4Q
. atA
. |.
r.

52 .51..

.
v‘. saIMvhf
0|. ‘2 n I
dodfi 1ri1.x..v .
p (a. .3 Ha ".I.I§.
bib 1-‘F-3'. u. .p

 

 

k.
f 7.5.0:.“ 3:»

05"! I‘ll

K

          

  
 

 

‘v I 1 ll I . L .
I . V . y . , In"! tn. 05:31- III! III...
.v. .. - A .5 . . . thnflfhnf.“ . ax. I I I I . 1-9.6:- . I Y0..-kr.chvlvtprliouvit...!
E . , t. I. . , y u t‘ . , . . 15.1.0.“ 35.”: I" .r'ouflx: ufin “fill .III... I .I.

 

If A . I . .
.u I. - I 6:10.. 7...: $232.... r. ..,.v y . a; 1..-. all, 3 1.- . . . 3, .. . V
In . t y I D. o A . It} , .l-.. I75. . . I v t I . . .1 Ifhl‘. ‘Tblpniodvtikr .
. v.51 P. H on}. . In...» . .h .5... . ya...) . I L .. q - t 5?. L.r , . I. L:.l..+ ‘ WK. 1;... . . , .19.“ I. . - . I
\
- I I ‘ H - ‘ ‘ ‘I‘I ‘ Il"‘.’ .

4 M to? 5931
111111111111111 1111111 111111111 111 LIBRARY

3 1203 00609 1197 Michigan Ste 3

University

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

This is to certify that the
thesis entitled
LIFE-HISTORY ASPECTS AND NATURAL DIET 0F LARVAL
COQUILLETTIDIA PERTURBANS (WALKER) (DIIPTERAzCULICIDAE)

presented by

EMILY JEAN OLDS

has been accepted towards fulfillment
of the requirements for

M. S - degree in ENTOMOLOGY

Kai/”W

Major profess“

 

 

 

93/25/57

0-7639 MS U i: an Affirmative Action/Equal Opportunity Institution

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or More data duo.

     

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmutivo Action/Equal Opportunity Institution

#7

 

LIFE-HISTORY ASPECTS AND
NATURAL DIET OF
LARVAL COQUILLETTIDIA PE RTURBANS (WALKER)
(DIPTERAICULICIDAE)

By

EMILY JEAN OLDS

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Department of Entomology

1989

51:71’5/é

ABSTRACT

LIFE-HISTORY ASPECTS AND NATURAL DIET OF LARVAL
WA W (WALKER) (DIPTERA:CULICIDAE)
By
Emily Jean Olds

Cmufllemdia mm is a medically important mosquito implicated in
the transmission of Eastern Equine Encephalitis to horses and humans in the

midwest and non-coastal parts of the eastern United States. Fundamental life-
history and larval-feeding data necessary to design effective control programs
using microbial pesticides are not currently available for this species. In this
thesis, methods for locating, sampling and sorting Qgpanumans larvae were
examined. The occurrence and abundance of larval 951. mm were
measured at four sites over two years, 1987 and 1988. Parasitism rates were
calculated at two sites for a previously undiscovered mermithid nematode
parasite. The natural diet of Cg. mm; larvae was examined and the

seasonal, site and instar variations in diet were determined.

ACKNOWLEDGEMENTS

I would like to thank the following people for their help and support in this
thesis. The following student workers provided much needed field assistance,
and to them I extend my thanks: Dell Richberg, Chris Edens, Paul McCann and
Cliff Leavitt. Thanks to Dr. Sheradan Haack for identification of bacteria. Dr.
Jean Stout provided excellent editorial and scientific guidance. Special thanks
for field assistance, and perspective go to Julie Greminger. I am greatly
indebted to Bill Morgan for his help in many ways. I am very grateful for the help
of Dr. Edward Walker during all stages of this study and would especially like to
thank Dr. Richard W. Merritt for his guidance and support. Finally, for his
love, tolerance, sense of humor and support, I would like to thank my husband,

Tony Davis.

This research was funded by a NSF Graduate Student Fellowship and by
NIH grant number Al-21884.

TABLE OF CONTENTS

Page

THESIS ABSTRACT .......................................................................................... ii
ACKNOWLEDGEMENTS .................................................................................. iii
TABLE OF CONTENTS ...................................................................................... iv
LIST OF TABLES ................................................................................................. vi
LIST OF FIGURES ............................................................................................... vii
PREFACE TO THE THESIS ............................................................................... ix
MANUSCRIPT 1. Evaluation of methods for sampling,

sorting and locating the habitat of larval

' ' ' We (Diptera2Culicidae) ..................................... 1

ABSTRACT ................................................................................................ 2

INTRODUCTION ....................................................................................... 3

METHODS ................................................................................................. 4

RESULTS .................................................................................................. 8

DISCUSSION ........................................................................................... 12

LITERATURE CITED ................................................................................ 14
MANUSCRIPT 2. The life-history and seasonal abundance

of larval Qqufllejfldia Was in south-central

Michigan ..................................................................................................... 16

ABSTRACT ................................................................................................ 17

INTRODUCTION ....................................................................................... 1 8

METHODS ................................................................................................. 19

iv

Page

RESULTS .................................................................................................. 24
DISCUSSION ........................................................................................... 37
LITERATURE CITED ............................................................................... 40
MANUSCRIPT 3. The natural diet and feeding behavior of larval
Qggumenjgia gamma: (Diptera .Culicidae) .................................... 44
ABSTRACT ............................................................................................... 45
INTRODUCTION ...................................................................................... 46
METHODOLOGY AND EXPERIMENTAL DESIGN ........................... 49
RESULTS .................................................................................................. 54
DISCUSSION ........................................................................................... 72
LITERATURE CITED ................................................................................ 77
BIBLIOGRAPHY ................................................................................................... 82

LIST OF TABLES
Page

MANUSCRIPT 2
Table 1. Number of samples, mean no. larvae per stem (lcg(x+1)),

standard deviation and estimated number of samples

needed for i 25% confidence intervals based on results from

18 sampling dates fromthree sites in 1987 ......................................... 32
MANUSCRIPT 3

Table 1. The number of each bacterial food type per larval gut on a
seasonal basis .......................................................................................... 57

Table 2. The number of each algal food type per larval gut on a
seasonal basis .......................................................................................... 58

Table 3. The number of each bacterial and algal food type per larval
gut on a site basis-Sites 1 and 2, using guts from fourth instars ..... 61

Table 4. The number of each bacterial and algal food type per larval
gut on a site basis-Sites 2 and 3, using guts from third instars ....... 62

Table 5. The number of each bacterial food type per larval gut for first
through fourth instars collected on 28 July 1987 ............................... 65

Table 6. The number of each algal food type per larval gut for first
through fourth instars collected on 28 July 1987 ............................... 66

vi

LIST OF FIGURES
Page
MANUSCRIPT 1

Fig. 1. Proportion of larvae removed by each sequential suction
sample of Walker and Crans (1986) pump sampler .......................... 10

Fig. 2. Percent of larvae separated from sediments by Morris et al.
(1985) funnel-separator in three time intervals, and
total separated and not separated after 20 hours .............................. 11

MANUSCRIPT 2

Fig. 1. Seasonal occurrence and abundance of pupae and larval
instars at three sites (Sites 1, 2 and 3) in 1987 .................................. 26

Fig. 2. Seasonal occurrence and abundance of pupae and larval
instars at Site 4 in 1988 .......................................................................... 28

Fig. 3. Total larval abundance in 1987 (Sites 1, 2 and 3) and
1988 (Site 4) ............................................................................................. 30

Fig. 4. Regression of number of samples required to provide
1; 25% confidence interval for the average number of larvae
per stem at p>0.05 against the percent of sampling stations on
each sampling date whose samples contained larvae ..................... 33

Fig. 5. Parasitism rates of: A. Total larval population; B. Fourth instar;
and C. Third instars (Site 3 will): from Site 3 (1987) and
Site 4 (1988) ............................................................................................. 34

Fig. 6. Percent of total population at Site 3 (1987) which exhibited

cuticular injury as a sresult of penetration by mermithid
preparasites .............................................................................................. 35

vii

Page

MANUSCRIPT 3
Fig. 1 Total bacteria, detritus and algae in guts of fourth instar larvae

on a seasonal basis ................................................................................. 56
Fig. 2 Total bacteria and detritus in guts of: A. fourth instar larvae

from Sites 1 and 2 collected from March to June 1987;

and B. instar larvae from Sites 2 and 3 collected on 2 August

and 28 July 1987, respectively .............................................................. 60

Fig. 3 Percent of total bacteria, detritus and algae in the guts
of first through fourth instar larvae collected on July 28, 1987 ........ 64

Fig. 4 Ratio of number of bacteria to number of detrital particles in
the guts of fourth instar larvae and the surrounding sediments ...... 68

Fig. 5 Scanning electron micrograph of lateral palatal brush of fourth
instar larvae showing microtrichia ........................................................ 70

viii

PREFACE TO THESIS

mm mm is a medically important mosquito. Recent

evidence has begun to eludicate its role in the transmission of Eastern Equine
Encephalitis (EEE). The EEE vims was first isolated from wild caught adult
females in 1949 (Howitt et al. 1949). Western EE has also been isolated from
wild-caught female Cg. gems (Sekla et al. 1980). Females transmit the
EEE virus orally under laboratory conditions (Chamberlain 1957, P. R. Grimstad
and RS. Copeland, unpub. data). They have a population transmission rate of
7% to suckling mice (Boromisa et al. 1987). Clark et al. (1985) showed that
infection takes place only in the adults and is not transmitted transovarially.
Adult female Qq. 9.9m are known to feed on a wide variety of hosts
including birds, mammals, reptiles, and humans (Fallis and Wood 1957, Snow
and Pickard 1957, Beckel and Atwood 1959, Hayes 1961, Downe 1962, Edman
1971 ). Females also are multiple feeders, biting more than one host to secure a
full blood meal (Downe 1962). It has been suggested that their status as an
EEE vector is enhanced by these two behaviors (Lounibos and Escher 1983),
and that Cg. new acts as the vector connecting the enzootic cycle of EEE
in birds with the epizootic cycle in horses, as well as the epidemic cycle in
humans.

95;. mug is a very cosmopolitan mosquito, found throughout most
of the United States and Canada (Carpenter and LaCasse 1974, Darsie and
Ward 1981 ). The larvae of the genera W and the closely related

Mansgnia are unique among the Culicidae because they are able to obtain

ix

their oxygen from the roots, rhizomes and stems of submerged vegetation and
need not come to the surface for oxygen (Smith 1908, McNeel 1932, Laurence
1960, McDonald and Lu 1973).

Currently, control measures for Cg. Mg: are aimed primarily
toward the adult stages; control measures usually involve the widespread
application of a broad-spectrum insecticide (often malathion) using ultra-low
volume spray apparatus (Allan at al. 1981). New microbial pesticides,
specifically Bacillus W var. W (B.t.i.), as well as insect
growth regulators (e.g. Methoprene) are becoming increasingly available for
mosquito control progams (Cheong and Yap 1985, Sjogren et al. 1986, Walker
1987). These have the advantage of being much more specific and, therefore,
less environmentally hazardous than older classes of pesticides. However, they
must be used against the larval stages. Fundamental life-history and larval
feeding data necessary to design effective control programs using microbial
pesticides are not currently available for this species.

The objectives of this study were: 1) to locate larval habitats, and
evaluate sampling and sorting methods for use in life-history and feeding
behavior studies; 2) to examine the seasonal occurrence, abundance and
several mortality factors of Cg. mm larvae; 3) to evaluate the amount of
intrahabitat variation in larval abundance; 4) to determine the diet of larval Cg.
mm in their natural environment, and elucidate possible larval feeding

mechanisms.

Chapter 1

Evaluation of Methods for Sampling, Sorting and Locating

the Habitat of Larval nguulemdia perturbans
(Diptera:Culicidae)

2

ABSTRACT

This paper describes a method employing the US Fish and Wildlife
Service's National Wetlands Inventory (NWI) maps for locating the larval
habitats of ngufllenjgia W. In addition, the larval sampling method of
Walker and Crans (1986) and sorting method of Morris et al. (1985) were
assessed for their precision. NWI maps were found to be much more useful for
locating larval habitat than standard U.S.G.S. topographic maps. It appeared
that the water regime of wetlands, as noted on NWI maps, was a good indicator
of suitable habitat, in the central Michigan area. The pump sampler of Walker
and Crans (1986) sampled larvae repeatably, with 4 to 6 sequential suctions
required to remove 80 to 90% of attached larvae from cattail stems. The funnel
sorting method of Morris et al. (1985) repeatably separated 94% of larvae from

sediments in 20 hrs.

3

INTRODUCTION

nguifletfldia mm (Walker) has been implicated as the vector for

Eastern Equine Encephalitis in both the Midwest and non-coastal parts of the
eastern United States (Nasci and Edman 1981, Francy 1982, Boromisa et al.
1986). However, studies of its larval stages have been largely neglected owing
to the difficulties associated with locating, sampling and sorting this species
(Hagmann 1953, Bidlingmayer 1954, Morris et al. 1985, Walker and Crans
1986)

The habitat requirements for Cg. mm larvae are rather specific;
many researchers have sought to quantify them (Hagmann 1953, Bailey 1984,
Batzer and Sjogren 1986). In general, Cg. Wlawae require a habitat
that is permanently wet,.and the water must be relatively clear and stationary.
The breeding area must support a high density of healthy, rooted, vascular
water plants, and the bottom must be covered with a layer of soft silt (Brower
1953). Several methods have been proposed for sampling and sorting
the larvae of Cg. mum (Bidlingmayer 1954, Morris et al. 1985, Walker and
Crans 1986, Hagmann 1953). This paper describes a qualitative method used
to locate Cg. W larval habitats, and evaluates the effectiveness of the
pump-sampling method of Walker and Crans (1986) and the inverted-funnel
sorting method of Morris et al. (1985). The degree to which these two methods

are quantitative also was evaluated.

4

METHODS

W

In 1974, the US. Fish and Wildlife service began the National Wetlands
Inventory (NWI) program, the mission of which was to provide a database on the
current status of wetlands in the USA, so that future trends could be evaluated.
The initial goal of the project was to completely map the wetlands using a
classification system devised by Cowardin et al. (1979). Mapping was begun in
1978 and by 1987, 45 percent of the wetlands in the lower 48 states were
completed. These maps are readily available to the public through the US.
Fish and Wildlife Service's National Wetlands Inventory, the US. Geological
Survey's National Cartographic Information Center, and some state agencies.
The map legend used on NWI maps provides information regarding a wetland's
hydrologic association (e.g., isolated, lake-side, river-side etc), the life-form of
the dominant vegetation in a particular area (9.9., aquatic bed, emergent,
forested etc.), and its water regime; i.e., the duration of wetland flooding on a
yearly basis. Examples of water regimes are: seasonally flooded,
semipermanently flooded,.permanently flooded, and temporarily flooded.

During the summers of 1986 and 1987, 18 wetlands were surveyed in
the central Michigan area for the presence of Cg. mm larvae. Based on
larval habitat descriptions in the literature, only marshes or mapping units with
an emergent vegetation life form (EM) were surveyed. Two water regime
modifiers were most commonly associated with emergent vegetation in the
central Michigan area: semipermanently flooded (F) or seasonally flooded (C).
Photocopies of NWI maps were obtained from the US. Fish and Wildlife
Service field office in East Lansing, Michigan. These maps are the same scale

as U.S.G.S. topographic maps. Map quadrangles were chosen for proximity to

5

East Lansing and for containing the highest number of wetlands per unit area.
Mapping units were color-coded and the wetlands on the NWI maps colored to
match the coding which facilitated site selection for surveying. Approximately
equal numbers of each of the two wetland types were selected. In addition, the
selected wetlands were within 1.0 km of a road to facilitate sampling. Using
these criteria, 18 wetlands were selected. All wetlands were located in the Bath
quadrangle (42°45'-52'30"N. lat., 84°22'30"-30'W. long.). Ten of the 18 had
semipermanently flooded water regimes and tended to be dominated by cattails
(Inna spp.), while eight were seasonally flooded and contained a variety of
plant species including mm W, and secondarily, some Cam
spp. and Lama spp.

