(PDF) Spatial distribution, territoriality and sound production by tropical cryptic butterflies (Hamadryas, Lepidoptera: Nymphal

Rev. Biol. Trop., 46(2): 297-330, 1 998 Spatial distribution, territoriaiity and sound production by tropical cryptic buUerflies (Hamadryas, Lepidoptera: Nymphalidae): implications forthe· '''industrial melanism" debate 2 Julián Monge-Nájera 1, Francisco Hernández María Isabel González 3, Javier Soley 4, José Araya4 and Stefano Zolla 5 Centro de Investigación Académica, UNED, Costa Rica. Mailing address: Biología Tropical, Universidad de Costa Rica, 2060 Costa Rica; fax (506)2075550;

. 2 Unidad de Microscopia Electrónica, 3 Escuela de Estadística, 4 Escuela de Física, Universidad de Costa Rica, 2060 Costa Rica 5 Centro Intemazionale Crocevia, Via Ferraironi 88G, 00172, Roma, ltaly. Received 23-V- 1 997. Corrected 1 9-1I- 1998. Accepted 1 3-1I1- 1 998. Absh:act: Neotropical buttert1ies of the genus Hamadryas, noted by the emission of sound, spend much lime perching on trees and are believed 10 be cryptically pattemed and colored with respecI lo Iree trunks and branches ¡hey use as perching siles, bul ¡he subject had not been studied previously. This paper describes spatial distribution, territoriality ¡¡nd sound production in five species, under natural conditions: Hamadryas amphinome (Lucas, 1 853), H. februa (Godart, 1 824), H. feronia (Fruhstorfer, 1 916), H. glaucollome (Bates, 1 864) and H. guatemalena (Bates, ! 864). Tree characteristics and use by bunerflies were recorded under natural conditions in open habitats (grassland thinly eovered with trees) in Costa Riea and Panama, avoidíng the problems thal affected previous natural selection studies in Biston betularía (the "industrial melanism" moth). Males perched on the trees and used Ihem as courting territories. The butterflies perched more oflen on sorne individual trees, and dia no! use olhers. The general Iree bark ("background") color tended to match wing coloration, while presence of food, position of teees along flight routes, Iree size, bark texture, and lichen cover were nol associated with the frequency of perching on ¡he trees. Mosl individuals Ihat perched in the study sites were males . Species differed in perching height and populations of H. februa perched al the same heights in both countries; H. leronia moves 10 higher perches near day's end. The relative use of branches and trunks is nOI related lO the time of day bul reflects the typical perehing height of each species. The northem side of trees is less used and cardinal side distribution is independent of time of day. Perches exposed to direct sunlight are less used in hot days. Al! species perch with the head downwards. Perching males frequently fly towards orher Hamadryas as well as towards tethel'ed cardboard models. Trees with experimentally removed males were taken by newcomers 32 times more ofien Ihan trees with resident males. Eaeh marked H. feronia male was seen perching on 1 -4 trees daily, without difference between seasons, and each tree used had a minimum daily mean of 1 .5 perching butterflies. Mos! H. leronia interaetions occur from 1 3:00through 1 5 :00 hours and are more fl'equent in tbe fmny season. At night males share perches. Sound emission was studied by using non-destructive experimental methods (N=858) and with a scanníng electron microscope. B oth sexes emit sound and the sound apparatus, located in the forewing, is percussive, no! stridulatory. At the end of the upward wingstroke, the wings are c1apped and modified [-m 1-2 veins meet al a speed of approximately 1 420 mm/s, producing the characteristic c1ícks . Wingbeat frequency of free-fiying individnals is 20-29 Hz. There is sorne wing deformatíon during movement. Clicks lasl a mean of 1 .3 8 ms with mean intervals of 43.74 ms and the component frequencies concentrate around 2.4 kHz, matching Hamadryas heariug capacity and being appropriate for the acoustic conditions of habitat. The swollen Se vein is present exclusively in Hamadryas; has a serpentine strueture inside 298 REVISTA DE B IOLOGIA TROPICAL and probably acts as resonance box. Growth of the sound apparatus may be checked by its effect on flight capacity, physiological costs and ecological reasons. AlI Hamadryas have a membrane, shaped as an elongated cupola, in the costal cell, that acts as ear. A second and smaller ear has fOUT chambers and may detect predatory bats when the insects are perching at nighl. Field observations showed that Hamadryas spp. emit audible c1icks when approached by potential predators, to defend territories from other Hamadryas and in at least one speeies also during eourtship. Severe wing damage, eommon in wild Hamadryas, almost never affects the sectíon with the sound mechanism. A review of Ihe Iiterature shows Ihat more than 50 species of lepidopterans ( 1 1 families) emit sound audible lo humans and suggests that sound mechanisms evolved several times. In general, lepidopteran sound is used basically as a warning to predators and for intraspecific communication. Key words: BehavioT, protective coloration, crypsis, industrial melanism, territoriality, spatial distribution, sound, stridulation, mechanism, wing structure, defense, mating, courtship, communication. Crypsis, a phenomenon In which interpretatíon and suggested that joint flights organisms receive sorne protection from simply are identification attempts by the predators and other enemies because they are pursuing maleo B asically, territoriality has been difficult to distinguish from their background, denied on two grounds: (a) alleged owners do is important in evolutionary theory (Cott 1946, not always return to the territory after aeriaJ Pasteur 1972, Steward 1977 b). B utterflies of interactions and (b) lepidopterans in general the genus Hamadryas spend much time lack harrnful fighting structures (Baker 1972). perching on tree trunks on which they are However, it is now known that there are several believed to be cryptic, but the consequences of perches in a territory (Bitzer & Shaw 1979, this behavior have not been studied (Young 1983); this can explain why sorne authors 1974, Jenkins 1983 ). Despite the fact that the beJieved that males did not always return to the genus has been known for two centuries, there same "territory" (perch, actually). are no studies of tree use; inforrnation is mostly limited to taxonomy and Iífe cycJes (reviews in Territoriality models which do not require Jenkins 1983 and Monge-Nájera 1992). physical aggression are now available (review in Thornhill and Alcock 1983 ) and Davies Spatial distribution and particularly (1978) found evídence that such is the case in vertical stratification has seldom been studied the butterfly Pararge aegeria. Nevertheless, in butterflies (De V ries 1988, Daily et al. Wickman (1983 ) proved with film analysis 1991). De Vries (198 8) lamented the lack of that physical aggression occurs in the species. detailed studies comparing species that are Aggression and wing damage during spira1 phylogeneticalIy c1ose. The relationship of flights is also known to exist in tropical spatial distribution with interactions that butterflies (Pinheiro 1990). appear to be territorial is also very poorly known (Monge-Nájera 1992). Neotropical butterflies of the genus Hamadryas are good subjects for territoriality Territoriality is one of the most debated study because of their large size and the topics in lepidopteran ethology. Darwin (1871) peculiarity of emítting audible sound (Monge­ proposed that the joint "spiral" flights of N ájera 198 8 , 1992, Monge-Nájera and butterflies were part of courtship, and Joy Hernández 1993). They align on trees and the (1902) associated these flights with species studied here have a wing calico pattern reproductive territoriality. that is cryptic on bark (Jenkins 1983). Adults feed on fermented liquíds and usually emit the Contemporarily, authors such as Scott sounds when two or more fly in circles ("spiral (1974) have rejected the territorial flights"), a behavior that has been varíously MONGE-NAJERA et al.: Spatial distribution, territoriality and sound in cryptic butterflies 299 interpreted as defense, territoriality or 1917, Monge-Nájera 1992, Monge-Nájera and courtship (Swinton 1877, Hampson 1892, Hernández 1993). Photographs of organs Darwin 183 9, 187 1, Swihart 1967, Monge­ proposed as the origin of sound show in faet Nájera 1992). rami, struetures associated with copulatíon (Jenkins 1983 , Monge-Nájera 1992). Production of sounds audible to humans has been reported for more than 56 species of In total, seven competing hypotheses have lepidopterans covering the families tried to explaín how the sound is produced in Agaristidae(Laithwaite eral. 1974); Arctiidae Hamadryas, but they were not subjected to (Haskell 196 1, Rothschild & Haskell 1966, experimental tests (Otero 1990, Monge-Nájera Lloyd 1974); Lymantriidae (Haskell 196 1); 1992). Morphological observations which Nymphalídae (Hampson 1892, Busnell 1963, compared "mute" and sound emitting speeies Dumortier 196 3 , Scott 1968 ); Noctuidae of several genera suggested that modified (Darwin 187 1, Hampson 1892, Hebard 1922, forewing veins allowed clicking by percussion Hannemann 1956, Haskell 196 1, Dumortier in Hamadryas (Monge-Nájera 1988 , Monge­ 1963, McCrae 1975, Wakamura 1977, B ailey Nájera & Hernández 1993). 