Click on a thumbnail to view the enlarged image
Figure 1 – Bauhinia seedling	Figure 2 – Mature Lianas	Figure3 – Mature Monkey Ladder

As Bauhinia glabra matures, the stems undulate as lianas form. They take on a stair-step configuration at an intermediate stage while the mature lianas form thick ladder-like structures resulting in the common name of monkey ladder vine. Other local names include "escalera de mono" in Columbia and "pata de vaca" in Peru, literally meaning "foot of cow." The large white flowers are a food source for nectarivorous bats (Janzen, 1983).

Bauhinia glabra is skiophytic. It is light-demanding and only tolerates shade when it is young. It needs the sunlight of the canopy to become mature. Hence, it is seldom conspicuous in primary forest and only becomes abundant when gaps appear in secondary forests. While the liana form is common in South America, it is not found on the African continent (Richards, 1952).

While the ease of identification and bizarre appearance of Bauhinia glabra were factors that we considered in choosing this species, its primary appeal was the skiophytic nature and preference for disturbed, secondary forest locations. As such, questions as to its distribution based on manmade gaps in the form of foot trails became the focus of this study. It is our belief that more Bauhinia will be present on the foot trail and fewer seedings will be found further off the trail a sample plot is located.

METHODS

Study site:

We selected study areas to research the effects of man made disturbances in the form of hiking trails on the presence of Bauhinia glabra. We specifically looked for human hiking trails that were reasonably flat and occurred in the secondary tropical dry forest that contained evidence of mature monkey ladder lianas. The specific trails that we selected were the Guayacan Trail and the Noguera Trail. Though the trails were approximately the same altitude (within 100m elevation) and both trails ran parallel to the ridge, there were several notable differences between the two. The Guayacan Trail contained considerably more understory. This edge growth was identified as Mansoa hymenea, or "little garlic" bush. In contrast, the Noguera Trail had less understory but more slope. To control for the slope we selected the flattest section we could find running parallel to the ridgeline. On the Noguera trail the right hand side was the downslope side.

Data Collection:

We collected data on the presence of Bauhinia glabra through the use of strip transects. The ten transects were oriented perpendicular to the hiking trails in paired data sets. We observed the number of Bauhinia glabra seedlings growing in a one-meter square sample plot. A seedling was defined as any plant having one or two leaves, each made of two fused leaflets. Any plant having evidence of a third fused leaflet was classified as too large and ignored.

Prior to collecting data on the presence or absence of Bauhinia glabra we ran a trial to standardize our identification of the seedlings. Our group split up into two pairs and each set up a random sample plot. We then individually counted the seedlings and compared our counts with that of our partner. The pairs then switched plots and determined the repeatability of our sampling method. Following this process, we felt confident that our counting methods were reliable.

Click on a thumbnail to see the enlarged image
Figure 4--Collecting Data	Figure 5 – Pam Measuring a Transect Line	Figure 6 – Pam and Julia counting Bauhinia seedlings

For each of the 20 meter transects we placed a measuring tape perpendicular to the trail extending into the tropical dry forest. At each site, pairs of researchers went to the far ends of the strip transect which was 10 m from the center of the trail and positioned a 1 m2 flexible frame on the forest floor. The frame was always placed on the same side of the measuring tape to ensure consistency, and the Bauhinia glabra seedlings were first counted independently by the researchers. Counts were compared and recensused if counts were inconsistent. The flexible frames were then moved to the 5 meter and 0 meter test sites respectively, and the test procedure was repeated. The 0 meter test plots for each transect were located adjacent to each other in the center of the trail. Finally, successive transects were spaced 10 meters apart on the trail and a paired design procedure was performed at each. Five strip transects were completed on two different trails for a total of ten pairs of data points. Measuring more than one trail allowed us to make a more "robust inference" (Langen, 2001) of the impact of human disturbance on Bauhinia glabra.

