Notalepid. 27 (4): 309-319 309 Prominent shoots are preferred: microhabitat preferences of & Maculinea alcon ([Denis Schiffermüller], 1775) in Northern Germany (Lycaenidae) & Annabell Küer' Thomas Fartmann^ ' SteinfurterStr. 502, D-48159Münster, email: [email protected] 2 InstituteofLandscapeEcology, DepartmentofCommunityEcology,University ofMünster, Robert-Koch-Straße 26, D-48149 Münster,email: [email protected] (corresponding author) Abstract. The egg deposition behaviouroftheAlcon Blue, Maculinea alcon ([Denis & Schiffermüller], 1775), was investigated in summer 2002 on a military training area near Osnabrück (North Rhine- Westphalia,NorthernGermany). The study aimstoanalyseovipositionpatternson the flowersofthehost plant Gentianapneumonanthe (MarshGentian) and microhabitatcharacteristics ofthe chosen shoots.All possiblehostplantsandalleggshellsofM.alconwerecountedandvariouscharacteristicsofthehostplant aswellasthesurroundingvegetationstructurewererecorded.Atotalof1,787eggswascountedon 124out of219 Gentiana shoots in an areaof3,200 wr.About70% were laidon the calyx and 25% on the flower. The vegetation {Juncus-Succisa pratensis association with patches of Ericion tetralicis) was generally denseandratherhigh. Occupiedspecimenshadalmostalways aluxuriantgrowthandweremostlyhigher than the surrounding vegetation. In a logistic regression model the height-difference between the plant andthevegetation (prominence),thenumberofflowerspershoot, andthenumberoffurthershoots in the surrounding explained 78% of observed plant occupancy patterns. Finally suggestions to maintain and stabilizeAlconBluepopulations aregiven. Zusammenfassung. Das Eiablageverhalten des Lungenenzian-Ameisenbläulings, Maculinea alcon ([Denis&Schiffermüller], 1775),wurdeimSommer2002aufeinemTruppenübungsplatzinderNähevon Osnabrück (Nordrhein-Westfalen, Norddeutschland) untersucht. Ziel der Arbeit war es, Eiablagemuster an den Blüten derWirtspflanze Gentianapneumonanthe (Lungenenzian) sowie Mikrohabitatpräferenzen beiderBelegungzustudieren. DazuwurdenalleLungenenziansprosseunddieEiervonM. alcongezählt. Verschiedene ParameterderWirtspflanzen sowie die Vegetationsstruktur im Umkreis von 50 cm wurden aufgenommen. Von insgesamt 219 Sprossen aufeiner Fläche von 3.200 m- waren 125 mit 1.787 Eiern belegt. Etwa 70% davon wurden an den Kelch geheftet und 25% an die Blüte. Die Vegetation {Juncus- Succisapratensis-A^?>oz\3X\on mit einzelnen Ericion tetralicis-Flccken) war meist sehr dicht und relativ hoch. Belegte Lungenenzian-Exemplare waren in der Regel üppig und überragten größtenteils die umgebende Vegetation. Ein Modell der logistischen Regression zeigt, dass mit Hilfe der Höhendifferenz zwischen Spross und Vegetation, der Zahl der Blüten und der Anzahl der umgebenden Lungenenzian- Sprosse78%derDatenkorrektvorhergesagtwerden.AbschließendwerdenVorschlägezumErhaltundzur Stabilisierung vonPopulationendes Lungenenzian-Ameisenbläulings gemacht. Key words. Alcon Blue, egg deposition, Gentianapneumonanthe,Maculinea alcon, Marsh Gentian, management, microhabitatpreferences,Germany. Introduction Habitat quality has been shown to be as important for the persistence of butterfly metapopulations as the degree of patch isolation and patch size (Anthes et al. 2003; Dennis & Eales 1997; Thomas et al. 2001). Thomas et al. (2001) and WallisDeVries (2004) pointed out that the criteria are not alternatives but should be considered complementarily. While many studies have recently addressed the effect ofpatch size and isolation on population dynamics for various butterflies, habitat