Temperature and Maternal Effects on Hatch Rate and Length at … · 2012. 9. 28. · North American...

This article was downloaded by: [Department Of Fisheries]On: 25 September 2012, At: 23:24Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

North American Journal of AquaculturePublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/unaj20

Temperature and Maternal Effects on Hatch Rate andLength at Hatch of Hybrid Bass (White Bass × StripedBass) LarvaeS. E. Lochmann a , K. J. Goodwin a & C. L. Racey ba Aquaculture/Fisheries Center, University of Arkansas at Pine Bluff, 1200 North UniversityDrive, Mail Slot 4912, Pine Bluff, Arkansas, 71601, USAb Arkansas Game and Fish Commission, 2 Natural Resources Drive, Little Rock, Arkansas,72205, USA

Version of record first published: 24 May 2012.

To cite this article: S. E. Lochmann, K. J. Goodwin & C. L. Racey (2012): Temperature and Maternal Effects on Hatch Rateand Length at Hatch of Hybrid Bass (White Bass × Striped Bass) Larvae, North American Journal of Aquaculture, 74:3, 283-288

To link to this article: http://dx.doi.org/10.1080/15222055.2012.672372

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form toanyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses shouldbe independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims,proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly inconnection with or arising out of the use of this material.

http://www.tandfonline.com/loi/unaj20http://dx.doi.org/10.1080/15222055.2012.672372http://www.tandfonline.com/page/terms-and-conditions

North American Journal of Aquaculture 74:283–288, 2012C© American Fisheries Society 2012ISSN: 1522-2055 print / 1548-8454 onlineDOI: 10.1080/15222055.2012.672372

ARTICLE

Temperature and Maternal Effects on Hatch Rate and Lengthat Hatch of Hybrid Bass (White Bass × Striped Bass) LarvaeS. E. Lochmann* and K. J. GoodwinAquaculture/Fisheries Center, University of Arkansas at Pine Bluff, 1200 North University Drive,Mail Slot 4912, Pine Bluff, Arkansas 71601, USA

C. L. RaceyArkansas Game and Fish Commission, 2 Natural Resources Drive, Little Rock, Arkansas 72205, USA

AbstractBroodstock selection is an approach that could improve aspects of hybrid bass (white bass Morone chrysops ×

striped bass M. saxatilis) production. We examined incubation temperature and maternal effects on modal period ofincubation (h), hatch rate, and length at hatch. Gametes from three to four white bass dams and one striped bass sirewere used to produce hybrids each week for 4 weeks. Eggs from each dam were incubated at 14, 16, 18, and 20◦C.The period of incubation, hatch rate, and length at hatch were determined for larvae from each dam × temperaturecombination. The relationship between incubation period and incubation temperature was exponential, and warmertemperatures resulted in progressively shorter incubation periods and less-variable hatching duration (i.e., the periodfrom first hatch to last hatch). The effects of incubation temperature and dam explained approximately equal amountsof the variability in length at hatch. Dam weight did not explain a significant amount of variability in modal incubationperiod, hatch rate, or average length at hatch. Selection for increases in length at hatch could offer improvementsto the hybrid bass industry by improving survival in ponds or by eliminating the need for rotifers at first feedingin tanks. The benefits of producing larvae that are larger at hatch should be compared with the effort and cost of abreeding program that selects for length at hatch.

The global seafood market is worth more than US$170 ×109 annually, but the USA is experiencing a $9 × 109 annualseafood trade deficit (FAO 2009). An increase in U.S. aquacul-ture production will be necessary to reverse this trade imbal-ance. The industry producing hybrid bass (white bass Moronechrysops × striped bass M. saxatilis) is an important compo-nent of U.S. aquaculture production. The striped bass and itshybrids ranked fifth in finfish production in the most recentU.S. aquaculture survey conducted by the National AgriculturalStatistics Service (NASS 2006). However, production in 2009fell below 4 million kg for the first time since 2000 (Turano2010). One possible reason for the decline in production is thesteady 2.7% increase in production costs experienced by thehybrid bass industry since 1997 (Turano 2010). Improvements

*Corresponding author: [email protected] April 4, 2011; accepted July 11, 2011Published online May 24, 2012

in production or reductions in production costs are crucial todeveloping the hybrid bass industry.

Broodstock selection is an approach that could improve as-pects of hybrid bass production. Garber and Sullivan (2006)outlined the potential benefits of selective breeding for the hy-brid bass industry. Categories of traits that could be selectedfor improvement include growth, health, morphology, and re-production. The benefits of fast growth or disease resistance ina culture setting are obvious. Larval length at hatch is anothertrait that could affect culture.

Larger larvae would have a wider prey size spectrum at firstfeeding, and they would also be better able to avoid predationin earthen ponds (Paradis et al. 1996). Bosworth et al. (1997)examined traits of white bass, striped bass, and hybrid eggs and

283

Dow

nloa

ded

by [

Dep

artm

ent O

f Fi

sher

ies]

at 2

3:24

25

Sept

embe

r 20

12

284 LOCHMANN ET AL.

larvae in a 3 × 3 diallelic cross study. Striped bass larvae werelarger at hatch than white bass, and hybrids were intermediatein length at hatch (Bosworth et al. 1997). Furthermore, therewas a significant positive striped bass maternal effect (relativeto white bass maternal effects) on length at 12 h posthatchand 7 d posthatch (dph) and on the percentage of larvae thatconsumed brine shrimp Artemia spp. at 7 dph (supporting thelink between length at hatch and feeding ability). Lochmannet al. (2009) demonstrated a maternal effect on length at hatchof hybrid bass larvae but were unable to determine whether theeffect was phenotypic or genotypic. There is evidence that somelarval traits have at least a partial phenotypic basis. Larvae fromstriped bass dams less than 4.5 kg had smaller mouths at firstfeeding than did larvae from larger dams (Zastrow et al. 1989).Nevertheless, these studies suggest length at hatch might beincreased through a selective breeding program.

In the Lochmann et al. (2009) study, there was also asignificant effect of the incubation period (h) on length at hatch.Larvae incubated at cooler temperatures hatched later and werelarger at hatch but had smaller yolk and oil volumes at hatch.This observation suggests that length at hatch or length atyolk absorption can be increased through the manipulation ofincubation temperature. Morgan et al. (1981) demonstrated thatstriped bass larvae from eggs incubated at 22–24◦C were longerat 24 h posthatch than larvae from eggs incubated at coolertemperatures. Conversely, Peterson et al. (1996) found thatstriped bass larvae were shorter at initial feeding when rearedat 18◦C than when reared at 14–16◦C. Reducing the incubationtemperature and increasing the incubation period to manipulatelength at hatch could negatively affect hatch rate. We wanted toexamine the importance of incubation temperature for influenc-ing hatch rate and length at hatch. We also wanted to comparethe incubation temperature effect versus the maternal effect onvariability in length at hatch. The objective of this study was toquantify the variation in hybrid bass hatch rates and length athatch among incubation temperatures and among dams.

METHODSWhite bass eggs and striped bass milt were used to produce

hybrids at Keo Fish Farm, Keo, Arkansas, in spring 2005. Whitebass dams came from the Arkansas River and were a mix of (1)wild fish that were spawned in 2004, held in ponds for a year,and spawned again in 2005 (dams A–F; weeks 1 and 2); and(2) wild fish that were caught early in 2005 and held for 1–3months before spawning (dams G–N; weeks 3 and 4). Stripedbass were the farm-raised progeny of fish from Lake Ouachita,Arkansas. Captive fish were fed an ad libitum ration of fatheadminnow Pimephales promelas and golden shiners Notemigonuscrysoleucas. Gametes from three to four white bass dams werecrossed with gametes from one striped bass sire each week for 4weeks to produce hybrids. Different striped bass sires and whitebass dams were used each week.

Final maturation and ovulation were induced with injectionsof human chorionic gonadotropin (Rees and Harrell 1990).

White bass were examined periodically before the predictedovulation time. Ripe white bass dams were weighed and thenstrip-spawned into individual plastic containers. The weight ofdams ranged from 0.4 to 1.3 kg during the study. Milt fromone striped bass and a small amount of water were added to thecontainer, and the gametes were thoroughly mixed. Eggs wereplaced in McDonald hatching jars and treated with a 150-mg/Lsolution of tannic acid for 10 min (Hodson and Hayes 1989) toreduce adhesion and clumping. After the tannic acid treatment,eggs were rinsed with well water for 30 min. Portable aeratorswere used to keep the eggs in suspension during transportation tothe Aquaculture/Fisheries Center at the University of Arkansas,Pine Bluff.

At the Aquaculture/Fisheries Center, eggs were incubated inrecirculation systems (Marine Biotech, Boston, Massachusetts).Four separate recirculation systems each had four 63-L acrylicaquaria; each aquarium contained a single 6.8-L McDonaldhatching jar and a larval collector sieve (300-µm mesh). Eachrecirculation system had an activated carbon filter, a cartridge fil-ter, a biofilter, and an ultraviolet sterilizer. Water flowed throughthe filters, into a hatching jar, into a larval collector, into anaquarium, and back to the filters. Water in the four recirculationsystems was maintained at 14, 16, 18, and 20◦C. The num-ber of eggs in four 1-mL samples was determined by manualcounting (mean ± SD = 1,623 ± 31 eggs/mL). Each damwas represented once at each incubation temperature by plac-ing 4,000 eggs/dam (number determined volumetrically) intohatching jars in each recirculation system.

Eggs or larvae were checked every 6 h until hatching wasobserved and every 3 h thereafter. When larvae were observedin a larval collector, the incubation period (i.e., the length oftime [h] from fertilization to appearance in the collector) wasrecorded for those larvae. The collectors were emptied every 3h after initiation of hatching until hatching had ceased. Larvaewere removed from the collector, euthanized with tricainemethanesulfonate (MS-222; FINQUEL, Argent Laboratories,Redmond, Washington), and photographed with a Spot InsightColor digital camera (Model 3.2.0; Diagnostic Instruments,Sterling Heights, Michigan) that was mounted on a dissectingmicroscope and interfaced with a computer. Five to fifteen larvaefrom each dam × temperature combination were photographedduring each 3-h period until approximately 30–50 larvae fromeach dam × temperature combination had been photographed.Unphotographed larvae from each 3-h period were preservedand later enumerated to determine the modal incubation periodand hatch rates (i.e., percent of eggs hatching) for each dam ateach incubation temperature. Standard length and finfold widthwere measured for each larva with Image-Pro Plus softwareversion 4.5.1.22 (Media Cybernetics, Silver Spring, Maryland).

