life tables for culex pippiens suman ds

Medical and Veterinary Entomology (2011), doi: 10.1111/j.1365-2915.2010.00939.x

Variations in life tables of geographically isolatedstrains of the mosquito Culex quinquefasciatus

D. S. S U M A N1, S. N. T I K A R1, M. J. M E N D K I1, D. S U K U M A R A N1,O. P. A G R A W A L2, B. D. P A R A S H A R1 and S. P R A K A S H11Division of Entomology, Defence Research and Development Establishment, Gwalior, India and 2School of Studies in Zoology,Jiwaji University, Gwalior, India

Abstract. Variations in the life tables and other biological attributes of four strains ofCulex quinquefasciatus Say (Diptera: Culicidae) from geographically isolated regionsof India that had been reared to the fifth generation in the laboratory were assessedunder a standardized rearing regime under constant laboratory conditions. Two strainsfrom arid habitats [Jodhpur (JD) and Bikaner (BKN)], one from a semi-arid inlandhabitat [Bathinda (BTH)], one from a semi-arid coastal habitat [Jamnagar (JMN)] anda standard laboratory strain (LAB) were compared. Horizontal life-table parameterswere measured for each strain. Egg mortality ranged from 4.4% (JD and BTH) to19.5% (BKN). The lowest rate of adult emergence and highest female : male ratiowere found in BKN, and the highest rate of adult emergence and lowest female : maleratio were recorded in BTH. The egg-hatching period was longest in BTH and shortestin LAB. The duration from oviposition to adult emergence was longest in JD andshortest in LAB. Females lived longer than males in all strains. The net reproductiverates (R0) of all field-derived strains (122.9162.2) differed significantly betweenstrains and were significantly greater than that of LAB (107.6). Similarly, both theintrinsic rate of increase (rm ) and finite rate of increase () were found to be lowerin LAB than in the field strains, but the mean generation time (T ) and doubling time(DT ) were longest in LAB. For several life-table attributes, JD and BTH clusteredtogether and were more similar to JMN than to BKN and LAB. The results indicatethat BTH, BKN and JD can be characterized as r-strategists, more so than JMN.Overall fecundity increased with age. Differences in annual temperature ranges andmean annual rainfall between locations were positively correlated (r = 0.460.97)with egg production, female life expectancy, R0, rm , and T . The results suggest thatstrains of Cx. quinquefasciatus from different geographical areas with contrastinghabitats vary in their survival and reproductive strategies accordingly.

Key words. Culex quinquefasciatus, ecological variation, life reproductive strategies,life-table attributes, strain development.

Introduction

Culex quinquefasciatus Say is the most common domesticspecies of mosquito in the tropics. It is the principal vector ofBancroftian filariasis (World Health Organization, 2006) and

Correspondence: Dr B. D. Parashar, Scientist F, Division of Entomology, Defence Research and Development Establishment, Gwalior, MadhyaPradesh 474002, India. Tel.: +91 751 223 1862; Fax: +91 751 234 1148; E-mail: [email protected]

breeds in diverse aquatic habitats, ranging from polluted tofresh water (Sirivanakaran, 1976).

The survival and fecundity of different mosquito species areaffected by factors such as nutrition (Day et al., 1994; Costeroet al., 1998), resistance to organophosphate insecticides (Jinfu

2011 The AuthorsMedical and Veterinary Entomology 2011 The Royal Entomological Society 1

2 D. S. Suman et al.

et al., 1998), temperature (Maharaj, 2003) and infection bymicrobial pathogens (Mahmood et al., 2004). Mosquitoesoccurring in distinct locations normally adapt to local con-ditions. Over time, ecological isolation and adaptation to localconditions may lead to the development of geographically iso-lated strains that differ in various biological traits, such asfecundity and survivorship, which gives rise to variation in arange of population parameters (Reisen et al., 1979).

Distinctive life-table attributes have been found in strainsof Cx. quinquefasciatus from Bangkok (Thailand), the south-eastern U.S.A. (Vero Beach, FL), the southwestern U.S.A.(Houston, TX) (Walter & Hacker, 1974) and the NorthwestFrontier in Pakistan (Suleman & Reisen, 1979). Recently,Suman et al. (2009) reported significant variations in morpho-metrics and morphology among eggs from the four strains ofCx. quinquefasciatus investigated in the present study, sourcedfrom arid, semi-arid and coastal regions of India. We hypoth-esized that these mosquito strains will differ in their life-tableand biological traits because they appear to be adapted to their

respective local environments. To test this notion, we assessedlife-table and other biological attributes under standardized lab-oratory environments in four strains of Cx. quinquefasciatusfrom a range of distinct isolated habitats in India and a standardlaboratory strain, and correlated the attributes with ecologicalvariables.

Materials and methods

Mosquito collection, rearing and experimentation

Culex quinquefasciatus larvae were collected from militarycantonments in Jodhpur and Bikaner on the western plain of anarid zone in the Thar Desert, an air force station in Jamnagaron the west coastal plain of a semi-arid zone close to the Ara-bian Sea coast, and Bathinda on a northern plain of the uppernorthwestern region of a semi-arid zone (Fig. 1). The annualrange of temperature (winter minimumsummer maximum)

km

Fig. 1. Locations of collections of Culex quinquefasciatus in India. BTH, Bathinda; BKN, Bikaner; JD, Jodhpur; JMN, Jamnagar; GWL, laboratoryat Gwalior.

2011 The AuthorsMedical and Veterinary Entomology 2011 The Royal Entomological Society, Medical and Veterinary Entomology, doi: 10.1111/j.1365-2915.2010.00939.x

Culex quinquefasciatus life tables 3

was intermediate in Jodhpur and Bikaner (245 C), mildestin Jamnagar (936 C) and most extreme in Bathinda(247 C). Annual rainfall ranged from 10 years. This is situated ina semi-arid region that receives 5001000 mm of rainfall peryear. Annual temperature ranges between 2 C and 48 C. Thevegetation of this region is tropical deciduous and thorny.