quality is still comparably ill-defined for many endangered species. Habitat preferences ofbutterflies are often largely determined by the requirements of the preimaginal stages since the eggs are not and the larvae are only slightly mobile (Fartmann 2004; Porter 1992). In caseofadverse weatherorrestrictedfood supply they are not able to escape. Therefore, Notalepidopterologica,07.06.2005, ISSN0342-7536 310 KüER&Fartmann: MicrohabitatpreferencesofMaculineaalcon the evaluation ofhabitat quality for a particular species requires a detailed knowledge ofthe preferences ofovipositing females and the survival ofpreimaginal stages under various conditions and across regions. Here we studied the larval habitat preference of the Alcon Blue {Maculinea alcon & ([Denis Schiffermüller], 1775)). Although the butterfly genus Maculinea van Eecke, 1915 has attracted considerable attention because ofits extraordinary relationship with ants ofthe genus Myrmica Latreille, 1804 (Munguira & Martin 1999; Als et al. 2004), recent descriptions of larval habitat requirements are scarce. The status of the Alcon Blue has been assessed as vulnerable in Europe (van Swaay & Warren 1999) and as endangered in Germany (Pretscher 1998); in North Rhine-Westphalia the species is critically endangered (Dudler et al. 1999). In many parts of Germany it is extinct (Fig. 1). Whereas quite a lot of research has addressed the relationship between M. alcon larvae and their host ants (e.g. Als et al. 2001; Elfferich 1988; Eimes et al. 1994; Liebig 1989; van Dyck et al. 2000), the relationship between M. alcon and its host plant Gentianapneumonanthe (Marsh Gentian) is much less studied. Krismann (2000) studied oviposition patterns on the host plant and found a preference for egg-laying on the calyx. WallisDeVries (2004) compared habitat characteristics of occupied and unoccupied sites atamesoscaleof 10 x 10 m. Thepreferredvegetation structureforegg deposition is mostly known for the sibling species Maculinea rebeli (Hirschke, 1904) (Dolek et al. 1998; Kockelke et al. 1994; Meyer-Hozak 2000) but not forM. alcon, yet Marktanner (1985) observed that it avoids dense and overshadowed vegetation. The status of both forms as distinct species is strongly questioned by the recent genetical data (Als et al. 2004). This study aims to increase our knowledge of the microhabitat structure at M. alcon oviposition sites. In particular we considered the following questions: (i) What are the oviposition patterns on the Marsh Gentian? (ii) Which kind ofvegetation structure does theAlcon Blue prefer for egg deposition? (iii) Which conclusions can be drawn for the management oftheAlcon Blue sites? Material and Methods Study species. Maculinea alcon ([Denis & Schiffermüller], 1775) has a scattered distribution across Europe up to EastAsia (Wynhoff 1998). In Germany theAlcon Blue is mainly found on the foothills oftheAlps and in the Northwestern Lowlands (Fig. 1). Its flightperiod in Germany extends fromearlyJulyto mid-August (Ebert& Rennwald 1991; Wynhoff et al. 1999). M. alcon thrives on moist meadows, wet heathland and fens with stands of its host plant, the Marsh Gentian {Gentiana pneumonanthe). In the Alps the Willow Gentian (G. asclepiadea) is used as well. The Alcon Blue has a complex life cycle and is dependent not only on the presence of its host plant but also on the presence of host ants of the genus Myrmica. In Southern Europe and the pre-alpine region of Germany (Nunner pers. comm.) only M. scabrinodis Nylander, 1846 serves as a host. In Middle and Northern Europe it is fully replaced by M. rubra (Linnaeus, 1758) and M. ruginodis Nylander, 1846 (Eimes et al. 1998). Females of M. alcon lay their eggs on the buds of G. pneumonanthe. Through basal hatching the larvae get into the flower and feed there until the fourth larval stage. After emerging from the flowerhead they letthemselves drop on the ground and wait to be carried into Notalepid. 