Title: Applied aspects of the thermal biology, ecology, and life history of the blue tilapia, Tilapia aurea (Pisces: Cichlidae) /
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00099493/00001
 Material Information
Title: Applied aspects of the thermal biology, ecology, and life history of the blue tilapia, Tilapia aurea (Pisces: Cichlidae) /
Physical Description: vii, 238 leaves : ill. ; 28 cm.
Language: English
Creator: Zale, Alexander V
Publication Date: 1984
Copyright Date: 1984
 Subjects
Subject: Forest Resources and Conservation thesis Ph. D
Dissertations, Academic -- Forest Resources and Conservation -- UF
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Thesis: Thesis (Ph. D.)--University of Florida, 1984.
Bibliography: Includes bibliographical references (leaves 214-237).
Statement of Responsibility: by Alexander V. Zale.
General Note: Typescript.
General Note: Vita.
General Note: Also published as Technical report no. 12.
 Record Information
Bibliographic ID: UF00099493
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000481796
oclc - 30754785
notis - ACP9413

Downloads

This item has the following downloads:

appliedaspectsof00zale ( PDF )


Full Text











APPLIED ASPECTS OF THE THERMAL BIOLOGY, ECOLOGY, AND LIFE HISTORY
OF THE BLUE TILAPIA, TILAPIA AUREA (PISCES: CICHLIDAE)





BY


ALEXANDER V. ZALE


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY



UNIVERSITY OF FLORIDA














ACKNOWLEDGMENTS


For guidance and assistance throughout my tenure at the University

of Florida, I extend to my graduate advisors, Dr. R. W. Gregory and Dr.

J. A. McCann, my sincere thanks and appreciation. Helpful suggestions

by Dr. M. W. Collopy and Dr. C. R. Gilbert substantially facilitated my

research.

Many individuals assisted me in my research. Without their help

this study would have been impossible. Special thanks go to Roxanne

Conrow, Kevin McKinstry, and John Wood who generously volunteered many

of their nights to assist me in the field. Others who provided

assistance or advice included C. Abercrombie, M. B. Bain, E. K. Balon,

J. Boccardy, P. G. Bohall, L. Brasel, N. A. Bruno, D. E. Canfield,

J. Carter, M. A. Clemons, J. P. Clugston, T. Edwards, R. K. Frohlich,

P. Germuska, T. Hingtgen, M. L. Hoffman, M. Hudy, D. Jennings,

M. Jennings, W. Kelso, R. F. Labisky, P. Layton, J. I. Maxwell,

S. McCann, C. L. Montague, C. Moore, R. Mulholland, J. M. Packard,

H. F. Percival, D. R. Progulske, K. Portier, K. Reddy, J. Richardson,

R. W. Rottmann, H. L. Schramm, P. L. Shafland, R. J. Wattendorf, and

D. K. Winter. I am grateful for their contributions and sincerely

apologize for any omissions.

This research was supported by the Florida Cooperative Fish and

Wildlife Research Unit and the National Fishery Research Laboratory -

Gainesville.
















TABLE OF CONTENTS


PAGE

ACKNOWLEDGMENTS .................................................... ii

ABSTRACT ....................................... ................... vi


CHAPTER

I. INTRODUCTION ................................................ 1

Background ............................................ 1
Objectives ............................................. 3
Nomenclature ......................................... 10

II. COLD TOLERANCE OF BLUE TILAPIA: EFFECTS OF SALINITY,
CHRONIC EXPOSURE TO ACUTELY-SUBLETHAL TEMPERATURES,
AND RATE OF TEMPERATURE DECLINATION ..................... 12

Introduction .......................................... 12
Methods ........................ ....................... 14
Effect of Salinity on Low Temperature Tolerance ........ 15
Effect of Chronic Exposure to Acutely-Sublethal
Temperatures ........................................ 15
Effect of Temperature Declination Rate ................. 16
Statistical Considerations ............................ 16
Results ................................................. 17
Effect of Salinity on Low Temperature Tolerance ........ 17
Effect of Chronic Exposure to Acutely-Sublethal
Temperatures ...... ................................23
Effect of Temperature Declination Rate ................. 23
Discussion ............................................ 27

III. PERIODICITY OF HABITATION OF A STENOTHERMAL SPRING
RUN IN NORTH-CENTRAL FLORIDA BY BLUE TILAPIA ........... 35

Introduction ........................................ 35
Study Area ........................................... 36
Methods .............................................. 42
Results .............................................. 42
Discussion ........................................... 45












IV. A LABORATORY COMPARISON OF GROWTH, SURVIVAL, AND FORAGING
ABILITIES OF EARLY LIFE HISTORY STAGES OF BLUE TILAPIA
AND LARGEMOUTH BASS ..................................... 51

Introduction .......................................... 51
Methods .............................................. 53
Survival and Growth Trials ............................. 54
Relative Foraging Ability Trials ...................... 56
Results ................................................. 58
Survival and Growth Trials ............................. 58
Relative Foraging Ability Trials ....................... 72
Discussion .. ............................................ 91
Advantages of Large Initial Size ....................... 95
Feeding Strategies ..................................... 98
Evolutionary Considerations ............................ 98
Potential Impacts .. ................................... 101

V. THE TROPHIC ECOLOGY OF EARLY LIFE HISTORY STAGES
OF BLUE TILAPIA IN LAKE GEORGE, FLORIDA:
OVERLAP WITH SYMPATRIC SPECIES ......................... 103

Introduction ......................................... 103
Methods ............................................. 103
Results ............................................. 104
Discussion .......................................... 110

VI. RELATIVE PREFERENCE OF LARGEMOUTH BASS FOR BLUE
TILAPIA AND BLUEGILL AS FORAGE ......................... 113

Introduction .......................................... 113
Methods ............................................. 114
Results ............................................. 115
Discussion .......................................... 117

VII. NEST-SITE SELECTION BY BLUE TILAPIA AND LARGEMOUTH BASS
IN SILVER GLEN SPRINGS RUN: CIRCUMSTANTIAL EVIDENCE
OF COMPETITIVE EXCLUSION ............................... 122

Introduction .......................................... 122
Methods ............................................. 124
Results ............................................. 125
Discussion .......................................... 135

VIII. REPRODUCTIVE BIOLOGY OF BLUE TILAPIA IN SILVER
GLEN SPRINGS RUN / LAKE GEORGE, FLORIDA ................. 142

Introduction .......................................... 142
Methods ............................................ 142
Results ............................................. 143
Discussion .......................................... 165












IX. AGE, GROWTH, AND MORPHOMETRIC RELATIONSHIPS OF BLUE TILAPIA
IN SILVER GLEN SPRINGS RUN / LAKE GEORGE, FLORIDA ....... 179

Introduction .......................................... 179
Methods ................................................. 179
Results and Discussion .................................. 180
Morphometric Relationships ............................ 180
Validation of Aging Technique ........................ 182
Age and Growth ....................................... 187

X. OVERVIEW AND CONCLUSIONS ................................. 198

Objectives .. .......................................... 198
Results and Implications .............................. 199
Recommendations for Further Research ................... 203
Thermal Biology ...................................... 204
Early Life History .................................... 207
Predator-Prey Relationships .......................... 207
Suppression of Largemouth Bass Spawning .............. 208
Reproductive Biology and Age-Growth Relationships ..... 208
Additional Comments and Recommendations ................. 209
Summary ............................................ . 212

LITERATURE CITED .............................................. 214

BIOGRAPHICAL SKETCH ........................................... 238
















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



APPLIED ASPECTS OF THE THERMAL BIOLOGY, ECOLOGY, AND LIFE HISTORY
OF THE BLUE TILAPIA, TILAPIA AUREA (PISCES: CICHLIDAE)


By


Alexander V. Zale


December 1984


Chairman: Richard W. Gregory
Major Departnent: Forest Resources and Conservation



Laboratory and field research was conducted on issues relevant to

the effective management of the blue tilapia, Tilapia aurea, an exotic

cichlid established in U.S. waters.

Cold tolerance of blue tilapia was enhanced at isosmotic

salinities. Time and temperature interacted to elicit cold death;

a model was developed to quantify this interaction. Considerable

heterogeneity in thermal tolerance among individual fish was observed.

Blue tilapia moved in and out of a stenothermal spring run in

direct response to seasonal changes in surface-water temperatures.

Survival, growth, and relative foraging abilities of early life

history stages of blue tilapia exceeded those of larvae of a

representative native centrarchid in laboratory experiments. The











success of the blue tilapia in colonizing systems and rapidly achieving

high abundances may be a result of this high relative proficiency in

exploiting available trophic resources. In Lake George, young tilapia

fed primarily on microcrustaceans and exhibited a high degree of

trophic overlap with larval shad.

In laboratory electivity experiments, largemouth bass offered

juvenile blue tilapia and bluegill of equal size exhibited a preference

for tilapia as forage.

Nest-site selection by blue tilapia was characterized in a

stenothermal spring run and compared to historical data on largemouth

bass nesting in the run prior to the presence of tilapia. Blue tilapia

preferentially selected nest sites in vegetated, shallow-water areas;

microhabitat characteristics corresponded to those formerly used by

largemouth bass. Because nesting by largemouth bass is behaviorally

suppressed under crowded conditions, the presence of nesting tilapia

may have been responsible for an observed decline in number of bass

nests in the run.

Blue tilapia in the Silver Glen Springs Run / Lake George system

spawned from March through June although nesting in the stenothermal

run was initiated in late December. Gonadal recrudescence was

correlated with increasing photoperiod. Fecundity estimates ranged as

high as 8599.

Maximum estimated age of blue tilapia in this system was 4+ years.

Growth was rapid; estimated total lengths at ages I, II, III, and IV

were 280, 359, 397, and 423 mm, respectively. Scale rings of males were

unreliable annuli.














CHAPTER I
INTRODUCTION



Background

Introduction of a new species has long been one of the fishery

biologist's most potent management techniques. Addition of an exotic

game fish can generate renewed public interest in a fishery, enhance

visibility of the responsible agency, and may even, on occasion,

improve the quality of fishing. Exotics have also been introduced to

control nuisance vegetation, insects, or fishes or to provide an

improved forage base for native species. Others have become established

accidentally, escaping from aquacultural facilities. Unfortunately,

successful, popular, ano ecologically 'benign' introductions are

infrequent. Rather, the expected benefits typically fail to

materialize, negative consequences arise, and reclamation proves to be

impossible (Elton 1958; McDowall 1968).

Florida is particularly susceptible to the introduction of exotic

fishes because of its subtropical climate, extensive environmental

perturbation, and abundance of ornamental aquarium fish farms

(Courtenay et al. 1974). Currently, 16 species of exotic fishes are

believed to be established in Florida waters and an additional 6

species are listed as possibly established (Shafland et al. 1983).

The blue tilapia, Tilapia ( = Sarotherodon = Oreochromis) aurea,

is an exotic cichlid of paramount concern in Florida. The species

naturally inhabits lowland lakes, ponds, rivers, and streams in Africa










from Senegal east to the Chad basin, the lower Nile, and Palestine

(Trewavas 1965; Philippart and Ruwet 1982). An important food fish, it

is propagated in aquacultural operations worldwide and supports

extensive commercial fisheries (Fryer and Iles 1972; Lowe-McConnell

1982; Philippart and Ruwet 1982; Trewavas 1982a). Blue tilapia were

first imported into Florida in August 1961, when the Florida Game and

Fresh Water Fish Commission acquired 3000 juvenile blue tilapia from

Auburn University to investigate the potential of the species as a

sport fish and as a Hydrilla control agent (Crittenden 1962). The fish

were stocked into naturally fertile, reclaimed phosphate pits at the

Pleasant Grove Research Station near Tampa. The species proved to be

neither an effective aquatic vegetation control agent nor a sport fish

(Buntz and Manooch 1968), and the experimental stock at Pleasant Grove

was exterminated. However, some of the fish were surreptitiously

transferred to nearby open waters by persons unknown (Buntz and Manooch

1968). The descendants of this transplant apparently comprise Florida's

current blue tilapia population. Early eradication attempts failed

(Ware 1973) and the species (occasionally aided by fishermen) expanded

its range through Florida's extensive canal and natural waterway

systems. It is now the most widely distributed exotic fish species

established in the state. Courtenay et al. (1984) recorded established

populations in the following 18 counties in central and south Florida:

Alachua, Brevard, Dade, DeSoto, Hardee, Hernando, Hillsborough, Lake,

Manatee, Marion, Orange, Osceola, Palm Beach, Pinellas, Polk, Sarasota,

Seminole, and Volusia. In addition, Foote (1977) reported the species

from Charlotte, Glades, Highlands, and Pasco counties. I have collected

the fish in the St. Johns River in Putnam County. Established










populations occur in two private ponds in Duval County (P. B, Southall,

personal communication). In light of its wide distribution and

inexpugnability the blue tilapia must be considered, at least for the

foreseeable future, a permanent component of Florida's ichthyofauna and

must be managed accordingly.



