Bol. San. Veg. Plagas, 25: 335-345, 1999

Coccinellids (Coleoptera: ) in citrus groves in Portugal: listing and analysis of geographical distribution

A. MAGRO, J. ARAÚJO & J. L. HEMPTINNE

Coccinellids were sampled in 88 citrus groves from the centre/south of continental Portugal. Analysis of distribution were performed by the usual indices but also with the help of a new method - SADIE. 38 species were identified, 36 of which are predators. Five coccinellids seem to have aggregated distributions; UTM maps of those species distributions are presented. Among these five, Nephus reunioni Fiirsch and Nephus includens (Kirsch) are the more abundant. Previous studies about these two coccinellids distributions are confirmed: N. reunioni is limited to a region of about 80 Km diameter around Lisbon while N. inclu­ dens appears only in Algarve. Coccidophagous species and a group of sp. and Nephus sp., with unknown food preferences in citrus, are the most important in terms of richness and abundance.

MAGRO, A.: Ecole Nationale Supérieure Agronomique de Toulouse, Av. de l'Agrobio- pole, Auzeville-Tolosane, BP 107-F31326 Castanet - Tolosan Cedex, France. ARAÚJO, J.: Centro de Ecologia Aplicada, Universidade de Évora, 7001 Évora Codex, Portugal. HEMPTINNE, J. L.: Ecole Nationale de Formation Agronomique, BP 87-F31326 Casta- net-Tolosan Cedex, France.

Key words: Coccinellids, citrus, biological control, geographical distribution.

INTRODUCTION attention is payed to the implementation of Integrated Pest Management (IPM). In fact, The culture of Citrus has a long tradition in the citrus have a complex phytosanitary Portugal and is still one of the most important situation that can be solved with this kind of crops in the country. In 1994 the orchards of approach putting together some of the alre­ various species of citrus covered 24.638 ha, ady existing biological control measures and mainly in the south of Portugal (Instituto the chemical treatments still needed against Nacional Estadística, 1995). some pests. Many phytophagous are reported Among the research projects in progress, to feed on citrus but only ten are considered biological control of coccids by coccinellids, as key pests in continental Portugal (CARVAL­ which are claimed to be the most rich and HO, et al. in press). Among these we find four abundant group of predatory insects in citrus Homoptera: Aleurothrixus floccosus (MAS- (LONGO & BENFATTO, 1987), is an important KELL) (Aleyrodidae), Planococcus citri (Ris- topic. The interest raised by this family is so) (Pseudococcidae), Saissetia oleae (OLI­ also supported by the fact that the first suc­ VIER) (Coccidae) and Lepidosaphes beckii cessful and often repeated case of biological (NEWMAN) (Diaspididae). control featured a ladybird , Rodolia The development of the crop is subsidised cardinalis (MULSANT) released against Icer- by the European Administration and great yapurchasi Maskell, a coccid, in 1888. Since then, other species of ladybird succe- llids present in citrus groves from the Cen- eded in controlling several species of coccids tre/South areas of continental Portugal. The around the world (HODEK, 1970, COPPEL & ladybirds present in citrus groves from the MERTINS, 1977, IPERTI, 1987, MAJERUS, most important regions of production were 1994, DIXON, et al, 1997). sampled and their geographical distribution The fauna of the coccinellids is well des- established. cribed in Portugal (see RAIMUNDO & ALVES, In order to accomplish these objectives we 1986) but their basic ecology is barely used some faunistic methodologies seldom known. However, the mechanisms driving employed for this kind of applied research. the variations in abundance of predatory spe- UTM (Universal Transverse Mercator) grid cies taken in isolation or in a community con- maps and analysis of distribution patterns are text are essential to define rational program- among them. To the traditional measures of mes of biological control. aggregation (Dispersion Indices) we The present work aimed to improve our added a new methodology -SADIE (Spatial knowledge on the distribution of the coccine- Analysis by Distance Indices)- which allo-