27 (4): 309-319 311 Fig. 1. Study area in Northern Germany (a) and distribution ofMaculinea alcon in Germany (b). Grid: 10' X 6' geographic grid. Grey dots: data before 1985, black dots: data since 1985, data from: BLfU (2001), de Lattin (1957), Ebert & Rennwald (1991), Habel (2003), Harkort (1975), Kinkier & Schmitz (1971), Kölligs (2003), natural history museum Muenster (own observation), Retzlaff (1973), Retzlaff et al. (1993), Stamm (1981),Wagener& Niemeyer (2003) andM. Goldschalt, H. G. Joger,A. Krismann, A. Nunner,T. Marktanner,R. Reinhardt,T. Schulte (ineachcasepers. comm.)- a nest oftheir host ant where they Uve 10 to 22 months until pupation (Schönrogge et al. 2000; Thomas et al. 1998). G. pneumonanthe (Gentianaceae) is distributed throughout Europe and Asia and has its core range in Westem Europe (Komeck et al. 1998). The perennial plant flowers between July and September on oligotrophia humid sites such as litter meadows ('Streuwiesen') and moist to wet heathland (Oberdorfer 2001; Sebald et al. 1996). The light-requiring G. pneumonanthe grows up to 50 cm high with up to 10 shoots and up to 25 flowers per shoot(Ellenberg 1996; Roseetal. 1998). Thereproductiononly takesplacebythe means of its smallseeds,whichareadaptedtoshortdistancewinddispersal(Oostermeijeretal. 1998). & Forsuccessful germinationthe seedsrequiremoistandbare soil (Kesel Urban 1999). InCentralEuropeandinGermanyG.pneumonantheisdeclining.Thisisduetoanintensified agricultural use on one hand. On the other hand the abandonment of smaller unprofitable habitats supports the succession on these sites. Both factors endanger the survival of the MarshGentian. Study area. The study area is located in the district Steinfurt in the north of North Rhine-WestphaliaadjacenttoLowerSaxony (Fig. 1). The study site is partofamilitary training area. The vegetation is dominated by wet grasslands {Molinietalia) with some patches ofAgrostietalia and Nardo-Callunetea. Gentianapneumonanthe was growing within two sites (7,500 m^ and 15,000 m^) of the study area, but only on the bigger site Maculinea alcon was present in 2002. This place was surrounded by wood and consisted of the dominating Juncus-Succisa pratensis association with patches of Ericion tetralicis. Within this site the host plant grew on an area ofabout 3,200 m^. 1 312 KüER&Fartmann: MicrohabitatpreferencesofMaculineaalcon Tab.1.Habitatstructureparametersatoccupied(n= 124)andunoccupied(n=95)Gentianapneumonanthe shoots(Mann-Whitney (/Test: *** P<0.001, * P<0.05). nG.pneumonanthe: numberofothershootsnear theobservedspecimen; prominence: gentian shootheightminus average vegetationheight. Parameter Minimum-Maximum Median U P occupied unoccupied occupied unoccupied Flowerspershoot 1-14 1-10 4 2 2847 *** Shootheight[cm] 22-60 22-51 39 35 3784 *** Averagevegetationheight[cm] 25-55 25-55 45 50 3397 *** Prominence[cm] -29-20 -30-6 -3.4 -15 2030 *** nGentianapneumonanthe 0-13 0-1 5 4 4981 * Horizontalherbaceouscover[%] in30cm 0-95 2.5-95 10 50 3751 *** Field Study. In summer 2002 after the flight period of Maculinea alcon all shoots of Gentianapneumonanthe were checked for eggs. For each gentian shoot we determined total height, the number of flowers and the