Larvae were brought to the Defence Research and Devel-opment Establishment, Gwalior and maintained in standardlaboratory conditions at 27 1 C, 75 5% relative humid-ity (RH) and light : dark (LD) 12 : 12 h. Larvae were kept ata density of 100 larvae per bowl, in enamel bowls contain-ing 2 L of dechlorinated water and were fed regular amountsof dried yeast on alternate days. After pupation, they weretransferred to cages (75 65 65 cm) to provide up to 50male and 50 female adults per cage after emergence. Afteremergence, adults were provided with 10% sugar solution adlibitum dispensed through a cotton wick in a Petri dish anda chick was provided as a source of blood twice a week.These colonies were maintained for up to four generations(approximately 2 months).Chicks were maintained in accor-dance with approved guidelines of the Committee for the Pur-pose of Control and Supervision of Experiments on Animals(CPCSEA), India.

Although laboratory rearing imposes a certain degree ofselection pressure on aspects of mosquito biology, all colonieswere exposed to standardized environmental conditions thatwere favourable to mosquito survival; hence, we assume thatestimates of life-table parameters derived from data collectedfrom the colonized wild strains represent a maximum expres-sion of their life-table parameters and are likely to reflect truedifferences between geographically isolated strains. Similarassumptions were made by Reisen et al. (1979) to comparethe life-table strategies of geographically distinct strains ofCulex tritaeniorhynchus after colonization in standard labo-ratory conditions for many generations (range: 63236).

To assess the adult parameters, 50 pairs of male and femalemosquitoes of the same age from the F5 generation werereleased in each of three rearing cages for each strain [i.e.four experimental strains from Jodhpur (JD), Bikaner (BKN),Bathinda (BTH) and Jamnagar (JMN), respectively, plus thelaboratory strain (LAB)]. Adults were provided with 10% sugarsolution and a chick for blood feeding ad libitum throughoutthe experiment. A glass Petri dish (140 mm in diameter)containing 400 mL water was provided for oviposition. Dailyobservations of egg rafts laid and the mortality or survival ofadult male and female mosquitoes were recorded until the deathof all mosquitoes. The numbers of egg rafts laid were counted,after which the numbers of eggs per raft were counted under

a microscope (Lieca MZ12; Leica Microsystems, Wetzlar,Germany). For each strain, the three largest egg rafts wereselected and placed individually into three bowls, eachcontaining 2 L of dechlorinated water. Subsequently, variousdevelopmental parameters of immature stages were recordedfor each strain. Hatching rates were calculated by comparingthe number of eggs per egg raft with the number of larvaeproduced 12 h later. This was carried out by observing thelarvae against a contrasting background and removing themby pipette to a new bowl. Egg rafts with unhatched eggswere transferred to separate bowls for further hatching andcounted at 24-h intervals until hatching of larvae was complete;numbers of late-hatching larvae were added to the initial count.

Statistical analysis of life-table and other biological attributes

Data analysis was based on mean values for the three repli-cated cohorts of 50 females for each strain. Data were anal-ysed for mean pre-oviposition days (mean number of daysfrom first exposure to blood host to the appearance of thefirst egg raft), oviposition days (mean number of days ofegg laying) and post-oviposition days (mean number of daysfrom the last egg raft laid to the death of the last female),mean number of eggs produced per cohort of 50 females,mean number of eggs per female lifespan [mean numberof eggs/female/oviposition (i.e. number of eggs/number offemales alive on each oviposition day) for all oviposition daysuntil all females were dead number of oviposition days]and eggs/female/day (eggs/female lifespan/number of ovipo-sition days). Various attributes of a horizontal life table wereestimated, including life expectancy at emergence (ex ), netreproductive rate (R0), intrinsic rate of increase (rm ), finiterate of increase () and mean generation time (T ) as per themethod of Howe (1953), Andrewartha & Birch (1954) andSlobodkin (1962). Population doubling time (DT ) was alsocalculated (Yurttas & Alten, 2006). The equations used forthese calculations were as follows:

1 age-specific survivorship: lx = yx/y0, where yx = thenumber of males or females alive on each day, x;

2 life expectancy at emergence: ex = Tx/lx , where Tx =wx=1 Lx and Lx = (lx + lx+1 )/2, where w = the day on

which the last individual died;3 net reproductive rate: R0 =

wx=1 lxmx , where mx =

mean number of eggs/2 (i.e. assuming a 1 : 1 sex ratiofor progeny, the mean number of eggs is divided by 2 togive the number of female eggs;

4 mean generation time: T =WX=1 xlxmx/R0;5 intrinsic rate of increase: rm = log e R0/T , where log e

is the natural logarithm;6 finite rate of increase: = anti-log erm , and7 doubling time: DT = (loge2)/rm.

Percentages of apparent mortality of each life stage (percent-age of dead individuals out of the total number of individualsthat entered that stage) and actual mortality (percentage ofdead individuals in a particular stage out of the total num-ber of individuals at the beginning of that generation) for each



developmental stage (i.e. eggs, larvae, pupae and adults) werecalculated (Southwood, 1978). The sex ratio (female : male)(Maharaj, 2003), egg-hatching period and periods from egglaying to development of larvae, pupae and adults were alsocalculated.

In order to compare findings among all the strains, all thebiological and life-table parameters of these strains were sub-jected to analysis of variance (anova) to obtain the leastsignificant difference (LSD) at P < 0.05. Regression analy-sis was conducted to reveal the relationship between fecundity(Y-function; i.e. eggs/female, egg rafts/female and eggs/eggraft) and age of female mosquitoes (X-function). The range ofannual temperature (difference between minimum winter andmaximum summer temperatures) and the mean annual rain-fall for the study areas were correlated with various life-tableattributes. Cluster analysis with the nearest neighbour methodand the squared Euclidean distance metric method (Krebs,1989) (statgraphics Plus 5.0; StatPoint Technologies, Inc.,Warrenton, VA, U.S.A.) was used to assess degrees of similar-ity among different strains according to 14 important life-tableand biological attributes of immature and adult stages, includ-ing actual mortality (percentages) of eggs, larvae and pupae,rate of adult emergence, female : male ratio, total egg produc-tion by cohort, eggs laid per female per day, life expectanciesof males and females, R0, rm , , T and DT. For the analysisof r-strategy, the ex of females, and the rm , and DT of allstrains including LAB, were considered.