1978, Alcock 1989); Papilionidae (Haskell 196 1, B usneU 1963); Pericopidae (Rothschild Otero ( 1990) described a procedure to and Haskell 1966); Satyridae (Kane 1982); elicit sound production in tethered individuals Sphingidae (Haskell 196 1, Rothschild and (an unclear remark in Seitz 1914 could be Haskell 1966) and Zygaenidae (Rothschild & interpreted to refer to the same procedure) and Haskell 1966). reported that sound was not produced after removal of the r-ml_2 veins, concluding that the In the XIX century, two hypothetical strike of both wings at the end of the upstroke mechanisms were proposed to explain (he was the probable cause of sound as originally evolution of sound in adult lepidopterans: hypothesized by Seitz (l914}. sexual selection and natural selection. In sexual selection (Darwin 187 1, Hampson However, the test (Otero 1990) did not 1892) sound may often be a male characteristic include su�gical controls, and the butterflies (Rothschild and Haskel! 1966). In contrast, may have ceased sound production because of White (1877) suggested natural selection the surgical shock even if the sound apparatus arguing that if sound is defensive it would be was located somewhere eIse. To avoid this more developed in the females because they problem, there was need for a non-destructive are more closely associated with the offspring. procedure that could inactÍvate any hypothetical sound structures intermittently. For the genus Hamadryas, the oIdest report of sound may be that of Langsddorf This paper describes Hamadryas spp. (1805, cited by Darwin 1839), for Hamadryas spatial distribution and its relation to wing februa in B razil. Early authors mentioned coloration under natural conditions, analyses several structures that they thought related to the territorial interpretation of aerial sound emission, but published no detailed interactions and presents field observations descriptions or illustrations (e.g. Swinton about sound, a solution to the experimental 1877, Hampson 1892), with the exception of problem of ablation, a computer analysis of Hampson ( 18 92) who illustrated "hooks and sonic emissions and an ultrastructural vesicles" whose existence was not supported description of sound-associated structures in by later observations (Calvert and Calvert Hamadryas spp. 300 REVISTA DE BIOLOGIA TROPICAL MATERIALS AND METHODS I Study sites: Tree use was studied in open areas (grassland with sorne trees) that previously had been rnoist "forest" in Panama and Costa Rica (Figs. 1-3 ). This habitat was selected because several Hamadryas species are associated with forest clearings, and with increasing deforestation they have become • established in open areas such as cattle pasture, ... where they fly along belts of riparian • vegetation and vegetated fence lines (Young 20m 1974, Jenkins 1983). • H. amphinome Sites 1 and 2 were located less than 100 m from the Panama Canal, in the Soberanía National Park, about 2 km NE of the town of Gamboa, Panama. Site 3 was located at mid way between the towns of TUITÚcares and Cebadilla, Alajuela Province, Costa Rica. Tree species appear in Tables 1 and 2, and spatial arrangements in Figs. 1-3. 20 10 ... I 10m . �. 20m • 10 _e .. • .. a. . . . H·feTonia I .,. Fig. 2. Use oC trees by perching Hamadryas butterflies at e Site 2 (Panama). Explanation in Fig. 1 e H·februa e • At Site 1 the ground was flat and covered by leaf litter and sorne grass. Tree trunks were mainly mottled gray, and were straight and • mostly branched 3 m or more aboye the 10m ground. Most branches were long and grew 20 upwards forming oval canopies. Sorne trees produced fruit during the study periodo The I le ground in Site 2 also was flat and horizontal, .,. but it was covered mainly by grass. Tree trunks 19 I were mottled gray or brownish, and were short l. H·feronia and not straight, often branching at less than 1 m from the ground. Canopies often had the Fig. 1. Use oC trees by perching Hamadryas butterflies at Site 1 (Panama). Tbe map is to scale; the ovals represent shape of inverted triangles because branches trees with no use records; and the bar height (scale at left) grew at angles of nearly 45 degrees. There indicates the total count oC butterflies perching on each were almost no fruits during the study periodo tree during the observation periodo Site 3 was on the side of a hill (about 20 MONGE-NAIERA et al.: Spatial distribution, territoriality and sound in cryptic butterflies 301 degrees inclination) covered b y grass and weeds. Tree trunk:s were mainly brownish or mottled gray and contorted, ofien branching about 1 m aboye the ground. Many branches / .. were long, and canopy shape was sinrilar to that � ¡; N' of Site 2. The site included a small pond, 20 muddy areas, and abundant fallen fruit that was I consumed both by the butterflies and the cattle •• • 10 / ... • that moved through the area almost daily. / .. .. . ¡ -. .. ,�-- í í Site 1 had the tallest trees, and Site 3 the ,1 • •• shortest; trunks were thicker in Site 3 , and / � thinner in Site 2. Only in Site 3 did some of the H. guatemalena trees lose foliage during the dry periodo Site 2 differed from the others in having two patches in which. the canopy was thick enough to produce areas of low light on trunks and the ground. Detailed study of some tree characteristics was done only in Site 3 . Data coUection: Field data were taken • during the early years of a study that began in .... . 1983 . The Panamanian sites were studied 5 .. ..... intensively (37 days within a 1.4 month • e • period), and the Costa Rican site over a 10nger period (30 visits distributed at convenience •• • during a five year period). At each visit, aH H·februa trees were checked with binoculars 2-5 times daily to record presence and location of the butterflies. From a subsample of 1 9 trees; each perched butterfly was qualified visually as " cryptic" or "noncryptic" according to Kettlewell's method. The data pernritted a basic assessment of the relationship between vlsltation frequency and background matching. The method scores a lepidopteran as cryptic ("inconspicuous") if the observer fails to separate it visually from its background at two yards or more, and non­ 5 ,. . . cryptic if he succeeds in doing so (Kettlewell •••• • .... • 1955). The method has "a marked degree of agreement and unifornrity" among observers (KettleweH 1955). Perch height was estimated visually in relation to reference mark:s placed 1 m aboye Fig. 3. Use oC trees by perching Hamadryas butterflies at ground on all trees. Site 3 (Costa Rica). Explanation in Fig. 1 . 302 REVISTA DE BIOLOGIA TROPICAL To identify factors associated wíth perch of black thread. The model was balanced frequency, pertinent variables (listed below) horizontally (approximately one arc per were identified by using an exploratory second) and if there was no reaction (flight multíple regression analysis, and then towards or away from the model) it was rested compared with Kruskal-Wallis ANOVA's. AH on the ground for 5 s until the next attempt. statistical analyses were repeated after Two sets of ten tests each were done and then excluding cases with any possibility of the distance between insect and model was counting an individual more than once (the reduced (range 8-0.5 m) and the test repeated. few cases in which a different result was obtained appear here as "purged data"). To test the hypothesis of territoriality U.e. presence of a resident on the tree reduces the Species characteristics: The butterflies occupancy by newcomers) sorne perching were Hamadryas amphinome (Lucas, 1853), H. males were removed to compare the behavior februa (Godart, 1824), H. feronia (Fruhstorfer, of newcomers with cases in which the resident 1916), H. glauconome (Bates, 1864) and H. was not removed. Removed males were guatemalena (Bates, 1864). Theír wing length is marked through the net to reduce manipulation about 4 cm, except for H. guatemalena (4.5 cm). stress and kept there for 20 min or until a new The five species have similar behaviors, male occupied the tree. Then the net was hostplants (the euphorbiaceous vines opened near the tree base to record the Dalechampia spp.), and wing shapes (Jenkins behavior of the original occupant. 1983, Monge-Nájera 1992). They díffer in color and pattern. In the study areas, H. februa is Hypotheses were tested with non­ mostly brownish, H. glauconome has a large parametric statistical tests. Al! data that were chalky wing area, H. amphinome is bluish, and not statisticalIy different were pooled for this H. feronia and H. guatemalena are brown with report (both species and variables); detailed whíte and light blue markings, the latter two tabIes are included in Monge-Nájera (1988 ). being difficult to distinguish except for theír size (Jenkins 1983, Monge-Nájera 1992). Voucher Field recordings and experiments: Field specimens are in The Natural History Museum, recordings of maje sonic emissions (free and London (No. 1987-269) and in the Invertebrate tethered individuals) were made in Atenas, Museum of the University of Panama, Panama. Alajuela, Costa Rica with a Sony TC-D5M tape recorder and a Sennheiser MKH 416TV 3 To distinguish individual behavior, monodirectionaJ microphone (power supply: butterflies were captured whenever possible Sennheiser MZA 14T). To test the hypothesis with nets and individually numbered with that sound is produced by percussion of the r­ quick drying blue enamel. Others were m'_2 veins, without damaging those veins, a collected with standard traps baited with piece of cotton was intermittently introduced banana pulp and beer and placed in the forest. between them during the upstroke while the tethered butterfly was emitting sound "Interaction" is here defined as "spiral (illustration in final section). Periods without flight of a Hamadryas that was originally the cotton served as controIs. This perching and another Hamadryas that flew in experimental design was selected after front of its perch". prevíous tests had shown that (1) a greater reduction in click intensity resulted from using Experiments : Response to a mode) thicker pieces of cotton and viceversa (these (Tinbergen 1958) was tested with a piece of were not thick enough to inhibít the full white cardboard cut to the size and shape of amplitude of wing movement); a piece that Hamadryas and hung from a stick with a piece measured about 5 mm in thickness when not MONGE-NAJERA el al.: Spatial distribution, teITÍtoriality and sound in cryptic butterflies 303 eompressed was appropriate (seleeted for the thecla (2 males). E. thecla and two specimens experiments deseribed hefe), (2) sound of H. februa from a population which does not reduetion was highest when the eoUon was emit audible sound (Otero 1987 pers. comm.) plaeed between the r-m¡_2 veins and the were from NE Venezuela, the rest are from damping effeet deereased as the eoUon was central Costa Rica. moved towards the tip or base of wing, (3) the animals eontinued clieking after experimental Specimens were fixed in KAAD for at rupture 01' the main forewing vein and both least a week and preserved in 75% ethano!. circular wingbase membranes (illustrated and Scales and debris were eliminated with a brush named "hearing memorane" and "Vogel's and by manual·· or ultrasonic agitation while organ" in Monge-Nájera and Hernández submerging the body (6 min) and the wings (3 1993), (4) normal clieking eontinued if the min) in commercial bleach (approximately eoUon was introdueed among the r-m¡_2 veins 2.5% sodium hipochlorite). After a rinse in during the downstroke, and (5) similar results distilled water, the specimens were dried with were obtained if a piece of synthetic sponge the help of a vacuum evaporator or in the airo was used instead of eotton (eotton was O bservations were done with a scanning pre1'erred because it is easier to mold out). eJectron microscope. The recordings were digítized and analyzed with the program MaeRecorder­ Farallon, with a lineal resolution of 8 bits and RESULTS a sampling speed of 11 kHz. Natural history of Hamadryas perching Analysis oC wing motion and behavior. The butterflies perched or "alighted" ultrastructure : To measure time between on tree trunks and branches. Almost all clicks, noise at 60 Hz and its harmonics were perching individualS were males, and they reduced with a digital high-pass filter centered appeared to use the trees as courting territories around 3 kHz. (Monge-Nájera 1988). They are cryptic on grayish and brownish bark, and on bark Wingbeat 1'requency and velocity 01' wings covered by gray lichens. Individuals entering were measured 1'rom a photograph of a tethered urban areas perch on gray or brown walls of a H. feronia, by illuminating with natural light variety of materials. They normalIy apress and a flash. As a result, a crisp record (flash) of them to the surface and hold their wings in the wing features and a cumulative record of planiform posítion (sensu Scoble 1992, see the moving wing (longer exposure of natural Young 1974). On windy days the wings may light) in known time were recorded flap slightly but that is not visible beyond a simultaneously on the film. Wing speed was few centimeters. On hot sunny days the wings measured by tracing the distance moved and may briefly be held elevated or even In a dividing it by the total exposure time (11250 s). veliform position (sensu Scoble 1992). For the ultrastructural analysis, 97 Perching and tree species . Almost aH individuals of eight species were studied, as individuals that perched in the study sites were follows: Hamadryas februa (5 females, 30 males. Thus, these results and the discussion males), H. feronia (2 females, 5 males), H. apply to them. Considering the number of g uatemalena (l female, 8 males), H. available trees of each species, H. feronia glauconome (1 female, 1 male), H. amphinome perched more frequently than expected on (l female, 1 maJe), S. stelenes (2 females, 24 Luehea seemani in Site 1, and Psidium males), A. fatima (3 females, 11 males) and E. guajava and Luehea sp. in Site 2; H. februa 304 REVISTA DE B IOLOGIA TROPICAL aligned most often on S. purpurea in both The butterflies perched more often on countries; H. amphinome on Mangifera indica, sorne individual trees and did not use others H. guatemalena on Psidium guajava and H. during the study period (Figs. 1 -3). General g lauconome on Hymenaea courbaril and tree trunk ("background") color was associated Bursera simaruba (Table 1 ). significantly with tree use (Kruskal-Wallis p=0.OO30). Relative use by site, and effect oC several factors : More butterflies were counted at Site Perching was more freguent on 1 , followed distantly by Site 2 and Site 3 backgrounds rated as cryptic (N=60 records): (TabJes 1 and 2, Appendix 1 ). in 1 4 trees where perching frequency was low (0-8 records), only 60% of sighted butterflies Sites 1 and 2 were visited by H. were considered "cryptic on the background", amphinome, H. februa and H. feronia and al! in four trees with 1 5-20 records, 72%, and in perched on the trees (Table 1 ). In Site 3 the the most used tree ( 1 1 7 records), 8 1 % (see perching butterflies were H. februa, H. Materials and Methods). glauconome and H. guatemalena (TabIe 3). Butterfly species differed by síte and among On the other hand, at Site 3 (where more themselves in use frequency (Kruskal-Wallis variables relating to the trees were recorded), p=0.01 48 for site and p=O.OOOI for species). presence of food and mud, and position of Hamadryas feronia was the most frequently trees in intersection sites along flight routes, recorded species in sites 1 and 2 and H. tree diameter and height, as well as lichen guatemalena in Site 3 (Tables 2-3, Appendix 1 ). cover, bark texture and bark basal color did not llave significant effects (p>0.47 or higher). TABLE 1 Tree perching by Hamadryas butterflies: tolal number oj trees observed, count of butterflies perching cm each tree, and Ilumber ofIrees oj each species used by the butterflies lo percho in /11'0 Panama sites. Eaeh row corresponds lo aIree species. (For laxonomic authorities see Monge-Nájera 1988) Butterfly species H. .!eronia Hfebrua H. amphinome Site and Iree species Trees Butterflies Trees B utterflies Trees B utterflies Trees observed perching used perching used perching used Site I 17 195 12 16 2 O O Guazuma 1Ilm!folia 7 32 4 2 O O Spondias purpurea 2 35 2 14 O O Luehea seemani 2 1 19 2 O O O O Mangifera indica S 7 3 O O O O Cecropia sp. 2 O O O O Site 2 S4 61 18 O O 28 5 Guazunta 1Ilm!folia 2 O O O O O O Luehea sp. 14 18 7 O O O O Mlll1gijera indica 2 S 2 O O 26 2 Psidium gllajava 3 19 O O O O Palmae 3 1 O O O O Unidenlified 32 16 7 O O 2 2 Means: Sile 1: 1 1.4 perching indívíduals per Iree, Site 2: 1.6 0 per Iree and Site 3 : 1.5 1 per (ree. MONGE-NAJERA et al.: Spatial distribution, territoriality and sound in cryptic buttertlies 305 TAB L E 2 Tree perching by Harnadryas butterjlies: total number of trees observed, count of butterjlies perching on each tree, and number of trees of each species used by the butterjlies to perch, in Costa Rica. Each row corresponds to a tree species B uttertly species H.guatemalena H.glauconome H.februa Tree species Trees B uttertlies Trees B uttertlies Trees Buttertlies Trees observed perching used perching used perching used Site 3 43 37 7 19 7 8 4 Guazuma ulmifolia 6 2 2 O O SpiJndiarputjuf feir 1 · 0 ··e· O O .5 ........ 1 Psidium guajava 6 21 1 1 1 O O Eugenia salamensis 3 O O O O O O Bursera simaruba 12 2 1 7 2 O O Cedrela mexicana 1 O O O O 1 Hymenaea courbaril 9 7 1 1 Rollinia sp. O O O O O O Cordia alliodora O O 2 1 O O Entherolobium ciclocarpum 2 2 1 O O O O Zanthoxylum sp. 1 O O O O O O Casearia aculeata O O O O O O Anacardium excelsum O O O O O O Pithecolobium sp. 1 O O O O O O Tabebuia rosea 2 O O O O O O Persea americana 1 O O O O O O Inga sp. 2 1 1 Vertical distribution: Vertical distribution 1 8%, Site 1 ; 65%, Site 2; 65%, Site 3). The of perches varied significantly. H. guatemalena relative use of branches and trunks is not related and H. februa represent a low-perching group to the time of day (Kendall Tau, p>0.05). (means under 3 m) (Fig. 4); H. feronia and H. glauconome also perch at similar heights Cardinal point distribution: The south (rniddle group, around 3 m); H. amphinome has and the east are more used In Site 1 , there is a the highest perches and the greater range and west preference in Site 2, and the east is more variance (mean 5.4 m). Members of each height used in Site 3 (Kendall Tau, p<O.01, Fig. 6). group do not differ among themselves, while both groups differ among themselves and from The use of each cardinal side of the tree is H. amphinome (Kruskal-Wallis ANOVA and independent of the time of day (Kendall Tau, Tukey's Test, p<0.0 1 ). Populations of H. februa p<O.O I ). perched at the same heights in both countries. No species shows preference for the sunny The perching height of H. feronia is related or shadowed side of trees on overcast days to time of day (Kendall Tau, p<O.O I ). B efore (Kendall Tau, p >0.05); on sunny days, H. 1 700 h, many perch less than 2 m from the feronia perches more frequently on the sunny ground, afier that time, most perch aboye that side (57%) and the other species do the height (Fig. 5). Species from aH heights have opposite (36% on sunny side, Pearson Chi­ underside eyespots and a similar wing length. square, p <0. 