Temperature was measured with a mercury thermometer ev-ery 6 h during incubation. Daily water samples were taken fromeach recirculation system. Dissolved oxygen (DO) was mea-sured with a YSI Model 52 oxygen meter, and pH was measuredwith an Orion Model 290A meter. Hardness was measured with

Dow

nloa

ded

by [

Dep

artm

ent O

f Fi

sher

ies]

at 2

3:24

25

Sept

embe

r 20

12

TEMPERATURE AND MATERNAL EFFECTS ON HYBRID BASS LARVAE 285

Hach reagents (Hach, Loveland, Colorado), and total ammonianitrogen (TAN) was determined by means of the Nessler methodand a Hach DR/2000 spectrophotometer. We report only theun-ionized ammonia based on the formula for the un-ionizedfraction of TAN (Emerson et al. 1975).

We determined whether modal incubation period and hatchrate were related to incubation temperature. We used a factorialanalysis of variance (ANOVA; GLM procedure in the Statis-tical Analysis System [SAS]; SAS Institute 1990) to analyzethe hatch data. Time (i.e., week of the study) and temperaturewere fixed effects, and dam weight was a covariate (i.e., ran-dom effect). Modal incubation period and hatch rate were theresponse variables. The effects of time and temperature weretested against the mean square of the time × dam weightinteraction term and temperature × dam weight interactionterm, respectively (Zar 1999). Hatch rates were arcsine–square-root transformed to normalize and homogenize the residuals.The relationship between incubation temperature and incuba-tion period for all hatched larvae was examined with linear andexponential least-squares regression models (REG and NLINprocedures in SAS).

We examined the effect of temperature and dam on length athatch. Length-at-hatch data were examined with a mixed-effectsnested ANOVA (MIXED procedure in SAS). The followingmodel was used to analyze length at hatch:

Yijkl = µ + TIMEi + TEMPj + (TIME × TEMP)ij+ DAM(TIME)ki + εijkl,

where Yijkl is an individual observation of length at hatch, µ isthe overall mean, TIMEi is the fixed effect of week i, TEMPj

is the fixed effect of temperature j, (TIME × TEMP)ij is theinteraction among time and temperature groups, DAM(TIME)kiis the random (genetic) effect of dam k nested in week i, andεijkl is the random residual (Bosworth et al. 1997; Wang et al.2006).

The effects of time, temperature, and the time × temper-ature interaction were tested against the mean square of theDAM(TIME)ki term (Zar 1999). To examine the effect of damweight on length at hatch, we determined the average length athatch of larvae from each dam × temperature combination. Weused a factorial ANOVA (GLM procedure in SAS) in which timeand temperature were fixed effects, dam weight was a covari-ate, and average length at hatch was the response variable. Asbefore, the effects of time and temperature were tested againstthe mean square of the time × dam weight interaction term andtemperature × dam weight interaction term, respectively. Anα level of 0.05 was used for all statistical tests.

RESULTSWater quality was similar between temperature treatments

(i.e., recirculation systems) and between weeks of the study.Mean DO was consistently at or above 7.6 mg/L (Table 1). ThepH of each treatment averaged 7.2–7.5 during the study. Waterhardness averaged 188 mg/L or higher in each treatment dur-ing each week. The un-ionized portion of TAN averaged lessthan 0.008 mg/L throughout the study. The maximum observedun-ionized fraction of TAN was 0.02 mg/L. Mean tempera-tures were always within a few tenths of a degree of targettemperatures and varied little within a treatment during anyweek.

TABLE 1. Average ( ± SD) water quality measures in different treatments (i.e., recirculation systems) during each week of the study to determine temperatureand maternal effects on the incubation period, hatch rate, and length at hatch of hybrid bass (female white bass × male striped bass).

WeekTreatment (nominal

temperature, ◦C)Dissolved

oxygen (mg/L) pH Water hardness (mg/L) Actual temperature (◦C)

1 14 8.0 (0.7) 7.4 (0.1) 202 (131) 14.3 (0.5)16 8.0 (0.6) 7.4 (0.3) 197 (126) 16.1 (0.3)18 7.9 (0.4) 7.4 (0.1) 194 (124) 18.1 (0.2)20 7.7 (0.4) 7.5 (0.1) 188 (126) 19.9 (0.2)

2 14 8.8 (0.3) 7.4 (0.2) 265 (13) 14.3 (0.5)16 8.7 (0.4) 7.4 (0.1) 295 (18) 15.7 (0.3)18 8.2 (0.4) 7.4 (0.1) 261 (15) 18.2 (0.3)20 7.6 (0.4) 7.4 (0.1) 269 (8) 20.2 (0.3)

3 14 8.9 (0.6) 7.4 (0.1) 264 (9) 14.3 (0.4)16 8.6 (0.5) 7.3 (0.2) 271 (7) 15.9 (0.4)18 8.1 (0.5) 7.5 (0.1) 261 (8) 17.8 (0.3)20 7.7 (0.5) 7.4 (0.1) 269 (8) 20.2 (0.2)

4 14 9.3 (0.3) 7.3 (0.2) 286 (12) 13.9 (0.3)16 8.8 (0.5) 7.2 (0.2) 289 (5) 15.9 (0.2)18 8.6 (0.5) 7.4 (0.2) 277 (11) 17.8 (0.2)20 8.3 (0.5) 7.4 (0.2) 282 (9) 20.1 (0.2)

Dow

nloa

ded

by [

Dep

artm

ent O

f Fi

sher

ies]

at 2

3:24

25

Sept

embe

r 20

12

286 LOCHMANN ET AL.

FIGURE 1. Relationship between incubation temperature and modal incuba-tion period (h) of hybrid bass (female white bass × male striped bass; solidline). The formula represents the model that provided the best fit to the datafrom this study. The dashed line represents the relationship reported by Reesand Harrell (1990).

The average hatching duration (period from first hatch tolast hatch) ranged from 19.1 ± 7.2 h (mean ± SD) at 20◦C to58.1 ± 7.5 h at 14◦C. Modal incubation period ranged from41 to 106 h and averaged 65.5 ± 21.2 h over the course ofthe study. Modal incubation period varied significantly amongtemperatures (F = 9.86, df = 3, P = 0.046). The time effect anddam weight effect were not significant. The model explained98% of the variability in modal incubation period (i.e., ex-plained variability = 1 − sum of squares [SS]error/SSmodel). Themajority of the explained variability was due to temperature(SStemp/SSmodel = 0.90). An exponential model (Figure 1)best fit the relationship between incubation temperature andincubation period (i.e., had the highest r2 value). As incubationtemperature increased, the incubation period decreased and thevariability in the incubation period also decreased. Hatch ratesranged from 10.5% to 64.2% during the study. The effects oftime, temperature, and dam weight were not significant in thefactorial ANOVA that examined variability in hatch rate.

Individual standard length at hatch ranged from 2.42 to3.34 mm. Length at hatch varied among times (F = 6.59,df = 3, P = 0.010) and among temperatures (F = 3.79,df = 3, P = 0.047). Length at hatch was greatest at thelowest incubation temperature. The time × temperatureinteraction was not significant; however, the random effectof DAM(TIME)ki was significant (F = 24.00, df = 10, P <0.001). The model explained 33% of the variability in lengthat hatch (1 − SSerror/SSmodel = 0.33). Approximately equalamounts of the model variability were explained by temperature(SStemp/SSmodel = 0.20) and DAM(TIME)ki (SSdam[time]/SSmodel= 0.18). Least-squares means for length at hatch varied amongdams by as much as 9.2%. Average length at hatch among dam× temperature combinations ranged from 2.80 ± 0.07 to 3.18± 0.07 mm (mean ± SD). The effects of time, temperature,and dam weight were not significant in the factorial ANOVAexamining variability in average length at hatch.

DISCUSSIONThe relationship between incubation temperature and incu-

bation period for hybrid bass was different from the relationshipreported for striped bass. Rees and Harrell (1990) presented alinear relationship between incubation period and incubationtemperature (Figure 1). The exponential model provided thebest fit to our data. At a low temperature (∼14◦C), the Reesand Harrell (1990) model would underestimate the period ofincubation for hybrid bass by almost 24 h. At temperatures of19–20◦C, the Rees and Harrell (1990) model slightly overesti-mates the incubation period. When spawning and hatching ofhybrid bass are extended beyond the typical spawning season orif incubation occurs at warmer or cooler temperatures than thosetypifying the spawning season, the relationship presented hereis likely to provide a better estimate of the incubation period.

Incubation temperature did not affect hatch rate of hybridbass across the range tested in this study. The results of our studywere similar to those reported by Geist et al. (2006) for Chi-nook salmon Oncorhynchus tshawytscha. In that study, initialincubation temperatures ranged from 13.0◦C to 16.5◦C. Lowerinitial incubation temperature increased the incubation periodbut did not influence the hatch rate of Chinook salmon (Geistet al. 2006). Conversely, incubation temperature did affect thehatch rates of striped bass (Morgan et al. 1981) and white perchMorone americana (Morgan and Rasin 1982) across wide tem-perature ranges (12–28◦C and 10–24◦C, respectively). The re-lationships were quadratic, and hatch rate was optimal at about18◦C for striped bass and 14.1◦C for white perch. A wider tem-perature range in our study would probably have resulted inreduced hatch rates outside of the typical incubation tempera-ture range. There appeared to be no phenotypic effect of whitebass dam weight on the hatch rate of hybrids. Our observationsare somewhat similar to those of Secor et al. (1992), who foundno effect of dam weight on egg size or embryo survival in stripedbass.