height of each flower. The eggshells were counted bud-wise distinguishing between top, middle and base of flower, calyx, leave and stalk. For microhabitat analysis the following parameters were collected within a radius of50 cm around the shoot: distance to the next shoot, numberofother shoots, the maximum and average vegetation height, the vegetation cover and horizontal vegetation coverin 10to45 cmheightabove soil surface (estimated in5%-steps).Agridof 10 x 10m was put on the study site to determine these parameters also on a bigger scale. For data analysis we calculatedthe difference ofthe shootheight and average vegetationheightto show the 'prominence'ofthe host plant. Negative values express anegative prominence, which means the shoot is smaller than the surrounding vegetation. Positive values show accordingly apositive prominence ofthe gentian shoot. Data Analysis. Literature data showed that a single Maculinea alcon female lays on the average 50-100 eggs (Maes et al. 2004). Meyer-Hozak (2000) found out that the sibling species Maculinea rebeli lays 100-150 eggs per female. The primary sex ratio in a population is 1 : 1 (for M. rebeli: Kockelke et al. 1994; Meyer-Hozak 2000). We therefore used the total egg count from this study to estimate the adult population size in 2002. Toassesstheexplanatory powerofdifferentvariablesontheoccupancyofgentian shoots we used a stepwise-forward logistic regression. All statistical analysis was performed with SPSS 1 1.0.1 statistical analysis package. Results We found a total of 219 Marsh Gentian shoots with 824 flowers. Ofthose, 124 shoots (57%) and 473 flowers (57%) were occupied with 1,787 eggs. Based on the total egg count the adult population size was estimated at 18-36 individuals. The preferred place of oviposition was the calyx in 70% of the cases, followed by the flower with 25%. The stem and the leaves played a minor role in egg deposition (Fig. 2). The height distribution of the eggs and the flowers were quite similar. About 2/3 of both flowers and eggs were found at 26 to 40 cm above ground (Fig. 3). 313 Notalepid. 27 (4): 309-319 14001 250 1200 1^ 00 200 S 150 S 800 0) 100 i 6oo^ E 50 c 400 200 11- 16- 21- 26- 31- 36- 41- 46- 51- >55 15 20 25 30 35 40 45 50 55 0 I I FT FM FB heightofflowers [cm] Fig.2.Ovipositionplaceson Gentianapneumonanthe(FT=flowertop,FM=flowermiddle,FB =flower base, C=calyx, S = stalk, L=leave). Fig.3. Heightofoccupied (grey) andunoccupied (white) flowersofGentianapneumonanthe. The vegetation at the egg-deposition places was relatively high and dense. Maximum vegetation height ranged from 80 to 100 cm. Vegetation cover was mostly 100%, only in some cases it was with 80-95% slightly less dense. The Marsh Gentians were scattered over the whole site but had a clustered occurrence in some places especially along old tank tracks. Most part of the study site was dominated by the Purple Moor Grass (Molinia caerulea) and was characterized by a vivid change of hummocks and hollows. Only some gentian shoots were found on a drier and more even area, which was dominated by the Tufted Hair-grass {Deschampsia cespitosa). Occupied gentians were generally higher and had more flowers than unoccupied specimen. High average vegetation height and low horizontal vegetation cover decreased the likelihood of a host plant to be accepted for oviposition (Tab. 1). Most occupied stalks were higher or only little lower than the average vegetation (Fig. 4). The more prominent a shoot the more eggs it received (Fig. 5). The distribution ofoccupied and unoccupied Marsh Gentian shoots was best explained by the combination ofheight difference (prominence), number offlowers per stalk and number ofother Marsh Gentian shoots in the proximity. The logistic regression model classified 78% ofthe data correctly by means ofthese three parameters (Tab. 2). Thehatchedgrids inFig. 