Given insect pollinators frequently have positive impacts on crop yields, the economic benefits of conspicuous pollinators such as managed Western honey bees, Apis mellifera L. (Hymenoptera: Apidae), are investigated often (Klein et al. 2007; Khalifa et al. 2021). However, increased global concern for Western honey bee health and management sustainability has brought increased attention to wild insect pollinators, such as bumble bees Bombus spp. Latreille (Hymenoptera: Apidae) and flies (Diptera) (Garibaldi et al. 2013; Orford et al. 2015; Rader et al. 2016). Notably, it has been found for many crops that pollinator diversity contributes more to increasing yields than pollinator abundance (Hoehn et al. 2008; Brittain et al. 2013; Rogers et al. 2014). This is because pollination services by multiple species can be synergistic, increasing cross-pollination among flowers due to pollinator variation in preference for flower height, foraging times, and behaviors for manipulating the pollen they carry (Hoehn et al. 2008; Rogers et al. 2014). For example, Cusser et al. (2021) found that butterflies (Lepidoptera) and flies (Diptera) visited flowers spatially and temporally separated from bees and ultimately contributed about one-third of pollinator mediated yield in upland cotton, Gossypium hirsutum L. (Malvaceae).

Overall, a better understanding of pollination services provided by diverse pollinators is expected to aid development of sustainable agricultural practices (Garratt et al. 2014; Dainese et al. 2019). This is particularly true for cotton production where pollination services remain underused despite findings that ambient pollinators contribute upwards of 20% of the final yield (Pires et al. 2014; Cusser et al. 2016, 2021; Muhammad et al. 2020). Upland cotton, G. hirsutum, is a major crop in the southern US; cotton was valued at $7 billion in 2019 (USDA-ERS 2022). Pollinators may have been overlooked historically in cotton agriculture because cotton is self-pollinating and cross-pollination is not necessary for cotton to develop fruit (McGregor 1976). It also is likely that pollination by insects in cotton fields has been limited in the past due to high pesticide use (Muhammad et al. 2020). Nevertheless, pollinators may become increasingly important contributors to improving cotton production, especially now that pesticide use has been significantly reduced in many regions (Free 1993; Ward & Ward 2001; Pires et al. 2014; Cusser et al. 2016). For example, Pires et al. (2014) observed a positive correlation between bee species richness and cottonseed yield. Additionally, manipulative studies show introduced honey bees can increase seed-cotton yield by 16.5 to 24.5% (McGregor et al. 1955; Rhodes 2002) and introduced bumble bees can increase seed-cotton yield by 17.0 to 46.6% (Saeed et al. 2012).

One area where information regarding pollination services in cotton remains sparse is the capacity for less conspicuous insects to be pollinators. For instance, in addition to Hymenoptera, Diptera, and Lepidoptera, insects such as beetles (Coleoptera), thrips (Thysanoptera), and true bugs (Hemiptera) are known to be pollinators in other systems (Ananthakrishnan 1982; Kevan & Baker 1983; Young 1986; Willemstein 1987; Wheeler 2001). The only published research the authors could locate on this subject in cotton was by Pierre and Hofs (2010); they reported that flower beetles, Astylus atromaculatus (Blanchard) (Coleoptera: Melyridae) were unexpected, but efficient pollinators of cotton in South Africa. Astylus atromaculatus frequently visited flowers and carried a similar pollen load as Western honey bees (Pierre & Hofs 2010). Although these beetles also are considered pests of cotton seeds and seedlings, they could provide some benefit to crop production once the plant reaches the flowering stage by cross-fertilizing flowers (McGregor 1976; Pierre & Hofs 2010). Cotton flowers are only open for 1 d for pollination before withering, and so opportunities to cross-pollinate with viable pollen grains are limited (Stewart et al. 2010). An improved understanding of which insects are visiting cotton flowers and are capable of cross-pollinating (i.e., Cusser et al. 2016, 2021) is needed to maximize opportunities for increasing cotton yield worldwide without increasing land-use for agriculture (Muhammad et al 2020).

