THE AGE AND GROWTH OF BROWN TROUT (SALMO mum) AND scumm (corms 39?.) AS :1 RELATES T0 EUTROPHTCATION m THE 10am AND AUSABLE RIVERS Thesis for the Degree of M. S. MICHTGAN STATE UNIVERSTTY ROBERT F. QUICK 19'21 .méucwn chic-3". , x“. I [1.1 [ L IITIIIIITITTTTTIIITITTTTTTZTTTITT v" M ‘ ABSTRACT THE AGE AND GROWTH OF BROWN TROUT (SALMO TRUTTA) AND SCULPIN (COTTUS SPP.) AS IT RELATES TO EUTROPHICATION IN THE JORDAN AND AUSABLE RIVERS BY Robert F. Quick Three sites on two rivers in Michigan were chosen to represent a gradient of eutrophication. Brown trout (§ElEQ trutta) were found in each site and 1,361 specimens were collected by electrofishing between March and November 1970. In the same period, 644 specimens of mottled sculpin (Cottus bairdi) were collected in the two most eutrOphic sites, and 556 specimens of slimy sculpin (Cottus cognatus) were collected in the two least eutrophic sites. In streams common to both sculpin species, mottled sculpin occurred in a relative abundance ratio of 10:1 with slimy sculpin. Brown trout were aged by scale annuli interpre- tation, and sculpin were aged by otolith analysis. Growth curves constructed for the seven fish pOpulations, indi- cated that for every species, fish of comparable age were larger in the more eutrophic streams. Instantaneous growth rates for the populations in each species differed only in Robert F. Quick the first year, with higher rates being exhibited by the populations in the most eutrophic environments. The population density of brown trout was 4,395 per hectare in the least eutrOphic site, and 887 per hectare in the most eutrophic, with standing crop esti- mates for the same fish indicating a greater biomass in the latter site. Brown trout condition (Ktl) increased during the spring to a peak in July, followed by a decrease in condition with the onset of winter. Condition of sculpin was highest in the spring and reached a low point in June, coincident with spawning activity. It was sug- gested by these findings that in the early stages of stream eutr0phication, these species might be used to identify changes in stream degradation through corresponding changes in the growth rate of successive year classes of under- yearlings. THE AGE AND GROWTH OF BROWN TROUT (SALMO TRUTTA) AND SCULPIN (COTTUS SPP.) AS IT RELATES TO EUTROPHICATION IN THE JORDAN AND AUSABLE RIVERS BY Robert Ff‘Quick A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Fisheries and Wildlife 1971 to my father the late Frank C. Quick whom I loved and respected ii ACKNOWLEDGMENT S I wish to express appreciation to Dr. Howard Johnson for the help and direction he has given me dur- ing this investigation, and to my committee members, Dr. Niles Kevern, Dr. Thomas Bahr, and Dr. Terry Haines for their review of this manuscript. I would also like to thank my fellow graduate students and especially Mr. Wayne Smith for their help in collecting and interpreting data. This investigation was supported in part by FWQA Training Grant STl—WP-109 and lT3-WP-264, and in part from Grant 14-31-0001-3153, provided by the United States De- partment of the Interior, Office of Water Resources Re— search, as authorized under the Water Resources Act of 1964. Thanks also go to my wife, Claudia, for her frequent assistance and helpful suggestions throughout this study. iii INTRODUCTION . . TABLE OF CONTENTS DESCRIPTION OF THE STUDY SITES The upper Jordan River station. The lower Jordan River station. The AuSable River station MATERIALS AND METHODS RESULTS PART I--BROWN TROUT. Length-scale relationship Length-frequency relationship . Calculated length at annulus formation Instantaneous rate of growth Length-weight relationship . Population density and age structure. Condition. . RESULTS PART II--SCULPIN. Length-otolith relationship. Length-frequency relationship . Calculated length at annulus formation Instantaneous growth rate Length-weight relationship Population density. Condition. . DISCUSSION. . . LITERATURE CITED. iv Page 12 18 24 29 29 30 33 39 42 43 47 51 51 51 57 63 66 66 71 77 83 Table l. 10. LIST OF TABLES Physical and chemical water parameters at study sites during growing season (collected 4/12, 5/3, 5/31, 6/26, 8/11, 9/4, 9/25/70). . Dates of fish sampling and numbers of each species collected . . . . . . . . . . Length-scale regressions and correlation coefficients for brown trout in the Jordan and AuSable rivers. . . . . . . . . . Total length at each age of brown trout from the upper Jordan River. Numbers in paren- theses indicate sample size. . . . . . . Total length at each age of brown trout from the lower Jordan River. Numbers in paren- theses indicate sample size. . . . . . . Total length at each age for brown trout from the AuSable River. Numbers in paren- theses indicate sample size. . . . . . Length-weight regressions of brown trout in the Jordan and AuSable rivers . . . . . . Calculated weight at annulus formation of brown trout in the AuSable and Jordan rivers . Estimated number per site (i 1 SE) and number and weight per hectare of brown trout at three study sites in August 1970 . . . . . . . .Length-otolith relationship of two species of sculpin in the Jordan and AuSable rivers with related correlation coefficients . . . . . Page 11 25 30 34 35 36 42 43 45 52 Table 11. 12. 13. 14. 15. l6. l7. 18. Page Age distribution for two species of sculpin from the Jordan and AuSable rivers in suc- cessive 5 mm intervals of total length . . . 53 Total length at each age for slimy sculpin from the upper Jordan River. Numbers in parentheses indicate sample size . . . . . 58 Total length at each age for slimy sculpin from the lower Jordan River. Numbers in parentheses indicate sample size . . . . . 59 Total length at each age for mottled sculpin from the lower Jordan River. Numbers in parentheses indicate sample size . . . . . 60 Total length of each age for mottled sculpin from the AuSable River. Numbers in parentheses indicate sample size . . . . . 61 Calculated mean lengths at annulus formation for four populations of sculpin in the Jordan and AuSable rivers (: 1 SE). . . . . . . 62 Length-weight relationships of two species of sculpin in the Jordan and AuSable rivers . . 67 Calculated weight (gm) at annulus formation for two species of sculpin in the Jordan and AuSable rivers . . . . . . . . . . . 70 vi LIST OF FIGURES Figure 1. Jordan River showing location of collecting stations . . . . . . . . . . . . 2. Study site on the upper Jordan River. . . 3. Seasonal variation of water temperature in the Jordan and AuSable rivers as diurnal maxima (A—A), diurnal minima (0-0), or as a single midday reading (CD) . . . . . . 4. Seasonal variation of dissolved oxygen in the Jordan and AuSable rivers as diurnal maxima (A-A) and diurnal minima (0-0). . . . . 5. Study site on the lower Jordan River. . . 6. AuSable River showing location of collecting Station 0 O O I O O O O O O O O 7. Study site on the AuSable River . . . . 8. Adult sculpin showing point of incision in preparation for otolith removal . . . . 9. Length frequency distribution of brown trout in the Jordan and AuSable rivers for fall 1970. Roman numerals designate age groups. Fish were grouped by 10 mm intervals of total length. . . . . . . . . . . 10. Mean annual increments of growth for brown trout in the Jordan and AuSable rivers. Vertical lines represent 95% confidence limits of the mean. . . . . . . . . ll. Instantaneous rate of growth for each year, of brown trout in the Jordan and AuSable rivers for 1970. . . . . . . . . . vii Page 10 14 16 20 23 27 32 37 41 Figure Page 12. Changes in mean condition (K) of yearling brown trout in the Jordan and AuSable rivers during 1970. Vertical lines repre- sent 95% confidence limits of the mean . . . 49 13. Changes in mean condition (K) of two-year- old brown trout in the Jordan and AuSable rivers during 1970. Vertical lines repre— sent 95% confidence limits of the mean . . . 50 14. Length-frequency distribution of two species of sculpin in the Jordan and AuSable rivers for August 1970. Roman numerals designate age groups. Fish were grouped by 2 mm intervals of total length . . . . . . . 56 15. Instantaneous rate of growth for each year of two species of sculpin in the Jordan and AuSable rivers for 1970 . . . . . . . . 65 16. Length-weight relationship. Logarithmic plot of mean weight against mean length for two species of sculpin from the Jordan and AuSable rivers . . . . . . . . . . . 69 17. Changes in mean condition (K) of two species of yearling sculpin in the Jordan and AuSable rivers during 1970. Vertical lines represent 95% confidence limits of the mean. . . . . 