University of Glasgow

Ecuador Expedition 2002

Final Report

Edited and Compiled by Stewart White

DEEB, Graham Kerr Building

Glasgow G12 8QQ

E-Mail: s.white@bio.gla.ac.uk

Introduction

The University of Glasgow Ecuador Expedition 2002 was the second of what we hope will be a series of research visits to Ecuador.

Our main contacts in the country are Dr Giovanni Onore and Darwin Garcia. Giovanni owns the Otonga Research Station in the Andes south-west of Quito and has been welcoming foreign researchers for some time.

The work at Otonga comprised both ornithological and entomological studies. The main part of the ornithological work was to continue building up the species list for the site. Two smaller projects studying hummingbird feeding behaviour and comparing bird numbers and species at different altitudes were also conducted. The insect project was looking at biodiversity of Diptera in Andean cloud forest habitats

Darwin Garcia is the representative of the Indian tribe that owns the Sumaco Reserve. He is the son of a missionary and a member of the local tribe, who has returned to the area after completing a teaching qualification, and now represents the tribe. The tribal lands have been made into the Sumaco Reserve and the tribe wants to generate income from Eco-tourism. They have already turned down offers from outside companies to run tours in their area and recently singed an agreement with Aalborg Zoo in Denmark for long term financial support. Darwin has recently started his own business bringing tourists into the area. Only the bird group traveled to Sumaco, the main task was to continue building up a species list for the reserve. The information we have discovered in 2000 and 2002 will be made available to Darwin to help him develop his eco-tourism business, support the tribe and also protect the reserve.

Bird Research Project Reports

The Neotropics

The Neotropical region has long been recognised as supporting a high level of biological diversity. Of the world’s 9,700 bird species no fewer than 3,600 (39%) occur in the non-Caribbean Neotropics (16 % of the planet’s land area) (Wege and Long, 1995). 290 of these species were, in 1995, considered at risk of extinction (Wege and Long, 1995) due to a range of anthropogenic disturbances: timber extraction, clearing for agriculture, drainage of wetland, mining, and other disturbances. Pristine habitats are becoming increasingly disturbed and fragmented. There is therefore increasing urgency to do something to halt or slow down the loss of species. One important task is to gain as much information as possible about the organisms dwelling in the remaining undisturbed areas of the Neotropics. The bird studies in the University of Glasgow Ecuador Expedition 2002 were an attempt to build on the information gathered in 2000 from two pristine sites in Ecuador, one in the Andes and one in Amazonia.

Ecuador

Ecuador has around 1,530 resident and migrant bird species (Ortiz and Carrión, 1991). Of these c.37 are endemic, 160 have restricted ranges, 6 are classified as critical, 13 as endangered, 43 as vulnerable, 46 as near-threatened and 3 as data deficient (Stattersfield et al, 1998; BirdLife International, 2000).

Endemic Bird Areas (EBA’s) have been identified by mapping the distributions of birds which have had, in historical times, breeding ranges of less than 50,000 km². An area supporting at least two of these restricted range species is identified as an EBA, a place where global extinctions are likely to occur unless the habitat can be protected (Wege and Long, 1995). Nine of these EBA’s lie partially or wholly within Ecuador. In addition 50 Key Areas, the most important places for globally threatened bird species in the Neotropics (Wege and Long, 1995), have been identified in Ecuador.

In summary Ecuador has an extensive, valuable avifauna much of which is under threat. The more that is known about it the more able we will be to protect it for future generations to enjoy.

Ornithological Survey of Otonga Forest Reserve

Introduction

Otonga Forest Reserve lies in the Andes to the South West of Quito, at 79^o 00’ 00” West and 0^o 25’ 00” South. It is at an altitude of 2100m and is an area of pristine cloud forest, disturbed only by narrow trails and lies in the Chocó EBA. The study had two main targets; first of all to capture and band as many birds as possible as part of what is hoped will be a long-term capture-recapture project; secondly to compile as complete a species list as possible to complement work already done in the reserve.