Results

Mortality in different stages of immature Cx. quinquefasciatus

The actual and apparent percentages of egg mortality(Table 1) were highest in BKN (19.5%) and lowest in JDand BTH (4.4%). On these parameters, JD and BTH differedsignificantly from LAB, BKN and JMN. The LAB straindiffered significantly from BKN, but not from JMN.

The rank order of the various strains by percentage of actualmortality in the larval stages was JMN < JD < LAB < BTH< BKN (Table 1). Statistically significant differences in per-centages of both apparent and actual mortality were observedamong LAB, BKN and JMN. However, LAB did not showsignificant differences from JD and BTH for either parameter.Differences between JMN and BTH were significant for thepercentage of actual mortality, but not for the percentage ofapparent mortality.

The highest percentages for actual and apparent pupal mor-tality were recorded in JD and the lowest in BTH (Table 1).Both of these strains differed significantly from the otherstrains in this respect.

The rank order of the various strains for adult emergencewas BKN < LAB < JD < JMN < BTH; rates ranged from42.0% to 74.0% (Table 1). The female : male ratio was >1.0in all strains except BTH (Table 1).

Developmental period for different stagesof Cx. quinquefasciatus

Only LAB and BTH differed significantly from one anotheron length of egg-hatching period. The minimum and maximumperiods from egg laying to the development of pupae occurredin BKN and JD, respectively. The minimum and maximumperiods from egg laying to adult emergence occurred in LABand JD, respectively (Table 1).

Survival of adult mosquitoes

Overall, females lived longer than males (Figs 2 and 3,Table 2). Life expectancies at emergence (ex ) of males dif-fered significantly among all strains, except between JMN andJD, and BTH and LAB. In females, ex differed significantlyamong all strains, except between BKN and LAB.

Table 1. Mortality rates and other developmental parameters in immature Culex quinquefasciatus of four strains from different geographical areasand a strain raised under laboratory conditions.

ParametersLAB strain,mean SE

JD strain,mean SE

BKN strain,mean SE

JMN strain,mean SE

BTH strain,Mean SE LSD

Eggs hatching, % 86.7 1.21a 95.6 0.55b 80.5 2.40a 85.9 8.50a 95.6 2.10b 7.50Mortality ratesEggs % actual and apparent 13.3 1.21a 4.4 0.55b 19.5 2.37c 18.1 3.90ac 4.4 2.10b 5.28Larvae % actual 14.5 0.60a 13.0 1.13a 32.3 2.89b 6.6 1.02c 17.2 3.05a 4.39

% apparent 16.7 0.71a 13.6 1.25ac 40.1 4.23b 8.1 1.65c 15.2 6.94ac 8.08Pupae % actual 7.7 0.43a 14.8 0.59b 6.3 0.88a 7.0 1.18a 4.4 0.48ac 1.83

% apparent 10.6 0.49a 17.9 1.68b 13.1 2.16a 9.4 2.04a 5.0 1.62c 3.93Developmental parametersEgg-hatching period, days 1.5 0.22a 1.8 0.51ab 1.7 0.1ab 2.0 0.36ab 2.4 0.58b 0.87Eggpupa period, days 10.0 0.71ab 11.8 0.51b 8.8 1.50a 9.5 1.52ab 9.3 0.51ab 2.55Eggadult period, days 11.6 0.22a 14.8 0.51b 11.7 0.58a 12.0 0.36a 12.2 0.36a 0.97% adult emergence 64.5 1.05a 67.7 1.57ac 42.0 4.46b 68.3 5.88a 74.0 5.56c 9.33Sex ratio (female : male) 1.1 0.05a 1.0 0.04ac 1.2 0.06b 1.1 0.06a 1.0 0.03c 0.12Mean values within rows indicated by same letter do not differ significantly.LAB, laboratory; JD, Jodhpur; BKN, Bikaner; JMN, Jamnagar; BTH, Bathinda; SE, standard error; LSD, least significant difference.



0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61Days

lx

Lab JD BKN JMN BTH

Fig. 2. Survival of males in the LAB, JD, BKN, JMN and BTH strainsof Culex quinquefasciatus.

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61Days

lx

Lab JD BKN JMN BTH

Fig. 3. Survival of females in the LAB, JD, BKN, JMN and BTHstrains of Culex quinquefasciatus.

Oviposition schedule

Only JD differed significantly from the rest of the strains inoviposition days. No significant differences were recorded forpre- and post-oviposition days in any strain (Table 2).

Fecundity

The rank order of the various strains for egg productionby cohort of 50 females was LAB < BKN < BTH < JD< JMN, with significant differences among strains (Table 2).The highest and lowest numbers of egg rafts were produced byJMN and LAB, respectively. Significant differences in egg raftproduction were observed among all strains, except betweenBTH and LAB. The number of eggs per egg raft was highestin BTH (n = 207.5) and lowest in JMN (n = 161.1), withsignificant differences among all field strains, but not betweenLAB and BKN or JMN. The largest (mean 317.3 eggs/raft) andthe smallest (mean 73.0 eggs/raft) were recorded in BKN. Nosignificant differences were recorded among all field strains forlargest and smallest egg rafts, except for BTH for the smallestegg rafts. The number of eggs/female lifespan was greatest inBTH (812.3 eggs) and least in LAB (512.4 eggs), and variedsignificantly among strains. Eggs/female/day ranged from 20.5(LAB) to 30.8 (BTH), showing some significant differencesbetween strains.

Effect of ageing on fecundity

Regression slopes indicate that the mean number of egg raftsper female (Table 3, Fig. 4) and the mean number of eggs perfemale (Table 4, Fig. 5) increased significantly with increasingage of females in all strains except JD. Reductions in the meannumber of eggs per raft with increasing age of females werevariable across the strains (Table 5, Fig. 6).

Adult life-table attributes

All strains differed significantly from one another in thenet reproductive rate (R0) (Fig. 7), with the maximum occur-ring in JMN (162.2) and the minimum in LAB (107.6). Theintrinsic (rm ) and finite () rates of increase were least inLAB and greatest in BTH, with significant differences between

Table 2. Comparison of biological attributes in Culex quinquefasciatus of four strains from different geographical areas and a strain raised underlaboratory conditions.


