometric input variables for all multivariate statistics were log transformed proportions unless noted. Flexible sorting strategy ( B =-0.25), an agglomerative polythetic method, was used for cluster analysis to group stations. The distance measure was an Euclidean metric; data were standarized t o normal deviates. It was necessary to analyze juveniles and adults separately due to allometric growth. Results of cluster analysis are shown in Figure 9. The first division is between the North Atlantic spring spawned juveniles and the fall and winter spawned young-of-the year Further separation of the so uthern group did not necessarily group stations into their respective geographical areas. Quite unexpecte dly North Atlantic spring spawners were grouped with South Atlantic spawners which peak in the winter months. Factor analysis (Dixon 1977) was performed on the standardized meristic and morphometric variables. Figure 10 shows the plot of the first two canonical variables which account for 58.5 % of the variance. These two variables are described as: First Canonical Variable= 103.65305-0.75413(TVERT) 0.31808 (CVERT) 76.38715(ln(HD/SL)) 46.68433(ln(OD/SL)) + 45.87560 (ln(CPD/SL)) -17.39236(ln(PDL/SL)) 0.34944(SCUTE)

PAGE 62

Figure 8 Frequency distributions of the morphometric body depth and transformations of its proportion to standard length for Barnstable Harbor.

PAGE 63

ooso vSL"O 51 _J en ....... c co Q) C) 0 _J ij z en (.) 0: <( _J en ....... c co c co

PAGE 64

Figure 9. Cluster dendrogram of juvenile and adult Atlantic menhaden.

PAGE 65

DISS I M ILARITY LEVEL Lazaretto I I CreeK -+---------......., Felgate ere e k -+--D-e-1-. _B_a_y_, White t 1-Creek L!_ Leipsic River River C ooper River-+-----. 1979 L_ 1978 Creek Creek Bath Creek Creek Peconic River -+--------, Childs -+---------" River Pond Barnstable Harbor I I I I I Fall and Winter Spawned North Atlantic Spring Spawned North Atlantic_ Fall Spawners 1-I c... c: < CD :::::s CD en Middle Atlantic_ )> Spawners a. c North Atlantic :::::+ Spring Spawners[___ CJ) South Spawners 53

PAGE 66

Figure 10. First and second canonical variables for juvenile Atlanti c menhaden. Solid circles represent scores for fall and winter spawned juveniles; the open circle is their mean. Filled triangles represent scores for spring spawned juveniles; the solid square is their mean. Open triangles represent the coincidence of scores of two fish from the different populations.

PAGE 67

l{) l{) 'V 'V /\ av ov Gs vG g a o r-g L-v G-G s-I I I I I I I I I IV .-I --, I I I I I I I I 9 L 8-I I 1 o o s- I 'f' 'f' 'f' 'f' 'f' 'f' 'f' I I w I "' : I 'f' 'f' 'f' I .... 'f' 'f' 'f''f''f' "' I ,.,.,. 'f' 'f' 'f' "' "' "' \ftf "' I ... ............. .. ............... W "fT 'Y 'Y Y ' .... w . 1 .., .... YY I ..... .... . .. j(;"2- -4'J L -... ... ... .... "' . . . . . .. . . . . .. '7----------W--..,_,_T--tl.., .... -"f' ---"f' ..... -"f'... .-..r--....-.-.()w TIT.-...,. .-.-T-r"r---r----... ... .... ... .. ..... .... . 0 'V ... ... WY ... ..... ... I 'f' 'f' W 'f' 'f' 'f' 'f' 'f' W I e T 'Y y 'f' ... ...... ... ... .. .. ....... .... ... v 'f' 'f' 'f' .-.., ..,. I '1 'f'...... ....... ....... 'Y 'Y 'Y I. 'f' I 'f' 'f' 'f' I "' "' I 'f' 'f' 'f' I 'V I I I I I I I 'f' L )9" L o o s 0 . 0 ..,.. .... l) r < ).;. :u ;. .. (.'1 I -rn 1\.)

PAGE 68

and Second C anon i c al V ariable = -36.56388 + 0 18238(TVERT) t 0. 10588(CVERT) -104.62449(ln(HD/SL)) t 55.63527{1n(OD/SL)) t 5.28719(ln(CPD /S L)) + 95.9 5 204(ln(PDL/SL))-O.l2068(SCUTE). 56 Discrimina n t anal ysis is a mult ivariate clas si f icat ion t ool which just recently i s being used in fisheries population studies {_Mes sieh 1 9 7 5 ; Manooch 1976 ; R akocinski 1980; Wilk et 1980). Generally, the analysis weights and linearily combines discriminating variables in such a way a s to maximize statistical differences between groups. Covariance matrices are used t o calculate the discriminating functions. Chi-square tests for homogeneity determine whethe r it is the within-group or pooled within-group matrices used. Discriminant s cores for all cases within a group on all functions are averaged withir the group. These means are referred to as group centroids. The pooled within-group covariance matrix and the group centroids are used to derive classification equations for each group. Classification scores for an unknown case are calculated according to each groups classifica tion equation. The unknown case is assigned to the group for which it has the highest score. Results of cluster analysis for juveniles (North Atlantic vs. South and Middle Atlantic) were used as groupings for discriminant analyse s In order to maintain a large sample size of North Atlantic juveniles the variables dor s al and anal ray supports were not used. A chi-square test for homogeneity between the with i n-group matrices was significant (p=O.OOOl). Despite significant differences between within-group covariance matrices, the analysis was conducted with pooled within-group matrices. The analysis is in

PAGE 69

57 fact robust and the consequence of pooling significantly different matrices is an increased likelihood of classifying a case into the group with the greatest overall dispersion. Pooled within-group co variance matrices are nece s sary to calculate predictative classification equations. Therefore,the consequences were accepted Stepwise discriminant analysis determined that the following variables, in decreasing order of discriminating ability, were important in maximizing the distance between the two groups: ln(HD/SL), ln(CPD/SL), ln(OD/SL), SCUTE, TVERT, CVERT, and ln(PDL/SL). Variable coefficients for classification equations are: constant ln(HD/SL) ln(CPD/SL) 1 n ( 00/SL) SCUTE TVERT CVERT ln(PDL/SL) spring spawned -12338.98 5605.62 -2256.06 631.33 15.88 183.46 171 91 14594.44 fall and winter spawned -12625.56 5818.32 -2383.80 761.32 16.85 185.58 172.80 14642.87 The above variables accurately discriminated 89.7 % of the spring spawned juveniles and 93.0 % of the fall spawned juveniles. A jacknifed classification did not increase the accuracy of the analysis. Interestingly, using the within-group covariance matrices for the analysis produced an increased classification accuracy of the fall and winter spawned juveniles (94.5 % ) and a decrease in the accuracy of classification of the spring spawned individuals (86.7 %). Should the original analysis have 11felt11 the significant differences between the withingroup covariance matrices which were pooled for the first run, an increase in classification accuracy of the spring spawned juveniles should have occurred in the second analysis. Therefore, pooling the

PAGE 70

58 within-group matrices had no detrimental effect on the accuracy of the analysis. The same variables were used to discriminate adults but could not accurately discern between the designated groups. Biochemistry Hemoglobin of Atlantic menhaden is composed of two fractions (Fig. 11). The faster fraction, the major of the two, exhibits a root effect (Saffran and Gibson 1978). Pokrywka and Gold (1980) applied menhaden hemoglobin to a DEAE-50 Sephadex column and resolved four components. The first component corresponds to a cathodal protein not resolved on disc-gels except as a hemoglobin component moving cathodally into the upper buffer chamber. The second component corresponds to the minor fraction observed by Sindermann and Honey {1963), Saffran and Gibson and in this study. Both the third and fourth components comprised the major fraction accounting for 77% of the hemoglobin. The data from this project and that of Sindermann and Honey indicate monomorphism for both observed fractions. However, in preliminary screening of fish hemoglobins by Dr. J. B. Sullivan (Duke University Marine Laboratory, personal communication), one individual B. tyrannus illustrated a polymorphism in the major fraction (Figure 11). Sodium dodecyl sulfate gel electrophosesis (SDS) indicated a denatured molecular weight of 15,600 for the globin chains. In addition to the hemoglobin fractions, the red blood cell lysate often contained another fraction (Figure 11). The excess of this protein allowing its visualization with normal hemoglobin isolation and electrophoretic techniques may be due to a gene duplication at the encoding locus or to a mutant regulatory gene. A likely

PAGE 71

Figure 11. Electropherograms of red blood cell lystates of Atlantic menhaden. a. normal red blood cell lysate pattern showing normal hemoglobin pattern and additional fraction. b. variate hemoglobin pattern observed only once by Dr. Sullivan. c. most frequent red blood cell lysate pattern.