Methodology

Mist netting, visual observations and sound recording were conducted over a twenty one day period in July and August 2002. Five 18m long x 2.5m high North Ronaldsay mist nets of standard 33mm mesh were used each day. Nets were erected on paths and in clear areas near the field station and opened at dawn, around 06.00 and left running until late morning, from 11.00 – 12.00 depending on weather conditions. In all over 50 net sites were used, at three distinct altitude levels, in a variety of positions, on tracks, across the side of and down slopes and on the ridge at the top of the reserve. The weather followed a regular pattern in that most days cloud cover would descend in early afternoon, soaking the nets and reducing bird activity to a minimum. There was no standardised regime for visual observation of birds. All members of the team kept a lookout for birds at all times. Sound recording was conducted for a period of 30 minutes on several days in the period just after dawn. All identifications were made using reference books on South American avifauna (Ridgely & Tudor, 1989, 1994; Ridgely & Greenfield, 2001a, b) and bird calls identified from reference tapes (Parker & English, 1992; Moore, 1994, 1996 & 1997).

Results

In all 379 birds in 48 species were captured and 490 birds in 75 species observed, as listed in Appendix 1. The tapes of bird calls are still being analysed.

Species of particular note observed were the Giant Antpitta Grallaria gigantea which is known from only a small number of locations in a very small range and is classified by the ICBP as endangered, the Moustached Antpitta Grallaria alleni which has a small and contracting range due to habitat destruction and fragmentation and is also classified as endangered, the Plate-billed Mountain-toucan Andigena laminirostris, the Toucan Barbet Semnornis ramphastinus and the Hoary Puffleg Haplophaedia lugens, all classified by the ICBP as near threatened (Birdlife International, 2000). In addition, the Brown Inca Coeligena wilsoni, Violet-tailed Sylph Aglaiocerus coelestis, Toucan Barbet, Dusky Bush-tanager Chlorospingus semifuscus, and Plate-billed Mountain-toucan are all restricted range species (Stattersfield et al., 1998).

Altitudinal Effects on Avian Diversity and Abundance

in an Ecuadorian Cloud Forest

Introduction

While on location at the Otonga Biological Reserve on the west slope of the Andes in Ecuador conducting the research described above a study was carried out to assess the variation in diversity and abundance of avian species with changes in altitude using a variety of techniques.

The aim of this study was to determine whether significant differences in the number of species and number of individuals of each species occur with changes in altitude, and if there are differences, investigate why they occur.

The study area was divided into three sites of varying altitude: the river site (1720-1880m), the field station site (1971-2027m), and the ridge site (2112-2193m). Species lists for each site were built up using mistnet capture data collected as a group and individual Mackinnon lists based on observations. Sound recordings were also taken at each site but due to complications at the identification stage they were not used in the analysis.

In addition to the avian diversity data, some information was collected about the habitat at each site. This included estimates of vegetation cover, tree height, circumference, architecture and density. This was used in addition to the other data as a comparison between sites to provide an insight into why certain species occurred where they did.

Results

On analysis of the results it was found that the field station produced the most species and individuals, followed by the ridge. This was to be expected due to the mid point location of the field station and overlap of habitats and also the fact that more time was spent in this area collecting data during the training period. However, relative species diversity was greatest at the ridge as it produced the most species from the fewest individuals. A species discovery plot was constructed and shows that for the same number of individuals detected the ridge would be likely to produce more species, this can be seen in graph 1.

The species from each site were also divided into feeding guilds and this information used, in relation to habitat, to explain why birds inhabited certain sites. It was found that insectivores vastly dominated the river site with all other guilds together making up less than 50% of the species found. This coincides with the information from an entomological project carried out in the same area, which found that insect diversity and abundance was greater at the lower altitudes around the river.

On analysis of the habitat data gathered it was found that habitats at each level were similar in terms of vegetation cover and tree height, architecture and diameter. However, tree density was much greater at the ridge providing a denser habitat for the avian species inhabiting this area.

Discussion

Overall, the findings from this study suggest that, although all three study sites provide suitable habitat for the eleven species that occur at all levels, significant variation is present between the lower and higher altitudes in both the species found and also diversity and abundance of species.

It was however noted from a previous study by Poulsen (2002) that high altitude regions in the Ecuadorian Andes are not representative of typical high altitude environments. In terms of avian richness and abundance, these areas are more similar to those of tropical sites than temperate ones.

Kirstin McGowan

Hummingbird feeding behaviour at Otonga Reserve

Introduction

Hummingbirds are very small endotherms that are found throughout the Americas. They can be seen all over the two continents, from Alaska in the north to Tierra del Fuego in the south. They live in almost any habitat between sea level and four and a half thousand metres as long as there are flowering plants in the area. They are known for their unique method of flight which utilises both the up and down strokes of the wing beat cycle, causing the characteristic humming sound of their flight and giving them the ability to hover. These adaptations of wing structure and flight muscles allow hummingbirds to exploit an ecological niche that is closed to most other endotherms, that of a specialised nectar feeder. Their hovering ability, as well as their small size, allows them to feed on hanging flowers that are only available to insects and hummingbirds. Like the insects that feed on nectar producing plants, the hummingbirds also carry pollen from flower to flower and thus help to cross pollinate the plants.