JD strain,mean SE

BKN strain,mean SE

JMN strain,mean SE

BTH strain,mean SE LSD

Pre-oviposition, days 7.0 1.00a 6.3 0.58a 6.7 0.58a 8.0 1.73a 7.3 0.58a 2.47Oviposition, days 25.3 0.58a 21.0 1.00b 25.7 1.15a 24.7 2.08a 26.3 0.58a 2.67Post-oviposition, days 7.3 0.58a 6.7 1.15a 4.7 0.58a 4.0 2.65a 5.3 0.58a 3.42Total egg raft production by cohort 63.4 2.08a 80.0 1.00b 74.3 3.51c 100.7 2.89d 65.3 0.58a 5.45Largest egg raft 276.3 35.80a 308.0 20.07ab 317.3 19.14b 287.7 2.31ab 310.3 8.08b 32.52Smallest egg raft 83.7 3.51a 74.7 0.58ab 73.0 5.20b 77.0 1.73ab 95.3 5.77c 9.82Eggs per egg raft 169.1 2.99ab 186.3 3.49c 170.2 5.16a 161.1 2.64b 207.5 0.92d 8.03Total egg production/cohort 10 763 196a 14 903 320b 12 639 213c 16 218 196d 13 560 129e 533Eggs/female/day 20.5 0.26a 28.2 0.77b 20.5 1.91a 27.9 2.13b 30.8 0.50c 2.38Eggs/female lifespan 512.4 6.64a 573.9 1.67b 552.4 12.80c 684.3 3.67d 812.3 4.66e 17.39Life expectancy (ex ) (at emergence), days 21.2 0.15a 19.2 0.21b 18.2 0.38c 19.8 1.02b 21.3 0.40a 0.97Life expectancy (ex ) (at emergence), days 27.4 0.16a 30.8 0.28b 28.0 0.40a 36.9 0.70c 25.2 0.60d 0.81Mean values within rows indicated by same letter do not differ significantly.LAB, laboratory; JD, Jodhpur; BKN, Bikaner; JMN, Jamnagar; BTH, Bathinda; SE, standard error; LSD, least significant difference.



Table 3. Regression analysis for number of egg rafts/female (Y) as a function of age (X) of females in Culex quinquefasciatus of four strainsfrom different geographical areas and a strain raised under laboratory conditions.

Strain Source of variation Mean square Regression equation d.f. F-ratio P -value

LAB Regression 0.0424 Y = 0.0140 + 0.0026X 1 12.67 0.0011Residual 0.00335 34

JD Regression 0.00848 Y = 0.0665 + 0.0011X 1 2.39 0.1327Residual 0.00354 29

BKN Regression 0.07664 Y = 0.0002 + 0.0038X 1 12.58 0.0012Residual 0.00609 34

JMN Regression 0.13987 Y = 0.00588 + 0.004X 1 22.33 0.0001Residual 0.00626 31

BTH Regression 0.20294 Y = 0.06088 + 0.0072X 1 21.79 0.0001Residual 0.00931 31

LAB, laboratory; JD, Jodhpur; BKN, Bikaner; JMN, Jamnagar; BTH, Bathinda.

(E)

Bathinda strain BTH

Egg

rafts

/fem

ale

0 10 20 30 40 500

0.1

0.2

0.3

0.4

0.5

0.6Y = -0.061 + 0.0072X

(C)

Bikaner strain BKN

Egg

raft/

fem

ale

0 10 20 30 40 500

0.1

0.2

0.3

0.4Y = 0.002 + 0.0038X

(D)

Jamnagar strain JMN

Egg

rafts

/fem

ale

0 20 40 60 800

0.1

0.2

0.3

0.4

0.5Y = -0.0059 + 0.004X

(A)

Laboratory strain LAB

Age of female, Days

Egg

rafts

/fem

ale

0 10 20 30 40 50 600

0.05

0.1

0.15

0.2

0.25Y = 0.0140 + 0.0026X

(B)

Jodhpur strain JD

Age of female, Days

Age of female, Days

Age of female, Days

Age of female, Days

Egg

raft/

fem

ale

0 10 20 30 40 50 600

0.04

0.08

0.12

0.16

0.2

0.24 Y= 0.0666 + 0.0011X

Fig. 4. Regression line between egg rafts per female and age of female (days) in Culex quinquefasciatus in (A) LAB, (B) JD, (C) BKN, (D) JMNand (E) BTH strains.



Table 4. Regression analysis for number of eggs/female (Y) as a function of age (X) of females in Culex quinquefasciatus of four strains fromdifferent geographical areas and a strain raised under laboratory conditions.


LAB Regression 594.577 Y = 5.7396 + 0.3168X 1 5.15 0.029Residual 115.374 33

JD Regression 142.008 Y = 13.8519 + 0.1612X 1 1.07 0.3107Residual 133.205 28

BKN Regression 840.04 Y = 4.6648 + 0.4075X 1 6.06 0.019Residual 138.514 34

JMN Regression 2328.9 Y = 2.5433 + 0.5467X 1 14.51 0.0006Residual 160.481 31

BTH Regression 7102.28 Y = 9.720 + 1.3603X 1 15.65 0.0004Residual 453.736 31


(E)

Bathinda strain BTH

Age of female, Days

Egg/

fem

ale

0 10 20 30 40 500

30

60

90

120

150Y = -9.720 + 0.1.360X

(C)

Bikaner strain BKN

Age of female, Days

Eggs

/fem

ale

0 10 20 30 40 500

10

20

30

40

50

60Y = 4.6648 + 0.4075X

(D)

Jamnagar strain JMN

Age of female, Days

Eggs

/fem

ale

0 20 40 60 800

20

40

60

80Y = 2.5433 + 0.5467X

(A)


Age of female, Days

Eggs

/fem

ale

0 10 20 30 40 50 600

10

20

30

40

50Y= 5.7396 + 0.3168X

(B)

Jodhpur strain JD

Age of female, Days

Eggs

/fem

ale

0 10 20 30 40 50 600

10

20

30

40

50Y = 13.851 + 0.1612X

Fig. 5. Regression line between eggs per female and age of female (days) in Culex quinquefasciatus in (A) LAB, (B) JD, (C) BKN, (D) JMN and(E) BTH strains.