Fig. I. - Sampling of coccinellids in 88 citrus groves in Portugal. Groves are located in 32 UTM squares of 10 km. corresponding to 7 main regions. wed us to take into account the available spa- A geographical database was created with tial information in the sample (PERRY, 1995 the help of Microbanque Faune-Flore, logi- a, b). ciel de gestión de banques de données biogé- ographiques (RASMONT, et al, 1993). The same program allowed the drawing of distri- MATERIAL AND METHODS bution maps. Maps were plotted on a 10 Km square basis derived from the UTM grid maps. Samples of adult coccinellids were collec- A first analysis of the distribution was ted from 1991 to 1994 in 88 citrus groves done using the dispersion index ID associa- located in 32 UTM squares of 10 Km in Por- ted with a statistical test d (LUDWIG & REY- tugal (Fig. 1). NOLDS, 1988): In 73 groves a single sample was taken during the whole study period, 3 others were inspected each two weeks for 3 years and the remaining 12 groves were sampled monthly for a year*. The groves sampled only once were visi- where x, is the average number of lady- 2 ted mainly during summer time, which birds of species / in the orchard samples, s f is corresponds to the period where the greatest the estimate of the samples variance and N number of species might be captured represents the total number of samples. If the (MAGRO, 1992, PAIXÃO, 1994). This is also samples follow a theoretical Poisson distri- the season where the key Homoptera are par- bution, this ratio is expected to be equal to ticularly abundant and we can so expect to 1.0. Significant departures of ID from 1.0 can find their predators more easily. be tested using d. lf\d< 1.96, agreement with To collect ladybirds a method -«Beating a Poisson distribution is accepted (P>0.05). method»- recommended by IOBC (AMARO If d <-1.96, the ladybirds are suspected to be & BAGGIOLINI, 1982) was adopted. 100 bran- regularly dispersed among the orchards, and ches uniformly distributed in each orchard if d> 1.96, a clumped dispersion is likely were selected and beaten with a stick. In (ELLIOT, 1973). order to standardise the sampling method, The study of distribution patterns was each branch was stricken three times. The completed by SADIE, created by PERRY & insects so dislodged fell in a jar attached at HEWITT (1991) and ALSTON (1994). In the the bottom of a tray in a form of a funnel. computations of measures and indices the This method mainly yielded adult ladybirds program Sadiec.for** was used. This pro- but few larval instars. gram analyses the spatial pattern of data that Coccinellids were identified using keys are in the form of counts at specified spatial published by CHAZEAU, et al. (1974), Gou- locations. Techniques and notation follow RREAU (1974), RAIMUNDO & ALVES (1986), those outlined in Perry (1995Z>). FÜRSCH (1987 a and b) and RAIMUNDO The techniques measure distance to com- (1992). Most identifications were achieved plete regularity/aggregation of the observed by considering external characters but there numbers and compare the result with those were still about 1.100 individuals for which based on randomised simulations of the same extraction and observation of genitals were data set. Tests and indices are so constructed. needed. Two types of simulation can be done:

* The periodical sampling meant to supply infomation to a second paper on temporal dynamics. N* The copyright in this software is vested jointly in Rothamsted Experimental Station, Harpenden, Herts AL5 2JQ UK and The Univesity of Tokyo, Bunkyo-ku, Tokyo 113, Japan. The software was supplied by Dr. J. N. Perry (IACR - Rothamsted). Table 1. - Coccinellid species captured in citrus groves from the center/south of Portugal in 1991-94 and their food preferences (Fp). 1 = Diaspididae, 2 = Coccidae, 3 = Pseudococcidae, 4 = Icerya purchasi, 5 = Aphididae, 6 = mites, 7 = Aleyrodidae, 8 = plants, 9 = fungus, 10 = predators with unknown food preferences in citrus, (*) = species captured only in periodical sampling, Tp = total number of individuals captured in punctual samples.

