FLIGHT BEHAVIOUR OF THE WOODLAND BUTTERFLY NEPTIS
RIVULARIS (SCOPOLI, 1763) (NYMPHALIDAE) WITHIN RESIDENTIAL
GARDENS
Karl Kral
Institute of Biology, Department of Zoology, University of Graz, Universitätsplatz
2, a-8010 Graz, austria; karlkral@aon.at; https://orcid.org/0000-0003-4793-0275
Abstract  Natural habitats of woodland butterflies are becoming more and more
reduced because of progressive changes due to human activities. For long-term sur-
vival, the butterflies may be required to adapt to cultural landscapes. We investigated
the woodland butterfly Neptis rivularis (Scopoli, 1763) as a good example of a cul-
tural follower. The focus was on the adaptation of the flight behaviour within a res-
idential garden as a new habitat. The results indicate that N. rivularis is adapted well
to small-scale habitat such as garden areas. Instead of flights dispersed over open
area accompanied by typically extended gliding, the butterfly maintained close con-
tact with shrubs and bushes. The flight paths were always similar. Thus, the butter-
flies circled around shrubs repeatedly and used corridors to reach other gardens.
They adapted their daily flight activity to the temperature of the garden area. It was
evident that the butterflies preferred ornamental spiraea shrubs as targets. From flight
behaviour, it is suggested that the permanent scanning of structures is necessary for
spatial orientation and host-learning. The small-scale flights at certain times of the
day may be a protection against insectivorous birds. The presence of ornamental spi-
raea shrubs is the prerequisite for the acceptance of the new habitat because they can
replace the natural hosts. residential gardens that fit the needs of woodland butter-
flies can contribute to their conservation.
Key WorDS: woodland butterfly, Neptis rivularis, residential garden, hostplant,
Spiraea spp., behavioural adaptation
Izvleček – leTalNo VeDeNJe GoZDNeGa MeTUlJa VelIKeGa KreS-
NIČarJa, NEPTIS RIVULARIS (SCoPolI, 1763) (NyMPHalIDae) V HIŠNIH
VrToVIH
21
ACTA ENTOMOLOGICA SLOVENICA
LJUBLJANA, JUNIJ 2022                    Vol. 30, øt. 1: 21–30
Zaradi človeške dejavnosti se naravni habitati gozdnih metuljev vedno bolj krči-
jo. Dolgoročno se bodo morda metulji morali prilagoditi kulturni krajini. Kot pred-
stavnika takšnih metuljev sem preučeval velikega kresničarja – Neptis rivularis
(Scopoli, 1763), s poudarkom na spremljanju adaptacije letalnega vedenja v hišnih
vrtovih kot novem habitatu. Izsledki kažejo, da se je N. rivularis sposoben prilagoditi
že majhnim vrtnim površinam. Namesto da bi metulji letali na relativno veliki povr-
šini, so raje ohranjali tesni stik z grmičevjem. Poti letenja so bile vedno podobne.
Metulji so večkrat krožili okoli grmovja in se posluževali koridorjev za doseganje
sosednjih vrtov. Svojo dnevno letalno aktivnost so prilagodili temperaturi na območ-
ju vrta. očitno je bilo, da so bili njihov cilj okrasni grmički medvejke (Spiraea spp.).
Izkazalo se je, da je za učenje in prostorsko orientacijo metuljev glede na gostiteljsko
rastlino potrebno trajno pregledovanje (skeniranje) okolice. letalna aktivnost na
majhni površini in ob določenem dnevnem času bi lahko bila učinkovita zaščita pred
žužkojedimi ptiči. Prisotnost okrasnih grmičkov medvejke (Spiraea) je predpogoj za
poselitev novega habitata, saj le-ti lahko nadomestijo gostiteljske rastline v narav-
nem okolju. Hišni vrtovi, ki ustrezajo potrebam gozdnih metuljev, lahko prispevajo
tudi k njihovemu ohranjanju.
