Content uploaded by Manuela Pinzari
Author content
All content in this area was uploaded by Manuela Pinzari on Jul 24, 2018
Content may be subject to copyright.
INTRODUCTION
The Marsh Fritillary butterfly Euphydryas au-
rinia (Rottemburg, 1775) has suffered a severe decline
in most European countries during the 20th Century
(Van Swaay & Warren, 1999; Van Swaay et al., 2010).
For this reason E. aurinia is listed in Annex II of the
BOLL. SOC. ENTOMOL. ITAL., 148 (3): 121-140, ISSN 0373-3491 15 DICEMBRE 2016
Riassunto: Deposizione delle uova, scelta delle piante ospiti e sopravvivenza delle larve di Euphydryas aurinia provincialis (Lepidoptera Nym-
phalidae) nell’area Mediterranea (Italia centrale).
In questo lavoro, sono riportati i risultati di uno studio di cinque anni sul campo finalizzato a individuare le piante nutrici delle larve e le loro
preferenze alimentari e valutare quali segnali siano utilizzati dalle femmine per la deposizione delle uova in Euphydryas aurinia provincialis
nell’area Mediterranea. Le femmine depongono su Gentiana cruciata, Scabiosa columbaria e Cephalaria leucantha e le larve si nutrono indif-
ferentemente su tutte queste piante, ma anche su Lonicera caprifolium e, in laboratorio, su Gentiana lutea. Le femmine non mostrano alcuna
preferenza per una specifica pianta ospite e le larve passano a nutrirsi dall’una all’altra pianta senza alcuna difficoltà. I fattori più importanti
nel determinare la selezione della pianta ospite per l’ovideposizione sono la visibilità, l’accessibilità e l’esposizione solare delle piante. Lo stato
vegetativo delle piante ospite è il fattore chiave che determina quale specie vegetale sia utilizzata dalle larve nei diversi stadi dello sviluppo du-
rante il periodo pre- e post-diapausa. G. cruciata e C. leucantha, piante ospite di grandi dimensioni, sono fondamentali per la crescita e la so-
pravvivenza degli stadi larvali pre-diapausa (I-III instar), mentre sono del tutto inadeguate per nutrire le larve in primavera dopo la diapausa
invernale a causa del loro ritardo nella ripresa vegetativa; invece le piante di S. columbaria, molto più abbondanti e in fase vegetativa sempre
molto più avanzata rispetto a G. cruciata e C. leucantha, sono utilizzate in primavera dalle larve post-diapausa per completare lo sviluppo. L.
caprifolium, seppur in misura minore per la minor diffusione e il relativo ritardo nello sviluppo dei germogli, svolge un ruolo analogo a S. co-
lumbaria. Il lavoro è completato da alcune considerazioni conclusive dedotte dalle osservazioni. Vivere in gruppo sembra essere evolutivamente
selezionato sin dalla deposizione delle uova sulle piante più grandi; infatti, le femmine di E. aurinia provincialis se trovano uova deposte da
un’altra femmina vi depongono a fianco o addirittura sopra. Nonostante la competizione trofica tra le larve, questo comportamento delle femmine
risulta evidentemente vantaggioso ed aumenta il tasso di sopravvivenza dei nidi sulle piante più grandi. La vita gregaria di un numero maggiore
di larve su una pianta di grandi dimensioni: nel periodo pre-diapausa, riduce i casi di starvation; durante la diapausa invernale, con nidi più
robusti e più stabili permette una maggiore resistenza ai predatori e ai rigori invernali; nel periodo post-diapausa, permette una maggiore effi-
cienza nella termoregolazione rendendo le larve indipendenti dalle basse temperature esterne.
Abstract: In this paper, we show the results of research that can inform conservation measures elsewhere in Europe for the endangered butterfly Eu-
phydryas aurinia. A five year field study was undertaken to identify the host plant preference of larvae of Euphydryas aurinia provincialis in the
Mediterranean and which signals are used by females to lay their eggs. The females oviposit on Gentiana cruciata, Scabiosa columbaria and
Cephalaria leucantha; the larvae feed on all these plants and additionally on Lonicera caprifoliumin the wild and on Gentiana luteain the laboratory.
The females do not show any preference for a specific host plant and the larvae move from one species of plant to another without any difficulty. The
most important factors in determining the female oviposition are the visibility, accessibility and sun-exposure of the host plants. The vegetative state
of host plants is the key factor in larval use of plants during the pre- and post-diapause period. The large-sized host plants, G. cruciata and C.
leucantha, are optimal for the growth and survival of the pre-diapause I-III larval instar, while they are unavailable to the larvae in Spring because
of their delayed vegetative growth. The post-diapause larvae preferentially feed on plants of S. columbaria, and to lesser degree L. caprifolium, as
they provide and abundant food source compared with G. cruciata and C. leucantha. The results also suggest that, there is an evolutionary advantage
in large numbers of caterpillars feeding together, with the females of E. aurinia provincialis preferring to lay their eggs nearby or above egg batches
laid previously by another female, and selecting large plants for oviposition. Despite the competition for food among caterpillars, the oviposition be-
haviour of females is advantageous and increases the larval survival rate on large plants. The gregarious larval behaviour provides several benefits
during both pre-diapause period (avoiding starvation) and post-diapause period (efficiency in thermoregulation).
Key words: Euphydryas aurinia provincialis, Female host plant choice, Larval host plant, Larval web survival, Larval aggregation behaviour.
121
Manuela PINZARI*- Mario PINZARI** - Valerio SBORDONI*
Egg laying behaviour, host plants and larval survival of Euphydryas aurinia provincialis
(Lepidoptera Nymphalidae) in a Mediterranean population (central Italy)
*Manuela Pinzari, Valerio Sbordoni, Dipartimento di Biologia, Università di Roma Tor Vergata, Via della Ricerca Scientifica 1,
00133 Roma, Italy. E-mail: manuela.pinzari@uniroma2.it, valerio.sbordoni@uniroma2.it
**Mario Pinzari, Dipartimento di Ingegneria, Università di Roma 3, Via della Vasca Navale 79, 00146 Roma, Italy.
E-mail: mario.pinzari@uniroma3.it
PINZARI - PINZARI - SBORDONI
European Community Habitats and Species Directive
(92/43/EEC) and is also given protection through the
Bern Convention.
In Italy the Euphydryas aurinia species com-
plex is represented by three taxa that were alterna-
tively considered as species or subspecies depending
on authors (Verity, 1950; Hartig, 1968; Prola et al.,
1978; Balletto & Cassulo, 1995; Parenzan & Porcelli,
2006; Balletto et al., 2014a). These taxa are: i) aurinia
(Rottemburg, 1775), occurring in wet meadows in the
Po Plains, ii) glaciegenita (Verity, 1928), in alpine re-
gion of Northern Italy, and iii) provincialis (Boisduval,
1828), which is limited to the Maritime Alps and the
Apennine Mountains (Balletto et al., 2014b). Accord-
ing to the most recent assessment (under Art. 17 of
Habitat Directive) the conservation status of E. aurinia
populations is judged to be “bad” in Central and
Northern Europe, where suitable habitat has become
strongly reduced in area and fragmented. In Italy, at
least 12 populations of E. a. aurinia have become ex-
tinct because of habitat destruction (Bonelli et al.,
2011). On the contrary, the conservation status of the
Mediterranean and Alpine populations is “favourable”
(Balletto et al., 2014b).
Most of the available information on ecology
and biology of the E. aurinia species complex comes
from E. a. aurinia in Western and central Europe
where it has been studied in some detail for over a cen-
tury (Ford & Ford, 1930; Porter, 1981; Warren, 1993;
1994; 1996; Warren et al., 1994; Klemetti &
Wahlberg, 1997; Wahlberg, 2001; Anthes et al., 2003;
Hula et al., 2004; Saarinen et al., 2005; Mazel, 2006;
Peńuelas et al., 2006; Svitra & Sielezniew, 2010;
Porter & Ellis, 2011). In contrast, very few observa-
tions on ecology and biology of this species have been
carried out in the Mediterranean area. The first eco-
logical observations on E. aurinia in the Mediter-
ranean are reported by Verity (1950) and Jutzeler
(1994) and only recently by Casacci et al. (2014). Ver-
ity describes E. aurinia provincialis as a butterfly that
usually inhabits both marshy meadows in the forest
glades and dry meadows. Jutzeler refers to his obser-
vations on caterpillars and their host plants from a Lig-
urian population. Casacci et al. (2014) reports the
results on metapopulation dynamics and dispersal
abilities of two populations of E. (a.) glaciegenita
from NW Alps and E. (a.) provincialis from Northern
Apennines, with mention of some larval food plants
(Tab. 1).
The main purpose of this paper is to understand
why the Mediterranean populations of E. aurinia in
Central Italy are in a better state than elsewhere by
looking at key aspects of its ecology and biology. In
this research we looked at a population of E. a. provin-
cialis in the Apennines of Central Italy with the aim
of identifying: i) host plant preference by ovipositing
females; ii) larval host plant preferences; iii) the inter-
action between larvae and host plants during their life
cycle; and iv) the larval survival after winter diapause.