PAGE 72

RBC LYSATE CA 60

PAGE 73

61 function of this protein was carbonic anhydrase (EC 4.2.1 .1}. However, specific staining for carbonic anhydrase (CA) following the methods of Shaw and Prasad (1970) and Drescher (1978) failed to verify this identification. Both methods_ did give positive results for a standard of bovfne CA. SDS gel electrophoresis results indicate a molecular weight of 26,800. Carbonic anhydrase is an enzyme physiologically important in the reversible hydration of co2 It has the highest molar activity of any known enzyme (Lehninger 1975). Mammals exhibit two kinds of CA but only one form exists in lower organisms (Dobzhansky et 1977) The molecular weights of both mammalian forms and of bacterial CA are close to 30,000 (Carter 1972). Maynard and Coleman (1971) reported elasmo branch CA as 36,000-40,000 daltons. The identity of this red blood cell lysate protein remains uncertain. Data from individuals collected in Bath Creek, North Carolina, and from juveniles screened from the Beaufort, North Carolina locality, indicate the occurrence of this fraction in 30% of the samples. This same frequency was observed for individuals captured in Winnepaug Pond, Rhode Island. Because of the equality of frequencies over such a large geographic distance juveniles were not screened for this protein. The frequency of this fraction in adult samples was much higher It occurred in 50% of North Atlantic fall spawners, 33.5 % of Middle Atlantic spawners and 45% of South Atlantic spawners. There were no blood samples taken for North Atlantic spring spawners due to the difficulty in locating fresh samples with an elevated maturity. Frequency between sexes within the areas were similar e x cept in the Middle Atlantic where females showed the fraction 39. 6 % of the time and only 19. 0 % of the males had it.

PAGE 74

62 Transferrin is a serum protein for iron transport. It has a monomeric structure and is encoded at one locus. Three codomi nant alleles were segregating at the transferrin locus of Atlantic menhaden (Figure 12). The most frequent allele, Tf(lOO) was intermediate in mobility to the other alleles. Tf(90) occurred only twice in 611 individuals and because of this low frequenc y it was not u sed in the statistical analyses. Transferrin phenotypes and gene frequencies are given in Table 15 and 16, respectively, for each juvenile sampling location and geographic area. Mean frequencies of the Tf( llO) allele of each juvenile collection location are plotted against latitude in Figure 13. Regression of the frequency of Tf(llO) on latitude gave a slight correlation (r=0.4274). Tests for Hardy-Weinberg equilibium at the transferrin locus of juvenile samples showed that random mating and equal survival of genotypes assumptions were not violated (Table 17). Chi-square tests for each each geographical area and all samples pooled were insignificant at the 95% confidence level. Stations were pooled into geogra phic areas and gene frequencies were tested for equality of variances (Sakal and Rohlf 1969). The variances were significantly different in all comparison between areas (Table 18). Because of the variance inequalities, independent t-tests of gene frequencies between geogra phic areas used weighted (by sample size) variances (Sakal and Rohlf 1969). Table 19 contains the results of the t-tests. Gene frequencies for Tf(llO) were not significantly different between the South and Middle Atlantic areas. However, tests comparing the North Atlantic with each of the other areas were significantly different indicating division of the gene pool.

PAGE 75

Figure 12. Electropherogram of transferrin phenotypes of Atlantic menhaden.

PAGE 76

TRANSFERR.IN 10 0 .. 1 1 0 -+64

PAGE 77

65 Table 15. Distribution of transferrin phenotypes of juvenile Atlantic menhaden. Phenotyp es frequency Location Tf(90,100) Tf(lOO,lOO) Tf(lOO,llO) Tf(llO,llO) North Atlantic Barnstable Harbor 30 2 Child s River 42 5 1 Winnepaug Pond 42 8 Peconic River 42 7 Summary 156 22 1 M1' ddle Atlantic Leipsic River 32 8 White Creek 39 10 Chester River 38 9 Onancock Creek 43 6 1 Felgate Creek 38 10 1 Sumnary 190 43 2 South Atlantic Bath Creek 1 38 10 1 Hancock Creek 40 9 Calabash Creek 1 38 9 2 Cooper Ri ver 1979 42 6 Sumnary 2 158 34 3

PAGE 78

66 Table 16 Transferrin gen e frequencies of juvenile Atlantic menhaden. Location Tf(90) Tf(lOO) Tf(llO) North Atlantic Barnstable Harbor 62 2 Childs River 89 7 Winnepaug Pond 92 8 Peconic River 91 7 Summary 334 24 (0.9330) (0. 0670) Middle Atlantic Le1'ps i c River 72 8 White Creek 88 10 Chester River 85 9 Onancock Creek 92 8 Felgate Creek 86 12 Summary 423 47 (0.9000) (0.1000) South Atlantic Bath Creek 1 87 12 Hancock Creek 89 9 Calabash Creek 1 86 13 Cooper River 1979 90 6 Summary 2 352 40 (0. 0051) (0.8934) ( 0.1015)

PAGE 79

Figure 13. Frequency of the Tf(llO) allele in juvenile Atlantic menhaden samples plotted against latitude of collection station.

PAGE 80

,... ,... . <.0 0 C\1 0 0 C\1 C') w c ::J ...,_ ...,_ <( ...J 68

PAGE 81

69 Table 17. Hardy-Weinberg tests of t ra n s ferrin phenotypes from (a). North Atlant i c juveniles (b) Middle ATlantic juveniles l c)_. South Atlanti.c juveniles and (d). All juvenile samples a. Phenotype Tf ( 1 00,11 0 ) if(.100, 100) Tf(llO,llO) obse rved 156 23 x 2 = o.oo2 x2 3 .84 155.82 23.18 = exp ected 05,1 b. Phenotype Tf( lOO,lOO) Tf(100,110} Tf(1 1 0,110) observed 190 43 2 x 2 = o .640 x 2 = 5 .99 exp ected 190.35 42.3 2.35 05,2 c. Phenotype ( 90,1 00) Tf(lOO,lOO) Tf(100,110) Tf(ll0,110) observed 158 34 5 x 2 = o.449 x2 = 5 .99 expected 157 .24 3 5.73 3 .82 .05,2 d. Phenotype Tf(90,100) Tf(l 00,100) Tf(1 00,110 ) Tf( 1 10,110) obse rved 504 99 8 x 2 = o. 22. 1 2 expected 503 100. 7 4 6.86 X .05.,2 = 5 .99

PAGE 82

70 Table 18 (a). Variances and (b). Tests for equality of variances among geogra phi cal areas for the transferrin alleles of juvenile Atlantic menhaden. 2 F 5, I s -where the variance of sample l is greater than the variance s 2 2 of sample 2. F.025(oo,oo) 1.000 a. b. area North Atlantic Middle Atlantic South Atlantic comparison test South Atlantic vs. Middle Atlantic South Atlantic vs. North Atlantic Middle Atlantic vs. North Atlantic p < 0.05 .063 .090 .095 1.013* 1. 460 1.440 n 358 470 394

PAGE 83

Table 19. Independent t-tests of Tf(llO) frequencies between geographical areas. t .OS,oo;:: 1.645 71 2 where fi, Si and ni are Tf(llO) frequency, variance and samples size of ith area compari son South Atlantic vs Middle Atlantic South Atlantic vs North Atlantic Middle Atlantic vs North Atlantic South Atlantic and Middle Atlantic vs North Atlantic p < 0.05 t 0.073 l. 700 1.683 1.853

PAGE 84

Casagrande and Pike (1978) giye an approxi mate formula for calculating a sample size, n, necessary for comparison of two binomial populations. n = A [ 1 1 + 4 ( P 1 + P 2) ] 2 + 4 ( p 1 A A = [Z1 Y2p q a 72 Alpha and beta were set at 0.05 and 0. 10, respectively. Using the Tf(llO) frequency of the North Atlantic juveniles as p 1 and the Tf(llO) frequency of the pooled South and Middle Atlantic juvenile samples as p 2 the number of alleles, n, necessary to statistically compare two populations with given allele frequencies is 1221 or 611 individual fish. Purely coincidental, the sample size of juvenile Atlantic menhaden assayed for the Tf allele was 611 individuals. Therefore, the sample size was large enough to statistically compare the two populations. Transferrin gene and phenotype frequencies of adult samples are given in Table 20. Hardy-Weinberg tests for each area and all adult samples combined were not significant. Adult and juvenile samples were pooled (Table 21). Again all tests for non-random mating and unequal survival of genotypes were insignificant. Combination of spawners with the offspring did not alter the significant results of t-tests of gene frequencies between areas. Initially, muscle samples of fish from the Beaufort locality were screened for polymorphic loci which could be of possible

PAGE 85

use in this study should muscle tissue be utilized. Table 22 gives those results. Of 22 enzymes assayed, only esterase, glucose-6phosphate dehydrogenase, glycerophosphate dehydrogenase, hexokinase, and phosphoglucomutase were polymorphic and could be scored. One enzyme (_isocitrate dehydrogenase) showed no activity despite attempts \'lith various buffers and stains and very fresh tissue homogenates. Superoxide dismutase was visualized as negative staining on gels stained for glutamate dehy drogenase. 73

