‘:_3_=E__:E___:__:::_E_Ea:

 

VERSITY LIBRARIES

lllllalll l H

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ll\llllllllllllllll

 

 

 

 

N

This is to certify that the

--J

‘ thesis entitled

Specialization, Foraging Efficiency
and Nutritional Relationships in a Solitary Bee,
Heriades carinata Cresson (Hymenoptera: Megachilidae)

 

 

. n
,.-
a-.. '

1‘ presented by

Jonathon Michael DeNike

has been accepted towards fulfillment
of the requirements for

MISS—degree in Entomology

flaw i/uC/L/t

Major professor

lo , -
Date [lg/j”! //‘.7. /‘( é/j

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

LIERAFW
Michigan Sim ”‘
Unlversiiv

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

i13‘1lSEIl‘

filjm
_=__l
‘_—l

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

fl

 

 

 

 

 

 

 

 

 

l_—7 3

MSU Is An Affirmative Action/Equal Opportunity Institution
cMmMJ

 

 

 

SPECIALIZATION, FORAGING EFFICIENCY, AND
NUTRITIONAL RELATIONSHIPS IN A

SOLITARY BEE. HEBJAQES QABJNAIA
CRESSON (HYMENOPTERA: MEGACHIUDAE)
By

Jonathon Michael DeNike

ATHESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Entomology

1991

ABSTRACT

SPECIALIZATION, FORAGING EFFICIENCY AND NUTRITIONAL
RELATIONSHIPS IN A SOLITARY BEE, HERIADES CARINATA
CRESSON (HYMENOPTERA: MEGACHILIDAE)

By

Jonathon Michael DeNike

Pollen foraging preferences of 3mm Cresson
were investigated by trap nesting in field sites near Pellston,

Michigan during the summers of 1988-1990. During the peak of
the nesting season, females of him provisioned
exclusively with pollen from manning L. (Anacardiaceae).
The flight season of 11.93me was synchronized with the
flowering of anus, and nesting activity by ligaflnata was only
found in study sites containing Elms. ligaflnam is at least
regionally specialized on films, and may be oligolectic on this
plant genus. .

Timed flights of ligaflnata females foraging for BILLS
pollen were not different from those foraging from other plants.
Survival by offspring of Jim to adulthood was higher on

provisions consisting of Elms Pollen than on other pollens.

ACKNOWLEDGEMENTS

I would like to thank Dr. Karen Strickler for her patience
and advice during the production of this research project. I would
also like to thank Dr. James Miller and Dr. Edward Walker for their
advice and encouragement; Dr. Roland Fischer for advice and
assistance in the identification of bees; Virginia Scott for the
same; Dr. James Teeri of the University of Michigan Biological
Station for three summers of financial assistance on the field
portion of this project; Margi Coggins for support, encouragement
and inspiration; and finally Dr. David Cowan of Western Michigan
University for introducing me to the study of entomology and

advice on the design and execution of this project.

TABLE OF CONTENTS

INTRODUCTION 1
Biology of damages carinata and anus naming 1
Specialization and Oligolecty 2
Nectar Foraging 5
Evolution of Oligolecty 6

Location and Assessment of Host Plants and Relationship

to Offspring Investment 8
Pollen Harvesting Efficiency 11
Diurnal Synchrony 12
Seasonal Synchrony 13
Morphological Adaptations to Host Plant Pollens 14
Nutritional Properties of Host Plant Pollens 15
Objectives 19

METHODS 21
Field Sites 21
Trap Nests 22
Vegetation Sampling 23
Sampling Chronology 24
Pollen Provision Sampling 24

Foraging Flight Timings 26

RESULTS

Nesting Activity and Pollen Provisioning
Pollen Provisioning by Other Bee Species
Foraging Flight Timings

Offspring Survival to Adulthood

DISCUSSION

Seasonal Synchrony and Nest Provisioning
Nectar Foraging

Nesting Locality and Host Plant Abundance
Pollen Foraging by Other Bee Species
Parental Investment

Foraging Efficiency

Offspring Survival to Adulthood

SUMMARY
LITERATURE CITED
APPENDIX: RECORD OF DEPOSITION OF VOUCHER SPECIMENS

27
27
31
36
36
38
38
40
4O
41
42
43

46
47
57

LIST OF FIGURES

Figure 1a. Weekly floral abundances for the moist site
during 25 June - 12 August, 1989

Figure 1b. Percent Rhys pollen in nests of mm
from the moist site from 25 June - 12 August, 1989

Figure 2a. Weekly floral abundances for the moist site
during 1 July - 18 August, 1990

Figure 2b. Percent BM pollen in nests of Heriadesgarmata
from the moist site from 1 July - 18 August, 1990

Figure 3a. Number of cells in relation to the proportion of
films pollen in nests of mm from the
moist site for 1989

Figure 3b. Number of cells in relation to the proportion of
Emu pollen in nests of mm from the
moist site for 1990

Figure 4a. Number of nests in relation to the number of cells
per nest of mm from the moist site from
1989

Figure 4b. Number of nests in relation to the number of cells

per nest of 5mm from the moist site from
1990

28

28

30

30

32

32

33

33

Figure 5a. Weekly average numbers of cells per nest of
Heriadesgarmata from the moist site from 25 June -
12 August, 1989 34
Figure 5b. Weekly average numbers of cells per nest of

3mm from the moist site from 1 July -
18 August, 1990 34

INTRODUCTION

Biology of mmmmand munition-
mm Cresson is a small megachilid occurring
throughout much of North America (Mitchell 1962, Matthews
1965). Two other species of the genus occur in eastern North
America (Mitchell 1962). The biology of ligaflnata was studied
by Matthews (1965). Females of this bee nest in pre-existing
holes of approximately 3-5 mm inside diameter, usually in hollow
twigs, and will readily accept trap nests of the appropriate size
as nesting sites, making it a convenient animal for study. After a
female has chosen a nest hole, she constructs a linear series of
cells, divided by partitions constructed of plant resins. Within
each cell, the female places a ball of pollen mixed with nectar,
obtained from multiple foraging flights. Once the appropriate
amount of pollen is in place, an egg is laid on the food mass, the
cell is sealed, and construction and provisioning of the next cell
proceeds. Larvae feed for an average of 30 days before molting to
a prepupa. As this species is univoltine, this stage will
ovenrvinter; pupation occurs in spring. Adult bees emerge in early
summer; mating occurs, and females begin to construct nests
throughout the rest of the summer, while males do not
participate in nesting activities. Females may live up to a month

1

2

or longer, while males live to a maximum age of around 12 days.

Matthews (1965) found that the pollen in his trap nests in
the vicinity of Kellogg Biological Station near Battle Creek,
Michigan during the peak of the nesting season was almost
exclusively from staghorn sumac (Bhustumma L.: Anacardiaceae).
Adult emergence of this bee also corresponded well with the
beginning of the blooming period for staghorn sumac. 3.1mm
grows as a tall shrub or small tree, often in thickets; other
members of the genus exhibit similar growth forms, although
some may be more prostrate or vine-like (Barkley 1937). E,
momma is dioecious; the inflorescences are terminal, usually
several per plant, and densely crowded with hundreds of small
(about 5 mm in diameter) flowers, presenting an obvious target
for pollen- and nectar-seeking insects. As this plant is usually
abundant in those Open field habitats where it occurs, it provides
an extremely important source of nectar and pollen when in
bloom, which is usually 2-3 weeks in a given habitat. This plant
can be considered a "cornucopian species” in the sense of Mosquin
(1971) in that it provides an extremely abundant source of pollen
and nectar and is attractive to several orders of insects,
including a diverse assemblage of Hymenoptera, Diptera,
Lepidoptera and Coleoptera.

Specialization and Oligolecty. Many insects exhibit
specialization in their foraging, in contrast to generalists which
exhibit a broad range of acceptable food choices. Trophic

relationships between specialists and their preferred food

3
sources can be very intricate and dynamic, and can vary
considerably in degree of specificity. Many of these relationships
can impact the overall ecology of the local ecosystems by
regulating populations of their food organisms, either through
predatory/parasitic interactions or by assisting in propagation by
dispersal of seeds or pollination (Prescott and Allen 1986).

Trophic relationships of specialists may be a driving force
in speciation of both insects and host plants (Ehrlich and Raven
1964, Cruden 1972, Cane and Eickwort in press). Many pest
insects are specialists, having both cued in on the characteristics
and overcome the defenses of their host organisms (Fraenkel
1959, Price 1984). Understanding foraging specialization in
insects is thus an important area of study for both basic and
applied research. Solitary bees (Apoidea) are a group of insects
that contains examples of all degrees of foraging specialization,
including opposite extremes within genera. Studying foraging
relationships of solitary bees is not only important in an
evolutionary sense (Cane and Eickwort in press), but also in
agriculture, where these bees are being increasingly used for crop
pollination (Bohart 1972, Torchio 1976, 1981, 1990a, b).

Overall, most solitary bee species (64.1%) exhibit
Oligolecty, or pollen specialization (Hurd 1979, Schemske 1983).
Oligoleges are generally distinguished from polyleges (bees that
do not specialize in pollen foraging) in that the former confine
the bulk (if not all) of their pollen foraging to a single plant

family across their entire range (and in the presence of other

4
pollen sources), whereas polyleges will collect pollen from
plants in more than one family, without demonstrating consistent
preferences across their range (Linsley 1958, Thorp 1979,
Eickwort and Ginsberg 1980). These categories have no sharp
distinction, as the pollen foraging habits of solitary bees form a
continuum from monolecty (exclusive use of a single plant
species for pollen) to broad polylecty (use of many plants in
different families for pollen) (Eickwort and Ginsberg 1980).

This issue is further confused by local specialization of
polyleges on highly abundant sources of pollen (Torchio 1976,
1981, Johnson 1984, Cripps and Rust 1989, Rust 1990). Even in
broadly polylectic species such as the social Apidae, individual
bees often exhibit floral constancy, foraging consistently from a
particular plant species and bypassing others (Grant 1950, Waser
1986). Cane and Payne (1988) base their decision that Habmnoda
W (F.) (Anthophoridae) is oligolectic on blueberries
(lasslnmm spp.) on criteria including visitation rates, floral
handling behaviors and efficiencies, pollen provisioning, and
phenology in relation to the host plants.

