wwvv

..."...

 

 

 

::¢-;;l$
. “>- 9»:
_. ".rl"'I\-'.

f:
0v
'1

' "P,

4‘ .4 r'
‘ , w.

n wL':$41
h..- :91“.

;4 "2'" 2
-« ‘3 rv
.L‘i \ Au:

" 9;“ :7 \. I‘Q
afzfiwfif
S

J IQ“:
EL".
. '1‘.)
WM." "““r'l‘r
I

fl'. Eve,

14:75:; ’5 J'- a; ‘« 1.2.655; ~-
, 1. ‘fi 1 A“ 3' Afi'fir. ‘ A’» N':
r“? .3“, ‘1? V3,? ‘9:

I 7“" k
.2"!- ”Que;

13"

fl}; 3%"!
' .J 2
)g‘g"u.:m,v
m...
IN-M‘
‘ An)" 5-“! \iat‘ 'I‘gfilv'
A! a???

: A'V'
Jf" I”
7? ‘ o
“‘r ."3 1" :‘E’; {£512. ,xnzh‘ "7 39*?"
I/ 4‘:‘I\{5 y_ 2 7 7‘1"}
by "fig; ‘1'?" Q r 7“ (I n' ‘ 2“ ”:2 “I”? 7',
nrfélli-Zj.‘ '4‘}: 11%.:- :‘x'v-J
. . ’
3;. 9 «
3’

1919'"
’Jult‘

.2"
V 2:;

‘ huh“

~
‘01-'-

1:1‘
.2...

9..., j

$4.327».
T‘LIS't;

,,..:"

 

¢
N: ”fir:- at nu
... "'1' .2 .
-rvp.-u‘- .
if"? 5;... .g'..f.a,fi‘:, .. _.
9' 73- " ”rm w- ..
'3'“ ”“1727: 2“
'5
“9.3."... I 5-3
M. ....
. ::.«:.«'JZL3' i’rflf'rt"
., .

:2 «:3;me

7%

j:

*a‘i. WWW flag.
«:2; 9.222532%» ... ,
.. 45%.... 32"” w .... ”£5; ~*2‘r"'v'-.t.'r’-...~g. .5,
’7' 331%“ ' "#:3573519 f"fi2§.,v-";z§%.... :zr'fi‘i
W» Wg‘xa’écgw‘ .
“5%: w: ~ ' ":2"
:33";

5}]:1 .

... .... ... $r:"£r ‘7'}:

'4‘"; 0 $2,.
"‘:,1‘-$‘.- m
0.2"..- £77... :

..‘t . ".3 . ...... 3.3”“
“ .. {wining-4; “"°?~-~*~".;.. . ‘ .
512;"? .. .m
rut-re:- W." 'Ikl'k '
Jun}; ‘11:”
1:15?

,1 >4 Jinntm.

‘ ""3:
«I: ...;
.,

II "D

1'.
c-

. mcmlom STlMHE UNwiERHSHjY Enzymes l 0\ f: {I ) 7 I 3‘1
s «Milli213g.1\)45ilglgli5|3llyrl|ll
LBBRARY " ' *
Michigan State
University ,

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

This is to certify that the

thesis entitled

NESTING BIOLOGY OF HYLAEUS ELLIPTICUS (KIRBY)
(COLLETIDAE: APOIDEA) IN NORTHERN MICHIGAN

 

presented by

VIRGINIA LOUISE SCOTT

has been accepted towards fulfillment
of the requirements for

MS degree in WY

 

flak/J) A ‘ Mk
I
Major professor

Date35%1:jlfg7

@7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

PLACE IN RETURN BOX to mnovo this checkout from your rooord.
TO AVOID FINES return on or baton duo duo.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.._. __ _____M_SU Is An Affirmative Action/Equal Opportunity Institution

NESTING BIOLOGY OF
HYLAEUS ELLIPTICUS (KIRBY) (COLLETIDAE: APOIDEA)
IN NORTHERN MICHIGAN

BY

Virginia Louise Scott

A THESIS

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

MASTER OF SCIENCE
Department of Entomology

1989

(pol-$2.01;

ABSTRACT

NESTING BIOLOGY OF
HYLAEUS ELLIPTICUS (KIRBY) (COLLETIDAE: APOIDEA)
IN NORTHERN MICHIGAN

BY

Virginia Louise Scott

Nest architecture of Hylaeus ellipticus (Kirby) is
described. Provision sources were mostly rosaceous species.
Nest site selection was influenced significantly by bore
diameter size. Number of primary cells per nest was
significantly greater in 1985 than in 1984. Sex of
offspring, bore diameter size, and cell position contributed
significantly to the variability of cell length as did site
and year for female cells only. Sex of offspring, site, and
bore diameter size contributed significantly to the
variability of cell volume as did cell position for male
cells only. Female offspring weighed 1.5 times greater than
male offspring. Secondary sex ratios were significantly
male biased when compared to expected sex ratios. Mortality
and parasitism did not appear sex biased. Parasites
included: Gasteruption kirbii kirbii (Westwood),
Coelopencyrtus hylaei Burks, Anthrax irroratus irroratus
Say, a dermestid beetle and phorid fly. Relationship
between nest architecture and parasitism is discussed, as

are seasonal trends.

I would like to dedicate this thesis to my mother,
Marilyn J. Scott,
who showed me how to persevere,
taught me true courage,
and never let me forget the value of thinking.

ii

ACKNOWLEDGEMENTS

This study was done in connection with the ELF
Communications System Ecological Monitoring Program,
Biological Study on Pollinating Insects: Megachilid Bees
subcontract number E0 6549-84-C-005.

Special thanks to Mark O'Brien at the University of
Michigan for identifying the Gasteruption, and Dr. George
Eickwort at Cornell University for identifying the
Coelopencyrtus. I would also like to thank Dr. Jerome Rozen
Jr. for housing voucher nests and reared specimens at the

American Museum of Natural History.

I would like to acknowledge Roland L. Fischer's help as
my major professor, and thank Dr. Karen Strickler and Mike
Arduser who also acted as sounding boards and provided

valuable comments on this manuscript.

Dr. Richard Merritt and his staff graciously provided
their electrobalance for my use, and Ken Dimoff provided

patient help with the statistical analyses.

Lastly, I would like to thank my family and Thomas

Czarny for emotional support throughout this study.

iii

TABLE OF CONTENTS

LIST OF TABLES . . . . . . . . . . . . . . . . . . . . vi
LIST OF FIGURES O O O O O O O O O O O O O O O O O O O O Vii i
LIST OF APPENDICES . . . . . . . . . . . . . . . . . . Xi

 

 

 

INTRODUCTION . . . . . . . . . . . . . . . . . . 1
METHODS AND MATERIALS . . . . . . . . . . . . . . . 3
LOCATION . . . . . . . . . . . . . . . . . . . 3
TRAP NESTS . . . . . . . . . . . . . . . . . . 3
OVERWINTERING . . . . . . . . . . . . . . . . 6
DATA COLLECTION . . . . . . . . . . . . . . . 6
POLLEN . . . . . . . . . . . . . . . . . 9
VOUCHER SPECIMENS . . . . . . . . . . . . . . . 10
NEST ARCHITECTURE TERMINOLOGY . . . . . . . . . 10
STATISTICAL ANALYSIS . . . . . . . . . . . . 13
RESULTS AND DISCUSSION . . . . . . . . . . . . . 15
NEST SITE SELECTION . . . . . . . . . . . . 18
Bore diameter size . . . . . . . . . 18

Nest height . . . . . . . . . . . . . 22

Nest entrance orientatio on . . . . . . 22

NEST ARCHITECTURE . . . . . . . . . . . . . 25
POLLEN . . . . . . . . . . . . . . . 32
PRIMARY CELL NUMBER . . . . . . . . . . . . . 33
CELL LENGTH . . . . . . . . . . . . . . . . . 36
CELL VOLUMES . . . . . . . . . . . . . . . 40
SEX PLACEMENT IN NESTS . . . . . . . . . . . 43
EMERGENCE . . . . . . . . . . . . . . 45
OFFSPRING WEIGHTS . . . . . . . . . . . . 45
SEASONAL PRODUCTION OF THE SEXES . . . . . . 50
SEX RATIOS . . . . . . . . . . . SO
MORTALITY AND PARASITISM . . . . . . . . . . . 55
Non-parasitic mortality factors . . . . . . . 55
Parasites . . . . . . . . . . . . 59
Gasteruption kirbii kirbii . . 59
Coelopencyrtus hylaei . . . . . . . . . 62

Anthrax Irroratus irroratus . . . . . . 67
Dermestids . . . . . . . . . . . . . . . 68

Phorids . . . . . . . . . . . . . . . . 68

SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . 69
GLOSSARY . . . . . . . . . . . . . . . . . . . . . . . 74

iv

APPENDICES . .
APPENDIX 1:
APPENDIX 2:

LITERATURE CITED

List 0

véuéhér' sée
List of Herbaceous Angiosperms

cimen

76
78
93

Table

Table

Table

Table

Table

Table

Table

LIST OF TABLES

Number and status of H. ellipticus nests and

 

number of primary cells constructed by year and
SiteOOOOOIIOOOOO0.00.00.00.00...O0.0.0.... p. 16

Number of complete H. ellipticus 1985 nests and

 

primary cells by season of nest completion and
site............... ....................... p. 34
Average number of primary cells per complete H.
ellipticus nest by year for site, bore diameter
size, and season of nest completion....... p. 35

Cell length analyses for H. ellipticus (General

 

Linear Model). .......... ............ ...... p. 39

Cell volume analyses for H. ellipticus (General

 

Linear Model) ............................. p. 42
Secondary sex ratios (male:female) for each cell

position and total number of live H. ellipticus

 

adult offspring emerging from that cell position.
(Cell 1 was the first cell constructed in any
nest) ..................................... p. 44

Dry weight analysis for 1985 H. ellipticus

 

offspring (General Linear Model) .......... p. 49

vi

Table 8

Table 9

Sex ratios (male:female) of H. ellipticus by

 

year ..................... . ................ p. 52
Percentage of H. ellipticus nests parasitized for
each primary cell number by the number of
Gasteruption offspring, 1984 and 1985 data

combined .................................. p. 63

vii

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

LIST OF FIGURES

Frontal view of a bundle of trap nests.... p. 4

Pair of hutches with bundles of trap nests in a

typical habitat........................... p. 4

Illustration of nest architecture terms... p.

Number of all nests of H. ellipticus for each

 

bore diameter size by year.......... ...... p.

Number of complete nests of H. ellipticus for

 

each bore diameter size by season of nest
completion (1985 data only)............... p.

Number of all nests of H. ellipticus for each

 

nest height by year ..... . ............ ..... p.

Number of complete nests of H. ellipticus for

 

each nest height by season of nest completion

(1985 data only) .......................... p.

Number of all nests of H. ellipticus for each

 

nest entrance orientation by year ......... p.

viii

.11

20

21

23

24

26

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Number of complete nests of H. ellipticus for
each nest entrance orientation by season of nest
completion (1985 data only)............... p. 27
Emergence patterns of male and female H.
ellipticus from 1984...................... p. 46
Emergence patterns of male and female H.
ellipticus from 1985...................... p. 47
Percent investment for each sex-season component

by season of nest completion for H. ellipticus

 

offspring‘(1985 data only)................ p. 51
Percent primary cells for each offspring type for

all nests by year. M = male H. ellipticus, F =

 

female H. ellipticus, X = dead H. ellipticus not

 

due to parasitism or handling errors, G =

Gasteruption, C = Coelopencyrtus, A = Anthrax, D

= dermestids, P phorids, K = H. ellipticus

 

killed during handling............ ........ p. 56
Number of primary cells for each offspring type
by season of nest completion (1985 data only). M

= male H. ellipticus, F = female H. ellipticus, X

 

= dead H. ellipticus not due to parasitism or

handling errors, G = Gasteruption, C =

 

Coelopencyrtus, A = Anthrax, D = dermestids, P =

phorids, K = H. ellipticus killed during

 

handling .......... ............. ........... p. 57

ix

Figure 15 Number of primary cells lost to each mortality

factor or parasite by site and year. M = male

III:

ellipticus, F = female H. ellipticus, X = dead H.
ellipticus not due to parasitism or handling
errors, G = Gasteruption, C = Coelopencyrtus, A =
Anthrax, D = dermestids, P = phorids, K = H.

ellipticus killed during handling... ...... p. 58

LIST OF APPENDICES

Appendix 1 List of voucher specimens deposited at Michigan
State University..................... ..... p. 76
Appendix 2 List of herbaceous angiosperms at the various

sites, excluding Poaceae and Cyperaceae... p. 78

xi

INTRODUCTION

Nesting biologies of a number of solitary bees have
been studied in detail, however some groups, including
Hylaeus, remain relatively untouched. The paucity of
biological information regarding the 50 North American
Hylaeus species may be attributed to many factors. Perhaps
the most obvious of these is their small size and lack of
distinguishing characters which make field observations
difficult. Unlike all other North American genera, these
small, relatively hairless bees carry pollen internally, and
have not been included in pollination studies. In addition,
the genus has an ill-deserved reputation for taxonomic
problems. Snelling (1966a, 1966b, 1966c, 1968, 1970, 1975,
1983) has contributed much to our understanding of Hylaeus
taxonomy. Many species are in fact easily identified when

both sexes are present.

This study is aimed at gathering information on the
nesting biology of Hylaeus ellipticus (Kirby). One
advantage to studying this species is that numerous nests
can be collected by trap nesting, and variability within and
between nests can be explored. Comparisons can be made with
what little is known of Hylaeus nesting biologies, and

parallels drawn with other trap nesting bees.

2

Hylaeus ellipticus ranges over most of North America

 

from Alaska to Nova Scotia and southward into parts of
Georgia, Arizona and California (Mitchell, 1960; Snelling,
1970; and Krombein et a1, 1979). To date, information on
this species consists of only a few short publications
(Davidson, 1895: Medler, 1966; and Fye, 1965a & 1965b). We

may conclude from these authors that H. ellipticus nests in

 

sumac twigs or trap nests made from sumac, Sambucus, or
Helga stems. Nest construction materials are limited to
silk produced by the adult female bee and pith scraped from
the inner walls of the trap nest. Nests either contain a
linear series of cells (small diameter nests, ie. < 6.4mm),
or overlapping cells (larger diameter nests, ie. Z 6.4 mm).
H. ellipticus is multivoltine in California, bivoltine in
Wisconsin and univoltine in Ontario. Known parasites of H.
ellipticus include: Habrocytus analis (Ashm.)
(Pteromalidae): Chrysis parvula Fabr. (Chrysididae);
Coelopencyrtus hylaeoleter Burks, Encyrtus sp. (Encrytidae);
and Gasteruption assectator (L.) (Gasteruptionidae). Many

floral records exist for H. ellipticus, but actual pollen

 

sources for nest provisions are unknown (Mitchell, 1960;

Krombein et al, 1979).

METHODS AND MATERIALS

LOCATION - This study took place during the summers of
1984 and 1985 in Dickinson and Iron Counties of Michigan's
Upper Peninsula. The five study sites, C5 (T42N R31W 814),
CH (T43N R30W 318), CL (T43N R3OW 819), F1 (T43N R29W $14),
and F2 (T43N R29W 814) were all open fields bordered by
forests. A list of the relative abundances of herbaceous
angiosperms, excluding the Poaceae and Cyperaceae, can be

found in Appendix 2.

