Biological Control of Aquatic Macrophytes by Herbivorous ...
Transcript of Biological Control of Aquatic Macrophytes by Herbivorous ...
ILLIN I SUNIVERSITY OF ILLINOIS AT URBANA-CHAMPAIGN
PRODUCTION NOTE
University of Illinois atUrbana-Champaign Library
Large-scale Digitization Project, 2007.
* kOGm L MAtS
O LLM\Sl9I- ('Q
r~ -t·t~
ILLINOIS, , NATURAL HISTORY
r SURVEY
BIOLOGICAL CONTROL OF AQUATIC MACROPHYTES BY HERBIVOROUS CARP
Part 2. BIOLOGY AND ECOLOGY OF HERBIVOROUS CARP
Aquatic BiologyTechnical Report
M. J. Wiley and R. W. Gorden
Principal Investigators
Final ReportFederal Aid Project F-37-R
Aquatic Biology Technical Report 1984(11)
Section
Tr~lr r·G~?Ptfj ·' rrr+r 'f r;i P;r .1,~I Cc rrt
f~l Ir~ ~ ·s'J Lu 1(X3~5
APr is
Aquatlc Biology Technical Report 1984(11)
B ILOGICAL CONTRCL OF AQUATIC MACROFHYTES BY HEFB IVOROUS CARP
Part II. Biology and Ecology of Herblvorouos Carp
Final Report, Federal Aid Project F-37-R
M. J. WIley and R. W. GordenPri ncl pal Investigators
R. W. Gorden, Ph.D., HeadAquatlc B lology
September 1984
M. J. Wl ey, PhhO., PrincipIlInvesti gator, 4atl c B iol
I
Project F-37-R, Effects of Using Hybrid Carp to Control Aquatic Vegetation,was conducted under a memorandum of understanding between the IllinoisDepartment of Conservation and the Board of Trustees of the University ofI IInols. The actual research was performed by the Illinois NaturalHistory Survey, a division of the Department of Energy and NaturalResources. The project was supported by funds from the U.S. Fish andWildl ife Service (DIngelI-Johnson) and the Illinols Natural HistorySurvey. The form, content, and data Interpretations made In this reportare the responsibility of the University of Illinois and the NaturalHistory Survey, and not that of the Illinois Department of Conservation.
ili
PRIMARY RESEARCH PARTICIPANTS
Research Scientists--
M. J. Wiley, Ph.D. - Project EcologistD. P. Philipp, Ph.D. - Project GeneticistR. W. Gorden, Ph.D. - Project Microbiologist
Supportive Scientists--
L. W. Coutant, M.S. - PhytoplanktonS. M. Pescitelll, M.S. - Fish ecologyT. F. Powless, M.S. - FisheriesS. T. Sobaskl, B.S. - Data management, fisheries, benthosP. P. Tazlk, M.S. - Microbiology, plant ecologyS. W. Walte, M.S. - Zooplankton, water chemistry, project coordinatorG. L. Warren, B.S. - BenthosL. D. WIke, MS. - Fish ecology
Consulting Sc entists--
D. H. Buck, Ph.D.R. W. Larlmore, Ph.D.R. C. HIltibran, Ph.D.W. F. Chllders, Ph.D.P. B. Bayley, Ph.D.
Research Ass stants--
P. A. Bukaveckas, B.S.C. M. Clark, B.S.L. L. Crossett, B.S.K. L. Ewing, B.S.
iv
RESEAR•I PARTICIPANTS, Arranged Alphabetically
Carl AideJohn BarlocwPeter BayleyPaul BeatyJanet BeckmannMaria I. BragaJoan BrowerCarolyn BrownHomer BuckPaul BukaveckasLarry CarolWillian ChlldersMark ChounardYuan-Ming ChowChristina ClarkLarry W. CoutantLorrie CrossettRobert DavisSusan DocimoMickey DcnagalaKatharyn L. EwingNancy FryeLeonard GackowskITan GonyonRobert GordenEric Hal I ermanRobert HI ti branMartin JennlngsJanice JohnsonChris KaminskiVictor KeithPawel KindlerAnn KirtsTan KwakAnos Kyse, Jr.Greg Larson
Dawn LutzShirely LoweSam LynchShel a MageePhil MankinGeorge MathieuChristine MayerDede MI llerMark NesbittDave NolteScott O'GradyRichard OlsenStephen M. PescitellAnnette PetreDavid P. PhillppChris PhillipsTodd F. PowlessTan RiceSherry RohlfingLinda Ruzev ckMary SananicSherri SandbergJohn Skel I eyJeff SmithStephen T. SobaskiMary StalTed StorckMike SuleSam SumPamela P. Taz kJoel lyn Von De BurStephen W. WaiteGary L. WarrenBill WestenhaverMark J. WetzelLynn D. WIkeMichael J. WtleyMarv in Young
TPBLE OF CONTENTS
Foreward.. ............. ............................................ v I
Acknowledgnents... ............................... ............ ...... vil
Guide to D-J Job Deslgnations ........................... ... .....* vill
Part II. Biology and ecology of herbivorous carp
Chapter 1.
Chapter 2.
Chapter 3.
Chapter 4.
Chapter 5.
Bloenergetics of herbivorous carp................ 1-1by M. J. Wiley and L. D. Wlke
Feeding preferences and consunptionrates of grass and hybrid carp................... 2-1by M. J. Wlley, S. M. Pescitelli,and L. D. WIke
Ecological Impacts of herbivorouscarp in pond ecosystems....................... 3-1by P. P. Tazlk, S. T. Sobaski,G. L. Warren, and M. J. Wiley1
Potential reproductive abilities ofdip old and triploid hybrid carp(female Ctenopharyngodon Idella xmale Hypophthalmichthvs noblils) anddiplold and triplold grass carp.................. 4-1by D. P. Philipp
Reproductive histology of grass carp(Ctenopharyngodon Ideal), bigheadcarp (Hypophthalmlchthys nobllls) andtheir F 1 hybrid.................... .............. 5-1by T. F. Powless, S. T. Sobaski, andM. J. Wiley
Appendix A3. Analysis of variance for ecologicalimpacts of herbivorous carp.....................A3-1
1With the technical assistance of K. L. Ewing
vl
FOREW ARD
This report consists of three separate parts to facilitate distributionand aid In comprehension. Part I contains the Executive Sunmary for thisproject; it gives a brief description of major results and a series ofrecommendations to the Illnois Department of Conservation regarding thestatus of the grass carp and its genetic derivatives In Illinois. Part IIcontains a series of chapters detailing our studies of the biology andecology of these fish. Part III contains stocking recommendations forIllinois and a description of the computer simulation model upon whichthey are based.
vil
ACKN QO LDG IENTS
Many people have contributed to this Investigation over the four and ahalf years since It began In 1980. A complete list of researchparticipants has been given and we are grateful for the cooperation andprofessional contributions of each and every person listed. Beyond thoseactually conducting the many research tasks involved in a study of thisscale, there are a number of people who deserve special thanks: J. M.Malone of J. M. Malone and Sons Enterprises, who provided us with fish andtechnical assistance throughout the study; Jana Walte and Sue Peratt, whoprovided secretarial, editorial and occasionally manual assistance forwhich we are most thankful; Jim Allen, Peter Paladino, and SteveHarrison of the Illinois Department of Conservation, with whom we haveInteracted throughout the study and for whose cooperation and patience weare grateful; Dr. A. R. Brigham and associates, who performed all of ourlaboratory water quality analyses; and Dr. Sue Wood and associates, whohandled most of our tissue analyses and calorimetry. We also thank B. F.Goodrich and Environmental and Process Equipment, Unlimited for supplyingthe PVC blof I tration media.
viii
GUIDE TO CHAPTERS BY D-J JCB DESIGNATION
In order to facilitate the identification of report contents with federalstudy and Job classifications used In the AFA for F-37-R, the followingguide Is presented. In addition to listing chapters In this final report,references to appropriate publications supported by this contract are alsogiven.
STUDY 101:
Job 101.1.
Job 101.2.
Job 101.3.
Job 101.4.
Job 101.5.
Job 101.6.
STUDY 102.
Job 102.1.
Job 102.2.
Job
Job
Job
102.3.
102.4.
102.5.
STUDY 103:
Job 103.1.
Job 103.2.
Job 103.3.
A comparison of the effects of hybrid grass carp and selectedaquatic herbicides on aquatic vegetation and the sport fishery.
Determination of existing aquatic macrophytes--Part 2: Chapter5.
Control--Part 2: Chapters 2 and 3; Gorden et al. 1982
Water qual ity--Part 2: Chapter 3
Establ Ish of fish populations--Part 2: Chapter 3
To determine growth characteristics--Part 2: Chapters 1, 2, and3.
Reproduction--Part 2: Chapters 4 and 5
The effects of the control of aquatic vegetation by hybridgrass carp and by herbicides on the rates of decomposition,nutrient cycling and flow and the subsequent effect on thebass, blueglll, and catfish populations.
Measurement of invertebrate and microbial populations--Part 2:Chapter 3
Rates of deocmposition of aquatic macrophytes--Part 2; Chapter3; Gorden et al. 1982
Nutrient cycl ing studies--Part 2: Chapter 3
Energy flow-Part 2: Chapter 1
Systems analysis--Part 3; Wiley et al. 1983
Genetic composition and reproductive capability of Fl hybridcarp.
Genetic composition and uniformity--Executive summary; Mageeand Philipp 1982
Gonadal developnent--Part 2: Chapter 5
Reproductive capability-Part 2: Chapter 4
Chapter 1
BIOENER3ETICS OF HEFBIVOROUS CARP
INTRODUCTION
Several aspects of energy balance In the grass carp (Ctenopharyngodon
idella Val.), Including consumption, food conversion, and growth rates,
have been examined (Fischer 1968, 1970, 1972a, 1972b, 1973, 1977; Fischer
and Lyakhnovich 1973; Fowler 1982; Negonovskaya and Rudenko 1974; ShIreman
et al. 1980; Stanley 1974a, 1974b; Urban and Fischer 1982; Venkatesh and
Shetty 1978). Despite this attention, we understand little about the
bloenergetlcs of these herbivorous fish. Most of the work to date has been
with fish less than 200 grams, not a particularly representative size for a
fish which can easily reach 1 kg in its first growing season and 10+ kg
within 5 years (Shireman and Smith 1983). Even less Is known about the
energetics of the various genetic derivatives of the grass carp now being
promoted and sold for aquatic plant control. The most prmainent among
these are the hybrid carp (female Ctenopharyngodon Idella x male
Hypophthalmlchthys .nob lls) (KIIambi and Zdlnak 1982; Shrreman et al.
1983), and the triploid grass carp (Malone 1983).
Interest in using herbivorous carp for control of aquatic macrophytes has
spurred much of the energetic research with grass carp to date (see
Shireman 1984). Energetic analyses are attractive In this context because
they can provide a quantitative basis for comparing the relative plant
control potential of various sizes and types of herbivorous fishes.
However, the energetics of these fish are of theoretical interest as well,
since most energetic work with fishes has been restricted to piscivorous,
Insectivorous, or omnivorous species. Herb vory Is a relatively rarefeeding strategy In fishes (Lagler et al. 1962) and the grass carp Is a
particularly enignatic herbivore. Despite Its apparent lack of special
adaptations for handling cel I ulose (Hickl ing 1966) and probable low
assimilation efficiencies (Brett and Groves 1979, Fischer and Lyakhovlch
1973, Urban and Fischer 1983), It can achieve near phenomenal rates of
growth (5-11 g/day; Shireman and Smith 1983).
1-2
In this paper we report the results of a series of energetic analyses with
the grass carp, the cammercially available triplold grass carp, and the
hybrid carp. Basic bioenergetic parameters of the three fishes are
compared statistically and used to formulate representative energy balances
for each. Because we are particularly Interested in the relative control
potential of the three types of herbivorous carp under differing climatic
regimes, a strong emphasis has been placed on examining both the size and
temperature dependency of basic bloenergetic processes.
E1THODS
Fish for all experiments were provided by J.M. Malone & Son Enterprises of
Lonoke, Arkansas. Studies were conducted In the Aquatic Field Laboratory
of the Illinois State Natural History Survey, Chanpalgn, Illinois. Fish
not Immediately used In the laboratory were kept In adjacent experimental
ponds. When handling was required, as for weighing and measuring, fish were
anesthetized with qulnaldine to reduce stress and aid In handling. When
temperature accl imatlon was required, changes were restricted to less than
2 degrees per day, and a 2-week period at the target temperature was
allowed before experiments were begun. PIoldy of hybrid carp was verified
by electrophoretic analysis (MaGee and Phil pp 1982). Grass carp ploldy
was verif ed by blood cell volume techniques (Beck and Blggers 1983).
Consumption Growth
Consunption and growth rate experiments were conducted in six 1500-liter
(400-gal) circular al uLnnun tanks (Fig. 1-1). Water qual ity was maintained
by circulating water from the tanks through two B.F. Goodrlch PVC
blofilters, each filter serving three tanks. A microcomputer controlled
filtration cycles, a 13-hour photoperloa, and tank water temperatures; It
al so recorded dally mean water and roa temperatures. All tanks were
aerated continually and dissolved oxyger was monitored regularly with a
YSI Model 54 dissolved oxygen meter. X•rlng temperature adjustment and
accl imation, fish were fed the plant se!ces to be used experimentally.
Fig. 1-1. Layout of laboratoryconducted.
1-6 Experimental tanksA Cold roomB Warm roomC Blof lIter (warm room)D Blof iter (cold romn)E Computer and electronic
control center
where feeding experiments were
F Window air conditionersG Central heating/air
conditioning unitH Well water sourceJ Make-up water filterK SinkLM
Chemistry laboratoryMunicipal water source
1-3
1-4
After acclimation, fish were weighed, measured, and Individually marked by
clipping soft fin rays (Welch and Mills 1981) before being returned to
their respective tanks. During each 2-3 day feeding period, each tank wassupplied with sufficient plant material to ensure constant foodavailab I ity. At the end of each feeding period, uneaten plants on thesurface were.renoved with a net; plants and feces renalning on the bottomof the tank were removed and separated with a modified swimming pool punp
and filter. Plants to be fed and those not eaten were weighed afterrinsing and removal of excess water by spinning 4 minutes in a washing
machine. Feces were weighed after removal of excess water and spinning,
then frczen for further separation and analysis. Sanples of fresh plants,uneaten plants, and feces were periodically retained for chemical and
caloric analysis.
Consunption experiments were conducted with fish ranging fran 34 to 2,133 g(I Ive weight) and at temperatures ranging from 15 to 34 C. Twenty-onefeeding experiments were conducted over a 2-year period (Table 1-1).Hybrid carp were used in eight experiments, normal diploid grass carp inseven, and tr plold grass carp in six. Foods used varied somewhat with
availability (Table 1-1), but Potamogeton crispus (curlyleaf pondweed) anda lettuce-endive combination were the most frequently used. Dallyconsumption rates were related to projected body size as a proportion ofbody weight per day. Body size was estimated throughout the duration of
each experiment by fitting a simple exponential growth model (Everhart andYoungs 1981) to the Initial and final weights In each experimental group.
Resplrometry
Standard metabol ism was determined using continuous flow resprancetry.
Fish were placed in tubular plexiglass chambers with inflow and outflow
ports of equal area. The chanbers were placed In a constant temperature
bath of oxygen-saturated water. Flow through the chanber was regulated by
a rheostat-controlled peristaltic pump and was maintained at a level
sufficient to prevent a greater than 20% drop In dissolved oxygen across
the chamber. Fine control and monitoring of flow rate was accomplished
1-5
Table 1-1. Summary of laboratory feeding studies
Fish Weight (g) TemperatureFood
Experiment Type Number type Range Mean Range Mean
1 HC 6 PC 111- 200 134.7 32.9-34.8 33.92 HC 6 PC 87- 252 143.2 27.2-30.7 28.93 HC 5 PC 85- 130 106.0 22.4-29.4 26.84 HC 6 PC 90- 138 106.3 20.9-23.8 21.95 HC 5 PC 75- 166 114.8 17.6-19.8 18.86 HC 6 PC 81- 154 116.7 14.8-18.3 16.27 HC 3 PC 865- 970 916.7 29.0-31.2 30.28 HC 4 PC 835-1305 1036.3 20.0-21.0 20.79 DGC 4 PC 930-2075 1412.5 27.0-30.0 28.5
10 DGC 3 PC 985-1090 1041.7 19.9-21.0 20.611 TGC 5 PC 72- 149 117.0 24.5-25.5 25.012 TGC 5 L 81- 177 135.4 24.5-25.5 25.013 TGC 10 L-E 48- 81 57.4 24.6-27.3 26.414 TGC 10 L-E 53- 66 60.5 24.3-27.6 26.415 TGC 10 L-E 43- 56 50.4 14.7-18.5 17.216 TGC 10 L-E 40- 87 56.2 15.5-19.2 18.217 DGC 10 L-E 43- 81 60.8 24.7-28.4 26.518 DGC 10 L-E 34- 69 52.0 17.1-20.0 18.919 DGC 2 MP 1485-2133 1809.0 - 24.020 DGC 1 MP 1365-1365 1365.0 - 24.021 DGC 2 NM 1055-1300 1177.5 - 24.0
For fishtri pl old
type: HC = hybrid carp; DGC dipl old grass carp, and TGC =grass carp
For food type: PC = Potamogeton crlsus; L = lettuce; and L-E =lettuce-endive; MP = mixture of odea canadensls, Ceratoohyv I umdemersump Mvrloohyllum sp., Potamogeton pectinatus, and Naasflexllls; NM = Naias m nor
1-6
wlth an adjustable flow meter. Use of a multiple head punp and valved
manifold assembly allowed use of up to four chanbers at a time without
Interruption of flow during measurement of dissolved oxygen. All dissolvedoxygen measurements were made with a Nestor Model 8000 BOD meter calibrated
with the azide modif ed W nkler method. Temperatures were maintained by aYSI proportional control er.
Resplranetry experiments were conducted between 14 and 30 C with 115 fish
of all three types ranging from 41 to 1,610 g (Table 1-2). Fish were
starved for a minimun of 18 hours before being placed in the chambers. The
fish were then allowed at least 12 hours to accl Imate to the chanbers
before readings were taken. Dissolved oxygen measurements were taken frcm
each chamber and the water bath every half hour for 3 to 5 hours. Fork
length, total length, and weight were recorded for each fish after removal
from the chamber. Equations for standard metabolism as a function of size
(weight) and temperature for each fish type were developed by multiple
regression (Minitab 1982) and compared by analysis of covarlance (Zar
1974).
Ca orImetry
Whole fish, samples of food plants, and feces were analyzed for energy
content. Samples were freeze-dried and pulverized. One-gram subsamples
were pressed Into pellets and combusted in a Parr oxygen bomb using anoxygen bomb calorimeter of the same manufacturer. Several replicateal Iquots were analyzed to give a representative mean value for each sampl e.
Energy content was calculated on both fresh-weight and dry-weight bases.
Wet to dry weight ratios were al so determined for each sampl e.
RESULTS
Standard metabolic rates for grass carp, triploid grass carp and hybrid
carp were measured; in each case, respiration could be described by a
simple power functions of body weight and temperature:
1-7
Table 1-2. Summary of fish type, size range and temperatures used inresplrametry studies.
Temperature (C)
Fish type 14 16 17 18 19 20 22 24 27 30 Total
Diploid grass carp<100 g - - - 6 - - 4 - 4 2 16
100-500 g - - 4 2 - - 4 - 4 6 20>500 g 2 - - - - 2 - 2 - 1 7
Total 2 - 4 8 - 2 8 2 8 9 43
Triplold grass carp<100 g - - - 9 4 - 1 - 8 4 26100-500 g - - - 3 - - 7 - - 1 11
>500 g - - - - - - - - - -
Total - - - 12 4 - 8 - 8 5 37.
Hybrid carp<100 g - 3 - - - - 4 1 - 5 13100-500 g - 1 - 4 - - 4 3 - 3 15
>500 g - 2 - - 2 - - 2 - 1 7
Total - 6 - 4 2 - 8 6 - 9 35
Grand total 2 6 4 24 6 2 24 8 16 23 115
1-8
In Q02 = a + B ( In temperature) + B2( n we ght) (1)
Multiple regression analysis of data for each type of fish suggested that
the dipl old and tr pl old grass carp had similar metabol Ic rates (Tables 1-3
and 1-4, Fig. 1-2), but the hybrid's metabolic rate was substantially
different (Table 1-5, Fig. 1-3). Analyses of covarlance found no
significant difference between diplold and triplold grass carp and so data
were pooled and used to generate a Joint estimator (Tables 1-6 and 1-7). A
similar analysis indicated that the hybrid's equation differed
significantly in slope frcm that of the diploid' s (Table 1-6). The
resulting power equations summarize the standard metabolic rates of the
grass carp and the hybrid carp:
Grass carp Q02 = 0.026 x welght0 . 64 5 x temperature1 .07 (2)
Hybrid carp Q02 = 0.001 x weight 0 . 819 x temperaturel . 6 8 (3)
An overlay of these two functions (Fig. 1-4) gives an Indication of the
relative difference in metabolism between the hybrid and the two grass
carp. Although the metabol Ic rate of the hybrid starts out lower than that
of the grass carp, Its rate of Increase with body size is greater and by 1kg exceeds that of grass carp by 4%. As size increases further, the
difference widens and by 10 kg the hybrid's standard metabolism would be
55% higher than that of grass carp.
Consumt i on
Consunption, standardized as percent body weight (fresh) consumed per day,
was estimated in each of the feeding experiments. Comparisons between the
three types of carp were made by regressing paired dally consunption values
of differing genetic types. Paired data were used only from those
experiments which ran simultaneously with different types of fish under
equivalent conditions (experiments 13-18 for diplold and triplold grass
carp; experiments 7-10 for hybrid and diploid grass carp). The hybrid
versus grass carp comparlson was supplemented with publ i shed data of
Sh reman et al. (1983) to provide the largest possible range in size. This
1-9
Table 1-3. Diplold grass carp standard metabol Ism as a function ofweight and temperature.
Log respiration = -3.70 + 0.665 (log weight) + 1.05 (log temperature)
Analysis of Variance
Due to d. f. SS
RegressionResi dual sTotal
R2 = 89.5
Variable
24042
Coefficient
17.87561.9800
19.8557
Standard DevI atl on
Log weightLog temperature
8.93780.0495
T-Rati o
-3.70260.665071.0490
0.54050.035510.1515
-6.8518.736.92
1-10
Table 1-4. Triplold grass carp standard metabol ism as a function ofweight and temperature.
Log respiration = -2.90 + 0.373 (log weight) + 1.21 (log temperature)
Analysis of Variance
Due to
RegressionResidual sTotal
R2 = 68.7
Variable
d. f.
23436
Coefficient
SS
2.64271.10973.7524
Standard Dev aton
Log weightLog temperature
1.32140.0326
T-Rati o
-2.89560.37291.2139
0.62960.11040.1611
-4.603.387.54
__
1-11
SoU.lr '\
0 0
- an' a
11 \
rP-A,1 a- Q j~stfa
5 ^ffi
13 (^
C13- c
:II
O"a
a a
t
2- C
L.I_
0 -._..QI ,..
== ',(
-r - ,-
CC
+ S4 -
JI-
-. -
-o
-a --
--=
o _n 3 I o r in o i" C2i
ui Z 52 CD \ - C-
1-12
Table 1-5. Hybrid carp standard metabol Isn as a function of weightand temperature.
Log respiration = -6.79 + 0.819 (log weight) + 1.68 (log temperature)
Analysis of Variance
Due to d. f.
RegressionResidualsTotal
R2 = 88.0
23234
SS
20.4062.595
23.002
10.2030.081
Variable Coeffici ent
Log weightLog temperature
-6.78690.81931.6796
Standard Deviation
0.79440.05350.2214
T-Ratl o
-8.5415.327.59
1-13
00
0
0
0i-)
Lr
Q_
LJ
mr
>--
O0
O0C+-t(0
C(0
U
+-
L
(o
.s
4-0
L
CO
I
-..
§-
CL
cnLLJ
-(\jm
I-?
_LJ
I-
L3r
U,u,
)
|
II z M: L:) ** =
1-14
Table 1-6. Analysis of covariance for standard metabol Ism equations.
In respiration =
Diplold vs. Triploldm=2
SS
DIploid 1.98Triploid 1.109Pooled 3.089Total 3.315
Diplold vs. Hybridm=2
SS
Diploid 1.98Hybrid 2.595Pooled 4.575Total 7.473
Analysis of covarlance
a + b(l n weight) + b2(ln temperature)
k=2
d.f.
40347477
k=2
d.f.
40327275
HO: Both regressions estimate thesame population
HA: The sample regressions do notestimate the same populations
SST - SSp SpFcalc -= -/- = 1.805
(m+1)(K-1 dfp
Fcrlt. 0.05(1),3,74 = 2.68
Therefore, accept HO:
HO: Both regressions estimate thesame population
HA: The sample regressions do notestimate the same populations
SST - SSp SpFcalc = = 7.60
(m+1)(K-1) dfp
Fcrlt. 0.05(1),3,72 = 2.68
Therefore, reject HO:
1-15
Table 1-7. Diploid and triplold grass carp standard metabol Isn as afunction of weight and temperature,
Log respiration = -3.66 + 0.645 (log weight) + 1.07 (log temperature)
Analysis of Variance
Due to d. f. SS
RegressionResi dual sTotal
R2 = 87.8
Variable
27779
Coefficient
24.5463.315
27.862
Standard Deviation
Log weightLog temperature
12.2730.043
T-Rati o
-3.65990.64501.0701
0.38420.02840.1107
-9.5322.739.67
1-16
C
-n40
1-'
S-I0*
E(
II -
FL
lo
II
a0
o o-
0)8
<0 Q>
0)
9)
U)
r°
r4.
(Al
m
I
'J.
mU,-3m
-- I
»--.
