.35 3.34114». 3
“#3 . .v fl... .w
. by! . ”aw? a. up?
_ .

»‘-! 'l
1.1.... x
2
3‘

. 2
ii... 2‘.

1&3»...

2
. . en...

«QmWQKRfi.
war... a

.3
a... La. .

. Sci...

‘ i 13h"?

. . . fit...

a

.1.

n:
6..

!! pr-
3!}!

i.”
1......«fia
E

2. 2...:
..h.....a. 3..
, If ’A\Jflvl“ . .
{I .I. V.

. . . V .i...$-.f.

. .14..)

):‘(¥' K

12:...

‘ 1 ‘2
.. .5; - .32....
Q‘.I.$.¢AS curl! a...

l....2.§. .1}.

H4...
I.It.....t..t..lt .
i. \(f! ‘ .

 

 

 

 

‘93:ng
.s .So‘rluvuiuli'l
: ‘VI-P: ,al’b n‘w'
)v..-‘ IV. I.
9.. .‘t .A . )\
lg<

u akin...»
,y .335: .
p. , 2 Trial

 

:l.h!.. : :1 Iii.

 

 
 

LIBRARY
Michigan State
University

This is to certify that the
thesis entitled

ISLAND POPULATIONS AND TRAIT COMPARISONS OF
TIGER SWALLOWTAIL BUTTERFLIES, P. CANADENSIS,
IN THE GREAT LAKES REGION

 

presented by

Gabriel J. Ording

has been accepted towards fulfillment
of the requirements for

14.8. . E 1
degree In ntomo ogy

WM

Major professor

 

Date J9 EEC gm]

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

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJClRC/DateDuepGSop. 15

 

ISLAND POPULATIONS AND TRAIT COMPARISONS OF TIGER
SWALLOWTAIL BUTTERFLIES, P. CANADENSIS, IN THE GREAT LAKES
REGION

By

Gabriel J. Ording

A THESIS

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

MASTER OF SCIENCE

Department of Entomology
and
Ecology, Evolutionary Biology and Behavior Program

2001

ABSTRACT
ISLAND POPULATIONS AND TRAIT COMPARISONS OF TIGER
SWALLOWTAIL BUTTERFLIES, P. CANADENSIS, IN THE GREAT LAKES
REGION
By

Gabriel J. Ording

The objectives of this thesis were to examine gene flow between geographically
isolated Great Lakes Island subpopulations of swallowtail butterflies, Papilio canadensis,
using wing trait morphometrics and allozyme electrophoresis. Specimens from Isle
Royale in Lake Superior, Beaver and South Manitou Islands in Lake Michigan, were
compared to adjacent mainland populations. There were no significant difi‘erences
between either Isle Royale or Beaver Island and their respective adjacent mainland
populations. South Manitou Island however, showed significant differences from
adjacent mainland populations for every character analyzed. These differences were
attributed to an introgression of genes from Papilio glaucus, a closely related species
who’s described range begins approximately 150 km to the south.

The extent of the P. glaucus introgression on and around South Manitou Island
was investigated through further analysis of morphometric, biochemical, behavioral, and
physiological traits. The Tiger Swallowtail butterflies on and around South Manitou
exhibit many characteristics making them appear hybrid-like, intermediate between P.
canadensis and P. glaucus. It is suggested that periods of increased thermal unit
accumulations along the western shore of Michigan may allow extended movement of P.

glaucus alleles significantly further northward from that observed inland.

To my family

iii

ACKNOWLEDGMENTS

Thanks to my co-major advisors, Dr. Mark Scriber for his never ending supply of
enthusiasm, guidance, patience and understanding; Dr. Larry Besaw for many years of
mentorship in the realms of science education. Thanks to my guidance committee, Dr.
Cathy Bristow, and Dr. Jim Miller for their willingness to participate in this project and
for providing helpful suggestions and revisions. Thanks also to Drs. Jim Smith and Guy
Bush for welcoming me into their laboratory to conduct the biochemical aspects of this
study.

Thanks to the National Park Service, at both Isle Royale and Sleeping Bear
Dunes, for allowing this research to take place, providing special use and specimen
collection permits. Special thanks to Jack Oeltke and Steve Yancho, the resource
management staff at both parks respectively.

Thanks to all of the graduate students and lab technicians of the Scriber Lab over
the past several years, for everything from showing me the ropes in the very beginning to
helping me with the finite details near the end. Thank you for the moral support and the
friendship. Special thanks to Aram Stump for an incredible amount of help with various

aspects of this research, especially sharing his expertise in allozyme electrophoresis.

Great gobs of thanks to my wife Betsy, for her patience and love.

TABLE OF CONTENTS

LIST OF TABLES .................................................................................. vii
LIST OF FIGURES ................................................................................. ix
CHAPTER 1:

INTRODUCTION ................................................................................... 1
CHAPTER 2:

ARE ISLAND POPULATIONS OF PAPILIO CANADENSIS ISOLATED FROM
ADJACENT MAINLAND POPULATIONS? ................................................... 5
Introduction ............................................................................................ 5
Materials and Methods ............................................................................... 9
Results ................................................................................................ 18
Discussion ............................................................................................ 30
CHAPTER 3:

ISOLATED “HYBRID SWARM”: INTROGRESSED GENES OF PAPILIO GLA UCUS
IN A P. CANADENSIS POPULATION FAR BEYOND THEIR HYBRID ZONE ....... 33
Introduction .................................................................................. ' ........ 33
Materials and Methods .............................................................................. 38
Results ................................................................................................ 45
Discussion ............................................................................................ 61
CHAPTER 4:

SUMMARY ......................................................................................... 69
APPENDIX 1:

RECORD OF DEPOSITION OF VOUCHER SPECIMENS ................................. 73
APPENDIX 1.1

VOUCHER SPECIMEN DATA .................................................................. 75
APPENDIX 2:

POPULATION WING MEASUREMENTS AND ALLOZYME DATA .................. 79
APPENDIX 2.]

OVIPOSITION PREFERENCE DATA ........................................................ 108
APPENDIX 2.2

LARVAL HOST PLANT SURVIVAL DATA ................................................ 112

LITERATURE CITED ............................................................................ 116

vi

LIST OF TABLES

Table 2.1. 1998 Island versus mainland forewing length comparison. Island locations
are printed in bold print with the corresponding comparative mainland populations
following. Sample sizes and mean forewing length values i standard deviation for each
location are presented. t-test P-values for each island versus mainland pair wise
comparison are listed following the corresponding mainland location. Values were
computed using Microsoft Excel statistical analysis ........................................... 19

Table 2.2. 1998 Island versus mainland black band comparison. Island locations are
printed in bold print with the corresponding comparative mainland populations following.
Sample sizes and mean black band width percentage values :1: standard deviation for each
location are presented. t-test P-values for each island versus mainland pair wise
comparison are listed following the corresponding mainland location. Values were
computed using Microsoft Excel statistical analysis. .......................................... 22

Table 2.3. 1998 population Pgd allele frequencies. Each sampled population is listed
with the sample size in parentheses below. Island populations are printed in bold type.
Alleles marked with an asterisk indicate a Papilio glaucus type allele ...................... 26

Table 2.4. Genotypic differentiation for each island and its adjacent mainland
populations at the PGD locus. Ho: the genotypic distribution is identical between
population pairs. Statistical analysis performed in Genepop Version 3.1d 1999.
Population pairs with P-values of significance <0.0S are in bold print ...................... 27

Table 2.5. Genie (allelic) differentiation for each island and its adjacent mainland
population at the PGD locus. Ho: the allelic distribution is identical between population
pairs. Statistical analysis performed in Genepop Version 3.1d 1999. Population pairs
with P-values of significance <0.05 are in bold print .......................................... 29

Table 3.1. Summary of selected species differences discussed between Papilio glaucus
and P. canadensis. (Modified from Table l in Scriber 1990) ................................ 34

Table 3.2. Additional sampling sites from 1998-2000. Male and female sample sizes and
dates of collection are provided for each sample location ..................................... 39

Table 3.3. South Manitou 3-choice oviposition preference for individual females 1998-
2000. Only females ovipositing >10 eggs were used for this analysis. The total number
of eggs oviposited by each female is indicated as well as the percentage of the eggs
placed on each host plant. Percentages >50% are in bold print indicating that that female
expressed a “preference” for that particular host plant. The other category includes many
eggs that were placed within 2 cm of the host plant and may have been intended for that
host. For the purposes of this study they have been scored as other... . . . ............48-49

vii

Table 3.4. South Manitou population total 3-choice oviposition preference 1998-2000.
The eggs of all females that laid any eggs (>10 and <10) were pooled to provide an
indication of the total population oviposition preference. Total number of eggs laid by all
females is provided as well as the number of females that were assayed .................... 50

Table 3.5. Allele frequencies for three diagnostic loci present in each of the study
populations from 1998-2000. Values presented are percentages of Canadensis versus
glaucus alleles detected. Dashes indicate that electrophoresis was not performed for that
locus for that population ........................................................................ 55-57

viii

LIST OF FIGURES

Figure 2.1. 1998 Sample sites and sample sizes ................................................ 11

Figure 2.2. Forewing length measurements (measurement A) are the distance from the
tip of the forewing to the thoracic wing base attachment (figure modified from Leubke et
a1. 1988) .............................................................................................. 13

Figure 2.3. Black band width measurements are the percentage of the hind wing anal cell
that is filled by the dark band labeled A. (Modified from Luebke et a1. 1988) ............ 15

Figure 2.4. Correlation of population mean forewing length by increasing latitude. The
populations sampled in 1998 are listed in order of increasing latitude (middle). Sample
sizes are indicated in each histogram bar. The top figure represents the same populations
arranged according to specific latitudes. The best linear fit is indicated, with R2 value =
0.897 and Prob > F value = 0.0004 ................................................................ 21

Figure 2.5. Mean black band width percentage and standard deviations for all
populations sampled in 1998 listed in order of increasing latitude. Population sample
sizes are indicated in histogram bars ............................................................. 23

Figure 2.6. Correlation of black band width versus latitude. Circles represent mainland
populations while squares represent island populations. quuare value = 0.788 and Prob
> F value = 0.003 ................................................................................... 25

Figure 3.1. The shaded region in Michigan roughly indicates what has historically been
considered the hybrid zone between Papilio glaucus and P. canadensis (Nielsen, 1999).
The shaded region across Wisconsin into Minnesota represents the 50 year average
degree day accumulation and northern limit allowing for two generations of Papilio
glaucus. The star indicates the location of South Manitou Island ........................... 36

Figure 3.2. F orewing length comparison between P. canadensis from South Manitou
Island, Oscoda co., and a population of P. glaucus from Lawrence co., Ohio. Mean
forewing length for each population is shown with error bars indicate :t s.e. (S. Manitou =
46.7 i 0.49, Oscoda co. = 46.0 i 0.92, P. g. Ohio = 49.0 i 0.53). Comparison using both
Tukey-Kramer and Student’s t-test at alpha value = 0.05 indicates that there is no
significant differences between S. Manitou and Oscoda co. while there is a significant
difference between S. Manitou and Lawrence co., Ohio ....................................... 46

Figure 3.3. Black band width comparison between P. canadensis from South Manitou
Island, Oscoda co., and a population of P. glaucus fi'om Lawrence co., Ohio. Mean black
band widths for each population is shown with error bars indicating :l: s.e. (S. Mnitou =
55.20 i: 1.29, Oscoda co. = 63.38 :t 1.71, P.g. Ohio = 35.50 i 1.94). Comparison using
both Tukey-Kramer and Student’s t-testat alpha value = 0.05 indicates that South
Manitou differs significantly fi'om both Lawrence co. and Oscoda co. (p-values <

0.001) ..................................................................................... . ............ 47

Figure 3.4. South Manitou population total 3-choice oviposition preference .............. 50

Figure 3.5. South Manitou larval host plant survival 1999-2000. approximately equal
numbers of the eggs from each female that laid >10 eggs were distributed on each of the
host plant Options (Tulip tree, Black cherry, and Quaking aspen). The mean family %
survival is the average percentage of larvae that survived through the first instar on each
host plant, where as the total % survival is the percentage of all assayed larvae that
survived on each host plant ........................................................................ 53

Figure 3.6. 1998 populations sampled. Pie charts indicate the percentage of glaucus
alleles in each population for the Pgd allozyme locus ......................................... 58

Figure 3.7. 1999 populations sampled. Pie charts indicate the percentage of glaucus
alleles in each population for the Pgd allozyme locus ......................................... 59

Figure 3.8. 2000 populations sampled. Pie charts indicate the percentage of glaucus
alleles in each population for the Pgd allozyme locus ........................................ 6O

CHAPTER 1:

INTRODUCTION

Gene flow is the movement of gametes, individuals, or groups of individuals from
one place to another (Slatkin 1987). Two important ecological concepts that are
significantly impacted by levels of organismal gene flow are island biogeography and
hybrid zone theory. This thesis investigates aspects of both of these theories, looking at

populations of Papilionidae butterflies in the Great Lakes region.

Islands are important in helping biologists understand the processes of evolution
and natural selection. Visiting the Galapagos Islands was of paramount importance in
Charles Darwin’s development of his theories. Likewise, it was while visiting the island
of Temate (Indonesia), that Alfred Wallace was struck by the idea of evolution
(Williamson 1981). Since then, island systems have been investigated extensively, and
used as natural laboratories, to further understand these processes. Key research
conducted by MacArthur and Wilson (1967) investigated aspects of island biogeography
correlating island sizes and distances from the adjacent mainland, with rates of species
colonization and extinction. Of course these rates vary for different organisms due to
differing dispersal mechanisms.

Island populations of organisms live in geographic isolation from their
mainland counterparts. This isolation can act as a barrier to gene flow between the island
and the mainland populations. The island population has a limited number of individuals
contributing to its gene pool, and is subject to genetic drift. Genetic drift is the

unpredictable and random change in gene frequency due to finite population size (Slatkin

1987). Genetic drift can act as a mechanism to drive genetic differentiation between

island and mainland populations.

As with island biogeography theory, the concept of gene flow heavily influences
hybrid zone theory. A hybrid zone is a geographic region where the populations of two
different species overlap and cross producing offspring of mixed ancestry (Harrison
1993). The width of a hybrid zone is determined by the counterbalancing forces of
selection and dispersal. There is great debate as to the importance that should be ascribed
to hybridization in evolutionary processes. Frequently, hybrid offspring are far less fit
than either parental strain, sometimes even sterile. However, hybridization may also
provide unique combinations of alleles that in certain environments prove superior to

either parent strain (Arnold & Hodges 1995).

A great deal of research has been conducted on dispersal and the levels of gene
flow in and between populations of many species of butterflies. Great variation exists in
the tendency of different butterflies to travel. Monarch butterflies (Danaus plexippus) are
well known for their annual migration from various locations in North America 2000
miles south to Mexico. Assisted by winds, monarchs have been known to fly 80 miles in
a single day. But, monarchs are the only species in their family known to migrate (Wolfe
1994). On the other hand, after many years of rigorous research it has been documented
that the Checkerspot Butterfly (Euphydryas editha) possesses intrinsic barriers to
widespread dispersal (Ehrlich 1961). “Butterflies (except those few species which are
migratory) seem to be quite sedentary as compared with what one might expect in view

of their powers of movement.” (Ehrlich and Raven 1969).

Barriers to gene flow in various Lepidoptera have been described, taking several
forms. Three closely related species of saturniid silk moths Callosomia promethea, C.
angulifera, and C. securifera can be hybridized by hand-pairing but are reproductively
isolated in nature by temporal differences in mating times (Johnson et al. 1996). More
commonly described as limiting gene flow are physical geographic barriers between
populations of the same species. For example, some Checkerspot butterfly populations in
mountainous areas have restricted gene flow (Britten et al. 1995). Also, the Great Lakes
themselves have been described as barriers reducing gene flow between populations of
butterflies (Waldbauer & Stemburg 1988).

There have been investigations of dispersal and gene flow done on species of
Papilionidae. Using methods of mark, release, and recapture, on Maryland populations
of Papilio glaucus, it was concluded that males tend not to disperse but that females do
tend to disperse more widely (Fales 1959). More recent research utilizing molecular
markers suggests high gene flow between populations of several Papilio species: P.
hospiton (Aubert et al. 1997), P. machaon (Auber et al. 1997, Hoole et a1. 1999), and P.
glaucus (Bossart & Scriber 1995).

In the Great Lakes region a great deal of research has been conducted
investigating gene flow between populations of Papilio glaucus and P. canadensr's,
within their respective ranges and across their narrow hybrid zone (Scriber 1990, Hagen
et al. 1991, Scriber 1994, Deering 1998, Scriber et al. 2001). An investigation was
conducted in Michigan solely on populations of P. canadensis to determine whether there
was restricted gene flow across the state. It was concluded that there were high levels of

gene flow between the populations sampled (Stump 2000).

