_196_

Bull. Ir. biogeog. Soc. No. 26 em2)

AN ANNOTATED LIST OF TIIE SYRPIIIDAE (DIPTERA) OF POLLARDSTOWN FEN, CO. KII,DARE, IRELAND

Martin C. D. Speight ResearchBranch, Naional Park and WiAW, 7 Ety Place, Dublin 2, Irelnd.

Introduction PollardstownFen, in Co. Kildare, may be known as a site of scientific interest, but little has beenpublished on its fauna and flora. This is inevitably particularly true of its invertebrate fauna. The presenttext provides a list of the speciesof one family of , the Syrphidae,or , known from PollardstownFen, togetherwith somecomment on that list. The hoverfliescollected from PollardstownFen tota;l77 ,as comparedwith 94 for Co. Kildare in general(Speight, 2000). The Irish list for this family now comprises178 species. Very few of the syrphid recordsfrom the fen have beenpublished previously. One exceptionis the occurrence of amplus Curran, which was, at the time, the frst published record for this speciesin Europe (Speightand Vockeroth, 1988). Detailed information on the Irish syrphid speciesis availableelsewhere, and has beenbrought togetherby Speight(2000), so the commentspresented here are more concernedwith what the faunaof the Fen may tell us about the Fen, than with the speciesthemselves. Essentially,the text is focussedon the questionof whether the speciesobserved reflect the habitas observed.The potential contribution of PollardstownFen habitatsto maintenanceof syqphiddiversity, within the local landscape,is also considered.These processes are mediated by the data coded into the Syrph the Net (StN) databasespreadsheets. Examples of the use of the StN databasein interpretationof specieslists are provided by Speightand Castella(2001a).

Methods Collection of hoverflies from PollardstownFen has beencarried out by direct methodsi.e. use of a hand net to collect the adult insectsfrom flowers etc, and a sweepnet to collect them from low-growing vegetation.These processes have beenaugmented latterly by use of Malaise traps and, to a minor extent, emergencetraps. The recordsspan the period 1975-2001and -t97-

Bull. Ir. biogeog.Soc. No.26

determinationshave been carried out by the author. The product of this collecting activity is a list of the syrphid speciesobserved on the Fen, which can be usedfor comparisonwith the list of speciespredicted to occur on the Fen. The nomenclatureused follows Speightand Castella (2001b), updatedas necessary. Given that habitatassociations of the Irish speciesare known and the Pollardstownsyrphid fauna can be regardedas a subsetof the Irish syrphid fauna, the Irish speciesassociated with the habitatspresent on the Fen togethercomprise a list of the syrphidsthat would be predicted to occur on the Fen. This list of predictedspecies can then be comparedwith the list of species observed,following the basic proceduredescribed in Speightet al. (2000\. Using the methodologyemployed here, productionof a list of the speciespredicted to occur on a site involves the use of a list of the habitatcategories present on that site. And, in order to use the StN databasefor predicting the fauna, the habitatcategories referred !o require to be thoseemployed in the database.All habitatcategories referred to in the databaseare defined in is Macrohabitatsglossary (see Speight a al.,200la) and habitatsurvey ofthe Fen has been carried out (during 2000 anri 2001) using thosedefinitions.

Results The list of syrphid speciesobserved is given in Appendix 1. The syrphid macro-habitat categoriesobserved are as follows:- forest categories: - Fraxinus, overmature/mature/saplings,with springs,drainage ditches, flushesand streams. - conifer plantation, with drainagedirches. - isolatedmature/sapling Fraxinus and Salix. - Atlantic scrub: Ulex thickets. wetland categories: - transitionmire, with springsand flushes. - acid fen, with flushes, pools and drainageditches. - rich fen, with springs, flushes, streams,pools and drainageditches. - reed beds, with springs, flushes, streamsand drainageditches. - tall sedgebeds, with springs, flushes, streamsand drainageditches. _198_