TRAP NESTS - consisted of white pine blocks 19 by 19 by
150mm in which a hole was bored lengthwise to a depth of
roughly 125mm. Six drill bits (4.5, 5.2, 6.0, 7.2, 9.4, and
11.0mm) were used to provide six bore diameter sizes.

Twelve trap nests were bound with plastic strapping into a
bundle so that one of each bore diameter size faced each
direction and no two trap nest entrances were adjoining,
Figure 1. Before nesting began, four bundles of trap nests
were randomly placed crosswise on each shelf of four shelved
wooden hutches located along the field edges, Figure 2.
Shelf heights of the hutches measured roughly 0.1, 0.4, 0.8,
and 1.1 m from ground level. A small roof covered the
uppermost shelf to keep off rain and provide shade. During

1984, four hutches were placed at each site. During 1985,

 

Figure 1. Frontal view of a bundle of trap nests.

 

Figure 2. Pair of hutches with bundles of trap nests in a
typical habitat.

5
six hutches were placed at all sites, except CH, which
remained at four. Hutches were grouped in pairs placed in
close proximity. Within each pair one hutch was oriented
North-South while the other was oriented East—West. This
arrangement resulted in equal numbers of trap nests whose
entrances faced North, South, East or West for each bore
diameter size and nest height, in a given area. When any
nest (Hylaeus or otherwise) was removed from the hutch, the

nest was replaced with the same bore diameter size.

During 1984, each individual H. ellipticus nest was

 

assigned a number, and nest entrance orientation, nest
height and bore diameter size recorded. Nests were taken to
site CH for storage once a nest cap was noticed. No
attempts were made to determine starting dates for these
nests. Nests without nest caps, but with obvious Hylaeus
nesting activity were collected at the end of the summer.
During 1985, all trap nests were checked at 9 to 12 day
intervals for Hylaeus nesting activity by reflecting
sunlight into the tunnels with a mirror. Newly found nests
were individually numbered and nest entrance orientation,
nest height and bore diameter size recorded. In addition,
nest contents or presence of nest caps were noted for all
nests. These nests remained in the bundles at the hutches

until after the first hard freeze (September 11, 1985).

6

OVERWINTERING - Each autumn, one-third of the nests
were placed in either glass or plastic centrifuge tubes
(depending on tube availability), with nest entrances facing
the closed end of the tube. The open end of the tube was
covered with a piece of tightly woven cotton cloth and held
firmly in place with rubber bands. Untubed nests were
stored in tightly sealed cardboard boxes and overwintered
with tubed nests in an unheated garage near site CH.
Untubed nests were subsequently put in plastic centrifuge
tubes the following April. The purpose of storing nests in

tubes was to prevent predation.

DATA COLLECTION - During the spring following nest
construction, while bees were in the prepupal and pupal
stages, nests were opened by splitting along the natural
grain of the wood with a chisel (from the non-entrance end
of the trap nest). Nest architecture measurements were
taken to the nearest .05mm using calipers. Color and
character of silk used in nest construction was recorded.
Cell volumes were calculated using the mathematical formula
for a cylinder. Cell volumes for cells whose shape differed

from true cylinders were not calculated.

Beginning in early June, nests were checked daily for
winged H. ellipticus or parasite adults still in their

cells. These were removed (ie. FORCED EMERGENCE), placed in

7
numbered gelatin capsules, and retained as were the nests.
Sex of H. ellipticus offspring, mortality factors or
presence of a parasite was recorded for each cell. Adult
dry weights (air dried over dry-rite) of 42 female and 58
male H. ellipticus 1985 offspring and all adult parasites
from 1985 nests were taken to the nearest 0.01 mg using a

Cahn 27 electrobalance.

In some instances one parasite offspring consumed the

contents of more than one H. ellipticus cell. PRIMARY CELL

 

NUMBER is an estimate of the number of H. ellipticus cells

 

originally constructed in a given nest, and is based on the
length of parasitized area compared to unparasitized H.
ellipticus cell lengths. SECONDARY CELL NUMBER refers to
the number of cells that were observed after parasitism.
E.g., if a given nest contained 2 unparasitized H.
ellipticus cells and 1 parasite that consumed 4 H.
ellipticus cells, that nest would have a primary cell number
of 6 (2+4) and a secondary cell number of 3 (2+1). CELL
POSITION refers to the position of the (secondary) cell as
numbered in sequence from the innermost (first cell

constructed) to the outermost (last cell constructed).

Information on sex and weight of H. ellipticus

 

offspring will be used to test hypotheses about sex ratio

theory. This theory predicts that equal investment in total

8

production of male and female offspring (Fisher, 1958;
Trivers and Hare, 1976). Since H. ellipticus cells are mass
provisioned and offspring consume the entire amount of food
provided for them, offspring weight can be considered a
reflection of parental investment. If the provisioning bees
invest equally in male and female offspring (ie number of
males x mean male weight = number of females x mean female
weight), it follows that the EXPECTED SEX RATIO
(males:females) should be (and for this paper was)
calculated using the formula:

mean adult female dry weight

mean adult male dry weight.

SECONDARY SEX RATIO (Frohlich and Tepedino, 1986) is

the sex ratio of adult H. ellipticus offspring that emerged

 

from nests. PRIMARY SEX RATIO (Frohlich and Tepedino, 1986)
is an estimate of what the sex ratio would have been had all
H. ellipticus offspring survived to adulthood. This was
estimated for each year by using a method based on Tepedino
and Torchio (1982a). The number and percentage of male
offspring, number and percentage of female offspring and
number of dead offspring (due to all mortality factors
combined) were calculated for each cell position. The
percentage of live male offspring that emerged at each cell
position was then multiplied by the number of dead offspring
at that cell position and added to the number of males for

that cell position, resulting in the primary number of males

9
for that cell position. The primary number of male
offspring was then totaled for all cell positions. The same
was done for female offspring. The ratio of primary number
of males for all cell positions to the primary number of

females for all cell positions equals the primary sex ratio.

The possibility of sex biased parasitism was
investigated by calculating a primary sex ratio for each
parasite species and non-parasitic mortality. The primary
sex ratio for each mortality factor or parasite was then
compared with the corresponding secondary sex ratio. This
method actually tests for cell position biased parasitism
rather than an actual sex biased parasitism, but since many
bees, including H. ellipticus place the sexes in different
positions within the nest (Krombein, 1967), this method

should provide an adequate test.

POLLEN - A reference pollen collection was made during
the summer of 1983 by taking pollen from flowers that had
been collected at the study sites and mounting them in
Kleermount on microscope slides. Pollen samples of H.
ellipticus provisions were taken as larval fecal pellets

from cells in which H. ellipticus offspring developed.

 

These were mashed with a coverslip in warm water on
microscope slides. Excess water was allowed to evaporate

and a temporary mount was prepared in glycerin. Pollen from

10

a total of 36 cells were sampled from all five study sites
at approximately 12 day intervals over the entire nesting
period. Identifications were made using a compound light
phase contrast microscope by comparing pollen samples from
H. ellipticus offspring fecal pellets with reference pollen
slides. Percentages of different pollen sources were
calculated by taking 4 transects of 25 grains across each
microscope slide. Nomenclature follows that of Gleason and

Cronquist (1963).

VOUCHER SPECIMENS - of Hylaeus ellipticus and
associated parasites will be deposited at Michigan State
University, Voucher# 1989—01. (See Appendix 1.) A
duplicate set, where possible, will also be deposited in the
Museum of Zoology at the University of Michigan, Ann Arbor.
Representative trap nests and associated reared material
will be deposited at at the American Museum of Natural

History, New York.

NEST ARCHITECTURE TERMINOLOGY - Basic architectural

terms used in this paper to describe nests of H. ellipticus

 

are shown in Figure 3 and briefly described here. A BASAL
SPACE is a space, usually empty, between the innermost end
of the trap nest and first cell constructed. A CELL is an
area lined with silk and provisioned with pollen and nectar

in which one individual offspring will develop. A CELL CAP

11

       

.mEhwu whaaowznohm «mm: *0 Cozmbm—z: .0 9.39"—

.jwo mbmsxal OwNEm<m<d

 

 

 

 

 

 

 

 

 

 

 

”.0 m2_<_2mm ZOFqud xizw
, . Amozaéom 0002c
moi—<3 m<0me2 / , ZO_.:._.m<n_ m<4344w0mmhz_
. 02E 5% / ..jmooaamma. L /
200000 ZORQDIMBWE ’ / mo<mm .395
1
lllT
._ 1... A
. .- in in
/ ’ \ _ _.
, aozaéom 600;. _ // _ \
mmoém mfiamfimg 03a 30 2253 _ __ mjmo 685$
_ _

mm0<mm mSDjm—OEMFZ

12
is a layer of silk used to close off the outermost end of a
cell after provisioning and oviposition. An INTERCELLULAR
PARTITION, located between two cells, consists of packed
wood scraped from the inner walls of the trap nest. An
INTERCELLULAR SPACE is a space, usually unlined and empty
between cells. An INTERCELLULAR WALL is a silk wall,
perpendicular to the bore (tunnel) length, between two
intercellular spaces. The TERMINAL CELL PLUG is packed wood
scrapings immediately following the final cell of a nest. A
VESTIBULAR SPACE is an empty space after either the terminal
cell plug and before a nest cap wall, or between two nest
cap walls. A SILK RING is a wall of silk, perpendicular to
bore (tunnel) length, in which a central hole, large enough
for the provisioning bee to pass through, has been left. A
NEST CAP is a series of nest cap walls, silk walls
perpendicular to the bore length beyond the last cell
constructed in a nest. (They are similar to silk rings,
except they contain no central hole.) The OUTERMOST NEST
CAP WALL is simply the outermost of all nest cap walls

present.

A COMPLETE nest is any nest with a nest cap present. A
RAIDED nest is any nest that had been observed as complete
during periodic field checks of 1985 (1984 has no raided
nest category since nests were removed as soon as a nest cap

was noticed), but whose nest cap and possibly some cells

13
were removed by another species which nested in the same
trap nest after H. ellipticus. An INCOMPLETE nest is any
nest lacking a nest cap, and showing no signs of another
species nesting beyond the last cell constructed. An
USURPED nest is a nest that was taken over by another
species while active or abandoned prior to completion.
(Intraspecific usurpation would not be detectable). Usurped
nests were never observed as complete. Each nest can be

placed into one, and only one, of these four catagories.

Complete nests constructed during 1985 were divided
into two catagories, early season nests and late season
nests. EARLY SEASON nests are defined as those nests
completed before July 22, 1985. LATE SEASON nests are those
nests completed after July 22, 1985. Completion dates were
used since beginning dates were sometimes difficult to
detect, and thus not accurate. July 22 was chosen since it
was closest to the middle of the nesting season and provided

a similar number of nests for both early and late season.

STATISTICAL ANALYSIS - X2 contingency table were used
to compare nest site selection factors and for sex ratio
analyses. Before data from the five study sites were
combined, G-heterogeneity tests were run to insure sites
were homogeneous (p > 0.05). Yate's correction was used on

all 2 by 2 contingency tables.

14

General Linear Model (GLM) procedure on SAS (Version 5)
was used to analyze sources of variability in cell lengths,
cell volumes and offspring weights. Type IV mean squares
were calculated and overall model error was used to test for
significance of variables. Site, year, sex, and bore
diameter size were treated as class variables. Cell
position and season were treated as covariates. Prior to
all GLM analyses, univariate tests were performed to
determine that data were normally distributed. In these a
Shapiro-Wilk statistic (N < 51) or Kolmogorov statistic (N >
51) were determined, and a significance level of 0.05 used.
Tukey's studentized range (HSD) tests were performed after
each GLM on variables which proved significant. The

significance level for all Tukey's tests were 0.05.

Categorical Data Modeling (CATMOD) procedure on SAS was
used to analyze sources of variability in number of primary
cells per nest. The analysis uses a Wald statistic which
approximates a X2 distribution to test hypotheses about
linear combinations of the parameters in the model. For
this procedure nests were grouped into 5 catagories: 1)
nests with 1 or 2 primary cells, 2) nests with 3 or 4
primary cells, 3) nests with 5 or 6 primary cells, 4) nests
with 7 or 8 primary cells, and 5) nests with 9 or more

primary cells. Variables tested included: site, year, bore

15
diameter size, season of nest completion, and parasitism by
Gasteruption. Analysis was done with a series of two
variable tests, since more variables resulted in sample
sizes that were too small. Any sample of a two variable
combination (ie. site-year) with a sample size 5 5 was

deleted from the analysis.

RESULTS AND DISCUSSION

During 1984, 90 nests were collected, 82 (91.1%) of
which were complete. During 1985, 181 nests were collected,
116 (64.1%) of which were complete. Number of secondary
cells collected was 400 in 1984 and 1183 in 1985. Number of
primary cells constructed was estimated to be 435 in 1984
and 1241 in 1985. Number of 1) complete nests 2) not
complete nests (including incomplete, raided, and usurped
nests) and 3) total of all nests, along with number of
primary cells constructed in those nests are listed by year
and site in Table 1. The increase in the proportion of
incomplete, raided and usurped nests during 1985 can mostly
be attributed to increased collection of incomplete and
usurped nests, since it became easier with practice to find

nests prior to nest cap construction.

16

Table 1. Number and status of H. ellipticus nests and
number of primary cells constructed by year and s1te.

 

 

 

NUMBER OF
Year COMPLETE NOT COMPLETE1 TOTAL
Site Nests Cells Nests Cells Nests Cells
1984
C5 7 29 0 0 7 29
CH 13 77 2 11 15 88
CL 13 67 0 O 13 67
F1 19 66 2 11 21 77
F2 29 142 3 14 32 156
? 1 8 1 10 2 18
Total 82 389 8 46 90 435
1985
C5 15 112 9 47 24 159
CH 3 33 11 97 14 130
CL 27 205 11 80 38 285
F1 16 91 6 20 22 111
F2 55 392 28 164 83 556
Total 116 833 65 408 181 1241

 

1includes incompléte, raidedfand usurped.

17

Besides H. ellipticus two other Hylaeus species nested
in the trap nests, Hylaeus basalis (Smith) and Hylaeus
verticalis verticalis (Cresson). Of these, only the latter
could be considered a major competitor for trap nests, since
it occurred at all study sites and was approximately two-
thirds as abundant as H. ellipticus. H. basalis was a
sporadic nester, absent from some study sites and only one-
tenth as abundant as H. ellipticus. These will be

considered in a later publication.

During 1985 (the only year nests were left in the field
for the entire summer) 64.1% of all nests were complete, and
18.9% of all nests were incomplete. Raided nests accounted
for 5.0% of all nests constructed. Raiding species
included: eumenids (22.2%), sphecid (22.2%), pompilids
(11.1%), Hylaeus verticalis (11.1%), and undeterminable
species (33.3%). Nest usurpation was noted in the remaining
12.2% of the 1985 nests. Usurpers included: spiders
(31.8%), pompillids (13.6%), sphecids (13.6%), Megachile

 

relativa Cresson (9.1%), Hoplitis spp. (9.1%), Hylaeus
verticalis (4.6%), Osmia spp. (4.6%), and undeterminable

species (13.6%). H. ellipticus was known to supersede H.

 

verticalis on two occasions. In one instance, a single H.
ellipticus cell was constructed in a trap nest after an H.

verticalis nest cap was complete. In the other instance, 3

H. ellipticus cells followed directly after the first and

18
only H. verticalis cell in the nest. Whether this was

usurpation of an active H. verticalis nest is unknown.