PAGE 86

74 Table 20. Tranferrin (a). G ene and (b). H a rd y-Weinbe r g tests on tranferrin phenot pe frequencies of adult Atlantic menhaden. Allele a. Tf(l 00) Tf(ll 0) North Atlantic 27 3 fall spawners 0. 9000 0.1000 Middle Atlantic 145 17 spawners 0.8951 0.1049 South Atlantic 5 4 10 spawners 0.8438 0.1562 Phenotype b. Tf(lOO, 100) Tf(lOO,llO) Tf(ll 0,11 0) North Atlantic 12 3 0 observed x 2 =o.o975 fall spawners 12.15 2.7 15 expected Middle Atlantic 66 13 2 observed x 2 =o.o940 spawners 64.90 15.21 0.8913 expected South Atlantic 22 10 0 observed x 2 =o.o927 spawners 22.78 8.44 0.78 expected

PAGE 87

Table 21. Transferrin (a.) Phenotype and (b.} Gene frequencies for adult and juvenile Atlantic menhaden. Tf(90,100) a. Tf(lOO,lOO) Tf(lOO,llO} Tf(JlO?llO} b. Tf(90) Tf(lOO) Tf(llO) North Atlantic 156 22 1 334 24 spring spanwed 155.82 22.32 0.86 2 0.9330 0.0670 juveniles x, ;:; 0.0020 North Atlantic 12 3 0 27 3 fa 11 spawners 12.15 2.7 0.15 2 0.9000 0.1000 and offspring xl ;:: 0.0940 Middle Atlantic 256 56 4 568 64 spawners and 255.22 57.54 3.24 ;:> 0 .2203 0.8987 0.1013 offspring South Atlantic 180 44 2+3 2 406 50 spawners and 181 55 44.70 2.91 2 0.0044 0.8904 0 1096 offspring x2 F 1.5176 Summary 605 126 2+8 2 1338 142 604.00 128.20 8.61 .. 0.2644 0.0013 0.9028 0.0958 "-J (.11

PAGE 88

Table 22 Results of enzyme specific stains on electrophoresed muscle homogenates of Atlantic .menhaden. Results are monomorphic (M), scorable (S), or polymorphic unscorable (U). The number of fish upon which these cone usions are based are given in parenthesis. inferred inferred stain cellulose starch polyacrylamide quaternary number reference acetate gels disc-gels structure of loci AAT, asparatate u u aminotransferase 2 {26) ( 16) ADH, alcohol M M dimer dehydrogenase l (33) ( 1 0) AKP, alkaline M phosphatase l (8) AK, adenlyate M M ... dimer 2 kinase 1 (26} (3} CPK, phospho-u u creatine kinase 1 {2} (3} ES, s s esterase 3 (25} ( 16) Monomer 2 FK, u fructokinase 4 (4) FUM, M M fumarase 1 (22) (3) tetramer 2 G6PDH, glucose-6-s s phosphase dehydrogenase 1 (4) (3) dimer 1 '-I (J)

PAGE 89

Table 22 (cont'd} inferred inferred stain cellulose starch polyacrrlamide quaternary number reference acetate gels disc-ge s structure of loci GDH, glutamate M M dehydrogenase 1 ( 19} (3) dimer 2 GPDH, glycero-s s phosphate dehydrogenase 1 ( 16} ( 3) dimer 2 HK, s hexokinase 4 (A} dimer 1 IDH, isocitrate 1 3, no no dehydrogenase 6,7 activHy activfty LDH, lactate M M M tetramer 2 dehydrogenase 8 (251 (7) ( 1 0) MDH, malate u u u dehydrogenase 3 (21) (7) ( 12) ME, malic M M enzyme 3 ( 19} (1} MK, M mannokinase 4 (4} 1 PGI, phosphogluco-M M -dimer 3 isomerase 5 (18} (24) PGM, phosephogluco-s s no -.....! mutase 1 2 1 activity monomer 3 -.....!

PAGE 90

Table 22 (contd) stain cellulose reference acetate SOH, sorbitol M dehydrogenase 1 (22) SOD, superoxi de u dismutase 1 (7} XOH, xanthine M dehydrogenase 1 ( 15) Stain References 1 Shaw and Prasad (1970) 2 Johnson, Utter, and Niggol (1972} 3 Turner (1979) 4 Jelnes (1971) 5 Martin (1979) 6 Chapman (1977) 7 Henderson (1965) 8 Markert and Faulhaber (1965) starch po 1 yacrrl ami de gels disc-ge s M M (8) (2) s (.7) M (8) inferred quaternary structure tetramer tetramer inferred numbe r of loci 2 2 ""-! co

PAGE 91

79 DISCUSSION On the basis of meri st1cs and gene frequencies of variants at the Tf locus, juvenile menhaden from the Mfddle and South Atlantic areas were similar to adults of the s e respective areas. There were no bfochemical analyses conducted on the North Atlantic spring spawners as these fish, from Pt. Judith pound net collections, were received frozen. Meristics of these fish were not comparable with those of juveniles collected in the North Atlantic or with data on North Atlantic spring spawning adults presented by June (1965). Gonad indices for the spring spawners, though than 4.0, were not as high as those of individuals collected in more southern areas It is likely that these northern individuals with maturing gonads would spawn in the North Atlantic but not until one to two months after their capture. The timing of their spawn would coincide with the late summer and fall peak in the North Atlantic. If this should be the case, pooling juveniles of the North Atlantic with these spring captured maturing adults is unjustified. Spawning of Atlantic menhaden occurs somewhere along the eastern seaboard every month of the year (Chapoton, R B. On the distribution of Altantic menhaden eggs, larvae and adults. Unpublished manuscript). In a compilation of 60 cruise reports, Judy and Lewis (Distribution of eggs and larvae of Atlantic menhaden, Brevoortia tyrannus, along the Atlantic coast of the United States. Unpublished

PAGE 92

80 manuscript) pointed out that menhaden eggs have been taken at some location every month but March. Although appeqring to spawn throughout the year, Atlantic menhaden reach spawning peaks inshore in the North Atlantic in spring, offshore in the Middle Atlantic in spring, near shore in fall in the North Atlantic, and offshore in fall and winter in the Middle and South Atlantic Bights (Higham and Nicholson 1964). Kendall and Reintjes (1975}, based on Dolphin reports of egg and larval collections, concluded that menhaden spawn during both the northward and southward migrations in addition to spawning while overwintering south of Cape Hatteras. These results confirmed the conclusions of Higham and Nicholson who inferred spawning times and locations from the state of gonadal maturity. Nelson et (1977) reiterated the Dolphin results and noted that the average size of larvae increased in an onshore direction and spawning distance from shore increased southward. Spawn ing in the North Atlantic appears to be inshore and probably somewhat inside the bays and sounds (Herman 1963; Massmann 1974; Ferraro l980a). Ferraro (1980b, c, personal communication) discovered much spawning activity in Peconic Bay, NY in early May 1972-1974. As May progressed and bay waters warmed, spawning activity moved eastward, farther towards the mouth. He found very little spawning in little Peconic Bay in late June, 1973 and none occurred in the Peconic Bay system by early July. He attributed the spawning migration within the Bay to spawners' pre ference for cooler waters. B. tyrannus eggs were found in temperatures of 12.lC to 25.0C. The effect of temperature on the meristics and morphometries of Atlantic menhaden is unknown. Environmental factors during early development, especially temperature, have been shown to affect the

PAGE 93

81 number of metameric parts of fishes (Vladykov 1934; Taning 1952; Lindsey 1954; Tatarko 1968; Fowler 1970). June (1958) addressed the problem, noting that there w as a l3C difference between the early summer temperatures north of Long I sland and mean w inter temperatures off the Carolinas, November to March; North Atlantic temperatures were warme r This difference was less than the range of temperatures November to March i n the Carolinas. These data were furnished by the U.S. Coastal and Geodetic Survey and were inshore temperatures. Hence, water temperatures offshore in the South and Middle Atlantic areas, where most of the spawning was occurring, were much warmer than the winter cooled inshore waters. Herman (1963) found eggs in Narragansett Bay in waters 12.0 to 24.4C May through August and in October. He found larvae June through July and October to February in temperatures 1.2 to 22.2C. Temperatures in June 1966 off Delaware Bay, where larvae were captured by the Dolphin, were 15 to l9 C and salinities varied from 30.3 to 32.1 /oo (Kendall and Reintjes 1975). They caught no larvae farther north in the previous 10 days of sampling. Temperatures in areas of capture off Long Island in August were 18 to 20C. In October, when they found larvae abundant from Martha's Vineyard to Currituck Beach, NC, water temperatures were about 3C cooler off northern New J ersey th a n they were in the August collection. Larvae collected on the Dolphin cruises were found in waters 0 to 25C and in salinities of 29 to 36 /oo. -The effect of salinity on the m e r istics of Atlantic m enhaden has not bee n investig ated Fis h spawned in near s hore waters and inside estuaries and sounds may well be subjected to different salinity regimes during early development Ocean spawned juveniles of the South