Objectives

Development of a sound management protocol for the blue tilapia

requires a thorough understanding of the life history, environmental

physiology, and ecology of the species. To this end, the Florida Game

and Fresh Water Fish Commission has conducted extensive research on the

species. However, considerable deficiencies remain in our knowledge of

the biology of the fish. My goal, in conducting the research presented

here, was to augment understanding of the ecology of the blue tilapia

on issues relevant to the effective management of the species in U.S.

wa ters.

When a species is introduced outside of its native range, it may

be subjected to physical conditions dissimilar to those to which it is

adapted. How the species is affected by these conditions, and how it

reacts to them, may have important implications in its management.

Because the blue tilapia is native to tropical and sub-tropical

regions, and because it is known to suffer mortality during winter in

temperate climates, I examined physiological and behavioral aspects of

the thermal biology of the species.

Not only is an organism affected by its physical environment, but

it must also, by necessity, interact with other biotic entities.

Therefore, introduction of an exotic species will alter the structure










and function of an ecosystem. The form and magnitude of these

alterations will influence how the introduced species is managed.

Tnerefore, I examined selected interactions between blue tilapia and

native organisms.

Finally, knowledge of the life history of a species in an exotic

environment must be obtained to effectively manage it there. Therefore,

I examined the most basic life history characteristics, reproductive

biology and age and growth, of the blue tilapia in north-central

Florida. The following objectives identify tne specific topics I

investigated.



1. To determine the effects of salinity, chronic exposure to

acutely-sublethal temperatures, and temperature oeclination rate on

the low temperature tolerance of blue tilapia.



A primary consideration in the management of an exotic species

concerns where the species can, and cannot, be expected to survive

outside of its native range. The sensitivity of blue tilapia to low

temperatures is apparently the most important, if not the only,

factor delimiting the eventual distribution of this species in North

America (Courtenay et al. 1980; Shafland and Pestrak 1982). However,

low temperature tolerance may be influenced by a number of factors. For

example, cold tolerance of euryhaline fishes may be enhanced at

isosmotic salinities. Additionally, extended exposures to acutely-

sublethal temperatures may ultimately result in death (Yashouv 1960;

Hauser 1977), and different rates of temperature declination may effect

manifold lower lethal temperatures (Fry 1947). To better characterize










the low temperature tolerance of the blue tilapia, and thereby

facilitate determination of its potential range in North America, I

examined the effects of salinity, cooling rate, and extended exposure

to acutely-sublethal temperatures on the cold tolerance of the species.



2. To determine the periodicity of habitation of a stenothermal spring

run by blue tilapia.



Blue tilapia have been observed congregating in heated effluents

(Buntz and Manooch 1968; Noble et al. 1975) and in stenothermal spring

runs (Shafland and Pestrak 1982) during periods when surface-water

temperatures are low. However, the phenomenon has never been

systematically documented. Preliminary speculations indicated that

these refugia may be critical to the survival of the species in north-

central Florida (J. A. McCann, personal communication). If, during

winter, blue tilapia actively select habitats where water temperatures

are elevated, then locations of these thermal refugia would have to be

considered when assessing habitat suitability on thermal criteria. To

elucidate this phenomenon, I documented the periodicity of habitation

of a stenothennal spring run by blue tilapia, testing whether

abunaances of blue tilapia in the run were inversely correlated with

surface-water temperatures.


I









3. To compare the relative growth, survival, and foraging abilities of

early life history stages of blue tilapia and largemouth bass

(Micropterus salmoides) at equivalent food concentrations.



The range of the blue tilapia in Florida has expanded rapidly

since introduction of the species in 1961; it is now the most

extensively distributed exotic fish established in the state.

Concurrently, populations of blue tilapia have demonstrated an ability

to attain high densities and dominate fish assemblages, often within a

few years of colonizing systems (Ware 1973; Germany and Noble 1977).

Phytoplanktivory by adults permits such densities to persist, but the

rapidity with which these abundances are achieved, often from very

small founder populations, and by a species with relatively low

fecundity, suggests that early life history survivorship in this

species can be relatively high. Because acquisition of sufficient food

is a major determinant of early life history survivorship in fishes, I

hypothesized that the success of the blue tilapia in colonizing systems

and rapidly achieving high abundances is a result of enhanced

survivorship and growth during early life history stages conferred by a

high relative proficiency in exploiting available trophic resources. I

therefore compared the growth, survival, and foraging abilities of

early life history stages of blue tilapia and a representative native

centrarchid, the largemouth bass (Micropterus salmoides), over a

range of food abundances in laboratory experiments. I predicted that if

tilapia were more proficient at exploiting available forage, then at

equivalent forage abundances survival and growth of tilapia would










exceed those of bass and that these differences could be traced to

interspecific differences in foraging performance.



4. To characterize and compare food selection of early life history

stages of blue tilapia and sympatric native species under natural

conditions.



A potential impact of the blue tilapia, heretofore not considered,

concerns exploitative competition between early life history stages of

blue tilapia and native species for trophic resources. If, as

hypothesized above, young blue tilapia are highly effective

zooplanktivores, then their presence may increase demand upon the

zooplanktonic resource of a system and thereby directly affect survival

of larvae of native fishes. However, evaluation of this possibility has

been hindered by the lack of information on food habits of young blue

tilapia in natural systems. I therefore examined the food habits of

young blue tilapia and sympatric species in Lake George, Florida, to

a) characterize the trophic ecology of young blue tilapia under natural

conditions, and b) determine the extent of interspecific overlap for

food between young blue tilapia and native fishes.



5. To determine the relative preference of largemouth bass for blue

tilapia and bluegill (Lepomis macrochirus) as forage.



Because blue tilapia have become a dominant component of fish

assemblages in many systems and may be displacing native forage fishes,

concern has developed regarding the effects of the species on predator-










prey interactions. The largemouth bass, a popular and economically

valuable sport fish in Florida, is a top predator and could potentially

be affected by changes in forage fish assemblages. Largemouth bass are

known to consume blue tilapia (Swingle 1960; Lewis and Helms 1964; Chew

1974; Ludbrook 1974; Noble et al. 1975; Shafland and Pestrak 1981;

V. P. Williams, personal communication) but the relative suitability of

blue tilapia as forage for bass has not been evaluated. I therefore

conducted laboratory electivity experiments to determine the relative

preference of largemouth bass for juvenile blue tilapia and bluegill, a

native centrarchid often prominent in the diet of largemouth bass.



6. To characterize nest-site selection of blue tilapia and largemouth

bass in a stenothermal spring run.



The possibility of competition between blue tilapia and native

centrarchids for spawning areas has commanded a great deal of public

attention (Harris 1978) and scientific speculation. Buntz and Manooch

(1968) collected spawning blue tilapia, bluegill, and redear sunfish

(Lepomis microlophus) sympatrical ly along the shoreline of Lake

Parker, Florida, and therefore inferred that competition for spawning

grounds occurred between these species. Noble et al. (1975) observed

that year classes of largemouth bass apparently failed to recruit in

Trinidad Lake, Texas, in the presence of high densities of blue

tilapia, and postulated that tilapia suppressed reproduction by bass.

Subsequent pond experiments (Noble et al. 1975; Shafland and Pestrak

1983) established that high densities of blue tilapia can inhibit or

suppress spawning by largemouth bass. Competition with centrarchids for









nesting habitat has therefore often been invoked as a major ecological

impact of blue tilapia in U.S. waters (e.g. Ware 1973; Courtenay et al.

1974; Taylor et al. 1984). However, no studies have been conducted to

investigate this interaction under field conditions. Aided by a pre-

tilapia study of largemouth bass nesting (Horel 1951), I examined nest-

site selection by blue tilapia and bass in a stenothermal spring run in

north-central Florida. Spawning by largemouth bass is behaviorally

suppressed in crowded environments (Chew 1972) in response to high

rates of interspecific encounter (Barwick and Holcomb 1976; Smith 1976;

Smith and Crumpton 1976). If microhabitat preferences of blue tilapia

for nest sites correspond to those of bass in allopatry, then

behavioral suppression of nesting by bass could be inferred.



7. To characterize the reproductive biology of blue tilapia in Silver

Glen Springs / Lake George, Florida.



8. To determine age, growth, and morphometric relationships of blue

tilapia in Silver Glen Springs / Lake George, Florida.



Knowledge of the life history characteristics of a species is

essential for its effective management. Because the life history of the

blue tilapia in open waters in the U.S. is poorly known, I described

the reproductive biology and age and growth relationships of the Silver

Glen Springs Run / Lake George blue tilapia population.









Each objective is addressed in a separate chapter. A final chapter

integrates the various sections and suggests promising avenues for

further research.



Nomenclature

Two taxonomic issues concerning the blue tilapia require

commentary. The original stock imported to the United States by Auburn

University was initially misidentified as T. nilotica, a closely

related species. The error was perpetuated in the early North American

literature concerning the species. Smith-Vaniz (1968) correctly

identified the fish, but many subsequent publications continued to

misidentify the species. Auburn University later acquired the real T.

nilotica, adding to the confusion. Caution should be exercised when

reading accounts regarding these species, especially those generated by

the early studies. The second issue concerns the generic nomenclature

within the tribe Tilapiini. Trewavas (1973) divided this complex into

two genera defined by structural and behavioral features. The genus

Tilapia (sensu strict) was differentiated from Sarotherodon (including

T. aurea (sensu lato)) structurally by the configuration of gill rakers

and pharyngeal and mesethnoid bones. Behaviorally, Sarotherodon

included the mouthbrooders, whereas substrate spawners were assigned to

Tilapia. Any reference to Sarotherodon aureus in the literature should

therefore be considered synonymous with T. aurea. Subsequently,

Trewavas (1982b, 1983) further divided the moutnbrooders. In

Sarotherodon brooding is paternal or biparental, in Oreochromis

maternal. Blue tilapia fall into the latter category. Reference to







11

Oreochromis is already common in the current literature (e.g. Edwards

et al. 1983; Gaigher and Krause 1983).













CHAPTER II
COLD TOLERANCE OF BLUE TILAPIA:
EFFECTS OF SALINITY, CHRONIC EXPOSURE TO ACUTELY-SUBLETHAL
TEMPERATURES, AND RATE OF TEMPERATURE DECLINATION



Introduction

The sensitivity of blue tilapia to low temperature is apparently

the primary, and possibly only, factor delimiting the eventual

distribution of this species in North America (Courtenay et al. 1980;

Shafland and Pestrak 1982). While the blue tilapia seems to be the most

cold-resistant of the mouthbrooding tilapias (Chervinski 1982),

considerable variation exists among reports of cold tolerance for the

species (Table 1). The lower lethal temperature (6-7 C) reported by

Shafland and Pestrak (1982) is probably the most relevant of the cited

temperatures pertaining to the blue tilapia in the southeastern United

States. In a controlled laboratory study using fish captured in

Florida, they decreased water temperatures by 1 C/day until death. Cold

tolerances of fishes may be altered by a number of factors, however,

including acclimation temperature, cooling rate, water quality,

duration of exposure, and experimental animal size. For example,

Chervinski and Lahav (1976) noted that blue tilapia maintained at 11 C

suffered less mortality in dilute sea water (5 ppt NaCl) than in

fresh water, and Allanson et al. (1971) observed that T. mossambica in

fresh water succumbed to secondary chill coma at 11 C, whereas fish

exposed to the same temperature in dilute sea water were unaffected.

