Title: Feasibility of using Toxorhynchites rutilus rutilus (Coq.) in the control of container-breeding mosquitoes
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00099258/00001
 Material Information
Title: Feasibility of using Toxorhynchites rutilus rutilus (Coq.) in the control of container-breeding mosquitoes
Physical Description: x, 121 leaves : ill. ; 28 cm.
Language: English
Creator: Focks, Dana A
Copyright Date: 1977
 Subjects
Subject: Mosquitoes -- Biological control   ( lcsh )
Entomology and Nematology thesis Ph. D
Dissertations, Academic -- Entomology and Nematology -- UF
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Statement of Responsibility: by Dana A. Focks.
Thesis: Thesis--University of Florida.
Bibliography: Bibliography: leaves 118-120.
General Note: Typescript.
General Note: Vita.
 Record Information
Bibliographic ID: UF00099258
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 - 000010019
oclc - 02896459
notis - AAB2131

Downloads

This item has the following downloads:

feasibilityofusi00fock ( PDF )


Full Text










FElASIBILITY OF USING Toxorhynchites rutilus rutilus (COQ.)
IN THE CONTROL OF- CONTAINER-nRlEEDING MOSQUITOES







By

DANA A. FOCKS


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












UNIVERSITY OF FLORIDA


1977















ACKNOILEDGEMENTS


The author wishes to acknowledge his sincere appre-

ciation to Dr. J.A. Seawright, his research advisor, for

his patience, assistance, support, and critical review ot

the dissertation. Special thanks are extended to Dr. D.W.

Hall, his Committee Chairman, for advice and encouragement

during the course of this study. To all the members of

his Ph.D. Committee, the author expresses appreciation for

their critical appraisal of the dissertation and their

helpful comments.

The author wishes to express his gratitude to the staff

of the U.S. Department of Agriculture Insects Affecting Man

Laboratory, Gainesville -- in particular, Drs. D.E. Weidhaas

and D. Haile, and C.E. Schreck, J.H. Gackson and M.Q.

Benedict.

The author thanks Dr. R. Darsie, Center for Disease

Control, USPHS, Atlanta, Georgia,for assistance on adult

identification and J.S. llaeger, Vero Beach Laboratories,

Inc., Vero Beach, Florida, for assistance with the coloni-

zation.

For his cooperation during the field trials, thanks are

given to B.C. Lemont of the University of Florida Pest

Control Auxillary.









Finally, deep gratitude is extended to his wife Debby,

for her patience, encouragement, and aid during both the

course of study and the preparation of this manuscript.















TABLE OF CONTENTS


Page

ACKNOWLEDGE ENTS . . . . . . . . . ii

LIST OF TABLES. . . . . . . .. . . v

LIST OF FIGURES . . . . .... ... . . vi

ABSTRACT . ... . . . . . . . . . . viii

INTRODUCTION. ...... . . . . .. . 1

LABORATORY COLONIZATION OF Toxorhynchites rutilus
rutilus (COQ.) . . . . ... .. ... 2

LABORATORY REARING OF Toxorhynchites rutilus rutilus
(COQ.) ON A NON-LIVING DIET .. . . . . . . 13

FIELD SURVIVAL, MIGRATION AND OVIPOSITIONAL CHARACTER-
ISTICS OF LABORATORY-REARED Toxyorhynchites rutilus
rutilus (COQ.). ..... . . . . .. 21

A DETERMINISTIC MODEL FOR SIMULATING THE PREDATION
OF Toxorhynchites rutilus rutilus (COQ.) ON
Aedes aegypti (L.) . . . .. . . . . 42

APPENDIX .. . . . . . . . . . 101

LITERATURE CITED. . . ... .... . . ... 118

BIOGRAPHICAL SKETCH . . . . . . . 121















LIST OF TABLES


Table Page

1. Duration of eggs, larval and pupal stages of
Tx. r. rutilus and total numbers of Ae. aeggyti
larvae eaten. . . .. . .. ... 6

2. Length of larval life of Tx. r. rutilus when
deprived of food . . . . . .. . . 10

3. Duration of larval instars of individually
reared Tx. r. rutilus at 28 1 1C when fed a
diet of Ae. aegypti larvae or TetraMin ... 18

4. Tx. r. rutilus oviposition observed after the
release of 350 adults into the central area of
the experimental area . . . . . . 30

5. The number and distribution of eggs recovered
within the experimental area. . . . . 31

6. Ae. aegypti and Tx. r. rutilus survival and
oviposition distribution parameters .. .. 45

7. Ae. aegypti immature development times and
average numbers (and proportions) per water
storage jar positive for Ae. aegypti
(Southwood et al., 1972) . . . .... 49

8. Larval development times for Tx. r. rutilus
in the laboratory when fed early (I i II) or
late (III, IV f, pupa) instar Ae. aegypti. .. 54

9. Comparison of the number of Ae. aegypti in-
dividuals per stage per container as reported
by Southwood et al., with model output. .. 67

10. Effects of contact adulticide application
causing 95% mortality with no residual action
on Ae. aegypti. .. ...... ....... .. 69















LIST OF FIGURES


Figure Page

1. Aerial view of the experimental area showing
the student housing project and the sur-
rounding woods . . .. . . .. 26

2. Map of the experimental area showing the 3
release points, ovitrap locations, and the
distribution and number of eggs recorded sub-
sequent to the release of 350 Tx. r. rutilus
adults. ............. ....... .. 29

3. A regression of the daily percent of ovi-
position occurring within the housing project
area on days after release. . . . . . 34

4. A regression of daily egg production on days
after release . . ......... ... 39

5. Ae. aegypti subroutine. . .. ... ... . . 48

6. Tx. r. rutilus subroutine . . ... .. . 52

7. Distribution of container types . ... .. . 57

8. Ae. aegypti adult density on a per container ba-
sis with no control measures applied. (No rain). 60

9. Ac. aegypti adult density on a per container
basis with no control measures applied. (Daily
rainfall into every container). . ... .. . 62

10. Total number of Ae. aegypti immatures/con-
tainer. (No rainT .. . . . . . 64

11. Total numbers of Ae. aegypti immatures/con-
tainer. (D)aily rainfall into every container) 66

12a. Effects on Ae. aegypti adults of a contact
adulticide application at day 90 causing 95%
adult mortality with no residual effects
(No rain) . . ... . .. . 71

12b. Effects on Ae. aegypti immatures when adulti-
cide applied as in Fig. 12a . . . . . 73









liST OF FIGURES
(Continued)

Fi gure l';age

13a. Effects on Ae. aegypti adults of a contact
adulticide application applied when Aedes
adult density exceeded 7 adults/container. . 75

13b. Effects on Ac. aegypti immatures when
adulticide applied as in Fig. 13a. . . .. 77

14a. Effects on Ac. aegypti adults of a contact
adulticide application applied when Aedes
adult density exceeded 4 adults/container. . 79

14b. Effect on Ae. aegypti immatures when adulti-
cide applied as in Fig. 14a. . . .. . 81

15a. Ac. aegypti adult density following an adult
Tx. r. rutilus release on day 98 producing
1 predator egg in 80% of the containers posi-
tive for prey. (No rain) . . . . . . 83

15b. The effect of 1 predatory larva/container in
80% of the containers positive for Ac. aegypti 86

16. The effect on Ae. aegypti adult density from
an adult predator release on day 98 resulting
in 3 predator larvae/container in 80% of the
containers positive for Ae. aegypti. (Daily
rain). . . . . . . . . . . 88

17a. The effect on Ae. aegypti adults of a predator
release on day 107 resulting in 1 predator
egg/container in 60% of the containers posi-
tive for Ac. aegypti. (No rain). . . . . 91

17b. The effect on Ae. aegypti adults of a predator
release on day 107 resulting in 1 predator
egg/container in 80% of the containers posi-
tive for Ac. aegypti immatures. (No rain). . 93

17c. The effect on Ae. aegypti adults of a predator
release on day 107 resulting in 1 predator
egg/container in 90% of the containers posi-
tive for Ae. aegypti immatures. (No rain). . 95

18. Adult Ae. aegypti densities resulting from an
adulticide application at day 90 followed by a
predator release on day 98. (No rain). ... .97

19. Adult Ac. aegypti densities resulting from a
predator release and 5 subsequent adulticide
applications on days 98, 101, 109, 113 and 117 99













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


FEASIBILITY OF USING Toxorhynchites rutilus rutilus (COQ.)
IN THE CONTROL OF CONTAINER-BREEDING MOSQUITOES

By

DANA A. FOCKS

March 1977

Chairman: D.W. Hall
Major Department: Entomology and Nematology

Toxorhynchites rutilus rutilus (Coq.) was successfully

colonized and studied in the laboratory to determine the

potential usefulness of this predatory species of mosquito

as a biological control agent for container-breeding mos-

quitoes. When Tx. r. rutilus larvae were reared at 28 + 10C

in individual containers with a surplus of larvae of Aedes

aegypti (L.) as prey, the duration of the immature stages

averaged 1.6, 15.6, and 6.0 days for eggs, larvae, and pupae,

respectively. Contrarily, with mass rearing conditions, the

duration of the larval stage was significantly reduced to

11.1 days and pupation was more uniform than in individual

containers. Adult females survive for 7 wk in laboratory

cages and oviposit an average of 1 egg/day. Fourth-instar

larvae of Tx. r. rutilus can survive for about 2 months

without food. Adult females preferred to lay their eggs

in water previously used to rear Ae. aegypti.


viii










For the first time, Tx. r. rutilus was reared on a

non-living diet of Tetraflin-Staple Food a commercially

available food for tropical fish. Larvae reared on TetraMin

required an average development time of 107.5 days from egg

to pupa compared to 15.6 days for larvae that were fed a

diet of larvae of Ae. aegypti. The daily survival rate of

larvae reared on the non-living diet was 0.9901 + 0.0098,

a value slightly less than the survival rate of 0.9973 t

0.0075 for larvae fed Ae. aegypti. Pupae reared on the

non-living diet were small (29.3 + 3.6 mg) compared to a

control group (50.0 + 2.0 mg), but surprisingly, adult

longevity and fecundity of the small adults did not differ

significantly from the control. The significance of the

above findings are discussed relative to the use of Tx. r.

rutilus as a biological control agent for container-breeding

mosque toes.