This thesis reports an investigation of gene flow between island and mainland
populations of P. canadensis in the Great Lakes. Chapter 2 emphasizes the initial
findings. Early in this investigation, high levels of genetic introgression from P. glaucus
were found in the South Manitou Island population of P. canadensis. This highly
unexpected finding shifted the long-term emphasis of this research project. Chapter 3
reports follow up research that investigated the extent to which introgression is present on

and around the South Manitou Island population of P. canadensis.

CHAPTER 2:
ARE ISLAND POPULATIONS OF PAPILIO CANADENSIS ISOLATED FROM

ADJACENT MAINLAND POPULATIONS?

Introduction

Island biogeography theory suggests that island populations experiencing reduced
levels of immigration have a tendency to become genetically differentiated from their
mainland counterparts (Johnson et al. 2000). This genetic differentiation can be the result
of a combination of mechanisms, including founder effect, genetic drifi due to finite
population size, or natural selection favoring adaptations to local environmental
conditions. These mechanisms leading to genetic differentiation can be counterbalanced
by gene flow (Slatkin 1987).

A great deal of research has been conducted on various species of Papilionidae
butterfly populations, including investigations of dispersal, gene flow, and estimations of
population differentiation (Fales 1959, Tong & Shapiro 1989, Bossart & Scriber 1995,
Aubert et al. 1997, Hoole et al. 1999, Stump 2000). Each of the investigations measuring
gene flow between populations suggested that there were sufficient levels of gene flow to
counterbalance genetic differentiation. However, none of these investigations sampled
island populations of Papilionidae.

Determination of genetic differentiation between island and mainland populations
can be accomplished using various techniques for trait analysis. One method to expose
divergence is statistical analysis of heritable quantitative phenotypic characteristics (Boag

& van Noordwijk 1987). However, a more powerful technique to identify genetic

divergence is the analysis of biochemical markers. Enzyme electrophoresis has become a
popularly utilized method to estimate levels of genetic variation and differentiation
(McKechnie et al. 1975, Leberg 1992, Bossart & Scriber 1995, Stump 2000). Combining
these two techniques can prove a powerful method by which to identify island and

mainland population genetic differentiation.

Butterfly forewing length is a direct indicator of adult size. This can be
influenced by host plant nutritional quality, impacted by thermal environmental
conditions, and has also been shown to be commonly selected upon in island populations
of insects. It has been shown that P. canadensis living in the interior of Alaska endure
significantly shorter, cooler summers than those living in northern Michigan. As
adaptations to these differing local environmental conditions, P. canadensis living in
Alaska were shown to lay smaller clutches of eggs but each egg a larger size, resulting in
larger first instar larvae, which then were observed to have 40% higher consumption
rates, that led to earlier pupation. These Alaskan pupae were significantly smaller, which
in turn produced smaller adult size and forewing length, than adults found in northern
Michigan (Ayres & Scriber 1994). It has been documented that the climatic conditions of
the three island locations under investigation in this thesis, are heavily moderated by their
respective surrounding bodies of water (Hatt et al. 1948, Allen 1979, Haswell & Alanen
1994). This could serve to act differentially for Great Lakes island versus mainland
populations, as a selective agent on the forewing length character.

Sometimes a reduction or complete loss of wings is a common insect adaptation

to island living. There are many examples of beetle (Coleoptera) populations living in

isolation on islands or mountain tops around the world, that are reported to exhibit
atrophy or complete loss of their wings (Darlington 1943). Tristan da Cunha, a group of
small volcanic islands in the South Atlantic Ocean, is home to 20 endemic species of
beetles, all but two of which have reduced wings, and also a flightless species of

Drosophilid (Scaptomyzafi‘ustolifera) (Williamson 1981).

Banding patterns on butterfly wings is a common method by which species can be
distinguished, as is the case between Papilio glaucus and P. canadensis (Hagen et al.
1991 ), or can also be used to discern differences between populations within a species.
The Monarch butterfly (Danaus plexippus) has a range that extends across the Western
Hemisphere, from Alaska in the north to Patagonia in the south. There are differences in
wing banding patterns that allow individuals coming from one region to be distinguished
from an individual coming from another (Williams et al. 1942).

P. canadensis has a distinct black band on its hind wing that partially fills the anal
cell (Fig. 2.3). This morphologic character is one used to distinguish between Papilio
canadensis and P. glaucus (Hagen et al. 1991). This dark band is wider in the northern of
the two species, and it has been suggested that increased dark melanic coloration could
serve as a thermal collection mechanism, helping increase possible metabolic rates, in
colder environments (Watt 1968, Kingsolver 1985, 1987, 1995). Again, the island
populations under investigation in this thesis exist under differing thermal climactic
regimes than do their mainland counter parts, owing to the lake effect. This could act
differentially, as a selective force on the black bandwidth, between island and mainland

populations.

Analysis of allozyme allele frequencies is extremely useful in identifying the
presence of genetic differentiation between populations. This can be especially true
when populations under scrutiny are of a finite size located on islands. Rare alleles that
might arise due to mutation can be amplified on islands, due to genetic drift. Also, there
is a tendency in isolated locations with finite effective population sizes, towards a
decrease in average heterozygosity and loss of alleles at any given locus (Carlquist 1974,
Jacquard 1974, Hartl 1988, Grant 1998). The most informative allozyme loci to use for
such an investigation would be those that are highly polymorphic. P. canadensr‘s
populations were assessed for differentiation using the Pgd (6-Phosphogluconate
dehydrogenase) allozyme locus. This locus was chosen as a result of it being highly
polymorphic (Hagen & Scriber 1991), relatively consistent and easy to interpret, and was

highly informative for the latter portion of this thesis.

MATERIALS AND METHODS

Papilio canadensis is an extremely common butterfly in the Great Lakes region,
including robust populations found on several of the islands of the Great Lakes. In order
to determine whether there was significant genetic differentiation between these island
populations and their mainland counterparts, samples were collected from three islands
(all at least 12 km from the nearest “mainland”) and five adjacent mainland locations.
Morphometric analyses were performed on two wing characteristics, forewing lengths
and hind wing black bandwidths. Multiple statistical analyses were performed to identify
significant differences. Allozyme electrophoresis analyses were used to determine allele
and genotype frequencies at a highly polymorphic enzyme locus (Pgd; phosphogluconate

dehydrogenase).

Specimen Acquisition and Transport

Papilio specimens utilized in this research were live-captured by net, from
selected wild populations throughout the State of Michigan (unless otherwise noted).
Specimens were most frequently captured while feeding on available sources of nectar,
puddling, and sometimes while in flight. Specimen collections were primarily made
during mid-day, between approximately ten o’clock am. and four o’clock pm. on warm
sunny days, these being the predominant hours for flight activity. Mainland specimen
collections were made using vehicle transportation. Butterflies were located along the
side of the road, puddling, in flight, or fluttering on nectar sources. This method of

specimen collection often led to the discovery of desirable puddling locations or high

concentrations of appropriate nectar sources, each often times with high densities of
nectaring butterflies. These locations could then be returned to multiple times in a season
for further collection.

Collections on island locations were similarly made, however foot travel was the
only available mode of transportation. Butterfly specimens were encountered while
hiking across the island locations. Again, specimens were found in flight, puddling, or on
nectar sources. Locations that offered high concentrations of butterfly activity due to
appropriate puddling conditions or high concentrations of nectar sources were returned
to, sometimes multiple times in a day.

The dates and locations of specimen collections for 1998 are as follows (see
associated map Figure 2.1): Cook Co. MN (n = 35 males; 29 May and 23 June), Isle
Royale National Park, Keweenaw Co. (n = 28 males, n = 1 female; 23-28 May), Gogebic
Co. (11 = 36 males, n = 4 females; 22 and 28 May), Dickinson Co. (n = 63 males, n = 19
females; 8 June), Beaver Island, Charlevoix Co. (n = 11 males; 14 June), Charlevoix Co.
(mainland) (n = 50 males, n = 17 females; 14-25 May), South Manitou Island, Leelenau
Co. (n = 32 males, n = 24 females; 17-18 June), Mason Co. (n = 50 males, n = 15
females; 23 May). The difference in the numbers of males collected compared to females
collected is likely the result of two factors. First, collections may have been completed
early in the flight period of the species. These earlier stages of annual flight are generally
dominated by males for which eclosion occurs somewhat earlier. The other likely reason
for the bias towards male sampling is the result of collecting large numbers of specimens

while they are puddling. Puddling is an activity that is performed almost exclusively by

10

7.. , ~_._wg,v T‘I w—jip‘? f3.“ 1"? .‘--m sis—.‘r-arx-rfi

. 4

.U: ~ ‘_..>.._

 

Figure 2.1. 1998 Sample sites and sample sizes.

Sample Location Males Females
1. Cook Co., Minnesota 35 0
2. Isle Royale National Park, MI 28 1
3. Gogebic Co., Michigan 36 4
4. Dickinson Co., Michigan 63 19
5. Beaver Island, Michigan 11 3
6. Charlevoix Co., Michigan 50 17
7. South Manitou Island, Michigan 32 24

8. Mason Co., Michigan 50 15

males. It is thought that the purpose for puddling is that males are collecting salts and
nutrients required for sperm production.

Upon capture, individual specimens were placed with their wings folded back into
2 oz. Glassine envelopes, which were appropriately labeled with specimen sex, date and
location of capture. Collections were transported alive to the laboratory in Tupperware®
plastic containers in ice coolers, which lowered specimen body temperatures and slowed
metabolism. While at the remote locations of South Manitou Island and Isle Royale
National Park, lowering of specimen body temperatures was accomplished by placing the
Tupperware® containers into large airtight zip-lock food storage bags. These were then
placed into a collapsible bucket containing cool water from either Lake Michigan or Lake
Superior. Isle Royale National Park specimens were overnight delivered (live) by the
US. Postal Service, from Houghton-Hancock to East Lansing, Michigan. Upon arrival in
East Lansing, specimens were preserved by freezing them alive in an -80° C ultra-low

biological freezer for later processing.

Wing Morphometrics

After the wings were detached from adult swallowtail specimens during the
allozyme electrophoresis preparatory protocol, they were assayed for two major
morphological features. Forewing length, from the distal tip of the wing to the basal
thoracic attachment (Fig. 2.2), was measured using a clear plastic metric ruler to the
nearest mm. On the ventral surface of the wings, the width of the anal black band was
- assessed as a percentage of the distance from the wing edge to the Cu2 vein. This

measurement was taken at a line of intersection with the junction of vein Cu2 and the

12

Forewing ‘
/\‘
\
\

 

Figure 2.2. Forewing length measurements (measurement A) are the distance
from the tip of the forewing to the thoracic wing base attachment (figure
modified from Leubke et al. 1988).

13

discal cell (Fig. 2.3). This anal black band measurement was taken to the nearest .Im
using a dissecting microscope and a WILD glass micrometer slip. For both of these
morphometric characters assayed, measurement values for both wings were taken when
available, and the mean values for each individual have been utilized for analysis. In
cases where wings were damaged and ripped, preventing an accurate measurement; if one
wing was undamaged a single measurement has been utilized for analysis; if both wings

were damaged, the specimen has not been included in the analyses.

Allozyme Electrophoresis

Allozyme electrophoresis was performed on adult male Tiger Swallowtail
Butterflies in this study. Electrophoresis protocol follows that of Hagen and Scriber
1991. Adult specimens were removed from -80° C and processed in a 4° C cold room.
Using a scalpel or razor blade, wings were dissected from the thorax at their place of
attachment and returned to Glassine envelopes for previously discussed morphometric
analysis. Tissue extracts were prepared by grinding one half of abdomen with 100 pl of
an extraction buffer. The lower half of the abdomen was utilized in male specimens. The
remaining abdomen portion, head and thorax, were returned to the -80° C freezer for
future use. The extract was centrifuged for 10 minutes at 14,000 rpm. At this point the
extract could be stored at -80° C until ready to continue the electrophoresis protocol.
Female specimens were not utilized in the electrophoretic portions of this study for
several reasons. First, the sample sizes for females were extremely low for many of the
populations sampled. More importantly however, the allozyme banding patterns

produced by females were frequently not as clear as those produced by male specimens

Hindwing Black band

 

Figure 2.3. Black band width measurements are the percentage of the hindwing
anal cell that is filled by the dark band labeled A. (Modified from Luebke et al.

1988)

15

and difficult to interpret. It is possible that the eggs contained in the abdomen somehow
disrupt the normal staining process. When the electrophoresis process was performed on
females, the upper portion of the abdomen was utilized so as to avoid possible sample
contamination from spermatophores contained in the lower abdomen from previous
matings.

Samples were removed from -80° C and allowed to thaw in 4° cold room for
approximately 10 minutes and were then centrifuged for 5 minutes at 14,000 rpm. 7.5 ul
of extract from each sample was applied to thin layer acetate plates (Titan 111 [94 by 76
mm], Helena Laboratories) for electrophoresis. The allozyme locus scored for this study
portion is Pgd (6-Phosphogluconate dehydrogenase). Staining protocol and solution
recipe is contained in Appendix A. Scoring of gel banding patterns was accomplished
following methods of Hagen and Scriber 1991 using photographs or original gels and

sketches.

Population Comparisons

Island populations, Isle Royale, Beaver, and South Manitou Islands, were each
matched for comparative analysis with specimens from the most closely adjacent
mainland populations available. Considerations were made related to distance fi'om
island to mainland location, comparable latitudes, and also direction for possible
immigration due to prevailing winds. Isle Royale was compared to Cook county,
Minnesota and Gogebic county, Michigan populations. Beaver Island was compared to
Charlevoix and Dickinson county populations. South Manitou Island was compared to

Charlevoix, Dickinson, and Mason county populations (Fig. 2.1).

16

Statistical Analysis

Multiple statistical analyses of both forewing length and black bandwidths were
performed using JMP statistical software version 3.2 by Altura Software, Inc. One way
anova was performed for all populations. The data sets for both wing measurements
were evaluated for normality. In addition, black band percentage values were normalized
using an arcsin transformation performed in Microsoft Excel 1997. Comparisons for all
population pairs was accomplished using both Tukey-Krarner HSD and Student’s t-test.

Statistical analyses were performed for population allele frequencies using the
program Genepop v3.1 (Raymond and Roussett 1995). Tests for both genotypic and

genie differentiation were performed for each island versus mainland comparison.

17

RESULTS

Wing Morphometrics

Two quantitative polygenic wing characteristics were chosen for analysis.
Genetic differentiation between two populations is possible through phenotypie analysis
because shared genes would be reflected in similar phenotypes (Boag & van Noordwijk
1987). The characteristics under investigation, forewing length and hind wing anal cell
black bandwidth, were chosen for a combination of reasons. Both traits are heritable and
polygenic (Luebke et al. 1988), thus conceivably impacted by mutation, genetic drift, and
natural selection under differing local environmental conditions. In addition, these two
traits were highly informative measurements to consider for indicating interspeeifie

hybridization (Scriber et al. 2001), the later focus of this thesis.

Using both Tukey-Kramer HSD and Student’s t-test at p-value of 0.05 (Table
2.1), P. canadensis forewing length measurements on Isle Royale have been shown not to
be significantly different from those of Cook or Gogebic populations. Beaver Island
showed no significant differences between either Charlevoix or Dickinson populations
using Tukey-Kramer HSD, but did show a significant difference from the Dickinson
population using the less rigorous Student’s t-test (p-value < 0.01). Analysis using
Tukey-Kramer HSD showed a significant difference for South Manitou Island from only
the Dickinson population. Student’s t-test indicated a significant difference between
South Manitou Island and both Dickinson and Mason populations (p-values < 0.001 and

0.036 respectively).

18

Table 2.1. 1998 Island versus mainland forewing length comparison. Island locations
are printed in bold print with the corresponding comparative mainland populations
following. Sample sizes and mean forewing length values :1: standard deviation for each
location are presented. t-test P-values for each island versus mainland pair wise
comparison are listed below the corresponding mainland location. Values were
computed using Microsoft Excel statistical analysis.

 

 

 

 

Location (11) Mean 1' Std. Dev. t-test P-value
Isle Royale 28 43.8 i 2.1

Cook Co. 35 43.6 i: 1.8 0.53
Gogebic Co. 34 44.2 i 1.6 0.38
Beaver Island 11 46.4 :1: 1.6

Charlevoix Co. 50 46.0 i 2.7 0.57
Dickinson Co. 63 44.7 i 2.0 <0.01
South Manitou 32 46.7 i 2.8

Charlevoix Co. 50 46.0 i 2.7 0.29
Dickinson Co. 63 44.7 i 2.0 <0.001
Mason Co. 50 47.9 i 2.3 0.036

19

Plotting the forewing length means for each population, in the order of decreasing
latitude (Fig. 2.4), shows an apparent trend of increasing forewing length moving in a
southerly direction. Plotting forewing length against latitude for each sampled population
indicates a strong correlation between latitude and forewing length (R2=0.897). This
correlation has been shown to be generally true of populations of Papilio from Florida to

Alaska (Scriber 1994).