These findings suggest that euryhaline thermophilic fishes may be more



















Table 1. Summary of reported effects of low temperature on blue tilapia.


Temperature
(C)


13-14

10

8

'8


6-7

5

8.9

5.0-8.9


Effect


Growth terminated

Ceased movement

Entered cold stupor

Sluggish gill
ventilation

Loss of equilibrium

Deati

Death

No mortality


S12.8 Death


11.7 No mortality


8.0-8.5 Death

9 Began to die


11 Began to die


10 Some mortality


6-9 Death




5.6 Median lower lethal
temperature

6 No mortality


13-16 Decrease in feeding
rate

10-12 Cessation of feeding

7-8 Loss of equilibrium

6-7 Death


Conditions and source


Ponds, laboratory (Yashouv 1960)












Ponds, fish <152 mm TL (McBay 1961)

Ponds, fish >152 mm TL, exposure of
less than 3 days (McEay 1961)

Ponds, fish 229-279 mn TL, exposure
of 32 days (McBay 1961)

Florida phosphate pits (Crlttenden
1965)

Laboratory (Sarig 1969)

Laboratory, acclimated to 18 C
(Chervinski and Lahav 1976)

Laboratory, acclimated to 28 C
(Chervinski and Lahav 1976)

Trinidad Lake, TX; short exposures
(Germany and Noble 1977)

Trinidad Lake, TX; total mortality
of population after exposure of
about one month (Germany and
Noble 1977)

Laboratory. 0.8 C/hour decrease
(Lee 1979)

Ponds, exposed for only a few hours
daily (Chervinski and Stickney 1981)

Laboratory, 1 C/day decrease
(Shafland and Pestrak 1982)









tolerant of low temperatures at isosmotic salinities than in hyper- or

hyposmotic media. Additionally, extended exposures to acutely-sublethal

temperatures may ultimately result in death (Yashouv 1960; Hauser

1977), and different rates of temperature declination may effect

manifold lower lethal temperatures (Fry 1947). These factors are

probably responsible for at least some of the variability in lower

lethal temperatures reported for blue tilapia. To better characterize

the low temperature tolerance of the blue tilapia, and thereby

facilitate determination of its potential range, I examined the effects

of salinity, cooling rate, and extended exposure to acutely-sublethal

temperatures on the cold tolerance of the species.



Methods

Experiments were conducted in a 500-liter fiberglass tank equipped

with a thermostatically-controlled chilling unit and viewing window.

The chilling unit maintained water temperatures within 0.5 C of target

but temperatures within the plexiglass compartments fluctuated only

+0.2 C. Dechlorinated tap water (pH 8.0-8.2) was used.

Experimental fish were selected from stock raised and maintained

in the laboratory in fresh water at 24-29 C. Fish were fed a commercial

pelleted feed once daily to satiation. Juvenile fish were used, as

juvenile blue tilapia are more resistant to cold temperatures than

adults (Shafland and Pestrak 1982). Fish were segregated individually

in plexiglass compartments within the tank. Twelve fish were used in

each trial.









Effect of Salinity on Low Temperature Tolerance

Fish from a single brood were used in this series of experiments.

Fish were acclimated to 20 C for at least one week prior to the

initiation of each trial. They were concurrently acclimated to

experimental salinities,at the rate of 1 ppt NaCl/day, by the addition

of synthetic saltwater aquarium salts. Trials were conducted at 0, 5,

11.6, 20, and 35 ppt NaC1 with corresponding conductivities of 250,

9300, 19800, 32000, and 53000 umhos/cm at 25 C, respectively. Following

acclimation, temperature in the experimental tank was decreased by 1

C/day. Temperatures associated with cessation of feeding, loss of

equilibrium, and death were recorded for each fish. Because only a

single tank and a limited quantity of salts were available, the trials

were run serially, over a period of 5 months, starting at 0 ppt and

ending at 35 ppt NaCl. An additional trial at 0 ppt was then conducted

to detect any shift in the thermal tolerance of the fish over the

course of the trials. A total of six trials (two at 0 ppt) was

therefore performed.



Effect of Chronic Exposure to Acutely-Sublethal Temperatures

Six trials were conducted to determine survival times of blue

tilapia subjected to acutely-sublethal temperatures (above 6 C) in

fresh water. In each trial, fish were initially acclimated to 20 C for

one week. Temperature in the experimental tank was then decreased by 1

C/day until a predetermined final temperature was reached. Temperature

was then held constant until all fish died. If no mortalities occurred

within 60 days, the trial was terminated. The number of days until loss

of equilibrium and until death, at each of the final temperatures, were









recorded for each fish. Final temperatures for the six trials were 7,

8, 9, 10, 11, and 12 C. Fish used in these trials were randomly

selected from a stock consisting of two broods of distinct parentage.



Effect of Temperature Declination Rate

In the salinity series, experimental temperatures were decreased

by 1 C/day. To determine the effect of alternative temperature

declination rates on the cold tolerance of blue tilapia, I conducted

two additional trials. In one, experimental temperatures were decreased

by 1 C every third day (1 C/3 days); in the other, temperatures were

decreased by 1 C every 4 hours (6 C/day). These temperature regimens

were designated the retarded and accelerated rates, respectively. Prior

to initiation of temperature declination, fish were acclimated to 20 C

for one week. Both trials were conducted in fresh water. Temperatures

associated with loss of equilibrium and death were recorded for each

fish. The two trials in fresh water (0 ppt) from the salinity series

(1 C/day) were integrated into the analysis of this component.



Statistical Considerations

Presence of treatment differences was evaluated using the

Kruskal-Wallis distribution-free one-way layout test (Hollander and

Wolfe 1973) for the salinity and declination-rate experiments.

Jonckheere's distribution-free test for ordered alternatives (Hollander

and Wolfe 1973) was employed for the chronic-exposure trials. To detect

which particular treatments differed from one another, a distribution-

free multiple comparisons procedure based on Kruskal-Wallis rank sums

was implemented (Hollander and Wolfe 1973).









Results

Effect of Salinity on Low Temperature Tolerance

The temperatures at which fish stopped feeding, lost equilibrium,

and died in the initial and final trials at 0 ppt NaC1 were not

significantly different (Tables 2-4), inferring that confounding due to

the serial nature of the experimental design was absent. However, the

median temperature at loss of equilibrium shifted from 7 C in the

initial trial to 6 C in the final trial, and the range of temperatures

at death was greater in the latter assay. Mean weight and range of

weight of fish were greater in the final than in the initial trial, but

temperature at death was not correlated with weight (P = 0.3312, r =

0.307; Spearman rank correlation; Snedecor and Cocnran 1980).

Salinity had a significant effect on the temperatures at which

blue tilapia ceased feeding, lost equilibrium, and died (all P <

0.005). The magnitude of the effect was slight, however (Fig. 1).

Temperatures at termination of feeding ranged widely within treatments

and overlapped broadly among the various salinities (Table 2). Fish

lost equilibrium at 6-7 C at 0, 5, and 11.6 ppt, but succumbed at

higher temperatures at the higher salinities (Table 3). Fish maintained

at intermediate salinities (5 and 11.6 ppt) all died at 5 C (Table 4).

The majority of the fish in freshwater died at 6 C; 3 died at 5 C and

one at 7 C in the final trial. Death occurred at 6-7 C and 9-10 C at 20

and 35 ppt, respectively (Table 4).

I expected blue tilapia to be most cold tolerant at the isosmotic

salinity (11.6 ppt, Beamish 1970) with resistance decreasing as

salinity deviated from isosmosity. Correlation of temperature at death

versus deviation of salinity from 11.6 ppt proved highly significant














Table 2 Temperatures at which blue tilapia terminated feeding at
salinities of 0, 5, 11.6, 20, and 35 ppt NaC1. Subscripts
for the 0 ppt assays denote the initial (i) and final (f)
trials. Bars below average ranks underline values not
significantly different (-=0.05; multiple comparisons
based on Kruskal-Wallis rank sums).



Salinity (ppt NaC1)

5 Of 0. 20 11.6 35


Median (C) 13.5 14 14 15 15 17

Range (C) 13-17 12-17 13-17 13-16 13-20 13-20

Average rank 26.96 29.62 30.42 34.92 39.46 57.62














Table 3. Temperatures at which blue tilapia lost equilibrium at
salinities of 0, 5, 11.6, 20, and 35 ppt NaC1. Subscripts
for the 0 ppt assays denote the initial (i) and final (f)
trials. Bars below average ranks underline values not
significantly different (OC=0.05; multiple comparisons
based on Kruskal-Wallis rank sums).



Salinity (ppt NaC1)

5 Of 11.6 0. 20 35


Median (C) 6 6 6 7 7 10

Range (C) 6-7 6-7 6-7 6-7 7-8 9-11

Average rank 21.67 21.67 23.96 37.71 47.50 66.50














Table 4. Lower lethal temperatures of blue tilapia at salinities of
0, 5, 11.6, 20, and 35 ppt NaC1. Subscripts for the 0 ppt
assays denote the initial (i) and final (f) trials. Bars
below average ranks underline values not significantly
different (o=0.05; multiple comparisons based on Kruskal-
Wallis rank sums).



Salinity (ppt NaCl)

5 11.6 Of 0. 20 35


Median (C) 5 5 6 6 7 9

Range (C) 5-5 5-5 5-7 6-6 6-7 9-10

Average rank 14.00 14.00 33.75 38.50 52.25 66.50

































Fig. 1. Median temperatures at which blue tilapia stopped feeding,
lost equilibrium, and died when subjected to a 1 C per day
decrease in temperature at salinities ranging from 0 to 35
ppt NaCl. Vertical bars denote ranges. Initial and final
assays at 0 ppt NaCl are designated i and f, respectively.


























25 Termination of Feeding
o Loss of Equilibrium
Death

20

0 I I
UJ 15



w
I- --



5 *



0 f
0 5 11.6 20 35
SALINITY (%o)









(P < 0.0001, r = 0.779; Spearman rank correlation), supporting this

hypothesis.



Effect of Chronic Exposure to Acutely-Sublethal Temperatures

Blue tilapia survived exposure to 12 C for 60 days (at which time

the trial was terminated), but eventually succumbed to cold death at

all temperatures below 12 C. The ultimate incipient lower lethal

temperature (the lowest temperature that can be survived indefinitely,

Fry 1947) for blue tilapia is therefore about 12 C. Highly significant

differences (P < 0.0002) in durations to loss of equilibrium and death

were present between treatments at final temperatures below 12 C.

Generally, lower temperatures resulted in decreased survival times

(Fig. 2). Median survival times were 2, 8, 6, 14, and 31 days at 7, 8,

9, 10, and 11 C, respectively (Table 5). Fish generally lost

equilibrium one day prior to expiring. The longer survival times of

fish tested at 8 C, compared to 9 C, may have been due simply to

experimental error; sample sizes were small and the difference was not

significant (Table 5). Conversely, the difference may have resulted

from genetic variability in thermal tolerance, as fish in these trials

were from two different broods. Considerable heterogeneity in survival

times was also apparent within the trials at 10 and 11 C (Fig. 2),

indicative of inherent variability in thermal tolerances among test

fish.



Effect of Temperature Declination Rate

The rate of temperature declination had a moderately significant

effect (0.010 < P < 0.025) on the temperature at which experimental






























Fig. 2. Survival times of blue tilapia in fresh water subjected to
7, 8, 9, 10, or 11 C following a 1 C/day decrease in
temperature. Letters denote frequencies of coincident
observations; a = 1 observation, b = 2 observations, etc.
























0.62838 C
DAYS = 0.02913 e6283

r2 = 0.843


9 10

TEMPERATURE (C)


7


qU *- i













Table 5. Days to loss of equilibrium and death for blue tilapia
subjected to chronic exposures to acutely-sublethal
temperatures. Bars below average ranks underline
values not significantly different (c(= 0.05; multiple
comparisons based on Kruskal-Wallis rank sums).