NEST SITE SELECTION - The methods of trap nesting
described earlier allow for analysis of factors important in
nest site (ie. trap nest) selection. Three bore diameter
sizes (4.5mm, 5.2mm, 6.0mm), four nest heights (1.1m, 0.8m,
0.4m, 0.1m), and four nest entrance orientations (North,
East, South, West) were used by this species. All nests
(complete, incomplete, raided and usurped) are included in
the first analyses to look at factors affecting overall nest
site selection. A second analysis was done to look for a
seasonal shift in nest site selection. Since complete nests
of H. ellipticus were not replaced during 1985, it is likely
that many of the more acceptable nest sites were used during
early season. Later in the nesting season, female H.
ellipticus might have had to select from the less acceptable
nest sites or nest elsewhere. Only the complete (dated)
nests from 1985 were used for seasonal analyses of factors

influencing nest site selection.

Bore diameter size - Of the 1984 trap nests selected,
82.2% were in the 4.5mm diameter trap nests, 14.4% in the
5.2mm diameter trap nests, and 3.3% in the 6.0mm diameter
trap nests. Of the 1985 trap nests selected 73.9% were in

the 4.5mm diameter trap nests, 18.2% in the 5.2mm diameter

19
trap nests, and 7.9% in the 6.0mm diameter trap nests,
Figure 4. H. ellipticus showed a strong preference for
nests of smaller diameters. The overall distribution of
trap nests selected among the various bore diameter sizes
available was significantly different from equal (X2 =
229.4, p < 0.001, df = 2). Of the nests completed early
season 1985, 85.7% of the selected trap nests had a bore
diameter size of 4.5mm, 14.3% were in 5.2mm diameter trap
nests, and 0 nests in 6.0mm diameter trap nests. Of the
nests completed late season 1985, 68.1% of the selected trap
nests had a bore diameter size of 4.5mm, 19.1% were in 5.2mm
diameter trap nests, and 12.8% were in 6.0mm diameter trap
nests, Figure 5. Use of 5.2mm and 6.0mm trap nests
increased late season as compared to early season. The
distribution of trap nests selected among the various bore
diameter sizes available by season of nest completion was
significantly different from equal when 5.2mm and 6.0mm trap
nests were combined (Xi: = 4.91, p < 0.05, df = l). Trap
nests with a bore diameter size of 4.5mm were the most
acceptable nesting sites in this study. Had trap nests with
a diameter smaller than 4.5mm been available it is probable
that they, too, would have been selected. Trap nests with
larger bore diameter sizes (5.2mm and 6.0mm) seem to be
selected when smaller diameter trap nests were not

available.

BORE SIZE USAGE BY YEAR

 

 

SLSEN :IO HEHWDN

BORE SIZE
Figure 4. Number of all nests of 1:1. ellipticus for eac

by year.

in bore diameter size

21

.35 See mam: 826388 «we: co common 3 mum
55:86 Son some .2 33:3 .3 .6 Emma 829:8 Lo .3632 .m 9:9...

mN_m mem

EE «6

 

 
 

l
S
SisaN so uaewnN

 

 

 

 

.. 4
ME: I . 8
5115 a
I - 8
- 8

mmmw .. 20m<mw >m m0<m= MN_w mem

22
Nest height - Of the 1984 trap nests selected 30.2%
were located 1.1m above ground level, 26.7% at 0.8m, 23.3%
at 0.4m, and 19.8% at 0.1m. Of the 1985 trap nests selected
25.6% were located 1.1m above ground level, 27.3% at 0.8m,
28.4% at 0.4m, and 18.7% at 0.1m, Figure 6. H. ellipticus

 

showed a slight aversion to nests only 0.1 m above ground
level, but the overall distribution of trap nests selected
among the various heights available was not significantly
different from equal (X2 = 4.9, p > 0.1, df = 3). Of the
nests completed early season 1985, 22.6% were in trap nests
located 1.1m above ground level, 25.8% at 0.8m, 37.1% at
0.4m, and 14.5% at 0.1m. Of the nests completed late season
1985, 17.0% were located 1.1m above ground level, 34.0% at
0.8m, 29.8% at 0.4m, and 19.2% at 0.1m, Figure 7. The
distribution of trap nest heights selected by season of nest
completion did not differ significantly from equal (X2 =
1.80, p > 0.05, df = 3). Nests on the lowest shelf were
often somewhat blocked and shaded by vegetation. Danks

(1971) found that the European species H. brevicornis prefer

 

nests with little or no vegetative cover and in either

little or no shade. H. ellipticus seems to follow suit.

Nest entrance orientation - Of the 1984 trap nests
selected, 18.6% had entrances that faced North, 24.4% faced
East, 27.9% faced South, and 29.1% faced West. Of the 1985

trap nests selected, 24.0% had entrances that faced North,

HEIGHT USAGE BY YEAR

I 1984
1985

%

RRW

m\\\\\\\\\\s _

 

 

OOOOOOO
QQQQQQ

SlSEN :IO HESWHN

NEST HEIGHT

ch nest height by year.

Figure 6. Number of all nests of 1214111911915; for ea

24

.35 Sen mom: cozoEEoo 1mm: .6 .633
3 £92 “we: come .2 «133:3 fl .6 9mm: 829:8 to 56:52 K 2:9...

.5..sz ...mmz

E F... E m6 E ed

         

 
         

    

 

  
   
     

 

    

    

 
 
 

 

\
ll
\
ll
\
ll
\

\\\\
\\\\

\\\\ ICN

\\\\

UFS I ssxs
>._m<m E .

lllll
lllll
lllll

 

 

I

\\\ \\\\
Illl III; III
\\\ \\\ \\\\
Illl III; III
\\\ \\\ \\\\
Illl III; III
\\\ \\\ \\\\
Illl III; III
\\\ \\\ \\\\
Illl Ill; III
\\\ \\\ \\\\
Illl III; III
\\\ \\\ \\\\
Illl III; III
\\\ \\\ \\\\
Illl III; III
\\\ \\\ \\\\
Illl III) III
\\\ \\\ \\\\
llll III; III
\\\ \\\ \\\\ 8
Illl III; III
\\\ \\\ \\\\
Illl Ill; III
\\\ \\\ \\\\
llll III; III
\\\ \\\ \\\\
III; III
\\ \\\\
III; III
\\\\ I
\I\I\I\ d

 

r on
mmmw . 20m<mm >m m0<m= ...—LEE...

25
19.4% faced East, 33.7% faced South, and 22.1% faced West,
Figure 8. The overall distribution of trap nests selected
among the various entrance orientations available was not
significantly different from equal (X2 = 7.2, p > 0.05, df =
3). Of the nests completed early season 1985, 29.0% of the
selected trap nests had entrances which faced North, 19.4%
faced East, 33.9% faced South, and 17.7% faced West. Of the
nests completed late season 1985, 25.5% of the selected trap
nests had entrances which faced North, 12.8% faced East,
38.3% faced South, and 23.4% faced West, Figure 9. The
distribution of trap nests selected among the various
entrance orientations available by season of nest completion
did not differ significantly from equal (X2 = 1.39, p >
0.05, df = 3).

NEST ARCHITECTURE - Materials used in nest construction
were limited to silk, a cellophane-like secretion produced
in the thoracic salivary gland of adult female Hylaeus
(Batra, 1972 & 1980) and wood fibers scraped from the inner

walls of the trap nest. Wood fibers used by H. ellipticus

 

would be analogous to pith reported in nests constructed in
twigs (trap nests or natural) of various Hylaeus species
(Davidson, 1895: Fye, 1965b; Medler, 1966; and Krombein,
1967).

Since actual nest construction was not observed, it is

impossible to know to what degree female H. ellipticus

 

ORIENTATION USAGE BY YEAR

R\\\\\\\\\\\\\\\\\N

‘W

 

 

SlSEN :IO HEEWRN

l-

ORIENTATION

Figure 8. Number of all nests of ti. ejflpjjgus for ea

orientation by year.

27

.38 28 mam; 8:an8 one: .o 888 .3 8sz28
8:88 L8: :86 .6. ago .fl .6 2mm: 229:8 .o .8832 .m 6.39“.

ZO_._.<._.Zm_m—O
...mm? Ihacm hm<m :hmoz

      

   

l
l

\\\ .1
\\\ \\\\

l l

  

l
l

    
    

I l l I l l
\ \ \ \ \ \ \

l\l\l\ \l\l\l\

l l l l l l

\
\
\

  

\
l

\
l

\

 
  

l
l

\ \ \

     

l
l

\ \
l l
\
l
\

l l

\
\

    

l l l l l l

               

l
l
l

 

\ \ \ \
l l l
\ \ \ \
l l l
\ \ \ \ \
l l l l l
\ \ \ \

 

 

\ \ \ \ \
l l l l l l
\ \ \ \ \
l l l l l l

\ \ \ \

    

   

 

 

SISSN :IO HEEWRN

w._.<._ I
>._m<w E

 

.. on
ma? . 20m<mw >m m0<m= ZO_._.<._.ZmEO

28
modified the trap nests prior to nest construction. There
was, however, no evidence to suggest that female bees
lengthen the tunnels prior to cell construction, though this
had been reported for H. ellipticus in sumac-twig trap nests
(Fye, 1965b). It was noted during the 1985 field checks
that once a nest was begun, silk rings were often placed
between the trap nest entrance and cell being constructed or
provisioned. Silk rings were present in many incomplete
nests, always between the nest entrance and the last cell
being constructed. Of 34 incomplete nests constructed
during 1985, 25 (73.5%) had at least 1 silk ring. Among
these 25 nests, there was an average of 6.4 i 4.1 (range 1-
18) rings per nest. Silk rings were occasionally visible in
completed nests, but with much less regularity. Of 116
complete nests constructed during 1985, only 17 (6.8%) had 1
or more silk rings in addition to a nest cap. These 17
nests averaged 3.75 i 2.9 (range 1-9) rings per nest. This
suggests that silk rings are either removed or nest cap
walls (and possibly cell caps) were constructed as an
extension of the silk rings by closing off the central
opening, thus obscuring their original form. Some nest cap
walls appear as if they were constructed in 2 parts, an
outer ring first and then the central hole closed off. A
possible advantage to the construction of silk rings will be
discussed in further detail under the section on

Gasteruption parasitism.

29

Basal spaces occurred in 4.4% of nests constructed
during 1984 and 18.4% of nests constructed during 1985.
They averaged 11.9 i 3.5mm (range 7.2 to 15.9mm) long in
1984 and 13.6 i 13.5mm (range 0.2 to 60.0mm) in 1985. The
first cell constructed in a nest was usually placed flush
with the innermost end of the trap nest (95.6% of 1984 nests
and 81.6% of 1985 nests). Cell construction as indicated by
various stages of incomplete cells appears consistent with
observations of Torchio (1984) for H. bisinuatus Forster.
Cells were placed linearly in trap nests with a bore
diameter size of 4.5mm or 5.2mm. In 6.0mm trap nests, cell
orientation was distorted, skewed (sometimes to the point of
being almost perpendicular to the nesting tunnel, or
overlapped, resulting in 2 or 3 cells at any given depth.
This irregular cell placement was previously noted for H.
ellipticus by Fye (1965b) and Medler (1966) in nests of

6.4mm and 8.0mm diameters.

Provisions filled 1/2 to 3/4 of the cell and consisted
of a viscous pollen-nectar mixture. The spring following
nest construction, cells in which no offspring developed
contained an eggshell to light tan colored pollen mixture
with the consistency of spun honey. This supports the
reported water retention properties of silk (Batra, 1972).

Eggs were apparently laid after cell provisioning was

30
complete, as evident by signs of desiccated eggs that failed
to hatch on the outer surface of some pollen masses. Eggs,
appear to have been laid lengthwise on the anterior end of

the provisions, agreeing with observations of H. bisinuatus

 

by Torchio (1984).

After a cell is capped with a wall of silk, an
intercellular partition of wood fiber scrapings was commonly
packed between cells, such that the fibers would not readily
separate unless prodded. These intercellular partitions
ranged in length from a size too small to measure to the
length of a cell, but usually measured between 0.5 and
1.5mm. After the last cell of the nest a terminal cell plug
of loosely packed woodscrapings was usually present (87.8%
of complete nests from 1984 and 88.8% of complete nests from
1985). The average length of the terminal cell plug was
3.42 i 2.06mm (range 0.75 to 11.75mm, N=72) for 1984 and 3.6
: 3.5mm (range 0.25 to 15.75mm, N=103) for 1985.

One or more intercellular spaces were present
approximately half way through cell construction in 12.2% of
1984 nests and 23.8% of 1985 nests. Intercellular spaces
may help to reduce parasitism by one species, and will be
discussed in further detail under the section on

Coelopencyrtus parasitism.

31

Following the terminal cell plug, nests were empty with
the exception of a silk nest cap. The silk used in making
the nest cap walls appeared similar to that used throughout
cell construction, only thicker. In smaller diameter nests,
nest cap walls were reminiscent of crinkly, thin, dull
grayish-white wax paper. In trap nests with larger bore
diameter sizes, nest cap walls were more transparent and
shiny, with an ever so slightly bluish tinge. They appeared
to be thinner and were more easily torn than those in
smaller diameter nests. The number of nest cap walls for
complete nests averaged 3.06 i 1.93 (range 1 to 11, N=82) in
1984 and 3.34 i 1.75 (range 1 to 8, N=116) in 1985. The
outermost nest cap wall was recessed from the nest entrance
Ian average of 6.97mm i 8.50 (range 0 to 46.1mm, N=82) in
1984 and 8.13mm 1 11.05 (range 0 to 56.6mm, N=116) in 1985.
The outermost nest cap wall was always attached to the inner
walls of the nesting here. Even when placed extremely close
to the nest entrance, it never extended onto the outer face
of the trap nest. The placement of the outermost nest cap
wall and the number of nest cap walls, in combination with
silk color and character made it possible to distinguish
nests of this species from nests of either H. basalis or H.
verticalis. As a result, species identification could be

made for complete nests in which no adult H. ellipticus

 

offspring deve10ped. This provided additional nests and

32
parasite information that would otherwise have had to be

omitted.

POLLEN - Of the 36 cells sampled from the 1985 nests,
29 (80.6%) were comprised entirely of rosaceous pollen.
Plants of the Rosaceae which bloomed at the study sites
during the H. ellipticus nesting period included (in
decreasing order of abundance): Rubus, Spiraea, Physocarpus,
Geum, Potentilla, Agrimonia, and Hggg. Those cells
containing provisions that were not entirely rosaceous
pollen occurred either at the beginning (2 cells) or the end
(5 cells) of the nesting period. Of the two cells that were
provisioned at the beginning of the summer, one was 5%
Heracleum lanatum (Apiaceae) pollen and 95% rosaceous
pollen, while the other cell contained 2% Hieracium spp.
(Asteraceae) pollen and 98% rosaceous pollen. Of the five
cells that were provisioned at the end of the summer, one
contained 3% Anemone spp. (Ranunculaceae) pollen and 97%
rosaceous pollen. Three cells provisioned at the end of the
summer contained 1, 1 and 9% Aster-Solidago (Asteraceae)
pollen with the remaining percentage being rosaceous pollen.
The final cell was comprised entirely of Cirsium spp.
(Asteraceae). This cell was provisioned extremely late in
the summer (during September) and occurred at the site with

the least amount of Rosaceous plants.

33

From these data it becomes evident that H. ellipticus
selects pollen from plants of the Rosaceae when possible,
but can use other pollen sources if forced to for either
part or entire provisions. This, in addition to Snelling's
(1970) reference of H. bisinuatus being oligolectic on
Legumes, suggests that Hylaeus spp. may be more pollen
specific than previously thought. Although they may visit a
wide range of plant species for nectar, at least some
Hylaeus spp. seem to restrict pollen collection to certain

plant families under certain conditions.

