2010 NORTHEASTERN NATURALIST 17(3):455–472
Cavity-nesting Wasps and Bees of Central New York State:
The Montezuma Wetlands Complex
Kevin M. O’Neill1, * and James F. O’Neill2
Abstract - Solitary nest-provisioning wasps and bees in North America include
species that naturally construct nests within existing cavities, such as hollow plant
stems or tunnels left by wood-boring insects. The materials used to construct brood
cells within nest cavities and the types of food provisions provided to offspring
vary considerably among species. Over five summers (2001–2002, 2005–2007),
we used trap nests to survey the cavity-nesting wasp and bee assemblage within
the Montezuma Wetlands Complex in central New York State. Over 350 trap nests
were occupied by 6 species of apoid wasps (Sphecidae, Crabronidae; 34% of nests),
7 vespid wasps (Vespidae: Eumeninae; 39%), 2 spider wasps (Pompilidae; 3%), and
12 bees (Megachilidae, Colletidae; 26%), as well as brood parasites and parasitoids
of the nest provisioners. The most common nest-provisioning wasp was Trypoxylon
lactitarse, followed by Ancistrocerus antilope, Isodontia mexicana, Symmorphus
canadensis, Symmorphus cristatus, and Euodynerus foraminatus. The only two bee
species with comparable incidences were Hylaeus annulatus and Heriades carinatus.
Natural enemies emerging from nests included at least 17 species from 10
families, the most common of which were brood-parasitic cuckoo wasps (7 species
of Chrysididae; 39 nests) and flies (Sarcophagidae; 11 nests). We also report brood
sex ratios of the seven most abundant species, finding them to be either male-biased
(A. antilope, T. lactitarse), female-biased (E. foraminatus), or not significantly different
from unity. We compare our survey results to others done in north-central
and eastern North America.
Solitary aculeate bees and wasps construct nests in a variety of locations,
using a wide range of nesting materials (Krombein 1967, O’Neill 2001).
Some species build free-standing nests of mud attached to rocks, plants, or
human structures. Others excavate tunnels in soil or plant materials, such
as rotten wood or pith-filled plant stems. Finally, the so-called “cavitynesters”
seek out existing cavities, commonly either hollow plant stems
or tunnels left by emerging wood-boring insects. Cavity-nesting females
usually modify nest cavities by adding partitions and plugs consisting, in
different species, of mud, plant resins, fresh or dried plant materials, or debris
gathered from the environment. Cavity-nesters have long been studied
with the use of “trap nests”, whose basic design consists of either natural
tubes made from hollow, dried plant stems or artificial tunnels such as paper
straws or holes drilled in wood (Krombein 1967). Trap nests are relatively
1 Department of Land Resources and Environmental Sciences, Montana State University,
Bozeman, MT 59717; 2188 Woodlawn Avenue, Auburn, NY 13021. Corresponding
author – email@example.com.
456 Northeastern Naturalist Vol. 17, No. 3
inexpensive, and can be placed in large numbers in appropriate locations
to provide nesting habitat for pollinators and biological control agents, aid
in detailed biological studies of individual species, or survey local communities
of cavity-nesters. Because of their habitat requirements, trophic
diversity, and roles as plant pollinators and hosts of natural enemies, cavitynesters
and other solitary nest-provisioners have been proposed as indicator
species for environmental change, including the effects of invasive species
(Barthell et al. 1998, Gayubo et al. 2005, Tscharnke et al. 2003).
From 2001–2007, we used trap nests to conduct ecological studies of an
assemblage of cavity-nesting wasps and bees and their insect natural enemies
at the Montezuma Wetlands Complex (MWC) in central New York State.
The MWC consists of ≈15,000 ha of public and private lands set aside to
preserve wildlife habitat. Most of the land is situated in Montezuma National
Wildlife Refuge (MNWR, administered by the US Fish and Wildlife Service)
and the Northern Montezuma Wildlife Management Area (administered by
the New York State Department of Environmental Conservation). Several of
our biological studies on particular species in this area have been published
(Jensen et al. 2007, O’Neill and O’Neill 2009, O’Neill et al. 2007). However,
one of our goals has been to provide an inventory of the overall assemblage
of insects occupying trap nests at the MWC. Therefore, we here report on
the abundance, species composition, and (for some species) offspring sex
ratios of nest-provisioning species and their insect natural enemies. To our
knowledge, our study is the first such survey reported for the MWC and one
of relatively few done in the northeastern US. We compare our results to
earlier surveys done in eastern and north-central North America (Fye 1965,
Jenkins and Matthews 2004, Koerber and Medler 1958, Krombein 1967,
Taki et al. 2008a).
Materials and Methods
The trap nests consisted of 16-cm-deep wood blocks drilled with 15-cm
long holes of different diameters, into which we inserted cardboard tubes
(Custom Paper Tubes, Inc., Cleveland, OH) with inside diameters of 3.2,
4.3, 5.0, 6.0, 7.0, 8.1, and 9.1 mm to provide potential nest sites for cavitynesting
species (Krombein, 1967); hereafter, we use nest size values rounded
to the nearest integer. The cardboard tubes were open at both ends, but the
inner ends abutted the back wall of the wood blocks. Each set of trap nests
consisted of multiple boards containing, in combination, 6–10 tubes of each
diameter. Krombein (1967) noted that trap nests were more likely to be occupied
when placed at the edges of woods, rather than in “dense, shaded
areas”. Therefore, we placed bundles of wood blocks on fence posts or attached
them to trees at heights of ≈1.5 m (also following Krombein 1967) on
the edges of groups of shrubs and trees, so that the southeast-facing openings
of the trap nests were exposed to full sunlight in the morning, but were at
least partially shaded in the afternoon. Such placement likely enhances the
2010 K.M. O’Neill and J.F O’Neill 457
ability of the insects to be active earlier in the morning, while protecting nest
occupants from intense insolation during the afternoon.