Three hundred and fifty, 6-day old laboratory-reared

Tx. r. rutilus adults were released into a sparsely wooded

13-acre residential area in Gainesville, Florida. Oviposi-

tion was monitored for 14 days using a grid of 64 oviposi-

tion traps located within the residential area and surround-

ing woods. Eighty percent of the ovitraps received eggs,

and despite the migration of the females into the surround-

ing woods at a rate of about 7% per day, 64% of the eggs

recovered were laid in the residential area. Equations were

derived that allowed estimating the daily adult survival

(Sa) and lifetime egg production per female released (F)










to be 0.79 and 3.99 eggs/female, respectively.

A deterministic computer model is presented detailing

the interaction of the container-breeding mosquito Ae.

aegyjti and the larval predator Tx. r. rutilus. Results of

simulation runs involving the release of Tx. r. rutilus

adults indicate that predator releases resulting in 1

predator larva per container are sufficient to reduce Aedes

adult density 75% in 20 days. The slow rate of immature

predator development enables control to be maintained for

several months. Simulations of predator release and the

use of adulticides indicate that it is possible to obtain

zero adult densities. Finally, the model indicates that

the most important parameter determining the degree of

control established is the distribution of predator eggs.














I NTRODUCT ION


Increased impetus to reconsider mosquito control

methodologies involving biological control agents is being

provided by the phenomenon of pesticide resistance.

Toxorhynchites (Theob.) is a genus of large, brilliantly

colored, non-biting mosquitoes that are predacious during

the larval stage on certain mosquitoes which breed in dis-

carded cans, bottles, tires, water cisterns and tree holes.

The research reported herein was undertaken to evaluate the

feasibility of using Toxorhynchites rutilus rutilus (Coq.)

as a biological control agent against the yellow fever and

dengue hemorraghic fever vector Aedes aegypti (L.). Each

of the 4 subsequent chapters are papers currently being

submitted to journals; they are complete in themselves.

Each introduces itself, reviews the relevant literature,

presents results of the work done and reports the signi-

ficance of these findings relative to the use of Tx. r.

rutilus in mosquito control.















LABORATORY COI.ONIZATION OF
T'oxorhlnchi ts rutilus rutilus (COQ.)



Abstract


Toxorhynchites rutilus rutilus (Coq.) was successfully

colonized and studied in the laboratory to determine the

potential usefulness of this predatory species of mosquito

as a biological control agent for container-breeding mos-

quitoes. When Tx. r. rutilus larvae were reared at 28 1 10C

in individual containers ivlh a surplus of larvae of Aedes

aegypti (L.) as prey, the duration of the immature stages

averaged 1.6, 15.6, and 6.0 days for eggs, larvae, and pupae,

respectively. Contrarily, with mass rearing conditions, the

duration of the larval stage was significantly reduced to

11.1 days and pupation was more uniform than in individual

containers. Adult females survive for 7 wk in laboratory

cages and oviposit an average of 1 egg/day. Fourth-instar

larvae of Tx. r. rutilus can survive for about 2 months

without food. Adult females preferred to lay their eggs in

water previously used to rear Ae. aegypti.



Introduction


Extensive use of synthetic pesticides has resulted in

the phenomenon of resistance to them. The ramifications of









this problem are providing increased impetus to reconsider

mosquito control strategies involving biological control

agents. Several of the more important mosquito vectors of

human disease breed in discarded cans, bottles, tires, water

cisterns and tree holes. Control of larvae of these species

is difficult because the larval habitats are small, dis-

persed and often inaccessible. Mosquitoes of the genus

Toxorhynchites are larval predators of container-breeding

mosquitoes, and possibly could be highly effective control

agents because the female Toxorhynchites might more effi-

ciently find these breeding sites than the mosquito-control

worker (Brown 1973). However, the literature is replete

with examples where Toxorhynchites spp. failed to control

prey species (Newkirk 1947; Paine 1934; Swezey 1930 and 1931;

Williams 1931). This has been attributed to intrinsic

factors such as long life cycles, low fecundity, and sur-

vival rate (Nakagawa 1963). Gerberg (1974) and Muspratt

(1951) contend these shortcomings can be overcome with

inundative releases of Toxorhynchites, a situation that

upsets the normal predator-prey relationship.

Currently we are conducting research on the feasibility

of using Toxorhynchites rutilus rutilus (Coq.) as a bio-

logical control agent against container breeding mosquitoes.

The present paper summarizes our progress in laboratory

studies and describes the colonization, mass rearing, and

other aspects of the biology of this species.

The genus, Toxorhynchites, in North America north of









Mexico is represented by two subspecies (Jenkins 1949) and

perhaps a third, the status of wlihich is uncertain (Zavortink

196l9). Tx. ruttili s septentrional is (Coq.) is known to

occur in the Eastern United States north to New Jersey and

Pennsylvania and west to the great plains of Kansas,

Oklahoma, and Texas. Tx. r. rutilus is known only from

the extreme Southeastern United States in peninsular

Florida, southern and coastal Georgia and coastal South

Carolina north to Myrtle Beach (Carpenter and LaCasse 1955).

Intergrades occur in the zone of overlap of the ranges of

these two subspecies (Jenkins 1949). The egg, larva, pupa,

adult stages (Carpenter and LaCasse 1955; Dodge 1964),

oviposition habits (Olinger 1957) and habitats (Seabrook and

Duffey,1946; Basham et al. 1947) of Tx. r. rutilus have

been described.



Colonization


We collected Tx. r. rutilus eggs from cavities in

various trees in Alachua County, Florida, during September

and October 1975. Larvae were reared to the pupal stage

individually in 8-dr glass vials. Each egg was individually

set with approximately 250 first stage larvae or eggs of

Aedes aegypti (L.); additional prey were added as required.

The prey larvae fed on 5-10 mg of TetraMin R(a commercially

available tropical fish food) supplied every other day.

Subterranean well water was used in all rearing. The

photoperiod was 14 hr of subdued light :10 hr dark. Reared









at 28 + 1"C, the development time from egg to pupation for

156 individuals was 17.2 + 3.3 days and total mortality was

s8' (Table 1).

The first cohort of 150 pupae reared in the above

fashion was placed in a 1 m x 1 m x 2 m high screened

aluminum cage covered with a clear plastic film to maintain

humidity. Water on sponge wicks, honey, apple slices and

black, 0.5-liter glass jars half filled with water for

oviposition sites were provided. The cage was within an

environmentally controlled room lacking windows. Light was

provided by 8 40-watt fluorescent tubes. The conditions

were: photoperiod 14 hr light : 10 hr dark, temperature

24 + 50C, relative humidity (RH) 85 + 10%. No attempt was

made to simulate twilight. No mating was observed and no

eggs were produced. No mortality was observed for the first

6 wk, but 75% died within the next 2 wk and no individual

lived longer than 10 wk.

A second cohort replaced the first under similar

temperature and humidity conditions and a number of things

were tried in an attempt to induce mating. J.S. Haeger (personal

communication) has had success in colonizing difficult

species by using either singly or in combination the fol-

lowing: (1) plants within the cage to act as swarm markers;

(2) twilight simulation using incandescent lights on a

rheostat; (3) the addition of a second species (Ae. aegypti);

and (4) casting shadows across the interior of the cage.

Using the above techniques llaegerhas been able to stimulate






























iLn o '0 c 0 o
'-I -^ft C~- i cc G i

S0 0 0 --

+1 +1 +1 +1 +1 +1




-)


r.




*H




O-s






































M 0
4,-

































S0
r-;






















0
Cl



O
4-












H.





o


0n













[-1


H




o



-e
0
E





*-I
44














O
0
S

















0




Q
0
0
















'o
/)









+1
0













0
a



0.

















'0
/
a
cc




ci/

+1

\x



0
z






2;






0
Car

LC]
44
(I


01 co o



+1 +1 +1 +i



0 c0 Co


C

t 44 '0 '4 c
co~d~ iur ~ b 0


4-1




t
-Z

-c



n


-e
0


h



















*O
Cl




0
0.





















44
H
C]




















to
',


C
nri












t4;
HO
a
e
N

0

r









o-

44


0'0
'05r
03C
-C



0 *-
in '0

I-so
o **



01-s
r 3M


-a 0
ko r








VI-
Vi-






cr40


rC]

rd









eating and oviposition in Tx. r. ruti lus. However, we tried

his techniques with no success. Varying the numbers of

Tx. r. rutilus within the cage (n = 25, 50, 100, or 150)

wlas also unsuccessful.

A third cohort of about 30 adults was placed in a

0.5 m x 0.5 m x 0.5 m plexiglass (acrylic plastic) cage.

The conditions were as for the first cohort. Mating was

observed on the second day after emergence, and oviposition

was observed 6 days postemergence. Females fly in a verti-

cal circle several inches above the oviposition jar and

eject the eggs singly onto the surface of the water (Olinger

1957). Sixteen females observed for a period of 3 wk after

oviposition began to produce an average of 1.23 eggs/female/

day. All the eggs were fertile but no ovarian cycle was

immediately apparent. Using the techniques described above,

Tx. r. rutilus has been maintained in the laboratory for

9 generations.

Rearing in outdoor cages was attempted during July and

August 1976 in an attempt to increase egg production. The

same I m x 1 m x 2 m high cage described previously but

without the plastic covering was used. This cage was

located within a longer 5.5 m x 7.3 m x 4.3 m screened

building. The enclosures were situated under large oak

trees which provided shade during the middle of the day.

An average daily temperature of 29C and RH of 85% were

recorded inside the smaller cage. Various numbers of Tx. r.

rutilus (n = 18, 30, 150, 250, or 450) were placed in the









inner cage with water wicks, honey, apple slices and ovi-

position jars (half filled with well water). At each

density of adults, first matings and oviposition were

observed at 2 and 6 days, respectively.