KlJUČNe BeSeDe: gozdni metulji, Neptis rivularis, hišni vrtovi, gostiteljska rastlina,
Spiraea spp., vedenjska adaptacija
Introduction
The Hungarian glider Neptis rivularis is a medium-sized butterfly native in the
open woodland of hills and mountains. Here in Styria, we observed N. rivularis
around the hills and mountains of Graz, in Sausal and in regions of the Koralpe
mountains in immediate proximity of rivers in sun-drenched valleys, woodland
clearings, or at woodland edges (see also Habeler, 1965; Kühnert, 1978). N. rivularis
is also present in the Vienna Woods and in the Karawanken mountains in undulating
deciduous forests, especially near brooks or springs in gorges and ditches (Höttinger,
1998; Hassler &Tschinder, 1998; Wieser, 2008). In Slovenia N. rivularis is present
in similar open woodland. The butterfly is most frequently found in the pre-alpine
mountain region like the Škofjeloško hills, the southern slopes of the Pohorje and
Kozjak Mountains and in the Koroška region. It is, however, also found in parts of
the Posavje region, in the lowlands of the ljubljana basin, and Vipava valley (see the
detailed distribution map by Verovnik et al., 2012; see also Čelik, 2014; Verovnik,
2019). The green leaves of the bridewort (Spiraea salicifolia), goat’s beard (Aruncus
dioicus) and meadowsweet (Filipendula ulmaria) are the native hostplants for the
larvae of N. rivularis. However, natural habitats of N. rivularis are becoming more
fragmented and are reduced because of progressive changes (e.g., increased density
of tall trees; increased canopy cover; drained wetlands; loss of hostplants) by human
activities (Fartmann et al., 2013). The increasing popularity of ornamental spiraea
shrubs in residential gardens and urban parks might be a reason for the increasing
immigration of N. rivularis into such artificial environments. The leaves of ornamen-
Acta entomologica slovenica, 30 (1), 2022
22
tal spiraea shrubs seem to provide an adequate food source for the larvae (räuschl,
2002; Straka, 2010). Verovnik et al. (2012) wrote that in recent years N. rivularis has
moved in Slovenia into urban areas, presumably because of the planting of ornamen-
tal spiraea shrubs as potential larval host plant.
Methods
We investigated the woodland butterfly Neptis rivularis (Scopoli, 1763) in the
study area (~ 1500 m2) within residential gardens in Seckau (47o17´N, 14o47´e; ~
850 m above sea level). The village is on a plateau on a sunny side of the Upper Mur
Valley in Styria, austria. In the study area there is a mixture of sunny and shady
green meadows with higher grasses and bare earth patches in between, flower beds
and various herbaceous perennials, trees, bushes, and shrubs, as ornamental spiraea
(six shrubs of Spiraea x vanhouttei, a Belgium hybrid of S. cantoniensis and S. trilo-
bata; four shrubs of S. x cinerea, a hybrid of S. hypericifolia x S. cana; two shrubs
of S. nipponica; three shrubs of Physocarpus opulifolius). Similar planted vegetation
is also present in gardens outside the study area. In each garden, a single-family
house is located. House driveways are paved or asphalted, and terraces are paved or
tiled. Pathways form free corridors also to gardens further away. Figure 1 shows a
Karl Kral: Flight behaviour of the woodland butterfly Neptis rivularis (Scopoli, 1763) (Nymphalidae) within residential gardens
23
Fig. 1: a row of various shrubs along which Neptis rivularis frequently flew
(photo: K. Kral).
row of various shrubs. at one side of the study area to the northwest, managed hay
meadows extend. 
We recorded the flight period of N. rivularis within the study area. The daily
flight activity was estimated by monitoring the number of specimens with the aid of
a digital voice tracer equipped with a timestamp. Flight duration and speed were
measured with an electronic stopwatch. For the latter, straight flights of known dis-
tance along a hedgerow were considered. The butterfly was followed visually from
a raised terrace with an almost all-around view. Where that was not possible, butter-
flies were followed by walking. The local weather station of the ZaMG/orF, which
was based next to the study area (https://wetter.orf.at/steiermark/seckau/) provided
the daily measured data of air temperature (oC) and wind condition (km/h; direction).
We calculated the natural illumination with a lux-meter, whereby the sensor upwards
to the free sky was always directed at the same position within the study area. 
Results
larvae, prepupae, and pupae of N. rivularis experienced a cool april and May in
the year 2021 with ambient temperatures of 4.3±3.5oC (mean ± SD) and 9.1±2.7oC
respectively. June was significantly warmer with 17.8±3.1oC. July temperature was
on average 18.2±2.0oC. adults of N. rivularis were seen for the first time on a sunny
afternoon in the middle of June and the last time towards the end of July. The flight
activity reached a peak during the second half of June (Table 1). N. rivularis speci-
mens with damaged and worn wings were sometimes seen already at the end of June,
but not until July (Fig. 2). The damaged wings did not deter the butterflies from fly-
ing around. 
