The text below is grayed out because it is not intended to be read. It is a necessarily imperfect OCR of the original and is only used by a search engine.
Journal of the Lepidopterists' Society 57(3), 2003, 235-238
TOUGH AFRICAN MODELS AND WEAK MIMICS: NEW HORIZONS IN THE EVOLUTION OF BAD TASTE
P. J. DeVries
Center for Biodiversity Studies, Milwaukee Public Museum, Milwaukee, Wisconsin 53233, USA
ABSTRACT. Mean hindwing toughness was measured experimentally and compared among three sympatric African nymphalid butterflies comprising an aposematic model, its Batcsian mimic, and a palatable, non-mimetic relative of the mimic. The unpalatable model species had the toughest wings and palatable species had the weakest. Implications for assessing butterfly palatability and mimicry are discussed in light of previous work, and a wing toughness spectrum is proposed as a potential correlate of the palatability spectrum.
Additional key words: butterfly mimicry, Amauris albimaculata, Pseudacraea lucretia, Cymothoe herminia.
Insectivorous birds have likely influenced the evolution of butterfly coloration and behaviors by attacking and eating adult butterflies (Poulton 1902, 1908, Carpenter 1932, 1937, 1938, Wourms & Wasserman 1985). Depending on where they fall on the theoretical palatability spectrum, some butterfly species are eaten by birds, while other species are avoided (e.g., Brower 1958a, b, Turner 1984, Turner & Speed 1999). Generally distasteful butterflies minimize predation by advertising noxious qualities with conspicuous color patterns and a slow flight, while palatable ones use cryptic coloration and rapid flight to evade predators (Fisher 1958, Chai 1986, 1996, Chai & Srygley 1990, Pinheiro 1996). Still other palatable butterflies diminish predation by mimicking distasteful species. The elegance of mimicry stems from the fact that mimics may show strong phenotypic and behavioral resemblance to their models, regardless of taxonomic relat-edness among the species involved (Fisher 1958, Turner 1987, Srygley 1994, Joron et al. 2001).
The evolution of warning coloration and mimicry requires differential survival of some individual butterflies following attacks and tasting by predators, and that the experience be memorable to predators (Fisher 1958). For example, the bodies of aposematic and unpalatable Danainae are well known to be more resilient to damage from bird attacks than cryptic and palatable Satyrinae (Poulton 1908, Carpenter 1942, Chai 1996, Pinheiro 1996). Here natural selection seems to have favored aposematic phenotypes that are resistant to handling by predators, and at the same time allowed for continued advertising of the unpalatable phenotypes. In sum, body toughness in butterflies appears to be correlated with unpalatability.
Recent experimental work extends our understanding of unpalatable traits in butterflies by showing that wings of aposematic African danaine and acraeine species are significantly tougher than those of cryptic, palatable nymphalines and satyrines (DeVries 2002). The study suggested that, in addition to body resilience, relative wing toughness may be correlated
with palatability, and that the spectrum of butterfly wing toughness needs to be documented more broadly. Accordingly this report explores palatability and toughness in a different light by asking whether African models are tougher than their mimics. To do so differential wing toughness was estimated among three sympatric nymphalid butterflies that represent an unpalatable model, a Batesian mimic, and a palatable, non-mimic.
Materials and Methods
The study was conducted from 12-25 August 2001 in western Uganda at the Kibale Forest field station that forms part of the 766 km2 Kibale National Park (0°13' to 0°41'N; 30°19' to 30°32'E) adjacent to the western arm of Africa's Rift Valley. The park lies between altitudes 1110 m in the south and 1590 m in the north. Classified as a moist evergreen forest, Kibale Forest has affinities with both montane forest and mixed tropical deciduous forest. The area around the preserve is a matrix of second growth forest, small agricultural plots, associated riparian edges, and has a long history of various human activities, including long-term studies of forest primates (summarized in Struhsaker 1997).