Incubation temperature affected length at hatch of hybridbass. A significant influence of incubation temperature on lengthat hatch was previously reported for striped bass (Morgan et al.1981) and white perch (Morgan and Rasin 1982). These resultsare dissimilar to those of Geist et al. (2006), who reported no ef-fect of initial temperature on length at hatch of Chinook salmon.Lochmann et al. (2009) observed longer incubation periods andlarger larvae at hatch when hybrid bass were incubated at coolertemperatures. Morgan et al. (1981) reported the opposite ef-fect of incubation temperature on striped bass; in their study,striped bass larvae that were incubated at warmer temperaturesexhibited longer lengths at 24 h posthatch. For moronids, incu-bation temperature appears to be critical for determining lengthat hatch. Therefore, experiments to examine the efficacy of max-imizing length at hatch through a broodstock selection programshould pay particular attention to consistency of the incubationenvironment. Otherwise, the influence of differences in incuba-tion temperature might confound the interpretation of maternaleffect experiments.

Dow

nloa

ded

by [

Dep

artm

ent O

f Fi

sher

ies]

at 2

3:24

25

Sept

embe

r 20

12

TEMPERATURE AND MATERNAL EFFECTS ON HYBRID BASS LARVAE 287

Hybrid bass eggs are susceptible to saprolegniasis (fungalinfection caused by Saprolegnia spp.) and low DO levels dur-ing incubation. Longer incubation times subject the developingembryos to greater risk. Nevertheless, under the conditions ofour study, increasing the length at hatch by reducing incuba-tion temperature was not detrimental to hatch rate. Producerscould maximize length at hatch by reducing incubation temper-ature without reducing embryo survival. However, Lochmannet al. (2009) reported that yolk and oil globule volumes at hatchwere smaller in larvae with longer incubation times. Reducedyolk volumes at hatch would probably decrease the time to yolkabsorption, may result in little change in larval length at yolkabsorption, and could influence the point at which hybrid basshave functional mouths. The transition to exogenous feeding isa critical period in larval development. Hybrid bass larvae heldat 18◦C are stocked into ponds or offered live feed in tanks at4–5 dph, coincident with yolk absorption. Decreasing the dura-tion of the yolk sac stage also decreases the time available for thedevelopment of exogenous feeding behavior. These possibilitiesshould be explored before the temperature (and thus incubationperiod) is manipulated to alter length at hatch.

Generally, length at hatch of hybrid bass from our studywas similar to results reported by Bosworth et al. (1997) andLochmann et al. (2009). The effect of DAM(TIME)ki was sig-nificant, indicating a maternal effect on length at hatch. Least-squares means in this study varied slightly more than the 6.5%variation in mean length at hatch reported by Lochmann et al.(2009). High variability in mean length at hatch among dams in-creases the likelihood that a broodstock selection program couldimprove this trait. Bosworth et al. (1997) found a significant ma-ternal effect among genetic groups (striped bass, white bass, andhybrids). They reported favorable striped bass maternal effectson length at 12 h posthatch and on ability to ingest live feed.However, Bosworth et al. (1997) did not report whether the ran-dom effect of individual dam nested within the time × damgenetic group interaction was significant. Wang et al. (2006)did report a significant maternal effect of white bass dams onlength of juvenile (

288 LOCHMANN ET AL.

in a northern environment. Transactions of the American Fisheries Society138:407–415.

Garber, A. F., and C. V. Sullivan. 2006. Selective breeding for the hybrid stripedbass (Morone chrysops, Rafinesque × M. saxatilis, Walbaum) industry: sta-tus and perspectives. Aquaculture Research 37:319–338.

Geist, D. R., C. S. Abernethy, K. D. Hand, V. I. Cullinan, J. A. Chandler,and P. A. Groves. 2006. Survival, development, and growth of fall Chinooksalmon embryos, alevins, and fry exposed to variable thermal and dissolvedoxygen regimes. Transactions of the American Fisheries Society 135:1462–1477.

Hodson, R. G., and M. Hayes. 1989. Hybrid striped bass: hatchery phase. South-ern Regional Aquaculture Center, Publication 301, Stoneville, Mississippi.

Lochmann, S. E., K. J. Goodwin, C. L. Racey, and C. C. Green. 2009. Variabilityof egg characteristics among female white bass Morone chrysops and therelationship between egg volume and length at hatch of sunshine bass. NorthAmerican Journal of Aquaculture 71:147–156.

Morgan, R. P., II, and V. J. Rasin Jr. 1982. Influence of temperature and salinityon development of white perch eggs. Transactions of the American FisheriesSociety 111:396–398.

Morgan, R. P., II, V. J. Rasin Jr., and R. L. Copp. 1981. Temperature and salinityeffects on development of striped bass eggs and larvae. Transactions of theAmerican Fisheries Society 110:95–99.

NASS (National Agriculture Statistics Service). 2006. Census of aquaculture(2005). U.S. Department of Agriculture, NASS, Washington D.C.

Paradis, A. R., P. Pepin, and J. A. Brown. 1996. Vulnerability of fish eggs andlarvae to predation: review of the influence of the relative size of prey andpredator. Canadian Journal of Fisheries and Aquatic Sciences 53:1226–1235.

Peterson, R. H., D. J. Martin-Robichaud, and A. A. Berge. 1996. Influence oftemperature and salinity on length and yolk utilization of striped bass larvae.Aquaculture International 4:89–103.

Rees, R. A., and R. M. Harrell. 1990. Artificial spawning and fry production ofstriped bass and hybrids. Pages 43–72 in R. M. Harrell, H. H. Kerby, and R.M. Minton, editors. Culture and propagation of striped bass and its hybrids.Striped Bass Committee, Southern Division, American Fisheries Society,Bethesda, Maryland.

SAS Institute. 1990. SAS/STAT user’s guide, version 6, 4th edition. SAS Insti-tute, Cary, North Carolina.

Secor, D. H., J. M. Dean, T. A. Curtis, and F. W. Sessions. 1992. Effect of femalesize and propagation methods on larval production at a South Carolina stripedbass (Morone saxatilis) hatchery. Canadian Journal of Fisheries and AquaticSciences 49:1778–1787.

Turano, M. J. 2010. U.S. hybrid striped bass production: 2010 industry up-date. Striped Bass Growers Association, Pine Bluff, Arkansas. Available:www.stripedbassgrowers.org/research.htm. (October 2010).

Wang, X., K. E. Ross, E. Saillant, D. M. Gatlin, and J. R. Gold. 2006. Quan-titative genetics and heritability of growth-related traits in hybrid stripedbass (Morone chrysops [female] × Morone saxatilis [male]). Aquaculture261:535–545.

Zar, J. H. 1999. Biostatistical analysis, 4th edition. Pearson Education, UpperSaddle River, New Jersey.

Zastrow, C. E., E. D. Houde, and E. H. Saunders. 1989. Quality of striped bass(Morone saxatilis) eggs in relation to river source and female weight. Rapportset Procès-Verbaux des Réunions, Conseil International pour l’Exploration dela Mer 191:34–42.

Dow

nloa

ded

by [

Dep

artm

ent O

f Fi

sher

ies]

at 2

3:24

25

Sept

embe

r 20

12



Conservation Aquaculture of Northern LeathersideChub and Effects of Temperature on Egg SurvivalMatthew S. Bartley a , Eric J. Wagner a & Randall W. Oplinger aa Utah Division of Wildlife Resources, Fisheries Experiment Station, 1465 West 200 North,Logan, Utah, 84321, USA

Version of record first published: 05 Jun 2012.

To cite this article: Matthew S. Bartley, Eric J. Wagner & Randall W. Oplinger (2012): Conservation Aquaculture of NorthernLeatherside Chub and Effects of Temperature on Egg Survival, North American Journal of Aquaculture, 74:3, 289-296








ARTICLE

Conservation Aquaculture of Northern Leatherside Chuband Effects of Temperature on Egg Survival

Matthew S. Bartley, Eric J. Wagner,* and Randall W. OplingerUtah Division of Wildlife Resources, Fisheries Experiment Station, 1465 West 200 North, Logan,Utah 84321, USA

AbstractWe present 4 years of data that refine aquaculture protocols for the northern leatherside chub Lepidomeda copei,

a species of conservation concern in the Intermountain West. Experiments examined life history traits (age at firstspawning and thermal limits to egg hatching success) and aquaculture techniques (brood density, spawning substratetype and surface area, and feeding methods for fry). Tests showed that leatherside chub can reproduce as early asage 2. Multiple spawns per female during a year were also documented. Survival of eggs was compared at incubationtemperatures of 18.4, 23.0, 24.6, and 26.8◦C. Eggs at 18.4◦C had the highest survival to hatching (54.0%); eggs at 26.8◦Chad significantly lower survival (1.5%). Egg survival at 23.0◦C and 24.6◦C (32–33%) was significantly lower thansurvival at 18.4◦C. Aquaculture experiments showed that the mean total number of eggs produced did not significantlydiffer between brood densities of 8.4 (1,246 ± 1,236 eggs [mean ± SD]) or 16.8 (2,224 ± 1,600 eggs) fish/m3. Studiesshowed that leatherside chub preferred spawning over natural cobble substrate to spawning over marble substrate.More eggs were recovered from a three-substrate tray treatment (1,350 cm2) than from a single tray treatment(450 cm2). Fry given brine shrimp Artemia spp. with probiotic bacteria or fed with an automated, more continuousdrip feeder did not show any advantages in growth over time. Juveniles at rearing densities of 800, 1,700, and3,400 fish/m3 did not differ significantly in growth rates, deformities, or mortalities. This research provides generalguidelines for rearing northern leatherside chub and some additional information on the species’ life history.

The recovery of threatened or endangered species has beenand continues to be an important aspect of fisheries manage-ment. Threats to these species include dewatering, increasingaverage seasonal temperatures, nonnative fish, habitat degrada-tion, and a host of other issues (Brouder and Scheurer 2007;Walser et al. 1999). One of the barriers to management is a lackof knowledge about basic life history traits. A lack of aquacul-ture techniques for rearing rare species can also be a barrier tospecies recovery. An understanding of basic life history traitsand methods for the aquaculture of a species can aid in conser-vation efforts for imperiled species (Rakes et al. 1999; Sarkaret al. 2006).

