Ecological Entomology (1987) 12, 139-148
Geographical variability in, and temperature effects on,
the phenology of Maniola jurtina and Pyronia tithonus
(Lepidoptera, Satyrinae) in England and Wales
PAUL M. BRAKEFIELD Department of Zoology. University College, Cardiff
ABSTRACT. 1. Geographical variability in. and temperature effects on,
the mean date of adult flight period and the SD about this date are analysed
for two univoltine, grassland butterflies in England and Wales from 1976 to
1985. Data were collected on the Butterfly Monitoring Scheme for
Maniola jurtina (L.) at twenty-nine sites and Pyronia tithonus ( L . ) at
twenty sites.
2. Substantial variability for mean date and SD occurs between years
and between sites. Changes in mean date between years tend to occur
consistently at different sites. The species show some parallel in variation
between sites, especially for mean date.
3. June maximum temperature accounts for 95% and 75% of the
varia-tion in mean date between years in M. jurtina and P.tithonus. respectively
(r=-0.97 and -0.87). Similar relationships occur for temperatures
cumul-ated over the period of post-winter development from March to July or
August.
4. Greater geographical variability in phenology, and a generally less
synchronized flight period in M.jurtina may be associated with broader
habitat preferences than in P. tithonus.
5. The mean date of adult flight period remains at roughly the same date
at more northerly latitudes. In M.jurtina the flight period becomes more
synchronized, begins later and ends earlier in the north. P. tithonus shows
little or no indication of such a response to latitude. This is discussed with
regard to changes in season length and factors limiting the species' range.
Key words. Maniola jurtina, Pyronia tithonus, butterfly, phenology,
geographical variability, temperature, development, emergence,
distribu-tion, latitude.
Introduction tribution of the British butterfly fauna (e.g. Dennis. 1977; Heath ft til.. 1984). Dennis There is renewed interest in factors which examined isotherms coinciding with the influence the historical and contemporary dis- northern limits of many species .ind showed that Correspondence: Dr Paul M Brakefield. Depart- a n u m h e r of ™ogeograph,cal and climatic
fac-ment of Zoology. University College. P.O. Box 78. tors- including July temperatures, act .is good
140 Paul M. Brakefield
species diversity. More recent studies have con-centrated on analysing patterns in 10 km square distribution maps given by Heath et al. (1984). Nearly 80% of the variation in species diversity is accounted for by summer temperature and sunshine together (Turner, 1986) or by latitude (Barbour, 1986). Turner et al. (1986) showed that climate was able to explain some part of the variation in diversity which is independent of latitude. Barbour ( 1986) also found that patterns of extinction since pre-1970 records were correl-ated with richness of the epiphytic flora and air pollution levels. Dennis & Williams (1986) emphasized the many covarying factors and the necessity of taking into account the great variability in the biology of the species of butterfly.
Ecological studies of small numbers of populations of certain species of butterfly in Britain have indicated that temperature can influence their population dynamics (e.g. Pollard, 1979a; Thomas, 1983; Warren et al., 1986; and see discussion in Thomas, 1984). Pollard's population study and survey work on
Ladoga Camilla strongly suggested that a
con-traction and subsequent expansion in its range was associated with changes in weather patterns and, in particular, in June temperature. Unusual weather frequently results in local extinctions or marked fluctuations in population size (Pollard, 1984; see also Ehrlich étal.. 1980). The larvae of certain species exhibit thermoregulatory behaviour, absorbing radiant energy from the sun to gain some independence from air temp-erature (Porter, 1982, and see Dennis, 1985). Such behaviour may enable populations to per-sist in more northerly latitudes. Geographical variability in the number of annual generations or changes in voltimsm between years has been documented for several species in Britain (see Heath et al., 1984; Dennis, 1985). Such phenomena seem to be associated with variation in the length of time available for development ('season length', sensu Roff, 1980). Systematic investigations of temperature effects on the development rate of British butterflies and hence on the timing of emergence or phenology of natural populations have not been carried out although such effects are ubiquitous in other insects. An understanding of them is likely to provide important insights into the constraints of season length on the butterfly fauna. A study by Hagen & Lederhouse (1985) has examined the
timing of the adult flight period in a population of a North American butterfly, Papilio glaucus, using data describing the degree-days above the developmental threshold required for complete development. The present study is an analysis of the influence of temperature during post-diapause, pre-adult development on the timing and duration of the flight period of two strictly univoltine, satyrine butterflies at grassland sites throughout England and Wales. The study species are the meadow brown Maniola jurtina (L.) and the gatekeeper Pyronia tithonus (L.). Geographical variability in their phenology is also examined. The analysis was made possible by data recorded over the period 1976-85 on the Butterfly Monitoring Scheme (BMS) and generously made available by Dr E. Pollard.