PRIMARY CELL NUMBER - Of the 113 complete nests
constructed during 1985 for which accurate completion dates
are known, 63 (55.8%) nests containing 363 (45.1%) primary
cells were completed during early season and 50 (44.2%)
nests containing 441 (54.9%) primary cells were completed
during late season. Number of complete nests and primary
cells constructed during early and late season 1985 are

given, by site, in Table 2.

The average numbers of primary cells per complete nest
are listed in Table 3 by year for site, bore diameter size,
and season of nest completion. The average number of
primary cells per complete nest in 1985 (mean = 7.18, N=116)
was significantly greater than in 1984 (mean = 4.74, N=82)

in a catmod analysis of year and site (p = .008). This is

34

Table 2. Number of complete H. ellipticus 1985 nests and
primary cells by season of nest completion and site.

 

 

SEASON OF NEST COMPLETION

 

 

EARLY LATE TOTAL
Sitetgf Nests Cells Nests Cells Nests Cells
cs 8 47 5 45 13 92
CH 2 24 l 9 3 33
CL 17 102 10 103 27 205
F1 8 39 8 52 16 91
F2 28 151 26 232 54 383

Total 63 363 50 441 113 804

35

Table 3. Average number of primary cells per complete H.
ellipticus nest by year for site, bore diameter size and
season of nest completion.

 

 

 

YEAR
1984 1985

_, N MEAN SD RANGE N MEAN SD RANGE
Site

C5 7 4.14 3.67 1-11 15 7.47 4.29 1-14

CH 13 5.92 2.72 2-10 3 11.00 3.46 9-15

CL 13 5.15 2.64 1-11 27 7.59 4.58 1-20

F1 19 3.47 2.72 1-10 16 5.69 4.33 1-13

F2 29 4.90 4.18 1-15 55 7.13 5.27 0-18
Total 82 4.74 3.41 1-15 116 7.18 4.84 0-20
Bore size
4.5mm 67 5.01 3.57 1-15 88 7.28 4.73 0-18
5.2mm 13 3.46 2.30 1-8 18 7.00 5.70 1-20
6.0mm 2 4.00 2.83 2-6 7 7.71 5.28 1-17
Total 82 4.74 3.41 1-15 113 7.12 4.90 0-20
Season
Early -- --- --- --- 63 5.76 4.23 0-15
Late -- --- --- --- 50 8.82 5.19 1-20
Total -- --- --- --- 113 7.12 4.90 0-20

 

36
probably due to weather and floral resource differences
between the two years. Site did not contribute
significantly to the variability in the number of primary
cells per complete nest in catmod analysis of year and site
(p = .32). During both 1984 and 1985 the mean number of
primary cells per nest was highest at CH, and lowest at F1.
Again this may reflect resource abundances at the individual
study sites. Bore diameter size did not contribute
significantly to the variability in number of primary cells
per complete nest in catmod analysis of year and bore
diameter size (p = .48). Season of nest completion also did
not contribute significantly to the variability in number of
cells per complete nest in catmod analysis of site and
season for 1985 data only (p = .11). The average number of
primary cells per nest was, however, greater for nests
completed late season. This may partially be an artifact of
the length of time it takes H. ellipticus to construct nests
in combination with the use of completion dates for season
definition, since all nests over 15 cells were begun early

season, but not completed until late season.

CELL LENGTH - During 1984, cells containing male H.
ellipticus offspring (all sites and bore diameter sizes
combined for all nests) averaged 5.47mm (N = 181) while
those containing female H. ellipticus offspring averaged

5.99mm (N = 77). During 1985 cells containing male H.

37
ellipticus offspring averaged 5.21mm (N = 551) while cells
containing female H. ellipticus offspring averaged 5.73mm (N

= 227).

Cells from 4.5mm diameter trap nests (all nests from
all sites and both years) averaged 5.43mm (N = 584) for
cells containing male H. ellipticus offspring and 5.90mm (N
= 230) for cells containing female H. ellipticus offspring.
Cells from 5.2mm diameter trap nests averaged 4.67mm (N =
129) for cells containing male H. ellipticus offspring and

5.40mm (N = 59) for cells containing female H. ellipticus

 

offspring. Cells from 6.0mm diameter trap nests averaged
4.50mm (N = 19) for cells containing male H. ellipticus
offspring and 5.76mm (N = 15) for cells containing female H.
ellipticus offspring. Cell lengths for each site (both
years and all bore diameter sizes combined) for male H.
ellipticus offspring averaged 5.28mm (N = 78) at C5, 5.26mm
(N = 110) at CH, 5.29mm (N = 165) at CL, 5.47mm (N = 79) at
F1, and 5.21mm (N = 300) at F2. Cell lengths for each site
for female H. ellipticus offspring averaged 5.80mm (N = 45)
at c5, 5.56mm (N = 34) at CH, 5.97mm (N = 48) at CL, 5.61mm
(N = 41) at F1, and 5.86mm (N = 136) at F2. Clearly, cells
in which female offspring were produced were longer than

cells in which male offspring were produced.

38

Sex of offspring, bore diameter size and cell position,
each contributed significantly to the variability in cell
length, while site, year, and a site-year interaction did
not (Table 4). Since sex of offspring had a significant
affect on the variability of cell lengths, a second set of
analyses were run for male and female offspring separately.
This should eliminate any compounding factors such as a sex
biased cell positioning. For male offspring, bore diameter
size and cell position proved significant, while site and
year did not (Table 4). Tukey's analysis showed cell
lengths in trap nests with a bore diameter of 4.5mm (mean =

5.43mm, N

584) were significantly longer than in 5.2mm
(mean = 4.67mm, N = 129) or 6.0mm trap nests (mean = 4.50mm,
N = 19), while the latter two were not significantly
different from each other. For female offspring, bore
diameter size and cell position were significant, along with
site and year, Table 4. Tukey's shows cell lengths in 4.5mm
diameter trap nests (mean = 5.90mm, N = 230) and 5.2mm trap
nests (mean = 5.40mm, N = 59) to be significantly different,
while cell lengths from 6.0mm trap nests (mean = 5.76mm, N =
15) were not significantly different from those in either
4.5mm or 5.2mm diameter trap nests. The 1985 data were used
separately to investigate the affects of season on cell
length. Season did not contribute significantly to the

variability of cell lengths, Table 4.

39

Table 4. Cell length analyses for H. ellipticus (General

Linear Model)

 

1984 & 1985, both sexes

 

 

 

 

some]: or 33 F PR > F
Sex 1 29.0226 55.85 0.0001***
Site 4 3.8300 1.84 0.1184
Year 1 0.6729 1.29 0.2554
Site*year 4 2.8263 1.36 0.2460
Bore size 2 71.3213 68.63 0.0001***
Cell position 1 36.1988 69.66 0.0001***
Model 13 186.7241 27.64 0.0000
Error 1022 531.0705

r2 = 0.2601 C.V. = 13.2819

1984 and 1985 data, males only

SOURCE DF 88_ F PR > F
Site 4 1.8750 0.91 0.4592
Year 1 1.5579 3.01 0.0829
Bore size 2 67.8729 65.67 0.0001***
Cell position 1 31.7230 61.39 0.0001***
Model 8 110.7870 26.80 0.0001
Error 723 373.6128

r2 = 0.2287 C.V. = 13.6337

1984 and 1985 data, females only

SOURCE OF SS F PR > F
Site 4 7.9321 3.98 0.0037**
Year 1 2.2361 4.48 0.0351*
Bore size 2 9.9607 9.99 0.0001***
Cell position 1 4.5793 9.18 0.0027**
Model 8 26.5290 6.65 0.0001
Error 295 147.1211

r2 = 0.1528 C.V. = 12.1758

1985 data, both sexes

SOURCE DF 88 F PR > F
Sex 1 13.4555 27.60 0.0001***
Site 4 13.4512 6.90 0.0001***
Season 1 0.0000 0.00 0.9994
Bore size 2 40.2249 41.26 0.0001***
Cell position 1 23.8262 48.87 0.0001***
Model 9 120.0262 27.36 0.0001
Error 550 268.1248

r2 = 0.3092 C.V. = 12.9848

 

40
Cell lengths tend to decrease with increasing bore
diameter size, in effect, keeping cell volumes somewhat
constant. The only exception was 6.0mm diameter trap nests,
but in these cells are no longer placed linearly in the trap

nest so that bore diameter is not equal to cell diameter.

CELL VOLUMES - Another way of looking at cell size is
to look at cell volume, since this considers cell diameter
as well as cell length. During 1984, cells containing male
H. ellipticus offspring (all sites and bore diameter sizes
combined for all nests) averaged 16.51mm3 (N - 181) while
those containing female H. ellipticus averaged 17.94mm3 (N =
77). During 1985 cells containing male H. ellipticus
offspring averaged 16.25mm3 (N a 551) while cells containing
female H. ellipticus offspring averaged 17.83mm3 (N = 227).
Cells from 4.5mm diameter trap nests (all nests from all
sites and both years) averaged 15.97mm3 (N = 584) for cells
containing male H. ellipticus offspring and 17.35mm3 (N =
230) for cells containing female H. ellipticus offspring.
Cells from 5.2mm diameter trap nests averaged 17.43mm3 (N =
129) for cells containing male H. ellipticus offspring and

19.08mm3 (N = 59) for cells containing female H. ellipticus

 

offspring. Cells from 6.0mm diameter trap nests averaged

19.55mm3 (N

19) for cells containing male H. ellipticus

 

offspring and 19.35mm3 (N = 15) for cells containing female

H. ellipticus offspring. Cell volumes for each site for

41
cells containing male H. ellipticus offspring (both years
and all bore diameter sizes combined) averaged 16.96mm3 (N =
78) at C5, 15.72am3 (N a 110) at CH, 16.19mm3 (N = 165) at
CL, 15.7lmm3 (N = 79) at F1, and 16.6mm3 (N a 300) at r2.
Cell volumes for each site for cell containing female H.
ellipticus offspring averaged 18.70mm3 (N a 45) at C5,
16.88mm3 (N = 34) at CH, 17.85mm3 (N - 48) at CL, 16.58mm3
(N = 41) at F1, and 18.05mm3 (N = 136) at F2.

Sex of offspring, site, bore diameter size, and cell
position each contributed significantly to the variability
in cell volumes, Table 5. As with cell lengths, separate
cell volume analyses were run for cells of male and female
offspring. For males, site, bore diameter size and cell
position were significant, Table 5. For females, site and
bore diameter size were significant, while cell position was
not, Table 5. Among 1985 nests, season of nest completion

was not significant with respect to cell volume, Table 5.

Cell volumes are affected by sex of offspring, site,
bore diameter size, and, for males only, cell position.
Cell position in the nest might be of greater consequence to
the volume of male cells because of the haplo-diploid sex
determination in combination with the tendency to place
female offspring in the inner cells and male offspring in

the outer cells. An unintended unfertilized egg in an inner

 

42

Table 5. Cell volume analyses for H. ellipticus (General

Linear Model)

 

1984 and 1985 data, both sexes.

 

 

 

 

SOURCE DF SS__ _ F PR > F
Sex 1 122.4455 17.92 0.0001***
Site 4 171.8624 6.29 0.0001***
Year 1 3.3513 0.49 0.4838
Site*year 4 33.9275 1.24 0.2915
Bore size 2 486.8853 35.64 0.0001***
Cell position 1 424.1207 62.09 0.0001***
Model 13 1686.1868 18.99 0.0000
Error 1022 6981.4344

r2 = 0.1945 C.V. = 15.6053

1984 and 1985 data, males only

SOURCE DF SS F PR > F
Site 4 113.7574 4.27 0.0020**
Year 1 8.1701 1.23 0.2684
Bore size 2 345.8542 25.97 0.0001***
Cell position 1 371.9474 55.85 0.0001***
Model 8 904.2604 16.97 0.0001
Error 723 4814.9163

r2 = 0.1581 C.V. = 15.8151

1984 and 1985 data, females only

SOURCE DF SS F PR > F
Site 4 113.6269 3.89 0.0043**
Year 1 16.6989 2.28 0.1318
Bore size 2 162.2312 11.09 0.0001***
Cell position 1 25.3356 3.46 0.0637
Model 8 328.1116 5.61 0.0001
Error 295 2157.0073

r2 = 0.1320 C.V. = 15.2031

1985 data, both sexes

SOURCE OF SS F PR > F
Sex 1 132.1339 20.35 0.0001***
Site 4 245.1261 9.44 0.0001***
Season 1 1.1234 0.17 0.6776
Bore size 2 250.7218 19.31 0.0001***
Cell position 1 188.7110 29.06 0.0001***
Model 9 1053.2997 48.02 0.0001
Error 550 3571.5270

r2 = 0.2277 C.V. = 15.5288

 

43
"female" cell might result in male offspring from an "extra-
large" cell. In addition to this, small females may be less
fit than small males as demonstrated in overwintering
mortalities among Qggig ligHaria (Tepedino and Torchio,
1982b). When resources are low or provisioning females old
(at end of the nesting period), it would be of more value to
produce only small male offspring (from small cells) than to
try and produce small female offspring. These two factors
could cause inner males to be from larger cells and outer
males to be from smaller cells, while females are produced
in cells of a more standard size despite their cell

position.

SEX PLACEMENT IN NESTS - Female offspring were
generally placed in the inner portion of nests (ie. the
first cells constructed), while male offspring were placed
in the outer portion of nests (ie. the last cells
constructed). Variations did occur, with some nests having
either all male or all female offspring. Also, an
occasional male offspring would be found in a series of
female cells, or visa versa. The secondary sex ratios
(male:female) for each cell position can be found by year in
Table 6. This occurrence of male offspring in outer cells
and female offspring in inner cells is common in many trap

nesting bees (Krombein, 1967; Frohlich and Tepedino, 1986).

44

Table 6. Secondary sex ratios (male:female) for each cell
position and total number of live H. ellipticus adult
offspring emerging from that cell position. (Cell 1 was the
first cell constructed in any nest).

 

 

 

 

YEAR
Cell 1984 __ 1985
Positiion H_ ‘gSR N SR
1 55 0.77:1 117 0.65:1
2 45 0.88:1 107 1.43:1
3 42 2.82:1 99 1.91:1
4 32 2.56:1 80 3.44:1
5 25 2.57:1 77 3.05:1
6 18 3.50:1 66 4.50:1
7 16 15.00:1 63 3.50:1
8 16 3.00:1 54 5.00:1
9 8 1.00:0 41 5.83:1
10 5 1.00:0 34 5.80:1
11 6 1.00:0 28 13.00:1
12 3 1.00:0 24 11.00:1
13 2 1.00:0 16 15.00:1
14 2 1.00:0 12 11.00:1
15 1 1.00:0 8 1.00:0
16 5 1.00:0
17 5 1.00:0
18 2 1.00:0
19 1 1.00:0

 

45

EMERGENCE - H. ellipticus was univoltine throughout
this study. Male offspring from 1984 began forced emergence
on June 1 and finished June 13, with a peak on June 8, 1985.
Female offspring from 1984 began forced emergence on June 1
and finished June 16 with a peak on June 10, Figure 10.
Male offspring from 1985 began forced emergence on June 6
(with the exception of 1 male on May 24) and finished on
June 18 with a peak on June 7, 1986. Female offspring from
1985 began forced emergence on June 7 and finished on June
16, with a peak on the first day of emergence June 7, 1986,
Figure 11. In 1984, male offspring clearly emerged before
female offspring. In 1985, peak emergence for the two sexes
were simultaneous, though female emergence did not decrease

as quickly as did male emergence.