0 1 ). Hamadryas perch chiefly on trunks in Site The proportion of Hamadryas on sunlit 1 and more on branches in the other sites perches is independent of the time of day (KendaH Tau, p<0.0 1 , percentage on branches: (Kendall Tau, p >0.05), except for the pooled 306 REVISTA DE mOLOGIA TROPICAL TABLE 3 Tree perching by Hamadryas bulterflies: number (!f observed trees, number (if butterflies perching on each and number �f trees l/sed !'O pereh in Cosía Rica, against occurrence {ltfood and mud Buuerfly species lLgualemalena Hglauconome H·februa Sites ¡¡na Iree species 'frees B ntterflies Tr!:!es Butterflies Trees Butterflies Trees observed perchi!1g ilscd perching used perching used With food 3 21 o O With mua 2 21 I I O o In meeting of trce rows O O O O O O Not in meeting 21 ¡ O O Wíthout mud O O O O O O In meeting of Irce rows O O O O O O O Not in meeting ¡ O O O O O O Without food 40 16 6 18 6 9 5 With mud 7 O O 5 1 In meeting of tree mws 2 O O 5 I No! in meeting 5 O O O O O O Witnout mud 33 16 6 13 5 8 4 In meeting of tree rows O O O O O O O Nol in meeting 33 16 6 I3 5 8 4 Total 43 37 7 19 7 9 5 Wi.tn mud 9 21 1 6 2 In meeting of tree rows 3 O O 5 1 No! in meeting 6 21 ¡ O O Wítho\11 mmí 3¿� 16 6 13 5 i5 4 In meeting of tree rows O O O O O O O No! in meeting 34 16 16 13 5 8 4 Food: Hamadryasfeea 011 iiquids f[om fermenting [mi!. Mud: for Hwnadryasmud is a source of water and possibly also of salts. Meeting of tree rows: sorne trees form natural or artificial rows and Hamadryasfly a10ng lhem. Where two fOWSmeet, Ihe number of passing females is higher and males might concentra!e ¡here. !O 8 6 4 lClg. 5. Perching height of Hamadryas feronia has two 2 significantly difieren! patterns: before and after 1700 h (N=208). Fig. 4. Perching height of Hamadl)'as butterflies: rnnge (vertical Iines), standard deviation (bars) and mean (horizontal lines).Numbers:studysites (see texI) (N=4 36). MONGE-NAJERA el al.: Spatial distribution, tem!Oriality and sound in cryplic butterflies 307 90 N=13 t 40 70 Non-Lepidopteran w 8 50 insects '" 20 � 30 w c.. lO <I"l ..l -< 70 :::> N=5 a 40 ;; i5 20 ¡s "" o z o z � 40 o o c.. �o o c.. � 20 c.. o � c.. Eas! NortlJ West Limit Fig. 6. Cardinal poínt orientatíon of Hamadryas spp. in the three sites (N=343l. ("aH species") data that show a preference for the shadowed side after 1 1 : 00 h (50% before 30 1 1 : 00 h; 67% in shadow after 1 1 : 00 h). 10 z z w -< :r: :r: Posture and seasonaHty: AH species o w w u u 6'" o:: � o:: o:: ¡: f-< W perch with the head downwards (N=523), U W W >- <: Z c.. W c.. <I"l "" ?i o parallel to vertical or slightly inclined trunks; w c<: o o f-< ttl W �¡..; z when perching on very inclined trunks or on ¡: tii z o:: ¡: z <: "" en :::> horizontal branches, they align strictly parallel o:: UJ o:: UJ ¡2 <: t;J o:: to gravity. Important deviations from that i5o :I: f-< :3 a o "" z orientation occur during feeding and z <: z <: <: z copulation (pers. observ.). ¡2 o f-< o :3 ¡¿, :><: ..l � c.. The height of perches is not seasonal � (Kendall Tau, p>0.05) unless purged data are used (slight increase in perch height for rainy Fig. 7. Reactíons of Hamadryas femnia and Hamadryas season, Kendall Tau, p<0. 05). Similarly, there guatemalena lo a variely·of stimuli, including falling is no seasonal difference on use of tree parts, leaves and human observers. Sorne perches were on cardinal onentation and perch illumination (H. roadside trees, so the reaction to vehicles passing in fron! of perches was also recorded. (N=201l. feronia, Kendall Tau, p>0.05). Reactions to moving objects : Hamadryas males generany react when a cause desertion of the tree, more often than a variety of organisms move in front of their change of perch or a flight toward the intruder perches (Fig. 7). A flight toward the newcomer (Fig. 7, Kendall Tau, p < 0.0 1 ) -and return to perch- is the most frequent reaction when another Hamadryas enters the Experiments with cardhoard models: field of view. Other stimuli are ignored or Response to cardboard model varies with the 308 REVISTA DE B IOLOGIA TROPICAL species, both for distance (Kendall Tau, p < TABLE 4 0.0 1 , A=4. 1 -8 m, B=4 m or less) and sequence Reslllts of aerial interactions of two species of of model presentation (Kendall Tau, p < 0.01 ; Hamadryas with members of the same species C= flrst six presentations, D= presentations 7¡h- ("intra-specific"), wifh Hamadryas of a different species 46th). H. guatemalena reacted (i. e. flew toward (Uintra-generic") and with lepidopterans of other genera model instead of ignoring it) more frequently ("inter-generic"). The tree residentfiies towards another that moves in Font of its percho After the aerial when the model was far (A 1 4%, N=43 ; B=3 % , interactÍo¡¡; o¡¡e or bOl/¡ individuals can stay in the free N= 1 30; Pearson's Chi-square, p < 0.0 1 ) and or abandon it (N=32). All under natural conditions. less frequently (C= 1 1 %, N=69; D=4%, N= 1 04; Pearson's Chi-square, p < 0.01 ) at the end of Type of interactíon each session of ten consecutive tests. On the Specíes and result lntra- Intra- Inter- contrary, H. februa reacted only if the model specific generic generic was at short range (A=O%, N=8 ; B =42% , H. gllatemalena N=28 ; Pearson's Chi-square, p<O.O l), Only one stays 4 8 independently of the part of the sequence (C B oth stay 7 2 O Both abandon O O 67%, N=6 ; D=36%, N=30; Pearson's Chí­ square, p<O.Ol). Distance and sequence H. glallcollol1le appeared to be unimportant for H. glauconome Only one stays 4 3 2 (A SO%, N=6 ; B=SO%, N= 1 2 ; C=SO%, N=6; D=SO%, N= 1 2 ; Pearson's Chi-square, p>O.OS for both) but the sample was small. Effed of time on interactions: Each Experimental removai oC resident marked H. feronia male was seen perching on males: When resident males were removed, 1 -4 trees daily, without difference between 82% of the trees were occupied by newcomers seasons (rainy season 1 .93 trees/day, N=29, within 20 min (N=26), but only 2.S% of trees dry season l . 4S trees/day, N=20, Mann were occupied if the resident was present Whitney Wilcoxon, p > O.OS). Each tree used (N=8 0) (Chi-squared test, p<O.OO l). In flve out had a minimum daily mean of 1 .62 residents in of 26 cases, the original resident participated in the rainy season (N= 1 08 ) and 1 .4 1 in the dry noisy aerial interactions with the new season (N=76). There is no difference in occupant, but mostIy it flew to a different tree number of residents between seasons if aH or stayed in the net after it was opened (see used trees are counted (Mann Whitney methods section for details). Manipulated Wilcoxon p > 0.05), but there is a difference males did not recover trees during the between the "preferred" (five most used) trees observation period (1 h after release) and often and the rest ( 1 .9 1 residents/day, N=82 and 1 .2 1 perched with the head pointing upwards. residents/day, N:::: 1 02 , respectively, Mann Whitney Wilcoxon, p < 0.0l). Result of natural interadions: Most spiral flights involve individuals of one species, After interactions (N=8 0), newcomers left but occasional inter-speciflc flights with sound the trees and the residents returned to their (intra-generic) and without sound (extra­ perches in 8 S % of the cases. Twice the generic) were a1so observed. Such encounters newcomer too k the perch and the other are rare and the small sample (Table 4) might individual left. account for the lack of correlation between result (resident abandons or retains perch) and Most H. feronia interactions occur from the taxonomic relation of the participants 1 3:00 through I S:00 hours and are more (Kendall Tau, p > O.OS). frequent in the rainy season (Mann Whitney MONGE-NAJERA ef a{:i Spati�,;JiSiribliti.on,;tel,TÍtoriality ,and soúnd in cryptic bUtterllies Wilcoxon. pc;:O.Ol, Fig; 8). :No sign.s of" J.s .,...- . �-----....- ... ---.., · aggression� �ere seeri &.t , night, when ,males RkINY ' peréh in gr01JPS hangíng with wings closed on SEASON the back from underneath leaves on branch tips. In Site 1 four males usedthe same tree to spend thenight; they ;f]ew tothe leavys in a very specific period,between 18:25 a�d l8:�5 (same,as in c aptive H. f�brua, pers,. pbs.)., Sound-producing mechanism: 'Fhe ", experimental inkoinission of softmaterial @5 2 (cotton) between the r-m1_2 veins during wing aJ' 1.5 clapping significantly reduced the intensity of the click sounds, which were produced � 1.0 normally when the cotton and rod were � removed (pig. 9). This procedure in<licated that thel'-ID1,2 veins are;percussive a,ndallowed us 7 ' , to intermittently silerice' the butterfly at will l1ME Of DAY (\1r) (N=396 claps, �1�3 with cotton and 243 Fig. 8. Aerial interactions/hour bytirrie of.day and season. contr()� Cor two, H. februa, aQ.d N=462, 217 Hamadryas feronia (N=89).