Hummingbirds and the plants they feed on have a close evolutionary relationship. As one has developed a new adaptation, the other has also. It is believed that the plants were first pollinated by insects which are more attracted to blues and purples and, therefore, it is thought that flower colouration has changed over time to allow for an ornithophilous relationship to develop more fully.

The position of these flowers are usually aimed at providing the birds with easier access to the nectar supply. The birds can locate these flowers by sight, either a direct visual, phyto- flagging or possibly with UV cues.

In order to access their food, hummingbirds use hovering flight. They can fly in any direction by varying the tilt of their wings. On average, a hummingbird beats it wings at a rate of about 25 times a second. Such a feat is very energetically expensive in terms of oxygen and food. To meet their energy requirements they feed on 1000- 2000 flowers in a day. In order to do this, hummingbirds of different species and morphologies adopt different feeding strategies to maximise their success. Some use a territorial approach, whereby they defend a patch of flowers, others are trap liners. This is a strategy that involves visiting scattered blossoms by a circuitous route. Territoriality and trap lining are opposite ends of a spectrum and a bird may use a feeding strategy that falls anywhere along that spectrum. Their differences in morphology mean that a particular strategy will be more efficient for that species than any other strategy.

There are many environmental variables that will determine whether or not a hummingbird can satisfy its daily energy requirements without resorting to torpor. One of the main factors is the plants production of nectar. This can vary in one of several ways. The main ways that it varies is in the volume and concentrations of the nectar produced. Changeable volumes bring varying rewards to hummingbirds and the energetics involved in obtaining them differ. There is also a variation in costs of different meal volumes for the different feeding strategies. Volume and concentration in wild plants may not reflect a hummingbird's feeding preferences. Most wild plants produce nectar with sucrose concentrations of about 20- 25%. Hummingbirds prefer much higher concentrations, up to 65%, despite the fact that this is a fairly viscous solution and more energetically expensive, per unit volume, to obtain.

It is possible that a plant will deliberately produce flowers with little nectar, or nectar of low concentrations, to create ‘blanks’. This encourages the birds to carry pollen to other plants as they will spent less time at plants with variable nectar production. Larger quantities of nectar have been found to be produced in plants that have been fed from earlier in the day.

Feeding has generally been found to decrease over the course of a day although feeding behaviour will vary with habitat, flower species available and between hummingbird species.

This study looks to answer three basic questions about hummingbird feeding:

1) Are there differences in feeding behaviour between hummingbird species? If so why?

2) Are they choosing plant species or is it random?

3) Does feeding rate change over time?

Methods and Materials

The study area used for this project was in an area of cloud forest to the south- west of

Quito in Ecuador, surrounding the Otonga field station. Within this area three observation posts were set up that allowed for maximal viewing of the hummingbirds feeding areas. There are a large number of flowering plants in a small area round the field station as the rainforest had been cleared from the area leaving an area exposed to more intense sunlight. This has encouraged a concentration of flowering plants into this small area. Over a period of three weeks data were collected about the hummingbirds feeding behaviours in several parts: the number of feeds over the course of the day; the types of hummingbird species feeding during the morning hours; the height at which the birds were feeding; the types of flowers that they were feeding on; the temperature at the different times of the day and the sucrose concentration of the plants they were feeding on.

During this study the weather made it hard for data sets to be collected for the whole day. Error was minimised by cutting down on the potential error sources. Observer error was limited as all observations were charted in the same manner, by the same person. Also the first few days in the area were spent in familiarisation with the area layout, the plant species and the hummingbird species. This kept observer error to the minimum.

Results and Discussion

Within the Otonga area, over the course of the study, ten hummingbird species were observed- seven regularly. Twelve different types of flower were identified as hummingbird flowers. These could not be identified by name and were, therefore, known by their colours.

Are there species differences?

Most of the seven main species showed a strong preference for the red flowers. The fawn breasted brilliant fed exclusively on this flower type. For all other species, except the booted racket tail, the red flowers made up the majority of their diet. The booted racket tail has a much stronger preference for pink/red flowers. The species have different feeding strategies and therefore feed within the area at different rates.