Bull. Ir. biogeog.Soc. No.26

- tall herb communities. openground categories: - unimproved,humid/flooded oligotrophic Molinia grassland,with springs, streams,flushes, temporarypools and drainageditches, lightly-grazedby horsesand cattle. - improvedgrassland, grazed by cattle, with drainageditches. - intensivegrassland. other categories: - hedges. Using the habitat associationdata coded into tle Macrohabitatsfite of the StN databasethe numberof observedspecies associated with eachof thesehabitat categories can be totalled. The number of specieson the Irish and Co. Kildare lists associatedwith eachof thesehabitat categoriescan similarly be totalled. Taking the Co. Kildare list as the speciespool from which the PollardstownFen syrphid fauna is derived, the numberof Kildare speciesassociated with eachof the Pollardstownhabiats is taken as the number of speciespredicted to occur with each of thosehabitats on any fen in the Co. Kildare region. Similarly, taking the Irish syrphid list as the speciespool from which the PollardstownFen syrphid fauna is derived, the number of Irish speciesassociated with eachof the Pollardstownhabitas is taken as the number of species predictedto occur with each of thosehabitats on any fen nationally. Of the 7'l syrphid speciesrecorded from the Fen, the occurrenceof only three (i.e. lessthan 5% of the species)would not be predictedthere from their habitatassociations. The three exceptions are: bergenstanuniBecker, C. variabilis (Panzer) and Paragas haemorrhousMeigen. They are excludedfrom further analysisof the PollardstownFen fauna but are consideredbriefly in Appendix 2. Basingprediction on all of the habitaS representedthere, PollardstownFen would be expected to support 85 of the speciesrecorded from Co. Kildare and fully 133 of the speciesknown from lreland in general.This would meanthat 85% of the expectedKildare speciesare recordedfrom the Fen (including its periphery), as comparedwith 557oof the expectedIrish speciesin general.But some of the habitatcategories observed on the Fen and at its periphery cannotrealistically be regardedas either natural/semi-naturalor wetland-related.Conifer plantation, Ulex thickeu, hedges,improved grasslandand intensivegrassland would fall into -r99-

BULL.Ir. biogeog.Soc. No.26

this ca@goryand have, for thesereasons, been excludedfrom further analysis.Suffice it to say that noneof the syrphidspecies recorded from PollardstownFen would be predictedto disappearfrom the Fen if the coniferplantation habitat were lost from the Fen. Similarly, loss of improved grasslandand intensivegrassland from the Fen's periphery would be predictedto causeloss of only one species,Eupeodes corollae (Fabr.), from the existingspecies list. The contributionof thesehabitats to maintainingthe presentfauna of the site could thusbe regarded as negligible. Loss of hedgesand lJlex scrub from the Fen periphery might be expectedto causeloss of four species:Baccha elongata (Fabr.), Dasysyrphus albostriatus (Fall6n), MelangynaLasiophthalnn (Zetterstedt) and Meliscaevacinctell,a (Zetterstedt). The proportionof the predicredsyrphid fauna that hasbeen observed at PollardstownFen tbr habitasother thanconifer plantation, improved and intensivegrassland, ULex scrub and hedges, is shownin Table 1, at both levelsof prediction- regional(i.e. Co. Kildare)and national.The habitatsrepresented by the more-or-lessisolated trees of Fraxinus and Salix found scattered aroundthe Fen are also omitted,since they are predictedto havevery few associatedspecies, all of which would be predictedto occuron the Fen in associationwith other habitatscovered by Table I.

Table 1. Numberof syrphidspecies in the observedand predictedlists for PollardstownFen, associatedwith eachof varioushabitats occurrinq on the fen.

Habitat number of species 7o of predicted spp observed observed predicted predicted Kildare IRL Kildare IRL Fraxinus forest 25 28 35 89 1l transition mire l-) 14 15 93 87 acid fen 35 38 44 92 '19 rich fen 44 48 54 92 82 reed beds l9 22 26 86 '73 tall sedgebeds JJ 36 40 92 83 tall herb communities r6 19 2I 84 76 unimproved Mo linia gr assland 42 48 63 88 <1 -200-

Bull. Ir. biogeog. Soc. No.26

In Table 2, microhabitatdata are summarised,for the observedand predictedsyrphid faunasof PollardstownFen, using the information coded into the Microsite Featuresfile of the StN database(see Speight et al.,200lb). The speciescovered are thoseassociated with the habitats coveredby Table I, but taken all together,rather than habitatby habitat. The categories recognisedare not true micro-habitats,but physical featuresof sites (micrositefeatures) that are usedas proxies for micro-habitats.The array of categoriesshown is a selectionof those coveredby the StN database,in which speciesassociations for more than 80 micro-sitefeature categoriesare coded.

