Research on the marsh fritillary suggests that the species requires a minimum area of suitable habitat on a metapopulation site to be viable in the long term.

On rhos pasture habitats, marsh fritillary larvae will only feed on devil's-bit scabious; the abundance of this food-plant is directly related to the survival of larvae. Sward height also affects survival. Very short swards are unsuitable as they provide no refuge for larvae in cold wet weather. Tall rank swards are also unsuitable because they do not support vigorous populations of the larval food-plant, devil's-bit scabious, and they do not provide basking opportunities for larvae. All larvae must bask in the sun to raise body temperature to allow feeding activity.

Sward structure and status of the larval food-plant in defining types of habitat quality:

Definition of Good Condition marsh fritillary habitat: Grassland with Molinia abundant where, for at least 80% of sampling points, the vegetation height is within the range of 10-20cm. cm (when measured using a Borman's disc) and Succisa pratensis is present within a 1 m radius. Scrub (>0.5 metres tall) covers no more than 10% of area. 

Definition of Suitable marshy grassland: Stands of grassland where Succisa pratensis is present at lower frequencies but still widely distributed (>5% of sampling points) throughout the habitat patch and in which scrub (>0.5 metres tall) covers no more than 25% of area. Alternatively, Succisa may be present at high density in close-cropped swards. Available habitat' is defined as the total of good condition and suitable habitat.

Extent and distribution of marsh fritillary habitat

Upper limit: not required

Lower limit: Within the SAC boundary:

26 ha of available habitat (the likely distribution is c 9 ha in the core compartments, c 12 ha in the low-density compartments and c 5 ha in the compartments for reversion)', including 10 ha of good condition habitat.

Lower limit: 50% of the marsh fritillary habitat within compartments 5 and 6 is

described as 'good condition habitat with dense Succisa'

And

50% of the marsh fritillary habitat within compartment 14 is described as good condition 

And

Scrub covers less than 10% of compartment 15

This is the only attribute that may be used as a performance indicator tor this species. Other attributes of quality, such as productivity and sex ratio, are difficult and time consuming to measure. In relation to determination of population size, it is most appropriate to consider the larval stage. Eggs are difficult to find and. for this reason alone, abundance cannot be systematically evaluated. Data collected on adult butterflies cannot always be relied on to indicate population size because of observer difficulties related to the mobility of the butterflies. The abundance of larval webs will therefore be considered as the sole attribute for this species.

Larvae feed communally in webs, which are easily observed and reasonably static. Historically, a large amount of sample data on webs ha^ been collected from a series of 53 transects in the current 'core' area for the butterfly on site - compartments 8, 5, 6, 7 and 10. Data has been analysed annually and an 'index' calculated which represents the total number of webs on all transects observed in any one year. In order to allow comparison with historical data, the performance indicators will include reference to the annual index collected on the 53 transects.

In addition, the performance indicators must take account of web abundance elsewhere on site, namely compartments 2, 3,4, 12 and 14, which have historically been occupied at low density, and compartments 16 and 18, which are currently semi-improved grassland but have potential for reversion.

Because of the host/parasite relationship with the wasp, the marsh fritillary population fluctuates significantly over time. Historical data from the site indicate that there is approximately a 10-year cycle between population peaks. If the butterfly population were not being affected adversely by other factors, it should be relatively high for half of that period.

The following site-specific performance indicator has been devised:

Attribute:       Larval webs

Upper limit:    not required

Lower limit:    over any 10-year period:

The web index count will be greater than 50 for at least 5 of the years, 

and annually:

• Larval webs continue to be present in all management compartments

• A minimum total of 50 webs are present in compartments 12 and 14

• A minimum total of 50 webs are present in compartments 16, 18, 22 and 24

Status of marsh fritillary: unfavourable recovering (Wilkinson 2005)

The 2005 revision of Favourable Conservation Status meant that, for the first time since 1992, the status of the feature was deemed to be unfavourable. This is because there is insufficient 'good' and 'suitable' habitat on the NNR at present. However, there is potential to increase the area of both categories of habitat through reversion of the improved pastures on the NNR and achieve Favourable Conservation Status in the future.