In 2001 (21 May) and 2002 (16 May), sets of trap nests were placed at
five locations within and near the MNWR, which straddles Seneca, Wayne,
and Cayuga counties. Four sets were within the refuge itself: 1) MNWR-1:
at the southern edge of the refuge’s “North Spring Pool” (42°58'49.00"N,
76°46'23.62"W); 2) MNWR-2: along the western edge of a meadow at the
refuge’s “Overlook” (42°58'29.84"N, 76°46'14.43"W); 3) MNWR-3: along
the western edge of a meadow on Lay Road (42°58'12.32"N, 76°46'48.08"W);
and 4) MNWR-4: along the northern edge of a meadow 160 m south of
MNWR Headquarters (42°57'56.44"N, 76°44'18.46"W). A fifth set, that we
refer to as MNWR-5, was placed 2.5 km SE of the refuge at the edge of a forest
clearing on private property (42°56'50.26"N, 76°43'1.22"W). Common
tree species in vicinity of trap nests included Tilia americana L. (Basswood),
Prunus virginiana L. (Chokecherry), Populus deltoides Bartram ex. Marsh
(Eastern Cottonwood), Fraxinus pennsylvanica Marsh. (Green Ash), and
Acer saccharinum L. (Silver Maple).
In 2005 (6 June), 2006 (13 June), and 2007 (21 May), trap nests were
placed at ten (2005) or six (2006–2007) locations within the Northern
Montezuma Wildlife Management Area, in a part of the area within Cayuga
County referred to as “Howland Island”, which is bounded by the Seneca
River and the Erie Canal. Habitat on the island contains a mix of marshes,
meadows, agricultural fields (some fallow and weedy, some planted with
corn), and woodlands with Eastern Cottonwood, Green Ash, Silver Maple,
Fagus grandifolia Ehrh. (American Beech), Juglans nigra L. (Black Walnut),
Salix nigra Marsh. (Black Willow), Rhamnus cathartica L. (European
Buckthorn), Rhus typhina L. (Staghorn Sumac), and Quercus bicolor Willd.
(Swamp White Oak). All trap nests were placed within 50 m of two dirt roads
on Howland Island referred to on trail maps as Hunter’s Home Road and
Wood Duck Loop; the roads are not open to the public, so receive little traffic. Traps were placed in an area bounded by 43°4'40.46"N, 76°42'2.17"W
on the west side of Howland Island to 43°5'17.32"N, 76°40’'19.43"W in the
center of the island.
Three to five times each year, from mid-June to late September, we
visited all sets of trap nests, removed nest tubes that had final plugs made
by the nest-provisioners, and replaced them with empty tubes, except on
the last visit of the year. Nests were kept at room temperature in ventilated
plastic bags until mid-November when they were transferred to Montana
State University and placed in cold storage (8 °C, 85% relative humidity).
All tubes were removed from cold storage in April of the following year
and placed individually in glass culture tubes with ventilated lids; a piece
of fine-meshed fabric was placed between the lid and tube to help prevent
wasps of the genus Melittobia (Eulophidae) from entering or leaving. Still,
some nests were parasitized by Melittobia, so any glass tubes containing
458 Northeastern Naturalist Vol. 17, No. 3
Melittobia were immediately placed in a freezer to kill the wasps and prevent
them from spreading to other nests; because this also killed surviving
offspring of the nest provisioners, these nests were not included in the survey
results; fewer than 10% of nests were lost in this way. All nests were
checked daily for emergence of the offspring of nest-provisioning bees and
wasps, or non-Melittobia parasitoids and predators. Emerging insects were
freeze-killed within vials labeled with a nest identification number and date
of emergence. After all insects emerged from the 2006 nests, we dissected
the nest tubes to identify any adults that did not exit nests, examine contents
of cells that did not produce offspring, and when possible, count the number
of cells constructed.
Insects were identified using published keys and web resources for bees
(Droege 2009, Michener et al. 1994, Mitchell 1962) and wasps (Bohart and
Kimsey 1982; Bohart and Menke 1963, 1976; Buck et al. 2008; Coville
1982; Sandhouse 1940; Townes 1957; Vincent 1979), and all identifications
were checked by one of us (K.M. O’Neill) against specimens in the Cornell
University Insect Collection and the Montana Entomology Collection at
Montana State University.
To examine offspring sex ratios within nests, we used chi-square
goodness-of-fit tests (1 d.f. each) to test the hypothesis that sex ratios deviated
from the null hypothesis of 1:1. We used chi-square contingency table
analyses to compare 1) sex ratios between nests of different diameters and
2) wasp:bee species ratios between sites or studies. All tests conducted had
Nest-provisioning wasps and bees within nests
Trap-nesting insects and their insect associates emerged from 379 nest
tubes. The 347 nests that could be attributed to one or more nest-provisioning
species were occupied by 27 species of solitary bees and wasps
(Table 1), including 6 apoid wasps (Sphecidae, Crabronidae; 34.3% of
nest tubes), 7 vespid wasps (Vespidae: Eumeninae; 38.6%), 2 spider wasps
(Pompilidae; 3.2%), and 12 bees (Megachilidae, Colletidae; 25.9%). All of
the nest-provisioning species are endemic to North America, with the exception
of Megachile rotundata. The seven most common species— Trypoxylon
lactitarse (19.3% of nests), Ancistrocerus antilope (17.3%), Hylaeus annulatus
(7.8%), Isodontia mexicana (6.3%), Symmorphus canadensis (6.1%),
Euodynerus foraminatus (5.8%), and Symmorphus cristatus (5.8%)—occupied
nearly 70% of the 347 nests. The seven rarest species were each found
in less than 1% of nests. Among species that emerged from at least 5% of
nests at a site, Heriades leavitti and Megachile centuncularis were found
only at MNWR, while E. foraminatus and S. cristatus were present only on
Howland Island. Among rarer species, three Megachile (M. centuncularis,
M. mendica, M. pugnata) and Osmia lignaria were collected only at MNWR,
2010 K.M. O’Neill and J.F O’Neill 459
and Ancistrocerus adiabatus, M. quadridens, M. campanulae, and O. pumila
only on Howland Island.