We observed that plant bugs, cotton fleahoppers, Pseudatomoscelis seriatus Reuter (Hemiptera: Miridae), are frequent flower visitors of cotton, G. hirsutum in Texas. Cotton fleahoppers are native to the southwestern US and are considered early season pests of cotton because they feed on flower buds and cause them to abscise (Stewart & Sterling 1989). Once the cotton is in its flowering stage, however, fleahoppers can be seen dusted with pollen while foraging within cotton flowers (L. G., personal observation). These fleahoppers likely are visiting flowers to feed on pollen grains (Burden et al. 1989; Wheeler 2001), but because fleahoppers are very mobile insects (Reinhard 1926), we hypothesized that fleahoppers might contribute to cross-pollination of cotton as they forage among flowers while carrying pollen grains.

Herein we present the results of our investigation into the pollination abilities of the cotton fleahopper. We quantified how frequently fleahoppers visited cotton flowers and how many cotton pollen grains they carried on their bodies (i.e., their pollen load). We also used fluorescent powder as a pollen analog to estimate fleahopper dispersal among flowers and pollen deposition behaviors. Finally, we determined their pollination efficiency by measuring the capacity of cotton fleahoppers to pollinate self-sterile flowers.

Materials and Methods

Results

FLEAHOPPER POLLEN LOAD

The majority of fleahoppers collected (27 of 32; 85%) were carrying pollen grains (Fig. 1). On average, fleahoppers carried 25.1 ± 40.8 pollen grains per insect (median: 5.5 pollen grains). The maximum number of pollen grains observed per fleahopper was 163 pollen grains.

Fig. 1.

Photographs of a cotton fleahopper, Pseudatomoscelis seriatus, taken under a dissecting microscope (A) foraging on a cotton, Gossypium hirsutum, stigma, and (B) carrying cotton pollen grains.

Discussion

Cotton fleahoppers are clearly anthophilous (flower-loving) insects. We observed adult cotton fleahoppers foraging frequently in cotton flowers and with pollen grains attached to their legs, abdomen, and antennae. In addition, cotton fleahoppers were able to disperse among flowers inside field cages after being dusted with a pollen analog. In our experiment, however, we observed that plants visited by pollen loaded cotton fleahoppers produced a similar number and mass of seeds, as well as similar lint weight as emasculated flowers. In contrast, plants able to self-cross or experience ambient pollination produced more seeds, seed weight, and lint weight; this result is similar to that observed by others (Saeed et al. 2012; Cusser et al. 2016).

Inadequate pollination likely is explained in part by our observation that cotton fleahoppers carried a pollen load much smaller than optimal for upland cotton flower fertilization. Approximately 50 viable pollen gains are needed to fertilize a full-sized cotton fruit (McGregor 1976), but cotton fleahoppers carried a median pollen load of 5.5 pollen grains. Relatively small pollen loads likely are due to the small size of this fleahopper (3.0–4.0 mm in length) and their smooth bodies. In contrast, Western honey bees are exceptionally hairy, up to 4 times larger than cotton fleahoppers (12–15 mm), and they can carry around 500 cotton pollen grains (Pierre & Hofs 2010). Moreover, furry bumblebees are up to 7 times larger than cotton fleahoppers and can carry thousands of cotton pollen grains (Berger et al. 1988). We did not observe active pollen collecting behaviors by cotton fleahoppers, unlike that typical of bees collecting food for their nests and which are reliant only on floral resources. Fleahoppers, in contrast, are also omnivores, feeding on other small insects such as lepidopteran eggs (L. G., unpublished data).

Nevertheless, we predict that the small size of cotton fleahoppers would not prevent them from being pollinators of other host plants with smaller or composite (i.e., clustered) flowers, like those in the plant family Asteraceae or Apiaceae (Willemstein 1987; Wheeler 2001). Levin et al. (1967), for example, found that plant bugs, Lygus hesperus Knight (Hemiptera: Miridae), were an efficient pollinator of safflower, Carthamus tinctorius L. (Asteraceae). In their experiment, L. hesperus transferred pollen to 28% of self-sterile safflowers inside field cages (Levin et al. 1967). Carthamus tinctorius has a composite flower head that is 15 mm in width and pollen grains that are 52 to 67 µm, approximately half the size of cotton pollen grains (Smith 1996). Like L. hesperus, cotton fleahoppers frequent many other host plants with much smaller flowers than cotton, such as wholly croton, Croton capitatus Michx. (Euphorbiaceae), silverleaf nightshade; S. elaeagnifolium; and horsemint, Monarda punctata L. (Lamiaceae) (Esquivel & Esquivel 2009). Our study, however, is the only study of cotton fleahopper pollination. How cotton fleahoppers affect the reproduction of their wild hosts is unknown.