CT
Ln
1-17
approach allowed a direct comparison of relative feeding rate over a wide
range of temperatures, sizes, and food types. Despite the highly variable
conditions represented by each data pair, regressions had surprisingly low
standard errors (Tables 1-8 and 1-9), suggesting that differences In
feeding rates of these three fishes can be adequately described in terms of
simple proportional differences in relative daily consunption. Hybrid carp
fed at a rate approximately 66% lower than did diploid grass carp (Fig. 1-
5). Triplold grass carp, while performing much better, still fed more
slowly than did normal grass carp, about 90% that of the diploid rate (Fig.
1-6). Therefore, the consumption rates of these three types of herbivorous
carp can be ranked as follows: diploid grass carp > triplold grass carp >
hybrid carp.
The relationship between temperature and consumption rate was examined for
each type of fish. Analyses were restricted to experiments with fish
snaller than 800 g (experiments 1-6 and 11-18) in order to avoid excessive
variance due to size differences (see below). Dally consunption for each
type of fish was plotted against daily mean temperature (e.g., Fig. 1-7)
and fit to a asymptotic log function (Tables 1-10 through 1-12). Analysis
of covarlance was then used to examine differences between the three types
of fish (Tables 1-13 and 1-14). The results of the analysis indicated that
each type had a statistically distinct temperature response (Fig. 1-8).
The diploid fish was the most active at lower temperatures and the hybrid
the least. The threshold temperature at which feeding begins can be
estimated from these regression equations, and was 9 C for the diploid
grass carp, 10 C for the triplold grass carp and 13 C for the hybrid carp.
The relationship between body size and consumption was examined by
correcting all experimental data to 20 C using the temperature functions
just described, grouping all values by 100 g increments, and then plotting
consumption against body size (Fig. 1-9). There was no statistically
significant difference in consunption (% body weight per day) across size
classes detectable by analysis of variance in either the hybrid or diplold
grass carp data (no analysis of triplold grass carp data was attempted
because of the snall size range available). However, both types gave some
1-18
Table 1-8. Regression of triplold consunption as a function ofdipl oid consunption.
Tripl old consunption = 0.898 (dipl od consumption)
Analysis of Variance
Due to
RegressionResi dual sTotal
d. f.
15556
SS
37.68612.2940
39.9801
MS
37.68610.0417
Var Iabl e Coeff icl ent
D plold consunption 0.89817
Standard Dev ati on
0.02988
T-Rati o
30.06
1-19
Table 1-9. Regression of hybrid carp consunption as a function ofdiploid grass carp consumption.
Hybrid consunptlon = 0.660(di pl old consumption)
Analysis of Variance
Due to d. f. SS
RegressionResidual sTotal
Variable
8788
Coeff Ic ent
23.8001.603
25.403
Standard Deviation
0.66037 0.01838
23.8000.018
T-Rati o
DIpl old consumption 35.94
1-20
- - C=) 0 > 1 C -=< =I
.7 r" w. , .n , ;,=
0l
II
0
J1
US
8Cm)
0
"I
? 800
io<8
QC3
0
I o
Cr-
000 a
0a I
0
o0
SCO
0
0 0o o S
0 -<
0\o C\\ ° -)
O°
%.w
1-21
a
0 -Co \ 0,
0_ 0
an
- )o
00
00 m
- \ -c
o Ioo o -
0 0 P- o -
0- "
0 0
0 0 0C)
0 C
\ o o °
00 o 0c 8
LO 00______________V gUo-S-S ~ ~ O
'" ^ Qr'- Q- l C3 »- C
1-22
- J W LR 01 % e (0
A-
O(c
00
a ^* 0
m
I a
.Sm
+ la m
0 0
0
c o
a M-m
-1- m
o
=^ s
w m^To
Os^»
§ s-
I - I I - 1 I I
O °°I 0 \ 0
0
0 °0 0
a 0
0
oBa ao
0
00
0 0
YI
1-23
Table 1-10. Regression of dip lold grass carp consunption as afunction of temperature.
Diplold consumption = -1.19 + 0.520(temperature)
Analysis of Variance
Due to
RegressionResidual sTotal
R2 = 35.6
d. f.
2122
SS
0.193140.307740.50088
0.193140.01465
Variable
Temperature
Coefficient
-1.18530.5202
Standard Deviation
0.44610.1433
Tabl efunction of
1-11. Regression of triplold grass carp consunption as atemperature.
Tripl old consumption = -1.19 + 0.499(temperature)
Analysis of Variance
Due to d. f.
RegressionResi dual sTotal
R2 = 56.7
13738
SS
0.475980.347420.82340
Variable
Temperature
Coefficient
-1.19380.4988
Standard Dev iation
0.21460.0701
T-Ratl o
-2.663.63
0.475980.00939
T-Rati o
-5.567.12
1-26
Table 1-14. Analysis of covarlance for dlploid-hybrid regressionsof consumption on temperature.
Analysis of Covarlance
Group d.f. SS MS
Hybrid 93 0.9394 0.01010Diplold 21 0.3077 0.01465
F = 1.45
Pooled regression114 1.2471 0.01094117 2.0483 0.01770
Difference 3 0.8012 0.26700
F (3,114) = 24.4
1-27
4,-OC0
tC
4-I
"3
a
.§
Om0.-L. h
(04 -
-0. V
8r
•b
00
-4-
L a
lIr C
LU
Lc
I--
-I
LD1SI-Q-
C)LJ.
LU
Lr
LU
)--
,Q
&n
d
-iCmI=a C
-< > C cm m amI C
I- -± i-. T T T T T V T T T
a A
------ -0---
O 0
0
CL .,. 1
01
0ol
Cn
01
••-
al
"I
0.
N(A
"I
0
01
0.U)
-9-
o
1-28
a.
8-
3
3
m
C0
6m
b
0-
H-
NAC3C3C2
-l
OC-
-0o
C-)
:zCrO
6-4
CO)
'-3
m
r 1
-I
a,0cc
cm
000
1-29
indication of a reduction In consumption at the largest size groupingsobserved (>1.5 kg). No regression analysis gave a satisfactory fit to thesedata, although parabol c functions came the closest.
Energv Budgets
Since consumption, growth, and standard metabol ism measurements were
available, energy budgets for each feeding experiment were calculated
(Table 1-15). Components of metabol ism not measured in this study, SDA and
metabolic costs of activity (active-standarad metabolism), were estimated
as k x standard metabol ism, where k ranged from 1 to 3, scaled In
porportion to feeding rate according to the appropriate temperature
function derived above. Total fecal and urine losses were calculated by
subtracting growth and total metabolism from consumption.
Respiration ranged fran a low of 6.2% of consunptlon for large diploid
grass carp fed mixed plants to a high of 88.6% for large hybrid carp fed
Potamoaeton crispus. As expected fecal and urine (F+U) losses were high,
ranging frcm a minimum of 44% with small hybrids at low temperatures to a
high of 91% with large diploid grass carp feeding on mixed species at 24 C.
The overall average loss due to F+U was 72%; the overall average loss due
to metabol sm was 26%, leaving an average of 2% available for growth. Of
course ccmbining all three types of fish provides only a very general
description of the energy balance In herbivorous carp. Nevertheless, this
average energy allocation would suffice to give a 1-kg carp feeding on ELcrispus a growth rate of 3 g/day. If the hybrid data are excluded, theaverage observed energy balance for the grass carp was: 1001 = 21M + 8G+71E; sufficient for a growth rate of 10 g/day.
Highest growth as percentage of consunption occurred with snall diploidand triploid grass carp fed lettuce and endive. Assimilation efficiencies
(1/U) varied greatly between experiments, ranging from 0.09 to 0.56. A
multiple regression analysis of factors Influencing assimilation efficiency
showed a significant (Table 1-16) decrease in assimilation with increasing
size (Fig. 1-10) and an increase with increasing temperature (Fig. 1-11).
anan
In
an
ra
aO
0
an)
0anI
0L
0O
(a
1o
qm
= o =2 0 Wa. a a
S
I,
ISL46
3:
a.r
a-
S
UU
cr
*o
a.
U
10an
IO
If
C;
in9Crs
w
i
o
*
a
a
a1
an
r
r
I
* a
ac
an
an
awe
ao
i
cm
a
0i
an0
(M
ana
i
I
anan
Ba
Ia
g
a
o1 o
I
o 0
.-.
cIar( a
0a 1a an
cu 03
9 . N
ia i
InS 0
i 0 0as a aan a
a aI
* *
9. 9 N
n Ra
ON i
* Ni
a N
0 9.
c c
k a
0 S t
4Ac
Id 16
bd toa u9=
1-30
a0
d
q
*C
tC
a0
a
0
d
0
en
N
0
a)
anan
aD
an
ao
0a
0cma
a
0c;*a
Ca
go
a
cucua
0
aca
ana
C"a
ca
a3
0*
so
-II II ?
I*
S
US
a .
S
: 0
-
a
S
S
ILa
*
06
&J
a
aao
a a a
a a a
a a a
a a a
a a a
a a a
aNl
I
CaI
S
T
I
a
S1
Ud
S;
II
i0
1ifI1I
I
an
usl !
I
I
Table 1-16. Summary of I Iterature val ues of 1/U, K1 , and K2.
FishAuthor size Food Tanp. 1/U K1 K2
Fowler (1982)
Urban & FIscher(1982)
Fischer (1970)
10-16 g Algae & chara
10-16 g Lettuce10-16 g Elodea
10-16 g Potamaoeton sp.
3- 612- 2219-124
365-545 d
23 g52 g
23+52 gcanbined
An ImalAnimalAnimalPlantMixedAnimalPlantMixed
PlantPI antPI ant
19-20 0.005-0.009
19-20 0.04219-20 0.016,
0.02519-20 0.022
2121212121212121
232323
0.2550.8100.4300.3850.3800.3400.1650.385
0.1220.1580.123
NR NR
NRNR
NR
0.0850.3850.1800.0800.1850.1350.0250.290
0.0200.0280.019
NRNR
NR
0.3400.4800.4000.2100.4800.4000.1400.760
0.1650.2100.165
Fischer (1972a)
FIscher & 86-107 gLyakhovlch (1973) 20- 55 g
Stanley (1974) 0.89-1.3 kg
AnimalProtei nCarbohydrate
PlantProtei nCar bohydrate
AnimalPlant
2222
2222
2222
0.261 0.152 0.6010.923 0.003 0.003
0.410 0.065 0.1590.754 <0.001 <0.001
0.395 0.1250.172 0.022
0.4040.145
22-24 0.500 0.329 0.658
1-31
mmmmlmwmmý mmmmommommý
1-32
There was a trend for hybrid assimilation to be lower than that of triplold
and diplold grass carp, but it was not statistically significant (Table1-17). Based on this multiple regression analysis, the relationship betweenthe proportion of consumed calories assimilated, temperature, and weightcan be sunmarized as follows:
1/U= 0.054 x weight-0 .436 x temperature .34 (4)
A decrease with increasing weight was also observed for K1 efficiencies(the percent consunption allocated to growth; WInberg 1956) (Table 1-18 andFig. 1-12). The highest K1 was achieved by grass carp fed lettuce-endive,and the lowest from large fish fed E. crlsus. The proportion ofassimilated consumption allocated to growth (K2) ranged from 0.03 to 0.48and showed no correlation with fish weight. Data from hybrid carpexperiments were excluded from the analyses of K1 and K2 because little orno growth occurred.
Using the energetic relationships developed in this study, sunmary energybalances can be computed for each of the three types of herbivorous carp.Fol lowing the convention of Brett and Groves (1979), and considering a 1-kgfish at 25 C, the energy balances are as follows:
Diploid grass carp 1001 = 14M + 80E + 6G (5)
Triploid grass carp 1001 = 15M + 80E + 5G (6)Hybrid carp 1001 = 22M + 86E - 6G (7)
For a fish feeding on P. crispus, these energy balances translate Intogrowth rates (with diploid consunption = 32% body weight/day) of 6, 5, and0 g/day for diplold grass, triplold grass, and hybrid carp, respectively.
DISJCUSSION
Even though herbivorous fish are economically important, their
physiologIcal ecology has received little attention (Brett and Groves
1979). Although there are proported to be over 500 papers dealing with the
grass carp (Fischer and Lyakhnovlch 1973), there seems to be I Ittle
I I I I a a I a a
1-33
LD
CJ
LC
<
-Jz
I
=ILDLii~
-J1
0C
Q4
0
C
au)
C
O
O9r
73
<»-
1-34
c Z _I-
"1 .• u -W rd ru . . ..
-i-- r-
Z --
0 01
* m
C-
<n ^
Ab
i
C
cm
II)
-4.
(A)
*. as-.
a r-
0 f §
aa
IC
Ii-
a >1
1--
a
a
--I
rnnIa
1-35
Table 1-17. Multiple regression analysis of In assimilationefficlency (1/U) as a function of In fish weight, In temperature, and fishtype.
In 1/U = -2.92 - 0.436(In weight) + 1.34(ln tenp.) - 0.373(flsh type)
Anal y sis of Variance
Due to d.f. SS MS
Regression 3 7.1973 2.3991Residuals 17 3.8757 0.2280Total 20 11.0730
S = 0.4775
1-36
Table 1-18. Regression analysis of K1 as a function of fish size.
K1 = 0.159 - 0.0157( log weight)
Analysis of Variance
Due to
RegressionResidual sTotal
R2 = 48.9
Variable
log weight
d. f.
11112
Coefficient
0.15891-0.01569
SS
0.00697730.00614580.0131231
Standard Dev ation
0.024740.00444
MS
0.00697730.0005587
T-Rat o
6.42-3.53
1-37
00
C0
C
U,
a
0
0
44
C
C8
C0
OI4-
00
00
0
L-
LLU
-J
LUL
_1J
aa
OO
1-38
agreement about many aspects of their biology. Most bloenergetic work to
date has been done by relatively few authors, primarily Fischer and herassociates (Fischer 1972a, 1972b, 1973, 1977; Fischer and Lyakhnov ch; Urban
and Fischer 1982), Stanley (1974a, 1974b), and Fowler (1982); principallyon fish under 100 g (see Table 1-16). There have also been a fewadditional studies examining daily growth and consumption rates(Negonovskaya and Rudenko 1974; Shireman et al. 1980; Tatral 1979;Venkatesh and Shetty 1978, Young et al. 1983); agaln mostly with very snal Ifish.
Two major general Izations about the bloenergetics of the these fIsh emerge
from this body of work. First, most authors agree that snall grass carp
and hybrid carp require certain amounts of animal protein for proper growth
and that animal and plant foods are eaten at different rates and util Ized
differently by the fish (Kllambi and Zdinak 1982; Shirenan et al. 1983;Fischer 1970, 1972a, 1972b, 1973, 1977, Fischer and Lyakhnovich 1973;Urban and Fischer 1982). Although this Is an Important observation, Itadds little to a general understanding of energy partitioning and use inadult fish, and over the majority of an Individual fish's I Ifespan. Asecond observation, more useful in that respect, is that the aerobicmetabol ic rates of these fish are unusually low (Winberg 1956; Yeh 1959;Fischer 1970; Stanley 1974). Winberg (1956) fit a general power functionfor metabol ism of the grass carp using the data of Chen and Shl (1955)arriving at the relationship:
Q02= 0.132 welgh10. 84 mL/h (8)
which is about half the average rate for fishes he examined and 61% of theaverage for tropical fishes reported by Sholander (1953). Stanley (1974),Incorporating the data of Fischer (1970), Yeh (1959), and Winberg (1966),
with additional data of his own, estimated routIne metabol Ism as:
Q02 = 0.220 weight 0 . 7 6 (9)
which was sl ightly lower than W nberg's equations. Convertlng our analysis
1-39
(Table 1-7) to comparable units (mL/h at 20 C) gives a standard metabol Ic
rate of
Q02= 0.444 welghtO. 6 4 , (10)
lower than even Stanley's estimate, although this is not surprising since
he Interpreted his data as representing routine metabol Ism. Therefore itseems clear that these herbivores have relatively low metabolic energydemands, which may contribute to their surprisingly high growth rates.
Reports of assimilation efficiency (U-1) in grass carp are more variable.They range fran 0.005 to 0.923, representing a wide variety of food types,sizes, and temperatures (Table 1-16). Our data Indicate that U-1 varieswith all three of these parameters so each should be specified for anygiven energy balance. For fish 100 g or larger feeding exclusively onplant material, Urban and FIscher (1982) and FIscher and LyanhhovIch (1973)reported that grass carp exhibit relatively low assimilation (<20%).Stanley (1974), on the other hand, estimated assimilation at 57%. U-1; inour experiments values ranged from 12 to 47% (excluding hybrid data) withan average of 29%. Stanley's estimate Is closer to other non-herbivorousfishes (73-80%: Winberg 1956, Brett and Groves 1979) but must be questionedon the basis of the overall reasonableness of his energy balance. He
reported 50% of the consuned calories were allocated to growth (k1 = 0.33);given that efficiency, a 1-kg fish feeding on . crispus would exhibit a
growth Increment of 50 g/day, which Is clearly impossible (literaturevalues range up to 12 g/day; Shlreman and Smith 1983). The preponderance
of the evidence available suggests that U-1 is In fact substantially lower,probably about one-third that of typical carnivorous fishes. Despite thislow assimilation, these fish are capable of high growth rates because oftheir high consunption rates (2-3 times that of carnivorous fish) andtheir ow metabolic demands.
It would be of considerable Interest to determine by experimental means
sane of the paraneters estimated or obtained by back calculation In this
study. In particular, a knowledge of SDA and activity costs would be
1-42
K lambi, R. V., and A. Zdinak. 1982. Food Intake and growth of hybridcarp (female grass carp, Ctenopharyngodon Idella x male bighead,Arlstichthys (Hypophthalmichthys) nobills) fed on zooplankton andchara. J. Fish Blol. 21:63-67.
Magee, S. M., and D. P. Phillpp. 1982. Biochemical genetic analyses of thegrass carp female x bighead carp male Fl hybrid and the parentalspecies. Trans. Am. Fish. Soc. 111:593-602.
Negonovskaya 1. T., and G. P. Rudenko. 1974. The oxygen threshold andcharacteristics of the respiratory metabolism of young herbivorousfish, grass carp (Ctenopharyngodon Idella) and the bighead(Arlstilchths nobi Ls). J. Ichthyol. 14(6):965-970.
Osborne, J. A. 1981. The size of grass carp as a factor In the control ofhydrllla. Aq. Botany, 11:129-136.
Pescitelli, S. M., M. J. Wiley, S. W. Walte, and L. D. Wike. 1983.Laboratory studies of hybrid carp consumption and assimilation rates.-In M. J. Wiley, R. W. Gorden, and S. W. Walte, eds. Effects of usinghybrid carp to control aquatic vegetation. Third annual report,Federal Aid Project F-37-R, Aq. Biol. Tech. Rep. 1983(1). IllinoisNatural History Survey, Champaign. 138 p.
Prosser. C.L. 1973. Comparative Animal Physiology. W. B. Saunders Co.,Philadelphia, PA. 966 p.
Ryan, T. A., Jr., B. L. Joiner, and B. F. Ryan. 1982. Minitab. MinitabProject, Statistics Dept., Pennsylvania State Unlv., University Park,154 p.
Shireman, J. V., D. E. Colle, and M. J. Macelna. 1980. Grass carp growthrates In Lake Wales, Florida. Aquaculture 19:379-382.
1-43
. R. W. Rottman, and F. J. Aldrldge. 1983. Consumption and
growth of hybrid grass carp fed four vegetation diets and trout chow
In circular tanks. J. Fish Biol. 22:685-693.
_ , and C. R. Smith. 1983. Synopsis of biological data on the grass
carp. FAO Fisheries Synopsis No. 135. Food and Agriculture
Organization of the United Nations, Rome, Italy. 86 p.
Stanley, J. G. 1974. Energy balance of the white amur fed egeria.
Hyacinth Control J. 12:62-66.
Urban, E., and Z. Fischer. 1982. Comparison of bloenergetic
transformations in grass carp (Ctenopharyngodon idella Val.) in two
age classes, under different feeding conditions. 2nd International
Symposlun on Herbivorous Fish, p. 111-119.
Venkatesh, B., and H. P. C. Shetty. 1978. Studies on the growth rate of
the grass carp Ctenopharyngodon _Idella (Valenciennes) fed on two
aquatic weeds and a terrestrial grass. Aquaculture 13:45-53.
Welch, H. E., and K. H. Mills. 1981. Marking fish by scarring soft fin
rays. Can. J. Fish. Aquat. Scl. 38:1168-1170.
Winberg, G. G. 1956. Rate of metabolism and food requirements of fishes.
Beloruss. State Univ., Minsk (Fish. Res. Board Can., Transl. Ser. No.
194 (1960).
Young, L. M., J. P. Monaghan, Jr., and R. C. Heldinger. 1983. Food
prefrences, food Intake, and growth of the F1 hybrid of grass carpfemale x bighead carp male. Trans. Am. Fish. Soc. 112(5):661-664.
Zar, J. H. 1984. Blostatistical Analysis. Prentice-Hall, Inc. Englewood
Cliffs, NJ. 718 p.
Chapter 2
FEEDING PREFERENCES AND CONSUMPTION RATES OF GRASS AID HYBRID CARP
INTRODUCTION
The rate at which grass carp consume various species of aquatic macrophytes
differs substantially (Verigin et al. 1963, Mehta et al. 1976, Shireman and
Smith 1983). Likewise, presented with a choice between several different
species, these herbivorous fish will express a consistent "preference" for
one species over another (Verigin 1963, Cure 1970, Fischer and Lyakhnovlch
1973). Since the use of this fish (or a genetic derivative) as a
biological control agent In Illinois Is being contemplated, an
understanding of its relative preferences for plant species commonly
problematic In Illinois is essential. More specifically, both the rate at
which these fish eat particular plants and the order In which they consume
particular populations In multi-species communities must be known to plan
stocking for macrophyte control In specific bodies of water.
In this paper we report the results of an investigation Into the relative
preferences and consumption rates of grass carp (Ctenopharyngodon Idella)
and its hybrid (female grass carp x male bighead carp, Hypophthalmichthys
nob lls) with respect to nine species of aquatic macrophytes identified as
problematic in Illinois (Tazlk and Wiley 1983). A species ranking
according to relative preference is developed and Is related to average
rates of consunption by both the hybrid and the grass carp. These results
are then compared statistically with several different plant
characteristics to examine underlying determinants of feeding preferences.
METHODS
Studies were conducted at the IHMS Aquatic Research Field Laboratory from
24 June to 17 October 1983. Fish were initially brought into the
laboratory on 12 April 1983 for acclimation. Large grass carp and hybrids
(>450 mm) from the 1979 year class, obtained fram J. M. Malone and Son
2-2
Enterprises, Lonoke, Arkansas, were reared to adult size In experimentalponds In Champaign and Kinmundy. Throughout acclimation andexperimentation, the fish were held In 1500-liter (400-gal) circularaluminum tanks that were screened from outside activity by opaque plasticbarriers. Six tanks were used: two for the preference study, three for
the consumption rate study, and one control. Each tank used for thepreference study was divided into three compartments that allowed freemovement of the fish but kept floating plants and falling plant fragmentsseparated; one tank held grass carp and the other hybrids. Tanks used inthe consumption rate study were left undivided. Two fish were placed intoeach tank Initially; two tanks contained hybrids and the other grass carp.The control tank, used to estimate macrophyte handling error, was alsodivided but contained no fish. Fig. 2-1 shows the general layout of thelaboratory and the experimental design. All fish were weighed and measuredbefore being placed Into the tanks, or whenever fish were lost.
The experiments consisted of consecutive 2-3 day feeding periods for eachset of tanks. Plants were presented to the fish In both studies byfloating them on the surface. In the undivided tanks, single species ofmacrophytes were fed In excess. In the divided tanks, a different speciesof macrophyte was placed Into each compartment. Placement of plantspecies was varied with respect to compartments to avoid compartment bias.After each feeding period, uneaten plants were removed fran both sets oftanks. Most of the uneaten plants rema ned floating on the surface; fecesand plant fragments on the bottom were retrieved using a modified pump andfilter that separated larger plant fragnents. Material passing through theinitial filter was then manually Inspected for smaller plant fragments.Before and after each feeding, plants were spun for 4 minutes atapproximately 500 rpm in a washing machine to remove excess water and thenwere weighed. Macrophyte species were used as they became seasonally
available. In the divided tanks, the plants were presented In two and
three species combinations until a definite order of preference was
establ i shed. If a species was selected over others for two consecutive
feeding periods, It was then compared to other species. In the consumption
rate study, an Individual plant species was fed until the consumption rate
2-3
Fig. 2-1. Diagram of laboratory and experimental design forpreference and consumption rate studies. 1 = hybrid carp, preferencestudy (divided tank). 2 = hybrid carp, consumption rate study (undividedtank). 3 = grass carp, consumption rate study (undivided tank). 4 =grass carp, preference study (divided tank). 5 = hybrid carp, consumptionrate study (undivided tank). 6 = control, no fish.
2-4
leveled off after an initial period in which the fish were adjusting to a
new food type. A total of nine macrophyte species (Table 2-1) were used In
both studies. Plants were collected fran local ponds and lakes and were
held for short periods (prior to feeding) in large tanks with flow-through
water.
During acclimation and experimentation, tank temperatures were maintained
with Immersion heaters controlled by an Apple II Plus microcomputer that
recorded hourly mean temperatures onto magnetic disks. Water quality was
maintained by pumping water from each tank through two 16-ft 3 B. F.
Goodrich PVC blof lters for two 15-minute periods each hour. Make-up water
was also added after each feeding. All tanks received continuous aeration
and dissolved oxygen was measured daily with a YSI dissolved oxygen meter
to Insure that adequate oxygenation was maintained. Water samples were
taken periodically and analyzed for ammonia, nitrate, and nitrite
concentrations using a Technlcon CSM-6 autoanalyzer (Table 2-2). Samples of
each macrophyte species were also analyzed for caloric content (Table 2-3).
Twenty-four different combinations of plants, presented in groups of 2-3
plant species each, were used to determine the order of preference for
nine species of macrophytes. A species was considered "preferred" when it
was consistently consumed before another species simultaneously presented.