PAGE 94

82 Atlantic would be subjected to a more constant salinity regime. Hettler (1976) hqs shown both temperature and salinity to affect the oxygen consumption rates and the growth rates of juvenile Atlantic menhaden. It is possible that either could affect the meristics and mor phometries of this species. Spawning in the South Atlantic is protracted throughout the winter. Most larvae in the South Atlantic Bight were captured December to March, with most taken in December. Despite a high sampling effort, none have been taken south of Cape Hatteras May to October (Judy and Lewis, Unpublished Manuscript). Recruits in Hancock Creek, a tributary of Pamilico Sound, NC, show a trimodal length frequency distribution (D. W. Ahrenholz, personal communication). These modes can be followed as the fish grow in the nursery areas. The central mode predominates and very few fish fall into length classes between the modes. Dr. Ahrenholz indicates that recruitment into the fishery is different for each of these modal groups. Fish taken from Hancock Creek for this study were from the shortest length frequency mode. McHugh et (1959) found indications of a bimodal distribution of juveniles in the Chesapeake Bay. They attributed the modes to the two spawning peaks in the Chesapeake region, one in the fall and one in the spring. It is not known whether fish from each of the modes are homogeneous with respect to meristics, morphometries or biochemical characters. It has been hypothesized that spawning during the northward spring migration is a continuation of the massive spawn south of Cape Hatteras and that spawning during the southward migration is precursory to the southern winter spawning activity (Nicholson 1972). There has

PAGE 95

been no indtcation of a pause in spawning before initiation of the northward migration through near cessation of spawning occurs during midsummer i n the North Atlantic prior to the southward migration Should the phenomena of South estuarine juvenne populations with multiple length frequency modes prove to exist outside Hancock Creek much indirect evidence could be shed on the subject of phasic spawning June (1965) found vertebral counts of adult North Atlantic spring spawners to be significantly lower than those of non-spring spawning North Atlantic adults and fall and winter spawning adults. These collections spanned three years; conclus ions based on each year separately were the same, documenting the temporal stability of the 83 difference. Most adult tag returns of the 1967, 1968, and 1969 North Atlantic, summer-released fish were in the North Atlantic (Nicholson 1978t Costen 1971). However, this was to be expected as the larger fish migrate further north. The fecundity and age of maturity need to be determined for the North Atlantic spring spawners and fall spawners separately. Because the differences observed by June (1965) were temporally stable, and juvenile offspring of the groups are homogeneous with respect to their parental population (June 1965) some mechanism must be isolating the two groups. The majority of spring spawners must return to the North Atlantic area to reproduce; this 11homing11 mechanism must be conferred to their offspring. Those offspring of spring spawners possibly reproducing earlier and farther south would be mixed with the winter spawned juveniles and be obscured by the larger numbers of the latter group. The key to this problem would be through examination of length frequencies of estuarine juveniles as these spring

PAGE 96

spawned juveniles would comprise a modal group shorter than the fall and winter spawned fish. 84 Nicholson (1978} reported on tag returns from juvenile Atlantic menhaden; thepoorest return percentages were from North Atlantic releases. From September through October, 1969-1973, 28,022 juveniles were tagged in the New York Area. Five were recovered the initial year of release (4 in the South Atlantic and 1 in the Chesapeake Bay), 126 were recovered one year after release (91 in the Chesapeake Bay, 35 in the South Atlantic), 77 in the third year after release (6 in the Middle Atlantic, 61 in the Chesapeake Bay, 8 in the South Atlantic); 11 recoveries after the third year of release were in the North and Middle Atlantic areas. Only 0.78 % of the tags were recovered. This pattern of increased northward movement with age was the same that he observed for fish tagged in more southern latitudes. Kroger et (1971) reported on tag returns of 1969 releases (1,020) from Winnepaug Pond, RI. These tagged fish did not enter the fall fishery the year of release, nor were they vunerable to the fishery the follow ing spring and summer. Based on three tag returns from inland waters, which are far too few, the authors concluded that the juveniles migrated southward and "integration of northern grown juveniles with those produced off the Mid and South Atlantic states" occurred. Should those fish enter an estuary after their southward migration, they would represent a length frequency mode of large fish with low meristic counts. Thus, there is no evidence, beside similarity of meristics between spawners and offspring, that juveniles spawned in the spring in the North Atlantic, upon reaching maturity, return to spawn in the North

PAGE 97

85 Atlantic in the spring or conversely, there is no evidence against it. Juvenile abundance surveys conducted by the National Marine Fisheries Service indicate a drastic decline in the relative abundance of juveniles in the North Atlantic (Ahrenholz, D. W. 1979. Evaluation of Atlantic and Gulf menhaden juvenile abundance surveys. Unpublished manuscript). Figure 14 illustrates this decline relative to abundance in other areas. The ordinate is the mean, log (x+l) transformed, catch e per tow in each area, which is actually the natural logarithm of the geometric mean catch per tow plus one. The North Atlantic is represented by 32 tows in 8 streams each year A datum point for the other areas represents 169 tows in 31 streams in the Middle Atlantic, Chesapeake Bay and South Atlantic. Surveys were conducted once a year at the same time in each area each South and Middle Atlantic surveys were conducted as close to July 1 as possible. This date represents the peak abundance of juveniles in the estuaries. Due to gear selectivity against small juveniles, North Atlantic streams were not surveyed until September 15. The exponential decline in abundance of North Atlantic juveniles can be attributed to: 1) 2) 3) decreased population size of larger, older fish and thus, fewer fish in the North Atlantic which would be spawning in that area, or higher vunerability of North Atlantic spawned individuals to the fishery, or a sampling artifact. Although a decreased population size of spawners seems plausible, a .clear spawner-recruit relationship has not beem demonstrated for Atlanti c menhaden (Schaaf and Huntsman 1972; Nelson et 1977; Loesch et 1979). Nelson et (1977) attributed a large portion

PAGE 98

Figure 14. Relative abundance of North Atlantic juvenile menhaden (solid line) and juvenile Atlantic menhaden from all other areas combined (broken line). Figure from Ahrenholz, D. W. 1979. Evaluation of Atlantic and Gulf menhaden juvenile abundance surveys (Unpublished manuscript).

PAGE 99

I I I l \ \ \ \ \ \ \ \ I I I "/ / / / / < ' MOl NV3V'J aE)Q1 8 7 co 1'-(j) 'r-1'-1'-(j) 'r-<0 1'-(j) 'r-lO 1'-0) 'r-'r-(\') 1'-(j) 'r-C\1 1'-0') or-

PAGE 100

88 of the recruitment variability to larval survival as a function of Ekman transport. June and Ntcholson (1964) further proposed another stock. From length-age data and comparisons with more northern areas of yea"r class strengths, recruitment patterns, mean lengths and weights at age, and total annual mortality rates they speculated that a southern population, upon which the summer South Atlantic fishery was dependent, existed south of Cape Hatteras The movement of fish out of the South Atlantic fishery, prior to the arrival of southward migrating fish from farther north, coincided with the arrival of large numbers of menhaden of the same length in waters off Fernandina Beach, Florida They hypothesized that the older fish moved from Florida to North Carolina in the spring and returned in the fall to Florida. The mechanism which maintained this stock in reproductive isolation was not identified. Dahlberg (1970} provided further evidence for a South Atlantic population. He found significant differences between North Carolina and Chesapeake Bay samples in three proportions: pectoral fin length, anal fin base length, and caudal peduncle depth North Carolina to Fernan dina Beach samples were uniform. Dahlberg does not give data on the lengths of the fish used in his analysis. Because proportions in general do not remain constant as standard length increases, differences in the lengths of fish from the two respective areas could have been the source of the significant.differences. Judging from his values in Tab 1 e 24, for cauda 1 pedunc 1 e depth of 9 1 % and 9. 5 % for the Chesapeake Bay and North Carolina, respectively, his samples were_adults. Dahlberg also defined a distinct race of Atlantic menhaden in Indian River, FL. This population had fewer vertebrae, smaller

PAGE 101

89 head depth; head length, pectoral fin length, upper jaw length, and mandible length, and greater body depth anal fin length than Fernan dina Beach samples. He attributed these features in part to introgressive hybridization with smithi. Again, significant differences in proportions could have been due to differences in lengths of the fish used in the analysis. The fact that this population has fewer vertebrae is interesting. Data from this study and that of June (1958) and Sutherland (1963) show vertebral number increasing southward Dahlberg (1970) found four clinal characters tyrannus (body depth, dorsal fin base length, caudal peduncle depth, and pelvic fin shape}. I found no clinal nature in either caudal peduncle depth or body depth; the other two characters were not used in this study He found head lengths, head depths, maxillary and mandible lengths to be homogeneous (excepting the Indian River population). In constrast, I found head lengths, head depths, orbit diameters, and predorsal lengths to be significantly smaller northward and counts for vertebrae, trunk vertebrae, ventral scutes, and anal ray supports to be lower northward. Hildebrand (1948) noted that several proportions increased to the south (head depth and head, maxillary, mandible, pectoral fin, and caudal fin lengths). Clines may often reflect a developmental response to environ mental factors. North-south clines are usually associated with tern-perature. A species response to linear changes in an environmental factor influencing its growth and survival may be linear, but more often is curvilinear (Westman 1980). All characters of juvenile Atlantic menhaden which I found to exhibit a clinal structure were significantly smaller to the north. If indeed, water temperatures at

