Material and methods
A hundred adult female butterflies were collected ‘fresh’ (i.e. shortly after onset of population emergence in the field, but already mated as shown by the deposition of fertile eggs later on) from the following localities (Table 1): the island of Sardinia, Italy (10 individuals: 7 M. nurag, 3 M. jurtina); vicinity of Illmitz (Burgenland), eastern lowland Austria (12 individuals in 2011 and 21 individuals in 2005: M. jurtina); the island of Krk, Croatia (28 individuals: M. jurtina); and Mariahilfer Berg/Unterberg (Lower Austria), eastern montane Austria (29 individuals: M. jurtina). Individuals from lowland Austria hereafter are referred to as ‘Pannonian’, individuals from Krk and Sardinia are termed ‘Mediterranean’, and individuals from Mariahilfer/Unterberg are referred to as ‘Alpine’ populations.
Butterflies were kept individually in 1-litre plastic containers lined at the bottom with humid paper towels, under 24h light-dark cycles with a photoperiod of L12:D12 at 24°C (light) and 16°C (dark). Sugar water (5% sucrose solution dissolved in tap-water) was available ad libitum to the animals as a source of carbohydrates, and fresh cut grass was provided as egg-laying substrate. For 40 of the individuals collected in 2005, a commercial mixture of twenty amino acids (Alanine, Arginine, Asparagine, Aspartic acid, Cysteine, Glutamic acid, Glutamine, Glycine, Histidine, Isoleucine, Leucine, Lysine, Methionine, Phenylalanine, Proline, Serine, Threonine, Tryptophan, Tyrosine, Valine) as used as human diet supplement was added to the sugar water as extra nitrogen source for these unusually long-lived butterflies. Containers were checked for eggs daily, and all eggs were removed after counting so that each count indeed reflected the number of eggs deposited at a given date.
Data were first analyzed separately for each population from the two different years. When we found no significant differences between years in populations from the Pannonian region, we pooled the data from 2005 and 2011 for subsequent analysis.
Only individuals that, once in captivity, had a reproductive period longer than five days and produced 10 or more eggs were included in the analyses.
For each individual we recorded the following five variables:
(a) timing of oviposition (= the number of days from capture to first egg laid)
(b) fecundity (= total number of eggs laid during entire lifetime)
(c) lifespan (= number of days from capture to death)
(d) reproductive period (= number of days from when first egg was laid till the day when the last egg was laid)
(e) post reproductive lifespan (= number of days lived after deposition of last egg)
To obtain normal distribution of the variables they were square-root transformed or log(e) transformed (for total number of eggs). Square-root or log transformed data were tested for significant effects of geographic provenance or diet in one-way ANOVAs followed by Tukey’s post hoc test for multiple comparisons. To test for the effects of diet and population membership on lifespan, fecundity and timing of oviposition in M. jurtina (M. nurag did not receive extra amino acids, as there were too few individuals to create a control group) we used two-way ANOVA. All statistical analyses were performed in the program Statistica 8.0 (StatSoft, 2005).