Northern leatherside chub Lepidomeda copei (formerly Gilacopei or Snyderichthys copei) is a small cyprinid species foundin the Bonneville Basin and the upper Snake River drainageof the western United States (Sigler and Sigler 1987; Wilson

*Corresponding author: [email protected] January 31, 2011; accepted September 12, 2011Published online June 5, 2012

and Lentsch 1998; Johnson et al. 2004). Recent population sur-veys have found reductions in abundance and distribution ofnorthern leatherside chub, resulting in part from water with-drawals and the presence of predatory brown trout (Walser et al.1999; Wilson and Belk 2001; Belk and Johnson 2007). Northernleatherside chub are currently considered a “species of concern”in Utah.

In an effort to preclude listing this species as threatened orendangered, an interagency recovery team has identified hatch-ery supplementation as part of a recovery plan. However, basiclife history information on northern leatherside chub on whichpropagation protocols can be based is limited, though recentefforts have provided some data (Johnson et al. 1995; Sigler andSigler 1996; Wilson and Belk 1996, 2001; Billman et al. 2008a,2008b). In this study we had two objectives: (1) to provide ad-ditional life history data on northern leatherside chub such as

289

Dow

nloa

ded

by [

Dep

artm

ent O

f Fi

sher

ies]

at 2

3:26

25

Sept

embe

r 20

12

290 BARTLEY ET AL.

age at first spawning and upper thermal limits for successful eggincubation, and (2) to provide better protocols for aquacultureof the species. The aquaculture tests explored effects of brooddensity, fry density, spawning substrate area and composition,automated drip feeders, and probiotic feeds for fry.

METHODSThe following studies were performed at the Fisheries Ex-

periment Station, Logan, Utah. These tests were conducted over4 years (2006, 2007, 2009, and 2010). Adult leatherside chubwere collected from Deadman Creek (October 2004) and Yel-low Creek (August 2005), tributaries to Mill Creek in the uppersection of the Bear River drainage, Summit County, Utah (2007studies). Fish were collected again in June 2008 (Yellow Creekonly) for the 2009 and 2010 studies.

All the spawning occurred indoors in fiberglass tanks(0.9 m × 1.8 m × 33 cm of water depth) and received 1.7–2.0L/min of hatchery well water. During winter (before the startof the spawning season), fish were maintained at temperaturesbetween 16.5 and 16.8◦C. At the beginning of each spawn-ing season (in late March or April), warmer well water wasadded to increase temperatures to 18.8◦C. This temperature wasmaintained throughout each spawning season (March throughSeptember). A 0.25- hp submersible pump was used to createa current, the output of which was directed though a horizon-tal polyvinylchloride (PVC) pipe with perforations to create alaminar flow. A range of velocities was created in each tank byputting a sheet of plastic in the center of the tank, perpendicularto the floor, but at an angle to the walls (see Billman et al. 2008afor full description). The resulting velocities provided higherflow areas for the spawning substrate and occasional use andexercise by the fluvial leatherside chub as well as resting areasof lower velocity. Automated feeders (Eheim “Feed-Air” digitalautomatic feeder and Rena LG100) were programmed to feedTetramin flakes 4 times per day, at 3% of total body weight perday (about 13 g per week). Phillips full-spectrum lights wereset on timers to match the outdoor photoperiod.

Spawning substrates (medium cobble or marbles, dependingon the experiment) were made available to the fish. When sub-strate was screened for eggs, it was done by removing a plastictray (15 × 30 × 3 cm) containing substrate, rinsing it with wellwater, and then examining both the rinse water and the substratefor eggs. If any eggs or fry were present, their numbers weredetermined. All handling of eggs was with a bulb pipette. Aftereach screening, the substrate was replaced with a fresh tray con-taining disinfected (1200 mg/L benzethonium chloride solution,>15 min exposure) substrate. Egg checks were performed 2–5times per week, concluding when no more signs of spawningactivity were observed.

Egg incubation temperature.—In 2010, survival to hatchwas compared among four temperatures: 18, 22, 24, and 26◦C.Eggs for the experiment were all derived from a single spawn(18.8◦C). Three replicates for each of the four temperature treat-

ments were used, with 44–45 eggs per replicate. The harvestedeggs were treated with 60 mg/L copper sulfate solution for 2min at 18.8◦C, and rinsed three times before being placed in theincubation chamber. The incubation chambers were made bycutting a 32-mm-diameter PVC pipe into a 5–cm-long section.The section was capped with mosquito mesh netting secured byrubber bands. This chamber was then placed (without temper-ing or acclimation) in a plastic tub (11 L) supplied continuouslywith water (about 1.6 L/min) at the desired temperature. To ma-nipulate water temperature, we used immersion heaters to heatwater in a separate tank. Water from this tank was mixed withnonheated water from the same source, and various water mix-ing ratios (warm/cold) were used to manipulate temperatures inthe plastic tubs. For the first 4 days of incubation, the eggs weredisinfected by placing the whole chamber into 60 mg/L coppersulfate solution for 2 min. The dead eggs and hatched fry in thechamber were counted after 7 d, when hatching was completed.

Age at first observed spawning.—In 2006, 36 northernleatherside chub were placed into a single fiberglass trough(244 × 61 × 30 cm) to determine the earliest age in whichleatherside chub can spawn. At this time the fish were 1 yearold, derived from hatchery spawning events the previous year.This tank was supplied with two trays of substrate, small cobble(see above) or medium cobble (3.5 cm diameter) mixed withyellow glass beads (1.8 cm diameter). The tank also contained aportion of a smaller trough (122 × 35 × 18 cm) suspended ona PVC frame above the bottom of the trough to create both coverand a riffle area where the small cobble tray was located (seeBillman et al. 2008a for further description). The substrate waschecked twice a week between April and September by usingthe procedure mentioned above. This tank was also monitoredin 2007 (when fish were age 2), using the same protocol.

Broodstock density.—This study took place during the 2007and 2009 seasons. Egg production (total eggs per season andnumber of spawning events) was compared between two brood-stock density treatments: either 4 or 5 (low density; 8.4/m3) or9 or 10 spawners (high density, 16.8/m3) per tank. Broodstockdensity treatments were limited by the availability of both adultfish and tank space. The low- and high-density treatments werereplicated by using two tanks per treatment in both seasons, pro-viding four replicates per treatment for the statistical analysis.

Spawning substrate tests.—Two substrate experiments wereconducted, one comparing the preference for either cobble orcolored marble substrate, and the second comparing two sub-strate surface-area treatments. In the first test (2007 spawningseason), fish chose between two spawning substrates: small cob-ble (21–48 mm in size, mean 31 mm) or colored glass marbles(mix of 1.5- and 2.5-cm-diameter marbles). Five spawning tanks(replicates) were used, four of which were also part of the brood-stock density experiment; the fifth tank had 14 broodfish present.The substrates were placed in a plastic tray (15 × 30 cm × 3cm tall), and 1-cm (bar measure) mesh was placed in the plastictray to create a “false floor” in the trays that collected eggs. Thefalse floor was intended to make egg collection easier and to

Dow

nloa

ded

by [

Dep

artm

ent O

f Fi

sher

ies]

at 2

3:26

25

Sept

embe

r 20

12

CONSERVATION AQUACULTURE OF LEATHERSIDE CHUB 291

protect the eggs from predation. Trays for both substrates wereplaced in the high-velocity area of the tank, side by side. Sub-strate was screened twice a week. If any eggs were present, thenumber found was recorded.

In the second experiment (2010 spawning season), egg pro-duction was used to compare two different spawning substratesurface-area treatments: 450 or 1,350 cm2. The substrate trayswere the same as used in previous years (plastic trays 15 ×30 × 3 cm of small cobble 21–48 mm; mean 31 mm). For thelarger surface-area treatment, three trays (1,350 cm2 total) wereplaced adjacent to each other in the tank. The small surface-areatreatment (450 cm2) received only one tray. Six broodfish wereput in each of five spawning tanks. Three tanks had a single trayand two tanks had three trays. The substrate was screened daily(except Friday and Sunday) for eggs. Total egg production wasrecorded for each spawning event and the number of eggs perspawn was compared between the two substrate surface sizes.

Fry diet comparison.—Growth rates of juvenile northernleatherside chub were compared among three feed types:Tetramin flakes, frozen brine shrimp Artemia spp. nauplii, andprobiotic frozen Artemia nauplii. Forty fry (mean weight, 291± 16 mg) were used in each of the three replicates of eachfeed treatment. The fry, pooled from eggs that had hatched dur-ing May and June 2007, were placed in nine separate tanks(Rubbermaid plastic tubs, 24 × 34 × 8.5 cm, with 250-µm mesh placed over effluent). The same water source wasused for all treatments (24.0◦C) and the flow to each tank was0.4 L/min. The fry were fed four and two times daily on week-days and weekends, respectively. Waste was siphoned off thebottom of the tanks as needed—usually weekly, sometimes morefrequently.

Each tank received a ration that was 8% of the total fishbiomass per day. The Tetramin flakes were sieved to provide thefry with consumable sized particles (

292 BARTLEY ET AL.

as the level of significance in all tests. One-way analysis ofvariance (ANOVA) was used to analyze percent hatch data inthe egg temperature test and percent mortality in the fry densitytest after arc-sine transformation (2 arcsine

√x; Kirk 1982).

ANOVA was also used to compare length and weight among frydensity or diet treatments. A t-test was used to compare annualegg production and the number of spawning events per yearbetween density treatments, eggs per spawn in the substrate sizetests, and growth variables between fry feeding methods. TheMann–Whitney U-test was used to compare eggs per spawn inthe substrate type test, since the variance was not homogeneous,as determined with the Kolmogorov–Smirnov test.

RESULTS AND DISCUSSION

Egg Incubation TemperatureThe final test temperatures were 18.4 ± 0.2, 23.0 ± 0.2,

24.6 ± 0.2, and 26.8 ± 0.4◦C (mean ± SD). The mean hatch-ing success at the highest temperature (1.5 ± 2.6%) was sig-nificantly lower than at the other temperatures (P = 0.001,F = 14.19, df = 11). Likewise, at the lowest temperature, meanhatching success (54.0 ± 10.6%) was significantly greater thanat the other temperatures (P = 0.001, F = 14.19, df = 11). Meanhatching rates for the two middle temperatures (23.0◦C: 33.3 ±4.4%; 24.6◦C: 32.2 ± 15.0%) were not significantly differentfrom each other (Figure 1).