Table 5. Regression analysis for number of eggs/egg raft (Y) as a function of age (X) of females in Culex quinquefasciatus of four strains fromdifferent geographical areas and a strain raised under laboratory conditions.


LAB Regression 4427.62 Y = 145.155 0.8647X 1 1.27 0.268Residual 3488.81 33

JD Regression 189.155 Y = 126.537 0.1788X 1 0.08 0.780Residual 2397.94 29

BKN Regression 20 985.8 Y = 173.792 2.0369X 1 7.79 0.008Residual 2694.96 34

JMN Regression 3614.83 Y = 141.675 0.6811X 1 1.56 0.2217Residual 2323.96 31

BTH Regression 1824.54 Y = 187.838 0.6894X 1 0.36 0.5540Residual 5097.64 31


(E)

Bathinda strain BTH

0 10 20 30 40 500

50

100

150

200

250

300Y = 187.83-0.6864X

Eggs

/egg

raft

(C) (D)

Jamnagar strain JMN

0 20 40 60 800

40

80

120

160

200

240Y = 141.675-0.6811X

Eggs

/egg

raft

Bikaner strain BKN

0 10 20 30 40 500

50

100

150

200

250

300Y =173.792-2.0369X

Eggs

/egg

raft

(B)

Jodhpur strain JD

Age of female, Days0 10 20 30 40 50 60

0

40

80

120

160

200

240Y=126.537-0.1788X

Eggs

/egg

raft

(A)


Age of female, Days

Age of female, DaysAge of female, Days

Age of female, Days

0 10 20 30 40 50 600

40

80

120

160

200

240Y = 145.155-0.8647X

Eggs

/egg

raft

Fig. 6. Regression line between eggs per egg raft and age of female (days) in Culex quinquefasciatus in (A) LAB, (B) JD, (C) BKN, (D) JMNand (E) BTH strains.



020406080

100120140160180

1 6 11 16 21 26 31 36 41 46 51 56 61Days

Net

repr

oduc

tion

rate

Lab JD BKN JMN BTH

Fig. 7. Cumulative net reproductive rate (R0) in Culex quinquefascia-tus in the LAB, JD, BKN, JMN and BTH strains.

JD and BKN and between BTH and JMN, but no significantdifferences between JD and BTH or between BKN and JMN.The mean generation time (T ) was longest in LAB and shortestin BTH. The LAB strain differed significantly from JD andBTH for T , but not from BKN and JMN. The longest doublingtime (DT ) was observed in LAB, followed by JMN, BKN, JDand BTH, but the differences between strains was not alwayssignificant (Table 6).

Cluster analysis

Analysis of all Cx. quinquefasciatus strains indicated twoclusters, consisting of BKN and LAB, and JD, BTH andJMN, respectively (Fig. 8). The distance between the clus-ters was 22.20. Cluster analysis also indicated a closerrelationship between JD and BTH than between JMNand BKN.

Correlation analysis between environmental factorsand life-table attributes

Table 7 indicates that annual temperature ranges in the fourfield locations were strongly correlated with egg production,female life expectancy, R0, rm , and T (0.770.97). Meanannual rainfall also correlated strongly with egg production,

Table 6. Comparison of adult life-table attributes in Culex quinquefasciatus of four strains from different geographical areas and a strain raisedunder laboratory conditions.


JD strain,mean SE

BKN strain,mean SE

JMN strain,mean SE

BTH strain,mean SE LSD

R0 107.6 1.96a 147.4 2.83b 122.9 7.89c 162.2 1.96d 135.6 1.29e 8.76rm 0.15 0.00a 0.20 0.00b 0.17 0.00c 0.16 0.01ac 0.21 0.02b 0.02 1.16 0.00a 1.23 0.00b 1.18 0.00a 1.17 0.01a 1.24 0.02b 0.02T 31.6 0.80a 24.5 0.38b 28.8 0.70a 31.4 2.08a 22.9 1.67b 2.94DT 4.7 0.11a 3.4 0.05b 4.1 0.05c 4.3 0.28c 3.2 0.24b 0.24Mean values within rows indicated by same letter do not differ significantly.LAB, laboratory; JD, Jodhpur; BKN, Bikaner; JMN, Jamnagar; BTH, Bathinda; SE, standard error; LSD, least significant difference; R0, netreproductive rate; rm , intrinsic rate of increase; , finite rate of increase; T , mean generation time; DT, doubling time.

female life expectancy and R0 (0.850.91), but relativelyweakly with rm , and T (0.460.59).

Comparison of r-strategy among the different strains

Among the different field strains, JMN had the longestfemale survival period, DT and the lowest rm and incomparison with BTH, BKN and JD, indicating that JMNwas superior in r-strategy. LAB appeared to be a weakerr-strategist than all the field strains because it required thelongest DT and had the least rm and .

Discussion

The life-table parameters and other biological attributes infive strains of Cx. quinquefasciatus originating from differentgeographical areas varied with ecological conditions.

Apparent and actual mortality in eggs was found to rangefrom 4.4% (JD and BTH) to 19.5% (BKN); values in the othertwo strains were relatively close to that in BKN. Rates of actualmortality of larval stages ranged from 6.6% (JMN) to 32.3%(BKN), with values for the other strains falling between theseextremes. The cumulative effect of mortalities of eggs, larvaeand pupae resulted in lower rates of adult emergence, rangingfrom 42.0% (BKN) to 74.0% (BTH); values for the otherstrains fell within 10% of that for BTH. The high mortality inBKN may be associated with its habitat, which is very proneto drought and is almost completely characterized by desertconditions. By contrast, JMN has a significantly higher rate ofadult emergence than BKN despite a non-significant differencein egg hatching because the conditions in which it lives providebetter support for larval and pupal development. In otherstudies, rates of adult emergence were low for four strains ofAnopheles sacharovi Favr (Diptera: Culicidae) (11.414.6%)(Yurttas & Alten, 2006), but exceeded 50% from first instarsof field-collected Anopheles gambiae Giles s.s. from BurkinaFaso (Lehmann et al., 2006). These studies revealed thatvariations in immature mosquito developmental rates dependon habitat conditions.

