top of page

A Guineo-Congolian lowland forest fruit-feeding butterfly assemblage

Adult butterflies from the majority of species from Afrotropical forests can be defined as either nectar or fruit-feeders. The predominant source of fruit at Semuliki is derived from the many species of figs (Ficus spp.) that are found within the forest. Other sources of food that can also supplement the diet of fruit-feeding butterflies can be animal scat, plant sap or honeydew. Butterflies that are attracted to fermented fruit and the volatile fermentation products (Utrio and Eriksson, 1977) as their primary food source contribute to the fruit-feeding butterfly assemblage, represented at Semuliki as a lowland forest fruit-feeding butterfly assemblage.

The number of recorded species making up this lowland forest fruit-feeding butterfly assemblage currently stands at 154 species or just under 30% of the total number of species (545) so far recorded at the park. Principal genera composition of the assemblage are all from the family Nymphalidae with four genera from the subfamily Limenitidinae: Euphaedra (25 spp.), Bebearia (16 spp.), Euriphene (5 spp.) and Cymothoe (16 spp.). Also making up a large component is a genus from the subfamily Satyrinae; Bicyclus (26 spp.) and a genus from the subfamily Charaxinae; Charaxes (40 spp.). A number of species of Charaxes are not attracted to fruit, demonstrated by the lack of captures of species from the former Euxanthe genus; C. crosslei, C. eurinome and C. trajanus with their presence more contained along the forest margins. Other species within these genera also prefer primarily alternative fruit sources so within the context of the assemblage these have not been included in the final total.

The four genera sub-set of research focus (Euphaedra, Bebearia, Euriphene and Cymothoe) comprise 62 species (although four species were not trapped during sampling: B. chloeropsis, Euriphene amaranta, E. doriclea and C. distincta ) and are predominantly composed of forest-restricted or restricted-range species. The optimum habitat preference for these butterfly species is provided by a dense canopy (in Semuliki provided by Cynometra alexandri), usually primary or older growth secondary forest providing a shady forest floor (Euphaedra, Euriphene and Bebearia) or understorey (Cymothoe). Here, within this forest substratum to around a height of around 2m climate conditions are stable with temperatures rarely diverging from +/- 2 °C from the daily mean; a relative humidity continuously greater than 90% and dappled subdued sunlight. 

This assemblage species sub-set will prove more ecologically sensitive to the impacts of environmental and climate change where the increasing intensity and frequency of even short events of extreme flooding or drought are affecting mean temperatures and thus the survivability of these species. Forest disturbance is also more likely to affect species with narrow geographic ranges (restricted range - equatorial) and this group will then also be more impacted by forest degradation, disturbance and edge effects (forest-restricted) where the canopy is opened for more generalist species to occupy. Species sensitive to the impacts of changes in forest conditions can either use the two strategies of thermal buffering (behavioural responses) or thermal tolerance (physiological responses), (Ashe-Jepson et al., 2023). For Afrotropical species there has been limited research on the effect of median temperature changes within the lowest forest stratum. However, using species occurrences (presence/absence) in measuring forest recovery and regeneration and effect of forest size, species within this assemblage sub-set are used regularly as an indicator of forest ecosystem integrity in African forest biodiversity studies.

Euphaedra and Euriphene are suspected of being unable to migrate over non‐forested areas like plantations or savannah (Fermon et al., 2000). This was demonstrated by Birket‐Smith (1970) when investigating the behaviour of several Euphaedra spp. which required a relative humidity above 80% to be active. A lower relative humidity of 40% was tolerated for up to six hours when inactive but in an active state this humidity was lethal after 30 minutes, and a relative humidity of 60‐70% was also seen to cause heliophobia in this genus. While in an Upper Guinean West African forest zone Bossart et al. (2006) found the two genera Bebearia and Euriphene as being the most sensitive to the effects of forest fragmentation.

So, the number of species from within these genera comprise an important assemblage sub-set which could provide an index analysis providing a proxy or surrogate value that can be used as one method to evaluate and monitor patterns of biodiversity during forest restoration, forest degradation or forest monitoring. Quantifying and monitoring the progress of forest restoration by trees and plants can be easily achieved by satellite imagery but invertebrate recolonisation or disappearance is more difficult to quantify. Thus the number of species of the forest-restricted genera sub-set within the fruit-feeding butterfly assemblage can be used as bioindicator models of forest ecological integrity or successful forest regeneration (and conversely forest destruction or degradation) within a pan‐African context, as importantly they contain both abundantly common and rare species. 

The six main genera comprising the assemblage contain species having distributions from west Africa to a general eastern range limit at Kakamega Forest in Kenya, the eastern most remaining remnant of Guineo‐Congolian forest. Species from the genus Cymothoe are confined to primary forests with the exception of Cymothoe caenis which has been found to penetrate more open and secondary forest habitats. They are predominantly mid-level forest butterflies in comparison to the forest floor genera Euphaedra, Euriphene and BebeariaEuriphene are strictly limited to forests in good condition with a full canopy and only a few species are found in more open or severely degraded areas, but they are never recorded in plantations. The genus Bebearia makes up a large component of forest‐floor butterflies and the majority of species are restricted to forests in good condition. They are considered more ecologically robust than Euriphene spp. and some species like B. cocalia are commonly found in secondary forest or more open savannah areas with palms, their host plant. Euphaedra spp. are predominantly limited to forests in good condition and in severely degraded forests (with a good canopy) they are usually limited to only two or three species. The only commonly trapped Euphaedra species is E. medon, although even then it is never abundant. The three most abundant species in the assemblage are all from the genus Bebearia: B. laetitiodes, B. cocalia and B. brunhilda. Individual species data can be found within the Individual Species Data tab. 


Cymothoe confusa male.

Bebearia sophus male.

Images by Lydia Messerschmidt.


Ashe-Jepson, E., Arizala Cobo, S., Basset, Y., Bladon, A. J., Kleckova, I., Laird-Hopkins, B. C., Mcfarlane, A., Sam, K., Savage, A. F., Zamora, A. C., Turner, E. C. & Lamarre, G. P. A., (2023). Tropical butterflies use thermal buffering and thermal tolerance as alternative strategies to cope with temperature increase. Journal of Animal Ecology, 00, 1–12.

Birket‐Smith, J., (1970). The behaviour of Euphaedra in relation to temperature, relative humidity and light (Lep. Rhopalocera). Entomologica Scandinavia 1, 123‐126.

Bossart, J.L., Opuni‐Frimpong, E., Kuudaar, S. & Nkrumah, E., (2006). Richness, abundance, and complementarity of fruit‐feeding butterfly species in relict sacred forests and forest reserves of Ghana. Biodiversity and Conservation 15, 333‐359.

Fermon, H., Waltert, M., Larsen, T.B., Dall’Asta, U. & Muhlenberg, M., (2000). Effects of forest management on diversity and abundance of fruit‐feeding nymphalid butterflies in southeastern, Côte d’Ivoire. Journal of Insect Conservation 4, 173‐189.

Utrio, P. & Eriksson, K., (1977). Volatile fermentation products as attractants for Macrolepidoptera. Annales Zoologici Fennici 14, 98‐104.

bottom of page