The data indicated that egg survival decreases at temperaturesabove 18–19◦C, but some survival was possible at temperaturesof 23.0–24.6◦C. Chronic temperatures approaching 27◦C werehighly lethal. Marsh (1985) found that for the Colorado Rivercyprinids Gila elegans and G. cypha and the pikeminnow Pty-chocheilus lucius, no eggs hatched when incubated at 5, 10,or 30◦C; at 15 and 25◦C, the prevalence of deformities wasgreater than at 20◦C, indicating that temperatures around 20◦C

FIGURE 1. Percent hatch (mean ± SD) of northern leatherside chub eggsincubated at one of four different temperatures. Treatments that share a commonletter are not statistically different (P < 0.05) from one another.

were optimal for incubation. In another study with Coloradopikeminnow (Bestgen and Williams 1994), however, a 62% av-erage hatch rate was observed for eggs incubated at 26◦C, com-pared with 72% and 67% hatching at 18 and 22◦C, respectively.This indicated a higher tolerance to elevated temperature duringegg incubation than was observed for leatherside chub in thisstudy.

It is possible that diel fluctuations in temperatures that ap-proach the same maxima would be more tolerable. For ex-ample, whitefish Coregonus lavaretus ludoga eggs subjectedto sublethal exposures to 21◦C for 2–3 h had increased resis-tance to subsequent exposure to 25◦C (EIFAC 1968). Schranket al. (2003) found that cutthroat trout Oncorhynchus clarki sur-vived in streams with fluctuating temperatures that exceededthe lethal limits established in constant temperature tests. Using20–40-mm-long spikedace Meda fulgida, Carveth et al. (2007)found that diel temperature fluctuations led to better survivaland growth than a use of a constant temperature did. Similarobservations have been recorded by Geist et al. (2010) for juve-nile Chinook salmon O. tschawytscha. However, Bestgen andWilliams (1994) did not find a significant difference betweenconstant and fluctuating ( ± 2.5◦C) temperatures for incubationof Colorado pikeminnow eggs. More research on effects of tem-perature fluctuation on egg survival is needed on many otherspecies, including northern leatherside chub.

Age at First Observed SpawningThe fish had two spawning events in 2007 with a total egg

production of 572 eggs, indicating that northern leatherside chubcan spawn at age 2. No spawning was observed at age 1.

Previous to this study, the basis for age of first spawninghad been the presence or absence of gonads and had beenrecorded only for southern leatherside chub L. aliciae (John-son et al. 1995). In this study, we demonstrated that northernleatherside chub can spawn at age 2. The egg production waslow but may have been influenced by the relatively high den-sity of fish in the tank and the fact that the fish were from thesame age-cohort. Several fish species are able to discriminatebetween kin and nonkin (McKaye and Barlow 1976; Barnett1981; Quinn and Busack 1985), preferring to mate with nonkinand showing less aggression toward kin (Brown and Brown1993). Barber et al. (1970) noted that spikedace begin spawn-ing at age 1. The warmer temperature regimen in southwesternU.S., where spikedace is native, probably gives the species moretime for growth and gonadal development than more northernspecies like northern leatherside chub. Spikedace is a small-bodied species as well, with adults typically less than 65 mmtotal length and not exceeding 4 years of age (Minckley 1973).

Broodstock DensityThe mean total number of eggs produced annually did not

significantly differ (P = 0.58, t = –0.58, df = 7) between the8.4 (1,246 ± 1,236 eggs [mean ± SD]) and 16.8 fish/m3broodstock densities (1,822.8 ± 1,651 eggs). Total production

Dow

nloa

ded

by [

Dep

artm

ent O

f Fi

sher

ies]

at 2

3:26

25

Sept

embe

r 20

12


in 2007 was 8,618 eggs from 37 different spawning events.The mean number of spawning events per year per tank didnot significantly differ (P = 0.80, t = –0.26, df = 7) betweendensities (8.4 fish/m3: 5.0 events ± 5.1 SD, 16.8 fish/m3: 5.8events ± 4.1). The total number of spawning events (summingacross all tanks) for the lower density was 17 in 2007 and 3in 2009, whereas the high-density treatment had 20 spawningevents in 2007 and 9 in 2009.

Density effects were not significant, indicating that brood-fish densities of 9.3–18.7 fish/m3 produced comparable numbersof eggs. The ideal density for broodstock is still not clear, butdoubling of density did not appear to double production. Therewas about a 50% increase at the higher density, on average,which could be biologically significant for conservation aqua-culture programs. The tank with the age-2 fish had 36 fish andlittle production, indicating that high density may deter spawn-ing or possibly increase cannibalism of eggs. We have observedleatherside chub cannibalizing their own eggs, so there is atrade-off between having enough fertile broodstock and havingso many that they will eat much of any production. For the size oftanks used in this study, a range of 9.3–18.7 fish/m3 would be suf-ficient. In larger systems, such as artificial stream reaches, moreresearch should help determine optimal densities. Given that wewere unable to identify gender when initiating a broodfish tank,we had to assume a 50/50 male-to-female ratio among the fish onhand and hope that some females were within each group. Indi-vidual pairing was not possible so that more fish per tank wouldprovide a higher probability of getting both sexes in a tank.

Evidence from one of the spawning tanks with five fishdemonstrated that northern leatherside chub spawn multipletimes in a spawning season. Since there were 12 different spawn-ing events in the tank between 7 April and 7 September 2007,and at least one of the five had to be a male because fertileeggs were collected, apparently at least three spawns per femalewere possible. Johnson et al. (1995) noted that mature ovariesof leatherside chub contain both immature and firm, yolked ova,also suggesting multiple spawns. In our studies, we counted oneskein of eggs in a female (107 mm total length) that had jumpedfrom the tank early in the spawning season: She had 510 largereggs (1 mm diameter) and 334 smaller eggs (0.6 mm). In an-other 125-mm-long female, the eggs in one skein totaled 550,or an estimated 1100 total per fish. Using the ratio from the onefish dissected (60% ripe) and the range of fecundity providedby Johnson et al. (1995; 938–2,573 eggs/female), the estimatednumber of eggs per spawning event in the spring would be 563–1,544 eggs. Blinn et al. (1998) similarly reported that LittleColorado spinedace L. vittata may spawn up to three times ayear, based on observations of mature ovaries and multiple co-horts in captive populations. Evidence in studies of spikedacealso indicates that at least two spawns per season are possible(Barber et al. 1970).

Spawning Substrate Type and Surface AreaIn 2007, the rocks received significantly more eggs per spawn

(217 ± 262 SD) than did the marbles (22 ± 60; P < 0.001,

Mann–Whitney U-test = 277, df = 1). Total production frommarbles and rocks was 894 and 8,870 eggs, respectively. Moreeggs would rinse off and were easier to see on the marbles, sowere significantly easier to collect than the eggs on the rocks.The effect of spawning substrate surface area was assessed in2010. The larger surface-area treatment with three trays puttogether produced significantly more eggs per spawning event(231.5 ± 249.5 SD) than a single tray did (90.29 ± 133.4;P = 0.03, t = –2.1, df = 36).

Although the use of marbles did improve egg collection effi-ciency as designed, fish still preferred the rock substrate, so wecannot recommend marbles as a spawning substrate. Whetherthe fish would spawn on marbles if not given a choice remainsto be determined. We are not aware of any other reports of usingthis approach for egg collection. In other substrate preferencestudies with cyprinids (Gibson et al. 2004; Gibson and Fries2005), Devils River minnow Dionda diaboli preferred gravel tolarger rocks, sand, plants or Spawntex. Blinn et al. (1998) notedthat Little Colorado spinedace spawned in gravel that averaged6.4 ± 0.37 mm in diameter. Spawning activity over gravel hasbeen observed in spikedace as well (Barber et al. 1970). How-ever, woundfin Plagopterus argentissimus chose larger spawn-ing substrates (5–10 cm diameter), rather than gravel, sand, orlarger-diameter rocks (Greger and Deacon 1982).

The success of the larger surface-area treatment could be theresult of the dispersal method when the fish spawns; that is, fora broadcast spawner, a bigger target is easier to hit. However,when the fish spawned on the three trays, a majority of theeggs were usually found in just one of the three. More probably,the production was higher for larger surface-area treatments as aresult of the reduction in cannibalism because of the greater areain which to hide the eggs. Cannibalism has been observed amongother cyprinids such as the spotfin chub Cyprinella monacha(Rakes et al. 1999). The rinse method used here collected theloose eggs efficiently and their physical removal with a pipettewas easy enough that three trays are recommended for futureuse. Other studies with cyprinids have used various sizes ofsubstrate to spawn. Clemment and Stone (2004) used substrateswith an area of 3496 cm2 for rosy red fathead minnows. Blinnet al. (1998) used four trays of gravel, each one covering 2500cm2, to spawn Little Colorado spinedace.

Fry Diet ComparisonNo significant growth difference was observed between fish

given the probiotic and the regular Artemia (P = 0.24 for lengthand P = 0.80 for mean weight). However, growth rates for fishfed Tetramin were significantly higher than for fish fed eitherArtemia or probiotic Artemia (all P < 0.001, F ≥ 12.0, df ≥ 8;Table 1). Differences among treatments in cumulative mortalityat 39 d were not statistically significant (P = 0.46, F = 0.88, df= 8). Mean survival percentages were generally high across alltreatments (91–100%; Table 1).

Probiotics are live microbial feed supplements that benefit ahost animal by improving its intestinal balance (Fuller 1989).In aquaculture, probiotics have been found to increase feed

Dow

nloa

ded

by [

Dep

artm

ent O

f Fi

sher

ies]

at 2

3:26

25

Sept

embe

r 20

12

294 BARTLEY ET AL.

TABLE 1. Comparison of mean length, weight, and survival of northernleatherside chub fed one of three different diets. Means within a column with acommon letter are not significantly different.