The percentage of eggs hatching has been reported tobe 82.1% in Culex pipiens fatigans (synonymous withCx. quinquefasciatus; Gomez et al., 1977) and 79.6% in



0

4

8

12

16

20

24D

ista

nce

Lab JD

BKN

JMN

BTH

Fig. 8. Dendrogram of the four field strains (JD, BKN, JMN andBTH) and a laboratory strain (LAB) of Culex quinquefasciatusshowing similarities in various life-table attributes. The distancecoefficient is measured using a squared Euclidean method.

Table 7. Correlation analysis of various life-table and biologicalattributes for the JD, BKN, JMN and BTH strains of Culexquinquefasciatus for temperature range and mean annual rainfall.

Correlation coefficient (r )Attributes Temperature range Mean annual rainfall

Egg production/cohort 0.79 0.90Female life expectancy 0.97 0.86R0 0.76 0.91rm 0.82 0.51 0.77 0.46T 0.87 0.59

R0, net reproductive rate; rm , intrinsic rate of increase; , finite rate ofincrease; T , mean generation time.

Cx. quinquefasciatus (Suleman & Reisen, 1979), which areclose to the values reported in the present study (Table 1).Variations in percentages of eggs hatching have also beenrecorded in Cx. tritaeniorhynchus, for which higher ratesemerged in Indian subcontinental strains (93.0%) than ina Japanese strain (78.8%) (Reisen et al., 1979). However,significantly lower egg-hatching rates were recorded in fourstrains of An. sacharovi from southeastern Anatolia, Turkey,leading to high overall real mortality rates (85.488.6%;Yurttas & Alten, 2006). The eggs of strains collectedin the present study from the most arid environments(BKN, JD) required a shorter hatching period than thoseof semi-arid strains (JMN, BTH), which suggests that shorthatching times may be a positive factor for survival in aridconditions.

In the present study, the mean development time from eggto adult was found to vary little, with four of the strainsshowing a range of


Numbers of eggs/raft varied significantly among strains ofCx. quinquefasciatus (Table 2), ranging from 161.1 (JMN) to207.5 (BTH), whereas the mean number of rafts/cohort wasnearly opposite, ranging from 65.3 (BTH) to 100.7 (JMN). Thisis not surprising because high numbers of eggs/raft probablyreflect the production of relatively few rafts/female and viceversa. The construction of large egg rafts may consume morereproductive energy and time than the construction of smalleregg rafts and thus females that laid small egg rafts wereable to produce more eggs over their lifetime. Significantdifferences in mean numbers of eggs/raft were observed amongall four field-collected strains, but values in BKN and JMNdid not differ significantly from that in LAB. Variations inthe mean number of eggs/raft among different strains of Culexmosquitoes have been reported previously (Walter & Hacker,1974; Reisen et al., 1979, 1984).

Interestingly, the strains from areas with the greatestrange of annual temperatures (Bathinda, Bikaner and Jodh-pur) face tougher conditions for survival in nature and thesemosquitoes typically had a shorter female life expectancy(25.230.8 days), laid eggs on most days (21.026.3 days) andconstructed larger egg rafts (170.2207.5 eggs/raft), indicatingthat these strains invest the most energy in rapid reproduction.By contrast, the Jamnagar strain, which comes from the areawith the smallest range of annual temperatures and hence moreoptimal conditions for mosquitoes, had the longest female lifeexpectancy (mean: 36.9 days), but a shorter oviposition period(mean: 24.7 days) and laid fewer eggs/raft (mean: 161.1). Fox& Czesak (2000) reviewed the evolutionary ecology of progenysize and concluded that more hostile environmental conditionsmay favour the production of a greater number of offspringin order to increase the probability that some will survive toadulthood. Thus, individuals from unfavourable environmentsare inherently committed to put maximum effort into the pro-duction of more offspring to ensure that an adequate numbersurvive to reproduce in their turn.

In the present investigation, the mean number of eggrafts/female (Fig. 4, Table 3) and the mean number ofeggs/female (Fig. 5, Table 4) increased with female age. Themean number of eggs/raft, however, decreased with femaleage (Fig. 6, Table 5). These observations indicate that cumu-lative fecundity increased as a function of female age asa result of an increase in the frequency of the depositionof progressively smaller rafts. Similarly, Walter & Hacker(1974) found the number of eggs/raft to decrease with increas-ing age in a range of geographically distinct strains ofCx. quinquefasciatus. Suleman & Reisen (1979) recorded sim-ilar observations in Cx. quinquefasciatus and Reisen et al.(1984) did so in Cx. tarsalis.

Species with relatively high net reproductive rates (R0) tendto have a competitive advantage over sympatric species thatoccupy similar ecological niches. Gubler (1970) observed thecompetitive displacement of Aedes polynesiensis Marks byAedes albopictus (Skuse) (Diptera: Culicidae), and showedthat the latter species was superior in female survival and netreproductive rate. In the present investigations, the R0 values inall wild strains (122.9162.2) were significantly greater than inLAB (107.6) and differed significantly amongst one amother,which suggests that varying degrees of fitness of the wild

strains may have been imposed by different levels of selectionpressure on the reproductive rate for each strain. Similarly,significant differences in R0 for Cx. quinquefasciatus havebeen reported in strains from Bangkok (8.1), Vero Beach (24.3)and Houston (37.7) (Walter & Hacker, 1974), from Caracas,Venezuela (80.7) (Gomez et al., 1977) and from Peshawar,Pakistan (58.6) (Suleman & Reisen, 1979); however thesevalues are all lower than those found in the present study.Reisen et al. (1979) observed variations inR0 in nine strainsof Cx. tritaeniorhynchus, with the highest values reported instrains from Karachi (98.5) and Taipei (84.9), and the lowestreported in three strains from Japan (14.431.2). The presentand referenced data show a consistent trend for mosquitopopulations in colder climates to have lowerR0 values thanpopulations in warmer climates. Yurttas & Alten (2006) alsofound significant differences in R0 (range: 3.212.8) amongdifferent strains of An. sacharovi, which are again, lower thanthose in the present study.