Based on their relative abundance during the study three butterflies were selected to represent palatable or unpalatable species. The trio was formed by a model species, its Batesian mimic, and a cryptic, non-mimetic species that is closely related to the mimic. Palatability and mimetic resemblance were assessed by direct field observations on their color pattern, flight behavior, sympatry, and inference from a detailed literature (Marshall 1902, Swynnerton 1915a, b, Carpenter 1941, Brower 1984, Ackery & Vane-Wright 1984, Turner 1984, Ackery 1988, Larsen 1991). These criteria strongly suggested that Amauris albimaculata Butler (Danainae) is an unpalatable model for the pu-tatively palatable Batesian mimic Pseudacraea lucretia Neave (Nymphalinae), and that Cymothoe herminia Grosse-Smith (Nymphalinae) is a palatable, non-mimetic species closely related to P. lucretia.
236
Journal of the Lepidopterists' Society
45 J V 40 J
^ 35 J
? 20 H
0-3-----.-----------------—----------------------------------------------------
Amauris Pseudacraea Cymothoe
albimaculata lucretia herminia
45 A
40 4 1 35 4
r 304
1-25-1
15 4
io 4
oJ—.--------------------------------------------------------------------------------
unpalatable palatable
Fig. 1. Box plot comparisons of wing tear weights. Each box spans the first to third quartile and the vertical bars extend to the maximum and minimum values of the sample. Within each box the median is shown by the dashed line, and the mean by the solid line. A, Comparison of wing tear weights for species. Sample sizes are as follows: Amauris albimaculata (N = 10), Pseudacraea lucretia (N = 23) and Cymothoe herminia (N = 14). B, Comparison of wing tear weights of all species grouped by palatable and unpalatable categories.
As done in DeVries (2002) an experimental bird-bill was fashioned using a small metal electrical clip with a small plastic weighing dish tied with thread opposite the clips jaws (hereafter, the clip assembly). A butterfly was killed by a pinch to the thorax, then immediately secured in the jaws of a wooden clothes peg attached to a rigid wire suspended from the center post below the legs of a photographic tripod. All individuals were secured with the wings closed in a natural resting position such that the clothes peg gripped all four wings. The clip assembly was then carefully attached to the hind-wing distal margins of the butterfly such that the jaws gripped the wings between veins Cux and 2A. This position closely approximates that of beak marks made by birds attacking resting butterflies (e.g., Carpenter 1932, 1937, 1938, 1941, Collenette & Talbot, 1928, PJD pers. obs.). The tripod center post was then raised slowly until the weighing dish was freely suspended about 20 mm above a receptacle. Once suspended, tiny ball bearings were slowly added to the dish until the clip as-
sembly tore free of the wing, falling into the receptacle below. The tear in the wing closely simulated wing damage inflicted by birds in the wild (DeVries 2002). The clip assembly and ball bearing weights were then weighed to the nearest 0.001 g on a model PB53 Met-tler-Toledo™ electronic balance. This weight established the force necessary to tear the clip assembly free of the hindwings, and provided a measure of relative wing toughness for each individual specimen.
Individual butterflies that had any wing damage or faded wing-patterns due to old age were not used. This avoided potential effects of wing condition on measures of wing-length or relative wing toughness. To estimate body size by species the distance from base to apex of one wing was measured with dial calipers to the nearest 0.1 mm for all individual specimens.
Differences in wing tear weights and forewing lengths among species were evaluated using a one-way AN OVA. The potential relationship between tear weight and wing length was tested for each species using linear regression. Significance levels for mean wing tear-weight and length in paired comparisons were adjusted for non-independence using the sequential Bonferroni-Dunn method (Rice 1989). Wing tear weights were evaluated using a one-way AN OVA for model, mimic and non-mimetic species, and for pooled palatable and unpalatable species.
Results
Mean wing tear weights differed significantly among the individual species (F = 35.523, p < 0.001, df = 2), where A. albimaculata had the toughest wings, P. lucretia less tough wings, and C. herminia had the weakest wings (Fig. 1A). Comparison of species pairs showed significant wing tear weight differences between species (Table 1A). As a group, unpalatable butterflies had significantly higher wing tear weights than palatable ones (Fig. 1A, B) (F = 51.135, p < 0.0001, df = 1). Tear-weights also differed among species pairs representing model, mimic and non-mimetic butterflies (Table 1A).
Wing lengths differed among species (F = 5.562, p = 0.007, df = 2), between species (Table IB), and unpalatable butterflies had greater mean wing lengths than palatable ones (F = 5.084, p - 0.029, df = 1). Although the largest species, A. albimaculata, had the highest tear weight (Fig. 1A, Table 1), linear regression showed no significant relationship between wing-length and tear weight among species; all probability values were between 0.8580 and 0.4599, and all R2 values were between 0.004 and 0.044.