Table 2. Number of syrphid speciesin the observedand predictedlists for Pollardstownfen, associatedwith eachof various micro-habitatsimicrositefeatures.

Microsite feature Nurnber of species olp as io observed (o) predlcted (p) terrestrial timber 2 3 67 overmature uees 4 JI tree foliage t4 23 6r understorey trees 5 8 63 shrubs/bushes/saplings l0 r6 63 lianas 4 5 80 tall herbJayer plants (on) 20 )R 11 low-growing herb-layer plants (on) 11 1< t5 within herb-layer plant tissues 8 t6 50 ground-surfacedebris I7 18 94 nests of social 2 50 grass-rootzone 10 13 77 aqualrc on emergent plants l3 20 65 on submerged plants 4 t2 33 submerged sediment/debris )1 JJ 67 water-soddenplant debris 22 27 82 -20r-

Bull. Ir. biogeog. Soc. No. 26

Discussion It was earlier suggestedthat conifer plantation, improvedand intensivegrassland, Ulex scrub and hedges,although present,together play but a minor role in maintenanceof the existing PollardstownFen syrphidfauna. In addition,it can be observedthat improvedand intensive grassland,with associatedhedges, is a predominantfeature of the Kildare landscapein the surroundto PollardstownFen, so the presenceof thesehabitats on the periphery of the Fen itself cannotbe said to add to the uniquenessof the contributionof the Fen to maintenanceof the Co. Kildare fauna. Excluding thesehabitats from considerationresults in reductionof the effective Pollardstownsyrphid fauna to 7l species,from the gross total of 77 specieson the observedlist. And this restrictedlist of habitashas a predictedfauna of 106 lrish species,78 of which are known from Co. Kildare. The effective total of observedspecies is further reducedto 68 speciesby exclusionof the three speciesobserved from, but not predictedto occuron, the Fen. With conifer plantation, improved/intensivegrassland, UIex scruband hedgesexcluded from consideration,64% of the lrish syrphid speciespredicted to occur in associationwith the array of habitatsoccurring on PollardstownFen have beenobserved there. This is not a particularly high proportion of the expectedIrish species,even though 86% of the expectedCo. Kildare speciesare represented.Essentially, this would suggestthat, in termsof biodiversity maintenance,Pollardstown Fen hasconsiderable regional significance, but, overall,would be of lessersignificance nationally. It follows that a closer examinationof the fauna is warranted,to seewhether particular habias of thosepresent on the fen might seemto be "underperforming", in comparisonto others. Table I shows that, when consideredseparately, the various habitas presentat Pollardstown apparentlysupport a higher proportion of their expectedsyrphid speciesrhan they do when consideredtogether, suggestingthere is more sharingof habitatsamong the speciespredicted and observedthan among the speciespredicted but absent.This is indeedthe case:on average, the speciesobserved are eachassociated with 3.4 of the habitatcategories observed there, while the speciespredicted but absentare associatedwith 2.3 of the habitatcategories observed. However,Table I alsoshows that more thanl1% of the Irish speciesassociated with eachof the habitatspresent on the Fen have beenfound there, exceptin the caseof the speciesof -202-

