Chapter Thirteen
INTRODUCTION
Sumatran towns are not nearly as interesting ecologically as they could be because their vegetation is largely foreign and therefore supports few birds, squirrels and other animals. Many of the larger animals that are able to live in Sumatran towns are shot as soon as they are in the sights of an air rifle and inappropriate pesticide use probably also takes its toll. Towns in India are often home to large numbers of animals, including monkeys and squirrels, because the human inhabitants have a religious respect for these forms of life. India is a useful comparison, because the high human population density and the effects of development there cause more serious environmental problems than currently exist in Sumatra.
Urban ecology is an underexploited field of study. For example, some trees are better at exploiting certain urban conditions than others, some urban animals and plants can be used as indicators of pollution, some urban rats carry transmittable diseases, and some trees support a wide range of other harmless plants and animals, thereby making the environment more interesting. This type of knowledge should be sought by urban planners for they are charged with the responsibility of creating a healthy and fulfilling human environment.
A great deal of ecology can be taught in towns and cities. Studies of common toads or house geckos (pp. 405 and 411) are unlikely to change any national policy, but they are useful for learning about population structure and dynamics, and thus increasing the awareness of the way animal populations function. Urban lakes can be used for studying phyto-plankton production. Bird watching is a recreation, but records kept over months or years can provide information on seasonality, new invaders, appropriateness of certain city trees, control of birds at airports and adaptations to a man-made environment. Some studies may be economically useful, such as the prevention of algae growing on white-washed and painted walls (p. 404).
As the proportion of Sumatra's population living in town increases, so urban ecologists will face great challenges to provide living space which is functional yet attractive, dynamic, and interesting.
VEGETATION
It may be felt that the vegetation of parts of Sumatran cities is really quite diverse. This is largely because we spend so much time in cities and thereby become familiar with the plants. If we spent an equivalent length of time in any of Sumatra's forested ecosystems it would soon become clear how poor city vegetation is in comparison.
How many tree species are there in Sumatran towns and where do they come from?1 The ubiquitous acacia Acacia auriculiformis comes from Thursday Island in the Torres Strait. The tall mahogany Swietenia macrophylla comes from tropical America. The teak Tectona grandis comes from Burma and Thailand. The origins of the tamarind Tamarindus indica are uncertain but are possibly east Africa and western Asia. Of the common urban trees probably only the she-oak Casuarina equisitifolia is indigenous to Sumatra.
The two most common urban trees, acacia and she-oak, were chosen for good reasons. They have strong roots, grow quickly, give shade, require little attention, are evergreen and quite attractive, can grow in dry exposed habitats and both possess the capability of fixing atmospheric nitrogen in root nodules. This last property means that the trees can grow on young soils and help to improve the soil for other trees. But these advantages are not confined to the two species above. All the leguminous trees fix nitrogen and drought-resisting properties are only necessary in the most open places. If just some of the thousands of indigenous Sumatran trees, including those with fruit eaten by birds, were brought into towns, the fauna might be rather more interesting. Try identifying roadside and garden trees - the total will probably surprise you - to acquaint yourself with plant characters and the classifications of plants. Identifying trees in natural ecosystems will then be much easier. Most of the epiphytic ferns are also quite easily identifiable (Piggott 1979).
The number of species of insect herbivores associated with a particular plant species depends on various factors, such as the plant's geographical range, its taxonomic isolation, growth form, palatability, structural complexity, successional status and also on its relative abundance in the prehistoric past and the length of time it has been available for colonisation (Wratten et al. 1981). The acacia used so much in recent re-greening programs originated on a small Australian island and has only recently been introduced to Sumatra, and so it would not be expected to have many herbivores. This may be thought of as a point in its favour, but native trees survive with perhaps hundreds of associated species of insects. It should be remembered that a herbivore does not necessarily devastate its host. In fact, acacia leaves2 in Medan are fed upon by larvae of bag-worm moths (Psychidae) (p. 398), some of which build their bags from near-entire leaves and others (Eumeta) which build small conical bags from circles of leaf epidermis removed from the lower leaf surface.