73 18. Changes in mean condition (K) of two species of two-year-old sculpin in the Jordan and AuSable rivers during 1970. Vertical lines represent 95% confidence limits of the mean . 75 viii INTRODUCTION Age and growth studies of fish pOpulations have, in the past, been generally of a descriptive nature. How- ever, recent interest in the problem of stream degradation has created the need to relate the results of this type of study, to the changes occurring in the aquatic environment. Beyerle and Cooper (1960) compared three streams in Pennsylvania and found that brown trout (galmg trutta) grew faster in enriched water. McFadden and Cooper (1962) found that Pennsylvania streams with high conductivity, indicat- ing a generally fertile stream, produced larger brown trout than did streams of low conductivity. While these studies demonstrated that a higher growth rate is related to higher levels of nutrients in a stream, changes in the physical environment can also lead to changes in the growth re- sponses of the fish pOpulation. For example, Hunt (1966) demonstrated an increase in production of wild brook trout (Salvalinus fontinalus) by artificial alteration of the stream channel, and Brown (1946) found that there are temperature Optima for maximizing growth in two-year-old brown trout. An example of biological factors which can affect growth is given by LeCren (1969) who noted a relationship between the growth rate of brown trout and pOpulation den— sity. These and other factors all operate together to determine the growth rate of a fish population. Brown trout exist on a relatively high trOphic level, and as such are not considered to be an accurate indicator of environmental change because physical and biological inter- actions operating on lower trophic levels tend to mask environmental changes before they are expressed in the fish. For the same reasons, however, the growth of these fish will not be influenced by short-term environmental variations and should, therefore, be a good indicator of the existence of long-term alterations in the stream ecology. A three-year investigation to develop predictive indices of stream eutrophication was initiated in 1970. Data concerning water quality, aquatic macrophytes, aquatic invertebrates, and fish growth and fecundity have been collected as part of the investigation. The stream sites studied were selected to represent a gradient of eutrOphication ranging from pristine to extensively per- turbed. This study was an initial part of the three-year investigation, and was concerned with the age structure and growth patterns of brown trout and sculpin in the three least perturbed stream sites. An attempt has been made to relate the growth of these fish to the level of eutrophication characterized by each study site. All of the stream data relating to this study, excluding data concerning fish, was collected by other researchers work- ing on the three-year investigation. DESCRIPTION OF THE STUDY SITES The upper Jordan River station The first site is located on the Jordan River near its source (T31N, R5W, section 31 of the Michigan P.M.) and represents unperturbed stream conditions (Figure 1). This study site has an average width of 13-16 meters, an average depth of 0.4-0.8 meters, and flow of 0.62 cubic meters per second. The stream bottom is predominantly sand and gravel and is bisected by fallen trees every few meters. These obstacles alter the stream flow, creating numerous depressions in the bottom (Figure 2). The shallow stream edges are silt covered and support dense growths of Chara sp. during the summer. Water temperatures varied throughout the sampling period with an estimated maximum of 20 C in July and a recorded minimum of 3.5 C in November. Diurnal temperature differences ranged from 18 C in the spring to 2 C in the fall (Figure 3). Water quality data pertaining to the site is located in Table 1. Phosphorus levels at this site were lower than at the other study sites, while nitrogen was the highest. Dissolved oxygen levels were relatively stable throughout the summer and ranged from .wcofiumum mcfluomaaoo mo GOHDMUOH mcfl3onm um>flm CMUHOhII.H musmflm ‘D {2821 5.... .2282 .m 598.. D 2.» 3:: 5.3.. .32.. x EM; \k \< ‘QQQB 2.» >2...» :83... .26.. X .um>flm common means on» so muflm mosumll.m mudmflm a lot-§.3.. te dads; , :0 . .J if... .......£T.. . . _... «WWIW. ..Wx‘~...ds.n .. _.—¢\Qd$.v Nw'fim. . A @v mcflwmmu mmpcwa mamcflm m mm Ho .Aouov mEHCHE Hmcnswo .Aflu manmmsm can Gammon may cw mnsumummfimu Hmum3 mo coflumanm> Hw:0mmmm|n.m onsmflm lO 2 5 . UPPER JORDAN 20- I5- IOb o 5; o E o @ lu 3 q “t 3:“ g tn 0 b \1 V no 000000 00 NN—- N (o) aanJadwal JawM Month ll Amo.|mo.v Amm.lmo.v Ammmummav Ammaummav . . I . I o m s mo.o hma mva ma 0 o H m 0 mm om an m 4 Amo.nao.v Aem.|mm.v Aommummav Aaomnanav . . u . u m eo.o cam mma mo 0 o H m 0 ma NH ampuon H 3oq Amo.lao.v Lem.ufimm.v Aoamlnmav Ammaavmav . . I . I am no Home mo.o mma NmH Ho 0 m o v o ma ma U h D msuosmmonm mcflaom mmmcoumc .uosam .vmum .vmum Hmuou Hmuou accommmm umOE umoE auow3 Hm>flm numme wmmmmww lumufla\mac Hmonamao Ase Hmoflmsnm .Ao>\mm\m .v\m .HH\m .mmxo .Hm\m .mxm .NH\¢ wmuomaaooc QOmmmm mcflzonm mcwuso mwufim mvsum um mnmumamumm Houm3 HMOflEmno cam Hmowmmnmnl.a manme 12 13 to 8.4 ppm (Figure 4). The major aquatic macrophytes present were Chara vulgaris, Potamogeton filiformis, and Elodea canadensis. Few fish species were encountered at this site. Slimy sculpin (Cottus cognatus) and brown trout comprised 50% and 30% of the March collection respectively. The remainder was composed of brook trout and rainbow trout (Salmo gairdneri). A single brook stickleback (Culea inconstans) was found early in the year. Coho salmon (Oncorhynchus kisutch) carcasses were observed, indicating that their spawning run from Lake Michigan had carried them to these distant reaches of the river. The first three fish collections from this site were taken from a stream section 31 meters long. The re- maining collections were taken from a lBZ-meter-long section which was a short distance upstream of the first location. The lower Jordan River station The second site is located in the lower reaches of the Jordan River (T31N, R6W, section 7 of the Michigan P.M.)(Figure l). The site is 601 meters long with an average width of 17.4 meters, a depth of 0.6-1.0 meters, and flow of 3.54 cubic meters per second. The stream bottom is predominantly medium to coarse gravel with short stretches of sand in the center of the stream channel (Figure 5). Dense mats of Elodea sp. occur along l3 . 3:3 mEHQHE HmcHsHo paw Awu wanmms< cam CMUHOU man ca cmmhxo ©m>aommwc mo COHHMHHM> Hmcommmmln.v musoflm 14 CO 5: 3. Q q 6 6 s 5 3 0: o: “D u. Lu ET 3: k S 3 3 222“° 322“° 329» ('l/‘bun uobflxo pamossgg Month 15 .Hm>Hm cMUHOh HmBOH may no muflm mpsumll.m musmam 16 . . ”4.... .. . firliyfldu . .4. . .... ’ 0.. 0.01.“... 17 the banks. Daytime water temperatures recorded during the sampling period reached a maximum of 18.5 C in July and a minimum of 5 C in November, with diurnal fluctuations of 10 C to 2 C (Figure 3). Total phosphorus and nitrogen values were intermediate to those of the other two sites (Table 1), while dissolved oxygen levels and diurnal pulses were similar to those of the other sites (Figure 4). The most abundant species of aquatic macrophytes were Elodea canadensis, Potamogeten filiformis, and Chara vulgaris with E. canadensis comprising the majority of the standing crOp. During the March sampling, mottled sculpin (Cottus bairdi) accounted for 75% of all fish collected. Brown trout and slimy sculpin comprised 17% and 5% of the col- lection respectively. Brook trout were next in abundance (3%) with the central mudminnow (Umbra limi) making up 1.6% of the collection. A single northern creek chub (Semotilus atromaculatus) was also collected, but this species was not encountered again during the remainder of the study. Many transient fish species were observed periodically throughout the year including trout-perch (Percopsis omiscomaycus) during June and July, rainbow trout from June through November, rock bass (Ambloplities rupestris) in June, and single specimens of coho salmon, steelhead trout (Salmo gairdneri), and sea lamprey (Petromyzon marinus) all apparently on spawning runs. 18 White sucker (Catostomus commersoni) were collected throughout the year but their distribution was restricted to a small pool at the lower end of the study site al- though a few white suckers were found throughout the site during the late summer. Stocking records of the Michigan Department of Natural Resources indicate that none of the study sites had been stocked with fish for at least three years prior to this study. Two thousand yearling brown trout were stocked in the Jordan River in 1970, three stream kilom- eters below the lower Jordan River study site, but there was no indication that any of these fish reached the study site. The AuSable River station The AuSable River flows across the north-central portion of Michigan's lower peninsula. It is used exten- sively for canoeing, fishing, camping, and streamside residence. Above the study site, the river receives pri— mary treatment sewage effluent from a community of 2000. This enrichment and other factors have created a condition of moderate pollution in the river. The study site was located 13 kilometers downstream of the community men- tioned, and 171 kilometers above the mouth of the river (T26N, R3W, section 10 of the Michigan P.M.)