Summaries of statistical analysis of hind wing black band widths using both
Tukey-Krarner HSD and Student’s t-test comparing island populations with mainland
populations were completed using JMP statistical software. Summaries of t-test p-values
are found in Table 2.2. The results are very comparable to those found for forewing
length comparisons with only one exception. The Tukey-Kramer HSD analysis indicates
that the Isle Royale population is not significantly different than those of Cook or
Gogebic counties. In contrast, Tukey-Kramer HSD suggests that Beaver Island is not
significantly different than either Charlevoix or Dickinson counties, where as the less
rigorous Student’s t-test indicates again that Beaver Island is significantly different from
Dickinson county (p-value = 0.018). Both Tukey-Kramer and Student’s t-test analysis
agree that South Manitou Island black band widths are significantly different than those
of both Charlevoix and Dickinson counties, but not from that of Mason county.

Again, comparable to that of forewing length, if mean black bandwidths for each
population are listed in order of increasing latitude, a trend seems apparent and suggests

that black bandwidth increases moving in a northerly direction (Fig. 2.5). If black

20

 

   

 

 

 

 

 

 

 

48 .

A R2=0.897

E 47 — Prob > F = 0.0004

2’

5'0 46 -

I:

3

E" ‘5 ‘

3

D

*5 .. —

LL.

43 I I I I 1
43 44 45 46 47 48 49
Degrees Latitude North
55

l
l
l
i - so
i

 

Mean Forowlnglangth (m

1’

 

 

 

 

 

  

     

 

 

 

 

l

l

r
4.

   

 

 

 

. "' n=50 n=63. ‘
.. ‘7 . , ‘ . , .7 i > . , 3;.“7' .‘g’Qfl, 3' J 40
Mason S. Manitou Charlevoix Beaver Dickinson Gogebic Isle Cooke Co.
Island Royale MN

 

 

Figure 2.4. Correlation of population mean forewing length by increasing latitude.
The populations sampled in 1998 are listed in order of increasing latitude (middle).
Sample sizes are indicated in each histogram bar. The top figure represents the same
populations arranged according to specific latitudes. The best linear fit is indicated,
with R2 value = 0.897 and Prob > F value = 0.0004. Statistical analysis was
performed using JUMP Statistical software.

21

Table 2.2. 1998 Island versus mainland black band comparison. Island locations are
printed in bold print with the corresponding comparative mainland populations following.
Sample sizes and mean black band width percentage values 1: standard deviation for each
location are presented. t-test P-values for each island versus mainland pair wise
comparison are listed following the corresponding mainland location. Values were
computed using Microsoft Excel statistical analysis.

 

 

 

 

Location (n) Mean i Std. Dev. t-test P-value
Isle Royale 28 69.5 :t 4.9

Cook Co. 35 71.4 :t 6.3 0.18
Gogebic Co. 34 67.6 :t 7.7 0.23
Beaver Island 11 60.8 i 7.3

Charlevoix Co. 50 62.4 :t 7.3 0.50
Dickinson Co. 63 67.0 :t 5.8 0.018
South Manitou 32 55.2 i 7.3

Charlevoix Co. 50 62.4 i 7.3 <0.001
Dickinson Co. 63 67.0 i 5.8 <0.001
Mason Co. 50 58.5 i 7.7 0.057

22

 

 

 

Percent

 

 

 

 

Figure 2.5. Mean black band width percentage and standard deviations for all
populations sampled in 1998 listed in order of increasing latitude. Population
sample sizes are indicated in histogram bars.

23

 

bandwidth is plotted against latitude for each sampled population (Fig. 2.6), a strong
correlation between latitude and black bandwidth is evident (R2=0.788). This too has
been reported to generally be the case in populations of Papilio extending from Florida to
Alaska. Overall, latitude may be playing a larger factor in black band differentiation for
P. canadensis populations than does any island effect, with the exception of possibly

South Manitou Island.

Allozyme Electrophoresis
Of the nine Pgd alleles that exist in the closely related species groups of

Papilionidae (Hagen & Scriber 1991), six were encountered in this thesis analysis. In
addition, one undeseribed rare allele was encountered in one of the study populations.
Population allele frequencies are listed in Table 2.3. Two statistical analyses were
performed to identify genetic differentiation between island and mainland populations:
l)genotypic differentiatidn tests whether the distribution of genotypes is identical
between population pairs; 2)genic differentiation tests whether the distribution of alleles
is identical between population pairs.

Analysis of genotypic differentiation indicates that both Isle Royale and Beaver
Island are not significantly different than their respective mainland counterparts. South
Manitou Island however, has been shown to be significantly different fi'om all of its
mainland comparison populations, Charlevoix, Dickinson, and Mason counties with p-
values of 0.00289, 0.00002, and 0.00728 respectively (Table 2.4). Analysis of genie
differentiation for each of the study populations directly correlates to the genotypic

differentiation, with Isle Royale and Beaver Island not being significantly different from

24

Black Band Width Percentage

 

  
    
    
 

 

 

 

75

R2=0.788

Prob > F = 0.003
70 D

Isle Royale
65 "
50 "' Bger
0 Island
S. Mfafinitou
55 I LI—V I I I
43 44 45 46 47 48 49
Latitude

Figure 2.6. Correlation of black band width versus latitude. Circles represent
mainland populations while squares represent island populations. quuare value =
0.788 and Prob > F value = 0.003.

25

Table 2.3. 1998 population Pgd allele frequencies. Each sampled population is listed
with the sample size in parentheses below. Island populations are printed in bold type.
Alleles marked with an asterisk indicate a Papilio glaucus type allele.

Allele

Population

Isle Royale
(28)

Dickinson
(43)

Gogebic
(36)

Cook
(35)

Beaver Island

(11)

Charlevoix
(50)

South Manitou
(32)

Mason
(50)

-150

0.018

0.069

0.029

0.026

-l37

-125

0.893

0,828

0.863

0.886

1.00

0.906

0.875

0.035 0.009 0.878

*-100 -90

0 0

0 0.009

0 0

0 0

0 0
0.009 0
0.109 0
0.009 0

26

-80

0.089

0.095

0.137

0.086

0.051

0.016

0.07

*-50

0

0.009

Table 2.4. Genotypic differentiation for each island and its adjacent mainland
populations at the PGD locus. Ho: the genotypic distribution is identical between
population pairs. Statistical analysis performed in Genepop Version 3.1d 1999.
Population pairs with P-values of significance <0.05 are in bold print.

Populations Compared

P-value i S.E.

 

Isle Royale & Cook
Dickinson
Gogebic

Beaver Island & Charlevoix
Dickinson

South Manitou & Charlevoix
Dickinson

Mason

1.0000 i 0.0000
0.3678: 0.0063
0.3073 :1: 0.0046

0.4321 1 0.0043
0.0556 :1: 0.0031
0.0029 i 0.0004
0.00002 i 0.00001

0.0073 i 0.0008

 

27

their mainland counterparts and South Manitou Island again being significantly different

from all of its mainland comparison populations (Table 2.5).

28

Table 2.5. Genie (allelic) differentiation for each island and its adjacent mainland
population at the PGD locus. Ho: the allelic distribution is identical between population
pairs. Statistical analysis performed in Genepop Version 3.1d 1999. Population pairs
with P-values of significance <0.05 are in bold print.

Populations

P-value 1 SE.

 

Isle Royale & Cooke
Dickinson
Gogebic

Beaver Island & Charlevoix
Dickinson

South Manitou & Charlevoix
Dickinson

Mason

1.0000 i 0.0000

0.5641 i 0.0061

0.4057 :1; 0.0056

0.8298 :1: 0.0044

0.2964 :1: 0.0053

0.0054 :1: 0.0009

0.0000 i 0.0000

0.0026 i 0.0005

 

29

DISCUSSION

Based upon the combined analyses performed, I found little evidence to suggest
that the Papilio canadensis populations living on either Isle Royale or Beaver Island have
genetically differentiated from their mainland counterparts. These findings are consistent
with the results of other investigations on gene flow in various other Papilio species
(Tong & Shapiro 1989, Bossart & Scriber 1995, Aubert et al. 1997, Hoole et al. 1999).
In addition, these findings further support the conclusion that there is little genetic
structuring in Great Lakes area P. canadensr‘s populations (Stump 2000). South Manitou
Island however, exhibits significant differences from mainland counterpart populations
for all analyses performed. This suggests that there is significant genetic differentiation
between P. canadensis on South Manitou Island from the surrounding mainland

populations.

The analyses performed comparing the Isle Royale population and the mainland
counterpart populations of P. canadensis indicate that there is no significant difference
for any of the characters under scrutiny. Isle Royale appears extremely similar to its
nearest mainland counter part, Cook County, Minnesota, for all characteristics. The most
powerful technique applied to these populations being allozyme electrophoresis indicates
that these two populations are nearly identical in allele frequencies. Intuitively, of the
two mainland populations compared to Isle Royale, Cook County is the most likely
location for gene flow to and from. These analyses would indicate that there is sufficient
gene flow between Isle Royale and the adjacent mainland to prevent any genetic

differentiation.

30

The Beaver Island population of P. canadensis is also very similar to its mainland
counterparts for each of the analyses performed. Beaver Island however differs
significantly from the Dickinson County population for wing morphometrics. Both
forewing length and black bandwidth have been shown to be significantly different (p <
0.001 and p=0.018 respectively) between these two populations. Beaver Island is not
significantly different from its more adjacent mainland counterpart population,
Charlevoix County, for these wing characteristics. This lack of consistent differences
between its mainland comparison populations suggests that any differentiation might not
be due to an island effect and might better be explained using an alternative hypothesis.

The striking feature of the P. canadensis assayed from the Beaver Island
population is the allele frequencies represented (Table 2.3). The Beaver Island samples
show the population being fixed at a single allele (-125) for the Pgd locus. This is in
great contrast to both of the mainland comparison populations that exhibit -150, -125,
-100, -90, -80, and -50. Dickinson and Charlevoix counties are both represented by the
largest allele diversity of any of the populations under scrutiny. It has been suggested
that a common phenomenon in isolated populations of finite effective size, is a marked
decrease of heterozygosity (Carlquist 1974, Jacquard 1974, Hartl 1988, Grant 1998).
However, the sample size taken from Beaver Island is relatively small. In fact it is the
smallest sample size in this study. Perhaps an increased sample size would show this

apparent homozygosity as being an artifact of small sample size.

31

The population of P. canadensis on South Manitou Island is the most intriguing
portion of this island study. This island population has been shown to be significantly
different from all of the mainland populations it has been compared to. Early on in this
investigation, allozyme electrophoresis on the Pgd locus suggested that there was a high
level of genetic introgression, from a closely related southern species, P. glaucus. The
Pgd -100 allele is diagnostic in distinguishing between P. canadensis and P. glaucus
(Hagen et al. 1991). This allele has been shown to occur at relatively high frequencies in
the South Manitou Island population (Table 2.3). High levels of introgression from P.
glaucus into this canadensis population would help explain the observed island versus
mainland differences for each of the characters studied. Further discussion and
investigation of this introgression on South Manitou Island and the adjacent populations

are the focus of Chapter 3.

32

CHAPTER 3:
ISOLATED “HYBRID SWARM”: INTROGRESSED GENES OF PAPILIO
GLA UC US IN A P. CANADENSIS POPULATION FAR BEYOND THEIR

HYBRID ZONE

Introduction

Introgression is the process by which alleles are exchanged from one species into
the gene pool of another. Introgression occurs as a result of hybridization. Hybridization
is the production of offspring by parents from populations that are diagnosably distinct
for one or more characters. A hybrid is an individual that is heterozygous (intermediate)
for any one or more of these characters. A “hybrid swarm” is a localized area of
individuals exhibiting a diverse array of recombinant types (Harrison 1993).

It has been generally assumed that hybrids are unfit relative to parental types, and
that they are evolutionary dead ends. However, recently this view has been challenged
(Amold & Hodges 1995). Hybrid vigor has been observed for certain traits in lab reared
crosses of Papilio glaucus and P. canadensis (Scriber et al. 2001). It is believed that
hybrid zones are maintained by a balance of selection and dispersal (Porter et al. 1997).

Papilio glaucus and P. canadensis are closely related butterflies, but are distinct
species (Hagen et al. 1991). Several diagnostic characteristics (morphologic, ecological,
physiological, and biochemical) are extremely useful in distinguishing between these two
species (Table 3.1; Scriber 1990). Where the ranges of these two butterflies meet, a very

narrow zone of hybridization occurs. In Michigan this hybrid zone is between 43° and

33

Table 3.1. Summary of selected species differences discussed between Papilio glaucus
and P. canadensis. (Modified from Table l in Scriber 1990).

 

 

Characteristic glaucus canadensis
(Morphological)

Adult size (forewing length) Long Short
Hindwing anal cell black band Narrow Wide

 

(Ecological/Physiological)

 

Tulip tree oviposition preference Yes No
Quaking aspen oviposition preference No Yes
Tulip tree detoxification ability High Low
Quaking aspen detoxification ability Low High
(Biochemical)

Pgd (X-linked) allozymes PGD -50, -100 PGD -80, -125
th (X-linked allozymes LDH 100 LDH 40, 8O
Hk(autosoma1) allozymes HK 100 HK 110

 

34

44° latitude and has been stable for at least two decades (Scriber 1982; Scriber et al.
1996; Figure 3.1).

Papilio glaucus and P. canadensis introgression has been documented for sex-
linked and autosomal diagnostic allozyme loci (Pgd, th, and Hk) (Hagen et al. 1991).
Introgression of P. canadensis into populations of P. glaucus has been suggested as a
possible explanation of the “spring form” of Papilio glaucus described throughout
Eastern North America (Scriber 1990). More recently, introgression of P. glaucus
mtDNA has been described in populations of P. canadensis (Stump 2000). However, the
majority of this introgressed hybridization between P. canadensis and P. glaueus
described has been noted in locations close to the described hybrid zone (e.g. Isabella
County and Mason County).

Chapter 2 indicated that the population of Papilio canadensis found on South
Manitou Island exhibits significant differences, both morphological and biochemical,
from the surrounding mainland populations of P. canadensis. Early on in this
investigation it was realized that the P. canadensis population on South Manitou Island
exhibited glaucus-like traits. South Manitou Island is approximately 150 kilometers
north of the canadensr's / glaucus hybrid zone in Michigan. High levels of introgression
have not yet been described at such a great distance from the canadensis / glaucus hybrid
zone.

The extent to which P. glaucus introgression was present on and around South
Manitou Island was investigated through analysis of several diagnostic characters

between P. canadensis and P. glaucus. A combination of morphologic, ecological and

35

 

Figure 3.1. The shaded region in Michigan roughly indicates what has historically been
considered the hybrid zone between Papilio glaucus and P. canadensis (Nielsen, 1999).
The shaded region across Wisconsin into Minnesota represents the 50 year average
degree day accumulation and northern limit allowing for two generations of Papilio
glaucus. The star indicates the location of South Manitou Island.

36

biochemical traits were considered using samples taken over a three-year period (1998-

2000).

37

MATERIALS AND METHODS

Specimen Acquisition and Transport

Techniques for Papilio collection and processing follow those used in Chapter 2.

Additional specimen collection sites and sample sizes are listed in Table 3.2.

Specimen Processing

Male specimens were preserved by freezing them alive in an -80° C ultra-low
biological freezer. Male specimens that were to be utilized for laboratory hand-pairings
were maintained in a 4° C refrigerator, and frozen (-80° C) alive at a later time. In order
to extend their mating potential and vigor, these were fed a solution of honey water,
sodium, and amino acids every other day. This was accomplished by extending their
proboscis with a straightened paper clip. The tip of the proboscis was then placed into a
teaspoon of the 20% honey/water/mineral solution held in a plastic spoon. Individuals
would readily feed for up to 15 minutes. Feeding individuals were left under a small
screen cage until they were finished drinking and left the feeding station, after which time
they were placed back into the Glassine envelopes and returned to the refrigerator. Prior
to and after feeding, these individuals were given approximately 30 minutes to acclimate
to room temperature and to digest their meals. After utilization for hand pairing, these
males were also frozen alive at -80° C.

Upon arrival, female specimens were processed by recording their capture date
and location, and given a “Mother Number ID” by which we could later label and

identify that individual’s offspring. Female specimens were fed daily, but on a 20%

38

Table 3.2. Additional sampling sites from 1998-2000. Male and female sample sizes and
dates of collection are provided for each sample location.

1998

1999

2000

Location

Oscoda Co., Michigan
Isabella Co., Michigan
Lawrence Co., Ohio

Emmet Co., Michigan
Charlevoix Co., Michigan
South Manitou Island
Benzie Co., Michigan

South Manitou Island
North Manitou Island
Leelenau Co., mainland
Charlevoix Co., Michigan
Benzie Co., Michigan
Oscoda Co., Michigan
Mason Co., Michigan
Isabella Co., Michigan

Males

13
50
22

24
8
120
7

100
96
68
33
11
21
l7
14

39

Females

fl

ooao—ooa—u.

Dates collected

17 May
18-24 May
14 May

30 May, 8 June
8 June

18-19 June

27 May

10-11, 26-27 June
9 June
3 June
3 June
3 June
9 June
2 June
4 June

honey water solution. Females were kept alive and utilized for oviposition preference

assessment for sometimes up to 10 days. When female specimens were too weak to hold
onto an extended finger or immediately after their deaths, they too were preserved at -80°

C.