· (". ( , , Fig:9.Bx�rimeiiia1 proéedure./� pi� of cotton attached to a wood rod was interrnittentlyifitrodil� betw een �yeJns ,'�Ypothesize�HoprOduce tJleso\UÍd. Sound intensity varied withcotton thickness, and normilfHiéks wete produce<hvÍlen " fue, c9tton v.:,as removed. ' " 3 10 REVI S TA DE BIOLOGIA TROPICAL cotton, 245 controls for two H. femnia). The intervals 43.74±6.05 (32.50-53.20) rns physical analysis of sorne representative tests (N=26). Component c1ick frequencies appears below. concentrate around 2.4 kHz (Fig. 11). Wing movements: Each click was emitted Ultrastructure of wing distal half: The at the end of the wing's upstroke movement. At sound apparatus is located in the mid-frontal the final part of the cycle, both wings become area of the forewings (Fig. 12). A comparison slightly curved longitudinally (Fig. lOA). of sound-producing and silent nymphalids Shortly before the cIap the wings are almost shows sorne clear differences. The transversal straight (Fig. lOB) and meeting begins al the veins of R3+4+5 and Mz (r-ml and ml-m2) welJ ' distal extreme (Fig. 1OC). As the contact area defined only in Hamadryas, are thicker in increases downward the r-m¡.2 veins collide males (Figs. 13-14), with the exception of H. producing the click and then the wings move februa, in which they are thin in both sexes and aparto There is a marked longitudinal bend of the in the Venezuelan subspecies (Fig. 14). In wings slightly aboye the r-m¡.2 veins (Fig. lOD). Hamadryas, the r-m¡.2 veins are devoid of In both upstroke and downstroke the hind scales and are connected to the R+Sc veins in margins of each wing paír appear to move ahead one end, and to the Cu in the other (Fig. 12). of the fore margins (Figs. 10 A-D). It is unknown if c1apping also occurs in nature at the end of the Ultrastrudure oC wing basal hal!: The downstroke, but it can be seen in tethered swolIen Se vein is present exclusively in individuals and no sound is produced. Hamadlyas; it has a serpentine structure inside (Fig. 18). Similar but less developed structures Wingbeat frequency (Hz) measured from occur also in the Cu and A veins. As a result, the photograph was 11 in H. feronia, and the pereussive r-ml.2 veins are connected to a according to the sound analysis (below), 10.80 systern of three tubes-within-tubes that ±3.37 (5.8 1-2 1.05; N=67) in tethered H. resemble a resonance box (Fig. 18). femnia and 23.45 ±2.41 (20.00-28.57; N=26) in a free-flying H. amphínome. Al! Hamadryas have a membrane, shaped as an elongated cupola, in the costal cell, here The ve!ocity of the wing at the apica! spot called MCC (l1}embrane of costal cell, Figs. is 2 100 mm/s (Fig. 10), while the heavier mid 15-16, 18 B ). part of the wing, which contains the percussive veins, moves at a lower speed of 1420 mm/s. Voge]'s organs (Figs. 15-17) are well developed in fore wings of all species and Sound charaderistics: When sorne occur in both sexes (not studied in A. fatima). expected c1icks are missing in a sequence, the This organ (Vogel 1912) occupies the base of distance between produced c1icks preserves Cu and its rear basal part contacts the base of the expected interval (Fig. 11 A). This A, it has a rigid cap (can be broken with a suggests that missing clicks represent cases in needle) mounted on a flexible ringo There are which the wings moved at the normal speed four chambers under the cap (Fig. 17); ink but were not c1apped. injected in them does not reach the adjoining veins, suggesting lack of an open connection. The etTect of the soft material io reducing The cavity beside Vogel's organ is covered by click intensity is obvious not only to the human a large amount of small papillae (Fig. 17). ear bu! also in sooograms (Fig. 11). Pmduction of sound in the field: The Overall, clicks last mean 1.38±SD 0.45 emission of sound during joint spira! flights (min. 0.70-max. 2.40) ms (N:::2 1) and the often is part of a male-male intraspecific . M ONGE-NAJERA et al.: Spatial distribution. terriloriality and sQund in cryptic butterflies 311 Fig. to. Sound is produccd during vigorous wingbeating. Al the fmal pan of the upstroke movemcnt. both wings become slightly cur\'ed longitudinally ( Al. S hortly before ¡he clap lhe wings are aimos! straight (E) and finalIy lhey meet, beginning by the distal exlTeme (C). The r-m]_2 vciDs (lower anow) collide, proc!ucing tbe dick. ancllhe wings move apart. Therc is a markcd longitudil1al bend of ¡he wing slightly ¡¡bove these veins (D). Uppcr arrow marks "white spot" used as ffiovement rcfcrcnce. 312 REVISTA DE BIOLOGIA TROPIC AL 250 250 • A ¡:c¡ B ¡:c¡ 200 � 200 � C/J C/J Z O � Z 150 ¡;.... 150 � SU SU 100 100 ::t: 5o ¡;... "'"'. 50 ::t: 50 ¡;... -. i3: � O.,.,.·.-� -ro�� 0.0 0.01 0.02 0.03 0.04 0.0 0.02 0.04 0.06 0.08 TIME (s) TIME (s) e 2.5 E �¡;.... V 2.0 . U ¡.r.:¡ "'" C/l ¡.r.:¡ 1.5 .f q.¡ �,� .. ... --11 li:fJ¡1�f$l$.+1f4' IOOmS $--!I O J ..� ;:¡ 1.0 b ,...¡ "'" 0.5 � <t: O 0.0 0.10 0.20 0.30 OAO 0.50 IOmS NORMALIZED FREQUENCY Fig. 1 1 . A. B . The experimental intromission of cotlon betwecn the percussive veins decreascs the illtensity of the clicking sounds. This is andible to the human ear and also shows in the sonogram. C. Wben some expected elides are missing in a seqnence, tbe distance between produced clicks preserves tbe expected intervals (T). Apparently the ioseet conlrols tbe emission of sound by chaoging wing movement: lo cliek, the wings are clapped. D. Whel1 2-3 iodividuals il1leracted, the intervals diminished. suggesting tJlal several individuals were simllltaneollsly emitting sound (detail el1larged). E. The peak amplitude of the elicle oeellrs al frequencies of 1 .3-3.8 kHz. Amplilllde in arbitral)' units for all figures. For lhe amplitude spectra: aboye: signal, below: noise. interaction (Kendall Tau, p < O. 0 1, Fígs. 19 apparatus exists in both sexes, but may be and 20). ExceptionalIy, we heard sound when slightly larger in males (Figs. 13 and 14) we approached the butterflies as well as when they interacted with Satyrinae, apparently in Although severe wing damage is not rare agonístic encounters for food (fallen fruit) as in wild Hamad¡�vas males, the part of the well as in the courtship of H. feronia. Females wings that contains the sound system was of this species can eITÚt sound when captured intact in al! of the hundreds of specimens (JMN pers. observo 1995). The sound which we have seen (Fig. 2 1). MONGE-NAJERA el al. : Spatial distríbution, tenitoriality and sound in cryptic butter11ies 313 Fig, 1 2 . The percussive part af the sound apparatus is lacated in the miel frontal area of (he farewing (B, detail belaw). Damping effect decreases as (he caUon is moved towards areas A and C . 314 REVISTA D E BIOLOGIA TROPICAL Fig, 1 3, Forewing veins in Siproeta ,s'telenes (aboye, len male, right femate), malc Ectil1w ¡heda (middlc IeIt), maje Ham.adryas g!aucollome (middle right) and Ha!nadrv{/s gllCllemalena (bdow: leH fema!e. right male ), MONGE-NAJERA er al. : Spatial distribution, territorialíty and sound in cryptic butterflies 3 15 Fig. 1 4. Forewing veíus in Hamadryas jebrua from Costa Rica (above: left female, right maJe) and f1'Om Venezuela (middle Jeft). Male Ha/lla dryas feronia r-m l and m¡-ffi2 veins have a longitudinal groove (middJe right: section marked in box appears enlarged). Below: venation iu female (left) and maje (right) Hal11a dryas feronia. 31 6 REVíSTA DE BlOLOGIA TROPICAL Eg. 1 :5 . B ase uf forcwings 01' Siproela stelenes (above, leIt fcmaie, right maje), maje Ectima ¡heela (midclk len), maje Hamadrras glauco!7ome (midclle ¡ight) ancl Hamadrvas guatemalcna (below: ldt femule, right male). lVrONGE-NAJERA el al. : Spatial distribution, tcrritoriality nnd soulld in cryptic b'lltert1ies 317 Fig. 1 6. B ase of forewings oi" Hamadryasferonia (above: lefl female, right male) afid Hamadryaslebrua (below: lef! female, right maleo with enl::lTgement of Venezuelall speeimen b y l OX). 318 REVISTA D E B IOLOGIA TROPICAL Fig. 1 7 . Details of forewing bases in Harnadryas feronia (above and middle) and Ectima theda (below) after removal of scales (middle: upper layers removed). VO Vogel's organ, probably a high frequency ear; CCM ear, Sc subcostal vein. MONGE-NAJERA et al. : Spatial distribution, territoriality and sound in cryptic butterflies 319 Fig. 1 8 . Details of Hamadryas februa forewings. B elow: spiral structure inside veins. Above left: intact ear. Above right: inner view of membranous spiral structure. 320 REVI S TA DE BIOLOGIA T RO PICAL SOUND PRODUCED (% CASES) 0 0 0 0 0 0 0 0 0 8 O - C'l ('<) "'i" /') '00 1"-- 00 O< Lepidopterans (N:: 1 9) (Except Hamadryas) Leaves (N::7) Vehicles (N= l O) Fig. 20. Pro ductio n of s Ol.md by Hamadryas sp p . in response to a variety of stimu li , i nclu di ng fa Jli ng Jeaves an d human o bservers . Sorne perches were o n roa dsi de trees,so the reactio n l Ov ehicles passi ng in fronl of perches Fig . 1 9 . Reco nstructio n of Hamadryas wi ng structure . was also recor de d. (N= 1 84). DISCUSSION human observers (Jenkins 198 3 , Monge­ Nájera 1992), (2) their natural behavior Spatial distribution aIll d wing continues in the presence of humans (Jenkins coloration: The wídely known case of crypsis 1983 , Monge-Náj era 198 8 ), (3) their in the industrial melanism moth Biston taxonomy is well established and they can be betularia (L.) is controversial because ( 1) most identified without capture (Jenkins 198 3 , data are experimental, (2) its natural history i5 Monge-Nájera 1992), and (4) their natural poorIy known and (3) the taxonomy of the history is relatively wel! known, so factors species is not c1ear (Keiper 1969, Sargent and Iikely to affect background choice can be Keiper 1969, Sargent 1973, B oardman et al. identified (Jenkins 1983, Monge-Nájera and 1974, Steward 1977a, Sargent 1985). Hernández 1991, Monge-Nájera 1992). In this discussion, comparisons are made with cryptic This study of Hamadryas has the moths because we are aware of no equivalent advantage of describing fieId behavior, and it data for butterflies. fits several important requirements no! met by the Biston studies: ( 1) Hamadryas butterflies The fact that a few trees of a few species alight on exposed surfaces and are large were used frequen tly as perching sites by the enough to be detected and counted with butterflies, while olher trees (sorne of the certainty by properly trained and equipped same species) were never used, is consistent MONGE-NAJERA et al. : Spatial distribution, territoriality and sound in cryptic butterflies 321 Fig. 2 1 . Severe wing darnage (arrow) is not rare in wild Hamadryas males, but the sound-producing section often is in good condition. 3 22 REVISTA DE B IOLOGIA TROPICAL with the hypothesis that certain tree experimental conditions, sorne moths are known individuals have characteristics that attract to perch on non-cryptic backgrounds (in 11-64% Hamadryas males. The underlyíng of trials) but no explanation has been suggested mechanism is unknown (see B itzer and Shaw for the observation (Kettlewell 1955, Steward 1979, B aker 198 3 , Aiello 1988). 