V. D ELUCCHI : Biology and control of Dicladispa gestroi Chapuis (Col., Chrysomelidae). Journal of Applied Entomology, 125 , 493–500
2002; Wiley; Volume: 126; Issue: 1 Linguagem: Inglês
10.1046/j.1439-0418.2002.00054.x
ISSN1439-0418
Tópico(s)Forest Insect Ecology and Management
ResumoJournal of Applied EntomologyVolume 126, Issue 1 p. 54-54 Free Access V. DELUCCHI: Biology and control of Dicladispa gestroi Chapuis (Col., Chrysomelidae). Journal of Applied Entomology, 125, 493–500 First published: 13 February 2002 https://doi.org/10.1046/j.1439-0418.2002.00054.xAboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat In the above paper 3, 4 were inadvertently printed in black and white. It is now republished on the following pages with the figures printed in colour. Figure 3Open in figure viewerPowerPoint . Adult of Dicladispa gestroi Figure 4Open in figure viewerPowerPoint . Patch of Dicladispa gestroi in a rice field Abstract The beetle Dicladispa gestroi is known only to come from Madagascar, where it is considered to be a pest of rice. Research was carried out from 1885 to 1994 in the Alaotra lake region, the main rice-producing area of the country, characterized by a warm rainy season from October to April and a cool dry season from April to October. The adult beetles invade the rice nurseries and the first direct-seeded fields at the beginning of the rainy season; they have a gregarious behaviour and their feeding activity, together with the mines bored by the larvae, determines a change of colour from green to pale yellow in the damage areas, which resemble outbreak areas of rice leafhoppers. Oviposition takes place only on young rice plants in the tillering stage. Females emerging after the end of February enter a reproductive diapause and leave the rice fields to `hibernate'. Temperature summations for the egg, larval, and pupal development, as well as for the preoviposition period have been calculated. There is no yield loss up to a larval density of 0.6 per leaf and this economic injury level is seldom exceeded in the Alaotra lake region. Life tables carried out under field conditions show that chalcid parasitoids are the main mortality factor and are responsible for the collapse of entire outbreak areas. Since the discovery of the rice yellow mottle virus in 1989 in the Alaotra lake region and the disease transmission by chrysomelids, the pest status of D. gestroi has changed and control measures have to be applied. However, to avoid interference with the action of the parasitoids, chemical applications should be limited to rice nurseries. 1 Introduction The beetle, Dicladispa gestroi, is a pest of rice (APPERT, 1967). The species has been found only in Madagascar, where it coexists with Trichispa sericea (Guérin), the African rice hispa. Although outbreaks of T. sericea are known from the highlands, particularly around Antananarivo (1200 m a.s.l.), those of D. gestroi have been observed in less elevated regions such as the Alaotra lake (≈ 700 m a.s.l.) and in the coastal rice areas of Marovoay. Of particular importance is the presence of D. gestroi in the Alaotra lake region (fig. 1), the largest rice-producing area of Madagascar (about 80 000 ha distributed around the lake and the marsh, roughly 300 km north-east from Antananarivo), where observations have been made and laboratory and field experiments carried out over a period of 9 years (1985–94). These results are included in the internal reports of the Swiss Development Cooperation to the Malagasy Ministry of Research Applied to the Development (MRAD) and the Ministry of Agriculture and Rural Development (MEADR), and also in diploma and Ph.D. theses, and are not easily accessible to interested persons. The present review gives a summary of the results. Figure 1Open in figure viewerPowerPoint . Alaotra lake region and its location in Madagascar. 