FeedMean

length (mm)Mean

weight (mg)Survival

(%)

Tetramin 39.1 z 550 z 90.8 zProbiotic Artemia 37.0 y 380 y 100.0 zArtemia 37.5 y 390 y 97.5 z

conversion and survivability in cutthroat trout (Arndt and Wag-ner 2007), turbot Scophthalmus maximus (Gatesoupe 1994), andmany other species (Gildberg and Mikkelsen 1998; Verschuereet al. 2000). In this study, probiotic Artemia did not significantlyimprove growth or survival. The lack of significant effects onsurvival is probably the result of high survival rates of the controlgroup and those treated with dry feed. Probiotics may be morebeneficial in situations where a bacterial pathogen is limitingsurvival of fry (Gatesoupe 1994).

The higher growth of the fry on the dry diet after 39 d mayindicate some variability in the Artemia biomass and energydensity. With both Artemia treatments it was difficult to get aprecise and consistent hatch, and harvest time varied by a fewhours. Also, juveniles used in the study may have outgrown thesize at which Artemia nauplii would be optimal forage. Other fryfeed studies have shown that Artemia feeding is usually superiorto dry feed alternatives (Harzevili et al. 2003; Tesser et al. 2005;Carvalho et al. 2006).

Automated versus Manual FeedingThere were no significant differences in length, weight, or

deformity rate for northern leatherside chub fed Artemia naupliieither manually four times per day or continuously over 7–8 hvia a drip system (Table 2; all P > 0.18, t < 1.61, df = 4).

Many studies have found that feeding multiple times per dayproduces higher growth rates, but a threshold is usually reachedat which the fish do not benefit from more frequent feedings(Andrews and Stickney 1972; Giberson and Litvak 2003). In thisstudy, the additional time and distribution of the ration did notresult in significantly greater growth. However, the automatedfeeder required just 10 min of set-up time and fed multiple tanks,reducing the overall labor required to feed the fry. One issue withthe automated feeder was that unhatched cysts tended to build upin the tubing over time, which could potentially lead to bacteria

TABLE 2. Comparison of average (SDs in parentheses) growth and deformityprevalence between northern leatherside chub fed Artemia either continuously(7–8 h/d) or manually (4 times/d).

Feeding regimenLength(mm) Weight (g)

Deformities(%)

Continuous 32.2 (0.90) 3.8 (0.40) 2.2 (0.85)Manual 31.7 (1.23) 4.0 (0.06) 1.1 (0.92)

and fungus problems. However, periodic flushing would helpcontrol this problem as well as transfer only hatched naupliito the system. In this study, pathogens were not a problem.Smaller drip-rate fittings clogged in earlier tests of the system,but the 7.6 L/min fittings alleviated those problems. Some of thesame concerns that arose from the probiotic experiment werepresent in this study, such as fish size and consistency in Artemiahatches. Overall, the automated feeder appeared to work welland is recommended for feeding multiple lots of newly hatchedfry until they are weaned onto dry feed.

Juvenile Rearing DensityThere were no significant differences in length, weight, or

percent mortality among the three densities (all P > 0.11; Ta-ble 3). Final lengths among all the tanks ranged from 45.4 to49.0 mm; final weights were 1.07–1.25 g. Percent mortality andspecific growth rate ranged from 0 to 10% and 1.77–1.99%/d,respectively.

The range of densities for our study was limited to the num-bers of fish we had on hand. We found no significant differencein growth or survival in the density range we examined. Ourhighest density, 3,400 fish/m3, was still well below or withinpreviously recommended densities for other cyprinid species inintensive aquaculture (Hepher and Pruginin 1981; Horváth etal. 1992; Wagner et al. 2006), but the above densities were typi-cally recommended for pond-reared cyprinids relying on naturalfoods (Stickney 1979; Horváth et al. 1992; Feldlite and Milstein2000).

Although the upper limit for juvenile northern leathersidechub rearing density has yet to be determined, densities of atleast 3,400/m3 can be used for intensive aquaculture withoutcompromising growth and survival. As seen with the least chubIotichthys phlegethontis (Wagner et al. 2006), leatherside chubtend to be a schooling species; therefore, high densities arepossible. The glass aquaria used for our study were adequate

TABLE 3. Comparison of average (SDs in parenthese) length, weight, percent mortality, and specific growth rate of northern leatherside chub reared at variousdensities for 73 d.

Fish/L (number) Length (mm) Weight (g) Mortality (%) Specific growth rate (%/d)

0.8 46.85 (0.87) 1.16 (0.05) 6.67 (4.62) 3.12 (0.18)1.7 47.71 (0.57) 1.13 (0.03) 4.00 (4.00) 3.07 (0.10)3.4 47.59 (0.74) 1.14 (0.02) 9.00 (2.64) 3.09 (0.06)

Dow

nloa

ded

by [

Dep

artm

ent O

f Fi

sher

ies]

at 2

3:26

25

Sept

embe

r 20

12


rearing environments for the densities we examined. If the fishare to be raised in rearing ponds, supplemental feed may benecessary.

SummaryThe work presented here demonstrated for the first time that

northern leatherside chub begin spawning at 2 years of age andproduce multiple batches of eggs during a spawning season. Inaddition, the research supplements previous work (Billman et al.2008a, 2008b) developing techniques for production of northernleatherside chub. These strategies will probably also be appli-cable to southern leatherside chub and perhaps other closelyrelated cyprinids. Our indoor and outdoor tank observations(unpublished data) suggest the optimal spawning temperature isaround 18.0–23◦C. Egg incubation temperatures were optimalat 18.0–19◦C. Spawning densities should be kept around 9.3–18.7 fish/m3. The glass marbles are not recommended, but largerspawning areas of gravel, which recovered more eggs, are. Toreduce labor, we developed an automated Artemia feeder thatworked well. Fry growth was not hindered at 3,400 fish/m3, andhigher densities may be possible. This rough guideline will helpto provide the numbers of fish needed to take conservation of thenorthern leatherside to the next level of stocking and assessmentthereof.

ACKNOWLEDGMENTSFunding was provided by the Utah State Wildlife Grant pro-

gram and the Utah Endangered Species Mitigation Fund. Addi-tional support was provided by the Utah Division of Wildlife Re-sources. We thank Paul Thompson, Aaron Webber, and SamuelMckay of the Utah Division of Wildlife for their help in collect-ing the broodstock used to provide eggs for this study. We alsothank Chris Wilson for his help in prophylactic treatment of thebrood on arrival for external and internal parasites.

REFERENCESAndrews, J. W., and R. R. Stickney. 1972. Interactions of feeding rates and envi-

ronmental temperature on growth, food conversion, and body composition ofchannel catfish. Transactions of the American Fisheries Society 101:94–99.

Arndt, R. E., and E. J. Wagner. 2007. Enriched Artemia and probiotic dietsimprove survival of Colorado River cutthroat trout larvae and fry. NorthAmerican Journal of Aquaculture 69:190–196.

Barber, W. E., D. C. Williams, and W. L. Minckley. 1970. Biology of the Gilaspikedace, Meda fulgida, in Arizona. Copeia 1970:9–18.

Barnett, C. 1981. The role of urine in parent-offspring communication in acichlid fish. Zeitshrift für Tierpsychologie 55:173–182.

Belk, M. C., and J. B. Johnson. 2007. Biological status of leatherside chub: aframework for conservation of western freshwater fishes. Pages 67–76 in M.J. Brouder and J. A. Scheurer, editors. Status, distribution, and conservation ofnative freshwater fishes of western North America: a symposium proceedings.American Fisheries Society, Symposium 53, Bethesda, Maryland.

Bestgen, K. R., and M. A. Williams. 1994. Effects of fluctuating and con-stant temperatures on early development and survival of Colorado squawfish.Transactions of the American Fisheries Society 123:574–579.

Billman, E. J., E. J. Wagner, and R. E. Arndt. 2008a. Reproductive ecologyand spawning substrate preference of the northern leatherside chub. NorthAmerican Journal of Aquaculture 70:273–280.

Billman, E. J., E. J. Wagner, R. E. Arndt, and E. VanDyke. 2008b. Optimaltemperatures for growth and upper thermal tolerance of juvenile northernleatherside chub. Western North American Naturalist 68:463–474.

Blinn, D. W., J. White, T. Pradetto, and J. O’Brien. 1998. Reproductive ecologyand growth of a captive population of little Colorado spinedace (Lepidomedavittata: Cyprinidae). Copeia 1998:1010–1015.

Brouder, M. J., and J. A. Scheurer, editors. 2007. Status, distribution, and con-servation of native freshwater fishes of western North America: a symposiumproceedings. American Fisheries Society, Symposium 53, Bethesda, Mary-land.

Brown, G. E., and J. A. Brown. 1993. Social dynamics in salmonid fishes: dokin make better neighbors? Animal Behaviour 45:863–871.

Carvalho, A. P., L. Araújo, and M. M. Santos. 2006. Rearing zebrafish (Daniorerio) larvae without live food: evaluation of a commercial, a practical and apurified starter diet on larval performance. Aquaculture Research 37:1107–1111.

Carveth, C. J., A. M. Widmer, S. A. Bonar, and J. R. Simms. 2007. An examina-tion of the effects of chronic static and fluctuating temperature on the growthand survival of spikedace, Meda fulgida, with implications for management.Journal of Thermal Biology 32:102–108.

Clemment, T., and N. Stone. 2004. Collection, removal, and quantification ofeggs produced by rosy red fathead minnows in outdoor pools. North AmericanJournal of Aquaculture 66:75–80.

EIFAC (European Inland Fisheries Advisory Commission). 1968. Water qualitycriteria for European freshwater fish: report on water temperature and inlandfisheries based mainly on Slavonic literature. EIFAC Technical Paper 6.

Feldlite, M., and A. Milstein. 2000. Effect of density on survival and growth ofcyprinid fish fry. Aquaculture International 7:399–411.

Fuller, R. 1989. A review: probiotics in man and animals. Journal of AppliedBacteriology 66:365–378.

Gatesoupe, F. J. 1994. Lactic acid bacteria increase the resistance of turbotlarvae, Scophthalmus maximus against pathogenic Vibrio. Aquatic LivingResources 7:277–282.

