A description of plant disease spatial distribution is the foundation for understanding epidemiology. When the pattern of disease is described, logical hypotheses about the factors driving the distribution can be explored. The first step for describing disease spatial distribution is determining what vines are infected. At present, there are no standard protocols for sampling vineyards for Pierce's disease, which is transmitted by xylem-feeding insects and is caused by the bacterium Xylella fastidiosa Wells et al. This study was initiated to improve sampling protocols for choosing grapevine foliage to identify X. fastidiosa-infected grapevines (Vitis vinifera L.) in California vineyards and is part of a larger program to develop Pierce's disease sampling protocols. We found that PD symptoms, defined by leaf necrosis and chlorosis, internodal distance, petiole length and weight, and extent of cane branching were not reliable indicators of X. fastidiosa infection. The matchstick symptom of petioles was the only consistent indicator of infection in X. fastidiosa-positive grapevines. Further study revealed that leaves selected from the most basal nodes of positive canes had the highest probability of X. fastidiosa detection, as determined by testing with enzyme-linked immunosorbent assay. We developed a symptom reliability index to evaluate the reliability of symptoms for PD diagnosis by node location and this index will be described.