Table 1: Number of Neptis rivularis in flight at the study area from the middle of
a sunny morning to a sunny noon
Acta entomologica slovenica, 30 (1), 2022
24
2021 Total number rel. frequencyat spiraea shrubs
rel. frequency
at other shrubs
rel. frequency
elsewhere Gender
June 16 1 1.0 - - ♂
June 20 8 0.75 0.25 - ♂♀
June 23 20 0.65 0.15 0.2 ♂♀
June 28 14 0.643 0.214 0.143 ♂♀
July 3 11 0.727 0.182 0.091 ♂♀
July 7 8 0.75 0.25 - ♂♀
July 13 4 0.75 - 0.25 ♀
July 19 2 1.0 - - ♂♀
July 23 3 1.0 - - ♂♀
Karl Kral: Flight behaviour of the woodland butterfly Neptis rivularis (Scopoli, 1763) (Nymphalidae) within residential gardens
25
Fig. 2: Neptis rivularis with completely intact wings resting on Spiraea x cinerea
(June 30, 2021) and with damaged and tattered wings ‘bleached’ by lost scales, rest-
ing on Spiraea x vanhouttei (July 9, 2021) (photos: K. Kral). 
Flight activity of N. rivularis could be observed only during sunny weather. The
measured sky illumination during flight activity was from 140 x 102 lux to 1400 x
102 lux (N = 50). The lowest measured air temperature at which specimens flew was
at 13oC. When the illumination was too low, as under an overcast sky, even optimal
air temperature could not stimulate the butterflies to take flight. In the early morning
(from an hour and a half to two hours after sunrise), even on sunny days no flight
activity could be observed. as Figure 3 shows, highest flight activity was from mid-
morning to noon and flight activity decreased over the afternoon. on the four days
in June when the afternoon temperature reached 30oC in the shade no flight could be
observed.
We observed that specimens, when flying around, always maintained close con-
tact with shrubs, bushes, and trees, often within less than 10 cm distance. The but-
terflies frequently flew around the same shrub. They quickly crossed open areas such
as meadows and paved or asphalted pathways flying close to the ground. otherwise,
they avoided bare features like walls and roofs of buildings. N. rivularis never flew
towards and over the adjacent hay meadows. 
The flight path was irregular in zigzags, turns, and loops. Short gliding phases
interrupted fluttery flight. The flight speed was at air temperatures of 20 to 25oC and
wind speed less than 5 km/h in average 2.7±1.4 msec-1 (mean ± SD, N = 30). The
Acta entomologica slovenica, 30 (1), 2022
26
Fig. 3: Daily flight activity of
Neptis rivularis in the study area
(standard time).
Fig. 4: Duration of flights of
Neptis rivularis in the study area.
slowest flight speed was 1.5 msec-1 and the fastest 5.0 msec-1. N. rivularis maintained
flight also under windy conditions, even with wind gusts of ~30 km/h. Figure 4
shows the number of flights as a function of flight duration (from take-off to land-
ing). For flight durations of less than one minute to two minutes the butterflies usu-
ally were surrounding one or two free-standing spiraea shrubs, whereby they also
performed hovering. longer flight durations were usually measured when specimens
were flying around in the garden area. Moments of rest lasted often only for a split
second to some seconds. For longer flight disruption, the butterfly usually disap-
peared within shrubs like spiraea, forsythia, or lilac where leaves offered protection.
We failed to see any specimen perching on exposed sites, neither on shrubs and trees
nor on the ground. To bridge larger distances N. rivularis preferred to fly close along
hedgerows and fences, which performed the function of corridors (Fig. 1). We did
not observe N. rivularis being attacked and caught by insectivorous birds, which
were present within the study area.
among all shrubs, ornamental spiraea were the primary target of males and
females throughout the entire flight period. We should note that spiraea formed one
third of the shrubs in the study area. The relative frequency of N. rivularis specimens
on spiraea shrubs was in all counts always above 0.6, in most cases above 0.7 (Table
1). We observed specimens at S. x vanhouttei, S. x cinerea and Ph. opulifolius,
although not at S. nipponica. However, S. x vanhouttei was three-quarters of times
more often the target than the other spiraea shrubs. only once could mating
behaviour be observed, and this on the yellow leaves of Ph. opulifolius. In six cases
we saw females laying single eggs on the upper side of green leaves of S. x vanhout-
tei. We found three brown larvae of the first stage (4-5 mm) on freshly nibbled leaves
of S. x vanhouttei at the end of July and at the beginning of august. 
