MOLECULAR ECOLOGY AND EVOLUTION Temperature-Dependent Development of Chilocorus bipustulatus (Coleoptera: Coccinellidae) 1 2 3 P. A. ELIOPOULOS, D. C. KONTODIMAS, AND G. J. STATHAS Environ. Entomol. 39(4): 1352Ð1358 (2010); DOI: 10.1603/EN09364 ABSTRACT The effect of temperature on development and survival of Chilocorus bipustulatus L. (Coleoptera: Coccinellidae), a predator of many scale insects, was studied under laboratory condi- tions. The duration of development of egg, Þrst, second, third, and fourth larval instars, pupa, and preovioposition period at seven constant temperatures (15, 17.5, 20, 25, 30, 32.5, and 35ЊC) was measured. Development time decreased signiÞcantly with increasing temperature within the range 15Ð30ЊC. Survival was higher at medium temperatures (17.5Ð30␱C) in comparison with that at more extreme temperature regimens (15 and Ͼ30␱C). Egg and Þrst larval instars were the stages where C. bipustulatus suffered the highest mortality levels at all temperatures. The highest survival was recorded when experimental individuals were older than the third larval instar. Thermal requirements of development (developmental thresholds, thermal constant, optimum temperature) of C. bipustulatus were estimated with application of linear and one nonlinear models (Logan I). Upper and lower developmental thresholds ranged between 35.2Ð37.9 and 11.1Ð13.0ЊC, respectively. The optimum temperature for development (where maximum rate of development occurs) was estimated at between 33.6 and 34.7ЊC. The thermal constant for total development was estimated 474.7 degree-days. KEY WORDS Chilocorus bipustulatus, biological control, development, temperature, thermal constant The armored-scale ladybeetle Chilocorus bipustulatus Although developmental thresholds and day-de- L. (Coleoptera: Coccinellidae) is one of the most gree requirements for development of many aphid and important predators of diaspidids, with intense pres- mealybug coccinellid predators have been estimated, ence not only in Greece (Argyriou et al. 1976; Argy- only a few data are available as far as diaspidid pred- riou and Katsoyannos 1977; Stathas et al. 2003, 2009) ators are concerned (Podoler and Henen 1983, Pon- but in the whole Palearctic region (Bodenheimer sonby and Copland 1996, Stathas 2000). 1951, Avidov and Harpaz 1969, Viggiani 1985). Apart The lack of experimental data about the effect of from armored scales, it may alternatively feed on other temperature on the development of C. bipustulatus, its scale families such as Coccidae, Asterolecaniidae (Yi- importance as a biocontrol agent against diaspidids, non 1969), and Pseudococcidae (Avidov and Yinon and the signiÞcance that the effect of temperature on 1969). insect development has for biological control made us Prey species of C.bipustulatus are among the most determined to estimate the lower and upper devel- important pests on perennials and ornamental plants, opmental thresholds, as well as the optimum temper- common in all regions in the world. The presence of ature and day-degree requirements, for development armored scales on fruits and ornamentals affects their of this predator. Estimation of lower developmental appearance, rendering them unmarketable. They may threshold (t) and day-degree requirements for devel- also cause discolorations, deformations, skin scleriÞ- opment (or thermal constant [K]) for each develop- cation, and cavities on fruit surface. On the leaves, mental stage was undertaken by Þtting experimental they cause chlorosis, reduce photosynthesis and tran- data on a linear model. The upper developmental spiration, and encourage leaf senescence and abscis- threshold (Tm) and optimum temperature for devel- sion, followed by branch dieback, as a result of plant opment (To) were calculated by application of a non- sap sucking through their long, Þlamentous mouth- linear model (Logan I). Most of these parameters are parts (Katsoyannos 1996, Gill 1997). novel for C. bipustulatus. 1 Corresponding author: Technological Educational Institute of Materials and Methods Larissa, Department of Crop Production, 41 110 Larissa, Greece (e-mail: [email protected]). Chilocorus bipustulatus was reared on the oleander 2 Benaki Phytopathological Institute, Department of Agricultural scale Aspidiotus nerii Bouche (Hemiptera: Diaspidi- Entomology and Zoology, 8 St. Delta Str., 14561 KiÞssia, Greece. ´ 3 Technological Educational Institute of Kalamata, Department of dae) infesting potato tubes. The initial population of Crop Production, 24 100 Kalamata, Greece. both predator and prey originated from olive orchards 0046-225X/10/1352Ð1358$04.00/0 ᭧ 2010 Entomological Society of America August 2010 ELIOPOULOS ET AL.: DEVELOPMENT OF C. bipustulatus 1353 in Attiki, Greece. The duration of separate develop- 2 2 ym s S.Eb mental stages (egg, Þrst, second, third, and fourth S.E ϭ ͱ ϩ ͩ ͪ t b N ϫ y 2 b larval instars, and