Plumage Ornamentation As an Indicator of Female Age and an Influence in Male Mate Choice in Protonotaria Citrea, the Prothonotary Warbler in Virginia

Virginia Commonwealth University VCU Scholars Compass

Theses and Dissertations Graduate School

2008

Plumage Ornamentation as an Indicator of Female Age and an Influence in Male Mate Choice in Protonotaria Citrea, the Prothonotary Warbler in Virginia.

Terry Smith Virginia Commonwealth University

Follow this and additional works at: https://scholarscompass.vcu.edu/etd

Part of the Biology Commons

Downloaded from https://scholarscompass.vcu.edu/etd/1559

This Thesis is brought to you for free and open access by the Graduate School at VCU Scholars Compass. It has been accepted for inclusion in Theses and Dissertations by an authorized administrator of VCU Scholars Compass. For more information, please contact [email protected].

School of Humanities and Sciences, Department of Biology Virginia Commonwealth University

This is to certify that the thesis prepared by Terry Lee Smith entitled PLUMAGE ORNAMENTATION AS AN INDICATOR OF FEMALE AGE AND AN INFLUENCE IN MALE MATE CHOICE IN PROTONOTARIA CITREA, THE PROTHONOTARY WARBLER, IN VIRGINIA has been approved by her committee as satisfactory completion of the thesis or dissertation requirement for the degree of Master of Science.

Leonard A. Smock, Professor and Chairman, Department of Biology

Robert J. Reilly, Professor, Center for Environmental Studies

Donald R. Young, Professor, Department of Biology

Michael L. Fine, Professor, Department of Biology

Dr. Robert Holsworth, Dean of the College of Humanities and Sciences

Dr. F. Douglas Boudinot, Dean of the School of Graduate Studies

Signed June 18, 2008

PLUMAGE ORNAMENTATION AS AN INDICATOR OF FEMALE AGE AND AN

INFLUENCE IN MALE MATE CHOICE IN PROTONOTARIA CITREA, THE

PROTHONOTARY WARBLER IN VIRGINIA.

A Thesis submitted in partial fulfillment of the requirements for the degree of Master of Science at Virginia Commonwealth University.

TERRY LEE SMITH Bachelor of Science, Christopher Newport University, 1995

Director: DR. LEONARD A. SMOCK, CHAIRMAN, DEPARTMENT OF BIOLOGY AND DR. ROBERT J. REILLY, PROFESSOR, CENTER FOR ENVIRONMENTAL STUDIES

Virginia Commonwealth University Richmond, Virginia June, 2008

Acknowledgements

This project could not possibly have been completed without the assistance and support of my advisors Dr. Robert Reilly, Dr. Leonard Smock, Dr. Donald Young and Dr. Michael Fine.

Dr. Charles Blem gave me my start on this project; his enthusiasm for avian research is infectious.

A number of people contributed field assistance, additional advisement, storage or equipment: Charles Blem and Robert J. Reilly provided USFWS bands in addition to birding and banding equipment, and training in handling and capturing birds. Mark Battista and the Dutch

Gap Conservation Area granted access to the gated study site and equipment storage as well as the use of canoes. Thanks also to Jill Reid and Emily Watkinson for advice and encouragement.

I received generous field work assistance from Heidi Clayton, Jenny Tarasova, Sara Felvey,

Nichole Winning, Ashley Tiedeman, Casey Seelig, Tommy Robinson, Steven Brantley, Donna

Kennedy, Rene Cabaniss, Maggie Butcher, Drew Garey, Jaime Fuest, and Crystal Nance. Julie

Nauman, Dr. Darcy May provided advice regarding statistics. Dr. Edward Crawford took on some extra grading work for the field assistants who wrote Independent Study papers based on their experiences on 'Team Warbler'. Dr. Thomas M. Reeves provided invaluable advice on statistics, graphics and content organization, and Dr. Linda Phillips graciously allowed me to sit in on her grantsmanship writing class.