Quite a wide range of urban and rural plants have some form of compound leaves which 'sleep' at night. That is, at night they hang down and/or fold up so that the leaf surfaces are held close or even touching. Common examples are the sensitive plant Mimosa pudica and young leaves of cassava Manihot utillissima. In 1880, Darwin suggested that this behaviour protected the plants from chilling, but this view did not gain wide acceptance. On page 290 it was stated that leaves with long edges relative to their length (such as compound leaves) would lose heat faster than simple leaves, and the chilling hypothesis has recently been shown to be valid. By experimentally preventing leaves from 'sleeping' it was found that the leaves became colder at night than those leaves which were allowed to sleep. The difference was only about 1°C but this could affect growth rates, and plants whose leaves sleep may have a competitive advantage over those plants with simple leaves (Enright 1982).
One of the most common orchids on big city trees is the pigeon orchid Dendrobium crumenatum. It is not frequently seen with flowers but if one plant is in flower, then all the plants in an area will be flowering. How is this coordination achieved? Similar gregarious flowering in bamboo was found to be genetic (p. 375) but the gregarious flowering of dipterocarp trees is probably a response to some environmental cue, possibly water stress (p. 221). The major trigger for pigeon orchids is a rapid fall in temperature, such as occurs during a rainstorm. Nine days after that temperature drop all the pigeon orchids will bear the white, sweet-smelling flowers. This was demonstrated in Medan during July 1983 and on the nights the stimulus occurred, the temperature fell below 20°C (fig. 13.1). Cold night temperatures alone (without a rapid drop) are not the primary stimulus, however, and analyses of meteorological data have been published (Burkill 1917; Coster 1926). The pigeon orchids presumably flower simultaneously to increase the likelihood of cross-pollination by the pollinating insects.
Plots of land awaiting development, roadsides, the gardens of unoccupied houses, and neglected corners of towns represent opportunities to study plant succession (Gilbert 1983). Which plants colonise open ground? How quickly does humus form? Assuming the first plants are grasses and herbs, how long until the first woody plant appears? Measure samples of the above-ground biomass at different times. Is the rate of increase constant? Does it show changes related to the seasons? What animals are associated with different stages of the succession?
Figure 13.1. Minimum night temperatures in Medan during part of 1983 and the start of flowering in the pigeon orchid.
LIFE ON WALLS
Algae
White-washed or emulsion-painted walls quickly discolour. Areas of green, black and orange appear in patches or streaks, some in shade, some in the open. The 'stains' are caused by green and blue-green algae and by diatoms (p. 134). In Singapore the distribution of these microscopic plants on buildings has been investigated by Chua et al. (1972) and by Lee and Wee (1982), who also provide an identification key. The black stains are usually blue-green algae which have caught airborne dust particles in their mucilaginous sheaths. Some of the algae are able to fix atmospheric nitrogen and as more dust and soil particles become attached, so a favourable habitat is formed for higher plants to colonise. This obviously takes time, and regular maintenance will prevent succession from proceeding.
Try to establish the ecological requirements for different species of algae. Note the type and age of the substrate, the compass direction it faces, the times it is in the shade, etc. Clear small circles from the middle of a patch of algae and 'transplant' another species. Follow the success of the two species. Reverse the experiment by transplanting some of the first algae into a cleared patch of the second algae. Do certain paints or colours remain free of algae for longer than others? Paint manufacturers would be interested in the results.
Geckos
There are few houses or other buildings in towns without a resident population of geckos. Their ecology is not very well known and among the earliest and most complete appear to be those of Church (1962), Church and Lim (1961), and Chou (1975, 1978). Geckos are unusual amongst the animals which live alongside man in that they are not, for the most part, regarded as dangerous or undesirable.
It is a frequent matter of debate how geckos manage to climb on vertical surfaces or even walk upside down on a ceiling. Gecko feet do not have suckers but instead have small, overlapping flaps of skin (fig. 13.2). These flaps are covered with minute, closely set hairs which make contact with the slight irregularities of a surface and enable geckos to cling where other animals would fall.
Geckos are also known for their ability to shed their tails when caught-a response designed either to enable them to break free if caught by the tail, or to distract the 'predator' with a wriggling tail, or perhaps both. Tail shedding is also known in some snakes and lizards and involves muscular contractions which cause a fracture to occur across a vertebra, not between two vertebrae. The tails which regenerate are not usually as long or as symmetrical as the original; for a discussion of the adaptive strategies and energetic costs involved in tail loss, see Vitt et al. (1977).