In the earliest experiments, It was observed that if species A was
preferred over species B and species B was preferred over species C,
species A was preferred over species C in all trials. Based on this
observation, order of preference was establ ished making a minimum number of
comparisons.
RESULTS
Preferences were clearly demonstrated by both types of carp when hybrids
and grass carp were presented the same combinations of plants. When it
became clear that there was no difference in preference between the grass
carp and the hybrid (approximately 4 weeks), they were given different sets
2-5
Table 2-1. Order of preference for grass carp and hybrids of ninecommon native Ill nois macrophyte species.
Rank Scientific name Cammon Name
1 Nazas flexills Slender naiad2 Na~s minor Brittle naiad3 Chara spp. Chara
Potamogeton follosus Leafy pondweed 1
4 Elodea canadensis American elodea5 Potamogeton pectinatus Sago pondweed6 Potamogeton crisps Curlyleaf pondweed7 My rophyllum spp. WatermlIfoll8 Ceratophylhu demersum Coontall
1 Not compared with all other species, placement inferred
2-6
Table 2-2. Physical and chemical parameters for experimental tanksand designation of experimental effect.
Temp.±SD Dissolved Ammonla-N Nitrite-N Nitrate-NTank Fish (C) oxygen (mg/L) (mg/L) (mg/L) (mg/L)
1-0 Hybrids 23.7±1.45 7.1 0.27 0.07 1.72-U Hybrids 24.5+1.02 7.5 0.21 0.07 1.13-U Grass carp 23.8+1.30 6.1 0.32 0.10 1.24-D Grass carp 24.1+1.63 5.9 0.36 0.12 1.35-U Hybrids 24.1+1.30 7.3 0.25 0.08 1.66-D Control 24.0±2.06 6.1 0.14 0.05 0.2
D = divided tanks used for preference studyU = undivided tanks used for consumption rate study
Table 2-3. Fresh to dry weight ratios and caloric values (cal/g) ofaquatic macrophytes.
Cal/g wet Cal/g dry Fresh to drySpecies Rank weight weight weight ratio
Naas flexills 1 517 3,390 0.16N. mior 2 328 3,640 0.10Potamaeton follsus 3 467 3,010 0.15Chara spp. 3 509 1,853 0.18Elodea canadensis 4 430 2,208 0.24P. pectinatus 5 658 3,603 0.09
. cr.lsaus 6 304 3,774 0.10Myr lohyllum spp. 7 813 3,980 0.20Ceratophyllum demersum 8 446 3,578 0.10
2-7
of species to test al I possible combinations while plant species were
available. The final preference hierarchy is given In Table 2-1.
Fish appeared to be unaffected by the dividers and swam freely between
compartments. Grass carp regularly consumed up to 70-80 percent of theirbody weight (BW). Hybrids ate less than 50 percent BW, even for the mostpreferred species. Preference was demonstrated clearly and within severaldays for most combinations. The most preferred species, particularly NaJasflexllis, were often chosen exclusively with other, less preferred speciesleft uneaten. For species at the lower end of the preference scale, thechoice was not always as clear. Potamogeton craspus and Myriophyllum spp.were very close in preference as indicated by the small difference Inamounts consumed. Ceratophyllum demersum was consistently the leastpreferred plant.
In single species feeding tanks, grass carp consistently consumed greaterquantities than the hybrids for all species tested (Table 2-4). For sameplant species, the consumption rate for grass carp was twice that for thehybrids. Nalas flexllIs and . dminor were among the most rapidly consumed
species by both fishes. ElQdea canadensis and Potamogeton pectinatus were
the only species which differed In relative rank between the hybrid and
grass carp. Chara spp. became unavailable locally before It could be fed
to hybrids but it was readily eaten by the grass carp and was among themore rapidly consumed species. Myrlophyllum spp. was consumed at very lowrates by both grass carp and hybrids, as was Ceratophy llum demersum. £Efoliosus was not abundant locally and disappeared early in the season, soit was not used in the consumption rate study.
Consumption rates were significantly correlated with relative preference(Table 2-5), Indicating that the species eaten first In mixed plantexperiments were those that could be consumed most rapidly. A positive
correlation was found regardless of whether consunption was measured as
fresh weight or in energy units (kcal/day). This was true of both hybrids
and grass carp, and when both groups were combined (Fig. 2-2). However,
neither preference nor consumption rate were correlated with food
2-8
Table 2-4. Mean daily consumption In percentgrass carp and hybrid carp during consumption ratetanks).
body weight per day forstudy (Individual
% body weight Number of Totalconsumed/day feeding number % loss/day
Plant species (mean ± SD) periods of days In control
Grass caral
Eodea canadensL s 42± 4 6 20 2Naias flexlls 41+11 4 9 4
. mLnor 38+ 6 4 10 6Chara 36+17 3 11 4Potamogeton pectfnatus 3Q+ 8 5 19 2Ceratophvl um demersum 22± 5 4 14 3Myrlophyllum spp. 10+ 3 7 7 2
Hybrid carp2
Naas flexlls 27± 8 8 10 4Potamogeton pectlInatus 23+ 8 7 16 2NaJas minor 19+11 4 13 6Edea canadensis 16 4 10 12 2Ceratophyllum demersum 13± 5 5 11 2Myrlophyl lum spp. 8± 4 11 11 3
SData from 1 tank2 Data from 2 tanks
2-9
Table 2-5. Correlation analyses of preference, consumption rate,and plant characteristics. Both simple amd Spearman rank correlations aregiven.
Comparison Pearson's r Spearman's r
Hybrid consumption (% BD/day) vs. preference 0.720* 0.662Hybrid consumption (kcal/day) vs. preference 0.443 0.512Diploid consumption (% BW/day) vs. preference 0.782* 0.810*Diplold consumption (kcal/day) vs. preference 0.705 0.810*Combined consumption (% BD/day) vs. preference 0.598* 0.582*Combined consumption (kcal/day) vs. preference 0.508* 0.523*Plant caloric content/g vs. preference -0.229 -0.143Plant wet:dry ratio vs. preference 0.256 0.244Plant caloric content vs. consumption (kcal/day) 0.178 0.304Plant wet:dry ratio vs. consumption (kcal/day) 0.289 0.131
*Significance at p = 0.05
0
0.
cc II )
a NQ) *
o
OD
CL -
0)
40
QI
0
0)
to
Q)01
a
a
:0
C-)
COcnc-D
r-1
>
m
-o
r
mrl
0
C0
('I5n
u,
2-10
0
-n1rm
r-P"1m
2-11
properties we examined, including percent water, caloric content per gram
(fresh and dry weights) and percent nitrogen content (an Index of protein
content) (Table 2-5).
DISCUSSION
Of the nine macrophyte species used in these studies, all are listed byState fisheries biologists as among the most abundant and problematicspecies In Illinois (Tazlk and Wiley 1983).
Feeding behavior was very similar between the grass carp and the hybrid,particularly in the preference study. Although the grass carp consumedgreater quantities, the final ranking of plants was similar for both carpIn the consumption study. Na.asf Ilexi lLs and 1. minor were highlypreferred species. Ch.ar spp., Eloda canadensls, Potamogeton .ectinatus,
and & follosus were also readily consuned. Ceratophyllum demersum andLyrlophyllum spp. were the least preferred species.
These results are similar to those reported by others for the hybrid andgrass carp. Young et al. (1983) found NaJas auadaluDensis (similar to JLflexllls) was chosen over Elodea nuttall l and Ceratophyllum demersum wasleast preferred by hybrids. In a study similar to the present one, Cassaniand Caton (1983) determined Chara spp., &L. guadalu ensis, and Potamogeton
Decttnatus were readily consumed by the hybrid and grass carp and were ofequal preference; SQ demersum and Myrlobhyllum brasl lense were least
preferred. Avault et al. (1968) reported that grass carp preferred Charaspp. > L auadalupensLs > I . demersun > Myrlophyllum spp.
The reason that fish consistently choose one plant species over anotherwhen given a choice is difficult to assess. Our study Indicated that there
was no correlation between food properties likely to be important and
relative preference. Caloric content, protein content, and water content
were not directly reflected In the patterns of choice observed. However,
because average consumption rate and preference were correlated, a viable
hypothesis would be that the fish choose to consune those plant species
2-12
first that they can consume most rapidly. This strategy would Indirectly
result In an optimization of caloric intake, corrected for relative
differences in handl ng time. We suggest then that handling time Itself Is
the criteria upon which the choice Is actually made. In the .ad L bitum
feeding experiments, handling time (pursuit time + ingestion time, sensu
Hol I Ing 1966) Is the primary component of consumption rate being measured,since searching time and digestive pause are likely to be zero for these
herbivores. Hickl ng's (1966) studies indicated that grass carp have noability to produce cellulase and must instead rely on a thorough grindingof cells by the pharyngeal teeth to physically fracture cells prior to
digestion. Thus, chewing for the grass carp, like many other herbivores,
becomes a major component of handling time, a component that could be
lessened by choosing plants that are easy to masticate. This Is consistent
with the observations (Verlgin 1961, Prowse 1971) that grass carp prefersoft, non-flberous plants, and that when feeding on single species oftenconsume "tender buds" first (L. WIke, personal observation). Other factorssuch as growth form may influence handl Ing time; we observed ratherelaborate oral manipulations of stems and leaves by the fish prior tochewing and have noted that some plant species seem to require more
manipulation than others. The value of a strategy that minimizes handling
time is well IIllustrated by energetic experiments performed In our
laboratory using common lettuce (Lactuca salva; see Part 1: Chapter 1).Lettuce Is a terrestrial plant that Is very low In caloric content and very
high in water content. It would seem to be a poor food for a herbivorerelative to most of our native aquatic macrophytes. Nevertheless, in moreor less identical experiments (experiments 11 and 12, Table 1-15), grasscarp grew substantially faster on lettuce than on Potamogeton crisgus;
although fish feeding on P. crispus had higher assimilation and K1efficiencies. The difference was entirely attributable to the fact that
fish consumed almost twice as much lettuce as EP crispus per day.
2-13
LITERATURE CITED
Avault, J. W., Jr., R. 0. Smitherman, and E. W. Shell. 1968. Evaluation
of eight species of fish for aquatic weed control. FAO Fish. Rep.
44:109-122.
Cassani, J. R., and W. E. Caton. 1983. Feeding behavior of yearling and
older grass carp. J. Fish. Biol. 22:35-41.
Cure, V. 1970. Dezvoltarea species (Ctenopharyngodon Idellal) in lazul
Frasinet (the development of grass carp in Fraslnet pond). Bul.
Cerc. Pisc. 29(4):31-49.
Fischer, Z., and V. P. Lyakhnovlch. 1973. Biology and bloenergetics of
grass carp (Ctenopharyngodon Idella). Pol. Arch. Hydroblol. 20(4):
521-557.
Hickl Ing, C. F. 1966. On the feeding process in white amur,
Ctenopharyngodon Idell. J. Zool. Lond., Proc. IPFC 12:236-243.
Hollings, C. S. 1966. The functional respone of Invertebrate predators to
prey density. Mem. Entcmol. Soc. Can. 45:1-60.
Mehta, I., R. K. Sharma, and A. P. Tuank. 1976. The aquatic weed problem
in the Chambal Irrigated area and Its control by grass carp, pp.
307-314.. In C. K. Varshney and J. Rzoska, eds. Aquatic weeds of
Southeast Asia. Dr. Junk, The Hague, Netherlands.
Prowse, G. A. 1971. Experimental criteria for studying grass carp feeding
in relation to weed control. Prog. Fish-Cult. 33(3):128-131.
Shireman, J. V., and C. R. Smith. 1983. Synposis of biological data on
the grass carp Ctenopharyngodon idella (Cuvier and Valinclennes
1844). FAO Fisheries Synposis 135. 85 p.
2-14
Tazlk, P. P., and M. J. Wiley. 1983. Problem aquatic macrophytes in
Illinois: a survey. In M. J. Wiley, R. W. Gorden, and S. W. Walte.
Effects of using hybrid carp to control aquatic vegetation. Third
annual report, Federal Aid Project F-37-R. Aq. Blol. Tech. Rep.
1983(1). III. Nat. Hist. Surv., Champaign. 137 p.
Verlgin, R. F. 1961. Results of work on acclimatization of Far Eastern
phytophagous fishes and measures for their assimilation and study in
new regions. Vopr. Ikhtlol. 1(4):640-649.
1963. The present state and perspectives of ut I Ization of silver
carp and grass carp in reservoirs of USSR. Ashkhabad 1961: 20-38.
Ashkhabad, Izdat. Akad. Nauk TurkmenskoJ SSSR.
, N. Viet, and N. Dong. 1963. Data on the food selectivity and daily
ration of white amur, pp. 192-195. Jn Problems of the fisheries
exploitation of plant-eating fishes in the waters of the USSR.
Ashkhabad, Akad. Nauk TurkmenskoJ SSSR.
Young, L. M., J. P. Monaghan, Jr., and R. C. Heldinger. 1983. Food
preferences, food Intake, and growth of the F1 hybrid of grass carp
female x bighead carp male. Trans. Am. Fish. Soc. 112:661-664.
Chapter 3
EC(LOSICAL IMPACTS OF HEFBIVOROUS CARP IN POND ECOSYSTEMS
INTRODUCTION
Herbivorous carp have received much attention as potential biological
control agents for aquatic plants. The grass carp (Ctenopharygnodon
IdelJla) has long been known to consune large amounts of vegetation, but
there has been continued concern over the ability of this Chinese carp to
reproduce in the United States. In 1978, a hybrid carp (female grass carp
X male bighead carp, Hypophthalmichthys nobtlls) was produced In Hungary
(Bakos et al. 1978). Many In the United States hoped that this sterile
hybrid could be used In place of the grass carp as an effective plant
control agent. More recently a presumably sterile triplold grass carp has
also been produced (Malone personal communication) with the claim that its
rate of vegetation consumption exceeds that of the hybrid and equals that
of the diplold grass carp.
Each of the aforementioned fish consune vegetation (Shireman and Maceina
1981), although at different rates (Part 2: Chapter 1). And in the veryact of removing vegetation, they affect the structure and function of
aquatic ecosystems (Buck et al. 1975, Fowler and Robson 1978, Lemnb et al.
1978, Mitzner 1978, Leslie et al. 1983). The extent to which various
ecosystem components are impacted has been difficult to assess. Field
studies designed to investigate the environmental impact of herbivorous
carp have produced many inconsistent results (see reviews by Pierce 1983,
Sh I reman 1984).
There are two principal factors contributing to the confusion about therelative impact of these herbivorous carp. First, effects on the aquatic
ecosystem are related to the extent to which plant populations are
suppressed and the degree of plant control achieved in studies of grass
carp Impact has varied widely. Second, the typically snalI nunbers of
lakes or ponds used in many studies make statistical differentiation
between natural variabili ty and fluctuations due to carp impacts extremely
3-2
difficult. Because of the high degree of natural variation between
different bodies of water and because of strong seasonal and annualvariability, highly replIcated experiments are necessary to separate
natural fluctuations from those resulting from plant managementtech niq ues.
Presented In this chapter are the results of 4 years of data collection
from a series of 31 pond experiments. The biological and chemicalcharacteristics of 12 control ponds (no treatment) are compared
statistically with a group of 19 ponds that were stocked with either hybrid
carp, grass carp, or both. Two levels of stocking are distinguished: lowdensity (LC) with rates ranging from 2.7 to 150 kg/ha in study ponds, and
high density (HC) with rates ranging from 160 to 360 kg/ha.
The unique strength of this study lies In the large amount of replication,In the range of stocking rates used, and in the variety of control levels
achieved. Our experimental design has provided us with a large data base
frcm which to draw sound statistical conclusions on the environmental
impact of herbivorous carp.
STUDY POND DESCRIPTIONS
Ponds from two separate sites were used in this investigation; the basic
morphcmetric data for each pond are sunmarlzed In Table 3-1.
Utterback Sit
Bluebird, Gar, and Turtle ponds are remnants of a former strip-mine gravel
operation located on private property approximately 9 km southeast of
Glbson City, Illinois (Table 3-1, Fig. 3-1). Gar, Turtle and half of
Bluebird are approximately 25 years old; the remainder of BI uebird was dug
In 1971-72. These ponds are completely closed systems that receive runoff
frcm adjacent banks and levees. Water levels are a function of the water
tabl e.
3-3
Table 3-1. Estimated pond surface areas, substrate areas,and vol umes. Estimates are based on plane table maps. C =control, LC = low stocking density of carp; HC = high stockingdensity of carp.
Pond Surface area (m2 ) Substrate area (m2 ) Vol une (m3 )
ClC2C304C5C6C7C8C9C10C11C12
4,3222,7742,8322,686
682690682682751689689689
4,7344,5884,344
580760740810720720730730730751751720
LC1abc
LC2LC3*LC4LC5LC6LC7LC8*LC9*LC 10*LC11LC12LC13
HC1*HC2*HC3HC4
4,4602,9022,9622,810
709718709709781717717717
4,8864,7354,473
603760790770842749760760760781781749
730730810740
760760842770
4,3693,8304,3843,480
695760695695663494494494
5,7505,0934,467
317611738561699506611611611663663506
611611699561
*Estimates of surface area, substrate area, and vol ume are means of otherstudy ponds at the Aquatic Research Field Laboratory that were used In1982. No plane table mapping was done for these ponds.
3-4
Fig. 3-1. Utterback pond site showing location of the ponds used inthis study. Pond 1 = Bluebird, Pond 2 = Gar, and Pond 3 = Turtle.
3-5
Annex Site
The experimental ponds used at this site are part of a man-made, 22-pond
complex constructed on 10 acres southeast of the Illinois Natural Resources
Studies Annex In Champaign, Illinois (Table 3-1, Fig. 3-2). Basins, built
as paired study and control ponds, were dug in 1971 and are drainable. In
June 1980, measurements by plane-table mapping indicated that surface areaswere approximately 0.20 acres. Water levels were maintained by periodicadditions of aerated city water. For more details on these study sites,see Gorden et al. (1981).
MATERIALS AND METHODS
Grass carp and hybrid carp were used In the experiments. Ponds receivingno treatment are designated as controls (Tables 3-2 and 3-3).
The Utterback ponds were rotenoned and stocked with largemouth bass,
Micropterus salmoldes (Lacepede), bluegill, Leomis macrochirus
(Raflnesque), and channel catflsh, Ictalurus punctatus (Raf nesque), In
spring 1980; ponds were rotenoned and censused In autunn 1980. The ponds
were restocked with these species in spring. 1981; Bluebird was also
stocked with hybrid carp (Tables 3-4, 3-5 and 3-6). Fish populations were
censused using Peterson mark-recapture with electrofishing and bag seining
in autumn 1981 and 1982. All ponds were completely censused by rotenoning
In autunn 1983.
Annex ponds were stocked with the appropriate rate of sport fish and/or
carp each spring and drained and censused each autunn (Table 3-4, 3-5 and
3-6). Sane Annex ponds were stocked at various rates with only herbivorous
carp (LC2, LC3, LC8, L C, LC10, HC1, HC2).
All ponds were monitored regularly fromn May through September 1980 using a
simple random sanpling design. In 1981, a stratified random samnpling
design was Implemented. The aquatic vegetation in all study ponds was
mapped as high (HMC) and low (LMC) macrophyte concentration areas.
3-6
Fig. 3-2. Annex pond site. Thirteen of the 16 larger ponds (0.20acre) were used during the 4-year study.
3-7
Table 3-2. Study pond designations by year based on treatment.
Pond 1980 1981 1982 1983
Site: UtterbacksBl uebird C1 LCla LC1b LClcTurtle C2 - C3 C4
Site: Annex1 - - LC -3 -- HC1 -4 - - LC6 LC136 - - LC5 HC37 - -LC10 -8 C5 C6 C7 C89 - - HC2 -
10 C9 LC3 LC11 LC1211 - - LC7 -12 C10 - Cll C1214 - - LC4 HC415 - - LC9 -16 - - LC2
C = control (no treatment), number assigned arbirtarllyLC = low density carp ponds ranked nunerically 1-13 according to
seasonal average carp bianassHC = high density carp ponds ranked 1-4 according to seasonal average
carp blamass
Table 3-3. Herbivorous carp stocked In study ponds in 1981-1983.
Fish
1980198119801981198019801981198119811981198019801979
1981198119811981
hybrid carphybrid carphybrid carphybrid carphybrid carphybrid carphybrid carphybrid carphybrid carphybrid carphybrid carphybrid carphybrid carp,
hybrid carphybrid carphybrid carpgrass carp
(mixed ploldy)(3N)
(3N)(3N)(2N)(3N)(3N)(3N)(3N)(3N)(3N)1980 hybrid carp (2N, 3N)
(3N), 1981 grass carp(3N), 1981 grass carp(3N)
Pond
LC1LC2LC3LC4LC5LC6LC7LC8LC9LC10LC11LC12LC13
HC1HC2HC3HC4
I
3-8
Table 3-4.Slow density
Stocking rates of sport fish (kg/ha). C = control pond; LCcarp pond; HC = high density carp pond.
Largemouth bass (mm) Bluegill (mm) Channel catfish (mm)
76-127 128-202 >203 >127 102 152 203
1980C1 0.71 - 16.90 18.40 0.99 2.50 4.40C2 0.87 -- 23.30 30.70 1.00 3.10 5.90C5 0.97 4.40 18.10 8.50 0.66 2.30 4.50C9 0.88 4.50 17.20 7.60 0.63 2.10 4.10C10 0.96 5.90 18.40 7.30 0.58 2.40 3.80
1981LCla 2.70 - 15.80 16.90 0.02 0.01 0.12
C6 3.00 - 19.10 19.70 1.70 2.90 6.80LC3 2.40 - 17.80 14.80 0.95 2.40 5.60
LC6 6.50 - 18.50 10.40 0.82 1.60 4.40LC5 6.00 - 16.00 10.20 0.83 1.50 3.50
C7 5.50 -- 13.90 14.50 0.78 1.90 4.20LC11 5.40 -- 16.70 10.10 0.83 1.30 3.70
C11 5.40 - 16.30 9.80 0.91 1.70 4.60
1983LC13 3.00 -- 11.30 9.00 1.00 2.60 4.40HC3 2.80 -- 10.70 8.00 0.64 2.20 4.10
C8 3.10 - 11.70 10.00 0.95 2.30 4.40LC12 2.80 -- 11.10 8.60 0.80 2.00 4.00
C12 3.00 - 13.40 9.20 1.00 2.10 4.60HC4 3.00 - 11.70 9.50 0.92 1.80 3.70
3-9
Table 3-5. Final census of sportfish (kg/ha). C = control pond; LC= low density carp pond; HC = high density carp pond.
Largemouth bass (In.) Bluegill (in.) Channel catfish (In.)
YOY 3-5 5-7 Breeder YOY Breeder 4 6 8
1980C1 4.1 12.3 -- 34.0 6.3 13.9 0.8 2.3 11.0C2 12.9 9.7 - 35.3 68.9 27.3 1.4 2.7 9.5C5 20.0 14.5 14.5 26.1 108.1 9.7 0.7 7.1 10.6C9 6.3 6.5 5.5 28.5 107.7 3.7 0.7 5.0 7.4C10 23.0 13.1 13.2 14.6 16.4 14.0 - 7.1 8.3
1981LCla 1 * 8.8 -- 33.8 * * - -
C6 10.0 44.9 -- 33.1 123.3 19.3 1.0 2.2 13.1LC3 74.8 33.1 -- 31.1 1.6 21.7 0.6 2.8 14.3
LC6 14.5 35.3 -- 34.3 34.2 12.4 3.6 9.9 15.1LC5 2.7 37.8 - 26.7 47.6 12.1 3.0 9.3 11.7
C7 18.3 49.7 -- 30.1 143.6 23.9 8.1 17.7 29.5LC11 16.6 35.8 -- 28.6 76.8 16.5 2.0 12.3 16.0
C11 27.6 35.3 -- 26.4 29.5 15.2 6.1 9.2 14.7
1983LC13 25.2 47.7 -- 333 14.5 16.4 9.7 13.7 14.6HC3 39.6 351 -- 25.2 55.1 15.4 9.8 13.6 17.1
C8 26.5 22.7 -- 18.2 51.9 20.6 8.9 10.6 18.8LC12 28.2 37.0 -- 26.3 102.0 20.8 9.8 14.6 21.8
C12 9.3 25.3 -- 21.8 81.6 17.3 9.7 14.5 19.5HC4 7.1 27.8 -- 20.7 10.1 15.8 11.8 16.1 18.4
1Peterson mark-and-recapture* <0.01 kg/ha
estimates
3-10
Table 3-6. Seasonal average, effective stocking, and harvest rates ofherbivorous carp (kg/ha). Number of fish stocked Is In parentheses.C = control pond; LC = low density carp pond; HC = high density carp pond.
Seasonal EffectivePond average stocking rate Harvest rate
LC1*a - 2.70 (48)b - -c - - 31.64
LC2 7.85 9.95 (25) 5.75LC3 11.82 4.65 (8) 18.98LC4 16.73 18.74 (75) 14.72LC5 17.95 11.25 (5) 24.64LC6 19.19 16.00 (5) 22.39LC7 24.20 38.27(150) 10.12LC8 31.43 6.73 (25) 56.13LC9 36.73 24.37 (75) 49.09LC10 51.15 45.10(150) 57.19LC11 91.96 59.39 (21) 124.53LC12 128.44 118.42 (20) 138.46LC13 145.31 149.00 (20) 141.61
HC1 163.39 110.76(260) 216.02HC2 188.80 119.60(217) 257.99HC3 239.34 198.57 (95) 280.10HC4 360.22 220.37 (28) 500.08
no rates*This pond was stocked in spring 1981 and rotenoned in 1983;are calculated for autumn 1981 or 1982.
3-11
Populations randomly sampled in each of these strata included macrophytes,
phytopl ankton, pi gnents, macrol nvertebrates, zoopl ankton, and bacterl a.
Water samples were collected (also by strata) to measure chemical and
physical parameters.