PAGE 102

90 the time of spring spawning activity and subsequent larval development fn the North Atlantic were lower than winter temperatures in the South Atlantic, then a direct relation to temperature is implied. However, as discussed before, the salinity regime to which North Atlantic spring spawned individuals were subjected could be different than that of South Atlantic individuals which generally were spawned further offshore The effect of genetic differences on the meristic and morphometric responses of individuals to environmental factors is unknown. Dahlberg (1970) cited genetic differences between t y rannus and patronus as the reason that meristic dfferences between the two species do not coverge in the southern fringes of their ranges where temperature regimes are similar. patronus is the Gulf congener of tyrannus. It is significant to note that the frequency of variants at the Tf locus also exhibited a clinal structure wi"th a step between the North and Middle Atlantic areas. Endler (1977) discusses geogra phic variation, speciation, and clines in great detail. Clines may be set up by hybridization, random genetic drift, selection, or random sampling error. The sharpness of geographic differentiation is increased by stronger selection and depressed by more frequent gene flow. A step may form in one generation by chance but will quickly be reduced or disappear in the next few generations unless it is selectively advanta geous. Simulations have shown that in a panmictic population with random mating, spatial differentiation can occur (Rohlf and Schnell 1971; Endler 1977). Changes in the fi.rst few generations determine the major features of the spatial distribution of gene frequencies. Endler concluded that "single-locus, one-dimensional cline models suggest that differentiation into two areas of high and low gene frequenciis

PAGE 103

can evolve smooth environmental gradients in spite of continuous gene flow, provided that gene flow distance is not large". More data 1s needed on the temporal and spatial stability of the transferrin gene frequencies. That natural selection is acting to maintain gene frequency differences in tyrannus between geographical areas cannot be documented here. Such docum entation does exist for transferrin in another species. Hershberger (1970) found that brook trout heterozygous for transferrin could bind iron faster and release it quicker than homozygous individuals, one homozygous condition was lethal. Heterosis maintained the lethal allele in the population. Further evidence for selection on alleles at a single locus is given by Koehn (1969, 1970) for Catostoma clarkii of the Colorado River system. The frequency of the ES-Ia allele decreased with increasing latitude. The role of selection was implicated. Physio chemical investigations showed the activity of allelic enzymes to vary at a given temperature. Selection was occurring for the allele with the highest activity in a given temperature environment. 91

PAGE 104

92 CONCLUSIONS Atlantic menhaden are not a panmictic population. Juveniles spawned in the North Atlantic 1 n the spring have significantly lower numbers of vertebrae, trunk vertebrae, ventral scutes, anal ray sup ports; shorter, shallower heads, smaller eyes, shorter predorsal lengths and lower frequencies of the Tf(llO} allele than fall and winter spawned juveniles of inore southern areas. Middle and South Atlantic juveniles were uniform with respect to all characters except vertebrae and dorsal ray supports. The finding of lower numbers of meristics in the north are consistent with the data presented by June (1958) and Sutherland (1965); morphometric variables exhibiting a clinal nature were not observed to do so by Dahlberg (1970). Discriminant analysis separated juveniles of the two populations with an accuracy of 89.7 % for the northern population and 93.0 % for the fall and winter spawned individuals. The following variables, in decresaing order of impor tance, are useful in discriminating between the two populations: lnHD/SL, lnCPD/SL, ln 0 ;sL, SCUTE, TVERT, CVERT, and lnPDL/SL. Data do not indicate a third population existing south of Cape Hatteras as proposed by June and Nicholson (1964) and Dahlberg (1970}. No data were taken from Indian River, Florida to substantiate Dahlbergs (1970) finding of another population in that estuary. Spawners and their offspring were homogeneous for all areas except North Atlantic spring spawning group. Maturing adults

PAGE 105

captured in this area in the spring were probably representative of the late summe. r and fall spawning group of that area, not of the spring spawning population. 93

PAGE 106

RECOMMENDATIONS I The effect of temperature and salinity on meristtcs and morpho metries needs to be investigated. Knowledge of the effect of these environmental variab les can help determine the extent to which meristic and morphometric differences are genetically controlled II. North Atlantic spring spawners and their offspring should be intensively tagged to determine whether relative reproductive isolation is maintained over several generations. Tagging can also give information on growth rates of the population. III. Determine population specific natural mortality, growth and fishing mortality rates; fecundity and age of maturity needs to be invest i gated for each population. IV. Menhaden eggs and larvae need to be sampled on a large scale along latitudinal gradients inland, inshore, and offshore. v. Multiple length frequency modes of estuarine juveniles should be investigated biochemically and morphologically. VI. Determine the frequency of transferrin alleles in the North Atlantic and Middle Atlantic spring spawners. VII. Through breeding e x periments document Mendelian heritability of alleles at the loci for hemoglobin, carbonic anhydrase (?) and transferrin. 94

PAGE 107

VIII. Document the temporal and spatial stability of gene frequencies. Stabtlity of meristics and morphometries is reaffirmed in this study by comparison with previous research. IX. For each population, determine the frequency of 'variants at ES, G6PDH, GPDH, HK and PGM loci. Investigate physiochemical properties of those alleles Cincluding transferrin} showing significant frequency differences between populations. 95

PAGE 108

96 LIST OF REFERENCES Allendorf, F. W and F. M Utter. 1979. Population Genetics In W S. Hoar and D. J. Randall, ed. 454. Fish Physiology Vol. VIII, 407Allendorf, F. W N Ryman, A. Stennek, and G. . 1976. Genetic variation in Scandi navian brown trout (Salmo trutta L ) evidence of distri.ct sympatr i c p opulations. Heredltas 83:73 -82. Atchley, W. R., C. T. Gaskins a n d D. Anderson. 1976. Statistical properties of ratios. I. Empirical results. Syst. Zool. 25:137-148. Avise, J. C. and. M. H Smith. 1977. Gene frequency comparisons between sunfish (Centrarchidae) populations at various stages of evolutionary divergence. Syst. Zool. 26(3):319 335. Boettcher, E. W., P. Kistler and Hs. Nitschmann. 1958. Method of isolating the B 1 metal-combining globulin from human blood plasma. Nature (land. ) 8 181 ( 4607): 490-491 Buth, D. G. 1979. Creatine kinase variability in Moxostoma macrolepidotum (Cyprinofoimes: Catostomi dae). Copeia (1979) (1) : 152-154. Carter, M J. 1972. Carbonic anhydrase: isozymes, properties, distribution, and functional significance. Biol. Rev. Camb. Philos. Soc. 47:465 513. Casagrande, J. T. and M. C. Pike 1978. An improved approximate fonnula for calculating sample sizes for comparing two binomial distributions. Biometrics 34:483-486. Chapman, R. W 1977. Plasma proteins and systematics of the genus Centro ristis (Pisces:Serranidae) M. S. thesis, University of West F orida. Christiansen, F. B., 0 Frydenberg, J. P. Hjorth and V. Simonsen. 1976. Genetic s of Zoarces populations. I X Geographic variation at. the three phosphoglucomutase loci. Hereditas 83:245-256. Christmas, J. Y. and G. Gunter. 1960. Distribution of menhaden, genus Brevoortia in the Gulf of Mexico. Trans. Am. Fish. Soc. 89(4}:33 8 -343.