Sites were surveyed for the presence of larval Cg. gem using the
pumping and sieving method described in Walker and Crans (1986). A site was
sampled by two people for 30 minutes or until larvae were found. When larvae
were found, they tended to be collected within the first 10 minutes of sampling
so that 30 minutes appeared adequate for assessing larval presence or
absence. Mud from the base of a cattail stem or other submerged plant part
(e.g., Cam; mound) was pumped into a basin with a modified bilge pump
(Walker and Crans 1986). The mud was examined for larvae on site, both by
pouring small amounts of mud into white enamel pans and by pouring mud
through sieves. In both cases, if larvae were present, they were readily seen
moving in an agitated fashion against the dark mud background.
W

The larval sampling method devised by Walker and Crans (1986)
seemed to best suit the needs and limitations of this project. A "Thirsty Mate‘RI"
bilge pump that was modified to function as a syringe by removing the distal

one-way valve, was used to collect larvae. The distal end of the pump was

6

placed at the base of a cattail stem and the pump handle was pulled out quickly.
The mouth of the pump was then moved rapidly into a position over a basin
where the sample was allowed to drain. The suction from the pump was
sufficiently strong to remove Cg. W larvae from the stem being
sampled, yet it did not appear to be strong enough to remove larvae from
adjacent stems.

In order to use the pump method semicuantitatively, it was necessary to
determine how many suctions should be taken from a stem to remove the
majority of attached larvae. Two methods were performed at Site 3 (see Olds
and Merritt 1989a) during the summer of 1987. For the first method, six suctions
were collected from each of three randomly chosen cattail (11203 W
stems. The water/sediment mix from each suction was poured through a funnel
into a 4-liter plastic milk jug. Each of the six suctions from each stem remained
separate to determine the average number of larvae per sequential suction.
The second trial involved taking a greater number of suctions per stem (10) to
see if any additional larvae which may have been more tightly attached, were
dislodged by further pumping. In the second trial, suction numbers one through
four were consolidated into one sample as were suction numbers five and six.
This facilitated the hand carrying of samples 2.0 km to a car. Samples from both
trials were taken to the lab and hand-sorted within 2 hrs of collection.
W

Hand sorting of Cg. mm larvae from the water/sediment mix
obtained using the Walker and Crans (1986) pump method was very time
consuming, with inherent error, owing to sorter fatigue and experience.
Therefore separation of larvae with funnel separators described by Morris et al.
(1985) was attempted. In order to use these separators for quantitative counts

of larvae, the number or percent of the larvae that were not separated from the

7

mud in a 20-hr period (the time period suggested in Morris et al. [1985]), was
ascertained in the following means: A sample was placed in the bottom of a
PVC pipe cylinder, (25.5 cm in length X 7.5 cm in diameter). The PVC pipe was
sealed at the bottom and three screws were screwed into the cylinder 8 cm from
the top at positions equidistant around the circumference of the cylinder. A
plastic funnel, (5.5 cm high with a top diameter of 7.5 cm and a tip diameter of
1.2 cm) was inverted and then rested on the screws. The cylinder was filled
with tapwater to at least 2.5 cm over the tip of the inverted funnel. Any debris
that appeared to be blocking the tip was removed. The cylinder contents were
kept in total darkness by covering cylinders with aluminum foil. This was
important since we had found Cg. Mus larvae to be negatively
phototactic. When the oxygen became depleted in the bottom of the cylinder,
the larvae rose through the inverted funnel to the surface for air, trapping them
in the clear water at the top of the cylinder. The water from the top of the
cylinder could then be siphoned off using a turkey baster, and the larvae easily
enumerated.

Two tests of the cylinder separators were made. For each test, five cattail
stems were sampled by the pump method described above. Four suctions were
taken at each stem for a total 20-liter volume of sampled material. This material
was condensed down to a volume of approximately 10 liters by filtering off
excess water with a 250-um sieve. This sieve was sufficiently fine to insure
retention of all instars (Nemjo and Slaff 1984). Eleven separators were used in
each trial. In Trial 1, the sample-filled cylinders were placed in a 5° C cooler for
16 hours since we have found that larvae often died within 24 hours if kept at
21°C (room temperature). Larvae were allowed to rise into the water above the
inverted funnel. After 16 hours at 5° C, the water above the funnel was drawn

off and replaced by fresh tapwater. The separators were then moved to room

8

temperature (21°C). After two hours at room temperature, water was again
drawn off and replaced. After a total of four hours at room temperature the water
above the inverted funnel was drawn off for a third and final time. The total
number of larvae per water sample for each time period was counted. The
separators were then emptied and the number of larvae remaining in the bottom
was counted. In Trial 2, the separators were held at 5° C for 16 hours, as in
Trial 1, then moved to room temperature (21° C) for four hours. The water
above the funnel was then drawn off. The number of larvae that had risen to the
surface and those that were left in the bottom of the cylinder were counted. This
test was done to determine if the greater degree of disturbance in Trial 1 biased

the results.

RESULTS

BMW

Marshes containing larval Cg. mm were relatively easy to locate,
using NWI maps to identify potentially suitable habitats. Three often
semipermanently flooded marshes contained Cg. mm larvae, but no
larvae were found in any of the seasonally flooded marshes. When compared
with a Fisher's Exact Test for small numbers, this result was significant at the
90% level (X2(1,.1)=2.71). This indicates that semipermanently flooded
marshes are more likely to be suitable habitats for Cg. mum larvae than

are marshes with a seasonally flooded water regime.

P - I'

The results of the pump-sampling method for Cg. gem larvae are
given in Figure 1. The total number of larvae sampled from three stems in Trials
1 and 2 were 95 and 123 larvae, respectively. Both trials gave similar results
with the greatest proportion of larvae being collected in the first four suctions. A
few larvae continued to be collected in sequential suction numbers 5 and 6. In
the second trial, virtually no larvae were collected in suction number 7, and a
small number were collected in suction numbers 8 through 10.

W

The results of the first Morris Funnel trial are given in Figure 2. In both
trials, 94% of the larvae were recovered in the water above the inverted funnel
after 20 hours. Seventy-five percent of the larvae rose to the surface in the 16-
hour period at 5° C while 21% rose to the surface in four hours at room

temperature (21° C).

10

 

 

1.0-

u:

< . A

E 0.8-

<

_1

LL -

o 0-6

2 1

9. 0.4- ,,

I'-

u:

2 0.2- :3";

0 I

n:

9- o.o- - '
1 2 3 4 5 6
SEQUENTIAL SUCTION SAMPLE

1,0.

Lu .

< , B

E 0.3:

<

.r

u.

0

z

9

.—

r:

O

o.

O

n:

a

 

 

1 to 4 5 - 6 7 8 9 1 O
SEQUENTIAL SUCTION SAMPLE

Figure 1. Proportion of larvae removed by each sequential suction sample of
Walker and Crans (1986) pump sampler. A. Trial 1. Six total suctions. B. Trial
2. Ten total suctions.

100 94%
80
60
40

20

  

PERCENT OF LARVAE

6%

INT.1 INT. 2 INT.3 TOTAL m

FRACTION OF TIME

Figure 2. Percent of larvae separated from sediments by Morris et al. (1985)
funnel-separator in three time intervals (INT. 1: 16 hrs, 5° C, INT. 2: 2 hrs, 21°
C , INT. 3: 4 hrs, 21° C), and total separated and not separated after 20 hrs.

12

DISCUSSION

WWW

Advantages of using NWI maps to locate Cg. mm habitat include
increased resolution for wetlands over the US. Geologic Survey topographic
maps. U.S.G.S. topographic maps are most often used by mosquito control
districts to locate marshes (E.D. Walker, C. Chilcote pers. comm). NWI maps
usually include wetlands as small as 0.25 acres, which are much smaller than
those on U.S.G.S. topographic maps. In addition, wetlands on NWI maps are
classified by dominant vegetation type so that marshes, shrub wetlands and
forested swamps can be easily differentiated, and only the relevant wetland
type surveyed. In the central Michigan area, there appears to be some
relationship between the water regime of a wetland, as presented on NWI
maps, and its suitability as Cg. Mums habitat. Marshes with a
semipermanently flooded water regime were somewhat more likely to contain
Cg. 91mm larvae than marshes with seasonally flooded water regimes.
More survey work using NWI maps from different sections of the United States
will be required to determine whether this result can be generalized to all
potential wetland habitats containing Cg. WJarvae.

lin

The pump sampling method of Walker and Crans (1986) was both
quantitative and simple to use. It is a very appropriate method for surveying or
sampling Cg. mm larvae in the central Michigan area where habitats
tend to be small, shallow (rarely deeper than one meter), and difficult to reach.
This research demonstrated that four to six suctions per stem were adequate to
provide a repeatable estimate of the number of larvae attached to an individual

cattail stem. Increasing the number of suctions beyond six seems to increase

13

the probability that larvae will be drawn from the surrounding areas, rather than
from the stem being sampled.

Lounibos and Escher (1983) have demonstrated that larvae are rather
loosely attached to cattails and when disturbed, readily release. It is therefore
unlikely that larvae collected in suctions eight through ten were larvae from the
stern being sampled. These probably were larvae that had detached from
neighboring stems and then transported into the region of the stem being
sampled when increasing amounts of water were removed. It is also

improbable that larvae would swim between cattails in the short sampling time,

as Cg. genumans larvae are very poor swimmers.
51-” 'E IIIIII

The separator funnels described by Morris et al. (1985) were a very time-
efficient method of sorting Cg. mm; larvae. To hand sort a four-liter pump
sample from one stem took a minimum of two hours when done by experienced
sorters. In addition, an unknown number of larvae were lost by the hand-sorting
method, owing to sorter fatigue. Using the funnel sorters, the processing time
per stem was reduced to approximately 45 minutes and a repeatable 94% of

the larvae in the samples were recovered.

1 4
LITERATURE CITED

Bailey, D.L. 1984. Comparison of diurnal and nocturnal Magsgma larval
populations on water lettuce plants. Mosq. News 44:548-552.

Batzer, D. P. “and R. D. Sjogren. 1986. Larval habitatcharacteristics of
gems (Diptera: Culicidae) in Minnesota. Can. Ent.
118:1193- 1198.

Bidlingmayer, W.L. 1954. Description of a trap for Mansonia larvae. Mosq. News
14:55-58.

Boromisa, R. D, R. S. Copeland and P. R. Grimstad.1987. Oral transmission of
Eastern Equine Encephalomyelitis virus by a northern Indiana strain of

WW. J. Am. Mosq. Control Assoc 3: 102- 104.

Brower, L. P. 1953. The distribution of M Mn; (Walker) in Morris
County. Proc. N.J. Mosq. Exterm. Assoc., 45th Annu. Meeting. 147-149.

Cowardin, C.M., V. Carter, F.C. Golet and ET. LaRoe. 1979. Classification of
wetlands and deepwater habitats of the United States. US Fish and
Wildlife Service, Washington DC. 131 pp.

Francy, DB. 1982. Eastern equine encephalomyelitis (EEE) and consideration
of some alternatives for dealing with EEE as a public health problem. pp.
15-18. m: J.G. Engemann (Ed.) Eastern equine encephalomyelitis (EEE)
and public health in southwestern Michigan. Sci. for Citizens Ctr. of SW
Mich., Western Michigan University.

Hagmann, LE. 1953. Biology of ”Mansonia perturbans' (Walker). Proc. N.J.
Mosq. Exterm. Assoc., 40th Annu. Meet. 141-147.

Lounibos, L. P. and R. L. Escher 1983. Seasonality and sampling of
W (Diptera: Culicidae) in south Florida.
Environ. Entomol.12:1087-1093.

Morris, C.D., J.L. Callahan and RH. Lewis. 1985. Devices for sampling and
sorting immature Coquillettidia perturbans. J. Am. Mosq. Control Assoc.
1:247-250.

Nasci, RS. and JD. Edman. 1981. Blood-feeding patterns of Culjseja
mglangra (Diptera:Culicidae) and associated sylvan mosquitoes in
southeastern Massachusetts eastern equine encephalitis enzootic fcci. J.
Med. Entomol. 18:493-500.

Nemjo, J. and M. Slaff 1984. Head capsule width as a tool for instar and species

identification of Mansmiaslxafl Manamiatitillans and
mmmiptera: Culicidae). Ann. Entomol. Soc. Am. 77: 633-635.

15

Walker, ED. and W.J. Crans 1986. A simple method for sampling W
perturbans larvae. J. Am. Mosq. Control Assoc. 2:239-240.

Chapter 2

The Life-History and Seasonal Abundance of

larval WW
(Diptera:Culicidae) in South-Central Michigan

16

17

ABSTRACT

The occurrence and abundance of larval nggjugnjgia W was
measured at four sites over two years, 1987 and 1988. First and second instars
were present from mid-July to September (1987) and from late June to October
(1988). Third and fourth instars were present for the majority of the year
including the winter months. Larval populations were greatest in late August or
September, with number of larvae per stem ranging from 42.4 to 7.7 in 1987
and 1988, respectively. Parasitism rates for a previously undiscovered
mermithid nematode parasite were calculated at two sites. The maximum rates
of parasitism for all larval instars were 21 % and 9% for 1987 and 1988,
respectively. The rates of parasitism for fourth instars alone (64% in 1987, 67%

in 1988) was much higher and less variable between years.

18

INTRODUCTION

mm mm is found throughout most of the United States
and Canada (Carpenter and LaCasse 1974, Darsie and Ward 1981). Larvae

belonging to the genera W and the closely related Mansonia, are
unique among the Culicidae in that they are able to obtain oxygen from roots,
rhizomes and stems of submerged vegetation and thus need not come to the
surface for oxygen (Smith 1908, McNeel 1932, Laurence 1960, McDonald and
Lu 1973).

Cg. W has long been known as an annoying biter (Headlee
1945), but recent evidence has begun to elucidate its role in the transmission of
Eastern Equine Encephalitis (EEE). Adult female Cg. Was are known to
feed on a wide variety of hosts including birds, mammals, reptiles, and humans
(Fallis and Wood 1957, Snow and Pickard 1957, Beckel and Atwood 1959,
Hayes 1961, Downes 1962, Edman 1971). Females also are multiple feeders,
biting more than one host to secure a full blood meal (Downe 1962). It has
been suggested that their status as an EEE vector is enhanced by these two
behaviors (Lounibos and Escher 1983), and that Cg. 0200:0305 acts as the
vector, connecting the enzootic cycle of EEE in birds with the epizootic cycle in
horses, as well as the epidemic cycle in humans.

Currently, control measures for Cg. Mus are aimed primarily
toward the adult stages; control measures usually involve the widespread
application of a broad-spectrum insecticides (e.g., malathion) using ultra-low
volume spray apparati (Allan at al. 1981). New microbial pesticides, specifically
Cagmus W var. W (B.t.i.), as well as insect growth regulators
(e.g. Methoprene) are becoming Increasingly available for mosquito control

programs (Cheong and Yap 1985, Sjogren et al. 1986, Walker 1987). These

19

have the advantage of being much more specific and, therefore, less
environmentally hazardous than older classes of pesticides. However, they
must be used against the larval stages.

The objectives of this study were to : I) examine the seasonal occurrence
and abundance of Cg. mm larvae at four sites over a two year period; 2)
evaluate the amount of intrahabitat variation in larval abundance; and 3)
investigate the impact on larval populations of two mortality factors, drought and

a previously undiscovered mermithid parasite.

METHODS

Four sites were chosen for a detailed study of the seasonal occurrence
and abundance of Cg. gem larvae, based on a previous survey of
breeding sites (Olds and Merritt 1989a). In March 1987 a severe drought
occurred In central Michigan, and as a result, only two sites were sampled in the
spring of 1987, a third site was sampled from June 1987 to December 1987 and
a fourth site was chosen for all work conducted in the summer of 1988.