(Figure 6). The site is 518 meters long, has a width of 20-25 meters, depth of 0.5-1.0 meters, and flow of approximately 4 cubic 19 .coflumum mcfluomaaoo mo :oHDMUOH mcH3ozm Hm>Hm wanmm5¢ul.m musmfim 20 ‘T ._E_ .9. 2.2 X 23m 2 €2.86: .3... 22m thxxt qufim. hv. ‘1. 21 meters per second. The stream bottom is composed of sand with a gravel riffle 30 meters in length near the midpoint of the site (Figure 7). Water temperatures ranged from 22.2 C to 5.6 C during the sampling period, and dissolved oxygen values were similar to those observed at the other sites. Total phosphorus was the highest and nitrogen the lowest when compared to the other sites (Table 1). Potamogeton filiformis was by far, the most abundant macrophyte with g. crispus occurring less frequently. Dense beds of Potamogeton sp. developed during the summer, covering the entire stream bottom. Mottled sculpin and brown trout were the most abundant species comprising 30% and 27% of the March fish sample respectively. Brook trout (14%) and creek chub (12%) were next in abundance while black nosed dace (Rhinichthys atratulus), white sucker, johnny darter (Etheostoma nigrum), Rainbow trout, and common shiner (NotrOpus cornutus) each occurred as less than 10% of the collection. The relative abundance of these fish changed in samples taken throughout the year due to the occurrence of transient fish Species. The most notable was the appearance of yellow perch (Perca flavescens) and white suckers in increasing numbers as the summer progressed. Small numbers of rock bass and golden shiner (Notemigonus crysoleucas) were also encountered during the summer. 22 .Hm>flm manmmsd may no muflm mpsumll.h musmwm T}..- Ln.» Lu... . .:I.. by»?... ‘0 ‘ s a ...1. w...b.. . .._.ac..—-~{.-.n._.. MATERIALS AND METHODS Each station was sampled monthly during the period of March 1970 to November 1970 (with the exception of October when fish samples were not taken). All fish were collected with an electrofishing apparatus, powered by a Homelite 115 volt alternating current generator, which was towed upstream in a small boat while fish shocking pro- ceeded. Each fish sample consisted of a single electro- fishing run through the study site for the purpose of col- lecting fish for growth determinations (Table 2). In addition, each site was sampled three consecutive times in August to obtain an estimate of fish population numbers. The AuSable River site was again sampled three consecutive times in November to obtain a new estimate of mottled sculpin population numbers. Population estimates were calculated by the Petersen method, and dorsal fin clips were employed to identify recaptured fish. The fish in each run were aged using the appropriate age-group length intervals constructed from the interpretation of scales collected in the final run of each collection. This 24 25 Table 2.--Dates of fish sampling and numbers of each species collected. . Brown Mottled Slim Dates of collection trout sculpin sculpin C 3-25-70 34 -- 57 3 4-17-70* -- -- 45 H 5—16-7o+ 44 -- -- g 6-26-70 44 -- 28 H 7-22-70 60 —— 59 0 8—25-70 110 —- 162 & 9-21-70 106 -- 79 D 11—14-70 64 -- 57 a 3-24/70 31 131 9 6 4-18—70* -- 18 11 B 5-16-7o+ 22 -_ -_ g 6-24-70 60 29 3 7-22-70 58 56 7 3 8-24-70 71 -- 24 5 9-20-70 77 100 13 a 11-14-70 72 60 2 3-23-70 38 42 -- 4-16-70* -- 42 —- g 5-15-7o+ 33 -- -— 3 6-23-70 44 13 -- 3 7-21-70 62 19 -— 3 8-21-70 171 37 -- 9-19-70 85 15 -- 11-12-70 75 82 -- *No attempt was made to collect brown trout on these dates. +Sculpin were collected on these dates but deteriorated during storage and were not included in calculations. 26 allowed a separate population estimate to be made for each age-group of a species. Captured fish were measured to the nearest milli- meter total length and weighed to the nearest gram. A minimum of twenty brown trout scales were removed from an area on the fish below the anterior edge of the dorsal fin at a point midway between the dorsal fin and the lateral line. The scales were impressed on acetate slides using a roller press without heat. This method distorts the scale impression slightly, but the distortion is constant at all points on the impression, resulting in a minimal bias to calculations derived from observations of scale annuli location (Butler and Smith, 1953). The scale im- pressions were magnified to 44 diameters on a Baush and Lomb scale projector and antero-lateral scale radii and annuli locations were recorded. Since the sculpin were to be aged by otolith examination, they were not returned to the stream after capture, but were placed in 40% formalin and after 24 hours were transferred to 40% alcohol for storage. The sculpin were dissected by making a transverse cut through the head from a point behind the skull to the posterior edges of the Opercular spines to gain access to the paired otoliths (Figure 8). These were removed with forceps, cleaned of connective tissue, and stored in glycerine. The otoliths became translucent within 24 hours revealing opaque banding. After this clearing 27 How coflumummonm cw coflmfloca mo ucflom .Hm>OEmH suflaouo maw3osm camasom uHSpdT|.m munmflm 28 process, the otoliths were magnified to 40 diameters on a dissecting microscope and observed under indirect lighting. The radius was measured on the concave side of the otolith from the central core to the most distant lateral edge. The annuli appeared as thin, transparent lines located more distant from the core than the thick opaque bands representing the previous spring growth (Bailey, 1952). Age and growth data for all species was calculated by computer with a program developed by Hogman (1969). The program was modified to permit the use of length- weight information from fish which had not been aged. RESULTS PART I--BROWN TROUT Length-scale relationship As described previously, total scale radius was recorded for most of the brown trout collected. When these values were plotted against the length of the fish from which the scale had been taken, a linear relationship resulted of the form: Y = a + bx where Y = total length in millimeters, a = the intercept value on the total length axis, b = the slope of the re- gression line, and X = the scale radius in millimeters (x 44). Some species of fish may have length—scale relationships which are slightly curved (Ricker, 1958). This was not true in the brown trout populations studied. Correlation coefficients for the equations ranged from .90 to .95, indicating a nearly perfect fit with the data (Table 3). 29 30 Table 3.--Length—scale regressions and correlation coef— ficients for brown trout in the Jordan and AuSable rivers. . . Length-scale Correlation Sampling Slte relationship coefficient Upper Jordan Y = 20.51 + 4.3158X .899 Lower Jordan Y = 14.87 + 4.4978X .948 AuSable Y = 23.85 + 4.5026X .901 Length-frequency relationship Histograms of the length-frequency of brown trout were constructed from data collected during August and are based on a minimum of 322 individuals per site (Figure 9). The modes of the histograms were assumed to represent the distribution of lengths in a given age-group based on reasoning given by Tesch (1968). The mean length at cap- ture for each age-group is also indicated. In each histo— gram, the calculated lengths for age-groups 0 through 3 are indicated and coincide with the distributional modes. Unregenerated scales could not be found for a number of fish over 220 mm long in the upper Jordan collection. Therefore the mean length at capture for age-group 3 seems to be biased downward in comparison with the modes of the histogram for that population. The mean length at capture of age-group 3 brown trout in the upper Jordan River dur- ing July, August, and September, when averaged, indicate that in August the mean length of age-group 3 lies closer 31 Figure 9.--Length frequency distribution of brown trout in the Jordan and AuSable rivers for fall 1970. Roman numerals designate age-groups. Fish were grouped by 10 mm intervals of total length. Number of Fish IZO- IOO. 80. 