Temperature (C)

7 9 8 10 11


Loss of equilibrium

Median (days) 1 4 7 13.5 31

Range (days) 1-2 3-5 4-8 10-29 22-61

Average rank 6.5 19.0 30.0 42.9 54.1



Death

Median (days) 2 6 8 14 31

Range (days) 2-3 4-6 4-9 11-30 23-62

Average rank 6.5 20.0 29.0 42.9 54.1









fish lost equilibrium (Table 6). The directionality of the effect was

unclear, however, as the average ranks of the accelerated (1 C/4 h) and

retarded (1 C/3 days) rates fell between those of the two assays at 1

C/day. All fish, except one, lost equilibrium at 6 or 7 C regardless of

declination rate. The lone exception lost equilibrium at 5 C when

subjected to the accelerated rate.

The effect of temperature declination rate on temperature at death

was much clearer. Highly significant differences (P < 0.005) were noted

among treatments. All fish subjected to the accelerated rate died at 5

C whereas the median of those cooled at the retarded rate was 7 C

(range 6-7 C) (Table 7). In the 1 C/day assays, median temperature at

death was 6 C (range 5-7 C). The temperature at death for fish in the

accelerated-rate trial was significantly lower than in the other assays

(Table 7).



Discussion

Varying degrees of salinity, particularly at or near isosmosity,

expand thermal tolerances in a number of euryhaline species (Loeb and

Wasteneys 1912; Gibson 1954; Arai et al. 1963; Craigie 1963; Alabaster

1967; Strawn and Dunn 1967; Garside and Jordan 1968; Allanson and Cross

1970; Allanson et al. 1971; Garside and Chin-Yuen-Kee 1972; Jordan and

Garside 1972). Upper lethal temperatures can be elevated as much as 6 C

in isosmotic media (Garside and Jordan 1968), but the magnitude of the

effect on lower lethal temperatures was heretofore unknown. My data

indicate that salinity does have a significant effect on the low

thermal tolerance of blue tilapia. Fish maintained in isosmotic media

generally succumbed to cold death at lower temperatures than tilapia in













Table 6. Temperatures at which blue tilapia lost equilibrium when
subjected to temperature declination rates of 1 C/4 h,
1 C/day, and 1 C/3 days. Two trials were conducted at the
intermediate rate. Bars below average ranks underline values
not significantly different (oc=0.10; multiple comparisons
based on Kruskal-Wallis rank sums).


Rate of temperature declination

1 C/day 1 C/4 h 1 C/ 3 days 1 C/day
(final) (initial)


Median (C) 6 6.5 7 7

Range (C) 6-7 5-7 6-7 6-7

Average rank 17.83 21.00 27.62 31.54













Table 7. Lower lethal temperatures of blue tilapia subjected to
temperature declination rates of 1 C/4 h, 1 C/day, and
1 C/3 days. Two trials were conducted at the intermediate
rate. Bars below average ranks underline values not signi-
ficantly different (=0.05; multiple comparisons based on
Kruskal-Wallis rank sums).


Rate of temperature declination

1 C/4 h 1 C/day 1 C/day 1 C/3 days
(final) (initial)

Median (C) 5 6 6 7

Range (C) 5-5 5-7 6-6 6-7

Average rank 8.00 24.38 28.00 37.62









water of higher or lower salinities, when subjected to a 1 C/day

decrease in temperature. Osmoregulatory failure is the direct mechanism

by which temperature-induced mortality is manifested (Houston 1962;

Heinicke and Houston 1965; Potts et al. 1967; Solomon and Allanson

1968; LUmminger 1969; Allanson and Cross 1970; Allanson et al. 1971;

Mackay 1971; Stanley and Colby 1971; Unminger 1971). At isosmotic

salinities, osmoregulatory stress is minimized, and temperature-induced

osmoregulatory failure is averted at temperatures that would elicit

death in hyposmotic or hyperosmotic media. The potential range of blue

tilapia in North America can therefore be expected to extend farthest

north along the coast, analogous to the restriction of T. mossambica to

estuaries at the southern extreme of its distribution in South Africa

(Allanson and Noble 1964). The difference in thermal tolerance between

fish in fresh water and isosmotic media was small (about 1 C); thus

additional habitable range may also be relatively small.

To predict where blue tilapia will survive, on the basis of

thermal requirements, necessitates an integrated analysis of the

thermal tolerance of the fish. Generally, evaluations of thermal

tolerances of fishes have been accomplished using the critical thermal

maximum/minimum (CTM) or upper/lower incipient lethal temperature (ILT)

methods. Fry (1947, 1967, 1971), Hutchison (1976), and Becker and

Genoway (1979) describe the methods in detail and review their relative

merits. Neither method is appropriate to determine if tilapia can

survive in a given system, as both employ environmentally-unrealistic

thermal schedules and discount the ability of fish to thermally

acclimate in waters undergoing slow cooling. Decreasing water

temperatures by 1 C/day was a convenient technique for evaluating the









effect of salinity on the low thermal tolerance of blue tilapia, and

unlike the CTM and ILT methods, the slow cooling rate (SCR) method

afforded test organisms opportunity to continuously acclimate to

temperatures decreasing at an environmentally relevant rate. However,

the actual temperatures at which the fish died are of limited use for

determining the suitability of a given body of water for overwintering

tilapia, unless the sequence of temperature declination in the system

closely follows the regimen enforced in the laboratory. Different rates

of temperature declination will result in mortalities at different

final temperatures. At declination rates commonly encountered in large

natural systems (0.3 to 1.0 C/day), however, this problem is minimized;

I did not detect significant differences in temperatures at death for

fish subjected to these rates. A more substantial deficiency of the SCR

method is demonstrated by the observation that blue tilapia will

succumb to cold death at temperatures appreciably above those at which

they die in an SCR trial, if subjected to such temperatures for an

extended period. For example, in SCR trials in fresh water, the median

temperature at death was 6 C. But when temperature declination was

terminated at 10 C, tilapia died nevertheless, albeit in about two

weeks. Cold death in fishes is not merely a function of temperature and

rate of temperature change, but also of time (Fry 1947; Brett 1956).

Defining a thermal limit in terms of temperature alone is therefore

errant; time must also be considered. In this respect, the ultimate

incipient lower lethal temperature may be a thermal milestone of some

consequence. By definition, it incorporates both temperature and time.

For blue tilapia in fresh water, this temperature is about 12 C.

Because cold death can occur at any temperature below 12 C, a safe









assumption might be that a blue tilapia occupying water cooler than 12

C is experiencing some degree of thermal stress. Estimating when such

stress will culminate in death may prove useful in evaluating the

potential of a given body of water to allow overwintering of tilapia.

I suggest here a methodology incorporating time and temperature

that evaluates the thermal stress accrued by an organism when subjected

to a given thermal schedule. The chronic-exposure technique

(essentially a hybrid of the SCR and ILT methods) provides the

necessary data. The cooling rate used in the chronic-exposure trials is

environmentally realistic (unlike the CTM) and allows fish to

continuously acclimate to decreasing temperatures (unlike the ILT).

Because cooling is terminated at a given level in each trial (unlike

the SCR), the impact of the final temperature on thermal tolerance can

be evaluated. The critical assumption of the methodology is that the

rate of dying at a given constant temperature below 12 C is linear with

respect to time, i.e. that each day at a certain temperature below 12 C

contributes equally to death. Acceptance of this assumption allows

calculation of 'daily coefficients of death' for temperatures below 12

C as follows. From the formula given in Figure 2 the predicted time to

death for tilapia held at 11 C is 29.3 days. The reciprocal of 29.3,

0.034, is the daily coefficient of death at 11 C. In essence, every day

that a tilapia is subjected to 11 C, it draws 3.4 % closer to death (if

the proceeding assumption is valid). The predicted time to death at 10

C was 15.6 days. Because these fish had been exposed to 11 C for one

day, the daily coefficient of death at 10 C is (1 0.034)/15.6 =

0.062. Similarly, the coefficients for 9, 8, 7, and 6 C are 0.108,

0.179, 0.260, and 0.357, respectively.










To evaluate the impact of a given temperature schedule on blue

tilapia survival, multiply the number of days at each temperature below

12 C by the appropriate coefficient of death. Only sets of consecutive

days below 12 C should be considered. Mortality of a significant

portion of a population of blue tilapia can be expected if the

summation of these products exceeds unity. With an adequate data base,

coefficients could be calculated for evaluating total mortality using

survival times of highly resistant individuals. Because of the nature

of the experiments from which the coefficients given here were derived,

application of the procedure (in the form presented above) is

restricted to environments where daily temperature shifts are less than

1 C.

Temperature schedules for large bodies of water.may be in accord

with this constraint, but in small systems temperatures may change

rapidly and fluctuate widely. Also, fish used in these trials were

acclimated to 20 C, then subjected to a 1 C per day temperature

declination; lower acclimation temperatures and slower rates of

temperature decrease may enhance thermal tolerance in this species.

Furthermore, the formula used to derive the coefficients may be

unreliable, as considerable variability was apparent in the data.

Clearly, the methodology requires further refinement and verification,

but the general approach of incorporating both time and temperature

appears promising, certainly more so than approaches based on

temperature alone.

While thermal tolerance undoubtedly affects the potential

geographic distribution of a species, evaluating the suitability of a

habitat on the basis of temperature requirements alone may be unsound.









Feeding, resistance to disease, successful reproduction, and sufficient

activity to permit existence in the face of competition or predation

are all necessary for the continued maintenance of a population.

Inability to maintain any one of these activities at moderately extreme

temperatures may be as decisive to continued survival as more extreme

temperatures are to immediate life (Brett 1956; Kinne 1963, 1970). The

tolerances reported here are therefore useful only for designating

those habitats that blue tilapia cannot invade, rather than those which

the species can successfully colonize.

As the range of blue tilapia extends into north Florida, the

opportunity for selection for enhanced cold tolerance may arise,

relegating the measures of thermal tolerance reported here obsolete.

Considerable heterogeneity in thermal tolerance was apparent among fish

sacrificed in the chronic-exposure experiments. Selection for cold

tolerance among wild fish is therefore likely.


~
















CHAPTER III
PERIODICITY OF HABITATION OF A STENOTHERMAL SPRING RUN
IN NORTH-CENTRAL FLORIDA BY BLUE TILAPIA



Introduction

Blue tilapia have been observed congregating in heated effluents

(Buntz and Manooch 1968; Noble et al. 1975) and stenothermal spring

runs (Shafland and Pestrak 1982) during periods when surface-water

temperatures are low. McBay (1961) observed that during winter, blue

tilapia in ponds occupied shallow water by day and retreated to deeper

areas at night in response to diurnal variation in temperature maxima.

Temperature is apparently an important directive factor regulating the

movements of this species. Fishes are attracted to areas where water

temperatures are at, or closest to, their preferred temperatures (Dendy

1945; Ferguson 1958; Gibbons and Bennett 1971; Beitinger 1977; Richards

and Ibara 1978; Winkler 1979; Brandt 1980); preferred temperatures

closely approximate metabolically-optimal temperatures (Crawshaw 1977).

The preferred temperature of blue tilapia is 30 C (Beamish 1970). Water

temperatures of spring runs in north-central Florida generally fall

within the range 21-24 C (Rosenau et al. 1977), whereas surface-water

temperatures can be much lower during winter. Habitation of

stenothermal spring runs during winter may therefore afford blue

tilapia physiological benefits, as well as offer thermal refugium from

lethal temperatures. Accordingly, abundances of blue tilapia in these

springs may be expected to be inversely correlated with surface-water









temperatures. If, during winter, blue tilapia actively select habitats

where water temperatures are elevated, then locations of these thermal

refugia would have to be considered when assessing habitat suitability

on thermal criteria.

The general consensus among people residing near springs in

north-central Florida, however, is that maximum abundances of blue

tilapia in the spring runs are attained not in January (when

surface-water temperatures are usually at their annual minimum in

Florida; Bradley 1974), but in late February and March, when surface

waters have already warmed appreciably. My preliminary observations

during 1980-81 led me to concur. In their native range, tilapia migrate

to clear waters to spawn (Lowe-McConnell 1953, 1959). Blue tilapia may

congregate in the spring runs because of the suitability of these

habitats for nesting, not because of thermal considerations. Therefore,

it was appropriate to test whether abundances of blue tilapia in

stenothermal spring runs were inversely correlated with surface-water

temperatures.



