Jurnal Natural Vol. 10, No. 1, 2010

FEEDING PREFERENCE S OF polytes L. L ARVAE ON RUTACEOUS HOST PLANTS

Suwarno

Department of Biology Faculty of Mathematics and Natural Sciences, Syiah Kuala University, Darussalam - Banda Aceh

Abstract . The food preference of the Papilio polytes L. larvae on four rutaceous host plants, aurantifolia (Chrism.) Swing, Citrus reticulata Blanco, Citrus hystrix DC, and Murraya koenigii (L.) Sprengle, were investigated on July to December 2006, in the laboratory of entomology School of Bio logical Science University Sains Malaysia Penang . The two-choice and four-choice food preference tests of the larvae on host plants leaves were conducted. The first instar of P. polytes larvae significant preferential feeding ( P < 0.05) for a host plant in all paired host plant leaves combinations tested, except for the C. hystrix and C. reticulata combination. Furthermore, the third instar larvae had significant preference for leaves of the three Citrus species over M. koenigii, while the third instar larvae showed no significant feeding preference among the Citrus spp. Indiscriminately feeding shown on the fifth instar larvae on leaves of all Citrus species but consistently consumed less M. koenigii compared to Citrus spp. in th e various combinations tested. The first and the third instar larvae of P. polytes were more selective in terms of feeding than the fifth instar. The nitrogen content in leaves of all four host plants differed significantly ( P < 0.01), with C. reticulata was the highest (4.52%), while the water content of leaves was significantly the lowest in M. koenigii (71.72%), compared to the three Citrus species (76.38 – 79.12%) among which the water content did not significantly differ.

Key words : Food preference, host plant, larvae, Papilio polytes , nitrogen and water content