60- 4o. 20. 60 4o- 80- 60 4O 20 60 32 UPPER JORDAN I II III 1 O 1 LOWER JORDAN I II III 1 i 6 AU SABLE I II III 6 i O I00 I40 ISO 220 260 300 340 380 420 Total Length (mm) 33 to 240 mm. This value fits the distributional mode for age-group 3 more closely. Since the modes of the histo- grams approximate the average length of each age-group as derived by the scale aging technique, it can be concluded that for this study, this method of aging brown trout was valid. Calculated length at annulus formation The total length of a fish at the time of annulus formation was calculated from the Lee-Lea equation: (Lt-a) Ln = a + ——§——— Sn t where Ln = the length of the fish at annulus n, Lt = the total length of the fish at capture, a = the total length intercept of the body-scale equation, Sn = scale radius at annulus n, and St = total scale radius at capture (Hogman, 1969). The mean length at the time of annulus formation for every year class, and a weighted mean length at annulus formation for the population as a whole was calculated (Tables 4 through 6 and Figure 10). The reliability of the calculated lengths for age-group 4 fish was reduced because few fish of this age group were collected. The resultant growth curves were compared at the first three annuli to determine if significant differences existed in growth between the three populations. An analysis of variance test was conducted for each comparison 34 o.mm m.om n.mm H.55 ucoEmHocH Amvv.nnm Ammvv.mam Aamavm.moa Amovva.nn momma wmugmflmz h.Hom «.nhm m.mmm m.vma Amy «.mn >H o.mmm m.mam H.5ma Ammv m.nm HHH o.mmH m.moa Amvavm.mn HH n.HvH Ammmvm.wn H musummo m w m m A no QB mzoum mod mommm>< AEEV cofiumEHOM msasccm um zumcma pmumasoamu .mnam mHQEmm muMOflccH mmmwnucmumm CH mumnesz .Hm>flm cmpHOb woman oz» Eoum usouu GBOHQ mo mom 30mm um aumcma Hmuoalt.v magma 35 m.mm m.nm o.moa m.o~H o.moa ucmfimuocH Avvm.mmv AHHVH.omv Ammvm.mmm Aamavo.nam onmvo.moa mammE cmuamamz m.vmm m.mmv m.nH¢ m.amm m.omm Av. o.Hm > v.0vq v.amv m.mmm N.5HN Anv «.mm >H m.mom «.mmm o.mHN Anmv H.om HHH m.mmm m.mam Amaavm.bm HH n.0ma Ammavm.HHH H 9.398 m 4 m m a um 49 msoum mmfi mmmum>¢ AEEV GOAumEHOM msHsccm um numcma pmumasoamo .muflm mHmEmm mumowpcfl mononucwumm cw mumnfisz .um>flm cmpH0h umzoq onu Eoum usouu c3oun mo omm comm pm spmcma Hmuoeul.m wabme 36 m.mm o.mn m.moa N.HNH ucmEmHocH Amvv.mnm Aqmvv.aam Amnava.amm Ammvvm.ama momma umunmfimz «.mmm «.mnm m.HHm m.me Am. m.mma >H N.vmm v.aam m.nmm Ammv m.mmH HHH m.mmm v.mmm Amvavm.ama HH w.HmH A>>NVN.ONH H mnsummo m e m N a pm AB moonm 00¢ mmmum>< AEEV coaumEHOM msasccm um zumcma Conwasoamo .mnwm meEMm mumoflpcw mommnunmumm cw muwnadz .Hm>flm wanmmsd mnu Eoum noon» GBOHQ wow wow 30mm um numcwa Hmuoall.o manna 37 500 - .b O O I \. .-‘~-. 0‘ o o 1 \K \ \ — Upper Jordan —.— Lower Jordan - - -. Au Sable Total Length (mm) Age Figure 10.--Mean annual increments of growth for brown trout in the Jordan and AuSable rivers. Vertical lines represent 95% confidence limits of the mean. 38 and in every case a significant difference was indicated (P=.01). This means that at each of the selected points on the growth curves, at least two of the three populations exhibited lengths which were significantly different. Kramer's modification to Duncan's New Multiple Range Test was employed to more accurately determine which pairs of growth curves were different (Kramer, 1956). This modifi- cation allows comparison of mean values which are derived from unequal sample sizes. The mean length at each annulus for each population was significantly different (P=.01) with the exception of the lower Jordan and AuSable rivers brown trout at the second annulus. This was ex- pected since growth curves for the two populations inter- sect during the third year of life. Based on the data just given, it is assumed that growth of brown trout in the upper Jordan River was re- stricted in comparison to the other two pOpulations. It is difficult to make a similar statement about the rela- tive growth of the brown trout in the lower Jordan and AuSable rivers since the AuSable population is initially the more rapidly growing of the two but becomes subordinate in growth to the lower Jordan brown trout after the second year. "Lee's phenomenon" is said to occur when the length at a particular annulus appears to be less when calculated from older fish, than when it is calculated from younger 39 fish (Lee, 1912). This has been observed previously in brown trout length calculations by Ball and Jones (1960), and a list of possible causes of this phenomenon is given by Tesch (1968). Lee's phenomenon occurred in the length calculations for lower Jordan trout, and a reversal of this phenomenon was exhibited in the length calculations for the AuSable trout population. The occurrence of this phenomenon in lower Jordan brown trout length calculations suggests that the actual lengths might be higher for young trout and lower for older trout than was calculated. Similarly, the reversal of Lee's phenomenon in length calculations for AuSable brown trout indicates that the actual lengths may be lower for young fish and higher for older fish than was calculated. If this is true, then the growth curves of the lower Jordan and AuSable brown trout differ less than is now indicated. Length calculations for the upper Jordan brown trout population did not exhibit Lee's phenomenon. Instantaneous rate of growth The instantaneous growth rate (or specific growth rate) is described by the general formula: loge Yt-logeYO G= t where G = the instantaneous rate of growth, Yt = the length at the end of the period, Yo = the length at the 40 start of the period, and t = the time interval between Yt and Y0 (Tesch, 1968). By choosing "t" to be one year, Y and Y0 can be t the estimate of length at two consecutive annuli. An estimate of the length of fry in the swim-up stage was necessary to determine the instantaneous growth rate dur- the first year. Beyerle and Cooper (1960) determined the average length of brown trout fry as 23 mm when they leave the redd. In Pennsylvania, brown trout are 25 mm long when the yolk sac is absorbed (Ball and Jones, 1962). The length of fish in age-group 0 used in growth rate calculations by Frost (1945) is 24 mm. The latter value was assumed to be the length of age-group 0 fish in the present study. The instantaneous growth rates calculated for the brown trout populations in this study are pre- sented in Figure 11. All three populations had the high— est growth rates as young fish and the AuSable River young grew the fastest of the three populations of young. In later years, however, this is reversed with the AuSable River brown trout growing at a rate equal to or slightly less than that of the other two trout p0pulations. There was little difference in instantaneous growth rates be- tween pOpulations of brown trout in the older age groups, regardless of the rate of growth of young. 41 |.6- Q \ \ \ \ \ O “ I 2 . \ \ T .2? E '5 3 2 .———0 Upper Jordan as 0.8 ‘ o— - —0 Lower Jordan 0, Ann-'4 Au Sable 3 a: c 2 c. 2 g 0.4r L l l I O I 2 3 4 Age Figure ll.--Instantaneous rate of growth for each year, of brown trout in the Jordan and AuSable rivers for 1970. 42 Length-weight relationship The expression: log W = log a + (b)(log TL) was used to describe the relation of fish length to weight where W = weight in grams, TL = total length in milli- meters, and a = the intercept value on the length axis (Carlander, 1969). "b" is almost always between 2 and 4, and is typically close to 3 for salmonids (Lagler, 1956). Determination of the correlation between length and weight was not attempted and is usually considered unnecessary (Tesch, 1968). The length-weight equations were deter- mined from the combined monthly collections to minimiZe the effects of growth stanzas and seasonal condition en- countered in calculations which are based on a single large sample (Beyerle and COOper, 1960)(Table 7). Table 7.--Length-weight regressions of brown trout in the Jordan and AuSable rivers. Sampling site Length-weight regression Upper Jordan log W = —4.9227 + 2.9641 logTL Lower Jordan log W = -4.7720 + 2.8981 logTL AuSable log W = -4.8556 + 2.9309 logTL 43 The length-weight regressions were almost identical for all three pOpulations and are similar to the regression; log W = -4.908 + 2.96 log TL which was determined for brown trout in the Pigeon River in Michigan (Cooper and Benson, 1951). The weight of the brown trout at the time of annulus formation was determined from the lengths calcu- lated for the same period, using the appropriate length- weight regression (Table 8). Table 8.