Wing Morphometrics

Two wing characteristics were chosen for analysis in order to identify the
presence of Papilio glaucus introgression into the South Manitou Island population of P.
canadensis. F orewing length and hind wing anal cell black bandwidth are morphometric
characters of adults which can be used in the field to help distinguish between P.
canadensis and P. glaucus. On the average, P. glaucus forewings are significantly larger
than those of P. canadensis. P. glaucus forewings have been shown to be 8-10 mm
longer, from thoracic attachment to tip, than P. canadensis forewings (Hagen et a1 1991).
The more powerful diagnostic morphometric wing character is the width of the black
band along the anal margin of the hindwing. For P. glaucus males, this band fills 10 to
50 percent of the width from the wing margin to the CuA2 vein; whereas for P.
canadensis the width of the band fills 50 to 90 percent of this anal cell (Hagen et al.
1991). Hybrid individuals display a black bandwidth intermediate between the two,
averaging 50 percent (Scriber et al. 2001). Interrnediacy for genetically based
morphometric traits is a good indicator of a hybrid individuals or populations (Harrison
1993).

The two wing traits were compared for all males captured from South Manitou

Island, a pure canadensis population in Oscoda County, and a pure glaucus population

40

from Lawrence county, Ohio. Methods used to score and statistically analyze these two

traits follow that used in Chapter 2.

Oviposition Host Preference Assessment

South Manitou Island female oviposition host preferences, for 1998-2000, were
assessed in a 3-choice arena (see Scriber 1993). Individual females (1998 n=24; 1999
n=52; 2000 n=50) were placed in clear polystyrene circular ventilated dishes, 10” in
diameter and 4” in depth. The bottom of the dish was lined with a sheet of Acclaim
“Natural” paper toweling. The dish contained leaves of three natural Tiger Swallowtail
butterfly host plants [Quaking Aspen (Populus tremuloides), Tulip tree (Liriodendron
tulipifera), and Black Cherry (Prunus serotina)], of approximately the same quantity,
equally spaced andrandomly placed around the outside edge of the arena. Host choices
were kept turgid in the arenas by supporting the petioles in rubber-capped plastic florist
aquapics containing water. Arenas were placed on turntables that rotated the dishes
approximately once every 5 minutes. These turntables were situated with 100 watt
incandescent lamps, approximately 0.5 meters away from one side, that were on a timed
photoperiod of six hours on, six hours off. This arena set up was housed in a room with
no natural lighting. While rotating, the females would be attracted to the side of the arena
that the light was then shining upon. She would there encounter each of the three host
plants in turn.

Approximately the same time each day, females were removed and fed. At this
time eggs that had been laid since the previous day were collected and recorded. Only

eggs that were directly laid upon one of the three host plant choices were recorded as

41

such. Eggs that were laid on the sides of the dish and/or the paper towel were recorded as
“other”. Many of these were within “reach” of the female abdomen while forelegs were
touching the leaf. Eggs were removed from the arena by cutting away the leaf surface or
paper towel that they were attached to using small scissors. Those attached to the sides
of the plastic dish were gently removed with the tip of a finger, after first loosening them
with a bit of fresh water expelled from a water bottle. At this time, host plants were
replaced as needed. Collected eggs were placed in appropriately labeled 150mm
diameter Lab-Tek® polystyrene petri dishes lined with paper towel. These were then
incubated in Percival® Growth Chambers at 23° C (18:6 photo/scoto-phase) and
monitored daily for hatching.

Host plants were collected from various sites near Michigan State University
campus in Ingham County, approximately every other day. Only the most vigorous and
least blemished leaves available were collected. In the lab, cut host plants were
maintained in a bucket of water, covered in large black plastic bags, and kept in a dark
cold (4° - 6° C) storage unit to prevent desiccation.

Oviposition “preference” was only assigned to females laying > 10 eggs total, and

> 50% of these eggs were laid on the same host plant.

Larval Survivorship

1999 and 2000 larvae were assayed for differential first instar survivorship on the
three host plants. Incubating eggs were monitored daily for hatching. Eggs hatched after
approximately 3 - 6 days. As eggs hatched, the neonate larvae were equally and

sequentially distributed on each of the three host plants, Black cherry, Quaking aspen,

42

and Tulip tree (1999 n=146 eggs; 2000 n=736 eggs). First instar larvae were carefully
picked up using the moistened fine tip of a camel hair paint brush. This was
accomplished using a gentle rolling motion of the brush tip. Larvae were then placed on
fresh host plants in appropriately labeled 150mm petri dishes lined with paper towel. No
more than 6 first instar larvae were placed in the same petri dish. Host plants were kept
turgid and fresh by placing them in water filled aquapics. Larvae were checked for
survival after 5-6 days for each petri dish, which generally included the duration of the
first instar.

Survival was analyzed in two different ways. The percent of first instar larvae
from each individual family surviving on each host plant was recorded, and then the
mean family percent survival was calculated across all families. In addition, the total
number of first instar larvae surviving on each host was recorded in order to calCulate the
total percent survival across the population. For P. canadensis it was expected to have
high first instar survivorship on both Quaking aspen and Black cherry. P. canadensis
does not however have a strong ability to detoxify Tulip tree. This being the case, any P.

canadensis survival on Tulip tree was considered to be significant.

Allozyme Electrophoresis

Allozyme electrophoresis was performed following the same techniques and
analysis outlined in Chapter 2. Analysis was performed for male samples from 1998,
1999, and 2000. In 1998 electrophoresis was performed for the diagnostic Pgd allozyme
for all populations. Additional diagnostic allozyme loci (th and Hk) were assayed for

South Manitou Island, Isle Royale, Beaver Island, and Oscoda County in 1998. Male

43

specimens collected from all populations in 1999 were assayed for three species
diagnostic allozyme loci. Male specimens collected from all populations in 2000 were
assayed only for Pgd and I-Ik loci. After previous analysis of the th locus for hundreds
of specimens that indicated absolutely no deviation from the expected canadensis allele,
it was decided to not spend valuable time and resources on further population wide
analysis of this locus. To be sure that no primary hybrids were present in any population,
only specimens that scored glaucus-like for Pgd were also analyzed and scored for the
th locus. Allele frequencies are reported as percentages of canadensis vs. glaucus

alleles for each allozyme locus encountered in the populations described.

RESULTS

Wing Morphometrics

F orewing length for South Manitou Island (mean = 46.7 i 0.49 s.e.) did not
significantly differ from that of the canadensis population in Oscoda County (mean =
46.0 :i: 0.92 s.e.) but did differ significantly from that of the glaucus population fi'om
Lawrence county, Ohio (mean = 49.0 :t 0.53 s.e.) using both Tukey-Kramer and
Student’s t test (Figure 3.2). South Manitou Island black bandwidth (mean = 55.2 % i:
1.29 s.e.) was found to be significantly different from those of both Oscoda (mean = 63.4

% i 1.71 s.e.) and Lawrence (mean = 35.5 % :t 1.94 s.e.) counties (Figure 3.3).

Oviposition Preference

The results for oviposition preference for 1998-2000 are presented in tables 3.3,
3.4, and figure 3.4. In 1998, of 24 South Manitou Island females assayed, eight females
laid >10 eggs in 3-choice arenas. Of these eight, only five displayed >50% preference for
any single available host plant option. Of these five, four females exhibited oviposition
preference for Tulip tree. From all 24 females assayed, 301 eggs were oviposited in total.
The greatest portion of this total (56%) was oviposited on Tulip tree. The next most
frequently chosen host plant was Quaking aspen (20%).

In 1999, of the 52 South Manitou Island females assayed, 18 females oviposited
>10 eggs. Of these 18, 15 displayed >50% preference for any single available host plant
option. nine of these 15 females preferred to oviposit on Quaking aspen, while six

preferred Tulip tree. From the 52 females assayed, 1161 eggs were oviposited in total.

45

Forewing Length (mm)

 

55

50" '

 

451 - -

Hr!
IIIIIH III-I

 

 

 

I I
Oscoda Lawrence co., Ohio S. Manitou

Population

Figure 3.2. Forewing length comparison between P. canadensr's fiom
South Manitou Island, Oscoda co., and a population of P. glaucus from
Lawrence co., Ohio. Mean forewing length for each population is shown
with error bars indicate :1: s.e. (S. Manitou = 46.7 :1: 0.49, Oscoda co. =
46.0 d: 0.92, P.g. Ohio = 49.0 :1: 0.53). Comparison using both Tukey-
Kramer and Student’s t-test at alpha value = 0.05 indicates that there is
no significant differences between S. Manitou and Oscoda co. while there
is a significant difference between S. Manitou and Lawrence co., Ohio
(p-value = 0.003).

46

 

 

so

70 - - '
at I I
g 60 ~ _
O
:3 so " . ' ii
.3 : '
3 .. I '
m C
a 30 "
.s
m I

20 - .

10

I I

 

 

 

Oscoda Lawrence co., Ohio S. Manitou

Population

Figure 3.3. Black band width comparison between P. canadensis from

South Manitou Island, Oscoda co., and a population of P. glaucus from
Lawrence co., Ohio. Mean black band widths for each population are shown
with error bars indicating :1: s.e. (S. Manitou = 55.20 i 1.29, Oscoda co. =

63.38 i 1.71, P.g. Ohio = 35.50 :t 1.94). Comparison using both Tukey-Kramer
and Student’s t-test at alpha value = 0.05 indicates that South Manitou differs
significantly from both Lawrence co. and Oscoda co. (p-values < 0.001).

47

Table 3.3. South Manitou 3-choice oviposition preference for individual female
1998-2000. Only females ovipositing >10 eggs were used for this analysis.
The total number of eggs oviposited by each female is indicated as well as the
percentage of the eggs placed on each host plant. Percentages >50% are in
bold print indicating that that female expressed a "preference" for that particular
host plant. The other category includes many eggs that were placed within 2 cm
of the host plant and may have been intended for that host. For the purposes of
this study they have been scored as other.

Percent of eggs per host

(n) # of
1998 eggs laid Black Tulip Quaking ‘ Other
cherry tree aspen
Mother ID
14220 13 0 23 62 15
14222 16 0 75 13 12
14223 36 25 47 28 0
14225 28 0 82 18 0
14226 46 28 41 28 3
14229 60 15 73 12 0
14230 71 13 55 18 10
14239 10 30 20 40
Mean i s.e. 13.9 :t 4.6 52.0 :h 8.3 27.4 :1: 5.9 6.3 d: 2.2
Percent of eggs per host
1999 (n) # of Black Tulip Quaking " Other
eggs laid cherry tree aspen
Mother ID
15227 98 13 61 17 8
15228 23 0 4 96 0
15234 13 0 69 8 23
15239 49 2 76 2 20
15242 22 0 9 82 9
15243 51 4 27 59 10
15250 10 0 10 70 20
15251 54 6 52 15 28
15253 117 18 40 21 21
15254 117 3 17 38 42
15261 103 6 81 4 10
15264 118 4 6 77 13
15265 14 7 14 50 21
15266 110 1 7 84 7
15267 69 O 39 39 22
15269 21 29 0 52 19
15272 62 15 73 5 8
15273 51 8 14 61 18
Mean t s.e. 6.4 i 1.8 33.3 a: 6.7 43.3 at 7.3 16.6 d: 2.3

48

Table 3.3 continued

Percent of eggs per host

(It) # of
2000 eggs laid Black Tulip Quaking "‘ Other
cherry tree aspen
MotherID
16136 56 0 50 32 18
16137 81 20 63 6 23
16175 21 19 71 5 5
16177 90 7 59 ll 23
16178 79 19 54 14 13
16181 14 7 7 36 50
16183 61 11 10 61 18
16185 36 ll 17 44 28
16190 41 5 41 20 34
16191 131 2 7 45 47
16192 100 19 39 12 30
16197 34 3 65 18 15
16198 64 28 41 25 6
16199 53 28 25 26 21
16202 190 14 59 13 14
16215 11 9 0 73 18
16219 58 3 34 41 21

Mean i s.e. 12.1 i 2.1 37.8 i 5.6 28.4 i 4.7 22.6 :t 3.0

49

Table 3.4. South Manitou population total 3-choice oviposition preference
1998-2000. The eggs of all females that laid any eggs (>10 and <10) were
pooled to provide an indication of the total population oviposition
preference. Total number of eggs laid by all females is provided as well
as the number of females that were assayed.

(n) # of
eggs laid

1998
Population
totals 301
n=24 females

1999
Population
totals 1 l 61
n=52 females

2000
Population
totals 1 120
n=50 females

Percent of eggs per host

Black Tulip Quaking Other
cherry tree aspen

16% 56% 20% 8%

7% 36% 40% 1 7%

12% 41% 25% 22%

50

Mona 95.6130. .-
>55 seem a
8; 9.3 a

 

 

ammo 02 T: ammo 5:“: ammo Sm":
mmwo Ea— moREou mm": ammo Ea— mo-«Eom an": mwwo Ea— muBEom 2n:
83:5.“ on u: 838$ an": 838$ van:

ooom mmmw

hi ... . .. 1 t. .
.. :3. .. 1:2...

mam?

II... .

   

 

   

 

 

 

 

 

.oocouomoa 536035 86:04“ 58 coca—smog 50:52 Snow in onE

 

$0

o\oON

$9
:3 Hana
2.3.—o.—

{com

{com

Ax500 F

51

The greatest portion of this total was oviposited on Quaking aspen (40%), closely
followed by Tulip tree (36%) with the rest on Black Cherry (7%) or “other” (17%).

In 2000, of the 50 females assayed, 17 females oviposited >10 eggs. Of these 17,
only eight exhibited >50% host preference. Of these eight females, seven displayed an
oviposition preference for Tulip tree. From the 50 females assayed in 2000, 1120 eggs
were oviposited total. Of this total, the greatest portion was placed on Tulip tree (41%)

followed by Quaking aspen (25%).

Larval Survivorship

In 1999, of the 146 larvae assayed for detoxification abilities across the three host
plant options, 52% of those placed on Tulip tree survived past the first instar, while 82%
and 75% survived on Black cherry and Quaking aspen respectively (Fig. 3.5). When
considering the average survival of all females, the mean family percent survival was as
follows: Tulip tree - 49%; Black Cherry — 82%; Quaking aspen — 68%. In 2000, of the
736 larvae assayed, 18% of those placed on Tulip tree survived, while 58% and 65%
survived on Black cherry and Quaking aspen respectively. Mean family percent survival

rates were Tulip tree —- 20%; Black cherry — 61%; Quaking aspen 69%.

Allozyme Electrophoresis

In 1998, the South Manitou Island population of Papilio canadensis exhibited
unusually high levels of glaucus alleles at the Pgd locus (10.9%). This is markedly
higher than any other Michigan population sampled in 1998; Charlevoix (1.8%), Mason
(0.9%), and Isabella counties (4.0%). The South Manitou Island population also

exhibited significant levels of glaucus Hk allele introgression (4.7%). Of the populations

52

Figure 3.5. South Manitou larval host plant survival 1999-2000. Approximately
equal numbers of the eggs from each female that laid >10 larvae were distributed on
each of the host plant options (Tulip tree, Black cherry, and Quaking aspen). The
mean family % survival is the average percentage of larvae that survived through the
first instar on each host plant, where as the total % survival is the percentage of all
assayed larvae that survived on each host plant.

Year Host Plant

1999 Tulip tree
Black Cherry
Quaking Aspen

2000 Tulip tree
Black Cherry
Quaking Aspen

Tulip tree

Total (n)
(n) % Survival # of families
larvae
50 52 10
56 82 10
40 75 7
255 18 18
232 58 18
249 65 18

Black Cherry

53

  

Quaking Aspen

Mean Family
% Survival

48.8:t10.7 s.e.
92.2i 3.8 s.e.
67.5i13.1 s.e.

20.2:1: 5.3 s.e.
58.3:t 7.5 s.e.
68.7:1: 6.2 s.e.

 

assayed for Hk, South Manitou was the only population that showed any levels of
introgression at this locus (Table 3.5, Fig. 3.6). No populations assayed for th,
including South Manitou Island, showed any level of glaucus type alleles for this locus.

The 1999 samples assayed appeared similar in allele frequencies to those in 1998,
with South Manitou Island exhibiting 7.9% glaucus alleles for Pgd (Table 3.5, Fig. 3.7).
In addition, the adjacent mainland population of P. canadensis of Leelenau County
exhibited unusually high levels of glaucus alleles (5.2%). None of the other surrounding
counties sampled and assayed had any glaucus alleles for the Pgd locus. There were
however, low levels of glaucus alleles found for the Hk locus in most of the populations
surveyed: South Manitou (5.4%); Leelenau (3.4%); Emmet (4.0%); Benzie (7.1%);
Charlevoix (0.0%). Again in 1999, there were no glaucus alleles (LDH 100) found for
the th locus in any population assayed including South Manitou Island.