PAGE 109

Clayto n, J. W and W. G Franzin. 1 970. Genetics of multipl e lactate dehydrogenase isozymes in m u sc l e tissue o f Lake Whitefi s h (Coregonus clup eafo r m i s) J. Fish. Res. B oard Can. 27;1115-1121. Costen, L. C. 1971. Summary of tag s released and recovered for the Atlantjc menhaden, 1966-69. Department of Commerce Data Report 66. 117 pp. Cross, T F. and R. H. Payne. 1978. Geographic variation in Atlantic cod, Gadus morhua, off eastern North America: A biochemical systematics approach J. Fish Res. Board Can. 35:117-123. Cushing, J E. 1956. Observations on sero logy of tuna. U.S. Fish. Wild1. Serv. Spec. Sci. Rep. Fish. 1 8 3, 14 pp. Dahlberg, M. D 1970. Atlantic and Gulf of Mexico menhadens, genus Brevoortia (Pisces:Clupediae). Bull. Fl. State Mus. Biol. Sci. 1513), l63 pp. 97 Davis, B. J. 1964. Disc electrophoresis II. Method and application to human serum proteins Ann. N.Y. Acad. Sci 121(2):404-427. de Ligny, W 1966. Polymorphism of se run transferrin in plaice. Xe. Cong. Eur. Groupes. Sang. Polymorphisme Biochem. Anim., Paris. 5-8 July, p. 373-378. de Ligny, W. 1969. Serological and biochemical studies on fish populations. Oceanogr. Mar. Biol. Annu. Rev. 7:411-513. de Ligny, W. 1971. Special meeting on the biochemical and serological identifi cation of fish stocks, Dublin 1969. Rap. P.-V. Reun. C ons. Int. Explor. Mer 161, 179 pp. Dixon, W J., ed. 1977. Biomedical Computer Programs P-Sereis. University of California Press, Berkeley, 880 pp. Dobzhansky, T., F. J. Ayala, G. L. Stebbins and J. W. Valentine. 1977. Evolution. W. H. Freeman and Co., San Francisco, 572 pp. Drescher, D. G. 1978 Purification of blood carbonic anhydrases a n d specific detection of carbonic anhy drase isoenzymes on pol y acryla m ide gel s wit h 5-dimethylaminonaphthalene-1sulfonamide (DNSA). Anal y Biochem. 90:349 -358. Dryfoos, R. L., R P. C h eek and R L. K ro ger. 1973. Prelimin a r y analy ses of Atlantic menhaden, Brevoortia tyrannus, migratio ns, population structure, s urvi val and e xplotation rates and availability as indicated b y tag returns. U .S. Nat. Mar. Fish. Serv. Fish. Bull. 7 1: 7 1 9-734.

PAGE 110

Echelle, A. A., A. F E chelle and B. A. Taber. 1976. Bioche mical evidence for conqeneric competition as a f actor restricting gene flow between populations of a darter (Percidae;Etheosto ma). Syst. Zool. 25(3 ):228-235. Endler, J. A. 1977. Mechanisms o f Speciation. Geographic variation, Speci ation and Clines. Princeton Univ Pre ss Prin ceton, N.J. 248 pp. 98 Ferraro, S. P 1980a. Embryonic development of Atlantic menhaden, Brevoortia t yrannus and a fish embryo age estimation method. U .S. Nat. Mar. Fish. Serv. Fish Bull. 77(4):943-949. Ferraro, S. P. 1980b. Pelagic fish eggs and larvae of the Peconic B a ys New York: 1972-1974. Ph.D dissertation, State Univers ity of New York at Stony Brook. Ferraro, S. P. 1980c. Daily time of spawning of 12 fishes in the Peconic Bays, New York. U.S. Nat. Mar. Fish. Serv. Fish Bull. 78(2):455-464 Fowler, J. A 1970. Control of vertebral number in teleosts -an embryological problem. Quart. Rev. Biol. 45:148-167. Franzin, W. G. and J. W. Clayton. 1977. A biochemical genetic study of zoogegraphy of Lake Whitefish clupeaformis) in western Canada. J. Fish. Res. Board Can. 3 (5):617 -62 5. Frydenberg, 0 and V Simonsen. 1973. Genetics of Zoarces populations V Amount of protein polymorphism and degree of gen1c heterozygosity. Hereditas 75:221-232. Frydenberg, 0., A 0. Gyldenholm, J. P. Hjorth and V. Simonsen. 1973. Genetics of Zoarces populations. III. Geographic variations in the esterase polymorphism Est III. Hereditas 73:223-238. Fujino, K. 1970. Immunological and biochemical genetics of tunas Trans. Am. Fish Soc., 99(1):152-178. Gauldie, R. W. and P J. Smith. 1978. The adaptation of cellulose acetate electrophoresis to fish enzymes. Camp. Biochem. Physiol. B Camp. Biochem. 61:421-425. Grant, W. S., G. B Milner, P. Krasnowski, and F. M. Utter. 1980. Use of biochemical genetjc variants for identificat ion of sockeye salmon (Oncorhynchus nerka) stocks in Cook Inlet, Alaska. C an. J Fish. Aquat. Sci. 37: 1236-1247. Grant, W. S and F. M. Utter. 1980. Biochemical genetic variation in walleye pollock, Jheragra chaleogramma: population structure in the southeastern Berin g S ea and the Gulf of Alaska. Can. J Fish. Aquat. Sci. 37:1093-1100.

PAGE 111

Hayashi, I. 1971. On the process of testicular maturation of the Japane se sea bass, Lateolabra x japonicus, Jpn. J. Ichthyol. 18(1): 39-50. 99 Healey, J. A. and M. F. Mulcahy. 1980. A biochemical genetic analysis of population s of the northern pike, Esox lu c ius L., from Europe and North Ameri ca. J, Fish. Biol. 17:3T?-324. Helwig, J. T. and K. A. Council, ed. 1979. SAS User's Guide 1979 Edition. SAS Institute Inc., Raleigh, NC, 495 pp. Henderson, N. S. 1965. Isozymes of isocitrate dehydrogenase: Subunit structure and intracellular location. J. Exp. Zool. 158:263-274 Henry, K. A. 1971. Atlantic menhaden (Brevoortia tyrannus) resource and fishery-analysis of decline. U.S. Dep. Cammer. Nat. Mar. Fish. Serv. Spec. Sci Rep. Fish. 32 pp. Herman, S. S. 1963. Planktonic fish eggs and larvae of Narragansett Bay. Limnol. Oceanogr. 8(1): 103-109. Hershberger, W. K. 1970. Some physiochemical properties of transferrins in brook trout. Trans. Am. Fish. Soc. 99(1):207-218. Hettler, W. F. 1976. Influence of temperature and salinity on routine metabolic rate and growth of young Atlantic menhaden. J. Fish. Biol. 8:55-65. Hewitt, R. E L. W. Ward, E. A. Halshrins and C. P. Haskins. 1963. Electrophoretic analysis of muscle proteins in several groups of poeciliid fis hes especially the genus Mollienesia. Copeia (1963) (2}:269-303. Higham, J. R. and W. R. Nicholson. 1964. Sexual maturation and spawning of Atlantic menhaden. U.S. Nat. Mar. Fish. Serv. Fish. Bull. 63(2):255-271. Hildebrand, S F. 1948. A review of the American menhaden, genus Brevoortia, with a description of a new species. Smithson. Misc. Collect. 107(18), 39 pp. Hildebrand, S. F. 1963. Genus Brevoortia. In Fishes of the Western North Atlantic. Memoir 1, Part 3. Sears-Foundation Mar. Res., Yale University, New Haven, Conn., pp, 342-380. Hills, M. 1978. On ratios-a response to Atchley, Gaskins and Anderson. Syst. Zool. 27(1):61-62. Hjorth, J. P. and V. Simonsen. 1975. Genetics of Zoarces populations. VIII. Geographic variation common to the polymorphic loci Hb I and Est III. Hereditas 8 1:1 731 8 4.

PAGE 112

Hubbs, C. L. and K. F. Lagler. 1958 Fishes of the Great Lakes Region. University of Michigan Press, Ann Arbor, 213 pp. 100 Jamieson, A. 1967. New genotypes in cod at Greenland. Nature (Lond.) 215:661-662. Jamieson, A. and B. W. Jones. 1967. Two races of cod at Faroe. Heredity 22:610-612. Jamieson, A. and R. J. Turner. 1978. The extended series of Tf alleles in Atlantic cod, Gadus morhua, L. In B. Battaglia and J. Beardmore. Marine Organisms II. Genetics, Ecology and Evolution, Plenum Press, NY, p. 669-730. Jelnes, J. E. 1971. Identification of hexokinases and localisation of a fructokinase and a tetrazolium oxidase locus in Drosophila melanogaster. Hereditas 67:291-293. Johnson, A. G., F M. Utter and K. Niggol. 1972. Electrophoretic variants of aspartate aminotransferase and adductor muscle proteins in the native oyster (Ostrea lurida). Anim. Blood Groups Biochem. Genet. 3:109-113. Johnson, M. S. 1973. An electrophoretic study of enzyme variation in fishes of the genus Menidia (Teleostei, Atherinidae) Ph.D. dissertation, Yale University. Johnson, M. S. 1974. Comparative geographic variation in Menidia. Evolution. 28:607-618. Johnson, M. S. 1975. Biochemical systematics of the Atherinid genus Menidia. Copeia (1975) (4):662-691. June, F. C. 1958. Variation in meristic characters of young Atlantic menhaden, Brevoortia tyrannus. Rapp.-Reun. Cons. Int. Explor. Mer 143:26-35. June, F. C. 1961. Age and size composition of the menhaden catch along the Atlantic coast of the United States, 1957. U.S. Fish. Wildl Serv. Spec. Sci. Rep. Fish. 373, 39 pp. June, F. C 1965. Comparison of vertebral counts of Atlantic menhaden. U.S. Fish. Wildl. Ser v Spec. Sci. Rep. Fish. 513, 12 pp. June, F. C. and W. R. Nicholson. 1964. Age and size composition of the menhaden catch along the Atlantic coast of the United States, 1958. U. S. Fish Wildl. Serv. Spec. Sci. Rep. Fish. 446, 40 pp. June, F. C. and J. W. Reintjes. 1 960. Age and size composition of the menhaden catch along the Atlantic coast of the United States, 1956. U. S. Fish. Wildl. Serv Spec. Rep. Fish. 336, 38 pp.

