Research mission

Research Mission

Research Mission 1999-2010

The Research Mission that the team at Las Casas de la Selva are carrying out started thirty years ago between 1983 and 1986 when the first trees were being line-planted. Over the years, various management teams have operated diverse projects at Las Casas de la Selva, including running a sawmill operation, sawing wood-falls from Hurricane Hugo, in 1989. The planted trees were growing. In 1998, Mark Nelson, John Allen, and Sally Silverstone, successfully proposed to the Earthwatch Institute a research plan to investigate how all the trees performed 15 yrs after planting, and after two hurricanes. When Category 3 Hurricane George blew in directly over the land in September 1998, the homestead was laid to the ground. The forest recovered, and slowly over the next ten years the homestead was rebuilt, alongside days and days of data collection on the slopes of the Las Casas de la Selva’s planted hardwood forests. The tabs above take you through the original write-up for this endeavor, by Dr. Mark Nelson and Sally Silverstone, to investigate tree survival and growth rate, along with a biodiversity study. Between 2000 and 2010, there has been a huge venture in scientific data collection at Las Casas de la Selva with Earthwatch Institute volunteers and volunteers from all over the world helping us. The results are presented in these publications:

RESEARCH MISSION from 2010 to present
  • Objective 1. To examine the effectiveness of thinning techniques in our current plantations of Mahoe and Mahogany.Objective 2. To examine the effectiveness of liberation thinning on maturation and volume of tree crops in secondary forest areas. To evaluate methods to transform the structure and species composition to increase productivity and value by applying silvicultural treatments such as improvement thinning or regulating natural regeneration.
  • Objective 3. To continue monitoring development of the line-planted areas to see impacts on economic viability, tree density and growth at later stages of their growth cycle.
  • Objective 4. With the results of the first three, to provide important information for the development of guidelines for the sustainable management of secondary forests in Puerto Rico and other tropical countries.
  • Objective 5. To conduct herpetological studies focused on identifying which species of reptiles and amphibians are present at Las Casas de la Selva; to determine the population density, population fluctuations, microhabitat utilization, and the effects of forest management on the herpetofauna of the forest.
  • Objective 6. To inventory the macrofungi of Las Casas de la Selva; documenting, describing, and photographing the fungi to create material that will aid in fungal identification for future studies.
  • Objective 7. Monitoring both Growth Rate and Seedling Abundance of Naturally Occurring Hardwoods in Secondary Forest at Las Casas de la Selva, Patillas, Puerto Rico.
  • Objective 8. Planting and monitoring critically endangered endemic tree species for recovery and habitat enhancement.
  • Objective 9. To conduct a preliminary bird survey on the Las Casas de la Selva land.