Some families contain more specialists than others. In the
Andrenidae, 83% of the species studied are oligolectic, while the
Megachilidae appears to contain the lowest proportion of
oligolectic species at 43% (Hurd 1979, Schemske 1983). The
degree of specialization is also variable: 15.3% of all solitary
bees studied specialize on pollen at the species level; 36.6% show

preferences for plants at the generic level; and 12.2% confine

5
their pollen foraging to a single family of plants (Hurd 1979,
Schemske 1983).

Examples of bees that forage only from a single plant
species and refuse all other sources of pollen include species
such as BMW (Schneck) (Megachilidae) which
refused to forage on all plants other than its preferred host,
Egbiumvulgam (Boraginaceae), in an experimental study

(Strickler 1979), and WW Cockerell (Halictidae)
which did not forage during a year when its host plant, Qenmnfla

Dallida (Onagraceae), did not flower (Bohart and Youssef 1976).
Other bees considered oligolectic will gather pollen from other
plant species only when their preferred host is not in bloom, as
does Andrenenflhmnfl Robertson (Andrenidae); this bee gathers
pollen from gums (Fagaceae) as well as other plants after its
preferred host, Eananum (Liliaceae), wilts (Michener and
Rettenmeyer 1956). Cruden (1972) discusses three species of
Alumna oligolectic on W (Hydrophyllaceae) that will, in
times of low host plant abundance, forage on other genera of the
Hydrophyllaceae or members of other plant families . Thorp
(1969) cites other examples and advances the idea that collecting
pollen from non-host plants is adaptive for survival of oligoleges
during periods of low host plant availability.

Nectar Foraging. Many, if not most, oligolectic bees will
also forage on plants other than their host for nectar (Linsley
1958, Eickwort and Ginsberg 1980). Although nectar is more

complex in composition than merely an aqueous solution of sugars

6
(Baker and Baker 1975), this composition is not as variable
between plant species as pollen (Cane and Eickwort in press) and
thus probably does not have as important an effect on the
nutrition of the bee. Nectar is primarily used as adult flight fuel
(and also in larval provisions to a varying extent), whereas pollen
is the primary protein source for growing larvae.

In many bees, assumptions of polylecty are based upon
flower records (Krombein et al 1979) which may in fact be nectar
foraging (Cane and Eickwort in press). Pollen preferences can
only be determined with certainty by examining pollen carried by
females, provisioned in nests, or in larval feces (Johnson 1984).
Foraging for nectar on non-host plants by oligoleges may lead to
”contamination" of otherwise pure host pollen loads with small
percentages of non-host pollen (Cripps and Rust 1989, Rust 1990,
Cane and Eickwort in press).

Evolution of Oligolecty. The evolution of oligolecty in
solitary bees has been a subject of debate for a long time.
Robertson (1914, 1929a) proposed that oligolecty evolved as an
adaptation to reduce competition by partitioning the pollen
resources, believed to be limiting, amongst many non-overlapping
specialists. This view was questioned by Lovell (1914, 1918),
who speculated that pollen was not limiting and that oligolecty
allowed for greater efficiency in foraging by focusing behaviors
on particular plant morphologies. This in turn would enable the
production of a greater number of offspring. Thorp (1979)

discussed the possibility that each hypothesis could apply to

7
certain situations and are not necessarily mutually exclusive. In
some habitats, such as arid regions with seasonally limited
rainfall, pollen resources may indeed be limited and oligolectic 1
species that are more efficient harvesters may have an advantage
(Linsley and Cazier 1963, Thorp 1979, Parker 1981). Michener
(1954) believes that oligolecty may have been derived through a
combination of competitive pressure and preadaptive morphology
or behavior that made particular polylectic bee species better
harvesters on some plants than on others, thus driving these
polyleges to specialize on plants to which they were most suited.
This in fact combines the theories of Robertson and Lovell in
explaining the evolution of oligolecty.

One could speculate that solitary bees had their
evolutionary beginnings when pollen resources were limiting, and
thus competition was a driving force in the evolution of
oligolecty. Oligolecty is much more common amongst bees that
live in arid regions, where pollen resources are likely to be '
limited (Linsley and MacSwain 1958, Linsley and Cazier 1963).
However, in other situations pollen does not appear to be a
limited resource (Cane and Eickwort in press). Some flowers are
visited by a large number of bee species, both oligolectic and
polylectic, in several families. Hangman; is visited by a total of
284 bee species, of which 39 species (in 12 genera) are
oligolectic on the genus and 92 more are oligoleges of the family
Asteraceae (Hurd et al 1980). Even in a single locality, such as a

commercial plantation (which does provide more than ample

8
resources), 7 species (all in different genera) of oligoleges visit
Heuamnusamuus (Parker 1981). MW (Fabaceae) is
visited by 22 oligoleges (in 9 genera) and 68 polyleges (Hurd and
Linsley 1975). W is visited by 20 different lineages of
bees, and 11 lineages each visit Osmium, Saljx and Sphaeralm
(Cane and Eickwort in press). Cripps and Rust (1989a, b)
concluded that a community of Qsmia (Megachilidae) species
partitioned their floral resources and overlapped on the most
abundant flowers, but found no evidence of competition for pollen.
Raw (1974) found that each of three Qsmja species nesting in the
same locality did collect pollen from different plant species.
Thus, pollen may be a limited resource in some situations but not
in others.

Location and Assessment of Host Plants and
Relationship to Offspring Investment. Host-plant finding
by oligolectic bees may rely on either visual or olfactory cues, or
a combination of stimuli, although literature on sensory cues in
solitary bees is limited. Learning may be involved (Linsely et al
1963, Thorp 1979) but evidence is scarce. Some authors have
speculated that bees may imprint on the chemical composition of
the pollen in their nest cells and thus use the same chemical cues
to locate their appropriate host plants (Linsely 1958, 1978, Thorp
1979). Some oligoleges investigate the unopened flower buds of
their host plants (Linsely 1958, Linsely et al 1963, Eickwort

1973), indicating that olfactory cues are important. Dobson

9
(1984) demonstrated that pollen coat lipids from Gflndglja alone
were as attractive as pollen and bagged flowers to a local
specialist of this plant, WWW
(CoHefldae)

Visual cues also appear to be important in some cases.
WM Say, a broadly polylectic species, prefers blue-
violet flowers and takes little pollen from flowers of other
colors, even when abundant (Michener 1953). Megagmlemjungata
Fabricius (Megachilidae), which is polylectic, prefers alfalfa
(Medicine) with normal purple flowers over yellow-flowered
variants (Golpen and Brandt 1975). Ultraviolet patterns also play
an important role, as speculated by Linsley (1958) about the
pollination of the green, odorless orchid WELLNE- This
apparently unattractive flower was nonetheless visited by four
species of solitary bees and one bumble bee species, while being
ignored by honey bees, in a garden filled with brightly colored and
aromatic violets (Leclercq 1945). Alteration of normal
ultraviolet patterns deters foraging behavior on normally
preferred plants (Jones and Buchmann 1974).

Oligolectic bees assess their pollen resources and respond
to the changing availability of host plant pollen. mm
W emerged but did not forage during a year in which its
host plant (W nailing) did not bloom, but instead the
adults returned to their natal burrows after filling up on nectar

(from other flowers) and overwintered a second year (Bohart and

Youseff 1976). Qammsismmms (Cockerell) (Andrenidae), an

10

oligolege of mm (Solanaceae), provisions its nest at
the astonishing rate of up to six cells per day (one to three is
typical for most solitary bees). Early in the day, when pollen is
abundant, QpeLsimist constructs cells for female offspring;
towards the end of the day, when the pollen of its host plant
becomes depleted, gimme, begins to produce cells for male
offspring, which require 33% less pollen than female cells '
(Danforth 1990). Male solitary bees are usually smaller than
females and require less food (Mitchell 1962, Krombein et al
1979). Production of more males late in the season as either
resources dwindle or female pollen harvesting efficiency declines
is considered to be a strategy for maximization of reproductive
success (Torchio and Tepedino 1980, Sugiura and Maeta 1989).

Torchio and Tepedino (1980) and Frohlich and Tepedino
(1986) speculated that increasingly male-biased sex ratios late
in the nesting season of Qsmialjgnariapmpjngua Cresson
(Megachilidae) were the result of declining resource availability,
and Torchio (1985) confirmed this. By experimentally increasing
provision masses, Plateaux-Ouenu (1983) caused females of
WW (Halictidae) to produce more female offspring.
Conversely, Torchio and Tepedino (1980) also speculated that
aging of females, and subsequent wear and tear, reduce the
abilities of females to gather pollen as effectively as earlier in
the season. Tepedino and Torchio (1982) proved this, and Sugiura
and Maeta (1989) demonstrated this with photographs comparing
young and old females of Qsmjaggmjims (Radoszkowski)

1 1
(Megachilidae), showing that age does indeed reduce the density
of scopal hairs and therefore pollen carrying capacity.

However, Strickler (1982) found that, even though Haalms
anmgggpgjgas handled flowers of its host Echimnxulgara more
slowly as females aged (and possibly due to declining quantities
of pollen), investment per individual offspring did not decrease
correspondingly. This bee is unusual in that there is little sexual
dimorphism in size (Strickler 1979). Males of H‘amhmnaigaa
are territorial (Eickwort 1973), and Strickler (1982) speculates
that smaller males would be at a disadvantage in territorial
competition. Small females would also be less efficient at
gathering pollen from the host plant (Strickler 1979, 1982).
Cowan (1981) found that larger females of two eumenid wasps,
Anclstmceulsadianatus (Saussure) and Eundmusmlninatus
(Saussure), had higher reproductive success than smaller females
by collecting more food and provisioning more offspring. This
may also be true for some solitary bees, but Johnson (1990) found
that larger females of the small carpenter bee Qaratinacalgarata
Robertson (Anthophoridae) did not produce either more total
offspring or more female offspring, and speculates that in this
species larger female offspring have a greater chance of
surviving through the winter. Tepedino and Torchio (1982) also
found no correlation between female size and fecundity in the
solitary bee Qsmjafianariampjnaua Cresson (Megachilidae).