During the outdoor cage work, 15 females were removed

after 12 days of being with 15 males and confined in indi-

vidual containers supplied with honey, water wicks and

oviposition jars. The containers, held at 280C, RI of

85 + 10% and a photoperiod of 14 hr light : 10 hr dark, were

checked daily for oviposition over a period of 25 days.

Thirteen (or 87%) of the females laid eggs. Fecundity was

0.83 t 0.66 eggs/female/day, a value not significantly dif-

ferent (p = 0.05) from 1.23 eggs/female/day reported above

for the indoor cage. The oviposition cycle can best be

described as highly irregular. Three of the females laid

1-3 eggs daily, and 5 of them oviposited 5-10 eggs on

successive days with 5-15 day intervals of no oviposition.

A third group of 5 females was intermediate in behavior.

As evidenced by the standard deviation, differences in total

egg production between females were great. The egg fertil-

ity was virtually 100% suggesting that females do not

oviposit unless they have mated.

Before colonization procedures were established for

Tx. r. rutilus, induced mating was used to maintain labora-

tory stocks. Techniques described by Gerberg (1970) and Trim-

ble and Corhet (1975) were successfully employed with the

exception of ethyl ether being used as the anesthetic.









Fertile eggs were laid 2 days after forced mating. Typi-

cally, however, oviposition stopped after only a few days.

T'he amount of time larvae of Tx. r. rutilus can

withstand fasting may be an important parameter in biological

control considerations. To investigate this, variously

aged larvae and eggs were placed in individual vials con-

taining clean well water and observed until death by

starvation occurred. Larvae were fed the usual diet of

Ae. acgypti before the test began. From the results in

Table 2, one can readily see that the ability of Tx. r.

rutilus larvae to survive fasting will indeed be of value

in the use of this predator for control. The first two

larval stages survived for about a week and third stage

larvae lived for 18 days without food. The fourth larval

stage was exceptional in its ability to withstand fasting,

and this noteworthy fact is all the more important because

this stage also is capable of eating more prey than the

first 3 instars.

The existence of an oviposition stimulant, e.g., the

presence or past presence of a prey species could be impor-

tant in biological control by Tx. r. rutilus. A simple

2-choice test between well water and water which had been

used in rearing Ae. aegypti from egg to pupa (hereafter

referred to as colony water) was conducted to determine

whether Tx. r. rutilus preferred one or the other as an

oviposition media. Four pairs of 0.5-liter black oviposi-

tion jars were placed in the 1 m x 1 m x 2 m high cage under
























r-




Cr-
'3 -u
'H 0





0
o
0
z









o
C)

0







CC











c









H-'0
C)
N.0




c c




o 0



cc




CC>C
> -1











-C

u
M ra


cy 0 0 CO







-4








t ) L n.

o O '0 (-i

+1 +1 +i +i



0 c C aD C:)
0 0 00 CT
-1 Ln









































oj 4 r t


0

t1

0
C1

Q3
0




VI
H

-D






4







ao
V)





0






czo0
F-:







HO

C5
Ul ^

n-4
H C









9 a')
o to a

S 0 H












0o 0 u
H H '
CC C '
F- -






- I 1-


the previously described outdoor conditions. Each pair

consisted of 1 jur half full of well water and 1 jar half

full of colony water. During the test, the colony water

did not contain any Ae. acgp. Each pair was placed in

a corner of the cage floor. Daily, the positions of the

jars within each pair were alternated to nullify any posi-

tional effects. At 3-day intervals, the water in each of

the 8 jars was replaced with new well or colony water as

appropriate. The cage contained approximately 750 adults

(ca. 375 females). The number of eggs in each type of

water was recorded daily for 6 days. The mean daily ovi-

position in the jars containing well water and colony water

was 74 + 47 eggs/day and 247 87 eggs/ day, respectively.

The difference between the 2 means is highly significant

(t-test).

The preference of Tx. r. rutilus females to oviposit

in the polluted water can be quite important in regard to

using this predator for control. Although the need for more

work is indicated, it appears that the females prefer to

oviposit in water similar to that found in natural settings

(treeholes, discarded trees, etc.) rather than in containers

(water bottles, rain barrels, and cisterns) commonly

employed to hold water for household uses.



Mass Rearing


Since the rearing of Tx. r. rutilus in individual vials

is a laborious, impractical process, experiments were









conducted to determine the feasibility of mass production.

Cannibalism is the principle problem encountered in mass

rearing Tx. r. rutilus in the same container; therefore,

all of these trials were conducted in complete darkness.

Fourteen plastic trays (38 cm x 51 cm x 10 cm) were filled

with well water to a depth of 4 cm, and the water tempera-

ture was maintained at 28 + 0.30C. To these trays were

added 110 Tx. r. rutilus eggs (less than 24 hr old), ca.

10,000 Ae. aegypti eggs and 1.7 g TetraMin. The same day

the Tx. r. rutilus eggs are set, a second identical tray is

set with ca. 10,000 Ac. aegypti eggs and 1.7 g TetraMin.

Each tray receives an additional 1.7 g Tetrallin on alternate

days. The second tray containing Ae. aegypti is added to

the first tray containing Ae. aegypti and Tx. r. rutilus

8 days after the trays were set. Overcrowding and under-

feeding the Ae. aegypti result in a mixture of third and

fourth instars being added to the Tx. r. rutilus. Seventy

+ 6% of the Tx. r. rutilus survived to the pupal stage with

cannibalism being the most likely cause of mortality. The

initial larval density of 18 cm2/larva changed to 25 cm2/

larva at pupation. Reared in this manner, the average

development time from egg to pupation was 12.66 + 1.22 days

for 547 larvae. The difference between length of develop-

ment (egg to pupa) when 156 larvae were reared individually

(17.99 + 3.31 days) and when reared in mass is highly

significant (t-test), but the reason for this observation

is not readily apparent.















LABORATORY REARING OF Toxorhynchites ruti lus
rutilus (COQ.) ON A NON-LIVING DIET



Abstract


For the first time, Toxorhynchites rutilus rutilus

(Coq.), a predatory species of mosquito, was reared on a
R
non-living diet of TetraMin-Staple Food a commercially

available food for tropical fish. Larvae reared on TetraMin

required an average development time of 107.5 days from egg

to pupa compared to 15.6 days fur l irvae that were fed a

diet of larvae of Aedes aegypti (L.). The daily survival

rate of larvae reared on the non-living diet was 0.9901 +

0.0098, a value slightly less than the survival rate of

0.9973 + 0.0075 for larvae fed Ae. aegypti. Pupae reared

on the non-living diet were small (29.3 + 3.6 mg) compared

to a control group (50.0 + 2.0 mg), but surprisingly, adult

longevity and fecundity of the small adults did not differ

significantly from the control. The significance of the

above findings are discussed relative to the use of Tx. r.

rutilus as a biological control agent for container-breeding

mosquitoes.


Introduction


Toxorhynchites (Theob.) is a genus of large, non-biting


-13-









mosquitoes that are predacious during the larval stages on

certain mosquitoes which breed in discarded cans, bottles,

tires, water cisterns and trceholes. Brown (1973) con-

siders the genus to be sufficiently promising as a biological

control agent of artificial-container breeding mosquitoes

to warrant continued research toward this end. In this

connection, inundative release of adult Toxorhynchites has

been proposed as necessary to upset the normal predator-prey

relationship, thus effecting control (Gerberg, 1974;

Muspratt 1951). Tnundative releases imply the mass rearing

of large numbers of mosquitoes and in determining the

practical utility of Toxorhynchites as a biological control

agent, the cost of mass rearing could be as important a

parameter as the biological aspects of the mosquito. In a

separate report, Focks and Seawright (1977) wrote that one

Tr. r. rutilus larva requires ca. 100 Ae. aegypti as food;

thus, a rather large colony of Ae. aegypti would be required

for the mass production of Tx. r. rutilus. In view of the

expense involved in maintaining a large colony of mosquitoes

as a prey species, the author is currently involved in a

study directed toward the development of alternative food

sources for Tx. r. rutilus. In the present paper the

author presents the results of investigations into the

feasibility of using a non-living diet. The fecundity,

daily adult survival, development time of immature stages

and pupal weights are reported for Tx. r. rutilus reared

solely on TetraMlin and compared with observations on









larvae fed solely on larvae of Acdes aegypti (L.)



Na trials and Methods


'TetraMin Staple Food' (manufactured in West Germany)

is a dried, flaky material consisting of fish and shrimp

meal, oat flour, fish liver, squid, fish roe, kelp, mosquito

larvae, brine shrimp, aquatic plants, agar agar, chlorophyll

and carotene. By weight it is 45% crude protein, 5% fat,

6% fiber and 44% water. It was selected for these

experiments for the following reasons: (1) Considering the

diversity of substances making up TetraMin, it was deemed

reasonable that TetraMin contained the required nutrients

for a predator; therefore, if Tx. r. rutilus failed to

develop on TetraMin, the feasibility of rearing this preda-

tor on a non-living diet would be dubious. (2) TetraMin

has been used as the sole food source in rearing other

mosquitoes, e.g., Culex spp., Aedes spp. and Anopheles

spp. (Pappas, 1973). (3) TetraMin does not promote as much

scum formation in water as is found with other materials

used as food for mosquitoes. (4) Since TetraMin is a

commercially available, fairly inexpensive material, the

cost of using it for rearing Tx. r. rutilus would not be

prohibitive.

The Tx. r. rutilus used in these experiments were taken

from the fifth and sixth generations of a colony maintained


Distributed in the U.S.A. by Tetra Sales of Hayward, CA.









in tile laboratory. The original material wus collected as

eggs from treeholes in Alachua County, Florida.

Tx. r. rutilus eggs were individually set in 8 dram

glass vials which were held in a water hath at 28 + iC and

a photoperiod of 14 hr light : 10 hr darkness. One group of

83 larvae was fed 5-10 mig of pulverized TetraMin at 3-day

intervals, and the control group of 48 larvae was fed larvae

of Ae. aegypti as required. At intervals of 10 days, the

Tx. r. rutilus larvae were transferred to clean water.