1, Complexe Agronomique CALA; 2, Ambatondrazaka; 3, Andingadingana; 4, Angoja; 5, Morarano; 6, Amparafaravola; 7, marsh zone; 8, to Andilamena; =, road; … , rice fields; , off-season dry area of the marsh zone; , dams of artificial lakes) 2 Climatic conditions and biology 2.1 Climate in the Alaotra lake region and the rice crop Some information about the climate and the cultural practices in the Alaotra lake region is necessary to under- stand the biology and behaviour of D. gestroi. The climate is characterized by a warm rainy season from October–November to April and a cool dry season from April to October–November. There are 1180 mm rainfall/year (average 1954–85), with a maximum in January (270 mm) and a minimum during the entire cool season (50 mm). The distribution of the rainfall as well as the monthly minima and maxima of temperature are given in fig. 2. The distribution and intensity of rainfall varies from year to year. The beginning of the rainy season is particularly important as it determines the appearance of the first patches of D. gestroi. In 1986 the first patches were noted in mid-October (exceptionally about 140 mm rainfall in 1 month), whereas in 1985 they appeared 3 weeks later (SDC, 1986–94, No. 3). Cyclones may also change the infestation pattern; during the rice season 1993–94, many rice fields were inundated when the tropical cyclones Géralda and Daisy crossed Madagascar. As a consequence, D. gestroi mass migrated to dryland rice, a crop which had been only rarely infested before by the beetle. Figure 2Open in figure viewerPowerPoint . Average monthly temperature (Temp °C) (1983–86) with maxima and minima, and average monthly rainfalls (R, in mm) (1954–85) recorded at the meteorological station of the Complexe Agronomique sur Lac Alaotra (CALA), Amboitzilaosana (from SEIFERT, 1992) About half of the wetland rice area is irrigated with water from small artificial lakes and the irrigation can be manipulated; the irrigation of the other half depends on the rainfall and its manipulation is difficult. In addition, dryland rice is cultivated on the neighbouring hills. Sixty percent of the rice area is transplanted and 40% direct-seeded. Because of the distribution of the rainfall and the lack of mechanization, there is only one crop per year, with the seedbed preparation and direct-seeding starting in November, and transplanting beginning in December. However, rice is direct-seeded and transplanted until March and is sown even off-season where the soil in marsh areas near the lake remains humid enough to allow the rice crop to grow. The most grown cultivar is Makalioka 34 (= MK34), a short-day javanica ecotype issued from a local variety; it is characterized by a high tiller production, thick stem and long grain. At the Alaotra lake there is panicle initiation at the end of February and flowering at the beginning of April. The potential yield is about 8 t/ha, but the yield obtained in the region does not exceed 2 t/ha, mainly because of weed infestations. To obtain maximum yield, MK34 has to be sown in November and transplanted in mid-December. There is a decrease of 1 t/ha for each month of delay (AHMADI et al., 1988). In recent years, a new cultivar Tché-Kouai (= no. 2798 of the Malagasy national collection) has been introduced, which is slightly less photosensitive, has a shorter life cycle and is less preferred by Dicladispa than MK34; in fact, in outdoor cage experiments with the same number of adults/cage, the infestation was 1.04–1.34 larvae/leaf in the case of MK34 and only 0.16–0.26 in the case of Tché-Kouai. Furthermore, in a choice experiment with five rice varieties, oviposition was less on Tché-Kouai (RANDRIAMANANTSOA, 1991). The marsh to the south of the lake has a surface of 60 000 ha; about 5000 ha of this marsh zone are cultivated with rice, but in years with heavy rainfall early in the season the fields are inundated and the whole yield is lost. 2.2 Hosts of D. gestroi RAVELOJAONA (1982) mentions seven species of Gramineae, three of Cyperaceae, and one of Compositae on which he found preimaginal stages of D. gestroi and of T. sericea. Most of these species (Leersia hexandra, Pycreus mundti, Ageratum conyzoides and species of Panicum, Eleocharis, Scirpus, and Echinochloa) belong to the weed flora associated with rice. Whereas these species have no importance as Dicladispa host plants during the rice season, they may play a role in the conservation of the Dicladispa populations during the cool season in the marshy marginal zones of the Alaotra lake region. 