The Project and Study Area

The Project and Study Area

Las Casas de la Selva, is a 409-hectare experimental forest enrichment project in Puerto Rico, was established in 1983. Located on some of the most challenging steep slopes and diverse ecology in the mountains of south-eastern Puerto Rico, the Las Casas de la Selva project uses a total systems approach to utilize a rainforest environment for profit without diminishing its species richness, biological diversity or total biomass. The project is located about one hour from the San Juan International Airport and within easy reach of major commercial centers and beaches. The forest’s elevation of 600 meters provides year round temperatures averaging 22°C/71.7°F. With the prevailing Easterly trade winds delivering an average annual rainfall of 3 m, the year-round humidity is high. To support the sustainable forestry practices, there is a nursery area that provides protection for saplings being raised on the site, and minimizing soil erosion techniques using bamboo and reforestation of barren soil with pine trees, are other experimental demonstration projects underway. The project maintains over half of the property uncultivated as a rainforest preserve for comparative study.
Las Casas is mostly representative of old secondary tabonuco forest. It is one of the very few secondary forests where its management addresses conservation and utilization issues, as most other ecological work is being done in mature (usually “climax”) rainforests. Human intervention has extensively disturbed most of the mature rainforests in Puerto Rico during the last few centuries including the use of rainforests for agriculture, coffee plantations and pastures. Increased standards of living on the island have led to a decline in agricultural land use and in the use of wood or charcoal for burning. The most marginal of these lands have reverted to secondary forests and they are left untended until it proves profitable to use them again for timber fuel or agriculture. However, it is clear that the rate of deforestation is still increasing.
In 1992 the FAO estimated that 8.3 million hectares of secondary forest were lost per year in Latin America alone (FAO, 1992). As far as the overall picture for tropical rainforests is concerned, it was estimated in 1990 that at least 5 million ha of tropical forests are cut per year for timber, paper pulp and other wood products, 2 million ha per year for cattle ranching and 8 million hectares for cultivation, much of it slash and burn agriculture (Corson, Ed. 1990). It is clear that research into the sound ecological management of timber production in tropical and subtropical secondary rainforest areas could make a valuable contribution to the future preservation of rainforest land and its species by providing a method of sustainable utilization of the land for profit without destruction of the local ecology. In the case of Puerto Rico, it could also make a considerable contribution to the island economy. It is estimated that in 1828, half of the lands previously owned by the Spanish Crown in Puerto Rico had been ceded to settlers whose main occupation was agriculture. At that time, approximately 72% of the island was still covered by forest. By the end of the 19th century 90% of the land had been ceded to settlers, and only 25% was still forested. 4% of this forestland was government owned and 21% privately owned. By 1935, 17% of the island was privately owned forestland; this percentage increased after World War II when much of the land that had been cleared for sugar cane became unproductive and was abandoned (Pico, 1974).
In general, the exploitable species found in the secondary forest areas are fruit trees in the old plantation areas and shade trees in the old coffee growing regions. Valuable timber is generally only found in government reserves and the less accessible parts of some farms. Frank Wadsworth has estimated that in order to adequately protect the soils and watersheds of Puerto Rico, 25% of the island would need to remain forested (Pico, 1974). Puerto Rico currently imports nearly all of its timber from the U.S.A. and Canada at an estimated value of $130 m. Wadsworth has suggested that at least 48% of this timber could be produced in Puerto Rico’s secondary forests without touching any land currently under agricultural use, thereby reducing the island’s net payments and providing approximately $20 million USD in employment (Pico, 1974). This would be particularly significant in rural undeveloped areas such as Patillas (the municipality in which the project is situated) where unemployment in some sectors is as high as 40%. The clearing of secondary rainforest in the area around the project site is caused mostly by agriculture and livestock grazing. Much of the land in this area is on very steep slopes which cause severe erosion, and production is poor without heavy use of chemical fertilizer. It is hoped that a demonstration of timber production on lands unsuited to long term agricultural use will encourage similar practices resulting in sustainable practices.

Forest Enrichment Line Planting Methodology

Forest Enrichment Line Planting Methodology

In 1984, Tropic Ventures started a program of line planting valuable hardwood timber trees within the existing secondary forest. Approximately 40,000 trees have been planted to date on 87 hectares, with a spacing of 3 m between trees in the rows. The planting density for the mahogany was approximately 365 trees per hectare. Final stocking with approximately 33% of the planted crop trees being carried through to maturity is expected to be 120 per hectare. Narrow 2 m (six foot wide) planting lines are cleared through the forest in an east-west orientation using hand tools which minimizes damage to the soils that are easily eroded when disturbed. In general, 10 m of rainforest is left undisturbed between tree lines. The plantings utilized economically valuable tree saplings, primarily Honduras Mahogany, Swietenia macrophylla, chosen because of its availability and proven success in previous plantings on the island (Weaver, 1983). Mahoe Hibiscus elatus, pine Pinus caribaea and other native species, Prunis occidentalis and Magnolia portoricencis, were planted in smaller numbers, accounting for around 10% of the planted trees, and are being evaluated for future plantings. Mahogany is a relatively shade intolerant (light demanding) species representative of an early to intermediate stage of natural succession (Wadsworth, 1992). It is known to respond well in natural or created gaps in the forest matrix. The trees mature rapidly with the combination of protection and competition afforded by the surrounding forest. It is also comparatively resistant to hurricane damage. (Mayhew, 1998)

Overall Research Objective

Overall Research Objective

The project’s overall research objective is to test the hypothesis that tropical hardwoods can be cultivated in a secondary growth rainforest using a line planting technique while maintaining plant biodiversity.