Notes were kept on the development time required by

each larval instar and on the comparative weights of pupae

for the mosquitoes reared on the living and non-living

diets.

Adult fecundity (eggs/female/day), longevity and daily

survival were recorded in an outdoor cage. The cage was a

1 m x 1 m x 2 m screened cage located within a larger

5.5 m x 7.3 m x 4.3 m screened building. These enclosures

were situated under large oak trees which provided shade

during the middle of the day. The inner cage was protected

from rainfall. The cage contained water wicks, honey,

apple slices and black 0.5-liter oviposition jars half-

filled with water. Eggs were removed and counted on a

daily routine to provide information on fecundity and

percent hatch of the two groups of adults.









Results


The development time for Tx. r. rutilus larvae reared

on letraMin and Ae. aegy2pti are shown in Table 3. Larvae

maintained on a diet of TotrMain required an average of

107.5 + 19.8 days compared to a relatively short 15.6 + 1.4

days for larvae reared on Ae. aegypti. Even this compari-

son is not complete, for after 160 days when the test was

terminated, 12% (or 10) of the 83 larvae reared on TetraMin

were in the 4th larval instar. These 10 remaining larvae

were offered (on day 190) Ae. aegypti as prey and pupated

within 10.4 + 3.0 days after consuming 87 + 34 1st and 2nd

stage larvae. During the 160 days, only 24% of the larvae

receiving TetraMin pupated, and 64% of the larvae died. In

comparison, 95.8% of the larvae fed Ae. aegypti pupated.

A daily survival rate (S.) for the larval stages was cal-

culated for the two groups of larvae by averaging the ratios

obtained by dividing the number of larvae alive on a given

day by the number of larvae alive on the previous day. The

S. values were 0.9901 + 0.0098 and 0.9973 + 0.0075 for
1
larvae reared on TetraMin and Ac. aegypti, respectively.

A comparison of the average weight of the two groups

revealed that pupae from the larvae reared on TetraMin were

40% lighter. Weights of the pupae were 29.3 + 3.6 mg

(n = 21) and 50.0 + 2.0 mg (n = 8).for TetraMin and Ae.

aegypti, respectively.

Daily survival (Sa) of the adults was calculated by

the same method used in calculating S.. Using this method,
1'






















C)










[ 3
U.
3















*I-




O *
SC
*r-
=


























-C)

OO
Mns
i-rC
* 4-i

t-li


C10

g I




riD
00)
>N C






H M.



C

44 0
^-1


00
I4i
ino


do,
4-SC)
ca
If) *H

C)T




1-

> 0
n3







00
4-i
>o3




= U
0 0






C)
3co




ic)



m
s-


(I 0 00 M 0)


















o 000
-I-- C 7 \

























*0
+r




























- -0 -1 > +0
co
r,















+-1




0

SU))
i-i '- 0 '
i- fI ^ >* c3
i 1 1 i *4 1
0 I


0
,c



























0









E0
4-1

U -




4-i
C,

C)




























0 -


or
UC)




. c
iC)










co
0
. 0i






















UC
0













c e)






)o-










4r--





C n






-19-


estimates of S were U.988 + 0.051 and 0.979 + 0.020 for

adults from larvae reared on TetraMin and Ac. aegypti,

respectively. There was no significant difference in adult

survival.

For convenience in comparing the fecundity (F) of the

two groups of adult females, fecundity was expressed as

eggs/female/day and was calculated by dividing the total

egg production by the number of female days. Using this

method, the fecundity averaged 0.98 + 0.23 and 1.03 + 0.28

eggs/female/day for adults reared on TetraMin and Ae.

aegypti, respectively. Given S = 0.979, F = 1.00 egg/
35
female/day and total egg production = nF a St, a cage
t=0 a
of n = 100 females would be expected to produce ca. 2500

eggs during a 5 week period. Percentage hatch for eggs

from the two groups of females was the same, 98%.



Discussion


The results presented herein document for the first

time that Tx. r. rutilus is not an obligate predator, albeit

the mortality of 64% of the larvae fed the non-living diet

would preclude the use of this technique for mass produc-

tion of this species. In consideration of the biological

parameters measured, the undesirable aspects of using

TetraMin centers on the extremely long development time and

the reduced size and weight of the pupae. Obviously the

larvae were not receiving an adequate diet, but whether









the malnutri t ion was due to the lack of essential nutrients

or ingestion is unknown. In his observations, the author noted

that the Tx. r. rutilis never consumed all the TetraMin

offered to them and this fact indicated that the larvae

simply do not feed vigorously on a non-moving diet. The

other parameters we measured, which included larvae survival

(Si), adult survival (S ) and fecundity (F) were not dif-

ferent from those of the control group. However, in the

context of mass'rearing, the extremely long development

time overshadows these bright points. For example, the

larval mortality exceeded 60% and pupation was highly

asynchronous due to the length of development. The cost of

rearing larvae with a 3 to 4 month development time would

be prohibitive.

One important fact concerning the results herein is

the ability of Tx. r. rutilus to survive over long periods

without prey. This aspect of the durability and flexibility

of this predatory species could be of some importance during

a biological control program. It may be possible to release

adult females and obtain a fairly high population of larvae

of Tx. r. rutilus in the environment preceding the expansion

of the prey population (Trpis 1972).

Further attempts to reduce the cost of mass production

of Toxorhynchites could center on using a prey species

which does not require an obligate blood meal in its life

history. The elimination of maintaining animals would

reduce the expenses incurred during culturing of the prey.














l:IEi.I) SURVIVAL, MI(TC[ I' ON ANI) OV POSll TIONAI.
LIIARALIT .Ilil I i'F LAIBORATORY-REAREDI)
Toxorhynchlitcs rutilus rutilus (COQ.)



Abstract


Three hundred and fifty, 6-day old laboratory-reared

adult Toxorhynchites rutilus rutilus (Coq.) mosquitoes

were released into a sparsely wooded 13-acre residential

area in Gainesville, Florida. Oviposition was monitored

for 14 days using a grid of 64 oviposition traps located

within the residential area and surrounding woods. Eighty

percent of the ovitraps received eggs, and despite the

migration of the females into the surrounding woods at a

rate of about 7% per day, 64% of the eggs recovered were

laid in the residential area. Equations were derived that

allowed estimating the daily adult survival (S ) and life-

time egg production per female released (F) to be 0.79 and

3.99 eggs/female, respectively.



Introduction


Past attempts to use Toxorhynchites spp. for control

of container-breeding species of mosquitoes failed to

achieve favorable results in Fiji (Paine, 1934), Hawaii

(Bonnet et al., 1951) and American Samoa (Peterson, 1956).









In these efforts, the imthod used involved the release of

small numbers of adult Toxorhynchites in the hope that these

adults would establish populations of the predator suf-

ficient to effect control of the prey species in containers.

These releases failed to reduce populations of the prey

species because the numbers released produced too few eggs

(Muspratt, 1951) and the subsequent progeny were not

numerous enough for control (Nakagawa, 1963; Newkirk,

1947).

In analyzing these attempts to use Toxorlynchites as

a control agent, the most obvious mistake was the failure

to consider the normal relationship between predator and

prey species. For example, Trpis (1972) reported a situa-

tion in East Africa where naturally occurring densities of

Tx. brevipalpus are sufficient to control Aedes aegypti

(L.), but only at the end of the rainy season. Trpis

attributed the lack of control early in the rainy season to

the slow population growth rate of Tx. brevipalpus. The

most recent attempt to use a species of Toxorhynchites as

a control agent was conducted on the island of St. Maarten

by Gerberg (1974). lie was able to eliminate Ac. aegypti

from houses by placing Tx. brevipalpus eggs in containers,

and albeit this is a laborious method the degree of control

was satisfactory.

The observations of Trpis (1972) and Gerherg (1974)

clearly indicate that the effective utilization of

Toxorhynchites spp. as a control agent will require multiple,










inundativc releases of the Toxorhynchi tes adults or eggs.

Ihe author is currently involved in an investigation

to study the feasibility of using Tx. rutilus rutilns, a

species endemic to the Southeastern United States, as a

control agent. In separate reports the author has reported

observations on the life history (Chapter 2) and rearing

on an artificial diet (Chapter 3), and this present paper

contains the results of a release of adults in a residential

area. Herein, the author reports observations of oviposi-

tion, migration, and adult survival for females of Tx. r.

rutilus.



Materials and Methods


The Tx. r. rutilus used in this experiment were mass-

reared in the laboratory on a diet of Ae. aegypti larvae by

methods described in Chapter 2. Pupae were transferred to

an outdoor screened cage 1 m x 1 m x 2 m located within a

screened building. The adults were supplied with water

wicks, honey, apple slices and black oviposition jars and

held in the outdoor cage for 6-8 days prior to release.

The adults were held for at least 6 days prior to the

release for 3 reasons: (1) the females do not begin ovi-

positing until they are 6 days old, (2) by holding them

until they were ready to oviposit, a better estimate could

be obtained on migration out of the release area, and

(3) maximum oviposition would be obtained from the released









Ifilmales. On the evening of July 26, 1970, 175 males and

175 females were released at 3 sites within the experimental

a l'e;I

The approximately square experimental area (Fig. 1)

covered about 31 acres and was located on the campus of

the University of Florida and a neighboring residential

section. The release was done in the center of a student

housing project that occupied a 13-acre plot more or less

centrally located in the experimental area. This housing

project plot was interspersed with various hardwood and

pine trees, shrubs, and open expanses of lawn. On 3 sides

the surrounding area was more densely wooded, and on the

fourth side there was a lake.

Oviposition by Tx. r. rutilus females was monitored

by means of 64 ovitraps placed at ca. 60 m intervals.

Thirty of the ovitraps were within the student housing

project and the remainder were in the densely wooded areas.