--Calcu1ated weight at annulus formation of brown trout in the AuSable and Jordan rivers. Calculated weight at annulus Site Sample formation (grams) Size 1 2 3 4 5 Upper Jordan 407 4 43 106 299 Lower Jordan 340 12 99 322 677 1017 AuSable 456 17 118 283 501 The differences in weight between populations reflected the differences in lengths between populations because the calculated lengths were used to compute weight. Population density and age structure Population densities were estimated by the Petersen method of mark and recapture and extrapolated to an area 44 of one hectare (Lagler, 1956). The Petersen method is based on the following assumptions: (1) marked fish retain identifying marks throughout the study period, (2) marked fish are evenly redistributed throughout the popu- lation, (3) marked and unmarked fish are equally sus- ceptible to subsequent capture, and (4) immigration or emigration does not occur (Robson and Reiger, 1968). Criteria of the first assumption are met by the short- term nature of the sampling. The fourth assumption in- cludes the requirement that mortality of marked fish is not greater than unmarked fish. Brynildson and Brynildson (1967) have shown that marking by fin clip does not intro- duce appreciable mortality, and Bouck and Ball (1966) have shown that mortality due to electroshocking is minimal under most conditions. The second and third assumptions could have been violated because a maximum period of 24 hours separated the collections, and this may not have been sufficient time for marked fish to recover and be re-distributed in the population. If these two assumptions were violated, then the population estimates have been biased upwards. This bias should have negligible effect on the use of the estimates as a means of comparing popu- lations between sites since the time intervals used for population estimate sampling were the same at each site, resulting in an approximately equal bias to all estimates. The fish caught in the first and subsequent sampling runs of the estimate were assigned ages according 45 to the length-age relationship of the population. This relationship was constructed from age determinations of fish in the final sample run. An estimate of the size of each age-group was then computed using the Peterson formula. In the upper Jordan River site, brown trout den- sity exceeded 4,000 fish per hectare compared to 699 per hectare in the lower Jordan and 887 per hectare in the AuSable River (Table 9). While the upper Jordan brown Table 9.-—Estimated number per site (+ 1 SE) and number and weight per hectare of brown trout at three study sites in August 1970. Number Of Number Of Kilograms of Site Age iii: per fish per fish per (I 1 SE) hectare hectare Upper Jordan 0 616:51.13 3075 12.3 River I 150:12.45 750 18.8 II 96: 2.98 480 36.0 III 17: 6.77 85 10.8 Lower Jordan 0 218:26.60 335 3.0 River I 166:10.46 255 17.3 II 57: 5.10 78 12.0 III 20: 9.70 31 12.1 IV -- -- -- V 2i 0.0 3 2.8 AuSable 0 42:17.68 69 0.6 River I 358125.06 587 35.2 II l4l:l7.20 231 43.2 trout maintained more than four times the pOpulation den- sity of the AuSable River brown trout, the upper Jordan brown trout exhibited a smaller standing crop in biomass because the majority of fish in the upper Jordan population 46 were underyearlings which did not contribute greatly to the total weight of the population. Brown trout of age-groups 3 and 4 in the AuSable River, and of age-group 4 in both Jordan sites were absent from the August fish sample and were not included in the pOpulation estimate. However, small numbers of fish in these age-groups were collected in other monthly samples. There was an inordinately small number of young- of-the-year fish in the AuSable brown trout population. It would seem reasonable that sampling efficiency was not responsible for this apparent lack of young fish since bias of this type would not affect the population estimate as long as it was consistent for each sampling run, an assumption which is acceptable. If the age—group 0 esti- mate is valid, two possible causes for its small size would be: (1) the study site will not support large numbers of young fish, or (2) recruitment was greatly restricted during 1970 only. If recruitment is con- sistently low, then a high rate of immigration of year- lings would be necessary to maintain the observed population structure. This does not seem likely since brown trout over 150 mm long do not tend to emigrate from their home range (Mense, 1970). In addition, high population densities have been assumed to be the chief cause of transients in a brown trout population (Jenkins, 1969). Yearling brown trout in the AuSable River, 47 attained the greatest length at the first annulus of any trout population of which I have knowledge. It is unlikely that these fish could grow as well their first year, in the presence of the high population density necessary to cause their emigration to the study site. The alternative cause of a small population of young fish, that recruitment was restricted for the 1970 year class only, can not be proven without determining the numbers of young brown trout in the AuSable River in subsequent years. Regardless of how they entered the population, the AuSable brown trout did grow rapidly, and had maintained the largest standing crop in biomass of the trout populations studied. Condition Condition factors of l- and 2-year-old brown trout, were determined for the months of May, June, July, Septem- ber, and November of 1970. The condition factor (Ktl) is a measure of the relative well-being of a fish, and is calculated from the equation: where W = weight in grams and L = total length in milli- meters. The constant 105 is used to bring the values near unity (Carlander, 1969). The condition of both age groups tended to in- crease during the spring to a peak in July, and then 48 declined progressively through the remainder of the sampling period (Figures 12 and 13). The same trends in condition were noted for brown trout in Spruce Creek, Pennsylvania (Beyerle and Cooper, 1960). The condition of yearling brown trout was greater than that of two-year-old trout in each population re- flecting a trend towards decreased condition with age. A similar tendency has been noted by other investigators (Ball and Jones, 1960 and Kathrein, 1951). Upper Jordan yearling brown trout had a higher condition early in the year, than did yearlings in the other two brown trout populations studied, but all exhibited the same condition by the end of the summer. When comparing the condition of pOpulations at a given length, as would be the case be- tween two-year-old upper Jordan brown trout (Figure 13), and yearling trout in the lower Jordan and AuSable rivers (Figure 12), the upper Jordan trout exhibit a consistently lower condition throughout the year than do the other two brown trout populations. Since age and length both affect condition in fish, to compare these three trout popu- lations at the same age and length might necessitate com- paring populations at different seasons, which would introduce yet another bias to the comparison. Obviously comparing fish populations by their respective condition factors is possible only when they exhibit the same growth rate throughout life, enabling fish of the same length, and age to be compared at the same season. 49 .cmmE mau mo mufiEwH mocmpwmcoo wmm ucmmmummu mmcfla Hmo twuum> .onma mcwusp mum>wu manmmsd paw cMUHOU may :H unouu czonn mcflaummm mo Amy coauwpcoo some CH mmmcmnoll.ma musmfim .4 5 . a q .mqmvfi. he. 5.62 z m .3 2 qqud-fiq EYQQQS QN§QQ >\ VQQQS kNanxV l “I 0°. 0 0. 9: 10400;) uompuao 50 .cmmE may mo muflEwH mocmpflmcoo «mm ucmmwumwu mmcwa Hmoflpum> .oan mcwusp mum>fiu manmmsd paw cuckoo on» ca unouu ozonn pHOIHammlo3u mo AMV cofluflpcoo some cw mmmcm3011.ma ousmflm £52 2 z m a. 2 2 ma! qu a a ..d . . - a . 4 d d 1 0°. 0 101003 uompuoo l l l O I l l N e — ‘ l ‘2 . E‘QQQS \< ‘QQQS .w «Qua. \V v. «(1qu .V thoib RESULTS PART II--SCULPIN Length-otolith relationship An otolith radius was determined for each sculpin collected and the relationship between total length and otolith radius was ascertained from the general formula: Y = a + bx where Y = the total length in millimeters, a = the inter- cept value on the total length axis, b = the slope of the regression line, and X = the otolith radius in milli- meters (x 40) (Table 10). Length-frequency relationship The length-age distributions of the four sculpin pOpulations demonstrate an extreme overlap of lengths be- tween members of adjoining age-groups older than one year (Table 11). This overlap tends to mask the distributional modes of the length-frequency histograms for these pOpu- lations of sculpin (Figure 14). However, the large sample from which upper Jordan River slimy sculpin length fre- quency histogram was made, does show distinct modes repre- senting the first three age—groups, and the calculated 51 52 mam. xenoa.a + NH.NT u m mabmms< camasom omfluuoz 4mm. xmva.H + an.m- n » cmpuoe um3oq cflaasom mmauuoz vom. xamma.a + va.v u w cannon H0304 chHsom heflam mam. vawH.H + mm.