None of the species show any significant variation in feeding rate over time when a chi squared test is used. Regression plots show that there is no significant correlation of feeding rate with time.

Are hummingbirds choosing flower types?

There is a distinct flower preference indicating that hummingbirds are choosing the flower types that they are feeding on. This result is also reflected in the individual species choices where many of the species in the area are choosing the flower types that make up their diet.

There is also a preference for flowers that are at a specific height. They fed most on flowers that were between one and two metres from the ground. There is no significant difference in height preference over time.

Does feeding rate change over time?

The mean value for the feeding rate data was calculated for each hour of daylight. This allowed for weather changes over the course of the study as this does affect their feeding behaviour. Feeding rates changed significantly over time. In this study their feeding rates were generally found to increase over the course of the day. Previous studies have found rate of feeding to decrease as the day progresses with a pre- dusk burst.

Possible reasons for this result were considered. Mean sucrose concentrations of the nectar and mean temperature were recorded to try to understand these results.

There is a significant change in both sucrose concentration and temperature with time. This would suggest that there is a temperature change throughout the day that causes a change in nectar production by the plants. Changes to the content of the hummingbirds' food could explain the differences seen in their feeding behaviour.

Hummingbirds compete for access to nectar resources so it is not surprising that there are species differences seen in their feeding behaviour. As shown in the results, there are some distinct differences to be found between the species and several explanations for these differences. One of the main differences between species is the birds morphologies, in particular their body size and the size and shape of their bill. The size and shape of their bill will determine which flower corollas that the bird can reach into. Most flowers that hummingbirds feed on have long, thickened, tubular corollas. It is from this part of the plant that the birds get nectar. Flowers with long corollas may prevent short billed hummingbirds from feeding on them. Short billed hummingbirds typically exploit flowers that insects often feed on. Bill shape determines which flowers can be fed on most efficiently. Distribution of such flowers will help to determine which feeding strategy each species uses.

In the Otonga area, during this study, the violet tailed sylphs and speckled hummingbirds were observed as those with the largest diet range. Both these species have short, straight bills that allow them to feed on almost all flower types. Neither species is typically thought of as being territorial, the sylphs in particular as most often trap liners, but in this area both species show a tendency for territoriality. The male sylphs are particularly aggressive in defending their feeding patches. A possible explanation for this may be that there is an extreme abundance of flowering plants in the field station area. This may make it energetically viable to defend these flowers whereas as more dispersed flowers are not easily defensible. The particularly dense aggregation of flowers may cause these species to change their strategies in order to optimise their efficiency in gathering their daily energy requirements.

From the results it is also clear that the flowers types being fed on are being chosen and that feeding is not just random. As hummingbirds and the plants they feed on have a close evolutionary relationship, it is possible that hummingbird preferences may cause their preferred plant types to become more abundant. Competition between species may have the potential to change the plant population and community structures, depending on the birds foraging strategies.

A staggered flowering strategy may help to explain why red flowers feature so predominately in most hummingbird species diets. The time of year in which the study took place may have coincided with the flowering period for this particular plant. The birds will choose plants that are most profitable to feed on, that is the ones requiring least energy output for the greatest energy gain. This is particularly applicable if it is the start of a flowering season as nectar production is a function of flower age in many plant species.

Flower choice may depend on that flower type supplying an optimal meal size. Having to visit more flowers increases energy expenditure so if the number of plants visited can be minimised it is predicted that a hummingbird would try to do this. Nectar production changes over time and results show that feeding rate varies over time. Both these changes may be accounted for by environmental factors. These feeding rate results are very different from observations from previous studies so it is likely that environmental factors are responsible for these differences.

Results also show that sucrose concentrations vary with time and temperature. Hummingbirds will decrease their food intake when sucrose concentrations in nectar are increased. This would suggest that hummingbirds will increase feeding rate as sucrose levels decrease. As sucrose levels vary with temperature it can be reasoned that feeding rate will also vary with temperature. Higher air temperatures cause sucrose concentrations to decline and in this area air temperature rise throughout the day. This would imply that sucrose concentrations will decrease as the day progresses and feeding rates will increase, the feeding rate increasing as food availability decreases.

During this study the climatic conditions of the area affected both the hummingbird behaviour and the ability to observe this behaviour. This limited the extent to which data could be collected. If a longer period had been spent on making observations it may have been possible to get more complete data sets to cover the whole of the daylight hours.

It is possible that some of the results found were a product of seasonality and that a study at a different time of year may have produced different results.

