Pollinators link a plant's floral morphology with its reproductive success, thereby shaping floral trait evolution. Accordingly, common functional pollinator groups create convergences in floral traits known as pollination syndromes (Faegri & van der Pijl, 1979). By appraising a plant's syndrome, we may infer its pollinator or pollinator guild. Although a single, dominant pollinator species or guild may contribute exclusively to a plant's reproductive output, the contributions of other less-effective or less-frequent floral visitors are also occasionally noted to contribute to pollination (hereafter called secondary pollinators; Rosas-Guerrero et al., 2014, Stebbins, 1970). Aside from some well-studied, generalized pollination systems, the extent of secondary pollinator contributions to plant reproduction and their potential influence on plant evolution are largely overlooked (Fenster et al., 2004). However, an estimated 70% of plants interact with at least one secondary pollinator (Ashworth et al., 2015). Thus, considering secondary pollinator roles in plant reproduction has the potential to explain perplexing reproductive trait variation that does not match a known primary pollinator, as well as to reveal important species interactions. Throughout most of its range, sweet sand verbena (Abronia fragrans, Nyctaginaceae) displays white flowers between dusk and dawn that emit a strong fragrance to attract nocturnal moth pollinators (Keeler & Fredricks, 1979). Yet, in a small region of northern Texas and southwestern Oklahoma, A. fragrans bears pink flowers (Figure 1). These pink flowers still close during the heat of the day and receive considerable nocturnal visitation by diverse moths; however, they remain open late into the morning and reopen early in the evening, and receive frequent diurnal visitation by butterflies and bees (C. Girvin, S. Jaeger, and E. LoPresti, personal observations, Figure 2). Whether this shift in flower color and floral closing behavior alters this wildflower's interactions with insects and whether diurnal floral visitors contribute any pollination services to pink-flowered populations are unknown. To quantify the independent contributions of nocturnal and diurnal pollinators to reproductive success of A. fragrans, we conducted a pollinator-exclusion study. We performed this experiment, coupled with floral visitor observations, in a pink-flowered population at the Gene Howe Wildlife Management Area in Hemphill County, Texas in May 2021. To parse diurnal and nocturnal pollination contributions, we excluded floral visitors during the day and/or night by covering inflorescences with small organza drawstring bags. We alternately assigned each of four pollinator-exclusion treatments (with a random order) to inflorescences as follows: open to pollinators overnight (“night” treatment), open during the day (“day” treatment), open both night and day (“always” treatment), and never open (“closed” treatment). When the population was in full bloom, we opened/closed bags during every dawn and dusk from the day of first flower until senescence of the inflorescence (usually 4–6 days). Upon senescence, we permanently bagged each inflorescence to prevent herbivory during fruit development. Although we tagged 360 inflorescences in this population, the actual sample size was lower due to herbivory and cattle trampling (with 322 surviving inflorescences on 103 plants, comprising 18,144 individual flowers). We collected all bagged inflorescences ~3 weeks later and scored pollination success. Since all Abronia set a single seed per fruit, we recorded whether each flower produced a seed or not. As expected, given the nearly complete self-incompatibility of most Abronia species (Jabis et al., 2011, Williamson & Bazeer, 1997), closed treatments had virtually no seed set (0.11%) and therefore were not analyzed further or discussed in what follows. We analyzed the data using a binomial (seed present or seed absent) mixed model with treatment (night, day, or always open) as a fixed effect and the plant ID as a random effect (as most plants had all three treatments); the model with treatment fit significantly better than one without treatment (likelihood ratio test: χ2 = 1153.2, df = 2, p < 0.0001). We then compared treatments using a Tukey's post hoc comparison of means and found that all treatments significantly differed from each other (Figure 3: day-night and day-always: p < 0.0001, night-always: p = 0.0027). In contrast to white-flowered populations farther north where nocturnal moths were the sole pollinators (Keeler & Fredricks, 1979, Appendix S1), diurnal visitors contributed considerable pollination to this pink-flowered