Vessel-length distributions in some trees, shrubs, and a vine have been calculated from measurements of particle penetration and of air-volume flow through the xylem. In shrubs and diffuse-porous species, longest vessels were about 1 m long, but most of them were much shorter, the largest percentage in the 0–10 cm length class. In the two ring-porous species investigated (Quercus rubra and Fraxinus americana), the longest vessels often were as long as the tree's stem, but most of them were much shorter. In the grapevine (Vitis labrusca) which has large-diameter vessels (ca. 300 μm) a small percentage of the vessels was 8 m, but most of them were less than 5 m long. In a given species, lengths of the longest vessel were quite variable, but the distribution of the short lengths was more constant. In general, vessel lengths are correlated with vessel diameters: wide vessels are longer. Even in diffuse-porous species, the slightly narrower latewood vessels are somewhat shorter than the wider early wood vessels. The method is a simplified version of that described by Skene and Balodis, but using a programmable desk calculator. It works best with diffuse-porous species in which vessels are randomly distributed in the stem, and less well in species with wide vessels, because as vessels reach the length of the stem itself, they cannot be randomly distributed.