Because of the different timing and intensity of sampling, few statistical
comparisons can be made between the results of the surveys at MNWR and
Howland Island, but one difference is notable: the ratio of wasp to bee nests
was 5.7:1 at Howland Island (n = 255), but just 0.9:1 at MNWR (n = 99;
chi-square contingency table analysis: χ2 = 53.24, d.f. = 1, P < 0.0001).
Most nest provisioners used a range of nest diameters spanning no
more than three of the seven tunnel diameters provided (Table 1). All eight
H. leavitti nests, for example, were in 3-mm tubes (Jensen et al. 2007). In
contrast, three species used a range of four (Dipogon sayi) or five (Ancistrocerus
antilope, E. foraminatus) of the seven sizes available (Table 1). The
most anomalous nest was in a 9-mm tube that had four cells of T. lactitarse
cells along with two of Hylaeus annulatus, which otherwise used only 3–4
mm tubes. The maximum number of cells within individual nests was ≥10
in seven species (Table 1). The greatest number of cells in any one nest was
21, found in a 3-mm tube that had 19 H. annulatus cells and 2 Passaloecus
cuspidatus cells. Our best quantitative estimate of the number of cells per
nest came from T. lactitarse nests dissected in 2006, which had a mean ±
standard error of 4.8 ± 0.3 cells (range = 3–8, n = 27).
Offspring of two different nest-provisioning species emerged from 14
nests (nest diameter in mm in parentheses): I. mexicana / T. lactitarse (8),
I. mexicana / M. pugnata (8), P. cuspidatus / S. canadensis (4), P. cuspidatus
/ Hylaeus annulatus (4), Trypoxylon collinum / Heriades carinatus (5),
Trypoxylon frigidum / Hylaeus annulatus (3), T. lactitarse / A. antilope (8),
T. lactitarse / Hylaeus annulatus (9), A. antilope / E. foraminatus (4), Heriades
carinatus / Heriades leavitti (3*), Heriades carinatus / M. campanulae
(5), Heriades leavitti / Hylaeus annulatus (3*), M. campanulae / O. pumila
(6), and M. centuncularis / M. pugnata (6); the two records marked with an
asterisk were previously given in Jensen et al. (2007). During 2003–2004,
when we conducted a focal study of I. mexicana at MNWR (O’Neill and
O’Neill 2009), 21% of 58 I. mexicana nests at the MNWR-1 site had been
originally occupied by T. lactitarse or M. relativa. However, one cannot
determine from such cohabitation data whether the interactions between
species involved usurpation or whether I. mexicana simply took over nests
previously abandoned by the other species.
Another case of co-habitation occurred between T. frigidum and
Passaloecus (probably P. cuspidatus) at Howland Island, and although no
offspring of the latter emerged from the nest, several resin partitions indicate
that Passaloecus occupied the nest tube after the Trypoxylon cells were
constructed. This particular case illustrates a possible hidden cost to offspring
developing in a nest taken over by another species. The single adult
offspring of the T. frigidum, as well as an adult Trichrysis doriae (a brood
parasite of Trypoxylon; Bohart and Kimsey 1982), were found dead behind
a hardened resin partition. Apparently, the offspring of Trypoxylon and
460 Northeastern Naturalist Vol. 17, No. 3
Table 1. Number of nests occupied by cavity-nesting species at the Montezuma Wetlands Complex. Numbers given for each species are frequencies of nests occupied
Montezuma National Wildlife Refuge (2001–2002)/Howland Island (2005–2007); sex ratios are for combined data.
Sex ratio of Maximum #
Nest diameter (mm) emerging offspring of offspring
3 4 5 6 7 8 9 TotalA as % females (n) from single nest
Isodontia mexicana (Saussure) - - - - 0/1 4/8 2/7 6/16 42.2 (83) 9
Trypoxylon collinum Smith 0/2 0/1 2/2 - - - - 2/5 62.5 (16) 3
T. frigidum Smith 3/4 0/6 - - - - - 3/10 36.7 (30) 5
T. lactitarse Saussure - - - - 6/5 2/26 1/27 9/58 43.1 (255) 8
Passaloecus cuspidatus Smith 1/0 5/2 - - - - - 6/2 50.0 (26) 6
P. monilicornis Dahlbom 1/0 1/0 - - - - - 2/0 0.0 (4) 3
Ancistrocerus adiabatus (Saussure) - - 0/4 - - - - 0/4 42.9 (7) 4
A. antilope (Panzer) - 0/1 2/11 1/6 2/8 0/17 0/12 5/55 36.1 (158) 9
Euodynerus foraminatus (Saussure) - 0/1 0/9 0/2 0/4 0/2 0/2 0/20 61.3 (80) 9
Monobia quadridens (L.) - - - - - - 0/2 0/2 0.0 (3) 2
Symmorphus albomarginatus (Saussure) - 1/0 3/0 0/1 1/1 - - 5/2 61.9 (21) 4
S. canadensis (Saussure) 7/9 0/4 - 0/1 - - - 7/14 47.1 (85) 11
S. cristatus (Saussure) 0/8 0/9 0/3 - - - - 0/20 46.0 (78) 10
2010 K.M. O’Neill and J.F O’Neill 461
Table 1, continued.