There are three common species of gecko in houses over most of Sumatra - the common house gecko Hemidactylus frenatus, the flat-tailed gecko H. platurus and the four-clawed gecko Gehyra mutilata which has no claw on any of its inner digits. The much larger (35 cm) tokay Gekko gecko, although quite common in Java, does not seem to be found in the northern half of Sumatra but would be expected to occur in the south as an introduction from West Java. The spotted gecko Gekko monarchus is intermediate in size (about 20 cm) but is not commonly seen. The tokay has a proper voice produced in a larynx but the common gecko call is produced by tongue-clicking. The tongue-clicking serves to space males to prevent overcrowding but has little or no effect on females (Marcellini 1977).
Geckos generally feed on insects but the tokay feeds on smaller geckos and even mice and small birds. Geckos in Medan have also been observed licking the juice from mangoes which have been part-eaten by bats.
Figure 13.2. The foot of a gecko to show the hair-covered folds of skin.
The three common house gecko species are more or less the same size ±11 cm - what forms of competition and niche separation allow them to coexist? Kill several of each species at intervals of several weeks over a period and analyse the stomach contents. Without necessarily identifying the animal food remains to species or even family, it is possible to detect consistent differences in composition of size of prey between the gecko species. Watch the geckos - does one species sit and wait for prey and another species search? Does one species become inactive and another active in the middle of the night? How do species space themselves around light bulbs? Is aggression shown within or between species? Are resting places (in cracks, behind pictures and mirrors, under stones, in atap roofing, in corners of ceilings or behind cupboards) different between species? For clues and guides see Church and Lim (1961), Pianka and Huey (1978) and Pianka et al. (1979).
LICHENS — MONITORS OF POLLUTION
Lichens are not strictly a group of plants but rather the results of a mutually beneficial relationship, or symbiosis, between a fungus and an alga. The dominant partner is the fungus and it derives its nutritional needs from the alga trapped within its strands. Despite their dual nature they behave like a single organism and for convenience are treated as such (see description of coral polyps on p. 126).
Figure 13.3. Five forms of lichen, a. leprose; b. crustose; c. squamulose; d. foliose; e. fruticose.
After Hawksworth and Rose 1976
Lichens of the Sunda Region are poorly known but without knowing their specific names they can be divided into various growth forms:
leprose - powdery collection of fungal hyphae (root strands) and algal cells with no organisation;
crustose - crust-like with algae situated below a distinct layer of fungal material;
squamulose - crustose but with the margins raised above the substrate surface;
foliose - like a leaf with distinct upper and lower layers, often attached by hair-like 'roots' but easily peeled from the substrate;
fruticose - erect or beard-like, attached to the substrate only at the base (fig. 13.3) (Hawksworth and Rose 1976).
Lichens are probably the longest-living organisms on earth. In the cold wastes of the Arctic, lichens have been found which are over 8, 000 years old, and in a part of Antarctica where it is warm enough for growth for only 300 hours per year, a lichen of about 10, 000 years old has been found. For comparison, the oldest known trees are not quite 5, 000 years old.
Even in the tropics, lichens grow slowly - 5 mm of radial growth per year would be a normal rate, although some grow faster. Since many species occur on exposed surfaces they have to withstand considerable extremes in environmental conditions. When conditions are moist, photosynthesis and respiration can occur rapidly but when lichens are dried out in hot sun, such processes halt because, unlike leaves, lichens have no protective cuticle. Mineral nutrients are either received from the rain or from the substrate (Hawksworth and Rose 1976).
Lichens occur on tree bark, rock and other substrates in most natural ecosystems, and different species or groups of species are confined to particular types of substrate. Tree bark, for example, varies in its acidity and so lichens of different tolerance will be found on different trees. In natural and urban ecosystems lichens are often the only organisms occupying a particular substrate. In towns, for example, lichens are often the sole inhabitants of certain types of wall, exposed tree bark, gravestones, building timber, etc.
The environmental interest in lichens stems from their sensitivity to pollutants: no organisms are more sensitive to sulphur dioxide than lichens. They can thus be used as indicators in programs of environmental monitoring.