Study Area

I conducted a limited survey of 22 stenothermal spring runs in the

central and lower sections of the St. Johns River drainage during the

winter of 1980-81 to identify suitable study sites. The water clarity

of these habitats allowed direct observation of blue tilapia and their

distinctive spawning nests from shore, boat, or by snorkeling. Blue

tilapia were observed in springs from Lake Jessup north to Welaka (Fig.

3). Major concentrations were found in Blue Spring, Silver Glen

Springs, and Mud Spring. These medium to high discharge springs have































Fig. 3. Locations of stenothermal springs in central and lower
sections of the St. Johns River drainage surveyed during the
1980-81 winter. Presence and absence of blue tilapia are
denoted by solid and open circles, respectively.

























































N








10 km















Pan.e d. L.on S.


Alexander S


Wekva S.


Wadesbora S











Gr.-n Cov S.


Salt S I







SJun ver

Jonsper S.










short runs (< 1.0 km) emptying into the river. Other sizable springs in

the drainage (e.g. Wekiva, Juniper, Alexander, Ponce de Leon) connected

to the river with long runs (> 5 ki) attracted no blue tilapia. Springs

with short runs but low flow (e.g. Satsuma, Seminole, Forest) also did

not attract tilapia. Small numbers of blue tilapia were observed in

Clifton Springs, Salt Springs, Gemini Springs, Green Springs, and

Beecher Springs. Additionally, a single blue tilapia nest was found in

Welaka Spring, but no specimens were observed at this site. I selected

Silver Glen Springs for this study because of its large blue tilapia

population, easy access, and available support facilities.

Silver Glen Springs is a first magnitude spring in Marion County,

north-central Florida (Fig. 4). Discharge averages 3170 liters per

second (Rosenau et al. 1977). Water from the spring flows eastward

about 1 km through semi-tropical forest to the western shore of Lake

George, a widening of the St. Johns River. Silver Glen Springs Run

(SGSR) ranges in width from 40 to 130 m with a maximum depth of about 3

m. Substrate is comprised of sand, shell, and muck. Vallisneria

americana and Hydrilla verticillata are the predominant submerged

macrophytes. Water temperatures at the spring vent ranged from 22.8 to

24.0 C during this study. Because water passage down the run is rapid,

temperatures generally do not change appreciably along the course of

the run; during exceptionally cool or warm weather, however, a shift of

up to 2 C was observed. Depending upon wind direction and force, spring

effluent mixes with lake waters at the mouth of the run or extends in a

plume up to several hundred meters out into Lake George. During winter,

the plume usually spreads out across the surface over the cooler lake

water. In summer, run effluent forms a cooler underlying wedge. Blue





































V)











'S
LA
C,
a



a4




















L,
co


















4-
a
Cd







CC













41
C;
Co




a,
0




-I-,
Cd



LA


rC
a


0)


+-a
a



C-,






L~L















w

< 0
-Jw
0C


U)
z
R
a-

ILI
Cf)
z

w
-J



7..
LU
c>


z l-




-q


42

tilapia were first observed in SGSR in 1976 or 1977, and have steadily

increased in abundance thereafter (J. Morgan, personal communication).



Methods

Relative abundances of blue tilapia in SGSR were estimated from

October 1981 through April 1983 by gill net catch-per-unit-effort

(CPUE) and enumeration along line transects. Sampling was conducted

approximately monthly. The two methods were generally employed within a

few days of each other. Blue tilapia were counted by snorkeling along

five standardized 100-m by 4-m transects (Fig. 4) marked by anchored

nylon ropes. I swam each transect three times on each sampling date.

Three gill nets (2, 3, and 4 inch bar mesh; 51, 76, and 102 mm,

respectively) were set overnight at standardized stations,

perpendicular to the current (Fig. 4). The three nets, fished for one

night, constituted one unit of effort. Gill nets were pulled at

midnight and again at dawn.

Mean monthly air temperatures were used to estimate mean monthly

water temperatures (McCombie 1959; Anderson 1975). Water temperatures

in Lake George were recorded on each sampling date.



Results

Visual line transects and gill net CPUE were both effective for

determining relative seasonal abundances of blue tilapia in SGSR.

Estimates derived by the two methods were highly correlated (P <

0.0001, r = 0.904; Spearman rank correlation) and showed close

agreement in discerning seasonal trends in relative abundances (Fig.

5).



























41


CLL

EE
GJ

co.-
U,

CLW





-~W
C-

ci



Ln
U)





ai
LC




cnw

S.
0) (u

LW

W 4-










a)
10








S-
A ci

-4-,






CL



u <
-04-'/)
r-



.00
'4-









Lr
CL;
Co








*0W0



COL

4-'
WOW
0) C












o.....o (0o) 3UnIV13dl/131 NV3lN
o o o




"<-- ^ + '--







". ....... .o ".
o 0 0

















. .. .1 0










-o









0 2









C 4i


I. ..... ...........

S0 0 0 0
0 o

SNV31V\ 13JiSNV81



o o 0 0 0

.................... f"N a'l 13N 7711









Relative abundances of blue tilapia in SGSR followed a seasonal

pattern (Fig. 5). In late 1981, blue tilapia abundances increased

during October and November and peaked in December. The number of fish

present declined slightly in January 1982 and was considerably lower in

February. Abundances continued to decrease in March and April, and by

May very few blue tilapia remained in the run. Tilapia remained scarce

in SGSR throughout the summer of 1982. Abundances accrued again during

the fall and winter of 1982-83, but peaked later than in 1981-82

correspondent with the difference in temperature trends between the two

winters (Fig. 5).

Relative abundances of blue tilapia in SGSR were inversely

correlated with surface-water temperatures (Fig. 5). Correlations

between relative abundances and mean monthly temperatures were highly

significant transectss: P < 0.0001, r = -0.933; gill net CPUE: P <

0.0001, r = -0.893), as were correlations between relative abundances

and date-specific Lake George water temperatures transectss: P =

0.0006, r = -0.799; gill net CPUE: P < 0.0001, r = -0.917).



Discussion

Temperature is a major determinant of fish distributions and

movements (Reynolds 1977; Giattina and Garton 1982). Because most

fishes are unable to thermoregulate physiologically, they do so

behaviorally instead, changing locations to inhabit waters where

temperatures are most conducive to optimizing metabolic processes.

Tilapia are capable of detecting fine thermal gradients (Kutty and

Sukumaran 1975), and accordingly, moved in and out of SGSR to occupy

waters where temperatures were at or closest to their preferred









temperature. The temperature differential between Lake George and SGSR

was apparently the proximate directive factor governing the habitation

of the run by blue tilapia. Tilapia entered the run in large numbers

when water temperatures in Lake George fell below those in the run, and

emigrated when Lake George became warmer than the run. Delayed

immigration in the winter of 1982-83, coincident with delayed onset of

cold weather, stressed the significance of temperature in regulating

movements of the fish. The apparent increase in abundance of tilapia in

the run during early spring, when emigration was already in progress,

was simply an artifact of the conspicuousness of male tilapia and their

nests in shallow areas of the run.

The few tilapia found in SGSR during summer months may have

remained in the run because of the suitability of the site as a nesting

area, despite the physiological disadvantages incurred. Non-thermal

stimuli (e.g. social interactions, predation, feeding, competition,

habitat) can subordinate the influence of temperature considerably

(Norris 1963; Javaid and Anderson 1967; Brett 1971; Neill and Magnuson

1974; Kelso 1976; Stuntz and Magnuson 1976; Reynolds 1977; Magnuson and

Beitinger 1978; Ross and Winter 1981).

Water temperatures in Lake George did not decline to levels

immediately injurious to blue tilapia during this study, but were low

enough to cause cessation of feeding and some thermal stress (see

Chapter II). The lowest water temperature recorded in Lake George was

11.0 C. For significant cold-induced mortality to result, maintenance

of this temperature for at least 3 weeks would be necessary. However,

11 days earlier, water temperature in Lake George was 16.25 C, and 13

days later it was 14.0 C. No evidence of cold-induced mortality was









apparent during this period. Habitation of stenothermal spring runs

does not appear to be critical to the survival of populations of blue

tilapia in large bodies of water like Lake George in north-central

Florida. However, seasonal occupation of SGSR, as opposed to continuous

residency in Lake George, would nevertheless appear beneficial. Fish

inhabiting SGSR during winter were able to continue to feed (and grow).

Gonadal recrudescence was probably accelerated for blue tilapia in the

run and habitation of the run extended the reproductive season of blue

tilapia, as nesting activity was observed there several months before

it commenced in Lake George (see Chapter VII). Also, fish migrating to

the run were probably less likely to succumb to disease or predation

than were tilapia remaining in Lake George.

The thermophilic tendencies in habitat selection of blue tilapia

could result in aversion of cold death during exceptionally cold

winters or at higher latitudes. Therefore, locations of potential

thermal refugia must be considered when assessing habitat suitability

on thermal criteria (Chapter II). Heated water discharges of human

origin (e.g. power plants) as well as stenothermal spring effluents may

allow considerable range expansion beyond climatically-dictated limits.

All, or even a majority of blue tilapia in the Lake George section

of the St. Johns River probably do not enter SGSR during winter. The

occurrence of a tilapia in the run depends entirely on its chance

encounter of the thermal plume. Tilapia may range widely (tagged

individuals have been recaptured as far as 240 km from their release

site; Lowe-McConnell 1959), but generally their movements are modest

and largely random (Philippart and Ruwet 1982; Rinne and Wanjala 1982).

In the course of its movements, a tilapia, upon encountering the









thermal plume in winter would enter the run because of its

thermophilic predeliction. A fish that failed to encounter the plume

would remain in Lake George. The proportion of Lake George tilapia

entering the run would therefore be dependent on the probability of

encountering the thermal plume. The size of the thermal plume (relative

to the magnitude of Lake George), and the likelihood that blue tilapia

movements in Lake George are limited, would suggest that only fish

residing within several kilometers of the run would be apt to occupy

the run.

The absence of blue tilapia in winter from stenothermal springs

connected to the St. Johns River by long runs may be traced to the

rheophobic tendencies of the microphagous tilapias (the blue among

them). These fishes avoid currents, preferring still waters and pools

(Philippart and Ruwet 1982). They would not be expected to ascend

swiftly-flowing spring runs in the absence of thermal stimuli; effluent

temperatures of long spring runs are not elevated above river

temperatures at their outflows into the St. Johns River. Salt Springs

Run appears to be the only exception to this rule; despite its length

(6.4 km), moderate numbers of tilapia have been observed there each

winter for several years. Current velocity in this run is negligible,

however, due to its width and low gradient, and may be insufficient to

deter tilapia.

Peak abundances of blue tilapia in SGSR during the two winters

were similar. Gill net CPUE for December 1981 and February 1983 were

199 and 204 blue tilapia, respectively. Corresponding mean transect

counts were 94.5 and 104.5 blue tilapia per transect. The similarity of

blue tilapia abundances in SGSR in the two winters may be indicative of









stability in blue tilapia abundances in the Lake George section of the

St. Johns River during this study. Conversely, space limitation may

restrict the blue tilapia 'carrying capacity' of SGSR. The lack of

territoriality among juveniles and females favors the former

explanation, however. It may be premature to infer that blue tilapia

have attained their maximum abundance in Lake George, on the basis of

these data, however.

The visual line transect method proved comparable to gill net CPUE

in estimating the relative abundances of blue tilapia in SGSR. This

technique has been successfully implemented in enumerating cichlids

(McKaye 1977; W. Courtenay, personal communication) and other fishes

(e.g. Brock 1954; Bardach 1959; Northcote and Wilkie 1963; Keast and

Harker 1977; Goldstein 1978; Colton and Alevizon 1981; Sale and Douglas

1981). The transect method required no costly or specialized equipment,

was efficiently accomplished, did not require sacrifice of blue tilapia

or non-target species, and did not evoke negative public reaction. The

main shortcoming of the method is the necessity of adequate visibility,

a requirement limiting its applicability in Florida waters. It is an

excellent technique for investigating fish abundances in springs. A

shortcoming was the standardization of selected transect locations.