1976, 1977b). B utterfly species differed by site, possibly This study of Hamadryas has shown that, because their ability to match trunk coJorations under natural conditions, these butterflies in the studied sites differed, as explained under perch differentially on a variety of Material and Methods (for details see Jenkins backgrounds, that perch selection is associated 1983 and Monge-Nájera 1992); for example, with background color, and that the the bluish H. amphínome perched more backgrounds chosen match wing color, at least frequently on dark bark under thicker canopy to human observers, thus supporting the (a patch of trees in Site 2, Fig. 2), while light interpretation that perching choice in brown H. februa perched on light, sun-exposed Hamadryas has been influenced by natural bark. The significant association between selection for crypsis. These results fit the general background color and perching theory associated with industrial melanism frequency is consistent with the hypothesis without having the methodological problems that the butterflies actively select backgrounds of the Biston studies. on which they will be cryptic, in nature. Vertical distribution : The genus In our data, background color overrides Hamadryas appears particularIy suited to other possible factors that have been suggested compare spatíal distribution because of the to influence the selection of trees as perching constancy of life histories among most species. sites: presence of mud where important salts Immatures use the same host plants and adults are found, presence of rotting fruit as food have identical behaviors and foods (Jenkíns (Jenkins 198 3 , Otero 1 988), and position of 198 3 , Monge-Nájera 1992). Sorne ecological trees in intersection sites along flight routes displacement ís expected to evolve under such used by females (Baker 1972). Hypothetically, circumstances and our basic hypothesis was males might prefer larger trees that may be that spatial distribution would show vertical visible to females from greater distances, or stratification, believed to favor mating they may avoid rough bark that could cause encounters (Daily et al. 1991) and the wing damage, but neither tree size (measured coexistence of more species. by diameter and height) nor bark texture were found to be associated with tree use. Probably, Several collectors have mentioned that (as Iichen cover and bark basal color did not have also found hefe) Hamadryas species can be a significant effect either because they, being classified in a "lower" group (e.g. H. februa, mostIy greenish or pink, frequently did not H. glaucoflome and H. guatemalena) and a match the butterfly colors. "higher" canopy group (e.g. H. fomax, H. rosandra and H. alida), although there were Under natural conditions, Hamadryas no quantitative studies previous to the present butterflies occasionally perch on inappropriate work (Jenkins 1983 ). backgrounds, mostly on smooth light brown Psidium guajava trees. The cost of an This study fits the requirements mentioned inappropriate background match seems to be by De Vries ( 1988 ) and found a statistically lower than the benefit of obtaining food, which significant vertical stratification consistent with is abundant in the form of rotten fruit that the "ecological displacement" interpretation, accumulates seasonally under these trees. Under but of course does not prove it. The five species pereh high�Ii iliap moits�that�s6' are cryptie on reaches thedlight positioll¡;simply ¡by skiptJing · · bark (Sargént @d 1<eip�i 1969). They peren (our unpublished analysi� ·ofvídeo im�ges). .: . higher at the �nd of the day possibly to reduee . flight di stan ee to the fo1iage 'where they $pend Re�ction to moving objects :. Flight the night. .'flie rel ativ e use of branches :·and . . towards organisms or. objeClts mgving. in front truhks apparently reflects ; the verti c al range oí ' of p erches is common in mány speci e s of . perohes typical of each species : trees in Si�e 1 1epidopterans . and seerns to re sul � frmn poor branehed at greater héight ahd thas "rtlore "ision (Tinbergen 1 958, Scott 1974, Bitzer and butterflies p�rched on the trun�. Shaw 19CJ9); ;,tlbe poor vision hyp othesi s is supportea:1Jy me reactiun:-of Hamadryas��:. Cardinal point distril1ution : · The rnooel s w h i c h are exainined- and insects such - cardinal d is tributi o n ofperéhes depends on the .• as'.ants Mfalking near their p erche s -which are s i te Jenkins ( 198 3 ) mentjtmed that sorne , ignored ór avoided by walking to an otber . s p e c i e s prefe r open, sunny ' places (e.g. H. percho The fre qu e n t walkin g :Qf tbese glaur;onome and H. guatemalena) while others butterfli es, which greatly ' surprised Darwin prefer dark parts Of the habitat (e.g. #. alicia (1839) ,an0 . has. ;ri�ver been expll!ined, may be . . � H. rost;lndra); As this study conceme4 energeticall�. �!c�ap�r", �d less conspicuous species assoeiated 'with open . areas 'l:luring · the to predators ·thal1 (tighí; ti me oí year wllert sun l i ght reached · the tre.e$ from .the sotlth¡ it could � predicted that: th� . Experiments with ,:ardboard models; ndFti�m !lide' w�>uld not be preferred¡ and thls, whe!1 themtodet w!lS presented . at a distánc� H. . " ' was ,fóünd in altcas,es. .gufltdlla lena reacted mo� often aJ!d 1 1 . ftbrua did ·. the' , oPP0site. 1?lle.' fotm�r!'\migb�: have ' a · ii' Youlíg 'aha' ;I3prkir¡. �J9g,5 ) proposéd that, .. bet�t vis.al capacity t� ��enfiiy tli;e mbde1 as ' /'!'tlmiídryas tlse 1es6 " 'liJtt1y the windy s.�4�; , "not'i�.esñng" aridl H:;'feb1{Ua IDay ·fail to ..' ·.· wili�h'ís less!cp pF<butterl}y couriship , : t¡ '�ge[ . l �nt�tspecific oetect';· : iti 'at . grea:ter , 'diffei���es ih v�l>�al: s�h�M'· .ar� �pwh ¡ti .. . 1¡;�:i11g�t. (Wi� Yklll an el. al. · 1;99��,. . . '.�'f;�in(W . .. . i,thtn S¡;t�l>; l\��d 2.futa .' . oth�r b�erflies;��itzer •" " ' amt Sh�w 1983). "i, - , We:: �yPotil,,�� ' \J- 1 . ', " - ;tb'A:; l1ortbe :�te, . 3, .'· thus ' , : ;'.: ;" �' :�� dí�t . b�: . . . .' . .. ':t&t ·Sités 1 , a�:2: Wlrén · her� was a teIla�ñcy of' f¡. gua,�m41e� 1 ��.�s�;t���1)n�¡,������y ttig�li��t";aild . / · in�iYidú�li to.stoP':�i¡lOting . #ter'pie' ��. t�s!S ' . . temp�ra�e roay be ill'i),re Importanf . . ' . e :tactOfS ' . , �Y r�sult Xt'rom wiíh ,:�tll e . m9'�eli';d'tlris. Jil:l H . . (Bit�r alld S�w 1995): habituation; as sug"s� �¡'b� �$'e :'� qbserVation '. that .a mate of the s�ies�s@p#�tClP� �eal\ltiij.g On suitn,y' days, ''l:l.' il� Sñ:óws. some lO leav�s falling Tepeat�y in '.onfÓf if. preference for sunny : �e , '. . . '. ' arid the �ther /$pecies for shad()W;�P P�rcfi�§';¡:Itf¡ the hot Santa , : . ..emoval óf'l'esid�.tr,nlales and hatuJ,1li .. Rosa Na�j:<1.i1al Hárk, :;9� o/6f Jof'>H. !érohiai, . . ;' · infe á i �ti.)�: The hjghW;.¡ti�����d frequency individual.s perch on. , .ith� í";'S}[adQw iÍl. :t�� . \'iVith whjch . tr�s were , .ta\Cei'í:.j,;y neWCOlflers mOl,lling (N=84¡ AP. Re";�fJt;Is� coiif. ��$'8) �ter: re$idents had !leen temo.'ved· supportsíthe prbballly to reduce he�t streSsl territ@FialiPte��tion �DirX1�� 1918). . PC)sture and �eas�D�ty: The specit1c; . . ':Althotlglj " Jnost Harfiádiyas interactions PQ�ition ,of the he/id €upward ot. sidewaYs} is " c:tln:si��red aggressive, .! .they are c�not ' be g!lÍnetically determliied;ill sorne rno1:lls (Sar.g'1llt . territorial:;�f:ca,QS* ; they l'e�:Q1t in exclusi? n of the intrm1ei'froln the tt���see,'�Q�m�g¡ 1978).; ." 1 969; " 'Satgent ; and." Keiper ' 1,969J. ,! ifit ; · '. HaTl'Jllilryas, perchiQ:g with the head, dí\)wttw.ard �le m8,ll>,"nla1e . int�rac�tin� m9sdy'en4' :wi� ma)' facilitat� eseapi flight, becaJ1,se the ins�¿t ' >' 1 " , .me · new�trie�" l�avJPg tbefiFe�;, 'O" '; ;2"10, f��iitlg" , ' '- >; _ " . , - �< , ' . . ' - ' '' '1 ' " _ ', " e_ ' " " 324 REVISTA DE BIOLOGIA TROPIC AL individuals can share a tree (Jenkins 1983). Hamadryas enters reproductive diapause Ross ( 1963 ) rejected the hypothesis of (Young 1974, Young and B orkin 1 98 5). territoriality in Hamadryas, but his descriptions show that he worked with feeding The disappearance of any apparent individuals, not with territorial males. aggression by night and the use of branch tips to perch may reduce predation on the Wing condition seems unrelated to butterflies, as suggested for other insects by territorial success, although data are scanty Shapiro ( 1977). (Davies 1988). In the case of H. feronia, males with very damaged wings can retain trees In conclusion, s piral flights vary in against intruders with fresh wings (N= 17). The frequency with season and time of day. Such damage itself may result from agonistic flights and the sound emission that often interactions (Young 1974, Pinheiro 1990) and accompanies them seem to have both warning may characterize highly aggressive males, thus (often territorial) and courting functions. solving Darwin' s ( 1 87 1) paradox (female butterflies copulate with males of poor wing Sound p roducti o n : From the many condition). In Hamadryas wing damage proposed mechanisms of Hamadryas sound actually favors crypsis by making wing outline production (Swihart 1967 , Jenkins 198 3 , less conspicuous (pers . observ.). Monge-Nájera 1992) our results ar e consistent only with the hypothesis of forewing collision Effect of time on interactions: Davies presented by Seitz ( 19 1 4) and supported by ( 1978 ) suggested that when territories are Otero ( 1990). However, our conclusion differs relatively abundant, a convention (such as from Otero' s hypothesis because the r-rn¡_2 