Pollen Harvesting Efficiency. Oligolectic bees have

variously adapted to the characteristics of their host plants

12
(reviewed by Thorp, 1979). Bees that specialize may have become
behaviorally adapted to the particular flower morphology of their
host plants so that they can harvest pollen more efficiently than
non-specialist, or polylectic, bees visiting the same plants.
flammaanmmmjgas was shown to be more efficient at
collecting pollen from its host plant, Eahjumlulgara, than four
polylectic species (Strickler 1979). Hahmgfialaharjgaa
(Anthophoridae), an oligolege of blueberries (mum), forages
more rapidly on its host plants than competing polylectic Banana
species (Cane and Payne 1988). This would appear to convey an
advantage to a specialist in localities where abundance of the
preferred host plant is high. Most solitary bees are believed to
limit their foraging ranges to within a few hundred meters of
their nests (Eickwort and Ginsberg 1980, Rust 1980, Neff et al
1982, Cane and Eickwort in press). One would therefore expect to
find oligolectic bees only in habitats containing sufficient
quantities of their host plants to sustain a population. Superior
efficiency of oligoleges on their host plants when compared to
polyleges supports Lovell's (1914, 1918) hypothesis about the
advantages of oligolecty, but evidence is still scanty.

Diurnal Synchrony. Many oligolectic bees show a high
degree of diurnal synchrony of activity with the anthesis
(shedding of pollen) of their host plants. This is especially true
in desert habitats, where the incidence of oligolecty is high and
floral resources are limited both temporally and in abundance
(Linsley 1958, 1978; Linsley and Cazier 1963). Several different

13
temporal foraging periods can be distinguished in desert habitats,
especially amongst bee species that forage from plants
presenting their pollen at specific times and for limited periods
such as members of the Solanaceae and Onagraceae (Thorp 1979).
Oligoleges often arrive at their host plants and extract pollen and
nectar before competing polyleges arrive (Linsley and Cazier
1963, Thorp 1979, Parker 1981). Learning may be involved with
diurnal synchrony (Thorp 1979) but many oligoleges exhibit
morphological adaptations to their temporal foraging periods,
such as crepuscular or nocturnal bees with large ocelli (Kerfoot
1967)

Seasonal Synchrony. Other oligoleges show seasonal
synchrony of emergence with the blooming phenology of their host
plants. Robertson (1929b) cites numerous examples of
oligolectic species with flight periods that are shorter than, and
overlap perfectly, the blooming period of their host plants. Many
oligoleges, however, are not perfectly synchronous with host
plant flowering. Annmaammgmfl females emerge before
Emnmnium begins blooming and continue to forage after the
plants are no longer in flower (Michener and Rettenmeyer 1956).
Emingadas (Megachilidae) species in California are not
synchronous in flight period with peak blooming of their host
plants in the genus Magma (Hurd and Michener 1955).
MW begins flight two weeks earlier than first
bloom of its host plant Sphaaralcaa (Eickwort et al 1977) while

Eerdilamaculigaramaculipamis Graenicher (Andrenidae) begins

1 4
foraging too late to catch peak bloom of its host Salixnjgra
(Michener and Ordway 1963). In many of these cases, it is
possible that mate finding and nest construction delay the onset
of actual foraging for nest provisions (Cane and Eickwort in
press).

The environmental cues used by oligolectic bees to achieve
synchrony of flight with the flowering of their host plants are
not well known. As solitary bee larvae are somewhat insulated
from the outside environment by their nest enclosures, the
stimuli used by host plants to flower and bees to pupate or
emerge may not be perceived in the same manner. Moisture may
be an important stimulus for emergence. A normally vernal,
univoltine desert oligolege, WW Crawford,
emerged as a second generation in October to forage on an unique
autumnal flowering of its host Elma (Asteraceae) after
unseasonally heavy rainfall (Hurd 1957). Other oligoleges did not
emerge during drought years when their host plants did not
flower (Thorp 1979). Temperature may also play an important
role; it is known to be the primary factor in the emergence of
Magaghjjammndata (Thorp 1979). Cane and Eickwort (in press)
suggest that loose synchrony with host-plant flowering is
adaptive in bracketing blooming which may be highly variable
between years.

Morphological Adaptations to Host Plant Pollens.
Some oligolectic bees have specialized structures for collecting

and transporting the distinctive pollen grains of their host

15

plants. Bees that specialize on members of the Malvaceae and
Cactaceae, which produce relatively large pollen grains, have
Ioosley plumose scopal setae to accommodate the large grains
(Linsley 1958). Some oligoleges of Solarium, species of which
produce relatively small pollen grains, have much more densely
plumose scopal hairs to securely hold the small pollen, while
others mix the pollen with regurgitated nectar to hold it in place
(Linsely and Cazier 1963, Roberts and Vallespir 1978). Species
of Xangglgasa, a genus oligolectic on species of manta, have
stout, unbranched, and grooved scopal hairs to hold the large, oily
pollen grains of their host plants (Roberts and Vallespir 1978).
Perhaps the best examples of morphological adaptation to host
plant pollen morphology involve oligoleges of the genera
Qangthara, Qamjsagma, and. Clarkja in the Onagraceae. Several
genera of bees, from different families, that specialize on these
plants all have unbranched scopal setae with hooked tips that
function well in holding the pollen grains of their hosts, which
are large and connected together by sticky threads (Roberts and
Vallespir 1978, Thorp 1979). Finally, there are examples of
oligolectic bees that have morphological adaptations of their
proboscis for extracting nectar from their host plants (Linsely
1958, Cruden 1972), although others are apparently not well
adapted to collect their host plant's nectar (Bohart and Youssef
1976).

Nutritional Properties of Host Plant Pollens.

Nutritional relationships between oligolectic bees and their host

16
plants are not well studied, but a few cases involving oligoleges
as well as polyleges deserve discussion. Bohart and Youseff
(1976) were able to rear larvae of WW. an
oligolege of Qanmnara, to adulthood on Madigagg, (Fabaceae)
pollen. However, larval growth was slower (8 days versus 6 on
W), and both prepupal (3 days versus 2) and pupal periods
(13 days versus 8) were longer on NEW. Andranaaatragafl
Viereck and Cockerell is oligolectic on Zigadanua (Liliaceae), a
plant with pollen that has proven toxic to honey bees (Ania
mailing L.) and possibly to other polylectic species as well
(Tepedino 1981).

Qsmialignafla is narrowly polylectic (foraging from a few
plant families) but its larvae do not develop equally well on
different pollens. Optimal development (maximum weight and
shortest larval period) occurs on Wm (Hydrophyllaceae),
growth is slower on BESSIE (Brassicaceae), Eiaum (Fabaceae)
and mum. and larvae die when fed on Gflndalja, 11a,
Sambatua, and Iaraxagum (Levin and Haydak 1957). This bee
also appears to do well on almond (Emma: Rosaceae) pollen, as
nests with provisions averaging 96% almond pollen yielded an
average survival rate of greater than 90% (Torchio 1981). Rust
(1990) reported a survival rate of only 47% for Qfignaria
offspring feeding on Iaraxagum pollen.

The alfalfa leafcutter bee Magaghflamjimdaja is also

considered to be narrowly polylectic, but prefers alfalfa
(M9199. saliva) pollen when present (Stephen and Torchio

17

1961). This bee is able to reproduce successfully on pollen and
nectar from other plants, such as carrot (Qauauacamta), but
when released in carrot fields in a pollination experiment the
bees flew to relatively distant (500 m) alfalfa fields for pollen
(Tepedino 1983). Larvae of this bee exhibited maximum growth
when fed alfalfa pollen, grew slightly less when fed on pollen of
the related legumes Maljlmlla and Galaga, and died when fed
pollen exclusively from the composite Haulapanaua (Guirgus and
Brindley 1974). Larval growth is faster on pollen from Madjgagg
than from Malibu: (Tasei and Masure 1978). The larvae of M.
Mandala are extremely efficient (in terms of energy and
nitrogen utilization) at using alfalfa pollen in nutrition
(Wightman and Rogers 1978).

It has long been known that the chemical composition of
' pollen varies between plant species and that these differences
affect nutritional value to bees (Todd and Bretherick 1942, Vivino
and Palmer 1944, Haydak 1970). The ability of insects to
evaluate the nutritional quality of their food sources was not
given much consideration (Fraenkel 1959) until House (1967,
1970) demonstrated that insects can choose the most nutritious
diet when given a choice among diets of varying nutritional
suitability. Essential nutrients, not merely secondary compounds
(Fraenkel 1959), have been demonstrated to be important
phagostimulants (Thorsteinson 1960, Beck 1965, Davis 1968).
Larval bees cannot chose their own diet, their food being provided

by their mothers. Adult honey bees do show preferences when

18
pollen foraging (Levin and Bohart 1955), and adult honey bees
demonstrate differential longevity on different plant pollens
(Schmidt et al 1987). Perhaps adult bees can detect nutritional
differences in pollen and provision only those pollens most
nutritious for their offspring, although it is unknown if there is
any correlation between larval and adult nutrition.

Conversely, it has been speculated that oligolectic bees are
imprinted on their host plant from the pollen in their nest
provisions (Linsely 1958, 1978). Pollens do indeed contain
components that have been shown to be attractive (Lepage and
Boch 1968, Hopkins et al 1969, 1975) or phagostimulatory
(Dobson 1984, Schmidt 1985) to bees. However, examples of
oligolectic bees that forage from non-host plants during years of
host plant flowering failure and then return to their host plants
the following year (Thorp 1969) do not support host plant

imprinting through pollen (Cane and Eickwort. in press).

 

19
Objectives. The objectives of my study were:
1) To determine if thaLinata specializes on matyahjna or
similar congeneric species for pollen for larval provisions in trap
nests.
2) To evaluate possible mechanisms of specialization with
respect to three hypotheses: a ”cornucopia" hypothesis, behavioral
specialization, and nutritional specialization.
a) "Cornucopia" hypothesis: ligarinatajemales forage on
films because it is the most obvious and abundant source of
pollen, and share similar foraging habits with other bees
that appear to be local specialists, foraging on other
plants in other habitats as well.
D) Behavioral specialization: ligaflnata females are able to
gather more pollen per unit time on films than from other
plants.
0) Nutritional specialization: tLgarjnata larvae survive to
adulthood at a higher rate when feeding on Bhua pollen than
on pollen from other plants.
These hypotheses are not mutually exclusive, as Elma pollen may
be a basically nutritious food for any pollen forager, and the
abundance of films at the study sites may affect the time
necessary to collect sufficient amounts of pollen for
provisioning. Other potential explanations for specialization by
ligarinaia on BULLS. such as morphological specialization and
diurnal synchrony, were not examined in this study. magnum

pollen is rather typical in size and shape, and it is unlikely that

20
any special morphological adaptations would increase

transportation efficiency of this pollen.