6 show the average heightdifference. The moreprominentthe gentian shoots were the more likely they were to be chosen for oviposition. Discussion Microclimatic aspects play an important role in butterfly oviposition (Fartmann 2004; Porter 1992; Thomas et al. 1998). As all gentians grew in sunny areas they were theoretically equally available for egg deposition. However, only about half of the shootswereoccupiedwitheggs. Notonlythesizeoftheplantandthenumberofflowers. 314 KüER&Fartmann: MicrohabitatpreferencesofMaculineaalcon Tab.2. Stepwise-forwardlogisticregressionmodelontheinfluenceofhabitatstructureparametersonthe eggdepositionpreferenceofMaculineaalcon(hostplantshootsn=219).n.s.=notsignificant;nGentiana pneumonanthe: number of other shoots near the observed specimen; prominence: gentian shoot height minus average vegetationheight. Parameter CoefficientB SD P R ModelimprovementChi^ Constant -0.027 0.497 n.s. Prominence [cm] 0.143 0.025 <0.001 0.326 76.49 Flowerspershoot 0.263 0.089 <0.005 0.150 10.83 nGentianapneumonanthe 0.139 0.052 <0.05 0.132 7.78 Shootheight[cm] 0.657 n.s. 0.000 Averagevegetationheight[cm] 0.657 n.s. 0.000 Horizontalherbcover[%] 0.159 n.s. 0.000 Modelsummary Chi'=95.1,df-3,P<0.001 Correctlyclassified 78.1% 20 80 70 10 60 50 40 E 1-10 30 20 -20 10 Ok -30 -30 -20 -10 0 10 20 30 124 95 occupied unoccupied prominence[cm] Fig. 4. Difference of hostplant height and average vegetation height (prominence) of occupied and unoccupied Gentianapneumonanthe shoots (IVlann-Whitney t/Test: U = 2030, P<0.001). CompareTab. 1 forstatistics. Fig. 5. Number of eggs per Gentiana pneumonanthe shoot in relation to the prominence (host plant shoots n = 124). r^=0.622, n = 124, P<0.01. Prominence: gentian shootheightminus average vegetation height. but especially the shoot height relative to the height ofthe surrounding vegetation are important. Results ofthe logistic regression model show that the preferred oviposition places are shoots with many flowers that protrude the vegetation and are surroundedby other Gentiana shoots. It is a common phenomenon that females choose large and conspicuous host plant individuals (Porter 1992). Maculinea rebeli prefers luxuriant specimens of its host plant Gentiana cruciata that are easy to reach (Dolek et al. 1998; Meyer-Hozak 2000). First, visual attraction is an important factor when searching for a suitable host plant 315 Notalepid. 27 (4): 309-319 — Average prominence wLm -16.5--12 a) -11.9--8 lüH -7.9-A -3.9-0 0.1-3 nohostplantexisting O Numberofshoots MarshG—entian shoots ^ O — unoccupied 1-5 occupied O 6-10 o 11-20 21-30 31-40 b) >40 Numberofdeposited eggs 1-50 51-100 I 101-200 201-400 I I >400 V m 10 Fig. 6. Occupied and unoccupied Marsh Gentian shoots (a) and number ofdeposited eggs (b) per grid in relationtopromincence.Theaverageprominence isthemedianofall recordedheightdifferencespergrid. 