Last but not least I must thank my parents Heidi and Robert Clayton for their support during my project.

iii Table of Contents Page

Acknowledgements...... ii

List of Tables ...... v

List of Figures...... vi

Abstract...... vii

Introduction and Objectives...... 1

Methods ...... 5

Statistical Methods ...... 7

Results ...... 8

Tail Spots and Age in Females ...... 8

Tail Spots and Age in Males...... 9

Correlation Matrix of Study Variables ...... 9

Age in Mated Pairs...... 9

Tail Spot Number in Mated Pairs ...... 10

Discussion …………………………………………………………………………….….11

Tail Spots and Age in Females ...... 12

Tail Spots and Age in Males...... 13

Age in Mated Pairs...... 14

Age and Weight ...... 15

Wing and Tail Length ...... 15

Weight and Tail Spots...... 15

Unexpected Correlations………...…………..………………………………...…16

Ornamentation and Sexual Selection.………….…………………...………….....16

iv Literature Cited………………………………………..……………………..………….22

Appendix A……………………………………………………………………………..27

Vita……………………………………………………………………………………...39

v List of Figures Page

Figure 1:Tail spot variation in Warbler species...... 28

Figure 2: Descriptive statistics for female birds...... 29

Figure 3: Descriptive statistics for male birds...... 30

Figure 4: Linear regression of Age on Number of Tail Spots in female birds ...... 31

Figure 5: Linear regression of Age on Number of Tail Spots in male birds...... 32

Figure 6: Linear regression of Female Age on Male Age within mated pairs...... 33

Figure 7: Linear regression of Number of Tail Spots within mated pairs...... 34

vi List of Tables Page

Table 1: Cross-tabulation of Age and Number of Tail Spots in female birds...... 35

Table 2: Cross-tabulation of Age and Number of Tail Spots, with the variable Age re-grouped into three ordinal categories...... 36

Table 3: Table of correlation coefficients between all pairs of study variables...... 37

Table 4: Cross-tabulation of Number of Tail Spots in males and in females of mated pairs...... 38

Abstract

PLUMAGE ORNAMENTATION AS AN INDICATOR OF FEMALE AGE AND AN

INFLUENCE IN MALE MATE SELECTION IN PROTONOTARIA CITREA, THE

PROTHONOTARY WARBLER, IN VIRGINIA.

Terry Smith, Bachelor of Science, Christopher Newport University, 1995

A Thesis submitted in partial fulfillment of the requirements for the degree of Master of Science at Virginia Commonwealth University.

Virginia Commonwealth University, 2008

Director: DR. LEONARD A. SMOCK,CHAIRMAN, DEPARTMENT OF BIOLOGY AND DR. ROBERT J. REILLY PROFESSOR, CENTER FOR ENVIRONMENTAL STUDIES

Flamboyant plumage and ornamentation is common and well-known in male birds; it serves as a sexual display to attract potential mates. While flamboyant plumage is less common and usually more subtle in female birds, it does occur in some species such as Dark-Eyed Juncos (Junco hyemalis) and Prothonotary Warblers (Protonotaria citrea). Prothonotary Warblers display relatively subtle sexual dimorphism. This study examines variations in tail spot patterns in Prothonotary Warblers and relates those variations to age in females. Females with fewer than six spots tend to be two years old or younger; females with six spots or more tend to be three years old or older. The tail spot numbers of mated pairs were also analyzed. Statistical analyses indicate that males mate with females with six tail spots more often than they mate with females with other numbers of tail spots. This suggests males prefer females who are at least three years old.

Introduction and Objectives

Tail spots are usually present in male Wood Warblers, a common moniker for the

group of singing, perching birds in the Order Passeriformes, Family Parulidae (Blake

1966, Burtt 1986). According to the otherwise comprehensive monograph by Edward

Burtt Jr. describing Wood Warbler plumage, 52 of 113 species of Wood Warblers have

tail spots. Burtt did not distinguish which species have tail spots in both genders versus

species in which only the males or only the females have tail spots (Burtt 1986).

