Flora of North America species comparison

Petals in Oenothera flava typically range from 7–32 mm with floral tubes 24–100 mm; however, plants from three disjunct areas: the Mogollon Plateau in Arizona to Catron County, New Mexico; Sacramento Mountains and Sierra Blanca, Lincoln and Otero counties, New Mexico; and the Sierra Madre Occidental from northern Chihuahua south to Durango, have much larger petals (30–55 mm) and longer floral tubes (80–265 mm). They were originally recognized as a distinct species or most recently as a subspecies (O. flava subsp. taraxacoides), but detailed study of the variation pattern suggests that the larger flowers occur in areas of high hawkmoth species diversity and higher rates of outcrossing, similar to the pattern discussed in detail by D. P. Gregory (1963, 1964). R. A. Raguso et al. (2007) and H. E. Summers et al. (2015) came to the same conclusion in an independent study of floral biology of sect. Lavauxia. Because populations from the three disjunct areas appear to have diverged independently from lower-elevation source populations, it seems best to treat the complex as one variable species without any formal subdivision.

(Discussion copyrighted by Flora of North America; reprinted with permission.)

P. H. Raven and D. P. Gregory (1972[1973]) found various levels of hybridization and intergradation between Oenothera filiformis and O. lindheimeri to occur where their ranges overlap. In the region of geographical overlap, most populations of O. filiformis are strigillose in the inflorescences and have evening-opening flowers, while O. lindheimeri is villous in the inflorescences, and has large morning-opening flowers. Moreover, O. lindheimeri occurs only on black clay prairie soil, while O. filiformis occurs in light, sandy soil, as it does throughout its range, and in more disturbed areas. Despite these differences, Raven and Gregory found hybridization between these species scattered across an area from eastern Texas across Louisiana to Alabama. At some locations there is apparently no hybridization, while at others hybrids were uncommon to relatively common. Intermediate morning-blooming plants in Alabama appear to represent evidence of past hybridization since O. lindheimeri does not occur there. Many of the individuals they tested had somewhat reduced pollen fertility (40–70% fertile). They suspected that habitat disturbance was primarily responsible in many cases, but they also detected what may have been intergradation resulting from past hybridization outside of the current distribution of O. lindheimeri. Many of these individual cases deserve further investigation to better understand the dynamics of the interactions between these species and if any of the interactions have led to stabilization of novel populations that might be recognized taxonomically.

Very few collections have been made from areas on the periphery of the range of Oenothera filiformis (southern Ontario, Connecticut, Maryland, Massachusetts, Pennsylvania), and populations in these areas probably represent recent human-based introductions. P. H. Raven and D. P. Gregory (1972[1973]) determined O. filiformis to be self-incompatible.

(Discussion copyrighted by Flora of North America; reprinted with permission.)