MATERIALS AND METHODS
Study animal. E. aurinia is univoltine butterfly with
adults emerging from May and flying through June to
July depending on altitude and microclimate. Eggs are
laid in batches of 300 or more on the underside of the
leaves of host plants (Kuussaari et al., 2004). The
species has six larval instars (three pre-diapause in-
stars and three post-diapause instars); in European
populations, the first five instars are gregarious and
the sixth instar is solitary. The first three instar larvae
are pale-brown and then change colour during the
moult from the third to the fourth instar, becoming jet-
black. In the early stages, larvae feed gregariously
within a communal silken web and move from a plant
to plant as their host plants are consumed; fourth instar
larvae do not feed prior to diapause and overwinter in
a larval web (hibernaculum), usually close to the
ground and incorporating a few dead leaves which are
spun tightly together (Porter, 1981; 1982). When the
fourth instar larvae emerge from the winter diapause
they still show the gregarious behaviour and use bask-
ing behaviour as a way of thermo-regulating (Porter,
1981; 1982). Larvae start to disperse in their fifth in-
star and feed solitarily (Warren, 1996; Kuussaari et al.,
2004). Finally they pupate close to the ground, usually
on plant stems (Warren, 1996).
Study area. E. a. provincialis is a widespread species
in the Central Apennines and in particular in Monti
Reatini (data from Osservatorio per la Biodiversità del
Lazio). The study population of E. a. provincialis in
Central Apennines was discovered during previous
survey work (Pinzari et al., 2010) on the Lepidoptera
in the area surrounding Vallemare (Rieti, Lazio, Cen-
tral Italy, Location WGS84: N42.4836°-E13.1148°).
Here, the species occupies habitat patches in montane
grassland and slopes, sometimes steep and rocky,
within a mosaic of different habitats including wooded
122
Host plants and larval survival of
Euphydryas aurinia provincialis
123
Tab. 1. Status of knowledge on host plants and larval host use in E. aurinia subspecies.
E. aurinia Egg laying Pre-diapause Post-diapause Instar not specified
subspecies I-III instar IV-VI instar
aurinia Succisa pratensis Succisa pratensis Succisa pratensis Succisa pratensis
(Porter, 1981; Porter & Ellis, (Porter, 1981; Sardet & (Porter, 1981; Porter & (Luckens, 1978; Betzholtz
2011; Anthes et al., 2003; Betremieux, 2006; Porter Ellis, 2011) et al., 2007; Casacci et al.,
Sardet & Betremieux, 2006; & Ellis, 2011; Botham et al., 2014).
Eeles, 2014) 2011; Masek, 1987,
In: Konvicka et al., 2003)
Scabiosa columbaria Scabiosa columbaria Scabiosa columbaria Scabiosa columbaria
(Sardet & Betremieux, 2006) (Sardet & Betremieux, 2006; (Lobenstein, 2008) (Porter, 1981; Warren, 1993;
Mazel, 2006) Mazel, 1984)
Knautia arvensis Knautia arvensis Lonicera periclymenum Scabiosa comosa
(Sardet & Betremieux, 2006) (Sardet & Betremieux, 2006; (Porter & Ellis, 2011) (Wahlberg et al., 2001)
Mazel, 2006)
Lonicera implexa Lonicera periclymenum Gentiana cruciata Knautia arvensis
(Stefanescu et al., 2006) (Warren, 1993) (Svitra & Sielezniew, 2010) (Porter, 1981; Mazel, 1982,
1984; Mazel, 1977
In: Jutzeler, 1994)
Gentiana cruciata Gentiana cruciata Valeriana dioica Gentiana cruciata
(Sardet & Betremieux, 2006; (Sardet & Betremieux, 2006; (Anthes, 2002) (Hafner, 2001 com. pers.
Lobenstein, 2008) Svitra & Sielezniew, 2010; In: Anthes, 2002; Perru &
Lobenstein, 2008) Sardet, 2005; Mazel &
Lutran, In: Jutzeler, 1994)
Gentiana asclepiadea Cephalaria leucantha Valeriana sambucifolia Gentiana asclepiadea
(Anthes, 2002; Anthes et al., (Mazel, 2006) (Betzholtz et al., 2007) (Ebert & Rennwald, 1991,
2003) In: Anthes, 2002)
Menyanthes trifoliata Gentiana lutea
(Anthes, 2002; Anthes et al., (Rey et al., 2004 in Sardet &
2003) Betremieux, 2006)
Cephalaria leucantha
(Mazel, 1982, 1984; Mazel &
Lutran, In: Jutzeler, 1994)
Valeriana officinalis
(Favretto 2009)
Dipsacum fullonum
(B. Gillam, com. Pers.
In: Warren, 1993)
beckeri Lonicera implexa Lonicera implexa Lonicera periclymenum Lonicera periclymenum
(Penuelas et al., 2006; (Stefanescu et al., 2006) (Junker & Schmitt, 2010; (Warren et al., 1994)
Stefanescu et al., 2006) Warren, 1993)
Lonicera implexa Lonicera implexa
(Stefanescu et al., 2006) (Munguira, com. Pers;
Templado, 1975, in Warren,
1993; Warren et al., 1994)
Lonicera etrusca
(Warren, 1993; Warren et al.,
1994; Kankare et al., 2005,
in Stefanescu et al., 2006)
Continua nella pagina seguente
PINZARI - PINZARI - SBORDONI
areas, hedgerows, fields and less steep areas which are
occasionally ploughed and cultivated. E. a. provin-
cialis inhabits discrete habitat patches that differ in
their larval host plants composition according to the
following altitudinal zones: i) 1500-1300 m, G. lutea
L., G. cruciata L. and S. columbaria L.; ii) 1300-1000
m, G. cruciata, S. columbaria and L. caprifolium L.;
iii) 1000-800 m, S. columbaria and L. caprifolium or
C. leucantha (L.) Schrad. ex Roem. & Schult. We de-
tected 13 habitat patches whose position, features and
land use are indicated in Fig. 1; they are adjacent to
each other and MRR sampling have shown (Pinzari
M., unpublished data) that flying adults can be easily
move between areas; in the study patches (Fig. 1) C.
leucantha and G. lutea were absent. Additional obser-
vations were also carried out in habitat patches located
outside the study area; they are at Monte Cagno and
Villa Camponeschi, where the species and its host
plants G. lutea and C. leucantha are recorded.
Adults of E. a. provincialis are regularly pres-
ent within the habitat patches and absent in forests and
scrub. They do disperse across other land parcels, such
as agricultural areas and pastures, but these lack suit-
able host plants and offer little in the way of nectar
sources. Caterpillars were observed only in patches of
habitat with uncultivated steeply sloping rocky areas
or road verges.
The field-work on adults of E. a. provincialis
began in 2010 and has been continued up to 2015. Ob-
servations on larval stages started in 2012 and contin-
ued until 2015. A total of 77 days over three years
were spent monitoring egg batches and larval webs,
focusing on: eggs (from May to the end of June); pre-
diapause gregarious phase (I-III instar on larval webs)
(from May to September); diapause phase (IV instar
in larval webs) (from September to mid-February);
post-diapause gregarious phase (IV instar) (from mid-
February to mid-March); and solitary caterpillars (V-
VI instar) (from March to mid-April).
Laboratory observations on larval feeding be-
haviour were carried out in captivity in Rome in arti-
ficial rearing facilities indoors in February 2014 (19
February to 13 March) during eleven days and in April
2015.
Host plants. The biology of pre-imagined stages of
E. aurinia complex has been studied by numerous sci-
entists in Northern and Central Europe. There are far
124
Tab. 1. Segue dalla pagina precedente.
E. aurinia Egg laying Pre-diapause Post-diapause Instar not specified
subspecies I-III instar IV-VI instar
glaciegenita Gentiana clusii
(Geiger, 1987
In: Warren, 1994)
Gentiana acaulis
(Geiger, 1987 In: Warren,
1994; Casacci et al., 2014)
Gentiana kochiana
(Gerber, 1972, In: Anthes
2002; Warren, 1996; Gerber,
1972, In: Mazel, 1982)
Gentiana alpina
(Warren 1996)
provincialis Cephalaria leucantha Lonicera etrusca Knautia arvensis
(Jutzeler, 1994) (Nel, 1983, in litt., (Casacci et al., 2014)
In: Mazel, 2006)
Cephalaria leucantha Cephalaria leucantha
(Jutzeler, 1994) (Casacci et al., 2014)
Centranthus ruber Centranthus ruber
(Jutzeler, 1994) (Bromilov In: Verity, 1950)
Host plants and larval survival of
Euphydryas aurinia provincialis
fewer observations on the species complex in the
Mediterranean area and in particular in Italy. For syn-
thesis, in Tab. 1, all known host plants of E. aurinia
subspecies are presented, with an indication where
possible the corresponding use by pre- and post-dia-
pause pre-imaginal stages. E. a. aurinia is the most
studied sub-species and the host plants used by
ovipositing females and by larvae are known. Several
authors suggest that larvae feed exclusively on a sin-
gle, main host plant; others state that the species can
utilize a wide spectrum of host plants. Succisa praten-
sis (Dipsacaceae) is the most important food plant of
E. a. aurinia across the European species range and
especially in the North; some other plants belonging
to Dipsacaceae, Caprifoliaceae, and Gentianaceae
families were also recorded for the species in Western
and Southern Europe. E. a. beckeri is dependent to
genus Lonicera for both oviposition and larval food
plant. Prior to the current study, information on egg
laying behaviour and female host plant preferences in
E. a. provincialis was not clear; observations on cater-
pillars and host plants referred to populations in rocky
125
Fig. 1. Habitat and position of the habitat patches in the study area: 1, 4, Crossroads Santa Maria del Monte – Fonte Brignola;
2, 3, Fonte Perara; 5, 6, 7, 8, 11, along the path between the road to Santa Maria del Monte and Hairpin curve Mattano-Malepasso;
9, 12, 13, Hairpin curve Mattano-Malepasso; 10, Pian Mattano. The patch 14, Villa Camponeschi, is not included on this map
due to its distance from the other localities (2 km from Fonte Perara). Features and land use of the study area: A) Semi-natural
dry grasslands and scrubland facies on calcareous substrates, where adults of E. a. provincialis were recorded in past studies
(Pinzari et al., 2010); B) cultivated areas for intense pasture and hay harvesting areas; C) thick scrubs; D) mineralized areas,
erosion or with very little vegetation cover; E) high trees oak and beech forests; where adults of E. a. provincialis were absent.