The 2000 samples (from a wider geographic area) showed populations exhibiting
glaucus alleles for both Pgd and Hit (Table 3.5, Fig. 3.8). South Manitou Island
continued to have the highest levels of glaucus Pgd alleles (9.5%). Again in 2000, for all

populations assayed, there were no glaucus alleles found for the th locus.

54

Table 3.5. Allele frequencies for three diagnostic loci present in each of the study
populations from 1998-2000. Values presented are percentages of Canadensis versus
glaucus alleles detected. Dashes indicate that electrophoresis was not performed for that

locus for that population.

1998

Population
(I!)

Cook Co., MN
(35)

Isle Royale
(23)

Gogebic Co.
(36)

Dickinson Co.
(43)

Beaver Island

(11)

Charlevoix Co.
(50)

South Manitou
(32)

Isabella Co.
(50)

Oscoda Co.
(13)

Mason Co.
(50)

1.00

1.00

1.00

1.00

0.983
0.018

0.891
0.109

0.96
0.04

0.992
0.009

Allele
(canadensis)
(glaucus)
th

1.00

1.00

55

Hk

1.00

1.00

0.953
0.047

Table 3.5 continued.

1999

Population Pgd
(n)

Emmet Co. 1.00
(25) .0

Charlevoix Co. 1.00
(8) .0

South Manitou 0.921
(120) 0.079

Leelenau Co. 0.948
(29) 0.052

Benzie Co. 1.00

(7) .0

Allele
(canadensis)
(glaucus)

th

1.00

.0

1.00
.0

1.00
.0

1.00
.0

1.00

56

Hk

0.96

0.04

1.00

0.946

0.054

0.966
0.034

0.929
0.071

Table 3.5 continued.

2000

Population
(I!)

North Manitou Island
(96)

South Manitou Island
(100)

Leelenau Co.
(68)

Benzie Co.

(11)

Charlevoix Co.
(33)

Oscoda Co.
(21)

Mason Co.

(17)

Pgd
0.937
0.063

0.905
0.095

0.949
0.051

0.909
0.091

0.96
0.04

0.971
0.029

Allele

57

(canadensis)
(glaucus)

Hk
0.932
0.068

0.93
0.07

0.912
0.088

0.955
0.045

0.97
0.03

1.00

0.912
0.088

  
   
     

  
 

   

       

._ ,A
' ‘1‘»..‘3

   
   

  
  

   

Isle Royale “-3 .' canadensis
“ ’ 2 I glaucus

 

(mu

4?
Dickinson
n = 48

Figure 3.6. 1998 populations sampled. Pie charts indicate the percentage of glaucus
alleles in each population for the Pgd allozyme locus.

58

    
     

 

 

 

 

 

 

 

 

 

 

 

 

“tum
rams:
wworo Insurance restoration oetuaw
11mm
anon IAKE escrou curt euowm
P—n—r
MY
OCEANA atcosn Ismua arouse
NEWAYGO ~1__1

 

 

 

 

 

Figure 3.7. 1999 populations sampled. Pie charts indicate the percentage of glaucus
alleles in each population for the Pgd allozyme locus.

59

    
 

El canadensis
I glaucus e
_____j

    

OIEWYGAN

    

 

 

 

 

   

 

 

 

 

 

 

 

 

 

 

  

Leelenau'
n=68 r
ntusn cuwroro
HISSAUKEE toscouuon comm
AIENM
oscrou am Gtaowm
,_..._
\ m
. \
ocmu utcosu Isabella .11 memo
NEWAYGO “=14 / ”'L__,

 

 

 

 

 

 

Figure 3.8. 2000 populations sampled. Pie charts indicate the percentage of glaucus

alleles in each population for the Pgd allozyme locus.

60

DISCUSSION

Based upon the combined analysis of morphological and biochemical traits, I
have found measurable evidence suggesting that there is a significant amount of P.
glaucus introgression in the South Manitou Island population of P. canadensis. The
presence of individuals heterozygous (intermediate) for even single diagnostic markers
suggests hybridization. The South Manitou Island population exhibits heterozygosity
(intermediacy) for multiple diagnostic characters, resulting in a “diverse array of
recombinant types”. This diversity of mixed ancestry can best be described as a “hybrid
swarm” (Harrison 1993). This hybrid swarm however, is unusual in that it appears to be
isolated at a latitude approximately 150 kilometers north of the historically described P.

glaucus and P. canadensis hybrid zone (Scriber 1982; Scriber et al. 2001).

The most readily available distinguishing morphological markers between P.
glaucus and P. canadensis are wing characters. Both forewing length and the relative
width of the hindwing anal cell black band are consistently described as being diagnostic
between the two species (Scriber 1982; Luebke et a1. 1988; Hagen et al. 1991; Nielsen
1999). The South Manitou Island population of P. canadensis exhibits interrnediacy for
both of these qualitative (autosomaly inherited) characters.

Papilio glaucus has a significantly larger forewing length than does P. canadensis
(Hagen et al. 1991). Forewing length in these Papilio species has been shown however
to be impacted by ecological environmental conditions (Ayres and Scriber 1994) and
generally correlated with latitude from Alaska to Florida (Scriber 1994; Chapter 2, Figure

2.4). Statistical analysis shows that in 1998 the forewing lengths of P. canadensr’s from

61

South Manitou Island were significantly smaller than a known population of P. glaucus
from Ohio but were not significantly larger than a population of P. canadensr‘s from
Oscoda county, Michigan (Figure 3.2). This lack of difference of forewing lengths,
between the South Manitou population and the Oscoda canadensis population does not
support the hypothesis of glaucus introgression, but should be further scrutinized.
However, the South Manitou Island population displayed the largest mean wing length of
any canadensis population surveyed in 1998 (Figure 2.4) except Mason county, which
lies at the northern most border of the known hybrid zone. A combination of
environmental selection pressures, latitude, and glaucus introgression could be
influencing this trait.

More indicative of glaucus introgression is the intermediacy of the hind wing anal
cell black bandwidth displayed in the South Manitou Island population. This black
bandwidth is the more consistent trait used to distinguish between P. glaucus and P.
canadensis in the field. It too may be influenced by environmental conditions and
latitude (Figure 2.6) but has been shown to have a significant genetic component that will
be intermediately expressed in lab reared interspecific hybrid crosses. Lab reared
primary hybrids from many families have a mean black bandwidths that fall between
42% and 57% (Scriber 1982). The South Manitou Island population black bandwidth
was significantly different from, and intermediate between both the Ohio glaucus
population, and the Oscoda county canadensis population located at the same latitude
(Figure 3.3). The mean black bandwidth was 55.2%. This intermediacy between the two
species for this morphologic trait is a documented method of identifying hybridization

and therefore would be indicative of glaucus introgression in the South Manitou Island

62

population. The selective advantage of such black band (melanic) widening with latitude
is not clear, although a role in thermoregulation can’t be ruled out (Watt 1968,

Kingsolver 1985, 1987, 1995).

Reciprocal host plant oviposition preferences and larval detoxification abilities
have long been described for P. glaucus and P. canadensis, even before they were given
distinct species status (Hagen et al. 1991). Given a choice of Tulip tree, Quaking aspen
or Black cherry, P. glaucus females prefer to oviposit on Tulip tree, whereas P.
canadensis prefers to oviposit on Quaking Aspen. This genetically determined choice
preference appears to be a sex-linked trait contained on the X chromosome (Scriber et al.
1991; Scriber 1994). Larvae of P. glaucus develop poorly on Quaking aspen, a good host
for P. canadensr’s, due to an inability to detoxify the phenolic glycosides that are present
(Scriber et al. 1989). Conversely, Tulip tree is toxic to P. canadensis larvae, preventing
growth, whereas it is successfirlly utilized by P. glaucus (Hagen et al. 1991). The
particular toxin for P. canadensis in Tulip tree is not known at this time. Hybrid larvae
can use both species (Scriber et al. 1995), and backcrosses have intermediate
detoxification and growth rate abilities (Scriber et. 1989; Scriber et al. 1999).

During the years assayed (1998-2000), the South Manitou Island population of P.
canadensis exhibited strong oviposition preference for Tulip tree and larval ability to
detoxify and utilize this host plant (Table 3.3, Figures 3.4 and 3.5). In 1998, 80% of the
females that exhibited a strong oviposition preference chose Tulip tree. 56% of all eggs
laid in oviposition preference tests were laid on Tulip tree. In 1999 the results were not

as striking but are still highly unusual for a canadensis population, with 40% of the

63

females exhibiting a strong oviposition preference choosing Tulip tree. Of the total eggs

laid from that population 40% were placed on Quaking aspen followed closely behind by

36% being placed on Tulip tree. Again in 2000 a strong Tulip tree preference is shown.
87.5% of all females exhibiting a strong preference, and 41% of the total eggs were
placed on Tulip tree. Most P. canadensis have much lower preference for Tulip tree in
the 3-choice arena and no P. canadensis larvae from Canada, Wisconsin or Michigan
survived on Tulip tree in previous studies (Scriber et al. 1991).

These population analyses indicate that a large portion of the total eggs oviposited
by females from the South Manitou island population are placed on Tulip tree. However,
upon closer analysis of the oviposition preferences of each individual female assayed, a
clear dichotomy becomes apparent. The vast majority of individual females that
oviposited more than 10 eggs showed a clear preference for either Tulip tree or Quaking
aspen (see Appendix Table 2.1). This strong dichotomous preference could be explained
through the population having mixed ancestry of P. canadensis and P. glaucus.

The results of the larval survival assay for South Manitou Island are also
suggestive of glaucus introgression. In 1999, a striking 52% of the total larvae assayed
survived on Tulip tree. In 2000, this number was only 18%. Though greatly reduced,
this is still of ecological significance as the expected survival of P. canadensis larvae on
Tulip tree is 0.0%. It should be noted that population larval survival would not
necessarily directly follow the trend in oviposition preference. Oviposition preferences
for Quaking aspen or Tulip tree is genetically determined basically by a single locus on

the X-chromosome, while larval detoxification abilities are more complex and polygenic

in derivation (Scriber 1994).

Like with the oviposition data, attention should be given to the survival rates of
individual fanrilies (see Appendix Table 2.2). In 1999, out of 10 families assayed for
larval survival on Tulip tree, there were only two families that had 0.0% survival. In
2000, only 40% of the families assayed had 0.0% survival rates. For both of the years
that larval survival was assayed on the three host plants, 1999-2000, there is a strong

indication that larvae from South Manitou females are able to survive on both Quaking
aspen and on Tulip tree. The ability for larvae to survive on these two host plants follows

the host plant utilization abilities of lab reared canadensis /g1aucus hybrids (Scriber
1994)

The most convincing evidence of significant levels of glaucus introgression into

the South Manitou Island population of P. canadensis are the results of the allozyme
electrophoresis analysis. Allozyme electrophoresis is frequently utilized to distinguish
between closely related species and to monitor introgression in and around hybrid zones
(Scriber et al.1992; Harrison 1993; Johnson et al.1996; Jiggins and Mallet 2000),
including P. glaucus and P. canadensis (Hagen et al. 1991; Hagen & Scriber 1991). The
South Manitou Island population of P. canadensis shows significant levels of glaucus
introgression at two key diagnostic loci, Pgd and Hk, for every year surveyed. The level
of glaucus introgression for these two allozymes is relatively comparable. It would be
easy to anticipate that any other diagnostic allozymes should likewise exhibit comparable
levels of introgression. It could be expected that this would be especially true of the th
diagnostic allozyme since, like Pgd, th is an X-linked allozyme. Given that these two

allozymes are linked on the X chromosome, it would seem as though the levels of

65

introgression at these two loci should be highly correlated. This however is not the case
at all. After the electrophoretic analysis of all of the samples collected north of the
hybrid zone, including South Manitou Island, there is no indication of any th

introgression at all between 1998 and 2000 (see also Hagen 1990.).

In summary, this investigation has focused on multiple diagnostic traits for the
South Manitou population of P. canadensis. Independent analysis of each of these traits
indicates P. glaucus introgression. There are however other diagnostic traits between P.
canadensis and P. glaucus, for which introgression is not expressed (i.e. th allozyme).
In fact, after analysis of over 250 male specimens from South Manitou Island, no primary
hybrids appeared. This suggests two things. First, the hybrid zone is a semipermeable
barrier to gene flow. And second that the introgression has been present for multiple
generations.

Specific evidence that the hybrid zone must act as a semipermeable barrier to
gene flow is readily apparent from the differential introgression of certain diagnostic
traits. Consider the differential introgression of the Pgd and th allozymes. This is
especially significant in that they are both X-linked and should be highly correlated.
Similarly, Tulip tree host use abilities have moved extensively northward with the last
few years of regional climate warming, but aspen abilities have not moved southward
(Scriber 2001). These examples of differential movement of genes must be enforced by
. strong environmental selection. Reasons for this differential selection are still under

investigation.

66

It is very likely that historic periods of warming along with any “lake effect”
might produce a moderated environmental passageway along the Lake Michigan coast,
allowing the movement of glaucus genes. After the return of “normal” temperatures
across the state the South and North Manitou island environments are able to maintain
certain alleles due to their moderated climates. The growing season of the Leelanau
Peninsula and associated islands is actually comparable to that of Lansing with an
average of 157 frost-free days. Grayling, which is at the same latitude, has a growing
season of only 114 days. (Haswell and Alanen, 1994). The exhibited mixed ancestry is
then maintained as a result of the isolation of the islands preventing genetic swarnping
from the surrounding mainland gene pool.

In fact, South Manitou Island, and likely North Manitou Island, represents old
genetic introgression. There are several indications that this is the case. First, as
previously described, there have never been any primary hybrids encountered. A primary
hybrid would be a first generation offspring resulting from an intraspecific pairing
between P. canadensis and P. glaucus. Secondly, there are several specimens that scored
homozygous for glaucus-like alleles at certain allozyme loci (see appendix).
Additionally, the relatively high frequency of glaucus-like alleles at the Pgd locus,
accompanied by the complete absence of any individuals with glaucus-like alleles for the
other X-linked diagnostic loci (th), can most simply be interpreted as a chromosomal
crossover event that occurred, producing offspring with glaucus Pgd alleles (-100) and
canadensis th alleles (80). Evidence for such a crossover event on South Manitou
Island was provided several years ago through analysis of offspring resulting from a lab

hand cross pairing, of a lab reared P. glaucus female and a single field captured P.

67

canadensis from South Manitou Island in 1991 (Scriber, 1994). At the time of the 1991
study, the significance of the find was not fully recognized. A more comprehensive
population investigation was required to identify the extent of the glaucus-like

introgression into the P. canadensis population of South Manitou Island.

 

68

CHAPTER 4:

SUMMARY

To more clearly determine whether or not there is genetic differentiation between
the island and mainland populations of P. canadensis I suggest the use of additional
biochemical and alternative molecular markers. This study only utilized one highly
polymorphic locus (Pgd). Use of multiple allozyme loci would be far more powerful in
identifying genetic differentiation. Likewise, the use of alternative molecular markers
such as RAPDs (Randomly Amplified Polymorphic DNA) or AF LPs (Amplified
Fragment Length Polymorphisms) might be better suited to detect genetic differences.

Several aspects of this preliminary study indicates that further research is
necessary in order to truly evaluate whether or not there is island versus mainland genetic
differentiation. First, the Beaver Island population exhibited complete fixity for the Pgd
allozyme for a single allele (125). This is in great contrast to Charlevoix co., the most
adjacent mainland population. In 1998 Charlevoix co. exhibited the greatest allelic
diversity of any of the populations that were assayed. It is possible that the apparent lack
of allelic diversity on Beaver Island is the result of a low sample size (n=11) however it
seems unlikely that this alone would account for the great difference.

The more compelling aspect of this research that indicates a need for further
investigation is the odd finding of a “hybrid swarm” on South Manitou Island. The
hybrids found there are of mixed ancestry of Papilio canadensis and Papilio glaucus.

This thesis merely describes the existence of this oddity. There are questions that

69

logically follow remain unanswered. How has this “hybrid swarm” come to exist
approximately 150 kilometers north of the described canadensis /g1aucus hybrid zone?
Why has this population of mixed ancestry been maintained on the island for multiple

generations?

After three years, focusing on the South Manitou population and adjacent
populations of P. canadensis, through extensive analyses of multiple diagnostic traits I
have shown that there is a significant amount of genetic introgression not only on South
Manitou Island, but also on North Manitou Island and the adjacent mainland. In addition,
there is substantial evidence to suggest that this genetic introgression has been present
and / or occurring for at least the past 12 years.

The South Manitou Island population of P. canadensis can best be described as a
hybrid swarm (Harrison 1993) as it is a localized area of individuals exhibiting a diverse
array of mixed ancestry. How the high levels of introgression came to be present in this
population was not the focus of this study, however it seems extremely possible that it
has been facilitated by historic climatic changes. As a result of a lake effect moderating
the western shore of the State of Michigan, accompanied by periods of regional warming,
glaucus-like genes would have a narrow passageway north along the western shore.
When “normal” cool temperatures returned to the region, these glaucus genes would be
maintained on South Manitou Island as a result of the moderated temperatures that that
portion of the state consistently experiences. The glaucus genes in the mainland
population of P. canadensis would be eliminated through a swamping effect, while the

island population retained the mixed ancestry due to isolation.