Randomization of transect locations would have allowed estimation of

the absolute abundance of blue tilapia present in the run.

The habitation of spring runs by blue tilapia affords sportsmen

the opportunity to utilize this species by bowfishing and

snatch-hooking. It may also provide fisheries managers the opportunity

to eradicate portions of blue tilapia populations. Best results, by







50

both groups, could be expected during, and immediately following,

periods when surface-water temperatures are lowest.















CHAPTER IV
A LABORATORY COMPARISON OF GROWTH, SURVIVAL, AND
FORAGING ABILITIES OF EARLY LIFE HISTORY STAGES OF
BLUE TILAPIA AND LARGEMOUTH BASS



Introduction

The range of the blue tilapia in Florida has expanded rapidly

since introduction of the species in 1961; it is now the most

extensively distributed exotic fish established in the state.

Concurrently, populations of blue tilapia have demonstrated an ability

to attain high densities and dominate fish communities, often within a

few years of colonizing systems (Ware 1973; Gennany and Noble 1977).

Phytoplanktivory by adults permits such densities to persist, but the

rapidity with which these abundances are achieved, often from very

small founder populations, and by a species with relatively low

fecundity (see Chapter VIII), suggests that recruitment in this species

can be relatively high.

Recruitment in fishes is primarily a function of survivorship in

early life history stages (LeCren 1962; Braum 1978; Hunter 1980), which

in turn is dictated principally by feeding, predation, and abiotic

factors (Jones 1973; Eipper 1975; Lett and Kohler 1976). Acquisition of

sufficient food not only averts death directly by precluding starvation

(Laurence 1977; Hunter 1981), but also deters predation; starvation can

increase susceptibility to predation (Ivlev 1961), and suboptimal diets

can slow growth rates (Riley 1966; Wyatt 1972; Houde 1975, 1977),

thereby protracting vulnerability to predation (Parker 1971; Cushing









1976; Taylor 1980; Hunter 1981). Effective exploitation of trophic

resources by young fishes is therefore an essential requisite for

survival.

Forage requirements of larvae of marine fishes have been

investigated extensively (e.g. O'Connell and Raymond 1970; Saksena and

Houde 1972; Wyatt 1972; Laurence 1974, 1977; Houde 1977, 1978)

primarily because of the long-standing (Hjort 1914, 1926), but

controversial (Marr 1956; May 1974; Laurence 1977), "critical period"

concept relating year-class strengths to prey densities at the

transition to exogenous feeding. Direct impacts of abiotic factors on

early life survival in the ocean tend to be ameliorated by the

environmental homogeneity and stability of marine systems (Moore 1966),

but heterogeneity of biotic variables (e.g. temporal fluctuations in

abundance or patchiness of zooplanktonic prey) can impact larval

survival. In freshwater habitats, climatic disturbances (e.g. storms,

cold fronts) significantly and rapidly alter aquatic conditions

(Ruttner 1966), and differences in year-class strengths of freshwater

fishes can often be traced to these abiotic perturbations (e.g. Kramer

and Smith 1962; Busch et al. 1975; Summerfelt 1975). Fluctuations of

edaphic productivity in freshwater systems are small in comparison to

abiotic variability; differences in year-class strengths due to abiotic

conditions are therefore much more pronounced. As a result, the forage

requirements of early life history stages of freshwater species have

received comparatively little attention and are poorly known (but see

Laurence 1971; Dabrowski 1975; Li and Mathias 1982). However,

limitations in zooplankton abundances can significantly affect survival

and growth of young freshwater fishes (Davis 1930; Langlois 1932;









Krumholz 1949; Noble 1975; Lemly and Dimmick 1982; Kashuba and Matthews

1984; Matthews 1984). Food-supply mediated mortality of young fishes in

freshwater habitats can be high, but because of relative constancy of

productivity between years does not manifest differential year-class

strength. However, differential recruitment of sympatric species with

dissimilar forage abundance requirements could occur; different species

vary in ability to survive under conditions of limited food

availability (Braum 1978; Saksena and Houde 1972; Hoagman 1974; May

1974; Houde and Schekter 1980).

I hypothesized that the success of the blue tilapia in colonizing

systems and rapidly achieving high abundances is a result of enhanced

survivorship and growth during early life history stages conferred by a

high relative proficiency in exploiting available trophic resources. I

therefore compared the growth, survival, and foraging abilities of

early life history stages of blue tilapia and a representative native

centrarchid, the largemouth bass (Micropterus salmoides), over a range

of food abundances in laboratory experiments. I predicted that if

tilapia were more proficient at exploiting available forage, then at

equivalent forage abundances survival and growth of tilapia would

exceed those of bass and that these differences could be traced to

interspecific differences in foraging performance.



Methods

Blue tilapia were spawned in outdoor pools at the Gainesville

National Fisheries Research Laboratory in Florida. Embryos were removed

from mouths of females shortly after fertilization. Embryos of

largemouth bass were collected from nests in spawning ponds at the









Welaka National Fish Hatchery, Welaka, Florida. Embryos of both species

were maintained in the laboratory at 27.0-28.0 C until the onset of

exogenous feeding.

Food organisms were collected from Lake Alice on the University of

Florida campus with 63-umn-mesh plankton nets. Use of wild zooplankton

insured the presence of a natural variety of sizes and species

duplicating that encountered by young bass and tilapia in Lake Alice.

The aggregate zooplankton concentration in each collection was

estimated by counting the numbers of organisms in 3 10-ml aliquots with

a dissecting microscope. The mean of the three samples was used as an

average concentration for the collection. Generally, several hundred

plankters were present in each 10-ml sample, and counts seldom differed

by more than 10 %. Appropriate volumes of the daily 'stock solution'

were measured and used as needed.



Survival and Growth Trials

Growth and survival of young blue tilapia and largemouth bass were

determined separately at prey concentrations of 0, 10, 50, 100, 500,

and 1000 zooplankters per liter in 25-liter glass aquaria. Two

replicates at each prey density were conducted for each species. Prey

concentrations were adjusted daily.

Because relative abundances of specific zooplankters procured from

Lake Alice changed daily, the biomass of zooplankton provided to each

fish species at numerically equivalent prey concentrations varied.

Survivorship was therefore analyzed by both number and biomass of

zooplankton provided. Formulae developed by Dumont et al. (1975) were

used to determine zooplankter weights. Furthermore, fish were able to









significantly alter nominal zooplankton concentrations between

adjustments during the final days of the trials when growth rates were

highest; virtually all zooplankters were consumed within 24 hours in

some trials. To standardize prey availabilities for growth analyses, an

index of prey biomass available to individual fish was derived by

dividing zooplankton biomass by the number of survivors in each trial.

Most of the water in each tank was removed daily during adjustment

of food concentration and replaced with filtered, dechlorinated tap

water. Water was aerated and gently agitated by compressed air supplied

through glass pipets. Water temperatures were maintained at 27.0-28.0 C

with immersion heaters. A 12:12 light:dark photoperiod was maintained

with fluorescent lighting, but other laboratory use precluded strict

adherence to the regimen.

Fish were stocked into the tanks at the onset of feeding (10 and 6

days after fertilization for tilapia and bass, respectively). Initial

stocking densities of fish were one per liter. Trials were of 16 days

duration. Surviving fish were counted every other day, but difficulty

was encountered in obtaining accurate counts during the first few days.

Dead individuals were removed when noted.

At the termination of the trials, survivors were counted, measured

to the nearest 0.05 mm with calipers, and individually weighed to the

nearest mg with an electronic microbalance. Interspecific comparisons

of growth were made only for weights as morphological differences

between species precluded viable comparisons of length. Survival and

growth of the two species were regressed against zooplankton

concentration and compared. Prey levels required to ensure 10 and 50 %









survival of each species were estimated from the regressions and by

probit analysis (Finney 1952).



Relative Foraging Ability Trials

Relative foraging abilities of young blue tilapia and largemouth

bass were determined by estimating prey consumption rates at various

prey densities; experimental procedures were similar to those of Houde

and Schekter (1980).

Fish used in these trials were maintained in a 500-liter

fiberglass tank following the onset of feeding. Water temperature was

maintained at 27.0-28.0 C. Copious quantities of wild zooplankton were

added to the tank daily to insure ad libitum feeding by fish;

zooplankton concentrations always exceeded 1000 per liter.

Prey consumption rates were estimated for blue tilapia and

largemouth bass at food concentrations of 10, 100, and 1000

zooplankters per liter. Fish were tested 3 days after feeding commenced

(DAFC) and every third day thereafter, up to 18 DAFC. Three replicates

of each combination of fish species, age, and zooplankton concentration

were performed. Trials were conducted in aerated, rectangular,

10-liter, Plexiglas aquaria. For each trial, a small number (3-20) of

randomly selected fish were transferred from the rearing tank to a test

aquarium. The prescribed quantity of zooplankton was introduced 2 hours

later and fish were allowed to feed for a specified interval (1-12

hours). Combinations of durations and numbers of fish were selected to

insure that enough prey were eaten to measure consumption rates without

reducing prey levels to less than half of initial concentrations. The









longest durations and largest numbers of fish were used in trials with

the youngest fish.

At the end of each trial, fish were removed from the experimental

aquaria; fish used in trials with prey concentrations of 1000 per liter

were fixed in 10 % buffered formalin, others were returned to the

rearing tank. At the conclusion of trials with prey levels of 10 and

100 per liter, the entire volume of each tank was filtered and the

remaining plankters were counted. The zooplankters remaining in two

1-liter samples were counted for the 1000 prey per liter trials.

Reductions in numbers of prey were used to estimate consumption rates.

No natural mortality of zooplankton could be detected in exploratory

trials without fish; I therefore inferred that reductions in numbers of

zooplankton were caused by fish predation. Confer (1971), Houde and

Scheckter (1980), and Drenner et al. (1932) similarly found the

methodology to be valid.

To convert prey consumption rates from number to biomass of

zooplankton ingested per fish per hour, stomach contents of the

preserved fish were examined. Zooplankters present were identified and

counted. Body lengths were measured to the nearest 10 pm with an ocular

micrometer mounted in a dissecting microscope, and dry weights were

calculated from these measurements using regression equations developed

by Dumont et al. (1975). Gut contents were examined only for fish from

the 1000 prey per liter trials; a paucity of test organisms precluded

sacrifice of fish from all trials. Because absolute abundance of prey

can affect foraging selection (Ivlev 1961; Estabrook and Dunham 1976;

Houde and Scheckter 1980; Rajasilta and Vuorinen 1983), the biomass

consumption rates calculated for the 10 and 100 prey per liter trials









may be unreliable. Converted consumption rates of bass and tilapia were

regressed against ages and weights of fish and compared by analysis of

covariance (Snedecor and Cochran 1980). Relative abundances of

zooplankters in stomach contents and in daily stock solutions were

compared to determine age-specific feeding electivities of both species

of fish.



Results

Survival and Growth Trials

Survival rates of both species increased in response to increases

in food concentration (Table 8). No individuals of either species

survived the 16-day trials in the absence of food, and total mortality

was observed among largemouth bass subjected to the 10 zooplankter per

liter prey level and in one replicate at 100 zooplankters per liter.

Survival of bass was low at or below 100 zooplankters per liter and

approached 50 % at the 500 and 1000 zooplankter per liter

concentrations (Fig. 6). The least squares fitted regression (arcsin

percent survivals on loglO zooplankton concentrations) predicted 10 and

50 % survivorship of largemouth bass at 44.8 and 1965.9 zooplankters

per liter, respectively. Probit analysis predicted 10 and 50 %

survivorship at 72.3 and 850.1 zooplankters per liter, respectively.