In the current study, differences in rm (0.150.21), (1.161.24) and DT (3.234.67) were observed amongthe various strains of Cx. quinquefasciatus and some ofthese differences were significant. Similarly, Walter &Hacker (1974) found significant differences in rm betweena Cx. quinquefasciatus strain from Bangkok and two otherstrains (Vero Beach and Houston), although the latter twostrains did not differ significantly in rm from each other.The rm values in LAB, BKN and JMN are closest to thatin the Caracas strain of Cx. p. fatigans (Gomez et al., 1977),but much higher than the value in the Peshawar strain ofCx. quinquefasciatus (Suleman & Reisen, 1979). Reisen et al.(1979) reported greater differences in rm among various strainsof Cx. tritaeniorhynchus from the Indo-Pakistani subcontinent(0.160.19) and Taiwan (0.18) than among Japanese strains(0.080.12). This appears to reflect a greater need on the partof the Japanese strains for a longer mean generation period(28.229.8 days) than the Indo-Pakistani (22.224.1 days) andTaiwanese (24.0 days) strains. A comparison of these dataindicates that warmer climates have a positive effect on rm ,increasing inherent capacity for reproduction.

In the present study, the mean generation time (T ) rangedfrom 22.9 days to 31.6 days among the various strains ofCx. quinquefasciatus. Higher values for T (39.6 days and44.7 days) were recorded by Gomez et al. (1977) and Suleman& Reisen (1979) and lower values by Walter & Hacker (1974)in the Bangkok (32.1), Vero Beach (26.8) and Houston (28.5)strains of Cx. quinquefasciatus.

The cluster analysis of various life-table and biologicalattributes of the four wild strains of Cx. quinquefasciatusrevealed that JD was biologically closer to BTH than to BKNand JMN, which may reflect cumulative differences in theirrespective ecologies.

Although Bathinda (BTH) and Jamnagar (JMN) are 1000km apart, they both lie in the same semi-arid zone. Averageannual rainfall is greater in Jamnagar than in Bathinda and RHis also greater for most of the year in Jamnagar as a resultof its coastal location. The range of annual temperatures isnarrower for Jamnagar than Bathinda. Accordingly, we mightexpect that environmental conditions in Jamnagar would favoursurvival and reproductive potential and would support a shorter



egg-hatching period (1.96 days), longer female life expectancy(36.9 days), a higher net reproductive rate (162.2) and a longermean generation time (31.4 days) than conditions in BTH(2.42 days, 25.2 days, 135.6 days and 22.9 days, respectively).

Jodhpur and Bikaner both lie in arid zones. Although Jodh-pur has relatively higher rainfall than Bikaner, these siteswould be expected to be less favourable than Jamnagar andBathinda for mosquito survival and reproduction. The positivecorrelations between rainfall and various life-table parame-ters for the wild strains indicate that environmental factorsdo indeed have an important evolutionary effect on mosquitolife strategies that can be detected even after several gen-erations of colonization. The JD, BKN and BTH strains ofCx. quinquefasciatus have relatively low DT as a result ofhigher rm and , which perhaps reflects adaptations to moreextreme environmental conditions than occur in Jamnagar,whereas the JMN strain had lower rm and values and higherDT, despite high R0 and T values. By contrast, the LAB strainhad a relatively high DT value as a result of having the low-est R0, rm and in comparison with the four wild strains.Crovello & Hacker (1972) also reported variation in the repro-ductive potential of 13 strains of Aedes aegypti L. from a rangeof habitats and found forest strains to show the lowest repro-ductive potential as a result of more homogeneous levels ofenvironmental stress than found in urban and semi-urban areas,where there is a greater degree of fluctuation in environmentalconditions. The evolution of an r-strategy has been describedfor Cx. quinquefasciatus by Rabinovich (1974) and supportedby Gomez et al. (1977) and Suleman & Reisen (1979). Reisenet al. (1979) also showed that strains of Cx. tritaeniorhynchusfrom Taiwan and the Indo-Pakistan subcontinent were morer-selected as a result of the more variable ecological sys-tems in which they live compared with Japanese strains, whichoriginated from more stable, cooler and more humid maritimeclimates. These strains of Cx. tritaeniorhynchus were lesser r-strategists than domestic species, such as Cx. quinquefasciatusand Ae. aegypti (Schlosser & Buffington, 1977). Analysis ofthe various parameters in the present study indicates that,among the wild strains, BTH, BKN and JD are more efficientr-strategists as they face more difficult environmental con-ditions than does JMN, which exists in more homogeneousenvironmental conditions. Interestingly, Suman et al. (2009)showed the JMN strain to differ from the JD, BKN and BTHstrains of this mosquito for various egg-related attributes. Fur-thermore, higher R0, rm and values and lower T and DTvalues were observed in the wild strains than in LAB, which isconsistent with the hypothesis that harsher environmental con-ditions select for high r-strategists in the field. Strains with highr-strategies may be more difficult to control because, when apopulation is diminished, it is able to re-establish more quicklythan a less able r-strategist.

The results of this study extend our knowledge of the impactof the environment on a wide range of life history traits in localpopulations of disease vectors. The study also provides detailedinformation which is valuable for increasing the accuracy ofthe population dynamic models that are an essential componentof the decision-making tools used to plan and monitor vectorcontrol programmes (World Health Organization, 2004).

Acknowledgements

We thank Dr R Vijayaraghavan, Director, Defence Researchand Development Establishment, Gwalior, for his support andinterest in this work.

References

Andrewartha, H.G. & Birch, L.C. (1954) The Distribution and Abun-dance of Animals. University of Chicago Press, Chicago, IL.

Bancroft, T.L. (1899) On the metamorphosis of the young formof Filaria bancrofti Cobb. Journal and Proceedings of the RoyalSociety of New South Wales, 82, 6265.