More work is required in this study site in order to test if seasonality affects which hummingbird species show territorial behaviour and which flower type is most dominant. It may also help in data interpretation if the factors affecting plant nectar production are fully identified.

It may be concluded that there are definite differences in feeding between hummingbird species. These differences are influenced by their morphology and their feeding strategy. Their feeding strategy will determine the extent to which they show preferences for particular flower types. Their feeding rate changes with time in order to maintain the hummingbirds' energy intake at a particular level. This rate is affected by costs incurred in foraging and the quality and quantity of the food available.

There are many factors that affect hummingbird feeding behaviour and much more work is required in order to establish exactly what all these factors are.

Alison Steel

Ornithological Survey of Sumaco Reserve, Napo Lowlands

Introduction

Sumaco reserve is a newly created reserve in an area of relatively undisturbed lowland rainforest in the Napo Lowlands area of Ecuador at approximately 77 10' W, 0 25' S. Mist-netting, visual observations and tape recording were used to survey the reserve during August 2002. The first aim was to continue with the work done in the 2000 Glasgow Expedition in building a species list for the area. In addition, all birds captured were marked with an individually numbered ring before release. We intend to conduct a study on longevity by returning to the area over the course of several years and obtaining data on recaptures. The third important aspect of the work was to generate further ecological information for our local contact, Darwin Garcia.

Methodology

Three study sites were investigated for five days each. Five 18m x 2.5 m., 33mm mesh size mist nets were erected at random each day between 06.00 and 12.30 on a grid of points. Each point was 50m from any other on the grid. All expedition members identified and made a note of all birds seen during the day. Tape recordings of calls were made for 30 mins on three mornings in each site. All identifications were made using reference books on South American avifauna (Ridgely & Tudor, 1989, 1994; Ridgely & Greenfield, 2001a, b) and bird calls identified from reference tapes (Parker & English, 1992; Moore, 1994, 1996 & 1997).

Results

A total of 49 Bird species were captured and 98 species observed as listed in Appendix 2. Tape recordings are still being analysed.

The students carrying out the above research were Caroline Blaikie, Neville Broadis, Harriett Chapman, Kirstin McGowan, Alison Steele and Fiona Stewart, supervised by Stewart White, University of Glasgow.

References

BirdLife International (2000) Threatened birds of the world. Barcelona and Cambridge, UK: Lynx Edicions and BirdLife.

Moore, J.V. (1994) More bird vocalisations from the lowland rainforest - Volume One. Nature Recordings.

Moore, J.V. (1996) More bird vocalisations from the lowland rainforest - Volume Two. Nature Recordings.

Moore, J.V. (1997) More bird vocalisations from the lowland rainforest - Volume Three. Nature Recordings.

Ortiz C.F., and Carrion, J.M. (1991) Introduccion a las aves del Ecuador. Quito, Ecuador: Fundacion Ecuatoriana para la Conservacion y el Desarrollo Sostenible.

Parker, T.A. & English, P.H. (1992) Birds of eastern Ecuador. Cornell laboratory of ornithology.

Poulsen, B.O. (2002) Acomparison of bird richness, abundance and trophic organisation in forests of Ecuador and Denmark: Are high altitude Andean forests temperate or tropical? Journal of Tropical Ecology 18: 615-636.

Ridgely, R.S. & Tudor, G. (1989) The birds of South America: The oscine passerines. Oxford. Oxford University Press.

Ridgely, R.S. & Tudor, G. (1994) The birds of South America: The suboscine passerines. Oxford. Oxford University Press.

Ridgely, R.S. & Greenfield, P.J. (2001a) The birds of Ecuador: Status, Distribution and Taxonomy. Ithaca, NY: Comstock Publishing Associates.

Ridgely, R.S. & Greenfield, P.J. (2001b) The birds of Ecuador: Field Guide. Ithaca, NY: Comstock Publishing Associates.

Stattersfield, A.J., Crosby, M.J., Long, A.J. and Wege, D.C. (1998) Endemic bird areas of the world: priorities for bird conservation. Cambridge, UK: BirdLife International (BirdLife Conservation Series 7).

Wege, D.C. and Long, A.J. (1995) Key areas for threatened birds in the neotropics. Cambridge, UK: Birdlife International (BirdLife Conservation Series 5).

Entomological projects in Otonga Reserve, 2002.