Survivorship of blue tilapia exceeded 50 % at all prey levels above 0

zooplankters per liter and became asymptotic at about 90 % at prey

levels above 50 zooplankters per liter (Fig. 6). Survivorship of 10 and

50 % was predicted at prey levels of 1.2 and 19.4 zooplankters per

liter, respectively, by the arcsin-transformed regression. Probit

analysis of tilapia survivorship indicated requirement of 0.5 and 12.7



















=3 4-)
cro
Es-





U 0)



4-
0r



ne

co





o 4

C3 4-)

4 C-
4-' C






0 -5 1o
10
OC






> m



U) 4-'
S- <




c a -E





a)L
3 u ai


0n 3
s-

04- 4-1
C
C-0 C)






Sc0 LU
WV




30u
C -i

4-1) Wi





o a)
N -

4- Co

I-
4 U)


4-100
UJr 3o


I- c

II




cc' cc c c



~1 1 C 4" COO '*=3 -* 00 *3"*- 0 0 0
cc cci cc! ccf ccU3n

-ii
cc cc
c

0
0
I O


> 0 00 00 00 00 00
{ '* c~o c~c oo ^rIr! -
o o 0 Ll -i i-i


2-O c: -1 1 0c



c; c c cc '

C cC rc -n








0 0 0' 0





cc cc c cc m c c










-cLn cc) cc i- c c



-~ 00 ccr '3- OC 00


Cc cc: c c 'nc'


OL~O
L1

Oii
O
i'
3 - I
erii,
3"~P






CI
lc~

L
c
u









N



,i



IZ


-- n-c cce cc c
cc Nc cc cc cc
cc cc ccCc cc




-c0 ccY ccm cu cc
cc cc cc ccY cc1




-- -0 00 CC 00






C'c ccO O- 00? CC



Cc cm ccr cc Nccc




cc Cc (ci- ahc cc c






















0
*r
ro




CD a
0 C 1.



ED 0
*s *r
0- a S-






N 0
CL

c4 1 41 -
r- 4-'
SC vI













a 0 0j

= Crd c
r0 0 *-
-'-5-




a iC

--- c


WOW









-0 C >
0C EC


) 0.0 Cd
O(UO












ar- 0 .
Cd C C 0



























(L 10
Su-m


0 .*r-
-04--



SC D0 3
S- *S 0
O4C C-







03 3



to- aQ--
4- 1 r-













W *












































































0 0

(%) IVAIAuns


0

C-
0)
0




q N,

en

C.)
cO


CO
M
U, c

V)
SW

01 CO
I-d





LU ,
fl N

o -:

wc
< a
I en
C~o


aJ C
'u

Uc.
C0


o
d
z


z
0


cc
I--
z

0
z
0
o



-






00
N
N









zooplankters per liter to achieve 10 and 50 % survival, respectively.

The elevations of the arcsin-transformed regressions for the two

species were significantly different (P < 0.0001), inferring enhanced

survivorship of blue tilapia over largemouth bass at equivalent prey

densities.

Small zooplankters comprised the bulk of potential forage supplied

to young largemouth bass. Mean relative abundances of cladocerans and

immature copepods (nauplii and copepodids) in the trials with bass were

55.9 and 22.7 %, respectively, whereas the larger adult copepods

comprised 19.6 % of potential forage. In trials with blue tilapia,

relative abundances of cladocerans, immature copepods, and adult

copepods were 28.9, 19.7, and 50.8 %, respectively. At each

experimental prey density, blue tilapia were thereby provided with a

greater concentration of available zooplankton biomass (Table 8). When

regressed against concentration of zooplankton biomass, the

survivorship function for blue tilapia was significantly (P < 0.0001)

elevated above that of largemouth bass (Fig. 7). The least squares

fitted regression predicted 10 and 50 % survivorship of largemouth bass

at 113.3 and 10496.0 pg of zooplankton (dry weight) per liter,

respectively. Probit analysis predicted 10 and 50 % survivorship at

biomass concentrations of 229.2 and 2705.0 pg per liter, respectively.

The probit relationship appeared to more closely approximate the data;

the least squares regression underestimated survival of bass at high

prey levels (Fig. 7). Zooplankton biomasses required to elicit 10 and

50 % survivorship of blue tilapia were 3.0 and 77.2 pg per liter,

respectively, as predicted by arcsin-transformed regression (Fig. 7).





















)V
cO








000
C) )








0 E

C4- 4-




C4 0 I



0 U





C
_aeo














u cU
0-

0*4- V)











U. 4- 4 )
(C)














-c S
4-L

















-0 U --- Q)
- I-




00 0(U1 f
S- *r- -C
-- a r- L>
In C- 2








U- (U &-
- 0 L 0





L0 *,- -
>.C) C- a




U3 ) C)





>- 4- 3
-a r=a
>4-) 0-U)
LVC*W
0 CJC

U) -a
4u-)C)>
aJ 5- c
C) (C C
00 0)
S.- ~ )U
OC) ) l(
a -'---





































C4

u) ;
. 1



0 0
sa:
i 1 U0
4-
C ^
\ >
\ ~ r 0 *


(%) 0 VAIA0

(%) IVAIAUflS


Cl
C
6

U




Lu
W 02
C


I. *


0


0,



0
'.
o

z
0
I-





o
I-





oO






Z
C0







N
z







go
* -o




N









Similarly, probit analysis predicted 10 and 50 % survivorship of

tilapia at 1.4 and 55.6 pg per liter, respectively.

Mortality schedules for bass and tilapia differed considerably

(Fig. 8). Especially striking was the difference in longevity of the

two species when starved (0 zooplankters per liter). All starved bass

died within 7 days (13 days after fertilization); median survival time

was 6 days. Total mortality of starved tilapia occurred in 16 days (26

days after fertilization); median survival time was 12 days. Both

species suffered some mortality at nearly all prey levels early in the

trials, but the rate of death among bass was considerably greater than

for tilapia during this period. All bass maintained at 10 zooplankters

per liter died within 9 days and the majority of deaths at higher prey

levels occurred within the same period. Appreciable mortality of

tilapia maintained at 10 zooplankters per liter occurred towards the

end of the trials, but at higher prey levels, the few deaths noted

occurred within the first week.

Weights and lengths of survivors were directly related to

experimental prey levels (Table 8). Mean weights of blue tilapia ranged

from 0.0110 g at 10 zooplankters per liter to 0.1006 g at 1000

zooplankters per liter. Mean bass weights ranged from 0.0076 to 0.0529

g at 50 and 1000 zooplankters per liter, respectively. The

relationships between survivor weight and forage level were linear for

both species (Figs. 9 and 10). Largemouth bass data gave poor fits

because of considerable heterogeneity within treatments, but

regressions for both species were highly significant (all P < 0.0001).

At equivalent prey levels, mean weights of surviving tilapia exceeded

those of bass (Figs. 9 and 10).






























Fig. 8. Mortality schedules of laboratory-reared largemouth bass
(upper) and blue tilapia (lower) maintained at six different
zooplankton concentrations (number per liter). DAFC = days
after feeding commenced.















100





80-
S\ LARGEMOUTH BASS




60





40









20 0 -----
--1 0 ,---- ---- ^**** ..- --- ""- -----
S................
2I0

100 ............................


,,. -- ,.A- E _-... ...... ....

80 *

BLUE TILAPIA



60




Prey Level
40
1000 -
Soo - -
100 ------
20 50 -s -

00 +



0 2 4 6 8 10 12 14 16

AGE (DAFC)


__



















S- 4-'



o0 C
u, a)c




S- S 4
A L)




CO
r- C4
0 0 c
no- o .
c 10
4-i i-



















00
C- 0-











-O Di0
C L-'
0 -* C



4 0









a ci a
4- 0 0












U 0 S"
.- N 5-





















0-i

























C;
0) QI
'cr-c





I 0 0

0 ci C '




fo ci .- -'





4-' L
















LL
























o n

o
+ 0





0
0 0









U,-
I-
0 0-


















C-


O
-N
a-
z











\ \
\ \
\ \
\ 8:


0
0


LU 0
< +

I
3 :

I-
w
O .
CD
sd


(B) IHOI3M




















0

M- >(3)


Sc) 0)
0 Cr--


10 0 1>
1 0 )-- 1-1
VI U
4.C r3- a
4-' S- V'
=$ 4-J (0
o C C C-
0O Cr C

C- 0 0



0 C 5-
C 4-1 0)co-
00 -. U S
C 010)
0 r- S -
0 Q. O 0
-- O -U

-- N V> C


C-
**- o c-o
r-0 0
0 o

C O 41 O
-.- n

S- 0
4014) 0
5-- N








S- a (U
0 C
aC '01
O GC4-'






I 0 >
> Ur- 0)
S- >U)0
o 0n C- 01


0 C


o 01 0


4- .

a) 01 (I. 3

C4)
-0 0a














LL








71


















0


c





0 0 0
oo I





( -
- 0
\ A






_w o







---



N





(5) 1H913M









Relative Foraging Ability Trials

Prey consumption rates (in zooplankters consumed per hour) of blue

tilapia increased rapidly with age (Fig. 11). At each age, consumption

rates were positively correlated with food concentration. Consumption

rates of largemouth bass increased slowly with age initially, but

accelerated rapidly after the ninth day of feeding (Fig. 11). At each

combination of age and food concentration, tilapia consumed more

zooplankters per unit time than did bass. Appreciable feeding by

tilapia was measurable at each food concentration and age, but

consumption rates of bass were very low or imperceptible (zero

zooplankters consumed in 3 trials) at 3 and 6 DAFC at the 10

zooplankter per liter food concentration.

The diet of blue tilapia at 3 DAFC consisted primarily of the

snall cladoceran Bosmina (Table 9). Accordingly, small zooplankters

dominated the size distribution of food items (Fig. 12). Mean weight of

forage at 3 DAFC was 1.16 pg. At succeeding ages, tilapia gradually

decreased intake of (and selection for) Bosmina and other cladocerans

and increased consumption of adult copepods (Table 9). Coincidentally,

the relative abundance of large items in the diet increased (Fig. 12).

Mean weights of forage at 6, 9, 12, and 15 DAFC were 3.04, 3.66, 6.06,

and 7.27 pg, respectively. By 18 DAFC, feeding electivity of tilapia

was positive and random for adult calanoid and cyclopoid copepods,

respectively, and negative selection was expressed for cladocerans

(Table 9). Mean forage weight increased to 10.08 pg.

Bosmina and small Daphnia were the major dietary components of

largemouth bass at 3 DAFC (Table 9), and cladocerans continued to be

important forage at later ages. Mean weight of bass forage at 3 DAFC



















-c
4-

0
E -


Wa)
CO) 4
5- 4-






4 10

0a L.

*- C


0o




0-v
' *a-




o

S4- S-

S.- C


-0

-0 -
L Ur








N 44
ova









-Q










03


CQ CE
E--
0-










C 44
'ar


















05-
OW
N a






















S-. E Q-
o a
aE
















>v ro 0)














0 0 0


-J o---z*
0. o0 0

P- 006


S 0 0 0 0






0 0 0 0


0 00


m 0


00 0


L n






0 0 0


000






O3D 03 -





0 CO o






00) .1


w ) 4) N


o 0o


CD 0


S..


nmn






0 O0O


(,4.sja splueldooz) 31iVl NOI-dldJnSNOD A3Ud


""`-


r.


















c -
T- C
CI


C S.- a) 4-
(O 0Q- C:
ro Coar

C E (In 1I1 4'-
0 0 Q1)
4-'0 I) V7 4-
lCe (V- 0

)O S.- 4-' 0 LI)
c X C W
N -r- .rr
w 0 wl



0o *i I--u-
o> 'o -

N.,- +.-

4- C
g- C C C
W a) -Zm .r
U, ro
11 4-> > a-)

*Q Li) CD (U C
LI '- fO
*C Lii () w)
4-' L- I- V)
:3 Q.L "0
CO E-0
Es- 0 ro
07) E1 r -Z-
S- -ro
SC0 f S- >
c- t > fO
MC 0,
0 C0
0 >,C C




4C -' CL
r- 0 4-
(0 *U C





>r -0 a) T
0 31 .fl C C-




4-. r Z3 Ur
*r-0S- C
4- J0 Q. 1I1 N
E E0
u (u0





3 U, >.*C C -
(U C c C




.- i- ( I
r o







CQ aja nii I.
c: 0 CD U, >
-c> -a 0j-i

C0:; Z' 4-) '
Z u
ou 4 a-a C
*i~- S-r 1








0 00
.-' C.Al C LI)





0
U C 0U 0T
0U*r-.*- 4-I-C
a- -' A (0 0

0) Li U r -
ajaj anQJQ








a 4- 41 rU

4- >, I) u Li aj

So a > Ir- o
(U -> > C
Q0*>, *
LI) 5- 4-' 4-' *-C"
I 0 1O T- LI' 0

C .0 4-' 10
C-'~S C













10
1-


I In | S




1 o C L] 5
6^ - 666W666


-,8
d d oo
i" l l ;
'-0 0 18 -


00 0 0 0 a







Gmoo o o









ooo 7 oo





























dodd d d
, I, S ,
B"o oo



98866


66G 6o~o6


ooDoo;oo




















odd ooo















0 0I cI I


d 0 dd 8d

o666W66G






























Fig. 12. Age-specific size frequencies of zooplanktonic forage
consumed by blue tilapia and largemouth bass in laboratory
feeding trials (DAFC = days after feeding commenced).