PINZARI - PINZARI - SBORDONI
habitat in Southern France and in NW Italy (Liguria).
Similarly, E. a. glaciegenita has been little studied and
its biology is poorly known; existing observations sug-
gest it uses species of Gentianaceae as food plants.
General. All study subjects (host plants, egg batches,
larval webs) were marked by an alphanumeric code
written on a stone that was positioned nearby. This en-
abled the relocation of plants at each sampling period
and from year to year. All qualitative and quantitative
observations have been documented with notes, pho-
tographs and short videos.
Female host plant choice. To identify the host plants
of E. a. provincialis, we searched and quantified the
target plant species, which were already known in lit-
erature for the E. aurinia complex species, and we ex-
amined these plants recording the presence of egg
batches and larval webs with feeding caterpillars.
G. cruciata, L. caprifolium and S. columbaria
were recorded in the study habitat patches, while G.
lutea and C. leucantha were confirmed as host plants
outside the study area. All individual plants of G. cru-
ciata and C. leucantha were located across meadows
and counted. S. columbaria was sampled along zigzag
transects (70 m2area, 35 m in length, 2 m in width).
For this purpose twenty-seven transects were assessed
in all habitat patches.
The presence of egg batches on host plants was
recorded during twenty-three, variable length, zigzag
transects. The use of variable length transects was
done to allow for the heterogeneity in the distribution
of host plants, especially S. columbaria and differing
areas of sample locations.
In each transect we counted the egg batches
(original and derived from larval webs) on each host
plant to assess the number of egg batches laid in the
studied habitats patches. The original site of egg batches
from larval webs were determined by following the
tracks of the webs spun by the caterpillars during their
transfers from one host plant to another as they fed. The
egg batches were recorded and the age of each batch,
as determined by difference in colour, used to distin-
guish the eggs of different females on the same host
plant. The eggs laid at different times by different fe-
males were easily recognizable by their different colour
(bright yellow, freshly laid and changing from orange
to purplish-brown, after a few days, Fig. 2). A female
can lay two separate egg batches on the same plant and
to avoid over estimation we considered all eggs with
same colour as laid by the same female.
Using the field data we calculated for the all
earlier stages (from eggs to pre-diapause larval stages)
the plant occupancy rate, as the ratio between the
number of plants with egg batches (original and de-
rived from larval webs), and the number of the host
plants examined along any transect.
Larval web survival and host plant phenology. To
detect the role of host plants in larval survival we
monitored the larvae behaviour and the larval webs
during pre-winter and spring phase for three genera-
tions, focusing our attention on the growth character-
istics of the host plants in the study area.The
vegetative status of the host plants and surrounding
vegetation was systematically recorded with photos,
short notes and videos; this was schematically shown
in Fig. 3A-C. We want to emphasize that C. leucantha,
reported in Fig. 3 together with the other host plants,
does not exist in the same place with G. cruciata; C.
leucantha is limited to a rocky slope in a small habitat
patch nearby Villa Camponeschi.
The pre and post-diapause larval webs were
counted at every visit in each locality recording the
presence/absence and documenting location by pho-
tos. For the post-diapause phase we counted both lar-
val webs (IV instar, gregarious stage) and solitary
larvae (V-VI instar). We calculated the survival rate of
the larval webs as the ratio between the number of lar-
val webs surviving the winter and the number of larval
webs recorded before winter diapause.
Post-diapause larval host plant choice. To detect the
larval host plant affinity of E. a. provincialis and the
use of each host plant by V and VI instar larvae, we
observed and counted the larval webs that overwin-
tered and survived and the activities of gregarious
(without webs) and solitary caterpillars in the study
area along zigzag transects in all study habitat patches.
In spring we observed the larval webs marked before
winter and gregarious caterpillars (IV instar) behav-
iour, we then looked at the activity of solitary cater-
pillars. The time for each activity was assessed in
proportion to the percentage of solitary caterpillars
found in a specific activity. The field data were
recorded when the weather was dry and sunny as this
was when larvae were active.
The behaviour of the solitary caterpillars was
126
Host plants and larval survival of
Euphydryas aurinia provincialis
127
Fig. 2. Eggs and larval stages of E. a. provincialis on host plants: A) double deposition on G. cruciata; B) single deposition on
C. leucantha; C) double deposition on the same leaf of S. columbaria; D) pre-diapause nest (II instar larvae) on G. cruciata; E)
pre-diapause nest (I instar larvae) on C. leucantha; F) pre-diapause (II instar larvae) on S. columbaria; G) pre-diapause nest (II
instar larvae) during transfer from S. columbaria to prostrate stems of L. caprifolium; H) and I) behavior of gregarious larvae
of IV instar after winter diapause: When it was sunny, larvae arranged more or less parallel and in a single layer closing ranks
as possible (H); when there was no sun, larvae piled up on each other in a more compact manner possible (almost spherical) (I).
PINZARI - PINZARI - SBORDONI
128
Fig. 3. Vegetation structure and host plants status during during larval stages in E. a. provincialis. A) Egg laying and pre-diapause
period; B) diapause; C) post-diapause period. 1) Lonicera caprifolium; 2) Gentiana cruciata; 3) Scabiosa columbaria; 4) Ce-
phalaria leucantha. This scheme reports all host plants present in the study area but G. cruciata and C. leucantha do not coexist
in any patch.
Host plants and larval survival of
Euphydryas aurinia provincialis
recorded against three behavioural categories (feeding,
moving and dorsal basking). In the feeding category
we recorded which host plant larvae were using.
To assess how the larvae of E. a. provincialis
can switch feeding between different host plants we
carried out feeding tests in laboratory on larvae of
each stage. The feeding tests were carried using 50 lar-
vae from a larval web on Gentiana cruciata collected
in February 2014. During these tests the host plants S.
columbaria, L. caprifolium, C. leucantha and G. lutea
were offered to caterpillars by random plant rotation.
Additionally, Plantago sp., reported in literature as
host plant of E. a. aurinia, and abundant in the study
area, was also tested during feeding experiments.
Observations on larval stages in captivity. To dis-
tinguish the instar of solitary caterpillars of E. a.
provincialis observed in the wild populations we col-
lected a sample of 27 larvae from natural populations
and their growth was monitored up to pupal stage in
captivity.
RESULTS
Host plants. In the study area we recorded three
species that were already cited as host plants for E. au-
rinia species complex in other countries. They were
S. columbaria, L. caprifolium and G. cruciata. A few
kilometres away from the study area C. leucantha and
G. lutea, were also found (see below). The plants of
S. columbaria were abundant in all habitat patches,
but with a patchy distribution (Tabs. 2 and 3).
This is in contrast to the very low density of S.
columbaria in the cultivated and recently ploughed
areas. The plants of G. cruciata were located in both
meadow and scrub in all habitats patches, but they
were uncommon and sparsely distributed (Tab. 4). The
129
Tab. 2. Records on single and multiple egg depositions in the surveys 2012-2013, 2013-2014, 2014-2015 and 2015-2016.
Patches Egg batches per plant Plants with Plants Plant
eggs occupancy
Single (Ns) Double (Nd) Triple (Nt) Egg masses (N) (D=Ns+Nd+Nt) (P) (D/P) %
Gentiana cruciata 4 0 1 0 2 1 1 100
5 1 2 0 5 3 7 43
6 2 0 1 5 3 5 60
7 10 2 1 17 13 14 93
8 4 1 0 6 5 14 36
9 2 1 0 4 3 5 60
10 1 0 0 1 1 2 50
11 2 2 1 9 5 6 83
12 2 1 0 4 3 7 43
13 2 3 0 8 5 9 55
Tot. (%) 26 (43) 13 (22) 3 (15) 61 (100) 42 70 60
Scabiosa columbaria 1 9 0 0 9 9 203 4
2 1 1 0 3 2 288 1
Tot. (%) 10 (83) 1 (17) 0 12 (100) 11 491 2
Cephalaria leucantha 14 5 0 0 5 5 18 28
Tot. (%) 5 (100) 0 0 5 (100) 5 18 28
Habitat patches: 1, 4, Crossroads Santa Maria del Monte – Fonte Brignola; 2, 3, Fonte Perara; 5, 6, 7, 8, 11, along the path between the road to
Santa Maria del Monte and Hairpin curve Mattano-Malepasso; 10, Pian Mattano; 9, 12, 13, Hairpin curve Mattano-Malepasso; (Fig. 1); 14,
Villa Camponeschi (see the location in Pinzari et al., 2010).