TAXA Fp Tp Chilocorini Chilocorus bipustulatus L. Exochomus nigromaculatus (Goeze) E. quadripustulatus L. Coccidulini Lindorus lophantae (Blaisdell) Rhizobius chrysomeloides (Herbst) R. litura Fabricius Coccinellini Adalia bipuncatata (L.) A. decempunctata (L.) Coccinella septempunctata L. Oenopia conglóbala (L.) O. doublieri (Mulsant) O. lyncea (Oliv.) Propylea quatuordecimpunctata (L.) Psyllobora vigintiduopunctata (L.) Tytthaspis sedecimpunctata (L.) Epilachnini Subcoccinella vigintiquatuorpunctata L. Hippodamiini Hippodamia variegata (Goeze) Hyperaspini Hyperaspis reppensis Herbst Noviini Rodolia cardinalis (Mulsant) Platynaspini Platynaspis luteorubra Goeze Scymnini Clitosthetus arcuatus (Rossi) Cryptolaemus montrouzieri Mulsant Nephus bisignatus Boheman Nephus includens (Kirsch) Nephus reunioni Fiirsch Nephus ulbrichi Fiirsch Nephus binotatus Brisout Nephus hiekei Fiirsch Nephus fuerschi Plaza Scymnus mediterraneus Khnzorian Scymnus auritus (Thunberg) Scymnus subvillosus (Goeze) Scymnus suturalis Thunberg Scymnus apetzi Mulsant Scymnus interruptus (Goeze) Scymnus levaillanti Mulsant Scymnus rufipes (Fabricius) Stethorus punctillum (Weise) • a: involves permutation of the actual case of groves subject to periodical sampling counts observed amongst the sample a single sample from summer time was cho- units; sen and used for the analysis mentioned abo- • r: involves totalling the counts over all ve. Analysis of distribution patterns were the sample units and redistributing each done for species having at least 12 captured individual of this total number, ran- individuals in punctual samples, because at domly, to the sample units. low population densities the distributions In type a simulation inferences relate only often approach the Poisson (MORRIS, 1960). to the spatial arrangement of the counts; in the type r simulation both spatial arrange- ment of the counts and the degree to which RESULTS their distribution is non-Poisson are taken into account. Table 1 presents the list of coccinellid spe- In the present work only tests and indices cies captured in citrus groves, their essential involving permutation (a) were used, for the food when known and total number of indivi- information about the degree to which the duals captured in punctual sampling. distribution of counts is non-Poisson is alre- 38 species were identified. A few indivi- ady given and tested by the dispersion index duals of the species Nephus peyerimhoffi ID. Only estimates based on distance to Sicard, claimed by RAIMUNDO (1992) as aggregation were analysed for the criteria of being present in Portugal, might be among distance to regularity are sensible to edge those identified as N. includens which would effects (PERRY, 1995a). raise the number of species to 39. We were As a result, the index J was calculated: not able to distinguish these two species due to the unclear descriptions of the genitalia of N. peyerimhoffi provided by RAIMUNDO (1992) and Fiirsch (1987 b). All species of the Nephus genus, with the where D is the distance to aggregation of exception of N. 2-notatus (i.e., 7 species) the observed samples and Ea, the average were described for Portugal only recently distance to aggregation for randomised sam- (MAGRO, 1992, RAIMUNDO, 1992, MAGRO, ples. 1997, RAIMUNDO, pers. com.). The index Ja, is followed by Qa, which is The identified species are divided into 9 the proportion of randomised samples with a tribes; 6 species appeared only in periodical distance to aggregation as large as, or larger sampling. than the observed value D. Several species are represented by 8 or For this computations the UTM co-ordina- less individuals. tes of each grove, in a scale of 1 x 1 Km were Exception being made of S. 24-punctata, taken and transformed to simple co-ordinates P. 22-punctata and T. 16-punctata all the coc- composed by two numbers corresponding to cinellids are predators. abscissae and ordinates. In Table 2 we can find the results of the Samples were grouped in 7 main regions analysis of distribution patterns. In what the (Fig. 1) and their Dominance (relative abun- ID index is concerned, all species with the dance - SZUJECKI, 1987) was calculated for exception of A. decempunctata and S. levai- each region. The Dominance of the different llanti, present significant variations in num- guilds of ladybird species in each region was bers from one sample to another (ID > 1). also evaluated. This spatial heterogeneity of the counts reve- In the plotting of maps and analysis of dis- als an aggregated distribution but does not tribution and dominance only data from the allow any inference about the spatial arran- punctual sampling was considered. In the gement of those counts. Fig. 2. - Distribution maps (UTM) of five species of coccinellids captured in citrus groves in Portugal. 341

Table 2. - Analysis of distribution patterns of coccinellid species present in citrus groves from the center south of Portugal. ID = dispersion index, Ja = aggregation index, d and Qa = statistical tests. ID or Ja = 1, random distribution; ID or Ja > 1, aggregated distribution; ID or Ja < 1, uniform distribution.