OFFSPRING WEIGHTS - Male H. ellipticus offspring from
1985 weighed on average (all sites, bore diameter sizes and
seasons of nest completion) 2.30 i 0.63mg (range 1.01 to
4.84mg, N=59), while female H. ellipticus offspring from
1985 weighed 3.44 i 0.60mg (range 2.35 to 4.68mg, N=42).
Thus, on average female H. ellipticus offspring weighed 1.50
times more than male offspring. Male H. ellipticus
offspring weighed on average 2.29 i 0.49mg (N = 44) from
4.5mm diameter trap nests, 2.12 i 0.46mg (N = 10) from 5.2mm
diameter trap nests, and 4.84mg (N = 1) from 6.0mm trap

nests. Female H. ellipticus offspring weighed on average

46

 

 

 

w._.<o
9 9 9 9 9 9 9 9 9 9 9 9
Ir Ir Ir L Ir 0 0 0 0 o 0 7w
I. 9 .V 8 O 6 8 I— 9 .7 Ir .7
b - hn - n - n n - n - ru°
:2.
new
rem
m._<_2mn_ .9. .
w._<_2 .n. -8
rom

 

«we. .. CZEn—mumo ”—0 mozmwmmfim

.33 89. 33:3 .I 28.2 86 26:. .o 8.223 8896:.m .9 6.39...

9N|HdS=I=IO SNIOHEWS :IO HEEWRN

47

.mmmw ES. 3% fl 2662 86 26:. .o 8.628 8:8.oEm .: 6.39“.

m._.<n_
9 9 9 9 9 9 9 9 9 9 9 C.
I I I I l I l I I l l I
Ir Ir Ir Ir Ir Ir Ir 0 0 0 0 7w
8 I— 9 8 7w Ir 0 6 8 I. 9 .7
. ) - n - n - n - I - -

 

so

..ch

i cow

 

 

SNIHdedO ONIOHSWE :IO HSEWRN

I can
mmmw ._ GZEn—mun—O ...—O woszm—msfi

48
3.44 i 0.60mg (N = 33) from 4.5mm diameter trap nests, 3.41
i 0.76mg (N = 5) from 5.2mm diameter trap nests, and 3.68mg
(N = 1) from 6.0mm diameter trap nests. Male H. ellipticus
offspring from nests completed early season weighed on
average 2.36 i 0.46mg (N a 49) while female H. ellipticus
offspring averaged of 3.44 : 0.60mg (N = 40). Male H.
ell§pticus offspring from nests completed late season
weighed on average 1.94 : 0.99mg (N - 10) while female H.
ellipticus offspring averaged 3.23 : 0.30mg (N = 2). Male
H. ellipticus offspring weighed on average 2.21mg (N = 7) at
site C5, 2.34mg (N = 5) at CH, 2.38mg (N = 15) at CL, 2.32mg
(N = 12) at F1, and 2.26mg (N = 19) at F2. Female H.
ellipticus offspring weighed on average 3.47mg (N = 8) at
site C5, 4.07mg (N = 2) at CH, 3.48mg (N = 10) at CL, 3.30mg
(N = 9) at F1, and 3.37mg (N = 13) at F2.

Sex of offspring and bore diameter size contributed
significantly to the variability in offspring weight, while
site, season of nest completion, and year did not, Table 7.
Bore diameter size may actually not be a genuine factor
affecting the variability in offspring weight, because of

the low sample sizes at larger diameters.

SEASONAL PRODUCTION OF THE SEXES - Number of males

produced in nests completed early and late season 1985 was

49

Table 7. Dry weight analysis for 1985 H. ellipticus
offspring (General Linear Model).

 

1985 data, both sexes

 

souncgg DF _ ss __ F PR > F
Sex 1 29.0353 88.42 0.0001***
Site 4 0.6895 0.52 0.7177
Bore size 2 4.4015 6.70 0.0020**
Season 1 0.0929 0.28 0.5962
Model 8 34.7343 13.22 0.0001
Error 85 27.9118

r2 = 0.5544 C. . = 20.6437

 

50
141 and 284, respectively. Number of female offspring
produced in nests completed early and late season 1985 was
87 and 68 respectively. This has been represented in Figure
12 as percent investment by sex and season. In this figure
the total of each of the 4 sex-season values equals 100%.
The figure shows more offspring are produced in nests
completed late season, as compared to those completed early
season. More male offspring were produced than female
offspring for nests completed both early and late season.
Production of female offspring decreases slightly, while
production of males increases substantially in nests
completed late season as opposed to those completed early
season. The proportion of offspring produced late season is

more male biased than that produced early season.

SEX RATIOS - are summarized in Table 8. Secondary sex
ratio of 2.46:1 (male:female) produced in 1985 differed
significantly from the expected sex ratio of 1.50:1 (Xi =
45.6, p < 0.001). Secondary sex ratios from 1984 and 1985
did not differ significantly from each other (Xi = 0.21, p
> 0.5, df = 1). Individual sites varied with respect to
secondary sex ratio, but within each year were not
significantly different from each other (1984: X2 = 6.85, p
> 0.05, df = 4; 1985: x2 = 7.72, p > 0.05, df = 4). The
secondary sex ratio of CL during 1984 was the only site to

differ from the overall secondary sex ratio of that year,

51

.38 22a mam: 8:88 «2232.3 .3 .6. 8:29:00 «me: .o
888 .5 28an8 888-86 :86 .2 E8882 38.6.. .3 6.29“.

29.5.5200 ...mmz u—O 20m<wm

 

 

 

m5... >._:<m .
_ _ o w
m
I . 3
I0
.. om ......
w
....—
.N.
I 8 8
.A
....“
- 8 x
V
N
w._<_2mu_ .9. G
was... 6. - 8 m...
. V
m.
.. 8. N

xmw >m OZEQmLu—O ".0 ZOFODn—Omn. 4<ZOw<mm

52

Table 8. Sex ratios (male:female) of H. ellipticus by year.

 

 

YEAR
HEX RATIO 1984 1985
Expected na 1.50:1
Primary 2.00:1 2.47:1
Secondary 2.03:1 2.46:1
Secondary Sex Ratios by:
Site
C5 3.75:1 1.86:1
CH 2.64:1 3.48:1
CL 5.60:1 3.06:1
F1 1.93:1 1.85:1
F2 1.76:1 2.43:1
Bore size
4.5 mm 2.05:1 2.56:1
5.2 mm 1.64:1 2.19:1
6.0 mm na 2.48.1
Season
Early na 1.62:1

53

but sample sizes were low, 28 males to 5 females. Secondary
sex ratios among the different here diameter sizes were not
significantly different from each other (1984: Xi - 0.05, p
> .75, df = 1, (5.2mm and 6.0mm combined): 1985: x2 = 0.55,
p > 0.5, df = 2). Among 1985 nests the secondary sex ratio
for late season was significantly more male biased than for
early season (Xi = 25.00, p < 0.001, df = 1). The
secondary sex ratio for nests completed early season was not
significantly different than the overall expected value of

1.50:1 (x: - 0.32, p > 0.5, df - 1) while that of late
season was (Xi =25.0 , p < 0.01 , df = 1) .

Secondary sex ratios of 2.03:1 for 1984 and 2.46:1 for
1985 were not significantly different from their respective
primary sex ratio of 2.00:1 for 1984 (Xi - 0.02, p > 0.90)
and 2.55:1 for 1985 (x: a 0.0026, p > 0.975), thus showing

that overall mortality per year were not sex biased.

SEASONALITY - During 1985, trap nests having a larger
diameter became more acceptable in the absence of many of
the smaller diameter trap nests later in the summer and were
subsequently used at a higher rate during late season as
opposed to early season. It is generally felt that in late
season, pollen resources may decrease and a provisioning
bee's ability to construct cells efficiently is diminished

due to her age (Tepedino and Torchio, 1982a & 1982b;

54
Strickler, 1982). Under these conditions it becomes more
economical to produce the smaller sex, in this case male
offspring. Seasonal shift in sex production observed here
has been observed in other bees (Torchio and Tepedino, 1980:
Tepedino and Torchio, 1982a & 1982b; and Frohlich and
Tepedino, 1986).

In this study, the secondary sex ratio increased from
not significantly different from expected during early
season to significantly male biased during late season. The
overall (early and late season combined) secondary sex
ratio's male bias in 1985 can be attributed solely to nests
completed late season, since those completed early season
had a secondary sex ratio that was not significantly
different from expected. Thus, it appears as though the
provisioning females may be producing offspring at the
expected sex ratio early in the season when resources are
high and the provisioning females in peak condition. Then,
as resources diminish and the provisioning bees age, create
extra individuals of the less expensive sex (in this case

males).

The combination of larger diameter trap nests becoming
more acceptable later in the summer and the bees producing

more of the less costly sex (males) late in the summer

55
results in the larger diameter trap nests containing mostly

offspring of the smaller sex (ie. males).

MORTALITY AND PARASITISM - Live adult H. ellipticus
offspring were produced in 63.4% of 1984 primary cells and
70.9% of 1985 primary cells. Percent primary cells of H.
ellipticus lost to mortality factors and parasites, along
with the percentage of surviving male and female H.
ellipticus offspring for 1984 and 1985, can be found in
Figure 13. Figure 14 shows 1985 mortality factors and
parasites for nests completed during early and late season.

Mortality factors by site and year are shown in Figure 15.

As presented earlier, all mortality factors combined
were not found to be sex biased. Since, offspring were
force emerged before natural emergence could occur, it is
unknown how many offspring, both H. ellipticus or parasites
would have been destroyed by other, earlier emerging

individuals.

Non-parasitic mortality factors - and not due to
handling errors, accounted for 25.0% of all 1984 mortality
and 42.3% of all 1985 mortality which is equivalent to 9.2%
of 1984 primary cells and 12.7% of 1985 primary cells. Of
the 1984 non-parasitic mortality, the leading factor was

larval death at 44.7%, followed by egg absence/failure at

56

85.22 65:6 8::

«99.88 .m u o. 82.9... .. a 86:88.86 .. o .8488 ... 14.98898
... o .30 n o .m.o..o 8:86: .6 62:26.8 2 8o Lo:

99% .m .88 u x .833 .2... mass. .. 1.3.38.3 29.. a _2 .8.

.5 28: :6 .o. 8.: 02.8.3 :08 .2 2.8 .888 88.3 .9 6.39...
GZEnmu—mo

2
u.
E

      

   

    

   

x n. D < 0 0

     

\\\. § m m V
m w \ 2 ./.
A m 8 m
m ....
82 s w 8 m
82 I m .8
-8

 

m<m> >m ZOFODDOEQ OZEQwLuO

57

$5.29. 8:3 .92... 9:83 ..u. .. v. 625%.

u n. 62.8.58 u a .3334 u < 9389833 ... o .8890 u

0 .225 9.293.. 5 62:35.. 2 can 3: 9.3.3.3 fl cwmu .- x 9.38:3
.3 239 u u. «3.33 .3 22. .. .2 .38 9% $2. 8.6358 .8:

.o .833 .5 3.: 9553.0 :03 .o. 2.8 ESE... .o .3632 .3 9:9“.

GZEanEO

 

\

Illl/IIIIIIII\I

\
\

\
\

\
\

\
\

\
\

\

xxxLxxLL
\

 

 

\
\

\\\\
\\\

\
\

\.\

ES I
>55 3

$1133 AHVWIHd :IO HSEWDN

 

 

.. com
mwmw . ZOm<mm >m ZOFODQOm—n— GZEn—muu—O

58
MORTALITY BY SITE - 1984

IDEAD

E G. KIRBII
I c. HYLAEI
E] AIRRORATUS
fl DERMES‘HDS

SITE

 

 

' l ' l ' I
50 100 150 200

NUMBER OF PRIMARY CELLS

MORTALITY BY SITE - 1985

F2

   

F1

   

DEAD
G. KlRBll

C. HYLAEI

A IRRORATUS
DERMESTIDS
PHORIDS

CL

SITE

CH

END-Inna

C5

 

o 5 o 1 (30 150 2 (I 0
NUMBER OF PRIMARY CELLS

Figure 15. Number of primary cells lost to each mortality
factor or parasite by site and year. M = male H. M1315. F =
female H. ellipticus, X= dead H $01.03: not due to parasitism
or handling errors, G= Wm, C= 9.030033%“: A =

Anthrax, D= dermestids, P= phorids, K= fl. BJLDJEZLL: killed
during handling.

59
39.5%, pupal death at 13.2% and unsuccessful ecdysis at
2.6%. Of the 1985 non-parasitic mortality, the leading
factor was egg absence/failure at 70.0%, followed by larval
death at 15.4%, pupal death at 11.5%, and unsuccessful
pupation at 3.1%. The increase in egg absence/failure
during 1985 is at least partially due to the increase in the
number of incomplete nests obtained that year, since many
incomplete nests end with an incomplete cell, which would be

included in this category.

Parasites - accounted for 67.5% of 1984 mortality and
56.1% of 1985 mortality which equivalent to 24.8% of 1984
primary cells and 16.4% of 1985 primary cells.

Gasteruption kirbii kirbii (Westwood) accounted for
38.4% of 1984 g. ellipticus mortality and 27.7% of 1985 g.
ellipticus mortality. This is equivalent to 14.0% of 1984
primary cells and 7.7% of 1985 primary cells. Mortality by
this species alone was not sex biased. There was no
significant difference found for parasitism rates between
nests completed early and late season 1985. g. kirbii
kirbii restricted its parasitism to Hylaeus spp.; other
species nesting in the trap nests were not parasitized. All
nests g. ellipticus parasitized by this species were in

4.5mm diameter bores.

60

Unlike many cleptoparasites, the number of host cells
consumed by one Q. kirbii kirbii offspring varies. The
number of g. ellipticus cells consumed in 1985 nests by male
g. kirbii kirbii offspring averaged 1.89 i 0.68 host cells
(range 1 to 3, N=18). Female Q. kirbii kirbii offspring
averaged 2.43 i 1.34 host cells (range 1 to 5, N=23).
Immature Q. kirbii kirbii overwinter in the prepupal stage,
with no apparent cocoon. There is, however, a distinctive
reddish brown crusty, skin-like "pseudococoon" apparently
made of the parasites fecal material, but possibly including
some Gasteruption silk, as well. This is plastered to the
bore walls in addition to forming a chamber base and cap.
In instances where more than one cell was consumed, remnants
of Hylaeus cell linings, intercellular partitions and cell
caps are evident in a space at either the inner or outer end

of the "pseudococoon".

Pupation occurs in the spring and emergence coincides
with host emergence. Male offspring from the 1984 nesting
season began forced emergence June 1 and finished June 13,
peaking June 8, 1985. Female offspring from 1984 began
forced emergence June 8 and finished June 16, peaking June
15, 1985. Male offspring from 1985 began forced emergence
June 7 and finished June 13, peaking on the first day June
7, 1986. Female offspring began forced emergence June 8 and

finished June 18, peaking June 17, 1986. As with the g.

61
ellipticus male Q. kirbii kirbii offspring emerge before

female Q. kirbii kirbii offspring.

Dry weights of adult Q. kirbii kirbii offspring from
1985 g. ellipticus nests averaged 3.22 : 0.93mg (N=16, range
1.27 to 4.91) for male offspring and 3.76 i 1.20mg (N=20,
range 1.71 to 6.45) for female offspring. The wide range in
weight seems to reflect the range in number of host cells

consumed by the immatures.