Comparative illustrations showing a representation of Warbler species sometimes fail to

indicate the gender represented, though by convention male plumage is considered

representational of a species (Fig. 1). However, the existence of tail spot differences by

sex and sometimes by age is well-known. The Identification Guide to North American

Birds: Part 1 by Peter Pyle is the Federal Bird Banding Laboratory's officially accepted

guide to ageing and sexing North American passerines (1997). According to Pyle and

others, some species can be aged and sexed in part by those tail spots (Burtt 1986, Pyle

1997, Schorre 1998). The Prothonotary Warbler (Protonotaria citrea) and the American

Redstart (Setophaga ruticilla,) have tail spots in both genders, though the variation is in number of spots in Prothonotary Warblers while the variation between male and female

Redstarts is in the color of their tail spots (Burtt 1986, Schorre 1998). In Black-throated

Blue Warblers (Dendroica caerulescens) the males alone possess tail spots on the outer

two retrices while the females remain unspotted (Blake 1966, Burtt 1986). Neither the

2 Kentucky Warbler (Oporornis formosus), nor the Ovenbird (Seiurus aurocapillus), have any tail spots at all in either sex (Burtt 1986, Schorre 1998)

Communication via tail spot flashing is a well documented phenomenon in birds

(Wible 1967, Bruce 1972, Mumme 2002 ). For example, the Slate-Throated Warbler

(Myioborus miniatus) uses its tail spots to flush prey out from under foliage and then catches its prey in flight (Mumme 2002). Wood Warblers will flash their tail spots during grooming activities, agonistic interactions, the take off and landing portions of flight and mating displays (Baird 1967, Ficken et al. 1968, Burtt 1986) .

Because plumage ornaments are used in mating displays it is generally accepted in the ornithological community that ornaments influence mating success (Darwin 1871,

Kristin et al. 2007). Mating success influences evolutionary fitness (Darwin 1871,

Kristin et al. 2007). This study examines which tail spot patterns in Prothonotary

Warblers are phenotypic variations that influence social interactions that lead to pairing and ultimately mating success.

While some Prothonotary females have tails nearly identical to males in pattern and color intensity, most will have solid dark grey tails with any of a number of spotting variations that include edging as well as spotting (Fig. 1). Some females have retrice feathers with a black base and ten spots identical in number and color intensity to male

Prothonotary Warblers. Female tail feathers with spots are sometimes adjacent to feathers with only white edging. For the purposes of this study 'spots' were distinguished from 'edges' by size and whether or not the spot was surrounded by black on all sides.

Tail spot patterns consistently start on the outer retrices and work their way inward. Any edges I observed were always the innermost of the tail markings with spots on the

3 outermost retrices (Fig. 1). The innermost pair of retrice feathers are always unspotted in both genders.

Male Prothonotary Warblers have black retrices, or tail feathers, with one white spot or patch on each of their outer retrices. Commonly, out of twelve retrice feathers, only the central pair will be solid black while the other ten will have a large white spot

(Fig. 1). While that is the most common tail spotting pattern in males, this study reveals individual variations occur and a male may have, for example, three or four spots per side.

The majority of data taken on Prothonotary Warblers over the course of the long term research project at VCU has been from females and nestlings (Blem and Blem

1992,1993 and Blem et al.1999, Podlesak et al. 2001) . This is primarily because males do not incubate the eggs so they do not enter their nest boxes and cannot be captured with the same methods used to capture females (Blem and Blem 1992,1993 and Blem et al.

1999, Podlesak et al. 2001). Males must be captured via mist-netting early in the breeding season and color banded. Later in the season it is possible to identify an individual male by color band combination when he comes to his nest box to feed hatchlings.

Aside from the differences in tail spot patterns, Prothonotary Warblers exhibit classic albeit subtle sexual dimorphism. Both males and females have blue-gray wings and yellow breasts, black feet and beaks, and white undertail coverts, which are feathers between the vent and the base of the tail. The males have a deeper yellow coloring on the head and nape, often darkening to an orange wash. The females have a yellow head matching their breast, with an olive colored wash on the crown and nape that connects

4 with their back. In females, the tail feathers are usually a medium to dark gray base-color and their spotting variation ranges from no spots at all to a full set of 10 spots: five spots on each side (Blake 1966, Burtt 1986, Blem and Blem 1992, Schorre 1998).