SI 0 . I'

Site 1. Drumheller Pond (42° 51' 50"N, 84° 26' 30"W). This pond was
located on the north side of Drumheller Rd. in Bath Township, Michigan, .8 km
east of Webster Rd. The marsh was 0.5 acres with cattails (11gb: latjjglja)
covering 0.21 acres. The cattails were interspersed throughout the marsh in
four main stands with open water areas in between.

Site 2. Inland Lakes Research Center (42° 40' 55"N, 84° 28' 50"W).
This natural marsh was located immediately to the west of College Rd. at the
most southern edge of Michigan State University's Inland Lakes Research

Center. The marsh was 1.3 acres in area with cattails forming a one to three

20

meter wide ring around the perimeter. The total cattail coverage was .34 acres
representing 27 percent of the total pond area.

Site 3. Gibbs Pond (42° 51' 50"N, 84° 26' 30"W). This pond was
located 1.0-km north of Drumheller Rd. and 0.6 km east of Webster Rd., in Bath
Township, Michigan. The marsh was 1.1 acres in area with a total cattail cover
of 0.94 acres. Sampling was confined to live cattail stands. Approximately 0.70
acres of the cattails were dead for an unknown reason. The living cattails were
located in a U-shaped pattern around an open water area. The average cattail
density was 14.7 stems rn'2 (S.D.=4.9). In 1988, there were no living cattails in
this pond. Possible explanations for the mortality of all cattails included muskrat
damage, a possible herbicide influx, a cattail blight or drought stress from the
previous summer.

Site 4. Maple River (43° 08' 10"N, 84° 33' 30”W). The section of marsh
sampled in the study was a small, 0.25 acre parcel on the Maple River State
Game Lands. It was located 10 km north of St. Johns, Michigan on Rt. 27 North.
The section of the marsh used in this study was located on the northern side of
the flooding, approximately 0.5 km from Rt. 27 North. This site was added in the

summer of 1988 when it became apparent that the three sites used in 1987 no

longer supported populations of Cg. genumans larvae.
E I I' S 1'

Sites were measured and mapped using a line and compass technique.
Total marsh area and vegetational cover was determined by cutting and
weighing constructed maps. Ten permanent stations were located at each site,
and stations were selected as uniformly as possible throughout the cattail
stands. If the cattails ringed the entire perimeter of a marsh, stations were
placed at regular intervals around the edge. Permanent stations were marked

by placing 2 m. long iron rod station markers into the sediments.

21

Random sampling was accomplished using a modified point-quarter
method of Mueller-Dombois and Ellenberg (1974). The section of the marsh
around each station was divided up into four quadrants using a compass. The
quadrants were located northeast, southeast, northwest and southwest of each
station marker. Weekly, biweekly or monthly samples were taken depending on
the season. For each sampling date, one quadrant was chosen beginning with
the northeast quadrant and moving counterclockwise with each succeeding
sampling date. Initially, the closest cattail to the station marker was sampled in
the chosen quadrant. Ten replicate samples, representing one sample per
station, were taken each sampling date. Over four sampling dates one cattail
stem in each quadrant of each station was sampled. Cattails were marked with
flagging after being sampled to exclude them from being resampled. Once one
cattail in each quadrant was sampled, the sampling cycle was reinitiated
beginning with the second closest stem in the northeast quadrant of each the
station marker. This sampling pattern was continued throughout the season.

Cattail stems were sampled using the method of Walker and Crans
(1986). Four suctions of the modified bilge pump were taken per stem. Material
was collected in a floating basin, and then poured into four-liter jugs for
transporting to the lab. Samples were hand sorted in the lab using a series of
Nalgene‘R) filters of decreasing mesh size (4 mm, 1 mm, and 0.25 mm mesh).
All plant debris in the samples was rinsed thoroughly to remove any entangled
larvae.

Samples were collected every two weeks at Site 1 and Site 2 from March
25 through June 30, 1987. By July, both ponds were almost dry, and could no
longer be sampled. Sampling at Site 3 began in June 1987. Weekly samples
were taken from June through September and biweekly samples were taken in

October to December. No samples were taken during ice-cover in January and

22

February. Sampling began in March 1988 but was discontinued after no larvae
were found for three sampling dates (March to May). Periodic checks during the
summer of 1988 showed that no larvae remained at this site.

Sampling at Site 4 began in May 1988, and only presence/absence data
were collected until July. In July, 1988 the funnel sorting technique of Morris et
al. (1985) was used to separate larvae from sediments for enumeration. Larvae
were counted from three buckets, each of which contained material sampled
from six stems at two suctions per stem.

W

To determine whether within-habitat larval abundance was related to any
obvious habitat features, the following measurements were taken at each cattail
sampled on each sampling date: 1) water temperature at a depth often cm; 2)
cattail stem diameter; 3) physiological state of the stern (i.e., living or dead); 4)
location of the stem (e.g., in a clump of stems or isolated [greater than 0.3 m
from any other stemII: and 5) proximity of the stem to open water. Water depth
was measured at all sampled stems and at all station markers. Cattail-stem
density was determined at Site 3 in October 1987 by averaging the number of
cattails per plot in ten 1-m X1-m plots.

An experiment was conducted on 20 August 1987 to determine if larval
numbers were correlated with cattail root volume. Lounibos and Escher (1985)
had previously found a positive correlation with the dry weight of the cattail root
mass and larval numbers. Ten cattail stems were randomly selected by the
modified point quarter system described above. However, in this case, two
stems each were taken from stations 1 to 5. Four suctions per stem were taken
using the method of Walker and Crans (1986). Samples were hand sorted for
larvae immediately after being returned to lab. Sampled stems were gently

removed from the mud and care was taken to collect all of the root mass of each

23

stem. The rhizome which tended to anchor the stem was broken at least 0.3 m
from the stem and the stems were labeled and transported to the lab for
processing. Root volume was measured by measuring the amount of water that
was displaced by submersing the stem in a known water volume.
11 _! I'll}! 'I' 'EII

A specific degree of accuracy of Cg. mm larval population
estimates was determined by using the sampling data from 18 sampling dates
from Sites 1, 2, and 3 and selecting a desired confidence interval. The number
of samples taken per sampling date was either five or ten using the method of
Walker and Crans (1986). The data on larvae per stem were transformed using
a log (x+1) transformation to normalize the data (Steele and Torrie 1980,
Southwood 1978). The mean and standard deviation were determined for each
date. The chosen confidence interval was i 25% of the mean, and the number
of samples needed to provide this level of accuracy was calculated according to
Steele and Torrie (1980).
Ram

In fall 1987, a mermithid nematode parasite was observed in the thorax
of a fourth instar Cg. Was larva. All third and fourth stage larvae collected
from Site 3 in 1987 and Site 4 in 1988 were examined for this parasite using a
dissecting scope at 120x. A record was kept of what appeared to be damage
(i.e., black spirals in the cuticle of the larvae) caused by the ore-parasites as
they entered the larvae (Petersen 1985).

In September 1988, a total of 11 parasitized larvae were collected from
Site 4 over a 3-week period. They were placed in 500-ml plastic containers
with water, a sand substrate and cattail roots to which the larvae could attach.
Larvae were kept in growth chambers at 12° C. Water was changed once per

week. Larvae were checked every two days for parasite emergence. Upon

24

emergence from the larvae, the post-parasitic juveniles were held in containers

with 500 ml water (20° C) and sand to see if they would molt to adults.

RESULTS

E I I' D .

Site 1 and Site 2.-- In spring 1987, both third and fourth instars were
found at Sites 1 and 2 (see Figure 1). This indicated that both instars were able
to overwinter in central Michigan. At Site 1, third and fourth instars were present
until June 23 when the marsh dried. Pupae were first observed at Site 1 on
May 28 and were present until the pond dried in late June. Owing to the
drought conditions, no further population sampling was possible after June 23,
1987. A study was conducted to determine whether larvae burrowed into the
mud to escape the drought, as some authors have suggested (Hagmann 1953).
However, no larvae were found, even after sorting 40 liters of mud from an area
of Site 1 known to have had very high larval densities (27 larvae per stem). At
that time, the pond still held five cm of water in the areas sampled.

Site 3.-- Beginning on 25 June 1987 populations continued to be
sampled at Site 3 (see Figure 1). Pupae were present from 9 July to 28 July;
first instar larvae were present from 14 July to 25 August; second instars from 14
July to 1 September, and third instars from 25 June to 20 November. Fourth
instars were present throughout the sampling period; however, owing to the
drought and high temperatures in 1987, larval numbers became very low in
winter samples and were below detectable levels in the spring 1988 samples.

Site 4.-- An additional site (Site ) was selected in May 1988. The pattern
of larval development at this site was similar to previous sites (See Figure 2).

Third and fourth instars were present in the spring samples, again indicating

25

they had overwintered in these stages. Pupae were present from 26 May to 28
June. First instars were present from 24 June to 12 August, second instars from
24 June through 13 September and third and fourth instars throughout the
sampling period from 26 May until sampling was terminated in November 1988.
LamaLAbundanQe

Larval abundance in Site 1 and Site 2 for the spring of 1987 is shown in
Figure 3. Site 1 was much more productive than Site 2 during the three-month
sampling period (April to June 1987). Site 1 contained anaverage of 11.4,
larvae per stem while Site 2 contained an average of 0.4 larvae per stem.

The seasonal pattern of larval abundance for Site 3 is shown in Figure 3.
Larval numbers peaked in mid-August at 42.4 larvae per stern but fell to 0.2
larvae per stem by December. Average total larvae for Site 3 was 114,000
larvae. This number was calculated using only the area of the living cattails.
The data obtained from Site 4 in 1988 is probably more characteristic of larval
habitats In non-drought years (see Figure 3). Larval numbers peaked in mid-
September at 7.7 larvae per stem but only tell to 6.0 larvae per stem by
November 1988. Although 1988 was the worst drought in the Midwest in more
than 30 years, it did not affect water levels at Site 4, since they were artificially
maintained.

III'IIII'II C III 151 I

Habitat variables were related to larval abundance by determining
correlation coefficients within each sampling date, with stations being
observations for each variable. The following variables showed no or few
correlations with larval numbers: stem diameter, root mass volume and stem
location, both center-edge and isolated-clumped. At Site 3 the stations with the
deepest water tended to have the greatest number of larvae, but no correlations

existed between water depth and larval number.

26

Figure 1. Seasonal occurrence and abundance of pupae and larval instars at
three sites (Sites 1, 2 and 3) in 1987.

27

43210

m<m=m

DEC.

OCT.

JUNE AUG.

APRl

 

-u- SITE1
+ SITE 2
+ SITE 3

 

 

w
R
P
A

F 552. .. N mfimz. a 552.
Suhm Ema m<>m<._

 

v m<.-.mz.

DATE

28

Figure 2. Seasonal occurrence and abundance of pupae and larval instars at
Site 4 in 1988. P: present (Sampled only for presence/absence on May 26,
June 24 and June 28).

29

D C.
DEC.

V

AUG.
AUG.

 

P——P
JUNE '
JUNE

 

 

0.10“

 

APRL '
APRL

 

.05 "‘
0.00

c .8
1 0

m< z.
F mfimz. «552. a 5

0.0

m<n5m

Suhm mun. m<>m<.._

 

DATE

 

.1 4 ‘1 ‘ ‘ q
6 4 2

v m<._.mz_

H.

R

P
.A
0

30

 
   
 
   

 

so -
2 4o -
It!
‘0 30 _ -=I- SITE 1
E + SITE 2
o. ‘ -I- SITE 3
E 20 - + SITE 4 (1988)
E
< 10 1
A I

o I _. _. -

APRIL JUNE AUG. OCT. DEC.

DATE

Figure 3. Total larval abundance in 1987 (Sites 1, 2 and 3) and 1988 (Site 4).

3 1
IIII'IIV'I"I IE! I
The results of the estimated number of samples required to provide an
degree of accuracy for larval abundance at P>F= 0.05, (with Cl. equal to $2570
of the mean) are given in Table 1. Estimates of numbers of samples required

ranged from 4.7 to 184.2 with a mean of 53.3. Although this suggests a wide

variation in larval numbers, it is mainly attributable to the large number of

Earl

samples that lacked larvae. When the estimated number of samples is plotted

against the proportion of station samples with larvae on each of the 18 sampling

[rm _

dates (Figure 4), it is clear that many more samples are required when most of
the samples lack larvae.
13.3mm

Mermithid nematode parasites were first observed on 14 July 1987 in
fourth instar larvae from Site 3 (Figure 5b). Parasites were also found in third
instars beginning on July 28 (Figure 5c). In addition, cuticular injury suspected
to be caused by pro-parasites was first noted on 14 July and 21 July for fourth
and third instars, respectively (Figure 6). This injury appeared to fit the
description of encapsulation and melanization of pro-pupae described by
Petersen (1985). The highest parasitism rates were recorded in mid-August .
They were 64% for fourth instars (Figure 5b) and 19% (Figure SC) for third
instars; the parasitism rate for the entire population was 21% (Figure 5a). In
early September, there was an abrupt decrease in the parasitism rate,
especially in fourth instars. The parasitism rate fell to zero in all instars after
October 8th (Figure 5a). Parasites were never observed in first instars and were
very rarely found in second instars. Pre-parasite injury in second instars was

rather common, with a maximum of 45% parasitized on 14 July 1987.

32

Table 1. Number of samples, mean number of larvae per stem(log(x+1)),
standard deviation and estimated number of samples required for 125%
confidence intervals (P>t=0.05) based on results from 18 sampling dates from

three sites in 1987.

 

 

No. Samples Mean Standard Estimated Number of
Taken Deviation Samples Required for
:1; 25% or. (P>t=0.05) l
l
10 0.10 0.22 98 ~
10 0.17 0.26 46
10 0.03 0.09 184
10 0.08 0.16 87
10 0.76 0.48 8.0
10 0.52 0.51 20
10 0.07 0.21 184
10 0.95 0.80 15
10 0.41 0.37 17
10 0.33 0.44 36
10 0.43 0.45 23
5 1.35 0.58 5.8
5 1.12 0.45 5.0
5 1.13 0.44 4.7
5 0.78 0.47 11
5 0.06 0.12 123
5 0.44 0.53 44
5 0.22 0.27 47

 

33

200‘

_ y=140.3136-1.6151X R=0.85

NUMBER OF SAMPLES
REQUIRED

 

 

 

0 20 40 60 80 100
PERCENT OF STATIONS WITH LARVAE

Figure 4. Regression of number of samples required to provide 125%
confidence interval for the average number of larvae per stem at P>t=0.05
against the percent of sampling stations on each sampling date whose samples
contained larvae.

34

 
  
 
  

 

 

 

E 3° ‘ TOTAL POPULATION A
t .
‘2 20 -
5 . 4} SITE 3-1987
1:. -- SITE 4—1988
5 10 a
I.“
O
u:
Ill (1 ,
0- JUNE
3 100 -
u . FOURTH INSTAR B
5, 80 ~
< i
E 60 1
a . 0 SITE 3- 1987
l;- 40 ~ + SITE 4- 1988
3 i
n: 20 1
LL! 4
D. 0 -

JU

DATE

3 3° ‘ THIRD INSTAR c
t d
0)
< 20 ‘
C
E -a- SITE 3-1987
5 10 «
Ill
0 .
C
E o -

JUNE AUG. OCT. DEC.

DATE

Figure 5. Parasitism rates of: A. Total larval population; B. Fourth instars; and C.
Third instars (Site 3 only); from Site 3 (1987) and Site 4 (1988).

35

   

 

 

E 5° : TOTAL POPULATION

2 4o -

E

if: 30 -

g . ~o SITE 3- 1987

E 20 -

1.u .

8

m 10 -

a 1
o d a a n -
JUNE Aug, OCT. DEC.

Figure 6. Percent of total larval population at Site 3 (1987) which exhibited
cuticular injury as a result of penetration by mermithid pro-parasites.