Aquatic macrophyte standing crop blcmass, I.e., the combined weights of
collected roots, stems and leaves, alive or dead, was measured by quadrat
sampl ng. A cylindrical hardware-cloth (5-mm mesh) sampler (0.25 m-2 area)with sheet metal support at the bottom was lowered to the pond bottom and arake was used to collect all plant material within the sampler. The samplewas then rinsed in a plastic sieve (5-mm mesh), spun dry in a salad spinner(30 revolutions), and weighed (fresh weight). In 1981 and 1983 all sampleswere sorted by species prior to weighing, whereas In 1980 and 1982 themajor species In each sample were recorded but samples were not sortedprior to weighing. Samples of each species were later dried in a forced-airoven at 65 C for 48 hours, weighed (dry weight), and ground using a Wileymill. Subsamples of the ground plants were re-dried, ashed at 500 C, andweighed (ash-free dry weights) (AFHA et al. 1976). Thus, conversion ratiosfor dry and ash-free dry weight were obtained for each species encountered.
In 1980 the number of macrophyte samples randomly collected from each ofsix study ponds was dependent upon pond size (Gorden et al. 1981). From1981 to 1983, six samples were collected from Annex ponds (3 HMC, 3 LMC)and eight from Utterback ponds (4 HMC, 4 LMC). In 1983, macrophyte sampleswere collected in ponds LC1, 04, and all ponds In which water chemistry wassampled (see below), In 1980 macrophytes were sampled 9 times; in 1981, 7times; and In 1982 and 1983, 3 to 5 times, depending on the pond site. Inaddition to estimates of standing crop, in 1982 and 1983 measurements ofplant production and mortal Ity were made using a hierarchical cohort method(Carpenter 1980). This method uses net production of shoots and subunitmortal ity to measure mortal ity and net production. Measurements were made
by enunerating shoots within a 0.25-m2 area and subsampling those shootsfor leaves and branches (subunits). Each species cohort within the samplearea was counted and 10-20 shoots were selected as a subsanmple; thoseshoots were examined for subunits present or lost, and then were separated
3-12
Into shoot, leaf, and branch subunits for weighing. For detalls onmethodology and general considerations, see Carpenter (1980). Samples werecollected according to the stratified sampl ing design in ponds C7 and LC11In 1982 and In ponds HC3 and C8 in 1983. Collections were made 5 times inC7 and 10 times In each of LC11, HC3, and C8 between 10 May and 22September.
Water samples were collected May through September between 0800 and 1400hours. Duplicate columnar water samples were collected biweekly fromrandom locations in 1980 and from both strata in each study pond in 1981.In 1982, ponds LC6, LC7, LC11, C7, and C11 were sampled biweekly and allothers were sampled monthly. In 1983, ponds LC12, LC13, C8, and C12 weresampled biweekly; ponds LC1c and C4 Were not sampled. Parameters measured
J1 sLtu were surface water temperature, dissolved oxygen concentration,hydrogen Ion concentration (pH), and light extinction coefficients; in allyears except 1983, alkalinity, conductivity, and total dissolved solidswere al so measured. Samples were preserved according to APHA et al. (1976)methodologies and transported to the INHS water chemistry laboratory forcarbon, nitrogen, and phosphorus analyses.
At the same time, repl Icate 1-Iiter samples of depth-integrated water werecollected from the euphotic zone to measure chlorophyll-sa, phaeophytin,and total corrected a plgnent concentrations. Pignents were extractedusing acetone and measured by spectrophotometer (APHA et al. 1976). Algalpignent samples were col lected all 4 years of the project.
Macroinvertebtrate fauna, both benthic and epiphytic cacnunity ccmponents,were sampled from ponds C7 and LC11 in late September 1982 and ponds C8,C12, LC12, LC13, HC3, and HC4 In late September 1983. Samples were takenat the end of each seasonal experiment to examine the cumnulative effects
of herbivorous carp. Four bottom and four macrophyte samples were taken in
each sample pond at every collection. Benthic samples were collected with
a piston corer (area sampled = 0.0036 m2) and extruded Into quart jars;
macrophytes and associated epiphytic macrol nvertebrates were col lected with
3-13
a 300-um mesh dacron net attached to a quart Jar. All samples were field
preserved with 80$ ethanol.
In the laboratory, macrolnvertebrates and detritus In core samples were
separated from Inorganic material by sucrose floatation (Wetzel and Wiley
1981). The resulting supernatant was poured through a 300-um mesh sieve;
retained organisms were then removed from detritus under a stereo-
dissecting microscope. Organisms were Identified to the lowest positive
taxonmnic level using current literature, enumerated, and wet weighed.
Results were expressed In nunber or milligrams of organisms per square
meter. Epiphytic samples were processed In three steps: (1)
macroinvertebrates were rinsed frcm plant material Into a 300-um mesh net.
The sample portion retained in the net was then examined under a dissecting
microscope. (2) A homogeneous subsample of the remaining plant mass was
taken and all adhering organisms removed under a dissecting microscope. The
total fresh weight of the plant mass was used to extrapolate totals for
adhering organisms. (3) Counts and blomass determinations from steps (1)
and (2) were combined to yield sample totals expressed In numbers or
mill grams of organi sms per gram of plant material. These total s were then
converted to nunbers or mill igrams of organisms per square meter of pond
bottom for canparison with sediment-associated samples.
Zoopl ankton was sampled w ith a submersibl e f I ter-pump apparatus (Walte andO'Grady 1980). Duplicate samples were collected randomly in 1980 and ineach strata in six study ponds In 1981. In 1982, ponds LC1b, LC6, LC7,LC11, C3, C7, C11 were sampled, and In 1983 ponds LC12, LC13, C8, and C12.An alcohol-formalin solution tinted with rose bengal was used to fix andpreserve the samples in the field. Taxonomic, gravimetric, andinterpretive analyses fol lowed those in Gorden et al. (1981).
Bacterial samples were collected in all study ponds In 1980 and 1981, in
ponds LC6, LC7, LC11, C7, and C11 in 1982, and in ponds HC3, HC4, C8, and
C12 In 1983. In 1980 and 1981, water and sediment samples were collected
biweekly; macrophyte samples were collected biweekly beginning In August
1980. In 1982, water, sediment, and macrophyte samples were collected on 1
3-14
June and 30 September only; in 1983 water samples were collected 1 June
and all three types of samples were collected on 1 and 20 September. An
autoclavable water col un sampler, a corer and sterile stainless steel
tubing, and sterile forceps were used to collect water, sediment, and
macrophyte samples, respectively. Standard plate count techniques were
used to measure bacterial populations. Methods of col lection, processing,and counting were consistent throughout the study. All results are
expressed as colony generating units (OGU) per unit sample. For more
deta iled descri ption, see Gorden et al. (1981).
Sedimentation rates were measured in control, lcw density, and high density
carp treatment ponds (C7, LC11, HC3, HC4, LC6, LC12, C8, C12). Sediment
traps collected seston and reference chambers were used to determine the
amount of attached and suspended materials. Traps and chambers were
constructed out of wide-mouth quart Jars and plexiglass. The Jar bottoms
were s I Iconed together with pl exigl ass in between. Four sets of traps and
reference chambers were suspended at a depth of .0.5 m below the waters'
surface or 0.5 m above the pond bottcm for 5 days. At the end of the
Incubation, the Jars were capped and the contents of each Jar were
homogenized and filtered through a pre-dried glass fiber filter (Gelman
A/E). The filters were then dried (105 C) for 24 hours and weighed (dry
weight). Some samples were ashed at 550 C (APHA et al. 1980). Nine
experiments were conducted in 1982 and three in 1983.
ExferImental Desgn
The 31 pond experiments were grouped according to treaiment: control (CN),
low density (LD), and high density (HD) carp ponds. Ponds stocked with
herbivorous carp were separated into LD and HD treatments by the seasonal
average blamass of carp (Tables 3-2 and 3-6). Grass carp and hybrid carp
have different consumption rates (Part 2: Chapter 1); therefore, to
accurately assess Impact with the different genetic types, all carp blcmass
is expressed In grass carp equivalents based on their differential feeding
rates. bMst statistical analyses were appliled to these three groups and
Included analysis of variance (ANOVA), B posterlorl tests, Kruskal-
3-15
Wall s test, and regression analyses. Usually SNK (Student-Newman-Keul s)procedure was used to test for differences between groups a posterlorl.
The SNK test uses range as a statistic to measure differences among means
(Sokal and Rohlf 1969). Scheffe's test, another a posteriodr test, is acomputation of confidence Intervals of variance and was used in analysis of
macro nvertebrate data (Sokal and Rohlf 1969). When standard analyses of
variance were of questionable validity due to a lack of hanogenlety of
variances (determined by Cochran's C and Bartlett's tests), the Kruskal-
Wall s test (nonparanetric analyses of variance by ranks) was also run to
test for inter-trealment differences; its results (p and H) are given withANOVA tables. Results are reported at p<0.05 unless otherwise stated.
Caomunity diversity indices were calculated for benthic data according toShannon (1948).
RESULTS
Impacts oD Macrophyte Populations
Macrophyte standing crops in the study ponds ordinarily peak in June,decline In late summer, and recover In autumn. This pattern represents thenormal development of Potamogeton crispus and NaJas flexi1ls populations(Fig. 3-3) and was usually exhibited by the control ponds (Fig. 3-4). Asexpected, herbivorous carp altered that pattern by reducing the summer
and/or autumn bianass levels (Table 3-7). Peak biomass was not
significantly different In control, low density, and high density pond
groups. However, average final standing crop bicmass In the high densityponds was significantly lower than that in low density and control pond
groups (Table 3-8), indicating that carp significantly reduced standing
crop biomass.
The ratio between peak and final (autumn) blcmass in a given study pond can
be used as an Indicator of the amount of control achieved by herbivorous
carp. Control ponds averaged 78% of peak bicmass In September, low density
ponds 70.5%, and high density ponds 5% (Table 3-9, Fig. 3-5). Seasonal
patterns of macrophyte blanass were variable, even in control ponds. There
am-m x m 30 c amr -4 4 -4 c
II
0
o 0O0C.Q6u) *C
to
0
t-
CO
0
3h
0.
Q
(0
000
0
C
O
-+
0
U)
O-
-I
IU,
(n
u,
3-16
U'
L--r€.n
OQO.
8
. C
C0 e
0C -
U)
0 a
O OLIn ,
-C.D 0OL O LL CJ .D- 0( Ln 0 E
0Q)
3-17
oC
XC
inr^
3-18
Table 3-7. Peak, final, and seasonal average standing crop bcanass(g dry weight m-3 )
Final Seasonal Final blaoassPeak (September) average as % of peak
Pond b anass* b cmass b anass bcmanass
LC1a 173.0 65.3 113.8 38b 112.6 25.3 76.1 22c 172.8 152.3 124.0 88
LC2 132.7(May) 54.9 79.6 41LC3 131.3(Sep) 131.3 95.6 100LC4 111.5(Sep) 111.5 47.4 100LC 68.2(Sep) 68.2 47.1 100LC6 93.3 68.0 73.8 73LC7 18.4 4.5 11.9 24LC8 81.5(Sep) 81.5 56.4 100LC9 120.0 94.7 100.3 79LC10 98.6 17.1 42.8 17LC11 139.9 78.7 90.8 56LC12 169.6 120.2 94.5 71LC13 145.5 123.7 82.9 85
HC1 17.7 0 8.7 0HC2 104.4 0 42.6 0HC3 137.5 28.0 52.8 20HC4 134.6 0 37.7 0
Ct 74.7 72.7 60.0 97C2 116.9 94.4 76.4 81C3 210.2 112.2 140.2 53C4 218.0 49.3 99.4 22CS 131.6 68.4 84.1 52C6 130.8(Sep) 130.8 81.5 100C7 69.0 37.7 40.4 55C8 132.0 131.6 80.4 100C9 117.4(Jul) 95.6 74.8 81C10 68.4(Sep) 65.6 44.1 96C11 72.3(Sep) 72.3 54.5 100C12 152.4 149.2 87.4 98
*Peak bianass in June unless otherwise noted.
Table 3-8. Comparison of final standing crop In control, lowdensity, and high density ponds.
Analysis of Variance
Source d.f. SS MS FS
Treatment 2 21525.8 10762.9 7.445Error 28 40477.8 1445.6Total 30 62003.5
p = 0.005
SNK Results
Control Low Density High Density
90.0±35.1 - 79.8+43.412 15
N 12 15
7.0±14.0*4
Mean+S EN
3-19
Table 3-9. Seasonal average plant biomass, final blonass as apercentage of peak blonass, and stocking rates of carp for all study.ponds.
Stocking rate of herbivorouscarp (kg/ha)
Seasonal ave. Final plantplant blonass Seasonal blamass as 5
Pond (g dry wt m-3) average Stocking Harvest peak biomass
LC1abc
LC2LC3LC4LC5LC6LC7LC8LC9LC10LC11LC12LC13
HC1HC2HC3HC4
ClC2C3C4C5C6C7C8BC9C10ClC12
113.876.1124.079.695.647.447.173.811.956.4
100.342.890.894.582.9
8.742.652.837.7
60.076.4
140.299.484.181.540.480.474.844.154.587.4
*
7.8511.8216.7317.9519.1924.2031.4336.7351.1591.96
128.44145.31
163.39188.80239.34360.22
2.7
9.956.734.65
18.7411.2516.0038.2724.3745.1059.39
118.42149.00
110.76119.60198.57220.37
*
*
31.645.7556.1318.9814.7224.6422.3910.1249.0957.19
124.53138.46141.61
216.02257.99280.10500.08
38228841
1001001007324
1007917567185
00200
9781532252
10055
1008196
10098
*Carp not stocked or censused
Table 3-10. Cmaparison of peak and final standing crop in control,low, and high density ponds.
Analysis of Variance
Source SS d.f. MS FS
Subgroups 53030.45 5 10606.09Carp 8753.00 2 4376.50 2.376 nsSeason 29491.05 1 29491.05 16.011***Interaction 14786.4 2 7393.2 4.014***
Within subgroup 103146.45 56 1841.90Total 156176.90 61 2560.28
u~cn C ro O o o> 3 s o nO*
* N o V s A A I I w e o w uC* C3 c C3 C3 C* C C2 C C C CI (0
- I
(0u,
0O
1 -,* .
0
NO
|I
-7
i a11 1
_ -0
'T (
*n aH^ C
3-20
0
,I
'i-
I-
-=
-4
-0m
"I"
r-
On
C,z
(n-4-4 ]-
I
I
19I
3-21
were years when peak bicmass occurred in autumn rather than in mid-smumer
(Table 3-7), generally the result of poor development of early cohorts of
&. crispus. A two-way analysis of variance comparing peak and finalbicmass (season) for the three pond groups showed a significant effect dueto season and a significant interaction between season and carp treatment(Table 3-10), again Indicating that the carp significantly affect normalseasonal biomass fl uctuations by reducing the average sunmer/autumn ratio.
The level of macrophyte control achieved was not directly proportional tothe nunber of carp stocked. Many variables affected the amount of plantblacass standing at the end of each experiment besides herbivorous fish,Including water temperature, plant species productivity and overwinteringbiomass, and shading by phytoplankton. Each of these factors or anycombination may complicate the relationship between stocking level andlevel of plant control. It is not surprising then, that In low stockingdensity ponds, carp provided highly variable amounts of plant control(Table 3-9), whereas high dens ty ponds produced more consl stent macrophytecontrol. Three high density treatment ponds (HC1, HC2, and HC4) wereconsidered over-grazed, because macrophyte standing crop fell to 0 duringthe experimental period (5 months) (Table 3-9). In the other high densitypond (HC3), the macrophyte standing crop was reduced over the season butnot totally el iminated; about 20% of peak biomass was standing InSeptember.
Production Studies. Production estimates provided a more detailed pictureof herbivorous carp feeding on aquatic macrophytes. Production studies wereconducted In ponds C7, LC1, CB, and HC3. The two major macrophyte speciesin those ponds were Nalas flexl lls and Potamogeton crIspus. Macrophytemortality due to herbivory can be estimated by comparing welght-specificdifferential mortal ity in adjacent treatment and control ponds. In .E
crLspus populations, the relative timing of mortal ity was altered I Ittle by
carp but mortal Ity Increased (Fig. 3-6 and 3-7). In the case of JL
fI exlLls, carp not only increased the mortal Ity but also changed the
mortal ity schedule by increasing the proportion of mortal ity occurring very
early in the species' life cycle. This effect was evident in HC3 but not
3-22
& -"< r»-»-- r* O X 0 < 0r"Clo 6-< 3 ac m
0<
3 -
"Io
0O
O
0 0
0
01
= M
-4o
10
O O*1
0.^
0,
0.4
3-23
q-O
C
Q.
0
L.
0
L.?" L.I OE o
C C
-J 0C
0)
-040
4 --
8C
> 04 -
0
-C
a
( 3 E - .04-P- > C -O L -:0 L -)
3-24
In LC11 (Fig. 3-8 and 3-9). In the high density carp pond, the Impact on
& fl.exiLls was greater than that on ., crispus, probably due to
temperature and preference effects. PE crlsous is a cool-water species and
begins growing early in the season before carp start feeding, making It
less likely that carp will control its early growth. Also, NL ifl&exl.s was
preferred over L. crLspus, Increasing the I kelihood that &L flexills willbe affected more than & crlspus,
High macrophyte production In LC11 masked the effects of carp In both
production and standing crop data. Seasonal growth rates are an Indicator
of how productive a plant population Is and those rates vary between ponds.Seasonal growth rates (g dry weight m-3) of E. crlspus in LC11 and its
adjacent control (C7) were 1.44 and 0.24, respectively. LN flexi 1s growth
rates for those same ponds were 1.79 and 0.07. In the case of the highdensity and control ponds studied In 1983, growth rates were higher in the
high density pond; 26.6 versus 3.66 g dry weight m-3 for P crlsaus In HC3and CB, respectively. Coaparison of standing crop estimates, which would
normally reflect those growth rates, show that the the peak and finalbiamass in the low density pond (LC11) was higher than that In the adjacent
control pond. Plant biamass levels were too high for the carp to control.In the case of HC3, the opposite was true; despite large growth rates and
initial blamass levels, control was attained.
Water Chemistry
The Impacts of using herbivorous carp on water qual Ity parameters were
mostly indirect, resulting primarily frcm the removal of macrophytes fromthe pond and a reduction in community photosynthesis. In ponds stocked at
lower densities, there were smae significant, although not necessarilydranatic, differences In water chemlstry. For 12 of 20 (60%) parameters
routinely measured (Tables 3-11 through 3-15), low density treatments had
mean concentrations significantly different frcm those in control ponds.
Differences were mainly attributable to reductions in plant productivity
and Included sl ight reductions in daytime pH and total and dissolved
carbon; and increases in alkalinity, carbon dioxide, nutrient (nitrogen and
3-25
%4-0
C
U,04
4-
L.
don%
00
E c
' e
4-
>bEC
< ~ e-
C.0)
u,0
*oQII
0O
L) E - 04-)> C M:O 0 L) V0-P-o4 >
3-26
Sr o 3 D < 0-Q 3 C0
5"8CO c
0-
0<
0 6
03=
T,-O
0
S S
Q*b
U,
S
0
3-27
Table 3-11. Results of SNK analysis for laboratory chemistryparameters (mg/L) in all ponds. Treatment means (± standard error) aregiven, with the number of samples in parentheses. A dash (-) Joins groupswith no significant difference and an asterisk(*) Indicates significantlydifferent treatment.
Low HighControl Density Density
n=216 n=196 n=46
Total carbon 27.36+0.77* 19.63+0.69 - 18.18+0.98
Particulate carbon 2.95±0.18 - 2.42+0.17 - 3.30,0.40
Dissolved organic 21.61±0.72* 13.240.43 - 14.88+0.73carbon
Inorganic carbon 23.67±0.51* 26.61+0.67* 35.38+1.85*
Total phosphorus 0.15+0.07 - 0.09±+.00 - 0.10±0.01
Soluble 0.05+0.00 - 0.04+0.00 - 0.06+0.01orthophosphate
Nitrate 0.04±0.00* 0.03+0.00 - 0.03+0.00
Ammonia 0.17+0.01* 0.10+0.01* 0.14+0.01*
Analysis of variance in appendix tables A3-1 through A3-6.
3-28
Table 3-12. Results of SNK analysis for laboratory chemistryparameters (mg/L) in Annex ponds. Treatment means (± standard error) aregiven, with the nunber of samples In parentheses. A dash (-) Joins groupswith no significant difference and an asterlsk(*) Indicates significantlydifferent treatment.
Low HighControl Density Densityn=160 n=158 n=46
Total carbon 26.20P0.85* 18.71±0.73 - 18.18±0.98
Particulate carbon 2.77±0.20 - 2.61±0.21 - 3.300.40
Dissolved organic 19.65+0.76* 13.51±0.51 - 14.88+0.73carbon
Inorganic carbon 22.50±0.50* 28.38±0.71* 35.38±1.85*
Total phosphorus 0.08+0.00 0.09+0.00 - 0.10±0.01
Soluble 0.04±0.00 - 0.04±0.00 0.06+0.01*orthophosphate
NItratea 0.04±0.00 0.03±0.00 0.03±0.00
Anmonlab 0.16±0.01 0.11+0.01 0.14±0.01
Analysis of variance in appendix tables A3-7 through A3-13.
a Low density and control ponds not significantly different from highdensity ponds control significantly different from low density ponds.b Control not significantly different fram high density ponds; lowdensity ponds significantly different from control and high density ponds.
3-29
Table 3-13. Results of SNK analysis on pignents (mg/L) for al Iponds and for Annex ponds only. Treatment means (± standard error) aregiven, with the number of samples In parentheses. A dash (-) Joins groupswith no significant difference and an asterisk(*) Indicates significantlydifferent treatment.
Low HighControl Densi ty Density
ALL aondsChlorophyll 11.90+0.84 - 12.67+0.87 - 15.51±1.48
(382) (456) (100)Phaeophytin 3.23+0.29 - 3.12±0.58 - 5.22±0.42
(382) (456) (100)Total chlorophyll 22.11±1.74 - 21.42±1.68 - 26.01+2.27
(228) (424) (100)
Annex ponds onlChlorophyll a 7.32±0.54* 12.01±0.97* 15.51±1.48*
(282) (372) (100)Phaeophy t n1 2.310.15 3.23±0.71 5.22+0.42
(282) (372) (100)Total chlorophyll 1 15.24+1.15 20.46±1.92 26.01±2.27
(188) (356) (100)
Analysis of variance in appendix tables A3-14 through A3-16.
1 Control pond not significantly different from low density pond;density pond not significantly different from high density pond;two groups were significantly different.
I owthose
3-30
Table 3-14. Results of SNK analysis for field chemistry parameters(mg/L unless otherwise noted) in all ponds. Treatment means (± standarderror) are given, with the number of samples in parentheses. A dash (-)Joins groups with no significant difference and an asterisk(*) Indicatessignificantly different treatment.
Low HighControl Densi ty Density
Surface watertemperature (C)
Surface dissolvedoxygen
Bottom dissolvedoxygen
pH
Free carbon dioxide
Alkal Inity
Total dissolvedsol I ds
Specific conductance(unhos/cm)
Vertical extinctioncoefficient 1
23.93+0.30(205)
8.55+0.15(207)
5.66+0.25*(206)
8.83+0.04*(184)
0.93+0.18*(168)
121.19+2.56*(178)
228.96+2.54*(168)
308.35±3.52*(168)
2.27+0.09(104)
- 23.12±0.27(215)
- 8.41+0.16(221)
6.36+0.22*(219)
8.48±0.05*(186)
2.86+0.35*(178)
144.23±3.60(178)
244.35±3.21*(178)
332.68±4.82*(178)
2.07+0.05(199)
- 24.15±0.57(44)
6.94±0.37*(44)
3.53±0.49*(44)
7.75+0. 12*(22)
9.02±2.03*(10)
- 151.00±10.49(20)
287.00+2.84*(10)
388.70+4.05*(10)
2.53+0.13(48)
Analysis of variance In appendix tables A3-17 through A3-24.
1 Low density pond significantly different frcm control and high densitypond.
3-31
Table 3-15. Results of SNK analysis for field chemistry parameters(mg/L unless otherwise noted) In Annex ponds. Treatment means (+standard error) are given, with the number of samples In parentheses. Adash (-) Joins groups with no significant difference and an asterisk(*)Indicates significantly different treatment.
Law HighControl Density Density
Surface watertemperature (C)
Surface dissolvedoxygen
Bottom dissolvedoxygen
pH
Free carbon dioxide
Alkalinity
Total dissolvedsol Ids
23.94+0.34(145)
8.41±0.17(147)
6.07+0.29(146)
8.85+0.06*(130)
1.03+0.25*(118)
117.54±2.48*(128)
216.63±2.24*(118)
- 23.310.27(173)
- 8.12±0.17(179)
- 6.68+0.24(177)
8.44+0.06*(150)
3.17+0.42*(146)
149.09±3.88(146)
242.87+3.76*(146)
- 24.15±0.57(44)
6.94±0.37*(44)
3.53+0.49*(44)
7.75+0.12*(22)
9.02±2.03*(10)
- 151.0010.49(20)
287.00±2.84*(10)
Specific conductance 291.20±3.12*(umhos/cm) (118)
331.34±5.69*(146)
388.70±4.05*(10)
Vertical extinctioncoeffici entl
2.25±0.09(84)
2.08+0.06(171)
2.53+0.13(48)
Analysis of variance in appendix tables A3-25 through A3-32.
1 Control not significantly different from low density pond; control notsignificantly different from high density pond; control and low densitypond significantly different frcm high density pond.
3-32
phosphorus) concentrations, and total dissolved sol ids. There were also
same increases In chlorophyll concentrations, Indicating a response in thephytoplankton popul atons.
In high stocking density ponds, similar but more dramatic trends wereobserved, with 90% of the measured water quality parameters showingstatistically significant differences from the control group (Tables 3-11through 3-15). The most important differences were decreased dissolved
oxygen and increased carbon dioxide concentrations (Fig. 3-10), planktonicalgae, turbidity (Fig. 3-11), and increased nutrient concentrations (Fig.3-12).