PINZARI - PINZARI - SBORDONI
robust growth from of G. cruciata meant that plants
were clearly visible above the general herbaceous veg-
etation during butterfly flight period. In grazed mead-
ows their visibility was increased due to their
unpalatability for livestock (Fig. 3A). In the areas
where adults of E. a. provincialis fly (Pinzari et al.,
2010) G. cruciata was only present in south-faced
meadows above an altitude of about 1000 m. L. capri-
folium was usually observed as a climber on scrub, or
prostrate on the ground in the edge of the meadows,
and occurred across all habitat patches in the study
area (Fig. 3). The plants of C. leucantha (thirty-two
plants) were 2 km from the study area on very steep
rocky slopes on the edge of a short segment of paved
road between the built-up areas of Villa Camponeschi
and Laculo (1000 m). In this xerothermic habitat,
where C. leucantha typically grows, adults of E. a.
provincialis do occasional find this host plant as they
fly between habitat patches. Egg batches and/or larval
webs were found on G. cruciata, S. columbaria, L.
caprifolium and C. leucantha (Tabs. 2-4), whilst no
eggs or larvae were observed on G. lutea in the wild.
Frequent mention is made of Plantago sp. as a
host plant in the scientific and popular literature. In
the study areas the genus Plantago is abundant and
represented by several species. However, no egg
batches or feeding larvae were found on any Plantago
spp. To the untrained eye, confusion is possible with
Melitaea cinxia (Nymphalidae) whose pre-diapause
and post-diapause larval instar could be mistaken for
E. aurinia. In laboratory, larvae of E. a. provincialis
did not eat Plantago spp. plants offered during the
feeding tests.
Eggs and host plant searching behaviour. The flight
period of adults of E. a. provincialis extends from May
to July in the study locality. Once mated, the females
fly slowly through the meadows, nectaring and basking
as well as searching for suitable oviposition sites. When
searching for host plants, mated females performed
very short flights, landing on plants and walking on the
leaves as part of a meticulous and long inspection of the
host plant (about 20 min on an area of not more than 9
cm in radius in one case). Once a plant is deemed suit-
able, the female stops on a selected leaf of the host
plant, perches on the edge of the upper side of the leaf,
and bends her abdomen to lay the eggs on the under-
sides of the lower and large outer leaves (Fig. 2A-C).
Over the period of the study we recorded sixty-
one batches on G. cruciata (single 43%, double 22%
and triple 15%), twelve on S. columbaria (single 83%,
double 17%), and five on C. leucantha (all single
batches) (Tab. 2). No egg batches were found on L.
caprifolium and G. lutea. Many plants were without egg
batches (40% G. cruciata, N=28; 98% S. columbaria,
N=480; 72% C. leucantha, N=13) (Fig. 4). The plant
130
Tab. 3. Records of eggs (or larval webs) on S. columbaria in all localities during the pre and post-diapause period in the surveys
2012-2013, 2013-2014, 2014-2015 and 2015-2016.
Patches Survey Plants (N) Plants with eggs Plant occupancy Plants with Larval web
or pre-diapause (%) post-diapause survival rate
larval webs (N) larval webs (N) (%)
1 2013-14 66 1 1 1 100
1 2014-15 65 2 3 2 100
1 2015-16 385 9 2 * *
2 2012-13 1000 20 2 0 0
3 2012-13 714 2 0.3 0 0
4 2013-14 50 1 2 0 0
4 2015-16 483 5 1 * *
12 2012-13 929 1 0.1 0 0
Tot./Mean 3692 41 1 3 11**
Habitat patches (Fig. 1): 1, 4, Crossroads Santa Maria del Monte – Fonte Brignola; 2,3, Fonte Perara; 12, Hairpin curve Mattano - Malepasso.
(*) The larval web survival rate was not calculated due to the 2015-2016 generation is yet in development. (**) The mean value was calculated
excluding the larval webs of 2015-2016 generation.
Host plants and larval survival of
Euphydryas aurinia provincialis
occupancy rate by egg batches differs in value between
the host plant species (Tab. 2); it was considerably
higher on: G. cruciata (60%, Chi square test, N=561,
χ2=232.216; d.f.=1, P<0.0001) when compared with S.
columbaria (2%) in the study areas; where C. leucantha
and S. columbaria occur together in area 14, C. leucan-
tha is used preferentially (28%, Chi square test, N=45,
χ2=5859; d.f.=1, P=0.0155) compared with S. colum-
baria (0%) in habitat patch 14 (Fig. 1).
Host plant use over three years. From 2013 to 2015
sixteen plants of G. cruciata (81%) were used for
oviposition and out of these 31% were used in two or
three consecutive years (Fig. 5). The data from the
131
Tab. 4. Records of eggs (or larval webs) on G. cruciata and C. leucantha in all localities during the pre and post-diapause period
in the surveys 2012-2013, 2013-2014, 2014-2015 and 2015-2016.
Patches Survey Plants (N) Plants with Plant Plants with Larval web
eggs or occupancy post-diapause survival rate
pre-diapause (%) larval webs (%)
larval webs (N) (N)
Gentiana cruciata 2 2012-13 1 1 100 0 0
3 2012-13 11 6 54 5 83
3 2014-15 11 7 64 - -
4 2015-16 1 1 100 *
5 2014-15 9 5 55 2 50
5 2015-16 7 3 43 *
6 2014-15 5 2 40 1 50
6 2015-16 5 3 60 *
7 2015-16 14 13 93 *
8 2015-16 14 5 36 *
9 2015-16 5 3 60 *
10 2015-16 2 1 50 *
11 2015-16 6 5 83 *
12 2012-13 7 6 86 2 33
12 2013-14 7 3 43 3 100
12 2014-15 7 4 57 1 25
12 2015-16 7 3 43 *
13 2013-14 9 3 33 2 67
13 2014-15 9 1 11 1 100
13 2015-16 9 5 55 *
Tot./Mean 146 80 55 17 45**
Cephalaria leucantha 14 2014-15 32 2 6 0 0
14 2015-16 18 5 28 **
Tot./Mean 50 7 14
Habitat patches: 1, 4, Bivio Santa Maria del Monte – Fonte Brignola; 2, 3, Fonte Perara; 5, 6, 7, 8, 11, along the path between the road to Santa
Maria del Monte and Hairpin curve Mattano-Malepasso; 9, 12, 13, Hairpin curve Mattano-Malepasso; 10, Pian Mattano (Fig. 1); 14, Villa Cam-
poneschi. (-) Lack of data due to the stationing of livestock that blocked the access to locality3 during 2014-15 survey. (*) The larval web
survival rate was not calculated due to the 2015-2016 generation is yet in development; out of eighty plants of G. cruciata with eggs (or pre-
diapause larval webs), only thirty were used in the data analysis; (**) The larval webs survival rate could not be defined for the five cases on
C. leucantha because the 2015-2016 generation is yet in development.
PINZARI - PINZARI - SBORDONI
2012-2013 survey includes records of egg batches on
the same two plants of G. cruciata for four successive
years. The plants of G. cruciata without eggs were lo-
cated in areas, which were in shadow or covered by
vegetation and so not available for females.
Larval host plants during pre-diapause phase. Fe-
males of E. a. provincialis laid the eggs on their host
plants when these plants were in full growth and flow-
ering. After hatching, we recorded larvae mainlyfeed-
ing on leaves of G. cruciata, rosette leaves and cortex
of stems of S. columbaria, and leaves of C. leucantha
and L. caprifolium.
The number of S. columbaria plants with egg
batches, or used in summer by pre-diapause larval
webs, was generally very low compared to the large
number of plants recorded in all habitats patches and
surveys (pre-diapause plant occupancy range: 0.1-
3%, mean value, 1%, Tab. 3). In G. cruciata, virtu-
ally all the available plants carried egg batches (or
pre-diapause larval webs), even if this plant was un-
common at a site (noting that all G. cruciata plants
were sampled). There were far fewer plants of G.
cruciata (range: 1-14 plants) compared to S. colum-
baria (range: 50-1000 plants) and the pre-diapause
G. cruciata occupancy was on average higher (ranges
from 11% to 100%, mean value: 55%, Tab. 4) than
that on S. columbaria (mean value, 1.11%, Tab. 3)
(Chi square test: χ2=1308.134, d.f.=1, N=3838,
P<0.0001). The same pattern is seen in C. leucantha
(Tab. 4) (Chi square test: χ2=2637, d.f.=1, N=77,
P=0.1044).
Furthermore, whilst the larvae will eat any suit-
able host plant, even moving from one plant species
to another when feeding (from G. cruciata to S.
columbaria, from G. cruciata to L. caprifolium, from
S. columbaria to L. caprifolium), in the wild the
change of host plant is a random event caused by the
total consumption of the plant by caterpillars. Larvae
from eggs laid on G. cruciata and C. leucantha almost
always complete their development on the same plant.
In the rare circumstances where the larvae did need to
move this was usually to the more abundant S. colum-
baria. Egg batches that were laid on S. columbaria
rapidly consume the host plant due to its small size
and are forced to move in search for a new food
source; this is usually another plant of S. columbaria
due to the abundance of this host compared to G. cru-
ciata. The use of different host plants and the fact that
all can act as interchangeable resources for larvae of
E. a. provincialis were confirmed by our feeding tests
in laboratory. In fact, larvae (I-VI instar) fed on all
host plants that were offered during trials without
showing any preference for the host species.