It is expected that the line clearing will have had some effect on insulation levels at the forest floor (Odum et. al., 1970) with possible impact on species diversity and that the importance values of the different plant species will be shifted with shade intolerant species increasing at the expense of shade tolerant species (Estrada et. al., undated).

Research tasks will be focused on the collection of data on biodiversity and tree growth at the experimental site. This data will be used in comparing the success of this forestry enrichment approach with production using other planting methods. Mayhew and Newton (1998) note that raw growth data from mahogany plantations are not widely available partly due to the commercial value of such data and the reluctance of forestry institutions to divulge their results. However it should be possible using the data available in the literature to make comparisons using commercial yield, comparing the mean annual diameter growth of planted trees at Las Casas with that of trees of similar age from clear-cut plantations, and other line-planting experiments. The ecological data will be used for an evaluation of the impact of the forestry enrichment program on ecological functioning, including species richness, dominance patterns and biodiversity.

Evaluating and documenting both successful and unsuccessful aspects of the project will increase its potential value to researchers, and increase our ability to integrate the data and modify the project as necessary to make it more successful and thus more valuable as a model. In some instances it may even help people avoid problems and less successful planting techniques.

The authors both recognize that a final evaluation of this method as a model for sustainable forestry will not be possible until trees have been harvested and sold. However as first major harvest is not due for at least a decade this aspect is outside the scope of the proposed research.

Evolution of Reforestation Methods at Las Casas

Evolution of Reforestation Methods at Las Casas

The planting design utilized has varied with time and site conditions. During the earliest plantings, lines were cleared following the contour of the land due to a concern for minimizing soil erosion. After that the design was changed and east-west running lines were established to increase the amount of direct light received by the new seedlings. This change was made after the erosion on the heavy acid clay soil was observed not to increase significantly regardless of the direction in which the tree lines were cleared. Seedlings were planted more densely in areas with younger secondary forest species composition and less densely in areas with more advanced species composition. In general the plantation design was to plant at approximately 3 m spacing within the lines and 10 m from line to line through the forest. This technique closely mimics a medium sized tree fall gap in which the forest floor litter and soils are undisturbed but the above ground nearby competition is removed, as is some of the overstory. The result is increased solar insolation and increased daily maximum temperatures at ground level, which in turn increase the decomposition rate of organic matter, releasing nutrients. These changed conditions promote the survival and growth of the new mahogany seedlings and also create microsite conditions suitable for the re-establishment of primary successional species including trees.

Early Study Plots 1986-1994

Twelve study plots each containing 40 trees were established to monitor growth of the different species and evaluate the significance of different site conditions. These plots were measured annually between 1986 and 1994. Previous studies carried out by Harry J. Scott, who received his M.S. in tropical forestry from the Yale School of Forestry and Environmental Studies have suggested that up to 1994 the most crucial factor affecting tree growth is aspect. The data from the study plots shows that the largest trees were found in lines that that lie between NW, W and SW in orientation. The smallest trees lie between NE, N, E, SE and S (Scott, 1994).

Hurricane impact 1991-1998

The effect of hurricane damage is expected to be significant as moderate to severe cyclonic storms periodically hit Puerto Rico (Walker, 1991). Hurricane Hugo in 1989 and Hurricane Georges in 1998 both caused severe damage to the plantations (the eye of Hurricane Georges passed directly over the project) and infrastructure. Hurricane events may be expected to increase the number of severely damaged trees and kill some trees in the lines, but may also increase growth rate of the remaining plants due to increased nutrient recycling (Lodge et. al., 1991).