70

As a result of the shift in focus of this research so early on, the original
question, whether or not there is significant genetic differentiation between the island

populations and adjacent mainland populations, was not fully attended to.

71

APPENDICES

72

APPENDIX 1:

RECORD OF DEPOSITION OF VOUCHER SPECIMENS

73

Appendix 1

Record of Deposition of Voucher Specimens‘
The specimens listed on the following sheet(s) have been deposited in the named museum(s) as
samples of those species or other taxa, which were used in this research. Voucher recognition
labels bearing the Voucher No. have been attached or included in fluid-preserved specimens.
Voucher No.: 2001-09
Title of thesis or dissertation (or other research projects):

ISLAND POPULATIONS AND TRAIT COMPARISONS OF TIGER
SWALLOWTAIL BUTTERF LIES, P. CANADENSIS, IN THE GREAT LAKES
REGION

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

Entomology Museum. Michigan State University (MSU)

Other Museums:

Investigators Name(s) (typed)
Gabriel J. Ordina

 

 

Date 10/4I01

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

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

Copies: Include 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.

74

APPENDIX 1.1:

VOUCHER SPECIMEN DATA

75

Appendix 1.1

Voucher Specimen Data

3 Pages

of

Page 1

2mm.

.anmas. 30.8.6.5
£935 23m $953 as c. .588
.6. 2063on 3%. 98m 05 3283.

.2230

 

room .m 52.2qu 200

 

 

 

 

moi—bow oz bozozo>

lmsEo ._. .953
82:: Ememz $23.82.
33608: z £35 .8868 $2

 

 

 

 

 

 

 

 

 

 

 

 

 

 

aan .<
82 .8 52
:92 w .8 xooo <._.Owwzz=2
9.65 .o
mam. .3 95..
.8 .6533
:22 m Ema. 38:22 5:8 z<0_10=2
QEBw .<
32 .8 as.
:92 u .8 :85on z<0=._o=2
9.20 .0
Saw .m 95..
:22 N .8 03.8 24.0.19:
9.65 .o
ooom .m 95...
:22 m .8 3:28..
95...... 22.5.2 582 z<0=._o=2 £26328 0:..an
m m 6 0+ e m e 3.583 new com:
3 s r .
m w m. m m m W W m w B 382.8 22:6on .8 Emu .33 :98. .050 .6 $6on
M w w o A A P N m E
«o 59:32

76

23 .2an

 

SON 1m .3900 200

 

Enema: 30.8.25
2.98.5 saw 58.8.... o... c. 888
.9 9.2.0on 02v... 0.63 on. 32801

 

 

8.20 ... .880

 

Appendix 1.1

Voucher Specimen Data

Page 2 of 3 Pages

moéoou oz .mzoao>

385 @952 30.3.7.9...
38800.. g 93...... .2363 $3.

 

:92

3m:

:92

:92

392

89.8 .2...
m8. .8 E...
25.5....
8.20 .o
82 .3 8....

8%. .258 25.5.5.
8.20 .o

88 .o 82.

.8 855m 2.5.10.5.
8.20 .o

88 .8 as.

.8 £33.

.00 imcmzfimx
{an .8262 «.83. 2.... 25.10....
8.20 .o
8.: Nu 8....

.8 9895 25.10....

£26358 9....me

 

Museum

where

deposited

 

Other

 

 

 

 

Nymphs

 

Larvae

Eggs

 

3.338 new com:
.o 8.02.8 88.8% .o. 98 .83

 

:96. 550 .o 3.0on

 

 

28 Adults 6
E Adults 9
z Pupae

 

 

77

Appendix 1.1

Voucher Specimen Data

Page g o

3 Pages

f

1.1".

0.00

3:03.). 30.0.5.5.
0.0.8.5 990 80.8.... 9.. c. .088
.2 0:0E.0000 00.0.. 0.600 0:. 0020001

.2050

 

poem .m .3200 0.00

 

 

 

moivocm 02 .0530)

m:.0.o .0 .0500
69...: .0982 898.88...
3000000: 0. 0.00:0 0:00.000 00:.

 

:92

am:

:92

Dws.

Dws.

.880 .2...
08. .3 .92

.00 8:92.64 O_IO
8.0.0 .0

82 .R .02

.00 30:900.. z<0....o__2
.880 .2...

82 .t .05.

.8 92.9.9.0 z<0.:0.s.
8.0.0 .0

88 .2 .02

.8 0880 25.10....
0000.0 .2 . 039.0 d

88 .2 .92

.8 88.2 z<0.:0.s.

00000.0 0.....00Q

0.0500000 0.....001

 

Museum
where

deposited

 

Other

 

Adults 6‘

 

 

 

Nymphs

 

Larvae

Eggs

 

09.00000 0:0 000:
.0 00.02.00 0:0E.0000 .0. 0.00 .000.

 

:0x0. .050 .0 00.0000

 

 

O
h

g Adults 9

z Pupae

 

 

78

APPENDIX 2:

POPULATION WING MEASUREMENTS AND ALLOZYME DATA

 

79

APPENDIX

Table 1. Specimen information by location and year. Each specimen
allozyme data presented represents two alleles. If only one allele is
indicated, the individual is homozygous for that allele. If two alleles are
indicated the individual is heterozygous for those two alleles. 1998 Pgd
allozyme data for Cook, taken from Stump 2000.

 

1998 Cooke County
Minnesota _
hi
Wings Allozymes
ID Forewing Hind Wing PGD
Male # Length Black Band F .

1 41 68.5 125

2 43 79 125

3 44 78.5 125

4 45 83.5 125

5 44 74 125/80
6 4O 65 150I125
7 46 75.5 125

8 44 70 125/80
9 42.5 68.5 125

10 44 72 125/80
11 43 69 125

12 43 80 125

13 43.5 59.5 125

14 44.5 76.5 125

15 44 72.5 125

16 48.5 66 125

17 43.5 66 125

18 42.5 63.5 125

19 41.5 65.5 125/80
20 46 78.5 125
21 44.5 67 125
22 4O 56 125
23 46 69 125
24 42.5 66 125
25 41 77.5 125
26 42.5 71 125
27 42 74 125
28 45 80 125
29 42 62.5 150/125
30 42.5 78 125

80

31
32
33

35

1998 Isle Royale

ID
Male #

NMNNNNNN-t-s-s-s—s-s-n—L—s—s
\ioamrsz—socoooxioamhwm-so‘om“@019de

29
Female

14109

46.5
43
42
45
44

46.5

41.5
46
44

46.5
45
43

44.5
40
44
40
40
47

42.5
46
43

43.5
45
46
44
41
43
43

42

45
42.5

45
43

Vans
Forewing Hind wing
Length Black Band

69
73.5
66.5
62.5

56
77.5

70
75.5

71
72.5

73
70.5

69

75

66
69.5
71.5
74.5

73
68.5
67.5

69
62.5
73.5
61.5

75

68

65

68

75.5
73.5
72
76
70.5

PGD

125
125
125
125
125
125/80
125
125
125
125
125
125
125
125
125/80
125
125/80
125
125

125/150

125
125
125
125/80
125
125
125
125/80

81

125

125
125/80
125
125/80
AMozwnes

LDH HK
80 110
80 110
80 110
80 110
80 110
80 110
80 110
80 110
80 110
80 110
80 100
80 110
80 110
80 110
80 *

80 110
80 110
80 110
80 110
80 110
80 110
80 110
40 110
80 110
80 110
80 110
80 110
80 110

 

1998 Gogebic County -

Data
ngs Allozymes
ID Forewing Hindwing PGD
Male # Length Black Band

1 45 73.5 125
2 42 71 .5 125/80

3 41 53 125
4 44.5 70.5 125

5 65 125/80

6 45.5 79.5 125

7 45 68.5 125

8 44.5 58 125

9 44.5 72 125/80
10 46 78 125

1 1 45 58 125
12 43 55 125
13 45.5 63.5 125
14 45.5 79 125
15 41 69 125
16 43.5 70.5 125
17 44 62.5 125
18 44.5 78.5 125
19 46 56 125/80
20 45 68 125/80
21 45.5 72.5 125
22 44 63 125
23 43 64.5 125
24 42 68 125
25 44 79 125
26 45 72 125/80
27 45 49 125/80
28 47.5 69.5 125
29 64.5 125/80
30 45 70 125/80
31 42 72 125
32 45 69 125
33 45 68 125
34 45 63.5 125
35 40.5 62.5 125/80
36 43 77 125

82

 

Female #

14265 48
14266 40
14267 46
14268 44.5
1998 Dickinson County -
Data
ID Forewing
Male # Length
1A 47
2A 45
3A 45
4A 44
5A 46.5
6A 40.5
7A 45
8A 42
9A 46.5
10A 45
1 1A 44.5
12A 45
13A 44.5
14A 43.5
15A 47
1 46
2 43
3 41
4 43
5 44
6 43.5
7 44
8 47.5
9 44
1O 45
1 1 45
12 43.5
13 45
14 45
15 43

74
73
70.5
83

Wings

Hindwing
Black Band

58
66.5
68.5

66
71.5

68
75.5
70.5

67
64.5

78
68.5

60
62.5

74
63.5
78.5

69
69.5

63

67
59.5
70.5

71

65
69.5
69.5

65

65
55.5

83

Allozyme
PGD

150l125
125/80
125
125/80
150l125
125
125
125
125
150l125
125
125
125/80
125
125

 

16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33

35
36
37
38
39
40
41
42
43

45
46
47
48

42
45
44.5
41.5
43
46
41.5
45.5
48.5
45
46.5
42
43.5
43.5
46.5
50

46
47
48
45
44.5
48
47
40
42
46
46.5
45
43.5
43
46

1998 Beaver Island -

ID
Male #
1

2
3
4

Data

Wings

57
68.5
73.5
62.5
67.5

66
59
55
71
69.5
74.5
66
69.5
77.5
63.5
82.5
56.5
64.5
62
61.5
72
72
64.5
68.5
72.5
63.5
70
57
64
71
68
68.5

Forewing Hindwing
Length

47
45
44
45

Length

62.5
60
60.5
72.5

PGD

125
125
125
125

34

150l125
125
125/80
125
125
125
125/90
125
125
125
125
125/80
125
125
125
125
125
125
125
125
125
125
125
125/80
125/80
125
125
125/80
150
125/80
125
150/80
125

Allozymes

LDH

80
80
80
80

HK

110
110
110
110

M

 

5
6
7
8
9
10
11

Female #

14203
14204
14205

1998 Charlevoix County - Data

ID
Male #

NNNNNNNNAA-L-L—t—t-AAA—t
mewa-socoooslmcnhooN-Ao‘om“°""““N“

46.5

49
45
46
49
47

46.5

47

48.5

51

46.5
44
47
47
45
57
45
42
44
45
47

40.5

44.5
45

42.5
46
46

44.5

48.5
46
47
46
50
42

46.5

47.5

45.5

54
50
62.5
49
65.5
64.5
67.5

77
71.5

Wings
FW Length HW BBW°/o

56.5
59.5
76.5
56.5
49.5
74
64.5
52
62
62.5
59
63.5
56.5
62.5
73
60.5
54.5
62
60.5
60.5
68
66
68.5
63.5
57.5
68.5
55

125
125
125
125
125
125
125

125

125

85

Allozyme

PGD

125
125
125
125

125/80

125
125
125
125
125
125
125
125
125
125
125
125
125
125

150l125

125
125
125
125

150/125

125
125

80
80
80
80
80
80
80

80
80

110
110
110
110
110
110
110

110
110

~LJ '

 

28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50

Female

#
14039
14042
14037
14041
14036
14035
14040
14033
14034
14038
14013
14015
14014
14009
14007
14002
14008

46
44
45.5

43
49
49
50
45
45

45
49
44.5
49.5
45
46
49
44.5
46
43.5
49
47.5

53
49.5
45
45
51.5
50.5
48.5
44.5
51

48.5
49
42

49.5

47.5

50.5

48.5

79
68
57
75
74.5
53.5
58.5
61
68
70
68
43
53.5
59.5
66
61.5
59
54.5
63
65
67
59
65

66.5
71
62

72.5

68.5

54.5
59
71.5
65.5
55.5
58
75
70.5
76
69.5
61

125
125
125/80
125
125
125
125/80
125
125/80
125
125
125
125
150l125
125/80
125
125/50
125
125
125
125
125
125

V... ‘I"! if“.‘-..'! ‘9 W

 

1998 South Manitou Island

ID
Male #

wwwNNNNNNNNNN-s—t—t—s—s—s—s—s—LA
N—scroooxioaoussz-soooosroamth—so‘om“0"”‘0’5’0

Vlfings
Forewing Hindwing
Length Length
49.5 61.5
49.5 55
38.5 52.5
47 54
39 53
46.5 60.5
44.5 57
49 53
49 50
48.5 61
50 73
47 38.5
48.5 60.5
46.5 62.5
49.5 54
49.5 49
45 52
44 43
46.5 38
45 62.5
45.5 53.5
48 52
48 60
48.5 57
46.5 57
46.5 53
48 56.5
45 53.5
47.5 62
49 64
42.5 60.5
46 47

PGD

125
125
125
125

125/100

125
125
100

125/80
125/100

125
125
125
125
125
125
125
125
125

125/100

125
125
125
125
125
125
125
100
125
125
125
125

87

Allozymes

LDH

80
80
80
80
80
80
40
80

* 80/40

80
80
80
80
80
80
80
80
80
80
80
80
80
80
*80
80
80
80
8O
80
80
80
80

HK

110
110
110
110
110
110/100
110
110
110
110
110
110/100
110
110
110
110
110
110
110/100
110
110
110
110
110
110
110
110
110
110
110
110
110

 

1998 South Manitou Island - Female Data

ID Forewing
Length
14219 51 .5
14220 51
14221 55.6
14222 44
14223 51
14224 50.5
14225 51
14226 53
14227 50.5
14228 50
14229 55
14230 51.5
14231 47
14232 53.5
14233 51
14234 50
14235 53
14236 49.5
14237 51.5
14238 53
14239 51.5
14240 54
14241 50
14242 48
1998 Mason County -
Data

ID Forewing
Male # Length

1 47

2 48

3 48

4 48

5 47.5

6 5O

7 51

Wings
Hindwing PGD
Black Band
59.5 125
57.5 150
55.5 125
61.5 **
47 it
64.5 80
57.5 100
59 **1 25
59.5 "125
62.5 125
66 *
64 125
60 "80
60.5 **
60.5 **
59.5 125
55 *t
58.5 125
69 125
53 125
55.5 100
67.5 125
64 80
52.5 125
VVIngs Allozymes
Hindwing PGD
Black Band
68 125
40.5 125/100
52 125
46 125
69 125
64 150l125
57 150/137

88

Allozymes
LDH

80
80
80
40
80
80
80
80
80
80
*80
80
80
80
80
80
80
80
40
80
80
80
80
80

HK

110

110

110
110

110

110
110
110
110
110
110
110
110

 

8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33

35
36
37
38
39
40
41
42
43

45
46
47
48
49
50

51
50
46.5
44.5
49
51
46.5
45
43
48
49
50
49
45
48
50
46
47.5
46.5
48.5
46.5
52
47
46.5
47.5
46.5
47
51
47
45
49.5
49
50.5
48
47.5
43
47.5
49
48

42.5
51
48.5

66
53
60.5
63.5
66.5
55

46
62
61
49
42
59.5
53.5
55
65
58
51.5
59
42
57
61.5
61
57
71.5
53.5
60
55
59
70.5
52

63
59
62.5
65.5
60
66
63.5
51
65
73
51

89

125
125
125/80
125
125
125
125
125
125
125
150l125
125/80
125
125
125
125/80
125
125/80
125
125
125
125
125
125
125
125
125
125
125
125
125
125
150/80
125/80
125
125
125
125
125
125
125
125/80
125

 

Female #

14065 48 71 .5
14062 49 58
14067 49 66
14068 55 75
14060 51 73.5
14064 48 56
14070 54 52
# 7 54.5 60
# 8 49 67
14061 49 59
14071 52 61
14072 51 59
14069 49 57
14073 52 62
14074 50 64.5

1998 Oscoda County - Male Data

Wings Allozymes
ID Forewing Hindwing PGD LDH HK
Male # Length Black Band
1 49 65 125 80 110
2 51 73.5 125 80 110
3 41 70.5 125 80 110
4 47.5 58.5 125 80 110
5 48 65.5 150l125 80 110
6 45.5 54.5 125 80 110
7 48.5 54.5 125 40 110
8 40 66 125 80 110
9 44 71 125 80 110
10 49 58.5 *150/125 40 110
11 46.5 64.5 125 80 110
12 45 63.5 125 80 110

13 43 58.5 125 80 110

90

 

1998 Isabella County -

 

Data
Wings Allozymes
ID Forewing Hindwing PGD
Male # Length Black Band

1 48.5 46.5 125
2 49 60 125
3 47.5 62 125
4 48.5 65.5 125
5 50 63.5 125
6 48 56 125/80
7 46 68 125
8 47 70 125
9 46.5 67.5 125
10 47.5 57 125
11 43 59 125
12 46 59.5 125
13 48 59 125/100
14 48 41 125
15 47 75 125/80
16 52 54 125
17 48 ' 33 125
18 47.5 63 125
19 48 50 125
20 45 59 125
21 50 60 125
22 43 63.5 125
23 45 74.5 125
24 45 64.5 125
25 45 60.5 125
26 45 65.5 125
27 49 61.5 125
28 51 64.5 125
29 46 70 125
30 48 67 125
31 46 61 125/80
32 47.5 50 125/80
33 47 68 125
34 47.5 71 125
35 48.5 61.5 125
36 49 70 125
37 43.5 61.5 125
38 47 69.5 125
39 45.5 57 125

91

‘81-- '

40
41
42
43
44
45
46
47
48
49
50
Female
#
14000
14020
14021
14022
14023
14024
14025
14026
14027
14028
14110
14111
14112
12—98

45.5
45
48
46

40.5
42
48
49
52

51.5

47.5
46
47
47
51
51

49.5

49.5
52
46
51
50
45

76.5 125
55.5 125
68.5 125
71 125/100
65.5 125
61 125
52.5 125/80
73.5 125/80
55.5 125
59 125
46.5 125

65.5
65
57.5
65.5
61.5
63
61.5
49
84.5
62.5
69.5
63
66
69

1998 P. glaucus Lawrence County, Ohio

Vans
ID Forewing Hindwing
Male # Length Length
1A 54 39
2A 48 37.5
3A 43.5 31
4A 51 19
1 49 37.5
2 47 38
3 52 36
4 50 39.5
5 45 46
6 48.5 46

92

 

 

i.