The purpose of this study is to document variations in plumage ornamentation, specifically tail spots, in Protonotaria citrea as a predictor of age.

Second year, or SY birds (one year old), are predicted to have fewer tail spots on average than older birds. A suite of other physical characteristics, including weight, retrice (tail feather) length and wing chord (the measurement of the wing length when folded against the body) were evaluated to determine if associations existed among those variables or with tail spot variation. This study documents these characteristics in mated pairs.

Additionally, nestlings were banded and weighed as part of data collection for the ongoing long term study on P. citrea at Virginia Commonwealth University.

Methods

Forested wetlands in the southeastern United States provide excellent breeding

habitat for Protonotaria citrea (Sawyer 1993, Zyla and Donovan 1993, Cartwright 1997,

1998, Petit 1999). This is one of two species of North American Wood Warbler that nest

in secondary tree cavities, the other being Lucy's Warbler (Vermivora luciae) (Blem and

Blem 1992, 1993). Prothonotary Warblers readily accept nest boxes as an alternative to

natural cavities (Sawyer 1993, Zyla and Donovan 1993, Cartwright 1997 and 1998, Petit

1999). Nest boxes (dimensions: 28L X 9W X 6D cm, entrance hole 3.8 cm in diameter)

were placed in the Dutch Gap Conservation Area in Chesterfield, Virginia beginning in

2002. Males and females were captured in the spring and early summer of 2004 and

2005. Additional birds used in this analysis were captured in 2006 in a related study

supervised by Dr. Robert J. Reilly. There are currently approximately 160 nest boxes distributed along the shorelines of the main waterway and its islands. Boxes are situated

atop metal poles to prevent predation. Boxes are 0 to 6.0 m from the river bank, and 0.5 to 1.5 m above mean high tide.

To determine which nest boxes were occupied, boxes were approached by canoe or kayak. This was done at regular intervals throughout the field work season to keep tabs on which box was in what stage of its nesting cycle. Females were caught while they were incubating their eggs. Small hand held nets were placed over the nest box entrance holes after approaching the box in a canoe or when warranted, hip waders.

When startled females flew into the nets they were banded, weighed, measured in length,

aged by plumage condition and coloration and their tail spots were counted. Males

6 do not incubate eggs, and therefore can not be captured using this method, which is why

they were caught via mist nets earlier in the season. All birds were weighed with an

electronic balance. Birds were aged by a suit of plumage characteristics or previous leg

bands if present. Plumage characteristics included the condition of the primaries, wing

coverts and tail feathers, and the peculiarities of the base color in those feathers. A

brownish-gray base color and worn feathers indicates an SY (second year) bird retaining

the feathers he or she developed as a fledgling. A blue-gray base color paired with

healthier, comparatively unworn feathers indicates an ASY (after second year) bird. It

can be inferred that the bird is at least two years old and possibly older. Birds banded as

nestlings in earlier years of the long-term portion of the study could be aged exactly and

were identified accordingly as anywhere from three to six years old. Any un-banded birds

were banded. Bird bands were provided by the United States Fish and Wildlife Service

(USFWS).

Female Prothonotary Warblers have between zero and five spots on each side of their tails (Fig. 1). It was decided to call these thinner, white areas 'edges'. Each time a female was captured the number of full spots and edges was recorded. For example, a

female with three spots and two edges on the right side of her tail was labeled as R 3/2.