Discussion
The cool april and May in the year 2021 induced a prolonged pupal stage and
was the reason for a delayed flight season of N. rivularis (see räuschl, 2002). This
time lag did not lead to a shift of the flight period into august (Table 1). The early
wing damage probably caused by frequent touch with branches, twigs, or fences
might have resulted from an increased flight activity. But it may also be that the
wings were not protected enough from damage by bad weather, such as thunder-
storms. obviously, oviposition and developmental processes must start on time to
ensure an optimal life cycle. Thus, synchrony with plant growth might be essential
to ensure that larvae have access to fresh leaves of young hostplant shoots before
hibernation (K. Kral, unpubl. obs.). 
an important factor was sufficient sky illumination to trigger flights, which was
only the case under sunny conditions. Interestingly, in the early morning despite
supra-threshold air temperatures and sky illumination, the butterflies remained inac-
tive. Perhaps they were still in an inactive phase due to insufficiently high metabolic
state for courtship, mating, and oviposition. However, it is known that woodland but-
terflies can even fly at suboptimal temperatures (Merckx et al., 2006). a positive
Karl Kral: Flight behaviour of the woodland butterfly Neptis rivularis (Scopoli, 1763) (Nymphalidae) within residential gardens
27
side-effect could be that the butterflies avoided flying during the early morning high
activity period of birds. For example, the house sparrow (Passer domesticus) has the
highest activity in the first hour after sunrise (robbins, 1981; K. Kral, unpubl. obs.).
N. rivularis avoided flying around in the blazing sun with high radiation, presumably
to prevent excessively high heat input. If too much heat is transferred from the heat
storage areas of the dark wing surface through the wing veins to the wing base and
thorax, the flight muscles could be damaged (liao et al., 2019). Thus, the residential
gardens and urban parks can only be an adequate habitat for N. rivularis and other
woodland butterflies when they provide sufficient shade (Kleckova & Klecka, 2016). 
It has to be kept in mind that the dark colour of N. rivularis is an adaptive trait
developed for living primarily in deciduous and mixed woodland and at woodland
edges, often in valleys and along rivers. There, the morning hours can be very cool
when little sun can penetrate the canopy. In the middle of hot, sunny summer days
the sheltering effect under the canopy, and the evaporative cooling effect of tree
leaves, makes the environment by some degrees Celsius cooler than is the case in
open grassland (Morecroft et al., 1998). as mentioned above, enough sky illumina-
tion is necessary for triggering flight activity. Thus, important prerequisites are open
forests and clearings with light providing adequate illumination levels also near the
ground. at this point, it is worth emphasizing again that N. rivularis and other wood-
land butterflies lose their habitats by reduced coppicing activity, leading to an
increase in forest density (Fartmann et al., 2013). 
The N. rivularis erratic fluttery flight with the ability to change the speed within
a relatively large range is much like in other butterflies (Johansson & Henningsson,
2021). However, the insistent narrow proximity to trees, shrubs, bushes and the
ground is remarkable. obviously, the observed flight behaviour is an adaptation to
the small-scale structures of the garden. It seems also to be a protective reaction to
resident birds. apart from the erratic fluttery flight, N. rivularis can also fly straight
forward but then parallel to hedgerows or fences, as known also from other lepi-
dopterans (Coulthard et al., 2016). 
Is the scanning of the spiraea shrubs while in flight a necessary host selection
behaviour with the purpose of learning? and is the purpose of touching the surface
of leaves observed here to learn the shape, structure, odour, and taste as cues for the
right oviposition site? learning may be necessary because cues are different from
those of the familiar natural hosts such as bridewort, goat’s beard, and mead-
owsweet. Host-finding by learning is not at all unusual in butterflies (e.g., Snell-
rood et al., 2011). Finding, with increasing certainty, the spiraea shrubs among all
the other shrubs, and then the increasing preference for distinct spiraea shrubs, as S.
x vanhouttei, might support this idea. a quantitative analysis would be necessary,
however, to confirm that. 
Acknowledgements
Thanks to Dušan Devetak and eddie John for corrections on an early version of
the manuscript. additional thanks to D. Devetak for translating the abstract into
Acta entomologica slovenica, 30 (1), 2022
28
Slovene language. Veronika Kral helped with the field studies. Thanks to the review-
er for the significant contribution to the improvement of the manuscript.
