The Bee Community of Cannabis sativa and Corresponding Effects of Landscape Composition

Abstract

Industrial hemp, Cannabis sativa (Cannabaceae), is a newly introduced and rapidly expanding crop in the American agricultural landscape. As an exclusively wind-pollinated crop, hemp lacks nectar but produces an abundance of pollen during a period of floral dearth in agricultural landscapes. These pollen resources are attractive to a range of bee species but the diversity of floral visitors and their use of hemp across a range of agricultural contexts remains unclear. We made repeated sweep net collections of bees visiting hemp flowers on farms in New York, which varied in both landscape context and phenotypic traits of hemp varieties. We identified all bee visitors to the species level and found that hemp supported 16 different bee species. Landscape simplification negatively impacted the abundance of bees visiting hemp flowers but did not affect the species richness of the community. Plant height, on the other hand, was strongly correlated with bee species richness and abundance for hemp plots with taller varieties attracting a broader diversity of bee species. Because of its temporally unique flowering phenology, hemp has the potential to provide a critical nutritional resource to a diverse community of bees during a period of floral scarcity and thereby may help to sustain agroecosystem-wide pollination services for other crops in the landscape. As cultivation of hemp increases, growers, land managers, and policy makers should consider its value in supporting bee communities and take its attractiveness to bees into account when developing pest management strategies.

Full text

197
© The Author(s) 2019. Published by Oxford University Press on behalf of Entomological Society of America.
All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
Pollinator Ecology and Management
The Bee Community of Cannabis sativa and
Corresponding Effects of Landscape Composition
NathanielRyan Flicker, Katja Poveda, and Heather Grab1,
Department of Entomology, Cornell University, Ithaca, NY 14853, and 1Corresponding author, e-mail: heather.grab@cornell.edu
Subject Editor: Theresa Pitts-Singer
Received 24 September 2019; Editorial decision 29 October 2019
Abstract
Industrial hemp, Cannabis sativa (Cannabaceae), is a newly introduced and rapidly expanding crop in the American
agricultural landscape. As an exclusively wind-pollinated crop, hemp lacks nectar but produces an abundance of
pollen during a period of floral dearth in agricultural landscapes. These pollen resources are attractive to a range of
bee species but the diversity of floral visitors and their use of hemp across a range of agricultural contexts remains
unclear. We made repeated sweep net collections of bees visiting hemp flowers on farms in New York, which varied
in both landscape context and phenotypic traits of hemp varieties. We identified all bee visitors to the species
level and found that hemp supported 16 different bee species. Landscape simplification negatively impacted the
abundance of bees visiting hemp flowers but did not affect the species richness of the community. Plant height, on
the other hand, was strongly correlated with bee species richness and abundance for hemp plots with taller varieties
attracting a broader diversity of bee species. Because of its temporally unique flowering phenology, hemp has the
potential to provide a critical nutritional resource to a diverse community of bees during a period of floral scarcity
and thereby may help to sustain agroecosystem-wide pollination services for other crops in the landscape. As
cultivation of hemp increases, growers, land managers, and policy makers should consider its value in supporting
bee communities and take its attractiveness to bees into account when developing pest management strategies.
Key words: hemp, wild bees, honey bees, landscape simplification, plant traits
Bees provide essential pollination services in both natural and
agricultural systems; yet, both wild and managed bees have been
adversely impacted by numerous characteristics of large-scale, in-
tensied agriculture, including the widespread use of chemical pes-
ticides and insecticides, persistent pathogens and parasites, and the
loss of seminatural nesting habitat and plant diversity (Goulson etal.
2015, Dicks etal. 2016). Landscape-scale loss of natural areas and
plant diversity, a dening characteristic of intensive agriculture, oc-
curs as a consequence of the increased size and connectivity of areas
devoted to agricultural production (Meehan et al. 2011). Habitat
loss associated with agricultural land-use change imposes nutritional
stress on bee communities (Naug 2009) by reducing the diversity of
oral resources and imposing temporal gaps in resource availability
(Di Pasquale et al. 2016). Changing land use patterns, therefore,
threaten the sustainability of the pollination services that our agri-
cultural systems relyupon.