Research needs from 2000

Research needs from 2000

Re-measuring these earlier plots will enable us to see if previous trends have continued and also to see if the rate of growth has been maintained under different site conditions. In addition at least 20 new study plots one acre in size, representing an initial planting of approximately 148 trees, will be established (more if time and resources permit). The trees in these plots will be both measured and tagged for future reference and will enable us to make a more complete assessment of the state of the plantation trees, some of which are now 15 years old. Creating a larger study area will give a better picture of overall mortality rates and the effects of hurricane damage and landslides.

The last significant survey of the tree growth in the plantations at Las Casas de la Selva took place in 1994. This survey was carried out in 12 study plots that had been established to monitor growth of the different species under different site conditions. The study to begin in 2002 will both continue and augment this work, providing data on tree growth that will enable the project to plan future thinning and replanting work, project future growth and potential yield and improve on the design of future new plantings. Production yields and costs will be compared with those of clear cut planting and with similar forestry enrichment trials in other areas (Ramos and del Amo 1992, Weaver 1989, Lugo et. al. 1990).

Within these larger plots we will be able to experiment in the future with different clearing and cleaning procedures using subplots and leaving some areas as a control.

In addition to the re-measurement of the tree growth and biomass, the following studies will be carried out to deepen understanding of the impact of the forestry program on important ecological vectors:

· Biodiversity: The impact of the program of forest enrichment on plant species richness will be determined by identification of plant species in areas containing the silviculture line planting vs. control areas of similar slope and aspect on the Las Casas property, which have not been disturbed.

· Frequency: Frequency will be determined for all species found in the study plots. Frequency is a measure of the probability of finding an individual species with the overall population sample (Brower et al, 1991). Determining frequency and canopy cover will enable us to calculate importance values of species and determine what changes if any have occurred in dominance patterns of line planted areas compared to the undisturbed secondary forest control.

Research Methods

Research Methods

(a) Forest growth

The methodology to be used in the study plots will be to measure total height (toht) and diameter at breast height (dbh at 1.37 m) of all surviving trees that are within the fixed area permanent study plots. Measurements will repeat those done in previous surveys. These measurements will include potential commercial height (coht, determined by the lowest major fork or branch), and the crown width or canopy (cano) in the direction of the line. An estimate of basal areas per acre will be made in each of the study plots as a measure of forest density using a 2.5 m2 Basal Area Prism. Note will be taken of any trees that have died or been severely damaged since the last measurement. These notes will include information on hurricane damage, for example windthrows and trees with broken stems. In the 12 study plots, these notes will be compared with the notes available in the raw data prior to Hurricane Georges.

(b) Site factors

In the newly established plots, slope aspect will be determined on which the plot stands and the direction of the line planting. The trees will be tagged, numbered, and measured using the same protocol as in the older study plots. An estimate will be made in each plot of the number of trees that have died since the date of planting. (Even if there are no remains of a dead tree evident it is possible to assess this fairly accurately as the trees were originally generally planted 3 m apart).

(c) Data Processing and Evaluation

The data will be collected in the field on data sheets. The raw data will then be uploaded onto computer spreadsheets ready for statistical analysis.

The plots will be compared with each other in terms of the measured variables and the data will also be compared with previously collected data to determine growth and mortality rates since planting and since 1994. Comparison and statistical significance of differences in paired measured variables within each plot (e.g. height, dbh, canopy cover) will be made using scatter plots and general linear regressions, and the results compared with such comparisons made in 1994 by Scott for a Yale School of Forestry and Environmental Studies paper. Growth rates will be assessed by calculating trend lines, determining coefficients of increase (k values) and goodness of fit (R2 values) of growth equations to the data set. Other statistical tests, such as ANOVA (Analysis of Variance), may be performed if the data suggests their potential value.