being resident) may decide who stays. That have been shown to be only part of the hypothesis is consistent with these data and the mechanism and becaus e sound is not abundance of territories for H. feronia: one produced by stridulation as he stated but by individual uses several trees daily, and each percussion (see Dumortier 1963 for detailed tree has several residents per day and may definitions). Our sonograms suggest that the remain vacant mos! of the day. two individuaIs can s imultaneously emit sound during joint spiral flights, possibly Preferred territories may be visited by more allowing an early decision about who stays in females and províde resident males with more the territory. copulations (Bitzer and Shaw 1 979, 1983); five trees apparently were more favored by H. The result of the evolutionary enlargement feronia and had twice as many residents per day and modífícation of veins has been the than the rest, but no copulations were observed, productíon of powerful sounds that can be as usual in butterflies (Pinheiro 1990). heard by humans at 30 m, probably louder than in any other lepidopteran. Stronger signals Mos! interactions took place in the first half may be more effective to deter predators of the afternoon, very similar to other butterflies (Busnell 1963), expel territorial intruders from tropical (Pinheiro 1990) and temperate (Alcock 1989) and attract mates (Wakamura areas (Baker 1972). A decrease in butterfly 1977, Spangler et al. 1984). territoriality at the beginning and at the end of the mating season is known in temperate species Growth of the sound apparatus probably is (Baker 1972). H. feronia interactions are less checked by its effect on flight capacity (Betts frequent in the dry season (tbis study) when and Wootton 1988 , B rodskii and Voroblev there is a reduction in abundance and 1 990), physiological costs and ecological MONGE-NAJERA et al.: Spatial distribution, territoriality and sound in cryptic butterflies 325 reasons (e. g. sound may attract predators, rather indicative to the passing male that he has Sharp 1 899, Surlykke and Gogala 198 6). intruded in a territory. What are thought to be real aggressive flights, more than 10 seco in The ear (cupola) of Hamadryas was duration (Wickman and Wicklund 198 3 ) are identified by Swihart ( 1967) but this is the first very rarely witnessed in Hamadryas and are ultrastructural description that has been characterized by an increase in the intensity published. It is basically a stretched membrane and frequency of the sound (pers. observ.), but that apparentIy vibrates with sound. The the territorial function seems clear from the possession oftwo earS f il!Y fª<:jlita� d�e!ection result of the male removal experiments. of sound origino What has been reported about Hamadryas hearing capacity (Swihart 1967) The variety of mechanisms (references matches click component frequencies and listed in the introduction) suggests that number of clicks per second. lepidopterans have evolved sound production several times, often for defense (Busne11 1963, Sound frequency and signal redundancy in Masters 1979) and through natural selection, Hamadryas are within the range for not through sexual selection. Sexual selection lepidopteran sounds (Haskell I 96 1, Loyd 1974, could only explain why male Hamadryas have Kane 1982, Surlykke and Gogala 1986) and fit larger sound-producing structures. For future the requirements to facilitate transmission work we suggest the hypothesis that through vegetation with minimal loss and intraspecific use to warn territory intruders and distortion (Michelsen and Larsen 198 3 ). to attract mates are secondary extensions of that function. As any signal (Pasteur 1972), The small size, rigidity and organization butterfly sounds may have evolved a complex of Vogel's organ suggests that it is also an ear variety of mimetical mechanisms that represent but specialized in ultrasound, possibly to virgin territory for biological exploration. detect predatory bats. B ats feed on night perching butterflies (Navarro 199 5). Ecology and evolution of sound: ACKNOWLEDGMENTS Despite the many speculations, the function of sound in Hamadryas had never been studied Many people have assisted us during the formal1y (Jenkins 198 3 ). Captive individuals decade that this project required, and we can are known to have courtship with sound (H. only name sorne of them. Alvaro Castaing R. feronia) and without it (H. februa) and sound began the project by suggesting the study of has been suggested to attract females to an Hamadryas to the senior author 15 years ago, area (Otero 198 8 ). and greatly assisted with early fi eld work, as did Patricia Val verde, B ernal Morera and The present field study showed that Sylvia Fernández. Adriana Herrera labeled the courtship is silent in H. februa and noisy in H. illustrations. Carlos Valerio, Alvaro Wille, feronia (pers. observ.) but sound may also Juan B . ChavaiTÍa, William Ramírez, D.C. have a warning function because it was used Robinson, EG. Stiles (University of Costa in male-male interactions (joint spiral flights) Rica UCR), P. J. DeVries (University of and when large animals approached. At least Texas), Luis D. Otero (Central University of the large animals were unlikely to be mistaken Venezuela), Douglas C. Robinson and E G. for females. Stiles (both UCR), Donald Windsor and Annette Aiello (both Smithsonian Tropical Davies (1 978) stated that short spiral Research Institute, STRI) gave valuable flights should not be considered aggressive but support, assistance and advise. Marco T. 326 REVISTA DE B IOLOGlA TROPICAL Saborío produced the photographs. We are Bailey, W. J. 1 97 8 . Resonant wing system in the Australian thankful for their help. For advise and whistling moth Hecatesia (Agaristidae, Lepidoptera). Nature 272 : 444-446. suggestions to improve sections of earlier drafts we deeply thank four annonymous reviewers as B aker, R. R. 1 972. Territorial behaviour of the nymphalid well as Annette Aiello (STRI), R. R. B aker, butterflies, A glais urticae (L.) and Inachis io (L.). 1 . Robert Dudley (University of Texas), WilIiam Anim. EeoL 4 1 : 453·469 Eberhard (UCR), Paul Ehrlich (Stanford Baker, R.R. 1 9 8 3 . Inseet territoriality. Ann. Rev. EntomoL University), Allen M. Young (Milwaukee 28 : 65-89. Public Museum), Royce J. B itzer and Kenneth C. Shaw (Iowa State University), Annemarie Betts, eR. & R. J . Wootton. 1 98 8 . Wing shape and f1ight Surlykke (Odense University), Dale Jenkins behaviour in butterflies (Lepidoptera: Papilionoidea (Florida), Robert Srygley (Cambridge and Hesperioidea): A prelirninary analysis. J . Exp. Bio!. 1 3 8 : 27 1 -2 8 8 . University), George Thomson (Dumbrieshire), and Per-Olof Wickman (Stockholm University). Bitzer, R . J. & K. e Shaw. 1 98 3 . Territorial behavior o f William B ussing (UCR) corrected the English Nymphalis antiopa a n d Polygonia comma text. This research was financed by a STRI (Nymphalidae). J. Lepid. Soe. 37: 1 - 1 3 . Exxon Corporation scholarship, the University Bitzer, R . J . & K . e Shaw. 1 995. Territorial Behavior of of Costa Rica (Vicerrectoría de Investigación the Red Admiral, Vanessa atalanta (Lepidoptera: 8 10-87-062), the authors, and Patricia Valverde. Nymphalidae) L The Role of Climatic Faetors and Early Interaction Frequency on Territorial Start Time. J. ¡nseet Behav. 8 : 47-66. RESUMEN Bitzer, R. J . & K. Shaw. 1 979. Territorial behavior of the red admiral, Vanessa atalallta (L.) (Lepidoptera: Las Hamadryas son mariposas neotropicales que Nyrnphalidae). J . Res. Lepid. 1 8 : 36-49. emiten sonido y pasan mucho tiempo posadas en árboles, donde son crípticas. Se estudió cinco especies en Costa Ri­ Boardman, M., R . R . Askew & L . M . Cook. 1 974. ca y Panama. Cada especie tiene una altura característica Experiments on resting site selection by nocturnal para posarse. Se posaron menos del lado sur de los árboles moths. J . ZooJ. London 1 72 : 343-355. y evitaron las partes soleadas en días calurosos. Los ma­ chos dejaron su percha para volar hacia otras Hamadryas Brodskii, A . K. & N . N . Voroblev. 1 990. The gliding flight y hacia "mariposas" de cartón. Cuando se retiró los ma­ of butterflies and role of the wing seale cover in flight chos de los árboles éstos fueron tomados con mucha ma­ (in Russian). EntomoL Obozr. 69: 241 -256. yor frecuencia por otros machos. La mayoría de las inte­ racciones aéreas de H. femnia se dieron entre las 1 3 : 00 y Busnell, R. G . 1 96 3 . (ed.). Aeoustie behavior of animals. las 1 5 :00 h y fueron más frecuentes en la estación lluvio­ Elsevier, Amsterdam. 933 p . sa. De noche, los machos compartían los árboles. El soni­ Calvert, A. S . & P. P. Calvert. 1 9 1 7. A Year of Costa Riean do característico del grupo es producido por venas modifi­ Natural History. Maemillan, New York. 577 p . cadas en las alas delanteras. Cott, H.B . 1 946. Adaptive eoloration i n animals. Methuen, London. 540 p . REFERENCES Daily, G . e , PR. Ehrlich & D. Wheye. 1 99 1 . Determinants of spatial distIibution in a population of the subalpine Aiello, A. 1 992. Dry season strategies of two Panamanian butterfly Oelleis chryxus. Oeeologia 8 8 : 587-596. hutterfly species, Anartia fatima (Nymphalinae) and Pierella luna luna (Satyrinae) (Lepidoptera: Darwin, e R . 1 839. A naturalist voyage. Murray, London. Nymphalidae), p. 573-575 In D. Quintero & A. Aiello (eds.). Insects of Panama and Mesoamerica: Selected Darwin, e R. 1 87 1 . The deseent of rnan and selection in Studies. Oxford University, Oxford. relation to sexo Murray, London. Alcock, 1. 1 989. Aeoustic Signaling, Territoriality, and Davies, N . B . 