2.3 Life cycle of D. gestroi The female lays her 0.8 mm long eggs singly through a slit within the epidermal layers of the dorsal and, more rarely, the ventral surface of the distal part of the leaf. At an infestation of 50% of the leaves of a rice hill (with a total of about 80 leaves/hill in cv. MK34), the distribution of the eggs in January may be as follows: 36% of the infested leaves have one egg, 23% two eggs, 18% three eggs, 11% four eggs, 6% five eggs, 3% six eggs, 1.5% seven eggs, 1% eight eggs and 0.5% nine eggs. This type of distribution does not change much during the months of January and February (VON CAPELLER, 1988). The eggs are often covered with a substance secreted by the ovipositing female. In captivity under laboratory conditions, the average fecundity of a female was 184 eggs: 50% of them were deposited within 1 week and 95% within 2 weeks, with some eggs still being deposited after 6 weeks (SDC, 1986–94, No. 3). In outdoor experiments in cages, the average fecundity was 126 eggs/female, 89% of which were deposited during the first week. The lowest developmental threshold temperature of the eggs has been estimated at 13.5°C; the temperature summation for the embryonic development varies between 53.7 and 59.2 degree–days at a temperature range between 22° and 31°C (SEIFERT, 1992). In January, 50% of the eggs hatch after 4.5 days in the field. The newly hatched larva starts mining the leaf from the oviposition point or it abandons the chorion and tries to find a convenient place to penetrate the leaf elsewhere. The mine is always oriented towards the leaf apex. If the leaf is damaged or the larval density is too high, the larva may abandon the primary mine and complete its development by boring a new mine in another leaf. There are four larval instars, which can be distinguished by measuring the head width (0.26–0.38; 0.40–0.48; 0.50–0.60; 0.62–0.82 mm). The lowest developmental threshold temperature of the larva is 12.5°C and the temperature summation necessary for the larval development varies between 130.6 and 144.9 degree–days at the temperature range indicated above. In January, in the field, 50% of the larvae pupate within 12 days after hatching. The quantity of the leaf surface which is destroyed depends on the population density and on the time it occurs: 3.3 cm2 (23% of the leaf surface) in December and 2.79 cm2 (17%) in January by an infestation of 1 larva/leaf, and decreases to 2.6 and 2.3 cm2, respectively, by an infestation of 2 larvae/leaf (measured on 6000 leaves, STEIGER, 1989). The maximum leaf surface destroyed by the larva is 6.09 cm2, i.e. about one-third of the leaf surface (SDC, 1986–94, No. 3). The fourth instar larvae pupate in the mine. The lowest developmental threshold temperature of the pupa is 14°C and the temperature summation necessary for the pupal degree-days development varies between 49.1 and 55.8 degree-days at the temperature range indicated above. In January it takes 4 days in the field. The optimum temperature for the preimaginal development has been estimated to be around 27°C, the shortest duration was observed at 35°C. The temperature values measured within the mine show less variation than outside; measurements made between 26 and 31 January indicated that temperature in the mine varied between 29 and 31°C, whereas the temperatures at 1 and 2 m above ground varies between 26.8 and 32°C and between 28.5 and 33.2°C, respectively (SEIFERT, 1992). The upper developmental threshold temperature is 37°C. Humidity is an important factor for preimaginal development, especially for the egg stage: indeed, a field humidity of 60% is already too low and causes an egg mortality of 20% (SEIFERT, 1992). A sample of 1000 adults were captured in 1988 and their longevity measured in cage experiments from 27 January until 5 July: 50% of this population was dead after 45 days and about 80% after 90 days, but some individuals were still alive after 6 months (SDC, 1986–94, No. 3). The following season, another sample of adults was caged in the field on 15 December: 50% of the population was dead after 1 month and 80% after 45 days; however, some individuals survived until 5 May (SDC, 1986–94, No. 4). The leaf damage caused by the adults is not of economic importance and is similar to that caused by other hispine species. Five generations of D. gestroi have been estimated to develop between October and March (SDC, 1986–94, No. 4). The sex ratio varies between 0.4 and 0.6. Newly emerged adults (fig. 3) migrate to the rice fields at a convenient stage of development and copulate. They prefer young rice plants with tender leaves. Females need an average temperature summation of 68.2 degree–days to start oviposition, which corresponds to about 5.5 days at 27°C or 3.5 days at 33°C. 3 Results 3.1 Dynamics of field invasion The problem of the `hibernation' of Dicladispa adults could not be completely clarified. Although all possible ecological situations were explored during the 8 years of the study (1985–93), and glued traps (different colours) were installed in small