The ovitraps were similar to those used previously in

determining the distribution and density of Ae. aegypti

during the U.S. Aedes aegypti Eradication Program (Tanner,

1969; Jakob and Bevier, 1969). Briefly, the ovitrap con-

sists of a pint (ca. 0.5 1) glass jar sprayed with flat

black enamel paint on the outside. The jar is 13 cm high

and the opening is 6 cm in diameter. The jars were filled

with ca. 200 ml of water which had been used in the routine

rearing of Ae. aegypti in the laboratory. The jars were

placed in wooden shelters designed to afford protection


























00


o a
CC





LO
to








C O
0 ',0
0





















T, c





0 A
13 4- 0
-'0i M





0 0







3
C) 0 r-



C A'





*H 0












*rl t
k 0l












from rainfall. The shelter consisted of a narrow shelf to

support the jar and a 15 cm" roof 15 cm above the mouth of

the jar. The shelters were attached about 2 m (6-8 ft)

above the ground on tree trunks and posts to minimize

tampering by children.

The ovitraps were checked daily for the presence of

Tx. r. rutilus eggs which were discarded after counting.

Water levels in the ovitraps were adjusted twice weekly.



Results and Discussion


Just prior to the release of the laboratory-reared

adults, the 64 ovitraps were monitored for 4 days to detect

the presence of indigenous Tx. r. rutilus adults. The

average number of eggs laid per day within the experimental

area was 1.75. Therefore, the best estimate of the con-

tribution of indigenous females to the total oviposition

observed during the 14 days subsequent to the release of

the laboratory material is 24.5 eggs (i.e., 1.75 eggs/day

x 14 days). A total of 407 eggs were recovered from the

experimental area during the 14 days: thus, indigenous

adults were responsible for about 6% of the total oviposi-

tion observed.

The distribution and number of eggs recovered sub-

sequent to the release are shown in Fig. 2 and Table 4 and 5.

The release of slightly less than 6 females/acre (175

females/31 acres) resulted in 51 (80%) of the ovitraps
































Figure 2. Map of the experimental area showing the 3
release points, ovitrap locations, and the
distribution and number of eggs recovered
subsequent to the release of 350 Tx. r.
rutilus adults.








-29-




















-- c-i I-i c- cc a Ln i- o a-.










r Ln -4 t- C-i c0 c 0 L CD 0 0
SLn o r- r- N- r- L- (i


O
C,

O








.l

0
U
0






4-a




0









cc











O
rj










0






o

Ca)








CC


me


r-N A i -
h omomece OH









IDM H H


' 1 t 0 -,I ci C D I-- C I N-
N-~ *- or a n In ^t i-H C- 1-4 0
~b a~~M ~ N -








-31-









*H
c)



O CJ






0C 0









**-- CD ) CO L


[:s 0 7





U Z0


*O CC







o Lc' 0 Lo "
w u









C U
OO criH
U C C













.0 3



0U Z 0





O C
CCC









.0 I L O0
H 3 0 H
tn u Uii.ii ^i u 'd






-32-


within the experimental urea receiving eggs.

Of the 30 ovitraps within the housing project where the

adults were released, 73% of the traps were positive for

eggs. Some of the ovitraps nearest the release points re-

ceived the greatest number of eggs (Fig. 2). It might have

been possible to reduce the number of ovitraps receiving no

eggs by releasing smaller numbers of adults at more loca-

t ions.

Prior to this experiment, the author was concerned with

the possibility of Tx. r. rutilus ovipositing the bulk of

its eggs in just a few containers. Examination of Fig. 2

visually demonstrates not a clumped, but rather a random

(or Poisson) distribution of eggs among the 64 ovitraps.

Mathematically, this is borne out by noting that the mean

and variance are approximately equal for the number of eggs

per container (4.40 and 4.69, respectively) when the 4

ovitraps immediately adjacent to the release points are

omitted from consideration and by a X2-test for goodness

of fit for a Poisson model.

Because Ac. aegypti and Tx. r. rutilus are typically

domestic and sylvan species, respectively, the author

expected to observe migration from the release site into

the surrounding wooded area. During the 14 days after the

release, the percentage of the total oviposition which

occurred within the housing project decreased steadily from

an initial 91% on day 1 to 10% on day 14. A linear regres-

sion analysis (Fig. 3) of the daily percent of oviposition



































GO
C

Cl
SCO

UCC
00





C

CO
C O C
U 04 0

k 1




ot

1-) *- >-











--, u
c rO
0 CO









cc o C-'--



In






[^






34 -
-34-





























+ U
U1














A -1HIIN i3<1 NI iN3 3d i
[L





+





S

+





- 3W| 7IIN2





-35-


occurring within the housing project area columnn 5, Tu'alc

-) on days after release showed tils trend to be signifi-

cant (R2 = .03). Moreover, the difference between the means

for the percent oviposition within that area for the 2

periods, day 1-6 and day 7-14 (0.73 + 0.13 and 0.17 + 0.19,

respectively) are highly significant (t-test). The slope

of the regression equation would indicate an overall rate

of migration out of the residential area and into the forest

of about 7% daily. Notice, that if the dispersion of

adults within the release area was due entirely to random

movement, the percent oviposition occurring within the

housing project area would have been expected to tend

toward 50% and not zero. Notice, also, that even though

the exodus of females was nearly complete, 64% of all the

eggs were recovered within the housing project area.

The determination of the daily adult survival (Sa)

is important because knowing Sa enables one to calculate

the expected number of eggs per lifetime per female re-

leased (F). Following is the derivation of some equations

useful in the estimation of S and lifetime fecundity (F)

from oviposition data.

If S is the probability of surviving from one day to

the next and this does not change with the age of the

mosquito, then the proportion of adults alive on any par-
t
ticular day t is S. It follows then, that the number of
females alive o day t can be represented b te expression
females alive on day t can be represented by the expression


Nt t
Nt = NO Sa










where N is the initial number of Iemales and N is the

number alive on day t. Alternatively, since the numbers

dying are proportional to the numbers alive, the change in

numbers over time can be expressed as


dn/dt = -k N (II)


which upon integration yields

-kt
N = N e (II)
t o


Taking the natural logarithms of equations (I) and (III)

yields the respective linear equations


In N = In N + t In S (IV)
t o a

and


In N = In N kt .(V)


Here, In N is the y-intercept, and In S and k are the
o a
slopes of (IV) and (V), respectively. Notice that setting

equation (IV) equal to (V) and simplifying yields

-k
S = ek or k = -In S
a a


Assuming that (1) the number of eggs recovered is

directly related to the number of surviving females,

(2) S in the field is constant over timea and (3) the
a

aEstimates from outdoor cage studies of daily fecundity (f)
and Sa for Tx. r. rutilus are constant over the time period
being discussed (Chapter 3).









pretreatment indigenous population is stable, an estimate
-k
of Sa (= k ) can be made by regressing the daily egg pro-

duction on days after release. The value calculated

(Fig. 4) for Tx. r. rutilus in this experiment was 0.785

with 95% confidence limits of 0.845 and 0.720 (R2 = 0.79).

This value of Sa compares favorably with estimates made for

other mosquitoes (Seawright et al., 1977; Sheppard et al.,

1969).

It should be obvious, given the foregoing assumptions,

that the regular decline in egg production in the study area

was due to daily adult mortality (1 S ) and migration out

of the experimental area. Since the data gathered in this

experiment does not permit us to partition out the decline

in oviposition that is due to migration, Sa = 0.785 repre-
a
sents a minimum value of adult survival. In larger release

areas where the effects of migration would be minimized,

the Sa term as derived here would more accurately reflect

the true adult survival.

The daily egg production (f) of Tx. r. rutilus females

in an outdoor cage is ca. 1.0 egg/female. By multiplying

equations (II) or (111) by f and then integrating, one can

obtain an estimate of the expected total egg production,

i.e.,



fN / St (Via)
t=0 a






































0







CO
U
0







S II

*HIo





H OC
o -


0 n




00

*H






0it




Ul-



















W
in
+ a:
NJ
W



+I
+
IL
IE

in

1_
.3 3






-40-


f N ekt (VIb)
o t=0


The expected egg production over the period of the release

(day 0 thru 14) obtained over the interval 0 to 14 is

699.8 eggs per 175 females released. Notice that the

approximation of total egg production


14
f N S (= 787.7 eggs)
0 t=0


overestimates the above integrals by 13%.

Of the expected oviposition, only 58% (407/699.8) was

recovered in ovitraps. Part of this may he attributed to

other oviposition sites within the experimental area. Five

days after the release, a tree stump flush with the ground

containing rain water was discovered in the densely wooded

area. During the remainder of the experiment, i.e., day 5

thru 14, this one natural site received eggs equal to 17%

of all the eggs recovered in the entire experimental area

during that period (36/215). This suggests that a good

deal of the egg production went to sites other than the 64

ovitraps.

In the context of using Tx. r. rutilus as a biological

control agent against container breeding mosquitoes, the

results reported herein are encouraging. Although migration

out of the area was observed, the preponderence of eggs

were laid within the urban area. The values of female






-41-


daily survival (Sa = 0.785) and lifetime fecundity (F =

3.99 eggs) are high enough that the cost of area t ratment

by inundative release ou adults should not prove pro-

hibitive. Finally, demonstrated here is the ability of

Tx. r. rutilus females to locate oviposition sites typical

of Ae. aegypti habitats and to randomly distribute their

eggs among them. Using these preliminary data, the author

is planning an adult release where the control of a prey

species will be monitored.














A DI:TERlR[NISTIC MODElI, FOR SIMULATING T'III IPREDATION OF
Toxorhynchites l s rutilus utilus (COQ.) ON Aedes aegypti (L.)



Abs tract


A deterministic computer model is presented detailing

the interaction of the container-breeding mosquito Aedes

aegypti (I..) and the larval predator Toxorhynchites rutilus

rutilus (Coq.). Results of simulation runs involving the

release of Tx. r. rutilus adults indicate that predator

releases resulting in 1 predator larva per container are

sufficient to reduce Acdes adult density 75% in 20 days.

The slow rate of immature predator development enables con-

trol to be maintained for several months. Simulations of

predator release and the use of adulticides indicate that

it is possible to obtain zero adult densities. Finally,

the model indicates that the most important parameter

determining the degree of control established is the dis-

tribution of predator eggs.