~- u 5 coupon “mam: cflaasom meflam ucofioflwwooo mflnmcowbmHmu :ofiumHmuHoo nufiaouOInumnmq mufim hpsum mmfiommm .mucmHOHmmmoo coflumamnuoo poumeu nuw3 pom ampHOh ecu ca camasom mo mmflommm o3u mo mflnmcoHumHmu mum>flu manmmsé nuflaouo-sumamqlu.oa magma 53 Table 11.--Age distribution for two species of sculpin from the Jordan and AuSable rivers in successive 5 mm intervals of total length. Slimy sculpin Interval of Upper Jordan Lower Jordan totaimlength age group age group O I II III 0 I II III 25-29 1 30-34 10 35-39 28 40-44 12 45-49 1 50-54 10 1 55-59 9 60-64 16 1 65-69 16 l 1 70-74 9 6 3 75-79 3 5 l 2 80-84 2 l 0 1 85-89 1 l 3 1 90-94 1 95-99 1 54 Table 11.--Continued. Mottled Sculpin Interval of total length (mm Lower Jordan AuSable age group age group O I II III 0 I II III 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 6 70-74 10 75-79 16 80-84 6 85-89 3 90-94 1 95-99 100-104 105-109 110-114 115-119 120-124 I—‘N I-‘I-‘xlmflwo I-‘OJU'IOU'ILD Howbaaw Hoowmmaoqoaon NONI-5H4) FHDBJH 55 Figure l4.--Length-frequency distribution of two species of sculpin in the Jordan and AuSable rivers for August 1970. Roman numerals designate age-groups. Fish were grouped by 2 mm inter- vals of total length. Number of Fish 60 56 PSL/MY SOUL P/N - UPPER JORDAN I III . III 1 20 30 40 50 60 70 80 90 I00 40 20 ,3L/MY SCULPIN - LOWER JORDAN b llilw 4O 20 20 30 4O 50 60 7O 80 90 IOO MOTTLED SOUL P/N - LOWER JORDAN I I III I l l l 4O 20 11.11.111.11... 20 30 40 50 60 7O 80 90 I00 IIO l20 MOTTLED SCULPIN - AU SABLE I' I II - I I I 1 W 20 30 40 50 60 70 80 90 I00 IIO Total LengthImm.) 57 lengths for these age-groups agree closely with the modal peaks. It was assumed that a similar reliability extends to age determinations for the remaining three sculpin populations. Ludwig and Norden (1969) found wide ranges of lengths within age-groups of mottled sculpin in Wiscon- sin and length-frequency histograms of 899 mottled sculpin in the West Gallatin River, Montana did not show clearly defined modes for fish two years old and older (Bailey, 1952). This indicates that sculpin do not grow signifi- cantly in length after the third year. Calculated length at annulus formation The lengths at annulus formation of each age-group for the slimy sculpin populations are given in Tables 12 and 13, and for the mottled sculpin populations in Tables 14 and 15. The weighted mean lengths at annulus formation calculated for each population are presented in Table 16. Student's t test was used to compare the calculated lengths between pOpulations of each species at each of the first three annuli (Mendenhall, 1967). The results of the six "t" tests indicated that a significant difference (P=.05) existed between the calculated lengths of the two populations of slimy sculpin and between the two popu- lations of mottled sculpin, at the first three annuli. The calculated lengths of slimy sculpin in Valley Creek, Minnesota collected during January, 1970 were close to the 58 v.~H m.mH H.mm N.mm ucoEoHocH Amvm.am Aamvm.oh Ahmavo.mm Avvmvm.mm momma wounmflmz m.vm m.Hm m.m> m.am Amy m.mm >H m.mm 0.05 H.mm Ammv m.vm HHH m.an H.mm Aooava.mm HH H.mm Anomvo.mm H ousummo v m N H pm as Qsoum mod amouo>< AEEV cofluoEnom msasccm um numama poumasoamu .mNHm mameom ouooflpcfl momonucoumm ca mquEsz .uo>wm common momma may Eoum camasom mafiam How mom comm um summed Hmuoetl.ma magma w.hN o.HN H.mN H.mv #cmEmHOGH 59 Aavo.moa Abvo.om Aamvm.on Ammva.mv mcmoe wounmfimz o.moa o.moa v.55 n.mm Adv >.mm >H m.om H.mm m.on va o.mv HHH H.mm o.Hn Awav~.nv HH o.mm Aavvo.mv H musumoo m v m N A no AB msoum 0mm ommuo>¢ AEEV COADMEHOM msasccm um numcoa popmasoamo .mnwm mHmEMm manuaocfi mononucmuom CH muonfisz .um>flm common umBoH onu Eoum Gamasom weflam How mom come no numcma Hmuoell.ma magma 60 m.mH m.Hm m.mm ucoEoHocH Amvm.nm Aomvm.mm Aommvv.Hn AOHmvm.mm mcooE nounmflmz m.HHH m.nm m.on m.mm Amv o.mm >H m.mOH m.mm m.on Aomv m.mm HHH m.mm m.Hn Avmmvn.mm HH v.5o Aommvv.ov H ousuamo m e m m H um AB msoum mod ommuo>< AEEV :oHuoEHOH msHscco um SumcoH woDMHsoHoo .oNHm onEmm wumoHp:H mononucmumm CH mumnEdz .Ho>Hm cmpHOh HoBOH may Eoum GHmHsom UmHuDOE MOM mom some um nbmcmH Hmuoall.va oHnme 61 n.mN ¢.mm m.mv ucmEouocH Ava.NOH Amva.NOH Ammvm.Nm Ammva.mv mcmoE emosmflmz o.mNH m.NOH o.vm m.om AHV m.Hm >H m.hOH o.m0H o.nh AVHV N.Nv HHH 0.5m n.vw Avvv o.mw HH m.Nh Avavm.mw H ousummo m w m N H um QB msoum mod mmmuo>< AEEV coHumEHOH msHsccm um npmcoH UoDMHsono .oNHm mHmEom mumOHch mononucoumm CH mHoQEdz .uo>Hm oHQmmsfi on» Eoum GHmHsom pmHuuoe How omm comm mo npmcmH Hmuoall.mH oHQme 62 m.~on mm.mnm.mon mm.nnm.~m he.onm.me mnnmms< unansom omnuuoz m.km eo.mnm.mm me.ofle.nn e~.onm.mm smonon nmzoa unansom nmnuuoz o.mon hm.mno.oa oo.n«m.on mm.onn.mv cannon nozoq unansom nannm m.nm nm.nnm.me ne.ono.mm ne.on~.mm cmcnon noon: unansom nannm e m N n coHnmEnOM mchccm um camcoH pouchonu ounm nosnm monoomm .Amm H My mnm>nn anmmc< 6cm cannon may cH chHcom mo mcoHUMHcmom n50m now coHqunOm mcHsccm um mcumcmH coma UmumHDOHmout.mH mHnme 63 lengths calculated at the first four annuli of slimy sculpin in the lower Jordan River (C. E. Petrosky, unpub- lished data). Detailed descriptions of sculpin growth were not found in the literature, but a February length- frequency histogram for mottled sculpin in a Utah stream yielded average lengths for age-groups 1 through 4 which twere similar to the growth of lower Jordan mottled sculpin (Zarbock, 1952). In this study, the faster growing populations in each species were located in the more eutrophic stream site, and in the lower Jordan River where the two sculpin species co-existed, the mottled sculpin grew faster than did the slimy sculpin. Instantaneous growth rate Instantaneous growth rates of the type described previously for brown trout were calculated for each scul- pin population (Figure 15). The growth rate of the 1970 year class was based on an estimated initial length of 7 mm. This is approximately the length reported by Ludwig and Norden (1969) and Bailey (1952) for mottled sculpin fry at the time they leave the nest. Each sculpin population exhibited a growth rate which was maximum in the first year, and which decreased sharply in later years. In each species, the population of sculpin which contained the most rapidly growing young was located in the more eutrophic stream. 64 .oan n0m mno>nn mHQamcm nca cannon ocu cn cancom mo monoodm 03» mo naom coao now cn3onm mo ouan mcoocaucaumcHll.mH oncmnm 65 q «I q 033 :< 9.10 .322. .23.. 42:0 \<\Q. “Qua. QM Nu k Q§ 522. .033 I4 522. Loan: I ‘50. V93... x ‘3 NW n L L w 1 l CNN O¢ .N omN elm qmom snoauolumsul 66 Length-weight relationship The length-weight regressions of the faster grow- ing populations in both sculpin species, had a much lower slope than did the slower growing populations, indicating that at a given length, the more rapidly growing popu- lations exhibited a greater weight than did the slower growing ones (Table 17 and Figure 16). The relationship of length to weight for mottled sculpin in a Montana stream was determined to be log W = -4.798 + 3.161 log TL (Bailey, 1952). This is similar to length-weight regressions for mottled sculpin in the AuSable River and slimy sculpin in the Lower Jordan River. The weight of sculpin at annulus formation was determined from the length-weight relationships given in Table 18. These calculated weights are only approximate estimates since the sculpin were weighed to the nearest gram, a coarse unit of measure for an organism of this size. Population density Population estimates of the four sculpin popu- lations were only partially reliable. There were an esti- mated 4,271 slimy sculpin in the upper Jordan River study site (SE 1 652) which represents a density of 85,045 fish per hectare. The combined population of both sculpin species in the lower Jordan River study site was estimated to be 4,787 (SE : 749). The relative abundance ratio of 67 an men mne~.m + nemm.m- u 3 men annmmsa unansom amnnnoz as man memm.~ + emmv.m- u 2 mon cannon nmzoq unansom amnuuoz dB moH mHoo.m + Hmvm.vl u 3 moH cannon nm3oq cancom AEHHm AB moH «mnH.N + OHmm.ml n z moH cannon nwmmb cancom hEHHm cOHmmwnmmn ncmnozlcnmcoq ounm ancum monoomm .mnm>nn oHQamcm nca cannon on» cH cHQHcom mo monoomm o3u mo mmncmcoHuaHon ncmnozlzumcoQIl.bH mHQaB 68 .mno>nn oHQamca nca cannon ocu Eonm cancom mo monoomm 03¢ n0m camcoH caoE umcnama ucmnos caoE no uon OHEcunnamoq .mncmconpaHmn ucmnoslcumcoqll.mH oncmnm 69 00. \. \\ \\ ‘ \\ .