Of particular Interest with respect to fisheries productivity is theobservation that dissolved oxygen levels were significantly decreased incarp treatment ponds (Tables 3-14 and 3-15). In analyses of surface and
bottom dissolved oxygen, control ponds had the highest levels, with
progressively lower levels In low density and high density ponds. Reduced
dissolved oxygen resulted in part from reduced macrophyte populations andIn part from Increased sedimentation. Although phytoplankton populationsincreased, photosynthetic production was probably reduced by increasedturbidity and concomitant increased vertical light extinction (Tables 3-14and 3-15, Fig. 3-13).
Within the group of high density ponds, there were three overgrazed pondsand one (HC3) in which macrophyte biomass was reduced but not eliminated.Comparing responses within the high density treatment, It is apparent that
total elimination produced much more dramatic changes than a major but notcomplete reduction of macrophyte biomass. Total chlorophyll in over-grazedponds was three times that of HC3 In August (Fig 3-14). Dissolved oxygenconcentrations dipped to much lower levels in HC4 than In HC3 (2.6 and 4.7
mg/L, respectively) (Fig. 3-15). Vertical extinction coefficient of light
(a measure of turbidity and related to total chlorophyll), total dissolved
solids, and sedimentation were markedly Increased In HC4 (Table 3-15, Fig.
3-16).
Co
sxLD^_j
3-33
CJ0
C
CiC)
cJ
ffl i11--C'KlJ
uo
4)
C--D
Cm
-0I-
0_J
0L.
U)C
LC0
0j;-
0
0
O
Sc
U)
L -
(D-C
.0)
0
OW
C
"SQC
cs00N
0L
S1C0OI
0
I I I I r
3-34
00
0
-I
-- "o -0 -* R) -aW C \ r0
CL
C oD
x oct
5 o
00 -,
O
o
w -+ -
**
4
o m
00 0
4 §3
S 2.
O c
I I0 :3
01 C1
1-.
mxCIO
bode:3
0 *
0
DO
ro
:3
-O0
4-)
-c
0)
O1
U)C)0
-J
LO 0
3-35
"00L
4JU
0a)§^OT
C30c0 o
c0.QoL
SOo r
lr••
-.J
E
c0L
u
0 -.0) C
4- e
C
E--
0)
0X C
0
0-c
I 8C
c·SU)
LC0
C
I
LOOJ
0L
0u
jNVL0
< CT-o.1-0-- C -4
-C
NJ unno e
w Ln
(n
OU)
Q.
8 SD 0
Q.
•0-t-0
-4--
> 0
0s.
OD
0%0
00
0.0
(0
cn«t
3-36
0
O n
mx
c-)
03
(O
0
<-»
c-,C3
CD
(m-<
CC
CU)
II
<4-00--------- I I"II r ¢tUtri X 3\\\\\\\\\\\\\\~I3
0(D 3~ r (r0 3 )
y•y1 y/^v/v/v/10 P"^,t==i
IerSrrrrrrr re rrrrrrr
^^^^^________
i7.7 7 'r J11 4i4i 4 lz o .0- Z li I l Z ! 11 iA 4 I-A .0 l F I U B ,A-I I I ,8,
IL
3-37
L-o.1E
4J
U)
U)- •,
E '
CLUL)CO=O4
LO Ch
Q, V'L
L U)..-C- LC)-• K
0*.-.>
0
>.-cJ,O.0L0
-cL)
4)0p-
C
Go
-c
0)
U)0
-.0
U)
5,-
CL- U
ao
(D
4- c
CO
OC C
'-0
4 On
Q)
L
(00r 1r 0
cD0(0
0rq<r
3-38
C,0M N o *n UI -f (- O
0 (A
C 0
1
00
OOD
NMI0
0
CL
(I
4 <
.0 4b
0I-s
Co
-4.
Or
(/» g
I I t~
-oL
I1
I
0
I2
va
><Sm
SD
EX
3-39
L
C8
L
c L
.1-') 06c
4-0)
CD
S-C
- €
0 Q
00
-£^
0) C
UO)C -^
Ln) o Ln o n o L O
CM C-C
- c
0)
!C
[
-A 0a
C: ucc [O Io
CO
- C L.
c
I I I I
3-40
Zooolankton
Zooplankton appeared to be more abundant in high density carp ponds (Tables
3-16 and 3-17). When all ponds (both sites) were Included In an analysisof variance, few differences between treatments were evident (Table 3-18);however, zooplankton abundance was generally higher at the Utterback sitethan at the Annex site. If only Annex ponds were compared, significant
differences in rotifer abundance and biomass were found between treatments
(Table 3-18). Naupl II abundance and blomass showed a significant
difference between pond groups in all analyses. Increases in naupl llrepresent added reproduction, probably In response to the Increased
phytoplankton populations in both low density and high density carp ponds
(Table 3-18). Groups of zooplanktors that did not Increase In carp
treatment ponds may have been controlled by Increased rates of fish
predation. Larger zooplanktors are preyed upon first (Hutchinson 1975),
and this differential util Ization of the food resource may have skewed the
size distribution of zooplankton populations In the carp ponds. For
example, In September 1983, rotifer abundance and blomass were
significantly higher In high density ponds; together with naupl I1,
rotifers comprised 61% of the zooplankton community bianass in high densityponds as opposed to only 33% In control ponds.
Sedimentat on
In control ponds, sedimentation rates (g dry weight m-2 per day) remainedfairly stable throughout the season (Table 3-19, Fig. 3-16). However, inthe In low and high density ponds sedimentation progressively increasedthroughout the summer. Treatment categories differed significantly withrespect to the sedimentation rates, particularly in September when thelargest effects due to carp would be expected (Fig. 3-17). Clearly, carpincreased rates of sediment fal I and that effect was very dramatic when the
carp were overstocked. Sedimentation rates in pond HC4 increased 7-17
times that in other ponds, primarily due to resuspenslon of material and
fecal output. Without rooted macrophytes to stabl ize the sediments,
sediments can be disturbed and material resuspended. Movement of sediments
Table 3-16.error) In control,given as n.
Mean zooplankton abundance (number m-3 ) (+ standardlow and high density carp ponds. Nunber of samples
Control Low High
AlI Annex AllI Annex AlI(Annex)n=297 n=217 n=151 n=127 n=16
Naupl I 58,107 57,717 74,802 79,387 152,361(±3,426) (+4,020) (+6,287) (±7,200) (+27,000)
Copepod da 16,622 14,939 11,528 11,030 24,405(±1,800) (+2,153) (±1,796) (+2,078) (+6,838)
Copepoda 12,017 12,358 17,603 18,675 3,267(±1,076) (+1,260) (±2,180) (±2,549) (±820)
Cladocera 75,319 96,772 99,450 108,078 79,570(±12,000) (+17,000) (±16,000) (+15,000) (±22,362)
Rotifera 252,876 105,774 319,094 300,718 252,046(±32,000) (±13,000) (±77,000) (+83,000) (±51,000)
Total 414,942 287,561 522,477 517,887 511,648(±33,000) (+23,000) (±76,000) (+82,000) (±66,000)
All = both ponds sitesAnnex = Annex study ponds only
3-41
3-42
Table 3-17. Mean zooplankton bianass (mg m-3 ) (+ standard error) Incontrol, low and high density carp ponds. Nunber of samples given as n.
Control
AlI Annexn=297 n=217
Naupl i 82 81(+4.9) (±5.7)
Copepodida 29 27(±3.2) (±3.9)
Copepoda 49 44(±5.9) (±6.0)
Cladocera 132 168(+18.3) (±24.2)
Rotifera 83 34(±10.5) (+4.2)
Total 375 354(+23.7) (±29.7)
All = both ponds sitesAnnex = Annex study ponds only
Low
AlI Annexn=151 n=127
106 113(+9.0) (+10.3)
21 20(±3.2) (±3.7)
51 50(±6.7) (+7.4)
191 177(+24.2) (±24.0)
106 100(±25.8) (+28.0)
475 460(±36.7) (±39.2)
High
Al I(Annex)n=16
218(±38.4)
44(+12.3)
14(+4.1)
135(±33.6)
80(±17.5)
491(+62.9)
--
3-43
0C- U
o .- ---
8.N cON e
Z U
0)I L
C C
--
> L.
0
U) U -
Sc c
a 0
*C .
V, -
XL L4
C <c0)4--h
000?
+0
0
04
0
turn
N
Ao
m
10
CM
L
&-CM r
00
C
0-
8 I
CO
.0
aC0
O
II
*
+1
10*
o
r E
4-
*
-I
*
-5 0-I
0
Ln
Q
II
CV
00
0*
CM
CN
Sr
"+
O X
C ^
*+
0
O
SN
LEK
0i
-
-i--
(00
*0
m -
*IGO
ra ^-
0 4
0-
* r94I
IA XU
. *«
0E
- 0)
COaC0(
r0
II
4-
0
0-
a
0 ^
I
10
C
0
CM
CO
0
+I1In
0
*E(D E4-O02
(0 Q
4-
-C4-O
m50)
i
Vm
CO
0,
L
m'4
-c 0
m
L
c c
8)
0.
L
O
O 0
I0)
O
I-J
4-
C8
00O
X
04-
x
C
a
(0
c
'-
O0
's-5
3-44
Table 3-19. Sedimentation rates (g dry weight m- 2 day - 1 ) incontrol, low and high density ponds. All data are means of fourrepl cates.
Seasonal averagePond carp bicmass June July September
C7 -0.48 0.72 0.99C8 0.91 0.81 0.53C12 - 0.29 0.52 0.39
LC6 145.31 - - 1.29LC11 128.44 0.21 0.28 3.11LC12 91.96 - - 3.15
HC3 239.34 3.49 2.53 3.26HC4 360.22 4.23 9.51 22.19
Table 3-20. Oxygen deficit (calculated as summed differences fromsaturation).
Cumulative Cunulative early Average carpPond daily deficit morning (5 am) deficit) density (kg/ha)
CB -8.4 190.2 0C12 -117.7 48.1 0LC12 -199.1 128.44LC13 -194.8 145.31HC3 15.4 234.4 239.34HC4 168.4 395.1 360.22
Pearson's r 0.875 0.884
3-45
4 •
U,
7'
O .-
-. - U 0
*O 3
Q>
7rtL O
0
Do z
0 Q
0 ®
8"
TO -
7%
0r.4
rm
3 >-
01
I
.- -
0
(O
(n
3-46
Into the water could have been caused by activities of fish as well as wave
action. In HC4, there was almost certainly continuous resuspenslon of
materials during August and September. In the other treatment ponds, where
the aquatic plant population was not completely removed, we also saw
Increases in sedimentation rates but these Increases were not as dramatic.
The principal concern over these increases in sedimentation Is that, as
turbidity of the water Increases, there Is less I Ight available for
phytoplankton photosynthesis, causing a reduction In the amount of oxygen
produced. At the same time, sediment resuspenslon dramatically increases
sedimentary respiration (mostly microbial) and puts additional demands on
available oxygen. Together these two processes can result in serious
reductions in dissolved oxygen and Increases in dissolved carbon dioxide.
This "overgraz Ing syndrome" is readily apparent in seasonal oxygen deficits
(Fig. 3-18 and 3-19, Table 3-20) and oxygen/carbon dioxide ratio (Fig.
3-10).
MicrobIal Populations
OGU in sediments were significantly different between the three treatment
groups (ANOVA); sediments In high density ponds had significantly more
CGU than control and low density ponds (SNK analysis) (Table 3-21),
probably due to Increased fecal output of the carp. The feces normally
settle to the sediments, Increasing the amount of organic matter available
to bacteria at the water-sediment interface. Analysis of OGU on macrophytes
showed no significant difference for the three pond groups. However, by
reducing macrophytic material, herbivorous carp reduced epiphytic bacteria
populations. When OGU In the water col unn was compared in control, low
density, and high density ponds during the study, no significant difference
was found. However, in 1983 high density carp ponds (HC3 and HC4) had
significantly higher OGU in the water col unn than did control ponds (C8 and
C12) (Table 3-22), primarily because the overstocked pond contained the
highest level of suspended material (Fig. 3-16). Suspended material serves
as substrate for bacterial colonization and can account for the elevated
bacterial numbers In the water column of the overstocked pond.
3-47
.-
8
Cd*
S > m6. o ?3Ln Ca
0 01 CD 10cc U3
U1
U--
0
a2u-
U'--I1-
Coal0
Co
CD
0.
-9(0
CDI--
r-
0
0
0
"1
~1
C-,
ac
P
to
U -
0-C26.
--
0
I=o'1
0LUS
a0
I I I c I I
2- -J Z:-5- Cw:
-o
4
IOta In ' ui,
3-48
I-
LLJ
LJ
> -
_j»-
ca
-a
-m
C2
C4
--"11'p^
(0
0)
.4-
0)
C
L
-J
*0
(0
+-
C
8)
i
cn
I IIr
3-49
Table 3-21. Comparison of sediment G0U In control, low, and highdensity pond groups. Treatment means (± standard error) are given, withthe number of samples In parentheses. A dash (-) Joins groups with nosignificant difference and an asterisk(*) Indicates significantlydifferent treatment.
Analysis of Variance
Source
BetweenWithinTotalp = 0.000Kruskal Wall is
d. f.
2311313
SS
4.5x10137.0x10 14
7.45x10 1 4
2.25x10132.25x1013
p = 0.000, H = 24.855
SNK Results
Low Density
1.8x10 6
(96)
High Density
2.9x106*(24)
Table 3-22. Comparison of water OGU in two high density ponds andtwo control ponds In 1983. Treatment means (+ standard error) are given,with the nunber of samples in parentheses. A dash (-) Joins groups withno significant difference and an asterisk(*) indicates significantlydifferent treatment.
Analysis of Variance
Source d.f. SS
BetweenWithinTotalp = 0.013Kruskal Wallls p =
18687
1.55x10 8
8.62x10118.72x10 1 1
1.55x10 8
2.40x10 7
0.084, H = 2.995
SNK Results
Control High Density
6.0x104 + 1x10 4
(40)3.4x104 + 4.4x103*
(48)
10.02***
Control
1.5x10 6
(194)
6.46*
--
3-50
Herbivorous carp are capable of altering aquatic bacterial populations,
particularly those in the sediments, and thus may alter functional
processes of decomposition and nutrient cycl ing. OGU in sediments
generally increased in high density carp ponds. The alterations in
bacterial populations as a result of herbivorous carp are among the more
subtle impacts observed. At this point It is not reasonable to predict
Iong-term impacts of herbivorous carp on aquatic bacterial populations.
Macro nvertebrates
Infaunal Benthos. Both abundance and bianass of the Infaunal organlsms/m2
were significantly greater In control and low density stocked (<150 kg/ha)
ponds than in ponds stocked at high densities with herbivorous carp (Tables
3-23 and 3-24). Densities of 5 of the 11 major taxonanic groups collected,
as well as the mean nunber of taxa per sample and community diversity,
decreased as stocking density increased, although many of these decreases
were not significant due to large variances. Statistically signif cant
decreases In numbers and bianass of Infaunal Chironomidae (non-biting
midges) with In high density ponds Indicates that sedentary tubicul us filter
feeders and detritovores were particularly affected by high levels of
herbivorous carp. Of all the major groups, only the Nematoda (round worms)
Increased nuner cal ly as stocking density Increased. Several groups,
Including aquatic Acarl (mites) and Ceratopogonldae (biting midges)
occurred in greatest nunbers and/or blomass In the low density stocked
ponds (Tables 3-23 and 3-24).
General reductions In Infaunal abundance and biomass associated with
Increased stocking rates were accompanied by changes in community
composition. Chironomidae dominated the abundance and blamass of all three
treatment types, but the percentage of the total community comprised of
chironomids decreased by half fraom control to high density ponds (Tables
3-25 and 3-26). Tublculus chironacmids which were among the daminants in
control and low density ponds (Ch ronomus, Cladotanyvtarsus, D crotend paes
and Tanvtarsus) were generally absent frcm high density ponds. Replacing
these omnivorous midges were predaceous Ceratopogoni ds, including
>0-0
U)
O-
(D
CL0
p b71-4 -
OC
z 0
5.'
-E
L 4Z 0
CE
O
0 0
-E
8 <
C 0 m L
- )S000 C44- ->t s-U)
CL
0i 0)
.0--0 C
00 (0
L CCM-.
LL
EP ,
U)
C C
a0E
C C
111111111
LmCNu ine 0 rno- n-0o 5- 5 5 0 0 5 m 0 so
N I+1 +1 +1 +1 +1 + 1 +1 +1 +1 +1
0- 0 M COVaO V N r* V
SI I I I I
* 5 5 I" 0 C 5 0I" 0 5 --
S+ 1 +1 +II +1 ~ + +1 ++1+1N oi C r- »n- n rn
N L
(D go-0Oto am8) (D20 C0 0 CO- 004- 0
-0 0 C0 L 05 a.O. 0-00.0c
40 4-0
p Rod ed d3 3iS - (LL QO O
- < . ,..8 S
N
NCM
I
N
+*o
OI
0)4-
OI-
0t aO N0
I 0 0 0 0 0 0
N %0
SC 0 0 0 0 *
o~ooor ooo
vO 0 0 v l .an 0 0 mn 0 w M 0 *f- 0 *0m
S + 1 + 1 + 1 + 1 + 1 ++1 +1 1
ooo 0 qw- fn %0ONCr po
0
+I
0c*O0a*
0OS41
3-51
W0N
G oo i
* O
r IO N9- V-
0-00
n -0
I +1 +1
* 6 0
I I
Coo
NOO'-000
-
I I I
.--*n *S- o
-
s .>Z:As LC
a cE
CD
U) C00
% C-
U S
0 0
b L
a am
S0 0 -0
E - -
L 004, -
.0 .0!IU) X L
S0 0- 0
cI 1 -S(O - )
0 Q) -I * E (0 4-.
X 4- CU) 400L
- >-L- C 0)"V) 0- 0SC C L-
S- Le-
C I r i= i= C U
CO U )C
4- - 0 U)
0 00 -0
2 E C C0
U)0 L. C E
0 II I I II II
~VC O ~ rO r 'OISO 0c ) *- O SO t 0~ 0~ 0
0.-.-000N00·0> c V~~I
L m r% 0 c0 0 00****(0 .0
0 0 m 00O %o ^t o ^ ^ o* * * *oC;c o
C
0(0V)
0) .
> 0
L.
0)
c
O
-0
ut o
(0.
4-4)
SEv,)
0 00L 1
( L
1) (
c C
O o
i U* E
0Cn
I0)
I 0
38
0
C C
4-
1-00·
0> C
0
OI
0
QO
+1
0
0
Ohn'mnCmLn t V N- 0
- 0 m 0
(D i +1 +1 +1 +1 + +1 +I 1 +1oo~mmco»-o oa
n v0 0 0 0 w 0
0- C - % QCN r
I 1III IlIl I
U)
C c
C
0 Vs'I
-N
85s
5-
+1fn
LCaNNcN
4- 4 .L 0
00 0. OL 0)
- 4- 0 00
L L - 6 9z48 0 + L ^ 4-
<D 0 < Q. 0 ®L. )- 0<0 <"0 0L. 4 <D E -
0)
C
E
*O 3
0
S+-
L
I U)% * 50 C
U)
0 04-C
S 0 0 (
+ -
X) xC- LoSU)0 -
0 0-
UD U) CLL ® Z
I 0 0e-
0 4-CO
U) I 0
-L 0.
C 0 4 C.
O-C
0 00- II I8 -0I
I- I- -
3-52
Ln n Ln sfcf v+I +I I + +I + +1 +I +I+I0A 00 0 L 00 00 0 0 0
o-0000 r'f
inininro~oo GO
LL.
Cmom m Moo oo
»- m CM <- cVim ~-NC - m
+1 +1 +1 +1 + + + +1 +I +I0n w0 0 n m% in o O
CN N CN M
3-53
Table 3-25. Percent total abundance of the major Infaunalmacrolnvertebrate taxoncmlc groups of the three treatment types.
Control Low density High densityTaxa (n = 12) stocked (n = 11) stocked (n = 8)
Cn daria * * *Nenatoda 4.79 7.71 28.4601gochaeta 2.49 2.48 9.75Anphipoda 4.27 4.90 0.87Gastropoda 0.52 0.44 *Aquatic Acarl 1.02 2.43 5.96Ephemeroptera 7.38 9.23 7.01Zygoptera 0.17 1.03 *Anisoptera * *Trichoptera 2.90 0.83 0.24Coleoptera 0.23 0.65 *Ceratopogonidae 4.83 16.44 11.71Chirona idae 70.12 53.11 33.68
Total percent ofmajor groups 98.72 99.25 97.68
Mean total organians(I/m2) 18,194 15,126 7,604
*Percent composition <0.01
Table 3-26. Percent total blamass of the major Infaunalmacrolnvertebrate taxonomic groups of the three treatment types.
Control Low density High densityTaxa (n = 12) stocked (n = 11) stocked (n 8)
Cnidaria --
Nenatoda 0.53 * *01 igochaeta 9.76 0.03 12.76Anphipoda 5.94 7.73 2.04Aquatic Acarl 1.97 3.50 15.86Ephemeroptera 5.78 20.17 8.20Zygoptera 0.57 8.12 *Anlsoptera * * *Trichoptera 2.44 1.01 *Col eopter a 3. 56 1.07 *Ceratopogonl dae 2.92 6.96 6.31Chirononmdae 64.32 51.12 33.74
Total percent ofmajor groups 97.79 99.71 82.06
Mean total blamass(mg/m 2 ) 2,256.83 3,154.00 323.00
*Percent composition <0.01-- taxon not present
3-54
Cul lcoldes, Sphaeromias, and Ceratopogon (Table 3-27). The NematodaIncreased from less than 5 to over 28% In control and high density ponds,
respectively (Table 3-27). The percent contribution of ol igochaetes and
aquatic Acarl also increased with Increased stocking rate. Taxa which wereconsistently among the five most abundant organisms of all three treatment
types included the predaceous midge Tanypus and the mayfly Caenis (Table3-27).
Increased sedimentation rates and physical disruption of sediments byintroduced carp may account for the depressed abundance and difference in
taxonomic composition of the Infaunal community of high density ponds. The
absence of sessile, tublculus midges (which typically Inhabit central
III inois ponds) and their replacement by free-swimming predators is
suggestive of unstable bottom conditions. Such conditions existed In pondHC4 where sedimentation rates were very high relative to other ponds (Table
3-19) and the bottom was completely denuded of macrophytes. Grass carp
have been reported to feed upon invertebrates once their macrophyte food
source Is depleted (Kilgen and Smitherman 1973) or In sparsely vegetated
situations (Gaevskaya 1969). Carp stirring up bottom sediments in search
of food may partially account for increased sedimentation rates and the
paucity of infaunal macroinvertebrates in high density ponds. Increased
predation by sport fish, resulting from reduced plant cover, may also bepartially responsible for decreased populations.
Epiphytic Macro nvertebrates. Herbivorous carp obviously reduced epiphytic
macroinvertebrate populations when consuming their preferred food;
previous studies have shown this consumption to be Incidental and at a rate
proportional to the plant matter consumed (Beaty et al. 1983). Our
results show at least a slight reduction in epiphytic community
concentrations. Chironcmidae were present on macrophytes in high density
ponds In significantly lower numbers than in the other treatment types, and
chlroncmid blomass and total cammunity abundance and blimass were also much
lower, although these trends were not always statistically significant.
Means and percent composition of abundance and bicmass of most other major
taxonomac groups, as well as total community abundance and biomass, were
d
IJa.
6a66
!
I.
!
|
r
5
b!
I
?JS
c
4,
6
0
*05
6
ao
a
Ca
c6
r,
U
ISaN
8
o
iI-*e
f
3-55
P-%
c
a m
8I,
*&x
I
C
86
A
cu
U
C
a0
C
0
c
-I8
C
>-
c
0Co
I.-I
I
C
a
0
I1
C0o
U
aU
c0KZ
I-
e
Iq- o m W " V mo w ocI o mo sO cu q o U) Cu
SD a- CO N O) 0 a0i «n cB l^ QO <» g^ CW CM
Cu
IU
D u I o a 0 a 0
I uw- C 0
Cu OB aD 0 m 16
a S S It 0S
4444aai«D~aad
W) W N w m amo w~
S 0 C 0 0 a o r- 0a
a- 0 N Nb so a ?»P ) U)
K
6 a o v a6
a-CuP300s0N00
a 0 so 4 N L3 1 Al L
iidn i iir V C13 qq V- W 0 COQ)
s ittm~ -t5 S, 8
0 cocoN CC U V a Na S a; s h l
l l
i-CuP3 + IaUhnUW0
U2
S0cr)O oN
a(00
ga.IQ
gm
C0
COV
6 Ui
V-
ar :
!62C
80611
^uOCii
3-56
greatest (although not significantly) In low density ponds (Tables 3-28
through 3-32). The mean number of taxa per sample and cammunity diversity
were greatest in high density ponds (Table 3-28); so, although total
abundance and bmanass within these ponds were low, the epiphytic community
was quite diverse, with few nunbers of Individuals spread over a relatively
large nunber of taxa.
The decl ne of epiphytic chlronnmid abundance in the low and high density
ponds is best illustrated in Table 3-30. Of the three chlronomids
dominating the epiphytic fauna of control ponds (CrLcotopus elegans,
Endochironomus and Pseudochlronomus), only -L. elegans maintained a
substantial percentage of the abundance in the high density ponds. It is
unl Ikely that carp were reducing these, and other, epiphytic populations by
selectively feeding upon then; however, bottom conditions such as those
existing in pond HC4 may Impair reproduction and recruitment of Insects
dependant upon aerial oviposition and the settling of eggs to the pond
bottmn. The three most abundant epiphytic organisms of high density ponds
were non-insect taxa that did not rely upon aerial oviposltlon.
Total Benthos (Whole Pond Communlty). Whole pond abundance and biomass ofeach major taxonamic group and species were calculated by sunming Infaunal
and epiphytic means (each converted to square meter of pond bottom).