rice fields sown along a transect from the hills to the marsh zone of the lake before the beginning of the rice season, and soil samples from previously infested rice areas were analysed, it was never possible to localize the refuges of the `hibernating' adults. The only occurrences of possible relevance were in July 1993, when adults of Dicladispa were observed on shrubs and gramineous plants in a valley near Andilamena (north of the Alaotra lake), and, in March 1994, when a kind of mass migration of adults from the rice fields to the neighbouring hills was noted in a valley near Andingandingana (east of Ambatondrazaka), where the vegetation is mainly composed of fruit trees. Both valleys are characterized by a high humidity and the microclimate of these sites allows the beetles to survive until the first rainfalls of October (SDC, 1986–94, No. 8B). From here they first colonize and feed on ratoon rice, on rice growing off-season in threshing places, and on various gramineous species, and then they migrate to the rice nurseries and the first direct-seeded fields. As these sites for oviposition are scarce at the beginning of the rice season, there is aggregation of the beetles and their feeding activity, together with the mines bored by the developing larvae, determines a change of colour of the damaged area from green to pale yellow or whitish (fig. 4), which resembles the outbreak areas of the rice leafhoppers. The beetles have gregarious behaviour. They oviposit only on young rice plants in the tillering stage. From December on, rice fields are readily available at the tillering stage until between the end of February to mid-March, when stem elongation and, subsequently, panicle initiation occur. These plant stages prevent oogenesis in ovipositing Dicladispa females, which enter a period of reproductive quiescence; it is resumed at any time when rice plants at the tillering stage are available again. This is possible at the Alaotra lake region because the farmers continue to transplant rice until February or even March and also because photo-insensitive cultivars are also used in addition to MK34. Oviposition by these females may be observed until the end of June, although fecundity is very low and the eggs generally do not hatch. According to observations made on adults caged in the field, after June they copulate no more and stop feeding, even though young rice plants at tillering stage are available. The low temperature seems to be the limiting factor here. Females that emerge after the end of February to mid-March enter a reproductive diapause and leave the rice fields to `hibernate'. The factor(s) inducing diapause is/are unknown, as well as the Dicladispa stage on which induction occurs; presumably, the photoperiod plays the main role. In any case, diapause is not influenced by the developmental stage of the plant. 3.2 Dynamics of selected patches of Dicladispa It is almost impossible to find common traits for the dynamics of the Dicladispa patches, each one having its own characteristics due to its geographic position and the time it occurs. Seven patches have been followed from the beginning of their appearance and two of them had to be abandoned (VON CAPELLER, 1988). A plot of 0.18 ha was selected in Angoja, on the eastern side of the lake near the marsh zone, where rice seedlings of the cultivar Rojofotsy (= no. 1285 of the Malagasy national collection, moderately photosensitive) were transplanted on 8 August. Eight samples were taken between 6 and 23 November and the density of the preimaginal stages varied from 7 to 20/rice hill (about 150–400/m2), with 81% of the individuals already in the larval stage. The first pupae appeared on 9 November. Because of the delayed rainfalls and the consequent lowering of the lake level, the soil dried up and the rice crop became unfavourable for the development of the eggs and larvae. However, most of the adults could emerge before the drought affected the crop. The total mortality of the preimaginal stages varied between 33 and 64%. Egg and larval parasitoids were already present and during the sampling period the degree of parasitization reached a maximum of 44% for the eggs and 68% for the larvae. In another plot of 0.14 ha in Ambaibo, also near the marsh zone, the situation was similar to the preceding one, with sampling starting on 6 November, but with seedlings belonging to the cultivar MK34. The seedlings were transplanted on 15 October and 75% of the preimaginal stages were already