TAXA ID Ja Qa Chilocorus 2-pustulatus Exochomus nigromaculatus E. 4-pustulatus Lindorus lophantae Rhizobius chrysomeloides Adalia 10-punctata Oenopia conglobata Propylea 14-punctata Rodolia cardinalis Clitosthetus arcuatus Cryptolaemus montrouzieri Nephus includens N. reunioni Scymnus mediterraneus S. subvillosus S. suturalis S. apetzi S. interruptus S. levaillanti Stethorus punctillum

The results of the Ja index point out that has a strong presence in the Muge region, O. half of the species have aggregated distribu- conglobata is mainly in the Santiago do tions of counts (Ja>\); 5 of them - P. Cacem region, N. reunioni was only caught 14-punctata, 0. conglobata, N. reunioni, N. in a 80 Km diameter region around Lisbon, includens and S. suturalis - show even very N. includens was only captured in Algarve high values of that index. In the maps repre- and S. suturalis was fundamentally restricted senting the distribution of those five species to the surroundings of Sintra/Mafra and (Figure 2a - e) we can see that P. 14-punctata Setúbal/Pegões. Nevertheless, the Qa values do not indicate that any of those species pre- sents a distribution significantly different Table 3. -Analysis of the distribution pattern of from a Poisson one. N. includens in Algarve, for three years. ID = dispersion index, Ja = aggregation index and its The remaining species have Ja, values clo- statistical test, Qa. ID or Ja = 1, random distri- se to or bellow 1, with no significant departu- bution; ID or Ja > 1, aggregated distribution; res from a Poisson distribution. ID or Ja < 1, uniform distribution. N. includens was, as said before, only recently described for Portugal. The possible Year D Ja Qa evolution of the spatial pattern of distribution of this species in Algarve from 1991 to 1994 was evaluated with the help of SADIE (Table 3). No evolution was detected, counts having for the three years a Poisson distribution.