Results of analyses of variance of the whole pond treatment means were
similar to previously presented results. Abundance and biamass of most of
these groups tended to be highest In low density ponds; however, only
ceratopogonid abundance was significantly higher (Tables 3-33 and 3-34).
Overall, ponds stocked at high rates with herbivorous carp had dranatic and
statistically significant reductions in chironomid and total community
abundances (Table 3-33). Ranking of the 15 most abundant taxa of the
whole pond assemblage in each treatment group again emphasizes the decl Ine
of Chlroncmndae and corresponding Increase of Nematoda as stocking density
increased (Table 3-35). Chironomid species ranking in the 15 most abundant
taxa decreased from 9 In control ponds to 6 In high density ponds. The
abundance of chlroncmids ranked in the top 15 taxa decreased from 11,879.55
m-2 (38.82% of total) In control ponds to 1739.47 m'2 (18.43% of total) In
3-57
S 00@
(0 (0 .0 0 .0 .0M O n - m m P %0M o r% 0 oow 0
·~ ~~ 0
NC
40.> a
tn4- 0
S0)
L
.-
E b
S-z
S)
COL.
8 0C -0
-L 0>C
oS +1O> 0
-E
-0- o
G)0
N
SU)
0 0CC
O U)
- ) 4-
*94-
> 0.>
Q) -
. +4 u
0-4-
LL
c O
U)
8
C
-J
O-NL.
S II
8~5
I-
CN
0
%0
+1%gIN
\0
N
N
CM
N
0 ) L W
000 0Q0 L0- 02(D +- 1 6 0 4)0 0 0 L aLP O O QO 0+ O O O C L, < )
" QOD L 0 .0 CL4- 2 >
~co u 8 o O >O5- . - - -I -3 E E><5l5^g.?£^i=<S855 a I-s
* 0 %
* . .000
m oo*
* 0 0
'-00Cr oo i
N 0
C4
I I
N'-0M 0 0
CM
o U" -, r"
G o m CMLn CNwl
C m rI - C I CN mo
c 0 5 0* * 0 0 9 I 0 0 0m0 0
0 0 9C4 0%0 0,*- V 0V 0 0 n
%tn m ooo V-o 9- W-on 'o d
C 0 0 0 f9 0 0 0 M n o0 0v 0
S Pf -%v --- v1NNN N0
S 0 • • .0 . • So_
+1 +1 +1+| + 1 + 1 +1 +| + 1 | +| +1 +1
aO P %-nornqm 6%ooo w nro
Fn 0Vm N* 0 r G
'-OO nOc K O O CMr - am
r.- »< 6O r KV i ' n o aO - r\
K^o no r\so oo ~o aoo ^#@@wr
0)C
00
E0)
L0a0. C
CD
4 -0 0
4) (0O CCL +
S- 0C0if 0
X E2C
0 - U
U 0 0Cc -L
00- - -
0 0 -
X II >I C
C- o +
(0 O *I
- u)m co-
0) CO£ (0 W
8 1 C
N
N0u
O
0
0.
.- 0'--
S)
4-U) 0
0
O-0
'--
X c- 0
LO
S4-
c
L
0 0(O L
C
0 )
8 +.
c.
0
- 4-
0 ,
4- -^ Jr
!i ·
U
-I
.C
-N0--
4---
C
I-X
q %
co
S % % 0C fn 0 * w 0 0 0m
n O C w- O%. .. . . . . . . . . .^
LL. Ofn•o-o •NUN •wN'- N
0t -I I 0 I0 h r ** * I*I ( ( (0 ( 0 I 0 3 0 0
ONNt1NO-DOO-N
+1 01 + ++0 +1+*+ 01 +I S% +oooooooooodo
n- QO CO r% -O -NV -V- r*,- V L r - CO w- Ch K) q
0 e Q 0 0 0* 0 0 0fnCco r %-cNr-oGoo
N m Q
III 1 1111111
** 0 * **C% * 0 0 %0 %0
CMG- C" O q-KOmi
III 1 I IlI ll
0 a 0
-0(D
r 00 CO C0.< OL. &t - 0)
eO OLO SoL 0 0.- 0. 0 0.0 C5C4 O) 0
8zo-4iS ^r^^^8 &UJ S 0
0I
.0
CI
+1
3-58
C
U)C
0® EC +0 C
> (D
0
F.4 -L48 0
> cL -
c 0
4 0 - -
.C O <L
U)
- c c
L eo.
L -4
- 4- V
-O -4u) >
ICO CO
le c Q)-
C ( 0 - U)
0- 0 5 c0 OLO>
$4- U - I
0 II I < II
ir
00
N
NI
00
CNN
4A0 0 " ' -' ", 0,
N mf~~1
c04-
~ u)
&5u
0·I.-
3-59
i4-0D
0
L
0
-c4-
04-0
4-0
L
0
SUr
0.0
4-C
0
£.00p
4-U)
0
0r
'4-
0
C
04-
§
L0C.
C0
Co0
0
I-
C0
8SU
0u
I
C0
(2
J6
cC0
4-
U
t
0
co
4-
u,
fr
1HI
v~ C)% w0 - P-- w w
Ii OIq0--
4
s % * - 0 0< V 0
0
* aa
N CF4 '0 0 ;0: 00% 0- !0 in > 1 0 N
N& 5 A<Sl£ <§»-cs~m fi^»o~coor
so~m or^O oo
< q
1 = -a.i. o
II
cV,
0r
8
cl
c
0
L4-I
%O tN r.- rke W"0
IO ,•
0In In o
o.'0
I-
4-4-.CC
4-04
3-60
Table 3-31. Percent total abundance of the major epiphyticmacroinvertebrate taxonmaic groups of the three treatment types.
Control Low density High densityTaxa (n = 12) stocked (n = 11) stocked (n = 8)
Cnl darla 0.42 0.46 10.49Nematoda 0.90 3.70 1.1101 igochaeta 2.44 4.06 10.30Anphipoda 5.17 19.50 16.79Gastropoda 5.47 5.44 4.79Aquatic Acarl 1.21 1.76 7.04Epheneroptera 5.98 6.70 8.57Zygoptera 5.16 3.05 5.30Anisoptera 1.01 0.11 0.09Trichoptera 7.53 6.50 1.84Coleoptera 1.72 0.98 0.77Ceratopogonl dae 7.84 10.40 7.69Ch ironom dae 53.88 36.36 23.98
Total percent ofmajor groups 98.73 99.02 98.76
Mean total organslsms(I/m2 ) 12,407 15,525 3,672
Mean total organisms(I/g) 15 16 19
Table 3-32. Percent total blanass of the major epiphyticmacroinvertebrate taxoncmic groups of the three treatment types.
Control Low density High densityTaxa (n = 12) stocked (n = 11) stocked (n = 8)
Cnidaria 0.49 0.12 5.34Nenatoda * 0.08 0.0101 Igochaeta 0.66 1.74 3.93mnphipoda 7.55 24.88 23.49
Aquatic Acarl 0.30 2.28 1.60Ephemeroptera 8.67 2.72 6.83Zygoptera 17.18 14.90 25.09Anisoptera 6.21 0.44 0.33Trichoptera 11.30 11.98 7.18Coleoptera 6.44 1,79 5.81Ceratopogonldae 0.69 4.10 1.34Chlronamldae 29.87 19.23 14.64
Total percent ofmajor groups 89.36 84.26 95.59
Mean total blanass(mg/m2) 2,688.23 4,288.49 584.94
Mean total bianmass(I/g) 2.82 4.60 3.02
*Percent comanposition <0.01
1-
3-61
d0C
8CC
Q)
OCA
EC
-j
C
ge
So-
§ 8E
Ce
C T.
(JC-U)
0.4-EU)
-0CO
C c
-0 0
O -
0 U)
CO
4-- 0
0
oJN (A
I t
E»g
U)-
C 0).-
liC
10
NO
248
a ci
g>. Q.-t
A 3C
=u^
SI-Z
m0 «
C
4-
U)
01O
U
ul
C
C
a.4
C
t5
15c3C0
C
4.
-J
N
U
C
0
C
dood oodood o
No .C.OMNa00 ..N .
••om-ocoocoo
' Nc0 IM NCý C ; O ooo ' CO 0
000000000000r CC O tC 0
rri r.- r..aoo r- w- w- qw«r % ^ 0 wo- oo q r00*00 O 00 s
C) CW Go M0 _C z I0
N M 0o
S o co o - %0 N
M Pt n in N N N V M
0(0O0ANN0I00IC 0
OON)00,-OO- 0
ov7Vn CM N •u00 0%Cc a 'CMcNw0%00 '0
UN V Q
w- »
V %0N QNNP-.CMO 'oo
M W% %0 n (74 qw W a 0w 0'0 a
UlC WO M coCV CV OI \ %0l
%W-'- fCor N \ r 0-
Q Fn00 % inr CO OOOM0 »_
•••N<N
2 w W % COr% 00 %0 fn0un ' In %0% W %0
e^ » < » d » C d MN. U o 4 .
QNFn0C;0 ; "- 0 '0rO c IO KS n 4 OCMO C 0 CV
o % o fn C D n N r-% o o M
00 %000U ')P ' " U'
%O % N qnO w'00%0 n r
N 0 G o
o 6 o B c o c
4. C( CL +- 0 L <0 'O (00 Ca 'o < 0. CL. ®0 L4 )- 0
CCC'e0 a O§.11 1° be 5 c cL>.C ' o q0 0
5z'Q.Ia ,2I CT n c
4.0
()
4.
C
4.C0--
U
0.0.w
C
CC
C
C
U)
0
I
3C
ul
L .
e )
I
0o
0 0
t_X ®
L. CL
" C
L.
O I
ru
LL.
-I<>
3-62
o0)E
Ug
=8(s
I-a
e.4-
"8
)i
CUC
4 >
0 -
c in
(-0
.4L
C
u c
C .
- 0
-0 08L.
U)-
(o w
4£4
)0 C
Z40
§2 9
<3
t e Q
v( CM- -a r- mNO O -a ov0 r-...-a M
0% 0% 0•, wO %O , 0
888000000000000
o -co v N- o o o a% Nq oo ODW% v in-. 0 -.
momN-I0 " -, M ,N I- n v I o v d
GO WM 400
oooo eoooooo
o0n %oo i coc0 0 0 n 0
o-m' 0 - 0 - 00
S.4
0 n ON, % N NW O 0 UC -CON ^tOO 00
00 O rO- 0uN o oo-- 0
r- N
C4 4 %0 C 4
N A ( Ac 4 0 0 f %O
UO rz 0; C; tA Nco o N c
0 "A-%0 04 V-- %0
40 40o0 r n CMnW C
0 4aOq C;O 0 C; O ) 0; 00
N C- CM
to L 60oc
L) (0< &.< L. Q)L.0-
I,CD
4-
i.5
II
C
4-
-J
C
0
1.
4-
4-
(0
U4-
(0
I4-
LU
C
40'S-
C
o4-0
0.Q.
C
40.--
40X40I
4 *U)
04-
]1
COUtO40
U)
0)cu)
(0
0
,5
T
L.S
c0
8C40
o30»
U)
U)
40
C
40
4-
U)
0
=1
e3»
o
I..
u.
c
o
a
>,
0
o0
0
C
4-15
ae
to
L
40
I5
o
I
00000t0qo00qC0 0 0 O V 0 0 0 N 0I
Ln ac u^ou anmmc
wN (A N CU 000 0
C930 N N 00 g " s 0 0 0 y
Y 9 0 W
0-o 0
NSS0 Cw ,-
• •
,..•
U
C
a8U
u
a.
Ui
U(I
6
St
8BcA3UI)
U
U
a
I
?
a
y
O
u
U
5
C
U
U
43
0
C
6S*»
C
I
tii
0
0
0
UA
U
I
«*
CT
ua*9-
0
ci
w
S
I
C
ai
a
0H
£-C
16
8
u
C
- -
CUOCU O000Cu~O u0 N 0O0 M2OO a O 1 q*N 0 0 0 0 0 0 0
" sI 'L . , 41
.CdV3 10 0 M 0C 0 0C;V 0 C4 04 aU- U- U- U-U-U-g
3-63
0l CU C O I00 C I O QCuD i0 000 q 0 Cu 0 Cu - 0 N 0 0 N Cu
0 0 N 0 N ~ 00000 0O a 'co OB It »s C rs r<? oo tn oc oC0 U -
I sU
CU0 IV 0 V 50 M
0 0 u - " s ON o 0 0 Cu 0 *0 N 10
0 0 0 i- 0u 000 C 00 NO 0 00 00 N O O . 0
a0 0l 0a a 0 W
^gS|^|5S°SS~oS5C W coD W c N O Qe 0 C3 00 V O 4
V 0 aB 0 3 M3 cu 0 a 0 0 0aU
a0 tsgs^ScS "
9 S Si 0; 5. i i §i a 5 S 9: j|W) U- C 0i M3M3
v00 0 0 m Kq
ao rz a ai io 1'. a a a i' j W ai a
P CU1% isr f%
a si 1-s i iSm 0 to ** 0 C0 V) 0 f t o 4 O
cu ao W Br Otuo D a ii
0 a c a 0 16 0 ( (B a'aB |go3 "u = 0 C- i ^ 0 Q
oDor~D~ mOrs aov aoc
a o a in 0 Q 0 3P 0 0
w acQ .a a o( s <
I~v cm u q g0 m00 %a i*
ce aII cc 0
0:
CO
co»s*
cm
0
aD
0
008
r0
N",
I
0
0
Cu
a
0
0
U43a9-
'I
U
IC
C
a
a
a!|
I
iS
e
ll
I
3-64
high density ponds, representing a significant decrease in available food
organisms to several fish species. Conversely, Nematoda Increased from
993.69 m- 2 (3.25% of total) In control ponds to an overwhelmingly
2737.26 m-2 (29.00% of total) In high density ponds. The decl Ine in whole
pond chironomid and total community biomass were evident, although not
statistically significant due to high sanple variances.
SQor± Fsh
The Impact of herbivorous carp on sport fishes varied between species.
Bl uegill production decl Ined dramatical ly (Table 3-36), particularly inhigh density treatments. Catfish production, on the other hand, increased
significantly; this contrast is indicative of the kind of shift in sport
fisheries that might be expected with large-scale introductions of
herbivorous carp. The response of largemouth bass populations was more
complicated, but was basically consistent with the predictions of our
trophic-level model (Wfley et al. 1983). Fingerling and breeder bass
production Increased significantly (Table 3-36) In low density carp ponds,
and decreased in high density ponds. Thus the highest average levels of
piscivorous bass production were observed in the ponds with intermediatelevels of plant control. Young-of-the-year bass had the highestproductivity in high density ponds although none of the treatments variedby more than 20%.
Average seasonal growth followed a pattern similar to that of production
(Table 3-37), except growth Increments of young-of-the-year bluegills
increased with carp stocking , presumably a response to lower population
densities. Condition factors increased with increasing carp densities
(Table 3-38) for all species; this result was unexpected and may have
resulted from a short-term increase in prey vulnerability as macrophyte
stands were el Iminated.
Survivorship of stocked fishes did not differ significantly across
treatments. However, young-of-the-year (YOY) recruitment of centrarchids
declined In ponds stocked with herbivorous carp, dramatically so In the
3-65
Table 3-36. Summary of analysis of variance results for sport fishproduction estimates (kg/ha) for control, low density, and high densityponds at the Annex site. Treatment means (+ standard error) of all years(except where noted) are given, with the number fish harvested over thatperiod In parentheses. A dash (-) Joins SNK groups with no significantdifference and an asterisk(*) Indicates significantly different treatmentat the 0.05 level.
Low High AlphaControl Density Density F value
Largemouth bassYoung-of-the-year 1 19.66±0.08* 16.99±0.18* 20.80+0.49* 79.499 0.000
(2474) (2601) (1226)Finglering (3-5") 2 33.17+0.32* 35.20+0.27* 31.00±0.58* 27.017 0.000
(92) (111) (35)Breeder (7"+) 9.86+0.23* 14.81±0.48 -13.33+0.61 42.688 0.000
(46) (36) (11)
Bl uegl IYoung-of-the-yearl 79.99±0.27* 59.76.0.14* 28.24+0.37* 511.566 0.000
(15,256) (19,138) (2251)Breeder (4"+) 9.37±0.18* 7.52+0.22 - 7.54+0.35 21.150 0.000
(56) (54) (19)
Channel catfish4" 5.18±0.13* 5.75±0.11* 10.01+0.30* 150.347 0.000
(21) (20) (12)6" 7.77+0.12* 9.23±0.15* 12.84±0.42* 118.674 0.000
(41) (31) (12)8" 10.76+0.17* 11.94±0.18* 14.91±0.59* 41.565 0.000
(47) (34) (11)
1 1982-1983 estimates only2 1981-1983 estimates only
Analysis of variance In appendix tables A3-51 through A3-58.
A L 0
a
W6
a136.
C6
Ce000
sa C
cai6.6
aS I
is
3* 55C-C
CL~4c
A . S* S U
5 C
4K3V
S cg
SC.
556c
S;:i
* 4-
343.
.SCC e
46
C45 3
o a6 S
043
0S. CS SX
4- S :
8. Ui
a^'
* 0
A a.1 s0B 0
3-66
CD
40w
S*
C
S5*£
I
0J
S**
S
0
0
60
Ce4a
V
J-
69
c
a
3.
£o
16
40
c
aJ. C
S
as
a ao a
sa 0
0 %Noco
0) C
O D
* a
cu
a aw-0 N-CT-5.S« a
S 0
in
* 6
0 •
0 0
C93N s-
C, .4
+1 cps+1 c00-N~
to ,(o D imga scu V
Cu 30 U-
0f 0c 0
+ P.
0 0N 0
*1*1a CU
Ns 0
C C
c Nq-*.
8 C;» lI 0
C;
aa
0
a0C;0
a
+1
+i
cc
00
0;
0C;
go
0
ci
co
N
0
+1
cu
U3
CL
ig+00
NW3
I
(0
+1
--, .S60I 3
4 6
So a
o * o o
a a & aS a 0 aa
0 10 . 9
+1 + *1 +U 1 C9 4' Cca t' ca
S 0 - 0
0 0 ?0 0
cu (a C 0q0 B 0- 0 ,-Cu V
C; -a - ;+1 ÷+ 1 ÷+ +I C*P 6
0* 3 r c UlO. n -
a . o 0
to -* rs cu o
• - , q- .-
a V C F qP
; a;a. m * Ci
0 0: 0
+, +1 +1 q,, T I-
I ( N 0l o 0
0 0
Cu 0 CM0-f
0 a NVl'.n. 0e 0 0
SS
01 10 +1
a P cu i
< 0 0" 0l
II • ÷5 ÷1
q- W Cu c
q 0F 0aCU3 Cuww
do 0
0 0 y0
; 0; 0; 0+ +1 + I- TI
C; 4.1; ;
+1 + 1 q- +1 Y-
C; Cu 1a
V- Cu 03
C03 q 0
CO +*1C03+1 0F
" cu U13
q- cu W0
N 0 0c
ca 0a c
+1 pq 1 +1 Tvcu q 0a a 4VM cu U3
0 003 0
* C ca+10 o 1V3 O - CU
N s
ola 00 0
ca oC
C1WN01
II n0
(.co
+1 08
0
N'
N;+
0U
a2a06.
co
0
6_.-4WD
816
a
4- 64-
i3
SC
4 a4-
C3
4-
4-
a .8
S-
41&
CS
*0
56.
4aC
0&
61,
T5•
5.
c u -
'U-» *
T g-S
J.3.2
£
I
5
S
AS
6
C
Io
aS5S«
*5-+
3 62 6VS
0, 0uL -
LL
6
0
.o
aS;Sp
atC0
cc60c
c
c
mu
.C
-45
060
0 60
lt
8 :
CL
06.cm
a
30%
qp-2 5S
0 S
16
a v a
16
I S
ma+ a310
X0
0
16
Pei
r^
a' c
Cc C
Oi 6
160
c o~06
cc0+ 6
i
9.069.0
0 0
LP!CO C
OI
6
6 6£
A > i!
666.9 0* 6. 0
@69.
0 0 0
N cu
m m |,Q
e--• Cu-- •o, o B np»
o o a
o »l qS. • 5. •
a m 0
9. 0 0
a as a
0
fa Cu 9. 3
S a
q- s p cu °
o C? a
f, o o
cu u 0 3
0 c 0
o o o
0 Ia 03
a u- C
a a a
a' N cu
au cu 0V
0 9.
00 Ow
qmo CY G€,
a o
q a
- 00
a *
CI wo C C 2
• •-1 CO
0 Mu
0 0
a aocu V3~
7 67 6
I ^C *- - 0£0 9 . 0. 3 -
ac s 'a g o
LL a 4C~0
3-67
a a0 0
So
o o
5 I
a a
sjj'I4- vs
C C
* a
«• •
I I
I "I
0 9'C#3
0 0
N0
C?
0 9*9 00 0
Cu0aa
* a9. CU(a a00 0
9. CU
00CI C*'
Ijji
0 0
a
I;0
j.Icu
* a 9.
cu 0
c! ! I
J 0
cu to
93 cu q.0. 0 0
C; C; C;t0 Cu (0
I I I0 0 0
cu 0 93
c -c -o
do- d-M d
; C ; 0
0 0 0.ca a a - n
(a a0 av0 0 9.
0 0 00 N 0 P-
P% N N
Cu 1 Cu5m qm 5 5
0 0 0
a 0 a
ai Ta a
0 0 0
0 i0 03
0 0 0
atao
c1C
O
0
i8c
6e
I1
S
uS0
E
Y
?
I(A
9.I -
o
c
16
L_
N*I
O
C
a
c
6
iS.
3% 3P
c
S
IS
e
coog 3 36
is6
*1:00069 9Q9 C
Oft °% If
3-68
overstocked pond (HC4) (Fig. 3-20). While high variances in the control
group of ponds obscured any statistical significance, these declines were
undoubtedly ecologically important; In fact, they were the basis for
observed statistically significant declines in YOY production In the carp
ponds (Fig. 3-21, Table 3-39). This reduced recruitment seems primarily
the result of increased predation with reduction in plant cover and was not
the result of any fal ure to successfully spawn.
Our experimental design, while giving us substantial statistical power In
analyzing single season responses of fish populations, did not provide any
direct Information on possible long-term impacts. To assess these
long-term consequences we are forced to rely on the predictions of a
theoretical model derived for this purpose (Wlley et al. 1983). Our model
suggests that major suppression of aquatic macrophyte populations over
several growing seasons will result in major reductions in centrarchid
production; Initially bl uegll Is wil I be most severely affected, and
finally bass populations as well. Since trends along these lines were
already apparent In the single season experiments, the model's basic
predictions seem accurate. It should be noted, however, that in
situations where there are pelagic forage fish available as a substitute
for sunfishes (e.g., gizzard shad), bass production may be Insulated from
the effects of macrophyte loss. In waters with pike populations, a similar
long-term reduction In productivity is likely If macrophyte populations are
el Imlnated entirely.
DISCUSSION
In discussing the possible environmental Impacts of using herbivorous fish
for biological control of aquatic macrophytes, It becomes essential to
distinguish both between long-term and short-term impacts and between low
to moderate levels of plant control and total plant elimination.
Short-term impacts were examined experimentally in this study and can be
specified with some precision. At low to medium standing crops of
herbivorous carp (<150 kg/vegetated ha), environmental Impacts were
generally snall and principally reflect changes in the carbonate-carbon
3-69
8
*-
q1
* 0
( 0
(o00(0Q.A
0.0
3-- I
e*
"Di+ '1
x e
-4.5
is-ia.0.
La,
3
C")
-1-
m
C:)
-o
30C3
,_
z2
3-71
Table 3-39. Summary of ANOVA comparisons and SNK groupings forsport fish survivorship In Annex ponds all years. Mean survivorship(calculated as the ratio of number harvested to nunber stocked) (±standard error) Is reported along with the alpha value and F statistic.Dashes (-) Join groups with no signf Icant difference.
n at Low High Alphastocking Control Density Density F Value
Largemouth bassBreeder 6 0.96+0.04- 1.00+0.00- 0.92+0.08 0.71 0.51Fingerlinga 20 0.81+0.08- 0.93±0.05- 0.88±0.03 0.79 0.47
BluegillBreeder 10 0.71+0.09- 0.90+0.04- 0.95+0.05 2.25 0.15
Channel catfish4" 6 0.44±0.12- 0.56+0.15- 1.00+0.00 2.15 0.166" 6 0.85+0.08- 0.86+0.11- 1.00±0.00 0.33 0.738" 6 0.98±0.02- 0.94±0.04- 0.92+0.08 0.70 0.51
a 15 fish were stocked In 1980
3-72
dioxide system affected by the removal of a large proportion of the
autotrophic camunity. While statistically significant changes could be
detected in water chemistry and sedimentation rates, these changes are of
limited ecological significance. The Impact of lower stocking densities on
sport fishes was generally positive. Bass production significantly
Improved In low density ponds, presumably because forage fish were more
available with lower levels of plant cover. Likewise, channel catfish
production increased significantly. Since Increased sedimentation in these
ponds consisted principally of carp feces, an enrichment of the sediment
may have been responsible for Improved catfish growth. Only bluegllI
production showed a significantly negative response In low density ponds;
both mean weight and seasonal production of young-of-the-year bluegill
declined about 5% relative to control ponds. Breeder bluegill production
was reduced by almost 25%. The reason for these declines is difficult to
ascertain, although they may be related to reductions in epiphytic
macrolnvertebrate production (see Wlley et al. 1983).
When herbivorous carp were stocked at densities sufficient to totally
eradicate macrophytes, the impacts on pond ecosystems were much more
dramatic. Beyond the expected changes in pH, alkalinity, carbon dioxide,
etc., sedimentation rates increased by more than an order of magnitude;
oxygen concentrations declined and turbidity increased. Chlorophyll
concentrations rose, suggesting increased production by phytoplankton, but
benthic macroinvertebrate populations decl ined dramatical ly. These
conditions were so consistently expressed In the three ponds in which
plants were el Iminated, that together they could be considered to
constitute a stereotypic "overgrazing syndrome."