Bull. Ir. biogeog. Soc. No. 26

Fraxinus forest, reed bedsand unimproved,oligotrophic grassland.This makesPollardstown Fen an exceptionalsite for the Irish syrphidsassociated with most of the habitas shown in Table 1. Once again grouping togetherthe Fen habitatscovered by Table 1, as a single entity, allows comparisonbetween the representationof syrphidsassociated with different fen microhabiats in the observedand predicted lists. This comparisonshows that syrphids with four different types of larval micro-habitatare paflicularly under-representedon the observedlist, the micro- habitatsinvolved being as follows: - overmaturetrees/rotten wood; - internal tissuesof herb-layerplants; - coloniesof social insects; - submerged,aquatic plants. The lack of speciesassociated with overmaturetrees and rotten wood would seema reasonablereflection of the condition of the ash woodlandand other treesaround the edgeof the Fen and probably also indicatesit is of recentorigin. Syrphids that feed internally in the tissuesof flowering plants are generally intolerantof flooding, since they nearly all passthe winter free in the soil in an immobile, resting phase(the puparium) of the life history that makesthem susceptibleto drowning. Thosepredicted for the Fen would mostly be expectedto occur in the peripheral grasslandsand may do so, further from the Fen and outsidethe influenceof seasonalhigh-water levels. Similarly, nestsof ants and other social insectswould be expectedmore in the peripheralgrasslands than the Fen itself and the under-representation of associatedsyrphids on the Fen list could be interpretedas a consequenceof the susceptibility to flooding of the nestsof their hosts. The recently-describedsyrphid Microdon myrmicae Schdnroggeet al. is an exceptionto this generalisation,living as it does in nestsof the ant Myrmica scahircdis Nylander in large tussockswithin the periphery of Fen. lt is one of the speciesobserved from Pollardstown.The apparentunder-representation of specieswhose larvae are associatedwith the submergedparts of aquaticplants (this microhabitatcategory excludes rotting vegetationon the bottom of a water body) might seemrather odd, until it is realisedthat areasof (permanent)standing water have'beenfew and far betweenon the Fen. Similarly, there is very little Typha on the Fen, a plant with which many of the syrphids with aquatic larvae are _203_

Bull. Ir. biogeog.Soc. No.26 frequentlyassociated. The situationof tle speciesassociated with that larval microhabitat contrastssharply with the specieswhose larvae are associatedwith water-soddenvegetation and peat (i.e. with temporarypools or situationswhere the ground-waterlevel is more or lessco- incident with the ground surface).The latter group of speciesis very well representedon the Fen, highlighting the significanceof this micro-habitatthere. Someof thesespecies can use cow dung as an alternativelarval micro-habitat,at leaston wet sites, so the contingentof speciesassociated with cow dung is also well-representedin the Pollardstownlist. Turning to the issueof the conservationstatus and managementof the PollardstownFen syrphid fauna, the great majoriry of the observed species is clearly associatedwith categories of habitatdependent for their survival upon the wetland statusof the site. Changein site hydrology, particularly drainageof the fen or reductionin the water supply to it, would thus be expectedto adverselyimpact upon is existing syrphid fauna. While some of the nationally scarcespecies that would be predictedto occur at Pollardstownseemingly do not do so, an examplebeing cryptarum (Fabr.), which may now have beenentirely lost from the Irish fauna, others do. The latter group comprises tarsata (Meigen), fulicrodon nryrmicae, Platycheirus amplus Curran and P. immarginntus (Tntterstrlt). Onhonevra geniculata (Meigen) is a borderline case.Some of the PollardstownFen syrphidsare sc:rcer elsewhere than Ireland in the Atlantic Region of the EU, examples being Melnnog(Eter aerosa (l,oew) and Parhelophilusconsimilis (Malm). But there are no syrphidswhose presence in Ireland is apparentlyconfined to PollardstownFen. PollardstownFen is probably best seenas one of the very few Irish siteswhere all of thesespecies have been found together.In this regard its "biodiversity maintenance"function has to be regardedas exceptional,supporting as it does75% and more of the Irish speciesassociated with various categoriesof wetland habitat. PollardstownFen is today an island of wetland habitatsisolated in a seaof farmland. Problemsof speciesdispersal through today's farmland landscapein Ireland have beenalluded to by Good (1998) and Speightand Good (2001), and while the Fen may be larger than the averagefarm, it is debateablefor how long it might maintain its presentflora and fauna without interchangeof wetland specieswith other sites. That it is an ancientsystem, with some of its presenthabitats continuously in situ for a time measurablein hundredsof years, rather than decades,is suggestedby someof the speciespresent there today, one examplebeing the snail -204-