(d) Biodiversity

Biodiversity will be determined by comparison of data from a series of 10 m x 10 m plots. Between 8 and 20 plots will be examined depending on availability of resources. Half the plots will be in areas that have been cleared for line planting (silviculture plots) and half the plots will be in areas of secondary forest of similar elevation aspect and previous land use on the property that have not been used for line planting (Control plots). An effort will be made to pair up the plots in terms of topography, for example for each plot that is examined in the line planted area in a valley slope or ridge top, a plot will be examined in a valley slope or ridge top site in the control area. Within the 100 m2 plots all plants with a stem width greater than 5 cm in diameter will be identified and data collected on dbh, height and canopy cover. (These plants will be tagged for future reference). Nested in each plot, five randomly chosen, 1 m2 plots will be examined and all individuals recorded and measured. In the case of grasses and ferns, a percentage of ground cover will be estimated on a scale from 1-5 with 1, being a single stem and 5, being over 50% cover. Then data will be analyzed for species richness and biodiversity determined using the Shannon biodiversity index (Shannon and Weaver, 1949; Brower et al, 1991):

H’ = -S pi log p i

where p i = n i /N

“p i ” is the proportion of species “I” in the total number of individuals in the population (N).

The Shannon biodiversity index will be calculated using the above formulas for log 2 and log 10.

(e) Frequency

Plant species’ frequency in the silviculture vs. control areas of secondary forest will be determined by examination of the data from the plots described above. Each individual plant stem will be counted as an observation in the site.

(f) Plant cover

Plant cover for each species will be evaluated by measuring canopy cover of the most prevalent species (over 5 cm in diameter) in silviculture and control plots.

(g) Importance values

Importance values will be calculated by combining frequency and cover data and dividing by two, so that the sum of all importance values for each system equals one. The graph of these data, called a dominance-density curve or species importance curve, will be plotted on a log/arithmetic scale against rank order (Brower et al, 1991).

These data will be examined to see whether the program of line-planting forest enrichment has resulted in increased dominance by light demanding plants as suggested by theory.

(h) Solar insolation and light interception

Solar insolation and light interception in the silviculture vs. control plots will be measured using a LI-COR LI-189 Quantum/Radiometer/Photometer equipped with a LI-COR Terrestrial Radiation Sensor, Type SA (LI-200SA) pyranometer sensor.

(i) Canopy closure

Canopy closure in silvicultural vs. control plots will be evaluated using analysis of hemispheric canopy photography (Rich, 1989). Photographic images of the canopies will be digitized and converted to a gray scale using Photoshop 2.0. Analysis for amount of canopy and light penetration will be determined with Map Factory software.

(j) Other Considerations

Using the expertise available to us through the University of Puerto Rico we will look at the most dominant species in the silviculture vs. control areas and assess what type of vegetation has been favored, for example vines, shrubs, ground-covers, shade tolerant or intolerant species, exotic versus native, pioneer species versus ones more typical of developed forest.

Application Of Results

Application Of Results

The study will enable us to evaluate how well the project is working and what modifications need to be made to improve the growth timber stands while continuing to preserve the surrounding rainforest ecology. Puerto Rico currently imports most of its timber; the results of this project could encourage other such enterprises on the island and elsewhere in the tropics providing income and employment. The results will also be made available to the staff of the Department of Natural Resources in Puerto Rico who have been interested in this approach as a model for other rainforest enrichment programs on the island. The Principal Investigators and other scientists who have been involved with the project will submit the compiled results to relevant journals for peer review publication, and also more popular media, with grateful acknowledgment for Earthwatch and its volunteers. Resources permitting, the results will also be written up for publication in Spanish.

Journals to which such articles can be submitted include: Ecological Engineering, Journal of Tropical Forestry, Bioscience, Biotropica, Ecological Applications and Ecology.