detoxification by larvae [1,4 ]. The compensatory I. INTRODUCTION response of P. polytes on host plant quality typically falls into three different categories : Papilio polytes L. ( : Papilionidae) is increased rates of food intake, consumption of known to feed on various genera of . In additional plant tissue that complements the the forests, the larvae of P. polytes was found on limiting nutrient, and increasing digestive Murraya, Triphasia, Glycosmis, Zanthoxylum, efficiency of limiting nutrient [6]. Toddalia, Evodia, and Poncirus . In recent years, this has adapted to the urban/suburban The ability of immature i nsects to discriminate host environments and is found to feed on citrus plants for feeding based on the quality of the host plant [1]; this feeding adaptation has made P. polytes a can be important in crop pests because of the role potentially serious pest in citrus orchards. of non-crop plants as alternate hosts [8]. The larvae of Manduca sexta L. that feed on an inferior host Most obviously, the host plant is one of the most plant would be moving to search for a more suitable important environmental aspects for the developing host [9]. Many herbivores change the quality of offspring [2]. Adults butterfly exhibit oviposition their host plants, affecting both inter - and intraspecific preferences corresponding to chemicals interactions [10]. experienced during development [3,4]; and hosts plants may differ chemically [5]. More subtle Understanding the feeding preference of various differences in nutrients, nitrogen and water content host species of P. polytes is an important are also present between host species, affecting life prerequisite for developing pest management cycle and larval performance of P. polytes [6,7]. strategies for this insect. Various studies have investigated the relationship between adult host The utilization of one or the othe r alternative host preference and larval performance by using this plant on is it potentially involves assumption [11, 12]. The objective of the present differences in host searching by females, host study was to eval uate the feeding preference and recognition by females and larvae, metabolization and Feeding preferences of Papilio polytes L. larvae on rutaceous host plants (Suwarno) ______selectivity of different larval stages of P. polytes on First instar larvae four host plant species. Two-choice test II. MATERIALS AND METHODS The methodology for this test was modified after The insect DiTommaso and Losey [13] and Ghumare and Mukherjee [14] and the test was conducted for a 24 Larvae of P. polytes were collected on July 2006 h period. All possible paired combinations for the from Citrus trees on the campus of Universiti Sains two-choice test of the host plant leaves ( C. Malaysia, Penang. The collected larvae were reared aurantifolia – C. hystrix, C. aurantifolia – C. on Citrus microcarpa in the laboratory condition at reticulata, C. aurantifolia – M. koenigii, C. hystrix 24-26 oC, 60-85% relative humidity, and a – C. reticulata, C. hystrix – M. koenigii and C. photoperiod of L12:D12, until adult emergence. reticulata – M. koenigii ) were evaluated using petri The larvae were placed in a screen cage measuring dishes (each 10 cm diameter). One of very young 50 x 50 x 50 cm. Ten pairs of the emerged P. leaf of each Citrus species (approximately 2 cm polytes adults were transferred to a field cage (4 m long, 1 cm wide, and weighed 10-12 mg), and three width, 6 m length and 2 m high) to facilitate mating young foliages of M. koenigii approximately equal and oviposition. of sp. were supplied in size and weight to the Citrus leaves were used as food for the caged adult P. polytes . The flowers for the first instar larvae experiment. The leaves contained in a glass jar filled with water were were collected from the respective plants and the sprayed with 10% sucrose solution twice daily to end part of their petiole was wrapped with ensure sufficient provision of nectar, and placed in moistened cotton to prevent dehydration prior to the cage; the were replaced every two days. placing them in the petri dish. Thereafter, twenty newly hatched P. polytes larvae starved for 4-5 h Host-plants were introduced in each petri dish containing the respective combination of leaves and allowed to Selection of the rutaceous host plants for this study feed on the leaves for 24 h period. The leaves in was based on their potential value as commercial each petri dish were weighed at the beginning of plants. Three commercially grown Citrus species, the experiment and reweighed 24 h post feeding by [C. aurantofolia (Christm.) Swing., C. hystrix DC ., the first instar . The water loss from the leaves C. reticulata Blanco], and a partially domesticated in each petri dish due to evaporation was estimated rutaceous species, M. koenigii (L.), were used in by placing a piece of young leaf of each Citrus spp. this experiment. Twenty young seedlings of each (and three young foliages in the case of M. species were used as food resources. The host koenigii ) in each of ten petri dishes (10 cm plants were separately planted in plastic bags (each diameter) prepared as in the two-choice bag 25 cm diameter and 35 cm high) containing a combination. The leaves in each of these petri 3:1:1 mixture of podsolic, compost and manure, dishes (without larvae) were weighed twice, at the respectively. Each plastic bag contained one beginning of the experiment and later at 24 h. Thus, seedling of a single species. Each seedling was the weight difference indicated water loss due to fertilized fortnightly with 10 g of artificial fertilizer evaporation. The larval consumed was recorded and 100 g manure. from weight of leaf before introduced to larvae subtract the weight of leaf after 24 h and mean of Feeding preference water losses. Both feeding preference of the first instar and leaves water content were replicated 10 Three larval stages, the first, the third and the fifth times. instars, were used in this experiment. Preliminary studies were conducted to determine the amount Four-choice test (weight) of leaves consumed in 24 h by each larval stage. Based on the average weight of leaves of The experimental design for the four-choice test each host plant consumed in the preliminary trial by was similar to the two-choice test, except that one the first, third, and fifth instar larvae (first instar: 5- very young leaf of all three Citrus (C. aurantifolia, 6 mg/10 larvae, third instar: 30-32 mg/5 larvae and C. hystrix and C. reticulata ) and three young fifth instar: 55-60 mg/larva), the weight of leaves foliages of M. koenigii were offered in the same for this feeding experiment was determined. All petri dish to the first instar larvae. Each of the ten larval instars were fed with leaves from the four petri dishes prepared for this experiment contained host plants using the two-choice and four-choice leaves of all four host plants approximately equal in experiments. size and weight. Twenty newly hatched P. polytes larvae starved for 4-5 h were introduced into each dish. The leaf host in each dish was weighed before the larval introduction and at 24 h post-introduction

