Information and Advisory Note Number 9 Back to menu

The distribution, ecology and conservation of the marsh fritillary butterfly (Eurodryas aurinia)

1. Introduction, Distribution and Status

1.1 The marsh fritillary butterfly {Eurodryas aurinia) is an attractive species which can be distinguished from Britain's other fritillaries by the row of black dots around the bottom edge on both sides of the hindwings with none on either side of the forewing. It is distributed throughout western Europe, Russia and across temperate Asia to Korea. The butterfly is declining in every European country and is now extinct in the Netherlands and Belgium. Britain and Ireland are believed to be major European strongholds for the species.

Marsh fritillary butterfly

1.2 The UK subspecies E aurinia hibernica is considered to be distinct by some authorities but biochemical and genetic differences between the UK and other European forms have not been investigated yet. A number of different forms have also been recognised in Britain and local races are common.

1.3 In Britain its range has declined by over 60% and it has recently died out over much of eastern England and Scotland. Although it is still quite widespread in parts of western England, Wales and western Scotland, colonies are estimated to be disappearing at a rate of well over 10% per decade.

1.4 A1990 study of marsh fritillary colonies across the UK identified a total of 228 definite colonies plus an additional 106 'possible* colonies. However, most of these colonies were small and at a high risk of extinction. Thirty-five of the definite colonies were in Scotland.

1.5 The exact status of the marsh fritillary in Scotland is not as clear as elsewhere in the UK due to under-recording. However, the best areas are in western Scotland such as Taynish in western Argyll and the Rhinns of Islay.

2. Ecology

2.1 In the UK, colonies of the marsh fritillary occur in two main biotopes: damp, neutral or acidophilous grasslands (usually open grassland but occasionally in woodland clearings); and dry, calcicolous grasslands. In both biotopes its main larval host plant is

devil's bit scabious Succisa pratensis, although other hosts such as field scabious Knautia arvensis, and small scabious Scabiosa columbaria are occasionally used in calcicolous grasslands. In Scotland all of the known colonies are on damp, neutral or acidophilus grassland (a recent study of larval sites in west Scotland recorded them occurring mostly on national vegetation communities M24 (Cirsio-Molinietum caeruleae fen meadow), M25 (Molinia caerulea-Potentilla erecta mire) and M23 (Juncus effusus/acutifloris-Galium palustre rush-pasture)). Breeding areas are generally very open and unshaded, though many are sheltered either by scattered scrub or by adjacent woodland.

2.2 Eggs are laid around late June on devil's bit scabious plants, up to 300 eggs in the first batch, with smaller numbers in second and third batches. After about three weeks they hatch and the caterpillars produce a communal silk web on the food plant. They overwinter during their fourth instar in a small hibernaculum web which insulates the caterpillars. On swampy sites the whole nest can remain submerged for several weeks without the caterpillars coming to any harm.

2.3 in early spring the caterpillars emerge and bask communally on top of the vegetation, forming a conspicuous dark mass that helps to raise their body temperature well above the ambient temperature. They begin to disperse in their fifth instar and are solitary by
their sixth and final instar. They pupate close to the ground beneath dead leaves or on plant stems.

2.4 The adult butterflies usually start to emerge in late May or early June, although this can vary between localities and years. The flight period can last until early July in the south and mid-July further north. Usually the marsh fritillary is a highly sedentary species but some dispersal from colonies does occur and individual adults are regularly seen in non-breeding habitats in the vicinity of existing colonies. Populations of marsh fritillaries can fluctuate tremendously in size from year to year with larvae occasionally reaching great densities. For example, around 1991 the butterfly under went a periodic expansion and was comparatively mobile and widespread in western Scotland. It is possibly more sedentary in lean years, contracting to core sites during sequences of poor seasons.

2.5 The population fluctuations appear to be dependent upon weather, food supply and the proportion of caterpillars killed by parasitic braconid wasps of the genus Cotesia. The two parasitoids in Britain are Cotesia melitaearum (Wilkinson) and Cotesia bignellii (Marshall).

2.6 The large fluctuations seen in marsh fritillary populations can cause problems if the habitats are small or of low quality and the butterfly population has become extremely small (or possibly so large that the food supply is wholly consumed, leaving the larvae unable to achieve hibernation). Under such circumstances an isolated colony can die out completely during troughs in the cycle. The periodic extinctions often give the appearance that colonies are shifting around fields or a group of sites. In fact the phenomenon is probably caused by local extinctions which are balanced by periodic colonisations.

2.7 Although the marsh fritillary is regarded as colonial and sedentary, it does display some emigration and movement and it may be that this butterfly, with its attendant parasitoids, is adapted to a 'metapopulation' structure and that this interacts with the levels of parasitism (a metapopulation being a collection of local populations, connected by occasional dispersal, in which there are local extinctions and colonisations).

2.8 Larval web counts around mid-September are considered to be most appropriate monitoring methodology although spring counts in February-April are also possible. These can be supplemented by adult surveys. During 1995, monitoring sites were set up at five sites in west Scotland through the Scottish Diurnal Lepidoptera Project funded by SNH.

3. Protective Status

3.1 The butterfly is listed in Schedule 5 of the Wildlife & Countryside Act 1981 for sale only, which means it is illegal to trade wild marsh fritillary butterflies in Britain without a licence (if taken to court, a seller of marsh fritillaries would have to prove they were captive-bred). It has been put forward for full protection on Schedule 5 in the 1995 Quinquennial Review by the JNCC. In Northern Ireland it is protected under the 1985 Wildlife Order (which is comparable to Schedule 5 of the 1981 Act).