Bull. Ir. biogeog. Soc.No.26

Vertigogeyei Lindholm. In Ireland this mollusc exhibis a relict distribution pattern (E. Moorkens,pers. comm.). The samecould be said for the syrphidsM nryrmicaeand P. amplus (seeSpeight, 2000 and Speight, 2002, this volume). Such data do not prove that speciescan survive for prolongedperiods of time within PollardstownFen, without re-inforcementof their populationsfrom elsewhere,but they do suggestit. However, the considerabledistances such specieswould now have to travel in order to re-establishthemselves on the Fen, were they once lost there, suggeststhat, unaided,their re-establishmentwould be virtually impossible. And there are no current trends in landscapechange that would suggestlandscape permeabiliry will increasefor such speciesin the forseeablefuture. Sadly, both M. nryrmicaeand P. amplus may have been lost from the Fen during the last few years. Attempts to relocatethem have failed and the parts of the Fen where they were found have changedin character,in one case probably due to cessationof grazing and in the other probably due to eutrophication. It was statedearlier that habitatsperipheral to the Fen (hedges,Uler scrub, improvedand intensivegrassland) support very few speciesadditional to thoseassociated with wetland habitatson the Fen. But there are other faunistic relationshipspotential to the interaction betweenthe Fen and its surround than that describedby consideringsimply what speciesmay be addedto the PollardstownFen list from its surround. In particular, speciesthat can be sharedby the Fen and surroundinghabitats would be expectedto survive on the Fen when affectedadversely in is hinterland. Similarly, were any of thesespecies to be lost from the Fen it would be expectedthat they could re-establishthemselves there from adjacentpopulations when conditionswere once more favourable. Of the fen-habitat(i.e. the habitatscovered in Table l) syrphidsobserved on the Fen, 23 specieswould be expectedto occur in associationwith improved/intensivegrassland and or hedgesand Ulex scrub in its sunound. To thosespecies, then, the surroundinglandscape may remain permeable.But for the rest of the speciesassociated with fen habitatson Pollardstown Fen - by far the majority of the syrphidsrecorded there - the conversewould seemmore true. For the 45 speciesin the latter group loss of the populationinhabiting the Fen could well mean more-or-lesspermanent loss of the speciesfrom the area, due to lack of other local populations from which to repopulatethe Fen. This group also representsalmost half of the syrphid species on the Co. Kildare list (which totals 94 species).While it would be unjustifiedto extrapolate _205_

Bull. Ir. biogeog. Soc. No.26

this statisticto other taxonomicgroups, it would none-the-lessseem possible that, at leastfor Syrphidae,Pollardstown Fen is responsiblefor maintainingthe presenceof nearly 50% of the diversity of the Kildare fauna within is local landscape.By contrast,the speciesfor which it sharesmaintenance with the surroundingfarmland amount to approximately25% of the Kildare syrphid fauna, while the farmland by itself is able to supportonly an additional 5%.

Acknowledgements I am most grateful to Jervis Good for constructivecomment on an earlier draft of this text and for the opportunity to include recordsof syrphid speciescollected by himself from PollardstownFen. Evelyn Moorkens kindly provided information on Vertigogeyei.

References Good, J. A. (1998) The potential role of ecologicalcorridors for habitatconservation in Ireland: a review. Insh Wildlift Manuals. 2.72 pp. Duchas,Dublin. Speight,M. C. D. (2000)Irish Syrphidae(Diptera), Part 1: speciesaccounts and distribution maps.pp. l-215. In Speight,M. C. D., Castella,E., Obrdlik, P. and Ball, S. (eds)Syfplr the Net, the daabase of EuropeanSyryhidae. lt. Syrph the Net publications,Dublin. Speight, M. C. D. (2001) Speciesaccounts of EuropeanSyrphidae (Diptera), 2001. pp. l-281. 1n Speight,M. C. D., Castella,E., Obrdlik, P. and Ball, S. (eds)Syrphthe Net, the databaseof European Syryhidae.27. Syrph the Net publications,Dublin. Speight, M. C. D. (2002) Two controversialadditions to the lrish insectlist: Microdon