PAGE 113

101 Kendall, A W., Jr. and J. W. Reintje s Geographic and h y dro distribution of Atlantic menhaden eggs and larvae along the m1ddle Atlantic coast from R/V Dolphin cruises, 1965-1966. U .S. Nat. Mar. Fish. Serv. Fish Bull., 73(2):317-335 Kirkpatric k M. and R K Selander. 1 979. Genetics of speciation in lake whitefishes in the Allegash Basin . Evolution 33(1):478 -485. Koehn, R. K. 1969. Esterase heterogeneity: D ynamics of a polymorphism. Science (vJash., D.C.) 163:943 -9 44. Koehn. R K. 1970. Functional and evolutionary dynamics of poly morphic esterases in catosto m id fishes. Trans. Am. Fish. S oc. 99(1):219-228. Kroger, R. L. and R. L. Dryfoos. 1972. Tagging and tag-recovery experiments with Atlantic menhaden, Brevoortia t y rannus. U .S. Dep. Cammer. Nat. Mar. Fish. Serv. Spec. Sci. Rep. Fis h Kroger, R. L., R. L. Dryfoos and G. R. Huntsman. 1971. Move ment of juvenile Atlantic menhaden tagged in New England waters. Chesapeake Sci. 12(2):111-120. Lehninger, A L. 1975. Biochemistry. Worth Publishers, Inc., New York, 1104 pp. Lindsey, C. C. 1954. Temperature-controlled meristic variation in the paradise fish Macropodus opercularis ( L.) Can. J. Zool. 32(2) :87 -98 Lodge, T. E. 1974. Multiple hemoglobin polymorphis m s of a catfish, the yellow bullhead (Ictalurus natalis (Lesueur)) in Florida. Ph.D dissertation, University of Miami. Loesch, J., chairman, G. Broadhead, C Grimes, W. Nelson, G Sakagawa, and K. West. 1979. Report of the Atlantic Menhaden Population Dynamics Subcommittee to the Atlantic Menhaden Scientific and Statistical Committee, 68 pp. Loudenslager, E J. and R. M Kitchin. 1979. Genetic similarity of two forms of cutthroat trout, Salmo clarki, in Wyoming. Copeia (1979) (4):673-678. Manooch, C S., III, G R. Huntsman, B. Sullivan and J. Elliot. 1 976. Conspecific status of the sparid fishes Pagrus sedecim and Pagrus pagrus Linnaeus Copeia (1976) (4):678-684. Markert, C L and I Faulhaber. 1965. Lactate dehydroge nase isozyme patterns of fish. J Exp. Zool. 159(3):319-332. Markert, C. L., J. B. Shaklee, and G. S Whitt. 1975. Evolution of a gene. Science (Wash. D.C.) 189: 1 02114.

PAGE 114

Martin, J. 1979. Biochemical. ecological and genetic studies on glucose -6-phosphate isomerase allozymes and acclimation studies on adductor muscle phosphofructokinase of the American oyster, Crassostrea virginica (Gmelin). Ph.D dissertation, Duke University. 102 Massman, w. H J. J. Norcros s .and E. B. Joseph. 1 962. Atlantic men. haden larvae in coasta l waters. Chesapea k e Sci. Maynard, J. R. and J. E. Coleman. 1971. Elasmobranch carbonic anhydrase. J. Biol. Chern. 246:4455-4464 McFarland, G. 1977. The staining of starch gels with coomassie brilliant blue G-250 perchloric acid solution. Experientia 33(12):1668. McGlade, J. and H. MacCrimmon. 1979. Taxonomic congruence in three populations of Quebec brook trout, Salvelinus fontinalis. Can. J. Zool. 57(10}:1998-2009. McHugh, J. L., R. T. Oglesby and A. L. Pacheco. 1959. Length, weight, and age composition of the menhaden catch in Virginia waters. Limnol. Oceanogr. 4(2):145-162. Menzel, B. W. 1970. An electrophoretic analysis of the blood proteins of fishes of the subgenus Luxilus (Notropis:Cyprinidae) Ph. D dissertation. Cornell University. Messteh, S. N. 1975. Delineating spring and autumn herring populations in the southern Gulf of St. Lawrence by discriminant function analysis. J. Fish. Res. Board. Can. 32(4):471-477. D. 1966. Lofoten area. Genetic difference s between cod groups in the Nature (Land.) 212:824. Morgan, R. P., II, T. S. Y. Koo and G. E. Krantz 1973. Electrophoretic determination of populations of the striped bass, Marone saxatilis, in the upper Chesapeake Bay. Trans. Am. fish. Soc. 102(1): 21-32. Nelson, w. R., M. C. Ingham and W. E. Schaaf. 1977. Larval transport and year-class strength of Atlantic menhaden, Brevoortia tyrannus. U.S. Nat. Mar. Fish. Serv. Fish. Bull. 75(1):23-41. Nicholson, W. R. 1972. Population structure and movements of Atlantic menhaden, Brevoortia tyrannus, as inferred from backcalculated length frequencies. Chesapeake Sci. 13(3):161-174.

PAGE 115

Nicholson, W. R. 1975. Age and size composition of the Atla ntic menhaden, Grevoortia tyrannus, purse sei n e catch 196371, with a brief discu ss ion of the fishery U .S. Dep Cammer. Nat. Mar. Fish Serv. Spec. Sci Rep. Fish. 684, 28 pp. 103 Nicliolson, H R 1978. Movements and population structure of Atlantic menhaden in dicated by tag returns. Estuaries 1(3):141-150. Nicholson, W. R. and J R. Higham, Jr. 1964a. Age and size composition of the menhaden catch along the Atlantic coast of the United States, 1959. U .S. Fish. Wildl. Serv. Spec. Sci. Rep. Fish. 478, 34 pp. Nicholson, W. R. and J. R Higham, Jr. l964b. Age and size composition of the 1960 menhaden catch along the U.S. Atlantic coast U S Fish Wildl. Serv. Spec. Sci. Rep. Fish. 479, 41 pp. Nicholson, W. R. and J R. Higham, Jr. 1 965a. Age and size composition of the menhaden catch along the Atlantic coast of the United States, 1961. U.S. Fish Wildl. Serv S pec. Sci. R e p Fish. 495, 28 pp. Nicholson, W. R. and J. R. Higham, Jr. 1965b. Age and size composition of the menhaden catch along the Atlantic coast of the United States, 1962, U.S. Fish. Wildl. Serv. Spec. Sci. Rep. Fish. 527, 24 pp. Nie, N. H., C. H. Hull, J. G. Jenkins, K. Steinbrenner, and D. H Bent. 1975. Statistical Package for the Social Sciences. McGraw-Hill Book Co., New York, 675 pp. Nishida, M andY. Takahashi. 1978. of a y u, Pleco g los sus altivelis. 44{10):1059-1064. Enzyme variation in populations Bull. Jpn. Soc. Sci. Fish Northcote, T. G s. N. Williscroft and H Tsuyuki 1970. Meristic and lactate dehydrogenase genotype differences in stream populations of rainbow trout below and above a waterfall. J Fish. Res. Board Can. 27:1987-1995. Odense, P. H., T. M. Allen and T. C. Luing. 1966. Multiple forms of lactate dehydrogenase and aspartate aminotransferase in herring. Can. J. Biochem. 44:1319-1326. O'Rourke, F. J. 1960. Observations on the serum proteins of the bass. Roy. In. Acad. Proc. 8.6 1 : 1 67176. Ornste i n, L 1964. Disc -I. Background and theory Ann. N.Y. Acad. Sci. 121(2):321-349. Pokrywka, G. S. and F. Gold. 1980. Separation and characterization of menhaden hemoglobin components. Hemoglobin 4:69 -88.