Sex ratio of Maximum #
Nest diameter (mm) emerging offspring of offspring
3 4 5 6 7 8 9 TotalA as % females (n) from single nest
Dipogon sayi sayi Banks - 0/1 0/2 1/0 1/2 0/3 - 2/8 43.5 (23) 6
Auplopus mellipes (Say) - - - - - - 0/1 0/1 0.0 (1)
Heriades carinatus Cresson 3/6 1/2 0/4 - - - - 4/12 46.2 (39) 10
H. leavitti Crawford 8/0 - - - - - - 8/0 41.9 (43) 14
Hoplitis spoliata (Provancher) - - 1/0 - - - - 1/0 0.4 (5) 5
Megachile campanulae Robertson - - 0/1 0/3 - 0/1 - 0/5 0.6 (20) 9
M. centuncularis (L.) - - - 2/0 5/0 - - 7/0 86.4 (44) 13
M. mendica Cresson - - - - 2/0 - - 2/0 88.9 (18) 10
M. pugnata Say - - - 1/0 - 3/0 - 4/0 0.0 (6) 2
M. relativa Cresson - - - - 2/7 1/1 - 3/8 48.0 (50) 9
M. rotundata F. - 2/0 - - - - - 2/0 57.1 (7) 6
Osmia lignaria Say - - - 2/0 - - - 2/0 0.0 (10) 6
O. pumila Cresson - - 0/1 0/4 - - - 0/5 69.6 (23) 7
Hylaeus annulatus (L.) 18/3 1/4 - - - - 0/1 19/8 54.9 (175) 19
AThe total number of nests from which at least one species emerged was 347, but 14 nests contained offspring of two species, so the grand total in this column
462 Northeastern Naturalist Vol. 17, No. 3
Trichrysis were unable to break through the resin barrier created by the
Offspring sex ratios
Among the seven species that occupied at least 20 nests (Table 1), two
displayed overall male-biased sex ratios among offspring (A. antilope: χ2 =
12.25, P < 0.001; T. lactitarse: χ2 = 4.80, P = 0.03), while one produced excess
female offspring (E. foraminatus: χ2 = 4.05, P = 0.04). In four species,
there was no sex-ratio bias: I. mexicana (χ2 = 2.04, P = 0.15), S. canadensis
(χ2 = 0.29, P = 0.59), S. cristatus (χ2 = 0.46, P = 0.50), and H. annulatus (χ2 =
1.65, P = 0.20).
For some species, sex ratios varied among nests of different diameter. For
A. antilope, 4–7-mm nests produced 17% females (n = 81), while 8–9-mm
nests produced 54% females (n = 72) (chi-square contingency table analysis:
χ2 = 22.9, d.f. = 1, P < 0.001). In T. lactitarse, 7–8-mm nests produced
28% females (n = 145), while 9-mm nests produced 63% females (n = 110;
χ2 = 30.3, d.f. = 1, P < 0.001). Such differences also occurred in species that
displayed no overall sex-ratio bias. For H. annulatus, 3-mm nests produced
48% females (n = 146), while 4-mm nests produced 93% females (n = 27;
χ2 = 18.3, P < 0.001). In I. mexicana nests, 7–8-mm nests produced 30%
females (n = 50), while 9-mm nests produced 61% females (n = 33; χ2 =
7.64, P = 0.006). However, for E. foraminatus, we found no difference in
sex ratios between 4–6-mm (58% females, n = 40) and 7–9-mm nests (65%
females, n = 40) (χ2 = 0.21, P = 0.65). Similarly, no differences were found
when comparing 3-mm nests to larger nests in either S. canadensis (χ2 =
0.19, P = 0.66) or S. cristatus (χ2 = 0.08, P = 0.78).
Along with the progeny of nest-provisioners, a diverse set of 102 other
insects from three orders emerged from nests (Table 2). All but one of these
insects are parasitoids, brood parasites, or predators of nest-provisioning
bees and wasps; the exception was Perilampus hyalinus, a known parasitoid
of sarcophagid flies.
Several researchers in the past 50 years have also conducted surveys
of trap-nesting bees and wasps in eastern and north-central North America
(Table 3). In all of the cited studies, trap nests were placed in multiple
locations, although the surveys varied in duration and in the types of
microhabitats in which nests were placed. Another major difference was
in the range of trap-nest diameters provided. A few studies provided no
trap nests with tunnel diameters <6 mm, while Krombein (1967) set out
12.7-mm diameter nests.
The two studies conducted closest to our site were those of Krombein
(1967) and Taki et al. (2008a). At Derby, NY, 180 km SSW of the MNWR,
2010 K.M. O’Neill and J.F O’Neill 463
Krombein placed trap nests on and near human structures and along creek
banks. Among 346 trap nests occupied, the ≈25:1 ratio of wasp to bee nests
was even more extreme than we observed at Howland Island. Among the
wasps, >69% were eumenines, including A. antilope (32%) and two species
of Ancistrocerus (≈20% combined) not found at MWC. All three eastern
North American Symmorphus were also present, as well as E. foraminatus.
Among the apoid wasps, P. cuspidatus, T. collinum, and T. frigidum were
recorded at both Derby and MWC, but I. mexicana was absent at Derby
and T. lactitarse was present in a smaller percentage (8%) than at the MWC
(19%). Four of six bees at Derby were also found at the MWC, but the four
did not include the most common bee at MWC, H. annulatus. Because
Krombein provided one size class of trap nests (12.7 mm) of greater diameter
than the largest in our study (9.1 mm), it is possible that we could
have missed or under-sampled some species that he found. However, the
only two common species using 12.7-mm tubes at Derby, A. antilope and
T. lactitarse, were the two most common wasps at MWC. On the other
hand, we may have underestimated the abundance of M. quadridens, whose
nests in Krombein’s (1967) studies were “almost all” in 12.7-mm tubes
(though never at Derby, NY).
Taki et al.’s (2008a) study in southern Ontario was conducted 310 km W
of the MWC, also within a mixture of forested and agricultural lands. The
results, however, contrast strongly with ours, perhaps partly because their
trap-nest sites and nest-box orientations were chosen randomly as part of
their particular experimental design, whereas our traps were placed specifi-
cally at the edges of open areas (e.g., roadways or forest clearings) and all
faced southeast. The Canadian study also included a test of the effect of
covering some sets of trap nests with burlap covers, but we consider their
combined data set here. The greatest difference between the two sites is that
no bees occupied nests in Ontario, suggesting major habitat differences;
nests in Ontario were also monitored throughout the summer, so any differences
between the studies are not likely to be related to temporal patterns
of sampling. Among the wasps using their trap nests (n = 531), eumenines
predominated (84% of nests), with A. antilope being most common (69%).
