A fractal analysis is presented for the detection of pathogens such as Francisela tularensis, Yersinia pestis (the bacterium that causes plague), Bacillus anthracis, Venezuelan equine encephalitis (VEE) virus, Vacinia virus, and Escherichia coli using a CANARY (cellular analysis and notification of antigens risks and yields) biosensor (Rider et al., 2003). This has biodefense applications. In general, the binding and dissociation rate coefficients may be adequately described by either a single- or a dual-fractal analysis. An attempt is made to relate the binding rate coefficient to the degree of heterogeneity (the fractal dimension value) present on the biosensor surface. Binding and dissociation rate coefficient values obtained are presented. Due to the dilute nature of the analyte(s) present, in some cases, a triple-fractal analysis is required to adequately describe the binding kinetics. This was noted at the lower end of the analyte concentration spectrum analyzed for the VEE virus (5000 pfu dual-fractal analysis; 500 pfu, triple fractal analysis), and for B. anthracis (10,000 cfu, dual-fractal analysis; 1000 cfu, triple-fractal analysis). Only two data sets are presented here. However, if this trend is observed for the detection of other pathogens, then this makes the detection of these pathogens at the lower end of the concentration spectrum more and more challenging. It should also be noted, and this is not entirely unexpected, that there is a lot of variation in the original experimental data when dilute concentrations of the analyte were analyzed by the CANARY biosensor (Rider et al., 2003). The data analyzed appears smoother since only discrete points at different time intervals were analyzed. The kinetics aspects along with the affinity values presented are of interest, and should along with the rate coefficients presented for the binding and the dissociation phase be of significant interest in help designing better biosensors for an application area that is bound to gain increasing importance in the future. In a general sense, fractal models are fascinating. Newer avenues are required to analyze and to help detect pathogens at very dilute concentration levels. The analysis of the studies of the boundaries (scale) over which the fractal behavior occurs should prove useful. The real interesting test of the fractal model would be if it can make a prediction that turns out to be correct. This would extremely valuable, especially in the detection of pathogens. Any increase in time that is made available to help in the evacuation process (for example, by making better biosensors) after the establishment of a pathogenic threat is invaluable.