population (Figure 3). Nocturnal pollination alone resulted in 93% and diurnal pollination resulted in 18% of mean pollination success for always-open inflorescences. Although most pollination was nocturnal, mean full seed set was not realized unless the inflorescence was exposed to both nocturnal and diurnal pollinators (Figure 3). This result demonstrates that secondary pollinators of the pink-flowered population contributed significantly to the plants' overall pollination success, in stark contrast to the white-flowered populations (Appendix S1: Figure S1). Therefore, we suggest not only that the atypical flower color and the floral closing and opening behavior alter this species' plant–insect interactions, but also that diurnal pollinator interactions may be a selective force behind the evolution or maintenance of pink floral color in these populations. What species are contributing to this diurnal pollination? Aside from flower color resulting from different betalain concentrations in perianth tissue, pink- and white-flowered plants share the same reproductive organ morphology. With stamens inserted shallowly below the floral tube opening and a single pistil deep in the tube (~1–2 cm), this floral shape prevents pollination by any insect lacking a long proboscis or tongue. Although a few small solitary bees (Halictidae) visited the pink sand verbena flowers, A. fragrans' floral shape inhibits effective bee pollination. However, even small butterflies like skippers are capable of Abronia pollination and carry substantial loads of pollen on their proboscises (Appendix S2: Figure S1). We observed butterflies—most commonly the skipper Lerodea eufala—as the most frequent and abundant diurnal visitors of the pink-flowered population with potential for pollination. Therefore, we suggest that butterfly pollination accounted for most diurnal pollination in the pink-flowered population in this year. In contrast, during a week of observations (also in the 2021 season) in a white-flowered population at the Cimarron National Grassland in Kansas, several butterfly species were common in the vicinity, yet we did not observe a single diurnal floral visitor! Across the genus, Abronia spp. flowers consistently exhibit narrow sepal tubes and emanate heavy fragrance—traits suggestive of “classic” moth pollination. In sand verbena species in which pollination systems have been investigated, moths are suggested to be primary pollinators. Still, many of these species' flowers are not white or day-closing. These species include the white-flowered and (nearly always) day-closing A. ammophila (Saunders & Sipes, 2006); pink-flowered, day-closing A. macrocarpa (Williamson et al., 1994); pink-flowered, always-open A. umbellata (Doubleday & Eckert, 2018); and the purple-flowered, always-open A. alpina (Jabis et al., 2011). Floral trait divergence from the typical moth–pollination syndrome is common among other sand verbena species as well, observed in floral color (pink, yellow, or magenta flowers) and open period (flowering through night and day). These Abronia species, and other plants that have evolved floral traits that deviate from a pollination syndrome, deserve further scrutiny into the potentially important reproductive contributions of secondary pollinators in driving these floral trait shifts. We hope that this small experiment will provide an impetus to pollination biologists to observe and quantify the pollination contributions of “less-important” floral visitors to pollination in many plants. We observe that our field sites are ancestral lands of several peoples, many of whom were forcibly removed from or moved to these areas, and we celebrate their tremendous stewardship and traditional ecological knowledge of these lands. These groups include Kiikaapoi (Kickapoo), Tawakoni, and Wichita tribes at our Texas site and Arapaho, [Gáuigú (Kiowa), Nʉmʉnʉʉ (Comanche), and “” (Osage) tribes at our Kansas site. We also thank Gene Howe Wildlife Management Area, Texas and Cimarron National Grasslands, Kansas, for maintaining quality and easily accessible A. fragrans habitat. We are very grateful to Kevin Keegan at the University of Connecticut for his valuable butterfly identification. The authors declare no conflict of interest. Data are archived in Figshare separately for two sand verbena populations studied: https://doi.org/10.6084/m9.figshare.21720989.v1 (Jaeger et al., 2022a) and https://doi.org/10.6084/m9.figshare.21721028.v1 (Jaeger et al., 2022b). All analyses are included in one script on Figshare: https://doi.org/10.6084/m9.figshare.21721109.v1 (Jaeger et al., 2022c). Appendix S1. Appendix S2. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.