Apoid wasps were rare (three Trypoxylon nests and one I. mexicana).
Overall, there are broad similarities between the studies at MWC, Derby,
and Ontario, in that 14 of 24 species at Derby and 8 of 12 species at the Canadian
site also occurred at MWC. The differences among the sites are likely
related to site-selection methods, as well as historical land-use patterns and
local site conditions, which have been shown to be important (Barthell et
al. 1998, Fye 1972, Gathmann et al. 1994, Steffan-Dewenter 2003, Taki
et al. 2008b, Tscharnke et al. 2003). At a smaller spatial scale, we observed
variation in species assemblages among the sampling locations in our study.
At MNWR, for example, the ratio of wasp to bee nests was 0.06:1 for the
combined data set for MNWR-1 and MNWR-3, but reversed to 3.9:1 at
MNWR-5 (χ2 = 39.9, d.f. = 1, P < 0.0001). Similarly, at Howland Island in
464 Northeastern Naturalist Vol. 17, No. 3
Table 2. Non-nest provisioning insects emerging from trap nests (both sites). Genus abbreviations: A. = Ancistrocerus, I. = Isodontia, T. = Trypoxylon, H. =
Hylaeus, S. = Symmorphus, M. = Megachile.
Species that Type of # of Number diameters
Natural enemy provisioned nest enemyA nests emerged (mm) Published host recordsF
Nemognatha sp. Megachile sp.B BP 1 1 7 Megachilidae
Unknown sp. Unknown Pr 1 1 3 cavity-nesting bees
Eustalomyia sp. UnknownC BP 1 2 5 Eustalomyia vittipes (Zett.) attacks I. mexicana
Amobia sp.D A. antilope BP 1 6 8 Vespidae (Eumeninae), Sphecidae, Crabronidae
I. mexicana 4 17 8–9
T. lactitarse 2 2 9
Unknown 4 7 4–6
Unknown sp. Unknown Pa 1 2 7
Leucospa affinis Say Megachile sp.B Pa 1 1 7 bees, including Megachile
Gasteruption sp. H. annulatus Pa 1 3 3 Gasteruption assectator L. attacks H. annulatus
Perilampus hyalinus Say Unknown Pa 1 1 4 Sarcophagidae
2010 K.M. O’Neill and J.F O’Neill 465
Table 2, continued.
Species that Type of # of Number diameters
Natural enemy provisioned nest enemyA nests emerged (mm) Published host recordsF
Chrysis cembricola Krombein S. canadensis BP 1 1 3 Symmorphus canadensis
C. coerulans F. Unknown BP 16 16 5–9 Ancistrocerus, Euodynerus, Parancistrocerus, S. cristatus
C. nitidula F. A. antilope BP 7 15 5–7, 9 Ancistrocerus, Euodynerus, Symmorphus
Unknown 6 9 5–8
Chrysis sp. A. antilope BP 1 1 9
Chrysura pacifica (Say) Osmia sp. BP 1 1 6 Osmia, including O. pumila,
Trichrysis carinatus (Say) T. lactitarse BP 4 4 8–9 Trypoxylon, including T. collinum and T. lactitarse
Unknown 2 2 4
Trichrysis doriae (Gribodo) T. frigidum BP 1 1 4 Trypoxylon, including T. frigidum and T. collinum
Coelioxys moesta Cresson M. centuncularis BP 1 1 7 Megachile, including M. centuncularis
M. relativa 1 1 7 Megachile, including M. relativa
Sapyga louisi H. leavittiE BP 2 6 3 Heriades carinatus
Sapyga sp. O. pumila BP 1 1 6 Sapyga centrata Say attacks O. pumila
ABP = brood parasite, Pa = parasitoid, Pr = predator.
BLikely one of the species listed in Table 1, but dead adult in nest was not identifiable to species.
CThis was clearly a Trypoxylon nest, with mud plugs and remains of spider prey.
DThese records count only emerged adult flies; some nests also contained many puparia that may well have belonged to Amobia.
EThis record reported earlier in Jensen et al. (2007).
FFrom Matthews (1965), Krombein (1967), Krombein et al. (1979), and Bohart and Kimsey (1982).
466 Northeastern Naturalist Vol. 17, No. 3
Table 3. Comparison of trap-nest surveys conducted in eastern and northern United States, listed in order from north to south.
Duration # of
North of survey nests Diameters # of Bee:wasp
Site latitude (years) occupied placed (mm)F species nest ratio Most common species (% of all nests)
Western OntarioA 49°20' 3 202 6, 8 9 -
WisconsinB 42°30'– 46°50' 1 778 6, 8 22 - Euodynerus foraminatus (21), Dipogon sayi sayi (18),
Ancistrocerus antilope (18)
MWC 42°56'–43°5' 5 347 3, 4, 5, 6, 7, 8, 9 27 0.34:1 Trypoxylon lactitarse (19), A. antilope (17), Isodontia
Southern OntarioC 42°37'–42°48' 1 531 3, 5, 7, 9 12 0:1 A. antilope (68), Auplopus mellipes Say (9), D. sayi (6),
Ancistrocerus adiabatus (6)
Derby, NYD 42°42' 8 372 3, 5, 10, 13 21 0.04:1 A. antilope (30), Symmorphus cristatus (17), Ancistrocerus
catskill (Saussure) (12)
Plummers Island, MDD 38°58' 7 762 3, 5, 10, 13 32 0.36:1 T. lactitarse (29), Osmia lignaria (18), T. clavatum Say
Kill Devil Hills, NCD 36°00' 3 252 3, 5, 10, 13 21 0.07:1 T. collinum (18), T. clavatum Say (15), Euodynerus
megaera (Lepeletier) (8)
Georgia and South CarolinaE 33°70'–35°02' 1 255 6, 10, 13 11 0.25:1 E. megaera (36), I. mexicana (19), Megachile frigida
Lake Placid, FLD 27°10' 5 780 3, 5, 10, 13 29 0.1:1 E. foraminatus (31), Pachodynerus erynnis (Lepeletier)
(11), M. quadridens (10)
AFye 1965 (paper gives results for wasps only).
