Subterranean Termite Foraging Puzzle

Subterranean termites are cryptic insects, spending most of their lives hidden from view. Understanding where they move and why they move requires extensive knowledge that is currently based on indirect evidence of termite foraging behavior and factors affecting that behavior. Here’s what we’ve been able to piece together so far.

ONE THEORY. Termite foraging behavior has been described as "random" or as having "random" components and as occurring continuously. The termite’s search for food has been described as a "random search" because there are no specific patterns to the search and no causal relationships except for broad categories such as moisture and guidelines. We might also expect that if termites were searching for food randomly, and in particular, tunneling randomly and continuously, then some stakes or monitors would be infested with termites at one time, while at another time others would be "hit." But anyone baiting or monitoring for termites can tell you that this is generally not what’s going on in the field.

WHAT WE SEE IN THE FIELD. Infestation rates for both the eastern and Formosan subterranean termites fall between 5 to 10 percent when wooden stakes are used around the perimeter of structures and landscape beds over observation periods of two to five years (Su and Scheffrahn, 1991, Oi, et al., unpublished data). Furthermore, it appears that over time, the same monitors are "hit" in general. Those PCOs who are baiting know that once they get to know an account, some monitors are going to be consistently active at each observation time, although theycontinue to check the other inactive monitors just in case.

METHODS OF TERMITE MOVEMENT. We know subterranean termites can move from place to place in three ways: 1) building mud tubes above ground; 2) freely walking the surface to explore it; and, 3) tunneling through the soil, which is the predominant method of termite movement. Tunneling uses a lot of energy for termites and is historically difficult to document for researchers. We also know that both the eastern and Formosan subterranean termites evenly divide an area when they are excavating tunnels in new territory (Robson, et al., 1995, Hedlund and Henderson, 1999).

FACTORS AFFECTING TERMITE TUNNELING. We know several factors affect termite tunneling, such as soil particle size, the availability of food, guidelines and moisture.

Soil Particle Size (Powell, et al., unpublished data). A foraging arena (23 by 23 cm/9 by 9 inches) was filled with either native soil, (a common soil in north central Florida), or clean fill dirt to a depth of 1 cm to ½ inch, burying one piece of cardboard and half a wooden tongue depressor (Fig. 1). The particle size for the native soil ranged widely from 0.002 to 2 mm, while the clean fill dirt particle size ranged only between 0.05 to 0.5 mm. Water was added to bring the moisture content to about 10 percent. Five hundred worker termites and five soldiers were added to each arena. These set-ups were replicated five times per soil type, using termites from a different colony each time. Tunneling progress was recorded daily with a high-powered scanner. Tunnel lengths were measured with a map measure. Data were analyzed by analysis of variance (SAS Institute 1988).

Termites were allowed to tunnel for a total of 14 days. Main tunnels were constructed between one and three days, during which time termites found the food source on the opposing end of the arena (Fig. 1). Branch tunnel formation began as main tunnels were being established and continued until day eight (Figs. 2, 3). After day eight, there was no increase in net tunneling (Fig. 4). Also, there was no significant difference in tunneling in either soil type, despite particle size differences.

In other words, the termites first built the main roads to their highway transportation system (main tunnels), and then built the bypass system (branch tunnels). This system of main roads and bypasses helped the termites move within the tunnels and back and forth to the food source on the other end of the arena. These tunnels evenly divided the search area. The main tunnels in particular were very stable structures, not changing throughout the course of the test.

This research design demonstrates that termite tunneling is not "random" because tunneling does follow a pattern of evenly dividing the foraging area. Also, since the main tunnels were stable structures, tunneling is not continuous under our experimental conditions.

Availability of Food Resources (Oi et al., unpublished data). Our second study tested the hypothesis that declining food resources would result in a burst of tunneling activity in search of new food. Our research design was similar to that of Powell, et al., described above except that we did not include food in the introduction site. A 1-inch by 2½-inch piece of cardboard was placed about 2 inches away from the edges of the arena and away from the introduction site. Termites were allowed to feed until the food was depleted.

We observed that the main tunnels were built first, then branch tunnels. We also observed that main tunnels are relatively stable structures, but there was no significant burst of tunneling activity with declining food resources. In fact, the termites starved to death in the arena around day 31 without building additional main tunnels.

Thus, after exploration of an area and even with declining food resources, tunneling still is not random and not continuous. Therefore, we cannot accept the "random foraging" theory as applied to tunnel construction. We instead propose that there is a pattern to tunnelingsince the search area is evenly divided, and that main tunnel construction is not continuous, but stable once tunnels are built, in the absence of disturbance.

We hypothesize that factors in the field such as soil disturbance due to construction, treatment or landscaping would prompt additional tunnel building and rebuilding. If termites in the field ran out of food, they would move on to a new area to search, not simply starve to death. However, our laboratory studies seem to explain why in the absence of disturbance, termites may infest one monitor, but not another, even if they are placed just three feet away from each other.

Guidelines. Well-trained termite technicians know where termite trouble spots will be around a home. Pipes and conduits or the outer foundation wall are common points of entry into a structure. These physical structures all serve as guidelines into a building. Remember, termites are blind, soft-bodied insects and like most animals, they will follow the path of least resistance to get to their goal.

We conducted a study in 1994 to examine the effect of guidelines on food consumption. We observed that food placed along a guideline was significantly more fed upon than food that was left free-standing, regardless of whether termites were allowed to establish tunnels beforehand. We also observed that when guidelines were placed in the Nunc dish bioassay arena, termites no longer evenly divided the foraging area and branch tunnels were almost nonexistent. Because termites had something to follow, they widened the tunnels along the guideline instead of building their bypass system using branch tunnels.

Moisture. The last factor we’ll discuss that can affect tunneling is moisture. When homes are inspected, probably more than 95 percent of the "conducive conditions" listed on inspection reports relate to moisture. If our goal is to protect structures, then the issue of moisture remains important once termites infest structures, but moisture may not be the most important factor when considering how termites get into structures in the first place. Subterranean termites can be found in dry conditions and very wet conditions, sometimes even surviving light flooding. We suspect that guidelines are more important than moisture in considering how termites infest structures in the first place. We look forward to new research to illustrate this point.

CONCLUSION. Our data shows that in the absence of disturbance, termite main tunnels are stable structures. If baits or other treatments miss these main tunnels, the likelihood of termites running into the treatment is small. In the case of baits and non-repellent termiticides, if termites aren’t in the treatment, then you can’t get the active ingredient in contact with the colony, which would affect control. Therefore, we propose that termite tunneling is not random because it does have a pattern in our laboratory tests, with termites evenly dividing the search area in the absence of guidelines, and termite tunneling is not continuous because the main tunnels especially are stable structures in the absence of disturbance.

Acknowledgments

The authors would like to thank Tracie Jenkins (University of Georgia) and Steven M. Valles (USDA-ARS CMAVE, Gainesville, Fla.) for their review of the manuscript. Part of this research was supported by USDA-SRRC Project No. 6435-32000-001-00D.

References

Hedlund, J. C. and G. Henderson. 1999. Effect of Available food size on search tunnel formation by the Formosan subterranean termite (Isoptera: Rhinotermitidae). Journal of Economic Entomology, 92:610-616.

Robson, S. K., M. G. Lesniak, R. V. Kothandapani, J.F.A. Traniello, B. L. Thorne and V. Fourcassié. 1995. Nonrandom search geometry in subterranean termites. Naturwissenschaften, 82:526-528.

Su, N.-Y. and R. H. Scheffrahn. 1991. Evaluation of bait-toxicants for suppression of subterranean termite populations. Sociobiology, 19:211-220.