\ \ \ ‘\ .\\\ \\ \ O. A 3 2243 52:8 3:3: 635 3. II. I £23m 3.202 :33: .23.. tttttt £23m 56.5 :33... .23.. II £23m a:.:...... :33: Loan: O. 00. 000. (WITH) ulbua'l IDTOJ. 70 on an m n man onnomcc cncncoo nonnnoz .. m n n onm cannon nozoc cnoncom nonnnoz nn on m n no cannon nozoc cnonsom manna n m N Hv nnm cannon nommo chHcom NEHHm n m N H z ouHm mncum monoomm coHuaEnom mchcca pa ucmHm3 nonaHdoHaU .mnm>Hn oHnamcd nca cannon as» cH chHcom mo moHoomm 03» nOM coHuaEnow mchcca ua AEmv pcmHm3 nonaHDOHaUII.mH oHnaB 71 mottled sculpin to slimy sculpin in this site was 10:1 in the total collection. The same ratio was used to apportion population estimates to the species. This resulted in an estimated density of 6,758 and 589 fish per hectare for mottled sculpin and slimy sculpin respectively. There were an estimated 4,608 mottled sculpin in the AuSable River site (SE 1 2,488); a density of 5,915 sculpin per hectare. While this estimate is of low reliability, it does indicate that the density of mottled sculpin is lower in the AuSable River, than in the lower Jordan River. This same density relationship was observed between the two sites during the sampling operations at the study sites. The slower growing populations of each species, the upper Jordan slimy sculpin and the lower Jordan mottled sculpin, exhibited the highest population densities for their species. Condition The sculpin on the three study sites, appeared to reach peak spawning activity in early May. Hann (1927) reported that mottled sculpin in Michigan spawn during late April and Ludwig and Norden (1969) found that in Wisconsin, this species spawned during the entire month of April with a peak activity near the middle of the month. A significant decline in condition occurred in each sculpin population during this period (Figures 17 and 18). Spawning does not appear to be the only possible 72 .caoE man no mnHEHH ooconnmcoo wmm ucomonmon mocHH HaoHnno> .oan mcann mno>Hn oHnamcm nca cannon mcu cH chHcom mcHHnaom mo noncomm can no Amy coHancoo came cH momcacolt.nH oncmHm 73 5.62 -O -O -< .5 -< -o -< d'? -< no do 14 -5 -< -O -O ‘4 5 4 .w whim. \Vv. EVQEQB tthq EYQQQS kwkl§ z_n_.50m own—PHD: ZETSUm >275 ‘0. o 0. '9. o. N ON 10:00:) uogtgpuoo 74 .camE man no muHEnH moconHmcoo wmm ucommnmon moan Hao IHuno> .oan mcann mno>Hn oHnamc¢ nca cannon ocn cH chHcom nHOTnaomlo3u mo monoomm 03“ mo AMV cenancoo caoE cH momcacUll.mH oncmnm 75 10 40 -T< -fi 44 -O 40 44 .3 < O -O -l< 1... d4 -0 -O -< -3 ‘< 1. N .anG. \VV. >2\th\. kmst.‘ \< YQQQS QMQQQ ZETSUm om..._.._.o_2 2.1.50m >215 0.0 '0. 0. 0. N 101303 uoglgpuoo 76 cause of a reduced condition since the same trends are apparent in yearling sculpin which did not spawn in that year. The peak condition occurs early in the spring, with another minor peak in the fall. Since sculpin were not collected in the winter, it cannot be ascertained if there are two peaks in condition annually or if sculpin maintain a high condition throughout the winter. Zarbock (1952) reported that condition of mottled sculpin increased with increases in length up to 55 mm standard length, after which condition decreased, and that these Ksl values were near 3.0 which is much higher than was calculated for sculpin in this study. The condition of age-group l sculpin was higher than that of age-group 2 during April, but two-year-old sculpin exhibited higher condition than yearlings after the spawning season. DISCUSSION The growth of brown trout has been related to a multitude of physiological and environmental factors, including, the natural fertility of the water (McFadden and Cooper, 1962), the quality and quantity of food avail- able (Ellis and Gowing, 1957), temperature (Purkett, 1951 and Swift, 1955, 1961), respiration (Fry, 1968), pH of the water (Frost, 1945), competition (McFadden and Cooper, 1962), and the presence of helminth parasites on the fish (Thomas, 1964). It is likely that these internal and ex- ternal factors act collectively upon the fish's genetic potential for growth. It has been the aim of this study to relate the growth of three fish species to the early stages of stream eutrophication. A river does not become eutrophic in the same sense that a lake does, but the term "eutrophic" can be applied to moving water to indicate the constant addi— tions of nutrients to the water and the resultant changes in the aquatic ecosystem. The most basic of these changes Would be an increase in primary production which in turn would create additional food and cover for many consumer species in the summer months. The increased primary 77 78 production would also alter the level of dissolved oxygen and increase diurnal dissolved oxygen fluctuations. The increase in primary production would be reflected in cor- responding qualitative and quantitative increases at all consumer trophic levels. These changes alone could account for a greater growth of fish in the more eutroPhic environ- ments by increasing the available food for the fish. At the same time, successive stages of eutrophication appear to be related to increases in the number of fish species which intensifies competition for food and predation. Coupled with competition for nesting sites, this could account for a reduction in pOpulation numbers of a given species. At some higher level of eutrophication one would expect to find mature fish growing at a maximum rate, coupled with a minimal population size and further in- creases in eutrophication would result in a lack of spawn- ing success and the species would be removed from the community. Water temperatures differed between the study sites and may have been responsible for some of the differ— ences in fish growth. Brown (1957) and Swift (1961) have shown that there are temperature optima for which brown trout growth is maximized and these optimal temperatures were found to exist in the more eutrophic sites. The three stream sites in this study were assumed to represent the early stages of stream eutrophication 79 where the upper Jordan, lower Jordan, and AuSable rivers represented pristine, minimally perturbed and moderately perturbed stream conditions respectively. Brown trout grew slower in the upper Jordan River than in either of the other two study sites. This was coupled with a high population density, due primarily to an abundance of young trout. Although the biomass of the upper Jordan brown trout population approached that of the brown trout in the AuSable River, the most eutrophic site, the average weight per individual was low as a re- sult of the preponderance of young fish in the population. The mechanisms responsible for this growth and density pattern were not identified, but were probably a combination of a paucity of available food coupled with a high rate of recruitment of young fish. It is possible that larger, faster growing brown trout had been produced in the upper Jordan River, and that they migrated out of the study area due to a lack of suitable food, cover, or inappropriate water temperatures but this is unlikely since the density of mature brown trout in the upper Jordan River far exceeded that of brown trout in the other study sites. It was expected that the growth of brown trout would be greater in the AuSable River than in the lower Jordan River, yet this relationship did not hold for mature brown trout. One cause for this seeming dis- crepancy has been given as the occurrence of "Lee's 80 phenomenon" in length calculations. It is also possible that calculated lengths of mature brown trout in the lower Jordan River were biased upwards by the inclusion of data from brown trout which had migrated into the study site from Lake Michigan. In addition, an extensive sports fishery exists on the AuSable River which could have been responsible for the removal of the faster growing mature trout from the p0pu1ation, resulting in a downward bias to calculated lengths of brown trout in that river. When the growth curves of 30 brown trout popu- lations in Europe and the United States were compared, two distinct growth patterns were evident. One growth pattern was characterized by an attained total length of 99 mm or more in the first year with relatively linear increases in length in the later years. Brown trout in the AuSable and lower Jordan rivers were representative of this type of growth. The second growth pattern was characterized by an attained length of less than 99 mm in the first year with length increments in later years which resulted in a