Volume 57, Number 3
237
Discussion
Butterflies are not discretely palatable or noxious to predators, but rather they encompass a theoretical palatability spectrum (reviewed in Turner 1984,1987). The palatability spectrum refers to the relative tasti-ness of potential prey that, depending on the species, is potentially distributed from delicious to positively noxious for particular predators. For example, groups of closely related butterflies (e.g., Danainae, Heliconi-inae) may include species that range from those eaten by birds to those that are always rejected because they possess a nasty taste (Turner 1984, Ritland 1991, Chai 1996, Srygley 1994, Pinheiro 1996). The concept of a palatability spectrum has challenged the traditional separation of Batesian and Mullerian mimicry in butterflies, and forces us to consider these discrete mimetic categories in a new light (Rothschild 1971, 1981, Huheey 1988, Turner 1984, 1987, Speed & Turner 1999, Turner & Speed 2001, Joron et al. 2001, Mallet 2001).
Empirical and theoretical work suggests that unpalatable butterflies should evolve physical attributes making them resistant to handling by predators (e.g., Poulton 1908, Carpenter 1938, 1941, 1942, Fisher 1958). By estimating the force necessary to tear wings this report corroborates the hypothesis that wing toughness may be a correlate of unpalatability in butterflies (DeVries 2002). Here the aposematic model (A. albimaculata) had significantly tougher wings than its putative Batesian mimic (P. lucretia) and a palatable non-mimic (C. herminia), and that the mimic had significantly tougher wings than its non-mimetic relative (Fig. 1, Table 1). If predators use wing toughness to help assess butterfly palatability, these observations support the idea that, in addition to sharing behaviors and color patterns with their models, some Batesian mimics may be to some degree unpalatable (e.g., Carpenter & Ford 1933, Rothschild 1971, 1981, Turner 1984, Ritland 1991). Using wing toughness as a metric, the cryptic species, C. herminia, would be the most palatable of the trio examined here. Obviously a larger study comparing many aposematic, mimetic and cryptic butterfly species is needed to help reveal evolutionary correlates and phylogenetic patterns of wing toughness. Nevertheless, in concert with other work (Carpenter 1941, DeVries 2002), the present investigation supports the concept of a wing toughness spectrum that has evolved in parallel with the palatability spectrum.
It seems likely that differential wing toughness is correlated with the category and location of damage marks left by predators on the wings of palatable and
unpalatable nymphalid butterflies. Because their wings are tougher, beak marks (impressions on the wings) should be observed more frequently among unpalatable species whereas wing tears (areas removed from the wing) should be observed with a higher frequency among palatable species than unpalatable ones. This indeed seems to be the case in specimens recovered from nature (e.g., Carpenter 1932, 1937, 1938, 1941, Collenette & Talbot, 1928), and it would be useful to compare predator damage among species that fall along a wing toughness spectrum. Bird attacks are most frequently directed to the hindwing in resting butterflies (Carpenter 1944), and in palatable species distinct patterns at the hindwing margin may function as targets that divert predator attacks away from vital body areas (Blest 1957; Wourms & Wasserman 1985); the attacked butterfly may escape leaving the predator with only a piece of wing. Thus, we might expect to find the location of wing tears to be biased toward the target areas (e.g., eyespots of Satyrinae) in palatable species, and greater variance in location of beak marks in unpalatable species without target areas. As pointed to previously (DeVries 2002), differential wing toughness raises the question as to whether hindwing target areas in palatable species are weaker than the wing areas surrounding them.
Our understanding of butterfly mimicry has depended on continued reassessment of theory in light of empirical observation (e.g., Carpenter & Ford 1933, Fisher 1958, Rothschild 1971, 1981, Benson 1977, Owen 1971, Cuthill & Bennett 1993, DeVries et al. 1999, Joron et al. 1999, Speed & Turner 1999, Turner & Speed 2001). This and a previous study (DeVries 2002) establish a motive for a comparative study on differential wing toughness as an evolutionary corre-
Table 1. A, Wing tear differences among species pairs. B, Wing length differences among species pairs. Bonferroni/Dunn comparisons are significant at p < 0.0167. Abbreviations: * = significant, n.s. = not significant.