Several investigations on insects were carried out by students. Two of these aimed to establish biodiversity differences within the cloud forest reserve using different sampling methods. These were pitfall traps and malaise traps. The former are effective for catching active ground-dwelling nocturnal predators such as Carabidae (ground beetles) and Staphylinidae (rove beetles). The latter sample flying insects particularly Diptera (flies) and Hymenoptera (wasps). The traps were sited in such a way as to provide comparisons between different altitudes within the area and also between primary forest (where no logging or agricultural practises have ever been carried out) and secondary forest (in which natural recovery was taking place after the exclusion of stock animals). The analysis of these data is not yet complete but it will be interesting to see if the different methods reflect the biodiversity equally.

The effect of altitude on insect biodiversity

within an Ecuadorian cloud forest

Altitude and its effect on species diversity has been the subject of many studies (Brown 2001, Janzen et al. 1976, Lawton et al. 1987, Olson 1994, Terborgh 1977, Wolda 1986). Whether the study is on birds (Terborgh 1977), mammals (Brown 2001) or insects (Janzen et al. 1976, Olson 1994) the conclusion of the majority of these studies suggest that diversity peaks at intermediate levels of altitude. However, the conclusions of many insect altitudinal studies are not consistent, with many also suggesting a gradual decrease in diversity with an increase in altitude (Lawton et al. 1987, Wolda 1986). Differences in the scale of altitudinal transects surveyed may account for these differences (Wolda 1986), with the majority covering a relatively short altitudinal range (Herbert 1980, Lawton et al. 1987, Janzen et al. 1976), and relatively few incorporating sites from sea level through to mountain peaks (Wolda 1986).

The present project also attempts to measure the effects of altitude on species diversity using pit fall traps to measure the abundance of insects at different altitudes within the Otonga cloud forest reserve. The hypothesis proposed is that a decrease in insect biodiversity will be observed as altitude increases.

Methodology

Fieldwork was carried out between the months of July and August in the summer of 2002 within the Otonga cloud forest reserve in Ecuador.

Twelve 1m² sites of primary forest on one side of the Otonga valley were chosen along an altitudinal transect, between 1720m to 2193m in height. The twelve sites were positioned so that each zone of the valley side (lower -slope, mid-slope and upper-slope) included four trap sites. Four pit fall traps (7.2cm diameter jam jars) were positioned randomly within each 1m² site using a quadrate. The jars were quarter filled with soapy water and were left in place for three days. Insects and collembolans were removed each day at the same time (11am). Other invertebrate groups were omitted. Once collected the samples were taken back to the field station and separated to order level. Samples from each day were combined later to give a total value for each site. Only the Coleopterans were kept and sorted to family level according to Booth et al. 1996 and then separated in to Identifiable Taxonomic Units (ITUs) or morpho- species back in Glasgow. A dissecting microscope was necessary due to the extremely small size (<2mm) of some Staphylinidae specimens. Separation into morpho -species was based on few characters, for example body length to the nearest mm from the tip of the labrum to the base of the last abdominal segment (according to Joy 1932). Colouration and pattern of the abdomen and thorax were also used to distinguish between morpho-species, as was the size (mm) and shape of the mandibles. Sexual dimorphism was taken into account whenever possible based on the shape of the last abdominal segment, males possessing a long thin terminal segment and females possessing a more rounded segment (Crowson 1981, Hancock pers. comm.).

In order to compare insect diversity between sites the Shannons (H’) and Margalefs indices were used to calculate values of diversity for each site. Unfortunately this was only possible with coleopteran samples as only these were kept and sorted into ITUs back at Glasgow.

Results

The most abundant insects within the samples were beetles of the family Staphylinidae, with 379 individuals collected. This was expected as this large family of coleopterans are, along with carabids and araneids, major components of pit fall samples (Work et al 2002), because of their carnivorous, mainly terrestrial existence. However, only one carabid beetle was collected within this study. A large number of collembolans (196 individuals) and orthopterans (226) were also collected along with lower numbers of dipterans (191), hemipterans (13), hymenopterans (115) and dictopterans (21).

It was found within this project that altitude had a significant effect (p=0.019, RSq= 43.6%) on species richness within the family Staphylinidae, which decreased with increasing altitude over a 473m transect. Altitude was not found to have a significant effect on the number of individuals or diversity of any other group tested. As stated within the introduction, reports on the effect of altitude on insect biodiversity within the literature are split between authors that conclude a gradual decrease with altitude (Herbert 1980, Lawton et al 1987, Wolda 1987) and those that conclude a peak in insect diversity at intermediate altitudes (Janzen et al 1976, Olsen 1994). The findings of this study (gradual decrease in Staphylinidae species richness with increasing altitude) therefore concur with those who support a gradual decrease in insect species diversity with increasing altitude.