The recent federal legalization and agricultural expansion of in-
dustrial hemp, Cannabis sativa (Schluttenhofer and Yuan 2017), may
inuence the spatial and temporal distribution of pollen resources
for bee communities in the United States. Industrial hemp offers
a unique oral resource to bees in agricultural landscapes. Hemp
owers late in the summer releasing an abundance of pollen during
a period of native and agricultural oral dearth (Dalio 2012, Koh
etal. 2016). As a result, hemp pollen may offer a vital subsistence re-
source to bees at a point in the season when they are resource-limited
(Dalio 2012), thereby helping to alleviate the pressures imposed by
spatial and temporal variation in resource availability that is charac-
teristic of simplied agricultural landscapes (Schellhorn etal. 2015).
Recent studies have documented the importance of hemp pollen
in supporting a diverse community of bees during periods of oral
resource scarcity (O’Brien and Arathi 2019). Hemp visitor commu-
nities may vary across the season (O’Brien and Arathi 2019) poten-
tially reecting changes in landscape-level oral resources over time.
Specically, hemp’s pollinator community may vary temporally as
bees respond to local declines in the availability of oral resources
when the owers of other late-season crops in the landscape senesce
(Grab etal. 2017). Additionally, we know little about the effects of
varietal traits, like plant height, which have been previously shown
to alter visitor preference in other systems (Parsche et al. 2011).
Plant height in hemp is highly variable and determined by both gen-
etic and environmental factors (Campbell etal. 2019). Furthermore,
ower visitors may respond to land use change not only in terms of
Environmental Entomology, 49(1), 2020, 197–202
doi: 10.1093/ee/nvz141
Advance Access Publication Date: 2 December 2019
Research
Downloaded from https://academic.oup.com/ee/article-abstract/49/1/197/5634339 by Cornell University Library user on 05 March 2020

abundance and species richness but also in community composition
(Rollin etal. 2015). By exploring changes in composition, we can
detect more subtle changes in community structure including species
turnover whereby the identity of the species and their relative abun-
dances change across the landscape gradient. However, prior studies
have been limited in both their spatial and taxonomic resolution.
Therefore, hemp’s ability to support a diverse bee community across
a variety of contexts remains unresolved.
While it is apparent that hemp, as a widely expanding, pol-
len-abundant crop, can provide resources for bees, the bee commu-
nity visiting hemp has not yet been identied in the northeastern
U.S. agricultural landscape. In this study, we characterize the bee
community visiting hemp along a landscape gradient to determine
how shift in landscape composition affects hemp’s bee community.
Furthermore, we investigate the effects of plant height and sam-
pling date on the abundance and species richness of the hemp bee
community.
Methods
StudySystem
Hemp is a dioecious, wind-pollinated crop. It lacks brightly colored
owers, extraoral nectaries, and sweet aromas, adaptations char-
acteristic of most insect-pollinated crops (Small and Marcus 2002).
Male plants grow taller than females and release mass quantities of
pollen for several weeks until they senesce. Bees visit male owers
to collect pollen; however, bees do not visit female owers because
they lack oral nectar and, therefore, they do not contribute to hemp
pollination.
Data were collected at 11 hemp farms in the summer of 2018 in
the Finger Lakes region of New York. This region is characterized by
a wide array of agricultural and natural land uses, including dairy,
eld crop, fruit, and vegetable production as well as deciduous for-
ests, woodlots, and old eld pastures. In the Finger Lakes region,
industrial hemp is produced for grain, ber, and cannabidiol (CBD)
extraction. Cannabidiol is a nonpsychoactive phytocannabinoid
produced by female hemp plants that is used in the manufacturing
of Epidiolex, a pure concentrate of CBD used to treat severe forms of
Epilepsy, as well as in the herbal supplements industry (Hilderbrand
2018). However, because CBD production only utilizes female
plants, all bee sampling was conducted on plots producing hemp for
grain and ber, which offer both male and female plants.
The hemp elds involved in this study varied in size, with the
smallest plot being ~317 m2 (<0.1 acres) and the largest operation
cultivating roughly 42,262 m2 (>11 acres). Hemp cultivars varied
across sites, depending upon whether the plot was being culti-
vated for ber, grain, or dual-purpose production. Fields generally
contained a single variety; however, when multiple varieties were
present, they were grown in separate areas and we focused our sam-
pling efforts on the variety with the greatest bee activity. The average
height of hemp plants was estimated as tall (≥2 m), medium (1–2 m),
or short (≤1 m).