larvae in the first sample. The first pupae developed on 11 November and about two-thirds of the population were in the pupal stage 10 days later. The total mortality varied in the samples between 27 and 53%, egg parasitism between 11 and 75% and larval parasitism between 3 and 41%. The dynamics of the two patches in Angoja and Ambaibo is typical of the early season and of fields near the marsh zone. A high percentage of the pre-imaginal population developed into adults, which are able to colonize the nurseries and the first direct-seeded fields. In a third plot of 0.36 ha located in Amdrarabary, seedlings of MK34 were transplanted on 15 December. Eleven days later the crop was infested by an average of 163 eggs/rice hill, equivalent to more than 1500 eggs/m2 and 58% of them were already parasitized. At the second sampling date (4 January) over 90% of the eggs were parasitized. The few larvae which could hatch were also parasitized or died, so that the Dicladispa patch collapsed and disappeared. A fourth plot of 0.22 ha was selected near Ambatondrazaka. Seedlings of the cultivar Tché-Kouai were transplanted on 24 December and samples were taken from 13 January to 2 February. The maximum density of the preimaginal stages were observed on 18 January, with 73 individuals/rice hill (about 1130 individuals/m2); 93% were eggs (of which 20% were dead and 33% parasitized) and 7% larvae. The larvae did not develop beyond the second stage and the whole preimaginal population was parasitized or dead by 2 February, so that the patch disappeared. Another plot of 0.18 ha in the same area presented similar dynamics. Seedlings of MK34 were transplanted on 15 January and maximum density of the pre-imaginal population was observed on 8 February with 179 individuals/rice hill (about 3400 individuals/m2) and the following distribution: 3% larvae (two-thirds already parasitized) and 97% eggs, of which 90% were parasitized and 3% dead. On 15 February, the total mortality of the preimaginal stages reached 99.8%. The above selected sites were investigated during the rice season 1987–88 (VON CAPELLER, 1988). During the season, the density of the preimaginal population per rice hill increased by about 15 times. Whereas most of the individuals could develop to adults in November, during the following months the population died out under the pressure of parasitoids before reaching the pupal stage. By comparisons made over the previous 3 years, it became evident that Dicladispa patches collapsed earlier each year and the parasitization degree increased at the same time. The most important mortality factor in Dicladispa populations is egg parasitism. From November 1987 to January 1988, the development of 416 patches (which correspond to an infested area of 200 ha) was followed in the region of Ambatondrazaka (fig. 5). Seventy percent of the patches were in nurseries, 24% in direct-seeded fields and 6% in transplanted fields. Egg parasitism did not exceed 15% in November or 30% in December, but it increased constantly from January on. Larval parasitism showed a first peak of 60% in December and a second one of 75% in January. During the period November to January the average density of emerging adults was about 3 individuals/100 leaves, as indicated in fig. 5. The collapse of Dicladispa patches, as shown in the selected examples, could be a frequent occurrence in January–February. Figure 5Open in figure viewerPowerPoint . Average density (D) of the preimaginal stages of Dicladispa gestroi per leaf in 416 patches of the Ambatondrazaka rice area from October 1987 (0–1) to the end of January 1988 (12–13) (pale grey, eggs; grey, larvae; black, pupae; 0–13: weeks) (from VON CAPELLER, 1988) Seven life-table studies carried out under field conditions from December to February (SDC, 1986–94, No. 2) and in January to March (NEMECEK, 1987; SDC, 1986–94, No. 3) confirmed the observations made on selected Dicladispa patches. Only the life-table which started on 11 December (1000 marked leaves, each with a single egg) on MK34 produced adults (26%). In the other life-table studies, which started on 6 and 7 January, 21 and 25 February, and 2 and 12 March, of a total of 1737 marked eggs, the preimaginal population did not develop beyond the larval stage, in most of the cases beyond the second instar. Parasitism of the egg (up to 83%) and of the larvae (up to 70%) was the main mortality factor. It appears therefore that at least in certain rice areas of the Alaotra lake region, and particularly in the south and south-east, development into adults is practically impossible owing to the impact of the parasitoids. This must be the main reason why `overwintering' adults could not be found in spite of research over several years. Only in areas where parasitoids are less frequent, such as the west and north-west of the marsh zone, is the development up to the adult stage possible during the whole rice season, as observed by the monitoring team (see section 3.4). 