36

In 1988, parasitized larvae were found in samples from Site 4 (Figure
5a). The dates of parasitism are similar to Site 3 in 1987, except one
parasitized larva was found on 3 November 1988. The highest parasitism rate
of fourth instars was also comparable (67%); however, the total population
parasitism rate was lower at Site 4 (9% compared to 21.7% for Site 3) (Figure
5a). This was attributed
to two factors: the lower parasitism rate of third instars, coupled with the greater
proportion of third instars over fourth instars for any given date.

Survival of parasitized larvae under laboratory conditions, until parasite
emergence, was very good. A total of five parasites emerged from the larvae.
In all cases, the larva from which the parasite had emerged was dead within 48
hours after the parasite's emergence. A single observation of parasite
emergence revealed that the parasite did not emerge from the thorax as had
been previously described by Petersen (1985), but appeared to emerge from
the distal end of the abdomen, possibly from the breathing tube. Emergence
occurred very quickly (less than 60 secs). The larvae from which parasites had
emerged were examined for emergence damage to the cuticle in either the
thorax or abdomen. No emergence holes were found at either location. This
suggested that the point of emergence was, indeed, the breathing tube.

Post-parasitic juveniles were observed for possible development to
adulthood. In two cases, definite changes were seen in the grainy appearance
of the nematode. The development of what might have been a vulva, as well
as, elongated projections on the anterior and posterior ends of the nematodes

were observed.

37

DISCUSSION

l ' mi

Qq. W had a univoltine life cycle in both ponds examined
throughout the year. These findings are in agreement with others who have
worked on fig. mm in mid-latitude regions (Brower 1953, Hagmann
1953, Carpenter and LaCasse 1974, Bidlingmayer 1968, Lewis and Bennett
1980, Allan at al. 1981). Pupae were present in Michigan ponds several weeks
prior to the times described by Lewis and Bennett (1980) and Allan et al. (1981)
as the onset of adult emergence in southwestern Ontario and the Nova Scotla-
New Brunswick area, respectively. This period probably corresponds to the
pupal to adult developmental time. The number of weeks pupae were present
in ponds in Michigan was also similar to the number of weeks adult emergence
occurred in southern Canada (Lewis and Bennett 1980, Allan at al. 1981). First
and second instars were present for a longer period of time, both earlier and
later in the season, than reported by Smith (1908) and Hagmann (1953) in New
Jersey. The longer season herein reported may have more to do with improved
sampling techniques and shorter sampling intervals than with any real
differences in seasonal pattern.

h in

Although the sampling of flu. mm larvae on a per-stem basis is
logical, as larvae always occur attached to macrophytes, there are certain
difficulties associated with this approach. Sampling of more than 10 stems in a
habitat is logistically difficult, and unless a great deal of assistance is available,
this method would be best used for hypothesis testing or for comparisons in
habitats with large, relatively uniform populations of 95;. mm larvae. An

additional modification to ensure presence of larvae in all, or most samples,

 

38

would be to pool material from two to five stems in one sample, ensuring all
samples would contain at least some larvae. This procedure would reduce the
number of samples required for estimating population densities with reasonable
confidence intervals.

W

The potential for using mermithid parasites of mosquito larvae for
biological control agents has been demonstrated repeatedly (Petersen and
Willis 1972, Petersen and Willis 1974a, Petersen and Willis 1974b, Petersen
1975, Nickle 1979). Although the mermithid parasite of Cg. W found in
this study does not appear to control the mosquito populations, it may be
influencing the voltinism of the population. In both 1987 and 1988, the group of
mosquito larvae with the highest parasitism rates and therefore highest mortality
rates was early maturing fourth instars. This group, in years of warm autumn
temperatures, could conceivably produce a second generation of adults (Allan
at al. 1981, Goshenko 1978). In this study, seasonal temperatures were
abnormally high; yet, no second pulse of pupae were detected. Mortality
caused by the mermithid parasites could have been a factor responsible for this
occurrence.

The mermithid nematode which has been used most commonly in
biocontrol programs is W W. B. W appears to be
limited by low dissolved oxygen (Brown and Platzer 1978b, Platzer 1981), high
ions or organics (Chen 1976 but see Brown and Platzer 1978a, Petersen 1979),
and polluted conditions (Mitchell et al. 1974). The habitat in which 951.
W larvae live is characterized precisely by these attributes (Batzer and
Sjogren 1986, Callahan and Morris 1987), suggesting that the mermithid
parasite of fig. W is well adapted to these conditions. If this is correct,

39

biocontrol programs using this parasite might be possible under conditions
unsuitable for B. guligjygrax.
E I' I' I II . r I

This study showed that the best time to survey 9g. gem larval
habitats is in the autumn. At that time, larval populations tend to be high and the
two larval stages present (third and fourth instars) are more visible. If a site
becomes dry or near desiccation, it can be excluded. Also, the modified bilge
pump method of sampling (Walker and Crans 1986)functions best in shallower
waters, which occur in autumn.

Interest exists in modeling the development and emergence of fig.
mm. These phenological events could be used to increase efficacy of
spray regimes (Slaff 1986). Since the abundance of fig. mm adults may
be related to outbreaks of EEE in horses and humans (Boromisa et al. 1987), it
would also be useful to model the abundance of Qg. Mus based on some
easily measured environmental parameter. Capotosto and Boyes (1984) found
a good correlation between the abundance of Cg. mum adults and mean
temperatures of the preceding winter. However, work by Rademacher (1979)
suggests that larvae are not killed by low winter temperatures. Based on the
larval survey and the population data collected in 1987 and 1988 presented in
this thesis, it appears that drought stress is an extremely important source of
mortality. Over the two and a half year period of this study, four sites containing
low to very high populations of Qg. Mn; larvae had their entire population
destroyed owing to pond desiccation. This undoubtedly resulted in a large
decrease in adult Qg. Mus abundance. Therefore, it is probable that
population models integrating both low temperatures and drought stress (as
measured by rainfall) would be much more accurate in predicting adult

abundance than would low temperatures data alone.

40

LITERATURE CITED

Allan, S..,A G. A. Surgeoner, B. V. Helson and D. H. Pengelly. 1981. Seasonal
activity of Manggnja mumps adults (Diptera: Culicidae) in
southwestern Ontario. Can. Entomol. 113: 133-139.

Batzer, D. P. and R. D. Sjogren. 1986. Larval habitat characteristics of
Qggujflgmdja (Diptera: Culicidae) in Minnesota. Can. Ent. 118: 1193-
1198.

Beckel, W.E. and H.L. Atwood. 1959. A contribution to the bionomics of the
mosquitoes of Algonquin Park. Can. J. Zool. 37:763-770.

Bidlingmayer, W.L. 1954. Description of a trap for Magsgnia larvae. Mosq.
News 14:55-58.

Bidlingmayer, W.L. 1968. Larval development of Manggnjg mosquitoes in
central Florida. Mosq. News 28:51-57.

Brower, L. P. 1953. The distribution of Mansgnia ggguflzans (Walker) in
Morris County. Proc. N.J. Mosq. Exterm. Assoc., 45th Annu. Meeting.
147-149.

Brown, B.J. and E.G. Platzer. 1978a. Salts and the infectivity of
W guljgixgmx, J. Nematol. 10:53-64.

Brown, B..J and E. G. Platzer. 1978b. Oxygen and the infectivity of
Bgmangmflmjscuflgimx. J. Nematol. 10:-110113.

Callahan, J.L. and CD. Morris. 1987. Habitat characteristics of Qgggmgmgja
mm in central Florida. J. Am. Mosq. Control Assoc. 2:176-180.

Capotosto, PM. and D. Boyes. 1984. Climatic effects on ' '
W in Barrington, Rhode Island. Mosquito News. 44:244-246.

Carpenter, S. J. and W.J. LaCasse. 1974. Mosquitoes of North America (North
of Mexico). Univ. of California Press, Berkeley. 360 pp.

Chen, PS. 1976. A study of Reesimermis nielseni for control of gum gigjgns
fatigans in Taiwan. Bull. Inst. Zool. Acad. Sinica. 15:21-28.

Cheong, WC and H.H. Yap. 1985. Bioassays of Bacillus sgnagn'gus (strain
1593) against mosquitoes of public health importance in Malaysia.
Southeast Asian J. Trop. Med. Pub. Hlth. 16:54-58.

Darsie, R.F.,Jr. and FLA. Ward. 1981. Identification and geographical
distribution of the mosquitoes of North America, north of Mexico.
Mosq. Syst. Suppl. 1:313pp.

7]

(TI—F" .

41

Downe, A.E.R. 1962. Some aspects of host selection by MW mug.
Can. J. Zool. 40:725-732.

Edman, J. D. 1971. Host-feeding patterns of Florida mosquitoes. I. Agdgg,
Anonhalas. Coquillettidia Mansonia. andEscanLa J Med Entomol.
8:687-695.

Fallis, AM. and 0M. Wood. 1957. Biting midges (Diptera: Ceratopogonidae)
intermediate hosts Haemoproteus of ducks. Can. J. Zool. 35:425-435.

Goshenko, V. A. 1978. Biotopes and times of development on Mansgnjg
figmamjj (Ficalbi) 1889 in the conditions of the Ukrainian steppes.
Medskaya Parazitol. (Moscow) 47. 36- 40. (in Russian)

Hagmann, LE. 1953. Biology of "Managua M" (Walker). Proc. N.J.
Mosq. Exterm. Assoc., 40th Annu. Meet. 141-147.

Hayes, RC. 1961. Host preferences of Culiseta melanura and allied
mosquitoes. Mosquito News 21:179-187.

Headlee, T.J. 1945. The Mosquitoes of New Jersey and Their Control. Rutgers
University Press, New Brunswick. 326 pp.

Laurence, BR. 1960. The biology of two species of mosquitoes, Mgnsgnjg
magma (Theobald) and Mansgnja gnjjgcmjs (Theobald), belonging to
the subgenus Mansonioides (Diptera: Culicidae). Bull. Entomol. Res.
51 :491-517.

Lewis, DJ. and GP Bennett 1980. Observations on the biology of Mansonia
Mum (Walker) (Diptera:_ Culicidae) in the Nova Scotia-New
Brunswick border region. Can. J. Zool. 58:2084-2088.

Lounibos, L. P. and R. L. Escher 1983. Seasonality and sampling of
Qggujflgnjgja (Diptera: Culicidae) in south Florida. Environ.
Entomol. 12:1087-1093.

McDonald, J.L. and LC. Lu 1973. Preference of mm m: (Theob.)
for specific water hyacinth plants Mosq. News 33:466-467.

McNeel, TE. 1932. Observations on the biology of Mansgnjg mm
(Walker) Diptera, Culicidae. Proc. N.J. Mosq. Exterm. Assoc., 19th Annu.
Meet, 125-128.

Mitchell, C.J., P. Chen and HG Chapman. 1974. Exploratory trials utilizing a
mermithid nematode as a control agent for Qujgx mosquitoes in
Taiwan. J. Formosan Med. Assoc. 73:241-254.

42

Morris, C.D., J.L. Callahan and RH. Lewis. 1985. Devices for sampling and

sorting immature Qggulllgmdla perturbans. J. Am. Mosq. Control
Assoc. 1:247-250.

Mueller-Dombois, D., and H. Ellenberg. 1974. Aims and Methods of
Vegetation Ecology. John Wiley and Sons, New York.

Nickle, W.R. 1979. Probable establishment and overwintering of a mermithid
nematode parasite of mosquitoes in Maryland. Proc. Helminthol. Soc.
Wash. 46:21-27.

Petersen, J. J. 1979. pH as a factor In parasitism of mosquito larvae by the
mermithid Bgmangmgrmlsgufigflmax. J. Nematol. 11 .105-106.

Petersen, J.J. 1985. Nematode parasites. pp.110-122. ll! Biological Control of
Mosquitoes. Eds. H.C. Chapman, A.R. Barr, M. Laird, and DE. Weidhaas.
Am. Mosq. Control Assoc. Bull. No.6.

Petersen, J. J. and O. R. Willis. 1972. Results of preliminary field applications of
W mglsgnl (Mermithidae: Nematoda) to control mosquito
larvae. Mosq. News 32 :-312 316.

Petersen, J.J. and OR. Willis. 1974a. ijjmgrmlg mm (Nematoda:
Mermithidae): A potential biocontrol agent of Angghglgs larvae. J.
Invert. Pathol. 24:20-23.

Petersen, J.J. and OR. Willis. 1974b. Experimental release of a mermithid
nematode to control Angghglgs mosquitoes in Louisiana. Mosq. News
34:316-319.

Platzer, E.G. 1981. Biological control of mosquitoes with mermithids. J.
Nematol. 13:257-262.

Rademacher, RE. 1979. Studies of ovenrvintering larvae of
W mosquitoes in Minnesota. Mosq. News 39:135-136.

Sekla, LH., W. Stackiw and RA. Brust 1980. Arbovirus isolations from
mosquitoes in Manitoba. Mosq. News 40:377-380.

Sjogren, R. D., D. P. Batzer and M. A. Juenemann. 1986. Evaluation of
methoprene, temephos and Bagfllusmgmglgngjg var. mglggsig
against Qagujllenjglapgnumanslarvae In Minnesota. J. Am. Mosq.
Control Assoc. 22276-279.

Slaff, M. 1986. Predicting the spring emergence of Qggulugnjgla perturbans.
J. Am. Mosq. Control Assoc. 2:227-228.

43

Slaff, M. and J. D. Haefner 1985. Seasonal and spatial distribution of Mansgnia

marl Mansonia titlllans and oadurbans
(Diptera: Culicidae) in the central Florida, USA, phosphate region. J.

Med. Entomol. 22:624-629.

Smith, J.B. 1908. Notes on the larval habits of Bulax gammans. Ent. News
19:22-25.

Snow, W.E. and E. Pickard. 1957. Correlation of vertical and horizontal flight
activity of Mansonia perturbans with reference to marked changes in
light intensity (Diptera:Culicidae). Ent. Soc. Am. Ann. 50:306-311.

Southwood, T.R.E. 1978. Ecological Methods with Particular Reference to the
Study of Insect Populations. Chapman and Hall, New York.

Steele, R.G. and J.H. Torrie. 1980. Principles and Procedures of Statistics: A
Biometrical Approach, 2nd ed. McGraw-Hill Book Company, New York.

Walker, E. D. 1987. Efficacy of sustained- release formulations of Bacillus

mumglansjs isLaalansis and methoprene for control of Qggumanigia
mailman; J. Am. Mosq. Control Assoc 3:97-99.

Walker, E. D. and W..J Crans1986.Asimple method for samplIng
Bcguiflanigiacagumanslarvae. J. Am. Mosq. Control Assoc 2:239-

Chapter 3

The Natural Diet and Feeding Behavior of Larval

W mailman: (Diptera:Culicidae)

44

45

ABSTRACT

The natural diet of larval chuilianigia ganumans was examined and the
season, site and instar variation determined. The numbers of bacteria, detritus
and algae were very consistent in guts of fourth larval instars , within season
and site. Average numbers per fourth instar gut of each food category were 1.0
X 105, 3.7 X 104, and 2.2 X 103, for bacteria, detritus and algae, respectively.
The number of individual bacterial and algal food types per gut varied
significantly over season and site. Proportions of bacteria and detritus were not
significantly different among instars, however, guts of fourth instars had a
greater proportion and diversity of algal food categories. Observation of larval
feeding behavior showed that fourth instars fed by suspension feeding and
browsing from sediments and plant surfaces. These feeding mechanisms were
supported by the presence of microtrichia on lateral palatal brushes of fourth

instars.

46

INTRODUCTION

chuillsmuja cactumans (Walker) has been implicated as the enzootic

and epidemic vector for Eastern Equine Encephalitis in horses and humans in
the Midwest and noncoastal parts of the eastern United States (Boromisa et al.
1987, Clark et al. 1985).

The feeding behavior of larval mosquitoes has been studied relatively

little; yet, its importance to effective larval control and to the vector status of

T_—_" _-____”I—l

adults has been demonstrated. Grimstad and Haramis (1984) showed that
Aadas msarlatus adults, developing from nutritionally deprived larvae, were
more efficient vectors of the LaCrosse Encephalitis virus than those from well-
fed larvae, and McComb (1980) found that an optimal laboratory larval diet
increased both the size and survival of adults. Although Cg. ganumans larval
control using Bacillus thu ringiansis israalansis (ELL) has been achieved in
some locations (Sjogren et al. 1986), repeatable results have not always been
demonstrated (Walker et al. 1985, Walker 1987), suggesting that some aspect
of the larval feeding ecology prevented effective ingestion of the B,1,i. (Walker
1987).