Geist, D. R., Z. Deng, R. P. Mueller, S. R. Brink, and J. A. Chandler. 2010.Survival and growth of juvenile Snake River fall Chinook salmon exposed toconstant and fluctuating temperatures. Transactions of the American FisheriesSociety 139:92–107.

Giberson, A. V., and M. K. Litvak. 2003. Effect of feeding frequency ongrowth, food conversion efficiency, and meal size of juvenile Atlantic stur-geon and shortnose sturgeon. North American Journal of Aquaculture: 65:99–105.

Gibson, J. R., and J. N. Fries. 2005. Culture studies of the Devils River minnow.North American Journal of Aquaculture 67:294–303.

Gibson, J. R., J. N. Fries, and G. P. Garrett. 2004. Habitat and substrate use inreproduction of captive Devils River minnows. North American Journal ofAquaculture 66:42–47.

Gildberg, A., and H. Mikkelsen. 1998. Effects of supplementing the feed ofAtlantic cod (Gadus morhua) fry with lactic acid bacteria and immunostimu-lating peptides during a challenge trial with Vibrio anguillarum. Aquaculture167:103–113.

Greger, P., and J. E. Deacon. 1982. Observations on woundfin spawning andgrowth in an outdoor experimental stream. Great Basin Naturalist 42:549–552.

Harzevili, A. S., D. De Charleroy, J. Auwerx, I. Vught, and J. Van Sly-cken. 2003. Larval rearing of chub, Leuciscus cephalus (L.), using decap-sulated Artemia as direct food. Journal of Applied Ichthyology 19:123–125.

Hepher, B., and Y. Pruginin. 1981. Commercial fish farming with special refer-ence to fish culture in Israel. Wiley, New York.

Horváth, L., G. Tamás, and C. Seagrave. 1992. Carp and pond fish culture.Halstead Press, New York.

Dow

nloa

ded

by [

Dep

artm

ent O

f Fi

sher

ies]

at 2

3:26

25

Sept

embe

r 20

12

296 BARTLEY ET AL.

Johnson, J. B., M. C. Belk, and D. K. Shiozawa. 1995. Age, growth, andreproduction of leatherside chub (Gila copei). Great Basin Naturalist 55:183–187.

Johnson, J. B., T. E. Dowling, and M. C. Belk. 2004. Neglected taxonomy ofrare desert fishes: congruent evidence for two species of leatherside chub.Systematic Biology 53:841–855.

Kirk, R. E. 1982. Experimental design: procedures for the behavioral sciences,2nd edition. Brooks/Cole Publishing, Belmont, California.

Marsh, P. C. 1985. Effect of incubation temperature on survival of embryos ofnative Colorado River fishes. Southwestern Naturalist 30:129–140.

McKaye, K. R., and G. W. Barlow. 1976. Chemical recognition of young by themidas cichlid, Cichlasoma citrinellum. Copeia 1976:276–282.

Minckley, W. L. 1973. Fishes of Arizona. Arizona Game and Fish Department,Phoenix.

Quinn, T. P., and C. A. Busack. 1985. Chemosensory recognition of sib-lings in juvenile coho salmon, Oncorhynchus kisutch. Animal Behaviour 33:51–56.

Rakes, P. L., J. R. Shute, and P. W. Shute. 1999. Reproductive behavior, cap-tive breeding, and restoration ecology of endangered fishes. EnvironmentalBiology of Fishes 55:31–42.

Sarkar, U. K., P. K. Deepak, R. S. Negi, S. Singh, and D. Kapoor. 2006. Cap-tive breeding of endangered fish Chitala chitala (Hamilton-Buchanan) forspecies conservation and sustainable utilization. Biodiversity and Conserva-tion 15:3579–3589.

Schrank, A. J., F. J. Rahel, and H. C. Johnstone. 2003. Evaluating laboratory-derived thermal criteria in the field: an example involving Bonnevillecutthroat trout. Transactions of the American Fisheries Society 132:100–109.

Sigler, W. F., and J. W. Sigler. 1987. Fishes of the Great Basin: a natural history.University of Nevada Press, Reno.

Sigler, W. F., and J. W. Sigler. 1996. Fishes of Utah: a natural history. Universityof Utah Press, Salt Lake City.

Stickney, R. R. 1979. Principles of warmwater aquaculture. Wiley, New York.Tesser, M. B., D. J. Carneiro, and M. C. Portella. 2005. Co-feeding of Pacu,

Piaractus mesopotamicus Holmberg (1887), larvae with Artemia naupliiand a microencapsulated diet. Journal of Applied Aquaculture 17(2):47–59.

Verschuere, L., G. Rombaut, P. Sorgeloos, and W. Verstraete. 2000. Probioticbacteria as biological control agents in aquaculture. Microbiology and Molec-ular Biology Reviews 64:655–671.

Wagner, E. J., E. J. Billman, and R. Arndt. 2006. Evaluation of substrate typeand density as factors in optimizing growth of least chub. North AmericanJournal of Aquaculture 68:306–312.

Walser, C. A., M. C. Belk, and D. K. Shiozawa. 1999. Habitat use of leathersidechub (Gila copei) in the presence of predatory brown trout (Salmo trutta).Great Basin Naturalist 59:272–277.

Wilson, K. W., and M. C. Belk. 1996. Current distribution and habitat use ofleatherside chub (Gila copei) in the Sevier and Beaver River drainages insouth central Utah. Utah Division of Wildlife Resources, Final Report, SaltLake City.

Wilson, K. W., and M. C. Belk. 2001. Habitat characteristics of leathersidechub (Gila copei) at two spatial scales. Western North American Naturalist61:36–42.

Wilson, K. W., and L. D. Lentsch. 1998. Current distribution and populationabundance of leatherside chub (Gila copei) in the Heber Valley. Utah Divisionof Wildlife Resources, Publication 98-13, Salt Lake City.

Dow

nloa

ded

by [

Dep

artm

ent O

f Fi

sher

ies]

at 2

3:26

25

Sept

embe

r 20

12



Nutritional Composition and Use of Common CarpMuscle in Yellow Perch DietsTravis W. Schaeffer a c , Matthew J. Hennen a d , Michael L. Brown a & Kurt A. Rosentrater b ea Department of Natural Resources Management, South Dakota State University, Box 2140B,Northern Plains Biostress Laboratory, Brookings, South Dakota, 57007, USAb U.S. Department of Agriculture, Agricultural Research Service, North Central AgriculturalResearch Laboratory, 2923 Medary Avenue, Brookings, South Dakota, 57006, USAc U.S. Geological Survey, Yankton Field Research Station, 31247 436th Avenue, Yankton,South Dakota, 57078, USAd Pacific Northwest National Laboratory, Ecology Group, Post Office Box 241, NorthBonneville, Washington, 98639, USAe Department of Agricultural and Biosystems Engineering, Iowa State University, 101Davidson Hall, Ames, Iowa, 50011, USA

Version of record first published: 13 Jun 2012.

To cite this article: Travis W. Schaeffer, Matthew J. Hennen, Michael L. Brown & Kurt A. Rosentrater (2012): NutritionalComposition and Use of Common Carp Muscle in Yellow Perch Diets, North American Journal of Aquaculture, 74:3, 297-305








ARTICLE

Nutritional Composition and Use of Common Carp Musclein Yellow Perch Diets

Travis W. Schaeffer,1 Matthew J. Hennen,2 and Michael L. Brown*Department of Natural Resources Management, South Dakota State University, Box 2140B,Northern Plains Biostress Laboratory, Brookings, South Dakota 57007, USA

Kurt A. Rosentrater3

U.S. Department of Agriculture, Agricultural Research Service,North Central Agricultural Research Laboratory, 2923 Medary Avenue, Brookings,South Dakota 57006, USA

AbstractHigh market demand for marine fish meals coupled with increasing costs and questionable sustainability of wild

stocks have led researchers to investigate a variety of alternative plant and animal protein sources for aquaculturefeeds. Our objective was to evaluate the use of common carp Cyprinus carpio, a locally abundant, nonnative fish species,to offset the cost of marine fish meal in fish feed. We completed analyses of common carp whole muscle, formulateddiets containing combinations of carp and Gulf menhaden Brevoortia patronus fish meal, and then evaluated test dietsin a feeding trial with yellow perch Perca flavescens. Composition (dry matter basis [dmb]) of common carp flesh(crude protein [CP] = 73.4%, crude lipid [CL] = 25.7%) slightly differed from menhaden fish meal (MFM; CP =71.0%, CL = 11.7%, dmb). Three experimental diets were formulated to include percentage ratios of 50:0, 25:25, or0:50 of common carp muscle meal (CCMM) to MFM to obtain similar crude protein (29.7 ± 0.9% [mean ± SD]),crude lipid (15.0 ± 3.7%), and digestible energy (14.2 ± 0.3 kJ/g) levels. Juvenile yellow perch (initial weight =18.1 ± 3.3 g) were randomly stocked (n = 7) in twelve 37-L tanks resulting in four replicate tanks per treatment.Fish fed 25% CCMM : 25% MFM had significantly higher weight gain, while fish fed 50% CCMM : 0% MFM hadsignificantly higher food conversion ratios and lower visceral somatic indices. No statistically significant differenceswere observed for protein efficiency ratios, Fulton condition factors, hepatosomatic indices, or feed intake of fish fedthe different diets. These results indicate that CCMM can be used to partially offset the use of MFM in yellow perchdiets.