Larval host plants during diapause phase.In the
end of summer, when larvae of E. aurinia provincialis
started the winter diapause inside larval webs, the host
plants G. cruciata and C. leucantha dried out, while
L. caprifolium began to lose their leaves as winter ap-
132
Fig. 4. Egg depositions on each host plants focusing on mul-
tiple events.
Fig. 5. Deposition frequencies on the same plant G. cruciata
from generation to generation (2013-2014, 2014-2015 and
2015-2016) in the habitat patches 12 and 13.
Host plants and larval survival of
Euphydryas aurinia provincialis
proached. In contrast, S. columbaria retains a basal
rosette of green leaves through the winter. In all habi-
tats patches we observed overwintering larvae in large
larval webs on G. cruciata and with small larval webs
on other host plants. Additionally, larvae could enter
diapause after spinning larval webs on other plants
that were not host plants, e.g. leaves of grasses, Car-
lina acaulis L., Eryngium sp., Prunus spinosa L. and
Sanguisorba minor Scop.
The larval webs on non-host plants were al-
ways recorded in two circumstances, i) during early
larval stages, when larvae move from a host plant to
another in search of new food and spin webs as part
of dispersal, and ii) after moulting from III to IV instar
when the larvae spin stronger webs in which to hiber-
nate; these webs can be made on any adjacent vegeta-
tion. These observations suggest that the non-host
plants may have been erroneously recorded as food
sources in past studies.
Larval host plants during post-diapause phase.
After the winter, the emergence of larvae from dia-
pause is dependent upon climatic conditions and there-
fore altitude and sun exposure. By February the fourth
instar larvae are actively basking and looking for food.
Each species of host plant differs in when they begin
to grow fresh leaves and develop sufficiently to pro-
vide a viable food source for the larvae of E. a provin-
cialis. In February, G. cruciata and C. leucantha are
still dormant and only beginning to sprout new leaves;
L. caprifolium is still leafless, while S. columbaria has
new shoots on small rosettes and an abundance of
leaves. In early February the surviving larval groups
of IV instar larvae on G. cruciata were still gregarious
but they did not live inside a silk larval web.
From February to April, larvae abandoned the
original larval webs and moved onto nearby plants,
usually S. columbaria plants, that were abundant and
had shoots and leaves that larvae can eat. The rarity
of the small shoots of other host plants is such that we
never found gregarious larvae (IV instar) feeding on
plants other than S. columbaria. During these transfers
the larvae follow silken “guide strands” spun by lead
caterpillars without having to build a larval web.
After emergence from diapause, the larvae
gradually disperse into small groups and can even feed
as solitary larva in IV instar. In March, when L. capri-
folium produces new leaves on prostrate branches, lar-
vae in V instar (all solitary) also started to feed on this
host plant. As growth develops on G. cruciata, the V
and VI instar larvae used all three host plants as food
source (S. columbaria, L. caprifolium and G. cruciata)
in the study area, although the majority still used S.
columbaria. L. caprifolium and G. cruciata were used
less frequently (Chi square test: χ2=22.231, d.f.=2,
P<0.0001). Out of forty-six larvae, 65% were recorded
on S. columbaria, 24% on L. caprifolium and 11% on
G. cruciata (Tab. 5).
During the journey from one plant to another,
the caterpillars often stopped to bask in the sun on a
silky “platform” which is spun on dry leaves on the
ground. Basking enables larvae to metabolise food
more rapidly than air temperatures would suggest they
spend much of their time on non-feeding activities (z-
score test: z=−3.7976, P=0.00014). We observed a
total of one hundred and thirty nine solitary larvae dur-
ing the three seasons 2012-2015: only 33% were in-
volved with feeding behaviour and the rest in
non-feeding types of behaviour (moving, 12% and
dorsal basking, 55%) (Tab. 5).
When it was sunny, larvae arranged themselves
in a single layer, more or less parallel to one another,
on dead leaves (Fig. 2H), closing ranks when possible.
When there was no sun (most of the day and night),
larvae took shelter among the leaves and dried herbs
and piled up on each other in a compact “ball” thus
the contact surfaces between larvae increased as much
as possible and reducing heat loss (Fig. 2I). This sec-
133
Tab. 5. Activities and host plants in solitary larvae (V e VI instar) of E. a. provincialis.
Activity Larvae (n) % Feeding on host plant Larvae (n) %
Feeding 46 33 S. columbaria 30 65
Moving 16 12 L. caprifolium 11 24
Dorsal basking 77 55 G. cruciata 5 11
Tot. 139 100 Tot. 46 100
PINZARI - PINZARI - SBORDONI
ond behaviour, unknown in literature, was displayed
by all gregarious caterpillars of fourth instar.
Larval web survival. Not all larval webs survived
winter diapause and many individuals probably died
(Tabs. 3 and 4). Out of forty-one egg/larval webs on
S. columbaria that were recorded during all surveys
only three overwintered giving a mean larval web sur-
vival rate equal to 11% (Tab. 3). Conversely, the larval
web survival rate on G. cruciata was usually high in
value and varied from 25 to 100% (Tab. 3). We found
a difference between the two host plants in larval sur-
vival rate approaching significance level (Chi square
test: χ2=2871, d.f.=1, N=121, P=0.09).
Gregarious and solitary larval stages. During our
observations on larvae feeding behaviour in February
we recorded groups of IV instar larvae simultaneously
with solitary larvae nearby. In March, all larvae were
solitary and according to Wahlberg (2000) would have
to be already at VI stage, but surprisingly they were
still quite small. By collecting some of these larvae
(N=27) and monitoring their growth up to pupa in cap-
tivity we revealed that twenty-four individuals were
at V instar.
DISCUSSION
Host plants. There are over twenty larval host plants
reported in literature for the Marsh fritillary Eu-
phydryas aurinia across its European range (Tab. 1).
Among host plants, Cephalaria leucantha is men-
tioned as the host plant of the pre-imaginal stages in
E. a. provincialis by Jutzeler (1994) and in E. a. au-
rinia by Mazel (1982, 1984); Gentiana cruciata is
cited alone or together with Scabiosa columbaria or
other plants by several authors (Mazel & Lutran In:
Jutzeler, 1994; Hafner, 2001 com. pers. In: Anthes,
2002; Perru & Sardet, 2005; Sardet & Betremieux,
2006; Lobenstein, 2008; Svitra & Sielezniew, 2010).
Lonicera caprifolium is reported as host plant of all
subspecies of E. aurinia, except provincialis ssp., by
Mazel (1977, in Jutzeler, 1994).
In Central Italy, our observations on E. a.
provincialis show that four species of plant are used
by larvae as host plants: Gentiana cruciata, Scabiosa
columbaria, Lonicera caprifolium and Cephalaria
leucantha. The larvae show no preference between
these plants. These host plants are consumed by cater-
pillars from I to VI instar but the role played by each
host plant in supporting larval growth is different and
depends on vegetative status of each host plant
species. The mean values of plant occupancy showed
that: G. cruciata (55%) and C. leucantha (14%) pre-
vail in the diet of the pre-diapause larval stages (Tab.
4), with S. columbaria providing the major food
source for the post-diapause fourth instar 65% to the
nourishment of the fifth and sixth instar. L. capri-
folium provides a lesser contribution as a food source
for the V-VI instars (24%, Tab. 5).
In laboratory larvae did eat Gentiana lutea,
no eggs or larvae were found using this host plant
in the wild.
Oviposition host species. Females laid their eggs on
three plant species, Gentiana cruciata, Scabiosa
columbaria and Cephalaria leucantha (Tabs. 2-4).
Whenever S. columbaria and G. cruciata coexist, the
pre-diapause plant occupancy on G. cruciata is on av-
erage higher than that on S. columbaria (Tab. 3) al-
though G. cruciata is uncommon at a site (noting that
all G. cruciata plants were sampled) and S. colum-
baria was very abundant. The same pattern was seen
where C. leucantha is together with S. columbaria.
Why do the females lay more egg batches on plants
that are less frequent and numerically rare? In E. au-
rinia, the main stimulus that encourages females to
alight on a plant is primarily visual, as observed in E.
editha (Ehrlich & Hanski, 2004). Therefore, the visi-
bility of plants is crucial in selecting oviposition sites
and depends on local features of the soil morphology,
the vegetation structure, and the physical size of the
host plant present within a habitat patch. This topic
has been studied by several authors and recognized as
related to the presence of eggs (Porter, 1981; 1982;
1992; Anthes et al., 2003; Konvicka et al., 2003;
Fowles & Smith, 2006; Liu et al., 2006; Stefanescu et
al., 2006; Betzholtz et al., 2007; Botham et al., 2011;
Pennekamp et al., 2013).
Our observations on E. a. provincialis have led
us to the same conclusion and show that there was no
preference in the choice of the host plant in this
species and the egg deposition was determined only
by host plants visibility and accessibility to females.
Females of E. a. provincialis mainly laid their eggs on
G. cruciata (and on C. leucantha) that are more easily
accessible than S. columbaria (Fig. 3) although those
plants were much less numerous than those of S.
columbaria (about 1:100). Taking into account the egg
134
Host plants and larval survival of
Euphydryas aurinia provincialis
batches alone, the plant occupancy rate were higher:
60%, G. cruciata, 28%, C. leucantha, 2%, S. colum-
baria, in each habitat patches (Tab. 2). This shows a
distinct preference for the large, clearly accessible
plants of G. cruciata. We want to emphasize that C.
leucantha does not exist in the same place with G. cru-
ciata; C. leucantha is limited to a rocky slope in a
small habitat patch nearby Villa Camponeschi there-
fore it is not ecologically relevant to compare cruciata
with leucantha.