Methods
The methods used in the BMS are fully docu-mented elsewhere (Pollard et al., 1975; Pollard, 1977). It is based on standardized counts usually made at least once a week along fixed transects at sites throughout Britain. Sites are visited from April until the end of September which covers the whole flight period of P. tithonus and all but the extreme tail for M.jurtina in a small propor-tion of its populapropor-tions. Estimates of the expecta-tion of adult life in a meadow habitat near Liverpool in 1976 and 1977 were similar for each species, varying from 3.5 to 8.5 days (Bra-kefield, 1979a, 1982a, b). Both species were also quite sedentary. Therefore, the probability of an individual insect being recorded along a transect on more than one date is likely to be similar for
M.jurtina and P. tithonus and although the BMS
FIG. 1. Map of the BMS sites (circles) in England and Wales used in this study. Map numbers refer to silc names in Table 1. The large circle shows position of the five sites with the prefix A. Squares indicate meteorological stations providing temperature data.
recorded in 1976). This choice enables a good coverage of England and Wales (Fig. 1) and yields the longest possible periods of continuous recording. It also provides a means of examining the effect of the exceptionally hot summer of 1976. Three of the sequences for individual sites (two for M.jurtina and one for P.tithonus) included a year with no records. These sequences were completed by substituting the overall means for the appropriate year and species. A total of twenty-nine sites were recorded in M.jurtina and twenty in P.tithonus. Their distribution is shown in Fig. 1. The nature of the individual data sets, particularly the variability between sites in the frequency of recording visits, means that some caution must be exercised when comparing estimates of tim-ing of the adult flight period for small numbers of particular sites. However, any trends present in the complete data are likely to have some bio-logical significance.
For each species at each site and in each year the (weighted) mean date of the counts is calcul-ated together with the standard deviation about
this date (SD). These represent estimates of the mean date of adult flight period and of the degree of synchronization of the flight period at each individual site. In P.tithonus there is no relationship between the site means for SD and total annual counts (r=-0.14). In contrast, for
M.jurtina this relationship is quite strong
(r=0.61, df=27, P<0.001). However, the latter correlation is entirely due to the five sites with counts substantially higher than the others (excluding the five sites: r=—0.03). Therefore there is some indication of a non-linear relation-ship in M.jurtina but the results described below are unchanged when these five sites are excluded from the analyses.
Both species overwinter as early to mid-mstar larvae exhibiting semi- (in M.jurtina) or wholly arrested development Regular night-time feed-ing on grasses commences durfeed-ing March. Larvae begin to pupate in late May (M.jurtina) or June. In both species nearly all adults have emerged by the end of August. Monthly mean daily max-imum and minmax-imum temperatures for the post-winter period of pre-adult development during March-August inclusive were abstracted from the Monthly Weather Reports (H.M.S.O., 1976-85) for a representative series of twelve meteorological stations distributed over the region covered by the BMS sites (Fig. 1). Cumulative daily temperatures were also calcul-ated for this period using the average of monthly means for daily maxima and minima. The latter statistics were used in the absence of the necess-ary data to calculate day-degrees. Statistical pro-cedures were performed using the MINITAB package on the mainframe computer at Univer-sity College, Cardiff.