Silk rings constructed by provisioning g. ellipticus

 

females may play a role in preventing or deterring at least

some parasitism by Gasteruption, since larger female

 

Gasteruption might not be able to fit through the small
central opening of the silk ring to reach cells under

construction. During 1985 field checks, Q. kirbii kirbii

 

females were noted entering and leaving trap nests as if
searching for suitable nests for oviposition. Oviposition

apparently occurs sometime when g. ellipticus is out of the

 

nest, although actual oviposition behavior is unknown
(Malyshev, 1968). It follows that female Gasteruptions must
be able to get into, oviposit and leave a nest in a
relatively short amount of time to limit detection by g.
ellipticus females. If they encounter a wall with an
opening that is too small for them to pass through easily,

they may not investigate further.

62

Table 9 shows the percentage of g. ellipticus nests
that were parasitized by Q. kirbii kirbii for each primary
cell number by the number of Q. kirbii kirbii offspring per
nest. To test whether a provisioning female 3. ellipticus
can detect parasitism by this species and cap the nest
earlier than would otherwise be done if the nest was not
parasitized a catmod analysis on the variability of number
of primary cells per nest was done using the presence of one
or more Q. kirbii kirbii offspring per nest and year.
Presence of a Q. kirbii kirbii offspring in the nests did
not contribute significantly to the variability in primary

cells per nest (p = 0.29).

Coelopencyrtus hylaei Burks (Encyrtidae) accounted for
19.5% of 1984 mortality and 14.9% of 1985 mortality. This
is equivalent to 7.1% of 1984 primary cells and 4.1% of 1985
primary cells. Mortality due to this species was not sex
biased. 0f the 18 nests parasitized by this species during
1985, 13 (72.2%) were completed early season, 2 (11.1%) were
completed late season and 3 (16.7%) were incomplete. Q.
hylaei restricted its parasitism to Hylaeus spp.; other

species nesting in the trap nests were not parasitized.

Numerous Coelopencyrtus hylaei larvae develop within

each host prepupae. Emergence was late relative to that of

63

Table 9. Percentage of g. ellipticus nests of each primary
cell number by the number o Gasteruption offspring, 1984
and 1985 data combined.

 

Primary Number

 

 

cell of Number of Gasteruption offspring per nest
number: nests: 0 1 2 3 4 5
0 2 100.0 - - - - -
1 40 77.5 22.5 - - - -
2 36 86.1 13.9 - - - -
3 25 80.0 4.0 8.0 8.0 - -
4 26 88.5 7.7 3.8 - - -
5 24 70.8 8.3 8.3 12.5 - -
6 13 92.3 7.7 - - - -
7 16 93.7 - - - 6.3 -
8 19 89.5 5.3 - - - 5.3
9 13 84.6 15.4 - - - -
10 15 93.3 6.7 - - - -
11 8 75.0 - 12.5 - 12.5 -
12 9 77.8 11.1 - 11.1 - -
13 10 100.0 - - - - -
l4 5 80.0 20.0 - - - -
15 4 100.0 - - - - -
16 1 100.0 - - - - -
17 2 100.0 - - - - -
18 1 100.0 - - - - -
19 1 100.0 - - - - -
20 1 100.0 - - - - -

 

64
g. ellipticus. Forced emergence from 1985 nests began June
23 and continued to July 03, 1986: all offspring from 1984
died prior to ecdyses. Q. hylaei were univoltine throughout
this study.

In this study ovipositing females apparently gain
access to cells by either entering the nest during cell
construction or by chewing through completed nest caps, as
indicated by tiny tell-tale holes at the perimeter of nest
caps of some parasitized nests. 0f the nests parasitized by
this species during 1985, 33.3% were judged to have been

entered after the nest cap was completed.

Burks (1958) gives an account by Krombein of
Coelopencyrtus hylaei adult female in-nest behavior and
larval development. He states that an adult female was
observed in a nest (thought to be g. modestus) near the nest
entrance. She waited in the nest until the Hylaeus larvae
consumed their provisions. Once this was complete and host
larvae were mature the Q. hylaei female parasitized all

cells in succession and was found dead in the nest.

In this study of g. ellipticus, nests containing Q.
hylaei parasitism tended to have all cells parasitized.
Only three things appeared to prevent the "spread" of Q.

hylaei parasitism through a series of cells: an undigested

65
provision, an intercellular space (the length of a cell or
more), or a intercellular partition of woodscrapings at
least 1.80mm thick. Few g. ellipticus nests observed in
this study contained a dead adult female as described by
Burks (1958), so apparently, at least some ovipositing
females are able to leave the g. ellipticus nest and

potentially seek out other nests to parasitize.

Parasitism methods of Coelopencyrtus hylaei may help to
explain their relatively late emergence. It would be
impossible for a Q. hylaei female to travel through a series
of Hylaeus cells while provision remained undigested. The
pollen-nectar mixture would act as a sticky trap.

Therefore, they must wait to parasitize until provisions are
consumed. A safe way to wait would be to emerge slightly

later than the host species, as long as the nest is entirely
parasitized. If the nest were not entirely parasitized, the

Q. hylaei offspring might be destroyed by emerging

unparasitized host offspring.

The presence of one or more intercellular spaces in a
nest might prove to be beneficial for the survival of some
g. ellipticus offspring in the event of parasitism by Q.
hylaei. Intercellular spaces may prove adaptive in that if
nests are parasitized, only outer cells are lost, while

inner cells are preserved. Furthermore, in the process of

66
emergence by inner unparasitized g. ellipticus, the later
emerging Q. hylaei offspring in the outer cells may be
destroyed, causing a further reduction of parasites. If,
however, no Q. hylaei attack the nest, all g. ellipticus
offspring survive. Intercellular spaces should not reduce
the total space used for reproductive cells since nests were
rarely entirely filled with cells. Intercellular spaces
would require less time or energy than constructing a long
intercellular partition. The intercellular spaces found in
these nests could be considered analogous to empty cells
constructed by some mud nesting wasps. Tepedino, et a1

(1979) found these empty cells could lower parasitism.

Of all parasites found for g. ellipticus in this study,
Coelopencyrtus appears to have the greatest potential for
sex biased parasitism, since its method of parasitizing
involves entering nests through the nest entrance and
parasitizing cells in succession until they reach a
blockage. Since male 5. ellipticus offspring are generally
located in the outer portion of the nests there may be a
greater potential for a disproportionate number of male 3.
ellipticus offspring to be lost, assuming some nests are not
completely parasitized. However, in this study, male and

female offspring were parasitized proportionately.

67

Anthrax irroratus irroratus Say (Bombyliidae) accounted
for 8.2% of 1984 mortality and 11.4% of 1985 mortality.
This is equivalent to 3.0% of 1984 primary cells and 3.1% of
1985 primary cells. A. irroratus irroratus parasitized 1985
nests completed during early and late season proportionally.
This species was a successful parasite of species other than
Hylaeus spp. that nested in the trap nests including:
Megachile relativa, g. inermis Provanchier, and gumegachile
puggata (Say), so it was not exclusively dependent on
Hylaeus for its survival.

Ovipositing females lay their eggs while on the wing by
shooting them into the bores, as observed during field
checks for new nests and as described by Marston (1970).
Beginning the spring after oviposition, each A. irroratus
irroratus larvae consumed either one Q. ellipticus or one
Q. kirbii kirbii prepupae. During feeding, pupation of the
host prepupae was prevented. A. irroratus irroratus
emergence was quite late with respect to that of g.
ellipticus. Forced emergence from 1985 nests began July 06
and finishing July 18, peaking on July 12, 1986. Offspring
from 1984 all died prior to ecdyses except for two offspring
which were force emerged July 13 and 14, 1985. This
emergence is not well timed with g. ellipticus. The larvae
and pupae of this parasite are very active and pliable and

it may be possible for at least some immature A. irroratus

 

4—0--. —.¢—-r:-»—_-

68
irroratus to escape destruction from emerging g. ellipticus

or Q. kirbii kirbii.

Dermestids accounted for only 1.9% of all mortality
among 1984 g. ellipticus offspring and 4.1% of all mortality
among 1985 g. ellipticus offspring. This is equivalent to
0.7% of primary cells constructed during 1984 and 1.1% of
primary cells constructed in 1985. One individual dermestid
caused the destruction of many 3. ellipticus cells. Of the
dermestids found in these nests only one lived to adulthood.

Identification is pending.

Phorids did not occur in 1984 nests and accounted for
only 1.2% of all mortality among 1985 g. ellipticus
offspring which is equivalent to 0.3% of primary cells
constructed in 1985. They occurred in the outermost portion
of only two nests. All 1985 offspring emerged during the
summer of 1985, before nests were collected for storage. As
a result, no adult specimens could be retained.
Identifications were based on pupal cases found in the

nests.

69
SUMMARY

Over a two year study, 271 g. ellipticus nests were
collected by trap nesting at five study sites in Michigan's
Upper Peninsula. Methods employed for collecting nests
allowed for analysis of three nest site (trap nest)
selection factors: bore diameter size, nest height and nest
entrance orientation. Bore diameter size was the only nest
site selection factor found to result in a distribution

significantly different from equal.

Nest architecture consisted of a linear series if silk-
lined cells in the 4.5mm and 5.2mm bores. In the 6.0mm
diameter bores the cells were placed in the nest so that
they overlapped. Nests within all bore sizes usually had
some cells separated by intercellular partitions of wood
fibers scraped from the inner walls of the nesting bore.
These fibers were usually between 0.5mm and 1.5mm in length.
Nest caps were composed of wax-papery silken walls which
average 3 nest cap walls per nest. The outermost nest cap
wall ranged from almost flush with the bore entrance to an
average of 6mm to 8mm recessed, depending on year. When
nest caps were almost flush, they were attached to the inner
walls of the trap nest. This in combination with the color
and character of the silk made species identification

possible for complete nests when no g. ellipticus offspring

 

survived to adulthood.

70
Pollen used for provisions was primarily Rosaceous.
Minor components included Apiaceae, Ranunculaceae and
Asteraceae and were used only when Rosaceous pollen was

scarce .

The number of primary cells per nest, for complete
nests, averaged 4.7 in 1984 and 7.2 in 1985. This
difference was significant, and most likely due to resource

differences between years.

Cell size was expected to vary with respect to several
variables. Cell length was significantly affected by sex of
offspring, bore diameter size and cell position. In
addition, site and year also had a significant affect on
length of female cells. Cell volume was significantly
affected by sex of offspring, site, and bore diameter size.
In addition, cell position was also significant for male

offspring only.

Female offspring were generally placed in inner cells
(those constructed first), while male offspring were
generally placed in outer cells. Male offspring tended to
emerge at least slightly earlier than female offspring
although peak emergence during 1985 occurred on the same day
for the two sexes. The majority of female offspring

produced were from nests completed early season, while the

71
majority of males produced were from nests completed late
season. The sex produced in greater numbers during early

and late season, however, was male offspring.

Female offspring dry weights were significantly greater
than male offspring dry weights. The expected sex ratio,
was therefore 1.5:1 (males:females). The secondary sex
ratios for 1985 (2.46:1) was significantly male biased when
compared to the expected value. The secondary sex ratios
for 1984 (2.03:1) and 1985 (2.46:1) were not significantly
different from each other. Although the secondary sex ratio
varied between study sites they did not vary significantly
among themselves; nor did those of the different bore
diameter sizes. Season of nest completion did affect the
secondary sex ratio during 1985. Secondary sex ratio from
early season completed nests (1.62:1) were significantly
different from that of late season completed nests (4.18:1),
but were not significantly different from the expected value
for 1985. Primary sex ratios were not significantly
different from the secondary sex ratios for either year,

thus showing mortality was not sex biased.

Parasites of g. ellipticus in this study included:
Gasteruption kirbii kirbii, Coelopencyrtus hylaei, Anthrax
irroratus irroratus, a dermestid and phorid. Q. kirbii

kirbii parasitized 14.0% of 1984 primary cells and 7.7% of

72
1985 primary cells. They often consumed more than one 5.
ellipticus cell per parasite offspring. Male Q. kirbii
kirbii offspring averaged 1.9 g. ellipticus primary cells
while female Q. kirbii kirbii offspring averaged 2.43 g.
ellipticus primary cells. Before pupation the Q. kirbii
kirbii prepupae forms a chamber (pseudococoon) of fecal
material and possibly silk. Emergence of this species was
basically simultaneous with that of g. ellipticus. Silk
rings placed in the nest by the provisioning g. ellipticus
may help to deter parasitism by this species by periodically
constricting the diameter of the bore, and thus potentially
making it difficult for (at least the larger) Q. kirbii

kirbii females to gain access to cells being provisioned.

Coelopencyrtus hylaei accounted for 7.1% of 1984
primary cells and 4.1% of 1985 primary cells. They would
enter the nest through the nest entrance while the nest was
being constructed or chew a small hole in the nest cap.
Usually the entire nest would be parasitized unless a
blockage prevented the ovipositing Q. hylaei to travel
between cells. Such a blockage consisted of either an
intercellular space (longer than a cell in length), an

intercellular partition of 1.8mm or more, or a g. ellipticus

 

cell in which the offspring did not develop and the
provisions remained undigested. Emergence of this species

was not well timed with that of g. ellipticus, occurring

 

73
several weeks later. The delayed emergence is not a problem
in nests that have been entirely parasitized, and may act as
a safe refuge for the Q. hylaei until the newly provisioned
g. ellipticus nests are at the proper stage for parasitism.

Anthrax irroratus irroratus accounted for 3.0% of 1984
primary cells and 3.1% of 1985 primary cells. The larvae of
this species consumed one prepupae of either g. ellipticus
or Q. kirbii kirbii. Emergence was delayed as compared with
g. ellipticus, occurring one month later. This species was
a successful parasite of several other species using trap
nests, and as such its survival did not depend on Hylaeus

parasitism exclusively, as did the Q. kirbii kirbii and Q.

hylaei.

74

GLOSSARY

BASAL SPACE - a space between the innermost end of the trap
nest and the first cell constructed in the nest.

CELL - an area lined with silk and provisioned in which one
Q. ellipticus offspring will develop.

CELL CAP - a wall of silk used to seal the outermost end of
the cell once it is complete.

CELL POSITION - position of a secondary cell in a nest as
numbered in sequence from the innermost (first cell
constructed) to the outermost (last cell constructed).

COMPLETE NEST - any nest in which a nest cap is present.
EARLY SEASON 1985 - before July 22, 1985.
EXPECTED SEX RATIO - ratio of male to female offspring such

that number of males x investment per male offspring =
number of females x investment per female offspring.

FORCED EMERGENCE - removal of winged adult offspring from
their cells before natural emergence could occur.

INCOMPLETE NEST - a nest in which a nest cap was never
observed and in which only 3. ellipticus nested.

INTERCELLULAR PARTITION - packed wood scrapings located
between two cells.

INTERCELLULAR SPACE - a space located between two cells.

INTERCELLULAR WALL - a silk wall partition located between
two intercellular spaces.

LATE SEASON 1985 - after July 22, 1985.

NEST CAP WALLS - silk walls, perpendicular to the bore, that
are between the final cell of a nest and the entrance to the
trap nest.

NEST CAP - all of the nest cap walls combined.

OUTERMOST NEST CAP WALL - outermost of all the nest cap
walls.

75

PRIMARY CELL NUMBER - an estimate of number of H. ellipticus
cells originally constructed in any given nest.

PRIMARY SEX RATIO - an estimate of sex ratio had no
mortality occurred among immature g. ellipticus offspring.