1 97 8 . Territorial defence in the speckled Mating in Whistling Moths, Hecatesia thyridion wood butterfJy (Parage aegeria): the resident always (Agaristidae). 1. Inseet Behav. 2: 27-37. wins. Anim. Behav. 26: 1 3 8 - 1 47 . MONGE-NAJERA et al.: Spatial distribution, territoriality and sound in cryptic butterflies 327 De Vrles, P.J. 1 9&8. Stratification of fruit-feeding (eds. ) . Neuroethology and Behavioral Physiology. nymphalid butterflies in a Costa Rican fores!. J. Res. Springer-Verlag, Berhn. Lepid. 26: 9 8 - 1 08 . Monge-Nájera, J . & F. Hernández. " 1 9 9 1 " ( 1 993). A Dumortier, B . 1 963. Morphology o f sound emlSSlOn morphological search for (he sound mechanism of apparatus in Arthropoda, p. 277-345 In R. G. B usnell Hamadryas butterflies. 1 . Res. Lepid. 30: 1 96-208. (ed. ) . Acoustic behavior in animal s . Elsevier, Amsterdam. 933 p . Monge-Nájera, J . 1 98 8 . Mariposas crípticas del género Hamadryas (Lepidoptera: Nymphalidae ) : uso de Hampson, G . F. 1 89 2 . On stridulation in certain sustrato, territorialidad, emisión de sonido y lepidoptera. Prac. Zool. Soc. London 1 892: 1 88 - 1 93 . morfología. M.Sc. Thesis, University of Costa Rica, San José, Costa Rica. 61 p . Hannemann, H. J . 1956. Uber ptero-tarsale Stridulation und einíge andere Arten der Lauterzeugung bei Monge-Nájera, 1. 1 992. Clicking Butterflies, Hamadryas, o f Lepidopteren. Dtsch. Entom. Z. 3: 1 4-27. Panama: Their biology and classification (Lepidoptera, Nymphalidae), p. 567-572. In D. Quintero & A Aiello Haskell, P. T 1 96 1 . Insect Sounds. Quadrangle B ooks, (eds.). Insects of Panama and Mesoamenca: Selected Chicago. 1 89 p . Studies. Oxford University, Oxford. Hebard, M . J 922. The stridulation o f a North American Navarro P., N. 1 99 5 . Incidencia de lepidópteros diurnos noctuid, Heliothis paradoxus Grote (Lepidoptera). e n l a dieta del murciélago orej udo (Macrotus Entorno!. News 33: 244. waterhousei minor Gundlach/Peters, 1 86 5 ). Garciana (Cuba) 2 3 : 1 1 - 1 2 . Jenkins, D. 1 98 3 . Neotropical Nymphalidae I. Revision of Hamadryas. Bull. Allyn Mus. 8 1 : 1 - 146. Otero, L . D . 1 990. The stridulatory organ in Hamadryas (NymphaJidae): Preliminary observations. J. Lepid. Joy, N .H. 1 902. (untitled). Proc. Entom. Soc. London 1 9 Soc. 44: 285-288 . Nov: 1 1 - 1 2 . Otero, L.D. 1 98 8 . Contribución a la historia natural del Kane, S . 1 982. Notes on the acoustic signals of a género Hamadryas (Lepidoptera: Nymphalidae). Neotropical satyrid butterfly. J. Lepid. Soc. 36: 200- Ph.D. Thesis, Universidad Central de Venezuela, 206. Maracay, Venezuela. 1 08 p. Keiper, R.R. 1 969. Behavioral adaptations of cryptic Pasteur, G. 1 972. Le mimétisme. Universitaires de France, moths. IV. Preliminary studies of species resembling Paris. 1 28 p. dead leaves. 1. Lepid. Soc. 2 3 : 205-2 10. Pinheiro, c.E.G. 1 990 (1 9 9 1 ) . Territorial HilItopping KettleweJl, H.B.D. 1 95 5 . Selection experiments on B ehavior o f Three Swallowtail B utterflies industrial melanism in the Lepidoptera. Heredity 9 : (Lepidoptera, Papilionidae) in Western B razil. J. Res. 323-142. Lepid. 29: 1 34- 1 42 . Laithwaite, E., A Watson, P. E. S . Whalley & W. D. Ross, G . N . 1 963 . Evidence for lack of territoriality in two Duckworth. 1 975. The dictionary of butterflies and species of Hamadryas (Nymphalidae). J . Res. Lepid. moths. Joseph, London. 296 p. 2: 241 -246. L1oyd, J . E. 1 974. Genital stridulation in Psilogramma Rothschíld, M. & P. T. HaskelL 1 966. Stridulation of the menephron (Sphingidae). J. Lepid. Soe. 2 8 : 349-3 5 1 . garden tiger moth, Arctia caja L. audible to the human ear. Proc. Roy. Entorno! . Soc. London (A) 4 1 : 1 67- 1 7 1 . Masters, W. M . 1 979. Inseet Disturbance Stridulation: Its Defensive Role. Behav. Eco!. SociobioL 5: 1 87-200. Sargent, T D . & R.R. Keiper. 1 969. Behavioral adaptations of cryptie moths. 1. Preliminary studies on bark-Iike MeCrae, A W. R. 1975. Clicking in flight by an African species. J.Lepid. Soc. 2 3 : 1 - 9 . fruit-piereing moth, Achaea obvia Hampson (Lep. Noctuidae). Ent. Month. Mag. 1 1 1 : 1 6 1 - 1 64 . Sargent, T.D . 1 969. Behavioral adaptations of cryptic moths. V. Preliminary studies on an anthophilous Michelsen, A. & O . N . Larsen. 1 98 3 . Strategies for species, Schiniaflorida (Noctuidae). J. N.Y. Entorno!. Acoustics: Cornmunieation in Complex Soc. 77 : 1 23 - 1 2 8 . Environments, p. 32 1 -3 3 1 . In F. Huber & H. Markl 328 REVISTA DE B IOLOGIA TROPICAL Sargent, T.D. 1 973. Behavioral adaptations of cryptic Swihart, S L 1 967. Hearing in butlerflies (Nymphalidae: moths. VI. Further experimental studies on bark-like Heliconius, Ageronia). J. lnsect Physiol. 1 3 : 469-476. species. J. Lepid. Soco 27: 8 - 1 2 . Swinton, A. H. J 877. On stridnlatiolJ in ¡he genus Sargent, T.D. J 9 8 5 . Melanism i n Phigalia títea (Cramer) Ageronia. Entomo!. Month. Mag. 1 3 : 207-208. (Lepidoptera: Geomeuidae) in southem New England: a response lO forest disturbance? J . N . Y. Entorno!. Thomhill, R. & J. Alcock. 1 98 3 . The Evolution of Insect Soe. 9 3 : 1 l 1 3 - 1 I 20. MalÍng Syslems. Harvard, Cambridge, Massachussetts 547 p . Scoble, M.J. 1 992. The Lepidoptera: form, function and diversity. Oxford University, Oxford. 3 8 9 p. Tinbergen. N. 1 95 8 . Curious naturalists. Country Life, London. p. 1 64- 1 7 8 . Seoll, F. W. 1 968. Sounds produced by Neptis hylas (Nymphalidae). J. Lepídop. Soco 2 2 : 254. Vogel, R . 1 9 1 2. Uner die Chordotonalorgane in der Wurzel der Schmeuerlingsflugel. Z. Wiss. Zool. 1 00: 2 10-244. Scott, J . A . ¡ 974. Mate-Iocating behavior of butteIflies. Amer Mid. Na!. 9 1 : 1 03- 1 1 7 . Walmmura, S. 1 977. SOllnd production of ¡he male moth A grotis fucosa (Lepidoptera: Noctuidae) during Seitz, A. 1 9 1 4 . The macrolepidoptcra of ¡he World. A. courtship behaviour. Appl. Entomo!. Zool. 1 2 : 202- Keman, Stuttgart. 203. Shapiro, A.M. 1 977. Avían predatíon on butterflies - White, p.E. 1 877. Strídulation in the genus Vánessa. Ent. Agaín. Entorno!. Ree. 89: 293-295. Month. Mag. 1 3 :208. Sharp, D. 1 899. Inseets. Dover, New York. Wickman, 1'.0, E. García B arro s & e. Rappe-George. 1 995. The location of landmark leks in lhe small heath Spangler, H. G. , M. D. Greenfield & K. Takessian. 1 984. butterfly, Coellonympha pamphilus: evídence against Ultrasonic mate calling in the Icsser was molt!. the hot-spot model. Behav. Eco!. 6: 3 9-45. Physiol. Entomo!. 9: 87-95. Wickman, P.O. & c. Wicklund. 1 98 3 . Ten'itorial defense Steward, R.e. 1 976. Experiments of resting site selection and its seasonal decline in the speckled wood butteIfly by the typical and meJanie formó of ¡he morh, (Pararge aegeria). Anim. Behav. 3 1 : 1 206- 1 2 1 6 . AllophyeS oxyacanthae (Caradrínidae). J . Zool. London 1 7 8 : 1 07- 1 1 5 . Young, A.M. & S . S . Borkin. 1 985. Natural history notes for some Hamadryas butterflies (Nymphalidae: Stewaxd, R.e. 1 977 a . Further experiments o f resting site Nymphalinae: Aegeronini) in Northwestern Costa selection by the typical and melanic forms of the mOlh, Rica during ¡he tropical dry season. J. Lepid . Soco 39: Allophyes oxyacanthae (Caradrinidae ) . J. Zool . 229-235. London 1 8 1 : 395-406. Young, A.M. 1 974. On Ihe biology of HamadlJas februa Steward, R.e. 1 977 b . Industrial and non-industria! (Lepidoptera: Nymphalidae) in Guanacaste, Costa melanism in ¡he peppered moth BistOll betularia (L.). Rica. Z. ang. Ent. 76: 380-393. Eco!. Entomo\. 2 : 23 1 -243 . Surlykke, A. & M. Gogala. 1 986. Stridulation and hearing in ¡he noctllid moth Thecophora fovea (Tr.). J. Comp. PhysioL 1 59 A: 267-273. MONGE-NAJERA el al.: Spatial distribution, territoriality and sound in cryptic butterflies 329 APPENDIX 1 Mulliple regression results for number of butterjlies perching versus several tree factors in Harnadryas butterjlies Species, Site ( 1 -3 = Variables in Coefficients Species, Site ( 1 -3 = Variables in Coefficients I-I1I), R adjusted and model I-I1I), R adjusted and model dependent variables dependent variables H. feronia 1 Constant 78.8* H. glauconome III Constant 1.14 0.33 V Species -75.8* 0.43 V Trunk diarneter 0.01 Height 4.2 . l..it:hell cº'{t�r..� 0.03 Color 30.7* Lichen cover-C 0. 1 0 Lichen cover-E -0.04 H. februa 1 Constant -2.25 Food 1 .70 0.39 V Tree -0.27* Position in -2.88 flight route Height 0.53* Color -0.45 H. feronia 1 Constant 1 0.46* H. februa III Constant 0.80 0.38 DV Trunk diarneter 23.95* 0. 1 5 V Trunk diarneter 0.02 Color 8.57** Height -0. 1 9 Lichen cover-C 0.01 Food 0.55 H. februa 1 Constant -0.40 Texture -0.46 0.5 1 DV Tree -0. 1 1 * * Basic color 0.27 Species 0.04* Color 0.20 Height 0. 1 2 * H. guatemalena III Constant 1 .08* H. feronia 11 Constant -0.78 0.35 DV Lichen cover-D 0.03 * * 0. 1 5 V Species 0. 1 3 Lichen cover-D -0.03* * Food -2. 1 6 * * H . amphinome 11 Constant -4.5 Position 1 .22* 0.40 V Trunk diarneter 14.0* * H. glauconome III Constant 0.40 0.3 1 DV Trunk diameter 0.01 H. feronia 11 Constant -0.47 Height -0.06 0.22 DV Height 0. 1 5 Lichen cover-B 0.01 Lichen cover-C 0.04** H. amphinome 11 Constant - 1 .34 Lichen cover-E -0.02 * * 0.58 DV Trunk diarneter 3.85 Lichen cover-H -0.01 Color 1 .84 Color -0. 1 5 H . guatemalena III Constant 4. 1 5 H. februa III Constant 0.63* Lichen cover-D 0. 1 7 * * DV Trunk diarneter 0.01 Lichen cover-H -0. 1 7 * * Height -0. 1 1 * * Trunk diarneter 0.02 Lichen cover-A 0.02 Food - 1 0.0** Lichen cover-C 0.02 Position in 8 . 1 6* * Lichen cover-E -0.02* flight route Mud 2.75* B asic color 0.53* B asic color 1 .30 Color -0.22 Color: general color of tree bark including epiphytes, B asic color: natural bark color without epiphytes, letters afier "Lichen" refer to different sides and heights in the trunk, 1.5 m aboye ground: C south, D east E north,F west; 0.5 m aboye ground G soutb, H east, 1 north, J west.