METHODS

Field Sites. Four study sites, each with distinctive
vegetational and ecological characteristics, were established on
the property of the University of Michigan Biological Station
(UMBS) near Pellston, Michigan. These sites were chosen because
of their differing plant communities related to soil moisture and
surrounding habitat.

The moist site (MS) was located in a clearing in the extreme
southwest corner of Section 5, Burt Twp., Cheboygan Co; it
consisted of an old field with a moderately diverse flora of
insect-pollinated plants, including staghorn sumac (BMW
L.), surrounded by mixed deciduous/conifer woodland.

The Colonial Point site (CP) was established in an
abandoned apple orchard located southeast of the intersection of
Lathers Road and Indian Point Trail in Colonial Point Forest. This
semi-virgin timber tract was located in Section 28, Burt Twp.,
Cheboygan Co., and consisted of old-growth mixed woodland with
a very rich community of spring ephemeral wildflowers. in
addition to apple trees, this site contained plant species not
found on the other three study sites, including a small population
othusalalzra.

The wet site (WS) was located in the extreme southeast
corner of UMBS property along Greenstar Trail in Section 35,
Munro Twp, Cheboygan 00.; it consisted of a seasonally wet
meadow bordered by cattail marsh, willow/alder swamp, and

21

22
farmlands to the south of the trail. This site had a high diversity
of insect-pollinated plant species, but no nearby firms.
The dry site (08) was located just north of Douglas Lake

Road near Van Creek in Section 26, McKinley Twp., Emmet Co.
This was an open, dry field habitat occupied by scattered white
pines and red maples, shrubby hawthorns, cherries,
serviceberries, and willows, and a low diversity of insect-
pollinated herbaceous plants. No arms was present within
approximately one-half mile of this site.

Trap Nests. A sampling grid was established in each site.
In 08, MS and WS, the grid measured 60 m by 60 m. In CP, the grid
measured 30 m by 40 m because of the smaller area of this site.
At each 10 m interval within the grid, a cluster or six trap nests
(Krombein 1967) held together by wire was positioned in a tree or
shrub branch or attached to a wooden stake, to provide artificial
nesting sites for twig nesting bees. Trap nest clusters were
located between 0.3 and 1.5 m above the ground; this both
approximates the likely foraging height of these bees and is
convenient for examination. Individual trap nests averaged 150
mm in length and 20 mm square, cut from pine or fir. To attract
as many different species of twig-nesting bees as possible, six
sizes of inside bore diameter were used (drill bits were in
English measurements; approximate metric equivalents follow):
5/32” (4mm), 3/16” (4.75mm), 7/32” (5.5mm), 1/4" (6.25mm)
9/32" (7.25mm), and 5/16" (8mm). One of each size was used per

cluster.

23

During preliminary work in 1988, 08, MS, and WS each
received a full compliment of 49 trap nest clusters. In 1989, i
decided to reduce the number of trap nest clusters at DS and WS
to 20 each, because of substantially lower nesting activity at
these sites compared with MS. CP also received 20 trap nest
clusters, because of the smaller size of the sampling grid in this
site. In 1990, only MS was used, because of the high amount of
nesting activity by mgarinata when compared with the other
sites. Only the two smallest sizes (4mm and 4.75mm) of trap
nest bores were used, as H.9arjna1a did not use the larger four
sizes. These small bores were randomly mixed in clusters of six,
and the site (MS) received a full compliment of 49 clusters.

Vegetation Sampling. lnsect-pollinated flowering
vegetation was sampled weekly by placing a 1 m square quadrat
twice at each grid point, with one corner touching the grid point.
Within each quadrat, flowering plants were identified to species
and the number of flowers or inflorescences (for plants such as
composites with highly compact inflorescences of minute
flowers) of each species was recorded. Voucher specimens were
collected for all insect-pollinated plants occurring within the
study sites. A reference collection of pollen from all insect-
pollinated plants occurring within the study sites was made by
collecting fresh anthers and preparing them using the same
techniques as outlinedfor the nest pollen samples (see Pollen

Provision Sampling below).

24

Sampling Chronology. During preliminary work in 1988,.
trap nests were placed in WS on 28 April, and in OS and M8 on 29
April. Weekly sampling continued for 4 weeks. Nests were not
removed from the field sites until 8 October. in 1989, trap nests
were placed in 08, MS, and WS on 4 May and in CF on 10 May;
weekly sampling was initiated immediately and continued until
29 August on all sites. Trap nests remained in all field sites
until 30 September. In 1990, only MS was used based upon the
abundance of nesting activity by liaarjnata in 1988 and 1989;
trap nest placement and weekly sampling began on 30 June (when
individuals of this species became evident in the site) and
continued until 18 August.

Pollen Provision Sampling. Each site was visited
weekly to examine the trap nests for occupation (and sample
flowering vegetation, previously described). Trap nests were
examined with an otoscope; completed nests were removed;
marked with a code indicating site, date, and grid point location;
and replaced with empty nests of the same bore size. The
occupied nests were brought into the laboratory and split open to
examine their contents. A small sample of pollen was removed
from as many cells as possible without disturbing the larvae.
These pollen samples were prepared for preservation on
microscope slides using the acetolysis method (Erdtman 1943)
with the following modifications: following acid treatment and

distilled water washing, the pollen was

25
. washed with a 1% solution of NaOH;
stained with a 0.2% solution of safranin;
washed again with distilled water;

washed with tertiary butanol to remove remaining water;

999N“

placed in glass vials with silicone oil for a few days to
allow remaining alcohol to evaporate.
These samples were then mounted on microscope slides and
labeled for cross-reference to adult bee voucher specimens from
nests. Pollen samples were examined using bright field
microscopy. Pollens were identified to species (when possible;
otherwise, to genus or family) using a prepared pollen reference
collection (see Vegetation Sampling), and approximate
proportions of different pollens in provisions were estimated.
Pollen samples were categorized with respect to the approximate
abundance of films pollen: 100%, 50-99%, 1-49% and 0%.

After removal of pollen samples, the number of cells
constructed in each trap nest was recorded for nests of ii,
carinata. The trap nests of all bees were then resealed and
capped with glass vials to allow for continued larval development
to obtain adult bee specimens for identification, and also to
determine the number of individuals of ligaflnata surviving to
adulthood on different pollens. Nests were overwintered outdoors
in southern Michigan to provide for normal diapause and
development. Emerging adults were killed, pinned, and labeled
according to the code of the trap nest from which they emerged;

they were retained as voucher specimens.

26

Foraging Flight Timings. In 1990, foraging flight
timings were taken on nesting females of H‘garinata. Timings
were taken on warm, sunny afternoons with minimal wind to
reduce variation due to climatic factors. An active nest was
located, and timing was begun when the female left the nest.
While she was away, a grass stem was carefully inserted to
determine the depth of the cell being provisioned; this
measurement was recorded. The timing was terminated when the
female returned with a pollen load (recognized by the bright
yellow or white mass of pollen on her abdominal scopa). Each
observed female was timed for up to 'five consecutive foraging
flights; flights for resin, nectar or other purposes were not
included. The depth measurement was used to determine which
cell was being provisioned during the timed flights when the
completed nest was removed and brought into the laboratory for
opening, and a separate pollen sample was prepared to identify

the plant species being foraged upon during the timed flights.

RESULTS

Nesting Activity and Pollen Provisioning. Forty—five
nests of ii.gar_i_na1a were constructed at the moist site (MS)
between 25 June and 12 August 1989. The onset and peak of
nesting activity of tLQaLmata was synchronous with the
flowering period of BMW (Figures 1a, 1b). Nests
constructed during the peak (first three weeks) of nesting
activity and Elma bloom were provisioned exclusively with films
pollen. Although nests continued to be constructed after firms, had
ceased blooming, nesting activity was highest during this plant's
flowering and declined steadily after Ems, was no longer in
bloom. The blooming period of films, was longer than these data
indicate, since some plants outside the actual sampling quadrats
were observed coming into bloom a week earlier and some may
have continued to flower a week later. This would explain the
ability of iigarinata females to find films, pollen after the floral
sampling data show it to be no longer available.

Only after films flowering had severely declined did 11.
garinata females begin to include other pollens in their nests.
Plant species used for provisioning after Elms included Satureia
vulgarja (Lamiaceae), WW (Guttiferae),
ILjIQLLum sp. (Fabaceae), and several st: acies of composites
(Asteraceae) that were difficult to identiy from pollen. In spite
of their abundance (Figure 1a), these plants were totally
unexploited as pollen sources by licarinata females while anus

27

 

NUMBER OF FLOWERS OR INFLORESCENCES

 

 

 

7/16 -

WEEK
Figure 1a. Weekly floral abundances for the moist site
during 25 June - 12 August. 1989.

 

 

 

 

 

 

 

 

 

 

 

 

15
El 0% Bugs
. El 149% Bum
§ . ,
8 seeev. Etta
10-
m
in I 100% BEE
'5
m 1
z
u 1
° l
m .-
g 5 d I 4:31: i t a
3 ‘ l 7/
,, , //
.— 0 N
'f S E g S 3 a
in ' ' ' ' 3 '
g S g g g S S
WEEK

Figure 1b. Percent Elms pollen in nests of Bananas carinata
from the moist site from 25 June - 12 August, 1989.

29

was in bloom. Nesting activity declined steadily after peaking in
early July, except for a minor resurgence the second week in
August; no nests were found after 14 August. It is doubtful that
this brief revival in nesting activity is due to bivoltinism; no
adults were observed to emerge from any nests brought in from
the field until the following summer during the three years of
this study.

Eight nests of fi‘garinata were constructed at the Colonial
Point site (CP) between 25 June and 12 August 1989. These nests
were temporally scattered throughout the season. This site
contained a small thicket of Engaglama which flowered in early
July. Except for a single nest found, partially completed, on 28
August, in the dry site, no nests of thaLinata were found in the
dry or wet sites. Neither of these sites contained any films.