Results
142 Paul M. Brakefield
TABLE 1. Overall means of mean date of adult flight period (days from 1 July) and standard deviation about this date (days) for Maniola jurtina and Pyronia tithonus at the sites indicated in the period 1976-85. Pooled means and SDs are given for all sites recorded in 1976. Map numbers as in Fig. 1. Site Map no. Al A2 3 A4 6 8 9 10 12 14* 15 16 17 A18 19 21 24 26 A27 28 30 31 32 39* 43* 44* 45* 47* 54* 60* Pooled mean Pooled SD Name Woodwalton Farm Bevills Wood Holkham Monks Wood Yarner Wood Old Winchester Hill Kingley Vale Oxwich Leigh Marshes Gomm Valley Foxholes Swanage Buttler's Hanging Wood hurst Studland Heath Castor Hanglands Dyfi Rostherne Mere Holme Fen Walberswick Aston Rowant S Waterperry Wood Wye Saltfleetby Leighton Moss Radipole Lake Gibraltar Point Skomer Nagshead Lindisfarne Maniola jurtina Mean date 29.1 22.4 23.9 27.8 19.5 41.2 36.0 39.8 18.1 21.0 20.8 38.1 20.5 21.8 -19.8 27.0 27.1 27.4 27.2 31.5 19.8 27.3 22.8 20.1 19.0 32.0 28.2 29.3 37.9 26.95 5.19 S I ) 11.8 13.8 12.5 14.5 14.4 20.0 17.6 15.8 13.2 13.8 13.9 18.0 16.1 12.0 _ 14.2 15.5 13.5 14.2 13.9 17.9 10.7 16.0 13.8 11.5 13.5 10.6 11.4 13.8 12.9 14.73 2.29 Pyronia tithonus Mean date 34.5 37.8 32.7 39.3 37.9 36.5 35.1 41.3 -_ -34.5 37.0 36.5 36.9 -34.2 37.5 33.9 36.3 38.8 34.8 -36.8 33.2 -36.51 3.56 S I ) 8.1 7.9 8.4 11.8 9.6 9.2 9.2 12.2 -_ -8.3 10.9 9.1 11.3 -10.5 9.8 9.2 10.4 10.3 9.2 -8.8 9.4 -9.78 1.72 'Sites not recorded in 1976.
substantially longer and less synchronized in M.jurtina. These differences in phenology are shown at nearly all sites common to both species (Table 1). When all individual sites in each year are analysed there is also a correlation between the mean date and SD (M.jurtina: r=0.32; P.tithonus: r=0.36, P<0.001 for each value). Therefore, although this correlation is not very high there is apparently a tendency for popula-tions with a later flight period to be less syn-chronized in emergence. This is probably a consequence of greater variability in the timing of the end of the flight period than of the begin-ning (analyses of all first and last records: M.jur-tma: F=1.60; P.tithonus: F=2.Q4, with P<0.001 for each value).
Variability between sites and years
TABLE 2. Two-way ANOVAs of (a) mean date of adult flight period, (b) change in mean date between consecutive generations, and (c) standard deviation about mean date for Mamola jurtina and Pvronia tnhonus at sites in England and Wales Results are given for both the 1976-85 and 1977-85 data set-.