Table 4 shows the four most dominant spe- human interventions add themselves to the cies in each region. Of the original 38 species disturbing factors. only 11 occupy these main dominant posi- The random distributions of A. 10-puncta- tions. S. mediterraneus, S. interruptus and L. ta and S. levaillanti might be the result of the lophantae are the most common. small number of captured individuals, as Among the 11 referred species, 6 are coc- defended by MORRIS (1960). In fact the two cidophagous, 3 have unknown predatory species were included in the analysis though regimes in citrus and the remaining species the number of captured individuals was still feed respectively on white flies and mites. low (12 and 14 respectively, against 20 for O. S. mediterraneus and S. interruptus are conglobata the next more abundant species). dominant in 6 of the 7 considered regions. In what distribution of counts are concer- In what guilds are concerned (Table 5), ned (Ja values, Table 2), the fact that all spe- coccidophagous species and species which cies distributions are not significantly diffe- food preferences are unknown in citrus are rent from a Poisson might be the result of the dominant. Aphidophagous are, in general, sampling strategy; sampling effort was not badly represented, E. nigromaculatus being uniform and samples were collected in a rat- the most abundant (Table 1). her discontinuous way. It is nevertheless pos- sible to say that SADIE results point out the tendency of some species for an aggregated DISCUSSION AND CONCLUSIONS distribution. Among those species N. reunio- ni and N. includens are quite abundant. The results presented in Table 1 confirm N. reunioni was first described in 1974 for that coccinellids are the most important the Reunion island (CHAZEAU, et al, 1974) group among citrus predators in Portugal: 36 and as been since then introduced in several species against only 12 species of Chrisopids countries for the biological control of mealy- (Chrisopidae: Neuroptera) (Pantaleão, et al, bugs, namely P. citri in citrus. In Portugal 1994), 7 of Coniopterigids (Coniopterygi- this coccinellid was first detected in 1990 in dae: Neuroptera) (CARVALHO & FRANCO, the Setúbal region. Its presence might be the 1994), 3 of Anthocorids (Anthocoridae: result of the introduction of some individuals Heteroptera) and 4 of Minds (Miridae: Hete- in a citrus grove at Oeiras (near Lisbon) in roptera)(SiLVA,e/a/., 1994). 1984 (Magro et al, 1992). The identified species correspond to 61% As we referred to, N. reunioni was restric- of the total coccinellids referred to Portugal ted to an 80 Km region around Lisbon but it by RAIMUNDO & ALVES (1986), RAIMUNDO might be spreading, with our sampling met- (1992) and RAIMUNDO (pers. com.). The hod being unable to detect newly established Scymnini tribe is the best represented. small populations. N. includens is a medite- The species mentioned as being represen- rranean species, the first individuals of which ted by 8 or less individuals are probably eit- were captured in Portugal (Algarve) in 1984 her rare species or species that do not use (MAGRO, et al, 1992). The results presented citrus as a typical habitat (HODEK, 1973). in Table 3 make it possible to conclude that The aggregated distributions observed for the interval of time between the introduction almost all species {ID index values, Table 2) of the species and 1991 is long enough to are not surprising for, as RABINOWITZ (1981) allow a random pattern of distribution of refers, changes of orders of magnitude in individuals in the all Algarve region. population sizes occur on the scale of meters TRANFAGLIA & VIGGIANI (1972/73) clai- without striking underlying heterogeneity. In med that this coccinellid presents biological fact, the number of abiotic and biotic condi- characteristics similar to C. montrouzieri, tions that can influence population numbers which leads us to think that its restriction to are huge. In agroecosystems the constant Algarve is mainly due to geographical barriers not allowing its progression to the population of coccid enemies is not sufficient North. The fact that this coccinellid has quite to keep some of those pests under the econo- a narrow range of food can also be in the ori- mical thresholds; it is the case in Portugal for gin of this restriction to Algarve; the region P. citri, S. oleae and/,, beckii. Either the auxi- immediately north of Algarve might not pre- liars are not effective or their action is limited sent the adequate conditions to its survival. by chemical treatments. In what effective- An interesting fact is that R. cardinalis and ness is concerned, MERLIN, et. al. (\996ab) C. montrouzieri, exotic coccinellids introdu- have shown that C. montrouzieri is able to ced in Portugal in 1888 and 1918 respecti- assess the quality of coccid colonies and vely for the biological control of/, purchasi change their oviposition decision accor- and pseudococcids, can now be found in all dingly. In the presence of larvae of their own the citrus regions. species that coccinellid can withhold eggs Although the taxocenoses structure is varia- and leave the colony. Identical results had ble among regions (Table 4 and 5), a basic pat- been previously found for Adalia 2-punctata tern seems to emerge, dominance main posi- (Hemptinne, et. al., 1992). These results tions and dispersion belong to a group of spe- point out the need to complete natural control cies among which we can find well known by the introduction of larvae of these preda- predators of citrus important pests. tors. The important place occupied by species In the second case, the use of specific pes- which food preferences are unknown in ticides only when economical thresholds are general and in citrus in particular is quite attained would possibly reduce greaty the interesting. All those species, belonging to level of those phytophagous and other pests the Scymnini tribe, can be potential predators in general. of pseudococcids, at least in the adult stage, The abundant group of coccinellids which but no specific studies were done until now food preferences are unknown in citrus (MAGRO, 1992). might constitute an amazingly important The overall small representation of aphi- potential for biological control in this crop. It dophagous species might be due to the fact would be important to investigate their role, that summer time is not favourable to their particularly in the case of S. mediterraneus presence in citrus. and S. interruptus. The present work shows the extreme rich- ness of the coccinellid fauna of citrus in Por- tugal. In this country the primary biotic and ACKNOWLEDGEMENTS abiotic conditions for the activity of those species seem to be present. The fact that exo- This paper was written while the first aut- tic species acclimatise and spread without hor was on a postdoctoral fellowship of the programmed introductions is very significa- Fundação para a Ciência e Tecnologia -Por- tif. This situation contrasts greatly with what tugal- (Programme Praxis XXI). The rese- has been observed for the countries of the arch was supported by a PhD grant from the close mediterranean region (e.g. LONGO & same programme. BENFATTO, 1987). We are pleased to acknowledge the assis- The captured predators cover an all array tance of Dr. J. N. Perry for supplying SADIE of preys, coccidophagous species being the software, S. Patiny for help to produce the best represented. This is quite important as UTM maps and A.M. Paixão and M.F. several coccids attack citrus. It is however Moura for the collaboration to collect some well known that the action of the natural samples. REFERENCES