In 1990, nesting by licarinata at MS began on 1 July, and by
18 August forty-nine nests had been completed. The same
seasonal synchrony between nesting activity and Elma flowering
was observed (Figures 2a, 2b). Peak nesting (i.e., first three
weeks) again corresponded with peak Bhus flowering, and nests
constructed during the peak of the nesting season were
provisioned exclusively with Elma pollel“. As in 1989, other plant
species were ignored in spite of their abundance (Figure 2b)
during Elana bloom and used only after Bhua pollen was declining
in availability. Nesting declined steadily through the season

although again there was a minor resurgence of nesting activity

NUMER OF FLOWERS OR INFLORESCENCES

 

 

 

 

r: 1 '- -° " 3: :3
'j e S S a; a a
5 E '2 :3 § 3 e
rs r~ 3

WEEK

Figure 2a. Weekly floral abundances in the moist site
during 1 July - 18 August, 1990.

NUMBER OF NESTS BEGUM

 

15
UmBfl-fi

E] 149% Bug

50997. 13115

 

lioov. awe

 

 

 

 

 

 

 

 

 

 

£1

 

 

v - c

t :- v- 0 a V v-

" e S S a: a

.- c; ' g n '

\ m N

'7 r: :- g‘ rs a "'
r~ r~ 8

WEEK

Figure 2b. Percent anus pollen in nests offiaflagas garmata
from the moist site from 1 July - 18 August, 1990.

31
in mid-August; some nests were still being provisioned when
sampling was terminated on 18 August.

The average number of cells in each nest of iigaflnata was
higher for nests containing exclusively Elms pollen than in nests
including any proportion of other plant pollens, even when the
non-Elms pollen component was less than 3%, in both 1989 and
1990 (Figures 3a, b). Chi-square tests comparing nests
exclusively containing Elma pollen with nests containing any non-
Bnua pollen in any proportion (Figures 4a, b) were significant for
both 1989 (Chi-square=22.8, df=2, p<.001) and 1990 (Chi-
square-10.5, df-2, p<.01). The average number of cells per nest
also declined steadily on a weekly basis in both 1989 and 1990,
with the exception in both years of a single anomalous nest late
in the season (Figures 5a, b).

Pollen Provisioning By Other Bee Species. Twenty-
two nests constructed by other bee species at the moist site
during 25 June - 12 August 1989 yielded adult voucher specimens.
Many additional nests were constructed but did not yield adult
voucher specimens for identification. Due to the difficulty of
identifying bee species other then tLgarjnaia on the basis of nest
architecture alone only those nests yielding adults for
identification were examined for pollen provisioning. Two
species of megachilids, WWa Cresson and Magaahija
ralatjxa Cresson, frequently nested in the moist site as well as in
the other sites; other species, including Magaamiamjungata
(Fabricius), Magaghilagugnata Say, and two Qamia sp. only

32

 

AVERAGE NUMBER OF CELLS PER NEST

 

 

 

100 50-99 1-49 0
%flflfl§POLLEN

Figure 3a. Number of cells per nest in relation to the proportion of
anuspollen in nests of tiarjagaggaflnata from the moist site for 1989.

 

12

10‘

on
1

AVERAGE NUMBER OF CELLS PER NEST
O\
I

 

 

 

100 50-99 1-49 0
SsflflflfiPOLLEN

Figure 3b. Number of cells per nest in relation to the proportion of
films pollen in nests of tiedadascarinata from the moist site for 1990.

33

 

Iiooec m
.60“ 315

NUMBER or NESTS
3 a
A 1

0|
I l A A

 

 

0d

 

Run-ER OF cELLs PER NEST
Figure 4a. Number of nests in relation to the number of cells per nest
of flariagas carinata from the moist site from 1989.

 

A4 I

NUMBER or NESTS
3

(II
A J k‘ A I .I l

 

 

 

o- -
1 - 4 5 - 8 9 or more

uuueER or was PER NEST
Figure 4b. Number of nests in relation to the number of cells per nest

of [images carinata from the moist site from 1990.

34

 

 

 

 

 

14
q
'5
m 12"
2 l
o:
w
“- 10‘
m
.1 i
.l
m
o 3-
IL
0 u
E 6-
g d
m 4‘1
3 .
‘4' 2-
<
o I I “I, l ' If; ' (L
:2 2 ,_ __
" " R § E a? a
'0 e 2 .', ' 8 J.
g '7 rs E g R 8
WEEK

Figure 5a. Weekly average numbers of cells per nest of fiafladascarinata
from the moist site from 25 June - 12 August, 1989.

 

 

 

 

 

 

 

14
q
512'
Ill
2 cl
3} ..
a 10
(n
—l
d
o 3"
II-
o 1
o:
w d
m 6
5
:t
2
III 4"
o
< .
o:
2
q
< 2
0 I I I I I l r
E E 2 8 S E :2
rs rs R . n B
: to up ' °’ Q. '
" r: r: 8 S a 53‘
N CD_
WEEK

Figure 5b. Weekly average number of cells per nest of Hariagaa garmata
from the moist site from 1 July -18 August, 1990.

35
constructed a single nest each that yielded adults for
idenflficafion.

ngljflscyljnfliga constructed twelve known nests at the
moist site; eleven of these were provisioned with >90% Bills
pollen (three exclusively). The earliest nest contained mostly
Hmaflcumnarfmatum pollen with a small amount of EINS- No
completed nests of inflmjflaa were found prior to 5 July, and
no completed nests were found later than 19 July at this site.

MagaghiLeLaLatjxa constructed six known nests at the moist
site; two of these were provisioned almost exclusively with 31315
pollen. The other nests were provisioned with mixtures of
composite (Asteraceae) species or composites mixed with yigja
flflasa (Fabaceae) or other legume pollens. The earliest of these
nests was completed by 29 June; the last two, which both
contained primarily firms, pollen, were completed by 11 July.

Both of these bee species constructed nests at the other
study sites during 1989. H‘Qflinmma made a total of nine known
nests at the other sites. Two nests from Colonial Point contained
Elms pollen, one completed by 18 July containing about 50% and
the other (completed by 26 July) with about 10%; a third nest
from this site, discovered 10 August, contained no Elma pollen. A
small thicket of Engaglama flowered briefly on this site in early
July. Four nests from the dry site, completed between 4 and 18
July, were provisioned mostly (three contained >90%) with

flxpaflcumgartuamm, which flowered abundantly in this site
throughout July and August. Two nests of Jim from the

36
wet site, both completed by 12 July, contained unknown (possibly
legume) pollens. Several legumes (ILiLQJium sp., Mammfiam,
Madigaggluaulina) flowered abundantly in this site during July.

Magachjiaralaflxa constructed eight known nests at the
other three sites, the earliest completed by 30 June (wet site)

and the last finished by 31 July (dry site). Six of these nests
contained mostly composite pollens. One nest from Colonial Point
contained about 50% pollen from Lamymalamqlma, which
flowered abundantly on this site in July. The last known nest
contained almost exclusively WW pollen.
Foraging Flight Timings. The foraging trips of ten 11,
carinata females were timed in the moist site between 21 July
and 13 August 1990. Five bees provisioned their timed cells with
a predominance (>50%) of films pollen, and five provisioned their
timed cells with mostly other (<50% Enlist) Pollen. Flight timings
were averaged for individual bees. Because foraging flight timing
data may not be normally distributed (Strickler, personal
communication), a Mann-Whitney U-test was conducted on the
two categories of bees. tLQaLinata females gathering
predominantly Elma pollen averaged 13.37 i 3.82 (SD) minutes per
foraging flight. Females gathering mostly other pollensaveraged
9.42 i 3.83 (SD) minutes per foraging flight. These average flight
times were not significantly different (U=19; p=0.175).
Offspring Survival to Adulthood. The number of adults
obtained was compared to the total number of cells constructed

in trap nests of ligaflnata from the moist site in 1989.

be'

D<

(in

EC

37
Trap nests were categorized by the predominance of BM (>50%)
or other (<50% Elms) pollen in the provisions of the whole nest;
nest cells were not sampled individually. Thirty-seven nests,
containing a total of 277 cells, were provisioned with
predominantly firms, pollen and yielded 58 adults for a total
survival of 20.9%. Eight nests, with a total of 39 cells, were
provisioned predominantly with pollens other than films and
yielded only 2 adults for a total survival of 5.1%. A Chi-Square
test conducted on the ratios of cells containing offspring
surviving to adulthood to cells containing offspring that died
before reaching adulthood was significant (Chi-square-143.6;
p<.001). Larval provisions of L-Lgar’ulata containing a
predominance of Elma pollen enabled a higher survivorship to

adulthood than provisions with a minority of, or entirely lacking
in, films pollen.

DISCUSSION

Seasonal Synchrony and Nest Provisioning. Adult
activity of oligolectic bees is frequently synchronized with the
blooming of their preferred host plants (Roberston 1929b, Thorp
1979, Cane and Payne 1988, Cane and Eickwort in press). The
observed synchrony between 11. carinata and Bhua, as well as the
exclusive use of films pollen by film during the peak of the
nesting season, provides evidence for specialization by Ii,
carinataon Elma. Matthews' (1965) observations indicated that
his study populations of man in the vicinity of Kellogg
Biological Station, over 200 miles further south, also used Elms
pollen almost exclusively for the provisioning of nests during the
peak of the nesting season. Matthews also studied a population of
ligarjnata in Oregon; he did not examine the pollen provisions in
their nests, but speculated that ”the source was probably garden
flowers in neighboring yards” (Matthews, 1965). As species of
Elma do occur in Oregon (Peck 1961, Hitchcock and Cronquist
1973), it is possible that Matthews' Oregon population of 11,
carinata was instead foraging on a source of Elma pollen outside
his study area. ligarinata is minimally a regional specialist on
films, and may be oligolectic as defined by Linsley (1958).

An alternate explanation is that ligarinata is a polylectic
species that is only alocal facultative specialist on Elma, due to
local abundance of this plant genus and the large quantities of

38

39
accessible pollen that these plants appear to produce. Some
known polylectic bees have been shown to specialize locally on
highly concentrated pollen sources while ignoring other plants no
matter what their abundance (Stephen and Torchio 1961, Torchio
1976, 1981, Cripps and Rust 1989a, b). Some of these examples
involve agricultural crops such as alfalfa or almonds that present
extreme instances of local abundance, while others represent
situations involving ”cornUCOpia species" (Mosquin 1971).