BKoerber and Medler 1958 (paper provides individual abundances for most abundant of 22 species).
CTaki et al. 2008a.
EJenkins and Matthews 2004.
FValues rounded to nearest 1 mm.
2010 K.M. O’Neill and J.F O’Neill 467
2007, all 27 completed nests at sites 1 and 2 were provisioned by eumenine
wasps, whereas 12 of 23 nests at site 3 were occupied by bees (χ2 = 27.0,
d.f. = 1, P < 0.0001), and only 2 by eumenines. Thus, characterizations of
overall cavity-nester assemblages within an area are improved by placing
multiple sets of trap nests in different locations.
In Wisconsin, Koerber and Medler (1958) sampled a large array of
sites with latitudes overlapping those of MWC; the three most common
nest-provisioning species at their sites were also found at ours
(Table 3). However, although the ratio of bee to wasp nests remains low
as one moves farther south along the eastern seaboard, the list of the
most common trap-nesters diverges increasingly from that for MWC. At
Plummer’s Island, MD, T. lactitarse was also the most common wasp,
but T. clavatum (absent from MWC) ranked third overall, and O. lignaria
(rare at MWC) was the most common bee (Krombein 1967). In North
Carolina, T. collinum replaced T. lactitarse as the most common wasp. In
disturbed habitats in Georgia, four of the five most common species (the
bee M. frigida, and the wasps E. megaera, S. plenoculoides, and A. campestris)
occupied 67% of the nests, but were absent from our sites (Jenkins
and Matthews 2004). Among the three most common species at MWC, two
were absent (A. antilope, H. annulatus) and one (T. lactitarse) was rare
(less than 2% of nests) in Georgia. The absence of such genera as Hylaeus from
the Georgia data (even though it has been collected in that state; Mitchell
1962) is not surprising because the researchers were not conducting a
broad survey, so they did not provide tunnels of less than 6 mm diameter.
Only I. mexicana was common at both MWC and in Georgia; I. mexicana
is endemic to much of North America, and has even spread recently to
southern Europe (Pagliano et al. 2001) and the Midway Atoll in the central
Pacific (Nishida and Beardsley 2002).
The most distant comparison to be made with our survey is that with
Krombein’s (1967) study from Florida. There, although E. foraminatus was
the most common trap-nester by far, the Florida population is E. foraminatus
apopkensis (Robertson), whereas the NY wasps are E. foraminatus
foraminatus (Saussure). Of the two species ranked next in Florida, Pachodyneris
erynnis is of a genus unrepresented at MWC (or any of the other
sites listed in Table 3), and M. quadridens was extremely rare (just two nests
at MWC). Two other wasp genera from Florida, Podium (one species) and
Stenodynerus (four), were also unrepresented at MWC, although two nests
of Stenodynerus pedestris (Saussure) were found at Derby.
Offspring sex ratios
Brood sex ratios of solitary bees and wasps are readily manipulated by
egg-laying females because of their haplodiploid sex determination system,
in which fertilized eggs produce females and unfertilized eggs males. Sex
ratios are often biased towards one sex (most often males) in solitary nestprovisioning
species, and variable among sites and between nests within
468 Northeastern Naturalist Vol. 17, No. 3
sites (Krombein 1967). The sex of the offspring in any given brood cell is
related to multiple factors, including (among others) 1) its position within a
nest (because males emerge before females, they must be in the outermost
cells), 2) the amount of food that can be provided (females are larger, so
require more food), and 3) the diameter of the nest (the largest females may
not fit within the smallest tunnels); see O’Neill (2001) for review of factors
influencing the evolution of sex ratios in solitary wasps.
For the common species we observed at MWC, emergence sex ratios are
typically either male-biased or unbiased, but they often vary among studies.
Male-biased sex ratios were found for A. antilope at Derby, NY (χ2 = 27.4,
P < 0.001; Krombein 1967); E. foraminatus at Derby, NY (χ2 = 14.5, P <
0.001; Krombein 1967); S. canadensis and S. cristatus in Ontario (Longair
1981); and T. lactitarse at Derby (χ2 = 18.8, P < 0.001; Krombein 1967) and
in Wisconsin (χ2 = 11.6, P < 0.001; Medler 1967). Sex ratios statistically indistinguishable
from unity (at α = 0.05) have been reported for H. annulatus
in Ontario (χ2 = 3.6, d.f. = 1, P = 0.06; Fye 1965), A. antilope in Ontario
(Longair 1981), E. foraminatus in Ontario (Longair 1981), S. canadensis
(χ2 = 0.2, P = 0.65) and S. cristatus (χ2 = 0.7, P = 0.40) at Derby, NY
(Krombein 1967), I mexicana at MWC in a different study conducted from
2004–2005 (χ2 = 2.2, P = 0.14; O’Neill and O’Neill 2009), and I. mexicana
in Montana (χ2 = 0.7, P = 0.40; O’Neill and O’Neill 2003); with the exception
of those by Longair (1981), all analyses above are our own, based on the
Thus, some previously reported sex ratios for H. annulatus, A. antilope,
S. cristatus, and E. foraminatus differ from those that we observed. For S. cristatus,
this may be due to the relatively small sample sizes reported. However,
the difference between studies of E. foraminatus at Derby and MWC may
be related to the sizes of tunnels used by wasps at the two sites: all nests at
Derby were within 4.8–6.4-mm tunnels (even though larger diameters were
available) and produced 24% females, whereas 40% of the nests at MWC of
were in 7–9-mm tunnels, where 65% of offspring were females. However,
unbiased sex ratios for H. annulatus were found both in Ontario (Fye 1965),
where females nested in 6.4–8.0-mm tunnels, and at MWC, where most nests
were within 3–4-mm tubes. Comparisons of sex ratios among populations are
complicated by the fact that sex ratios may vary not only with the particular
frequency distribution of nest sizes made available, but with the length of the
tunnels provided, the generation sampled (in bivoltine species), and the quality
and quantity of resources available to provisioning females (Danks 1983,
Longair 1981, O’Neill 2001).