ALSTON, R., 1994: Statistical analysis of animal popula- MAJERUS, M. E. N., 1994: Ladybirds. London, Harper tions. PhD Thesis, University of Kent, (cit.: PERRY, Collins, 367 pp. 1995a). MERLIN, J.; LEMAITRE, O. & GRÉGOIRE, J. C, 1996a: Ovi- AMARO, P. & BAGGIOLONI, M., 1982: Introdução à Pro- position in Cryptolaemus montrouzieri stimulated by tecção Integrada. Vol. 1, Lisboa, FAO/DGPPA, 276 pp. wax filaments of its prey. Entomologia Experimentalis CARVALHO, B. A. H. P. & FRANCO, J. C, 1994: Coniopteri- etApplicata, 79: 141-146. gídeos associados aos citrinos - Estudo realizado em MERLIN, J., LEMAITRE, O. & GRÉGOIRE, J. C, 19966: Che- dois pomares da região de Setúbal. In AMARO, P. & mical cues produced by conspecific larvae deter ovipo- FRANCO, J. C. (Eds.), Actas I Congresso de Citricultura. sition by the coccidophagous ladybird beetle, Cryptola- Silves-Algarve, 20-22 de Janeiro de 1993, Câmara emus montrouzieri. Entomologia Experimentalis et Municipal de Silves: 433-442. Applicata, 79: 147-151. CARVALHO, J. PASSOS; FRANCO, J. C; AGUIAR, M. F. & SOA- MORRIS, R. F., 1960: Samplin- populations. Annu. RES, A. O.: Insect Pests of Citrus in Portugal. In Proc. Rev. Entomol, 5: 243-264. Vill Congr. Int. Soc. Citric, 12-17 May 1996. (in press). PAIXÃO, A. M. T. P., 1994: Coccinelideos em citrinos no CHAZEAU, J.; ETIENNE, J. & FÜRSCH, H., 1974: Les Cocci- Algarve: cartografia e estrutura da taxocenose. Rela- nellidae de Pile de la Reunion (Insecta: Coleóptera). tório de Estágio em Recursos Faunísticos e Ambiente, Bull. Mus. Hist. nat. Paris, 3e série, n.° 210: 265-297. FCUL, Lisboa, 91 pp. COPPEL, H. C. & MERTINS, J. W., 1977: Biological insect PANTALEÃO, F.; CARVALHO, J. PASSOS; FRANCO, J. C. & pest suppression. Germany, Springer-Verlag, 314 pp. MAGRO, A., 1994: Crisopídeos associados aos citrinos. DIXON, A. F. G.; HEMPTINNE, J. L. & KINDLMANN, P. 1997: In AMARO, P. & FRANCO, J. C. (Eds.), Actas I Congres- Effectiveness of ladybirds as biological control agents: so de Citricultura. Silves-Algarve, 20-22 de Janeiro de patterns and processes. Entomophaga, 42 (1/2): 71-83. 1993, Câmara Municipal de Silves: 427-431. ELLIOT, J. M., 1973: Some Methods for the Statistical PERRY, J. N., 1995a: Spatial analysis by distance indices. J Analysis of Samples ofBenthic Invertebrates. Scientific Anim. Ecol. 64: 303-314. Publication No. 25, Freshwater Biological Association, - 19956: Spatial Aspects of Animal and Plant Distribu- Ambleside, Westmorland, Great Britain, (cit. LUDWIG tion in Patchy Farmland Habitats. In GLEN, D. M.; GRE- & REYNOLDS, 1988). AVERS, M. P. & ANDERSON, H. M. (Eds.), Ecology and FÜRSCH, H., 1987a: Übersicht Über die Genera und Sub- Integrated Farming Systems. United Kingdom, John genera der Scymnini mit besonder Berücksichtigung Wiley & Sons Ltd.: 221-242. der Westpalaerktis. Entomol. Abb. Mus. Tierkd. Dres- - & HEWITT, M., 