7 50 40

 

8 49 39
9 52.5 10.5
10 46 49
11 49 25
12 48.5 52
13 50 42
14 49 25.5
15 49 44
16 51 25.5
17 50 41
22 46.5 33.5
1999 Emmet County
Wings Allozymes
ID Forewing Hindwing PGD LDH HK
Male # Length Length
1 45 65.5 125 80 110
2 47.5 63.5 125 80 100
3 45 64.5 150l125 80 110l100
4 49 59 125 80 110
5 46 66 125 80 110
6 49 65 125 80 110
7 44.5 61.5 125 80 110
8 43.5 67 125 80 110/100
9 43.5 49 125 80 110
10 46 61 125 40 110
11 48 72.5 125 80 110
12 47.5 63 125 80 110
13 58.5 125 80 110
14 45.5 56.5 125 80 110
15 46.5 63 125 80 110
16 44.5 75 125 80 110
17 46.5 54 150l125 80 110
18 48 61.5 125 80 110
19 47 70 125 80 110
20 48 72 125 80 110
21 48.5 67.5 125/80 80 110
22 46 63 125 80 110
23 46 64.5 125 80 110

24 48 66.5 125 40 110

93

 

Female ID

15119 51 75.5
15120 50 65
15120 46 60
15153 50 78.5
15154 45 59
15155 47 75
15156 48 60.5
15157 49 62.5

1999 Charlevoix County - Data

Wings Allozymes

ID Forewing Hindwing PGD LDH HK

Male # Length Black Band

1 48 50.5 125 80 1 1 O

2 48 74 125 80 1 10

3 44 58 125 40 1 10

4 46 67 125 80 1 1O

5 43.5 70 125 80 100

6' 48 64 125 80 1 10

7 45 59.5 125 80 1 10

8 47 73.5 125 80 1 1O
9 47.5 61.5 125/80 80 *
10 41.5 59 125 80 *
11 44.5 72.5 125 40 *
12 45 59 125 40 *
13 47.5 54 125 40 *
14 48 56 125/80 80 *
15 43 75 125 80 *
16 44 71 125 80 *
17 45 62.5 125 80 *

Female #

15149 45 70.5 125 * *

15151 49 55.5

94

 

1999 South Manitou Island - Male

ID

wwwwNNNNNNNNNNd—t—td—s—t—L—s—s—t
fiwM-xocoooxloaouowmaoomwmmowNAo‘omNm‘”*“N‘

00000000
(”VODO'I

Dana
Vans
Forewing Hindwing
Length Black Band

48 51
46.5 65
51 55
46 55
46 67
44.5 55
40.5 60
45 56
52.5 36
45 53
46 64
42.5 52
46 62
50 60
47 51
46 57
57

48 66
49.5 52
49 66
46 56
44.5 56
49.5 52
51 .5 50
52.5 60
47.5 64
49.5 57
48 58
48 58
47 72
48.5 66
50 60
47 55
50 53
54.5 54
43.5 54
50 65
51 52

PGD

125
125
125
125/80
125
80
150/100
125/80
125
125
125
125
125
125
125
125
125
125/100
125/80
125/100
125
125
125
125
125
125
125
125
100/50
125/100
125
125
125
100/50
125
125
125
125/80

95

AMozynufi;
LDH HK
80 110
80 110
80 110/100
80 110
80 110
80 110
80 110
80 110
80 110
80 110
80 110
80 110
80 110
80 110
80 110
80 110
40 110
80 110
80 110
80 110
80 110
80 110
80 110
80 110
80 110
80 110
40 110
80 110
80 110/100
80 110
40 110
80 110
80 110
80 110
80 110
80 110
80 110
*100/80 110

39
40
41
42
43

45
46
47
48
49
50
51
52
53

55
56
57
58
59
60
61
62
63

65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83

48.5
47.5
47.5
52
46
50
46.5
48
52.5
47
49.5
49.5
50.5
46
50
50
48
48
52.5
49

49'

49.5
51
50
50

49.5

48.9
46
45
47
49

47.5

46.5

46.5

49.5
51

48.5

50.5

46.5
48

48.5

52.5
50

52.5

57
59
58
62
61
59
66
68
60
45
54
62
52
58
66
54
61
52
53
61
49
45
58
51
48
59
57

60
74
63
48
62
67

55
61
59
44
55
47
62
45
63
50

125/100
125
125/80
125
125
125/100
125/100
125
125/80
125
125/80
125
125
125/80
125
125
125/100
125/80
125
125
125
125
125/100
150l125
125
125/80
125
125
125
125
125
125
125
125
125
125
125/100
125
125
125
125/100
125
125
150/80
125/80

96

80
80
80
80
80
80
80
80
40
80
80
80
80
80
80
80
80

80
80
80
40
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
*80
80
80
80
80

110
110/100
110
110
110
110
110
110
110
110
110
110
110
110/100
110
110
110
110l100
110
110/100
110
110
110
110/100
110
110
110
110
110
110
110
110
110
110
110
110
110/100
110
110
110
110
110
110
110/100
110/100

 

34
85
86
87
88
89
90
91
92
93

95

96

97

98

99

100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120

49
47
46
45.5
46
50
48
49
45
51
47.5
50.5
46
47.5
49
47
48.5
49
49.5
50
45
49.5
46
51
51
44.5
48

47
46
49
51
47.5
48
50
47.5
47

56
54
49
59
54
50
47
50
55
59
56
43
58
65
56
62
58
53
52
46
52
50
63
58
58
53
59
50
48
46

53
49
56
73
39
59

125
125/80
125
125
125
125/100
125/100
125
125
125
125/100
125
125
125
125
125
125
125
125
125
125/100
125
125/80
125
125/80
125
125
125
125
125
125
125
150l125
125
125
125
125

97

80
80
80
80
40
80
80
40
80
80
80
40
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
40
80
80
40
80
80
80
80
40
80

110/100
110
110
110
110
110
110
110
110
110
110
110 l
110 *j
110 j

110 ‘-

 

 

110/100
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110

110l100

 

1999 South Manitou Island Female
Data

Wings Allozymes
ID Forewing Hindwing PGD LDH HK
Female # Length Black Band

1F-99 49 60 125 40 1 10
15222 50 50 125 80 1 10
15223 53 59 100 80 110
15224 48 62.5 125 80 110
15225 46 68 137 80 1 10
15226 49.5 54 125 80 1 10
15227 50 57.5 125 80 1 10
15228 46 71 125 80 110
15229 44 60.5 125 80 110
15230 50 64.5 125 80 "110
15231 53 67 125 80 110
15232 49 52.5 80 40 1 10
15233 45 63.5 125 80 110
15234 51 68 ** 80 **
15235 49 67 125 80 1 10
15236 50 62.5 80 80 1 10
15237 53 69 100 80 110
15238 48 47 ** ** "1 10
15239 49 65.5 ** 80 1 10
15240 50 66 125 ' 80 1 10
15241 52 62.5 125 80 110
15242 48 69 80 80 1 10
15243 48 71 125 80 110
15244 50.5 65.5 125 80 110
15245 50.5 74 125 80 1 10
15246 54 65.5 125 40 110
15247 50 48 125 80 1 10
15248 47 64.5 125 80 110
15249 48 70 125 80 110
15250 52.5 60.5 80 80 110
15251 50.5 56 125 80 110
15252 48.5 62.5 125 80 110
15253 53 70 125 80 1 10
15254 48 59 125 80 1 10
15255 48.5 125 80 "
15256 51 51.5 100 80 110
15257 54 49 125 80 1 10

15258 49 68 125 80 110

98

 

1 5259 51 41 .5
1 5260 49 66
1 5261 50 68
1 5262 54 62.5
1 5263 49.5 59.5
1 5264 46 60
1 5265 52 47.5
1 5266 51 78
1 5267 49 71 .5
1 5268 52 63
1 5269 49.5 66
1 5270 51
1 5271 52 69.5
1 5272 49 59
1 5273 49 65
1999 Leelanau County Data
Vans
ID Forewing Hindwing

Male# Length Black Band

NNNN—SA-AA—‘LAA—t—s-A.
wmaommwmmthAo‘om‘o’U‘lin-t

43
44
47
50
48
47.5
49
45
46.5
48.5
39.5
47
45.5
43
48.5
49.5
49
48.5
42
48
49.5
46.5
46.5

56
69
61.5
53
63.5
39
77
68.5
63
67
54
57
47.5
56
60
58.5
51.5
59.5
69.5
71.5
58
67
64.5

125
80
125
100
125
125
125
125
80
125
125
125
125

PGD

125
125/80
125
125
125
125/80
125
125/100
125
125/80
125/80
125
125/80
125
125/80
125
125
125/80
125
125/100
125
125/80
125

99

80 110
80 1 10l100
80 110
80 110
80 110
80 110
80 110
80 110
80 110
80 110
80 110
80 110
80 110
40 *
* 110
AMozwnes
LDH HK
80 110
80 110
80 110
80 110
80 110
80 110
80 110
*80/40 1 10
80 110
80 110
80 110
80 110
80 110
80 110
80 110
80 110
80 110
80 1 10/100
80 110
80 110
80 1 10/100
40 110
80 110

 

 

 

24 47 48 125
25 46 57 125
26 42.5 60 125
27 46 64.5 125
28 44.5 62.5 125
29 46 125/100
Female #
15105 51 66
15106 45 67.5
15107 52 73
1999 Benzie County Data
Wings

ID Forewing Hindwing PGD
Male# Length Black Band

1 64.5 125/80
2 46 57 125
3 44.5 61 125
4 46.5 55.5 125/80
5 45.5 69 125
6 47 75 125
7 46.5 59.5 125
2000 Benzie
County
Allozymes
ID Pgd Hk th
Male #

1 125 110

2 125 110

3 125 110

4 150l125 110

5 125/80 110

6 100l80 110

7 125 110

8 125 110

9 125 110
10 125/100 110/100 80
11 125 110

100

80 110
80 110
80 110
80 110
80 110
80 110
Allozymes
LDH HK
80 110
80 110/100
40 110
80 110
80 110
80 110
40 110

 

 

 

2000 Leelenau

County
lD Pgd
Male #

1 137

2 125

3 125

4 125/100
5 125

6 125

7 125

8 125

9 125
10 125/80
11 125/100
12 125
13 125/100
14 150l125
15 125
16 125
17 125/100
18 80

19 125
20 125
21 125/80
22 125
23 80
24 125
25 125
26 125
27 150/125
28 125/80
29 125
30 125
31 150l125
32 125
33 125
34 125
35 125
36 125
37 125
38 125

Allozymes

Hk

110l100
110
110
110

110/100
110
110
110

110/100

110/100
110
110
110
110
110
110
110
110

110/100
110
110
110
110
110
110
110
110
110

110/100
110
110
110
110
110
110
110

110/100
110

th

80

80

80

80

101

 

 

 

39 125 110/100

40 125 110

41 125 110

42 125 110

43 125 110

44 125/80 110

45 125 110

46 125/100 110 80
47 125 110

48 125/100 110 80
49 125 110

50 125 110

51 125 110

52 125/80 110

53 125 110

54 125 110

55 125 110

56 125/100 110 80
57 125 110

58 125 110

59 125 110/100

60 125 110l100

61 125 110/100

62 125 110

63 125 110/100

64 125 110

65 125 110

66 125 110

67 125/80 110
68 150l125 110

2000 Mason
County
Allozymes
ID Pgd Hk th
Male #
1 125/80 110
2 125 110
3 125 110
4 125 110
5 125 1 10l100
6 125 110
7 125 110
8 125 110

102

10
11
12
13
14
15
16
17

125
125
125
125
150l125
125
125/100
137
125

110
110
110
110
110
110
110
110/100
110/100

2000 North Manitou Island

ID
Male #

NMNNN—L—l—t—t—L—L—h—t—A—t
gfiggwa-socooouoamwa—xo‘omVO’m““N‘

Pgd

150l125
150l125
125
125
125/100
125/80
125
125
125
125
125/100
125
125/80
125
125
125
125
150l125
150l125
125
150l125
125
125/100
125
125
125
125
125

Allozyme

HR

110
110/100
110
110/100
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110

80

th

103

29
30
31
32
33

35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53

55
56
57
58
59
60
61
62
63

65
66
67
68
69
70
71
72
73

125/80
125
125
125
125
125
125
125
125
125

125/80
125
125

125/100
125

125/80
125
125
125
125
125
125
125
125
125
125

125/80
125
125
125
125
125
125
125
125
125
125
125
125
125
125
125

125/80

125/80

125/100

110
110
110
110
110
110
110/100
110
110
110
110
110
110
110
110
110
110
110l100
110
110
110
110
110
110
110
110/100
110
110
110
110
110
110
110
110
110
110
110
110/100
110/100
110
110
110
110
110
110

104

 

 

 

74
75
76
77
78
79
80
81
82
83

85
86
87
88
89
90
91
92
93

95
96

125
125/100
125
125/100
125
125/100
125
125
125
125
125
125
125
125
125
125
125/80
125
125/100
125
125/100
125/100
125/100

110l100
110/100
110
110
110
110
110
110
110
110
110l100
110
110
110/100
110
110/100
110
110
110
110/100
110
110
110

2000 South Manitou Island

ID
Male #

fijacoooxloamoonu‘

13

Allozymes

Pgd

125
125
125
125/100
125
125
125/100
125
125/100
125
125/100
125
125

Hk

110
110/100
110
110
110
110
110
110
110
110
110
110
110

40

80

105

 

 

14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33

35
36
37
38
39
40
41
42
43

45
46
47
48
49
50
51
52
53
54
55
56
57
58

125
125
125
150l125
125
125
125
125/100
125
137/125
125
125
125
125/80
125/100
125
125/80
125
125/100
125
150/100
125/100
125
125/100
137/125
150l125
125
125
150/125
125/100
125
125/80
125/100
125
125/100
125
125
125/80
125
125
150l125
125/80
125/100
125
125

110
110
110
110
110
110
110
110
110
110
110
110
110
110l100
110
110/100
110I100
110
110
110
110/100
110
110
110
110
110l100
110
110
110
110
110
110l100
110
110/100
110
110l100
110
110
110
110
110
110
110
110/100
110

106

 

 

59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
1 00

125
125
125
125
125
125
125
125
125
125/100
125
125/80
125/100
125/100
125
125/80
125/80
125
125
125
125
125/100
125
125
125
150l125
125
150l125
125
125/80
125
125
125
125
125/80
125
125
125
125
125/100
125
125

110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110/100
110/100
110
100
110
110
110
110
110
110
110
110
110

107

APPENDIX 2.1

OVIPOSITION PREFERENCE DATA

 

108

Appendix Table 2.1. South Manitou Island female 3-choice oviposition preference
data for 1999-2000. Total numbers of eggs laid by each female are given as well

as the percent of the total number of eggs placed on each of three host plants. The
other category includes eggs that were placed < 2 cm from the host plant. These wer
likely intended to be placed on the plant, but for the purposes of this study have been
scored as other. Oviposition was only scored for females laying >10 eggs.

Female oviposition preference was assigned only to females that layed 50% or more
of her eggs on any single host plant.