Belen, A. & Alten, B. (2006) Variation in the life-table characteristicsamong populations of Phlebotomus papatasi at different altitudes.Journal of Vector Ecology, 31, 3544.

Chandra, G., Chatterjee, S.N., Banerjee, B.D. & Majumdar, G. (1997)Effect of seasonal variations on the development of Wuchererialarvae in Culex quinquefasciatus. Basic and Applied Biomedicine,5, 2124.

Costero, A., Edman, J.D., Clark, G.G. & Scott, T.W. (1998) Life tablestudy of Aedes aegypti (Diptera: Culicidae) in Puerto Rico fedonly human blood plus sugar. Journal of Medical Entomology, 35,809813.

Crovello, T. & Hacker, C. (1972) Evolutionary strategies in life tablecharacteristics among feral and urban strains of Aedes aegypti.Evolution, 26, 185196.

Day, J.F., Edman, J.D. & Scott, T.W. (1994) Reproductive fitness andsurvivorship of Aedes aegypti (Diptera: Culicidae) maintained onblood, with field observations from Thailand. Journal of MedicalEntomology, 3, 611617.

Fox, C.W. & Czesak, M.E. (2000) Evolutionary ecology of progenysize in arthropods. Annual Review of Entomology, 45, 341369.

Gomez, C., Rabinovich, J.E. & Machado-Allison, C.E. (1977) Popu-lation analysis of Culex pipiens fatigans Weid. (Diptera: Culicidae)under laboratory conditions. Journal of Medical Entomology, 13,453463.

Gubler, D.J. (1970) Comparison of reproductive potentials of Aedes(Stegomyia) albopictus Skuse and Aedes (Stegomyia) polynesiensisMarks. Mosquito News, 30, 201209.

Howe, R.W. (1953) The rapid determination of the intrinsic rate ofincrease of an insect population. Annals of Applied Biology, 40,134151.

Jinfu, W., Shaohong, L., Rui, C. & Lingling, W. (1998) Relativefitness of three organophosphate-resistant strains of Culex pipienspallens (Diptera: Culicidae). Journal of Medical Entomology, 35,716719.

Krebs, C.J. (1989) Ecological Methodology. Harper Collins Publishers,New York, NY.

Lehmann, T., Dalton, R., Kim, E.H., Dahl, E., Diabate, A., Dabire, R.& Dujardin, J.P. (2006) Genetic contribution to variation in larvaldevelopment time, adult size, and longevity of starved adults ofAnopheles gambiae. Infection, Genetics and Evolution, 35, 410416.

Maharaj, R. (2003) Life table characteristics of Anopheles arabiensis(Diptera: Culicidae) under simulated seasonal conditions. Journal ofMedical Entomology, 40, 737742.

Mahmood, F., Reisen, W.K., Chiles, R.E. & Fang, Y. (2004) Westernequine encephalomyelitis virus infection affects the life tablecharacteristics of Culex tarsalis (Diptera: Culicidae). Journal ofMedical Entomology, 41, 982986.


Culex quinquefasciatus life tables 13Manson-Bahr, P.E.C. & Bell, D.R. (2003) Mansons Tropical Disease.

ELBS Publishers, London.Rabinovich, J.E. (1974) Demographic strategies in animal populations:

a regression analysis. Tropical Ecological Systems, EcologicalStudies, Vol. 11 (ed. by F. B. Golley & E. Medina), pp. 1940.Springer-Verlag, New York, NY.

Reisen, W.K. & Milby, M.M. (1987) Studies on autogeny in Culextarsalis : 3. Life table attributes on autogenous and anautogenousstrains under laboratory conditions. Journal of the AmericanMosquito Control Association, 3, 619625.

Reisen, W.K., Siddiqui, T.F., Aslam, Y. & Malik, G.M. (1979) Geo-graphic variation among the life table characteristics of Culex tri-taeniorhynchus from Asia. Annals of the Entomological Society ofAmerica, 72, 700709.

Reisen, W.K., Milby, M.M. & Bock, M.E. (1984) Effects of immaturestress on selected events in the life history of Culex tarsalis.Mosquito News, 44, 385395.

Schlosser, I.J. & Buffington, J.D. (1977) The energetics of r - and k -selection in two African strains of Aedes aegypti. Annals of theEntomological Society of America, 70, 196202.

Sirivanakaran, S. (1976) Medical entomology studies III. A revisionof the subgenus Culex in the Oriental region (Diptera: Culicidae).Contributions of the American Entomological Institute, 12, 1272.

Slobodkin, L.B. (1962) Growth and Regulation of Animal Populations.Holt, Rinehart & Winston, New York, NY.

Southwood, T.R.E. (1978) Ecological Methods, with Particular Refer-ences to the Study of Insect Populations, 2nd edn. John Wiley &Sons, New York, NY.

Suleman, M. & Reisen, W.K. (1979) Culex quinquefasciatus Say:life table characteristics of adults reared from wild-caught pupaefrom North West Frontier Province, Pakistan. Mosquito News, 39,756762.

Suman, D.S., Shrivastava, A.R., Parashar, B.D., Pant, S.C., Agrawal,O.P. & Prakash, S. (2009) Variation in morphology and morphomet-rics of eggs of Culex quinquefasciatus mosquitoes from ecologicalregions of India. Journal of Vector Ecology, 34, 191199.

Walter, M.N. & Hacker, C.S. (1974) Variation in the life tablecharacteristics among three geographic strains of Culex pipiensquinquefasciatus. Journal of Medical Entomology, 11, 541550.

World Health Organization (2004) Global Strategic Frameworkfor Integrated Vector Management, WHO/CDS/CPE/PVC/2004.10.WHO, Geneva.

World Health Organization (2006) Global programme to eliminatelymphatic filariasis. Weekly Epidemiological Record, 81, 22232.

Yurttas, H. & Alten, B. (2006) Geographic differentiation of lifetable attributes among Anopheles sacharovi (Diptera: Culicidae)populations in Turkey. Journal of Vector Ecology, 31, 275284.

Accepted 1 November 2010


life tables for culex pippiens suman ds

Documents

life tables

lifetable attributes

female life expectancy

life reproductive strategies

correlated r

bth clusteredtogether

bth to19

strain development