Sulphur dioxide is a by-product of combustion of coal and some types of oil, and in its various forms (solutions of sulphate, sulphite and bisulphite ions, sulphorous acid, gaseous sulphur dioxide and sulphur dioxide and sulphur trioxide) affects many plants but particularly lichens. The effect of sulphur dioxide is primarily through disrupting photosynthesis. This can cause a reduction in the reproduction and growth rates or in death. This can be recognised by chlorosis, a bleaching of the lichen, and a tendency for it to peel away from the substrate. The centres of lichens (the oldest parts) generally die first. Whitewash, cement containing lime and certain tree barks tend to neutralise the acidic effects on the lichens they bear, whereas substrates with acidic properties such as some other tree barks and sandstones, will quickly lose much of their lichen flora. Some lichens are more resistant than others to the damaging effects of sulphur dioxide, but the mechanism of this resistance is complex (Hawksworth and Rose 1976). Thus, if a survey transect is established outwards from a source of sulphur dioxide pollution, the following effects are likely to be found:
• the quantity of sulphur in the lichen tissue will decrease;
• the size of a given species of lichen will increase (fig. 13.4);
• the species assemblage found on a particular type of substrate will change and probably become less diverse.
Lichens show similar accumulation of, and sensitivity to, other pollutants such as fluorides, radio nuclides and heavy metals. Many species accumulate heavy metals without overt effects and thus metal content in lichens will generally show a gradual reduction with increasing distance from, for example, a metal smelter.
Figure 13.4. Changes in size for a species of lichen at different distances from a source of pollution.
After O.L Gilbert in Hawksworth and Rose 1976
The study of lichens, let alone their use as pollution monitors, has barely begun in Sumatra or even elsewhere in Southeast Asia. It is obviously preferable to know the identity of different lichen species before beginning environmentally oriented studies. Even if scientific names have not been determined, temporary names referring to colour, shape or location, or simple codes, can be given.
It is important that a reference collection be made so that, at least within one group of researchers, names can be standardised. Whole specimens, not fragments, should be collected, dried gently, and placed in individual air-tight containers containing silica gel or some other desic-cant.3 The containers and specimens should be handled carefully because lichens can be very brittle when dry.
The following preliminary studies are suggested:
1) Count the number of lichen species on a single type of substrate. Repeat in other areas or along a transect suspected of being a pollution gradient. Plot the number of species on a graph with distance from a possible pollution source along the horizontal axis. Suitable substrates for study include tombstones, trunks of a single tree species, walls of buildings, roof tiles, and milestones. Care should be taken that the samples of substrates chosen are of more or less similar age and aspect to the sun. If it is possible to conduct a number of transects around a possible pollution source, a 'contour' map may be drawn of the number of lichen species.
2) Decide on one particular and relatively abundant species of lichen and measure the concentration of sulphur or metal in this species along a transect from a possible pollution source. Display results graphically as in 1).
3) Measure the size of one particular and relatively abundant species of lichen on one substrate of similar age along a transect as above (Hawksworth and Rose 1976).
Sulphur dioxide levels in Sumatran towns are not particularly high (± 1 ppm was measured in Medan) but lichens on trees are clearly killed by some agent near busy and confined roads. The death of lichens can have unexpected effects, as is shown by the following story of the peppered moth. Soot originating from uncontrolled gaseous factory effluent killed many lichens on tree trunks around Britain's industrial centres up to the middle of this century. The death of one particular light-coloured lichen species resulted in interesting changes in the peppered moth Biston betularia. This is generally a pale-coloured moth with flecks of black on its wings which afford good camouflage when it rests on certain lichens. Black forms of this species, however, were found to be more common than the normal pale form in the industrial areas where pollution had killed the lichens. Experiments by Kettlewell (1955, 1959) demonstrated that the variation in the abundance of the two colour forms was caused by natural selection, one of the main mechanisms by which evolution occurs. He placed equal numbers of light and dark forms onto tree trunks in two forests - one in a polluted area and the other in an unpolluted area. He observed the frequency with which predatory birds caught the moths and found that light forms in the polluted area and dark forms in the unpolluted area were more susceptible to being eaten (fig. 13.5). His results are shown in table 13.1.
Thus, over a long period, pollution caused changes in the genetic corn-position of a moth population, and insectivorous birds acted as the agents for the change. Controls of factory emissions in Britain began to come into effect in the early 1950s and, as one might predict, the light form of the moth and the lichen are becoming common again (Cook et al. 1970).