Landscape Analysis
Measures of landscape composition were used to assess the inuence
of habitat characteristics on the abundance and diversity of bees
visiting hemp plots. Using the 2018 National Agricultural Statistics
Service Cropland Data Layer for New York State in ArcGIS, we es-
timated the proportion of land in agricultural uses (row crops, or-
chards, fruit and vegetable elds) at radii of 500 and 1,000 m.The
cover of forests (wooded wetlands, deciduous, coniferous, and
mixed forest stands), seminatural habitats (fallow elds, shrublands,
hay elds, and wetlands), and urban lands were also quantied at
each scale.
Bee Community ofHemp
Bees were netted from the owers of male plants in 20-min sam-
pling transects through hemp plots and around plot perimeters.
Twenty-minute sampling periods were roughly split between tran-
sects through the eld and walks around plot perimeters, depending
on the farm and eld layout. Any bee seen landing on or collecting
pollen from a male hemp ower was collected. Sampling was fo-
cused on areas of the eld that had the greatest prevalence of open
owers and accessible pollen. Each of the sites were visited four
separate times over the course of the sampling period, amounting
to a total of 80 min of sweep-netting per site throughout hemp’s
owering period. All sampling were conducted between July 30 and
September 15 of2018.
Dry ice was used to freeze captured bees on site, and collected
bees were cleaned and pinned for ease of identication. All pinned
bees were identied to the species level using the DiscoverLife.org
keys, published revisions (Gibbs 2011, Gibbs etal. 2013), and refer-
ence materials maintained in the Cornell University Insect Collection
(http://cuic.entomology.cornell.edu/).
Statistical Methods
Total abundance and species richness of wild bees through time
was calculated for each sampling site and date across the sampling
period. The European honey bee, Apis mellifera L.(Hymenoptera:
Apidae), is a nonnative, highly managed species and their abun-
dance on hemp is likely to represent local hive density rather than
landscape context. Consequently, we evaluated the effects of land-
scape composition, sampling date, and plant height on honey bees
and wild bees separately. Although the common eastern bumble
bee, Bombus impatiens Cresson (Hymenoptera: Apidae), can also
be commercially managed, use of managed colonies in this region is
uncommon and we did not observe managed bumble bee colonies at
any of our eld sites. The effects of variation in landscape compos-
ition at each spatial scale on the abundance, and diversity of hemp’s
bee community were analyzed using generalized linear models with
Poisson error structures as implemented in the lme4 package (Bates
etal. 2015). Separate models were t for each land cover type and
scale and ranked based on Akaike Information Criterion scores cor-
rected for small sample size (AICc). The effects of plant height and
sampling date were included in each model as separate xed effects.
We then used model averaging to calculate estimates and P-values
across models with a cumulative weight ≥0.95. Post-hoc Tukey tests
were used to evaluate whether signicance of differences among
plant height categories. Additionally, we evaluated the variation in
the composition of the community using NMDS based on Gower
distances and tested whether community dissimilarity was explained
by gradients in land cover variables by permutational multivariate
analysis of variance (Oksanen etal. 2018) with 1,000 permutations.
Results
The Bee Community ofHemp
Throughout the sampling period, hemp supported a total of 16 bee
species (Table 1) and 355 individuals were captured from the 11 sites
over the sampling period. The most abundant species were A.mellif-
era, constituting nearly 60% of all captured individuals, and B.impa-
tiens, which constituted nearly 30% of hemp’s bee community (Fig. 1).