Introduction


Historically, mosquito control involved source reduc-

tion and insecticides. Predicting the outcome of these

measures was simple and straightforward. Currently, addi-

tional methods of mosquito control are being studied, some










of which involve the interaction of 2 species. Difficulty

in understanding the dynamics of the interaction between

the 2 species makes predicting the outcome of new methods

complicated. Attempting to optimize a strategy utilizing

2 different methods in conjunction, e.g., predators and

insecticides, is particularly difficult. Therefore, com-

puter simulation models are becoming increasingly popular

as a tool in the development and evaluation of new control

strategies.

The purpose of this paper is to: 1) present a determin-

istic computer model examining the interaction of Aedes

aegypti (L.) and a larval predator, Toxorhynchites rutilus

rutilus (Coq.); and 2) simulate the population dynamics of

Ae. aegypti under different control strategies involving the

release of Tx. r. rutilus adults and/or the use of adulti-

cides. Pertinent to this discussion is an explanation of

methods used in life-history analysis by Mertz (1970). The

utility and rationale of simulation in developing control

strategies is presented by Conway (1970), Haile and Weidhaas

(1976), and Weidhass (1974).

It is desirable to emphasize here, that while many of

the values for Toxorhynchites parameters are from laboratory

experiments conducted under conditions obviously different

from those expected in the field, there remains a vbry real

value in the model building exercise. Characterizing the

interaction of the 2 species highlights areas where better

information is needed. Additionally, simulating with the









present model gives us insight into how best to approach the

first large-scale release experiment where prey control is

to be attempted. Finally, the preliminary model gives us a

framework within which to interpret the resulting, experi-

mental field data.



Model Description


The model described herein may be called a compartment

model (Miller et al., 1973). Each stage of each species is

represented by a number of storage registers within an

array (an array is a group of registers). The registers

represent 1-day age classifications of the various stages;

a particular array represents the age distribution of a

particular stage or instar. Except for the Toxorhynchites

larval stage, the length of the various stages and instars

is determined by the number of registers within a particular

array. The model is made to cycle on a daily basis by "re-

placing the contents of the next storage register with the

contents of the previous register multiplied by the daily

survival for that particular stage and species; the output

of the terminal register of one array (stage) is the input

to the next array (stage). Most values of daily survival

for both species are fixed during any particular simulation

run (Table 6); the daily survival for Ae. aegypti 1st and

2nd instars is a density dependent variable. The length of

larval life of Tx. r. rutilus is also a variable -- it being

a function of the amount of prey available. Details of






























.-)
"-
H


-,C

.-







4-'
H
C1




0

0.

3

a


r1t
>
2,
2i


i



-













0
-Cl


c


cO 0

c c



S0 0 E
0 C E




C- 4
*) 'c) '-







U) m <

SO O E
c' o-C ). *-i













S 1, M 4
Cto *, n




C 0 C) 0 > 0n
u c 0 a) i
U -V)








U r c U 0 4
0 (











:5 4- 0' C)
) 0







v a :5 0 0
0 )> -H H *-
n c s .i o > "
Oe H *t 1 C o













0 O0 0 0O
*H 'r C)


C O 4J -r-I O










oo o l-i o *r rt
CO -H 0 4) *

S- C












CC
Go CD 0 0 C 4
0 L 0C 2 ) CD C) 0




















>^ < r H C) C

23 C- 11 2 1u 'n 0 <
S 1- 4







C)


z
cc
H--v
H --HCC
ccCU CCc 01


2-






m C0


73m >
.0C *H











S2 0*



L- ra o

> O
Sto 0
0 0O CZ
:3 > C









*- r-1 (0 *




c O rC)










00 C) CD
3 -. t-' C)






-46-


these variables will be presented in detail presently.

The data enabling the modeling of Ac. acgypti were

largely derived from Southwood et al. (1972) (immatures)

and Sheppard et al. (1909) (adults). These life-table and

ecological studies were conducted in Bangkok, Thailand in

hopes of correlating Ac. acgypti population dynamics with

the known seasonal incidence of dengue haemorrhagic fever.

Surprisingly absent from both studies was any information

on fecundity and ovipositional patterns; estimates for these

parameters were derived from data on Ae. aegypti in northern

coastal Florida (J.A. Seawright, personal communication). Data

for Tx. r. rutilus came from laboratory, outdoor cage, and

field experiments conducted by the author (Chapters 2, 3

and 4).

Briefly, the Ac. aegypti situation to be modeled is as

follows (Figure 5): according to Sheppard et al., the main

source of all Ae. aegypti breeding in Bangkok is the earthen

or ceramic water storage jar (ca. 100-200 1 capacity) which

is found in association with all types of housing. They

usually contain water throughout the year and are replenished

with rainwater, tapwater or riverwater. Southwood et al.,

state the water storage jars number about 150 per acre and

about 53% of these are positive for aegypti immatures at

any particular time. Interestingly, both studies reported

remarkably stable densities and distributions of all stages

from season to season. Adult densities of 1100 per acre

(or 7.3 adult/container) were reported; Table 7 presents
































0

*H r-





-n>
H










CC
0 Ir,
i-7













0

0 0 ,H
c



*H

0 ) 1
(C










cC
*[-









48-





c.


Al- II
0

0.)W

0.)O
-o 0.




































E




0 C0
rt





r..) o
o o
0 0




S0















3 1
r-
















0 4-
0 0

O


*

00



c1 0
0 *C






30 0


000


*-' 0








0 r0
*r- +-


r- I
*r) f
nI1C ~


1 O C CO 0 D
















































0 + U
2 0 C


I~ i U- t- E









imlimature development times and average numbers of i mnatuires

per container. Southwood et al. note that 20% of the

embryonated eggs hatch without the flooding stimulus de-

scribed by Christophers (1960), and that app roximately 50%

of the remaining eggs hatch with each subsequent reflooding.

Since they recorded no data on the frequency of flooding

for the jars, the computer model was simulated for no-rain

and daily rain situations. A final feature of the popula-

tion dynamics of Ac. aegypti in Bangkok is the density

dependent survival during the first 2 larval instars. The

stability of the mean number of immatures/container (Table

7) appears to stem from a paucity of larval food; the num-

ber of new 3rd instar larvae is a consequence of the avail-

able food and not the number of newly hatched 1st instars.

An algorithm for producing the observed larval densities

which is largely independent of oviposition or frequency

of hatch is detailed in Figure 5.

The subroutine describing Tx. r. rutilus, with several

exceptions, is similar to that for Ae. aegypti. Reference

to Figure 6 reveals: 1) the 3 last immature stages are

separated into 3 arrays; 2) that eggs hatch on the second

day after oviposition and independently of rainfall; and,

3) that oviposition begins 6 days after eclosion and occurs

daily thereafter. Notice that Figure 6 shows laboratory-

reared adult Toxorhynchites being released at 6 days of

age (Chapter 4).

The nature and mechanics of modeling the predation of

































,:c

-H r-




C 0




-3


S0 r
*H 0


l-.










0 >
*I







3t1
H
-1 Cl
+cT o








4- )









3 c





































C-.





-53-


T.\. r. rutilus on Ac. aegypti are presented below. Table 8

presents development times and the numbers of prey devoured

hicen the various instars are offered either early (1st and

2nd instar) or late (3rd, 4th instar, and pupae) Ac. aegypti

immatures. Table 2 details how long the various Tx. r.

rutilus instars can fast before death and Table 3 presents

the duration of larval instars when fed on organic, non-

living diet. The situation to be modeled is as follows:

1) Tx. r. rutilus 3rd and 4th instars develop faster on a

diet of late instars and pupae than when fed early instar

prey; 2) all larvae develop at a rate proportional to the

amount of prey available until the amount of prey exceeds

that given in Table 8; 3) all predator instars can survive

without prey for a period of time and if provided detritus,

can develop at a very slow rate through to eclosion; and

4) the daily immature survival (SIT) is constant, irrespec-

tive of diet (Chapter 4).

In the model, 1st and 2nd instar predators (array T12)

are assumed to eat before 3rd and 4th instar predators (T3

and T4, respectively) on any particular day. This assump-

tion simplifies the programming of predation, and does not,

considering the relative amounts of prey consumed by each

instar, introduce significant errors. The number of storage

registers within the arrays T12, T3 and T4 (16, 32 and 96,

respectively) do not correspond directly to 1 day each.

This is because the larvae are moved incrementally within

an array as a function of prey size and numbers of prey






















C)


0



0




0
*I














0




0
CC





a,













4
c
0
Oc




































-.I
*>
o

















HI
..



















00
rl














o -
0







-n


CC

CC-*





0
-o
CC
clut
c -
a i,





H
n



c-u









4 t-<


O












0
n








S::









C









O
0



CO




I-)








01






















C
H
un


C^


IC LO Hi












^ ,0

e c

























oCO o
oo
n N I) n oo




























'0 10 .0 0
10









l C 0 C)

co,
i-I (-J 00 ('-
(C


0









cr)












H U
t-*

4n




















0
in E



o O









7) C,
u 0







0U 0




4 c
f 0


U


0 C


I I I -









available. If no prey are available, all the predators are

incremented 1 register forward. If the number of prey

equals or exceeds the numbers presented in Table 8, the

larvae are incremented 4, 8 and 12 registers forward in

arrays T12, T3 and T4, respectively. Intermediate amounts

of prey result in incrementing the predator arrays an amount

proportional to the available prey. The Ac. aegypti im-

mature arrays in containers positive for Tx. r. rutilus

(XA and YA) are updated daily for the effects of predation.

Since cannibalism does not occur among the predators at the

densities considered here, even when there is no prey,

cannibalism is not modeled.