From Kettlewell 1959
Figure 13.5. Both forms of peppered moth on unpolluted (left) and polluted (right) bark.
If Sumatra's lichens have been barely studied, studies of their associated fauna have not even begun. The example above of the peppered moth comes from Europe but it serves to illustrate the perhaps unexpected dependence of one component of an ecosystem on another. Soot is not a serious pollutant in Sumatra but the death of lichens caused by other pollutants might well be expected to have more widespread effects than simply a change in the distribution of that lichen.
DITCHES
To an engineer, urban roadside drains are simply means of preventing floods and of removing household water to large water courses. To an ecologist, a ditch is a simple, small river, the life in which can give indications of water quality.
Perhaps the most obvious animal in many ditches is the guppy Poecilia reticulata (fig. 13.6). This small fish is a native of South America but was probably first introduced to Indonesia as an unwanted aquarium fish. It is mainly an algae eater but its growth rate is greater if the diet includes animal material such as insect larvae (Dussault and Krammer 1981). Guppies do not lay eggs like most fish but give birth to small fry.
Guppies are most obvious in urban ditches when they mouth at the surface, looking as though they are breathing air. Some fish can use atmospheric oxygen but the guppies, and many other fish which exhibit similar behaviour, are in fact taking in oxygen-rich water from the air-water interface (Kramer and Mehegan 1981). This aquatic surface respiration is generally only used where oxygen levels are low (such as slow-moving ditch water with a high organic content), and when oxygen levels are raised the fish respire normally. Aquatic surface respiration does not allow guppies to stay alive for long periods in highly deoxygenated water (Kramer and Mehegan 1981) but it confers an advantage such that guppies can survive where other fishes cannot. Thus guppies can be used as an indicator species.
Investigate different urban ditches, and try to catch one of each of the fish species present. It is more informative if the fishes caught can be given a scientific name (Kottelat et al. 1993), but simply the number of different species provides useful data. Plot a graph of the number of species against the biological oxygen demand or the oxygen concentration of the water. Repeat in a different section of town and in irrigation canals. You now have a means of determining approximate oxygen levels in the field using biological indicators instead of resorting to expensive laboratory tests. Johnson (1968) suggests some other organisms which could be used as pollution indicators.
Another ubiquitous inhabitant of ditches and their surroundings is the common toad Bufo melanostictus. An average-sized garden in Medan can harbour nearly 30 toads and these can be individually recognised by clipping toes and noting colour (variable in this species), weight or length. These data can be used to estimate the population size (Caughley 1978). Do the toads grow at a constant rate throughout the year? Is there a clear breeding season (as indicated by large numbers of small toads appearing at about the same time) or not? Kill a few and examine the stomach con-tents- does the diet change through the year? Take a stretch of road and remove toad corpses. Make daily counts of the toads freshly killed. What proportion of the toad population is killed by cars each year (Gittins 1983)? What other forms of mortality are important?
Figure 13.6. The guppy Poecilia reticulata; female above (about 4 cm), male below (about 2.5 cm). The males found in ditches are very variable but not as fancy as those bred for aquaria.
BATS
Some cities, such as Medan, have huge populations of resident bats, but others, such as Palembang, have rather fewer. Some bats, usually insectivorous ones, often roost in roofs of houses, churches and mosques. Others, both insectivorous and frugivorous species, live amongst palm fronds, particularly the hanging, dead leaves of oil palms, and in tree holes. A Sumatran city is home to probably 15-30 bat species.
At first sight, bats in flight all look the same but with a little patience different groups can be distinguished (Gould 1978b). The medium-sized, roof-dwelling, long-winged tomb bat Taphozuos longimanus is usually the first to start flying. Before it leaves its roost it becomes quite vocal and this is easily heard in the house below. It and other smaller insectivorous bats fly rather erratically as they swoop to catch insects. Frugivorous bats generally fly in a more direct manner.
Stand outside at dusk and watch for bats and note the time the different types appear. Where do they seem to come from? Finding roosts is sometimes easier at dawn when bats tend to fly round and round their roost.
Over a series of days try to pin down the actual roosts. Do the bats emerge at the same time each day? What effect does rain have? Are roosts available all over the city or are they concentrated in one area? How many bats emerge from each roost? Is it the same number day after day? Take a series of 1 ha plots in a town and collect data to estimate the number of resident bats. From data in Lekagul and McNeely (1977) and Medway (1969), estimate the biomass.