198 Environmental Entomology, 2020, Vol. 49, No. 1
Downloaded from https://academic.oup.com/ee/article-abstract/49/1/197/5634339 by Cornell University Library user on 05 March 2020

Effect of Landscape, Plant Height, and
SamplingDate
The proportion of agricultural land cover had the strongest effect on
the bee community visiting hemp (conditional model-averaged esti-
mate=−0.66, z= 2.779, P=0.005). Specically, increased agricul-
tural cover at the 1,000-m scale reduced the average number of wild
bees visiting hemp owers by as much as 76% (Fig. 2a, Supp Table 1
[online only]). Urban cover at the 1,000-m scale was associated with
an increase in bee visitation (estimate=7.41, z=2.285, P=0.022);
however, the model including urban cover was not different from a
model without any land cover variables (ΔAICc=1.27, Supp Table 1
[online only]). The height of hemp plants also had a strong inuence
on the abundance of wild bees visiting hemp owers, with tall plants
attracting nearly 17 times the number of visits compared to short
plants (estimate= −3.49, z=4.139, P<0.005, Fig. 2a, Supp Table
1 [online only]). Additionally, the average number of wild bee vis-
its increased across the sampling period (estimate=0.46, z=2.908,
P=0.003, Supp Table 1 [online only], Fig. 2b).
Abundance of A.mellifera did not vary with landscape compos-
ition (Supp Table 2 [online only]) but, similar to wild bees, increased
with plant height (conditional model-averaged estimate = −2.88,
z=2.80, P=0.005, Supp Table 2 [online only]) and sampling date
(estimate=0.48, z=2.84, P=0.004, Supp Table 2 [online only]).
The species richness of hemp visitors was greatest on tall plants
(conditional model-averaged estimate=−1.16, z=2.25, P=0.023,
Fig. 3, Supp Table 3 [online only],) but was not inuenced by sam-
pling date (estimate = −0.01, z = 0.102, P =0.918, Supp Table 3
[online only]). Landscape composition also did not have a signi-
cant effect on the bee species richness of hemp ower visitors (Supp
Table 3 [online only]). However, both forest cover (F(1,28) = 2.76,
P=0.05) and seminatural habitat cover (F(1,28)=4.38, P =0.014)
at the 1,000-m scale were associated with variation in community
composition (Fig. 4).
Discussion
Crops serve as critical resources for pollinators in agricultural land-
scapes (Westphal etal. 2003, Le Feon etal. 2010). Hemp, a newly
introduced and rapidly expanding crop in U.S. agricultural land-
scapes, offers an abundance of pollen resources to bees during a
period of oral resource scarcity in agricultural landscapes (Dalio
2012). In this study, we sought to identify the bee community vis-
iting hemp and to analyze the effects of landscape composition on
bee visitation to hemp. Hemp supported a total of 16 different social
species of bee pollinators. We found a negative impact of agricul-
tural cover on the abundance of wild bees visiting hemp. The average
number of bees visiting hemp owers increased across the sampling
period and both the abundance and diversity of the bee community
visiting hemp increased with plant height.
Table 1. Species visiting male hemp flowers in New York and
counts for each species observed
Species Count
Apis mellifera 210
Bombus impatiens 105
Lasioglossum hitchensi 11
Augochlora pura 5
Xylocopa virginica 4
Lasioglossum ephialtum 4
Lasioglossum spp. (male) 3
Lasioglossum zonulum 2
Lasioglossum paradmirandum 2
Lasioglossum zephyrum 2
Halictus confusus 1
Lasioglossum imitatum 1
Lasioglossum laevissimum 1
Lasioglossum planatum 1
Lasioglossum versatum 1
Lasioglossum oblongum 1
Lasioglossum perpunctatum 1
Fig. 1. Honey bees, Apis mellifera (a and b) and bumble bees, Bombus impatiens (c and d) collecting pollen from male hemp flowers.
Environmental Entomology, 2020, Vol. 49, No. 1 199
Downloaded from https://academic.oup.com/ee/article-abstract/49/1/197/5634339 by Cornell University Library user on 05 March 2020

Hemp provided pollen resources to important pollinators
in the NYS agricultural landscape. Overall, the community was
not dissimilar to prior work describing hemp visitors in northern
Mississippi (Lago and Stanford 1989), which reported primarily
A. mellifera, B. impatiens, and Lasioglossum (Dialictus) spp.
(Hymenoptera: Halictidae) but did contrast with recent work in
Colorado (O’Brien and Arathi 2019) which found a large number
of Melissodes bimaculata Lepeletier (Hymenoptera: Apidae) and the
cucurbit specialist, Eucera (Peponapis) pruinosa Say (Hymenoptera:
Apidae). In New York, nearly 60% of hemp’s bee community was
represented by A.mellifera, which may be considered the most im-
portant and widely relied-upon species for agricultural pollination
services in the United States (Morse and Calderone 2000). Nearly,
30% of the visitors captured on hemp were B.impatiens, a generalist
pollinator that, like A.mellifera, is intensively relied upon in wild
and managed contexts for agricultural pollination services (Kleijn
etal. 2015). Hemp also supported a diverse community of nonman-
aged bee pollinators. Wild pollinators enhance the effectiveness of
pollination services through functional complementarity with man-
aged species (Chagnon etal. 1993, Hoehn etal. 2008, Frund etal.