In high density ponds, piscivorous largemouth bass production declined
relative to low density ponds, but It was still significantly higher than
that in the control ponds. Bass young-of-the-year production was actual ly
highest in high density ponds, although it exceeded production In the
control ponds by only 5%. Channel catfish production increased
dramatically, almost doubling that in control ponds for 4- and 6-inch fish
(size at stocking), and Increased significantly for 8-Inch fish as well.
3-73
Bluegill production was reduced, catastrophically so for young-of-the-year.
Although young-of-the-year recruited at the end of the season were
significantly larger than those in controls, production was 65% lower in
high density ponds. In the one overgrazed pond for which young-of-the-year
data are available, their production was reduced to less than 20% of the
average for control ponds.
Long-term eradication of macrophytes and concomitant maintenance of
conditions associated with the overgrazing syndrome would undoubtedly prove
deleterious to centrarchid sport fishes In general. On the basis of
trophic level organization alone, our model predicts major declines In
sunf sh and bass production as macrophyte populations approach zero. The
necessarily negative Impacts of oxygen depletion and high sedimentation
rates are additional stresses that would seem likely to guarantee large
reductions in productivity. On the other hand, because the removal of
macrophytes will result In a more or Iess plankton-dom nated system, more
pelagic species like shad may do relatively well. Likewise, catfish may
benefit from Increased organic deposition rates as observed in our
experimental ponds. The desirability of these changes depends on one's
viewpoint. However, the long-term effect of maintaining high densities of
herbivorous carp will be to drastically alter the structure and function of
lentic ecosystems. The implications of these alterations for the sport
fisherman are difficult to predict with certainty, but the ratio of bass to
catfIsh production observed in our experimental ponds Is probably
indicative of the kinds of long-term changes that can be expected (Fig.3-22).
LITERATURE CITED
Amerl can Publ c Health Association, American Water Works Association, and
Water Pol I lution Control Federation. 1976. Standard methods for the
examnination of water and wastewater, 14th ed. APerican Public Health
Association, Washington, DC.
3-74
RATIO OF CATFISH TO BLUEGILL PRODUCTION1. A-
1.0-
.8-
.7-
.6-
.5-
.4-
.3-
.2-
.1-
CONICONTROLI
LOV OENSITY
TREATMENT
HIGH DENSITY
Fig. 3-22. The ratio of total catfIsh and total bluegl I productionby treatment In INHS experimental ponds.
RATI0
I I L I MMMMMMM.M..llm
I I
*
3-75
American Publ c Health Association, American Water Works Association, and
Water Pollution Control Federation. 1980. Standard methods for the
examination of water and wastewater, 15th ed. American Publ ic Health
Association, Washington, DC.
Bakos, J., Z. Krasznal, and T. Marian. 1978. Cross-breeding experiments
with carp, tench and Asian phytophagus cyprlnlds. Aquacult. Hung.
(Szarvas) 1:51-57.
Beaty, P. R., R. G. Thiery, R. K. Fuller, J. E. Boutwell, J. S. Thullen,
and F. L. Nibling, Jr. 1983. Impact of hybrid amur In two California
Irrigation systems: 1982 progress report.
Buck, D. H., R. J. Baur, and C. R. Rose. 1975. Comparison of the effects
of grass carp and the herbiclde diuron In densely vegetated pools
containing golden shiners and bluegills. Prog. Fish-Cult. 37(4):
185-190.
Carpenter. S. R. 1980. Estimating net shoot by hierarchical cohort method of
herbaceous plants subject to high mortality. Am. Midi. Nat.
104(1):163-175.
Fowler, M. C., and T. 0. Robson. 1978. The effects of the food
preferences and stocking rates of grass carp (Ctenopharyngodon Idella
Val.) on mixed plant communities. Aquat. Bot. 5(3):261-276.
Gaevskaya, N. S. 1969. The role of higher aquatic plants in the nutrition
of the animals of freshwater basins. Rol'vysshIkh vodnykh rastenil v
pitanll zhibotnykh presnykh vodoemov, vol. 2. National Lending
Library for Science and Technology, Yorkshire, England (translated
from Russian).
3-76
Gorden, R. W., S. W. Walte, and M. J. Wiley. 1981. Effects of using hybrid
carp to control aquatic vegetation. First annual report, Federal Aid
Project F-37-R Illinois Natural History Survey, Champaign, Illinois.
147 p.
Gorden, R. W., M. J. Wiley, and S. W. Walte. 1982. Effects of using hybrid
carp to control aquatic vegetation. Second annual report, Federal Aid
Project F-37-R. Illinois Natural History Survey, Champaign, lliInois.
122 p.
Hutchinson, G. E. 1975. A treatise on limnology, vol 2. John Wiley and
Sons, NY. 1115 p.
KIlgen, R. H., and R. 0. Smitherman. 1973. Appendix F. Food habits of
the white anur (Ctenopharyngodon Ide•la) stocked In ponds alone and
In combination with other species, pp. F1-F13. In E. 0. Gangstad,
ed. Herbivorous fish for aquatic plant control. Aquatic Plant
Control Program. Tech. Rep. U.S. Army Eng. Waterways Exp. Stn.,
Vicksburg, MS.
Lembi, C. A, B. G. Ritenour, E. M. Iverson, and E. C. Forss. 1978. The
effects of vegetation removal by grass carp on water chemistry and
phytoplankton in Indiana ponds. Trans. Am. Fish. Soc. 107(1):161-171.
Leslie, A. J., Jr., L. E. Nail., and J. M. Van Dyke. 1983. Effects of
vegetation control by grass carp on selected water quality variables
In four Florida lakes. Trans. Am. Fish. Soc. 112:777-787.
Mitzner, L. 1978. Evaluation of biological control of nuisance aquatic
vegetation by grass carp. Trans. Am. Fish. Soc. 107(1):135-145.
Pierce, B. A. 1983. Grass carp status in the United States: a rev few.
Env. Manage. 7(2):151-160.
3-77
Shannon, C. E. 1948. A mathematical theory of communication. Bell System
Tech. J. 27:379-423, 623-656.
Shireman, J. V. 1984. Control of aquatic weeds with exotic fishes. jI
W. R. Courtenay, Jr., and J. R. Stauffer, Jr. Distribution, biology
and management of exotic fishes. Johns Hopkins Univ. Press,
Bal timore. 430 p.
#, and M. J. Macelna. 1981. The utilization of grass carp(Ctenopharyngodon Idella) for hydrilla control In Lake Baldwin,
Florida. J. Fish Blol. 19:629-636.
Sokal, R. R., and F. J. Rohlf. 1969. Blcmetry. W. H. Freeman & Co., San
Francisco. 766 p.
Stanley, J. G., W. W. Miley II, and D. L. Sutton. 1978. Reproductive
requirements and likelihood for naturalization of escaped grass carp
In the United States. Trans. Am. Fish. Soc. 107:119-127.
Walte, S. W., and. S. M. O'Grady. 1980. Description of a new submersible
f l ter-punp apparatus for samplIng plankton. Hydroblologla 74:187-
191.
Wetzel, KM J., and M. J. Wlley. 1981. Benthos, pp. 110-119. J. R. W.
Gorden, S. W. Walte, and M. J. Wiley, eds. Effects of using hybrid
carp to control aquatic vegetation. First annual report, Federal AidProject F-37-R. Illnois Natural History Survey, Champaign.
Wiley, M. J., R. W. Gorden, and S. W. Walte. 1983. Effects of using
hybrid carp to control aquatic vegetation. Third annual report,Federal Aid Project F-37-R. Aq. Biol. Tech. Rep. 1983(1). III. Nat.
Hist. Survey, Champaign.
Chapter 4
POTENTIAL REPRODUCTIVE PBILITIES OF DIPLOID AND TRIPLOID HYBRID CARP
(female Ctenopharyngadon Ide1l l x male Hypophthalmlchthys nob I s)
AND DIPLOID AND TRIPLOID GRASS CARP
The possibility of successful reproduction and subsequent establishment of
naturalized populations remains the major perceived danger in the use of
exotic fish species in the United States. As witnessed by the explosive
naturalization of the common carp (OQgrlnus carplo) throughout the country
and of the many species of Tilapla in Florida and Texas, this fear Is well
Justified. Conservative approaches to the use of exotic fishes must be
maintained to avoid ecological catastrophes.
The grass carp, Ctenopharyngodon Ideia, has long been known to be a very
effective consumer of aquatic macrophytes (Swingle 1957). Although a fewstates allow the production and distribution of grass carp, many others,
fearing the potential for establishment of naturalized populations, do not.Successful reproduction of an introduced species requires two distinctcomponents. First, sexually mature, reproductively capable Individuals ofboth sexes need to be present. Second, adequate habitat needs to beavailable for spawning, egg incubation, embryo development, and fry and
fingerllng survival. Clearly, diploid grass carp develop viable
reproductive organs and are quite capable of producing normal gametes.
However, it has been questioned whether acceptable habitat exists among the
large river systems in the United States. Stanley et al. (1978) concluded
that habitats did exist that would allow limited reproductive success. The
Increasing catches of grass carp throughout the southern Mississippi Riversystem seem to indicate that this prediction was correct.
As a means to prevent possible establishment of naturalized grass carp in
this country, genetic manipulative techniques have been employed in an
attempt to produce a variation of the normal diploid grass carp that is
sterile but retains the capacity to effectively consume large quantities of
aquatic macrophytes. One such manipulation has been to hybridize grass
4-2
carp females with bighead carp (Hypophthalmlchthys nobIlls) males (Bakos etal. 1978, Marlan and Krasznal 1978). This hybrid cross resulted in the
production of both diplold and triplold Individuals (Magee and Phlllpp1982). These two fish have been taxonomically classified into distinctgenera. Intergeneric fish hybrids, even when viable to adulthood, haveproven to be reproductively incapable (Phllipp et al. 1983, Parker et al.
1984). However, an electrophoretic evaluation of the two carp specieshybridized (Magee and Phlllpp 1982) revealed that their genetic distancewas much less than that proposed by Avise (1976) for distinct genera. They
more closely, resembled subspecies or sibling species, taxonomicclassifications known to readily hybridize and to produce fertile F1hybrids. Since the taxonomic status of these fish is uncertain,predictions of sterility for the diplold hybrid based upon Its Intergenericstatus would be highly speculative. Predictions of sterility for thetriplold hybrid based upon Its ploldy has some merit. However,
reproduction of diploid species (originally produced through a naturalhybridization event) through a unique melostic procedure, termedhybridogenesis, has been demonstrated (Cimlno 1972, VrlJenhoek 1975).
Clearly, the most rel lable way of determining the reproductive capabl lti es
of these hybrids would be to raise some Individuals to sexual maturity andexamine their gonads. This was done In the present study and the resultsshowed that although both sexes of grass carp and bighead carp haddeveloped gonads, virtually no gonadal development had occurred In the
hybrids (Part 1: Chapter 5). It might be argued that older, larger hybridsmay develop gonads; however, studies by Krasznal (personal communication)on a few larger individuals revealed highly abnormal gonadal development,as well. Based upon this study and those of Krasznal, we feel that boththe diploid and triplold hybrid carp have been shown to be functionallyster II e.
Recently, though, another genetic manipulative technique has resulted in
the production of triploid grass carp. Since these fish appear to be much
more prol if Ic consumers of aquatic vegetation than the hybrid carp (Part 2:
Chapter 1), and since these fish appear easier to produce and raise In
4-3
large numbers, their potential for use as biological control agents appears
greater than the hybrids at this time. Based solely upon the ploidy of
this fish, it has been predicted to be sterile (J. M. Ma one, personal
communication). Not enough time has elapsed since the formation of thesetriplold grass carp to physically assess sexual maturation. Thus, only a
prediction of its reproductive abilities can be offered based upon resultswith other triplold fishes.
Triplold Individuals have been naturally reported (Cuellar and Uyeno 1972,Gold and Avise 1976, Allen and Stanley 1978, Thorgaard and Gall 1979) andartificially formed (Swarup 1959, Purdum 1972, Valenti 1975, Refstle et al.1977, Stanley and Allen 1979, Chourrout 1980, Gerval et al. 1980, Lemolne
et al. 1981, Wolters et al. 1981) In a number of fish species. Not all of
these fishes, though, were assessed for gonadal development. Wolters et
al. (1982) showed that trIplold channel catfish (Ictalurus punctatus),
produced through cold shock (Wolters et al. 1981), experienced quiteabnormal gonadal development. These authors concluded that both sexes of
this triplold fish were functionally sterile. Similar conclusions havebeen reached regarding triplold rainbow trout (Salmo galrdneri) producedthrough heat shock (Thorgaard and Gall 1979, Thorgaard et al. 1981).
Predictions regarding the reproductive capability of trlplold fishes ingeneral must be based upon those very few examples cited. These examplesindicate that functional sterility Is at least widespread among triploid
species. However, to state that sterility is universal among trlplolds ispremature. Discussion among members of the Grass Carp-Hybrid CarpSubcommittee, Exotic Fishes Ccmmittee, American Fisheries Society, duringthe 1984 annual meeting held at Cornell University focused on this topic.
They concluded that a statement guaranteeing sterility of the trlploldgrass carp was not warranted, since it would have been totally speculative.
However, based upon results only with other species of fish, it was agreed
that the probability for similar abnormal ties In gonadal development
occurring among triplold grass carp was very high. It was also unanimously
agreed upon that the triplold grass carp certainly poses no greater threat
for naturalization than the diploid. Although proof of functional
4-4
sterll ty awaits further growth and aging among existing specimens of
triplold grass carp, the overwhelming opinion of most involved fisheriesbiologists is that the reproductive capability of the triploid grass carpwill be considerably less than that of the d plold grass carp.
LITERATURE CITED
Allen, S. K., Jr., and J. G. Stanley. 1978. Reproductive sterility inpolypl od brook trout, Salvel lnus fontlnalls, Trans. Am. Fish.Soc. 107:473-478.
Avise, J. C. 1976. Genetic differentiation during speclation, p. 106-122.
1JL F. J. Ayala, ed. Molecular evolution. Sinauer Associates,Sunderland, MA.
Bakos, J., Z. Krasznal, and T. Marian. 1978. Cross-breeding experimentswith carp, tench, and Asian phytophagus cyprinids. Aquacult. Hung.(Szarvas) 1:51-57.
Chourrout, D. 1980. Thermal induction of diploid gynogenesis and
triploldy In the eggs of the rainbow trout (Salmo galrdneri
Richardson). Reproduction, Nutrition, Development 20:727-733.
Cimino, MC. 1972. Melosis in triplold all-female fish (Poec LLLopsis.Poec iIdae). Science 1975:1484-1486.
Cuellar, 0., and T. Uyeno. 1972. Triploldy In rainbow trout.Oytogenetics 11:508-515.
Gerval, J., S. Peter. A. Nagu, L. Horvath, and V. Casnyl. 1980. Induced
triploldy in carp, Cyprlnus carpio L. J. Fish BIol. 17:667-671.
Gold, J.R., and J. C. Avise. 1976. Spontaneous triploidy in the
Ca lfornia roach Hesperoieucus svymetricus (Pisces: OfprlnlIdae).
Oytogenetics 17:144-149.
4-5
Lemolne, H.L., Jr., and L. T. Smith. 1980. Polyploidy induced in brook
trout by cold shock. Trans. Am. Fish. Soc. 109:626-631.
Magee, S. M., and D. P. Philipp. 1982. Biochemical genetic analyses of
the grass carp female x bighead carp male F1 hybrid and the
parental species. Trans. Am. Fish. Soc. 111:593-602.
Marian, T., and Z. Krasznal. 1978. Karyological investigations on
Ctenopharyngodon jIdJ a and Hypophthalmichthys nobiL s and their
cross-breeding. Aquacult. Hung. (Szarvas) 1:44-50.
Parker, H. R., D. P. Phillpp, and G. S. Whitt. 1984. Gene regulatory
divergence among species estimated by altered developmental patterns
In interspecific hybrids. Molecul. Biol. Evol. (in press).
Phllipp, D. P., H. R. Parker, and G. S. Whitt. 1983. Evolution of gene
regulation: Isozymic analysis of patterns of gene expression during
hybrid fish development, p. 193-237. jI M. Ratazzi, J. G.
Scondal iss, and G. S. Whitt, ed. Isozymes. Current topics in
biological and medical research, vol. 10.
Purdom, C. E. 1972. Induced polyplolds in plaice (Plueronectes
platessa) and its hybrid with the flounder (Plaltlchthys flesus).
Heredity 29:11-24.
Refstle, T., V. Vassvik, and T. GJedrem. 1977. Induction of polyploidy in
salmonids by cytochalasin B. Aquaculture 10:65-74.
Stanley, J. G., W. W. Miley II, and D. L. Sutton. 1978. Reproductive
requirements and likelihood for natural izatlon of escaped grass carp
In the United States. Trans. Amn. Fish. Soc. 107:119-127.
4-6
Swarup, H. 1959. Effect of triploldy In the body size, generalorganization and cellular structure in Gasterosteus aculatus (L.).J. Genet. 56:143-155.
Swingle, H. 1957. Control of pond weeds by use of herbivorous fishes.Proc. South. Weed Conf. 10:11-17.
Thorgaard, G. H., and G. A. E. Gall. 1979. Adult triploids in a rainbowtrout failly. Genetics 93:961-973.
, M .E. Jazwin, and A. R. Stler. 1981. Polyploidy Induced by heatshock in rainbow trout. Trans. Am. Fish. Soc. 110:546-550.
Valenti, R. J. 1975. Induced polyploidy In Tllapla aura (Steindachner)by means of temperature shock treatment. J. Fish Biol. 7:519-528.
Vrljenhoek, R. C. 1975. Gene dosage In diploid and triplold unisexualfishes (Poecllopsls, Poecillidae), p. 463-475. In C. L.Markert, ed. Isozymes IV: genetics and evolution. Academic Press,NY.
Wolters, W. R., G. S. Libey, and C. L. Chrisman. 1981. Induction oftriploidy in channel catfish. Trans. Am. Fish. Soc. 110:312-314.
--- -- and '. 1982. Effect of triploidy on growth and gonaddevelopment of channel catfish. Trans. Am. Fish. Soc. 111:102-105.
Chapter 5
REPRODUCTIVE HISTOLOGY OF GRASS CARP (Ctenopharyngodon iJdeLa),
BIGHEAD CARP (Hypophthalmichthys noblis) AND THEIR F1 TRIPLOID HYBRID
INTRODUCTION
Because of the controversy over the reproductive potential of the grass
carp in the United States, a hybrid carp (female grass carp,
Ctenopharyngodon Idella x male bighead carp, Hypophthalmichthys nobIlls)
(Bakos et al. 1978) has been proposed as a sterile alternative.
Investigations were begun to empirically determine Its reproductive
potential relative to Its parent species.
METHODS
Hybrid carp, grass carp, and bighead carp used In these experiments were
received from Malone and Son Enterprises, Lonoke, Arkansas (Table 5-1).
All fish were from the 1979 year class. Prior to histological examination,
fish were held In a 0.08-ha pond at the Illinois Natural History Survey
pond complex, Champaign, Illinois. Water temperature was 23 C (73.4 F)
prior to and during Injections.
In 1982, subsamples of all three genetic types of carp were Injected with
human chorlonic gonadotropin (HCG) at a rate of 1 mL/0.37 kg of fish per
day for 2 days. Fish were Injected on 23 and 24 June. Individuals of all
three genetic types of carp were used as controls (not Injected). Gonads
of all fish (both Injected and controls) were removed and fixed In Bouln's
fluid. Standard histological samples were obtained using an AO Spencer
"820" microtome, sectioning from 8 to 10 microns. Tissue sections were
then stained in haemotoxyl In and eosin using the ethyl-alcohol schedule
described by Humason (1972, p. 46).
5-2
Table 5-1. Carps used for reproductive histology experiments In 1982and 1983. All fish are from the 1979 year class.
Bighead carp Grass carp Hybrid carp
1982 (Age 3+)n 4 4 2TL (mm) 330-350 310-430 350-400
1983 (Age 4+)n 7 6 4TL (mm) 599-702 510-612 395-465
5-3
In 1983, subsamples of each genetic type of carp were Injected with LH-RHa
(I uteniz ng-rel easing hormone) (Argent Chemical Laboratories, Redmond,
Washington) at a rate of 10 ug/kg of fish per day. Fish were Injected on
20 and 21 June using bacterlostatic water to form the hormone solution.
All fish were sacrificed and examined for gross signs of gonadal
development on 21 and 22 June. Histological procedures were similar tothose followed In 1982. Stained tissue sections were examined under 400xmagnification. Microscopic examination results were reported as: welldeveloped but not fl owng; developed; and not developed.
RESULTS
Upon autopsy, gonads of examined fish appeared as paired, white strands
located above the swim bladder and in direct contact with the dorsal
membrane Ilning the abdomen. The degree of development varied
independently of size and sex (Table 5-2). In 1982, three of fourbighead carp analyzed were females with oocytes present; the male had no
sperm present. Grass carp analyzed in 1982 consisted of three males withno sperm present and one female with oocytes present. In 1983, four of
seven bighead carp exhibited well developed gonads but without free-flowingsex cells; three of those four were females. The remaining three bighead
carp had developed gonads. Two of six grass carp in 1983 exhibited welldeveloped gonads with a 1:1 male to female ratio. The remaining four grasscarp showed poor to no development (Table 5-2). In both 1982 and 1983,hybrid carp examined showed no signs of gonadal development, nor was itpossible to determine the sex of the fi sh In either experiment (Table 5-2).
DISCUSS ION
Results of 1982 and 1983 experiments consistently showed a complete lack of
gonadal development In the hybrid carp, Indicating that the hybrids are
almost certainly functionally sterile, and as such, can be used for aquatic
macrophyte control without fear of reproduction In the wild. Similar-aged
parental carps did show saome development of sex organs, as expected.
5-4
Table 5-2. Total length (mm), sex, type of treatment, and gonadaldevelopment of bighead, grass, and hybrid carp In June 1982 and 1983. HGC= human chorlonic gonadotropin; LH-RHa = lIutenizing-releasing hormone.IND = sex Indeterminate. Gonadal development Is reported as + for welldeveloped but not flowing, - for developed, * for no development ofgametes, and 0 for no development of gonads.
GaneteCarp Year Age TL (mm) Injection Sex Development
Bighead 1982 3+ 360 None F3+ 330 HOG F -3+ 480 None F3+ 530 None M *
1983 4+ 702 LH-RHa M +4+ 618 LH-RH F -4+ 603 LH-RHa M -4+ 599 None F +4+ 675 None M -4+ 613 None F +4+ 605 None F +
Grass 1982 3+ 310 None M *3+ 310 HG M *3+ 430 None M *3+ 420 HG F -
1983 4+ 606 LH-RHa M +4+ 574 LH-RHa F +4+ 510 LH-RHa M *4+ 612 None M *4+ 570 None F *4+ 558 None F *
Hybrid 1982 3+ 350 None IND 03+ 400 HOG IND 0
1983 4+ 465 LH-RHa IND 04+ 445 LH-RHa IND 04+ 395 None IND 04+ 437 None IND 0
5-5
There are several physiological and environmental factors related to
development of sex cells and spawning activity, including nutrition, watertemperature, age and size of fish, and In the case of these experiments,
type and timing of hormonal Injections. These factors were not entirelyoptimal in this study. For example, good nutrition Is Important In the
development of sex cells prior to spawning; therefore, an adequate supplyof high quality vegetation and zooplankton must be available to grass andbighead carp, respectively, for proper development of sex cells (Stanl ey etal. 1978, Shireman and Smith 1983). Food supplies were adequate in theholding pond Initially. However, as time progressed, plant biomass wasreduced, possibly resulting in Insufficient food for optimal rates ofmaturation of sex cells.
On the other hand, experimental water temperatures (23 C) were within thedesired range for artificial spawning (personal communications, BobWattendorf, Jim Malone, and Jerome ShIreman). Induced spawning of carps islargely dependent upon cumulative degree days, i.e., the total heat sum
after 1 January (Hulsman 1979). Degree days must be within 1200-1400 forsuccessful hypophysation, a condition met in both years of our experiment.
Reproductive development in carp can be achieved by treatment with avariety of hormones. When hormones are injected into "brood stock," thepituitary gland is stimulated to secrete gonadotropins and initiatespawning. LH-RH~ Is not species-specific. If fish are Injected at primeseason for maturation, this compound works very rapidly (Bob Wattendorf andJim Malone, personal communications), speeding up the maturation process.However, if the fish are not "primed," HOG and carp pituitary are betterto initiate flowing sex cells. Bardach et al. (1972) also found that HOGalone does not work well on grass carp, but additions of small amounts ofcarp pitultary preparation promote stimulation of sex cells.
Wattendorf, Malone, and Shireman (personal communications) use fish
weighing 4.5-6.8 kg (10-15 Ibs) for brood stock. The carp used In our
experimental Injectlons may have been marginal brood stock, since the
largest fish were 3.8 (bighead carp), 2.85 (grass carp), and 1.2 kg
5-6
(hybrid carp). They were, however, within the size range for reproductive
maturity In the parental fish (Shireman and Smith 1983).
Generally. Injection of the above hormones into well nourished, mature carp
at an appropriate time should initiate spawning. This did not occur In our
experiments, possibly because one or more essential reproductive
requirements were not met. However, because there was no development of
sex cells in the hybrid carp In either experiment, there seems little
possibility of natural reproduction. In grass carp, gonadal development
Is sufficient to distinguish the sex histologlcally In age 1+ fish
(Shireman and Smith 1983). Since our 3- and 4-year-old hybrids were not
sexually distinguishable, we can conclude that the hybrid Is functionally
ster I e.
ACKN LEDGMENTS
Many thanks to Dr. Jo Ann Cameron, Physiology and Biophysics Department,
and Dr. Janice Bahr, Animal Science Department, University of Illinois, for
their contributions and thoughtful criticism of the methods employed.
REFERENCES
Bakos, J., Z. Krasznal, and T. Marian. 1978. Cross-breeding experiments
with carp, tench and Asian phytophagous cyprinids. Aquacult. Hung.
