Numerous authors have attempted to quantify the physical properties of leaves in relation to aspects of leaf ecology, including decomposition, sclerophylly, herbivory, and leaf function and longevity. This paper examines the relative merits of the punch-and-die, tearing and shearing tests for assessing leaf physical properties. We conducted a series of these three mechanical tests on leaves of Solanum laciniatum, and determined the effect of various test parameters on the measurement of fracture properties. For the punch-and-die test, the parameters considered were machine speed, clearance between the punch and the die, edge definition of the punch, and area of the punch. Aspects of the tearing test examined were notch length, end effects, and length-to-width requirements of test strips, and for shearing tests the effects of blade proximity, angle and sharpness were investigated. All the test parameters investigated were found significantly to affect the assessment of leaf-fracture properties. In addition, fracture properties were found to vary significantly within leaves. Some general principles for designing and implementing tests are outlined. This study suggests that while punching and shearing tests are useful means of quantifying leaf fracture properties, the value of the tearing test may be reduced as it is most constrained by the biological nature of the test material.

Leaf biomechanical properties have the potential to act as antiherbivore defences. However, compared with studies on chemical defences, there are few studies that have demonstrated that the physical or biomechanical structure of plants can prevent or influence herbivory. This difference in focus by ecologists may relate to the dominant paradigm of plant chemical defences in ecological research and the perceived difficulties that ecologists have with the engineering principles embedded in biomechanics. The advantage of using materials engineering concepts is that each property is precisely defined and quantifiable, although the latter may be difficult in leaves because of their composite and anisotropic nature. Most herbivory studies have used simple penetrometers to measure leaf properties, often termed ‘toughness’. As defined in materials engineering, the measured properties are ‘force to fracture’ and ‘strength’, not toughness. Measurement of strength, the resistance to crack initiation, is relevant to understanding herbivory. Measurement of ‘toughness’ as defined by materials engineering is also relevant. Toughness is the resistance to crack propagation and is a measure of the energy required to fracture the leaf. This requires more sophisticated equipment than simple penetrometers because it requires a simultaneous measure of the punch displacement. In addition, purists would argue that a punch cannot be used to measure true toughness because the crack is not controlled and plastic deformation is also involved. However, it may be the only method that allows detection of fine‐scale pattern in mechanical properties across a leaf surface at a scale that is relevant to herbivory. There is very little work on the scale at which these properties vary, particularly with regard to different sized herbivores. In addition, few studies have investigated a broad range of relevant biomechanical properties in relation to herbivory. Therefore, it is not possible yet to be definitive about the relative merits of the various types of tests. A single test might show a pattern in relation to herbivore damage at a gross level. However, to really understand the functional and ecological significance of leaf texture in relation to herbivory, a more reductionist approach is needed. Only then can we move on to the larger scales of pattern that many ecologists are seeking.

Trail traffic of the leaf-cutting ant Atta cephalotes involves intermingled flows of outbound and returning foragers. Head-on encounters between workers from the opposite flows are a common occurrence in this traffic. Each encounter momentarily delays the two ants involved, and these small delays might pose a significant cost to the colony's foraging performance when summed over thousands of workers along many metres of trail. We videotaped outbound and returning foragers over a 1 m course, and measured the encounter rates they experienced and their velocity. Our analysis indicates that locomotion speed is diminished by increasing encounter rate, but that the effect is small relative to the effects of ant body size and load mass. Head-on encounters allow exchange of information and leaf fragments between workers, and we consider how the benefits of such encounters may make this form of traffic organization superior to segregated outbound and returning lanes, despite the measurable cost of encounters in mixed traffic.

No abstract

Colonies of Atta cephalotes (Myrmicinae: Formicidae) construct cleared paths between their nest and the vegetation sources at which they harvest leaf tissue. Here, we employ ideas from traffic engineering to study streams of laden and unladen ants on these paths. The relationship between average traffic speed and the concentration of workers on the road surface follows a relationship similar to what is expected by analogy to fluid dynamics. Although the traffic is composed of eusocial organisms with a common interest in group success, the coarse-grained behavior of Atta traffic displays little more coordination than a moving fluid. The relationship between speed and concentration implies that maximum flow rates (which are likely to be closely tied to colony-level rates of resource acquisition) occur at a relatively high concentration that keeps individual speeds well below their "free flow" maximum. We predict that this optimal concentration will characterize peak traffic throughout a trail network, and we propose a simple behavioral mechanism that would allow trails to be cleared to the correct width to provide the optimal concentration. Collisions (including encounters for antennation) are common in leaf-cutting ant traffic because traffic is not segregated into unidirectional streams. Nonetheless, we find a counterintuitive suggestion that flow rates (with concentration differences statistically removed) are higher when traffic is near a 50:50 mix of outbound and returning ants than when it contains majority flows in a single direction. Mixed-direction traffic may help disperse laden ants with reduced agility, thereby preventing inhomogeneities in the traffic stream that could clog the trail.