Flowering plants predominantly conduct water in tubes known as vessels, with vessel diameter playing a crucial role in plant adaptation to climate and reactions to climate change. The importance of vessels makes it essential to understand how and why vessel diameter, plant height, and other ecological factors are interrelated. Although shoot length is by far the main driver of variation in mean vessel diameter across angiosperms, much remains to be understood regarding the factors accounting for the abundant variation around the y-axis in plots of mean species vessel diameter against shoot length. Here, we explore the potential role of porosity types, wood density, leaf phenology, background imperforate tracheary element type, vasicentric tracheids, vascular tracheids, perforation plate type, and successive cambia in causing variation in the y-intercept or slope of the mean species vessel-diameter– and vessel-density–shoot-length associations at the shoot tip and base. We detected numerous cases of ecologically significant variation. For example, latewood vessels of ring porous species scale with a lower slope than earlywood, i.e., latewood vessels are relatively narrow in taller plants. This pattern is likely the result of selection favoring freezing-induced embolism resistance via narrow vessels. Wood density was negatively associated with vessel diameter, with low wood density plants having wider vessels for a given height. Species with scalariform perforation plates scale with a lower shoot base vessel-diameter–shoot-length slope, likely reflecting selection against scalariform plates in wide vessels. In other cases, functional groups scaled similarly. For example, species with successive cambia did not differ from those with conventional single cambia in their mean vessel-diameter–shoot-length scaling, rejecting our prediction that species with successive cambia should have narrower vessels for a given shoot length. They did, however, have fewer vessels per unit shoot cross-sectional area than plants of similar heights, likely because vessels have longer functional lifespans (and therefore are fewer) in species with successive cambia. Our methods illustrate how vessel diameter can be studied taking shoot length into account to detect ecologically important variation and construct theory regarding plant adaptation via the hydraulic system that includes plant size as a vital element.