1:51-57.
Bardach, J.E., J. H. Ryther, and W. 0. McLarney. 1972. Aquaculture: the
farming and husbandry of freshwater and marine organisms. Wiley
Interscience, NY. 868 p.
Huisman, E. A. 1979. The culture of grass carp (Ctenopharyngodon idella
Val.) under artificial conditions. In J. E. Halver, ed. Finfish
nutrition and fishfeed technology, vol. 1. Proc.., World Symposium,
Hamburg, 20-23 June 1978.
5-7
Humason, G. L. 1972. Animal tissue techniques, 3rd ed. W. H. Freeman and
Co., San Francisco. 641 p.
Shireman, J. S., and C. R. Smith. 1983.
the grass carp, Ctenopharyngodon
1844). FAO Fisheries Synopsis No.
Synopsis of biological data on
Idel la (Cuvler and Valenciennes,135. 86 p.
Stanley, J. G., W. W. Miley III, and D. L. Sutton. 1978. Reproductive
requirements and I ikel ihood for natural Ization of escaped grass carp
In the United States. Trans. Am. Fish. Soc. 107(1):119-128.
Appendix A3. Analysis of variance for ecological Impacts ofherbivorous carp.
Table A3-1. Total organic carbon comparison between control, lowdensity, and high density groups for all ponds.
Analysis of Variance
Group d.f. SS MS F
Between 2 7395.636 3697.818 35.302Within 455 47660.583 104.749Total 457 55056.220
p = 0.000Kruskal-Wal Is p = 0.000, H = 56.307
Table A3-2. Particulate organic carbon ccmparison between control,low density, and high density groups for all ponds.
Analysis of Variance
Group d.f. SS MS F
Between 2 43.757 21.879 3.318Within 455 3000.385 6.594Total 457 3044.143
p = 0.037
Table A3-3.density, and high
Dissolved organic carbon cnmparison between control, lowdensity groups for all ponds.
Analysis of Variance
Group
BetweenWi th nTotal
d.f.
2455457
SS
7498.61532437.72539936.341
3749.30871.292
F
52.591
p = 0.000Kruskal-Wal I is p = 0.000, H = 77.606
A3-2
Table A3-4. Inorganic carbon comparison between control, lowdensity, and high density groups for all ponds.
Analysis of Variance
d.f.
2455457
SS
5284.07236243.98041528.053
2642.03679.657
33.168
p = 0.000Kruskal-Wal I s p = 0.000, H = 38.965
Table A3-5. Nitrate comparison between control, low density, andhigh density groups for all ponds.
Analysis of Variance
Group
BetweenWithinTotal
d. f.
2455457
SS
0.0160.2820.297
0.0080.001
F
12.535
p = 0.000Kruskal-Wal I s p = 0.000, H = 64.473
Table A3-6. Anmonia canparison between control, low density, andhigh density groups for all ponds.
Analysis of Variance
Group
BetweenWithinTotal
p = 0.000Kruskal-Wal I Is
d. f.
2455457
p = 0.000, H = 60.854
Group
BetweenWithinTotal I
SS
0.3943.5183.912
0.1970.008
25.471
- --
A3-3
Table A3-7. Total organic carbon comparison between control, lowdensity, and high density groups for Annex ponds.
Analysis of Variance
BetweenWithinTotal
d.f.°
2361363
SS
5208.88233546.19238755.074
2604.44192.926
28.027
p = 0.000Kruskal-Wallis p = 0.000, H = 44.878
Table A3-8. Dissolved organic carbon comparison between control, lowdensity, and high density groups for Annex ponds.
Analysis of Variance
d. f.
2361363
SS
3118.25422221.90825340.161
1559.12761.557
F
25.328
p = 0.000Kruskal-WalI Is
Table A3-9.density, and high
p = 0.000, H = 38.115
Inorganic carbon comparison between control, lowdensity groups for Annex ponds.
Analysis of Variance
Group
BetweenWithinTotal
p = 0.000Kruskal-Wal Is
d. f.
2361363
6735.64425805.79432541.439
p = 0.000, H = 61.102
Group
Group
BetweenWI thinTotal
SS
3367.82271.484
F
47.113
A3-4
Table A3-10. Total phosphorus comparison between control, lowdensity, and high density groups for Annex ponds.
Analysis of Variance
Group d.f. SS MS F
Between 2 0.025 0.013 4.191WithIn 361 1.091 0.003Total 363 1.116
p = 0.016
Table A3-11. Soluble orthophosphate canparison between control, lowdensity, and high density groups for Annex ponds.
Analysis of Variance
Group
BetweenWithinTotal
d. f.
2361363
SS
0.0260.7950.821
MS
0.0130.002
F
5.846
p = 0.003Kruskal-Wal I is p = 0.85, H = 4.940
Table A3-12. Nitrate comparison between control, low density, andhigh density groups for Annex ponds.
Analysis of Variance
Group d.f. SS MS F
Between 2 0.008 0.004 6.710Within 361 0.212 0.001Total 363 0.219
p = 0.001Kruskal-Wal I s p = 0.000, H = 36.261
A3-5
Table A3-13. Amonia comparison between control, low density, andhigh density groups for Annex ponds.
Analysis of Variance
Group
BetweenWlthinTotal
d.f.
2361363
SS
0.1962.5882.783
0.0980.007
13.66
p = 0.000Kruskal-Wal Is p = 0.000, H = 34.181
Table A3-14. Chlorophyll a comparison between control, low density,and high density groups for Annex ponds.
Analysis of Variance
d.f.
2751753
SS
6175.077175904.346182079.424
3087.539234.227
13.182
p = 0.000Kruskal-Wal I s p = 0.000, H = 73.635
Table A3-15. Phaeophytin comparison between control, low density,and high density groups for Annex ponds.
Analysis of Variance
Group
BetweenWI thinTotal
p = 0.039Kruskal-Wal is s
d.f.
2751753
p = 0.000, H = 69.020
Group
BetweenWithinTotal
SS
630.75372751.36473382.117
MS
315.37696.873
F
3.256
- ---
A3-6
Table A3-16. Total chlorophyll comparison between control, lowdensity, and high density groups for Annex ponds.
Analysis of Variance
d.f.e
2641643
SS
7918.604565497.213573415.818
3959.302882.211
4.488
p = 0.012Kruskl-al-WalI s p = 0.000, H = 21.069
Table A3-17. Surface dissolved oxygenlow density, and high density groups for all
comparison between control,ponds.
Analysis of Variance
Group d.f. SS MS F
Between 2 95.571 47.785 9.261Within 469 2419.953 5.160Total 471 2515.524
p = 0.000
Table A3-18. Bottom dissolved oxygen comparison between control,low density, and high density groups for all ponds.
Analysis of Variance
Group d.f. SS MS F
Between 2 296.189 148.095 12.715Within 466 5427.711 11.647Total 468 5723.901
p = 0.000
Group
BetweenWithinTotal
A3-7
Table A3-19. pH comparison between control, low density, and highdensity groups for all ponds.
Analysis of Variance
Group
BetweenWithinTotal
d. f.
2389391
SS
28.649174.184202.833
14.3250.448
31.991
p = 0.000Kruskal-Wal I is p = 0.000, H = 51.004
Table A3-20. Free carbon dioxide camparisondensity, and high density groups for all ponds.
between control, low
Analysis of Variance
Group
BetweenWithinTotal
d.f.
2353355
SS
811.5445190.1646001.708
405.77214.703
F
27.598
p = 0.000Kruskal-WalI s p = 0.000, H = 44.953
Table A3-21. Alkalinity canparison between control, low density,and high density groups for all ponds.
Analysis of Variance
SS
53605.484654883.346708488.830
26802.7421755.719
p = 0.000Kruskal-Wal I Is p = 0.000, H = 25.287
Group
BetweenWithinTotal
d. f.
2373375
15.266
A3-8
Table A3-22. Total dissolved solids comparison between control, lowdensity, and high density groups for all ponds.
Analysis of Variance
Group d.f. SS F
44870.585505862.190550732.775
22435.2931433.037
p = 0.000Kruskal-Wal I s p = 0.000, H = 33.042
Table A3-23. Specific conductance comparison between control, lowdensity, and high density groups for all ponds.
Analysis of Variance
BetweenWithinTotal
d.f.
2353355
SS
95904.2891082160.8231178065.112
47952.1443065.611
p = 0.000Kruskal-Walls p = 0.000, H = 33.605
Table A3-24. Vertical I Ight extinction coefficient comparisonbetween control, low density, and high density groups for all ponds.
Analysis of Variance
Group
BetweenWithinTotal
p = 0.001Kruskal-Wal I Is
d.f.
2348350
8.869225.370234.238
p = 0.006, H = 10.112
BetweenWithinTotal
2353-355
15.656
Group F
15.642
SS F
4.4340.648
6.847
A3-9
Table A3-25. Surface dissolved oxygen comparison between control,low density, and high density groups for Annex ponds.
Analysis of Variance
Group d.f. SS MS F
Between 2 72.731 36.366 7.627Withtin 367 1749.760 4.768Total 369 1822.491
p = 0.001
Table A3-26. Bottom dissolved oxygen canparlson between control,lwc density, and high density groups for Annex ponds.
Analysis of Variance
Group d.f. SS MS F
Between 2 345.066 172.533 15.378Within 364 4083.981 11.220Total 366 4429.047
p = 0.000
Table A3-27. pH camparison between control, low density, and highdensity groups for Annex ponds.
Analysis of Variance
Group d.f. SS MS F
Between 2 27.811 13.905 30.556Within 299 136.066 0.455Total 301 163.876
p = 0.000
Table A3-28. Free carbon dioxide comparisondensity, and high density groups for Annex ponds.
between control, low
Analysis of Variance
d.f.
2271273
SS
745.9114965.2555711.166
F
372.95618.322
2.356
p = 0.000Kruskal-Wal I Is p = 0.000, H = 44.245
Table A3-29. Alkal Inity cmaparison between control, Iow density,and high density groups for Annex ponds.
Analysis of Variance
Group
BetweenWithinTotal
d.f.
2291293
SS
73063.210460232.611533295.822
36531.6051581.555
p = 0.000Kruskal-Wall Is p = 0.000, H = 40.584
Table A3-30. Total dissolved sol I ds canparlson between control, lowdensity, and high density groups for Annex ponds.
Analysis of Variance
d.f.
2271273
SS
75006.117369280.121444286.237
37503.0581362.657
p = 0.000Kruskal-Wal Ils p = 0.000, H = 52.947
A3-10
Group
BetweenWithinTotal
F
23.099
Group
BetweenWithi nTotal
F
27.522
wmwmNwmRmmmý
A3-11
Table A3-31. Specific conductance comparison between control, lowdensity, and high density groups for Annex ponds.
Analysis of Variance
d.f.
2271273
SS
159768.770820242.095980010.865
79884.3853026.724
p = 0.000Kruskal-Wallls p = 0.000, H = 53.284
Table A3-33. Vertical I lght extinction coefficient comparisonbetween control, low density, and high density groups for Annex ponds.
Analysis of Variance
d. f.
2300302
SS
7.671183.540191.211
MS
3.8350.612
p = 0.002Kruskal-Wal Its p = 0.010, H = 9.183
Table A3-33. Comparison of naupl abundance between pond groups.All ponds are included In the analysis.
Analysis of Variance
d. f.
2461463
SS
1.49x10 1 12.10x10122.25x1012
7.46x10 1 0
4.55x10 9
FO.05 (2,infinity) = 3.0Kruskal Wallls p = 0.000, H = 18.476
Group
BetweenWlthinTotal
F
26.393
Group
BetweenWi thinTotal
F
6.269
Source
BetweenWi thi nTotal
F
16.38***
mwm m-==
A3-12
Table A3-34. Canparlson of naupl iI blomass between pond groups.All ponds are Included In the analysis.
Analysis of Variance
Source
BetweenWithinTotal
d. f.
2461463
SS
3.07x10 5
4.30x106
4.61x106
F
1.54x10 5
9.34x10316.44***
FO. 0 5 (2,fnflnity) = 3.0Kruskal-Wal I Is p = 0.000, H = 18,466
Table A3-35. Canparison of rotifer abundance between pond groupsfor Annex ponds only.
Analysis of Variance
SS
3.13x10121.19x10 1 4
1.22x10 14
1.56x10 1 2
3.32x10 1 1
FO.0 5 (2, nf nlty) = 3.0Kruskal-Wallls p = 0.001, H = 14.252
Table A3-36. Ccmparison of rotifer bicmass between pond groups forAnnex ponds only.
Analysis of Variance
Source d.f. SS MS F
Between 2 3.54x10 5 1.77x10 5 4.705*Within 357 1.34x107 3.76x10 4
Total 359 1.38x10 7
FO. 0 5 (2, nfinity) = 3.0Kruskal-WalI s p = 0.004, H = 10.886
Source
BetweenWithinTotal
d. f.
2357359
F
4.712*
A3-13
Table A3-37. Canparison of rotifer abundance (nunber m-3 ) incontrol, low density, and high density pond groups In September 1983.
Analysis of Variance
d. f.Source
BetweenWithinTotal
22931
SS
1.88x10 1 1
6.20x10 1 18.08x10 1 1
0.94x10 1 1
0.21x10 1 14.402
p = 0.021Kruskal-Wal I s p = 0.081, H = 5.034
Table A3-38. Camparisondensity, and high density pond
of rotifer blamass (mg m-3) in control, lowgroups in September 1983.
Analysis of Variance
SS32758.21962350.75095108.969
16379.1092150.026
p = 0.001Kruskal-Wal I Is p
Table A3-39.Nematoda abundancedensity ponds).
Source
TreatmentErrorTotal
= 0.044, H = 6.259
Results of one-way analysis of variance of Infaunalby treatment type (control, low density, and high
Analysis of Variance
SS
15.66469.34285.007
d.f.
22830
7.8322.477
FS*
3.163
F0 .05 (2,28) = 3.34
*Analysis performed upon transformed (log(x+1)) data.
Source
BetweenWl th nTotal
d. f.
22931
7.618
A3-14
Table A3-40. Results of one-way analysis of variance ofChironcmidae abundance by treatment type (control, low density,density ponds).
Infaunaland high
Analysis of Variance
Source SS d.f. MS FS
Treatment 13688533.990 2 6844266.995 4.419Error 43368392.848 28 1548871.173Total 57056926.839 30
F0 . 0 5 (2,28) = 3.34
Table A3-41. Results of one-way analysis of variance of totalInfaunal community abundance by treatment type (control, low density, andhigh density ponds).
Analysis of Variance
Source SS d.f. IMS FS
Treatment 0.761 2 0.381 3.173Error 3.360 28 0.120Total 4.121 30
F0 . 05 (2,28) = 3.34
Table A3-42. Results of one-way analysis of variance of nunber oftaxa per infaunal sample by treatment type (control, low density, and highdensity ponds).
Analysis of Variance
Source SS d. f. MS FS
Treatment 254.926 2 127.463 3.216Error 1109.784 28 39.635Total 1364.710 30
F0 . 0 5 (2,28) = 3.34
A3-15
Table A3-43. Results of one-way analysis of variance of InfaunalChironamidae bioaass by treatment type (control, low density, and highdensity ponds).
Analysis of Variance
Source SS d.f. MS FS*
Treatment 9.796 2 4.898 6.611Error 20.743 28 0.741Total 30.539 30
F0 . 0 5 (2,28) = 3.34
*Analysis performed upon transformed (log(x+l)) data.
Table A3-44. Results of one-way analysis of variance of totalInfaunal community blcmass by treatment type (control, low density, andhigh density ponds).
Analysis of Variance
Source SS d.f. MS FS*
Trealment 5.335 2 2.668 8.960Error 8.336 28 0.298Total 13.671 30
F0 . 05 (2,28) = 3.34
*Analysis performed upon transformed (log(x+1)) data.
Table A3-45. Results of one-way analysis of variance of epiphyticChironomldae abundance by treatment type (control, low density, and highdensity ponds).
Analysis of Variance
Source SS d.f. MS FS*
Treatment 1.751 2 0.876 4.828Error 4.352 24 0.181Total 6.104 26
FO. 0 5 (2,24) = 3.40
*Analysis performed upon transformed (log(x+1)) data.
A3-16
Table A3-46. Results of one-way analysis of variance of number oftaxa per epiphytic sample by treatment type (control, low density, andhigh density ponds).
Analysis of Variance
Source SS d.f. MS FS
Treatment 872.523 2 436.261 5.746Error 1822.144 24 75.923Total 2694.667 26
F0 .05 (2,24) = 3.40
Table A3-47. Results of one-way analysis of variance of epiphyticAmphipoda bcanass by trealment type (control, low density, and highdensity ponds).
Analysis of Variance
Source SS d.f. MS FS*
Treatment 5.236 2 2.618 3.665Error 17.144 24 0.714Total 22.381 26
F0 .0 5 (2,24) = 3.40
*Analysis performed upon transformed (log(x+1)) data.
Table A3-48. Results of one-way analysis of variance of wholecanmunity Gastropoda abundance by trealment type (control, Iow density,and high density ponds).
Analysis of Variance
Source SS d.f. MS FS
Trea ment 749666.697 2 374833.348 10.676Error 175547.348 5 35109.470Total 925214.045 7
F0 .0 5 (2,5) = 5.79
A3-17
Table A3-49. Results of one-way analysis of variance of wholecanmunity Chironanidae abundance by treatment type (control, low density,and high density ponds).
Analysis of Variance
Source SS d.f. MS FS
Treatment 265449859.596 2 132724929.798 10.703Error 62004529.367 5 12400905.873Total 327454388.963 7
F0 .0 5 (2,5) = 5.79
Table A3-50. Results of one-way analysis of variance of wholecommunity total abundance by treatment type (control, low density, andhigh density ponds).
Analysis of Variance
Source SS d.f. MS FS
Treatment 681851621.709 2 340925810.855 10.756Error 158485808.002 5 31697161.600Total 840337429.711 7
F0 . 0 5 (2,5) = 5.79
Table A3-51. Young-of-the-year largemouth bass production (kg/ha)camparison between control, low density, and high density groups for Annexponds in 1982 and 1983.
Analysis of Variance
Source d.f. S.S. IM.S. F Ratio F Prob.
Between 2 15225.021 7612.511 79.499 0.000WithIn 6287 602017.979 95.756Total 6289 617243.000
n = 6301
A3-18
Table A3-52.canparison betweenponds (1981-1983).
Finglering Iargemouth bass production (kg/ha)control, low density, and high density groups for Annex
Analysis of Variance
Source d.f. S.S. M.S. F Ratio F Prob.
Between 2 499.050 249.525 27.017 0.000Within 222 2050.330 9.236Total 224 2549.380
n = 238
Table A3-53. Breeder largemouth bass production (kg/ha) comparisonbetween control, low density, and high density groups for Annex ponds (allyears).
Analysis of Variance
Source d. f. S.S. M.S. F Ratio F Prob.
Between 2 421.370 210.685 42.688 0.000Within 74 365.225 4.935Total 76 786.596
n = 93
Table A3-54. Young-of-the-year bluegill production (kg/ha) comparisonbetween control, low density, and high density groups for Annex ponds(1982 and 1983).
Analysis of Variance
Source d.f. S.S. M.S. F Ratio F Prob.
Between 2 6130872.011 3065436.0064 511.566 0.000Within 36641 24889359.172 679.462Total 36643 31020231.183
n = 36,645
A3-19
Table A3-55. Breeder bluegill production (kg/ha) comparison betweencontrol, low density, and high density groups for Annex ponds (all years).
Analysis of Variance
Source d.f. S.S. M. S. F Ratio F Prob.
Between 2 94.100 47.050 21.150 0.000Within 110 244.706 2.225Total 112 338.805
n = 129
Table A3-56. Four-inch channel catfish production (kg/ha) comparisonbetween control, low density, and high density groups for Annex ponds (allyears).
Analysis of Variance
Source d.f. S.S. M.S. F Ratio F Prob.
Between 2 151.900 75.950 150.347 0.000Within 34 17.176 0.505Total 36 169.076
n = 53
Table A3-57. Six-inch channel catfish production (kg/ha) comparisonbetween control, low density, and high density groups for Annex ponds (allyears).
Analysis of Variance
Source d.f. S.S. M.S. F Ratio F Prob.
Between 2 198.139 99.070 118.674 0.000Within 65 54.263 0.835Total 67 252.402
n = 84
A3-20
Table A3-58. Eight-inch channel catfish production (kg/ha) comparisonbetween control, lcw density, and high density groups for Annex ponds (allyears).
Analysis of Variance
Source d.f. S.S. M.S. F Ratio F Prob.
Between 2 128.506 64.253 41.565 0.000Within 73 112.847 1.546Total 75 241.353
n = 92
Table A3-59. Ccmparlson of breeder bass change in weight fromstocking to harvest for Annex pond groups, 1980-1983.
Analysis of Variance
Source d.f. S.S. M.S. F Ratio F Prob.
Between 2 98624.300 49312.15 6.969 0.002Within 87 615563.271 7075.44Total 89 714187.571
n = 93
Table A3-60. Comparison of fingerl Ing bass change in weight fromstocking to harvest for Annex pond groups, 1981-1983.
Analysis of Variance
Source d.f. S.S. M.S. F Ratio F Prob.
Between 2 20496.895 10248.447 8.316 0.000With In 199 245247.479 1232.399Total 201 265744.374
n = 205
A3-21
Table A3-61. Comparison of bass young-of-the-year change In weightfrcm stocking to harvest for Annex pond groups, 1982-1983.
Analysis of Variance
Source
BetweenWithinTotal
d.f.
210731075
S. S.
427.60111953.24712380.848
M.S.
213.80011.140
F Ratio F Prob.
19.192 0.000
n = 1076Kruskal-WalI s p = 0.000, H = 32.973
Table A3-62. Comparison of breeder blueg ll change in weight frcmstocking to harvest for Annex pond groups, 1980-1983.
Analysis of Variance
Source d.f. S.S. M.S. F Ratio F Prob.
Between 2 9313.002 4656.501 2.607 0.078Within 123 219734.691 1786.461Total 125 229047.693
n = 130
Table A3-63. Comparison of blueglll young-of-the-year change inweight from stocking to harvest for Annex pond groups, 1982-1983.
Analysis of Variance
Source
BetweenWI th inTotal
n = 1139Kruskal-Wal 1
d.f.o
211361138
SS.
59.1071521.6451580.752
M.S.
29.5531.339
F Ratio F Prob.
22.063
Is p = 0.000, H = 31.359
0.000
-- --
A3-22
Table A3-64. Comparison of 4-inch channel catfish change In weightfrom stocking to harvest for Annex pond groups, 1980-1983.
Analysis of Variance
Source d.f. S.S. M.S. F Ratio F Prob.
Between 2 7653.731 3826.865 3.863 0.028Within 47 46555.426 990.541Total 49 54209.157
n = 53
Table A3-65. Canparison of 6-Inch channel catfish change In weightfrom stocking to harvest for Annex pond groups, 1980-1983.
Analysis of Variance
Source d.f. S.S. M.S. F Ratio F Prob.
Between 2 42275.331 21137.666 10.010 0.000Within 78 164704.809 2111.600Total 80 206980.140
n = 84
Table A3-66. Comparison of 8-inch channel catfish change In weightfrom stocking to harvest for Annex pond groups, 1980-1983.
Analysis of Variance
Source d.f. S.S. M.S. F Ratio F Prob.
Between 2 48489.138 24244.569 5.066 0.008Within 86 411576.665 47857.775Total 88 460065.804
n = 92
A3-23
Table A3-67. Sunmary of ANOVA of change In condition factor (K) forbreeder largemouth bass at the Annex site, all years.
Analysis of Variance
Source d.f. S.S. M. S. F p
Between 2 0.1588 0.0794 2.752 0.694Within 87 2.5100 0.0289Total 89 2.6688
n 90
Table A3-68. Summary of ANOVA of change In condition factor (K) forfingerl Ing largemouth bass at the Annex site, all years.
Analysis of Variance
Source d.f. S.S. M.S. F p
Between 2 0.0448 0.0224 0.612 0.5432Within 232 8.4948 0.0366Total 234 8.5396
n = 238
Table A3-69. Sunmary of ANOVA of change in condition factor (K) foryoung-of-the-year largemouth bass at the Annex site, 1981-1983.
Analysis of Variance
Source d.f. S.S. M S. F p
Between 2 1.9374 0.9687 17.867 0.000Within 1070 58.0102 0.0542Total 1072 59.9476
n = 1076
A3-24
Table A3-70. Summary of ANOVA of change In condition factor (K) forbreeder bluegill at the Annex site, all years.
Analysis of Variance
Source d.f. S.S. M.S. F p
Between 2 0.9167 0.4584 2.408 0.0942Within 124 23.6018 0.1903Total 126 24.5185
n = 130
Table A3-71. Summary of WOVA of change in condition factor (K) foryoung-of-the-year bluegill at the Annex site, 1982-1983.
Analysis of Variance
Source d.f. S.S. M. S. F p
Between 2 41.1724 20.5862 44.793 0.000Within 1133 520.7147 0.4596Total 1135 561.8871
n = 1139
Table A3-72. Summary of ANOVA of change in condition factor (K) for4" channel catfish at the Annex site, all years.
Analysis of Variance
Source d.f.S S.. M.S. F p
Between 2 0.0733 0.0367 1.433 0.2489Within 47 1.2026 0.0256Total 49 1.2759
n = 53
A3-25
Table A3-73. Sumary of ANOVA of change In condition factor (K) for6" channel catfish at the Annex site, all years.
Analysis of Variance
Source d.f. S.S. M.S. F p
Between 2 0.0910 0.0455 0.744 0.4786Within 78 4.7723 0.0612Total 80 4.8633
n = 84
Table A3-74. Summary of ANOVA of change In condition factor (K) for8" channel catfish at the Annex site, all years.
Analysis of Variance
Source d.f. S.S. M. S. F p
Between 2 0.0080 0.0040 0.400 0.6719Within 86 0.8631 0.0100Total 88 0.8711
n = 92