References
Čelik, T., 2014: Supplements to the atlas of butterflies (lepidoptera: rhopalocera)
of Slovenia. Hacquetia, 13: 1-40.
Coulthard, E., McCollin, D. & Littlemore, J., 2016: The use of hedgerows as
flight paths by moths in intensive farmland landscapes. Journal of Insect
Conservation, 20: 345-350. 
Fartmann, T., Müller, C. & Poniatowski, D., 2013: effects of coppicing on but-
terfly communities of woodlands. Biological Conservation, 159: 396-404.
Habeler, H., 1965: Die Großschmetterlinge von Graz und seiner Umgebung (I).
Mitteilungen des naturwissenschaftlichen Vereins für Steiermark, 95: 16 - 76.
Hassler, L. & Tschinder, M., 1998: ein Beitrag zur Schmetterlingsfauna von
Unterkärnten (Insecta: lepidoptera). Carinthia II, 108: 437-451.
Höttinger, H., 1998: Die Tagschmetterlinge der Stadt Wien (lepidoptera: Diurna).
Studie im auftrag des Magistrates der Stadt Wien, Ma 22 – Umweltschutz.
Wien, 82 pp.
Johansson, L.C. & Henningsson, P., 2021: Butterflies fly using efficient propulsive
clap mechanism owing to flexible wings. Journal of the Royal Society Interface,
18: 20200854.
Kleckova, I. & Klecka, J., 2016: Facing the heat: thermoregulation and behaviour
of lowland species of a cold-dwelling butterfly genus, Erebia. PLoS ONE, 11:
e0150393. 
Kühnert, H., 1978: Über die Verbreitung einiger interessanter Tagfalterarten in der
Südsteiermark. Zeitschrift der Arbeitsgemeinschaft Österreichischer
Entomologen, 30: 49-61.
Liao, H., Du, T., Zhang, Y., Shi, L., Huai, X., Zhou, C. & Deng, J., 2019:
Capacity for heat absorption by the wings of the butterfly Tirumala limniace
(Cramer). PeerJ, 7: e6648. 
Merckx, T., Karlsson, B. & Van Dyck, H., 2006: Sex- and landscape-related dif-
ferences in flight ability under suboptimal temperatures in a woodland butterfly.
Functional Ecology, 20: 436-441. 
Morecroft, M.D., Taylor, M.E. & Oliver, H.R., 1998: air and soil microclimates
of deciduous woodland compared to an open site. Agricultural and Forest
Meteorology, 90: 141-156.
Räuschl, G., 2002: Zu Ökologie und artenschutz des Schwarzen Trauerfalters
(Neptis rivularis SCoPolI, 1763) (lepidoptera: Nymphalidae) in Wien-
ottakring, Österreich. Beiträge zur Entomofaunistik, 3: 81-85.
Robbins, C.S., 1981: effect of time of day on bird activity. Studies in Avian Biology,
6: 275-286.
Snell-Rood, E.C., Davidowitz, G. & Papaj, D.R., 2011: reproductive tradeoffs of
learning in a butterfly. Behavioral Ecology, 22: 291-302. 
Karl Kral: Flight behaviour of the woodland butterfly Neptis rivularis (Scopoli, 1763) (Nymphalidae) within residential gardens
29
Straka, U., 2010: Tagfalter in Stadtgärten: Beobachtungen von Tagfaltern im Garten
der Universität für Bodenkultur (BoKU) Wien in den Jahren 2006–2010.
Verhandlungen der Zoologisch-Botanischen-Gesellschaft in Österreich, 147:
31-49.
Verovnik, R., 2019: Prenovljeni seznam dnevnih metuljev (lepidoptera:
Papilionidea) Slovenije. Acta Entomologica Slovenica, 27: 5-15.
Verovnik, R., Rebeušek, F. & Jež, M., 2012: atlas dnevnih metuljev (lepidoptera:
rhopalocera) Slovenije. atlas of butterflies (lepidoptera: rhopalocera) of
Slovenia. Atlas faunae et florae Sloveniae, 3: 1-456.
Wieser, C., 2008: Narzissenwiesen und Biodiversität in den Karawanken –
Schmetterlingsfauna der Golica, lllitschalm und Umgebung. Kärntner
Naturschutzberichte, 12: 55-77.
Received / Prejeto: 24. 8. 2021
Acta entomologica slovenica, 30 (1), 2022
30