3.3 Damage and its impact on yield From 1987 to 1991 cage experiments were carried out in the rice fields to define the impact of the damage caused by adults and larvae on the yield and possibly the economic threshold. Every year, two crops were considered, one transplanted `in time' and another 4–6 weeks later. In Table 1 the cultivars used are indicated as well as the main steps necessary to perform the experiments and the timing of the operations. Table 1. . Important data concerning the cage experiments carried out from 1987 to 1991 (SDC, 1986–94) The installation of cages was necessary during 4–5 weeks to prevent parasitization of eggs and larvae. Copulating adults were collected in the rice fields and remained in the cages for only 4–7 days to minimize the damage to the leaves. The simulation of the damage by adults and larvae occurred in uninfested cages; it involved the elimination (by cutting) of 30% and subsequently 100% of the leaves 7 and 25 days, respectively, after the infestation date. In addition to the cage experiments, the impact of Dicladispa on rice yield was evaluated in a naturally infested field of MK34 during the season 1989–90 (SDC, 1986–94, No. 6). From all the cage experiments it appears that up to a larval density of 0.6/leaf (i.e. 60% of the leaves infested, or 20% of the total surface of the leaves) there is no negative impact on the yield and the damage may even be beneficial to the crop. The rice plant compensates for the damage by producing more tillers and therefore more panicles. As the panicles of damaged rice hills are generally lighter in comparison with the control, the yield does not differ in infested and in pest-free crops. However, during the rice season 1987–88, the crop transplanted in December overcompensated the damage at a larval density of 0.2/leaf and the yield increased by 20%. Larval densities lower than 0.6/leaf are rather common in the Alaotra lake region (see section 3.4), so the benefit to the crop should prevail in infested fields. The more the transplanting date is delayed, the less the crop is able to compensate for the damage, even though the leaf surface destroyed by single larvae is smaller in late-transplanted or direct-seeded crops (STEIGER, 1989). The yield decreases by 6 and 10% at a larval density of 1 and 1.3 larvae/leaf, respectively. The simulation of 100% damage indicates that, even in the case of the total destruction of the leaves, the yield reduction varies between 20 and 40% only, depending on the transplanting date. In a field of MK34 transplanted on 16 December, the infestation by Dicladispa was observed on 13 January and 500 rice hills were marked 3 weeks later. The development of the infestation was followed until harvesting date (16 May). The density of the larvae on the marked hills varied between 0.5 and 4 larvae/leaf. The highest number of panicles/hill and the highest weight of paddy/hill was observed at a larval density of 3 larvae/leaf. Even at a density of 4 larvae/leaf the yield was higher than the control (uninfested hills). It was observed that the highest density of the larvae occurs on hills that are particularly vigorous, which are probably preferred by females for oviposition. Considering the larval density in patches of rice fields and its impact on yield, D. gestroi does not constitute a pest problem in the Alaotra lake region, except in areas contaminated by the rice yellow mottle virus and in nurseries. Here the damage is sometimes so heavy that it is necessary to re-prepare the seedbeds. 