316 KüER&Fartmann: MicrohabitatpreferencesofMaculineaalcon (Dolek et al. 1998). Prominent shoots can be seen easily and are hence predestined as oviposition site. Second, shoots that grow higher than their surrounding vegetation are less shaded and offer better microclimatic conditions for a quick development of eggs and larvae. Besides it is risky to lay eggs on gentians hidden in dense vegetation because ofthe orb-web spiders, which are common in unmown meadows in this time ofthe year (Nunnerpers. comm.). Third, biggergentians may produce biggerbuds and therefore offer more food resources for the larvae. Over occupation of the shoots is seldom. Only 10 out of 124 shoots had more than 6 eggs per flower. About 4 to 6 Maculinea alcon larvae can feed on one flower (Eimes &Thomas 1987; Ebert & Rennwald 1991), which means thatforalmostall larvaethere are sufficient food resources. Despite the impression that the striking white eggs of M. alcon are mainly laid on the blue flowers of Gentiana pneumonanthe we found almost three quarters of the , eggs on the calyx but only one quarter on the flower. On the foothills of the Alps, Krismann (2000) found a similar distribution pattern on both G. pneumonanthe and G. asclepiadea. This ovipositionpattern makes sense taking into accountthatthelarvae of M. alcon leave the eggs through basal hatching (Thomas et al. 1991). They bore through the calyx and directly move to their food resource, the plant ovary. Since M. alcon larvae hatch rapidly, the strong exposition of the eggs might be contributing to a fast larval development (Porter 1992). Within 3-4 weeks the larvae have to reach the fourth larval instar and need therefore optimal microclimatic conditions such as on the concealed flowers are provided. Although the eggs are more or less unprotected and very conspicuous, they are rarely parasitised. This may be due to the thick eggshells, which effectively prevent perforation by parasitoid wasps (Thomas et al. 1991). Thick eggshells may further protect against bad or hot weather. Thestudysiteliesfallow;thereisnoregularutilizationorcare.Thereforeitisdominated by the Purple Moor Grass (Molinia caerulea). Vegetation cover was mostly 100%. Nevertheless the Marsh Gentian is able to survive in such dense vegetation because of its longevity (Rose et al. 1998). On the study site mostly adult G. pneumonanthe grow. Oostermeijer et al. (1994) call this population type 'senile'. However, a senile population provides advantage forM. alcon regarding the suitability ofegg deposition as most ofthe plants are relatively high and mostly covered with several flowers. This is reflected in the proportion ofoccupied gentians, which was with over 50% very high (compare Habel 2003; Krismann 2000). Still for a long-term survival of the gentian population and also of the Alcon Blue population a rejuvenation of the host plants is necessary which means creating gaps ofbare soil. There are different ways to assure the regeneration of G. pneumonanthe populations. & Kesel Urban (2000) andWallisDeVries (2004) suggestthat small-scale sodcutting is best to promote existing gentian populations. Gaps are created in which the seeds can germinate and the young seedlings can grow protected. Though mowing is probably the better alternative to prevent from floristic impoverishment, to keep the vegetation open and to support the growth of accessible gentian shoots (Nunner pers. comm.). The best time is in October when the gentian seeds are mature and the M. alcon larvae 317 Notalepid. 