In my field sites, several other plants were in high floral
abundance during BULLS bloom (Figures 1a, 2a). ligarjnata females
did forage upon these other plant species when films was no
longer flowering. While some strictly monolectic bees may
totally refuse to forage on other plant species when the preferred
host is unavailable (Eickwort 1973, Strickler 1979, Rust 1980,
Neff et al 1982), many oligolectic bees will collect pollen from
non-host plant species when the preferred host species or genus
is of low abundance or unavailable (Michener and Rettenmeyer
1956, Thorp 1969, Cruden 1972, Bohart and Youseff 1976).

Matthews (1965) estimated that females of tLgaLLnata may
live longer than a month. This is certainly supported by my field
studies in 1990, when some nests were still being provisioned on
18 August, more than a month and a half after the beginning of
nesting activity. Long-lived females that still have eggs to lay
may have higher fitness if they lay their eggs on something, even
if it is not the preferred host's pollen, than if they abstain from

further nesting; this is especially true if the potential life

40
expectancy of some females exceeds the host's blooming period.
Thorp (1969) considers the ability of oligoleges to collect other
plant pollens during periods of ”stress” (i. 9., lack of host plant
bloom) to be an adaptive trait.

Nectar Foraging. The exclusive use of films pollen in
nests of tLgaLinata during the peak of the nesting season
indicates that these bees may be foraging exclusively on Erma for
nectar as well. Nectar foraging on other plants would probably
lead to "contamination" of host- plant pollen provisions with 1-
3% of other pollens, as speculated by Rust (1990). As the
blooming period of films was finishing, several nests of ii
carinata began to show "contamination” by pollen from Saturaja
maria and Hypariaumnartgmum, in amounts varying from
traces to a few percent. it is probable that fiminata females
were still collecting residual pollen from senescing male Elms
flowers that were no longer producing nectar, forcing the bees to
collect nectar from other plants. However, pollen foraging on
Bhua appeared to continue until even residual supplies from
senescing flowers had dried up.

Nesting Locality and Host Plant Abundance. ii,
carinata was only found in the proximity of BM and was not
present at study sites DS and WS which did not contain species in
this genus (with the exception of a single, very late nest in 08).
It is believed that most solitary bees forage only within a few
hundred meters of their nests (Eickwort and Ginsberg 1980).

Thus, tLgarinata females only nest in areas with populations of

41

films within their foraging ranges, indicating specialization on
this plant genus. It is known that thaLinata females will use
hollowed-out twigs of EMS. as nesting sites (Matthews 1965).

Pollen Foraging by Other Bee Species. Other bees
occurring in the same habitats as iimrjnata do not exhibit the
same restriction in host plant usage. Hapfltjacylimma appears
to be a local specialist on arms in MS and not truly oligolectic.
Although the flight season of this bee also correlates well with
flhya flowering and almost all of the nests constructed in MS
contained primarily EMS. pollen, mailman also nested in sites
containing little or no arms, and provisioned these nests with
other locally abundant sources of pollen. This bee probably does
not forage on Elma exclusively for nectar even when it is
specializing on films pollen, as only three of the twelve nests
constructed by this bee in MS contained Rhys, provisions that were
completely free of contamination by other pollens. Magagmla
ralatjya begins nesting before films, begins flowering, constructs
nests in habitats containing little or no Elma, and uses a variety
of pollens in nest provisioning, including Elms, when locally
abundant. Minflorescences are highly attractive to several
orders of insects and no doubt provide a highly visible and copious
source of both pollen and nectar. Since the flight seasons of
several solitary bees begin right around the onset of Elma
flowering, it is logical to assume that some, if not most, of these

bees will exploit this abundant food source when available.

42
The foraging habits of these other bee species do not support the
hypothesis that tLgarjnata is foraging on Baas, simply because it
provides the most abundant source of pollen. lf mm were
merely a local specialist on films, because of the abundance of this
plant, it would be expected that this bee would probably switch
to other sources of pollen fairly quickly when anus, bloom was
waning, that it would forage on other plants for nectar while
Elms was in peak bloom (as other bee species appear to), and that
it would also occur in other habitats which did not contain Ema
and use other pollens in nest provisions. Even though other bee
species will locally specialize on Em; when it represents a
highly abundant pollen source, suggesting that ligaL'mata is
merely doing the same, these other species do not show the same
absolute constancy in foraging and nesting location that is
demonstrated by ligarinata.

Parental Investment. The decline in the average number
of cells per nest of ti._cari_na1a as other pollens begin to be
included in the nest provisions (Figures 3a, b) represents a
decrease in investment per individual nest. Other bees have been
shown to reduce the number of cells per nest and/or increase the
number of male progeny (which require less food) as the season
(or even day) progresses, either in response to dwindling floral
resources (Torchio and Tepedino 1980, Torchio 1985, Danforth
1990) or because of decreasing efficiency of aging bees from

wear (Sugiura and Maeta 1989). Thus, it is likely that fewer cells

43
in late season nests is indicative of a behavioral response by Ii,
carinata to the decreasing availability of films, pollen.

l was not able to assess intended sex ratios for ligarmaja
from weighed provision masses, a typical measure of offspring
investment and a possible indicator of sex in dimorphic bees
(Torchio and Tepedino 1980, Sugiura and Maeta 1989, Danforth
1990), because I did not weigh the provisions. Sex of a cell can
also sometimes be determined by size and position, with females
generally found in the innermost, larger cells and males in the
outermost, smaller cells (Krombein 1967, Cowan 1981). I
attempted to estimate sex ratio by cell size and position after
the larvae had constructed their cocoons and discovered that
there was considerable variation in cell size (6-10 mm in length)
without respect to cell position. Although Matthews (1965)
found that the inner nest cells of tLQaflnata had a higher
probability of being female and the outer cells were more likely
to yield male offspring, I encountered many nests with cells
containing dead larvae in central positions and other nests with
very few entirely dead cells, and determining the sex of these
cells was not possible with certainty.

Foraging Efficiency. Elms inflorescences are quite
densely packed with flowers, are presented in an obvious manner
at the apices of well elevated branches, and are usually quite
abundant given the thicket-forming growth habit of this genus.
All of these features would seem to potentially enhance

efficiency at finding and collecting pollen for a specialist

44
forager. It is possible that foraging on other plant species for
pollen is more costly in terms of foraging time, and that this
increases nest construction time which in turn provides more
opportunities for invasion by parasites such as sapygid wasps or
cleptoparasitic megachilids such as Stalls sp. (Matthews 1965)
before a nest is finally closed. If this were the case,
construction of nests containing fewer cells that took less total
time to complete and seal more effectively against parasites
would be advantageous. However, females of H‘gafluata do not
take significantly longer to gather pollen from plants other than
Elma. The number of bees and the total number of flights timed
were low, and timings were not obtained during peak Elms, bloom;
thus, these observations may not be conclusive.

Offspring Survival to Adulthood. ligarinata offspring
appear to have a significantly better chance at surviving to
adulthood when feeding primarily on Elma than on other pollens,
supporting the hypothesis that there is a nutritional basis for
specialization on films by this bee. Other bees have demonstrated
differential survival and/or development on different pollens
(Levin and Haydak 1957, Guirgus and Brindley 1974, Bohart and
Youseff 1976, Tasei and Masure 1978), and it appears that films
pollen confers such a survival or developmental advantage to 11.
carinata larvae. This would suggest that adult behavior evolved
in response to larval nutritional requirements, those adults that
feed their larvae Em pollen having a higher fitness than those

feeding other pollens to their larvae. Thus, foraging for pollen on

45
anus, is selected for in individuals of igarinata. However, the
converse argument could be made that the larvae have adapted to
adult Elms provisioning behavior. The possibility that tLQamea
females are imprinted on films pollen as larvae is not supported
by my observations, as larvae from late-season nests would be
imprinted on other plants. ligarinata has a fairly short
emergence period of approximately one to two weeks: the first
emergence in the laboratory in 1990 was a male that emerged on
26 June, and by 4 July all individuals had emerged. Matthews
(1965) also observed that his populations emerged within a
period of approximately one week. Thus, all females provision
exclusively Elma pollen during the first week or so of their lives

and cannot have been imprinted on any other plants.

SUMMARY

iigarjnata may be oligolectic on am. This bee forages
exclusively for both pollen and nectar on this plant genus when in
bloom, taking pollen from other plants only when Elms, pollen is
no longer available. The onset and peak flight period of this bee
is synchronized with the flowering of its host plants. ligarinata
nests only in localities containing Elms species, and is
completely absent in areas where its host plants are not present.
Larvae of ligarinata more likely to survive to adulthood when fed
on Elana pollen.

46

LIST OF REFERENCES

Baker, H. G. andl. Baker. 1975. Studies of nectar constitution and

plant pollinator coevolution. In W
giants, L. E. Gilbert and P. H. Raven (eds..)

Barkley, F. A. 1937. A monograph of the genus films and its allies
in North and Central America including the West Indies.

AnnalLoIJDLMLsseurLBetanieaLfiardens 24: 265-498

Beck, S. D. 1965. resistance of plants to insects. AnnuaLBayjaw
dLEnterneledx 10: 207-232-

Bohart, G. E. 1972. Management of wild bees for the pollination
of crops AnnnaLBexiellLeLEntemeledx 17: 287 312

Bohart, G. E. and N. N. Youseff. 1976. The biology and behavior of

Ewaausgalninaiae Cockerell (Hymenoptera: Halictidae).
flasmannJeurnalfleledx 34(2): 185 234

Cane, J. H. and G. C. Eickwort. In press. Oligolecty, oligophily, and
the role of specialization in bee-flower coevolution. In S.

Buchmann (ed), W, University of

Arizona Press.

Cane, J. H. and J. A. Payne. 1988. Foraging ecology of the bee
Habmpgdalabarigsa (Hymenoptera: Anthophoridae), an
oligolege of blueberries (Ericaceae: Madeinnrrn) in the
southeastern United States. AnnaanEntmnalggical
SecietLeLAmeriea 81(3) 419 427

Cowan, D. P. 1981. Parental investment in two solitary wasps.
Aneistreeerus adiabatus and Enedxnenls teramrnatlrs
(Eumenidae: Hymenoptera). BehayiQLaLEcelegund

Seeiebieleelr 9: 95- 102.
47

48

Cripps, C. and R. W. Rust. 1989a. Pollen preferences of seven
Qamia species (Hymenoptera: Megachilidae). Enytmnmantal
Entgmalagx 18(1): 133-138.