2013) and have become increasingly important in buffering the im-
pacts of widespread declines in managed A.mellifera populations on
agricultural pollination services (Kremen etal. 2007, Garibaldi etal.
2011, Brittain etal. 2013). As a late-season crop that blooms during
a critical period of oral resource scarcity in agricultural landscapes,
hemp may facilitate pollination services for crops in the following
year by lling periods of late-season resource scarcity (Waser and
Real 1979, Moeller 2004, Riedinger et al. 2014, Grab etal. 2017)
and reducing the spatial and temporal variation in oral resources in
agricultural landscapes (Schellhorn etal. 2015).
The height of hemp plants provided a strong indicator for both
the diversity and abundance of hemp’s bee community. In addition
to potentially providing a greater overall amount of pollen, tall
plants are more apparent than shorter varieties, which may explain
why they attracted a more abundant and species rich community of
pollinators (Russo and Shea 2017). The abundance of bees visiting
Fig. 3. The average species richness of bees sampled from hemp varied with
plant height. Letter values indicate significant differences (P<0.05) based on
post hoc contrasts with a Tukey correction.
Bombus_impatiens
Xylocopa_virginica
Augochlora_pura
Halictus_confusus
Lasioglossum_spp.
Lasioglossum_zonulum
Lasioglossum_mitchelli
Lasioglossum_ephialtum
Lasioglossum_paradmirandum
Lasioglossum_laevissimum
Lasioglossum_versatum
Lasioglossum_zephyrum
Lasioglossum_perpunctatum
Agriculture (1000m)
Agriculture (500m)
Forest (1000m)
Forest (500m)
Urban (1000m)
Urban (500m)
Seminatural (1000m)
Seminatural (500m)
−0.2
−0.1
0.0
0.1
0.2
−0.25 0.00 0.25 0.50
NMDS1
NMDS2
Species
Land cover
Fig. 4. Ordination plot showing the position of hemp visiting species (points)
relative to the vectors of environmental variables (lines). Solid lines indicate
that the environmental variable was significantly associated with variation
in the community.
Fig. 2. The average number of wild bees visiting hemp flowers varies depending on a) plant height, the proportion of agricultural land cover at 1,000 m
surrounding the field and b) the sampling date.
200 Environmental Entomology, 2020, Vol. 49, No. 1
Downloaded from https://academic.oup.com/ee/article-abstract/49/1/197/5634339 by Cornell University Library user on 05 March 2020

hemp plants also increased throughout the sampling period, which
may reect the increasing importance of hemp as a pollen-forage
resource as the owers of other late-season crops senesce at the end
of the summer. Hemp varieties that are taller and later owering
varieties may therefore better support the pollinator communities of
other crops in the agroecosystem.
At the landscape level, hemp plantings located at sites with mod-
erate agricultural cover are also likely to support a more abundant
population of crop pollinators, as we found that the number of wild
bees visiting hemp declined as the proportion of agricultural cover
in the landscape increased. Landscape simplication has been shown
to restrict the availability of foraging and nesting resources for pol-
linators in agricultural landscapes through reduced diversity of oral
resources and seasonal periods of resource scarcity (Di Pasquale
et al. 2016), which imposes nutritional stress on bee populations
(Naug 2009). Our ndings conrm the negative impacts of agricul-
tural land use change on the abundance of wild bees visiting crop
owers. Because high agricultural land cover negatively impacts the
abundance of hemp’s bee community, hemp may provide a more ef-
fective resource for bees in more complex landscapes, where it is
utilized by a greater number of bees. However, the average number
of species utilizing hemp did not vary with landscape context sug-
gesting that hemp will support a broad array of crop pollinators
across a range of landscape contexts. We observed shifts in commu-
nity composition that were driven mainly by differences in forest and
seminatural habitat cover. These patterns are likely due to the avail-
ability of suitable nesting habitat as species like Augochlora pura Say
(Hymenoptera: Halictidae), which nest in rotting logs (Stockhammer
1966), were associated with forest cover and species like Xylocopa
virginica L.(Hymenoptera: Apidae), which often nest in man-made
wooden structures (Gerling and Hermann 1978), were associated
with urban cover. These ndings suggest that hemp will support a
different suite of species in landscapes with more open natural habi-
tats, such as meadows and shrublands, compared with landscapes
with greater forested cover, reecting variation in habitat require-
ments, and life cycle characteristics of different bee species.