27 (4): 309-319 are adopted (Briemle & Ellenberg 1994; Nunner pers. comm.). The hay should be taken away to prevent litter accumulation (Fartmann & Mattes 1997). Trautner et al. (2004) and WallisDeVries (2004) suggest that extensive grazing is also appropriate as management for M. alcon habitats with G. pneumonanthe as host plant. The gentian plants grow less luxurious but the small flowers seem to offer enough food for the Maculinea larvae until adoption. However, grazing is not an alternative when G. pneumonanthe sites are small because of the risk of local overgrazing or when G. asclepiadea is the host plant because it is more sensitive to browsing. As most Marsh Gentian plants on the study site grow along old tank tracks it may be supposed that occasional mechanic disturbance through tanks or other means can have positive effects ifit happens only every couple ofyears. Acknowledgements The idea for this study came from C.Artmeyer. H. Mattes gave expert advice about the manuscript. We thank N.Anthes, K.Arnold, G. Hermann, A. Nunner and M. WallisDeVries for valuable comments. Distribution data of Maculinea alcon were provided by H. Dudler, M. Goldschalt, H. G. loger, A. Krismann,T. Marktanner,A. Nunner,R. ReinhardtandH. Retzlaff. References Als, T. D., D. R. Nash & J. J. Boomsma 2001. Adoption of parasitic Maculinea alcon caterpillars (Lepidoptera: Lycaenidae) bythreeMyrmica ant species. -Animal Behaviour62: 99-106. Als,T. D.,R. Vila,N. R Kandul,D. R. Nash,S.-H.Yen,Y.-F. Hsu,A.A. Mignault,J. J. Boomsma&N. E. Pierce2004.Theevolutionofalternativeparasitic lifehistories in largebluebutterflies. -Nature432: 386-390. Anthes, N., T. Fartmann, G. Hermann & G. Kaule 2003. Combining larval habitat quality and meta- population structure-the key forsuccessful managementofpre-alpineEuphydryasauhnia colonies. -JournalofInsectConservation7: 175-185. Bayerisches Landesamt für Umweltschutz (BLfU) 2001. Artenschutzkartierung Bayern: Arbeitsatlas Tagfalter-Augsburg. Briemle,G.&H.Ellenberg 1994.ZurMahdverträglichkeitvonGrünlandpflanzen.-NaturundLandschaft 69: 139-147. deLattin,G. 1957.DieLepidopteren-FaunaderPfalz.I.Teil.-MitteilungenderPollichia(ser.4)3: 51-167. DennisR. L. H. &H. T.Eales 1997. PatchoccupancyinCoenonymphatullia(Müller, 1764)(Lepidoptera: Satyrinae):habitatqualitymattersasmuchaspatchsizeandisolation.-JournalofInsectConservation 1: 167-176. Dolek,M.,A. Geyer&R. Bolz 1998. DistributionofMaculinearebeliandhostplantuseonsitesalongthe riverDanube. -JournalofInsectConservation2: 85-89. Dudler, H., H. Kinkier, R. Lechner, H. Retzlaff, W. Schmitz & H. Schumacher 1999. Rote Liste der gefährdeten Schmetterlinge (Lepidoptera) in Nordrhein-Westfalen. Pp. 575-626. - In: Landesanstalt für Ökologie, Bodenordnung und Forsten/Landesamt für Agrarordnung Nordrhein-Westfalen (LÖBF), Rote Listedergefährdeten Pflanzen undTiere inNordrhein-Westfalen, 3. Fassung-LÖBF- Schriftenreihe 17, Recklinghausen. Ebert,G. &E. Rennwald 1991. Die SchmetterlingeBaden-Württembergs2(2).-Ulmer, Stuttgart. 535 pp. Elfferich, N.W. 1988. Zuchterfahrungen mit Maculinea alcon (Denis & Schiffermüller, 1775) (Lepi- doptera,Lycaenidae). -MitteilungenderEntomologischenGesellschaftBasel38: 134-150. Ellenberg,H. 1996.VegetationMitteleuropasmitdenAlpeninökologischer,dynamischerundhistorischer Sicht.-Ulmer, Stuttgart. 1059 pp. Eimes, G.W. & J.A. Thomas 1987. Die Gattung Maculinea. Pp. 354-368. - In: Lepidopterologen- Arbeitsgruppe,TagfalterundihreLebensräume. -SchweizerischerBundfürNaturschutz, Basel. 318 KüER&Fartmann: MicrohabitatpreferencesofMaculineaalcon Eimes, G.W., J.A. Thomas, O. Hammarstedt, M. L. Munguira, J. Martin & J. Van Der Made 1994. Differencesinhost-antspecificitybetweenSpanish,DutchandSwedishpopulationsoftheendangered butterflyMaculineaalcon (Deniset. Schiff.) (Lepidoptera).