Cripps, C. and R. W. Rust. 1989b. Pollen foraging in a community
of Qsmja bees (Hymenoptera: Megachilidae). Enyjmnmantal
Entgmgjggy 18(4): 582-589.

Cruden. R. w. 1972. Pollination biology of ngnzigsn
(Hydrophyllaceae) with comments on the evolution of

oligolectic bees. ExaLutjan 26: 373-389.

Danforth, B. N. 1990. Provisioning behavior and the estimation of
investment ratios in a solitary bee, Cameosis
(HimmaemteLa) eersirnilis (Cockerell) (Hymenoptera

Andrenidae) MW 27 159-
168.

Davis, G. R. F. 1968. Phagostimulation and consideration of its

role in artificial diets Bulletin_ei_the_Eniemeledieal
MAW 14: 27- 30

Dobson, H. E. M. 1984. Pollen lipids in bee visited flowers. In J.

N Tasei (ed) WWW
Versailles, 1983. Les Colloques INRA no. 21, Paris.

Ehrlich, P. R. and P. H. Raven. 1964. Butterflies and plants: a
study in coevolution. muting 18: 586-608.

Eickwort, G. C. 1973. Biology of the European mason bee, Hagljtja
W (Hymenoptera: Megachilidae) in New York

State. Search—Adm 3(2) 1-32

Eickwort, G. C., K. R. Eickwort and E. G. Linsley. 1977.
Observations on nest aggregations of the bees Qiagasja
9.1113293 and Dtdjmmuta (Hymenoptera: Anthophoridae).
WW! 50:1- 17

Eickwort, G. C. and H. S. Ginsberg. 1980. Foraging and mating

behavior in Apiodea. AnnualjayjeLQLEntQmelegy 25: 421-
446.

49

Erdtman. G. 1943. An__ln1rodnetien_td_Eellen_Analxsis Chronica
Botanica Co., Waltham, Mass. 239 pp.

Fraenkel, G. 1959. The raison d'etre of secondary plant
substances. Sam 129: 1466-1470.

Frohlich, D. R. and V. J. Tepedino. 1986. Sex ratio, parental
investment, and interparent variability in nesting success
in a solitary bee. EmLutigm 40(1): 142-151.

Golpen, P. B. and S. A.'Brandt. 1975. Alfalfa flower color
associated with differential seed set by leaf-cutter bees.

Admemueurnal 67: 804-806.

Grant, V. 1950. The flower constancy of bees. BetanjeaLBeyiew
16: 379-398.

Guirgus, G. N. and W. A. Brindley. 1974. Insecticide susceptibility
and response to selected pollens of larval leafcutting bees,
Magaghflagagjflga (Panzer) (Hymenoptera: Megachilidae).
EntrirenmentaLEntemeledx 3 691-694

Haydak, M. H. 1970. Honey bee nutrition. AnnuaLBaxieLot
W 15: 143-156.

Hitchcock, C. L. and A. Cronquist. 1973. W
mum. University of Washington Press, Seattle. 730 pp.

Hopkins, C. Y., R. Boch andA. W. Jevans. 1975. Honeybee

attractant compositions. Qanadian_flate_nt (1975) No.
970209, ii+9 pp.

Hopkins, C. Y., A. W. Jevans and R. Boch. 1969. Occurrence of
octadeca- trans- 2, cis-9, cis-12-trienoic acid in pollen
attractive to the honey bee. CanadiaLJQumaLgt
Bimbamistry 47: 433 -.436

House, H. L. 1967. The role of nutritional factors in food
selection and preference as related to larval nutrition of an

insect, Eseudesareeenadaarfinis (Diptera Sarcophagidae)
on synthetic diets. W 99: 1310- 1321.

50

House, H. L. 1970. Choice of food by larvae of the fly, Anga
atfinis, related to dietary proportions of nutrients. moat
diJnsectflxsieLoex 16: 2041-2050.

Hurd, P. D. Jr. 1957. Notes on the autumnal emergence of the

vernal desert bee, WSW (Hymenoptera,

Apoidea) Jeumal_ei__tne_lSansas_Enterneleeieal_Seeient 30:
1 o.

Hurd, P. D. Jr. 1979. Superfamily Apoidea. Pp. 1741 -2209 in K. V.
Krombein, P. D. Hurd Jr., D. R. Smith and B. D. Burks, eds,

WWW.VDIHZ

Smithsonian Institution Press, Washington, D. C.

Hurd, P. D. Jr., W. E. LaBerge and E. G. Linsley. 1980. Principle
sunflower bees of North America with emphasis on the
southwestern United States (Hymenoptera: Apoidea).

W 310 158 pp

Hurd, P. D. Jr. and E. G. Linsley. 1975. The principleLaLLea bees of
the southwestern United States (Hymenoptera: Apoidea).

W19331'74

Hurd, P. D. Jr. and C. D. Michener. 1955. The megachiline bees of

California Bulletin..ei_ttte__0aliiernia_lnsect_§unrex3= 1-
248.

Johnson, M. D. 1984. The pollen preferences ofAngrana

LMaLandLenalgmngi Cockerell (Hymenoptera: Andrenidae).
Jeumal_o.f_tne_lSansas_Entemeledieal_Sdeietx 57(1) 34- 43

Jones, C. E. and S. L. Buchmann. 1974. Ultraviolet floral patterns
as functional orientation cues in hymenopteran systems.
AnimaLBenairier 22: 481-485

Kerfoot, W. B. 1967. Correlation between ocellar size and the
foraging activities of bees (Hymenoptera: Apoidea).

ArnerieanJlanrralist 101: 65- 70

51

Krombein K V 1967 WM
HistoriesJestandJsseeiates Smithsonian Press
Washington D. C. 507 pp.

Krombein, K. V. P. D. Hurd Jr. D. P. Smith and B. D. Burks. 1979.
QataledelelxrneneDtera_rn_Amenea_Nenn_ei_Mexm vol 2

Smithsonian Institution Press, Washington D. C.

Leclercq, J. 1945. Les butineurs de I'Hellebore verte.
Lambillienea 45: 28-29.

Lepage, M. and R. Boch. 1968. Pollen lipids attractive to honey
bees. Lipids 3: 530-534.

Levin, M. D. and G. E. Bohart. 1955. Selection of pollens by honey
bees. W 95: 392-393, 402.

Levin M. D. and M. H. Haydak. 1957. Comparative value of different

pollens in the nutrition of Qsmjaljgnana. Mild 38:
221-226.

Linsley, E. G. 1958. The ecology of solitary bees. HilgaLdja
27(19): 543-599.

Linsley, E. G. 1978. Temporal patterns of flower visitation by
solitary bees, with particular reference to the
southwestern United States. JaumaLQLtnLKansas
EntemeleeieaLSeeietx 51:531-546

Linsley, E. G. and M. A. Cazier. 1963. Further observations on bees
which take pollen from plants of the genus S_o_|an_nn1. Ban;

W342 1-18

Linsley, E. G. and J. W. MacSwain. 1958. The significance of floral
constancy among bees of the genus Djadasia (Hymenoptera:
Anthophoridae). EmLutjan 12: 219-223.

Linsley, E. G., J. W. MacSwain and P. H. Raven. 1963. Comparative
behavior of bees and Onagraceae. l. Qangtnata bees of the
Colorado Desert. ll. Qangtnata bees of the Great Basin.

52

II. °| [CIT . EII°|‘ 'EI I 33:1_
58.

Lovell, J . H. 1914. The origin of oligotropism. Entgmaiggjgai
News 25: 314-321.

Lovell, J. H. 1918. W. C. Scribner's Sons,
New York. 286 pp.

Matthews, R. W. 1965. The biology offianagasgannata Cresson.
Wm 1(3):

Michener, C. D. 1953. The biology of a leafcutter bee (Magagnjte

braids) and its associates LinixersinLeLlSansasSeienee
Bulletin 35: 1569- 1784.

Michener, C. D. 1954. W. Bulletin of the American
Museum of Natural History No. 104.

Michener, C. D. and E. Ordway. 1963. The life history of Bandita

maeuflgeramagujjpannis (Hymenoptera: Andrenidae).
leurnaLeLtne_lSansas_Enterneledieal_Soeietx 36 34- 45

Michener, C. D. and C. W. Rettenmeyer. 1956. The ethology of
mm with comparative data on other species

(Hymenoptera: Andrenidae). WW
BHUBIILI. 37. 645- 684.

MItChBII. T. B. 1962. WWW Volume

II. The North Carolina Agricultural Experiment Station
Technical Bulletin No. 152. 557 pp.

Mosquin, T. 1971. Competition for pollinators as a stimulus for
the evolution of flowering time. Qikga 22: 398-402.

Neff, J. L. B. B. Simpson and L. J. Dorr. 1982. The nesting biology
oijagaaLaatfljgta Cress. (Hymenoptera: Anthophoridae).
JeumaLerJnLKansasfintemeleeieaLSeeietx 55 499- 518

53

Parker, F. D. 1981. Sunflower pollination: abundance, diversity,
and seasonality of bees and their effect on seed yields.

JeumaLLAeicnnuraLBesearcn 20:49-61

Peck. M. E. 1961. A_Manual_o.i_tne_l:lidneLElants_QLQreden. 2nd ed.
Binfords and Mort, Portland. 936 pp.

Plateau-Quenu, C. 1983. Does the volume of a pollen ball
influence the sex of the egg laid on it? An experimental
study dealing with the first brood of Eulaanscalgaatnts
(Seep) (Hym Halictinae) Annales.des.§eience.l:laturelles.
Wang, 13e serie, 5, 41 -.52

Prescott C and R Allen 1986 Ine_Eirst_Besenrce;_lllLild_SDeeies
in_tne_Nenn_ArneLiean_Eeenemx Yale University Press New

Haven, Conn.

Price, P. W. 1984. W. John Wiley and Sons, New
York. 607 pp.

Raw, A. 1974. Pollen preferences of three Osmia species
(Hymenoptera). ijga 25: 54-60.