If on the left side of her tail she had three spots and one edge that was labeled L 3/1 and

so on. These data were collected for both the right and the left sides of the tails for the

males and the females in the 2005 and 2006 field work. Because potential asymmetry

was not initially taken into consideration when collecting the 2004 data, all data used in

these analyses are from the right side of the tails. Because of this distinction birds are

referred to in this paper as 'three spotted male' or 'four spotted female', when in fact they

7 would likely have had a total number of six or eight tail spots, respectively. Only spots were used in the statistical analysis as it is believed the edges may be hidden in displays and therefore inconsequential to interactions among this species (Burtt 1986).

Statistical Methods. Statistical analyses were conducted in SPSS v11.5.

Ordinary least-squares linear regression was used to evaluate the relationship between pairs of variables. Variables were selected for regression analysis which addressed structural changes over time (e.g., age-related changes in tail spot number), or were theoretically relevant to mate selection (e.g., tail spot number in mated pairs) (Field

2000). Some datasets used for regression analysis were also examined using cross- classification analysis. In one application of this approach, the interval variable Age was recoded to an ordinal scale of measurement, and a statistic appropriate for ordinal, non- parametric correlation (Kendal tau-b) was applied to the resulting cross-tabulation (Field

2000). A matrix of Pearson Product-Moment Correlation Coefficients (r) was computed for all pairings of variables in the study, to facilitate the identification of significant associations within sexes and between mated pairs of birds (Hanke and Reitsch 1986,

Field 2000).

Results

Data were acquired from a total of 118 Protonotaria citrea (59 mated pairs). A summary of descriptive statistics for all study variables is shown for females (Fig. 2) and for males (Fig. 3).

Tail Spots and Age in Females. A linear regression analysis was used to

examine whether tail spot number is a significant predictor of age in female Protonotaria

(Fig. 4). The resulting regression equation [Age = 2.01 + (0.158 * Tail Spots)] was not

statistically significant (p=0.26; R2 = 0.023). The presence of significant non-normality

of the error residuals in the regression model (Kolmogorov-Smirnov statistic = 0.21;

p<0.001) suggested that the initial linear regression approach was not optimized for these

data, or that the fundamental relationship between age and tail spots may be nonlinear

(Field 2000). Therefore, the relationship of age to tail spots in females was further

examined using measures of association in two-way cross-tabulations (Table 1). Chi-

2 2 square (χ ) and Kendall tau (τb) statistics did not indicate significant differences: χ (20) =

21.53, (p = 0.366); τb= 0.188, (p = 0.089). Because a large number of cells had a low

number of expected observations, (26 cells with expected < 5), the variable Age was

transformed to an ordinal classification of three groups: birds 1, 2 and 3 or more years

old (Table 2). When female birds were grouped into the three ordinal categories, Number

2 of Tail Spots was significantly related to Age for both χ 8 = 16.51, (p<0.05) and

Kendall’s τb = 0.244, (p<0.05).

Tail Spots and Age in Males. Linear regression of Age on Number of Tail Spots,

in male birds, did not lead to a significant regression model: Age = 2.05 + (0.081 *

Tailspots). This regression model was not significant

(R2 = 0.009), (p=0.51) (Fig. 5). Also, analysis of cross-tabulations of these variables,

repeating the ordinal age grouping of Table 2, did not demonstrate significant

2 associations [χ (6) = 8.20, (p=0.22); τb =0.15, (p=0.24)].

Correlation Matrix of Study Variables. This study provided data on five

variables for each sex, corresponding to 45 unique pairwise combinations of variables

which may be considered for correlation. Some of these pairings are between similar

variables of the mated pairs (for example, tail spot number in the male and the female of

each mated pair), and some pairings are different variables within the same sex which

may have physiological significance (for example, age and weight). Table 3 shows the

Pearson Product-Moment Correlation Coefficient (r) and the significance level of the

correlations. Eight pairings were statistically significant and are discussed later, with

regression analyses conducted on selected pairings.

Age in Mated Pairs. The age of males significantly correlates with that of their

mates (r=0.40; p<0.01). The paired age data also generated a significant linear regression

model (F(1,56) = 10.48; p <0.01) (Figure 6). The least-squares equation line was: Female

Age = 1.02 + (0.662 * Male Age), R2 = 0.158.

