Two taxa interacting half a world away

Observations were made by sitting/walking in a residential lawn in Baton Rouge, Louisiana, USA, during late September 2013 through early October 2013 (Fig. 1). Occurrences of honey bees visiting centipede grass were documented with both video and still imagery. All observations were between 10:00 AM and 2:00 PM, consisting of video, and one session of macro-photography (Fig. 2). The grass was sampled for identification and a specimen sheet was created at Louisiana State University Herbarium (Fig. 3). Other plants at anthesis that could provide potential forage for bees were also noted. (Table 1).

Table 1. Download as CSV 

Other plants at anthesis in association with E. ophiuroides.

Species	Family
Duchesnea indica (Andrews) Focke	Rosaceae
Mikania scandens B.L.Rob.	Asteraceae
Ligustrum sinense Lour.	Oleaceae
Lablab purpureus (l.) Sweet	Fabaceae
Kyllinga brevifolia Rottb.	Cyperaceae
Oplismenus hirtellus (L.) P. Beauv.	Poaceae
Digitaria ciliaris (Retz.) Koeler	Poaceae
Ruellia simplex C.Wright	Acanthaceae
Brugmansia sp. Pers.	Solanaceae

Table 2. Download as CSV 

Pollen analysis after acetolysis from bee corbiculae

Bees – using one corbicula	Sampled pollen grains	Percentage Poaceae pollen
1	252	100%
2	266	100%
3	270	100%

Figure 2.

Honey bee moving pollen up the culm, while also spreading pollen through biotic winds; here with pollen visible at left and below bee

Figure 3.

Eremochloa ophiuroides specimen collected at observation locality

A total of three bees were sampled for taxonomic identification and examined by curators at the Louisiana State University Arthropod Museum (Fig. 4). One honey bee pollen basket was then sampled for homogeneity at Louisiana State University Center for Excellence in Palynology (Fig. 6). Three honey bee corbiculae pollen contents were then sent to and processed by at Washington State University via acetylosis. Pollen identification was later performed by the Palynology consultants at University of Arizona.

Figure 4.

Collected honey bee with body dusted in pollen and packed pollen baskets or corbiculae

Figure 5.

Centipede grass at anthesis

Figure 6.

Pollen sample at 20× from one bee corbicula demonstrating homogeneity. Image by: Sophie Warny

a

b

Figure 7.

Inflorescences demonstrating color change in anthers after dehiscence from purple to brown

a: anthers before and after dehiscing
b: another culm showing same anther color changes

Equipment used:

• Galaxy Note I cell phone for video
• Nikon D300 DSLR camera with a 1:1 macro lens for still images
• Olympus Microscope with slaved digital camera for microscopy images

Rainfall prior and post-observations; a wet summer/early fall and not a time of drought stress/starvation (Table 3, Suppl. materials 1, 2, 3, 4, 5, 6)

Table 3. Download as CSV 

Precipitation amounts for summer and early fall 2013, in Baton Rouge, Louisiana, USA. http://www.ncdc.noaa.gov/

Month (2013)	Precipitation (cm)
June	10.4
July	11.9
August	10.9
September	19.3
October	7.9
Total	60.4

The honey bees were exclusively gathering unifloral Poaceae pollen (Table 2). Macro-photography revealed that as the bees traveled from inflorescence to inflorescence, they generated biotic winds that moved the Poaceae pollen significant distances (Fig. 8).

a

b

c

d

Figure 8.

"Pollen biotic winds" caused by honey bees are denoted by arrows showing dispersal of grains between culms – original images available at http://www.herbarium.lsu.edu/keys/pensoft/bees3/

a: Pollen dispersal caused by bee
b: Spreading of pollen as result of bee
c: Pollen both above and below bee
d: Distribution and distance of pollen travel from bee

The graminoids are treated in botanical literature as using the pollination syndrome of anemophily (Niklas 1985), or abiotic wind pollination (c.e.g., Walters and Keil 1996). A reason for this abiotic relationship are that the flowers are small and drab in appearance rather than showy (Knuth 1909). In contrast, though diminutive and lacking petals and sepals, most graminoid inflorescences are quite colorful when blooming, plus present ultraviolet visual cues that are visible to the bees but cannot be seen by humans (Baby et al. 2013). This dichotomy reveals an interesting question: is this just a scale problem for attractiveness? Insects are resourceful feeders, and will take advantage of pollen feeding opportunities that are acceptable and provide visual signatures of readiness for anther dehiscence (Fig. 7).

Honey, a well documented economic commodity that is studied and sampled for purity and origins, tells a different story from botanical literature. Melissopalynology, or the study of pollen in honey, describes the collection of graminoid pollen by honey bees as commonplace. This literature is not isolated but found from across the globe, describing collection of pollen from all graminoids: Poaceae (Keller et al. 2005), Cyperaceae​ (Song et al. 2012), and Juncaceae (Huang et al. 2013). Interestingly, additional observations of other non-graminoid anemophilous plants of the Angiospermae, are also found to have associations with honey bees; including willows (Salicaceae) (Puusepp and Koff 2014, Salonen et al. 2009), oaks (Fagaceae) (Stawiarz and Wroblewska 2010, Bryant 2001), and even glassworts (Sarcocornia) (Adam et al. 1987).

Thanks to Chris Reid for editing and correction of my Rosaceae identification; and to Sophie Warny, Vaughn Bryant, and John Jones, concerning all things about pollen identification; and to Mary Barkworth for the advice to catch some of those bees.

Tim Jones observed the process and collected all samples. Dr. Victoria Bayless identified the honey bees as honey bees. The fly was identified by Tim Jones using the LSU Entomology collection. Dr. John G. Jones identified the pollen samples collected from the bees. Dr. Sophie Warny and Dr. Vaughn Bryant provided in-person and email palynological advice, respectively.