Source of variation Mamola jurtina df SS MS F P Pyronia tithonus df SS MS F f
(a) Mean date, 1976-85 Site Year Error Total 1977-85 Site Year Error Total (b) Change Site Period Error Total 1977-85 Site Period Error Total 20 9 180 209 28 8 224 260 in mean date 20 8 160 188 28 7 196 231 (c) SD about mean date,
Site Year Error Total 1977-85 Site Year Error Total 20 9 180 209 28 8 224 260 9743.6 7360.6 4840.3 21944.5 14262.7 3153.2 10109.2 27525.1 between 175.7 10143.8 7992.1 18311.6 231.2 3703.5 8265.5 12200.3 1976-85 1059.0 270.2 946.2 2275.5 1522.7 312.1 1372.2 3207.1 487.2 817.8 26.9 5094 394.1 45.1 18.1 30.4 11.3 8.7 «0.001 «0001 «0.001 «0.001 16 9 144 169 9 8 152 179 771.8 7066.6 1821.0 9659.4 843.9 3651.9 1668.9 6164.7 48.2 785.2 12.6 44 4 456.5 110 3.8 62.3 4.0 41.5 <0.001 «0.001 <0.001 «0.001 years. 1976-85 8. K 1268.0 50.0 8.3 529.1 42.2 53.0 30.0 5.3 54.4 39.0 6.1 0.2 25.4 0.2 12.5 10.0 5.7 8.9 6.4 NS «0.001 NS «0.001 «0.001 <0.001 «0.001 <0.001 16 8 128 152 19 7 133 159 16 9 144 169 19 8 152 179 71.9 9122.4 3938.9 131332 72.4 3872.4 3530 3 7475.0 261.6 107.0 424.7 793.4 247.4
111.3
395.5 754.3 4.5 1140.3 3(1.8 3.8 553.2 26.5 16 .4 11.9 3.0 13013.9
2.6 0.1 37.0 0.1 2 0 9 5.5 4.0 5.0 5.3 NS «0.001 NS «0.001 <0.001 <0.001 <0.(X)1 <0.001relatively more of the total variance is accounted for by variation between years in P. tithonus. The analysis of changes in mean date between con-secutive generations (Table 2b) shows that less than 2% of the total variance is due to
differ-ences between sites. This strongly suggests that in each species the sites are behaving similarly with regard to changes in timing of the adult flight period between years. Such consistency is indicative of general effects of weather.
Further-TABLE 3. Spearman rank correlations between Kititudc and the indicated mean statistics describing the adult flight period of
Mamola jurtina and Pvronia tithonus at sites in England and Wales.
Values are given for the 1976-85. and the 1977-85 (larger no. of sites) data sets.
144 Paul M. Brakefield
more, analysis of the data for individual years at the nineteen sites common to both species shows a highly significant correlation between species for mean date (r=0.59. df=184). The correla-tion for SD is much weaker (r=0.17, P<0.05).
Relationships to latitude
The correlation coefficients describing the relationships between phenology and latitude are given in Table 3. There is no evidence of any change in the mean date of adult flight period with latitude in Britain for either species although the correlations are all negative sug-gesting, if anything, earlier mean dates further north. Analysis of the first and last dates on which each species was recorded indicates that in M.jurtma emergence begins earlier, and adults fly until later in the south. There are no corres-ponding relationships for P.tithonus. Associated with the changes in dates in M.jurtina is a more synchronized flight period in the north. The cor-responding correlation is also negative in P.tithonus but does not reach formal signifi-cance. The four sites for M.jurtina with the longest flight period are all near the south coast or in southern England (8, 9, 16 and 30). Three of the corresponding sites for P.tithonus are in southwest England (6) or west Wales (10 and 24), and the other is on the south coast (19). Relationships to temperature
The interpretation of correlations between statistics describing phenology and monthly temperatures is not greatly complicated by auto-correlations between the temperature variables. Thus for the matrix of sixty-six correlation coefficients between the mean monthly max-imum and minmax-imum temperatures for the 10-year period there were only six (all positive) significant values (i.e. r>0.63) and many nega-tive values. Four of the six significant values are for maximum x minimum values for the same month. This matrix and that for cumulative monthly temperatures suggest that there is a trend towards consistent temperatures during each of the periods March-May and June-August.
Examination of the correlation coefficients given in Table 4(a) for mean date of adult flight period against temperature variables reveals similar patterns in each species. Cooler
22 O 2 20 o> o. E S E D 0) C 3 -3 18
-ie
a)
-,82 ,83 ' 7 8 20 40Phenology of satyrine butterflies 145
I. 24 o • • ra & 22 E O 20 .83 ,82 • 20 40
Mean date of adult flight period, days from 1 July
FIG. 2. Relationships between mean date of adult flight period of Maniola iiiriuni (•) and of Pyroma
inlwnwt (o) and (a) mean maximum temperature in June and (h) cumulated daily temperatures for March
to August. Figures indicate years.
tures in the three summer months, especially June (Fig. 2a), are associated with later flight periods. The earlier months have little influence on timing of flight. The similarity of the effects on the two species is emphasized by the values for cumulative daily temperatures. The relation-ships for average temperature over the whole period of post-winter development are shown in Fig. 2(b). Exclusion of the exceptionally hot year of 1976 with a very early adult flight period in each species has little effect on the correlation matrix. The difference between the earliest and latest years is about 3 weeks in each species (Fig. 2).
