Butterflies of the Sierra Nevada de Santa Marta, Colombia
by Bruce Purser
Key words : Andes, bamboo, Colombia, endemism, Morpho, Santa Marta, montane forest, neotropical
(Click on map and pictures for a larger size !)
Santa Marta, the highest point within the Colombian Andes, has an exceptional setting not only because of its proximity to the Caribbean Ocean but, especially, because of its geographic isolation (Fig. 1). Separation from the parental Andean chain is the consequence of major geological fault-lines whose mutual obliquity causes the characteristic triangular form of this structural block. The massif is cut by deep, radially disposed valleys which dissect the mountain into a series of high, sharp-crested ridges whose topography is favoured by the relatively soft nature of the deeply weatherd granite which composes most of the massif. The culminating points appear to be less jagged in outline and numerous small lakes and wide depressions suggest a glacial influence. The culmination, Pic C. Colon (5800 m) carries permanent snow.
Geographic isolation has resulted in a considerable degree of endemism ; there are 120 species of endemic vertebrates (ref. 5), and probably many more endemic butterflies, notably those inhabiting the higher mountain forests. For example, 20 new species of ithominid have been collected recently by Dr W. WINHARD of Bogota (pers. comm.) and other relatively high altitude genera such as Perisama include a number of endemics (ref. 2). The most reputed endemic is the mountain Morpho rhodopteron (Fig. 2 ).
A peripheral highway around the massif, together with at least four good roads into the interior, favour ready access. However, the central parts of Santa Marta are reached only by 4 x 4 vehicle and, finally, by mule. Most of the sierra is a national park within which collecting is prohibited. Sierra Nevada de Santa Marta is renowned historically for its « Lost City », a series of pre-Colombian constructions erected by the Tayrona Indians within the forested NE parts of the massif (ref. 5). The flora and fauna of the region has received only limited study mainly due to the presence of armed militias and drugs. However, the situation, having improved in recent years, one may now visit the peripheral N, W and S parts of the sierra ; the eastern extremity remains dangerous.
Climate and vegetation
The relatively small massif (Fig. 1) is extremely varied ; both its climate and vegetation range from tropical rain-forest to semi-desert. Here, in a relatively limited space, exist virtually all the environments of a major Andean chain but, because these important variations are imbricated, one should examine Santa Marta as a whole. A marked asymmetry related to an ENE prevailing wind carries humidity onto the NE slopes. These are clad in anthropised rain-forest (Fig. 3) which passes upwards into typical pre-montane vegetation. These humid forests pass abruptly into exceedingly dry vegetation (Fig. 4) as one approaches the NW corner of the Sierra, the town of Santa Marta being fringed with cactus. The prolongation of xerophytic vegetation along the lower, western slopes of the massif probably expresses a « shadow effect » along the leeward side of the massif . However, as one proceeds towards the south, the band of cactus diminishes in width being replaced progressively by open forest (savanna) which occupies most of the southern and south-eastern parts of the mountain. To the east, the dry forests of Santa Marta pass into the Guajira desert.
A series of radial transects up the slopes of the massif is facilitated by paved roads to an altitude of about 1000 meters prolonged by well defined tracks frequented by motor-cycles and mules. Because of marked lateral and vertical variations in climate and vegetation, each profile is unique.
Transect 1 (Fig. 1), beginning from the north coast, leads up to the archeological « lost city ». Regulations require a local guide and a permit is required to capture butterflies. Vegetation is humid (Fig. 3) and, apart from the lower slopes, hill forest seems to be well preserved and butterflies probably abound.
Transect 2, from the NW extremity of Santa Marta begins in Tayrona Park which comprises two contrasting types of lowland vegetation : the eastern half is clad in evergreen coastal rain-forest, the western in exceedingly dry deciduous forest (Fig. 4 ). In this dry region open spaces are populated by cactus, euphorbias and ascelepia (milk-weed), the host plant for the numerous danaid butterflies (D. eressimus). During the month of April the occasional Morpho peleides, numerous Battus and Hamadryas and the occasional Heliconius erato flew through this leafless forest . The transect continued up the road towards the mountain villages of Minca and Campano, dry vegetation of the lower slopes passing rapidly (around 200 m) into savanna and pre-montane forest with abundant groves of bamboo which surrounded the village of Campano (1500 m). An occasional Morpho rhodopteron flew around the high bamboo but because of persistent light rain, butterflies were limited to the occasional Actinote and ithominid. We were advised by the local population not to venture further than Campano.
Transect 3, on the western flank of the sierra, follows the road up to San Pedro, a small village perched on the knife-edged crest of a very narrow ridge. The village is surrounded by high (25 m) montane forest (Fig. 5), bamboo and coffee plantations, generally bathed in heavy mist. A gravel road leading westwards from the village passes through primary forest frequented by Morpho helenor and ithominids. About 3 kms from the village, at an altitude of about 1500 meters, a sunlit valley flanked with large stands of bamboo (Fig. 6) was frequented by about a dozen male Morpho rhodopteron, all flying close to the bamboo. This remarkable insect, whose behaviour closely resembles that of Morpho sulkowskyi, flies only in bright sunshine, settling instantly when clouds appear. Males fly between 9 and mid-day, females being exceedingly discreet. M. rhodopteron is very seasonal, flying in April and May and, again, in September and December. Other butterflies included frequent Heliconius clysonimus and an unidentified, pale-grey Hamadryas which fed under the numerous mango trees.