Future Work

It would be interesting to increase the length of altitudinal transect used within the project to see if the intermediate peak in species diversity, reported by some authors (Janzen et al 1976, Olsen 1994) actually occurs. However, to achieve this the transect would have to breach the boundaries of the reserve. It would also be beneficial to incorporate another sampling method, possibly swap-netting or malaise trapping, combined with pit-fall trapping so that a larger sample of insects could be collected. This would provide a more accurate representation of the effect of altitude on insect biodiversity within the Otonga reserve and would not be limited to terrestrial diversity.

Kevin William Conway

Insect Biodiversity In The Cloudforest Of Otonga, Equador.

Introduction

The vast majority of biodiversity, biological diversity, on our planet is contained within the small, uncharismatic and thus far understudied taxa such as the Arthropods. Insects make up over half of the worlds described species yet the majority of them still remain undescribed, and the identification of those that have been described requires a trained specialist. The Dipterans are the third most diverse Order of insects, after the Coleoptera and Hymenoptera, with roughly 120,000 species. Furthermore they are vital to various ecological functions such as decomposition, pollination and pest control.

For these reasons it is necessary to devise better methods of sampling wholesale biodiversity quickly and cheaply in order to make intelligent decisions regarding conservation measures. Current methods include using well known species as indicators of underlying diversity and higher taxon richness as a predictor of wholesale diversity.

A habitat-based approach was used to investigate the community patterns in the Otonga cloud forest reserve during the dry season (4/7/02 – 3/8/02) concentrating on the Diptera. The study examines the differences in insect diversity between primary and regenerating secondary forest. It also includes a comparison of the diversity and abundance of insects along an elevational gradient (1736 – 2138m). The study also focuses on the underlying reasons for the observed patters within a wider biological context.

Methods

The reserve is situated on the western foothills of the Andes in the Western Cotopaxi Province. Two malaise traps were used in paired comparisons to passively sample the Dipteran community in the aforementioned habitats. At each new site only one variable was ever changed to reduce confounding factors. Traps were placed in the primary and secondary forest at ridge top, hut level, and riverside positions.

The samples were sorted to Order level whilst in the field and the Dipterans collected for sorting to family level once back home in Glasgow. The resultant data was then combined in such a way as to give reasonably long term, comparable representations of the 6 sites under study.

The traps were emptied daily and left in place for 3 days at a time before changing their position. As a further study data was also collected to examine the differences between day and night catches and finally catch composition throughout the day.

A fairly broad ecological survey was also carried out to highlight the differences between habitats and sites quantitatively. This allowed more meaningful interpretation of the results.

Results

The ecological data revealed that the mean tree diameter and height was larger in the Primary forest. It also revealed that the secondary forest site had a greater proportion of its overall vegetation nearer ground level and received a greater amount of direct sunlight. Furthermore trees at higher elevation were found to be on average shorter and wider suggesting a slower growing environment.

The Order data revealed that the abundance of insects was considerably larger in the secondary forest when compared to the primary. It also showed an increase in abundance with decreasing altitude.

Detailed statistical analysis of the Dipteran data and the resultant diversity indices produced very similar values for the diversity of the primary and secondary forest. Different indices produced slightly different patterns for the diversity of the sites along the elevational gradient. A general trend of increasing diversity with decreasing altitude was observed (Hills numbers). The diversity index of the log series distribution, alpha, suggested an intermediate altitude diversity maximum.

Discussion

Most of the results can be understood in relation to the spatial structure of the study sites and in relation to the local microclimate. Sampling artefact may also be responsible for some of the observed patterns.

The unique position of the reserve combined with the prevailing winds and tropical climate produced a varied climate, even over the relatively small elevation studied. The mist was most common at higher altitudes resulting in fewer hours of sunshine and therefore lower temperatures when compared to lower altitude sites. Diptera activity is closely related to ambient air temperature so this may explain the differences in abundances of insects.

The spatial structure of the vegetation surrounding the trap may also help explain the observed differences in abundance and diversity at all of the sites. The secondary forest vegetation contained a more diverse array of microhabitats and received a greater amount of direct sunlight. The primary forest however received little direct light and was structurally fairly uniform. These features alongside the large distance between the trap and canopy in the primary forest suggest that sampling artefact and habitat structure had an important role in affecting abundance and diversity.