-MemorabiliaZoologica48: 55-68. Eimes, G.W., J.A. Thomas, J. C. Wardlaw, M. E. Hochberg, R. T. Clarke & D. J. Simcox 1998. The ecology ofMyrmica ants in relation to the conservation ofMaculinea butterflies. -Journal ofInsect Conservation2: 67-78. Fartmann, T. 2004. Die Schmetterlingsgemeinschaften der Halbtrockenrasen-Komplexe des Diemeltales - Biozönologie von Tagfaltern und Widderchen in einer alten Hudelandschaft. -Abhandlungen aus demWestfälischenMuseumfürNaturkunde66 (1): 1-256. Fartmann, T. & H. Mattes 1997: Heuschreckenfauna und Grünland - Bewirtschaftungsmaßnahmen und Biotopmanagement. Pp. 179-188. - In: H. Mattes (ed.), Ökologische Untersuchungen zur HeuschreckenfaunainBrandenburgundWestfalen.-ArbeitenausdemInstitutfürLandschaftsökologie Münster3. Habel, J. C. 2003. Auswirkungen der Fragmentierung von Lebensräumen am Beispiel von Maculinea alcon (Denis & Schiffermüller 1775). - Diplomarbeit, Institut für Ökologie und Umweltchemie, UniversitätLüneburg. 83 pp. Harkort,W. 1975.SchmetterlingeinWestfalen(ohneOstwestfalen).FundortkartenundFundortlisten(Stand: Ende 1974).-DortmunderBeiträgezurLandeskunde: NaturwissenschaftlicheMitteilungen9: 33-102. Hochberg, E., J. A. Thomas & G. W. Eimes 1992. A modelling study of the population dynamics of a Large Blue butterfly, Maculinea rebeli, a parasite ofred ant nests. - Journal ofAnimal Ecology 61: 397^09. Kesel, R. & K. Urban 1999. Population dynamics of Gentianapneumonanthe and Rhynchosporafusca during wetheathlandrestoration. -AppliedVegetation Science2: 149-156. Kesel, R. & K. Urban 2000. Species dynamics in wet heathland restoration. - Proceedings lAVS Symposium: 286-290. Kinkier, H. &W. Schmitz 1971. DieTagfalterdes BergischenLandes. -Jahresberichte des Naturwissen- schaftlichenVereinsWuppertal 24: 20-63. Kockelke, K., G. Hermann, G. Kaule, M. Verhaagh & J. Settele 1994. ZurAutökologie und Verbreitung des Kreuzenzian-Ameisenbläulings,Maculinea rebeli(Hirschke, 1904).-Carolinea52: 93-109. Kölligs, D. 2003. Schmetterlinge Schleswig-Holsteins.-Wachholtz, Kiel. 212pp. Korneck, D., M. Schnittler, F. Klingenstein, G. Ludwig, M. Takla, U. Bohn & R. May 1998. Warum verarmt unsere Flora? Auswertung der Roten Liste der Farn- und Blütenpflanzen Deutschlands. Pp. 299^44. - In: Bundesamt für Naturschutz (eds.), Ursachen des Artenrückgangs von Wildpflanzen und Möglichkeiten zur Erhaltung derArtenvielfalt. - Schriftenreihe für Vegetationskunde 29, Bonn BadGodesberg. Krismann, A. 2000. Zum Eiablageverhalten von Maculinea alcon (D. & S., 1775) an Gentiana pneumonanthe und Gentiana asclepiadea. Pp. 103-110. -In: J. Settele & S. Kleinewiedtfeld (eds.), Populationsökologische Studien anTagfaltern2. UFZ-Bericht.-UFZLeipzig-Halle,Leipzig. Liebig, W.-H. 1989. Bemerkungen zur Biologie von Maculinea alcon Schiff. (Lep., Lycaenidae). -Entomologische Nachrichten undBerichte33: 117-121. Maes, D., W. Vanreusel, W. Talloen & H. van Dyck 2004. Functional conservation units for the endangeredAlconBluebutterflyMaculineaalconinBelgium(Lepidoptera: Lycaenidae).-Biological Conservation 120: 229-241. Marktanner, T. 1985. Die Bedeutung des Schwalbenwurzenzians als Futterpflanze von Maculinea alcon (Lep.: Lycaenidae) im oberschwäbischenAlpenvorland und die Verbreitung derLycaenide in diesem Raum. -EntomologischeZeitschrift95: 257-272. Meyer-Hozak, C. 2000. Population biology ofMaculinea rebeli (Lepidoptera: Lycaenidae) on the chalk grasslands of Eastern Westphalia (Germany) and implications for conservation. - Journal of Insect Conservation4: 63-72. Munguira, M. L. & J. Martin 1999. Action plan for the Maculinea Butterflies in Europe. - Nature and Environment97. Council ofEurope Publishing, Strasbourg. 63 pp. Oberdorfer, E. 2001. Pflanzensoziologische Exkursionsflora für Deutschland und angrenzende Gebiete. -Ulmer, Stuttgart. 1051 pp.