Evidence for strong temperature effects on the synchronization of the adult flight period is less apparent (Table 4(b)). However, the flight period is probably more spread out in M.jurtina in years when July and, possibly, August are relatively cool. A similar effect may operate in P.tithonus but is not significant for the present data. In M.jurtina there is an indication th.it .1 warm April, and perhaps May, tends to produce a more synchronized flight period. The flight period was most synchronized in 1983 in both species and most spread out in 1985 and 1980 (M.jurtina only). There are no clear relation-ships between either mean date of adult flight
period or SD and the temperature variables for the period March-August in the preceding year.
Discussion
This study shows that 94.5% of the variation between years in mean date of adult flight period in M.jurtina, and 75.5rr in P.tithonus, is
146 Paul M. B rake field
of adult flight period than those in the last 2 or 3 months. It is possible that spring temperatures are relatively more important to the synchro-nization of the flight period. These findings indi-cate that substantial temperature effects on total development time in natural populations occur during the period of late instar larvae and pupae. These may be most marked in the pupal stage because any damping of such effects by modification of activity and feeding patterns is not possible. Pupation occurs from late May
(M.jurtina) or June until early or mid-August.
The length of the pupal stage is temperature-dependent in the laboratory. In M.jurtina its duration is 15-18 days at 15°C and 9-11 days at 25°C(Brakefield, 1979a). Cooler periods in mid-summer will produce delays in emergence and, therefore, the observed later and more pro-longed adult flight period. Interestingly. Turner
et al. (1986) showed that the relationship
between summer climate and species diversity in Britain was closely similar for ectothermic but-terflies and endothermic moths (as adults). One possible explanation suggested by Turner et al. for this similarity is that the relationship is not a direct one through adult thermorégulation but may be associated with climatic effects on larval development.
The great variability in the length of the flight period in M.jurtina has been recognized for a long time. Localities with the longest flight peri-ods include many with shorter turf, especially on chalk soils (Pollard, 1979b). They have also been reported on the Isles of Scilly (Ford, 1975) and the Isle of Wight (Thomson, 1971). In some populations there may be more than one over-lapping peak of emergence of each sex (Bra-kefield, 1982b). Males tend to emerge before females and the degree of protandry can be extremely marked. It would be interesting to investigate the relationships between variation in protandry, the form of emergence and habitat. Breeding experiments in which sibs of
M.jurtina with differing wing spotting
phenotypes showed differences in timing of emergence strongly suggest that there is genetic variability for development rate (Brakefield, 1984). Such variability was demonstrated by selection experiments on the pattern of volti-nism in stocks of the satyrine Coenonympha
pamphilus (Lees, 1962, 1965). In many
popula-tions of M.jurtina, butterflies emerging later in the season have less extensive wing spotting and
are also smaller in size (references in Brakefield, 1984; Brakefield & Macnair, unpublished observations).
The variability between years in mean date of adult flight period and in SD about this date is similar for M.jurtina and P.tithonus. Why then is the flight period of M.jurtina both substantially less synchronized and more variable between populations? The answer may lie in the wider habitat preferences of M.jurtina (Brakefield, 1979a, 1982a). In a population study in a meadow near Liverpool, P.tithonus was largely restricted to rough grassland immediately adja-cent to shrubs or scrub while M.jurtina was much more widely distributed throughout the habitat. This difference is characteristic of the species at all stages of the life cycle (Brakefield, unpub. data; and see e.g. Heath et al., 1984). The grea-ter variability in microhabitat shown by
M.jur-tina is likely to result in more heterogeneity in
development time and emergence both within and between populations. At a site in the Isles of Scilly post-winter larvae of M.jurtina from sheltered grass gullies between thickets were more advanced than those of open grassland (Brakefield, 1979a).