RAIDED NEST - a nest observed as complete, but whose nest
cap and possibly cells were removed by another species.

SECONDARY CELL NUMBER - number of cells that were observed
after parasitism.

SECONDARY SEX RATIO - ratio of male to female offspring that
emerged as live adults.

SILK RING - a ring of silk (wall with a central hole) placed
perpendicular to the bore, between the nest entrance and the
cell being constructed.

TERMINAL CELL PLUG - packed wood scrapings immediately
following the final cell in the nest.

USURPED NEST - a nest that was never observed as complete
and has been taken over by another species.

VESTIBULAR SPACE - a space between the final cell of the
nest and the entrance to the trap nest.

APPENDIX 1

76
APPENDIle

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.: 1939-01

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

Nesting Biology of Hylaeus ellipticus (Kirby)

(Colletidae:Apoidea) in Northern Michigan

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

Entomology Museum, Michigan State University (MSU)

Other Museums:

Investigator's Name (5) (typed)

 

___Yirginia_Lnnise.Scott

 

 

Date April 29 1939

*Reference: Yoshimoto, C. M. 1978. Voucher Specimens for Entomology in
Nerth 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.

77
APPENDIX 1.1
VOucher Specimen Data

9 Pages

of

..l._

Page

mama

r .wwumuso

 

8mm: .om 93na¢ mum:

 

 

r. s .,..wamaasxxmw
q. .0 829......

zuwmuo>fi== oumum cowacowz mzu c: uwmoamo

 

 

90u mcosaoodm ovum“: m>onm mzu pm>awumm

 

Adiawafl

.oz nocoso>

1:86:63

quUm umwzaé awanmufi>

Amvmemz m.uoumw«umm>cw

Ahummmoooc u: mumocm Hmcofiuwocm mmzv

 

am "umoz
uuoom A.ouswwuu>

 

msofiumwaao mammazz

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

am: 9684 «am sow: zmse .66 666639696 "a: Aznusxc
Ah “umwz
uuoom a aficfiwufi>
am: gnu: «Hm zaNm zmse .66 666699696 .92 193039. mauswmsfifim mammfiam
us .6662
yucca a cflafiwufi>
am: am «Hm zm~m,zmsa .66 666693696 ”A: Amnufixv m=UfiumeHm mammfim:
4 em .6682
aboum a asaawua>
am: new: 83m 36mm zmse .66 6668.963: "A: assuage maofiuaasfim mammfi:=
no “6682
“boom A uscamuw>
am: An»: saw some zmse .66 6686.369: .8: Amaufisc macauasflfim mammmwr
x< “umoz
buoom a macamua>
amz. am «am 29mm zasa .oo couH “a: “mousse «26.683938 m:mm~::
Juo+ vouamoamo new mom: uo omuuoHHoo coxmu umzuo no mofiowdm
. s s s e acme omdm no mum w m
maramnanum
no 933852

mum" om Hanna mama

 

 

 

mama I. owwuso
«nuwm. _/. mw<vagummmwitu

Esmmsz xonOEoucm
huwmuo>wc= mumum cowasowz mnu cw uwmoawp

 

uuoum mmfimqq macamuw>

78

APPENDIX 1.1
Vbucher Specimen Data

Pages

._2_.

Page 2 of

you msmswomam swam“: m>onm may oo>wmomm

 

Haiamma .oz umzoso> Ammahuv

A>ummmwooc u: mummzm Hmcowuwoom wmav

Amvmsmz n.90umwfiumm>cw

 

sz

am:

am:

am:

N

m "HHoo an “new:
com: .5 cash "powuoam
buoom a aaaamus>
«Hm 3m~¢ zmqe .oo nomcaxoao "a:
m "Hamo Am "ummz
mam: .m mash “vowuosm
uuoom a aficfiwuas
«Hm saw“ amee .oo somcaxown “w:
H: "Haou no ”umoz
mama .n mash “wowuoam
uuoum A aficfimufi>
mam Zena Zneh .oo somcaxoaa "a:
n "Haoo no "ummz
cum: .5 mash "powuoam
aboum A macawufi>
mam 36mm ZMSH .oo comcfixoan "a:
H "Haou M< ”umoz
emmH .5 each "powuoam
buoum a aquam69>
«Hm 3am: ZNqH .oo couH "ax

 

Aznufixv msoaumwaao momma»:

 

Aznuaxv msoaumwaam mammamm

 

Ambufixv msoaumfiaam mammamm

 

Ambuaxv msoaumuaao moomHNw

 

Ahnufixv msoaumwfiam msomHm:

 

Museum
where

depos-
ited

 

Other

 

Adults 0'

 

 

 

 

Adults 9
Pupae

Larvae

 

Eggs

omufimoaov pom pom: mo omuomHHoo
msoE:omam pom mumc Hmnmg

 

coxmu umzuo no mofiowdm

 

 

“mo nonssz

 

 

79

APPENDIX 1.1
VOucher Specimen Data

Pages

 

of

_§_.

Page

Snwmno>ac= oumum cmwAcofiz wan cu uAmodoo

mum:
meuw

\

~
I

nmmmwso

 

mama .o~ Anna<

mama

 

.,. 39

 

sums: xwvoEoucm

 

now msmsAomam ponmAA m>onm ozu oo>Aoowm

 

AcimwoA .oz nososo>

Aomaznv

uuoom omAsOA chAwnA>
Amvmsmz m.noummAnmo>cH

Azncwmmom: MA madman AmcoAnAmpm omav

 

am:

am:

am:

am:

N “Aaoo M< unmoz
omoa .AA ossn "vownoam
unoom A mucamna>
«Am 3amm Zach .oo conH "A:
m "AAoo om “ummz
cwmA .oA mash onomon "vownmam
uuoom A mAusna>
mam sang ZMQH .oo somsAonn “A:
m "Hamu UM numoz
cam: .0: 06:6 onomon "vmwnmam
6.66m a ananmnn>
mAm 30mm ZMSH .oo sochonn "A:
AA “AAoo Ma unmoz
emmA .5 was: ”newnmam
unoum A macawna>
«Hm samm qua .00 somcwxown “A:
e "AAoo MB "ummz
cam: .n mash "mownoam
uuoom A mAcAwna>
«Am sown 2mea .ou concaxoaa "A:

 

Asannxv maunuawnnm

mammAmz

 

Azanfiev maofiumnfinm

mommAhz

 

AAEnnev magnuawnnm

mammAmm

 

Ahnnnx. muonuannfim

msomAmm

 

Anonnsc maunuanflnm

mammAmm

 

Museum
where

depos-
ited

 

Other

Adults Cf >4
Adultsrg

 

 

 

Pupae
Nymphs

 

 

Larvae
Eggs

 

omuAmoaoo can own: no omnquAoo
mcwsAomam now mnmo AobmA

 

COXNU Hwfiu O .HO

mvomam

 

 

"mo nobssz

 

 

80

APPENDIX 1.1
Vbucher Specimen Data

of 2 Pages

-4—

Page

mum:

 

aman .oN Annm«I mum:

 

 

xuwmnm>wcs mumum :mwaoAz man :« namedmo

FQRMIM. 333W...

 

sons: amvoEoncm

 

now mcmEAomdm mono“: m>oam ozu om>Amomm

 

Helmmofl .Oz nonoso>

Aswanuc

unoom omA30A mficAwnA>
Amvmemz m.n0umwfiumw>cH

A>nmmmmomc MA muomnm AmcoAUAoom mmav

 

am:

sz

sz

am:

A "Ange on .66602
baa: .o:,ocsh onomop ”cownmam
buoum A «Acnwnn>
mam some zmsn .oo somanxuna “n:
o “AAoo MH ”ummz
baa: .n mash "townoaw
“boom 3 ananwnn>
«Hm sang zmsn .oo comanxonn "A:
N unnou en ”nmmz
baa: .n ouch “momnoam
unoum a ananmnn>
«Hm sans zmsn .oo comanxuna "A:
N unnoo an ”ummz
swan .nn can: .emmnoam
Locum a uncnwnn>
«Hm sans zmsn .oo comaaxunn "n:
m .nnuu x< “umoz
swan .nn maze “sewage:
“boom 4 «Aaawnn>
snm 3Hmm zusn .ou noun "A:

 

Amnnfixv msounmwAAo mammam:

 

Amnnfixv msOAumwAAm mammAmm

 

Aznnwxv msoAHNWAAo mammAmw

 

Ahnnaxv msoaumwAAo momma»:

 

AhanAMv msofiumAAAo osmMAmm

 

Museum
where

depos-
ited

 

Other

 

 

 

 

 

Adults 0'
Pupae
Nymphs

Larvae

 

Eggs

omnwmonoo was com: no omnquAoo
mcmEAomdm now mumo AmbmA

 

coxmu nmzno no monwam

 

 

“5 Adults 9 ...,

nonEsz

 

 

81
APPENDIX 1.1

Voucher Specimen Data

5 of 9 Pages

Page

5nAmnm>Ac= onmum cmmwzowz man :A uAmodmc

mum:

k

3.5. ,. 3

.925

 

awofi .om Anna< mama

 

 

.Esmmsz wwo oucm

 

now mamEAooam ponmAA m>onm ozu om>Aoo~m
HeimmmA .oz nozoso> Ammazuv AmeEmz m.n0umwAnmm>:H

Aanwmmmoo: mu mucosa AmcoAuAmmm omav

 

unoom omasoA mAcmMnA>

 

 

Ahnnaxv msownaAHAo msmuAhm undo:
A "AAmu Hz uuuoz
baa: .5A ossn omwnoam
uuoom A naenwnn>

 

 

 

Wcooaumozv «Annex
AAAnAx cownnsnmummo

 

 

 

 

 

 

 

 

 

 

 

 

2m: x «Hm 3m~m 2meh .oo concuxoan "A:
Ahnnfimv msoAnaAAAo mammahm "umom
A "AAoo QA uumoz .
was: .mn «use .eownoam .cooaumozv Annnns
am: x «Hm 36mm 2mm: .60 aomcmxuno "A: «Anna: connasnmnmmo
Amnnwxv msoAnAAAAo msooahm "noon
N "AAoo am "nmoz
ommA .5 0:96 "cownoam
uuoum a ananmnn> Aeooaumozv nAannx
3m: x «Hm sown ZmNH .oo somcaxoun "A: AAnnAx coAuasnmummo
Ahnnuxv ozoAuaAAAm msomAhm "umo:
A "AAmu em “umoz
0mm: .5 «can "vownosm
“boom 3 ananmnn> Aeooaumozv nnpnnx
mm: N «Am 3¢Nm zmee .oo somcfixowa "Ax «Annfix sownasnoummu
Juo+ vwnAmoaoc can poms no omnooAAoo coxmu nonuo no monon
m e r r m m e .m & mcmsuomnm nOm pump AobmA
“awamaummflm
mwaumnnmuma .

 

 

"mo nmnEsz

 

 

82
APPENDIX 1.1
Vbucher Specimen Data

Page 5 of 2 Pages

mama
d N. do Gig

zuumum>wcs mumum :mwwzoqz may 6% ufimoamv

ruoumuso

 

mmafi .cm ngm« mama

 

 

.anmsz %on050u=m

 

no» mcofiwomam mmumwa m>onm on» vm>wmoom

 

Helmmma .oz Mucuso>

uuoom mmfizoa macfiwufi>

Amma>uv Amvamz m.uoumwfiumm>cm

Auummmwom: «a mummnm Hmcowuammm mmsv

 

am:

am:

am:

am:

N

 

Ahauaxv naofiuawaao mammHN: uuoom
H "Have an “umoz
omsfi .HH «can ”cowuoam
uuoom a macawufi>
mam 3bmm ZmQH woo somcaxofin "Hz
Ahnuaxv oaoaumwaao mammamm "umo:
m ”Haoo um "umoz
ommfi .oH «can "umwumam
uuoom a macamufi>
mam 30mm zmce .oo cooaaxoao "a:
khnuaxv ozoaumafiam momma»: "umom
m ufifimu 3H "umoz
emafi .BH mesa "umwumam
uuoum a macawufi>
«Hm Smwm ZmQH .oo somafixoaa “a:
fizpuaxv maowunfiaam maomfimm "umo:
H ”HHoo Am "umoz
ommfl .NH mans "umwuoam
uuoum a aficamuas
«Hm smug ZmQH .ou comaaxowo “a:

 

hcooaummzv afinuwx
awnufix cowumzumummo

“moozumozv Hanuwx
Hanuax cofiumsuoummu

fiwooaummzv Hanuwx
“annax cowumduoummo

mooaumo3v «anufix
Hapuax doaumwuoummo

 

Museum
where

depos-
ited

 

Other

 

 

 

 

 

Adults 0'
Pupae
Nymphs

Larvae
Eggs

 

kuamoawv mam mom: no mouooaaoo
mamanoam pom mumm Huang

:oxmu uwzuo no mmwomam

 

 

 

“5 Adults 9

umaesz

 

 

 

omme+o~ Hnno¢ mama

 

 

mama b n0um so
[QQSOs/A 3%

.Esmmaz zonOEOucm

 

 

 

>uHmno>Hca wumum :me30H: mzu :H nHmoamm ‘luuoum‘uwfiamq «HannH>
now maweHuoam vmumHH m>onm wcn vo>Hmuwx
nllllllHdflddfiHl .oz nozoso> Ammaauv AmeEmz m.noumeumm>:H

Aznmmmwom: mH muwmzm HmcoHuHmmm omav

 

of 2 Pages

83
APPENDIX 1.1
Voucher Specimen Data

7

Page

 

A%nonv mooHanHHm maomHhm,uumo=
H uHHoo ox unmoz
ommH an mash “momnmam

 

 

 

 

 

 

 

 

 

 

nuoom a ananwnn> “woosummzv finnnnx
x «Hm 3nmm zmen .oo aonn “n: «nannn conuaanaummo
4u0¢ monHmoaom new now: no mwuooHHoo :oxmu nozuo no mmHowam
m e r r m m e .m % mcmEHumam now mnmv Hman
e r o.d e .1 .1 a p. w s
figmnmwmam
"M w A_4i nu .A .A P. N" 1“ nu t

 

"mo nonficz

 

 

 

84
APPENDIX 1.1
voucher Specimen Data

Page 8 of 9 Pages

mum:

nODmnsu

 

swan .CN nnna< mama

 

 

qumnm>Hcs mnmum :mecoH: onn :H nHmoamm

1x. :1 31W»

 

.Esmmaz hmoHOEonam

 

now mcwEHomam mwumHH m>oam as» mo>Hmum¢

 

chommH‘ .oz nmsozo>

flan“ Mn “:0; Qfiflnmhfi>

AmeAuv Amvmsmz m.noumeumm>cH

Haanmoom: HH mummcm HmcoHquvm mmav

 

am:

am:

am:

am:

mum

muiau+ x
332 x

m: anvf N

 

Ahanxv msoHuQHHHo mommHm: "woo:
mH uHHmu no uumoz
omsn .m Anna "eumnoem
unoum a ancnwnn>
mHm sand ZmQH .oo aomconHn "Hz
Ahnonv msoHnaHHHm mamlom “noon
0 uHHoo x< “umoz
Swan .m anus "umwnmam
uuoum A ananwnn>
«Hm 3Hmm chs .ou :onH "Hz
Ahnonv mnoHuQWHHo mammHmm “poo:
H ”HHmo mcH unmmz
mmmn .om mesa "nmwnmam
unoum a onannmn>
«Hm zomm ZMQH .ou nomstoHa "Hz
Ahnonv maoHumHHHo msoonw “noon
c uHHoo oHo uumoz
mmsn .oN mama "nmmnuam
uuoum A ananmnn>
mHm Zomm ZmQH .oo comaonHn "Hz