Disturbance hypotheses also help explain the results although this is not a major factor throughout the reserve as human disturbance is minimal and the secondary forest is in the latter stages of succession. Neither of these factors were measured so their importance cannot be verified.

There are many alternative explanations for elevational gradients in species diversity that are all associated with 4 main features: elevational gradients in climate; elevational gradients in area; geographical isolation; and feedback among zonal communities. There is no clear way of disentangling these processes so one can only conclude that the resultant patterns of insect diversity are the result of a number of a number of ecological processes whose importance differs both temporally and spatially.

It would be of interest to see if the observed patterns hold for different seasons, years and localities. Furthermore, a cheap and efficient method of sampling biodiversity, in particular of insects, is required for future assessment of diversity in order to discover and protect species before its too late.

Robert P. Ellis

Two other projects examined the insects associated with particular plant types, namely tree ferns and bromeliads. Each of these took samples of those insects that seemed to be dependent on the plants as a result of primary feeding of living tissue, herbivory, and/or secondary utilisation of dead or decaying plant tissue, saprophagy. The results of the tree fern weevil (Curculionidae; Cossoninae) faunas will be used for comparison with similar samples from New Zealand that are available in the collections of the Hunterian Museum (Zoology) and the Natural History Museum (London). The Gondwanaland distribution of the plants may be paralleled by the insects that feed on them. This analysis also is not yet completed at the time of writing.

The work of the students engaged in this project was done with the support and involvement of Dr Giovanni Onore, of the Università Cattolica Pontificia of Quito, Ecuador, who is also an entomologist. Clearly there is much potential in this area of the Andes for biological investigations. The students carrying out insect research were Kevin Conway, Carrie-Ann McCulloch, Tracey Begg and Bob Ellis, assisted by Lucy Webster, Stacey Erwin and Oscella McKnight. Supervision was provided by Geoff Hancock, Hunterian Museum (Zoology) and Graham Rotheray (National Museums of Scotland). A sample of the insects that were collected for identification are to be returned to Ecuador for the collections of the zoological departmental museum in the Catholic University, Quito.

References

Booth. R. G, Cox. M. L, Madge. R. B, (1990). IIE Guides To Insects Of Importance To Man: 3. Coleoptera. International Institute of Entomology (Institute of C.A.B International) The Natural History Museum London.

Brown, J. H. (2001). Mammals on mountainsides: elevational patterns of diversity. Global Ecology & Biogeography 10: 101-109.

Crowson. R. A. (1981). The Biology of the Coleoptera. Academic Press.

Herbert, P. D. N. (1980). Moth communities in montane Papua New Guinea. Journal of Animal Ecology 49: 593-602.

Janzen, D. H., Ataroff, M., Farinas, M., Reyes, S., Rincon, N., Soler, A., Soriano, P. & Vera, M. (1976). Changes in the arthropod community along an elevational transect in the Venezuelan Andes. Biotropica 8: 193-203.

Lawton, J. H, MacGarvin, M. & Heads, P. A. (1987). Effects of altitude on the abunbence and species richness of insect herbivores on bracken. Journal of Animal Ecology 56: 147-160.

Olson, D. M. (1994). The distribution of leaf litter invertebrates along a neotropical altitudinal

gradient. Journal of Tropical Ecology 10: 129-150

Terborgh, J. (1977). Bird species diversity on an Andean elevational gradient. Ecology 58: 1007-1009

Wolda. H. (1986). Altitude, habitat and tropical insect diversity. Biological journal of the Linnean Society 30, 313-323.

Work, T. T, Buddle, C. M., Korinus, L. M. & Spence, J. R. (2002). Pitfall trap size and capture of three taxa of litter-dwelling arthropods: implications for biodiversity studies. Environmental Entomology 31(3): 438-448.

Finances

In order for the expedition to go ahead a large of amount of fund raising was required. Each member made a personal contribution of £600, and everybody helped in letter writing, t-shirt selling, running the snack bar, raffle ticket selling, and other various activities, the University gave a generous grant of £1700, and the Carnegie and Dennis Curry Trusts and Fyffes Bananas all contributed generously. Below is the complete set of accounts for the Expedition.

Income

Personal contributions £8450

University of Glasgow £1700

Carnegie Trust £2000

Dennis Curry Trust £1000

Fyffes Bananas £ 750

Exploration Society - First Aid course £ 600

Raffle £1014

T-shirt sales £1234