Considering the presence of eggs and larval
webs together, although different factors (i.e. mortal-
ity, starvation, etc.) can affect the conclusions above,
a similar pattern in host plant use is observed (55%,
G. cruciata, 14%, C. leucantha, 1%, S. columbaria
(Tabs. 3 and 4). During the flight period of E. a.
provincialis the vegetative status of the host plants and
surrounding vegetation was systematically recorded.
This could provide an approximate index of the visi-
bility of the host plant.
In all altitudinal zones in the study area G. cru-
ciata is represented by prominent, large-sized individ-
uals that stand out clearly above the surrounding
vegetation. The ground rosettes leaves of S. columbaria
are almost always completely hidden, and therefore in-
accessible to searching females. In north-facing and wet
grassland patches, S. columbaria is very frequent but
patchily distributed and completely hidden by the sur-
rounding tall grasses, with the flower heads of S. colum-
baria concealed by vegetation (Fig. 3A). In these places
there are neither webs nor eggs despite the local high
host plant density. On contrast, the eggs batches are
found on S. columbaria where slope, aspect, sun expo-
sure, humidity and soil, do allow good visibility of host
plants. In these situations the host plants may be larger
in size, or more apparent due to grazing activity.
Some authors suggest that the presence of egg
batches is linked to host plant density and other factors
such as host plant size, morphology and growth habits,
and vegetation structure of the habitat (Anthes et al.,
2003; Fowles & Smith, 2006; Liu et al., 2006; Ste-
fanescu et al., 2006; Betzholtz et al., 2007), or micro-
climatic predictors (sun exposure and topography,
Botham et al., 2011; Pennekamp et al., 2013). Our ob-
servations are not inconsistent with those results. In
fact, the egg and larval occupancy of S. columbaria is
very low in all habitat patches, both where large and
small host plants coexist, and where only small plants
are present.
Some plants of G. cruciata, even if visible and
accessible for females, occur in shady or dense vege-
tation cover. These plants were never used for egg lay-
ing in contrast to plants in open sunny conditions that
held egg batches year after year (Fig. 5). These results
show how the location (sun exposure, microclimate)
of plants of the same species can influence the female
choice of oviposition site. In addition, the exposure to
sunlight within the habitat could shape the female
choice for oviposition plant; for example, large, visible
plants would not be considered for egg laying if they
were in shadow. For this reason there were plants that
were highly visible, but shaded, on which egg batches
were never observed compared to “optimally” placed
plants that were used for oviposition for consecutive
seasons (Fig. 5). The observed pattern of host plant
use is of great relevance to the conservation of the
species E. a. provincialis, with the data showing se-
lection of the same G. cruciata plants year after year.
Accordingly to observations on E. a. aurinia (Porter,
1992; Anthes et al., 2003; Sardet & Betremieux, 2006;
Singer, 2003 In: Stefanescu et al., 2006) females of E.
a. provincialis also lay their eggs on the same plant
where other females have previously laid. The multi-
ple batch placements seem unrelated to host plant
availability. In the study area there were many plants
without egg batches (40%, G. cruciata, N=70; 97%,
S. columbaria, N=396) (Fig. 4). In fact, females
seemed to prefer the plants with egg batches rather
than “empty” ones despite these being available. The
presence of eggs as a stimulant for further egg batches
being added is known in other butterflies (Ehrlich &
Hanski, 2004). In E. a. provincialis the presence of
multiple egg laying events on the same plant could not
be random because during oviposition behaviour the
females make a long and meticulous inspection of the
host plant and then lay their eggs nearby or on the egg
batches laid previously by a different female (Fig. 2C).
If this inspection of host plants was aimed at prevent-
ing the competition for trophic sources, females would
avoid laying on host plants that already held batches.
The hypothesis that the multiple egg laying events on
the same host plant can be due to the lack of host
plants is unsupported. For example, excluding the
plants located in non-optimal conditions for butterfly
activity, a moderate percentage of Gentian plants were
not used by females.
Given the above, why do females oviposit on
small plants and not continue to search for large host
135
PINZARI - PINZARI - SBORDONI
plants? We think that females of E. a. provincialis do
not use size in selecting host plants, but they oviposit
on the first host plant found during their search; then,
they check it, not to evaluate the plant (prerequisite
for a choice whether or not to abandon it) rather to
search for eggs laid by other females before adding
their batch. Although this female behaviour is seem-
ingly not favourable for larvae in terms of competition
for food source, it provides benefits in terms of sur-
vival by creating large numbers within a feeding group
during the post-diapause stages (see below).
Larval host plant choice. Pre-diapause and diapause
phase. The young larvae (I-III instars) have a distinct
behaviour and live in a group inside a silk larval web
feeding on the four above-mentioned plants. The key
point is that, accordingly to the data on egg batches,
large plants were mostly occupied by the pre-diapaus-
ing stages. In the study area, the pre-diapause plant oc-
cupancy on G. cruciata was on average always higher
than that on S. columbaria (respectively, 55% vs. 1%,
Tabs. 3 and 4). This was also true in the habitat patches
nearby Villa Camponeschi where G. cruciata is absent
and C. leucantha takes over as large sized host plant;
here, the plant occupancy on C. leucantha was 14%
and higher than that on S. columbaria (on this plant no
larval web was observed). This suggests that, when
eggs laying occurred on large-sized plants, the young
caterpillars remain aggregated, thanks to the abun-
dance of food provided by plant. These larvae also built
large, robust larval webs, spun around two or three
stems of host plant, that easily withstood the ravages
of winter (Fig. 2D,E). These larval webs may contain
some hundreds of larvae. In contrast, if egg laying oc-
curred on small host plants, such as the rosettes of S.
columbaria, the larvae quickly consumed the whole
plant and then were forced to move onto new food
sources. In these cases, the original feeding group of
larvae dispersed in the meadows and often divided into
small groups, not all of which survived due to starva-
tion (see below in Larval survival paragraph). The sur-
viving larvae made small and less robust larval webs
(Fig. 2F) and these contained a small numbers of indi-
viduals. Only two examples were found where this also
occurred on G. cruciata where larvae consumed the
original host plant and the original gregarious group
dispersed as they searched for a new food source. The
overwintering larval webs were recorded on the four
host plants cited above and also on some Graminacee
and other plants as Carlina acaulis, Eryngium sp.,
Prunus spinose, Sanguisorba minor.
Post-diapause phase. After the diapause pe-
riod, in February, gregarious larvae (IV instar) became
active; they moved on the ground in search of food,
using a silk thread to guide the group of caterpillars.
In February, available food consisted solely of small
but abundant shoots of S. columbaria, with minimal
contribution from small shoots of the few plants of G.
cruciata and C. leucantha (Fig. 3). The plants of L.
caprifolium are, at this time of year, without leaves
and cannot be used as a food source. From March on-
wards, the fifth instar larvae were feeding as individ-
uals with no gregarious behaviour and start to utilise
L. caprifolium as it produces fresh leaves and shoots.
In the study area all three host plants provided food
source for E. a. provincialis caterpillars post-diapause,
but with S. columbaria as main food source (Tab. 5).
Larval survival after winter diapause. It is clear that
the larger host plants have an important role in the larval
development during the pre-diapause period and reduce
winter mortality during the early larval instars and a
high survival through the winter (diapausing larvae).
The survival rate of overwintering larval webs
on G. cruciata was on average higher than on S. colum-
baria, though not reaching significance. This result be-
comes more significant when the differences in web
size are considered; although this rate was impossible
to determine directly without causing irreparable dam-
age to the larval webs, it was easily inferable as the lar-
val webs on G. cruciata and C. leucantha contained
many more larvae (150-200 larvae) than those on the
S. columbaria (20-50 individuals). The better survival
rate on G. cruciata can be related two points: the size
and biomass of G. cruciata and the benefits of large
overwintering webs. The large plant size and abundant
leaves of G. cruciata, enable the first three larval in-
stars to develop on a single host plant. This means that
larvae do not need to disperse in search of food in con-
trast to larvae using small plants as S. columbaria. The
high numbers of larvae on G. cruciata are able to con-
struct large and robust larval webs (a larval web on
gentian: 10-15 cm in length and 4-5 cm in diameter),
which provide protection from climatic pressures
(snow, rain, low temperatures) and predators. The pred-
ators includes true bugs (Hemiptera, Heteroptera) as
Deraeocoris schach (Fabricius, 1781) (Pinzari, 2016)
and Picromerus bidens (L., 1758) that attack the pre-
136
Host plants and larval survival of
Euphydryas aurinia provincialis
diapause larvae, and flies Erycia furibunda (Zetterst-
edt, 1844) (Diptera, Tachinidae) that are parasitoid of
the pre-diapause larvae (Pinzari, in press).
At this point, it is legitimate to ask why E. a.
provincialis continues to egg lay on S. columbaria if it
seems to be advantageous in terms of larval survival
to oviposit on G. cruciata and large plants in general.
Our results show that, on their own, large plants of G.
cruciata do not ensure the survival of E. a. provincialis
larvae as they are unavailable as a food source for post-
diapause larvae. S. columbaria and L. caprifolium are
key sources of food during the post-diapause period.
Additionally, in patches where G. cruciata was absent,
the larval webs survival of E. a. provincialis was ex-
clusively dependent upon S. columbaria and L. capri-
folium, although with low survivorship.