The absence of a strong relationship between latitude and mean date of adult flight period is unexpected because of the south-west to north-east decline in mean annual temperature (Tout, 1976). However, the pattern is less clear-cut in mid-summer. In July most of England south of a line from sites 26 to 39 in Fig. 1 has a mean temperature of about 16°C (reduced to sea-level). The northward decline is only marked in the region above this line which includes but two study sites (where only M.jurtina occurs). Coas-tal areas are generally cooler than inland areas at t h i s time of year (Tout, 1976). The apparent relationships between latitude and synchroniza-tion of the flight period (Table 3) are unlikely to result from a latitudinal change in adult longevity since estimates of expectation of adult life of M.jurtina provide no evidence for this (Brakefield, 1982b).
Although the peak of adult emergence in
M.jurtina remains at roughly the same date at
becoming more stabilizing in nature towards the margins of the species' range (see Brakefield, 1979b). A contributing factor could be that grassland habitats become less diverse north-wards in England and Wales. Such effects could play an important role in determining northern limits in range. M.jurtina occurs up to the north of Scotland while P.tithonus reaches its limits along the Cumbrian coast and in Yorkshire not far north of site 39 (Fig. 1 : Heath et al.. 1984). The later and shorter flight period of P.tithonns may be related to more restrictive thermal con-straints than in M.jurtina. There is no evidence that the adult flight period of P.tilhonu\ becomes more synchronized in northern popula-tions although the correlation is negative (Table 3). There is no change in the timing of the begin-ning or end of the flight period. Thus this species may be responding more passively to climatic change than M.jurtina and consequently be less able to extend its range so far northwards. While the influences of climate during the adult flight period on the population dynamics and distribu-tion of butterfly species are clearly important, those acting through development rate should not be neglected.
Acknowledgments
It is a pleasure to t h a n k Ernie Pollard for help and advice at all stages of this project. The efforts of all individual BMS recorders and the support for the scheme by the Institute of Ter-restial Ecology and the Nature Conservancy Council are also gratefully acknowledged. R. L. H. Dennis, N. A. C. Kidd. J. A. Thomas. J. R. G. Turner and an anonymous referee provided useful comments on the manuscript. I n i t i a l impetus for this study came from an examination of data for mean date in M.jurtina at five south-western sites (1976-84) supported by a grant (No. GR3/4909) to M. R. Macnair from the Natural Environment Research Council.
References
Harbour. D . A . (1986) Why arc there so few butterflies in Liverpool ' An .inswcr. Antenna. Bulletin of the
Rovul Entomological Societ\ of London. 10,
72-75.
Brakefield. P.M. (1979a) An cxpenment.il study of the maintenance of variation in spot pattern in
M.jurtina. Unpublished Ph.D. thesis. University
of Liverpool
Brakefield. P.M. (1979b) Spot-number in Mamola
nir-initi Variation between generations and selection
in marginal populations Heredit\. 42, 259-266 Brakefield. P.M. (1982a) Ecological studies on the
butterfly Mamola jurtina in Britain I. Adult behaviour, microdistribution and dispersal.
Jour-nal of Animal Ecology, 51, 713-726.
Brakefield. P.M. (1982b) Ecological studies on the b u t t e r f l y Mamola jurtina in Britain II Population dynamics: the present position Journal of Animal
Ecology. 51, 727-738.
Brakefield. P.M. (1984) The ecological genetics of quantitative characters of Mamola \urnna and other butterflies. The Biology of Butterflies.
Sym-posium of the ROMI/ Entomological Society of Lon-don. 11 (ed. by R I Vane-Wright and P. R.
Ackery). pp. 167-190 Academic Press. London. Dennis. R.L H (1977) The British Butterflies. Their
Origin and Estahlishment. E. W. Classes.
Faringdon. Oxon
Dennis. R.L.H. (1985) Voltmism in British Aglaix
urticae ( L . ) (Lep.: Nymphalidae): variation in
space and time. Proceedings and Transactions of
the British Entomological and Natural History Society. 18, 51-61.