myrmicaeSch0nrogge et al. and, festiva Meigen (Diptera: Syrphidae). Bull. Ir. biogeog.Soc.26: 143-153. Speight, M. C. D. and Castella,E. (2001a) An approachto interpretationof lists of insects using digitised biological information about the species.J. Conservation5: 131-139. Speight, M. C. D. and Castella,E. (2001b) Rangeand statusdata for EuropeanSyrphidae (Diptera),2001.pp. l-215. In Speight,M. C. D., Castella,E., Obrdlik,P. andBall,S. (eds) Syrph the Net, the daabase of EuropeanSyrphidae.30. Syrph the Net publications, Dublin. Speight,M. C. D. and Good, J. A. (2001)Farms as biogeographicalunits: 4, conceptand _206-

Bull. Ir. biogeog. Soc. No. 26

realiry. Bull. Ir. biogeog. Soc.25:2y2-308. Speight, M. C. D., Castella,E. and Obrdlik, P. (2000) Use of the Syrph the Net database 2000. pp. l-99. In Speight,M. C. D., Castella,E., Obrdlik, P. and Ball, S. (eds)Syryft the Net, the duabase ofEuropean Syrphidae.25. Syrph the Net publications,Dublin. Speight, M. C. D., Castella,E. and Obrdlik, P. (2001a) Macrohabitatpreferencesof European Syrphidae(Diptera),2001.pp. l-565. In Speight,M. C. D., Castella,8., Obrdlik, P. and Ball, S. (ed$ Syrp, the Na, the daabase of EuropeanSyrphidae.2t. Syrph the Net publications,Dublin. Speight, M. C. D., Castella,E. and Obrdlik, P. (2001b) Microsite featuresused by European Syrphidae(Diptera), 2001. pp. l-240. In Speight,M. C. D., Castella,8., Obrdlik, P. and Ball, S. (ed$ Sylr the Nd, the databaseof EuropeanSyrphidae.29. Syrph the Net publications,Dublin. Speight, M. C. D. and Vockeroth, J. R. (1988) Platycheirusanplus: an insectnew to Ireland not previously recordedfrom Europe (Dipt.: Syrphidae).Ir. Nat. J. t2:518-521. _2W_

Bull. Ir. biogeog. Soc.No. 26

APPENDIX 1. List of the Syrphidae(Diptera) recordedfrom PollardstownFen. Nomenclature usedfollows Speight(2001) updatedas necessary. * : threatenedspecies (lRL)

Anasinryialineata (Fabricius, 1787) Baccha elongaa (Fabricius, 1775) Cheilosia bergenstammi Becker, I 894 CheilosiaiLlustrata (Harris, 17E0) CheiLosiaimpressa Loew, 1840 Cheilosianebulosa (Verrall, 1871) Cheilosiapagana (Meigen, 1822) Cheilosiavariabilis (Panzer, 1798) Cheilosiavernalis (Fall6n, 1817) Chrysogastercoemite iorum (Linnaeus, 1758) Chrysogastersolstitialis (Fall6n, 1817) Chrysotoxumbicinctum (Linnaeus, 1758)

Chrysotoxum fasciuum (Miiller, l7 64) Dasysyrphusalbostiatus (Fall6n, 18 17) Episyrphusbalteatus (DeGeer, 1776) Eristalinus sepuhhralis (Linnaeus, 1758) Eristalis abusiva Collin, 1931 (Linnaeus, 1758) Eistalis interrupta (Poda, 1761) Eistalis inticaria (Linnaeus, 1758) Eistalis lineua (Harris, 1776) Eistalis pertinar (Scopoli, 1763) Eistalis tenox (Linnaeus, 1758) Eumerusstigatus (Fall6n, 1817) Eupeodesbucculatus (Rondani, 1857) Eupeodescorollae (Fabricius, 1794) -208-

Bull. Ir. biogeog. Soc. No.26

MPENDIK 1 (continued)