BIRDS
Introduction
In general the bird fauna of towns tends to have a lower species richness and diversity than nearby forests, but a higher biomass and density, and a very few dominant species. In addition, the major feeding niche shifts from bark- and canopy-insect eaters to ground-feeders (Ward 1968).
No comprehensive list of birds appears to have been compiled for Sumatran towns despite the ease of observation and data collecting, but lists for Kuala Lumpur and Singapore can be used as indications of what might eventually be recorded. A two-year study of birds in Kuala Lumpur revealed the presence of 24 species of common resident birds, nine species from rural areas which visited towns irregularly, one species which remained only between April and September, and eight species which were resident only between September and April. In addition to these, about 20 other species were seen only occasionally (McClure 1961). A similar number and composition of species was recorded for Singapore and it was noted that the total number of species was lower than for towns in other tropical areas, for instance west Africa (Ward 1968). An examination of the natural habitats of urban and suburban birds in Peninsular Malaysia and Singapore suggests that over half originate in coastal and riparian vegetation, only about 5% from lowland forest and a similar percentage are normally cliff or cave-mouth nesters. About 25% have been introduced or are recent immigrants. The similarities between cliffs and buildings are obvious and swifts have taken advantage of that. The similarity between coastal and riparian vegetation and towns is less clear, but a common factor is their simple plant communities with few species (p. 402) (Ward 1968). Because of this, generalised foragers are the major town invaders. The most common invader from the above habitat is the yellow-vented bulbul Pycnonotus goiaveri. Unlike Africa, the Sunda Region has no large areas of natural open country or savannah which might be expected to form a source of urban birds, and so the number of urban birds originating from indigenous natural habitats is limited.
For certain birds from the dryer land north of the Sunda Region, roads, railways, disturbed vegetation and towns present opportunities for colonising areas which had previously been closed to them because of the intervening large areas of species-rich forests (Ward 1968). This has yet to be documented for Sumatra but birds from the dryer, more open land in Java are starting to spread into Sumatran towns (Harver and Holmes 1976; Holmes 1977).
Sumatra and Peninsular Malaysia have witnessed a spectacular invasion by the tree sparrow Passer montanus. This bird is a native of the Palaearctic Realm (fig. 1.23, p. 41), that is Europe, Russia and China, and probably arrived at ports aboard ships in the sixteenth and seventeenth centuries. It has now spread to every urban area on Sumatra. A similar spread of a new species may occur if the introduced house crow Corvus splendens (not to be confused with the shy forest crow Corvus enca) or the common mynah Acridotheres tristis spreads from the island of Penang across the Malacca Straits to northern Sumatra (Charles 1978). The crow fills the scavenging niche which in most other tropical regions is also filled by various crows and birds of prey.
Those concerned with environmental affairs in Sumatra should keep an eye open for species new to a region. They are likely to occur first in the coastal urban areas. Even if they are unlikely to compete severely with indigenous birds, they may be considered undesirable for other reasons and the time to undertake a management program is at the beginning of an 'invasion', not when the 'invaders' have established a firm foothold. House crows are somewhat unpopular because of their raucous calls, gregarious nesting habits, fouling of public places, and stealing of food or young chicks (Charles 1978), but they do process urban waste, which might otherwise be utilised by rats which, unlike crows, harbour and transmit several important diseases (Hadi et al. 1976) (p. 390). Charles (1978) has concluded that one of the best ways to control these birds is by controlling the disposal of household waste, which would clearly have other benefits.
It is sometimes remarked that the total number of birds, not just the number of species, is low in Sumatran towns. One reason, of course, is the thoughtless shooting by the air-rifle toting, motor-bike riding, urban cowboys on Sunday afternoons. Additional ecological reasons are that few of the urban trees produce fruit suitable for birds and there are few insects able to utilise the 'foreign' trees and therefore there is less food for insectivorous or partially-insectivorous birds (see p. 223).
Birds represent excellent subjects for a study of urban ecology. Observation conditions are as near ideal as one could wish for and the number of food species and competing bird species are relatively few. The study of an urban bird community in Sumatra is unlikely to be of island-wide environmental significance, but it will furnish those involved in such a study with an awareness of ecological complexity and principles (see, for example, Ward and Poh 1968). These will be of immediate practical use when those involved turn to a more complex ecosystem, an environmental impact assessment or similar study.