Transect 4, located on the southern slopes of Santa Marta, follows a paved road up to the town of Pueblo Bello and thence up-hill, along a very bad « road » to the village of San Sebastian. Cactus disappeared near the southern extremity of the sierra the lower slopes of which were covered in open savanna while scattered patches of hill forest became more frequent closer to Pueblo Bello. Although groves of forest occured near the bottom of valleys, primary pre-montane forest comparable to that of the Andean cordilleras does not exist and the aridity of the southern flanks of Santa Marta is expressed by widespread scrubby vegetation. In spite of this relatively unattractive aspect, brown Danaus eressimus and Morpho rhodopteron were surprisingly abundant. The morphos flew around the edges of open fields (Fig. 7) and near the banks of the Rio Molino fringed with bamboo, situated at an altitude of about 1100 meters. As such, the Pueblo Bello locality is considerably lower and much dryer than that of San Pedro. Over a period of two days, about 25 Morpho rhodopteron were observed, only two of which were females. Their behaviour changed somewhat according to the hour ; between 8.30 and 10 they flapped and glided at heights rarely exceeding two meters but towards mid-day they tended to frequent the tops of bamboo thickets within which they settled immediately following the disappearance of sun. The high hills between Pueblo Bello and San Sebastian some 20 kms. to the north are relatively arid and patches of forest are limited to the bottoms of several valleys. Careful inspection north of Pueblo Bello failed to locate Morpho rhodopteron, butterflies being limited to the occasional Colias.
Transect 5, on the SE slopes of Santa Marta, follows the Rio Guatapuri and its tributary, the Donachui. This transect was studied during an expedition (1985) led by Bernard Courtin who has generously offered verbal information. In common with transect 4 (Pueblo Bello), the region traversed by Courtin consisted of savanna and dry scrub at lower altitudes but eventually attained well preserved montane forest between 1500 and 1800 meters. These forests were frequented by numerous Morpho rhodopteron belonging to the same sub-species encountered by the author at 1100 meters near Pueblo Bello. These observations imply that Morpho rhodopteron occupies an altitudinal range of at least 1000 meters and in view of comparable ranges of Morpho sulkowskyi (1200 m in the Cosnipata valley, Peru), this is not surprising.
Fig. 8. Male sub-species Morpho rhodopteron schultzei (alive) on bamboo, photographed at San Pedro. Photo : Bruce Purser
Morpho rhodopteron (GODMAN & SALVIN, 1880) and endemism on the Sierra Nevada de Santa Marta
The limited number of studies of butterflies at Santa Marta all stress the high degree of endemism which characterises this geographically isolated massif. The author confirms these observations, the best example being Morpho rhodopteron. This remarkable insect closely resembles Morpho sulkowskyi, a butterfly which also frequents bamboo growing between 2000 and 3200 meters throughout much of the Andean cordillera. Morpho sulkowskyi occurs in isolated colonies, there being at leats 9 sub-species (ref. 3). However, M. sulkowskyi has not been observed on the Sierra Nevada de Santa Marta where it is replaced by Morpho rhodopteron whose two subspecies have been observed by the author. The NW form, M. rhodopteron schultzei (ex. nevadensis) (LE MOULT & RÉAL, 1962) observed at Campano and San Pedro, is a slightly darker (Fig. 8) and the marginal bands on both wings are slightly broader than those of the SE subspecies M. rhodopteron rhodopteron (GODMAN & SAVIN, 1880) (Fig. 9). However, differences between the two subspecies are more marked in the females (Fig. 10). Again, the northern form (schultzei) is considerably darker and the marginal bands on both the dorsal and ventral surfaces of each wing are considerably wider in schultzei. Furthermore, the shapes of the fore-wings differ slightly, those of the southern form, rhodopteron, being more pointed. Together with their somewhat different habitats, the two variants of Morpho rhodopteron are fairly easy to distinguish. One may ask therefore whether these variants are subspecies or true species ?
M. schultzei and rhodopteron do not seem to be sympatric. However, if these two variants are only subspecies there should exist a physical barrier preventing mixing. Because both variants occur on the same massif, there being no major topographic barriers, there is no obvious reason why the two presumed subspecies should not inter-breed. However, individualisation may not necessarily be due to geographic isolation but rather to climatic or vegetal barriers. The pedology (soils) on most of the NW, W, and SE parts of Santa Marta are developed on crystalline rocks whose deep (20 m), pink-coloured weathering, even within the arid, SE parts of the massif, is typical of hot, humid climates. Today, these conditions are confined to the northern parts of Santa Marta where they are associated with Morpho rhodopteron schultzei. These observations are pertinent for they imply major climatic changes, notably in the S and SE. Therefore, one may assume that the Sierra de Santa Marta was clad originally in tropical forest during Quaternary times. A climatic change involving retreat of the tropical forest and widespread aridity, even at relatively high altitudes (2000 m) in the SE, was accompanied by a minor evolutionary change leading to the development of Morpho rhodopteron rhodopteron.
The two existing species of Morpho rhodopteron seem to be related to specific biotopes and climates, schultzei occurring on the humid montane forests in the NW, rhodopteron on the dryer, SE slopes. Faunas of the humid pre-montane forests occupying the northern slopes of Santa Marta (Fig. 1) do not appear to have been studied and the possibility of additional subspecies of Morpho rhodopteron can not be excluded.
1. ADAMS M.J. (1973), « Ecological zonation and the butterflies of the Sierra Nevada de Santa Marta, Colombia », Journal of the Natural History Soc., London, 7, 699 – 718.
2. ATTAL S. & CROSSON DU CORMIER A, (1996), The genus Perisama, Science Nat. Venette, France, 149 pp.
3. BLANDIN P. (2007), The Systematics of the Genus Morpho, Hillside Books, Canterbury, 277 pp.
4. COURTIN B. (2009), verbal information.
5. URIBE et al., La Sierra Nevada de Santa Marta, Editions Mayr & Cabal Ltd., Bogotà.