Following a release of Tx. r. rutilus adults, one would

expect on the basis of the distribution of Ae. aegypti

larvae (53% positive) and an assumed predator distribution

of 80% (from Chapter 4) to obtain 4 types of containers:

42.4% positive for both species, 10.6% positive for Ae.

aegypti only, 37.62 positive for Tx. r. rutilus only, and

9.4% containing neither species. Since it is assumed that

no Toxorhynchites adults result from oviposition into con-

tainers devoid of prey, and the 4th type of container pro-

duces neither species, the latter 2 types of containers are

not specifically modeled in the program but are accounted

for with scaling factors (Figure 7). To represent the

salient features of the interaction in the field and to

represent the resulting Aedes adult density, 2 containers

both positive for Ae. aegypti are included in the model,

























a)


t0

C mi
a'o



,D 0
ioaC)



S.co





* O *
'Cu
3 U

'-' <| e











,)a
n


(I +-






















4-' 0) 1|
) 4-) CC






oi n O
u h p~

(U (Ui
C)l^H > *

0CD ^ -
^ ^






-57-


I


v -
*LYlIf
O U
fc* J


LZ
I.. ~-,
U (
Z i,


Stl
'*
,,
L
v
ii









only 1 of which is interfaced with the predator subroutine.

Ac. aegypti cclosion and Tx. r. rutilus oviposition are

scaled by factors of 'DISTBN' and 0.53, respectively, to

depict the frequency of the 2 types of containers in the

field. Notice, that because of the scaling factors, the

number of Aedes and Toxorhynchites adults are in terms of

1 container and that multiplying the output of adults by the

number of containers per area gives the. absolute population

estimate.



Results of Simulation


Figures 8 and 9 represent model generated Ae. aegypti

adult densities on a per container basis for the 'no rain'

and 'daily rain' into every container situation, respec-

tively. Figures 10 and 11 depict the corresponding total

number of Aedes immatures/container for the 2 rainfall

situations. Tn each instance and in all subsequent simula-

tion runs, the model was initialized on day 1 with 7.4 1-day

old adult Ae. aegypti. The resulting numbers within each

stage are largely independent of the number and stages used

in initializing the model. The graphs have been smoothed

by plotting 4-day moving averages. Notice that the cyclic

nature of the 'daily rain' situation in Figure 9 dampens

with time.

Table 9 presents a comparison of the number of Ae.

aegypti individuals per stage per container observed by

Southwood et al., and the number of individuals per stage




























I)
*H

rt
cC





C.

C-







0%











e0
(C
U



CC











0,
1)i

T3
.U



t-1


i-
3 r








-60-


* -.- i- -- -^ -- ------ 4 4 -- --- ----- ---- --- ----- -------- ---







Ci
I I C



ro













-. ----- ~ ~ ~ ~ --- -- -- ---- ---
0(














SI -
I I







0)
I 4
I I















I I I
oo




I I: I I
I I






I












SI I I 00 I -



o CDcr a
N -N1



































*H






4J H
Ur-








C)

v)
oe
















-1 H

0 $
CC
u'-4 '













ZO
U,
Cl


















n
[ I-U
*HC
ox 0>















S* -- -- *--


a
I





I*




III
:; (M--T ;
I a a Ift t
I a pI a a

























g.-
II

II
* ftftf


















*
I a













* ft + f .w_____________ a
I~ a














a- a


















* f
*


a aa .. a N <
a


















*
ft 4 ft. f
a a a a a
a a aa a~
a~~~ a af
a a a a -* a-~
a a a af t t
a a a aftft p
a a a a ft a a
a~ a .
a~ a. a a a
ftI ftftf f
P a a a .
a~ a a a ft

a a- a a ft-- -- ;- a a

a~ a .. a



a-- a a a .. a a
a a

a~ a f
a a f
a a f .
a- a a a .. a a
ft -fI ft ft -- t t tf




a a a:- a f-- t ft---- a

I ~~ a I

ft --- ft- ft-- ------ ~----- ft - ft f --- ft -
a aO a.
aU aU a .










































4-J*
0
U


C(

0










4J









o








b-
J3
EL-
C3


C*



T-c



-I










&C

Lr


LI,





















.aCi











IM
-I I - -M




























to
----- ----- ---- --- ----









aIn














*I I tI










o






























Li
ci



cc





rC
Li


Cci


CJ
0cc
04-
C,
0








~ccrc



ciC)








00



Cri




'4rl











-66-









**





I



N














































* I I-- - - I-- -
*




































-- ---- ------- 0,
is








ear

S*













I*









I I 4. 0)


























I I


I I ,









-67-


r-
0





*4-










S
u










































0 0


oM
a













































0 0
0




























UC
"-












r 0
0 0A






























u o
00
C4
- T










00
4o4
















-3
C


LO O 0C, LC
0 *


'0

















CO






























O0 0 l 0- 0
I)r- r-























0 n ol C- a) 4

oo





















0 H H 1 0 3

-, 4-


4-

0
0
O
0










C4
0














0


4lJ
U)









per container generated by the model when no control mIea-

sures are applied. Notice that the model generated total

number of immatures/container exceeds the total reported by

Southwood et al. This disparity was allowed to remain

because the difference was not large (especially in the

'no rain' situation) and would make the model more conserva-

tive in subsequent control simulations. Density-dependent

1st and 2nd instar daily survival (SI) for the 'no rain'

and 'daily rain' situation averaged ca. 0.75 and 0.50,

respectively.

The effects on Ac. aegypti population dynamics of

contact adulticide application causing 95% mortality among

the adults with no residual action are presented in Table lP

and Figures 12-14. Notice in Figure 12a, as a consequence

of 1 application, the overshoot of adults 1 generation

later. Notice also (Figures 13a and 14a) the increased

frequency of application required to maintain an increasing-

ly lower adult density. Because of this, population sup-

pression has usually involved the use of larvicides in

addition to adulticides (Gould et al., 1970; Gould et al.,

1971).

Figure 15a shows the effects of a predator release

resulting in 1 predator egg in 80% of the containers on day

98 during a period of no rainfall. Notice that a 75% re-

duction in Ae. aegypti adult density occurs within 20 days.

The Aedes adult blip beginning about day 170 results from

the eclosion of the predator. The effects of the subsequent






























0
O



I--



0 Ci


*0c







4c 0





0 0




01
C7




O
Cr
4*H










u
0
oH


4 _









o





U"

0n 0



0 -0 V)

) r)
lI .f-
*-' C
cu o
'i03r


C3
U 0






0

C 0 +-'


03

r(1

























C














Li
00 Co o-







o







C:)
-lO


































Lnf



4O1 C 4

0 )*H
000






0 CY| m
0
g) rd *H


'0 W
03 0
C) 4-1 -

0




.1 0



m u







S0H 0
|C 2*
Hp -H r
J4cO t









.-I
























or





to

: I







N




*- ------------------- y,-



OT
















------ ---- -------"--.--.---iL ----_ _
I





I~ Q. a




















o o










I
*

-- -- ---
.. <0
I~~ *
II* *




I* f

I~ *o


t~~~-----~-.-5----'^--



























a)
-t

u

,-J






*0












0 c
ci

(3









Or
23























C0
C













-A









i*
ci


(C1


**r-o







































tt
* - ----- ------ .. 4 -4-- ----- -- ----- ---- ---
I I I I 4
















iL
I -- --- I ------- I - --- -- ------' ---------
I I I I I















:
I I


























_------------ --------------i------:---------- ----*----- C
I I I
* I
























SII

























i








*
i i I
I I I I I




































I I I I 4 II






II I
I I
* II


* II
* I I*
I II I I r






I I
































































***
SI I I 4





I
* I I I 4 II
* I I I 4



I I I 4 II
* I I 4 I
I I II 4 I






IIr 4 I
I II









II 4 t





SI I I I
II I
* I I I

* I I 4
* I I I 4 II


II I I I




















0 CM I* (0 CO
I I I 10 I
I I I I.
I I I I
I I I I I I

I I I I
I II I I

I I I I I I.
* I I S u
*I I I
I I I I II
* I I I 4 I


* I I I

* II *4 I

I I I I I ~ I
I I I II

II I I I I


I CI I IO 4
I I) IC .44



























4-a

0 C -H





-0 0.H
o3 U

C C-
Ot u U









u r
o < -









0 o:C o)
4 0 0-




tl 4-i Hrf

C (1) 4 -
4 -H 0





iH U
0 0 43 *H







ir n
4-) 0 -




OH
U H4



O 0

( 0 t 0 (





1 3C r1





rto





Ob




















** a
----------------- --- -- -


:.ro






a j


















- -- - - -: -. -
* a


* *aa
:'
.a a....











.--- .. ..---.------ - ._ ...






















c- ------ ----------




















N--------- --------------------
+ a







N. D.
a









0 )------- -- --.. -- _ .- _
a I a ..












-- -- - - - ..____.-- ._ .. -- - L~ -..--





















g '" w~ "~~^~~~~~~ao~~~'~^~---- '0i
a aa ..a









a ------ ----a -.----*----- a-,- -- ---

*u a a I












































C


o t
Oc



0 M






EA
iC)






Ot



(o),








o o

0; U
4-) H/
r; *rt












U 4
UC)

~iz

*In
*rel
*rl




'c;


* &.
O ft

(4r














-- -- --- -------- -- ..



