22

Feeding preferences of Papilio polytes L. larvae on rutaceous host plants (Suwarno) ______of the larvae. The water loss from each petri dish young leaves of M. koenigii similar in size and due to evaporation was estimated by weight to the Citrus species were used. Each of the simultaneously running additional 10 petri dishes ten petri dishes prepared for this experiment without larvae as described above for the two- contained leaves of all four host plants choice test. approximately equal in size and weight. The vegetation in each dish was weighed before the Third instar larvae larval introduction and at 24 h post-introduction of the larva. The water loss from each petri dish due to The procedures used and the experimental design evaporation was estimated by simultaneously for evaluating the feeding preference of the third running additional 10 petri dishes without larvae as instar larvae for two-choice and four-choice test of described above for the two-choice study. four host plants were similar to that employed for the first instar. However, in this experiment, five Nitrogen content of host plant leaves newly-moulted third instar larvae were introduced per petri dish and larger petri dishes (15 cm The Kjeldahl method was used to determine the diameter and 2.5 cm high) were used. In each dish percentage of total nitrogen in the leaf [15]. one young of leaf each Citrus spp. (each Approximately 0.10 g of dried and homogenized approximately 4 cm long, 2 cm wide, and weighing young leaf materials were used for this test. 30-32 mg) was used; M. koenigii leaves used in the experiment were similar in size and weight to any Water content of host plant leaves one of the leaves of the Citrus species. The water loss from each petri dish due to evaporation was Twenty young leaves of each Citrus species and M. estimated by simultaneously running additional koenigii were collected from each host plant. All petri dishes without larvae as already described leaves of each species were separately weighed for above. The replication for two-choice and four- wet weight (WW) and dried in an incubator at 80 oC choice test was 10 times each. for 24 h and then weighed again for dry weight (DW). The leaves water content of each species Fifth instar larvae was calculated as a percentage {WW – (DW/WW)} x 100% [16]. Two-choice test Statistical analysis The experimental design for the two-choice test for the fifth instar larva was very similar to the two- The weight of leaves consumed on feeding choice test conducted for first and third instar preference of the P. polytes for each combination of larvae except that one newly-moulted fifth instar P. the host plant in two-choice test were analyzed polytes larva per petri dish (each 15 cm diameter using an independent t-test. The feeding preference and 2.5 cm high) was used. Two young leaves on the four choice test, larval performance (each approximately 5 cm long and 3 cm wide and parameters and chemical compound of leaves were weighing 55-60 mg) of each Citrus species and 5-7 evaluated by using the one-way analysis of variance young foliages of M. kenigii similar in size and (ANOVA). Means associated with each variable of weight to the Citrus species were used for all the host plant were further separated using the possible paired combinations (six pairs) for the Tukey’s test. All data were analyzed using the two-choice test. The six combinations of the leaves SPSS software version 12 [17, 18]. were replicated 10 times. The leaf in each dish was weighed before the larval introduction and at 24 h III. RESULTS AND DISCUSSION post-introduction of the larva. The water loss from each petri dish due to evaporation was estimated by Feeding preference simultaneously running additional 10 petri dishes without larvae as already described. Feeding preference of the first instar P. polytes larvae showed significant preferential feeding ( P < Four-choice test 0.05) for a host plant in all paired host plant leaves combinations tested, except for the C. hystrix and The experimental design for the four-choice test for C. reticulata combination (Table 1). It is obvious the fifth instar P. polytes larvae was similar to the from the data in Table 1 that C. reticulata leaves four-choice test for the first and third instars, except were generally the most preferred by the first instar that one newly-moulted fifth instar P. polytes larva larvae. The third instar larvae had significant per petri dish (each 15 cm diameter and 2.5 cm preference for leaves of the three Citrus species high) was used. Two young leaves (each over M. koenigii, while the third instar larvae approximately 5 cm long and 3 cm wide and showed no significant feeding preference among weighing 55-60 mg) of each Citrus species and 5-7 the Citrus spp. The fifth instar larvae fed

23

Feeding preferences of Papilio polytes L. larvae on rutaceous host plants (Suwarno) ______indiscriminately on leaves of all Citrus species but feeding than the fifth instar. The younger instars consistently consumed less M. koenigii compared to preferred and consumed more leaves of the Citrus Citrus spp. in the various combinations tested. spp. having a higher nitrogen and water content than the leaves of M. koenigii . The younger instar Table 1. Mean consumption (g) by the first, third larvae generally prefer host plants of higher quality and fifth instars larvae of Papilio polytes during a because they are important determinants of their 24 h period when provided with six different performance [19, 20]. In addition, young larvae are combinations of leaves of host plants, Citrus more sensitive to mechanical leaf traits (e.g. aurantifolia (Ca), Citrus hystrix (Ch), Citrus toughness) owing to their smaller mouthparts and reticulata (Cr) and Murraya koenigii (Mk) in the less developed musculature [21]. The low laboratory. preference for M. koenigii leaves may have been due to their tougher texture compared to the leaves of Citrus spp. The stronger odor of the curry leaves could also be a feeding deterrent for the younger instar of P. polytes . The first instar always prefer young leaves which contain less of the secondary metabolites, less fiber [20], and toughness [22].