Post-diapause larval activities and thermoregula-
tion. The fact the shoots of small plants were the main
food source for larvae during the post-diapause period
means that they need to move frequently to find food.
This is facilitated by the high density of S. columbaria
plants but also includes opportunities to bask. This
matches Porter’s studies (1982) where larvae need an
optimum body temperature (not less than 30°C) to me-
tabolize food efficiently. When they feed their body
temperature decreases for the ingestion of cold food;
consequently, they need to bask in sunshine to raise
their body temperature above 30°C. In February, the
temperatures reached as low as 0°C and never exceed
10-15°C until April. Consequently, it is easy to under-
stand how, at low temperatures, it is difficult for cater-
pillars to maintain the optimum body temperature for
rapid food digestion and how long periods of dorsal
basking are needed to (especially in the fourth instar)
to facilitate rapid digestion. The results of our obser-
vations on the fifth and sixth solitary caterpillars ac-
tivities do show that around 55% of their time is
devoted to body thermoregulation compared the time
spent moving or feeding (Tab. 5).
In E. a. aurinia, larvae use dorsal basking be-
haviour to heat up in the sun and raising their body to
temperatures above 30°C (35-37°C) (Porter, 1982).
The change of larval habitus from pale-brown in post-
diapause feeding stages to jet-black in post-diapause is
presumably related to individual body thermoregula-
tion. The colour helps both absorbing the solar radia-
tion and limit heat losses. Porter (1982) showed that
the larval body temperature decreases as they feed low
down in the cooler parts of the plant and suggested that
larvae move closer together to reduce the lateral dis-
persion of heat. This behaviour was also observed dur-
ing our field work; when it was sunny, larvae arranged
themselves more or less parallel in tightly aligned
groups and in a single layer (Fig. 2H). Additionally, we
observed another behaviour used by caterpillars for re-
ducing heat loss from their body in absence of sun (in
cloudy days or at sunset). At these times larvae took
shelter among the leaves and dried herbs and piled up
on each other in a tight, almost spherical cluster, thus
the larvae are in close contact with all their neighbours
and this will reduce heat loss from each larva (Fig. 2I).
The basking and clustering behaviours are reinforced
by the female behaviour of adding eggs to existing
batches to create larger groups of larvae. This prefer-
ence to lay eggs on a plant with existing egg batches
will have evolutionary advantage. The aggregation of
individuals is an effective way to promote warming,
and the greater the number of larvae in a group the bet-
ter. Hence, multiple egg deposition on the same host
plant and large larval webs on large host plants give
greatest advantage to survival of larvae.
As spring progresses, the climatic conditions
change and temperatures rise. In these conditions gre-
garious behaviour is not needed as a way of maintain-
ing the temperature body; the need to find new food
sources is now more important than the need to keep
warm by group basking. At this time the fresh shoots
of host plants are soon exhausted as the larvae grow
larger, and larval groups begin to split up to search for
new food sources. By the fifth instar the gregarious
behaviour ceases and larvae become solitary.
CONCLUSIONS
Females of E. a. provincialis oviposit on three
host plants (G. cruciata, S. columbaria, C. leucantha)
without any preference and locate them by using vi-
sual stimuli. More visible and accessible plants (G.
cruciata and C. leucantha) show a higher number of
egg laying events.
Female prefer to lay their eggs on the plants
where other females have previously oviposited as in
E. aurinia aurinia (Porter, 1992; Anthes et al., 2003;
Sardet & Betremieux, 2006; Singer, 2003 In: Ste-
fanescu et al., 2006). In E aurinia, living in larval
groups seems to be evolutionary selected through lay-
ing egg batches. Double and triple egg laying events
on the same plant contribute to form single, large
137
PINZARI - PINZARI - SBORDONI
groups of larvae. This female behaviour can be seen
as disadvantageous because of the high competition
for food sources among young larvae on the same host
plant; however, gregarious behaviour can offer several
benefits to larvae during both pre-diapause period and
post-diapause period.
E. a. provincialis is a polyphagous species in the
study area. The larvae feed on three plants G. cruciata,
L. caprifolium and S. columbaria, and in other patches
nearby our study area they also eat C. leucantha. A fifth
species, G. lutea is eaten under laboratory conditions
but needs further confirmation in the wild. The larvae
show no preference for a specific host plant, readily
moving from one to another one when feeding. All host
plants are important for the larval survival in E. a.
provincialis, supporting their growth and development
during one or more of their larval stages.
The success of E. a. provincialis and its per-
sistence over forty years in the study area may be
helped by the larvae being polyphagous. The prefer-
ential use of different host plants by larvae at each
growth stage ensures the survival of the species.
Larval survival is often the key determinant of
population size and distribution, so understanding the
sources and variation in larval mortality is essential
to understanding and predicting butterfly population
dynamics.
ACKNOWLEDGMENTS
The present study is a part of the projects of the
Osservatorio per la Biodiversità del Lazio (OBL), co-
ordinated by the senior author (VS) at the Department
of Biology, Tor Vergata University of Rome on behalf
of the Assessorato Ambiente e Cooperazione tra i
Popoli della Regione Lazio and intended to provide
guidelines to help monitor species protected under the
Habitat Directive 92/43/CEE (Art. 17, par. 1). We
would like to thank Anna Appolloni and Giuseppe Fa-
rina for their help in the field work; Arch. Valeria Pin-
zari for her graphic support; Dr. Stefano Valente for
the identification of some plants. Finally, we are very
grateful to Dr. Keith Porter (Natural England, UK) for
his helpful comments on the original submission and
language review of the final version.
138
REFERENCES
ANTHES N., 2002 - Lebenszyklus, habitatbindung und Populationsstr Englandtur des Goldenen Scheckenfalters Euphydryas
aurinia Rott. Im Alpenvorland. Diplomarbeit Westfalische Wilhelms-Universität Münster, 87 pp.
ANTHES N., FARTMANN T., HERMANN G., KAULE G., 2003 - Combining larval habitat quality and metapopulation structure, the
key for successful management of pre-alpine Euphydryas aurinia colonies. Journal of Insect Conservation, 7: 175-185.
BALLETTO E., CASSULO L.A., 1995 - Lepidoptera Hesperioidea, Papilionidea. In: MINELLI A., RUFFO S., LAPOSTA S. (eds.),
Checklist delle specie della fauna italiana, 89. Calderini, Bologna.
BALLETTO E., BONELLI S., ZILLI A., 2014a - Lepidotteri. In: GENOVESI P. et al. (eds), Specie e habitat di interesse comunitario
in Italia: distribuzione, stato di conservazione e trend. Rapporti ISPRA.
BALLETTO E., CASSULO L., BONELLI S., 2014b - Annotated Checklist of the Italian Butterflies and Skippers (Papilionoidea, Hes-
periioidea). Zootaxa, 1: 1-114.
BETZHOLTZ P.E., EHRIG A., LINDEBORG M., DINNETZ P., 2007 - Food plant density, patch isolation and vegetation height determine
occurrence in a Swedish metapopulation of the marsh fritillary Euphydryas aurinia (Rottemburg, 1775) (Lepidoptera,
Nymphalidae). Journal of Insect Conservation, 11: 343-350.
BONELLI S., CERRATO C., LOGLISCI N., BALLETTO E., 2011 - Population extinctions in the Italian diurnal Lepidoptera: an analysis
of possible causes. Journal of Insect Conservation, 15: 879-890.
BOTHAM M.S., ASH D., ASPEY N., BOURN N.A.D., CULMAN C.R., ROY D.B., J. SWAIN, ZANNESE A., PYWELL R.F., 2011 - The ef-
fects of habitat fragmentation on niche requirements of the marsh fritillary, Euphydryas aurinia, (Rottemburg, 1775)
on calcareous grasslands in southern England. Journal of Insect Conservation, 15: 269-277.
CASACCI L.P., CERRATO C., BARBERO F., BOSSO L., GHIDOTTI S., PAVETO M., PESCE M., PLAZIO E., PANIZZA G., BALLETTO E.,
VITERBI R., BONELLI S., 2014 - Dispersal and connectivity effects at different altitudes in the Euphydryas aurinia com-
plex. Journal of Insect Conservation, 19(2): 265-277.
EELES P., 2014 - Marsh Fritillary Euphydryas aurinia (Rottemburg) F. hibernica (Birchall) oviposition observations in Dublin,
Ireland in 2013. Entomologist’s Record and Journal of Variation, 126(1): 15-19.
Host plants and larval survival of
Euphydryas aurinia provincialis
EHRLICH P.R., HANSKI I., (eds) 2004 - On the wings of checkerspots: a model system for population biology. Oxford University
Press, Oxford, 371 pp.
FAVRETTO J.P., 2009 - Valeriana officinalis, a new plant host of Euphydryas aurinia (Rottemburg, 1775) (Lep. Nymphalidae).
Oreina, 7: 7-9.
FORD H.D., FORD E.B., 1930 - Fluctuation in numbers, and its influence on variation, in Melitaea aurinia, Rott. (Lepidoptera).
Transactions of the Royal Entomological Society of London, 78(2): 345-352.
FOWLES A.P., SMITH R.G., 2006 - Mapping the habitat quality of patch networks for the marsh fritillary Euphydryas aurinia
(Rottemburg, 1775) (Lepidoptera, Nymphalidae) in Wales. Journal of Insect Conservation, 10(2): 161-177.