1991: A new index of aggregation for den, 51(4): 57-74. animal counts. Biometrics, 47: 1505-1518. (cit.: PERRY, - (1987b): Die Scymnini der Kanaren, Azoren und 1995a). Madeiras. Acta Coleot., 3: 1-14. RABINOWITZ, D., 1981: Seven forms of rarity. In HUGH, S. GOURREAU, J. M., 1974: Contribuition à l'étude de la tribu (Ed.) The Biological Aspects of Rare Plant Conserva- des Scymnini (Coccinellidae). Ann. Zool. Ecol. anim., tion. Chichester, John Wiley & Sons Ltd.: 205-217. n.° Hors-série, 223 pp. RAIMUNDO, A., 1992: Novas espécies de Scymnini para a HEMPTINNE, J. L.; DIXON, A. F. G. & COFFIN, J., 1992: fauna de coccinelideos de Portugal. Bom. Soe. Por. Attack strategy of ladybird beetles (Coccinellidae): fac- Ent., supl. 3, Vol. 1:373-384. tors shaping their numerical response. Oecologia, 90: - & ALVES, M. L., 1986: Revisão dos Coccinelideos de 238-245. Portugal. Évora, Univ. Évora, 103 pp. HODEK, I., 1970: Coccinellids and the Modern Pest Mana- RASMONT, P.; BARBIER, Y. & EMPAIN, A., 1993: Microban- gement. BioScience, 20(9): 543-552. que Faune-Flore. Logiciel de gestión de banques de - (1973): Biology of Coccinellidae. The Hagen-Holland, donnés biogéografiques. Belgique, Université de Dr. W. Junk, 260 pp. Mons-Hainaut et Jardín Botanique National, 200 pp. INSTITUTO NACIONAL DE ESTATÍSTICA, 1995: Estatísticas SILVA, A. C. P. P., FRANCO, J. C. & CARVALHO, J. PASSOS, Agrícolas. Lisboa, INE: 59. 1994: Mirídeos e Antocorídeos associados aos citrinos. IPERTI, G., 1987: Insetti utili: i coccinellidi. Informatore Estudo realizado em dois pomares da região de Setúbal. Fitopatol, 2:21-27. In AMARO, P. & FRANCO, J. C. (Eds.), Actas I Congres- LONGO, S. & BENFATTO, D., 1987: Coleotteri entomofagi so de Citricultura. Silves-Algarve, 20-22 de Janeiro de presentí sugli agrumi in Italia. Informatore Fitopatol, 1993, Câmara Municipal de Silves: 421-426. 37(7-8): 21-30. SZUJECKI, A., 1987: Ecology offorest insects. LUDWIG, J. A. & REYNOLDS, J. F., 1988: Statistical Eco- Dordrecht/Boston/Lancaster, Dr. W. Junk, 601 pp. logy. New York, John Wiley & Sons Inc., 337 pp. TRANFAGLIA, A. & VIGGIANI, G. 1972/73: Dati biologici MAGRO, A. 1992: Os Coccinelideos na luta biológica con- sullo Scymnus includens Kirsch (Coleóptera: Coccine- tra os Pseudococcideos associados a citrinos. Disser- llidae). Boll. Lab. Ent. Agr. Portici, 30: 9-18.. tação de Mestrado em Protecção Integrada, ISA-UTL, Lisboa, 198 pp. - 1997: Os Coccinelideos dos citrinos: estudo compara- tivo do seu interesse em luta biológica. Tese de Douto- ramento, Univ. Évora, Évora, 177 pp. FRANCO, J. C. & RAIMUNDO, A., 1992: Dados preliminares sobre a distribuição de duas espécies do género Nephus (Col.: Coccinellidae), novas para Portugal, referencia- das em ecossistemas citrícolas. Bolm. Soc. Por. Ent., (Recepción: 9 abril 1999) supl. 3, Vol. 2: 281-289. (Aceptación: 13 agosto 1999)