3.4 Survey of the rice area Table 2 summarizes the data obtained by a monitoring team of 20 men during a period of 13–14 weeks starting towards the end of October, soon after the first patches became visible. The survey was conducted during 6 years (1987–93), but the data of the last 3 years have been disregarded because they concern the nurseries only. The monitored rice area extended from the south to the north-west of the lake and its marsh zone. Table 2. . Results of the surveys carried out between 1987 and 1990 during the first half of the rice season Although the data show a considerable variation according to the rice areas, the situation did not change much over the years. During the first half of the rice season, the highest number of patches always occurs in the nurseries (over 50%), but the largest infested area is observed in direct-seeded fields (63–71%). Patches in these fields are much larger than in nurseries. In general, in the most infested areas the direct-seeded fields dominate. In transplanted fields, the number of patches is relatively low. In the second half of the rice season, from the end of January onwards, patches are almost equally distributed in transplanted as well as in direct-seeded fields (SDC, 1986–94, No. 4). From the average density of the larval and pupal stages is appears that D.gestroi exceeds the economic injury level of 0.6 larvae/leaf only in 2.2% (1987–88) to 4.4% (1988–89) of the patches. Therefore, D. gestroi is seldom a rice pest in fields that are free of the rice yellow mottle virus. 3.5 Parasitoids Parasitoids are the most important mortality factor. They may be so important that the Dicladispa population of a patch dies out before the generation is completed. Such an extreme situation is typical for the patches which appear in January and especially in February, particularly in the south and south-west areas of the Alaotra lake region. The preimaginal stages of the first patches of Dicladispa, which appear in nurseries and in direct-seeded fields, are already parasitized. In the rice areas to the north-west and east of the Alaotra lake the parasitoids are less efficient. The eggs are attacked by an unidentified species of Ufens (Chalc., Trichogrammatidae), which is able to parasitize almost all of the eggs of a patch, as happened in Andrarabary in January 1988 and near Ambatondrazaka 1 month later, with 93.7 and 93.2% parasitization, respectively (see section 3.2). It is interesting to note that the efficiency of the parasitoid is density independent (fig. 6) and the degree of parasitism does not change with host densities varying from 1 to 11 eggs/leaf (NEMECEK, 1987). In a report by RANDRIAMANANTSOA (1991), the species has been attributed to the genus Paracentrobia; however, the species of this genus are parasitic on leafhoppers. Figure 6Open in figure viewerPowerPoint . Degree of egg parasitism (P%) in relation to egg density (1–11) of Dicladispa gestroi per leaf. The columns represent the frequency (n) of the leaves for each egg density (from NEMECEK, 1987) The complex of the parasitoids of the larvae is composed of the following six species (all Chalc., Eulophidae): Sympiesis (Notanisomorphella) somalica Masi, two unidentified Sympiesis (Notanisomorphella) sp. and Pediobius vigintiquinque Kerrich (all ectoparasitoids), and two unidentified species of Chrysonotomya (endoparasitoids). The genera Sympiesis and Notanisomorphella have been synonymized by BOUCEK and ASKEW (1968), but in all the documents on D. gestroi the name Notanisomorphella is retained and is therefore mentioned here. Sympiesis somalica is the most abundant species; of the 4103 parasitoids reared from larvae in 6 years, 73% belonged to it. A Chrysonotomya species is also important, with 19% frequency (SDC, 1986–94, No. 7A, 7B, 8A and 8B). The pupae of D. gestroi are not parasitized. 3.6 The rice yellow mottle virus The rice yellow mottle virus (RYMV) disease has been known in Madagascar for 30 years, but the Alaotra lake region was free from it until 1989, when it was discovered near Ambatondrazaka, south of the marsh zone. Within 5 years, the disease spread over the entire area (SDC, 1986–94, No. 8A and 8B). Insects play a major role in the disease transmission; among them an Oxya sp. (Orthoptera), the chrysomelid Chaetocnema pulla Chapuis (Halticinae) and D. gestroi. Of particular importance is the role of C. pulla, a rather small insect which is difficult to detect in the rice fields. The species is most abundant in fields to the east of the lake and the RYMV impact is greater there than in other areas. Outside the Alaotra lake, the disease is also transmitted by other chrysomelids, such as Sessilia pusilla Gerst. (Galerucinae) and the hispine T. sericea. Cage experiments were carried out in 1993/94 to evaluate the relationship between the density of D. gestroi adults infected with the virus and the impact of the disease on the yield. The results show that the greater the density the lesser