The native range of this species, in its strict sense, may have been largely restricted to the Coastal Plain and Piedmont from Va. to La., in contrast to macrostachya, but some authors have not considered these two taxa distinct (Sm, F; Isely 1998; W). It is possible that some plants in southern Appalachian regions are intermediate, but no definitive study has been made. Reports of typical frutescens from Ky. (e.g. Duncan 1967) deserve further investigation, and may be based on the plants mapped here under macrostachya from more sandy alluvium along Appalachian rivers: in LAUR (BEREA), MCRE (KY), PULA (KY) and WHIT (MM). [There is similar biogeography within Isotrema and Catalpa.] Typical frutescens has much smaller racemes, mostly 4-12 cm long (versus 15-30 cm); pedicels and calyx have few or no clavate glands (versus abundant). Leaflets tend to be smaller, less acuminate and darker glossy bluish green; hairs on stems, petioles and pedicels are generally denser, less appressed, and more whitish (versus slightly yellowish). In recent decades, several cultivars of both taxa have become widely grown and appear quite distinct: for example, macrostachya "Blue Moon" versus frutescens "Amethyst Falls" (Valder 1995). Flower color, fragrance, growth form and hardiness vary much among cultivars and wild types.