 

 

 

 

mxnsm Hmmth mounhocoaoHooo

 

mxnsm HmmHNn maunhucmmonoo

 

mxnsm HmmHkfi mannhocmmonoo

 

mxnsm HomHm: mannhoamnonoo

 

Museum
where

depos-
ited

 

Other

 

Adults 0v

 

 

 

 

Adults 9

8
NW
N

Pupae

Larvae
Eggs

 

canHmoawm mam mom: no vmuooHHOo
mamEHomam now mumm Hman

 

coxmu nozno no mmHooam

 

 

”mo noasaz

 

 

85
APPENDIX 1.1
voucher Specimen Data

9 of 9 Pages

Page

mama

neumnau

 

mmmn .om nnna< mum:

 

 

1 any. is...“
flV, .Esmmsz hono ucm

anmnm>Hca mumum :me30H: wan :H UHmoamm

 

 

no“ mamEHooam vwanH m>onm o5» mo>Hmoom

 

HoummaH .oz nm£o=o> Ammaxuv

unoom omHnoH chHwnH>

Amvmsmz m.noumeumm>cH

Aanmmmooo: uH mummzm HmcoHuHmvm mmzv

 

 

AhanHMv maoHumHHHo mamMth "woo:
H uHHou ma ”ummz
swan .nn mean "emwnmam

uuoum a ananmnn>

 

AmmcHEnmnomcav mmmHnmmanma

 

 

 

 

 

 

 

 

 

 

am: uHav< «Hm 3Hmm ZNQH .oo comaonHa “Hz
AhnnHMv maoHumHHHm maomHmm‘uumo:
q uHHoo Hm uumoz
emmH .m AHDH "umwnmam
nuoom A anHwnH> how moumnonnH
am: nH=v< «Hm SaNm che .00 comawaHo "Hz maumnonnH xmnSnc<
moo+ vmnnmoamm mam mom: no monomHHoo coxmu nwcno no mmHooam
m e r r m m e .m % mamEHomom now want Hoan
e r 0.0 e .1 .1 a D. w s
wmwmmmmwmaw
“M w.d.1 .u .A .n p. n" r“ w“ t

 

 

"mo nunszz

 

 

APPENDIX 2

86

List of herbaceous angiosperms excluding the Poaceae and Cyperaceae
1985'

Not at site
Rare

uncommon

>OCFU

Abundant

SPECIES
.Achillea.udllefolbun
.Actea rubrum
AenfinmnxLa striata
Anadandhier
.Amelandhier Bartramiana
.Anaphalis nmmgamitacea
Anenome canadensis
Anenome quinquefolia
Anenome‘virginimma
Aquilegia canadensis
Arabis glabra
.Arctiumlndnms
Asarum.canadensis
Aster cordifolius
Aster lateriflorus
Aster‘macrophyllus
Aster puniceus

Aster simplex

Aster umbellatus
Barbarea vulgaris
Berteroa incana
Bidens ????
Caltha.pa1ustris
Campanula rotundifolia
Centaurea unculosa

Ceras tium vulgatun

87
w
Asteraceae
Ranunculaceae
Rosaceae
Rosaceae
Rosaceae
Asteraceae
Ranunculaceae
Ranunculaceae
Ranunculaceae
Ranunculaceae
Brassicaceae
Asteraceae
AristOIOChiaceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteracece
Brassicaceae
Brassicaceae
Asteraceae
Ranunculaceae
Campanulaceae

Asteraceae

Caryophyllaceae

IWWWO

CO

CCWFUCFUOC‘.

C125

CG>O$UFUC=W>O

FUDGE:

’3!

>757!

C/U

SPECIES
Chrysanthemumihaxxnuimmtnn
Cirsium.arvense
Cirsiun palus tre
Cirsiun nuticun
Cirsiun vulgare
Claytonia caroliniana
Clematis virginiana
Clintonia borialis
Convolvulus spithamaeus
Coptis trifloiuu

Cornus canadensis
Cornus stolonifera
Coronilla varia
Corylus cornuta

Daucus carota
Diervilla lonicera
Epilobium angustifoliun
Epilobiunl

Erigeron axmuus/s'trigosus
Erigeron canadensis
Erigeron philadelphicus
Eupatorium maculatum
Eupatorium perfoliatun
Eupatorium.rugosum
Fragaria vesca

Fragaria virginica

8 8
FAMILX

 

Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Portulacaceae
Rzmmculaceae
Liliaceae
Convolvulaceae
Ranunculaceae
Cornaceae
Cornaceae
Fabaceae
Betulaceae
Apiaceae
Caprifoliaceae
Onagraceae
Onagraceae

Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Rosaceae

Rosaceae

”CW7!

0

CW”

- U
C/A R
C A
U U
- R
U/C -

R

R
- R
U R
- R
U U
R R
U U
U C
R ..
C _
U -
R ?
U C

SPECIES

Galeopsis tetrahit
Galliunt
Gaulthemiilynxxnxdxaus
CkanIaleppicmnl
Helenium.autumna1e
Hepatica.anemicana
Heracleum.lanahun
Hieraciumtauranticum
Hieracium canadense/scabruv
Hieracium.canadense
Hieracium f lorentinun
Hieraciun pratense
Hypericumlperforatum
Hypericumlpynmnkknnxn
Iris

Krigia‘biflora
Ledumtgroenlandicum
Linaria vulgaris
Lonicera tatarica
Lotus corniculatus
Lychnis alba
Lysmachia ciliata
Tiaianthemum.canadense
'Mbdicago lupulina
Mbdicago sativa

Meli lotus alba

8 9
FAMILY

Iandaceae
Rubiaceae

 

Ericaceae
Rosaceae
Apiaceae
Rammculaoeae
Apiaceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Hypericaceae
Hypericaceae
Iridaceae
Asteraceae
Ericaceae
Scrophularia
Caprifoliaceae
Fabaceae
Caryophyllaceae
Prinulaceae
Liliaceae
Fabaceae
Fabaceae

Fabaceae

CSCLCHFlFZ

_ 2 9
U U U
U .. -
A A! A
R R '7
u -/R c
- - '2
U/R u -
R _ -
- R ..
_ U/R ..
U .. ..
- - U
- - U
R R 7
U U U
U U U
- - .-/R

COWGfP

5t!

PUWNC:

SPECIES
lflelilotus officonalis
kauuniazfistulosa.
Cenothera

Pastinaca sativa
Physocarpus opulifolius
Polygala paucifolia
Potentilla.norvegica
Potentilla recta
Prunella vulgaris
Prunus pensylvanica
Prunus serotina
PrunUS‘virginiana
PyrUS‘malus
Ranunculus acris
Ranunculus arborvitus
Rhamnus alnifolia
Ribes americanun
Ribes cynosbati
Ribes #2

Ribes triste

RObinia pseudoacacia
Rosa.b1anda

Rubus allegheniensis
Rubus canadensis
Rubus'hispidus

RUbus idaeus

9O
FAMELY

Fabaceae
Lamiaceae
Onagraceae

Apiaceae

 

Rosaceae
Polygalaceae
Rosaceae
Rosaceae
Lamiaceae
Rosaceae
Rosaceae
Rosaceae
Rosaceae
Ranunculaceae
Ranunculaceae
Rhamnaceae
Saxifragaceae
Saxifragaceae
Saxifragaceae
Saxifragaceae
Fabaceae
Rosaceae
Rosaceae
Rosaceae
Rosaceae

Rosaceae

9.2. 9.: E 11. .12
-IR- - - -
-IR- - .. ..
R R R R R
- - ‘U/C - -
_ - - U -
R R R U

2 R ? R R
R R R - -
- R U R U
U A U? R R
U - R? U R
U - c U R
- .. R .. ..
U R U R R
- - - R, -
U R 9 U -
R, - - _ _
c R 9 U R
_ R - - -
R. - - - -
_ - - R/U -
c - - - U
_ c - - -
- C/U- - -
c c c c A

SPECIES

RUbus parviflorus
RUbus pUbescens
Rudbeckia.hirta
SaliXLbebbiana

Salix hunilis

Salix lucida

Salix petiolaris
Sambucus pubens
Sanguinariancanadensis
Saxifrage
Scrophularia lanceolata
Scutellaria

Senecio aurens
Senecio paupercula
Silene cucdbalus
Smilacina stellata
Sparganium.

Spiraea alba
Solidago canadensis
Solidago gigantea
Solidago graminifolia
Solidago missourense
Solidago nemoralis
SolidagO‘uliginosa
Sondhus

Stellaria longifolia

91
FAMILY

 

Rosaceae
Rosaceae
Asteraceae
Salicaceae
Salicaceae
Salicaoeae
Salicaceae
Caprifoliaceae
Papaveraceae
Saxifragaceae
Scrophulariaceae
Lamiaceae
Asteraceae
.Atseraceae
Caryophyllaceae
Liliaceae
sparganiaceae
Rosaceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae

Asteraceae

Caryophyllaceae

.CEEEQifl_
UU?R
-U..-
CCCR?
?UU?
CR-R/U
AACR
RR-U
C - - -
- R - -
---c
U--'R
---U
---R
- R _ -
---R
R. - _ -
-URU
cccc
C-UA
Ac--
UUCC
uccc
?RR-
--U_
--U-

SPECIES

Taraxacun officinale
Thalictrum.dasycarpum
'I‘ragopogon pratensis
Trientalis borealis
Trifoliun agrarian
Trifoliun hybridun
Trifolium pratense
Trifoliun repens
Trillium cernuun
Trilliun grandiflora
Triosteum

'I‘ypha latifolia
Urtica dioica
Vaccinium angustifoliun
Vaccinium myrtiloides
Verbascun thapsus
Veroinca anagallis-aquatica
Viburnum

Viola conspersa
Viola cucullata
Viola

Viola

Viola pubescens

Viola septentrionalis

9 2
FAMILY

 

Asteraceae
Ranunculaceae
Asteraceae
Primlaoeae
Fabaceae
Fabaceae
Fabaceae
Fabaceae
Liliaceae
Liliaceae
Caprifoliaceae
Typhaceae
Urticaceae
Ericaceae
Ericaceae
Scrophulariaceae
Scrophulariaceae
Caprifoliaceae
Violaceae
Violaceae
Violaceae
Violaceae
Violaceae

Violaceae

CL CH
C "C
U U
U ?
U U
C C
C C
R C
R ..
R ..
C U?
C U?
R ..
- R
C -
C -

F1 F2
c R
c R
U R
R ..
U U
R A
-.- c
- c
R ..
- R
- U
c R
C/U‘R
C/U R
U C/U
c R
C ..
c -

93

LITERATURE CITED

Batra, S. W. T. 1972. Some properties of the nest-building
secretions of Nomia, Anthophora, Hylaeus and other
bees. J. Kansas Entomol. Soc. 45:208-218.

 

. 1980. Ecology, behavior, pheromones, parasites and
management of the sympatric vernal bees Colletes
inaeggalis, g. thoracicus, and g. validus. 3. Kansas
Entomo . Soc. 53:509-538.

Burks, B. D. 1958. A recharacterization of the genus
Coelo enc rtus, with descriptions of two new species.
J. WasE. Acad. Sci. 48:22-26.

Danks, H. V. 1971. Populations and nesting-sites of some
aculeate hymenoptera nesting in Rubus. J. Anim. Ecol.
40:63-77.

Davidson, A. 1895. On the nest and parasites of Prosopis
varifrons Cresson. Psyche 7:315-316.

Fisher, R. A. 1958. The genetical theory of natural
selection, 2nd ed. New York.

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

Fye, R. E. 1965a. The biology of the Vespidae, Pompilidae,
and Sphecidae from trap nests in northwestern Ontario.
Canad. Ent. 97:716-744.

. 1965b. Biology of Apoidea taken in trap nests in
northwestern Ontario. Canad. Ent. 97:863-877.
Gleason, H. A. and A. Cronquist. 1963. Manual.of vascular
plants of northeastern United States and adjacent
Canada. Willard Grant Press, Boston, Mass. 810p.

Krombein, K. V. 1967. Trap nesting wasps and bees.
Smithsonian Press, Washington D.C.

Krombein, K. V., P. D. Hurd, D. R. Smith, and B. D. Burks.
1979. Catalog of Hymenoptera North of Mexico.
Smithsonian Institution Press, Washington D.C.

94

Malyshev, S. I. 1968. Genesis of the Hymenoptera and the
phases of their evolution. 319p.

Marston, N. 1970. Revision of new world species of Anthrax
other than the Anthrax albofasciatus group.
Smithsonian ContriE. ZooI. 43:148p.

 

Medler, T. 1966. A note on H laeus Fabricius in trap-nests
in Wisconsin. Proc. Ent. Soc. Wash. 68:131.

Mitchell, T. B. 1960. Bees of the eastern United States,
vol. 1. North Carolina Ag. Expt. Sta. Tech. Bull. No.
141:538p.

Snelling, R. R. 1966a. Studies on North American bees of
the anus H laeus 1. Distribution of the western
spec es of t e subgenus Prosopis with descriptions of
new forms. Los Angeles Coun y Museum Contrib. in
Science 98:18p.

. 1966b. Studies on North American bees of the genus

H laeus 2. Description of a new subgenus and species.
Proceed. Biol. Soc. Wash. 79:139-144.

. 1966c. Studies on North American bees of the genus
H laeus 3. The nearctic subgenera. Bulletin So. Calif.
Academy Sciences 65: 164-175.

 

 

. 1968. Studies on North American bees of the genus
H laeus 4. The subgenera Cephalylaeus, Metziella and
Herana. Los Angeles Coufity Museum Contrib. in Science
144: 6p.

. 1970. Studies on North American bees of the genus
Hylaeus. 5. The subgenera H 1aeus, s. str. and
garaprosopis. Los Angeles County MUseum Contrib. in
Science 180: 59p.

 

. 1975. Taxonomic notes on some colletid bees of
western North America with descriptions of new species.
Los Angeles County Museum Contrib. in Science 267:9p.

 

. 1983. Studies on North American bees of the genus

H laeus 6. An adventive palaearctic species in southern
California. Bull. Southern California Acad. Sci.
82:12-16.

 

95

Strickler K. 1982. Parental investment per offspring by a
specialist bee: does it change seasonally? Evolution
36: 1098-1100.

Tepedino, V. J., L. L. McDonald, and R. Rothwell. 1979.
Defense against parasitization in mud-nesting
hymenoptera: Can empty cells increase nest reproductive
output? Behav. Ecol. Sociobiol. 6:99-104.

Tepedino, V. J. and P. P. Torchio. 1982a. Temporal
variability in the sex ratio of a non-social bee, Osmia
li aria ro in a: extrinsic determination or the
EracEIng o an op imum? 0ikos 38:177-182.

. 1982b. Phenot ic variability in nesting success
among Osmia li ar a ro in a females in a glasshouse
environmenE. Eco . EntomoI. 7: 453-462.

Torchio, P. F. 1984. The nesting biology of Hylaeus
bisinuatus Forster and development of its immature
forms. 3. Kansas Ent. Soc. 57: 276-297.

Torchio, P. F. and V. J. Tepedino. 1980. Sex ratio, body
size and seasonality in a solitary bee, Osmia lignaria
propingga Cresson. Evolution 34: 993-1003.

Trivers, R. L. and H. Hare. 1976. Haplodiploidy and the
evolution of social insects. Science 191: 249-263.

"TIEWMAAMMMAA’1'“