the yield; a yield reduction of 9% is already noted at a density of 1 infected adult/rice hill and of 50% at a density of 4 adults/hill (SDC, 1986–94, No. 8A and 8B). Under natural conditions, the presence of such densities of adults are certainly not rare. 3.7 Control measures Until 1989, cultural practices such as the elimination of ratoon rice and of the volunteer rice in threshing places as well as the accurate timing of pesticide applications in the nurseries were considered sufficient to maintain the D. gestroi populations below the economic injury level of 0.6 larvae/leaf. With the application of such measures, chemical treatments of rice fields can be avoided and the action of the parasitoids is entirely preserved. With the discovery of RYMV, the situation changed drastically and the only way to control the disease is the introduction of resistant cultivars. If this is achieved, the practices mentioned above would be effective. The most effective chemicals against eggs and young larvae are those based on ethophenprox, and against the old larvae those based on deltamethrin (SDC, 1986–94, No. 6). In addition to these practices it has been proposed that the patches are submerged, as it is well known that heavy rainfalls during the tillering stage may inundate the fields and cause a heavy mortality to the pre-imaginal stages of Dicladispa on the submerged leaves. Experiments carried out in 1988 show that 50% of the larvae survive after 2 days submersion of the leaves, 38% after 4 days, and 15% after 8 days. As the rice plant cannot tolerate a submersion longer than 4 days, the survival rate is too high to use this cultural practice for the control of D. gestroi. Footnotes Author's address: V. DELUCCHI, Strada da Vissin, 6822 Arogno, Switzerland Acknowledgements The author is grateful to Professor Dr MARTIN WOLFE, Fressingfield, Suffolk (UK), for his assistance with linguistic problems and for improvement of the manuscript. He is also indebted to Dr CESARE GESSLER and to Dr PHILIPPE BLAISE, Swiss Federal Institute of Technology, Zürich (CH), for the preparation of the figures 2, 5 and 6. The Swiss Federal Institute of Technology, Zürich, generously accepted to cover the printing cost of the coloured figures 3 and 4. References 1 AHMADI, N.; CHARPENTIER, H.; FEAU, C.; RABARY, E., 1988: Amélioration variétale du riz pour la région du Lac Alaotra à Madagascar. L′ Agron. Tropicale 43 , 91– 98. 2 APPERT, J., 1967: Les insectes nuisibles aux cultures de Madagascar. Bull. agron. 22, 177 pp. 3 BOUCEK, Z.; ASKEW, R. R., 1968: Palearctic Eulophidae (excl. Tetrastichinae) (Hym. Chalcidoidea). In: Index of Entomophagous Insects. Ed. by DELUCCHI, V.; REMAUDIE`RE, G. Paris, France: Le François, 254 pp. 4 NEMECEK, TH., 1987: Lebenstafeln und Untersuchungen über das reproduktive Verhalten von Dicladispa gestroi Chap. (Coleoptera: Chrysomelidae) in Abhängigkeit vom Pflanzenstadium. Diploma Thesis. ETH, Zürich 81 pp. 5 RANDRIAMANANTSOA, R., 1991: Mise au point d′une méthode d′élevage de Dicladispa gestroi Chapuis, Hispine, Coléoptère nuisible à la culture du riz à Madagascar et étude de son comportement vis-à-vis des variétés de riz. Mémoire de stage. Montpellier, France: CIRAD-IRAT, 36 pp. 6 RAVELOJAONA, G., 1982: Contribution à une monographie de Trichispa sericea Guérin, coléoptère nuisible au riz à Madagascar. Terre Malgache 21, 57 pp. 7 SDC (SWISS DEVELOPMENT COOPERATION), 198694: Protection intégrée en riziculture au lac Alaotra. Rapport d′activité no. 2 (88 pp.), no. 3 (166 pp.), no. 4 (217 pp.), no. 5 (170 pp), no. 6 (91 pp), no. 7A & 7B (136 pp & 82 pp), no. 8A & 8B (79 & 91 pages) au Ministère de la Recherche Appliquée au Développement (MRAD) et au Ministère d′Etat chargé de l′Agriculture et du Développement Rural (MEADR), Madagascar. 8 SEIFERT, M., 1992: Analyse der Populationsdynamik des madagassischen Reisblattkäfers Dicladispa gestroi Cha- puis (Coleotera: Chrysomelidae) am Alaotra see (Madagascar). Ph.D. Thesis. Nr. 9661. ETH Zurich, 117 pp. 9 STEIGER, S., 1989: Schädlichkeit von Hispa gestroi Chap. Evaluation einer Methode zur Erzeugung von künstlichem Befall und Quantifizierung des larvalen Frassschadens. Diploma Thesis. ETH Zurich, 23 pp. 10 VON C APELLER , E., 1988: Untersuchung der Populationsdynamik des Reisblattkäfers Hispa gestroi Chap. (Coleoptera: Chrysomelidae) und seiner Parasitoiden in Madagascar. Diploma Thesis. ETH Zurich, 63 pp. Volume126, Issue1February 2002Pages 54-54 FiguresReferencesRelatedInformation
Referência(s)