PAGE 116

Purdom, C. E., D Thompson and P R. Dando. 1976. Genetic anal ysis of polymorphoisms in plaice (Pleuronectes platessa). Herdity 37(2} : 193-206. Rakocinski, C F. 1980. Hybridization and introgression between Campostoma oliqolepis and C. anomalum pullum (Cypriniformes: Cyprinidae). Copeia ( 1980T (4):584 594. 104 Reintjes, J. W. 1964. Annotated bibliography on biolog y of menhadens and menhadenlike fishes of the world. U.S. Nat. Mar. Fish. Serv. Fish Bull 63(3):531 549. Reintjes, J. W 1966. Classification and distribution of North American menhadens. In Annual Report of the B ureau of Commercial Fisheries Biolog i cal Laboratory Beaufort, NC for the fisca l yea r ending June 30, 1965. U.S. Fish. Wildl. Serv. Circ 240, p. 16. Reintjes, J W., J. Y Christmas, Jr., and R A. Collins. 1960. Annotated bibliography on biology of American menhaden. U.S. Fish. Wildl. Serv. Bull. 60(170):297-322. Reintjes, J. W. and P. M. Keney. 1975. Annotated bibliography on the biology of the menhadens, genus Brevoortia, 1963-1973. U.S. Dept. Cammer. Nat. Mar. Fish. Serv. Spec. Sci. Rep. Fish. 687, 92 pp. Ridgeway, G. J., S. W. Sherburne and R D. Lewis. 1970. Polymorphism in the esteras e s of Atlantic herring. Trans. Am. Fish. Soc. 99(1): 147-151. Robertson, W. V. 1959. The osteology of Brevoortia patronus and gunteri (Pisces: Clupeidae). M.S. thesis, Agricultural and Mechanical College of Texas. Rohlf, F. J and G. D Schnell 1971. An investigation of the isolation by distance model. Am. Nat. 105(944):295-324. Roithmayr, C M. 1963. Distrib. ution of fishing by purse seine vessel s for Atlantic menhaden, 1955-5 9 U.S. Fish. Wildl. Serv. Spec. Sci. Rep. Fish 434, 22 pp. Saffran, W. A. and Q. H Gibson. 1978. The effect of pH on carbon monoxide binding to menhaden hemoglobin. J. Biol. Chern 253(9): 3171-3179. Schaaf, W. E and G R. Huntsman. 1972. Atlantic menhaden stock: 1955-1969. 101:290-297. Effects of fishing on Trans. Am. Fish. Soc. Scopes, R K. and C. Gosselin. l968. Polymorphi s m in carp muscle Creatine kinase J. Fish. Res. Board Can. 25(12):2715-2716.

PAGE 117

l 05 Shaw, C. R. and R. Prasad. 1970. Starch gel electrop horesis a compilation of recipes. Bioche m Genet. 4 : 297-320, Sick, K. 1961. Haem oglobin polymorphisms in fishes. Nature (Lond.) 192:894-896. Sindermann, C. J. 1961. Serological techniques in fishery research Trans. 26th N. Amer. Wildl. Natur. Resour. Conf. 298309. Sindermann, C. J. 1962. Ser ology of Atlantic clupeoid fishes Am. Nat. 96:225-231. Sindermann, C. J. and K. A. Honey. 1963. Electrophoretic analysis of the hemoglobins of Atlantic clupeoid fishes. Copeia (1963) (3):534-537. Sindermann, C. J. and D. F. Mairs. 1959. A major blood group system in Atlantic sea herring. Am. Nat. 96:225-231. Sindermann, C. J. and D. F. Mairs 1961. spiny dogfish, Sgualus acanthias L. 120:401-410. A blood group system for Biol. Bull. (Woods Hole) Smithies, 0. 1959. An improved procedure for starch gel electrophoresis. Further variation in the serum proteins of normal individuals. Biochem. J. 71:585. Smith, P. J., R. I. C. C. Francis and L. J. Paul. 1978. Genetic variation and population structure in the New Zealand snapper Chrysophrys auratus Forster. N. Z J. Mar. Freshw. Res. 12:343-350. Sakal, R. R. and F. J. Rohlf. 1969. Biometry. W. H Freeman and Co., San Francisco, 776 pp. Sutherland, D. F. 1963. Atlantic menhaden. 435, 21 pp. Variation in vertebral numbers of juveniles U. S. Fish. Wildl. Serv. Spec. Sci. Rep. Fish. Suzuki, A., T. Morio, and K. Mimoto. 1959. Serological studies of the races of tuna. II. Blood group frequencies of the albacore in Tg-s ystem Part I. Comparison of the Indian and the north western Pacific Ocean. Nankai Reg. Fish. Res. Lab. (Japan) Rept. 11: 17-23. Suzuki, A:, Y. Shimizu and T. Mario. 1958. Serological studies of the races of tuna . I. The fundamental investigations and the blood groups of albacore. Nakai Reg. Fish. Res. Lab. (Japan) ReD t. 8: 1 0411 6

PAGE 118

Taniguchi, N. 1969. Comparative electropherograms of muscle proteins of three species of lizard fishes referable to the genus Saurida Bull. Jpn. Soc. Sci. fish. 35(9):885-890. Taniguchi, N and I. Nakamura. 1970. Comparative electropherograms of two species of frigate mackerel Bull Jpn. Soc. Sci. Fish 36(2):173 -176. A. V. fishes. 1952. Experimental study of meristic characters in Biol. Rev. Camb. Philos. Soc. 27:169-193. Tatarko, K. I. 1968. The effect of temperature on the meristic characters of fishes. J. Ichthyol. 8:339-350. Tatum, F. M., H. Boschung and R Lindahl 1979. An isozymic com parison of two distinct chromosomal races of Fundulus notatus (Cyprinodontidae). Assoc. S E. Biol. Bull. 26(2), Abstract 8 1. Thompson, D. and S Mostert. 1974. Muscle esterase genotypes in the pilchard, Sardinops ocellata J. Cons. Int. Explor. Mer. 36(1}:50-53. Turner, K. 1979. Protein polymorphisms in the mud crabs, Eur ano eus deeressus, Panopeus herbstii forma simhsoni, and ano eus erbst1i forma obesa, at North Inlet, Sout Carolina Decapoda:Brach yura Xanthidae) M.S. Thesis, Clark University. Utter, F. M. 1969. Transferrin variants in Pacific lake, Merluccius productus. J. Fish. Res. Board Canada 26(12):3628-3271. Utter, F. M. and H. 0. Hodgins. 1972. Biochemical genetic variation at six loci in four stocks of rainbow trout. Trans. Am. Fish. Soc. 101(3):494-502. Vladykov, V. D. 1934. Environment and taxonomic characters of fishes. Trans : R. Can. Inst. 20:99-140. Weber, K. and M. Osborne. 1969: The reliability of molecular weight determinations by dodecyl sulfate polyacrylamide gel electrophoresis. J. Biol. Chern. 244:4406. Weinstein, M. P. and R. W. Yerger l 976a. Protein taxonomy of the Gulf of Mexico and Atlantic ocean seatrouts, genus Cynoscion. U.S. Nat. Mar. Fish. Serv. Fish. Bull. 74(3):599 607. Weinstein, M. P. and R. W. Yerger. 1976b. Electrophoretic investigation of s ubpopulation s of spotted seatrout, Cynoscion nebulosus, in the Gulf of Mexico and Atlantic coast of Florida. U.S.A. Comp. Biochem. Physiol. B Comp. Biochem. 54(1):97-102.

PAGE 119

Westman, W. E. 198 0. Gaus sian analysis: identifying environmental factors influencing bell-shaped species distributions. Ecology 107 Wilk, S. J., W. G. Smith, D. E. Ralph and J. Sibunka. 1980. Population structure of summer flounder between New York and Florida based on linear discriminant analyses Trans. Am. Fish. Soc. 1 09( 3): 265-271. Wilkins, N. P. and T. D. Iles. 1966. Hemoglobin polymorphism and its ontogeny in herring (Clupea harengu s ) and sprat (Sprattus sprattus). Comp. Biochem. Physiol. 17:1141-1158. Wiseman, E. D., A. A. Echelle, and A. F. Echelle. 1978. Electro phoretic evidence for subspecific differentiation and intergradation in Etheostoma spectabile (Teleostei:Percidae) Copeia (1978) (2):320-327 Wishard, L. N., F. M. Utter and D R. Gunderson. 1980. Stock separation of five rockfish species using naturally occurring biochemical genetic markers. U.S. Nat. Mar. Fisn. Serv. Mar. Fish. Rev. 42(3-4):64-73. Wright, J. E. and L. M. Atherton. Polymorphisms for LDH and transferrin loci in brook trout populations. Trans. Am. Fish. Soc. 99(1):179-192. Zenkin, V. S. 1966. Analysis of Atlantic herring (Clupea harengus haren us L.) populations by blood group frequency. J Ichthyol. 13 5 :662-668.


printinsert_linkshareget_appmore_horiz

Download Options

close
Choose Size
Choose file type
Cite this item close

APA

Cras ut cursus ante, a fringilla nunc. Mauris lorem nunc, cursus sit amet enim ac, vehicula vestibulum mi. Mauris viverra nisl vel enim faucibus porta. Praesent sit amet ornare diam, non finibus nulla.

MLA

Cras efficitur magna et sapien varius, luctus ullamcorper dolor convallis. Orci varius natoque penatibus et magnis dis parturient montes, nascetur ridiculus mus. Fusce sit amet justo ut erat laoreet congue sed a ante.

CHICAGO

Phasellus ornare in augue eu imperdiet. Donec malesuada sapien ante, at vehicula orci tempor molestie. Proin vitae urna elit. Pellentesque vitae nisi et diam euismod malesuada aliquet non erat.

WIKIPEDIA

Nunc fringilla dolor ut dictum placerat. Proin ac neque rutrum, consectetur ligula id, laoreet ligula. Nulla lorem massa, consectetur vitae consequat in, lobortis at dolor. Nunc sed leo odio.