130 Notul. odonatol., Vol. 1, No. 8, pp. 125-140, December 1, 1981 Butterflies as prey for Orthetrum austenti (Kirby) (Anisoptera: Libellulidae) T.B. Larsen IPPF, 18-20 Lower Regent Street, London, SW1Y 4PW, England Abstract — In O. austent 14.21 Sortie after E. malathana con- Lagos, Nigeria, ap- ; physical pears highly specialised as a predator of but- tact observations terflies. Field are recorded, and 14.21 Sortie after A. eponina;physical contact taken a list of Rhopalocera spp. as prey or 14.22 Sortie after E. malathana; physical con- observed under attack is presented. tact Introduction In the untidy and not very interesting bush cum kitchen gardens of the Airport Hotel in I Lagos. Nigeria came across (22 Nov. 1980) male a dragonfly eating a specimen of the butterfly Acraea eponina (Cramer) (Acraei- dae). It was kindly identified as Onthetrum austeni (Kirby) by Mr S.J. Brooks of the British Museum (Natural History). The observations made during this and the recorded below. next few days are Record of the observations on feeding behaviour in a male specimen Fig. I. (Kirby) eating 13.50 Encountered eating Acraea eponina. (Cramer) (Lagos. Nigeria, Photographed (cf. Fig. I) December 1980). 13.51 Contact lost Refound 14.22 Sortie after Pieridae abortive 14.03 on same perch eating a speci- sp.; A. men of Colotis evippe 14.23 Sortie after eponina; no contact 14.24 Junonia 14.07 Sortie after Euchrysops malathana; Sortie for oenone; physical physical contact contact 14.09 Unidentified sortie 14.25 Brief unidentified sortie 14.11 Sortie for E. malathana; abortive 14.25 Sortie after Eurema brigitta; no contact Unidentified sortie 14. If> Sortie after Neptis sp.; physical contact 14.25 14.18 E. abortive 14.26 Sortie after Sortie after malathana; Catopsiliaflorella; no con- 14.19Sortie after A. eponina:physical contact tact 14.21 Brief sortie, false alarm 14.27 Sortie for A. eponina; physical contact Vol. No. December 1981 Notul. odonatol., I, 8, pp. 125-140, 1, 131 14.28 Sortie for J. oenone; no contact between the first and second senes of obser- 14.29 Sortie for J. oenone. physical contact vations, with the dragonfly using the very the I 14.29 Two attempts at same E. malathana; same perch, was uncanny, was momentarily physical contact left wondering whether I had actually cap- 14.30 Unidentified sortie tured the first specimen! 14.33 J. oenone nearly captured; dragonfly I had contact with the first specimen for I bit through leaf in misapprehensionit was 85 minutes during which time physically of this scared it it 4 butterflies. The second prey; my investigation saw eating speci- away; contact lost till 15.10 men made I kill in 40 minutes. Thus it would 15.10 Refound on its perch eating A. eponina appear that on a reasonable daya male of this False alarm 15.13 sortie dragonfly makes a kill every half hour, or 15.14 Sortie for J. terea; physical contact about twenty kills a day. 15.15 Unidentified sortie During 36 minutes of direct observation 15.16 E. and off the made Captured malathana flew to first specimen a total of 28 sorties. a perch 10 m distant The second made 23 sorties in 40 minutes, at 15.25 Dragonfly captured for identification. a time when the weather was less propitious. The total number of sorties made by a male Further observations discussion thus and during a nine hour working day would 1 — On I December 1980 at 17.10 came across seem to be about 400 one in twenty of another male dragonfly of the same species which are successful. sitting at the favourite perch of the first The following points list some additional specimen. It was eating a male Bicyclus doro- observations based on field notes compiled thea (Cramer) (Satyridae); beneath the perch immediatelyafter the observations were made. were the wings of a male Appias epaphia (I) Ofa total of 3 males and I female dragon- The observed the (Cramer) (Pieridae). specimen was more fly in field, all were interested restless than the first and although I met with it exclusively in butterflies (Rhopalocera). from time time insects to duringthe following week, Bees, flies, wasps and other were I had the chance of but evoked In never protracted obser- plentiful no response. the vation. However, pattern appeared gener- particular, small, black-winged Odonata ally similar. — which I sometimes mistook for butter- On 6 December 1980 at about 11.00 I was flies — were never attacked. able to manage another series ofobservations (2) All species of butterfly were attacked for of about 40 minutes from of for a period the irrespective size, except Papilio I observed the same perch. following activ- demodocus Esper, and irrespective of ities. 