BLUE TILAPIA LARGEMOUTH BASS
100


60o
40
3 DAFC




60-
L






* 9 DAFC
40
6 DAFC
20


>- 60o
w 401

S59 DAFC










40L I
I Ic 15 DAFC
20


20 18 DAFC

1 5 9 13 17 21 25 29 1 5 9 13 17 21 25 29

DRY WEIGHT (pg)









was 0.92 pg. Intake of larger forage increased at 6 and 9 DAFC as

consumption of adult copepods intensified (Table 9), but then regressed

through 15 DAFC (Fig. 12). Mean weights of bass forage at 6, 9, 12, and

15 DAFC were 1.56, 4.05, 3.39, and 2.88 pg, respectively. Consumption

of adult copepods increased sharply at 18 DAFC (Table 9) and mean

forage weight rose to 5.94 pg coincidentally, but modal forage weight

remained < 2.0 pg (Fig. 12).

Rotifers and ostracods were rarely present in the diets of either

fish as availabilities of both items were limited (Table 9). Immature

copepods were available in appreciable quantities, but were generally

minor components of diets of both bass and tilapia (Table 9). Selection

for immature copepods was negative by tilapia at all ages, but they

comprised a significant portion (14.7 %) of the diet at 9 DAFC. Bass

selected against immature copepods at all ages except 6 DAFC when

electivity was random for this forage.

Determination of forage weights allowed conversion of prey

consumption rates from number to biomass of zooplankton consumed per

hour per fish (Fig. 13). Prey consumption rates were described as power

functions of age (Table 10) based on the regression of natural

logarithm-transformed consumption rates on natural logarithm-

transformed ages (Fig. 14). This fitting procedure gave linear

regressions from which the back-transformed lines in Figure 13 were

derived. The power functions provided excellent fits for the blue

tilapia data but underestimated feeding rates of bass at 18 DAFC. The

coefficients of determination (Table 10) indicated that the regressions

explained a sufficient proportion of total variation in all cases,

however. Analyses of covariance and multiple comparisons tests (Zar


























a) -


.D)0W
S -
-CWC
E 4-


0 a'











4- 4- 0
~0


cr v
-Q U
C II C













(-)



Q) U 0
~o U

~o a)
raL














0- c
a) "0













0)a
~oa)














S-~c
> -i

















E 4- u 4
0~--L


0L~J
.C 0
















:3'a c,

cil~
0 ) F-'
SLa












4-'0 0)
= 4-'
4)


a) 0
>, 0-ci




11- m
Ca)


U)U


5- a).
Ca)E


ELO
r D.



>-, 4-'







C'.

0)

















4) *= 0)

0 4) 0

















C),




I


OC





o o 0
t C)


0 0 0


c oo nnj ,








dc an a




















1o





0 0 o


4) 0 0


oU.
4
CO

















in



















toLti

Q
- U
4


(.-B6ri) 3.1lV NOI.dldPnSNOO A3ld


r














Table 10. Prey consumption rate fish age relationships of blue
tilapia and largemouth bass. Parameter estimates are for
data fitted to the power function C = mtb, where C = prey
consumption rate (pg prey. h-' ), t = age in days after
feeding commences, b = power relating the rate of change
of prey consumption rate to age, and m = regression
constant.




Prey
concentration Parameter estimates
Species (no./liter) m b r2



Blue tilapia 10 0.0346 2.9292 0.984

100 0.5158 2.6156 0.993

1000 3.3173 2.1277 0.990



Largemouth bass 10 0.0032 3.2164 0.846

100 0.1978 2.3910 0.956

1000 1.2818 1.9256 0.926
























4-1
ZC


0 S- C


0)4- c






0 C


s- aa












S-s
.-o v*so









C- 4a

4- 0)1






5- 0
0 .0 -0
OZ E =

C 0-
Ca,



5-- U
a,




0 I
Ql---*-









s5- ,
- 5- E
0W1


CD
LE






0 E-
Ln > *r-




a, -0

-s- Ca




C

cc


0 .-- 0



0- O 4- o-



0--




Ll_








83






*'\ \ 0









m 'u






rM
ED D

















\








*
LU
I-:

( \o
m -







4









\w















ci ., ^\ 'i


(,q-6.rl 8601) 3IVH NOIJddWnSNOO A3Hd









1974) were performed on the log-log linear regressions to test for

differences in the consumption rates at different food concentrations

within species and at equivalent food concentrations between species.

Heterogeneity existed among slopes (b values, Table 10) of the 3

regressions for blue tilapia (P < 0.0001). Multiple comparisons

indicated that slopes at each prey concentration were significantly

different from each other (P = 0.0025, < 0.0001, and < 0.0001 for 10

vs.100, 10 vs.1000, and 100 vs.1000 zooplankters per liter,

respectively).

Heterogeneity existed among slopes of the regressions for

largemouth bass (P = 0.0009), but the slopes of the regressions for 100

and 1000 zooplankters per liter were not significantly different (P =

0.1201). The slope of the 10 zooplankter per liter regression was

significantly different from each of the others (P = 0.0131 for 10

vs.100, P = 0.0002 for 10 vs.1000). Elevations of the regressions for

100 and 1000 zooplankters per liter were significantly different (P <

0.0001).

Interpreted, these results indicate that prey consumption rates

were greater at higher prey concentrations (m increased as prey

concentration increased), but that the rate of prey consumption

generally increased more rapidly with respect to age at low prey

concentrations than at higher prey concentrations (b increased as prey

concentration decreased). The slopes of the regressions for largemouth

bass at 100 and 1000 zooplankters per liter followed this trend (b =

2.3910 and 1.9256, respectively), but sufficient deviations from the

regressions existed to preclude demonstration of a statistically

significant difference.









Slopes of the regressions for the two species at equivalent food

concentrations were not significantly different in all comparisons (P =

0.4223, 0.1162, and 0.1736 for 10, 100, and 1000 zooplankters per

liter, respectively). In all 3 cases, the elevations of the regressions

were different, however (all P < 0.0001). Elevations (m values, Table

10) of regressions for blue tilapia exceeded those of bass functions at

each prey concentration. Therefore, the developmental response (the

rate of change in prey consumption mediated by increasing age) at each

prey concentration was similar for both species. At a given age,

however, tilapia were capable of higher prey consumption rates than

bass at equivalent food concentrations.

The interspecific disparity in age-specific foraging abilities was

attributable largely to differences in size between tilapia and bass.

Wet weights of individual eggs of tilapia and bass used in these

experiments were about 7.4 and 2.8 mg, respectively. At 3 DAFC, mean

wet weight of tilapia was 12.1 mg whereas bass larvae weighed about 3.3

mg each. At subsequent ages, the disparity in sizes expanded as growth

of tilapia exceeded that of bass under ad libitum feeding conditions

(Fig. 15). Cubic regression models (Table 11) provided optimum fits for

regressions of prey consumption rates on weights (Fig. 16). The cubic

term significantly improved accuracy of the regression model for each

data set; inclusion of a quartic term did not (a = 0.05). Prey

consumption rates of bass and tilapia were reasonably comparable at

equivalent weights (Fig. 16), but the regression lines for bass

generally were elevated above those of tilapia, particularly in the 10

and 100 zooplankter per liter trials. Tests for coincident multiple

regressions (Zar 1974) indicated that each pair of sample regression





























Fig. 15. Mean wet weights (g) of blue tilapia (squares) and
largemouth bass (circles) used in laboratory foraging
ability trials in relation to age (days after feeding
commenced). Vertical bars represent +1 SD.









87




.25












.20












.15








w



.10
0-




















.05 BLUE TILAPIA










S I LARGEMOUTH
BBASS

0 3 6 9 12 15 18

AGE (DAFC)












Table 11. Prey consumption rate fish weight relationships of blue
tilapia and largemouth bass <0.06 g. Parameter estimates are
for data fitted to the cubic function C= aW3+bW2+cW+d,
where C = prey consumption rate (gg prey. h-'), W = weight
in g, and a, b, c, and d are regression constants.





Prey
concentration Parameter estimates
concentration
(no./liter) a b c d r2


Blue tilapia

10 1716462.8 -151668.6 4807.0 -38.1 0.997

100 9130631.3 -852207.3 29452.3 -238.4 0.974

1000 25503611.1 -2600564.5 90808.8 -732.5 0.996



Largemouth bass

10 1688994.6 -94615.3 2115.5 -5.9 0.957

100 6822369.1 -347586.9 9700.4 -24.3 0.997

1000 20323581.7 -1255617.4 28927.3 -70.4 0.989





















S-




.0 Q) 4-
41 4d

-C O




0a) F_
d- 4-
QJ *- 4-
41 *C.
r---C In
C SO








*0-^-0
-C 0
Cd 4-1 0 In
c or

















5-'
)a 4w I-a




*0
-4-CO 4-

01 r- C-1
3 0 .0 C



4- 0C 4-

S*r
r 0 C "
5-r- r0

01--
.C 01
E S- a)


1--.-Ia0
C Cn
S- 4 a -c

S 4-- 0

.CM 0 +- >
0-*r- C C
* r-- I- (U



>L Q.( O
C- -0

S- 0
u 1- 1-






I- ID 1-

> OW 0



0 r- I Cd
+ >> 0 Ce

4-' >, In Cd
O. (U 3-- *
E 1- C d In
3 Qar- a r-
n 0 0

000 Cd
- 0 c.-




LO~ ,





















V)
U)




1 3
0d


a U
c4
< -


i
(6


0I











0b


0 0 0 0
0 0 0 0
on t o


(l -46Br) 31VI: NOIldw~nSNOD A3Hd


0

~~0-~










functions did not estimate the same population regression (all P <

0.001). Plateaus in the developmental response coincided approximately

with metamorphosis to the juvenile period for both species.



Discussion

My results show that at equivalent ages, young blue tilapia were

more proficient than largemouth bass at exploiting zooplanktonic food

and thereby were conferred enhanced survival and growth in the

laboratory during the critical early life history period. Extrapolation

of these findings to natural waters may explain, at least in part, the

success of the species in colonizing new habitats and rapidly achieving

high abundances in Florida. Essentially, tilapia are able to survive

under conditions of food availability detrimental to survival of young

bass.

The advantages of enhanced foraging proficiency would be largely,

if not altogether, inconsequential in influencing survivorship in the

presence of abundant forage. However, zooplankton concentrations in

most Florida lakes (Fig. 17) encompass a range wherein survivorship of

blue tilapia would be expected to exceed considerably that of

largemouth bass and other native species with similar food abundance

requirements.

Even in lakes with high zooplankton concentrations tilapia may

still have the advantage. Zooplankton assemblages in these eutrophic

systems are typically dominated by rotifers (Reid and Squibb 1971;

Cowell et al. 1975; Blancher 1979), an infrequent and unpreferred

forage of centrarchid larvae (Chew 1974; Lemly and Dimmick 1982).

Abundances of microcrustacea in these lakes are often similar to those


I































Fig. 17. Frequency distribution of mean annual zooplankton concen-
trations (number per liter) for 165 lakes in Florida. Data
courtesy of D. Canfield (by permission).























40









30
vn
N = 165






20
._










0
I-

















0



0 500 1000 1500


MEAN ZOOPLANKTON CONCENTRATION (no./liter)




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs