v ‘ .
M; In: ‘11:.
L I
,n $9.2m y. e
... . ... h. 4
x _ I
. . 54m
, ..

’6.

..v
t
a.

a
t .
355 r...

v;

 

in...

“a ,

2. av . i...
4:, . L
VA .

‘a...x. 4.11:3»..‘1
1 13:11.5)...
. :11 a.

”ﬁnk“ 1...»me

 

w: ... N
. sﬁﬁﬁuﬁ m. mmﬁxgw

 

.Hgggg

[u

SIHNTYUBA

llllllllllllllWillIllllllllllllllll ll

3 1293 01688

Will

 

 

 

 

 

 

 

This is to certify that the

thesis entitled

Testing the 'Cold-Pocket' Hypothesis:
Oviposition Preference in the Canadian
Tiger Swallowtail, Papilio canadensis

 

presented by

Piera Y. Giroux

has been accepted towards fulﬁllment
of the requirements for

M. S . degree in Entomology

30. mm

Major professor

 

Datg 26 August 1998

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

 

LIBRARY

M’Chigan State
University

 

 

 

 

 

 

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

 

MTE DUE

DATE DUE

DATE DUE

 

FEB 2 4 2097

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1M COMM-p.14

 

TESTING THE “COLD POCKET” HYPOTHESIS:
OVIPOSITION PREFERENCE IN THE CANADIAN TIGER
SWALLOWTAIL, PAPILIO CANADENSIS
By

Piera Y. Giroux

A THESIS

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

MASTER OF SCIENCE

Department of Entomology

1998

ABSTRACT
TESTING THE “COLD POCKET” HYPOTHESIS: OVIPOSITION
PREFERENCES OF THE CANADIAN TIGER SWALLOWTAIL, PAPILIO
CANADENSIS
By

Piera Y. Giroux

Three areas in Northern Michigan; Vanderbilt, Pellston and Cross Village were
compared for climate differences; host plant phenology differences and Papilio
canadensis Rothschild and Jordan (Lepidoptera: Papilionidae) oviposition preference
differences. The ‘cold pocket’ hypothesis predicted that these sites were climatically
distinct, with Vanderbilt the coolest site and Cross Village the warmest; that phenology in
the ‘cold pocket’, Vanderbilt area, would be delayed; and that oviposition preference by
‘cold pocket’ P. canadensis butterﬂies would be for white ash.

Every year with regard to total degree-day accumulations, Vanderbilt was the
coldest site. Vanderbilt was cooler than the other sites only sixteen times in twenty—nine
years during the time period when P. canadensis butterﬂies were actively selecting host
plants (March lst — July 5th). In the years of this study, 1996 and 1997, host plant
phenology was not delayed in the ‘cold pocket’. In 1996 and 1997, P. canadensis
butterﬂy populations from across Northern Michigan did not show oviposition preference
differences. In 1996 and 1997, butterﬂies from the ‘cold pocket’ did not show an
oviposition preference for white ash. These results indicated a greater depth and
complexity to climate/ plant! herbivore interactions than previously assumed by the ‘cold

pocket’ hypothesis.

Acknowledgements

With great appreciation, I would like to thank the following people for their
assistance and advice throughout my graduate career. First, my committee: Jim. Miller,
Cathy Bristow, and Muralee Nair. Second, fellow graduate students: Nathan Siegert,
Heather Govenor, Sheila Ebert, Chrissy J arzomski, Dylan Parry, and Heather Rowe.
Third, lab support: Vanessa Serrao, David Tesar, Keegan Keefover, Aram Stump, Mark
Deering and Jessica Deering. Fourth, ﬁnancial support: MAES Project Numbers 1644
and 1722, a 1997 Department of Entomology Hutson Grant, and UMBS ﬁnancial
support. Fifth, I would like to thank the faculty, staff, and students of the University of

Michigan Biological Station. Finally, I would like to thank my advisor, Mark Scriber.

iii

TABLE OF CONTENTS

LIST OF TABLES vi

LIST OF FIGURES viii

CLIMATE, HOST PLANT PHENOLOGY AND OVIPOSITION PREFERENCE
OF THE CANADIAN TIGER SWALLOWTAIL, PAPILIO CANADENSIS

Introduction 1
Climate, Plant and Herbivore Interactions 1

‘Cold pocket’ Hypothesis 5
Papilio canadensis 7

Sites 11
Methods 12
Climatology 12
Phenology 13
Oviposition Preference 14

Five Choice Array 15

Young and Old White ash Host Plant Array 15

Phenology Array 16

Chemical Array 16

Larval Growth 17
Statistical Analyses 18
Results 21
Climatology 21
Phenology 22
Oviposition Preference 24

Five Choice Array 24

Young and Old White ash Host Plant Array 25

Phenology Array 26

Chemical Array 26

Larval Performance 27
Overall Survival 27

Survival in Each Instar 27

Pupal Weight 27

Days to Pupation 28

Discussion 29

iv

APPENDIX A

APPENDIX B

APPENDIX C

BIBLIOGRAPHY

66

68

7O

89

LIST OF TABLES

Table 1: Scientiﬁc and common names of host plants examined for P. canadensis
oviposition preference

Table 2: Accumulated degree-days °C (threshold temperature 10 °C) at three sites
in Northern Michigan

Table 3: Stepwise regression of accumulated degree—days at three sites, 1969-1997

Table 4: Stepwise regression of water content of leaves of ﬁve species collected at
Pellston and Vanderbilt, 1996

Table 5: Stepwise regression of water content of leaves of ﬁve species collected at
Pellston, Vanderbilt and Cross Village, 1997

Table 6: Stepwise regression of water content of leaves of ﬁve species collected at
Pellston and Vanderbilt, 1997

Table 7: Stepwise regression of water content of leaves of ﬁve species collected at
Pellston and Vanderbilt for 1996 and 1997

Table 8: ANCOVA results for 1996 oviposition preference of P. canadensis on ﬁve
host plant species

Table 9: ANCOVA results for 1997 oviposition preference of P. canadensis on ﬁve
host plant species

Table 10: ANCOVA results for 1998 oviposition preference of P. canadensis on
young and old (fully expanded) white ash

Table 11: ANCOVA results for 1997 oviposition preference of P. canadensis on
white ash collected from four sites, resulting in phenological differences in
white ash

Table 12: ANCOVA results for 1997 oviposition preference of P. canadensis on
white ash extracts from four sites

vi

38

39

4O

41

42

42

43

45

46

47

48

Table 13: 1997 P. canadensis larval survival in each instar for larvae reared on
black cherry, paper birch, white ash, basswood and quaking aspen

Table 14: Mean and percent differences in pupal weights of P. canadensis reared
on black cherry, paper birch, white ash, basswood and quaking aspen, 1996

Table 15: Mean and percent differences in pupal weights of P. canadensis males
and females reared on black cherry, paper birch, white ash, basswood and quaking
aspen, 1997

Table 16: Mean and percent differences in days until pupation of P. canadensis
males and females reared on black cherry, paper birch, white ash, basswood and
quaking aspen, 1997

Table 17: Stepwise regression of water content of leaves of ﬁve species collected at
Pellston and Vanderbilt adjusted for degree-day differences, 1997

Table 18: Stepwise regression of water content of leaves of ﬁve pecies collected at
Pellston and Vanderbilt adjusted for degree-day differences, 1996 and 1997

Table 19: 1996 Five choice oviposition preference by P. canadensis
Table 20: 1997 Five choice oviposition preference by P. canadensis
Table 21: 1998 Young and old oviposition preference by P. canadensis

Table 22: 1997 P. canadensis oviposition preference for white ash collected from
four sites

Table 23: 1997 P. canadensis oviposition preference for extracts of white ash
collected from four sites

Table 24: 1996 P. canadensis pupal weight data

Table 25: 1997 P. canadensis pupal weight data

vii

49

49

50

51

68

68

7O

73

75

76

76

77

84

LIST OF FIGURES

Figure l: Accumulated degree-days at three sites; Cross Village, Pellston and

Vanderbilt, 1969-1997. A is seasonal (March lst — October 3150 degree-days; _B_ is

ﬂight and larval season (March lst - July 3 lst) degree-days; Q is ﬂight season

(March lst - July 5th) degree days. Linear trends are illustrated for Vanderbilt data.
The trend is statistically signiﬁcant in A. (r2 = 0.27, n = 29), but not in g (r2 = 0.18),

not in _(_: (:1 = 0.09).

Figure 2: Heat/ precipitation indices for Vanderbilt, Michigan. May (diamonds), June
(squares), July (triangles) and August (circles), 1987 — 1997. July, 1989 and June,

1991 are extreme outliers.

52

53

Figure 3: Percent water content of leaves from ﬁve species collected at Pellston (squares)

and Vanderbilt (circles). A_ is black cherry; B is paper birch; Q is
white ash; _Q is basswood and E is quaking aspen. Data from the 1996 season

Figure 4: Percent water content of leaves from ﬁve species collected at Cross
Village (triangles), Pellston (squares) and Vanderbilt (circles). A is black cherry;
_B_ is paper birch; Q is white ash; _D is basswood and E is quaking aspen. Data
from the 1997 season.

Figure 5: Preference of butterﬂies collected in Pellston (n = 31, diagonal striped
bars), collected in Vanderbilt (n = 17, dotted bars), or collected in the Upper
Peninsula (11 = 23, cross-hatched bars) for oviposition on leaves of ﬁve species,
or no leaf (paper). Represented means and standard errors of means for any one
bar type are adjusted for a balanced design. Leaves were from the Pellston area.
Data from the 1996 season.

Figure 6: Preference of butterﬂies collected in Pellston (n = 7, diagonal striped
bars), collected in Vanderbilt (n = 17, dotted bars) for oviposition on leaves of
ﬁve species, or no leaf (paper). Represented means and standard errors of means
for any one bar type are adjusted for a balanced design. Leaves were from the
Pellston area. Data from the 1997 season.

viii

54

55

56

57

Figure 7: Preference of butterﬂies collected in Pellston (n = 3, diagonal striped
bars), collected in Vanderbilt (n = 2, dotted bars), or collected in the Upper
Peninsula (n = 4, cross-hatched bars) for oviposition on leaves of ﬁve species, or
no leaf (paper). Represented means and standard errors of means for any one bar
type are adjusted for a balanced design. Leaves were from the Vanderbilt area.
Data from the 1997 season.

Figure 8: Preference of butterﬂies collected in Pellston (n = 4), Vanderbilt (n = 2),
the Upper Peninsula (n = 3), Oscoda (n = 2), Charlevoix (n = 1) and Mason (n = 3)
for oviposition on unexpanded white ash leaves (narrow horizontal line bars), fully
expanded white ash leaves (dark vertical line bars), or no leaf (paper) (solid

bars). Represented means and standard errors of means for any one bar type are
adjusted for a balanced design. Data from the 1998 season.

Figure 9: Preference of butterﬂies collected in Pellston (n = 6, diagonal striped
bars), collected in Vanderbilt (n = 5, dotted bars), the Upper Peninsula (n = 1,
cross—hatched bars) and the combined populations (n = 12, gridded bars) for
oviposition on leaves of ﬁve species, or no leaf (paper). Represented means
and standard errors of means for any one bar type are adjusted for a balanced
design. Data from the 1997 season.

Figure 10: Oviposition preference of Papilio canadensis on methanol extracts of
white ash leaves collected from Cross Village, Pellston, Vanderbilt and Okemos
(n = 9). A leaf sprayed with water, a leaf sprayed with the acetone solvent, and
eggs laid on paper all served as controls. Represented means and standard errors
of means for any one bar type are adjusted for a balanced design. Data from the
1997 season.

Figure 11: Total survival of Papilio canadensis on ﬁve host plants: black cherry,
paper birch, white ash, basswood and quaking aspen. (No larvae survived on paper
birch or basswood). Data from the 1997 season.

Figure 12: Mean pupal weights (g) of Papilio canadensis reared on ﬁve host plants:

black cherry, paper birch, white ash, basswood, and quaking aspen. (No larvae
survived on paper birch). Data from the 1996 season.

Figure 13: Mean pupal weights (g) of male (n = 72, dotted bars) and female

(11 = 78, bricked bars) Papilio canadensis reared on ﬁve host plants: black cherry,
paper birch, white ash, basswood, and quaking aspen. (No larvae survived on
paper birch or basswood). Data from the 1997 season.

Figure 14: Mean days until pupation for male (n = 72, dotted bars) and female

(n = 78, bricked bars) Papilio canadensis reared on ﬁve host plants: black cherry,
paper birch, white ash, basswood, and quaking aspen. (No larvae survived on
paper birch of basswood). Data from the 1997 season.

ix

58

59

60

61

62

63

65

Figure 15: Preference of butterﬂies collected in Pellston (n = 10, diagonal striped
bars), collected in Vanderbilt (n = 19, dotted bars), or collected in the Upper
Peninsula (n = 4, cross-hatched bars) for oviposition on leaves of ﬁve species, or
no leaf (paper). Represented means and standard errors of means for any one bar
type are adjusted for a balanced design. Leaves were from the Pellston and
Vanderbilt areas. Data from the 1997 season.

69

Introduction

Climate, Plant and Herbivore Interactions

“The action of climate seems at ﬁrst sight to be quite independent of the struggle
for existence; but in so far as climate chieﬂy acts in reducing food, it brings on the most
severe struggle between the individuals whether of the same or of distinct species, which
subsist on the same kind of food.” (Darwin 1859). The interaction between climate,
plants and herbivores has been at the center of a great deal of ecological, environmental
and evolutionary research. Basic aspects of biology are rooted here; population
dynamics; nutrient ﬂow and stabilizing mechanisms in ecology (Hairston et al. 1960); and
coevolution of host plants and their herbivores (Thompson 1994) are ecological
phenomenon where understanding is advanced by studies of climate, plant, and herbivore
interactions.

Examining effects of predation and climate, in addition to phytochernistry, might
further clarify plant/ herbivore relationships, as some scholars have suggested that
phytochemical coevolution theories do not fully express the depth or variation found in
plant/ herbivore relationships (Smiley 1978, Janzen 1988, Bemays and Graham 1988). A
recent variant on the topic of plant] herbivore relationships and the role of climate is the
geographic mosaic theory of coevolution, which posits a coevolutionary relationship
continuum in which interactions vary in intensity and expression within a species range
(Thompson 1994). This theory incorporates the effects of abiotic variance, such as

climate differences, on plant-herbivore interactions. Although much ecological research

has delved into climate/ plant! herbivore interactions, understanding is far from complete.

Plausibly, a variety of climatic factors could inﬂuence plant! herbivore
interactions, of which regional warmth or coolness, humidity and precipitation are but
two examples (Barbosa 1988). A convenient measure of the former characteristic is
thermal-unit accumulation. Host plant phenology has also been implicated in affecting
herbivore selection, especially in cases where there are changes in environmental
conditions (Barbosa 1988). Constraints on thermal units have been shown for latitudinal
clines that can effect host choice (Scriber and Lederhouse 1992). This ‘voltinism-
suitability’ hypothesis has been extended to local ‘cold pockets’ not simply latitude
(Scriber 1996a).

The voltinism-suitability hypothesis has its wide-reaching roots in basic tenets of
plant! herbivore interaction theory. The tenets include factors that drive plant! herbivore
interactions and herbivore range. There are questions as to whether secondary
phytochemistry, predators, and the environment are more important to herbivore
population control and dynamics. Often, the range of suitable host plants can limit the
distribution of associated herbivores. Host plant distribution can be limited by
environmental conditions, particularly temperature. This could effect the distribution,
development time, and ﬁtness (Cockrell et al. 1994) of the associated herbivores.

Temperature can also affect the number of generations an herbivorous insect can
complete in a growing season. Butterﬂies may make behavioral and physiological
adjustments to prevailing weather conditions (Cockrell et al. 1994). In areas where the
herbivore may not be able to complete one or two generations, as the area is thermally

constrained, there are selection pressures on the herbivore to feed on the plant that will

most enhance growth. For monarch butterﬂies, it was shown that latitude and oviposition
date can inﬂuence the maturation time and the number of generations. Earlier oviposition
dates had greater inﬂuence on maturation time of larvae than later oviposition dates
(Cockrell et al. 1994). In addition to latitude, oviposition date, and climate differences
affecting herbivorous insect behavior, growth, distribution and survival, host plant quality
is also important. Not all host plants are equal in suitability for larval growth. Growth of
many insect larvae is nitrogen limited (Mattson 1980, Scriber 1984a and b, Mattson and
Scriber 1987). Since foliar nitrogen content and leaf water are correlated (Scriber and
Slansky 1981, Mattson and Scriber 1987), larvae on leaves with low leaf water tend to
grow more slowly (Scriber 1977). In areas that are thermally constrained, a herbivore that
feeds on a more suitable host plant has an increased chance of pupating before the end of
the season. In areas where the number of generations is not thermally constrained,
selection pressures are lifted and herbivores are able to feed on a wider number of host
plants successfully.

The interaction of thermal units and host plant distribution may create a dynamic
interaction in which herbivore/ plant interactions vary across space and time, causing
local specialization patterns for a polyphagous species. Evidence has indicated that some
species of Papilio have an extremely localized oviposition preference in relation to
thermal accumulation or phenology. These butterﬂies oviposit on leaves that are in full
sun, or that may have higher water content (Grossmueller and Lederhouse 1985). In
summary, in areas with a short growing season, there is selection pressure for a herbivore
to consume a high quality food source that allows it to reach maturity earlier, albeit of

smaller size (Ayres and Scriber 1994). The voltinism—suitability hypothesis is the direct

predecessor to the cold-pocket hypothesis.

Papilio canadensis (Rothschild and Jordan) butterﬂies, their oviposition host
plants and their larval performance have been studied as an example of climate! plant!
herbivore interactions. These butterﬂies are excellent research organisms because they
are common, showy, strong ﬂiers, and have a variety of interactions with different host
plants from extreme speciﬁcity to a great deal of polyphagy (Scriber 1995). Oviposition
preferences within the Papilio group form a particularly intriguing way by which to test
interactions, as part of the oviposition preference is genetically based and some of the
genes effecting oviposition preference have been localized to a single chromosome
(Thompson 1995). Oviposition preferences may be inﬂuenced by a variety of factors.

Not all oviposition sites afford similar nutrition, cover, and protection for larvae
and adult butterﬂies. In the landscape of available oviposition sites, some sites are more
rewarding. Since larvae generally do not move between sites, ovipositing female
butterﬂies that choose oviposition sites that ensure the greatest ﬁtness for offspring and
survival of their genes would be reproductively successful. Because of their catholicism,
choice of oviposition sites might be cued by the environmental situation during the ﬂight
season. The cues used by the butterﬂies could include visual ones, such as leaf shape
(Rausher 1980, Papaj 1986, Renwick and Chew 1994), tactile ones, such as leaf
toughness, and sensory responses to leaf chemical components (Renwick and Chew
1994). Larval growth potential need not be the only important consideration. Larvae may
also be susceptible to host-speciﬁc predators or parasitoids (Thompson and Pellmyr
1991). In order to reduce the probability of being attacked by predators and parasitoids,

some larvae use a form of crypsis (Thompson and Pellmyr 1991), but tree characteristics,

such as secondary phytochemistry that reduce effects from parasitoids and predators,
might be important selection factors (Thompson and Pellmyr 1991).

The direct measure of larval ﬁtness and hence reproductive success of a test
species follows the rearing of larvae to adults in the natural environment and
determination of survival to sexual maturity. Survival in the ﬁeld would also provide an
estimate of natural levels of mortality. Indirect measures, more amenable to controlled
experimentation, include pupal weight, length of time until pupation, and survival of the
larvae. If oviposition preferences were being driven by qualities intrinsic to the host plant
there could be a correlation between larval performance and host plant quality. If the
system is being driven by extrinsic factors, those that increase the survival for the
butterﬂy, but not necessarily for the larvae (Thompson 1988, Thompson and Pellmyr
1991), there should be less correlation between larval performance and oviposition

preference for a host plant.

‘Cold Pocket’ Hypothesis:

In the Northern Lower Peninsula of Michigan, and in the Western Upper
Peninsula of Michigan, there are areas of lower average annual frost-free days compared
to nearby areas, known as ‘cold pockets’ (sensu Scriber 1996a). In these areas with a
constrained growing season, it is implied that plant phenology and bud-break are delayed
(Johnson and Scriber 1994, Scriber 1996a). It was observed that in these ‘cold pocket’
areas, P. canadensis butterﬂies preferred F raxinus americana L., white ash, as an

oviposition host plant (Johnson and Scriber 1994). Studies outside ‘cold-pockets’ had

shown white ash to be of poor quality for hosting larval growth (Johnson and Scriber
1994) because ash quickly declines in some forms of soluble nitrogen and increases in
leaf toughness after bud-break (Hunter and Lechowicz 1992). The ‘cold pocket’
hypothesis posits that white ash would not be as poor a host inside as it was outside of the
‘cold pocket’ as white ash, a late bud-breaking plant, would be even further delayed in
bud-break in the ‘cold pocket’ (Scriber 1996a). White ash would be younger, with higher
water content at the time that P. canadensis is ﬂying. P. canadensis, if selecting leaves
that increase larval performance, would choose these delayed white ash leaves as they
would be more nutritive, with higher water content, an increase in some forms of soluble
nitrogen, and with a lower leaf toughness. The increase in larval performance relative to
other host species would not be seen on white ash outside the ‘cold pocket’ (Scriber
1996a).

Other rationales to explain the localized P. canadensis white ash oviposition
preference have been proposed. These include the possibility that P. canadensis
competes for resources with major forest defoliators such as Malacosoma disstria
(Hﬁbner), forest tent caterpillar, and Lymantria dispar L., gypsy moth (Scriber 1996a;
Scriber and Gage 1995). As gypsy moth is known to avoid white ash as a host plant P.
canadensis might be driven to utilize white ash in the face of such competition (Scriber
1996a). However, gypsy moth is a recent arrival to Michigan (Scriber and Gage 1995)
and it seems unlikely that in 6-8 years gypsy moth would have driven P. canadensis to a
white ash preference. P. canadensis preference for white ash has been observed in
competitor—free laboratory trials (Scriber 1996a, Johnson and Scriber 1994), also

suggesting that the choice was not due solely to forest pest outbreaks. Additionally, these

forest pests do not occur in all of the ‘cold-pockets’, such as in the Upper Peninsula of
Michigan, in which the oviposition preference shift was observed. Lastly, these two pests
do occur in conjunction with P. canadensis outside of ‘cold pocket’ areas and where

white ash is not a preferred host.

Papilio canadensis:

Papilio canadensis is a species in the Papilio glaucus L. group, and as a species
only recently has been separated from P. glaucus (Hagen et al. 1991). The range of P.
canadensis corresponds to the Pleistocene glaciation area of Northern America and
constitutes a signiﬁcantly distinct ecotone (Scriber and Gage 1995), extending from the
Appalachian mountain range into the Great Lakes area, and north across Canada and
Alaska (Hagen et al. 1991). The adaptations of P. canadensis for life in cold climates
(Kukal et al. 1991, Ayres and Scriber 1994) as well as their ability to detoxify a great
variety of plant allelochemicals, such as tremulacin from quaking aspen and other
Salicacious plants, and prunasin from black cherry, demonstrates the successful escape
from its tropical ancestry of the Papilionidae (Scriber 1995).

P. canadensis is a univoltine butterﬂy (Hagen and Lederhouse 1985), spending
four to eight months of the year as a pupa, often buried under snow. In order to avoid
eclosing before the temperature is sufﬁcient to maintain metabolic and dietary needs, P.
canadensis must be tuned in to local climate factors, such as precipitation and day length.
Once it has emerged, it spends its three to six week adult life span (Scriber 1996b)

feeding, mating and ovipositing. P. canadensis emerges from its puparium in late May

(Scriber 1996b). P. canadensis is protandrous in most years, with males emerging
slightly before females (Lederhouse et a1. 1995). Early emergence allows males greater
access to females (Ae 1995), as well as salts and minerals (Lederhouse et a1. 1990) that
are possibly used for spermataphore construction (Lederhouse et al. 1990). A mature
male patrols from site to site seeking receptive females (Brower 1959), chases, courts and
attempts to copulate with a receptive female. Papilio butterﬂies are polygamous, with
females sometimes mating ﬁve to six times (Scriber 1996b). After a mating, the female
stores the spermataphore of the male in her bursa copulatrix, and may utilize the sperm of
the most recent mating to fertilize her eggs (Scriber 1996b). After transfer of the
spermataphore, females search for oviposition sites.

Upon alighting on a host plant, a female swallowtail uses her forelegs in a
drumming behavior to ascertain host plant quality (Nishida 1995). She approaches the
leaves, and curling the tip of her abdomen forward (Nishida 1995), deposits a single egg
on the plant surface (Scriber 1996b). Eggs when freshly laid are a deep green, blending
into the leaf surface color. As the embryo within the egg matures, the egg becomes
deeper in color, and is almost brown at the time of ecdysis. After ecdysis, the larva eats
the chorion of the egg in order to obtain some early nutrition, or perhaps to remove
evidence of its presence from potential natural enemies (Scriber 1996b). The larva feeds
on the leaves of the plant on which it was oviposited. If the plant has toxic chemicals, is
low in nutrition, subject to desiccation, signals predators to feed on the larva, or affords

little protection, the larva is less likely to survive.

P. canadensis is a polyphagous butterﬂy, unusual in the butterﬂy world because of
the high degree of polyphagy, with adults and larvae utilizing a variety of host plants for
feeding and oviposition (Scriber 1984a). P. canadensis can utilize plants from the
families Salicaceae, Oleaceae, Rosaceae, Tiliaceae, Lauraceae, and others (Scriber
1984a). As P. canadensis host plants are trees, they are usually apparent and enduring,
ensuring that P. canadensis can actively seek and oviposit on a host plant rather than lay
eggs haphazardly (W iklund 1984).

The trees investigated for oviposition preference‘by P. canadensis are listed in
Table 1. For each, the northern portion of Michigan is roughly in the middle of its range.
There are differences among the trees in preference for soil type, tolerance for shade,
tolerance for water stress and other characteristics, as might be expected (Voss 1985,
1996, Crow 1990, Marquis 1990, Safford et al. 1990, Schlesinger 1990, Perala 1990). Of
particular importance for this project is that bud-break depends on thermal accumulation
with quaking aspen (Michaux) (Perala 1990) and paper birch (Marshall) (Safford et al.
1990) breaking bud early; and basswood (L.) (Crow 1990) and white ash (L.)
(Schlesinger 1990) breaking bud late. Black cherry (Ehrhart) breaks bud intermediately
(Marquis 1990).

The present study investigated P. canadensis oviposition preference in relation to
larval performance using host material from areas with decreased thermal accumulation
and from areas of greater thermal accumulations. Climatic differences at three different
locations in Northern Michigan were characterized. Water content of host leaves in 1996
and 1997 was measured. Oviposition preference and larval performance experiments

were carried out with a variety of P. canadensis populations in Northern Michigan in

1996 and 1997. The 1996 growing season was climatically typical for the region, while
1997, an El Niiio year, was dryer and colder across the state. A few updating

observations were made in May and June of 1998; both months were hot and dry.

10

Sites:

The surveyed sites covered a range of growing seasons based on the average
number of freeze-free days as described by the Michigan Climatic Atlas (Eichenlaub et a1.
1990, Scriber 1996a). The ﬁrst site was the Pellston area (Pellston Plains; on Catsmanﬂs
comer; to the intersection of Riggsville Rd. and Bryant Rd., Emmet and Cheboygan
counties). Pellston averages 90-100 freeze-free days in the growing season. The
Vanderbilt area, a ‘cold pocket’, (near Vanderbilt in Pigeon River State Forest, on the
border of Cheboygan and Otsego counties) was the second site. Vanderbilt averages 70
freeze-free days in the growing season. Thumb Lake was ‘added’ to the Vanderbilt site
only for occasional collection of ‘cold pocket’ butterﬂies. For some oviposition
preference trials, yields of Vanderbilt test organisms were inadequate for the experimental
protocol and Thumb Lake specimens were taken to supplement Vanderbilt ones (Thumb
Lake averages around 90 freeze-free days in the growing season). A third area beside
Lake Michigan, near Cross Village, Wycamp, Hardwood State Forest (Emmet county)
was examined. Cross Village averages 140-150 freeze-free days in the growing season.
No butterﬂies collected from Cross Village laid any eggs. Most of the butterﬂies
collected from Cross Village were collected early in the ﬂight season and may have been
unmated. In order to compare oviposition preferences of an outlying population of
butterﬂies, butterﬂies were collected from across the Upper Peninsula and employed.

The Upper Peninsula (Chippewa County) averages 110-130 freeze-free days; (Iron
County) averages 70-90 freeze-free days; (Dickinson County) averages 100-110

freeze-free days.

11

Methods:

Climatology:

Daily degree-days were calculated using the averages method (Pedigo and Zeiss
1996) with a general insect threshold temperature of 10°C. Climate data from the three
sites, from 1969-1997, were obtained from the Department of Geology, Climatology Lab,
Michigan State University. These years had almost complete data sets across the sites.
Three time periods were examined. The ﬁrst time period was the seasonal accumulated
degree-days (March lst - October 3 lst). The ‘cold pocket’ hypothesis assumed climate
differences across Northern Michigan based on seasonal freeze-free day differences,
roughly correlated to seasonal accumulated degree-day differences. In order to make
comparisons within the framework of the ‘cold pocket’ hypothesis, it was necessary to
examine climate differences at this level. Degree-day accumulations that occur after leaf
senescence and after larvae pupate, may contribute to overall degree-day accumulations,
but are not very interesting biologically. Early season degree-day accumulations
however, may be very important to the biological systems studied here. Early season
degree-day accumulations can inﬂuence bud-break, leaf ﬂush, and butterﬂy eclosion. For
the next two analyses of degree-day accumulations, late season degree-day accumulations
were excluded, and early season degree-days were included. Another time period
examined was the ﬂight and larval development season accumulated degree days (March
lst - July 3 lst). This time period included early season degree-day accumulations, the
degree-days accumulated during the P. canadensis ﬂight season (usually conﬁned to

June, sometimes occurring earlier in May), and during the time period when larvae were

12

developing. The third time period examined was the ﬂight season accumulated degree-
days, (March lst - July 5th). This time period included early season degree-day

accumulations, and the degree-days accumulated during the P. canadensis ﬂight season.

Degree-days (threshold temperature 10°C) accumulated in Vanderbilt for May,
June, July, and August for each year, 1987-1997, were divided by the amount of
precipitation in Vanderbilt in millimeters. Vanderbilt was the only site that had both
reliable precipitation and thermal unit data. Such heat! precipitation indices are good for

indicating drought stress conditions (Gage, 1998).

Phenology:

Leaves were collected in 1996 from Pellston and Vanderbilt. Leaves were
selected without conscious bias from several trees of each of ﬁve species: black cherry,
paper birch, white ash, basswood and quaking aspen. Collections were made on nine
dates between June 3 and July 23, although not all hosts were sampled from both sites on
each date. There were at least four collection dates per species per site. Leaves were
immediately placed into plastic, airtight bags and stored on ice. They were categorized by
site, date and species. For water content determination, leaves were weighed the same
day as they were collected, placed into a drying oven set at 50°C, for 3-4 days. Dry and

wet weights were used to calculate percent water content.

In 1997, leaves were collected from all three sites. There were 38 collection dates
between May 23 and August 13. Due to the labor-intensive nature of the sampling

regime, Vanderbilt leaves were usually collected a day later than Pellston and Cross

13

Village leaves. Ten leaves of each species per date, per site, were processed.

Relationships among leaf water weight, tree species, site of origin, and date were
examined using a stepwise regression analysis, with yearly data analyzed separately, then

combined, to determine if there were year to year differences in phenology.

Oviposition Preference:

Four oviposition assays were carried out. In each, there was one butterﬂy, and
one leaﬂet, leaf, or set of leaves per treatment per chamber. Lifetime assays were run
(until the butterﬂy was weak or exhausted). Forewing length measurements and age
estimates, as described by Lederhouse and Scriber (1987), were made on ﬁeld-collected
females before each was assigned a brood number and distributed to an oviposition
preference trial. Leaves, of approximately equal surface area, refrigerated less than seven
days were used. The leaf petiole was placed into a water-ﬁlled plastic aquapic. Random
placement of all host plants in each array, around a clear plastic multi-choice oviposition
chamber (25 cm diameter by 9 cm height) ensured that oviposition results were
uninﬂuenced by sequence. Oviposition dishes were stacked on a rotating turntable (6
turns/h) lit by 60-watt incandescent bulbs (6h light-dark cycles) (Scriber 1993).
Temperature inside the oviposition dishes was maintained near 30°C during peak
oviposition times, when the oviposition dishes were illuminated to simulate daylight.
Butterﬂies were removed from oviposition dishes and fed a 20% honey solution daily
while eggs were collected and counted. Eggs on the paper liner, or plastic chamber were

counted as on a leaf if they were within 1 cm of the leaf. If the egg was more than 1 cm

14

from the leaf, they were counted as laid on a plastic or paper surface, which was
considered a ‘leaf type’ in analysis. Leaves with eggs present were removed and stored
(27°C) for larval assays. Positions were reﬁlled with fresh foliage of the same species.
The replacement foliage was not necessarily from the same tree or collection date. In all

cases, the foliage was from the same site.

Five choice array: Adult female P. canadensis were presented simultaneously with
leaves of white ash, basswood, paper birch, black cherry and quaking aspen collected
either in Pellston (1996 and 1997) or Vanderbilt (1997). Butterﬂies were collected from
the Vanderbilt area (17 in 1996, 19 in 1997) and from outside the Vanderbilt area (54 in

1996, 16 in 1997).

Young and old white ash array: In 1998, an oviposition array consisting of two types of
white ash foliage was tested. The two types were older, fully expanded leaves (collected
from Okemos, Michigan) and young unexpanded leaves (collected from near the
‘Mystery Spot’ in Chippewa county). Butterﬂies were collected from Vanderbilt (n=2)

and ﬁve other sites across Northern Michigan and the Upper Peninsula (n=13.)

15

Phenology array: In 1997, oviposition arrays consisting of white ash foliage collected
from the three principal sites, plus one more, were tested. The fourth site, in Okemos,
near Michigan State University, was outside the geographic region, and south of the range
of P. canadensis. Butterﬂies were collected in Vanderbilt (n=6) and outside this area

(n=6).

Chemical extract array: White ash leaﬂet material was collected from four sites:
Pellston, Vanderbilt, Cross Village and Okemos. Leaves from Pellston were collected on
10 June 1997 and 24 June 1997; and extracted on 16 June 1997 and 27 June 1997.
Vanderbilt leaves were collected on 4 June 1997 and 11 June 1997; and extracted on 5
June 1997 and 16 June 1997. Cross Village leaves were collected on 10 June 1997 and
24 June 1997; and extracted on 16 June 1997 and 30 June 1997. Okemos leaves were
collected on 12 June 1997 and 24 June 1997; and extracted on 16 June 1997 and 26 June
1997. The leaﬂet material (petiole and rachis not included) from each site, on each
extraction date was placed in a sterile liquid nitrogen cooled mortar and pestle and
roughly ground. This material was then placed in a sterile Electric Coffee and Spice
Grinder (Regal, Kewaskum, WD and ground until the material was homogeneous, and
ﬁne. Thirty to forty g’s of the dispersion was placed in a ﬁltration column (149 mm x
450 m) that had been packed with cotton swabbing, and methanol (175 mL) was added.
(An oviposition assay in 1996, testing Papilio glaucus oviposition preference for white
ash extracts found a higher response to methanol rather than hexane or ethyl acetate.
Extracts in 1996 were also solubalized in acetone and sprayed onto quaking aspen leaves

with a plant sprayer.) After 30 min., the column stopcock was opened and efﬂuent was

16

collected. The stopcock was closed, and the collected efﬂuent was added back to the
column. This process was repeated two times. After the efﬂuent was collected a third
time, the solubalized extract was concentrated in a rotovap (Brinkmann Instruments Inc.,
Westbury, NY) at 100 °C (there was not a successful vacuum created by the rotovap set-
up used), until all volatile components had been removed. The residue was weighed and
acetone was added to make a 1g/1L, or 10% suspension. Using a plant sprayer, this was
sprayed to saturation onto quaking aspen leaves that were placed in oviposition arenas.
Ovipositional responses to such extracts, from Pellston, Cross Village, Vanderbilt and
Okemos, were measured and compared to the response to leaves sprayed with acetone
alone and water alone. Butterﬂies were collected from Vanderbilt (n=4) and from outside

Vanderbilt (n=5).

Larval growth:

Eggs were placed in dishes marked with a brood number and the host plant
preference of the mother. Mother preference was deﬁned as the oviposition host plant
with the highest percentage of eggs. Dishes were stored in a Percival growth chamber at
27 °C (18 L: 6 D) and checked daily for eclosion. When neonates emerged, all larvae
from the same brood were distributed randomly to a feeding assay on black cherry, paper
birch, white ash, quaking aspen or basswood leaves. Few larvae were set up on paper
birch or basswood in 1997, as these were found to be poor hosts in 1996. No more than
ﬁve or six larvae per dish were assigned to initial feeding assays. Larvae were reared at

27 °C (18 L: 6D) in Percival growth chambers. Larvae were checked every two to three

17

days (or more frequently if leaf material was rapidly consumed), the dishes were cleaned,
leaf material replaced, and the date, number of surviving larvae, and the instar of each
larvae were recorded. When the larvae reached the third instar they were separated and
reared in individual dishes to reduce crowding effects. After pupation, they were weighed
to the nearest 0.0001 g and sexed. Weight, length of time to pupation, length of time in
each stage of metamorphosis and overall survival were recorded. Overall survival was
the percentage of neonate larvae that pupated relative to the number set up on the host

plant.

Statistical analyses:

Data were analyzed in spreadsheet format using Microsoft Excel 5.0 (Microsoft,
1994). Normality was conﬁrmed with the Shapiro-Wilkes tests in the proc univariate
program (SAS Institute Inc., 1989). Climate data were analyzed with proc glm in SAS to
observe statistical differences in mean degree-day accumulations between sites; and using
proc reg in SAS to investigate relationships between year, site and accumulated degree-
days. As Cross Village data were not complete for the 1985-1997 period, missing years
were excluded from regression and analyses of variance. Phenology data were analyzed
using proc reg in SAS for 1996, 1997 and the two years combined to probe relationships
between site, date, accumulated degree-days, host-plant species, year, and foliar percent
water content. Oviposition data ratios were arcsine transformed and analyzed using proc
glm in SAS with an ANCOVA where approximate butterﬂy age and winglength were

covariates. Statistical signiﬁcance was assigned at or = 0.05 using Fishers least signiﬁcant

18

difference test.

The most important contributing factor to signiﬁcant interactions was determined
by slicing the interactions in SAS. All reported means and standard errors are least
square means as these means and standard errors are adjusted as if the design had been
balanced. They provide a population marginal mean, and allow that the sum of
oviposition preference ratios will add to one. Mean pupal weight differences were
analyzed using proc glm in SAS with an AN OVA to uncover signiﬁcant differences in
pupal weight attributable to pupal sex or host-plant. Mean days to pupation differences
were analyzed using proc glm in SAS with an AN OVA to discover signiﬁcant differences
in the days to pupation attributable to pupal sex or host-plant. Difference in survival of
larvae per instar (where the larval host plant, the mother’s oviposition preference, the
instar the larva was in, and the length of time the larva spent in that instar were variables)
was analyzed with a repeated measures analysis in proc mixed in SAS, with the

covariance parameter estimate as a diagonal arcsine model.

As there is some concern as to how a butterﬂy’s oviposition preference should be
weighted, the oviposition assays were examined using an additional protocol. In this
analysis, only butterﬂies that laid a minimum of ten or more eggs were included. The
cut-off value of ten was used, because this was the historical cut-off value in prior
examinations of the ‘cold pocket’ hypothesis. While this analysis may skew the results in
favor of butterﬂies that lay more eggs, it minimizes the chance that the results may be
skewed by butterﬂies that lay few eggs, and may not really exhibit host plant preferences.

This statistical analysis was conducted in the same manner as the above, with the

19

exception that butterﬂies laying fewer eggs were excluded.

20

Results:

Climatology:

Over the time for which comparable data were available (i.e. 1969-1986),
generally, Cross Village had the largest average number of accumulated degree-days and
Vanderbilt the smallest. This was true whether or not one was examining accumulated
degree-days for the season (March lst - October 3 lst); early season to the time of
pupation (March lst - July 3 lst); or the early season and the butterﬂy ﬂight period (March

lst - July 5th) (Table 2).

Mean seasonal accumulated degree-days, March lst — October 3 lst, were
signiﬁcantly different between Vanderbilt and the other two sites (p<0.0001); and
Pellston and Cross Village (p < 0.05). Mean ﬂight and larval development accumulated
degree-days, March lst - July 3 lst, were signiﬁcantly different between Pellston and
Cross Village (p <0.002) and Vanderbilt and Cross Village (p <0.0001). There was no
statistically signiﬁcant difference for this period between Pellston and Vanderbilt. Mean
ﬂight season accumulated degree-days, March lst - July 5th, were not statistically
signiﬁcantly different between Vanderbilt and Cross Village; Vanderbilt and Pellston;
and Pellston and Cross Village. In 5 years of 29 the difference in seasonal degree-days
between Vanderbilt and Pellston exceeded 200 at July 3 lst. In 5 years of 29 the

difference between Vanderbilt and Pellston at July 5th exceeded 100 (Figure 1).

Vanderbilt showed a signiﬁcant warming trend in total seasonal accumulated

degree-days (Figure 1). For the two shorter periods, trends in the Vanderbilt accumulated

21

degree-days, while positive, were not statistically signiﬁcant. Neither Cross Village nor

Pellston showed any persistent trends over 29 years.

For all three time intervals, the year to year variance in accumulated degree-days
was greater than site to site variation in accumulated degree-days. Stepwise analysis for
the ﬂight season degree-day accumulations, March lst - July 5th, removed site
differences from the model, as it didn’t add to the power of the regression, showing that

site differences were not signiﬁcant (Table 3).

As precipitation differences and water stress could inﬂuence host plant quality for
larval growth, heat! precipitation indices are of particular relevance to this project. Heat!
precipitation indices (Figure 2) for the Vanderbilt area for years 1987-1997 indicated that
May, 1997, was the driest of all Mays and that 1997 had the second most drought-like
June. (June 1991, had a higher heat! precipitation index and was both hot and dry.) July

and August of 1997 had heat! precipitation indices similar to those of other years.

Phenology:

In 1996, leaf water content varied among tree species, i.e. interspeciﬁcally, and
collection date, i.e. seasonally (Table 4). However, within a given plant species, and on a
particular date there were no site differences (Table 4). When data were adjusted for
degree-day accumulations, stepwise regression analysis kept all factors in the regression
model, but site differences were the least important (Table 4). Water content declined in
tree leaves throughout the season, with quaking aspen and paper birch having high water

content early in the season, with water contents declining earlier, and basswood and white

22

ash maintaining a high water content longer (Figure 3).

A greater number of phenological assessments were made in 1997, using three
sites, ﬁve host plant species, and nineteen dates. Leaves were indexed early, before bud-
break of some species, so that bud-break and early leaf ﬂush water contents could be
recorded for some species, providing a clearer picture of water content and suitability for
larval nutrition. Stepwise regression analysis on these data showed that leaf water
content varied with tree species (interspeciﬁcally), collection date (seasonally), and with
site (Table 5). The contrast with 1996 results was explored by stepwise regression on leaf
water contents at the Sites (Pellston and Vanderbilt) common to both years. Similar
regression results were found, both with and without adjustment for degree-day
contributions when Pellston and Vanderbilt were compared, and when Pellston,
Vanderbilt and Cross Village were compared. Site contributions were the least
meaningful contributor to the regressions (Table 6). As in the previous set of
measurements, leaf water content declined throughout the 1997 season for all species.
Across all three sites, quaking aspen and paper birch had high water content early in the
season and water content declined rapidly early; basswood and white ash did not break
bud as soon, but maintained a high water content later (Figure 4).

Data for Pellston and Vanderbilt for the years 1996 and 1997 were compared
(Table 7). Stepwise regression indicated that leaf water content varied seasonally,
interspeciﬁcally, geographically, and annually. When these data were degree-day

adjusted, annual variation remained a signiﬁcant factor, second in importance to seasonal

inﬂuence (Table 7).

23

Oviposition Preference:

Five choice array: In 1996, oviposition preference was attributable to species of host, but
not butterﬂy origin (Table 8). P. canadensis preferred to oviposit on quaking aspen, with
29.8% of eggs laid on these leaves when the data was pooled (Figure 5). Signiﬁcantly
more eggs were laid on quaking aspen by the P. canadensis test group than on any other
leaves. For pooled data, quaking aspen and white ash did not have a signiﬁcantly
different number of eggs laid on them. Black cherry and basswood did not have a
signiﬁcantly different number of eggs laid on them. Paper birch and the chamber paper
were also not Signiﬁcantly different from each other in the percentage of eggs laid on
these substrates. All three groups were signiﬁcantly different from the other two groups
(Table 8). Using a cut-off value of ten eggs per female for inclusion in the analysis did
not change the results, or effect the signiﬁcance of any of the factors.

In 1997, neither the origin of the butterﬂy, nor the site of origin of the leaf
material was correlated to host preference (Table 9). Again there was an oviposition
preference attributable to tree species. Oviposition preference was greatest for black
cherry, with 29% of the total eggs laid on these leaves when data were pooled for leaves
from the Pellston (Figure 6) and Vanderbilt (Figure 7) sites and for butterﬂy location.
Mean percent eggs laid on a treatment were similar for: white ash, basswood and quaking
aspen; basswood, quaking aspen, paper birch and chamber paper; black cherry was

signiﬁcantly different from all other treatments (Table 9).

24

Using a cut-off value of ten eggs for inclusion in the analysis did slightly change
the results. In this case, the origin of the host plant, and the origin of the butterﬂy did not
signiﬁcantly affect the oviposition preference. However, the species of tree, the
interaction of the butterﬂy origin and the species of tree, and the interaction of the origin
of the host plant and the species of tree were all found to be signiﬁcant. When the
butterﬂy origin by species of tree interaction was examined, it was found that tree Species
contributed most signiﬁcantly to oviposition preference except for butterﬂies from
Vanderbilt; the butterﬂy origin contributed signiﬁcantly to the interaction on black cherry
host plants. When the tree collection site and tree species interactiOn was compared, it
was determined that the tree species contributed most signiﬁcantly to the interaction.
Mean percent eggs, with a cut-off value of ten, laid on a treatment were similar for: black
cherry, white ash, and quaking aspen; basswood and quaking aspen; basswood and paper

birch; and paper birch and chamber paper.

Young and old white ash array: There was no difference in preference attributable to
butterﬂy collection site. There was a signiﬁcant difference in preference for young,
unexpanded white ash leaves, versus fully expanded white ash leaves, versus a paper (no
leaf) control. When data were pooled, the most eggs (Table 10, Figure 8) were laid on the
unexpanded white ash, and the least eggs laid on chamber paper (Table 10, Figure 8). An
intermediate number were laid on the expanded older white ash leaves (Table 10, Figure
8). The interaction of butterﬂy collection site and white ash phenology was also
signiﬁcant. When this interaction was examined, it was determined that the white ash age

contributed most strongly to every interaction, except that white ash age did not affect

25

oviposition preference for butterﬂies collected from Charlevoix or butterﬂies collected
from Vanderbilt. Both of these populations consisted of two or fewer butterﬂies. It was
also determined that the butterﬂy collection site did signiﬁcantly affect oviposition
preference for old white ash leaves. When the data was examined with a cut-off value of

ten eggs, there was no difference in the results.

Phenology array: There was no signiﬁcant difference in preference for different
phenological stages of white ash as indexed by four collection sites by all butterﬂies
tested (Table 11). Differences in preference by butterﬂies from the Vanderbilt area
versus the Pellston area were not signiﬁcant (Figure 9). The sole signiﬁcant difference
between treatments was that fewer eggs were laid on chamber paper than on any of the
foliage treatments (Table '11). When analyses were performed with a cut-off value of ten

eggs, there were no differences in the results.

Chemical array: There was no preference difference by all butterﬂies tested for
methanol extracts of white ash collected from four different sites or the controls (Table
12, Figure 10). When analyses were performed with a cut-off value of ten eggs, there

were no differences in the results.

26

Larval Performance:

Overall survival from neonate to pupa in 1997 was low. No larvae reared on paper birch
and basswood survived to pupation. The survival of larvae on black cherry was 33%,
followed by quaking aspen, 13.5%, then followed by white ash, 7.3% (Table 15, Figure

11).

Survival in each instar in 1997, with host plant, host plant preference by the ovipositing
female, larval instar, and days per instar showed that the host plant and the days per instar
were important and signiﬁcantly different in percent larval survival in each instar (Table

13).

Pupal weight varied depending upon the host plant species in 1996. Pupal weights on
black cherry, quaking aspen, and white ash were not Signiﬁcantly different, although the
least square means were higher on quaking aspen, followed by black cherry, followed by
white ash (Table 14). Pupal weight on basswood was similar to weight on white ash
(Table 14, Figure 12).

In 1997, there was a (1: 1) ratio of pupal males to females (72 males: 78 females).
Variation in pupal weight was examined by looking at differences explained by the
rearing host plant, and the sex of the individual. The pupal weight was mainly explained
by host plant (Table 15). Pupal sex, and the sex by host plant interaction were not

signiﬁcant contributors to pupal weight. Pupal weights were highest for larvae reared on

27

black cherry, and were signiﬁcantly different than the weights of larvae reared on white

ash and quaking aspen (Table 15, Figure 13).

Days to pupation: Time to reach pupation was also examined as a ﬁtness indicator. In
1997, the length of time it took to reach pupation was not statistically dependent on host
plant, pupal sex, or the sex by host plant interaction. Although duration to pupation was
not explained by host plant, individuals on quaking aspen seemed to reach pupation

slightly faster than other individuals, and the males even faster than the females, but this

trend was not signiﬁcant (Table 16, Figure 14).

28

Discussion:

Papilio canadensis, the Canadian tiger swallowtail, is common throughout the
Northern United States. Adults emerge in early summer, nectar, mate and females lay
eggs on a variety of plants. The eggs ecdyse and the neonates feed, develop and pupate
all within a few weeks (Scriber 1996b). The larvae usually stay on the same leaf, at least
in the ﬁrst two instars and thus selection of oviposition sites by the egg-laying female is
important (W atanabe 1995). If she selects a site less suitable for the growth, development
and survival of offspring, her ﬁtness, in an evolutionary sense, is inferior. Oviposition
preferences may be driven by intrinsic factors such as chemical cues of the host plant that
reﬂect nutritional quality or that are feeding deterrent toxins, and by extrinsic factors such
as protection from predation (Thompson 1988, Thompson and Pellmyr 1991). Many
studies implicate allelocherrricals in the process (Feeny 1995).

Some areas of the Northern Lower Peninsula of Michigan have fewer degree-days
of thermal accumulation and fewer frost-free days over the growing season than other
areas. It is proposed that P. canadensis has an oviposition preference for white ash as a
host plant in colder areas, in contrast to warmer areas, as delayed white ash bud-break and
leaf development would be better suited relative to other hosts to nurture rapid larval
growth (Scriber 1996a). This ‘cold pocket hypothesis’ was a local modiﬁcation of the
voltinism-suitability hypothesis (Scriber and Lederhouse 1992). The purpose of the
present project was to search for an effect of climate on oviposition preference of P.
canadensis. It was supposed that host plants, which depend on accumulated degree-days

in seasonal development, would be of different attractiveness in warmer and colder areas

29

during the P. canadensis ﬂight period, and that egg-laying females would differentially
select among hosts in such areas.

White ash was predicted to be the preferred host plant for oviposition in the
‘cold pocket’, and larvae were predicted to perform better on white ash. Oviposition
preference did vary with inter-speciﬁc differences in host plant, and with large differences
in host plant phenology (newly ﬂushed vs. older, tougher foliage). As host plants may
vary in suitability for larval growth, inter-speciﬁc differences in host—plant quality are
well documented and not surprising. Newly ﬂushed leaves tend to have a higher percent
water content and decreased concentration of certain ‘quantitative’ defensive compounds
(Feeny 1976). As water can be a limiting factor for larval growth and defensive
compounds can reduce or slow larval growth, that butterﬂies would prefer to oviposit on
newly ﬂushed leaves is also not surprising. There is some evidence in the literature that
indicates that leaf age and bud burst phenology can play a role in oviposition preference
(Hunter 1992, Hunter et al. 1997, Scriber and Slansky 1981). In a study with winter
moths, Hunter et al. (1997) determined that local population variation was seemingly
related to plant quality and budburst phenology. Other studies have indicated that
oviposition preference is inﬂuenced more by the over-riding importance of inter-speciﬁc
plant differences than intra-speciﬁc differences in plant quality (Schultz 1988).

Larval performance on the ﬁve host plants was examined. Signiﬁcant differences
in pupal weight on host plants were compared to differences in oviposition preference. In
1996, oviposition preference was the same for quaking aspen and white ash; black cherry
and basswood; paper birch and chamber paper. In 1996, larval performance was similar

on quaking aspen, white ash and black cherry; white ash and basswood. No larvae

3O

survived on paper birch. Similar differences existed in oviposition preferences and larval
performance. In 1997, oviposition preference was not different on: quaking aspen, white
ash and basswood; or quaking aspen, basswood, paper birch and chamber paper.
Preference for black cherry was greater than in all other treatments. In 1997, larval
performance was the same on quaking aspen and white ash. No larvae survived on paper
birch or basswood. Performance on black cherry was higher than on all other treatments.
Again oviposition preference and larval performance hierarchies were similar. Both
oviposition preference and larval performance hierarchies on host plants changed between
1996 and 1997. This change occurred in all populations, and may have been attributable
to the dry May and June of 1997.

Neither oviposition preference nor larval performance followed the ‘cold pocket’
hypothesis predictions in either 1996 or 1997. I tested the assumptions of the
‘cold pocket’ hypothesis, to determine if the initial conditions had been met, and to obtain
a better picture of what was occurring in the ‘cold pocket’ in 1996 and 1997. These
assumptions included determining if bud-break and host plant phenology were delayed in
the ‘cold pocket’ and if the ‘cold pocket’ was a thermally unique area.

In 1996 and 1997, foliar percent water content for ﬁve test plants was measured as
an index of plant nutritional quality. Foliar water content varied by host plant, time of
season and year, but variance due to site was minimal. These data are in agreement with
the climate data in that neither data set found site to site variation, but both detected
yearly variation. These phenological data support the contention that partial (that is,
early) season measurements of climate and foliar water content are important in this

biological system as values are high and then taper off.

31

~

Twenty-nine years of weather data were available for three areas in the Northern
Lower Peninsula. Vanderbilt was always cooler than Pellston and Cross Village when
summing total seasonal degree-days (March lst - October 3 lst) and ﬂight and larval
development seasonal degree-days (March lst - July 3 lst ). The differences in
degree-day accumulations between sites were statistically signiﬁcant for these two time
periods. The whole-season difference justiﬁed calling the Vanderbilt area a ‘coldpocket’.
However, accumulated degree-days through the ﬂight season up to July 5th alone,
Vanderbilt was the coolest site only ( 16 times in 29 years), and site differences over this
period were not signiﬁcant. Early degree-day accumulation is most important to the
biological processes I exarrrined and differences over ﬂight season among sites were
obscured by the magnitude of year-to-year differences within a site. To the extent that
climate indirectly inﬂuences oviposition preference, one rrright expect P. canadensis to
show as much, or even more, lability in host plant choice across years at a given site, than
across sites for a given year. It is also unclear how many catastrophic ‘cold’ years out of
29 years are enough to exert signiﬁcant selection on host choice of P. canadensis. When
climate data was examined across Northern Michigan, it was determined that the
Vanderbilt area was not cooler when compared to nearby areas at a biologically
signiﬁcant time, (March lst — July 5th). Also, when host plant bud-break and phenology
were assessed, it was determined that they were not delayed in the Vanderbilt area in
1996 and 1997. The lack of thermal unit accumulation differences, and similar water
content data supported the contention that Vanderbilt was not remarkable as a ‘cold
pocket’ during recent ﬂight seasons. As such, the ‘cold pocket’ hypothesis would predict

no difference in the oviposition hierarchy amongst butterﬂies from these sites; or among

32

butterﬂies for foliage from these sites. Subsequently, white ash should not be preferred
as an oviposition host plant, and larvae should not have increased performance on white
ash. My results with P. canadensis, were consistent with these observations.

Heat! precipitation indices usefully depict one aspect of climatic variability. I
prepared such indices for four months of Vanderbilt data from 1987 to 1997. The ﬁnal
year stood out as being unusually dry in both May and June. Dry conditions lead to water
stress on a plant. As plant water content decreases, soluble forms of nitrogen increase
(Mattson and Scriber 1987, Scriber 1977, Thomas and Hodkinson 1991). White (see
Thomas and Hodkinson 1991) hypothesizes that water—stressed plants suffer increased
herbivory; this theory is based on the observation that climatically disturbed areas often
have insect outbreaks (see Thomas and Hodkinson 1991). Bultman and Faeth (1987)
tested the hypothesis by examining leaf miner populations as an indication of predation
pressure under water stress conditions (drought achieved by cutting off roots), where
water had been added (irrigation) and in control conditions. Their ﬁndings contradicted
the supposition: Cameraria Sp. B, predicted to decrease, increased in the irrigated
treatment; and Camerarr'a sp. A, predicted to increase, decreased in water stress
conditions. Other studies also have shown that insect larvae perform less well on plants
with low leaf water levels. Larvae on leaves with low leaf water grow more slowly and
are less efﬁcient at utilizing nitrogen, and water content of leaves may limit larval growth
(Scriber 1977). Water stressed conditions can have consequential impacts on plants and
their herbivore communities. The oviposition differences that we saw in 1996 and 1997
may, in some part, be due to the dramatic differences in May and June precipitation

between the two years. The aspen seemed especially sensitive.

33

In summary, thermal unit differences from March lst - July 5th in Northern
Michigan differ between years rather than between sites. Plant quality and phenology
differences vary by species, by seasonal ﬂuctuations and by year-to-year ﬂuctuations. In
comparison, climatic and phenological spatial variations for the three test sites were not
remarkable during the period of biological signiﬁcance for the present project.
Vanderbilt was not meaningfully a ‘cold pocket’ in the 1969-1997 interval at an
appropriate time period, and the ‘cold pocket’ hypothesis could not apply here.
Generalist oviposition preferences were labile and showed year to year ﬂexibility. For
Papilio canadensis, oviposition preference and larval performance hierarchies were
similar. The ‘cold pocket’ hypothesis will have to be evaluated by comparing areas that
differ more in climate during the early part of the year than did Vanderbilt, Pellston and
Cross Village over the recent twenty-nine years. Although the whole season
(March lst — pupation) selects against oviposition mistakes on the wrong host plant, it is
early season differences in host plant quality that the female must evaluate.

Vanderbilt had the lowest seasonal (March lst - October 3 lst) accumulated
degree-days over a 29 year period, as compared to Pellston and Cross Village. The ﬂight
season (March lst - July 5th) accumulated degree-days did not differ between Vanderbilt,
Pellston and Cross Village. Vanderbilt was generally the coldest site (March lst - July
5th), (21 times in 29 years); (March lst - July 3 lst), (23 times in 29 years);

(March lst - October 3150, (29 times in 29 years). Year to year ﬂuctuations in
accumulated degree-days were greater than site to site variation. Although there were
year to year variations, the Vanderbilt site showed that there was a trend to the variance.

This may be indicative of long term climate trends with short-term variation. Hypotheses

34

based on supposed climate averages for a site may not usefully predict the outcome in any
but the most average of years.

The assumptions of the ‘cold pocket’ hypothesis did not hold true in 1996 or
1997. The ‘cold pocket’, during the behaviorally critical time period (March lst - July
5th), did not exist. There were no differences in intra speciﬁc host plant differences in
phenology between the sites. Even considering the limited availability of butterﬂies, the
white ash preference did not exist in 1996 or 1997 (only 19% and 18% in a ﬁve choice
study respectively). There were no differences in oviposition preference of butterﬂies
from different p0pulations. Larval performance and oviposition preference hierarchies
were similar in 1996 and 1997. Larvae did not perform well on white ash in either 1996
or 1997. This leads to interesting speculation regarding the interactions of plants,
herbivores, the prevailing climate conditions, and the evolutionary signiﬁcance of these
interactions. The ‘geographic mosaic’ theory of coevolution describes the evolutionary
landscape as dynamic, where coevolutionary relationships are not static across a host
species range, but rather are labile in response to host plant distribution, competition and
environmental differences, among other factors (Thompson 1994).

Given the nature of the relationships discovered and tested in this project, P.
canadensis oviposition preference in the ‘cold pocket’ readily conforms to the deﬁning
principles of the geographic mosaic theory of coevolution. Under a given set of
conditions, reduced number of accumulated degree-days and delayed bud-break,
generalist herbivore oviposition preference was for a normally poor quality host plant,
white ash. When these conditions varied, host plant preference varied. Snapshots of

three different climatic conditions resulted in three different oviposition preferences. In

35

cold years (1992-1995), white ash was preferred. In a thermally average year (1996),
with average precipitation, quaking aspen was preferred. In a thermally average year
(1997) with low precipitation, black cherry was preferred. Deterrrrining whether or not
these relationships between localized climate and localized preference are true
relationships, or artifacts due to either experimental procedure, low number of butterﬂies,
or another source of variation, is an important priority in continuing this line of research.
First, one would have to examine the preferences across years and sites and try to
distinguish what, if any, trends exist.

This project indicates that both temperature and precipitation can be important
factors inﬂuencing plant-herbivore interactions, and subsequently could be important
evolutionary selective factors. In addition, responses to varying soil type, and other
factors such as geographic variance and photoperiod could be controlled by common
garden experiments or other studies. One example would include not only examining
oviposition preference of ﬁeld caught butterﬂies on ﬁeld collected foliage, but also
butterﬂy preference on foliage from trees reared in speciﬁc conditions. Through this
combination of oviposition arrays, one might be able to determine if localized
populations exhibit any variance in oviposition preference, or if the differences in
oviposition preference are a species-wide response to differences in host plant quality.

My research showed that ash preference of butterﬂies from the Lower Peninsula
‘cold pocket’ of Michigan was less than 20% in ﬁve choice arenas in 1996 and 1997.
This is a decline from the observation of these same populations in 1991 to 1995 that
showed ash preferences of 92%, 71%, 60%, 39% and 34% respectively (Scriber 1996a

and unpublished). This is especially interesting in view of the increase in seasonal

36

degree-days observed during this period (e.g. 700, 800, 900, 1000 from 1992-1995;
Figure 1) which could allow inﬂux into the ‘cold pocket’ from surrounding areas, and the
survival on most host plants during this period. Since 1991, there were no severely

constrained years that could select out non-ash preferring females.

37

Table 1: Scientiﬁc and common names of host plants examined for P. canadensis

oviposition preference

 

 

Scientiﬁc name Author Common name
Betula papyrifera Marshall Paper birch
Fraxinus americana L. White ash
Populus tremuloides Michaux Quaking aspen
Prunus serotina Ehrhart Black cherry
Tilia americana L. Basswood

 

 

38

Table 2: Accumulated degree-days °C (threshold temperature 10 °C) at three sites in

 

 

Northern Michigan
Site Seasonal Mean Flight Season and Larval Flight Season
Accumulated Development Period Mean Mean Accumulated
Degree-Days, Accumulated Degree-Days, Degree-Days,
March I - October March I - July 31 5; SD. March 1 - July 5 _-r_-
31 i SD. 5.0.
Cross Village 1047.1 1 87.3 620.4 _-1_-_ 238.5 424.3 :1; 342.2
Pellston 982.7 i 88.2 521.3 3 56.2 269.3 i 45.9
Vanderbilt 856.2 _-t_- 119.1 462.3 i 69.4 253.3 1; 45.4

 

_

 

ANOVA
F value 20.39 12.40 1.67
p value <0.0001 <0.0001 0.19

39

Table 3: Stepwise regression of accumulated degree-days at three sites, 1969-1997

 

Regressor p Value Contribution to r2 Remaining in model?

 

March 1 - October 31
year <0.0001 0.28 yes
site <0.004 0. 15 yes

adjusted r2 = 0.34

May 1 - July 31
year < 0.0001 0.28 yes
site <0.004 0.08 yes

adjusted 12 = 0.34

May 1 - July 5
year <0.08 0.06 yes
site <0.18 0.04 no

adjusted r2 = 0.07

 

 

40

Table 4: Stepwise regression of water content of leaves of ﬁve species collected at

Pellston and Vanderbilt, 1996

 

 

A.

Regressor: p Value: Contribution to r’: Remaining in model?
Date <0.0001 0.42 yes

Site <0.4 0.0007 no

Tree Species <0.009 0.006 yes

 

 

r2 = 0.42, adjusted [’2 = 0.42

Adjusted for Degree Day Differences, 1996

B.

 

Regressor: p Value: Contribution to r2:

Remaining in model?

 

Degree-Day <0.0001 0.41
Site <0.02 0.02

Tree Species <0.01 0.004

 

yes

yes

yes

 

r2 = 0.42, adjusted :3 = 0.42

41

Table 5: Stepwise regression of water content of leaves of ﬁve species collected at

Pellston, Vanderbilt and Cross Village, 1997

 

 

Regressor: p Value: Contribution to r2: Remaining in model?
Date <0.0001 0.35 yes
Site <0.0001 0.009 yes
Tree Species <0.0001 0.02 yes

 

 

r2 = 0.38, adjusted r2 = 0.38

Table 6: Stepwise regression of water content of leaves of ﬁve species collected at

Pellston and Vanderbilt, 1997

 

 

Regressor: p Value: Contribution to r2: Remaining in model:
Date <0.0001 0.29 yes
Site <0.0001 0.007 yes
Tree Species <0.0001 0.01 yes

 

 

r2 = 0.32, adjusted r2 = 0.32

42

Table 7: Stepwise regression of water content of leaves of ﬁve species collected at

Pellston and Vanderbilt for 1996 and 1997

 

 

Regressor: p Value: Contribution to r2: Remaining in model:
Date <0.0001 0.33 yes
Site <0.005 0.0003 yes
Tree Species <0.0001 0.008 yes
Year <0.0001 0.06 yes

 

 

r2 = 0.36, adjusted :3 = 0.36

43

 

 

.850 some 88% 850$:me bﬁoumzﬂm 8: 0.3 Bao— oEam 05 53, 823/ 08550... mod n 8 a “56$:me ...
Gas 93 2 Sea... Eda so: a Ewes €525
G sou H 3 ceded .8820
< sou ...r. new can 9.230
m seam H ﬁﬂ nooERm
< Sod m oém awe. 32>?
o sod H3 :23 ceded
m secs +2: \Eeea scam .. _Sodv 3.2 m macaw ass ,8:
as; e: _ sausage Essen
$2 .o mo.m _ some .5335 Ban—i899»
as? 42 N Ewes €33
«3.0.03 Q3 ”knack Am
amoewameﬁgzmgaeeh
gem wwwm Suﬁsm \o 2325 ESE :6: 3:3» a 2:5» k 4% 8.56m

 

 

860% ESQ 60: gm no watchers.» .& mo 00:29.8.“ 528920 33 .8 8.3.8 <>OUZ< ”w 038.

44

.850 :08 Soc 8853?. bauzmzﬂm so: can 8:2 2:8 2: 53» 82.; 039260 36 u 8 an Enocﬁwmm ...

 

45

 

...
880% :33
83¢ 98 m as: a Ewes sad as: a 53.5 €88
mmwmd 84 v 888% .53 H8: x Ewto :83 “8:
$3 93 w 88% 2% seem a 53.5 €28
8o; 2 .o N Ewes 22a 8: a ease €88
0 seem H 8 8am
om s? H 22 =83 mango
Om Sod ....m 9m _ given
m Asian + QR :8 83>?
0 seem H 8 e23 ceded
< sewn H ode Edna ace—m ... 586v Nod v 888% :83 80:
ammo :3 a tendons; Essen
owwod cod _ a. own xccozsn Beamxoame.
as; 2 .o N Ewes €88
33.0 :3 _ Ewes saga as:
«36"»: 93 “.185: am
neozmawaggzwgcmh
8% 3mm casuammsso 58:5 EB: 36% 33> m 85:; k \V 8.56m.

 

 

8.8% Ema .8: 02C no £33638 .& Lo 3:80.88 553830 33 .8.“ 3:58 <>OUZ< no 035.

850 :08 Bob “505:3,“ £30583 8: 0.8 8:2 2:3 05 53, 825/ 238.60

86 n a a 385:? ..

 

 

o
.886 onN e own a... 233 x ENE €058
85¢ 23 _ .v smegma? €28
mead cod _ ... own mﬁoznn ouaﬁmxoam’x
I Looodv 3.8 N cm“ 5.“ 333
0 $5 + 9N 39$
m NEW m mam we;
< $56 + how 20 83.0 3.0 m 53.8 ~5:525
«3.an 93 mini: .3
amu=88§§=m§8h
8Q ammm 3.88% \e 2825 ESE 98% 85:; a 853 k km 8.3%.

 

 

:8 233 38:38 3:5 Bo 93 was?» so £88828 .& mo ooaouomoa 52893 32 ..8 £82 <>OUZ< A: 2an

46

 

 

 

.850 :03 Bob Ego—mama b33693 8: 0.3 3:2 083. 05 53 mos—Sr 83550... mod n d as 835:»? *
59.5
am? So w :33 as: x :35 @025
385 R; M a. Esau? €23
whmnd mmd ﬂ .9 own zcuozsn oumemxoamax
on? 84 N 538 $395
m $3 H 3. 59a
< chw H wdm 888.0
< $3 H q: 03:3 :89
< exoww M ”2 2382:;
< $3 + tam SEE “”885 93 v are Ea so:
«Edna» Q3 hgmﬁi Am
nwuzmgwggzuggh
hi 3mm 38ng Kc 5325“ ESE 33¢ 3:3, Q 2:? K Kn 92.8%

 

:3 BE? 5 mooceobﬁ 33385;;

E was—38 .85 So“ 89c 638:8 can 853 no amazuﬁguu .m we oocouomoa 558330 39 .8 838.. <>OUZ< ”2 03a...

47

wouatg’oo

 

 

 

.9
an; N; _ a. 38353 €23
othd 3 .o _ o own xuugsn oamﬁﬁoamxx

$3. H on “an“.
§Nd H m8 occuoo<
§Né W v.2 333
§~é H 5.2 888.0
$3 M 3m ems; 220
§~é H NE :3..ou§>
§N€ + v.2 scam—Em good wmd c 882m
$253:
gum ”mum Euuxum \e Equﬂ E33 :62 3:?» A wig E kn 3.30m.

 

was .58 Set $8.58 can 233 no £33328 .m mo coup—£03 .838an 82 ..8 3.32 <>OUZ< ”Q 035.

00
4

Table 13: 1997 P. canadensis larval survival in each instar for larvae reared on black

cherry, paper birch, white ash, basswood and quaking aspen

 

 

Source Denominator df F value p value

Host plant 1202 16. 12 <0.0001
Mother Host plant Preference 519 0.11 <0.75
Instar 519 0.01 <0.92

Time 519 99.00 <0.0001

 

 

Table 14: Mean and percent differences in pupal weights of P. canadensis reared on black

cheny, paper birch, white ash, basswood and quaking aspen, 1996

 

 

Host Plant Total Survival Weight (g) 1 SE Pupal Weight Diﬁ’erences By
Fisher ’3 LSD (a=0.05)

Black cherry Not measured 0.84 i 0.01 A

Paper birch 0

White ash Not measured 0.77 i 0.02 AB

Basswood Not measured 0.59 i 0.13 B

Quaking aspen Not measured 0.90 i 0.02 A

 

 

Values with the same letter are not statistically signiﬁcant from each other.

49

Table 15: Mean and percent differences in pupal weights of P. canadensis males and

females reared on black cherry, paper birch, white ash, basswood and quaking aspen,

 

 

1997
Initial Overall Weight (g) i Pupal Weight
Number Survival SE Differences by F isher’s
LSD (a=0.05)

Females 78 0.69 + 0.02 A
Males 72 0.66 i 0.02 B
Black cherry 303 33% 0'73 i 0-01 A
Paper birch 23 0%
White ash 411 7.3% 0.66 _-l; 0.02 B
Basswood _ 32 0%
Quaking 141 13.5% 0.64 i 0.03 B
aspen

 

 

Values with the same letter are not statistically signiﬁcant from each other.

50

Table 16: Mean and percent differences in days until pupation of P. canadensis males and
females reared on black cherry, paper birch, white ash, basswood and quaking aspen,

l 997

 

Treatment Overall Survival Days Until Pupal Weight Differences By

Pupation 1; SE F isher's LSD (a=0.05)

 

Females 27.00 i 0.66 A
Males 26.39 i 0.65 A
Black cherry 33% 27.37 3!; 0.45 A

Paper birch 0

White ash 7.3% 27.32 3; 0.83 A
Basswood 0%
Quaking aspen 13.5% 25.37 i 1.02 A

 

 

Values with the same letter are not statistically signiﬁcant from each other.

51

 

1400

1300 . A Seasonal Degree-days: March

1200ql-October31 ‘ I
I

1100 ‘

3%

l

800 -
700 a
600 q
500 .

 

Accumulated Degree Days
(Base 10 C)

 

 

 

I T T I T

400

1965 1970 1975 1980 1985 1990 1995
Year

I Pellston O Vanderbilt A Cross Village—Linear (Vanderbilt)

 

800
B Flight and Larval Development

700 ‘ Season: March 1 - July 31 3
600 ‘ . I

500 *

 

400«
3004 '

 

Accumulated Degree Days
(Base 10 C)
1..
o I
o
c
O
o
n
o
I
- .
on

 

 

2m l I I I T T
1965 1970 1975 1980 1985 1990 1995

Year

 

450
4w ..
350 - 2 I n
300 - '

250 .
200 ~
150 -

C Flight Season: March 1 - July 5 I

 

 

 

 

Accumulated Degree Days
(Base 10C)

 

l m I I I T I I
1965 1970 1975 1980 1985 1990 1995
Year

Figure l: Accumulated degree-days at three sites; Cross Village, Pellston and Vanderbilt,
1969-1997. A is seasonal (March 1st - October 3150 degree-days; I; is ﬂight and larval
season (March lst - July 3130 degree-days; _(_3 is flight season (March lst -July 5th)

degree-days. Linear trends are illustrated for Vanderbilt data. The trend is statistically

signiﬁcant in A (r2 = 0.27, n=29), but not in _12 (r2 = 0.18),nor in Q (r2 = 0.09).

52

 

 

 

 

 

 

 

 

100
/// l93.06
/ ‘ 30.43
20 e
I
§ .5
.5
:3
8
:2
.9-
b
3 10
:1:
o
I
I
5 I O C
A
2 .
A . A I
. ' o O
I
o S 0 A
o I ' o
. O
1986 1988 1990 1992 1994 1996 1998
Year

Figure 2: Heat/ precipitation indices for Vanderbilt, Michigan. May (diamonds), June
(squares), July (triangles), and August (circles) 1987-1997. July, 1989 and June, 1991 are

extreme outliers.

53

 

é

 

 

 

 

 

 

 

 

 

 

 

A Black cherry
‘E 90 .
8
g 80 ..
‘3 I
§ 70 - " 0 I
E 3 .
3 6° ‘
£ 50 «
40 r r r 7
140 160 80 200 220
Julian Date
100
B Paper birch
‘5 9o «
8
I:
5 so «
§ “ o
a 70 ‘ I
3 . O
i“
a: 50 -
40 l r Tr I
140 160 180 200 220
Julian Date
100
.. E Quaking aspen
§ 90‘
t:
e
U 80 ~
I- I
0
3 70 "
3
E 50 . °
§ ' :
a 50 - '
w 7 I T I
140 160 180 200 220
Julian Date

é

 

8

Percent Water Content

00
C
1

8

LII
C

C White ash

\I

C
L

I

 

S

 

140

§

8

l

on
O

s)
O

.L

Percent Water Content
8

VI
0

l

8

160 l 200 220
Julgtn Date

 

 

D Basswood

'o.

l

l

 

 

 

f 1

1 80 200
Julian Date

160 220

g

Figure 3: Percent water content of leaves from ﬁve species collected at Pellston (squares)

and Vanderbilt (circles). A is black cherry; Q is paper birch; Q is white ash; Q is basswood;

and E is quaking aspen. Data from the 1996 season.

54

 

 

 

 

 

 

 

 

 

 

 

 

100
390 q A Black cherry
8
38°10. . .
*- A
‘W i ‘ “ ‘ -
E I ~ A ‘ '
860‘ I . ' I ‘
h IIII
9350*
40 J Y I T
140 160 180 200 220
Julian Date
100
BPaperbirch
39°“
3
530.
o _
II
a , J! II
'70 ‘ A.“ ro“
3 I . . i 2 I [r
i“ '
£504
40 T T I T
140 160 I80 200 220
Julian Date
100
EQuaking aspen
390‘
3
830.
E3 ’-
3 0"
«70‘ ‘.
3 IA ‘ A
g... ' .‘ I; a r
I I ' I
£50+ '0 o
40 I J T r
140 160 0 200 220
Ju8 Date

Figure 4: Percent water content of leaves from ﬁve species collected at Cross Village

 

é

 

 

 

 

 

.. C White ash
3%»
r:
O C A
80 4
: J I. '
370* I O .0
3
550 ~
40 I I l'
140 160 .180 200 220
Juhan Date
100
DBasswood
390‘
g . .-
80 ~
8 A
l- ‘ A O 0
£70“ 0 lb.“ : 0 °
3 n f I
56°“
3150 -
4o . r .
140 160 180 200 220
Julian Date

(triangles), Pellston (squares) and Vanderbilt (circles). A is black cherry; g is paper birch;

Qis white ash; _D is basswood; and E is quaking aspen. Data from the 1997 season.

55

 

 

683m 03— 2: EB... Sun .83 EVE—om 2: an So? 823..— .5330
v8.8.3 a he 3.3.6.“ 95 8b :3 25 as" .8 238 ..o Echo c8237. EB 88:. uoEomoEom A333 .32 o:

8 .86on 2,: mo «.98. :o 5:32—36 .8 $3.. 3:82.308 .3 HE 285:?— 3qu 2: E 880:8 8 .939 3:26
.2 "5 =_€oc§> E 880:8 ABS coats. 3.53:0 ._m n5 sew—Em E 3.00:8 85.523 Co 8:285 “m 2&5

:3..— “no:
human— :oama wee—30 vookmmam can 33>» :83 eon—am

mam .V\

Pogo gum—m

 

 

 

 

 

 

 

 

 

 

 

 

o—

 

 

 

 

 

 

 

 

 

 

 

 

O
N

 

 

 

 

 

 

O
M

 

 

 

I:—

 

 

}——
O
<-
“omsodng waned

 

O
in

 

 

 

 

on

56

homun—

.=o.maom 32 05 EB» Sun—
.33 :o.m=on_ 2: 89¢ 22: 823... .533 30:23 a 8.. 36:35 Ba on»... an 28 >5 .8 23:.

mo Echo 2356 98 £32: Bagged damn: .30— o: no .3627. o>¢ ho 323— :0 5:62.30 he 3:5 3:8
.2 "5 2332:; E 380:8 Ago 3&5 :28me .h "5 Sam—Em E 380:8 8553.5 mo 02.2895 no Esmﬁ

«Sm—m “me:
593 $5.30 803mmum :8 323 :85 SEE Dunno mun—m

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2

 

 

 

57

 

 

 

 

 

 

 

 

 

8

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

O
M

 

 

 

 

3
uomsodpo 103:)an

 

 

 

 

 

 

 

o
n

 

 

 

 

 

on

4.8an 32 on. :88 5.5 do... ._B.oc§> 05 :58 0.03 82.3
.cwﬁov 32.23 a .8 3.3:... 0.3 an». .3 25 b8 .8 23:. .80 .25 83:8... E... 2.3:. BEomoEom don—«3 «no. 2.
.o .86on 28 L6 828— :0 551.0930 .8 $53 3:82-203 .v "5 £3.55.” .RED 2.. E 8.00:8 .0 A39 v2.8.
.N "5 £88.53 5 3.00:8 A83 82... 3.8%.? .m "5 EVE—om E 3.00:8 moﬁbzsn .0 35.80.; K eswi

2.2.— .8:
.09.; :25 9.330 .603QO :8 32>» :85 .09..— ~0...... ..an

 

 

 

 

 

 

 

 

 

 

 

 

@—

 

 

 

 

 

 

 

 

 

ON

 

 

 

 

 

 

 

 

 

on

 

 

 

 

 

 

 

 

 

on

 

 

 

 

 

 

 

 

 

 

 

nomsodng manna

 

 

oh

 

 

ow

 

 

 

 

 

 

oo—

58

.330. mg. 05 :88 Sun :33: 30:23 a .8 3.3.8: 0.: 0%. .3 0:: b... .8 0:008 06

m.o..0 83:8. :5 33:. v0.5.0.3”. A83 :28. @033 8:0. 0: .0 A83 0:: 30...? .33 00.50. :8 0.53 3:598
38 A83 0:: 3.38.3 3055 «.030— :m: 0.23 33:98:: :0 5:683: .8 Am "5 :88). :5. CHE 332.30
ANNE 3:80 Am "5 235:8. .033 0:. AN "5 580:5; .? "5 :o.m:0m E 3.00:8 8.5.2.3 ..o 00:0.805 ”w 0.:wE

3.0 558:5 $.33:
:83). £92.30 3080 as ._B.0..::> :o.m=0m

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

-2.

 

 

 

-ow

 

 

3,——
cl: o
n V?
uomsodng gunned

 

 

 

 

 

59

.:0.00. 30. 0... :88 San. .:m..00 0005...... a .8 00......00 0.: 0a.... .3 0:0
as. .8 .50.: .0 ..0..0 0.00:8. 0:: .50... 02:80.03. A300... ..00. 0: .0 £0.00... 02:. ..0 .030. :0 :0....0...>0 .8
...... 0000..» .N. u :. 30.3.0000 00:39.00 0... 0:: 3.0.. 00:30:-..08 ..n5 0.35:0. .0003 0... .....3 00:00

.m u 5 £80053 :. 00.02.00 .....5 00...... 0303.0 .0 "5 553.. :. 00.02.00 38.0.25 ..0 00:20.0... .0 20w...
3% 5:00:00 ...: 0.0.3

.00.... .0585 0w....> ..80 50.005; :0..=0n.

O o
N v—

C
M

5% 8
uopgsodyso waned

8

o...

 

cm

60

.538 32 2: Eat San dwmmow 30:23
a .8 33:?“ 0.8 v.52. .3 £83 Emu—gm 98 £808 Bucomoaom $3.250 mm 338 =~ .592. :0 Eu— mmwo
was .2828 0:908 2: .23 @289. .32 a £033 .23 358% .32 < .aucv 888.0 25 =5398> 603:0; awn—.5

380 Got 380:8 838— :3 823 .3 30838 3552: :o ﬂarwngau 325$ mo oocaomoa 55330 ”9 Esmi

:8 323 assim— bauEEG .... 35 538:8
895 ocouoo< 533 8on0 emu—.5 820 =_£on=u> gem—Em

 
 

 

 

 

I
52

 

l
V}

 

 

 

 

 

uoggsodyxo waned

 

.__.
O
N

 

 

 

 

Om

61

.538 39 05 89... Sun 5803an Ho :85 59E so wuztsm 8?.“— ozv dogma maﬁa—5

Ea 89$an .an 823 £23 Baum ..Eono x85 ”Sam—Q .8: o>c co 9.35333 953$ .«o 3:28 .88. u: 8amE

unar— «ma—.—

 

 

 

 

 

594“ was—£5 803QO :8 323 :83 59$ Pogo gum—m
. § _ . V\\ 95¢
\\\U §n

 

 

 

 

 

 

$2

 

 

 

 

 

 

 

a?
o
N

 

 

oxcmm

 

 

 

9&9.

 

 

§\\\\\\\\\\\\\\\\\\\\\\

 

 

 

$3

62

 

.538 83 05 an Sun 225 human :0 3:33. out“. czv dogma wcimsc 600333 .53

8E3 .585 human .Eoco x83 ”SEW—m .8: 3: .5 3.82 ﬂatmwgco Paint no 3 ﬂaws? _nmsa :32 ”S oBmE

25 a8
comma 35.2.0 803QO _n— I in 223 atone gum—m

, Nd

4». 4,, «A.
o O O
(3) "Imam

l

t
«3.
O

 

\
l‘.
o

 

 

 

|- wd

 

 

 

63

.538 So— oﬁ 50¢ San— Aﬁooamman 8 :83 59a :0
633.5» oats— ozv dogma mac—25 68333 .5“ 833 525 human 5.55 x83 ”853 “no: gm :o 332

3.3%38 9:..un Gan vow—0E .3. u 5 255% E5 9:3 686“. .3 .I. 5 BE: .3 Amy 2:303 39a :32 ”2 oSwE

«an: «no:
:83 9.32.0 :3 2:5» Pogo :83

1M.
«ﬁdo

V?
C
an [a

!

d (3%“?

°9
o

0.0

 

.588 32 05 Sat Sun @8333 he :25 human
:0 BZEQ untu— 36 .593 act—«.6 6833.3 .nma BE? £23 Bag .905 x83 ”£53 “mo: u>c co c933

3:38:39 Smack 23 Riot: db "5 oEE£ c5 $.39 wagon .Q. n 5 29: 8% conga :25 "than :32 “E Bsmﬁ

«Ea—.— 30:
5%“ 95.30 =3 SE?

.928 sou—m

uonndnd mun ska

 

 

 

 

 

 

aw

65

APPENDICES

APPENDIX A

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.: 1993'5

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

Testing tre 'Cold-Pod<et' Hypothesis: Ovipositim Preference of the Canadian Tiger
Sellmtail

Museum(s) where deposited and abbreviations for table on following sheets:
Entomology Museum, Michigan State University (MSU)

Other Museums:

lnvestigator's Name(s) (typed)
E' I! G'

 

 

Date m 1cm

 

‘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.

Page:

2

 

of

2

APPENDIX 1. l

Voucher Specimen Data

 

Page

 

3.5 t 3.950 mam H umsmqa m N 8.5

\L

%§\§\N .2... ex

b.3323 o w 535:2 05 E .383 .
.8. 9.2592.» coda: 33a 05 3283. 5.5.35 M 9."on

clmooﬁ .oz eonoao> .3252 «1639.33—

 

 

 

 

:Qanmuuo: .= 38:» 333.23 one

 

soudo «swam .mmmunmuH> Hz .suesoo
xHo>oHumno Boom xUOum powwow nma

 

 

 

 

 

 

 

 

n

a

8

.1

.m )

m w souﬁu muoﬁm .wmmHImIC/ Hz 33550

..m comcﬁxoﬁm 80.5 xo0um powwow nma

W Y x3930 .3on .mmmﬁlamiw

% ....L. Hz $3550 commz .douwcﬂosa

m m xsouﬂo mamum .mmmﬁunmu> Hz .zuasoo

Yw comcwxoﬁo aoum xUOum powwow nma

mrm xaouﬁu mumum .w¢¢H-m~u> Hz .suasou

M. U comaHxUHo Eoum xuoum powwow 93

m e

w n 39.50 ..w muoﬁm .mmlmmlw Hz 33:309.; mﬁmdovmamo ,

h “we 3 3 domawxowa Boum xooum powwow £3” 0.3.“ mm

m w d O m 3:396 93 can: .8 unto—.8 :98. .350 3 3.on
a M m e D. e ado—Sauna 3.. Sat .33

mmwm u u m m W m

de M M .m N E

 

 

 

 

”..o .8952

 

 

APPENDIX B

APPENDIX B

Table 17: Stepwise regression of water content of leaves of ﬁve species collected at

Pellston and Vanderbilt adjusted for degree-day differences, 1997

 

 

Regressor: p Value: Contribution to r2: Remaining in model?
Degree-Day <0.0001 0.29 yes
Site <0.0001 0.007 yes
Tree Species <0.0001 0.01 yes

 

 

r2 = 0.32, adjusted :1 = 0.31

Table 18: Stepwise regression of water content of leaves of ﬁve species collected at

Pellston and Vanderbilt adjusted for degree-day differences, 1996 and 1997

 

 

Regressor: p Value: Contribution to I}: Remaining in model?
Degree-Day <0.0001 0.32 yes
Site <O.2 0.0003 yes
Tree Species <0.0001 0.008 yes
Year <0.0001 0.06 yes

 

 

r2 = 0.35, adjusted r2 = 0.35

68

.588 3a— oﬁ Soc Sec .32.“ :33ch EB zoom—Em 2: So: 203 8234 .cwmmou
noose—B a .8 33:5 2a 090 has one be é 33.: mo Echo pave—3m new 33.: 333.893— A3926 .32 o:
.5 .86on 2E no 838. :o sausage pom A23 3:82-380 .v "5 28555 395 2.: E 380:8 .5 3.23 ooze—o
.2 "5 .__€oc=e> E 380:8 .93 para Enema—o .2 "5 53:?— E Boon—.8 woe—8:5 mo 082035 ”3 oEmE
.693 :88 mes—«=0 wooing—«ca: .8: :8 323 :23 Band .905 x85

— P —

Wmmmmmm mm” \

 

   

 

 

 

 

 

 

 

 

 

 

o—

 

 

 

 

 

 

 

 

om

 

 

 

 

 

 

 

C
m

 

 

 

 

3
yo maxed

 

[sod

 

 

 

 

O
'11
non

 

8

 

E.

 

 

 

 

Ow

69

APPENDIX C

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

w _od _o.o TR Hod _owm _o.o —o._ seam— 8&—
_~: .3 _2: _S: TR 2 _gm _o._ seam— 2&—
_$_ 2. a: «.8 _SN E _w.: m: 33821 S :—
_ww 3“ Sn w: _oam 3 _mé w: $3806— a:—
c: n. as 3 TS. .: _3 _ _o._ seam— ONE—
: .3 a: 2 TR _o.o _aam —o._ seem— 2 E—
v: __ s m. _ n 3 _w.n _oi _3... _3 sesam— : E—
S _3 Sn 3 _Nem _o.o _Sm _o._ Beam— 2 E
3. _3 «R .3 3: _ed _3. _o._ Beam. 3 E—
; w. _ _ ”.8 _os _3m _2_ _3 _3 555m— : E—
; _b m 3:. 3 T2 _3 _EN 6.. 555m. s E—
.8 .o o _S S: _m.m _o.o _mﬁ _3 seam— . _ :—
_§ _N. _ _3: 2 3m _3. 3.2 _o._ seam— e. :—
_S _o o .03 3.2 2 _o.o _34. _o._ -355- 8%
m: _3 ..2 Te 3 _eé _wd _o._ seem— 8E—
xm To 2;. _S S». _E —o.e _3 segm— 8:—
E _o o Ea _o.o 5% _od w; _3 555m— 8:—
3. _o _ .93.. _3; s. _ m _2 _ 93 _o._ seam— 8E—
? _E H: _2 ...? TN _2; _o._ seam— SE—
— wwwm _m.o,_.— eaam— :33 mcigc— too3mmam :3 BE?— gouﬁ hamm— atone xosm— 8:».me cos—woo; 25— 89:52 8522—

 

 

.:mesneeeoﬁ .3 02.2803 838930 8650 3E 82 no. 2an

U N—Dzmmm<

70

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3“ _od _gn w. : _3 _S; _o.n _o.~ 5223i 8&—
v To _o.% _o. o _oﬂ _o. o _c 2 _3 £923.6— 8:—
v _od TR _o.m~ _o. o _o. on _o. o EN 52235— $5—
3 To #2 _a: _o. R _Z __.2 _3 52236— $5—
? _ _N.m T. on T: _3 _~.m _2 _o.~ 59,235— 3:.—
_¢ _o.o —o. o _odm _od .3 .8 _o.N 522.20— S:—
_o_ _ n2 _2: _§_ _Em _m.m w: _o.~ 52251 SE—
3. on _odm __ .2 .05 _3; 3 _o.~ 59,255— 33—
M: 2: _wé v.2 v.2 _o. S Nam _o.~ €3.35— 8%
mm mm .2 _Eu. to _o.o v.8 _3 €5.35— am:—
3 5: N8 _2 won To 6.3 _o.~ 52,256— R:—
.o is So _o.o .3 To _m.mm _o.~ 52235— RE
2 =3. 5. _ v _5: _od _5: _3 _o. N 592.20— ﬁz—
_ _od .3 _o.o _o.o _o.o _o. 9: _o. N 59256— 8:—
_ _o.o .35. To _cd 75 _o. o _o. _ E3826— 8:
8 _od .93. —o. ca 38 _3. .2 _o._ $885. no;
2 _o.o ..om _ .a TR _od 3. _3 5382.0— 3:
a: _E 5.2 ..2 _v. _ ... __ s E _o._ $3826— 8E
—$ _va :3 § _o.o _~._ «:8 _o._ 5&826_ RE
_: T: .9: .3 _od To 33 _3 53821 R r:
Fox _odm _v.: _o. 2 _m.m~ _2: 2.. _o._ cmﬁsogu_ Va:—
_cn #3 —o. 8 _o. v _3. _o.o 3.8 _3 $3801 SE—
. _od _o. 8. _o.o _od _o.o _o.o _o. _ 5&821 can;
g _E __ .3 _2 _ _SN __. _ _w: _c. _ signs; $7.—
3 _od _3 _38 _nS _o. o _3 _o._ 555— $5—
5 _3 _N. _ N _2 _ _Em _Sw _~.: _3 Beam— 2.:—

 

 

 

71

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

w o... 3 o... o... nNo 2m 3 $2.96 Na...
. o... 9o 98. o... 3 3 3 $2.95 8:
N3 ...N ......N ..N 2m 3. 3. 3 $2.96 a...
m o... 3 o... 98. o... o... 3 $2.96 3....
mm o... 3 98 98 ..a 3 3 $2.95 Sr.
2 3 NN 3.. 2. SN No. 3 $2.95 NM...
2. ..0 2:. EN N..N 3 ..N. 3 $2.95 2:.
oN 3 3N 3m 3. 3 98 3 $2.95 ......
.N. 3 N... 3.. 3. 3 SN 3 $2.96 0:.
S 3 3.. 2 3 9.. _ ...NN 3 $2.96 E...
N... Na. EN 3. N.NN 3. .... 3 $2.95 E.
a N... ...m. 3. EN 8. 3N 3 $2.95 E:
8 N... nN. m. .N 3. we 3.. 3 $2.95 .8;
..N o... nN. S: NdN NdN nN. 3 $2.95 we...
8 ...2 3m 3 Nd. 5 tn. 3 «$395 SE
an 3 oNN 3 3. 3 3 3 2.9.2.2 2:.
E 3 EN SN 3. ..N. N... 3 02:39). NN:
.... 3. ..N o... QNN N... no 3 89.82, .N. E
N. 3 3o. 3 o... 3 3 3 89.82 cm...
NM. 3 N3 .... . .3 .3 no. 3 $2.95 ON...
N 3 3o. 3 3 3 o... 3 $2.95 MN...
0N 3 ...: 9: SN 0.... 3. 3 $2.96 NNE
3 ..o 3 N... NNN EN com 3 $89.6 0N:
Nm ...: N.N.. ...m. .3». 3. I. 3 $35 No...
o. 0... ON... o... N... o... EN 3 cacao 8:.
NW 3. ...Nm 8. n2 2N ...m 3 $8.0 ow:

 

 

72

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

m . 0.0 0.0 0.0 0.00. 0.0 ..mn 55:00—0N 0N.o>0...0..Q om~0
N 0.00. 0.0 0.0 0.0 0.0 0.0 55:00 0d 0.55.0.5 th0
0. ...... 0.0 OR 0.2 0.0 ...... 55:00 0d v03010.5 mmm0
0N w.m v.2 m.~v w.0m ma M...” 55:00 0.N 50.6.3.5 3N0
w 0.0 0.0 0.00. 0.0 0.0 0.0 55:00 0d U2.32.0.5 NmNn.
0m 0.0m M... . Own 0.0m 0.0 w... 55:00 0.N 5.05.0.5 .80
E. 0.0 Ev. m0 0..... 0.N. 0.0. 55:00 QN 0.55.0050 0m~0
mm 0.0 .0.» m.w~ 0.0 0.0 QNN 55:00, 0.m 0.55.0.5 m~m0
0.0 0.0 0.0.V 0.0 0.0 0.0 0.00 55:00 ,0.m x_o>0_0:U >30
0 0.0 .... ... . 0.0 0.0 0.2. 55:00 0.N 0N5.5.00.5 m-0
we 0.0 0.2 w.0m ..N ON. 0.; 55:00 0.. 59.2.0.5 N80
0.. 0.0 0.3 m. mam 0.0 0.0.q 55:00 0.. 00:50.5 2N0
00 0:.” Wm. 0.0 9mm Na 06m 55:00 0.. 0030020 2N0
. 0.0 0.0 0.0 0.00. 0.0 0.0 0005.0 0.. 00350.5 EN0
5 0.0 .... 3 .... o... 0N0 003.0 0.. 50385, 2N0
v 0.0 0.0 0.0 0.3 0.0 0.2. owoﬁo 0.. .0EEm. N.N0
_ 0.0 0.0 0.0 0.0 0.0 0.00. 55:00 0.. .0EEm. 8N0
0.. 0.. 0.x 5w. 0...‘ 0.0 0.~m 55:00 0.. .0550 m0~0
. 0.0 0.0 0.0 0.0 0.00. 0.0 55:00 0.. 5:55 008
. 0.0 0.00. 0.0 0.0 0.0 0.0 55:00 0.. =0wxon0zu m0~0
mm 0.0 0.... 0.0 ......» m... v.0m 55:00 0.. 5352.0 .0N0
: 5 b5... 2% 50054 .095:

mwmm .050. .0000 3.0.0: 503500 . .50 0.53 5.5 00000 0.00.0 5050‘. 00:35.0 ozm 0050.).

 

 

0.500.038 .0 .3 02.0.0080 50500.5 00.9.0 030 30. ”om 0.00...

73

 

 

030_ 4

 

$00— 7

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

_00 _0.0 __.00 _0.0. _0.00 _0.0 _0.0. 00000—0... 80F
_0 _0.0 _0.00 _0.0 _0.0 _0.0 _0.00 00820.0 :2.—
_0.. _0.0 _0.0 _0.00 0.0 _0.00 _0.00 0005—0... :9..— 0000—
—0_ _0.0 _0.0 _0.0 5.0 _0.0 _0.0 0005—0... .5..— 000,.—
_00 _0.0 _0.00 __.00 20 _0.0 0.0. 0005.0... .50— 0000—
_0 _0.0 _0.0 _0.0 _0.00 _0.0 0..... 0035.00 0005‘ 0000—
?. _0.0 _0.0. _0.0 00.00 _0.0 0.0.. emcee—_0.0 003.5. 000..—
0 .00 _0.00 _0.00 _0.0 _0.0. 0..: 8.23.000 0035‘ 0000
0 0.0.0 _0.0 __0.0 _0.0 .0 0.00. 8.23000 00050 0000—
00 _0.0 _0.0 _0.0 _0.00 _0.0 _0.0». 80:00:00 0.92.55— 500—
00 _0._ _0.0 _0._ .0 _0.0 _0.00 80:00.00 003.351 0000—
. _0.0 _0.0 _0.0 _0.0 5.0 _0.00_ 02200—00 0925.0— 3.00—
.00 _0.0 _0.0. _0.0 __.00 _0.0 _0.0 80:00—00 005.3..0— 0000—
_0 40.0 .0 _0.00 _0.00 .0 _0.0 80:00—00 09,235— :00—

 

Table 21: 1998 Young and old oviposition preference by P. canadensis

White oung otal Eggs
'te ash

28. 71
17. 82.
53 46.
9. 90.
8. 91.
45. 33.
38. 61
40. 60.
46. 48.
34. 59.
33. 66.
66. 33.
81.
56. 33.
40. 60.

 

5.
l.
l.
1.
5.
4.
6.
4.
4.
1.
7.
7.
7.
2.
2.

75

 

Table 22: 1997 P. canadensis oviposition preference for white ash collected from four sites

Ilston anderbilt otal Eggs

32.1 39.
0. 0.
36. 7.
26. 31.
12. 33.
23. 15.
40.
ll 35.
26.
35.
0.

11
30.

999““99990‘5‘99

 

Table 23: 1997 P. canadensis oviposition preference for extracts of white ash collected

from four sites

anderbilt

33.
22.
4.
O.
O.
O.
26.

 

1.
l.
2.
2.
2.
4.
4.
4.
2.

76

Table 24: 1996 P. canadensis pupal weight data

Date ost Plant

other pal weight

106
106
106
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
108
108
108
108
108
108
108
109
109
109
110
110
110
110
112

7 Jul
7 Jul
7 Jul
15 Jul
20 Jul
21 Jul
21 Jul
2 Jul
19 Jul
19 Jul
21 Jul
2 Jul
23 Jul
23 Jul
30 Jul
30 Jul
04-Au
26 Jul
29 Jul
30 Jul
19 Jul
23 Jul
25 Jul
27 Jul
19 Jul
20 Jul
21 Jul
10 Jul
17 Jul
19 Jul
23 Jul
26 Jul
27 Jul
31 Jul
30 Jul

1996
1996
1996
1996
1996
1996
1996
1996
1996
1996
1996
1996
1996
1996
1996
1996
-96
1996
1996
1996
1996
1996
1996
1996
1996
1996
1996
1996
1996

1996 .

1996
1996
1996
1996
1996

lack che
lack che
lack che
lack che
lack che
lack che
lack che
lack c
'te ash
hite ash
'te ash
'te ash
'te ash
’te ash
'te ash

'te ash

as

lack che
lack
lack c
lack c
'te ash
'te ash
'te ash

)

77

O.

 

 

1996 lack che
1996 Black c
1996 Black che
1996 Black che
1996 ite ash
1996 'te ash
—96 asswood
1996 akin as
1996 akin as
1996 'n as
1996 akin as
1996 'n as
1996 'n as
1996 lack che
1996 lack che
1996 lack c
1996 lack che
1996 lack c
1996 lack c
1996 lack c
1996 lack c
1996 lack c
1996 lack c
1996 lack c
1996 lack c
1996 lack c
1996 lack c
1996 'te ash
1996 'te ash
1996 °te ash
1996 'te ash
1996 'te ash
1996 'te ash
1996 ' as
1996 ° as
1996 ' as
1996 ' as
1996 ' as
1996 '

 

78

 

lack c
lack che
lack che
lack che
lack c
lack che
lack che
lack c
lack c
lack che
lack che
lack che
lack che
lack che
lack c
lack c
lack c
lack c
lack che
lack c
lack che
lack che
lack c
lack c
lack c
lack c
lack c
lack c
lack c
lack c
lack c
lack c
lack c
lack c
lack c
lack c
lack c
lack che
lack c

 

79

 

'te ash

'te ash
hite ash
hite ash

'te ash

as
'n as n
'n as n
lack c
lack c
lack che
lack che
lack che
lack che
lack c
lack che
lack c
lack c
lack c
'te ash
'te ash
°te ash
'te ash
'te ash
°te ash
'te ash
'te ash
' as

lack c
lack c
lack c
lack che
lack c
lack c
lack
lack c

 

80

Jul 1996 lack c
1 Jul 1996 lack c
1 Jul 1996 lack c
1 Jul 1996 lack c
Jul 1996 lack c
4 Jul 1996 lack c
6 Jul 1996 lack che
Jul 1996 lack c
18 Jul 1996 'te ash
19 Jul 1996 'te ash
0 Jul 1996 ’te ash
Jul 1996 '
18 Jul 1996 lack c
18 Jul 1996 lack c
18 Jul 1996 lack c
19 Jul 1996 lack c
19 Jul 1996 lack
19 Jul 1996 lack c
Jul 1996 lack c
Jul 1996 lack c
1 Jul 1996 lack
7 Jul 1996 lack
Jul 1996 lack c
Jul 1996 'te ash
0 Jul 1996 'te ash
1 Jul 1996 'te ash
1 Jul 1996 'te ash
1 Jul 1996 'te ash
1 Jul 1996 'te ash
1 Jul 1996 ’te ash
Jul 1996 'te ash
Jul 1996 'te ash
19 Jul 1996 '
Jul 1996 ' as
4 Jul 1996 '
Jul 1996
Jul 1996
Jul 1996
19 Jul 1996

 

81

 

 

82

lack c
lack c
lack c
lack c
lack c
lack c
lack c
lack che
lack c
lack c
lack c
lack c
lack 0
lack c
lack c

lack c
lack
'te ash
'te ash

 

83

 

Table 25: 1997 P. canadensis pupal weight data

ost Plant
urvived urvival

lack 30 1 330033003
birch

'te ash 7.29927007

wood

' 13.475 177

other #: reared on: Until weight:
'on:

us serotina O.

' us americana 31 0.635
us tremuloides 2 0.601

us serotina 0.91

us serotina 0.933

us serotina 0.

us serotina 0.93
us tremuloides 0.61

us serotina 0.

us serotina 0.5901

us serotina 0.55

us serotina 0.72

us serotina 0

us serotina 0.55
us tremuloides 0.
us tremuloides 0.598
us serotina 0.84
us serotina 0.64
us serotina 0.84

' us americana O.
' us americana 0.5
' us americana
raxinus americana
' us americana
serotina

 

84

inus americana

lus tremuloides

lus tremuloides

us serotina

us serotina

us serotina

'nus americana emale
lus tremuloides emale

nus serotina emale

us serotina
us serotina r
us serotina

inus americana

‘7— —‘ __ ‘o—l

us serotina
us serotina

 

' us americana
' us americana
' us americana
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
raxinus americana
us serotina
us serotina
us serotina
us serotina
americana

 

85

raxinus americana
raxinus americana
raxinus americana
raxinus americana
raxinus americana
lus tremuloides
lus tremuloides
us serotina
nus serotina
us serotina
lus tremuloides
raxinus americana
lus tremuloides
us serotina
us serotina
us serotina
us serotina
us serotina
‘ us americana
raxinus americana
us tremuloides
us serotina
us serotina
us serotina
us serotina
raxinus americana

raxinus americana

us tremuloides
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
raxinus americana
us tremuloides
us tremuloides
us tremuloides

86

0.658

0.75

0.6
0.57

emale
e

emale
e

0.654 female

0.62
0.633
0.629
0.67

0.781
0.517
0.765
0.623
0.642
0.507
0.631
0.727
0.777
0.8041
0.76
0.61
0.8
0.808
0.71
0.754
0.
0.54

0.
0.5431
0.5991
0.61

0.61
0.727
0.5
0.593
0.706
0.5521
0.75
0.83

emale
emale

emale
emale
emale
emale
emale

emale
emale
emale
emale
emale
emale

emale

 

 

us tremuloides
us serotina
us serotina
us serotina
us serotina
us serotina
raxinus americana
lus tremuloides
us tremuloides
us serotina
us serotina
us serotina
us serotina
us serotina

 

us serotina
us tremuloides
raxinus americana
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
raxinus americana

raxinus americana
serotina

 

87

us serotina
us serotina
us serotina
us serotina
us serotina
us serotina
raxinus americana
us serotina
us serotina
us serotina

us serotina

88

emale
emale
emale

emale
emale

emale
emale

 

‘N-—x--—a _I-I :1

 

BIBLIOGRAPHY

Tr" 9".rﬂc. ‘.

 

BIBLIOGRAPHY

Ae, S. A. 1995. Ecological and evolutionary aspects of hybridization in some Papilio
butterflies. pp. 229-235. In Swallowtail butterﬂies: their ecology and evolutionary
biology. (J. M. Scriber, Y. Tsubaki, and RC. Lederhouse eds.). Scientiﬁc Publishers,

Gainesville, FL.

Ayres, M. P., J.M. Scriber. 1994. Local adaptation to regional climates in Papilio

canadensis (Lepidoptera: Papilionidae). Ecological Monographs 64(4): 465-482.

Barbosa, P. 1988. Some thoughts on “The evolution of host range”. Ecology 69(4):

912-915.

Bemays, E., M. Graham 1988. On the evolution of host speciﬁcity in phytophagous

arthropods. Ecology 69(4): 886-892.

Brower, LP. 1959. Speciation in butterﬂies of the Papilio glaucus group. H. Ecological

relationships and interspeciﬁc sexual behavior. Evolution 13: 212-228.

89

Bultman, T.L., S.H. Faeth. 1987. Impact of irrigation and experimental drought stress on

leaf-mining insects of Emory oak. OIKOS 48 (5-10): 5-10.

Cockrell, F.J., S.B. Malcolm L.P. Brower. 1994. Time, temperature, and latitudinal
constraints on the annual recolonization of eastern North America by the monarch
butterﬂy. pp 233-251. In Biology and Conservation of the Monarch Butterﬂy (Malcolm,

SB. and MP. Zalucki eds) Natural History Museum, Los Angeles County.

Crow, TR. 1990. Tilia americana. pp. 784-791. In Silvics of North America volume
2. Ed. Burns, Russell M. and Honkala, Barbara H. USDA Forest Service, Washington

DC.

Darwin, C. 1859. The origin of species. pp. 54-55. Washington Square Press, New

York.

Derridj, 8., BR. Wu, L. Stammitti, J.P. Garrec, A. Derrien. 1996. Chemicals on the leaf
surface, information about the plant available to insects. Entomologia Experimentalis et

Applicata 80: 197-201.

Eichenlaub, V.L., J.R. Harman, F.V. Numberger, H.J. Stolle. 1990. The climatic atlas of

Michigan. University of Notre Dame Press, Notre Dame, Ind.

90

Feeny, P. 1976. Plant apparency and chemical defense. pp. 1-40. In Recent Advances

in Phytochemistry, Volume 10. (R.L. Mansell, ed.). Plenum Press, New York.

Feeny, P. 1995. Ecological opportunism and chemical constraints on the host
associations of swallowtail butterﬂies. pp. 9-16. In Swallowtail butterﬂies: their
ecology and evolutionary biology. (J. M. Scriber, Y. Tsubaki, and RC. Lederhouse eds.).

Scientiﬁc Publishers, Gainesville, FL.

Gage, Dr. Stuart. Personal Communication. 4 May 1998.

Grossmueller, D.W., R.C. Lederhouse. 1985. Oviposition site selection: an aid to rapid
growth and development in the tiger swallowtail butterﬂy, Papilio glaucus. Oecologia

66: 68-73.

Hagen, R.H., R.C. Lederhouse. 1985. Polymodal emergence of the tiger swallowtail,
Papilio glaucus (Lepidoptera: Papilionidae): Source of a false second generation in New

York state. Ecological Entomology 10: 19-28.

Hagen, R.H., R.C. Lederhouse, J .L. Bossart, J .M. Scriber. 1991. Papilio canadensis and

P. glaucus are distinct species. Journal of the Lepidoptera Society 45: 245-258.

Hairston, N.G., F.E. Smith, L.B. Slobodkin. 1960. Community structure, population

control, and competition. The American Naturalist 44(879): 421-425.

91

Hunter, A.F., M.J. Lechowicz. 1992. Foliage quality changes during canopy

development of some northern hardwood trees. Oecologia 89: 316-323.

Hunter, M. D. 1992. A variable insect-plant interaction: the relationship between tree
budburst phenology and population levels of insect herbivores among trees. Ecological

Entomology 16: 91-95.

Hunter, M. D., G. C. Varley, G.R. Gradwell. 1997. Estimating the relative roles of top-
down and bottom-up forces on insect herbivore populations: A classic study revisited.

Proceeding of the National Academy of Science USA 94: 9176-9181.

J anzen, D. H. 1988. On the broadening of insect-plant research. Ecology 69(4): 905.

Johnson, K. S., J .M Scriber 1994. Geographic variation in plant allelochemicals of

signiﬁcance to insect herbivores. pp. 7-31. In Functional dynamics of phytophagous

insects. Ed. Ananthakrishnan, T.N. Oxford and [EH Publishing Co. Pvt. Ltd., New

Delhi.

Kukal, O., M. P. Ayres, J. M. Scriber 1991. Cold tolerance of the pupae in relation to the

distribution of swallowtail butterﬂies. Canadian Journal of Zoology 69: 3028-3037.

92

Lederhouse, R.C., J .M. Scriber. 1987. Ecological signiﬁcance of a postmating decline in

egg viability in the tiger swallowtail. Journal of the Lepidopterists’ Society 41(2): 83-93.

Lederhouse, R.C., M.P. Ayres, and J.M. Scriber. 1990. Adult nutrition affects male

virility in Papilio glaucus. Functional Ecology 4: 743-751.

Lederhouse, R.C., M.P. Ayres, and J .M. Scriber. 1995. Physiological and behavioral
adaptations to variable thermal environments in North American swallowtail butterﬂies.
In Swallowtail butterﬂies: their ecology and evolutionary biology. (J. M. Scriber, Y.

Tsubaki, and RC. Lederhouse eds.). Scientiﬁc Publishers, Gainesville, FL.

Marquis, DA. 1990. Prunus serotina. pp. 594—604. In Silvics of North America
volume 2. Ed. Burns, Russell M. and Honkala, Barbara H. USDA Forest Service,

Washington DC.

Matsuki,, M., S. F. MacLean. 1994. Effects of different leaf traits on growth rates of

insect herbivores on willows. Oecologia 100: 141-152.

Mattson, W.J. 1980 Herbivory in relation to plant nitrogen content. Annual Review of

Ecological Systematics 11: 119-161.

93

Mattson, W.J., J. M. Scriber. 1987. Nutritional ecology of insect folivores of woody
plants: Nitrogen, water, ﬁber, and mineral considerations. pp. 105-146. In Nutritional
ecology of insects, mites, and spiders. Ed. Slansky, F., Rodriguez, J .G. John Wiley and

Sons, Inc., New York.

Microsoft Corporation. 1994. Microsoft Excel. Microsoft Corporation, USA.

N ishida, R. 1995. Oviposition stimulants of swallowtail butterﬂies. pp. 17-26. In
Swallowtail butterﬂies: their ecology and evolutionary biology. (J. M. Scriber, Y.

Tsubaki, and RC. Lederhouse eds.). Scientiﬁc Publishers, Gainesville, FL.

Papaj, DR. 1986. Interpopulation differences in host pereference and the evolution of

learning in the butterﬂy, Battus philenor. Evolution 40(3): 518-530.

Pedigo, L. P., MD. Zeiss. 1996. Analyses in Insect Ecology and Management. Iowa

State University Press, Ames.

Perala, DA. 1990. Populus tremuloides. pp. 555-569. In Silvics of North America
volume 2. Ed. Burns, Russell M. and Honkala, Barbara H. USDA Forest Service,

Washington DC.

Rausher, MD. 1980. Host abundance, juvenile survival, and oviposition preference in

Battus philenor. Evolution 34(2): 342-355.

94

Renwick, J .A., F.S. Chew. 1994. Oviposition behavior in Lepidoptera. Annual Review

of Entomology 39: 377-400.

Safford, L.O., J .C. Bjorkbom, J. C. Zasada. 1990. Betula papyrifera. pp. 158-171. In
Silvics of North America volume 2. Ed. Burns, Russell M. and Honkala, Barbara H.

USDA Forest Service, Washington DC.

SAS Institute, Inc. 1989. SAS/STAT Users Guide, Version 6, Fourth Edition, Cary, NC:

SAS Institute Inc.

Schlesinger, RC. 1990. Fraxinus americana. pp. 333-338. In Silvics of North
America volume 2. Ed. Burns, Russell M. and Honkala, Barbara H. USDA Forest

Service, Washington DC.

Schultz, J. C. 1988. Many factors inﬂuence the evolution of herbivore diets, but plant

chemistry is central. Ecology 69(4): 896-897.
Scriber, J. M. 1977. Limiting effects of low leaf-water content on the nitrogen

utilization, energy budget, and larval growth of Hyalophora cecropia (Lepidoptera:

Satumiidae). Oecologia 28: 269-287.

95

Scriber,J.M. 1984a Host-Plant Suitability. InChemical Ecology of Insects. Ed. Bell,

W.J, Carde, R.T. Sinauer Associates, Inc.

Scriber,J.M. 1984b. Nitrogen nutritiOn of plants and insect invasion. pp. 441-460. In

Nitrogen in Crop Production. (R.D. Hauck eds.). ASA-CSSA-SSSA, Madison,

Wisconsin.

Scriber, J. M. 1993. Absence of behavioral induction in oviposition preference of

Papilio glaucus (Lepidoptera: Papilionidae). Great Lakes Entomologist 26(2): 81-95.
Scriber, J. M. 1995. Overview of Swallowtail butterﬂies: taxonomic and distributional
latitude. In Swallowtail butterﬂies: their ecology and evolutionary biology. (J. M.

Scriber, Y. Tsubaki, and RC. Lederhouse eds.). Scientiﬁc Publishers, Gainesville, FL.

Scriber, J. M. 1996a. A new ‘cold pocket’ hypothesis to explain local host preference

shifts in Papilio canadensis. Entomologia Experimentalis et Applicata 80: 315-319.

Scriber, J. M. 1996b. Tiger tales: natural history of native North American swallowtails.

American Entomologist: 19-32.

96

Scriber, J. M., and Gage, SH. 1995. Pollution and global climate change: Plant
ecotones, butterﬂy hybrid zones and changes in biodiversity. pp. 319-344. In
Swallowtail butterﬂies: their ecology and evolutionary biology. (J. M. Scriber, Y.

Tsubaki, and RC. Lederhouse eds.). Scientiﬁc Publishers, Gainesville, FL.

Scriber, J .M., R.C. Lederhouse. 1992. The thermal environment as a resource dictating
geographic patterns of feeding specialization of insect herbivores. pp. 429-466. In
Effects of Resource Distribution on Animal-Pant Interactions. (Hunter, M.D., T.

Ohgushi, and P.W. Price eds.). Academic Press, Inc., San Diego. ‘

Scriber, J. M., S.F. Slansky. 1981. The nutritional ecology of immature insects. Annual

Review of Entomology 26: 183-211.

Smiley, J. 1978. Plant chemistry and the evolution of host speciﬁcity: New evidence

from Heliconius and Passiﬂora. Science 201: 745-747.

Thomas, AT, 1. D. Hodkinson. 1991. Nitrogen, water stress and the feeding efﬁciency

of Lepidopteran herbivores. Journal of Applied Ecology 28: 703-720.
Thompson, J .N . 1988. Evolutionary ecology of the relationship between oviposition

preference and performance of offspring in phyophagous insects. Entomologia

Experimentalis et Applicata 47: 3- 14.

97

Thompson, J .N . 1994. The coevolutionary process. University of Chicago Press,

Chicago.

Thompson, J. N. 1995. The origins of host shifts in swallowtail butterﬂies versus other
insects. pp. 195-204. In Swallowtail butterﬂies: their ecology and evolutionary biology.
(J. M. Scriber, Y. Tsubaki, and RC. Lederhouse eds.). Scientiﬁc Publishers, Gainesville,

FL.

Thompson, J. N ., O. Pellmyr. 1991. Evolution of oviposition behavior and host

preference in Lepidoptera. Annual Review of Entomology 36: 65-89.

Voss, E. 1985. Michigan Flora Part II: Dicots. Regents of the University of Michigan,

Ann Arbor.

Voss, E. 1996. Michigan Flora Part III: Dicots Concluded. Regents of the University of

Michigan, Ann Arbor.

Watanabe, M. 1995. Population dynamics of Papilio xuthus larvae in relation to th life
history of the host tree. pp. 101-106. In Swallowtail butterﬂies: their ecology and
evolutionary biology. (J. M. Scriber, Y. Tsubaki, and RC. Lederhouse eds.). Scientiﬁc

Publishers, Gainesville, FL.

98

Wiklund, C. 1984. Egg-laying patterns in butterﬂies in relation to their phenology and

the visual apparency and abundance of their host plants. Oecologia 63: 23-29.

99

 

"llllllllllllllllllf