HARTIG F., 1968 - Einige neue Lepidopterenrassen und -formen und eine wiederentdeckte Noctuide aus Süditalien. Reichen-
bachia, 12(l): 1-13.
HULA V., KONVICKA M., PAVLICKO A., ZDENEK F., 2004 - Marsh-Fritillary (Euphydryas aurinia) in the Czech Republic: moni-
toring, metapopulation structure, and conservation of an endangered butterfly. Entomologica Fennica, 15: 231-241.
JUNKER M., WAGNER S., GROS P., SCHMITT T., 2010 - Changing demography and dispersal behavior: ecological adaptions in an
alpine butterfly. Oecologia, 164: 971-980.
JUTZELER D., 1994 - Quelques observations de terrain sur l’ecologie d’Eurodryas aurinia ssp. provincialis (Boisduval, 1829)
en Ligurie (Italia) et dans les Alpes-Maritimes (France) (Lepidopteres: Nymphalidae). Bulletin de la Société Ento-
mologique de Mulhouse, 4: 77-81.
KLEMETTI T., WHALBERG N., 1997 - Punakeltaverkkoperhosen (Euphydryas aurinia) ekologia ja populaatiorakenne Suomessa.
Baptria, 22: 87-93.
KONVIČKA M., HULA V., FRIC Z., 2003 - Habitat of pre-hibernating larvae of the endangered butterfly Euphydryas aurinia (Lep-
idoptera: Nymphalidae): what can be learned from vegetation composition and architecture? European Journal of En-
tomology, 100: 313-322.
KUUSSAARI M., VAN NOUHUYS S., HELLMANN J.J., SINGER M.C., 2004 - Larval biology of checkerspots. In: EHRLICH P.R., HANSKI
I. (eds), On the wings of checkerspots. A model system for population biology. Oxford University Press, Oxford, 371,
pp. 138-160.
LIU W., WANG Y., XUR., 2006 - Habitat utilization by ovipositing females and larvae of the Marsh fritillary (Euphydryas aurinia)
in a mosaic of meadows and croplands. Journal of Insect Conservation, 10(4): 351-360.
LOBENSTEIN U., 2008 - Life history of the marsh fritillary Euphydryas aurinia (Rottemburg, 1775) in the Weser region of Lower
Saxony, Germany (Lepidoptera: Nymphalidae). Entomologische Zeitschrift, 118(5): 217-220.
LUCKENS C.J., 1978 - Euphydryas aurinia Rott. in Britain: notes on distribution and life history. Entomologist’s Record and
Journal of Variation, 90(4): 108-112.
MAZEL R., 1982 - Exigences trophiques et evolution dans les genres Euphydrias et Melitaea sensu lato (Lep. Nymphalidae).
Annales de la Société Entomologique de France, 18(2): 211-227.
MAZEL R., 1984 - Trophisme, hybridation et speciation chez Eurodryas aurinia Rottemburg (Lepidoptera, Nymphalidae). Thèse
d’Etat, Académie de Montpellier, Université de Perpignan, 335 pp.
MAZEL R., 2006 - Déterminisme et stratégie de choix des plantes hôtes chez Euphydryas aurinia Rott. (Lepidoptera, Nymphal-
idae). Revue de l’Association Roussillonnaise d’entomologie, 15(2): 78-83.
PARENZAN P., PORCELLI F., 2006 - I macrolepidotteri italiani Fauna. Lepidopterorum Italiae (Macrolepidoptera). Phytophaga,
XV: 1-1051.
PENNEKAMP F., MONTEIRO E., SCHMITT T., 2013 - The larval ecology of the butterfly Euphydryas desfontainii (Lepidoptera:
Nymphalidae) in SW-Portugal: food plant quantity and quality as main predictors of habitat quality. Journal of Insect
Conservation, 17(1): 195-206.
PEŃUELAS J., SARDANS J., STEFANESCU C., PARELLA T., FILELLA I., 2006 - Lonicera implexa leaves bearing naturally laid eggs of
the specialist Herbivore Euphydryas aurinia have dramatically greater concentrations of iridoid glycosides than other
leaves. Journal of Chemical Ecology, 32: 1925-1933.
PERRU S., SARDET E., 2005 - Biologie de la conservation de Maculinea rebeli Hirschke (Lepidoptera, Lycaenidae) et de sa
plante-hote Gentiana cruciata L en region Lorraine (Nord-Est de la France). Linneana Belgica, XX(4): 123-134.
PINZARI M, 2016 - Deraeocoris schach, a new predator of Euphydryas aurinia and other heteropteran feeding habits on caterpillar
web (Heteroptera: Miridae; Lepidoptera: Nymphalidae). Fragmenta entomologica, 48(1): 77-81.
139
PINZARI - PINZARI - SBORDONI
PINZARI M, PINZARI M, ZILLI A., 2010 - Deep lepidopterological exploration of Mt Cagno and surroundings (Central Italy), a
restricted mountain massif and hotspot for butterfly and moth diversity (Lepidoptera). Bollettino dell’Associazione Ro-
mana di Entomologia, 65(1-4): 3-383.
PORTER K., 1981 - The population dynamics of small colonies of the butterfly Euphydryas aurinia. PhD dissertation, Oxford
University, UK.
PORTER K., 1982 - Basking behavior in larvae of the butterfly Euphydryas aurinia. Oikos, 38: 308-312.
PORTER K., 1992 - Eggs and egg-laying. In: DENNIS R.L.H. (ed.), The ecology of butterflies in Britain. Oxford University Press,
pp. 46-72.
PORTER K., ELLIS S., 2011 - Securing viable metapopulations of the Marsh Fritillary butterfly, Euphydryas aurinia, (Lepidoptera:
Nymphalidae) in Northern England. Journal of Insect Conservation, 15: 111-119.
PROLA C., PROVERA P., RACHELI T., SBORDONI V., 1978 - I Macrolepidotteri dell’Appennino Centrale. Parte I. Diurna, Bombyces
e Sphinges. Fragmenta Entomologica, 14: 1-217
SAARINEN K., JANTUNEN J., VALTONEN A., 2005 - Resumed forest grazing restored a population of Euphydryas aurinia (Lepi-
doptera: Nymphalidae) in SE Finland. European Journal of Entomology, 102: 683-690.
SARDET E., BETREMIEUX P.A., 2006 - Distribution et conservation du Damier de la succise (Euphydryas aurinia) en Lorraine
francaise (Lepidoptera, Nymphalidae). Linneana Belgica, 20(5): 163-179.
STEFANESCU C., PEÑUELAS J., SARDANS J., FILELLA I., 2006 - Females of the specialist butterfly Euphydryas aurinia (Lepidoptera:
Nymphalidae: Melitaeini) select the greenest web leaves of Lonicera implexa (Caprifoliaceae) for oviposition. European
Journal of Entomology, 103: 569-574.
SVITRA G., SIELEZNIEV M., 2010 - The first observation of Euphydryas aurinia (Lepidoptera, Nymphalidae) developing on Gen-
tiana cruciata in Lithuania. Polish Journal of Entomology, 79: 195-201.
VAN SWAAY C.A.M., WARREN M.S., 1999 - Red Data Book of European Butterflies (Rhopalocera). Nature and Environment
Series No. 99. Council of Europe, Strasbourg. 260 pp.
VAN SWAAY C., WYNHOFF I., VEROVNIK R., WIEMERS M., LÓPEZ MUNGUIRA M., MAES D., SASIC M., VERSTRAEL T., WARREN M.,
SETTELE J., 2010 - Euphydryas aurinia. In: IUCN 2013. IUCN red list of threatened species. Version 2013.2. www.iuc-
nredlist.org
VERITY R., 1950 - Famiglie Apaturidae e Nymphalidae. In: Le Farfalle diurne d’Italia. Marzocco, Firenze, Vol. IV, Tav. 38-54,
XV-XX, 380 pp.
WAHLBERG N., 2001 - The phylogenetics and biochemistry of host-plant specialization in Melitaeine butterflies (Lepidoptera:
Nymphalidae). Evolution, 55(3): 522-537.
WAHLBERG N., KULLBERG J., HANSKI I., 2001 - Natural history of some Siberian melitaeine butterfly species (Nymphalidae:
Melitaeini) and their parasitoids. Entomologica Fennica, 12: 72-77.
WARREN M.S., 1993 - Observations on the use of honeysuckle (Lonicera periclymenum) as a natural food-plant of the Marsh
Fritillary (Eurodryas aurinia) (Lepidoptera: Nymphalidae) in southern Britain. Entomologist’s Gazette, 44: 241-244.
WARREN M.S., 1994 - The UK status and suspected metapopulation structure of a threatened European butterfly, the marsh frit-
illary Eurodryas aurinia. Biological Conservation, 67: 239-249.
WARREN M.S., 1996 - Euphydryas aurinia. In: VAN HELSDINGEN P.J., WILLEMSE L.P.M., SPEIGHT M.C.D. (eds.), Background in-
formation on invertebrates of the Habitats Directive and the Bern Convention. Part I - Crustacea, Coleoptera and Lep-
idoptera: 121-126. Council of Europe, Strasbourg, Nature and environment, No. 79.
WARREN M.S., MUNGUIRA M.L., FERRIN J., 1994 - Notes on the distribution, habitats and conservation of Eurodryas aurinia
(Rottemburg) (Lepidoptera: Nymphalidae) in Spain. Entomologist’s Gazette, 45: 5-12.
140