Most work on larval mosquito diets occurred in the early part of this
century (e.g. Coggleshall 1926, Boyd and Foot 1928, Senior-White 1928,
Hinman 1930, and Howland 1930); yet, many questions about larval feeding
ecology still remain. Until recently, techniques for visualizing gut contents were
inadequate, resulting in certain food-types being under-represented.

Benthic-dwelling mosquito larvae can gather their food in a variety of
ways by employing the following mechanisms: 1) filter-feeding; 2) browsing
from a substrate surface; or, 3) actually ingesting the substrate. Many authors

have described each of these feeding behaviors for different genera of

47

mosquito larvae (Renn 1941, Surtees 1959, Pucat 1965, Harbach 1977).
Mouthpart morphology of mosquito larvae is a good indicator of their feeding
strategy (Harbach 1977). For example, container-breeding Aadas, which
mainly feed by browsing material rather than filtering, appear to have shorter,
and thicker mandibular brushes than do pond-living, eddy-filtering mosquitoes;
e.g., many ADM larvae (Harbach 1977, Merritt and Craig 1987).

Although mosquito larvae are not generally viewed as selective feeders
(Renn 1941 ), studies have demonstrated a relatively high degree of selectivity
for particle sizes (Clements 1963, Dadd 1971a,b, Merritt et al. 1978, Merritt
1987). This ability to select among potential food items is an important
adaptation for benthic-dwelling animals. Unlike the epilimnion, where food
resources are often scarce, ingestible material in the benthos is rarely lacking;
the problem for benthic-dwelling organisms is that of selection for edible
particles.

Larvae of Cg. pauumags and other members of the genera chuillanigja
and Magsggia are unique in that they do not need to come to the surface for air,
but attach by means of a modified anal syphon to the submerged parts of rooted
or floating aquatic macrophytes. They use the oxygen in the aerenchymous
tissue of these plant parts for their respiration (McNeel 1932, Laurence 1960,
McDonald and Lu 1973). In particular, Cg. ganumags larvae occur in the highly
organic benthic zone of eutrophic ponds and marshes (Hagmann 1953, Bailey
1984, Batzer and Sjogren 1986, Smith 1908).

The habitat requirements of larval Cg. canumans are very specific and
the location, sampling and culturing of this species has been difficult to
investigate (see Olds and Merritt 1989a for detailed discussion). As a result,

little is known about many aspects of the larval biology of Cg. panuLQans,

48

including its diet and feeding behavior. In addition, direct behavioral
observations of this species have never been published.

Walker (1987) dissected the guts of larvae collected in his study and
found nonliving particulate material and algae. The feeding behavior of the
Eurasian counterpart to ngujLamgla perturbans. Cg. fighlardii, has been
investigated further. Guille (1976) suggested that the organic, flocculent
material on pond bottoms not only serves as a food source for Cg. [igniargii
larvae, but also forms the medium for bacterial colonies. He suggested that
bacterial colonies were the major food source of the larvae. Goshenko (1985),
also working with Cg. flgluaggii, found that feeding varied with instar. First
instars were periphytophageous on the substrate while second through fourth
instars were filter feeders. However, some feeding from the substrate was
observed in later instars. Goshenko (1985) implied that larvae moved their
points of anchorage more frequently in response to food shortages in the water
column. He also felt that substrate feeding by third and fourth instars occurred
primarily under these conditions, and an analysis of the gut contents of Cg.
flQllifiElll larvae confirmed that periphyton and benthic algae were present in
guts.

Laurence (1960), working with members of the genus Mansguia, showed
that first instars consumed particles less than 2 um in diameter, consisting
primarily of yeast, bacteria and small flagellates. Second Instars ingested
particles up to 20 um in diameter, including rotifers and large protozoa, and
third instars consumed larger flagellates such as the chrysophycean algae,
Enacus, and small rotifers up to 50 um in diameter. Fourth instars ingested
plant debris up to 200 um long. Dunn (1918) and Cheong and Yap (1985)

reported similar findings for members of the genus Mansgnla of their regions.

49

This paper describes the diet of larval Cg. garlurbarls in their natural

environment and discusses the feeding mechanisms employed by the larvae.
METHODOLOGY AND EXPERIMENTAL DESIGN
Sampling Methods and Gut Content Analysis

Larvae were collected from three sites in lngham and Clinton Co.,
Michigan between March and December 1987. Two sites, Drumheller Pond
(Site 1) and Inland Lakes Research Center Pond marsh (Site 2) (see Olds and
Merritt 1988 for site descriptions) were sampled from March 1987 through June
1987. After that, the ponds became unsuitable habitat for Cg. mumans
larvae, except at the end of July at Site 2, owing to drought in the Midwest in
1987. Larvae from a third site, Gibbs Pond (Site 3), were collected from June
1987 to December 1987. Larvae were collected from areas of the ponds known
to have high densities of larvae, using the modified bilge-pump method of
Walker and Crans (1986). Five or 10 stems were sampled on each sampling
date, the material was then transported to the lab in 20-liter buckets where the
larvae were handsorted within 2 hours of sampling. Earlier observations
showed that larvae did not feed while in the mud/water slurry in the buckets:
thus, the amount of elapsed time between sampling and sorting did not affect
the gut contents. Samples were also placed in a 5° C cooler upon returning to
the lab to slow any digestion or other processes that might have altered the gut
contents. After sorting, larvae were preserved in 10% formalin and stored in the
dark at 5° C until dissection.

Dissections were performed with a modification of the Cummins et al.

(1973) method. In this method, the guts were dissected out and the contents

50

removed from the peritrophic membrane in several washes of filter-sterilized
distilled water. The contents were then placed in an acid-washed shell vial, and
filter-sterilized distilled water was added to bring the total volume to 2 ml. The
vial was then sonicated for 15 seconds to break-up large aggregates of material
formed in the digestive tract, but not long enough to rupture any cells. The
contents were then stained with DAPI (4'6 diamidino-2-phenylindole) (Porter
and Feig 1980) at a final concentration of 2 ug ml'1 for 20 minutes in the dark.
The stained material was then filtered on to a black, 0.22 um pore-size,
Nalgenel“) filter using a low pressure hand-operated Milliporel“) filtration
apparatus. The area of the filter covered by the sample was calculated by the
standard formula for area of a circle. Filters were then mounted on slides using
Cargille B immersion oil. Cover slips were applied for sample enumeration
under epifluorescence microscopy.

Counts of bacteria, algae and detritus were made using the DAPI
staining method of Porter and Feig (1980), which has successfully been used to
identify the contents of mosquito guts (Walker et al. 1988). Algae, protozoa
and bacteria were identified using keys in Ward and Whipple (1959).

Counting was performed using a Leitz Laborlux 11 microscope with
epifluorescence light fittings (Walker et al. 1988). At least 15 fields containing
bacteria and detritus were counted, for statistical reliability as suggested by
Kirchman et al. (1982). If fewer than 200 bacteria were counted in the initial 15
fields, additional fields were counted until at least 200 bacteria were counted
per slide. Counts for algae and other large, rare particles were performed by
counting the particles in five transects across the widest diameter of the filter,

equal to 4.5% of the total filter area.

5 1
Natural Diet

WW

Seasonal variation in the natural diet of larval Cg. panumans was
examined using fourth instars collected from Site 3 from 25 June through 2
December 1987. Forty-two larval guts were examined for bacteria and detritus,
and 15 guts were examined for algae. Data were analyzed on a monthly basis

to determine the effect of season on the quantity and diversity of bacteria,

-l——— -

detritus and algae present. In June and December, larval numbers were two
low to provide a reasonable sample size (n<3). As a result, larvae sampled in
June and December were included in the July and November data,
respectively.

All data from gut contents counts were transformed (log(x+1)) to
normalize the data (Sokal and Rohlf 1981). Data were tested for normality,
skewness and kurtosis, and homogeneity of variance using Statistical Analysis
System (SAS) software. MANOVA tests were performed to determine if the total
food composition present in the guts varied seasonally. One MANOVA used
each food category of the bacteria, as well as detritus, and a second MANOVA
used only the algal food categories. The analyses were compartmentalized this
way owing to the three order of magnitude difference between the number of
bacteria or detritus and algae per larval gut. An ANOVA on each individual food
type was performed using the SAS glm procedure for unbalanced data (SAS
1985). Tukey's Mean Separation test was used to determine which means

were significantly different.

52

2] El D'II

It was possible to make between-site comparisons of gut contents on two
occasions: 1) fourth instars sampled from Sites 1 and 2 during the spring of
1987; and 2) third instar larvae collected on 28 and 30 July from Sites 2 and 3.
Data from these ponds were analyzed in the same manner as the gut contents
data from the seasonal samples. Algae, except for blue-green algae, were not
enumerated in gut contents of larvae from these trials.
W

All four instars were collected from Site 3 on July 28, 1987. For that
period, it was possible to compare the diet among instars using gut-contents
data. Counts were made Of the number of each food type per gut, then
transformed into the proportion of each food type, for instar comparisons.
Although gut diameter and length were measured, calculations of gut volume
and subsequent food type per ml of gut proved to be very difficult, especially in
smaller instars. This was due to the inherent limits of resolution of the ocular
micrometer used to measure gut dimensions. The proportion of each food-type
per gut were normalized using an arcsin transfdrmation (Steele and Torrie
1980), then analyzed using MANOVAs, ANOVAs, and Tukey's Mean

Separation tests, where appropriate.

FEEDING BEHAVIOR

1) Natural Eaagigg Silas

In June 1988, an experiment was conducted at the Maple River Marsh
(Site 4) (see Olds and Mern’tt 1988a for description), to determine whether
larvae fed directly on sediments. Ten stems were sampled using the method of

Walker and Crans (1986) and immediately sorted in the field. For each stem

53

that had larvae, a sample of the sediments surrounding that stem was taken,
using a 2 cc sterile syringe. The sediment sample was carefully taken as close
as possible to the surface of the sediments to prevent mixing when the sample
was removed from the water. Six larval and six sediment samples were
preserved immediately with formalin in the field, transported to the laboratory in
a cooler, and stored at 5° C until analyzed.

The guts were dissected from the larvae, and the contents stained and
counted as described above, except sediments were diluted for more reliable
counts. Total detrital and bacterial particles also were counted from the diluted
samples.

ZIEI'IQI Ii 'IIIII

Observations of the feeding behavior of fourth instar Cg. ganumans were
conducted in a chamber constructed of clear glass on the front and back, and
plexiglass on the sides for support (6.7 cm high X 6.7 cm wide X 0.7 cm deep).
A 2-cm layer of pond mud was placed in the bottom, and three small cattail roots
(approx. 1 mm in diameter), were placed in the chamber with one end
immersed in the mud and the other and extended and secured to the top of the
chamber. Tap water was added to within 0.5 cm from the top of the chamber,
resulting in a volume of 29.0 cm3.

For each observation period, four larvae were added to a chamber which
was placed inside a 16° C water bath. The water bath was placed inside a dark
hood so that the light levels could be controlled carefully, since we had
previously observed larvae to be negatively phototactic under high light
conditions. Light was provided using a high intensity microscope illuminator
with the tips covered with red plastic to further simulate the subdued light

conditions of a pond bottom (Wetzel 1983).

54

After larvae were allowed to equilibrate for 30 minutes, observations
were made using a Wild dissecting scope mounted horizontally in a boom.
Both 60X and 120x magnifications were used. Larvae were observed for two
hours per session and during that time, larval movements, location, orientation,
and other feeding behaviors were recorded. The phagostimulants carmine red,
charcoal and yeast extract were added on two separate occasions per
phagostimulant to see if filter-feeding could be enhanced. This experimental
addition was based on the work of Dadd (1970) who noted that some
mosquitoes will not filter-feed in the lab unless a phagostimulant was added to
their growth medium.

BLSEM

Scanning Electron Micrographs (SEM) were taken of the lateral palatal
brushes and mandibular brushes of fourth instar Cg. gammans. After larvae
were collected, they were immediately sorted and stored in 3.5%
gluteraldehyde. The heads were removed and dehydrated in a series of
increasing ethanol concentrations, critical-point dried (with C02 as the
transitional fluid), mounted, and gold coated. Scanning electron micrographs

were taken with a JEOL JSM-356 scope at 600x, 1200X and 2400X.
RESULTS

Natural Diet

WW
Three categories of particulate food were found in larval guts: bacteria,
detritus and algae. Protozoa were also found, but were present in few larvae at

low numbers. Four types of bacteria were identified on the basis of morphology:

55

cocci, rods, purple sulfur bacteria, and spirochetes. Four types of algae were
identified: desmids, diatoms, euglenoids (flagellated unicells) and blue-greens.
When blue-green algae were particularly abundant, they were enumerated by
fields (rather than in transects) with the bacteria.

21W

Results of counts of total bacteria, detritus and algae are presented in
Figure 1. All three food types were present and the abundance of each was
similar throughout the season. Bacteria were the most numerous food type and
ranged from a monthly average of 5.37 X 105 to 1.00 X 105 cells gut", with a
maximum number of cells per gut of 1.66 X 105 and a minimum of 3.16 X 105.
Detritus was the second most abundant food type, with averages ranging from
1.05 X 104 to 3.72 X 104 particles gut“. The maximum and minimum number of
detrital particles in an individual larva was 1.58 X 105 and none, respectively.
Algae were the least abundant food type of the three categories (including blue-
greens), with monthly averages ranging from 1.12 X 102 to 2.29 X 103.

Cocci were always the most abundant bacteria food type, usually
followed by spirochetes and then rods (Table 1). The most abundant algal type
was bluegreens, followed by euglenoids, although the latter were more
abundant in August (Table 2).

Bacteria and detritus varied significantly in larval guts among months
(MANOVA, Wilk's Criterion p<0.0031), but algae did not (MANOVA, Wilk's
Criterion p<0.1145). Among types of bacteria, rods varied significantly (p<0.05)
seasonally, whereas cocci, purple sulfur bacteria and spirochetes, as well as
total bacteria and total detritus did not (p>0.05) (Table 1). The number of rods
was significantly higher in June/July than in August, September and November
but was not significantly different from October (Tukey's Mean Separation

test,p=0.05).

56

 

'5 7
0

6 u
a: .
I.l.l c:
o. v; 5 MA!-
{fl 5,, 4 -u- BACTERIA
(I, 8 3 -I- DETRITUS
I: d + ALGAE
I 2
<
D- 1
O o
z I # I ' 1 ' I ' fl

JU-JUL AUG SEPT OCT NOV-DEC

DATE

Figure 1. Total bacteria, detritus and algae in guts of fourth instar larvae on a
seasonal basis.

57

Table 1. The number of each bacterial food type per larval gut on a seasonal
basis.a

 

MQNItl bl BACTERIA
COCCI RODS PURPLE-SULFUR SPIROCHETES

 

 

Ju./July 10 5.7 5.23 4.0 5.1
(0.08) (0.12) (0.67) (0.20)
August 12 5.5 4.6b 4.2 4.7
(0.06) (0.07) (0.41) (0.38)
September 7 5.6 4.3b 3.5 4.8
(0.05) (0.12) (0.66) (0.1 1)
October 4 5.5 4.7a,b 4.6 5.2
(0.03) (0.06) (0.17) (0.15)
Nov./Dec. 9 5.7 4.5b 2.0 4.7
(0.05) (0.12) (0.61) (0.14)

 

Column means followed by the same letter are not significantly different
(P>0.05, Tukey's Mean Separation Test). If no letter by any means in column
then no significant differences were found between any means. Analysis
conducted on log(x+1) transformed data.

a Data are means (SEM).