 

RESULTS

There was a significant decrease (P < 0.05) in the number of Bauhinia seedlings 5 meters from the center on either side of Trail 1. Significance was determined by a two-tailed paired T-Test.

In contrast, there was an overall increase in the number of Bauhinia seedlings between 0 – 5 meters on Trail 2. However the distribution on either side of the trail was uneven. The left side of Trail 2 decreased consistently for the entire 10 meters, while the right side was much more variable, in that it increased significantly

(P < 0.05) from 0 – 5 meters.

On both trails, the trends reversed from 5 – 10 meters from the center of the trail. An additional comparison of all transects on the Guayacan Trail indicates an initial decrease at 5 meters followed by an increase in the number of Bauhinia seedlings at 10 meters. In contrast, the Noguera trail recorded an increase in total numbers at 5 meters and a decrease at 10 meters.

Click on a thumbnail to see the enlarged image
Table 1  Raw data	Table 2  T-Test	Chart1-trail 1 seedling	Chart 2-trail 2 seedling count	Chart 3-trails 1 and 2 seedling counts

DISCUSSION

Though we expected a higher presence of Bauhinia glabra on the hiking trail in the highly disturbed area and fewer seedlings 10 m into the forest, we found their presence to be highly variable especially along the edge of the trail.

There are several observations that might provide insight into the results. The two trails, though in similar locations, differed in several aspects. As described in the introduction, Trail 1, the Guayacan Trail, contained copious amounts of Mansoa hymenea, or "little garlic" bush. Trail 2, the Noguera Trail, had less understory, but more slope. The only transects that seemed to have a consistently decreasing relationship were on the left side of Trail 2 which was the only set of transects to run uphill. This area also contained less understory.

In comparing all of the transects, it was reassuring the at 0 or 10 meters large discrepancies did not exist between Trails 1 and 2. However, dramatic differences appeared between the two trails at a distance of 5 meters, reflecting observable habitat differences at the edge of the trails.

It could be speculated that the absence of Bahinia seedlings, although shade tolerant, were explained by the dense growth of Mansoa hymenea or "little garlic" bush. It should also be noted that several of the transects for Trail 1 contained large rocks in the sample areas. Trail 2, the Noguera Trail, contained considerably less understory than Trail 1. Perhaps this decreased competition for light and space could be explained by the unpredictable dense growth of a competing species.

Results from our study sites seem to indicate that optimum levels of light for the Bauhinia seedlings occurred 5 meters in from the center of Trail 2. The disturbance of the trail did not significantly effect the presence of Bauhinia seedlings ( p>0.05). However, the resulting edge effect from the initial disturbance produced variations in a number of seedlings present at 5 meters. It was observed that the edge effect was highly variable depending on the location.

In conclusion, disturbance does seem to have variable effects on the presence of Bauhinia glabra seedlings in the secondary growth tropical dry forest due to the variable edge effect. Future studies should include the use of light meters to quantify the edge effects and ensure higher level of consistency among trail selection.

ACKNOWLEDGEMENTS

Key to the completion of this study was the exceptional rapport developed over a very short period of time by the members of team three.

Many thanks are due to our team leader, Tom Langen, for without his tremendous guidance and patience, our group may have spent many hours in confusion and chaos.

A well-deserved "gracias!" also goes to our colleague and fellow team member, Bob Kuhn for his expertise and instruction with the Microsoft Excel program in the organization and development of our graphics.

We are especially indebted to the employees of Palo Verde National Park and The Organization for Tropical Studies for their assistance in our research. Special thanks goes to Nicole Turner for introducing us to the park and it’s trails, Manrique Montes for his assistance in identifying the unknown garlic plant, now known as Mansoa hymenea, and Mauricio Castillo for helping us to obtain maps to incorporate into our presentation.

Thanks to all our team members for their continued assistance and support during our research!

 

Nancy, Sharon, Pam, & Julia

July 19, 2001