Flowers that are pollinated by birds are generally recognisable by the following features:
1) open during daylight;
2) vivid colours, often scarlet or striking contrasting colours;
3) lip or margin absent or curved back;
4) hard flower wall, filaments stiff or united, nectar retained at rear of flower;
5) no odour;
6) abundant nectar;
7) no nectar guides (lines running along the petals indicating nectar source);
8) a relatively large distance between nectar source and sexual parts (Faegri and van der Pijl 1979);
9) relatively low sugar concentration (± 25%) in the nectar in comparison with nectar in flowers pollinated by bees (± 75%) (Baker 1975).
A common urban flower which fulfils all these criteria is the hibiscus Hibiscus rosa-sinensis. It is visited by sunbirds (Nectariniidae) but, because most hibiscus plants are sterile, the plants do not seed (van der Pijl 1937; Prendergast 1982). What other plants do sunbirds drink nectar from? Do all the plants conform to the characters listed above? Are sunbirds the only birds to visit the flowers? Design experiments to ensure that only sunbirds gain access to certain flowers. Repeat the experiment but exclude all possible pollinators. Are the birds actually pollinating the flowers? Where on the bird's body is pollen carried and what features of the flower design ensure that pollen is transferred?
Swifts
Swifts and swift-like birds belong to three families: the swifts and swiftlets (Apodidae), treeswifts (Hemiprocnidae), and swallows (Hirundinidae). Only the first and last of these are particularly common in towns and the two most common species can be distinguished from each other as shown in table 13.2 and figure 13.7. Other species occur in towns, however, and for proper identification reference should be made to King et al. (1975).
Swifts are masters of the air. A swift in Africa has been recorded as flying at 170 km/hr and is easily the fastest-moving vertebrate. In normal flight, however, swifts usually fly at about 50 km/hr. Swifts feed, drink, mate and even sleep while flying. As dusk approaches, non-breeding birds rise steadily in the sky and probably reach a few thousand metres above the ground where they sleep and beat their wings irregularly. When a young swift leaves its nest it may be at least two years before it lands again. Some swifts migrate to and from the cooler northern climates. The oldest known wild swift was ringed as an adult 15 years before it was found dying in England. It was calculated that it had flown over 7 million km during its life. Swifts normally live about six years (Bromhall 1980).
Figure 13.7. House swift (left) and barn swallow (right) in flight.
After King et at. 1975
Swifts feed by chasing or filtering insects and airborne spiders from the air. They are, in effect, feeders on 'aerial plankton'. The food that adult European swifts give to their chicks has been examined and produced some surprising figures (Bromhall 1980). A single mouthful collected over half an hour or so might contain over 500 animals of many species. During a fine day a pair of swifts with two or three young to feed may catch up to 20, 000 insects and spiders (Bromhall 1980). Under the eaves of the grand Central Post Office in Medan, there are some 100 swift nests which are occupied during the breeding season (approximately April to June). Swift chicks generally leave their nest when about six weeks old and a simple calculation shows that the breeding swifts and their chicks account for about 40, 000, 000 insects and spiders during that six-week period. It is not difficult to estimate how many insects would be consumed in one year by this one colony of swifts alone and thus how important the birds are to insect control.4
If the nest sites chosen in Medan are any indication, it is clear that modern buildings are not favoured. Most of the nest sites were built at least 30 years ago and are used repeatedly. When the Medan Central Post Office was repaired in 1982, all the nests were scraped away from the eaves. It was not many months, however, before nests were rebuilt in the same places.
Examine the locations used by swifts for nesting. What features do they have in common? How many nests are there in each location? Is there any correlation between number of nests and age of buildings? Is there any connection between direction faced by the wall and the number of nests? Can you suggest ways in which modern buildings could be made more attractive to swifts? Do the numbers of nests occupied vary from year to year? How many nests are occupied in which months? How does the number of occupied nests relate to rainfall or other patterns? Do the patterns tie in with the cycles discussed on page 222?
It is hoped that these few examples will instill some enthusiasm in readers to venture into the urban environment with newly opened eyes, a questioning mind and a notebook. Sumatra desperately needs a wider and deeper data base of ecological information and students, school teachers, and university lecturers should take up the challenge now.