58

Table 2. The number of each algal food type per larval gut on a seasonal
basis.a

 

MQNIH N. ALGAE
CYANOPHYTA EUGLENOIDS DIATOMS DESMIDS

 

Ju./July 3 2.9 2.2 0.5 0.9
(0.18) (0.14) (0.46) (0.46)
August 3 2.4 1 .9 1 .6 1.1
(0.1 7) (0.04) (0.14) (0.60)
September 3 1.4 2.2 1.3 1.3
(0.72) (0.09) (0.75) (0.75)
October 3 2.1 1.7 0.5 1.5
(0.13) (0.15) (0.46) (0.15)
Nov./Dec. 3 2.7 1.8 0.0 1.0
(0.33) (0.21) (0.0) (0.51)

 

Column means followed by the same letter are not significantly different
(P>0.05, Tukey's Mean Separation Test). If no letter by any means in column
then no significant differences were found between any means. Analysis
conducted on Iog(x+1) transformed data.

a Data are means (SEM).

59

3] SI [III

The number of total bacteria and total detritus in larval guts from Sites 1
and 2 are presented in Figure 2A, and from Sites 2 and 3 in Figure 2B. Total
bacteria and detritus from Sites 1 and 2, which were sampled in the spring of
1987, were very similar to those in the seasonal samples from Site 3 (see
Figure 1 and Figure 2A for comparison). As found in the gut contents data from
the seasonal samples in Site 3 (Table 1), cocci were the most abundant type of
bacteria (Table 3). and the second most abundant was again spirochetes in
Site 2. However, gut contents from Site 1 had an unusually large number of
sulfur bacteria, an average of 1.26 X 105 gut‘1 (see Table 3).

The MANOVA for Site 1 and Site 2 using cocci, rods, blue-green algae,
sulfur bacteria, spirochetes and total detritus, was significant (Wilk's criterion
P>F=0.0059). ANOVAs using the SAS glm procedure, on each food type as
well as total bacteria showed significant differences only in rods and sulfur
bacteria. Rods were significantly greater in samples from Site 2 while the
opposite was true for sulfur bacteria.

The MANOVA for samples from Site 2 versus Site 3, using the same
variables as in previous analyses, was significant (Wilk's Criterion P>F=
0.0199). ANOVAs showed significant differences in rods and blue-green algae
(see Table 4). Total bacteria and detritus were not significantly different
between ponds (Figure 2). In samples from both ponds, cocci were the most
abundant food type (Table 4).

Wm

The percent of total bacteria and detritus were fairly constant among
instars (Figure 3). Total bacteria was 82 to 92% of the total particles in the gut of
all instars, while detritus ranged from 6 to 17% of the particles. Algae in the guts

exhibited the greatest relative range among instars, with means ranging from

6O

 

 

s .- A
0
m
In A
n. ,7 IQIAL
tn +
In Li, I BACTERIA
6' 0 DETRITUS
" O
E e,
<
n.
O
2
IQIAI.
I BACTERIA
DETRI'IUS

 

NO. PARTICLES PER GUT
(LOG [X+1])

 

SITE

Figure 2. Total bacteria and detritus in guts of: A. fourth instar larvae from Sites
1 and 2 collected from March to June 1987; and B. third instar larvae from Sites
2 and 3 collected on 2 August and 28 July 1987, respectively.

61

Table 3. The number of each bacterial and algal food type per gut on a site

basis-Sites 1 and 2 using guts from fourth instarsa

 

SJIES fl BACTERIA _.ALGAE__
COCCI RODS PURPLE-SULFUR SPIROCHETES CYANOPHYTA

 

SITE1 11 5.8 4.5a 5.1a 5.0 3.8
(0.05) (0.08) (0.1 1) (0.15) (0.39)

SITE2 9 5.7 4.8b 4.7b 4.9 4.3
(0.05) (0.07) (0.15) (0.15) (0.07)

 

Column means followed by the same letter are not significantly different
(P>0.05, T-test, two-tailed). If no letter by any means in column then no
significant differences were found between any means. Analysis conducted on
log(x+1) transformed data.

a Data are means (SEM).

62

Table 4. The number of each bacterial and algal food type per gut on a site
basis-Sites 2 and 3 using guts from third instars.a

 

SEES N BACTERIA __A|_QAE_
COCCI RODS PURPLE-SULFUR SPIROCHETES CYANOPHYTA

 

SITE2 6 4.9 3.5a 3.5 2.9 2.8a
(0.06) (0.17) (0.72) (0.91) (0.58)
SITES 4 5.2 4.30 2.3 3.9 0.6b
(0.18) (0.20) (0.95) (0.21) (0.62)

 

Column means followed by the same letter are not significantly different
(P>0.05, T-test, two-tailed). If no letter by any means in column then no
significant differences were found between any means. Analysis conducted on
log(x+1) transformed data.

a Data are means (SEM).

63

0.1 to 1%. The MANOVA, using all bacterial types and total detritus, performed
on the arcsin transformed data was not significant (Wilk's Criterion P>F=
0.5654). ANOVAs of the proportion of each bacterial type, total bacteria and
total detritus showed no significant differences in any of the food types among
instars (Table 5). The MANOVA performed on the algal food types was not
significant (Wilk's Criterion P>F= 0.3062). However, ANOVAs of the proportion
of each algal food type plus total algae showed significant

, differences at the 0.05 level in desmids, diatoms and euglenoids. Total algae
was significantly different between instars at the 0.10 level (see Table 6).
Tukey's Mean Separation Test showed that fourth instars contained
proportionately more desmids and euglenoids than first, second or third instars,
while fourth instars had more diatoms than first instars, but neither were different

from second or third instars (Table 6).

Feeding Behavior

IIIII IE |' El
Using the paired larval and sediment samples collected on 24 June 1988
from Site 4, the ratios of bacterial particles to detrital particles were calculated
for larval guts and their surrounding sediments. Data were analyzed using a
paired t-test . The proportion of bacteria in larval guts was significantly greater
than in the sediments (p<0.0001) (Figure 4).
2] E I . I Q! I'
The observed behaviors of Cq. perturbans larvae were divided into two
classes: those when the larvae were attached to a root and those when

detached. Larvae tended to spend the majority of time attached to the root or to

64

IQIAI.

I BACTERIA
7, DETRITUS
** I ALGAE

PERCENT OF PARTICLES
PER GUT

 

 

1 2

3
INSTAR

Figure 3. Percent of total bacteria, detritus and algae in the guts of first through
fourth instar larvae collected on July 28, 1987. *= significant differences
(p>0.05). Columns with same letters are not significantly different (p>0.05,
Tukey's Mean Separation Test).

65

Table 5. The percent of each bacterial food type per larval gut for first through
fourth instar collected on 28 July 1987.3

 

 

 

INSIAB N BACTERIA
COCCI RODS PURPLE-SULFUR SPIROCHETES
1 5 59.2 15.4 5.2 9.3
(4.9) (2.5) (3.3) (1 .9)
2 5 63.5 14.7 3.7 13.5
(1.8) (2.4) (2.3) (1.7)
3 5 53.7 17.4 6.6 12.9
(2.6) (2.3) (3.6) (3.3)
4 3 50.3 20.5 15.7 16.7
(4.5) (5.9) (1.6) (7.4)

 

Column means followed by the same letter are not significantly different
(P>0.05, Tukey's Mean Separation Test). If no letter by any means in column
then no significant differences were found between any means. Analysis
conducted on proportion of each particle type per gut transformed by an
arcsin(sq.rt.(Y)) transformation.

a Data are means (SEM).

66

Table 6. The number of each algal food type per larval gut for first through

fourth instar collected 28 July 1987.a

 

INSIAB N

ALGAE

CYANOPHYTA EUGLENOIDS DIATOMS DESMIDS

 

1 3 1.1
(0.57)
2 3 2.5
(0.31)
3 3 1.3
(1.26)
4 3 4.0
(1.64)

0.3a
(0.33)

0.3a
(0.28)

0.2a
(0.18)

1.30
(0.19)

0.0a
(0.0)

0.38,!)
(0.28)

0.2a,b
(0.18)

1.1 b
(0.23)

0.46
(0.39)

0.3a
(0.26)

0.03
(0.0)

1.70
(0.13)

 

Column means followed by the same letter are not significantly different
(P>0.05, Tukey's Mean Separation Test). If no letter by any means in column
then no significant differences were found between any means. Analysis
conducted on proportion of each particle type per gut transformed by an

arcsin(sq.rt.(Y)) transformation.

a Data are means (SEM).

67

roothairs of provided cattail stems. They usually attached with their ventral
surface oriented toward the water surface, a position that placed their mouth in
the path of any material settling from the water column. While attached, larvae
often bent toward the cattail root and appeared to graze from around the point of
attachment. Larvae never attached themselves low enough to be fully covered
by mud, yet they attached with the posterior half of the body in the mud, and the
mouth still in water. On several occasions, it appeared that larvae had
excavated a funnel-shaped depression around the root by swimming around
the area where it entered the sediments. They then attached to the root in the
bottom of the depression with dorsal surfaces oriented toward the sediments.
Their heads would be approximately at the level of the sediment-water
interface, and their mouths were oriented to the water column.

Larvae often swam immediately to the bottom when detaching from the
cattail roots. Once detached, larvae swam up or down along a root or through
the water for an extended period of time, up to 10 mins. They also swam along
the surface of the sediments with mouth bmshes touching the actual sediment
surface. Occasionally, the mouth brushes moved while larvae were swimming
in this fashion.

Digging into the sediments was another common larval behavior,
accomplished this by swinging the posterior halves of their bodies back and
forth through the sediments. Although the sediments were 2 cm deep, they did
not appear to dig more than 1 cm deep. Larvae never stayed completely
covered with sediments for more than 3 to 4 minutes. They did not completely
cover themselves with sediments, but rather burrowed into the sediments
leaving their heads and mouth parts exposed to the sediment/water interface.
While doing this, the larvae tended to orient their bodies almost parallel to the

sediment surfaces with dorsal surfaces directed down.

68

 

 

('5 25' MEAN DIFFERENCE: 18.5
> > ‘ PR>T= 0.0001
0:
Lu In
a. o.
< In
- a
I:
u:
< E
m a
0' /—fi__
2

 

GUTS SEDIMENTS

Figure 4. Ratio of number of bacteria to number of detrital particles in the guts of
fourth instar larvae and the surrounding sediment.

69

The larvae did not filter feed when clear tapwater was provided in the
observation chamber. The phagostimulants charcoal powder and carmine red
did not promote or enhance filter-feeding behavior, even when added at high
concentrations. However, when a slurry of yeast was added, larvae readily
filtered for prolonged periods of time. Similar results have been obtained using
solutions containing protozoa (Copeland, pers. comm.)

SLEEM

Micrographs of the lateral palatal fans of fourth instar Cg. panumans

larvae showed the presence of microtrichia. The microtrichia were 0.2 um at the

base, 1.0 um in height with a maximum gap of 0.2 um (Figure 5).

70

Figure 5. Scanning electron micrograph of lateral palatal brush of fourth instar
larvae showing: A. microtrichia.

71

I!

f

- , l
5“ -‘., ’I
fat ,. XIII
' a J. 5‘ . v . ~,
I ( - .
‘ I I V ‘
’ . , .
. .
v
a\ , L
.

 

7 2
DISCUSSION

The diet of larval Cg. gammans was very consistent regardless of
season, site or instar. Bacteria were the most abundant food type, followed by
detritus, then algae. Cocci were always the most abundant type of bacteria
found in the guts. Regardless of time of year, the number of bacteria in the guts
of fourth instar larvae ranged from an average of 5.37 X 105 to 1.00 X 106 cells
gut". In comparison, Nilsson (1987) found the number of bacteria per gut
ranged from 6.68 X 106 to 2.18 X 107 for a variety of mosquito species. Also,
Walker et al. (1988) found that the number of bacteria per gut for field-collected
Aadas m'saciatus and Anggnalas guagflmaculams larvae averaged 2.2 X 106
and 2.0 X 105 bacteria, respectively per gut. These latter results were obtained
using the DAPI technique for visualizing bacteria (Walker et al. 1988). Larval
Cg. gammans appeared, therefore, to have fewer bacteria per gut than other
mosquito larvae investigated. Possibly, they are able to compensate for
consuming fewer bacteria per unit time, by having longer larval development
times.

Although total bacteria in guts did not vary significantly with season or
between sites, individual types of bacteria did. This is probably related to local
changes of microbial populations within the small areas that the larvae browse.
Several authors have shown that there is a definite seasonal component in the
abundance of certain bacteria (Paterek and Paynter 1988, Holmes and Vennes
1970),and different sites can have differing abundances of microbe types within
microsites (Lopez 1988, Hines and Buck 1982). Therefore, it seems improbable
that larvae were selecting different types of bacteria at different sites or months.
Several authors have demonstrated that mosquito larvae are able to select

particles on the basis of size (Merritt et al. 1978, Merritt 1987), but only within

73

fairly broad categories (i.e., less than 2 um, 2 to 10 um etc.). Virtually all
bacteria found in the guts of Cg. panumags larvae were less than 5 um in the
largest dimension, excluding filamentous forms (spirochetes), which were
longer. Larvae did not appear to distinguish among types of bacteria in the
same size range. I

There was considerable variation found in certain food categories of the
diet of Cg. ganumags larvae. Often, the abundance of some bacteria in larval
guts, (e.g., spirochetes or purple sulfur bacteria) varied as much as 200% within
site and date. This is strongly suggestive of the patchy nature of bacterial
abundance and larval feeding. It is well known that many species of bacteria
grow in patches in the environment (Franklin et al. 1988, Jones and Paynter
1980), and several other invertebrates and protozoans are known to feed on
patches of bacteria (e.g., Sibbald and Albright 1988). Behavioral observations
of larval Cg. gammans revealed that they moved only short distances over
time. Thus, differing bacterial patches may be encountered and ingested rather
infrequently. This could result in the high variation in the numbers of some
bacteria ingested.

Unlike Goshenko (1985), who found distinct diet differences among
instars of Cg. fichiargii under laboratory conditions, few major differences were
found for diets of different instars of field-collected larval Cg (magmas. Fourth
instars had a higher proportion and more types of algae in their guts than did
earlier instars. They may'have an expanded repertoire of feeding behaviors in
later compared to earlier instars, as suggested by Goshenko (1985). However,
it could also be a stochastic process based on the development time of each
instar and the probability of encountering and retaining algal particles in that

time period. Fourth instars develop much more slowly than first and second

74

instars (see Olds and Merritt 1989b), and therefore would have an increased
probability of finding algal particles, given a patchy distribution of algae.

The findings of this study complement similar studies conducted on the
European counter-part , Cg. Lighlargii. Guille (1976) examined the gut contents
of Cg. ficniargii without using an epifluorescence, technique but concluded that
bacteria represented the major food source for that species. Both Walker
(1987) and Goshenko (1985) also found algae and non-living particulates in the
larval guts of Cg. canumans and Cg. M, respectively. Neither used a
method that allowed visualization of bacteria and, as a result, no information
was available regarding bacterial abundances.

Larvae did not appear to feed directly on sediments, but rather managed
to be water-column feeders and browsers of sediment and plant surfaces even
while dwelling in a detritus enriched environment. Bacteria are more readily
digested and assimilated than detritus (Cammen 1980, Lopez et al. 1977,
Hargrave 1970, Kofoed 1975) and therefore the energetics are more favorable
for feeding on bacteria rather than on detritus. This feeding strategy is not
uncommon among dipteran larvae. Many chironomidae live in depositional
habitats; yet, feed on resources in the water column. They do this by
constructing tubes in the sediments, and weaving silken nets that cover one end
of the tubes. The nets trap particles as larvae undulate their bodies, thereby
pumping water and food through the tube (Merritt and Cummins 1984). Algae
may have contributed greatly to larval nutrition although the total number of
algal cells in larval guts was low. Algae tended to be at least an order of
magnitude larger than bacteria in one dimension. On the basis on biovolume,
algal volume in guts would be very similar to total bacterial volume.