An important consideration that should be addressed in future
research is the relative value of hemp pollen in supporting bee repro-
duction. Mass owering crops can support pollinator populations
(Westphal etal. 2003, Jauker etal. 2012), but the incorporation of
novel pollens into the diets of generalist bees has been shown to
have detrimental effects on larval development (Williams 2003). The
presence of cannabinoids, particularly tetrahydrocannabinol (THC),
in hemp pollen (York etal. 1975) is not likely to have an impact on
bee development due to the loss of cannabinoid receptors in insects
(McPartland etal. 2001).
Conclusion
Hemp is a high pollen producing crop owering during a period
of oral resource scarcity and supports a diverse array of bees in
the northeastern U.S.landscape. The rapid expansion of hemp pro-
duction in the United States (Schluttenhofer and Yuan 2017) may
have signicant implications for agroecosystem-wide pollination dy-
namics. The potential for hemp to serve as a oral resource for bees
is inuenced by landscape composition, the height of hemp plants,
and temporal factors. Growers, extension agents and policy makers
should consider risks to bees as pest management practices are de-
veloped for this crop (Cranshaw etal. 2019). As a late-season crop
owering during a period of seasonal oral dearth, hemp may have
a particularly strong potential to enhance pollinator populations and
subsequent pollination services for crops in the following year by
lling gaps in late-season resource scarcity.
SupplementaryData
Supplementary data are available at Environmental
Entomologyonline.
Acknowledgments
We would like to gratefully acknowledge the hemp farmers who participated
in this study and to Dr. Larry Smart of the Cornell Hemp Research Program
for allowing pollinator sampling at Cornell research sites. N.F.would espe-
cially like to acknowledge the CALS Charitable Trust research support pro-
gram, which provided grant funding to support this research project.
ReferencesCited
Bates,D., M.Mächler, B.Bolker, and S.Walker. 2015. Fitting linear mixed-ef-
fects models using lme4. J. Stat. Softw. 67: 1–48. doi:10.18637/jss.v067.i01
Brittain,C., C.Kremen, and A.M. Klein. 2013. Biodiversity buffers pollin-
ation from changes in environmental conditions. Glob. Chang. Biol. 19:
540–547.
Campbell, B. J., A.F.Berrada, C. Hudalla, S. Amaducci, and J.K. Mckay.
2019. Genotype × environment interactions of industrial hemp cultivars
highlight diverse responses to environmental factors. Agrosyst. Geosci.
Environ. 2: 180057. doi:10.2134/age2018.11.0057.
Chagnon,M., J.Gingras, and D.deOliveira. 1993. Complementary aspects
of strawberry pollination by honey and indigenous bees (Hymenoptera). J.
Econ. Entomol. 86: 416–420.
Cranshaw,W., M.Schreiner, K.Britt, T.P.Kuhar, J.McPartland, and J.Grant.
2019. Developing insect pest management systems for hemp in the United
States: Awork in progress. J. Integr. Pest Manag. 10: 26.
Dalio,J.S. 2012. Cannabis sativa – an important subsistence pollen source for
Apis. J. Pharm. Biol. Sci. 1: 1–3.
DiPasquale,G., C.Alaux, Y.LeConte, J.F.Odoux, M.Pioz, B.E.Vaissière,
L.P.Belzunces, and A.Decourtye. 2016. Variations in the availability of
pollen resources affect honey bee health. PLoS One. 11: e0162818.