The results will be made freely available to visiting scientists, which in the past have included the Yale School of Forestry and Environmental Studies, the University of Vermont School of Forestry, Royal Botanic Gardens at Kew personnel and US Forestry officials. Our most recent visit was from a group of forestry students at Carolina State University. The project at Las Casas is also attracting a great deal of interest from other school students from both Puerto Rico and the continental United States. The results of these surveys will be made available to all of these visiting groups to evaluate, and through website and publications, popular and scientific, to a much wider audience.

Previous Research with Line Planting and Rainforest Enrichment in other parts of the world.

The need to find alternatives to clear cutting as the main source of economy has led to experimentation with various forms of reforestation strategies. Indigenous forest populations have successfully managed sustainable production and harvesting of food, fuel and construction wood from tropical forests for thousands of years (Gomez-Pompa and Kaus, 1988). For example, a forestry management plan was devised using Yanesha Indians of Eastern Peru’s knowledge of forest dynamics and gap regeneration. 30 to 40 m wide strips were clear cut in the natural forest so that the seeds naturally regenerated in the clear-cut strip. The regeneration of native tree species was found to be excellent with good financial returns (Hartshorn, 1988).

However the main forms of experimentation have been either enrichment planting (the desired species is planted in lines or parcels with a long term aim of maintaining a mixture of planted species), or conversion (the long term aim is that the planted species will eventually dominate and replace the natural forest). Dawkins’ (1965) study of conversion sites in Africa suggested five necessary conditions for successful planting:

  • The species used must be self-pruning, straight and fast growing.
  • There must be no tree canopy over the planted area.
  • There must be little or no sale for thinnings in the area concerned.
  • The regrowth between the lines must be non-inflammable.
  • Browsing animals must be absent.

The line planting project of mahogany and other hardwoods at the Las Casas site would seem to satisfy these conditions if the canopy remains sufficiently open over the planted lines after the initial clearing (see methodology below).

Line planted mahogany was extensively studied at sites in the Luquillo Mountains of Puerto Rico near Rio Chiquito (Weaver, 1983). In the area studied, abandoned agricultural and pastoral land had regrown into brush and secondary forest. Line plantings were established in 1963 using 3 x 3 m spacing. Maintenance practices included control of competition and in 1975, about 30% of the trees were thinned. Sample lines selected along topographical features were measured in 1981. The 1981 measurements showed a mean annual increment (MAI) of 1.4 cm/yr in diameter and 1.0 m/yr in height. Weaver concluded that line planted mahogany could be successfully used to enrich secondary forest and the practice could provide local employment.

Enrichment planting in Uxpana, Mexico tested several different treatments. Mahogany grew most rapidly in height in completely open areas but ultimately had the highest survival rate in areas where all vegetation with a dbh (diameter at breast height) less than 12 cm were removed allowing a 37% light transmission. This treatment minimized competition from weeds during the first year of growth (Ramos and del Amo, 1992).

In southern Para, Brazil, mahogany logging companies tested line planting with the trees planted in 2-3 m wide swathes at 10 m intervals. However, growth was very slow which was attributed to the fact that lianas (thick vines) covered the gap between trees at the canopy height. Other problems were due to the fact that they did not remove enough of the natural forest canopy (Verissimo et al., 1995).

In reviewing the various experiments carried out in line planting techniques, Mayhew and Newton (1998) concluded that line planting can be an effective method to convert logged or secondary forest into mahogany plantations. They recommended that each tree be planted with cleared space of 2-3 m and overhanging trees removed. Maintenance of undisturbed forest vegetation between the lines provides suitable conditions of lateral shade and overhead light conducive to the growth of light-demanding mahogany tree seedlings.

The Las Casas enrichment experiments were designed to test whether such line-planting enrichment was viable in the secondary forest of Southeast Puerto Rico, and to evaluate the growth of mahogany and a variety of other valuable exotic and native trees in such a planting regime.

This Research Mission since 20oo resulted in these publications.