Dennis. R.L.H. & Williams. W . R . (1986) Butterfly 'diversity' Regressing and a little latitude.
Ante-nna. Bulletin of the Royal Entomological Society of London. 10, 108-112.
Ehrlich. P.R.. Murphy. D.D. Singer. M.C.. Sher-wood. C.B.. White. R.R & Brown. I.L. (1980) Extinction, reduction, stability and increase: The responses of checkerspot butterfly (Euphydr\as> populations to California drought Oecologni
(Berlin). 46, 101-105.
Ford.E.B. (1975) Ecological Genetics, 4th edn. Chap-man & Hall. London
Hagen. R.H. & Lederhouse. R.C (1985) Polymodal emergence of the tiger swallowtail. Papilio glaucus (l.epidoptcra Papiliomdae) source of a false second generation in central New York State.
Ecological Entomology. 10, 19-28.
H e a t h . .1 . Pollard. E. & Thomas. J A ( 1984) Atlas of
Butterflies in Britain and Ireland. Viking.
Harmondsworth
1 ecs. E. (1982) On the voltmism of Coenonympha
pamphilus (L.) (Lep.. Satyridae). Entomologist.
95, 5-6.
Lees. E. ( 1965) Further observations on the voltmism of Coenonympha pumpliilux ( L . ) (1 ep . Satyridae). Entomologist, 98, 43-45.
Pollard. E. (1977) A method of assessing changes in t h e abundance of b u t t e r f l i e s Biological <.
'on.ierva-non. 12, 115-134.
Pollard. E. (1979a) Population ecology and change in range of the white admiral butterfly Ladoga
Camilla L. in England. Ecological Entomology. 4,
61-74.
Pollard. E. (I979h) A national scheme for monitoring the abundance of butterflies the first three years. Proceedings und Transaction* of the British
I inonio/ogical and Natural History Sociel\. 12,
148 Paul M. Brakefield
Pollard. E. (1984) Fluctuations in the abundance of butterflies. 1976-82. Ecological Entomology, 9, 179-188.
Pollard, E., Elias, D.O., Skelton, M.J. & Thomas, J A . (1975) A method of assessing the abundance of butterflies in Monks Wood National Nature Reserve in 1973. Entomologist") (iazette, 26, 76-88.
Porter, K. (1982) Basking behaviour in larvae of the butterfly Euphydryas aurmia. Oikos, 38, 308-312. Roff, D (1980) Optimizing development time in a seasonal environment: the 'ups and downs' of cli-nal variation Oeco/ogia (Berlin), 45, 202-208. Thomas, J.A. (1983) The ecology and conservation of
Lysandra bellargus (Lepidoptera Lycaemdac) in
Britain. Journal of Applied Ecology, 20, 59-83. Thomas, J. A. ( 1984) The conservation of butterflies in
temperate countries Past efforts and lessons for the future. The Biology of Butterflies, Symposium
of the Royal Entomological Society of London, 11
(ed. by R. I. Vane-Wright and P. R Ackery), pp. 333-353. Academic Press, London
Thomson, G. (1971) The possible existence of temp-oral sub-speciation in Maniola jurtina ( L . ) .
Entomologist's Record and Journal of Variation,
83, 87-90.
Tout, D. (1976) Temperature. The Climate of the
Bri-tish Isles (ed. by T. J. Chandler and S. Gregory),
pp 96-128. Longman, London.
Turner, J.R.G. ( 1986) Why are there so few butterflies in Liverpool? Homage to Alfred Rüssel Wallace.
Antenna, Bulletin of the Royal Entomological Society of London, 10, 18-24.
Turner, J.R.G.. Gatehouse, C.M. & Corey, C.A. ( 1986) Does solar energy control organic diversity? Butterflies, moths and the British climate ()iko\ (in press)
Warren, M.S., Pollard, E. & Bibby, T.J. (1986) Annual and long-term changes in a population of the wood white butterfly Leptidea smapis. Journal
of Animal Ecology, 55, 707-719.