While individual cotton fleahoppers may not be important cotton pollinators, cotton fleahoppers could contribute to cotton cross-pollination as part of a community of flower visitors (Wheeler 2001; Parys et al. 2020). In the future, a treatment group in which pollen carrying cotton fleahoppers are released onto flowers also exposed to ambient pollinators may illuminate the possibility of community contribution (see for example: Cusser et al. 2016, 2021). In addition, our results are consistent with others who found that self-crossed flowers and ambient pollinated flowers produced similar yield outcomes (Saeed et al 2012; Cusser et al. 2016). Other experiments, however, highlight that cotton is pollen limited. Cusser et al. (2016), for instance, found that hand pollinated flowers produced more seed-cotton weight than other treatment groups, including ambient pollination. Likewise, Saeed et al. (2012) found variety specific increases in seed-cotton with hand pollination compared to ambient pollination. These results alongside findings of increased yield following pollinator augmentation (i.e., McGregor et al. 1955; Rhodes 2002; Saeed 2012), suggest that cotton is not only pollen limited if only able to self-cross but also pollinator limited within ambient pollinator communities (Rhodes 2002). Therefore, efforts to not only conserve diverse pollinator populations, but to increase their abundance in cotton fields may be paramount to realize the economic benefits of pollination in cotton agroecosystems (Pires et al. 2014; Cusser et al. 2016; Muhammad et al. 2020).

Fig. 2.

Pollination of cotton, Gossypium hirsutum, by cotton fleahopper, Pseudatomoscelis seriatus: (A) number of seeds per fruit among pollination treatments; (B) seed mass per fruit among pollination treatments; (C) lint weight per fruit among pollination treatments. Bars represent treatment means and error bars represent ± standard error of the mean. Treatment means listed with the same letter are not significantly different (P > 0.05).

Avenues of continued research in cotton pollination include: (1) documenting flower visitor diversity and pollinator efficiency in different agricultural cotton growing regions and in wild G. hirsutum populations (Mayes & Petrillo 2017; Parys et al. 2020); (2) observing effects of landscape management on ambient pollinator populations (Cusser et al. 2016; Dainese et al. 2019); (3) determining effects of genetically modified cotton traits on pollination services (Hofs et al. 2008; Niu et al. 2017, 2018); (4) investigating pollinators as mediators of gene flow between genetically modified cotton and wild cotton plants and the ultimate impacts of gene flow on wild cotton ecology and evolution (Vázquez-Barrios et al. 2021); (5) determining effects of cotton variety on yield outcomes for self and out-crossed flowers (Saeed et al. 2012; Stein et al. 2017); (6) observing the impacts of pollinator augmentation to cotton fields (Rhodes 2002, Saeed 2012); and (7) determining effects of nectar robbing by ants attracted to nectaries and aphids on pollinator visitation (Levan & Holway 2015). Overall, improving pollination services in cotton agriculture is expected to be a sustainable approach to increase yield without increasing the area of cultivated land (Aizen et al. 2009; Kevan et al. 2009).

To date, few authors have investigated the pollination abilities of plant bugs (Wheeler 2001). Plant bugs are inefficient pollinators for some plants (Bohart & Nye 1960; Lindsey 1984), but the notion they could be minor pollinators for others should not be disregarded without experimentation (Scott 1983; Wheeler 2001). We found that cotton fleahoppers can readily move and transfer a pollen analog to cotton stigmas and we suggest that they could contribute to cotton pollination as part of a community of pollinators, or pollinate some of their wild hosts. When studying pollinator communities, plant bugs should not be overlooked because bugs carry pollen too. Frequent flower visitors like plant bugs could contribute to the pollinator diversity that we rely on to increase crop yields.