Eupeodeslatifasciaus (Macquart, 1829) Eupeodesluniger (Meigen, 1822) F erdinandea cuprea (Scopoli, l7 63) Helnphilus hybridus loew, 1846 Helaphiluspendulzs (Linnaeus, 1758) Lejogastermaallina (Fabricius, 1781) Lej ogast e r tarsata (Meigen, 1822)* Leucozonalaternaia (Mtiller, 1776) Leucozonalucorum (Linnaeus, 1758) Melangyna lnsiophthalna (Zetterstedt, I 843) Melnngynn umbeLlatarum(Fabricius, 1794) Melnnogaster aerosa (l,aew, 1843) M elnnogaster hirtelln (l-oew, 1843) Melnnostomamellinum (Linnaeus, 1758) M elanostomascalare (Fabricius, 1794) Meliscaeva cinctella (Zetterstedt, 1843) Microdon myrmicaeSch6nrogge et al,,2N2* Neoasciagmiculata (Meigen, 1822) Neoasciapodagn'ca (Fabricius, 1175) Neoascia tenur (Harris, 1780) Onhonevra geniculata (Meigen, 1830) Paragw haemonhousMeigen, 1822 Parhelophilusconsimilis (Malm, 1863) Pipiza rcctiluca (L., L758) Plntycheirusahimanus (Fabricius, 1781) Plntycheirus anplus Cur'ran, 1927* P latycheirus angustatus (Zetterstedt, 1843) Platycheirusclypeatus (Meigen, 1822) -209- Bull. Ir. biogeog. Soc. No.26

APPEI\IDD( I (continued)

Playcheirusfulviventis (Macquart, 1829) Platycheirusgranditarsus (Forster, 1771) Platycheirus immargirwus (Zetterstedt, 1849)* Platycheirus nnnicatus (Meigen, I 822) Plarycheirusoccultus Goeldlin, Maibach and Speight, 1990 Plarycheirusrosarum (Fabricius, 1787) Platycheirus scambus (Staeger, 1843) Plnrycheirusscutatus (Meigen, 1822) Mingia campestrisMeigen, 1822 Riponnewia splendens(Meigen, I 822) Scaevapyrasti (Linnaeus, 1758) Seicorryia silentis (Harris, 1776) Sphaerophoia iwenupta (Fabricius, 1805) Sphaerophoia philnwha (Meigen, 1822) Spheginaclunipes (Fall6n, 1816) Spheginaelegans Schummel, 1843 Syina pipiens (Linnaeus, 1758) Syrphusibesii (Linnaeus, 1758) Syrphusvitipennis Meigen, 1822

Tichopsontyia flavirarsls (Meigen, 1822) Tropidia scita (Harris, 1780) Volucellabombylnns (Linnaeus, 1758) Xylota segnis(Linnaeus, 1758) _2lo_

Bull. Ir. biogeog. Soc. No.26

APPEI\DIX 2. The syrphids recordedfrom PollardstownFen but whosepresence would not be predictedon the basis of known habitat associations.

Cheilosiabergenstammi C. bergenstammjis associatedwith ragwort (Seneciojacobaea), its laryae feeding internally on the tissuesof this plant. This hoverfly wits common and widely distributed in Ireland until recently, but has becomescarcer since ragwort control measureshave been more rigorously enforced(the plant is toxic to both cows and horses).There is less needto control ragwort on sheep-grazedpasture. The predictive procedurewould not predict C. bergenstammifrom cattle- grazed, improved grassland,but would predict the speciesfor sheep-grazedgrassland. The recordsof C. bergenstammifrom the Fen date from when sheepwere present.It is not known whetherthis syrphidstill occursthere, but its presencewould be expectedin the vicinity, given the propensityof ragwort to occur along the hedgemargins of minor roads.

Cheilosia vaiabilis

C. vartabilis also has larvae thar feed internally in plant tissues,in this instancein scrophulariaspp., which is not found on the Fen and would be expectedonly at its margins. C. variabilis was collected from the edge of the fen, but the record remainsanomalous.

Paragushaemorrhous

P. haemorrhou,rwas recordedfrom a part of the periphery of the fen that has since changed in character.It was then (some 15 yearsago) grazedand burnel Molinia grassland,whereas today this area is fencedfrom grazing and no longer burned. Since thesechanges took place there have been no recordsof P. haemorrhousfrom the Fen.