Our observations on natural enemies emerging from nests confirm
many previous host records (Bohart and Kimsey 1982, Krombein 1967,
Krombein et al. 1979). The most common group of natural enemies were
cuckoo wasps (Chrysididae), with 50 individuals of 7 species emerging
2010 K.M. O’Neill and J.F O’Neill 469
from 39 nests. No host offspring emerged from 24 of the 39 nests, so we
cannot always draw definite conclusions about host associations. Particularly
striking is the fact that Chrysis coerulans never emerged from nests
that also produced host offspring; however, all of those nests contained
mud plugs and partitions typical of eumenine nests, and all previous host
records for C. coerulans are eumenines (Bohart and Kimsey 1982). For
Chrysis nitidula, the case is stronger that its hosts were A. antilope, because
that wasp emerged from seven of the nests that also produced C. nitidula;
in four nests, C. nitidula attacked multiple (2–5) cells. The second most
common group of natural enemies were flies of the genus Amobia, which
are well-known brood parasites of cavity-nesting apoid wasps (O’Neill
2001). Krombein (1967) and Taki et al. (2008a) also found C. coerulans,
C. nitidula, and Amobia to be the most common natural enemies in trap
nests. Natural enemies emerged from just 7 of 85 nests that produced bees,
but these represented a diverse array of brood parasites (Meloidae, Megachilidae,
and Saygidae) and parasitoids (Leucospidae).
Trap nests placed at the MWC from 2001–2007 attracted a diverse array
of solitary nest-provisioning species and their natural enemies. Overall,
wasps using the trap nests are known to take a diverse set of prey, including
spiders (Trypoxylon, Auplopus, Dipogon), crickets and katydids (Isodontia),
aphids (Passaloecus), moth caterpillars (Ancistrocerus, Euodynerus,
Monobia), chrysomelid beetle larvae that feed externally on leaves (Symmorphus
albomarginatus, S. cristatus), and leaf-mining caterpillars and
beetle larvae (S. canadensis) (Krombein 1967, O’Neill 2001). Based on
published flower-visitation records and pollen records, bees of the genera
emerging from the trap nests may all be polylectic (Krombein et al. 1979,
Matteson et al. 2008).
The composition of the assemblage of nest-provisioning species was
generally similar to those from studies done at similar latitudes, but differed
increasingly from those documented in other surveys as one moves further
south along the east coast of North America. Some of the variation between
studies may be due to differences in methodology (e.g., sizes of nests made
available and sampling intensity). Thus, comparisons among existing studies
cannot replace those that could be made with simultaneous surveys
undertaken along a latitudinal gradient using standardized trap-nesting and
site-selection methods, but they do suggest that such studies would likely
reveal clear geographic trends in the composition of cavity-nester species assemblages.
They could also provide a basis for determining the future effects
of climate change and habitat disturbance or restoration on the distribution,
abundance, and diversity of trap nesters and their natural enemies. Overall,
it is clear from this and previous studies that the use of multiple trap-nest
diameters is important if one’s goal is a full assessment of the species composition
of an assemblage and the brood sex ratios of a population.
470 Northeastern Naturalist Vol. 17, No. 3
We thank Tracy Gingrich of the Montezuma National Wildlife Refuge and Dave
O’Dell of the Northern Montezuma Wildlife Management Area for help in obtaining
research permits and locating research sites. Susan Stubbs provided access to her
property near the Montezuma National Wildlife Refuge. The following provided
help in identifying specific taxa: Richard S. Miller (Heriades, Hylaeus, Leucospa,
Perilampus, Sapyga), James Pitts (Pompilidae), and Bryan Danforth (Osmia). Jessica
Fultz, Richard Miller, Megan O’Neill, Ruth O’Neill, and April Pearce assisted with
field work and monitoring insect emergence. James Liebherr and Richard Hoebeke
gave us access to the Cornell University Insect Collection.
Barthell, J.F., G.W. Frankie, and R.W. Thorp. 1998. Invader effects in a community
of cavity-nesting megachilid bees (Hymenoptera: Megachilidae). Environmental
Bohart, R.M., and L.S. Kimsey. 1982. A synopsis of the Chrysididae in America
north of Mexico. Memoirs of the American Entomological Institute 33:1–266.
Bohart, R.M., and A.S. Menke. 1963. A reclassification of the Sphecinae with a revision
of the Nearctic species of the tribes Sceliphronini and Sphecini. University
of California Publications in Entomology 30:91–182.
Bohart, R.M., and A.S. Menke. 1976. Sphecid Wasps of the World: A Generic Revision.
University of California Press, Berkeley, CA. 695 pp.
Buck, M., S.A. Marshall, and D.K.B. Cheung. 2008. Identification atlas of the
Vespidae (Hymenoptera, Aculeata) of the northeastern Nearctic region. Canadian
Journal of Arthropod Identification 7. Available online at http://ejournals.library.
ualberta.ca/index.php/CJAI/issue/view/106. Accessed April 2009.
Coville, R.E. 1982. Wasps of the genus Trypoxylon Subgenus Trypargilum in North
America (Hymenoptera: Sphecidae). University of California Publications in
Danks, H.V. 1983. Difference between generations in the sex ratio of aculeate Hymenoptera.
Droege, S. 2009. Bee genera of Eastern North America. Available online at http://
www. discoverlife.org/mp/20q?guide=Bee_genera. Accessed April 2009.