pupa) and adult preoviposition pe- m riod of C. bipustulatus was measured at seven constant S.E Ϯ Ϯ Ϯ Ϯ Ϯ ϭ b temperatures (15 1, 17.5 1, 20 1, 25 1, 30 and S.EK 2 1, 32.5 Ϯ 1, and 35 Ϯ 1ЊC). The relative humidity was b 65 Ϯ 5% and the photoperiod was 16-h light: 8-h where ym is the sample mean (mean value rate of darkness at all temperatures. All experimental indi- development as derived from the linear equation y ϭ viduals were kept in growth chambers with the desired a ϩ bT), b is the linear equation parameter, s2 is the temperature regimen according to experimental de- residual mean square, and ⌵ is the sample size (num- sign. All dishes with individuals of the same temper- ber of different temperatures used during study). Cal- ature treatment were kept in the same chamber and 2 culation of ym,s, and SEb was conducted with the were observed at the same time. statistical program JMP (SAS Institute 2007). To get experimental individuals, mated females Experimental data from all temperatures were transferred to oviposition cages and left undis- (15Ð32.5␱C) were Þtted to Logan I nonlinear model turbed for 24 h to lay their eggs on A. nerii. Afterward, (Logan et al. 1976): females were taken away, and eggs were collected and Tm Ϫ T put individually on petri dishes (9 cm in diameter and ␳T ␳Tm ͑ ͒ ϭ ␺ ϫ Ϫ ⌬ 1.6 cm high). Special care was taken so that experi- d T ͫe e T ͬ mental individuals were always supplied with an ex- cess of prey. Experimental individuals were checked where d(T) is the rate of development at temperature daily (every 12 h for individuals kept at 32.5 and 35ЊC), ⌻ (oC) (dÐ1), ␺ is a directly measurable rate of tem- and developmental stage was recorded. Data were perature-dependent physiological process at some ␳ taken from 25 individuals except those at 15, 32.5, and base temperature Tb, is a composite Q10 value for 35ЊC, where 50 individuals were studied (because of enzyme-catalyzed biochemical reactions, ⌻m is a max- increased egg mortality). imum temperature at which life processes can no For the preoviposition period study, newly emerged longer be maintained for prolonged periods of time, females were collected from the culture, paired with and ⌬⌻ is a temperature range, above the optimum 7- to 10-d-old males, and put on petri dishes with temperature for development and below Tm, over excess of prey. Pairs were checked daily, and initiation which the “thermal breakdown” becomes the over- of egg-laying was recorded. The study on the duration riding inßuence. of preoviposition period was carried out on 12 females The base temperature Tb is the minimum temper- Њ Њ ature used in the experiments, is inserted as a Þfth at 15 and 35 C, 14 females at 17.5 C, 15 females at 20, ␱ 30, and 32.5ЊC, and 16 females at 25ЊC. equation parameter, or can hypothesized as ⌻b ϭ 0 C. Lower and Upper Developmental Thresholds. Es- As stated by Lactin et al. (1995) and Got et al. (1997), timation of the lower developmental threshold was the model effectiveness is not affected by the hypoth- based on a linear model. Data obtained from the ex- esis of zero Tb. Taking also into account that the model periments were described by the linear regression is simpliÞed and easier for application, Tb was zero in ␺ ␳ ⌻ ⌬⌻ equation of the form: this study. Parameters , , m, and are not mea- sured directly but are estimated as parameters of non- y ϭ a ϩ bT linear regression where the Logan I equation is used as model. This nonlinear regression was carried out with the statistical program JMP (SAS Institute 2007). where y is the rate of development at temperature T The optimum temperature for development (⌻␱) and a and b are constants. Equation constants (a, b) was calculated by the equations of Logan et al. (1976): were estimated with least squares method using the statistical program JMP (SAS Institute 2007). Estima- ␧ ϫ ln( bo) tion of constants was based only on data obtained at To ϭ Tm ϫ ͩ1 ϩ␧ϫ ͪ, ␱ 1 Ϫ ␧ ϫ b 15Ð30 C, because the assumed linear relationship is o known to hold as an approximation for the median ⌬T ␧ϭ ϭ ␳ ϫ temperature range only (Campbell et al. 1974). Out- where: and bo Tm side of this range, the relationship is nonlinear (Mills Tm 1981). Statistical Analysis. Data (duration of develop- The regression line was extrapolated back and met ment) were subjected to analysis of variance the absicca at the developmental zero t, which was (ANOVA) with ␣ ϭ 0.05. Means were separated using calculated from t ϭϪa/b. The total quantity of ther- the TukeyÐKramer honestly signiÞcant difference mal energy required to complete development, the (HSD) test (Sokal and Rohlf 1995). The differences in thermal constant K, was calculated from the reciprocal survival percentages during total development among of the slope of the regression line (1/b) (Wiggles- treatments were determined by GLIMMIX procedure worth 1953, Campbell et al. 1974). (SAS Institute 2008).