The fifth instar larva was less selective about its food than the earlier instars because at this stage the larva needed a lot of nutrients (volume) to transform into . It has been reported that 80% of food consumed during the larval stage occurs in the fifth instar. Young larvae are generally more Means in a combination for each instar marked with an sensitive to food quality, including nitrogen and asterisk ”*” are significantly different ( P < 0.05, n = 10, t- secondary metabolites; also, the older larvae are test), ns = not significant. able to feed on a wider range of host plants [23].

The trend in the feeding preference of P. polytes Although the pupal stage is non-feeding, there are larvae in the four-choice test was similar to the two- massive physical transformations that occur during choice test where first instar larvae were relatively pupation. Therefore, the fifth instar larva consumes more selective than the third and fifth instar larvae a lot of food and consequently becomes less (Figure 1). All tested instars of P. polytes larvae discriminate in terms of the plants it consumed. consumed C. hystrix leaves more than the leaves of Early instars often feed on young leaves, while later the other host plants offered; M. koenigii was the instars can also utilize mature leaves, because the least preferred. young instar need to the higher nutrition for growth [24].

Naturally, the oviposition preference of P. polytes is very much related to the placing of off-springs on the most suitable host plant. This is very important especially for the first instar larva because of its fragile body and poor mobility [19]. There are three important factors, which determine the affinity of young larvae on suitable food choice. First is the provision of energy and the size of the larvae because the ability to cut through leaves depends on the size of head capsule, mass of chewing muscles and mandible morphology [21]. Second is the break down of ingested nutrient into fragments. Toughness of food material may affect the end size of the food particles. Finally is the absorption of food across the gut epithelium. The absorption efficiency of the digested nutrients is also positively Figure 1. Weight of Citrus aurantifolia, Citrus hystrix, correlated with the gut or body sizes. Citrus reticulata and Murraya koenigii consumed by the first, third and fifth instar Papilio polytes larvae during a The later instars could move and feed on many 24 h period in the laboratory . parts of the plant. The fifth instar larvae for example are bigger and stronger; therefore, they This study showed that the first and third instar feed on both young and mature leaves and on larvae of P. polytes were more selective in terms of

24

Feeding preferences of Papilio polytes L. larvae on rutaceous host plants (Suwarno) ______young stems. The older larvae of Cotacola spp. (Lepidoptera: Noctuidae) could consume all parts ACKNOWLEDGEMENTS of host plants because they could move about easily [25]. I acknowledge Assoc. Prof. Dr. Che Salmah Md Rawi and Prof. Dr. Abu Hassan Ahmad (School of Nitrogen and water contents of host plants Biological Sciences, Universiti Sains Malaysia) and Prof. Dr. Arsyad Ali (University of Florida) for The nitrogen as well as water content in the leaves editing this manuscript. We also thank Mr. of M. koenigii were the lowest and significantly Syanmugam for assisting in maintenance of host different ( F3,8 = 7.20, P < 0.01) from all three plants used in various experiments. Citrus spp. analyzed. Nitrogen content in the leaves of C. reticulata was the highest and significantly different from the other two citrus species; REFERENCES however, the water content in the leaves of all three citrus species was somewhat similar (Figure 2). 1. A.S. Corbet and H.M. Pendleburry, 1992, The butterflies of the Malay Peninsula . 4th edition. The hierarchy of water content of all four-host Malayan Nature Society. Kuala Lumpur. plants was similar to the nitrogen content (Figure 2). Consequently, the weights of leaves consumed 2. S. Nylin and N. Janz, 2009, Butterfly host by P. polytes larvae on Citrus species were higher plant range: an example of plasticity as a than those consumed on M. koenigii . The higher N 2 promoter of speciation?. Evolution Ecology level and water content in the host plants increased 23, pp. 137-146. the developmental rates of herbivorous . In contrast low nitrogen and water contents in hos