Leafcutter Ants

Hypothesis: We hypothesize that the removal of natural vegetation from primary forest would decrease canopy cover and humidity and increase air and soil temperature. Such alterations in the abiotic factors surrounding an A. cephalotes colony would raise the temperature of the colony, causing an increase in evaporation of moisture. As these ants farm fungi by planting it on a vegetative substrate produced from resources carried into the colony, the increase in evaporation will cause the ants to seek resources high in moisture to support the growth of the fungi. We predicted that the least favorable abiotic factors for maintining colony moisture would be found in the developed area, with the abandoned agroforest being more favorable, and the primary forest the most. Therefore, we expected to find that the level of moisture in resources carried to the colony to be highest in the developed area, moderate in the agroforest, and lowest in the primary forest.

Materials: Tweezers Ziplock bags Colored Markers Stop watch Counter Compass Balance

Three habitats within La Selva were selected to compare. These three habitats include old growth forest, abandoned agroforest, and developed areas around the research compound. Within each habitat two or more colonies were chosen for observation. These colonies were the first two colonies found in each habitat based on recalling where we had seen active colonies on previous days. A point less than 5 m from the colony entrance along an incoming ant trail was designated for measurement. Data was collected from ants moving toward the colony entrance for 1 minute out of every 5 for a total of 30 minutes. Having observed these colonies previously, we noted that the frequency of ants is fairly steady over a several hour period within one colony trail. Therefore, observing more than 30 minutes was not necessary. Sampling only 1 minute out of every 5 increased the possibility of correcting for slight variations in ant frequency and collecting all of the types of resources that may be carried. It also enabled us to pause between observation periods in order to record data, take measurements of the abiotic factors, and discuss our progress. During each observation period, all cuttings carried by the ants were collected with tweezers and stored in ziplock bags and the quantities of workers and hitchhikers were counted. Soil and air temperature were taken over the entrance to the colony, using a standard thermometer. Relative humidity was taken by use of the sling psychrometer. The canopy cover was measured by holding a transparency with a 64-square grid above the colony entrance. The number of squares in the grid which were covered by vegetation was counted and divided by 64 to obtain a percentage. Moisture content of resources was measured by massing samples from each colony, drying them in a dessicator oven for 10 hours at 21^o C, and massing them again. The difference in mass was divided by the original wet mass in order to obtain a percentage. It is important to note that our schedule forced us to keep resource samples in plastic bags for several hours before massing and drying them. The plant cuttings continued to transpire in the bag over these hours, losing some of the moisture we intended to measure. More accurate results would have been obtained if the resource samples were separated, massed, and dried immediately after collection.

Abiotic Factors: Our data shows that there does not appear to be any significant difference in atmospheric conditions or canopy cover across habitat types on the sampling day (See Figure 1). Variation in the data from site to site is more likely a reflection of time of day and imprecision of measuring instruments than any other factor. A longer-term study could illustrate significant differences in abiotic factors between different habitats and how they may affect A. cephalotes.

Worker Frequency: Looking at the data, there does not appear to be a correlation between worker frequency and habitat type or time of day. Trails with a high and low frequencies of workers were found in all habitat types and throughout the day with no apparent overall pattern (See Figures 2 & 3). There may be daily patterns of varying activity level in each colony. However, it would require sampling each colony at several different times of day over a period of time in order to discover this. Worker frequency is probably dependent most on the total population of the colony, the availability of desired resources, daily weather conditions, and disturbances such as trail blockages. We observed during our sampling that even minor trial blockages, such as fallen twigs across the path of the ants, caused the outgoing workers to stop and remove the obstacle or reroute the trail before resuming harvesting efforts.

Hitchhiker Frequency: There is no relationship between the frequency of hitchhikers and the type of habitat. The frequency of hitchhikers directly correlates with the frequency of workers (See Figure 4), and in some ways may depend on the type of resource being cut and carried. Hitchhikers were observed on both leaves and flowers, but not on any other type of resource. Qualitative observations demonstrate a relationship between the size of the resource and the number of hitchhikers, with small globs of fruit or pollen being both unable to support hitchhikers and unnecessary to require them. Whereas, larger leaves and flowers tended to have hitchhikers, with the highest number of hitchhikers on the largest pieces. Apparently, there have been competing hypotheses as to the role of these hitchhikers. Some researchers claim that these smaller ants clean the resource as it is being carried. Others claim that these ants are martyrs, willing to sacrifice their lives in order to protect the workers below. However, we observed several times that when workers carrying large cuttings encountered difficult terrain, they were helped by the hitchhikers. Hitchhikers crawled off the large cutting and assisted in lifting it over twigs or around obstacles. In fact, the highest hitchhiker frequencies were all along trails in which the primary resource collected was green leaves cut into relatively large pieces.

Resource Type: It was observed that A. cephalotes cut and carried several types of plant material used to farm their fungus. Although green leaves were the overwhelming majority of the material, small purple flowers, parts of bright pink flowers, brown and yellow leaves, seeds, seed husks, stems, stamens, pistils, globs of fruit, and even pollen were harvested. Colonies usually have many trails leading to several different entrances, each of which may be bringing in slightly different mixture of resources. We also noted, in a few cases, that multiple ant trails lead to different types of plants. These trails converged into one larger trail and the final mixture was brought into the colony. Since we tended to take samples less than 5 m from the colony entrance, our collected samples represent this final mixture of materials coming into the colony.

Resource Moisture: Moisture of resources was consistent across all three habitat types (See Figure 5). Disregarding the outlying data point for Trail 4, the average resource moisture was 66% regardless of habitat type or resource type mixture. The value for Trail 4 was 86%, which is interesting to note, because the resources carried on this trail were over 90% composed of flowers and flower parts. Although fresh leaves must be available to this colony, due to their proximity to two other colonies in the same habitat, this colony seemed to strongly prefer harvesting flowers. One idea resulting from this is that colonies may choose flowers over leaves due to the high moisture content, if given a preference. However, much further study needs to be done to explore this idea. For instance, do the two nearby colonies also have access to the flower resource? If so, why do they prefer leaves? If not, is it because the colony of Trail 4 is dominating this resource and preventing other colonies from accessing it? We do not know how available this resource is in terms of resource plant population density in the area or frequency of blooms. When discussing flower versus leaf preference with a local researcher, another idea was raised. As flowers are ephemeral in nature, trees may not invest much energy in creating toxic anti-herbivorous compounds in them. Therefore, not only do the flowers have high moisture content, but also they may be desired by A. cephalotes for farming fungus because of the lack of certain chemical compounds.

Conclusions: Our hypothesis was not supported. There were no significant differences in abiotic factors between the three habitat types, despite differences in their degree of deforestation. Likewise, there were no significant differences in the moisture content of resources collected by ants between the three habitat types. However, our data raises the question of whether, regardless of colony habitat, A. cephalotes will show a preference for resources high in moisture content when given a choice.




BACK TO PROJECT LIST