Cr
N














*
------- -


















O(0
0































i 0 1 M
N d cr o.
Q':



:. m cu




























4+

*


C3 Or
'0 'H
U) 1 1 z)



U C

U $ .r





),



U 0 0

.-i J ) U) 4i C7)
C- u-



A14- [
of) 3 CO


0|U 3

r'H 40 a

-' 4-i n
U
C0cl0




U -- H -
E CnX

*r-l r
11,~


















-_-----,- ----.--______---- -----* -;--------~-- --

._ : -
S SS


a





l
* S S S S S

S S S S* Og


.------+---_----*-_-__^--r--- ---- ---


,J



I""T I
S S
~* * S




----- --*-_--- ---------+ -------------,- -------- ,-
S S S 5 C






.:* S SS
I I ^
* *


~ ~~ S 5a
~~~ S
IIII


_ _ _ _ - ^ _

S S S S a a
~ ~~ S S t C
~ ~~ S S


* * * *













Sa


:S a









o
*--- - ________ _----- ---- -SSa S ^


















* S a a















.
i, ,i i I

~I I I

I~~I I I I
























-: . . . . .
o I I I


















* a a
~ ~ ~ 5 5 $
~ ~ ~ S s

* 5 aa
I-2 E5L I




























S S a

* a
I~~ I




r---------- ~ -------*-----+--^---.-r--- --~--0+
S S S 5 5 *ar S
* S S 5 5 55i 5
* S S S S a
* S S C Sr
*I S S S S a -
* S S S S C
*----:---- S S S S a 5--';--:
* S SCa

S S
* S SI S
S S S S a 5
* S S a a S
*--------- S---------- S-- - - S a a a S--r
* S S a a S
* S C S ~


* 5 5 14 5 5 I


* * ** -- If
p 5
*1 5 5 SS
S 5-- -- 5-_ __ 5-- -- - -~ aaaa S S
* S S *55 I S
* S 5 5 55 4
S S S a s
* S S CC S
S S S saC S S
S S S S as 5
* S Cs 5 5
* S S S~ aI
S 5 5 5 ~ 4 *5
*------- S- __+--- S S 5 555-----
SI 5 5, 5 5 .
S 5 a




o1 t
N N 0 '
































u
U
*H
.CJ

*H

4)























In





T)
0


r-t
S*H
C3



















U|



















a g

__- -- ,----- t----t--- --.:; r


S4 S
* S I 54 4-
*I *I 4
* I I S





___-_-_--- --r--------- --------- ---
S S I S





,n
___---- -- --------,-- --
** a S


S

* 4
..* S S5








S S 5n 4n 44



I
* S S S I 4 5


S. 5















o
r
,-------' -------- -------' ------ --*----''~
S S 5 5 4








I







o
.

-- -- --- ----_- < (ft----+
S 4 --

*~~ S S
I 5






'* S
* I S 54






-4*
* S S I 5 44

*I S S . 5 .
* 5 5 I 5














I
----- -_-----*--- ---------~ -- ---------.".--*. ^-------
*--_---_- -- --- ----4
S S 5 4















~ ~ ~ S 5 4
* I I 5 54
* S I 54

*~ I S 4

* I 5 5 4
















- --- -- -_ ^ ^ ^--~- -- --
*~~ S S




























"
5 5 5

.. .. .. .. .. . . .. . .. . . "_. .
0 S I 54
5 5 4 44 I


S I 4 5











;----_.-- _____.-----:-_----- -----:-----------r -:--------1----- Q
*I 5 4
S I 4
5 5 .4~ 5
II 5 5. 44 5

- -- In


5 5 44 I S




S I 5 4 4

S 5 5 I 54


'----'--- -- -------'------- 0


1* 0 in cN


































00 C
>





g-_

Um

3 tD o

C *k

C)o c
'H +2

























rL :
0 0
r1 U






oC co )
r +
+2i 0 Y











1 + ( '-H
+C )




















-4















*----- ---- ----t ----------* -- ------4.-.

























D
~~ a
_ __ -----4 _-._ -.

















a








a a a a

~~~ a a a
~ ~ ~ a a.
** a aa



, I
* a p a































~* a aa
~~~ a p aa
* a






























* a a a a
* a a a





























*I


~ ~~ a aa
~~~ a a aa
a aa*
----------- ---------- ----- --1"^---
* a a a a
Saa a a
a p aa
Sa a aaa a
a a a aa a
a a a
a a a
*aaa
p a a a a. a

I I a a





SI I


*---- 4- ---4--- --a---- -------- --a----








I .


a a_. a aq
* a a a a.
* a a a a. a a
* aa .


*---- a- a---- a---- a------- a---- --
a a a
a a a a a a
*I a a *4 a
* a a a. a a
a~~ a .
* a a a. a a





U C I Ca a
aI a a a 4at a



a a a a ~ 44 I
*~ a a 1

a a a a a. a
a~ a a *
a a a ~ a a a


oI t


N-- N- ---- --- - r.









2nd generation of predators resulting from the release are

shown in Figure 15a after day 185. Until actual field

trials are made, tie reality of the effects of a 2nd genera-

tion will remain uncertain. The model does depict, however,

smaller and smaller subsequent generations of predator --

results which are consistent with known equilibrium densi-

ties of predator and prey in nature. Figure 15b reveals

that 1 predator larva is sufficient to deplete a water

storage container in 6-8 days. The model further shows

continued oviposition by declining numbers of Ae. aegypti

provides food allowing the predator to develop and eclose

in approximately 53 days.

Simulations involving larger adult releases producing

greater numbers of predator eggs/container did not appreci-

ably alter the Ae. aegypti decline. This is: reasonable in

that 1 predator/container is sufficient to stop prey breeding

in that particular container, and additional predators are

superfluous. Larger numbers of predators/container further

exacerbates the prey shortage producing even longer develop-

ment times. In the field, the variable number of predators/

container would likely result in asynchrony of predator

eclos ion.

Since rainfall increases the number of young prey

larvae, 5 predators/container are required to eliminate

aegypti breeding in containers positive for Toxorhynchites.

Figure 16 refers to the failure of 3 predators/container to

establish control. Furthermore, the increased food supply

























VI
C)

0 C 0
CO H 1 C

C)



.,I cc
0 Nl Cl 0
col
000


O 0
0 u| ro
o o

\ <1 o.0












0 0- O








Ol *r
F- 1i











eC
3rfl C

U Cr1

















F-




















S-------- -------* --------- --------- ---------.




10
























I *







4
*~ U




** i
























l o
S *
S











*: I D
o



I I i



S S ,
*- --* ----
























I

























o
S ,
S S S S






I
*~~ S S






I




al o
























o
S S*










































0 1 1 (0 C
* S S S S
* S S S 5
* S S S S
* S SS S
*1 -- I~ -

* * *
* SS
* S SS S
S SS
* S S S S
* S S S S
* SS
*I S S S S
* S S S S

*~ S S 5
* S S S
S~ S S S





* *--,------ *-------
* S S S Su
*I S S S S S
* S Sr S S ~
* S e S 5 44*



* ------------------r----------4--r---- -4 ------ 5 4 ------4--
* S S S 5 45 5
*I S S S 5
* S S S 5 45
*~ S Sr S
*~ S S S
*1 S S
*~ S Sr ~ S
* SS
S SS S

* S S, S
* *p * *
































O c

Q 4J








co k
41 0CC
Cm



r*
') C



.c 7 Zc
CI C0 >
3 >NCC 0













*H



*k o'


0 4- >















CC




















*----------- ------S --I--- -* _

* I S I 44 9
* I 5 4
* S I I 4-
* I I I 41

* I *49 S
* I S I S I
- c-_- _-_ --~-4 ---*4tt--- --.---
* I 4
: _ro

I I i I. S(
I I C 1441
S I
I II I l 4
I I I 4 4
I I 445
I I I 4
r---- --- -- ----- -- -.-- -- - -- -_----- _ -_-

I S I 4
I I
. .I ..

I I I I 44
II I



r ----r--__----- --_---_. ______ __:l_-
I I S 5 4 .
I I 5 C%
I I I 5 4S
o









I S. 14j
I I 14 I
~ ~ ~ I 4
I I I I 44
S I S I 44 5
I I I I 4
I I 44 I
*-----+-------+---------------*-
*I S I 144
I I I 4 4 I
I I *4e P S
SI I I 4 4
I SI I 44 I I
S I 444 I S
I S I 44 I I








*. I 4 4


I S 5 I 4 5
141







--* ------ ------ -----_---------- ---- _- ^
S I 4 ,
I I I 5 4
P I I I 5 4
I I I 4

I I I II 4 5
II S I 4
S I I I 4
S I 44 5
I I I 4 I
SI I I .4







:" ;
I I *
I I I 4


SI I I
I I I I 4 5
I S I I 4 I
I I I 4 I I

*I S I I :r ,r
0 I I I I
I 5 I *4 I
I I I 4 I
I I I 44 I
SI I 5 5 44 I
SI I 4144

I I Ir 44 I
SI I I 44 I
I 144 5 I
* I 1 4 I I
* I I 5 4 I I
I I 44 I I
I S I 44 5 I
* I I I 4 44
-------- --------r------. -- __---c- --- *



*I S I 544
I I 5 4 I
* I I 5 5l I
I I I 4

* I I I 44 I I
* I S 4 ..I I I
I S 44 I I

*-`-------- -----4--------,-*----- --------*-
S S 5 5 4 4454 I
SI 5 5 5 4. 4 I
S I 5 5 4445)
* I S I I 41
* I I I I 44
I S I 5 44 5
5 5 I I 444
I I SI 4. 44 I
* I I I 4 441 I

o to

N-N






-89-


results in shorter preda itor developinent times res ulting in

shorter periods of control from the first cohort of preda-

tors. ''he ramifications of this depend on the results of

subsequent generations of predators resulting from the

original release. At any rate, the lower numbers of preda-

tors required in the dry season to initiate control indicate

the logical time to attempt control.

Figures 17a, b and c represent expected Ae. aegypti

adult densities when the proportion of containers receiving

1 predator egg/container is 0.6, 0.8 and 0.9, respectively.

The proceeding discussion and these figures lend support to

the idea that the most important parameter determining the

degree of control established is the distribution of preda-

tor eggs (= DISTBN). It seems likely that energy expended

to improve predator distribution by releasing smaller

numbers of adults at more sites would be well spent.

Simulations involving the use of adulticides and

predators reveal the following: 1) One adulticide appli-

cation shortly before or after a predator release does not

significantly decrease the resulting prey density nor in-

crease the rate at which it is achieved (Figure 18).

2) Several applications 2 or 3 days apart after a predator

release should result in immediate and sustained prey

densities near zero (Figure 19). The maintenance of con-

trol is again, a function of the efficacy of the resulting

subsequent predator generations.




























0



0 *n







O *HH
moo









000
(D 0






0 7D Q





u 0 M -
0) 4




0 '-'
TCo



cric









0tC
C) 0 0j


O Cf
0 441












1i-




ca




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