 Literature Cited

Literature Cited

National Forest, Puerto Rico. Forest Service Research Paper. Institute of Tropical Forestry, Rio Piedras, Puerto Rico.

Brower, J.E., Zar, J.H., and C.N. von Ende. 1991. Field and Laboratory Methods for General Ecology, 3rd ed., Wm. C. Brown Publ., NY.

Corson, W.H. (Ed.). 1990. The Global Ecology Handbook, Beacon Press, Boston

Dawkins, H.C. 1965. Problems of natural regeneration, plantations and research. Unpublished report for the Forestry Division of the Ministry of Natural Resources, Sierra Leone.

Estrada-Pinto, A., R. W. Garrison, D.P. Regan, R.B. Waide, and C.P. Zucca, undated. Flora and fauna of El Verde field station. CEER-T-159. Center for Energy and Environment Research, Rio Piedras, Puerto Rico.

FAO. 1992. The forest resources of the tropical zone by main ecological regions. Forest Resources Assessment 1990 project, Rome, Italy

Gomez-Pompa, A. and A. Kaus. 1988. Traditional Management of Tropical Forests in Mexico, In: Alternatives to Deforestation, A.B. Anderson (ed.), Columbia University Press, New York.

Hartshorn, G. 1988. Natural Forest Management by the Yanesha Forestry Cooperative in Peruvian Amazonia. In: Alternatives to Deforestation, A.B. Anderson (Ed.), Columbia University Press, New York.

Lodge, D. J., F. N. Scatena, C.E. Asbury and M.J. Sanchez. 1992. Fine litterfall and related nutrient inputs resulting from Hurricane Hugo in subtropical wet and lower montain rainforests of Puerto Rico, Biotropica 23: 336-342

Lugo, A.E. 1992. Comparison of tropical tree plantations with secondary forests of similar age. Ecological Monographs 62 (1):1-41

Mayhew, J.E., and A.C. Newton. 1998. The Silviculture of Mahogany. CABI Publishing U.K.

Milton, J.S.1992. Statistical Methods in the biological and Health Sciences, McGraw-Hill, NY.

Odum, H.T., G. Drewery, and J.R. Kline.1970. Climate at El Verde, 1963-1966, pp. B347-418 In: A tropical rainforest, H.T. Odum and R.F. Pigeon (eds.). NTIS, Springfield, Virginia.

Pico, R. 1974. The Geography of Puerto Rico Aldine Publishing Co.

Ramos, J.M. and S. del Amo. 1992. Enrichment Planting in a tropical secondary forest in Vera Cruz Mexico, Forest Ecology and Management 54:289-304.

Rich, P.M. 1989. A Manual for Analysis of Hemispherical Canopy Photography, Los Alamos National Laboratory, Los Alamos, NM.

Shannon, C.E. and W. Weaver.1949. Mathematical Theory of Communication, University of Illinois Press, Urbana.

Verissimo, A., Barreto P., Tarifa, R. and C. Uhl. 1995. Extraction of a high-value natural resource in Amazonia: the case of mahogany. Forest Ecology and Management 72: 39-60.

Weaver, P. L.and G.P. Bauer. 1983. Growth of Line Planted Mahogany in the Luquillo Mountains of Puerto Rico. Forest Service Research Paper. Institute of Tropical Forestry, Rio Piedras, Puerto Rico.

Weaver, P.L.1989. Taungya plantings in Puerto Rico, Journal of Forestry 87(3): 37-41.

Wadsworth, F.W.1992. Temperate zone roots of silviculture in the tropics, pp. 245-256 In: The ecology and silviculture of mixed-species forests, M.J. Kelty, B.C. Larson and C.D. Oliver (eds.), Kluwer Academic Publishers, Dordrecht, Netherlands.

Walker, L.R.1991. Tree damage and recovery from Hurricane Hugo in the Luquillo Experimental Forest, Puerto Rico, Biotropica 4: 379-385