Larvae appeared to have two principal mechanisms of feeding:

suspension feeding from the water column, and browsing from sediment and

75

plant surfaces. Although Cg. ganumans larvae live in the benthos of ponds,
their feeding behaviors seemed to be more similar to other water-column
dwelling mosquitoes, than to other benthic-substrate dwelling organisms. The
ratio of bacteria to detritus was significantly greater in larval guts than in the
surrounding sediments. This indicated that larvae were not simply ingesting
sediments, and supported the behavioral observations that larvae also feed by
browsing on sediment and plant surfaces, and by filtering from the water
column.

The results of observations of the different feeding modes used by Cg.
mm is supported by SEM micrographs of their lateral palatal brushes
(LPB) bearing microtrichia. Several authors have found that microtrichia on the
rays of the lateral palatal brushes or mandibular fans coincide with a browsing
and a suspension feeding mode (Merritt and Craig 1987, Aly 1983, Harbach
1977). Merritt and Craig (1987) found that the larvae of Asgastcisadatus,
species that browses substrates, had lateral palatal brushes with long
microtrichia (5 um height X 1 um basal width). Microtrichia on Cg. panumans
were much smaller than those found on As. misaLiatus, indicating that while Cg.
gamms may employ browsing, its primary means of food acquisition was
suspension feeding. Alternatively, the size of the microtrichia could be related
to the hardness of the substrate being browsed, with harder substrates requiring
larger, stiffer microtrichia. Asgas Malus larvae often browse from very hard
substrates (i.e., bark and leaves In treeholes); whereas, Cg. canumans larvae
primarily browse the sediment surface and the surface of soft aquatic plant roots
and stems.

This study illustrates the importance of examining the natural diet and

feeding behavior of mosquito larvae. It is clear from this study that Bacillus
muuggiansis var. israalansis and Bacillus sghaacicus are not excluded from the

76

larval gut owing to some aspect of larval feeding behavior. If a proper
formulation of 5.1.1. were used that took advantage of the tendency of these
larvae to both filter and to browse from plant and sediment surfaces, larval
control using B,l,i. may be possible. Because of their unique mode of
respiration, traditional methods of larval control are not possible for Cg.
Mm. making the development of effective microbial pesticides even
more imparative. The complexity of the habitat in which Cg. parlumans larvae
live (see Olds and Merritt 1988a, Batzer and Sjogren 1987) may be responsible
for the failure of BM. to control Cg. mils (Walker 1987). These factors will
always pose a challenge to effective larval control regardless of control agent.
With a thorough understanding of the larval biology of this species, control

measures that rely on less toxic, more specific larvicides should be possible.

7 7
LITERATURE CITED

Aly, C. 1983. Feeding behavior of Aadas vegans larvae (Diptera: Culicidae) and

its influence on the effectiveness of Bacillus mumgiansis var. isLaalansis.
Bull. Soc. Vector Ecol. 8: 94-100.

Bailey, D.L. 1984. Comparison of diurnal and nocturnal Mansonia larval
populations on water lettuce plants. Mosq. News 44:548-552.

Batzer, DP. and RD. Sjogren. 1986. Larval habitat characteristics of
' ' i canumags (Diptera:Culicidae) in Minnesota. Can. Ent.
118:1193-1198.

Boromisa, R. D. R. S. Copeland and P. R. Grimstad. 1987. Oral transmission of
Eastern Equine Encephalomyelitis virus by a northern Indiana strain of

nguillaltidiacanumans. J. Am. Mosq. Control Assoc. 3: 102- 104.

Boyd, M. and H. Foot. 1928. Studies in the bionomics of American Anggnslcs. J.
Prev. Med. 2: 219-242.

Cammen, L.M. 1980. The significance of microbial carbon in the nutrition of the
deposit feeding polychaete Narcis sucsinaa. Marine Biology 61 :9-20.

Cheong, WC and H.H. Yap. 1985. Bioassays of Bacillus sgnaacicus (strain
1593) against mosquitoes of public health importance in Malaysia.
Southeast Asian J. Trop. Med. Pub. Hlth. 16:54-58.

Clark, G. G. W. J. Crans and C. L Crabbs. 1985. Absence of Eastern Equine
Encephalitis (EEE) virus in immature '
associated with equine cases of EEE. J. Am. Mosq. Control Assoc. 1:540-
542.

Clements, A.N. 1963. The Physiology of Mosquitoes. Pergamon, Oxford. 393pp.

Coggleshall, LT. 1926. Relationship of plankton to anopheline larvae. Am. J.
Hyg. 6: 556-569.

Cummins, K.W. 1973. Trophic relations of aquatic insects. Ann. Rev. Ent.
18:183-206.

Dadd, RH. 1970. Relationship between filtering activity and ingestion of solids
by larvae of the mosquito Culax ginisns: a method for assessing
phagostimulant factors. J. Med. Ent. 7: 708-712.

Dadd, R.H. 1971a. Effect of size and concentration of particles on rates of
ingestion of latex particulates by mosquito larvae. Ann. Entomol. Soc.
Am. 64:687-692.

78

Dadd, R.H. 1971b. Size limitations on the infectibility of mosquito larvae by
nematodes during filter-feeding. J. Invert. Path. 18:246-251.

Dunn, LR. 1918. The lake mosquito, Mansgniatilillans Walk, and its host plant,
Bisua lealiglas Linn., in the Canal Zone, Panama (Diptera: Culicidae).
Entomol. News 29:260-269.

Franklin, M.J., W.J. Wiebe, and W.B. Whitman. 1988. Populations of
methanogenic bacteria in a Georgia salt marsh. Applied and
Environmental Microbiology 54:1 151-1 157.

Goshenko, U.A. 1986. Variability in morphological and ecological features in
larvae of Mansgnia richlarmi Fic. (Diptera :Culicidae) in DA. Skarlato
(ed). Systematics of the Diptera. Oxonian Press, New Delhi, India.
pp 15-17.

Grimstad, RR. and L.D. Haramis. 1984.Aadas1[isa[ialus (Diptera: Culicidae)
and La Crosse virus. Ill Enhanced oral transmission by nutritionally
deprived mosquitoes.J. Med. Entomol. 21 :249-256.

Guille, G. 1976. Recherches eco-ethologiques sur nggillgmgia ' '
fignjargii (Ficalbi), 1889 (Diptera-Culicidae) du littoral mediterranean
Francais ll. -Milieu et comportement. Ann. Sci. Nat. Zool., Paris 22:5-112.

Hagmann, LE. 1953. Biology of ”Mansgnia gammans" (Walker). Proc. N.J.
Mosq. Exterm. Assoc., 40th Annu. Meet. 141 -147.

Harbach, RE. 1977. Comparative and functional morphology of the mandibles
of some fourth stage mosquito larvae (Diptera:Culicidae). Zoomorph.
87:217-236.

Hargrave, B.T. 1970. The utilization of benthic microflora by Hyaialia aziaca
(Amphipoda) J. Anim. Ecol. 39:427-437.

Hines, ME. and JD. Buck. 1982. Distribution of methanogenic and sulfate-
reducing bacteria in near-shore marine sediments. Appl. Environ.
Microbiol. 43:447-453.

Hinman, EH. 1930. A study of the food of mosquito larvae (Culicidae). Am. J.
Hyg. 12: 238:270.

Holmes, H.W. and I.W. Vennes. 1970. Occurrence of purple sulfur bacteria in a
sewage treatment lagoon. Appl. Microbiol. 19:988-996.

Howland, L.J. 1930. The nutrition of mosquito larvae, with special reference to
their algal food. Bull. Entomol. Res. 21: 431 -440.

79

Jones, U.J. and M.J.B. Paynter. 1980. Populations of methane-producing
bacteria and in vitro methanogenesis in salt marsh and estuarine
sediments. Appl. Environ. Microbiol. 39:864-871.

Kirchman, D., J. Sigda, R. Kapuscinski and R. Mitchell. 1982. Statistical analysis
of the direct count method for enumerating bacteria. Appl. Environ.
Microbiol. 44: 376-382.

Kofoed, L.H. 1975. The feeding of WW (Montague). 1. the
assimilation of different components of food. J. Exp. Mar. Biol. Ecol.
19:233-241.

- .mI,’

Laurence, BR. 1960. The biology of two species of mosquitoes, Mansgnia
alricana (Theobald) and Mansnnia unilgrmis (Theobald), belonging to
the subgenusMansonioides (Diptera: Culicidae). Bull. Entomol. Res.
51 :491-51 7.

 

“T. If.

Lopez, GR. 1988. Comparative ecology of the macrofauna of freshwater and
marine muds. Limnol. Oceanogr. 33:946-962.

Lopez, G.R., J.S. Levinton and LB. Slobodkin. 1977. The effect of grazing by
the detritivore gamma gg'llus on Ssanina litter and its associated
microbial community. Oecologia 30:1 1 1-127.

McCombs, SD. 1980. Effect of differential nutrition of larvae on adult fitness of
Aadas msaijatus. MS. Thesis, University of Notre Dame. 123pp.

McDonald, J.L. and LC. Lu 1973. Preference of Mansgnia uniigimis (Theob.)
for specific water hyacinth plants. Mosq. News 33:466-467.

McNeel, TE. 1932. Observations on the biology of Mansgnia canumans
(Walker) Diptera, Culicidae. Proc. N.J. Mosq. Exterm. Assoc., 19th Annu.
Meet, 125-128.

Merritt, R.W. 1987. Do different instars of Asgas msaciatus feed on particles of
the same size? J. Am. Mosq. Control Assoc. 3:94-96.

Merritt, R.W. and D. A. Craig 1987. Larval mosquito (Diptera:Culicidae) feeding
mechanisms: mucosubstance production for capture of fine particles. J.
Med.Entomol. 24:275-278.

Merritt, R.W. and K.W. Cummins. 1984. An Introduction to the Aquatic Insects of
North America, 2nd ed. Eds. R.W. Merritt and K.W. Cummins.
Kendal/Hunt Pub. Co.,Dubuque, Iowa.

Merritt, R.W., M.M. Mortland, E.F. Gersabeck and DH. Ross 1978. X-ray
diffraction analysis of particles ingested by filter-feeding animals. Ent.
Exp. and Appl.24:27-34.

80

Nilsson, C. 1987. Feeding and food utilization by mosquito larvae. Dept. of
Zoology. Univ. Uppsala. 269 pp.

Paterek, J. R. and M.J.B. Paynter. 1988. Populations of anaerobic phototrophic
bacteria in a BpanLria allamiflgra salt marsh. Appl. Environ. Microbiol.
54:1360-1364.

Porter, KC. and Y. S. Feig. 1980. The use of DAPI for identifying and counting
aquatic microflora. Limnol. Oceanogr. 25: 943-948.

Pucat, AM. 1965. The functional morphology of the mouthparts of some
mosquito larvae. Ouaest. Entomol. 1:41-86.

Renn, C. E. 1941. The food economy of Anggnalas guaggmacuiatus and A.
cmgians larvae. pp. 329- 341. lgA Symposium on Hydrobiology. U. of
Wisconsin Press, Madison.

Senior-White, R. 1928. Algae and food of anopheline larvae. Indian J Med. Res.
16: 11-30.

Sibbald, M.J. and L.J. Albright. 1988. Aggregated and free bacteria as food
sources for heterotrophic microflagellates. Appl. Environ. Microbiol.
54:613-616.

Sjogren, R. D., D. P. Batzer and M. A. Juenemann. 1986. Evaluation of

methoprene, temephos and Bacillusmuungiansisvar. israaiansis
againsthguillanigiaganumanslarvae in Minnesota. J. Am. Mosq.

Control Assoc. 22276-279.

Smith, J.B. 1908. Notes on the larval habits of Culax ganumans. Ent. News
19:22-25.

Sokal, RR. and F.J. Rohlf. 1981. Biometry, 2nd ed. W.H. Freeman and
Company, New York.

Statistical Analysis System. 1985. SAS users guide: statistics, ver. 5. Statistical
Analysis System, Inc. Cary, NC. 956 pp.

Steele, R.G. and J.H. Torrie. 1980. Principles and Procedures of Statistics: A
Biometrical Approach, 2nd ed. McGraw-Hill Book Company, New York.

Surtees, G. 1959. Functional and morphological adaptions of the larval
mouthparts in the sub-family Culicinae with a review of some related
studies by Montschadsky. Proc. R. Entomol. Soc. Lond. 34:7-16.

Walker, E. D. 1987. Efficacy of sustained-release formulations of Bacillus

mumgiansis isLaalansis and methoprene for control of nguillanigia
Qanumans. J. Am. Mosq. Control Assoc 3:.97-99

81

Walker, ED. and W.J. Crans 1986. A simple method for sampling QQQllillenidia
canumans larvae. J. Am. Mosq. Control Assoc. 2:239-240.

Walker, E.D., E.M. McDonald and JP. Kobayashi. 1985. Small-scale field

evaluations of teknar, abate, and dursban as nguillanidia panumans
larvicides. Proc. N.J. Mosq. Control Assoc. 72:169—172.

Walker, E.D., E.J. Olds and R.W. Merritt. 1988. Gut content analysis of mosquito
larvae (Diptera: Culicidae) using DAPI stain and epifluorescence
microscopy. J. Med Entomol. 25:551-554.

Ward, HE. and GC. Whipple. 1959. Fresh-Water Biology. W.T. Edmondson, ed.
John Wiley and Sons, New York.

Wetzel, R.G. 1983. Limnology, 2nd ed. Saunders, New York, 767 pp.

BIBLIOGRAPHY

82

83

BIBLIOGRAPHY

Allan, S.A., G. A. Surgeoner, B. V. Helson and D. H.Pengelly. 1981. Seasonal
activity of Magsggia ganumans adults (Diptera:Culicidae) in
southwestern Ontario. Can. Entomol. 113:133-139.

Aly, C. 1983. Feeding behavior of Aagasyaxans larvae (Diptera: Culicidae) and
its influence on the effectiveness of Bacillus mumgisnsis
vanisraalansis. Bull. Soc. Vector Ecol. 8: 94- 100.

Bailey, D.L. 1984. Comparison of diurnal and nocturnal Mansgnia larval
populations on water lettuce plants. Mosq. News 44:548-552.

Batzer, D. P. ”and R. D. Sjogren. 1986. Larval habitatcharacteristics of
(Diptera: Culicidae) in Minnesota. Can.
Ent.118:1193—1198.

Beckel, W.E. and H.L. Atwood. 1959. A contribution to the bionomics of the
mosquitoesof Algonquin Park. Can. J. Zool. 37:763-770.

Bidlingmayer, W.L. 1954. Description of a trap far Mansgnja larvae. Mosq. News
14:55-58.

Bidlingmayer, W.L. 1968. Larval development of Mansgnia mosquitoes in
centralFlorida. Mosq. News 28:51-57.

Boromisa, R. D, R. S. Copeland and P. R. Grimstad 1987. Oral transmission of
Eastern Equine Encephalomyelitis virus by a northem Indiana strain of

chuillanigiacaclumans. J. Am. Mosq. Control Assoc 3:102104.

Boyd, M. and H. Foot. 1928. Studies in the bionomics of American Angnnalas.
J. Prev. Med. 2: 219-242.

Brower, L. P. 1953. The distribution of Mansgniacsnumans (Walker) in Morris
County.Proc. N.J. Mosq. Exterm. Assoc., 45th Annu. Meeting. 147-149.

Brown, B..J and E.G. Platzer. 1978a. Salts and the InfectIVIty ofBgmangmsLmis
culicixgrax. J. Nematol. 1053-64

Brown, B.J. and E. G. Platzer. 1978b. Oxygen and the infectivity of
Bomammermlsculicixorax J Nematol 10:-110113

Callahan, J.L. and CD. Morris. 1987. Habitat characteristics of ' '
pgnumans in central Florida. J. Am. Mosq. Control Assoc. 2:176-180.

Cammen, L.M. 1980. The significance of microbial carbon in the nutrition of the
deposit feeding polychaete Narais succinaa. Marine Biology 61 :9-20.

84

Capotosto, P. M. and D. Boyes. 1984. Climatic effects on nguillcmdia
perturbansin Barrington, Rhode Island. Mosquito News. 44 :-244 246.