Robertson, C. 1914. Origin of oligotropy in bees. Entamalggjgal
um 25: 67-73.

Robertson, C. 19298. W. Science Press
Printing 00., Lancaster, Penn. 221 pp.

Robertson, C. 1929b. Phenology of oligolectic bees and favorite
flowers. Ma 36: 112-118.

Roberts, R. B. and S. R. Vallespir. 1978. Specialization of hairs
bearing pollen and oil on the legs of bees (Apoidea:

Hymenootera) Annals.eLthe_Enterneieeieal_Seeiety_et
Amanga 71: 619-627.

Rust, R. W. 1980. The biology offlumnnxbammmnmja
(Hymenoptera: Anthophoridae). W
W 53 427-436

 

54

Rust, R. W. 1990. Spatial and temporal heterogeneity of pollen
foraging in ernialignariaerenineua (Hymenoptera:
Megachilidae). EnflmnmantaLEntgmptggy 19(2): 332-338.

Schemske, D. W. 1983. Limits to specialization and coevolution
in plant-animal mutualisms. Pp. 67-109 in M. H. Nitecki
(ed.), W. University of Chicago Press, Chicago.

Schmidt, J. O. 1985. Phagostimulants in pollen. JaunaLpt
AdieultrtraLBesearcn 24: 107- 114

Schmidt, J. O., S. C. Thoenes and M. D. Levin. 1987. Survival of

honey bees, ApLsmeJljjaLa (Hymenoptera: Apidae), fed
various pollen sources.

eLArneriea 80:17 6- 183.

Stephen, W. P. and P. F. Torchio. 1961. Biological notes on the

leafcutter bee,MedaenileLEutLi.cnaraea1rg.tundata
(FabriciUS) Ean;Eaeitie_Entemeleeist 37: 85- 93.

Strickler, K. 1979. Specialization and foraging efficiency in
solitary bees. E92110! 60(5): 998-1009.

Strickler, K. 1982. Parental investment per offspring by a
specialist bee: does it change seasonally? Exglntjan 36(5):
1098-1100.

Sugiura, N. and Y. Maeta. 1989. Parental investment and offspring
sex ratio in a solitary mason bee, Qamiacmnjtmna
(Radoszkowski) (Hymenoptera, Megachilidae). Japanese

JenrnaLQLEntemelenx 57(4): 861-375-

Tasei, J. N. and M. M. Masure. 1978. Sur quelques factuers

influensant le developpement de Magaghflapagiflga Panzer
(Hymenoptera, Megachilidae). Apidalpgja 9: 273-290.

Tepedino, V. J. 1981. Notes on the reproductive biology of

Zieadenuseanieulatus. a toxic range plant GreaLBasin
Natutaflst 41: 427- 430.

 

55

Tepedino, V. J. 1983. An open-field test of Magaanjjamjundata
as a potential pollinator in hybrid carrot seed fields.

WW 22(1) 64- 68

Tepedino, V. J. and P. F. Torchio. 1982. Phenotypic variability in

nesting success among Qamialjgnaflagmpmaua females in
a glasshouse environment (Hyemoptera: Megachilidae).

EcologicaLEmgmglggx 7 453- 462

Thorp, R. W. 1969. Systematics and ecology of bees of the
subgenus Qiandrana (Hymenoptera: Andrenidae). unjxarsjty
g.t_C.aliIg_Lnia_Eub.ligatig.ns_.in_Emgmglg.ox 5221-146

Thorp, R. W. 1979. Structural, behavioral, and physiological
adaptations of bees (Apoidea) for collecting pollen. Annals

Wm 66: 788- 812.

Thorsteinson, A. J. 1960. Host selection in phytophagous insects.

WM 5: 193-218.

Todd, F. E. and O. Bretherick. 1942. The composition of pollens.
JgumaLoLEangmiLEntgmglggx 35(3) 312 317

Torchio, P. F. 1976. Use of Qamjajjanafla Say (Hymenoptera:
Apoidea. Megachilidae) as a pollinator in an apple and prune

orchard W491
475- 482.

Torchio, P. F. 1981. Field experiments with Qamjaljanafla
W Cresson as a pollinator in almond orchards: I.
1975 studies (Hymenoptera: Megachilidae). W

W 54 815- 823

Torchio, P. F. 1985. Field experiments with the pollinator
species, Qamiaflgnaflagmainaua Cresson, in apple
orchards: V (1979- -,1980) methods of introducing bees,
nesting success, seed counts, fruit yields (Hymenoptera:

Megachilidae) W
58: 448- 464.

 

 

56

Torchio, P. F. 1990a. Qamiaflbiflarja, a native bee species
developed as a commercially managed pollinator of highbush
blueberry (Hymenoptera: Megachilidae). JgumaLQLma
WM 63 427- 436

Torchio, P. F. 1990b. Diversification of pollination strategies for

U. S. crops. WM 19(6): 1649-1656.

Torchio, P. F. and V. J. Tepedino. 1980. Sex ratio, body size and
seasonality in a solitary bee, Qamiajjgnaflanmmgua
Cresson (Hymenoptera: Megachilidae). EALQJMIIQD. 34(5): 993-
1003.

Vivino, A. E. and L. S. Palmer. 1944. The chemical composition
and nutritional value of pollens collected by bees. Amhmas

Mochamism 4: 129- 136

Waser, N. M. 1986. Flower constancy. definition, cause, and
measurement. Wang 127(5): 593- 603.

Wightman, J. A. and V. M. Rogers. 1978. Growth, energy and
nitrogen budgets and efficiencies of the growing larvae of
Magaghflaaagiflga (Panzer) (Hymenoptera: Megachilidae).
Qacalagja 36: 245-257.

APPENDIX 1

Record of Deposition of Voucher Specimens*

The specimens listed on the following sheet(s) have been deposited in
the named museum(s) as samples of those species or other taxa which were
used in this research. Voucher recognition labels bearing the Voucher
No. have been attached or included in fluid-preserved specimens.

Voucher No.: 1991-02

Title of thesis or dissertation (or other research projects):

Specialization, Foraging Efficiency, and Nutritional Relationships
in a Solitary Bee, Heriades carinata Cresson (Hymenoptera:
Megachilidae)

 

Museum(s) where deposited and abbreviations for table on following sheets:

Entomology Museum, Michigan State University (MSU)

Other Museums: None

Investigator's Name (5) (typed)

Jonathon Michael DeNike _

 

 

Date 15 Mav 1991

*Reference: Yoshimoto, C. M. 1978. Voucher Specimens for Entomology in
North America. Bull. Entomol. Soc. Amer. 24:141-42.

Deposit as follows:

Original: Include as Appendix 1 in ribbon copy of thesis or
dissertation.

Copies: Included as Appendix 1 in copies of thesis or dissertation.
Museum(s) files.
Research project files.

This form is available from and the Voucher No. is assigned by the Curator,
Michigan State University Entomology Museum.

APPENDIX 1.1

Vbucher Specimen Data

Pages

_2__

of

__1._

Page

mum:

HOU H:

 

$214.: w. 38

 

 

\sE «380

c 44.6“... a. a. ._

huwmuo>acb madam cmwficofiz may cw ufiwoaov

 

 

you mcoEHoodm woumaa m>onm ocu vo>aouum

 

8:83 .62 65:2, 82;:

oxwzon Homnowz coanGOH
Amvmsmz m.uoumwaumo>aH

Ahummmoooc «a mucosa Hmaowufivvm omav

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

owofi um=w=<lmash
.auwz ..oo commononu ..Q3P uham .m
3m: m w .uom Scum umu: may» seem condom :ommmuo m>fiumaou oawnomwoz
owofi um=w=<lhanh .nuwz
..oo uoaam ..n3H moacfluuz .oN
am: q .oom scum umo: amuu aonm woumom :ommoho mofiuvnflamu mfiuHHmo:
owoa um=w=<lhash .50fiz
..oU commonono ..a:e chug: .mm
pm: a .oom aoum umo: nmuu aoum woumom :ommouo mUfluuaflamo mwuwaaom
am: m N .gowz ..oo commopoau ..93H uham .wN
.uom scum ummc menu scum woumom :ommouu mowuvcwfiww mwuwaqom
oon umaws<lhaah
Laue: ..ou dmwmonozo ..n39 uuam .m
2m: m 0H .oom aoum and: noun aouw condom commono wofiuwgwdww mwufiamo:
; owofi um=w=<lmaah
new: ..oo :mwmononu ..n39 uusm .m
am: am mm .oom scum umm: menu scum condom commouu mumawumo movmfluom
466+ wouwmommv can wow: Ho wouooaaoo GONMu umnuo no mofiumnm
w e r r m m e .m % mdoEHomam now want Honda
wawwmuummWw
mmnnmmmmumm

 

 

“mo nonaaz

 

 

APPENDIX 1.1

Voucher Specimen Data

2 of 2 Pages

Page

 

Sggbb 1N... 3.

huamuo>fica madam :mwfiLUfiz us“ ca ufimoamv

\ mama

HO NH:

 

$.16: m? 86a

 

 

honoEo

 

now maoafiuwam vmumwa m>onm gnu wo>fimumm

 

NOIHooH .oz nonoso> Acmmzuv

mxgzma Homage: 66:66:65

Amvmamz m.uoumwaumo>cH

Ahummmmom: ma mumocm Hmcoaufivwm omDV

 

owofi um=w=<ImH=h .nowz
..oo goaam ..a:a eoficgeoz .om

 

:ommouo w>flumHmu mafinumwmz

 

 

 

 

 

 

 

 

 

 

 

:m: N .oom aouw umo: noun aoum wouwom
oon um2w5<lhaah .aowz
..oo :mwmonono ..n39 owns: .mm
mm: m .oom scum umo: nonu aoum condom :ommouo m>Humaou ofifinomwmm
owoa umnww<lhash .nofiz
..oo commonoao ..a39 unam .wN
am: H q .uom scum umo: awn» Scum condom commouo m>wumaou oawgogwoz
Juh¥ wouwmomow cam mom: no wouooHHoo coxmu nozuo no mofioonm
m e r r m m e .m % mcoefioomm you dump Honda
erOdellapWS
wmwmmmnwmag
"n w.d.1 .u .A .A p. n" r“ we

 

 

"mo Monasz

 

 

 

“ll llllllllllllllllllllll