1999 Percent Eggs LaId
Total Eggs Quaking Tulip Black Other

Female ID (n) aspen tree cherry

1F-99 0

15222 0

15223 5 20 40 20 20

15224 0

15225 0

15226 0

15227 98 17 61 13 8

15228 23 96 4

15229 2 100

15230 3 100

15231 2 50 50

15232 5 100

15233 0

15234 13 8 69 23

15235 0

15236 0

15237 0

15238 1 100

15239 49 2 76 2 20

15240 1 100

15241 0

15242 22 82 9 9

15243 51 59 27 4 10

15244 0

15245 0

15246 7 28 71

15247 7 14 71 14

15248 0

15249 0

15250 10 70 10 20

15251 54 15 52 6 28

15252 2 50 50

15253 117 21 40 18 21

15254 117 38 17 3 42

109

 

Appendix Table 2.1 continued.

110

15255 0
15256 7 43 57
15257 5 40 60
15258 0
15259 1 100
15260 0
15261 103 4 81 6 10
15262 0
15263 2 50 50
15264 118 77 6 4 13
15265 14 50 14 7 21
15266 110 84 7 1 7
15267 69 39 39 22
15268 3 67 33
15269 21 52 29 19
15270 0
15271 6 17 83
15272 62 5 73 15 8
15273 51 61 14 8 18
2000
Percent Eggs Laid
Total Eggs Quaking Tulip Black Other
Female ID (n) aspen tree cherry
16136 56 0 50 32 18
16137 81 20 63 6 23
16173 3 0 67 0 33
16174 1 0 0 100 0
16175 21 19 71 5 5
16176 0 0 0 0 0
16177 90 7 59 11 23
16178 79 19 54 14 13
16179 0 0 0 0 0
16180 0 0 0 0 0
16181 14 7 7 36 50
16183 61 11 10 61 18
16184 0 0 0 0 0
16185 36 11 17 44 28
16186 0 0 0 0 0
16187 1 0 100 0 0
16188 0 0 0 0 0
16189 9 0 0 44 56
16190 41 5 41 20 34
16191 131 2 7 45 47
16192 100 19 39 12 30

Appendix Table 2.1 continued.

0

16193
16194
16195
16196
16197
16198
16199
16200
16201
16202
16203
16204
16205
16206
16207
16208
16209
16210
16211
16212
16213
16214
16215
16216
16217
16218
16219
16220
16221

a)
c>c>44
c>c>c>c3

 

 

a:
"I
.3
m

:Sjfifgfic3c3<»<3
.5
—I
h)
or

N
01
N
0')

A

000000000000800

 

‘
u.
Cit;00C)<3<3(O'CICJCDCDCDtDlCICDCDCDCDCD:K’C>C>Eg:§lu>C>CDCD<D

arc:Eic><3<:;;lc»c>c>c:<3<3lc»c>c>c3<3<3;§;c>c>ficngglc>c>§3<3

<3 k§§§<3 c>c3<3 c>c><3<3 c>c><3<3 c>c><3<:»g;c><3

<33: fl:c:cac:£3131c>c>c>c3<3<31c>c>c><3<3

an
caco cI,c>c:><::o

 

111

APPENDIX 2.2

LARVAL HOST PLANT SURVIVAL DATA

 

i l

 

112

Appendix Table 2.2. 1999 and 2000 South Manitou Island larval host plant survival
larval host plant survival data. The number of larvae from each mother placed on
each host plant is provided next to the mother ID number, followed by the percenta
percentage of those larvae that survived through the first instar. Survival values

> 0% for Tulip tree are in bold print as they are significant values. P. canadensis is

not normally able to detoxify Tulip tree.

1999 %Larval Survival on each Host Plant

“3 l1 CHI n
Female #
1 F -99
15222
15223
15224
15225
15226
15227
15228
15229
15230
15231
15232
15233
15234
15235
15236
15237
15238
15239
15240
15241
15242
15243
15244
15245
15246
15247
15248
15249
15250
15251
15252
15253
15254
15255
15256
15257
15258
15259

75

100

50

100

80

6715

ooooogooaoooooooAooo—soooonooooo-sooogooo
ooooo‘onoAaooooooomooomoooowoooAomoooB-soo

113

TT

100
50

50

33

100

50

48.77778

 

76

100

100

100

100

100
67

 

87
100

92.22222

c>c>c>c>c>3§c>ass:c>c>c>c>c>c>c>onnac>c>-sc>c>c>c>nac>c>c>naca-scacacafigcacaca

Appendix Table 2.2 continued.

15260 0 0 0
15261 0 0 0
15262 0 0 0
15263 0 0 0
15264 0 0 0
15265 0 0 0
15266 0 0 0
15267 0 0 0
15268 0 0 0
15269 0 0 0
15270 0 0 0
15271 0 0 0
15272 0 0 0
15273 0 0 0
2000 %Larval Survival on each Host Plant
n QA n 11' n BC
16136 9 44 9 0 11 73
16137 15 87 20 35 8 88
16173 0 0 0 2 100
16174 0 0 0 0
16175 0 0 0 0
16176 0 0 0 0
16177 1 100 5 20 2 67
16178 25 68 25 12 24 10
16179 0 0 0 0
16180 0 0 0 0
16181 4 100 0 5 40
16183 16 50 15 0 18 56
16184 0 0 0 0
16185 9 89 5 60 7 29
16186 0 0 0 0
16187 0 0 0 0
16188 0 0 0 0
16189 0 0 1 0 6 67
16190 9 67 14 0 6 67
16191 37 76 34 24 31 81
16192 22 46 20 45 26 58
16193 0 0 0 0
16194 0 0 3 0 0
16195 0 0 0 0
16196 0 0 0 0
16197 2 50 0 2 100
16198 16 63 15 40 13 46
16199 13 85 16 13 12 67
16200 0 0 0 0
16201 0 0 0 0
16202 60 52 63 4.8 54 41

114

 

Appendix Table 2.2 continued.

16203
16204
16205
16206
16207
16208
16209
16210
16211
16212
16213
16214
16215
16216
16217
16218
16219
16220
16221

60

c>c>ggc>c>c>c>c3c>c3c3c3c><3<3

.3
CO
C

100 100

Ob—B-‘OOUIOOOOOOOOOOOO
ONNOOOUIOOOOOOOOOO-‘O
OQNOOOOOOOOOOOOOOOO

115

 

LITERATURE CITED

Allen, D.L. 1979. Wolves of Minong: Isle Royale’s Wild Community. University of
Michigan Press, Ann Arbor, Michigan.

Arnold, ML. and SA. Hodges. 1995. Are natural hybrids fit or unfit relative to their
Parents? TREE, 10(2): 67-71.

Aubert, J ., B. Barascud, H. Descimon & F. Michel. 1997. Ecology and genetics of
interspecific hybridization in the swallowtails, Papilio hospiton Géné and P.
machaon L., in Corsica (Lepidoptera: Papilionidae). Biol. J. Linn. Soc. 60: 467-
492.

Ayres, M.P. & J.M. Scriber. 1994. Local adaptations to regional climates in Papilio
canadensis (Lepidoptera: Papilionidae). Ecological Monographs, 64 (4): 465-
482.

Boag, P.T. & A.J. van Noordwijk. 1987. Quantitative Genetics. Chapter 2 of Avian
Genetics: A population and ecological approach. Edited by F. Cooke & P.A.
Buckley. Academic Press. Orlando.

Bossart, J .L. & J .M. Scriber. 1995. Mainte nance of ecologically significant genetic
variation in the tiger swallowtail butterfly through differential selection and gene
flow. Evolution 49: 1163-1171.

Britten, H.B., P.F. Brussard, D.D. Murphy & P.R. Ehrlich. 1995. A test for isolation—
by-distance in central Rocky Mountain and Great Basin populations of Edith’s
Checkerspot (Euphydryas editha). J. Hered. 86: 204-210.

Carlquist, SJ. 1974. Island Biology. Columbia University Press.

Darlington. 1943. Carabidae of mountains and islands: data on the evolution of isolated
faunas and on atrophy of wings. Ecological Monographs 13: 37-61.

Deering, MD. 1998. Preferential mate selection by males as a reproductive isolating
mechanism between the swallowtail species; Papilio glaucus and P. canadensis
(Lepidoptera, Papilionidae). MS. Thesis, Michigan State University, East
Lansing, Michigan.

Ehrlich, PR. 1961. Intrinsic barriers to dispersal in Checkerspot butterfly. Science,
134:108 — 109.

Ehrlich, P.R., Raven, RH. 1969. Differentiation of populations. Science, 165: 1228-
1232.

”6

 

Fales, J .H. 1959. A field study of the flight behavior of the Tiger Swallowtail
Butterfly. Announcements of the Entomological Society of America,
52: 486-487.

Grant, RR. 1998. Evolution on Islands. Oxford University Press. New York.

Hagen, RH. 1990. Population structure and host use in hybridizing subspecies of
Papilio glaucus: (Lepidoptera: Papilionidae). Evolution, 44: 1914-1939.

Hagen, R.H., R.C. Lederhouse, J.L. Bossart & J .M. Scriber. 1991. Papilio canadensis
and P. glaucus (Papilionidae) are distinct species. Journal of Lepidopterists’
Society, 45 (4): 245-258.

Hagen, R.H. & J .M. Scriber. 1991. Systematics of the Papilio glaucus and P. troilus
Species groups (Lepidoptera: Papilionidae): Inferences from allozymes. Annals
of the Entomological Society of America, 84: 380-395.

Harris, AG. 1977. Geology of National Parks, 2"d ed. Dubuque, Iowa.

Harrison, R.G. 1993. Hybrid zones and the evolutionary process. Oxford University
Press. New York.

Hartl, D.L. 1988. Primer of population genetics. Sinauer Associates, Inc. Sunderland,
Massachusetts.

Haswell, SO. and Alanen, AR. 1994. A garden apart: An agricultural and settlement
history of Michigan’s Sleeping Bear Dunes National Lakeshore region. Midwest
regional office, National Park Service, Omaha, Nebraska & State Historic
Preservation Office, Michigan Bureau of History, Lansing, Michigan.

Hatt, R.T., Van Tyne, J ., Stuart, L.C., Pope, CH, and Grobman, AB. 1948. Island life:
A study of the land vertebrates of the islands of eastern Lake Michigan.
Cranbrook Institute of Science Bulletin No. 27. Cranbrook Press, Bloomfield
Hills, Michigan.

Hoole, J .C., D.A. Joyce & A.S. Pullin. 1999. Estimates of gene flow between
populations of the swallowtail butterfly, Papilio machaon in Broadland, UK and
implications for conservation. Biological Conservation 89: 293-299.

Jacquard, A. 1974. The genetic structure of populations. Springer-Verlag. New York.

Jiggins, CD and J. Mallet. 2000. Bimodal hybrid zones and speciation. Tree, 15 (6):
250-255.

Johnson, K.P., F.R. Adler & J .L. Cherry. 2000. Genetic and phylogenetic consequences
of island biogeography. Evolution, 54(2): 387-396.

117

Johnson, K.S., D. Snider & J .M. Scriber. 1996. Estimates of genetic differenctiation
among Callosamia species and Hyalophora cecropia (Saturniidae) using
allozyme electrophoresis. Journal of the Lepidopterists’ Society, 50 (3): 217-225.

Kingsolver, J .G. 1985. Thermoregulatory significance of wing melanization in Pieris
butterflies: physics, posture, pattern. Oecologia, 66: 540-545.

Kingsolver, JG. 1987. Evolution and coadaptation of thermoregulatory behavior and
wing pigmentation pattern in pierid butterflies. Evolution, 41: 472—490.

Kingsolver, J .G. 1995. Viability selectioin on seasonlly polyphenic traits: Wing melanin
Pattern in western white butterflies. Evolution, 49(5): 932-941 .

Leberg, PL. 1992. Effects of populations bottlenecks on genetic diversity as measured
by allozyme electrophoresis. Evolution, 46 (2): 477-494.

Luebke, H.J., J .M. Scriber & B.S. Yandell. 1988. Use of multivariate discriminant
Analysis of male wing morphometrics to delineate a hybrid zone for Papilio

Glaucus glaucus and P. g. canadesnis in Wisconsin. The American Midland
Naturalist, 119(2): 366-379.

MacArthur, R.H. & E.O. Wilson. 1967. The theory of island biogeography. Princeton,
N.J., Princeton University Press.

McKechnie, S.W., P.R. Ehrlich & R.R. White. 1975. Population genetics of Euphydryas
butterflies. 1. Genetic variation and the neutrality hypothesis. Genetics, 81: 571-
594.

Nielsen, MC. 1999. Michigan butterflies & sippers: A field guide and reference.
M.S.U. Extension, Michigan State University.

Porter, A.H., R. Wenger, H.J. Geiger, A. Scholl, & A.M. Shapiro. 1997. The Pontia
daplidice-edusa hybrid zone in northwestern Italy. Evolution 52: 1561-1573.

Raymond M & Rousset F. 1995 GENEPOP (version 1.2): Population genetics software
for exact tests and ecumenicism. J. Heredity, 86:248-249.

Scriber, J .M. 1982. Food plants and speciation in the Papilio glaucus group.
Proceedings of the 5th International Symposium of Insect-Plant Relationships.
Pudoc, Wageningen, 307-313.

Scriber. J .M. 1990. Interaction of introgression from Papilio glaucus canadensis and
Diapause in producing “spring form” eastern tiger swallowtail butterflies, P.
Glaucus (Lepidoptera: Papilionidae). The Great Lakes Entomologist. 23 (3):
127-138.

118

 

Scriber, J .M. 1994. Climatic legacies and sex chromosomes: Latitudinal patterns of
voltinism, diapause, size, and host-plant selection in two species of swallowtail
butterflies at their hybrid zone. Insect Life-cycle Polymorphism. 133-177.

Scriber, J .M., R. L. Lindroth and J. Nitao. 1989. Differential toxicity of a phenolic
glycoside from quaking aspen to Papilio glaucus butterfly subspecies, hybrids
and backcrosses. Oecologia 81: 186-191.

Scriber, J .M., J .L. Bossart & D. Snider. 1992. Diagnostic alleles from electrophoresis
dstinguish two noctuid pest species, Hydracecia immam's and H. micacea
(Lepidoptera: Noctuidae). The Great Lakes Entomologist, 25 (2): 91-98.

Scriber, J .M., R.C. Lederhouse & R.V. Dowell. 1995. Hybridization studies with North E
American swallowtails in Scriber, J.M., Y. Tsubaki, & R.C. Lederhouse (eds).
swallowtail butterflies: Their ecology and evolutionary biology. Gainesville, FL:
Scientific Publishers.

 

Scriber, J.M., R.H. Hagen and RC. Lederhouse. 1996. Genetics of mimicry in the tiger
swallowtail butterflies, Papilio glaucus and P. canadensis (Lepidoptera:
Papilionidae). Evolution, 50(1) 222-236.

Scriber, J .M., M. Deering and A. Stump. 2001 in press. Hybrid zone ecology and
swallowtail speciation: Geographic and genetic distance influence influence
behavioral, biochemical, and ecological trait clines in North American
Papilionid butterflies. In Ecology and Evolution Taking Flight: Butterflies
As Model Study Systems. (eds. C. Boggs, W. Watt and P. Ehrlich) University
Of Chicago Press, Chicago, IL.

Scriber, J .M., G. Ording, K. Lamphire. 2001 In Review. Latitudinal trends, seasonal
thermal unit accumulations and inheritance of a diagnostic wing trait in two
hybridizing species of tiger swallowtail butterflies (LepidopterazPapilionidae).
Amer. Midl. Naturalist

Scriber, J .M., K. Weir, D. Parry, & J. Deering. 1999. Using hybrid backcross larvae of
Papilio canadensis and Papilio glaucus to detect induced phytochemical
resistance in hybrid paplar trees experimentally defoliated by gypsy moths.
Entomologia Experimentalis et Applicata, 91: 233-236.

Slatkin, M. 1987 . Gene flow and the geographic structure of natural populations.
Science, Vol. 236 (789).

Stump, AD. 2000. Lack of cryptic reproductive isolation between Papilio canadensis
and Papilio glaucus; and population genetics near their hybrid zone. MS.
Thesis, Michigan State University, East Lansing, Michigan.

119

Tong, M.L. & A.M. Shapiro. 1989. Genetic differentiation among California
populations of the anise swallowtail butterfly, Papilio zelicaon Lucas. J. Lepid.
Soc. 43: 217-228.

Waldbauer, G.P. & J.G. Sternburg. 1988. Lakes Michigan and Huron limit gene flow
between the subspecies of the butterfly Limenitis arthemis. Can. J. Zool. 66:
1790-1795.

Watt, W.B. 1968. Adaptive significance of pigment polymorphisms in Colias
butterflies. Variations in melanin pigment in relation to thermoregulation.
Evolution: 22: 437-458.

Williams, C.B., G.F. Cockbill & M.E. Gibbs. 1942. Studies in the migration of
Lepidoptera. Transactions of the Royal Entomological Society of London.

92: 101-283.
Williamson, M. 1981. Island populations. Oxford University Press.

Wolfe, A. 1994. Migrations — Wildlife in motion. Hillsboro, Oregon.

120