Dicks, L. V., B. Viana, R. Bommarco, B. Brosi, M. D. Arizmendi,
S.A.Cunningham, L.Galetto, R.Hill, A.V.Lopes, C.Pires, etal. 2016.
Ten policies for pollinators. Science. 354: 975–976.
Fründ, J., C.F.Dormann, A.Holzschuh, and T. Tscharntke. 2013. Bee di-
versity effects on pollination depend on functional complementarity and
niche shifts. Ecology. 94: 2042–2054.
Garibaldi,L.A., I.Steffan-Dewenter, C.Kremen, J.M.Morales, R.Bommarco,
S.A. Cunningham, L.G. Carvalheiro, N.P.Chacoff, J.H.Dudenhöffer,
S. S. Greenleaf, et al. 2011. Stability of pollination services decreases
with isolation from natural areas despite honey bee visits. Ecol. Lett. 14:
1062–1072.
Gerling, D., and H. R. Hermann. 1978. Biology and mating behavior of
Xylocopa virginica L. (Hymenoptera: Anthophoridae). Behav. Ecol.
Sociobiol. 3: 99–111.
Gibbs, J. 2011. Revision of the metallic Lasioglossum (Dialictus) of eastern
North America (Hymenoptera: Halictidae: Halictini). Zootaxa. 216: 1–216.
Gibbs, J., L. Packer, S.Dumesh, and B. N. Danforth. 2013. Revision and
reclassication of Lasioglossum (Evylaeus), L. (Hemihalictus) and L.
(Sphecodogastra) in eastern North America (Hymenoptera: Apoidea:
Halictidae). Zootaxa. 3672: 1–117.
Goulson,D., E.Nicholls, C.Botías, and E.L. Rotheray. 2015. Bee declines
driven by combined stress from parasites, pesticides, and lack of owers.
Science. 347: 1255957.
Grab,H., E.J.Blitzer, B.Danforth, G.Loeb, and K.Poveda. 2017. Temporally
dependent pollinator competition and facilitation with mass owering
crops affects yield in co-blooming crops. Sci. Rep. 7: 45296.
Hilderbrand, R. L. 2018. Hemp & Cannabidiol: what is a medicine? Mo.
Med. 115: 306–309.
Environmental Entomology, 2020, Vol. 49, No. 1 201
Downloaded from https://academic.oup.com/ee/article-abstract/49/1/197/5634339 by Cornell University Library user on 05 March 2020

Hoehn, P., T.Tscharntke, J. M. Tylianakis, and I. Steffan-Dewenter. 2008.
Functional group diversity of bee pollinators increases crop yield. Proc.
Biol. Sci. 275: 2283–2291.
Jauker, F., F. Peter, V. Wolters, and T. Diekötter. 2012. Early reproductive
benets of mass-owering crops to the solitary bee Osmia rufa outbalance
post-owering disadvantages. Basic Appl. Ecol. 13: 268–276.
Kleijn,D., R.Winfree, I.Bartomeus, L.G.Carvalheiro, M.Henry, R.Isaacs,
A.M.Klein, C.Kremen, L.K.M’Gonigle, R.Rader, etal. 2015. Delivery
of crop pollination services is an insufcient argument for wild pollinator
conservation. Nat. Commun. 6: 7414.
Koh,I., E.V.Lonsdorf, N.M.Williams, C.Brittain, R.Isaacs, J.Gibbs, and
T.H.Ricketts. 2016. Modeling the status, trends, and impacts of wild bee
abundance in the United States. Proc. Natl. Acad. Sci. USA. 113: 140–145.
Kremen,C., N.M. Williams, M.A. Aizen, B.Gemmill-Herren, G. LeBuhn,
R.Minckley, L.Packer, S.G.Potts, T.Roulston, I.Steffan-Dewenter, etal.
2007. Pollination and other ecosystem services produced by mobile or-
ganisms: a conceptual framework for the effects of land-use change. Ecol.
Lett. 10: 299–314.
Lago, P. K., and D. F. Stanford. 1989. Phytophagous insects associated
with cultivated marijuana, Cannabis sativa, in Northern Mississippi. J.
Entomol. Sci. 24: 437–445.