1 had no the method of which varies from the pencil on me, so degree flight very of detail is less: fast Hesperiidae to the slow and fluttering I unsuccessful sortie for Pelopidas sp.; species such as Eurema (Pieridae). See 1 unsuccessful sortie for for list of Tagiadesflesus; Table I a observed prey. 2 unsuccessful sorties for Coeliades fores- (3) Captured prey would be brought to the tan; perch normally used by the dragonfly, 2 rather unsuccessful sorties for Appias epaphia; but occasionally to some more 1 unsuccessful sortie for A. sylvia; hidden perch. The whole buttefly was I unsuccessful sortie for the which fluttered Catopsilia florella; eaten except wings, Junonia when the I unsuccessful sortie for sophia; to the ground body was eaten. 3 unsuccessful sorties for Zizeeria the knysna: There was no attempt first to cut off I unsuccessful sortie for Euchrysops ma- wings as some other predators are known lathana; to do. 9 unidentified sorties, mostly false alarms; (4) Species of Acraea (Acraeidae) were read- I These successful sortie, capturing Acraea ence- ily eaten. butterflies are aposematic don which was eaten at another perch. and act as models in mimicry complexes. mantis will The similarity of the general pattern However, many praying eat 132 Notul. odonatol., Vol. I, No. 8, pp. 125-140, December I, 1981 There be Table I — Butterfly species (Rhopalocera)as (7) appeared to an optimum range Orthetrum for location of between radius of prey or under attack by austeni prey a 1.5 to 2.5 m of the perch. Butterflies Wingspan moving around under the veryeyes of the bpeC,eSSpecies (mm)(mm) dragonflywere never attacked and those than three more metres away only rarely. Hesperiidae However, prey was sometimes chased for Coeliades forestan (Stoll) 45 a considerable distance. Tagiadesflesus (Fabr.)(Fabr.) 3535 (8) Movement from the perch was usually Baoris restricted sorties of PelopidasPelapidas sp. (? Baoris fatuellusfatueHus to 1.5 to 10 metres, but Hopffer) 28 sometimes a particularly promising butterfly would be pursued for longer Pieridae than that. The altitude varied from *Colotis*Cololis evippe (L.) 4242 ground level to two or three metres how its *Appias epaphiaepaphia(Cramer)(Cramer) 4545 depending on the prey sought A. sylviasylvia (Fabr.) 50 escape. Eurema brigittabrigilla (Stoll)(Stoll) 40 (9) The species is obviously territorial. I than male CatopsiliaCalopsiliaflorella(Fabr.) 55 never saw more one within the of the gardens hotel, though once a Nymphalidae further male was seen in an adjacent Junonia 48 Junonia oenoneoenone(L.) 48 territory occasionally trespassing briefly. *J. sophiasophia (Fabr.) 3535 The area was about 150 x 100 m and 40 rather well J. lereaterea (Dru.)(Dm.) 40 delineated. The two succes- Imelicerta 40 sive males could be found the Neptis sp. ((?melicerta)) usually on perch where the observations were made, SatySatyridaeridae but there were other perches, at least two which used both *BicyclusBicyclus dorothea (Cramer) 40 of were by specimens. At the end of visit I could find my usually AcAcraeidaeraeid ae the male within less than half an hour *Acraea eponina (Cramer) 38 somewhere in the garden. They clearly *A. encedonenredan (L.) 4040 had a good topographical knowledge of their territory, moving from one perch Lycaenidae to another over a distance of 50 m + • EuchrysopsEuchrysops malathanamalalhana without needing to search. (Boisduval) 22 (10) On one occasion I witnessed an amusing Zizeeria knysna (Trimen) 18 sight. The dragonfly thought it had suc- ceeded in capturing a specimen of Juno- • butter- Observed being eaten. — A further 10 nia oenone, a large and powerful It species were attacked. A total of 50 species fly. which managed to escape. chewed the off entire leaf which it plus were seen in garden. the on had settled with increasing frenzy, presumably won- Acraea meal tasted butterflies which are obviously dering why such a likely not protected against insect predation. decidedly vegetarian. I inspected the leaf butterflies Butter- (5) Only flying were taken. later; it was cut. as if with a dull knife, flies which visible and were active on for a length of 2 cm. flowers were ignored. There were at least half a dozen other (6) On most occasions the dragonfly would species of dragonfly active in the garden, all make physical contact with the prey — of which seemed to have their own ecological — oftenaudibly so — and given the speed niche, and none of which were ever seen at- victims kill of in of some of its a rate one tacking a butterfly. One was seen eating a fly. twenty seems very impressive. On the basis of the observations given Notul. Vol. No. December 1981 133 odonatol., I, 8, pp. 125-140, 1, it to above would appear safe conclude that level ofpredation was not too great. However, Orthetrum austeni is other the highly specialised as a at times of the year level of preda- predator of butterflies, at least as far as the tion by this species ofdragonfly could become At Lagos population is concerned.