curvilinear growth curve. The growth of brown trout in the upper Jordan River was of this latter type. The rate of growth of underyearlings was the only factor which consistently differed between p0pu1ations in the two growth types, and it is possible that the general pattern of growth throughout life is genetically predetermined to be one of the two types shown, and that the expression 81 of one growth pattern over the other depends on the rate of growth of underyearlings. The response of the sculpin populations to changes in stream eutrophication was similar to that of the brown trout. Slimy sculpin occurred in large numbers in the upper Jordan River but with restricted growth, while in the lower Jordan River this species displayed more rapid growth with an apparent corresponding decrease in density. Mottled sculpin responded in a similar manner to changes in stream eutrophication between the lower Jordan and AuSable rivers. The condition of yearling brown trout was con- sistently higher than that of two-year-old trout in each stream. Thomas (1964) suggested that condition factors could be used as an index of growth when p0pu1ations of the same age at the same season were compared, a high condition being indicative of slow growth. This relation- ship held for yearling brown trout only, while two-year- old trout in all three river sites exhibited approximately the same condition. The average condition of sculpin was higher for populations in the more eutrophic sites, than it was for populations in the less eutrophic ones. In all three fish species studied, there was a sharp decline in condition during the Spawning season which seemed at least partially due to genetic factors since it occurred in both mature and immature individuals. 82 It appears that the decline in condition was not entirely related to season since the brown trout and sculpin both exhibited a decline in condition during the spawning season although these species spawned at different times of the year. In each of the three fish species studied, the instantaneous rate of growth of the fish in the first year was directly related to the level of eutrophication in the stream. That this relationship did not hold for mature brown trout suggests that immature fish are better indi- cators of stream eutrophication in its early stages than are the mature fish. Frost (1945), Ball and Jones (1960), and Thomas (1964) have found that the growth of brown trout in later years was directly related to the rate of growth as under- yearlings. If this is correct and if the growth of young fish is augmented by increases in stream eutrophication, as was indicated by this study, then the year-to-year changes in the rate of growth of young fish may be a valuable index to changes in stream eutrophication. LITERATURE CITED Bailey, Jack E. 1952. Life history and ecology of the sculpin Cottus baridi punctulatus in southwestern Montana. Copeia. 4:243-255. Bailey, Reeve M. (chairman). 1960. A list of common and scientific names of fishes from the United States and Canada. 3rd. ed. Amer. Fish. Soc. Special publ. No. 6. 150 p. Ball, J. N., and J. W. Jones. 1960. On the growth of the brown trout on Llyn Tegid. Zool. Soc. (London), Proc. l34(l):l-4l. Beyerle, George B., and Edwin L. Cooper. 1960. Growth of brown trout in selected Pennsylvania streams. Amer. Fish. Soc., Trans. 89(3):255-262. Bouck, G. R., and R. C. Ball. 1966. Influence of capture methods on blood characteristics and mortality in the rainbow trout (Salmo gairdneri). Amer. Fish. Soc., Trans. 95(2):l70—l76. Brown, M. E. 1946. The growth of brown trout (Salmo trutta Linn.) II. The growth of two-year-old trout at a constant temperature of 11.5C. J. Exp. Biol. 22:130-144. Brown, M. E. 1957. Experimental studies on growth, Vol. 1, p. 361-400. £2 M. E. Brown (ed.) The physiology of fishes. Academic Press, Inc., New York. Bulter, R. L., and Lloyd L. Smith, Jr. 1953. A method for cellulose acetate impressions of fish scales with a measurement of its reliability. Prog. Fish-Cult. 15(4):175-l78. Brynildson, O. M., and C. L. Brynildson. 1967. The effect of pectoral and ventral fin removal on survival and growth of wild brown trout in a Wisconsin stream. Trans. Amer. Fish. Soc. 96(3):353-355. 83 84 Carlander, K. D. 1969. Handbook of freshwater fishery biology. Vol. I. Iowa State University Press, Ames, Iowa. 752 p. COOper, Edwin L., and Norman G. Benson. 1951. The coef- ficient of condition of brook, brown and rainbow trout in the Pigeon River, Otsego County, Michigan. Prog. Fish-Cult. 13(4):181-192. Ellis, Robert J., and Howard Gowing. 1957. Relationship between food supply and condition of wild brown trout, Salmo trutta Linnaeus, in a Michigan stream. Limnol. Oceanogr. 2(4):299-308. Frost, Winifred E. 1945. River Liffey Survey, VI. Dis- cussion on the result obtained from the investi- gations on the food and growth of brown trout (Salmo trutta L.) in alkaline and acid waters. Roy. Irish Acad., Proc. (B)50:321-342. Fry, R. E. J. 1968. The aquatic respiration of fish, p. 1-79. I2 M. E. Brown (ed.) The physiology of fishes. Academic Press, Inc., New York. Hann, H. W. 1927. The history of the germ cells of Cottus bairdi Girard. J. Morphol. and Physiol. 43: 427-497. Hogman, Walter J. 1969. Documentation of a computer pro- gram for multispecies fish p0pu1ations. Center for Great Lakes Studies. Spec. Report No. 9. University of Wisconsin-Milwaukee. Milwaukee, Wisconsin. Hunt, R. L. 1966. Effect of habitat alteration on pro- duction, standing crOps and yield of brook trout in Lawrence Creek, Wisconsin, p. 281-312. £2 T. G. Northcote (ed.) Symposium on salmon and trout in streams. University of British Columbia. Jenkins, T. M., Jr. 1969. Social structure, position choice and microdistribution of two trout species (Salmo trutta and Salmo gairdneri) resident in mountain streams. Anim. Behav. Mono. 2(2):57-123. Kathrein, Joseph W. 1951. Growth rate of four species of fish in a section of the Missouri River between Holter Dam and Cascade, Montana. Amer. Fish. Soc., Trans. 80:93-98. Kramer, Clyde Young. 1956. Extension of multiple range test to group means with unequal numbers of repli- cations. Biometrics. 12:307-310. 85 Lagler, Karl F. 1956. Freshwater fishery biology. 2nd ed. Wm. C. Brown Co., Dubuque, Iowa. 421 p. LeCren, E. D. 1969. Estimates of fish populations and production in small streams in England, p. 269- 280. IE T. G. Northcote (ed.) Symposium on salmon and trout in streams. University of British Columbia. Lee, Rosa M. 1912. An investigation into the methods of growth determination in fishes. Cons. Perm. Int. pour l'Epler. de la Mer. Publ. de Circ. 63:34 p. Ludwig, Gerald M., and Carroll R. Norden. 1969. Age, growth and reproduction of the northern mottled sculpin (Cottus bairdi bairdi) in Mt. Vernon Creek, Wisconsin. Milwaukee Public Mus. Occas- sional Papers No. 2. McFadden, James T., and Edwin L. Cooper. 1962. An ecological comparison of six populations of brown trout (Salmo trutta). Amer. Fish. Soc., Trans. 91(1):53-62. Mendenhall, William. 1967. Introduction to probability and statistics. 2nd ed. Wadsworth Publishing Co., Inc., Belmont, California. 393 p. Mense, James B. 1970. Relation of density to brown trout movement in a Michigan stream. (Unpublished Ph.D. thesis, Michigan State Univ. Library, East Lansing, Michigan.) Purkett, Charles A., Jr. 1951. Growth rate of trout in relation to elevation and temperature. Amer. Fish. Soc., Trans. 80:251-259. Ricker, W. E. 1958. Handbook of computations for bio- logical statistics of fish populations. Fish. Res. Board Can. Bull. No. 119. 300 p. Robson, D. S., and H. A. Regier. 1968. Estimation of population number and mortality rates, p. 124-158. £2 W. E. Ricker (ed.) Methods for assessment of fish production in fresh waters. International Programme Handbook No. 3. Blackwell Scientific Publications, Oxford and Edinburgh. Swift, D. R. 1955. Seasonal variations in the growth rate, thyroid gland activity and food reserves of brown trout. J. Exp. Biol. 32(4):751-764. 86 Swift, D. R. 1961. The annual growth-rate cycle in brown trout (Salmo trutta Linn.) and its cause. J. Exp. Biol. 38:595-604. Tesch, F. W. 1968. Age and growth, p. 93-123. In W. E. Ricker (ed.) Methods for assessment of fiEH pro- duction in fresh waters. International Biological Programme Handbook No. 3. Blackwell Scientific Publications, Oxford and Edinburgh. Thomas, J. D. 1964. Studies on the growth of trout, Salmo trutta from four contrasting habitats. Zool. Soc. TLondon), Proc. 142(3):459-509. Zarbock, William M. 1952. Life history of the Utah scul- pin, Cottus biardi semiscaber (Cope), in Logan River, Utah. Amer. Fish. Soc., Trans. 81:249-259.