A
Mean
wing Critical
Ornnparison tear difference p Significance
albimaculata x herminia 24.433 7.299 <0.0001 *
albimaculata x lucretia 16.897 6.678 <0.0001 *
herminia x lucretia -7.536 5.976 <0.0030 *
B
Mean
wing Critical
Comparison length difference p Significance
albimaculata x herminia 3.327 2.562 0.0023 *
albimaculata x lucretia 1.346 2.343 0.1599 n.s.
herminia X lucretia -1.981 2.097 0.233 n.s.
238
Journal of the Lepidopterists' Society
late among many palatable and distasteful butterflies. They also suggest new ways of assessing the palatabil-ity spectrum among butterflies that have been traditionally considered palatable mimics. Finally, the methods used here provide a means for asking whether model butterflies are tougher than mimics, and if non-mimic butterflies are the weakest of all. By exploring the parallel between the palatability spectrum and wing toughness we may potentially open new horizons in the evolution of bad taste.
Acknowledgments
I am grateful for the student and faculty interactions on the 2001 Tropical Biology Association course at Kibale, and in particular P. Brakefield, T. Clutton-Brock, F. Molleman for field assistance, discussions and enthusiasm. Special thanks are due to Rosie Trevelyan for facilitating my work in Kibale and introducing me to Sparky s place. I thank A. L. Freitas, R. Hill, N. Luebke, R. Lande, N. Martin, C. M. Penz, and J. R. G. Turner for discussions and comments relevant to this manuscript. This study was supported in part by the National Science Foundation (DEB 980679), and is dedicated to the magnificent body of work left behind by the late Tommy Flanagan.
Literature Cited
Ackery, P. R. 1988. Hostplants and classification: a review of nymphalid butterflies. Biol. J. Linn. Soc. 33:95-203.
Ackery, P. R. & R. I. Vane-Wright. 1984. Milkweed butterflies: their cladistics and biology. British Museum (Nat. Hist.), London.
Benson, W. W. 1977. On the supposed spectrum between Batesian and Mullerian mimicry. Evolution 31:454-455.
Blest, A. D. 1957. The function of eyespots in the Lepidoptera. Behaviour 11:209-256.
Brower, J. V. Z. 1958a. Experimental studies of mimicry in North American butterflies part 1. the monarch, Danaus plexippus, and the viceroy, Limenitis archippus archippus. Evolution 12:32-47.
----------. 1958b. Experimental studies of mimicry in North American butterflies part 3. Danaus Berenice and Limenitis archippus floridensis. Evolution 12:273-285.
Bkower, L. P. 1984. Chemical defenses in butterflies. Symp. Roy. Entomol. Soc. Lond. 11:110-134.
Carpenter, G. D. H. 1932. Attacks of birds on butterflies. Trans. Roy. Entomol. Soc. Lond. 81:21-26.
----------. 1937. Further evidence that birds do attack and eat butterflies. Proc. Zool. Soc. Lond. A. 1937:223-247.
----------. 1938. The attacks of birds upon butterflies. Proc. VIII International Ornithological Congress, Oxford. 1934:265-276. —. 1941. The relative frequency of beak-marks on butterflies of different edibility to birds. Proc. Zool. Soc. Lond. A 111:223-230.
----------. 1942. Observations and experiments by the late C. M. F.
Swynerton on wild birds eating butterflies and the preferences shown. Proc. Linn. Soc. Lond. 154:10-46.
Carpenter, G. D. H. & E. B. Ford. 1933. Mimicry. Methuen, London.
Chat, P. 1986. Field observations and feeding experiments on the response of rufous tailed jacamars (Galbula ruficauda) to free-flying butterflies in a tropical rainforest. Biol. J. Linn. Soc. 29:166-189.
----------. 1996. Butterfly visual characteristics and ontogeny of responses to butterflies by a specialized tropical bird. Biol. J. Linn. Soc. 59:37-67. -----& R. B. Srygley. 1990. Predation and the flight, morphology, and temperature of Neotropical rain-forest butterflies. Am. Nat. 135:748-765.
Collenette, C. L. & G. Talbot. 1928. Observations on the bionomics of the Lepidoptera of Matto Grosso, Brazil. Trans. Royal Entomol. Soc. London. 76:392-416.