Carpenter, S. J. and W.J. LaCasse. 1974. Mosquitoes of North America (North
of Mexico). Univ. of California Press, Berkeley. 360 pp.

Chamberlain, R.W. 1957. Virus-vector host relationships of the American
arthropod-home encephalitides. Proc. Tenth Int. Congr. Entomol. 3:
567-572.

Chen, RS. 1976. A study of Baesimsrmis nialsani for control of Culax gigigns
latigans in Taiwan. Bull. Inst. Zool. Acad. Sinica. 15:21-28.

Cheong, WC. and H.H. Yap. 1985. Bioassays of Bacillussghaaricus (strain
1593) against mosquitoes of public health importance in Malaysia.
Southeast Asian J. Trop. Med. Pub. Hlth. 16:54-58.

Clark, G. G. W. J. Crans and C. L. Crabbs. 1985. Absence of Eastern Equine
Encephalitis (EEE) Virus in immature
associated with equine cases of EEE. J. Am. Mosq. Control Assoc. 1:540-
542.

Clements, A.N. 1963. The Physiology of Mosquitoes. Pergamon, Oxford. 393pp.

Coggleshall, LT. 1926. Relationship of plankton to anopheline larvae. Am. J.
Hyg.6: 556-569.

Cowardin, L.M., V. Carter, F.C. Golet and ET. LaRoe. 1979. Classification of
wetlands and deepwater habitats of the United States. US fish and
WIldlife Service, Washington DC. 131 pp.

Cummins, K.W. 1973. Trophic relations of aquatic insects. Ann. Rev. Ent. 18:
183-206.

Dadd, RH. 1970. Relationship between filtering activity and ingestion of solids
by larvae of the mosquito Culax gigians: a method for assessing
phagostimulant factors. J. Med. Ent. 7: 708-712.

Dadd, RH. 1971 a. Effect of size and concentration of particles on rates of
ingestion of latex particulates by mosquito larvae. Ann. Entomol. Soc.
Am. 64:687-692.

Dadd, R.H. 1971b. Size limitations on the infectibility of mosquito larvae by
nematodes during filter-feeding. J. Invert. Path. 18:246-251.

Darsie, R.F.,Jr. and RA. Ward. 1981. Identification and geographical distribution
of the mosquitoes of North America, north of Mexico. Mosq. Syst.
Suppl. 1:313pp.

85

Downe, A.E.R. 1962. Some aspects of host selection by Mansgnjaganumans.
Can. J. Zool. 40:725-732.

Dunn, L.H. 1918. The lake mosquito, Mansgnia millags Walk. and its host plant,
Bislia silatigtas Linn. in the Canal Zone, Panama (Diptera: Culicidae).
Entomol. News 29 :260- 269.

Edman, J. D. 1971. Host- -feeding patterns of Florida mosquitoes. I. Aadas,

Accolades. chufllsttldla Mansonia andEEanbQLa J Med Entomol.
81687-695.

Fallis, AM. and D.M. Wood. 1957. Biting midges (Diptera: Ceratopogonidae)
intermediate hosts Haemoproteus of ducks. Can. J. Zool. 35:425-435.

Francy, DB. 1982. Eastern equine encephalomyelitis (EEE) and consideration
of some alternatives for dealing with EEE as a public health problem.
pp. 15-18. lg: J.G. Engemann (Ed.) Eastern equine encephalomyelitis
(EEE) and public health in southwestern Michigan. Sci. for Citizens Ctr.
of SW Mich., Western Michigan University.

Franklin, M.J., W.J. Wiebe, and W.B. Whitman. 1988. Populations of
methanogenic bacteria in a Georgia salt marsh. Appl. Environ. Microbiol.
54:1151-1157.

Goshenko, V. A. 1978. Biotopes and times of development on Mansgnia
ficniaLgii (Ficalbi) 1889 in the conditions of the Ukrainian steppes.
Medskaya Parazitol. (Moscow) 47: 36- 40. (in Russian)

Goshenko, U.A. 1985. Variability in morphological and ecological features in
larvae of Mansonia ficniargii Fic. (Diptera :Culicidae). pp. 15-17. in DA.
Skarlato (Ed.) Systematics of the Diptera. Oxonian Press, New Delhi,
India.

Grimstad, RR. and L.D. Haramis. 1984. Aagcs msaijalus (Diptera: Culicidae)
and La Crosse virus. "I Enhanced oral transmission by nutritionally
deprived mosquitoes. J. Med. Entomol. 21:249-256.

Guille, G. 1976. Recherches eco-ethologiques sur W ' ' '
n'chjaLdii (Ficalbi), 1889 (Diptera-Culicidae) du littoral mediterranean
Francais ll. -Milieu et comportement. Ann. Sci. Nat. Zool., Paris 22:5-112.

Hagmann, LE. 1953. Biology of ”Mansggia csfiufizans” (Walker). Proc. N.J.
Mosq. Exterm. Assoc., 40th Annu. Meet. 141-147.

Harbach, R. E. 1977. Comparative and functional morphology of the mandibles
of some fourth stage mosquito larvae (Diptera: Culicidae). Zoomorph.
87 :21 7-236.

86

Hargrave, B.T. 1970. The utilization of benthic microflora by flyalalla aztaga
(Amphipoda) J. Anim. Ecol. 39:427-437.

Hayes, RC. 1961. Host preferences of Culisata melanura and allied
mosquitoes. Mosquito News 21 :179-187.

Headlee, T.J. 1945. The Mosquitoes of New Jersey and Their Control. Rutgers
University Press, New Brunswick. 326 pp.

Hines, ME. and JD. Buck. 1982. Distribution of methanogenic and sulfate-
reducing bacteria in near-shore marine sediments. Appl. Environ.
Microbiol. 43:447-453.

Hinman, EH. 1930. A study of the food of mosquito larvae (Culicidae). Am. J.
Hyg. 12: 238:270.

Holmes, H.W. and I.W. Vennes. 1970. Occurrence of purple sulfur bacteria in a
sewage treatment lagoon. Appl. Microbiol. 19:988-996.

Howitt, B..F, H. R. Dodge, LK. Bishop and R. H. Gorrie 1949. Recovery of the
virus of Eastern Equine Encephalomyelitis from mosquitoes (Mansgnia
canturbans) collected in Georgia. Science 110: 141- 142.

Howland, L.J. 1930. The nutrition of mosquito larvae, with special reference to
their algal food. Bull Entomol. Res. 21: 431-440.

Jones, U.J. and M.J.B. Paynter. 1980. Populations of methane-producing
bacteria and in vitro methanogenesis in salt marsh and estuarine
sediments. Appl. Environ. Microbiol. 39:864-871.

Kirchman, D., J. Sigda, R. Kapuscinski and R. Mitchell. 1982. Statistical analysis
of the direct count method for enumerating bacteria. Appl. Environ.
Microbiol. 44: 376-382.

Kofoed, L.H. 1975. The feeding of WW (Montague). 1. the
assimilation of different components of food. J. Exp. Mar. Biol. Ecol. 19:
233-241.

Laurence, B. R. 1960. The biology of two species of mosquitoes, Mansgnja
alnsana (Theobald) and Mansonia unjlgrmis (Theobald), belonging to
the subgenus Mansonioides (Diptera: Culicidae). Bull. Entomol. Res.
51:491-517.

Lewis, DJ. and GE Bennett 1980. Observations on the biology of Mansgnia
ganumans (Walker) (Diptera: Culicidae) in the Nova Scotia-New
answick border region. Can. J. Zool. 58:2084-2088.

87

Lopez, GR. 1988. Comparative ecology of the macrofauna of freshwater and
marine muds. Limnol. Oceanogr. 33:946-962.

Lopez, G.R., J.S. Levinton and LB. Slobodkin. 1977. The effect of grazing by
the detritivore Crghcslia gfillus on Spanina litter and its associated
microbial community. Oecologia 30:1 11-127.

Lounibos, LP. and R.L. Escher 1983. Seasonality and sampling of
nguillanigia (Diptera:Culicidae) in south Florida. Environ.
Entomol. 12:1087—1093.

McCombs, SD. 1980. Effect of differential nutrition of larvae on adult fitness of
Aadasm'saciatus. MS. Thesis, University of Notre Dame. 123pp.

McDonald, J.L. and LC. Lu 1973. Preference of Mansgnia unjlngis (Theob.)
for specific water hyacinth plants. Mosq. News 33:466-467.

McNeel, TE. 1932. Observations on the biology of Mansgnia cagurbans
(Walker) Diptera, Culicidae. Proc. N.J. Mosq. Exterm. Assoc., 19th Annu.
Meet., 125-128.

Merritt, R.W. 1987. Do different instars of Aadas msariatus feed on particles of
the same size? J. Am. Mosq. Control Assoc. 3: 94-96.

Merritt, R.W. and D. A. Craig 1987. Larval mosquito (Diptera:Culicidae) feeding
mechanisms: mucosubstance production for capture of fine particles. J.
Med. Entomol. 24:275-278.

Merritt, R.W. and K.W. Cummins. 1984. An Introduction to the Aquatic Insects of
North America, 2nd ed. Eds. R.W. Merritt and K.W. Cummins.
Kendal/Hunt Pub. Co., Dubuque, Iowa.

Merritt, R.W., M.M. Mortland, E.F. Gersabeck and DH. Ross 1978. X-ray
diffraction analysis of particles ingested by filter-feeding animals. Ent.
Exp. and Appl. 24:27-34.

Mitchell, C.J., P. Chen and HG. Chapman. 1974. Exploratory trials utilizing a
mermithid nematode as a control agent for Culax mosquitoes in Taiwan.
J. Formosan Med. Assoc. 73:241-254.

Morris, C.D., J.L. Callahan and RH. Lewis. 1985. Devices for sampling and

sorting immature chuillanigiaganumans. J. Am. Mosq. Control Assoc.
1 :247-250.

Mueller-Dombois, D., and H. Ellenberg. 1974. Aims and Methods of Vegetation
Ecology. John Wiley and Sons, New York.

88

Nasci, RS. and JD. Edman. 1981. Blood-feeding patterns of Culisaia malanuLa
(Diptera:Culicidae) and associated sylvan mosquitoes in southeastern
Massachusetts eastern equine encephalitis enzootic foci. J. Med.
Entomol. 18:493-500.

Nemjo, J. and M. Slaff 1984. Head capsule width as a tool for instar and species

identification of Mansania dxati Mansanlatitillans and
ganumans (Diptera: Culicidae). Ann. Entomol. Soc. Am. 77:633-635.

Nickle, W.R. 1979. Probable establishment and overwintering of a mermithid
nematode parasite of mosquitoes in Maryland. Proc. Helminthol. Soc.
Wash. 46:21 -27.

Nilsson, C. 1986. Collection and digestion of bacteria by mosquito larvae. Ph.D.
Dissert., Univ. Uppsala. 269pp.

Paterek, J. R. and M.J.B. Paynter. 1988. Populations of anaerobic phototrophic
bacteria in a Spaniga allamiflgra salt marsh. Appl. Environ. Microbiol. 54:
1360-1364.

Petersen, J.J. 1979. pH as a factor in parasitism of mosquito larvae by the
mermithid Bgmangmarmis culiciygrax. J. Nematol. 11:105-106.

Petersen, J.J. 1985. Nematode parasites. pp.110-122. lg Biological Control of
Mosquitoes. Eds., H.C. Chapman, A.R. Barr, M. Laird and DE. Weidhaas.
Am. Mosq. Control Assoc. Bull. No.6.

Petersen, J. J. and O. R. Willis. 1972. Results of preliminary field applications of

Baasimarmis nialsani (Mermithidae: Nematoda) to control mosquito
larvae. Mosq. News 32 .312 316.

Petersen, J.J. and OR. Willis. 1974a. Diximarmis gatarsani (Nematoda:
Mermithidae): A potential biocontrol agent of Anccllalas larvae. J. Invert.
Pathol. 24:20-23.

Petersen, J.J. and OR. Willis. 1974b. Experimental release of a mermithid
nematode to control Angmalas mosquitoes in Louisiana. Mosq. News
34:316-319.

Platzer, E.G. 1981. Biological control of mosquitoes with mermithids. J. Nematol.
13:257-262.

Porter, K.G. and Y. S. Feig. 1980. The use of DAPI for identifying and counting
aquatic microflora. Limnol. Oceanogr. 25: 943-948.

Pucat, AM. 1965. The functional morphology of the mouthparts of some
mosquito larvae. Ouaest. Entomol. 1:41-86.

89

Renn, CE. 1941. The food economy of Angcnalss guagflmaculatus and A.
crucians larvae. pp. 329-341. in A Symposium on Hydrobiology. U. of
WIsconsin Press, Madison.

Sekla, L.H., W. Stackiw and RA. Brust 1980. Arbovirus isolations from
mosquitoes in Manitoba. Mosq. News 40:377-380.

Senior-White, R. 1928. Algae and food of anopheline larvae. Indian J Med. Res.
16: 11-30.

Sibbald, M.J. and L.J. Albright. 1988. Aggregated and free bacteria as food
sources for heterotrophic microflagellates. Appl. Environ. Microbiol. 54:
613-616.

Sjogren, R. D., D. P. Batzer and M. A. Juenemann. 1986. Evaluation of
methoprene, temephos and Bacillus munngiansis var. isLaalansis
againsthgulllanigiacsnumags larvae in Minnesota. J. Am. Mosq.
Control Assoc. 2 :-276 279.

Slaff, M. 1986. Predicting the spring emergence of chujllanigia W. J.
Am. Mosq. Contol Assoc. 2:227-228.

Slaff, M. and J. D. Haefner 1985. Seasonal and spatial distribution of Mansgnia
gm Mansonia titlllans and Coauillattidia 0.6mm (Diptera Culicidae)
in the central Florida, USA, phosphate region. J. Med. Entomol. 22 :624-
629.

Smith, JB 1908. Notes on the larval habits of Culcx gammans. Ent. News 19:
22-25.

Snow, W.E. and E. Pickard. 1957. Correlation of vertical and horizontal flight
activity of Mansgnia canumans with reference to marked changes in light
intensity (Diptera:Culicidae). Ent. Soc. Am. Ann. 50:306-311.

Sokal, RR. and F.J. Rohlf. 1981. Biometry, 2nd ed. W.H. Freeman, New York.

Southwood, T.R.E. 1978. Ecological Methods with Particular Reference to the
Study of Insect Populations. Chapman and Hall, New York.

Statistical Analysis System. 1985. SAS users guide: statistics, ver. 5. Statistical
Analysis System, Inc., Cary, NC. 956 pp.

Steele, R.G. and J.H. Torrie. 1980. Principles and Procedures of Statistics: A
Biometrical Approach, 2nd ed. McGraw-Hill, New York.

Surtees, G. 1959. Functional and morphological adaptions of the larval mouth-
partsin the sub-family Culicinae with a review of some related studies by
Montschadsky. Proc. R. Entomol. Soc. Lond. 3427-16.

90

Walker, E. D. 1987. Efficacy of sustained- release formulations of Bacillus

munngiansis W and methoprene for control of chuillatligia
nadurbans. J. Am. Mosq. Control Assoc 3:.97-99

Walker, ED. and W.J. Crans 1986. A simple method for sampling ngufllcmgia
canumans larvae. J. Am. Mosq. Control Assoc. 2:239-240.

Walker, E. D, E. M. McDonald andJ. F. Kobayashi. 1985. Small- scale erld
evaluations of teknar, abate, and dursban as
larvicides. Proc. N. J. Mosq. Control Assoc. 72: 169-172.

Walker, E.D., E.J. Olds and R.W. Merritt. 1988. Gut content analysis of mosquito
larvae (Diptera: Culicidae) using DAPI stain and epifluorescence
microscopy. J. Med Entomol. 25:551-554.

Ward, H.B. and G.C.Whipple 1959. Fresh Water Biology. Ed.,W.T. Edmundson.
John Wiley and Sons, New York.

Wetzel, R.G. 1983. Limnology, 2nd ed. Saunders, New York. 767pp.

"IillIlllllllllllllllllllES