3.2 It is listed on Annex II of the EC Habitats Directive as a species whose conservation requires the designation of Special Areas of Conservation (SACs). SACs have been proposed for six sites in the UK for the marsh fritillary, including Taynish Woods in Scotland.

3.3 Of the known colonies in Britain, 44% are encompassed by statutory Site of Special Scientific Interest (SSSI) notification. Of the counties with the highest number of colonies, 62% of colonies in Dorset are on SSSIs, 54% in Strathclyde, 52% in Dyfed, 48% in Cornwall, 25% in Devon and 16% in Glamorgan. The marsh fritillary is also known to occur in 11 National Nature Reserves (NNRs).

4. Threats and Conservation Aims

4.1 The major threats to the marsh fritillary are the wholesale destruction of its habitat caused by development and agricultural improvement; inappropriate management of sites including abandonment of grazing; and the increasing fragmentation and isolation of its habitats. It is a species that depends especially on extensive farming systems, and appropriate habitat management in its unimproved grassland habitat.

4.2 A particular problem is that many rough grassland areas are unsuitable for modem commercial breeds of cattle. Also, the grazing regimes required for the butterfly may, at present, be impractical for the farmer. The use of rare and traditional breeds may provide
an efficient and effective means of management but this needs further investigation.

4.3 In Britain, most colonies (47% in 1983) occupy very small patches of habitat, typically less than 2ha; only 15% of occupied patches are greater than 10ha. Together with the huge yearly fluctuations in abundance, this makes the butterfly very prone to natural extinction. The natural pattern of existence is in metapopulations, the survival of which depends upon having a network of patches within which there is periodic extinction and re-colonisation.

4.4 As a result, only a proportion of the suitable patches within the metapopulation area will be occupied at any one time; there will be empty patches which will be used as the individual populations are formed and become extinct. This natural pattern is now broken throughout most of its UK range and the butterfly is often restricted to single habitat patches, or ineffectual groupings of small patches. Many of the present colonies are isolated, not only from other nearby colonies but also from potentially habitable sites.

4.5 The population dynamics of this species means that the management of small areas of land with existing populations of the butterfly by themselves will not work in the long term. The conservation aim should be to maintain large areas of land with networks of colonies and potentially suitable, but unoccupied habitat between. An effective network of habitats must have suitable patches of habitat within dispersal distances of 5-1 0km of each other so that the process of successful dispersal at times of high population levels compensates for the periodic local extinctions to which local populations are prone. Wider countryside measures are therefore required in addition to the current network of SSSIs.

5. Basic Site Management Requirements for the Marsh Fritillary

5.1 Type of Stock: Traditional breeds of cattle and horses are best at coping with coarse vegetation and are less selective. Sheep are generally unsuitable although colonies persist on some lightly sheep grazed areas in Scotland. Deer grazing occurs at many of the main west Scotland sites.

5.2 Intensity of Grazing: In general a long period of light grazing is better than short periods of heavy grazing. Heavy grazing is unsuitable. Maximum grazing level tolerated is about 0.4-0.7 livestock units/ha. Some sites may be maintained in suitable condition at lower grazing levels. The general aim should be to produce a sward with a reasonable proportion within the 10-15cm height range on damp grass (a recent study in Argyllshire recorded a summer sward height of 11-14cm as optimal) and 5-15cm on calcicolous grassland. Areas with a high density of devil's bit scabious may be the key requirement within the grassland habitat.

5.3 Timing of Grazing: Effects poorly known. Devil's bit scabious regenerates in the spring so it is important to have an open sward at that time. Therefore some winter/spring grazing, where possible, is likely to be beneficial. Some light summer grazing is probably needed to keep coarse vegetation in check, and may be the only suitable time on very wet sites. At Taynish there is light grazing by cattle during autumn and early winter but a longer period of light grazing commencing in June is being considered.

5.4 Scrub Control: Periodic scrub control may be needed on some sites to prevent scrub invasion under low grazing levels, although scattered scrub provides shelter.

5.5 Cutting: Generally unsuitable, especially regular cutting or haymaking. Rotational cutting may be suitable but experiments are needed on frequency and timing.

5.6 Burning: Periodic burning may be useful in maintaining suitable habitat on some sites (though most larvae will probably die in the actual area burnt). Burning should only be carried out where it is traditional (as burning can be detrimental to many other invertebrates and plants), and in rotation so that only part of each field is burnt each year. Care should be taken in rank (neglected) sites, and bracken areas avoided. The main breeding areas on a site should be identified and avoided.

5.7 Releases: Clandestine releases and support breeding should be avoided and discouraged. They may only serve to mask the decline of viable colonies. Translocations may have a role where suitable habitat has been restored on isolated former sites. However, such releases should only be part of regional recovery strategies, they should follow the JCCBI guidelines on translocations
and they must be fully documented (Oates and Warren 1990).

Author

Dr Martin Gaywood (with acknowledgement to the Butterfly Conservation action plan, Barnett and Warren (1995), on which this advice note is based)

Contacts for advice and information

Dr Martin Gaywood, Habitats Directive Species
Co-ordinator
Dr David Phillips, Invertebrate Ecologist
Jane Mackintosh, Grassland Ecologist
John Ralston, Licensing Officer

Research and Advisory Services Directorate
Scottish Natural Heritage
2 Anderson Place
Edinburgh EH6 5NP
Tel. 0131-447 4784

John Halliday, Site Management and Information Officer for Taynish

Scottish Natural Heritage
Kilmory Estate
Lochgilphead
Argyll PA31 9RR
Tel-01546 603611

Butterfly Conservation
PO Box 444
Wareham
Dorset BH20 5YA
Tel 01929 400209

Back to menu