Cuthill, I. C. & A. T. D. Bennett. 1993. Mimicry and the eye of the beholder. Proc. Roy. Soc. Lond. B 253, 203-204
DeVries, P. ]. 2002. Differential wing toughness in distasteful and palatable butterflies: direct evidence supports unpalatable theory. Biotropica 34:176-181.
---------, R. Lande & D. Murray. 1999. Associations of co-mimetic
ithomiine butterflies on small spatial and temporal scales in a Neotropical rainforest. Biol. J. Linn. Soc. 62:343-364.
Fisher, R. A. 1958. The genetical theory of natural selection. 2nd ed. Dover, New York.
HUHEEY, J. E. 1988. Mathematical models of mimicry. Am. Nat. 131:S22-S41.
Joron, M., I. R. Wynne, G. Lamas & J. Mallet. 2001. Variable selection and the coexistence of the mimetic butterfly Ileliconius numata. Evol. Ecol. 13:731-754.
Larsen, T B. 1991. Butterflies of Kenya and their natural history. Oxford University Press, Oxford.
Mallet, J. B. L. 2001. Causes and consequences of a lack of co-evolution in Mullerian mimicry. Evol. Ecol. 13:777-806.
Marshall, G. A. K. 1902. Five years observations and experiments (1896-1901) on the bionomics of South Africa insects, chiefly mimicry and warning colours. Trans. Entomol. Soc. London 1902:287-584.
Owen, D. F. 1971. Tropical butterflies. Oxford Univ. Press, Oxford.
Pinheiro, C. E. G. 1996. Palatability and escaping ability in Neotropical butterflies: tests with wild kingbirds (Tyrannus melancholicus, Tyrannidae). Biol. J. Linn. Soc. 59:351-365.
Poulton, E. B. 1902. The colours of animals. International Scientific Series 67. Appleton, New York.
----------. 1908. Essays on evolution. Clarendon Press, Oxford.
Rice, VV. R. 1989. Analyzing tables of statistical tests. Evolution 43:223-225.
Ritland, D. B. 1991. Revising a classic butterfly mimicry scenario: demonstration of Mullerian mimicry between Florida viceroys {Limenitis acrhippus floridensis) and queens {Danaus gilippus Berenice). Evolution 45:918-934.
Rothschild, M. 1971. Speculations about mimicry with Henry Ford, pp. 202-223. In Creed, E. R. (ed.), Ecological genetics and evolution. Blackwell Scientific, Oxford.
----------. 1981. The mimicrats must move with the times. Biol. J.
Linn. Soc. 16:21-23.
Speed, M. P. & J. R. G. Turner. 1999. Learning and memory in mimicry: II. Do we understand the mimicry spectrum? Biol. J. Linn. Soc. 67:281-312.
Srygley, R. B. 1994. Locomotor mimicry in butterflies? The associations of positions of centeres of mass among groups of mimetic, unprofitable prey. Phil. Trans. R. Soc. Lond. B 343:145-155.
Struhsaker, T. T. 1997. Ecology of an Africa rainforest: logging in Kibale and the conflict between conservation and exploitation. Univ. Florida Press, Gainesville.
Swynnerton, C. M. F. 1915a. A brief preliminary statement of a few results of five years special testing on the theories of mimicry. Proc. Entomol. Soc. Lond 1915:21-33.
----------. 1915b. Birds in relation to their prey; experiments on
Wood-Hoopoes, Small Hornbill and a Babbler. J. South African Ornithologists' Union, Pretoria 11:32-108.
Turner, J. R. G. 1984. Mimicry: the palatability spectrum and its consequences. Symp. Roy. Entomol. Soc. Lond. 11:141-161.
----------. 1987. The evolutionary dynamics of batesian and muel-
lerian mimicry: similarities and differences. Ecol. Entomol. 12:81-95.
Turner, J. R. G. & M. P. Speed. 2001. How weird can mimicry get? Evol. Ecol. 13:807-827.
Wourms, M. K. & F. Wasserman. 1985. Butterfly wing markings are more advantageous during handling than during the initial strike of an avian predator. Evolution 39:845-851.
Received for publication 23 April 2003; revised and accepted 30 May 2003.