Le Feon, V., A. Schermann-Legionnet, Y. Delettre, S. Aviron, R. Billeter,
R.Bugter, F.Hendrickx, and F.Burel. 2010. Intensication of agriculture,
landscape composition and wild bee communities : a large scale study in
four European countries. Agric. Ecosyst. Environ. 137: 143–150.
McPartland, J., V. Di Marzo, L. De Petrocellis, A. Mercer, and M. Glass.
2001. Cannabinoid receptors are absent in insects. J. Comp. Neurol. 436:
423–429.
Meehan,T.D., B.P.Werling, D.A.Landis, and C.Gratton. 2011. Agricultural
landscape simplication and insecticide use in the Midwestern United
States. Proc. Natl. Acad. Sci. USA. 108: 11500–11505.
Moeller,D.A. 2004. Facilitative interactions among plants via shared pollin-
ators. Ecology. 85: 3289–3301.
Morse, R. A., and N. W. Calderone. 2000. The Value of Honey Bees As
Pollinators of U.S. Crops in 2000. Bee Cult. 128: 1–15.
Naug,D. 2009. Nutritional stress due to habitat loss may explain recent hon-
eybee colony collapses. Biol. Conserv. 142: 2369–2372.
O’Brien,C., and H. S.Arathi. 2019. Bee diversity and abundance on ow-
ers of industrial hemp (Cannabis sativa L.). Biomass Bioenergy. 122:
331–335.
Oksanen,J., F.G. Blanchet, M.Friendly, R.Kindt, P.Legendre, D.McGlinn,
P.R.Minchin, R.B.O’Hara, G.L.Simpson, P.Solymos, etal. 2018. vegan:
Community Ecology Package. R package version 2.5-2. https://CRAN.R-
project.org/package=vegan.
Parsche,S., J.Fründ, and T.Tscharntke. 2011. Experimental environmental
change and mutualistic vs. antagonistic plant ower – visitor interactions.
Perspect. Plant Ecol. Evol. Syst. 13: 27–35.
Riedinger,V., M.Renner, M.Rundlöf, I.Steffan-Dewenter, and A.Holzschuh.
2014. Early mass-owering crops mitigate pollinator dilution in late-ow-
ering crops. Landsc. Ecol. 29: 425–435.
Rollin, O., V. Bretagnolle, L. Fortel, L. Guilbaud, and M. Henry. 2015.
Habitat, spatial and temporal drivers of diversity patterns in a wild bee
assemblage. Biodivers Conserv. 24: 1195–1214.
Russo, L., and K. Shea. 2017. Deliberately increased network connectance
in a plant-pollinator community experiment. J. Complex Networks. 5:
473–485.
Schellhorn,N.A., V. Gagic, and R.Bommarco. 2015. Time will tell: resource
continuity bolsters ecosystem services. Trends Ecol. Evol. 30: 524–530.
Schluttenhofer,C., and L.Yuan. 2017. Challenges towards Revitalizing Hemp:
AMultifaceted Crop. Trends Plant Sci. 22: 917–929.
Small,E., and D.Marcus. 2002. Hemp: a new crop with new uses for North
America, pp. 284–326. In J.Janick, and A.Whipkey, (eds.), Trends New
Crop. New Uses. ASHS Press, Alexandria, VA.
Stockhammer,K. A. 1966. Nesting habits and life cycle of a solitary sweat
bee, Augochlora pura (Hymenoptera: Halictidae). J. Kansas Entomol. Soc.
39: 157–192.
Waser,N.M., and L.A.Real. 1979. Effective mutualism between sequentially
owering plant species. Nature 281: 670.
Westphal,C., I.Steffan-Dewenter, and T.Tscharntke. 2003. Mass owering crops
enhance pollinator densities at a landscape scale. Ecol. Lett. 6: 961–965.
Williams,N.M. 2003. Use of novel pollen species by specialist and generalist
solitary bees (Hymenoptera: Megachilidae). Oecologia. 134: 228–237.
York,N., B.Garden, and E.Botany. 1975. The constituents of Cannabis sativa
pollen. Econ. Bot. 29: 245–253.
202 Environmental Entomology, 2020, Vol. 49, No. 1
Downloaded from https://academic.oup.com/ee/article-abstract/49/1/197/5634339 by Cornell University Library user on 05 March 2020