Fye, R.E. 1965. Biology of Apoidea taken in trap nests in northwestern Ontario (Hymenoptera).
The Canadian Entomologist 97:863.877.
Fye, R.E. 1972. The effect of forest disturbances on populations of wasps and bees
in northwestern Ontario (Hymenoptera: Aculeata). Canadian Entomologist
Gathmann, A., H.J. Greiler, and T. Tscharntke. 1994. Trap-nesting bees and wasps
colonizing set-aside fields: Succession and body size, management by cutting and
sowing. Oecologia 98:8–14.
Gayubo, S.F., J.A. González, J.D. Asis, and J. Tormos. 2005. Conservation of
European environments: The Spheciformes wasps as biodiversity indicators
(Hymenoptera: Apoidea: Ampulicidae, Sphecidae, and Crabronidae). Journal of
Natural History 39:2705–2714.
Jenkins, D.A., and R.W. Matthews. 2004. Cavity-nesting Hymenoptera in disturbed
habitats of Georgia and South Carolina: Nest architecture and seasonal occurrence.
Proceedings of the Entomological Society of Washington77:203–214.
2010 K.M. O’Neill and J.F O’Neill 471
Jensen, P.J., K.M. O’Neill, and J.F. O’Neill. 2007. Biological Notes on Heriades
carinatus Cresson, Heriades leavitti Crawford, and Heriades variolosus (Cresson)
(Hymenoptera: Megachilidae). Proceedings of the Entomological Society of
Koerber, T.W., and J.T. Medler. 1958. A trap-nest survey of solitary bees and wasps
in Wisconsin, with biological notes. Wisconsin Academy of Sciences, Arts, and
Krombein, K.V. 1967. Trap-nesting Wasps and Bees: Life Histories and Nest Associates.
Smithsonian Press, Washington, DC. 570 pp.
Krombein, K.V., P.D. Hurd, D.R. Smith, and B.D. Burks. 1979. Catalog of Hymenoptera
in America North of Mexico. Volume 2: Apocrita (Aculeata). Smithsonian
Institution Press, Washington, DC 2209 pp.
Longair, R.L. 1981. Sex-ratio variations in xylophilous aculeate Hymenoptera. Evolution
Matteson, K.C., J.S. Ascher, and G.A. Langellotto. 2008. Bee richness and abundance
in New York City urban gardens. Annals of the Entomological Society of
Matthews, R.W. 1965. The biology of Heriades carinata Cresson. Contributions of
the American Entomological Institute 1:1–33.
Medler, J.T. 1967. Biology of Trypoxylon in trap nests in Wisconsin. American
Midland Naturalist 78:344-358.
Michener, C.D., R.J. McGinley, and B.N. Danforth. 1994. The Bee Genera of North
and Central America (Hymenoptera: Apoidea). Smithsonian Books, Washington,
Mitchell, T. 1962. Bees of the Eastern United States, Volume II. North Carolina Agricultural
Experiment Station Technical Bulletin 152. Raleigh, NC. 557 pp.
Nishida, G.M., and J.W. Beardsley. 2002. A review of the insects and related arthropods
of Midway Atoll. Records of the Hawaii Biological Survey for 2000, Bishop
Museum Occasional Papers 68:25–69.
O’Neill, K.M. 2001. Solitary Wasps: Natural History and Behavior. Cornell University
Press, Ithaca, NY. 406 pp.
O’Neill, K.M., and R.P. O’Neill. 2003. Sex allocation, nest structure, and prey of
Isodontia mexicana (Saussure) (Hymenoptera: Sphecidae). Journal of the Kansas
Entomological Society 76:447–454.
O’Neill, K.M., and J.F. O’Neill. 2009. Prey, nest associates, and sex ratios of Isodontia
mexicana (Saussure) (Hymenoptera: Sphecidae) from two sites in New York
State. Entomologica Americana 115:90–94.
O’Neill, K.M., J.F. O’Neill, and R.P. O’Neill. 2007. Sublethal effects of brood parasitism
in Isodontia mexicana (Hymenoptera: Sphecidae). Ecological Entomology
Pagliano G., P.L. Scaramozzino, and F. Strumia. 2001. Introduction and spread of
four Aculeate Hymenoptera in Italy, Sardinia, and Corsica. Pp. 290–295, In A.D.
Austin and M. Dowton (Eds.). Hymenoptera. Evolution, Biodiversity, and Biological
Control. CSIRO Publishing, Collingwood, Australia. 512 pp.
Sandhouse, G.A. 1940. A review of the Nearctic wasps of the genus Trypoxylon (Hymenoptera,
Sphecidae). American Midland Naturalist 24:133–176.
Steffan-Dewenter, I. 2003. Importance of habitat area and landscape context for species
richness of bees and wasps in fragmented orchard meadows. Conservation
472 Northeastern Naturalist Vol. 17, No. 3
Taki, H., P.G. Kevan, B.F. Blandina, F.O. Silva, and M. Buck. 2008a. Artificial covering
on trap nests improves the colonization of trap-nesting wasps. Journal of
Applied Entomology 132:225–229.
Taki, H., B.F. Viana, P.G. Kevan, F.O. Silva and M. Buck. 2008b. Does forest loss
affect the communities of trap-nesting wasps (Hymenoptera: Aculeata) in forests?
Landscape vs. local habitat conditions. Journal of Insect Conservation
Tscharnke, T., A. Gathman, and I. Steffan-Dewenter. 2003. Bioindication using
trap-nesting bees and wasps and their natural enemies: Community structure and
interactions. Journal of Applied Ecology 35:708–719.
Townes, H. 1957. Nearctic wasps of the subfamilies Pepsinae and Ceropalinae. Bulletin
of the United States Natural History Museum 209:1–161.
Vincent, D.L. 1979. A revision of the genus Passaloecus (Hymenoptera: Sphecidae)
in America north of Mexico. Wasmann Journal of Biology 36:127–198.