Aquaculture is the fastest growing animal food-producingsector of the world economy (FAO 2009); however, increasingglobal production will probably contribute to sustained exploita-tion of marine fisheries for feed ingredients used for farmed fishand terrestrial livestock feeds (Naylor et al. 2000). Production ofcultured fish requires large inputs of fish meal and oil, primarilyderived from marine sources (Naylor et al. 2000), which leads

*Corresponding author: [email protected] address: U.S. Geological Survey, Yankton Field Research Station, 31247 436th Avenue, Yankton, South Dakota 57078, USA.2Present address: Pacific Northwest National Laboratory, Ecology Group, Post Office Box 241, North Bonneville, Washington 98639, USA.3Present address: Department of Agricultural and Biosystems Engineering, Iowa State University, 101 Davidson Hall, Ames, Iowa 50011,

USA.Received April 29, 2011; accepted December 5, 2011Published online June 13, 2012

to feed costs that often account for over 50% of total productioncosts for aquaculture species (Coyle et al. 2004). Marine fishmeal provides an important source of protein in aquacultureand livestock diets because of its well-balanced profile ofamino acids, essential fatty acids, digestible energy, vitamins,and minerals (Bimbo and Crowther 1992; Abdelghany 2003).However, Watson and Pauly (2001) suggest that global marine

297

Dow

nloa

ded

by [

Dep

artm

ent O

f Fi

sher

ies]

at 2

3:27

25

Sept

embe

r 20

12

298 SCHAEFFER ET AL.

fisheries experienced a decreased harvest of nearly 0.66 millionmetric tons/year (0.72 million short tons/year) from 1988 to1999, and no dramatic increase in landings is anticipated.Furthermore, the Food and Agriculture Organization of theUnited Nations indicated that 28% of current marine fish stocksare overexploited, depleted, or recovering from depletion andan additional 52% are fully exploited (FAO 2009).

The primary source of fish meal in the United States comesfrom four species of menhaden Brevoorita spp. which comprisedabout 15% of the total annual commercial fish landings between2003 and 2007 (NOAA 2009). On one hand, Gulf menhadenB. patronus stocks are maintaining themselves near historic lev-els (Powell 1994; Vaughan et al. 2000), while exploitation ofAtlantic menhaden B. tyrannus stocks has led to a truncation inage structure and reduction in lifetime fecundity (Powell 1994).Owing to the questionable sustainability, potential negative eco-logical impacts, and rising costs of menhaden fish meal (MFM)(Mora et al. 2009), considerable research has been focused ondeveloping feeds derived from alternative plant and animal pro-teins as potential replacements for marine fish meals (Jaunceyand Ross 1982; Tacon 1994; Fontaı́nhas-Fernandes et al. 1999;Coyle et al. 2004; Schaeffer et al. 2011).

Fish meal derived from nonnative freshwater fish species,such as common carp Cyprinus carpio, is an understudiedprotein source that could be used to offset or replace marine fishmeal. Common carp, which originated from the Black, Caspian,and Aral Sea drainages, is one of the most widely distributed fishspecies, occupying every continent (Panek 1987; Balon 1995).Despite being important recreationally and commercially as anaquaculture species in Europe and Asia (Naylor et al. 2000;Arlinghaus and Mehner 2003), populations of wild-type com-mon carp have had extensive negative impacts on ecosystemsin other parts of the world (Koehn 2004; Weber and Brown2009). In particular, common carp are a detrimental invasivespecies in North America and Australia, where they often attainhigh densities and negatively affect aquatic ecosystems (Koehn2004; Weber and Brown 2009). Numerous studies in the U.S.Midwest have described populations of common carp exceeding300 kg/ha biomass with some estimates as high as 1,000 kg/ha(Neess et al. 1957; Lubinski et al. 1986; Fritz 1987; Bonneau1999; Weber et al. 2011). Thus, aquatic resource managersgenerally aim to reduce the abundance of common carp usingvarious means. Massive biomass removals (>40 metric tons)have occurred in areas throughout the upper Midwest region ofthe United States (Ricker and Gottschalk 1941; Rose and Moen1952; Neess et al. 1957; Lubinski et al. 1986). For example,common carp harvest by one commercial fishing crew in SouthDakota alone yielded between 147,000–513,000 kg annuallyfrom 5 to 10 lakes from 2005 to 2010 (Weber et al. 2011). Inaddition, one of the largest fish processing plants in the Midwest(Schafer Fisheries, Illinois) processes 1.36–2.26 million kgof common carp annually primarily for human consumptionand fertilizer (Schaefer Fisheries, personal communication).Despite a relatively low market price for whole common carp(US$0.22–$0.44/kg) and both their processed muscle fillets and

minced product ($1.43–$2.20/kg), none of the common carpcurrently processed at that plant are used in fish meal products(Schaefer Fisheries, personal communication). However, thecost of whole menhaden ($0.33–$0.52/kg) used in commercialfish meals potentially makes common carp a viable replacement(Buguk et al. 2003).

Currently, a growing fraction of common carp research is fo-cused towards increasing biological knowledge, with the goal ofadvancing reduction techniques and removal programs (Penneand Pierce 2008; Weber and Brown 2009; Weber et al. 2011). Itis likely that removal strategies will continue to become moreefficient and cost effective; thus, information regarding uses ofthe removed biomass is necessary. Potentially, a large availablebiomass and the relative low cost could make common carpan ideal source to replace or offset MFM and the developmentof markets for common carp meal would in turn provide anincentive for their commercial removal.

Common carp meal could provide an inexpensive, local feed-stuff for regionally cultured fishes in the Midwest. To our knowl-edge, no studies have used common carp as a protein source inaquaculture diets or as a MFM replacement. Therefore, the ob-jectives of this study were to determine the composition andenergy content of common carp found in the Midwest region ofthe United States and to determine if processed common carpcould potentially provide a suitable replacement for MFM inaquaculture diets, specifically for yellow perch Perca flavescens.

METHODSNutritional composition and energy content of common carp

flesh.—Common carp were collected during fall from BrantLake, Lake County, South Dakota, with boat electrofishing andstored whole at −20◦C, pending processing. Common carp werepartially thawed and total weight (grams), total length (millime-ters), sex, fillet weight (grams), and visceral weight (grams)were measured (Table 1). Five common carp (three males, twofemales) ranging in total length (TL) from 640 to 745 mm andweighing between 3,986 and 6,129 g were randomly selectedto excise whole-muscle fillets for analyses (Table 1). Analyseswere done on homogenized muscle tissue to determine mois-ture, ash, crude protein, crude lipid, fatty acid, mineral, andamino acid compositions and were performed by the Oscar E.Olson Biochemistry Laboratories, Analytical Services Labora-tory, Brookings, South Dakota (Tables 2–5). Moisture contentwas determined by drying samples in a VWR 1350G dryingoven (VWR Scientific, West Chester, Pennsylvania) for 84 h at60◦C (AOAC 2009, method 950.46B; Table 2). Ash content wasdetermined by ashing dried samples at approximately 600◦Cusing an Isotemp model 184 muffle furnace (Fisher Scientific,Pittsburgh, Pennsylvania) for 48 h (Table 2). Crude protein wascalculated from nitrogen content (6.25 conversion factor) mea-sured by a combustion method (AOAC 2009, method 992.15),while crude lipid was measured using ether-extractable fatcontent by solvent extraction (AOAC 2009, method 991.36;Table 2). The amino acid profile was determined using the

Dow

nloa

ded

by [

Dep

artm

ent O

f Fi

sher

ies]

at 2

3:27

25

Sept

embe

r 20

12

YELLOW PERCH DIETS 299

TABLE 1. Summary of common carp individuals used for tissue nutrient and energy content analyses. Muscle weight was derived from one fillet. M = male,F = female, E = energy content analysis, N = nutrient content analysis.

Carcass Total weight Total length Muscle weight Viscera weight Analysisidentification (g) (mm) Sex (g) (g) performed

CC01 4,784 707 M 752 792 ECC03 4,976 677 F 680 1,327 ECC04 4,491 705 M 700 806 ECC05 5,040 716 M 768 850 Ea, NCC07 5,527 735 F 717 1,171 E, NCC09 3,986 640 F 536 766 Ea, NCC11 7,402 750 F 958 1,390 Ea, Na

CC12 6,288 744 F 703 1,220 Ea

CC13 4,040 665 M 534 770 E, NCC16 6,129 745 M 926 1,074 E, N

aAnalysis was performed, but data not reported owing to high variability.

Association of Official Analytical Chemists method (AOAC2009, method 982.30) (Table 3). Fatty acids were determinedon common carp flesh, prior to the freeze drying process, via aone-step extraction–transesterification procedure (Sukhija andPalmquist 1988; Table 3) and mineral analysis was performedwith a modified procedure described by Dierenfeld et al. (2005)(Table 4).

Composition of MFM was provided by the Scoular Company(International Proteins, Minneapolis, Minnesota). The MFM(IPC 740 Gulf menhaden meal) was commercially processedfrom whole menhaden; fish were frozen after capture, thencooked, pressed (i.e., most of the oil and water was removed),and air dried at low temperature (Mike Cici, The Scoular Com-pany, Minneapolis, Minnesota, personal communication). Incontrast, analyses of common carp muscle meal were performedon homogenated muscle fillets removed from thawed, wholecommon carp that had been frozen until time of processing.

The combustible energy content of flesh from individualcommon carp was measured with a Parr model 1108 oxygenbomb calorimeter (Parr Instrument Company, Moline, Illinois)using the procedure described by Hartman and Brandt (1995)(Table 5). For each dried muscle sample, the tissue was homog-enized, and two 1-g (range, 0.93–1.61 g) subsamples were com-busted to determine caloric content. Subsamples that differedby more than 10% of the mean energy density were excludedfrom the results (Hartman and Brandt 1995).

TABLE 2. Proximate composition (mean ± SD, 100% dry matter basis) ofcommon carp flesha (CCF) and IPC 740 Gulf menhaden mealb (MFM).

Component CCF (%) MFM (%)

Crude protein 73.4 ± 6.2 71.0Crude lipid 25.7 ± 7.9 11.7Ash 8.8 ± 0.8 16.7

aAnalyses conducted by Oscar E. Olson Biochemistry Laboratories, Analytical ServicesLaboratory, Brookings, South Dakota.

bAnalyses provided by The Scoular Company, Minneapolis, Minnesota.

Diet formulation and processing.—Three semipurified ex-perimental diets were formulated to replace 0, 50, and 100%MFM with processed common carp muscle tissue. Frozen, bone-less, skinless fillets were cut into approximately 2.

Date post:	25-Jan-2021
Category:	Documents
Upload:	others
View:	0 times
Download:	0 times

Temperature and Maternal Effects on Hatch Rate and Length at … · 2012. 9. 28. · North American...

Documents