10 Tail Spot Number in Mated Pairs. Tail Spots in mated pairs were significantly correlated (r = -0.35; p<0.01) (Table 3). Figure 7 shows the linear regression analysis for

Tail Spot Number in mated male and female pairs. A significant (p<0.01) regression equation was obtained: Female Tail Spots = 6.10 – (0.514 * Male Tail Spots). Table 4 also provides correlation of tail spot numbers between birds of mated pairs. The cross- tabulation of the tail spot numbers makes clear the dominant pattern of 5-spot males paired with 3-spot females. However, the table also shows exceptions to this pattern. For example, 22/23 (96%) of the 3-spot females paired with 5-spot males, but only 10/17

(59%) of the 5-spot females were mated with the 5-spot males. In addition, 4/5 (80%) of

3-spot males were paired with 5-spot females which also runs counter to the dominant patterns seen in the 5-spot males.

Referring again to the matrix of correlations among all study variables (Table 3), a significant negative relationship between Age and Weight (r = -0.41; p<0.01) in female birds contrasted with a lack of a significant correlation of these variables in males (r =

0.16; p=0.28). The strongest correlations observed in the study pertained to the wing and tail lengths within individual male (r = 0.68; p < 0.001) and female (r = 0.67; p< 0.001) birds.

Discussion

The existence of tail spot differences by sex and sometimes by age is well known.

The Identification Guide to North American Birds: Part 1 by Peter Pyle (1997) is the federal Bird Banding Laboratory’s officially accepted guide to ageing and sexing North

American passerines. It provides drawings for several species of warblers in which tail spot patterns are shown to differ by sex and sometimes by age. In his overview section on ageing and sexing the Dendroica warblers, his illustrations, similar to Figures 1A, B and C in this document, show clearly that the number and size of white tail spots increases with age and from female to male.

Direct measurement of selective pressures is possible where the characteristic in question varies within the population (Burtt 1986). Age is one common parameter of fitness communicated by plumage variation (Jeffrey et al. 1993, Krištín et al. 2007).

Observations indicate Prothonotary Warblers do not lose spots with age. The phenomenon of females becoming more 'Male-like' has been recorded in the Hooded

Warbler (Wilsonia citrina). Hooded Warbler females might become more Male-like and then at some point stop, but those individuals have not been observed to become less

Male-like (Morton 1989). Hooded Warblers reach a certain plumage ‘class’ and remain there, even through several molts (Morton 1989). If this is also the case for Prothonotary

Warblers, the number of tail spots in females either increases or remains the same from molt to molt, but does not decrease.

12 Tail Spots and Age in Females. A linear regression analysis was used to examine

whether tail spot number is a significant predictor of age in female Protonotaria (Fig. 4).

Although the regression analysis of Age and Tail Spots in female birds did not produce a

significant relationship, there were at least two reasons to further explore the dataset for

specific associations between these variables. First, theoretical reasons pertaining to age-

related plumage changes suggest that a regression analysis spanning all ages of birds is

not an optimal statistical approach. Specifically, age-dependent change in plumage is

likely to be most detectable when 1-year-old birds are compared to older birds. This is

because the first year molt in most passerines is incomplete and hence produces a duller

plumage than is characteristic of older birds (Pyle 1997). Secondly, visual inspection of

the data plots suggested some dependency of tail spot number on age; all female birds

with fewer than three spots were less than three years old. In addition, the presence of

significant non-normality of the error residuals in the regression model (Kolmogorov-

Smirnov statistic = 0.21; p<0.001) also suggested that the initial linear regression

approach was not optimized for these data, or that the fundamental relationship between

age and tail spots may be nonlinear (Field 2000).

The relationship of age to tail spots was further examined using measures of association in two-way cross-tabulations. An initial analysis, using the full dataset with no re-grouping, did not result in statistically significant differences (Table 1). However, a relatively large number of cells had a low number of expected observations, which presents interpretive problems for the χ2 test. To simplify the two-way classification, and

to reduce the number of cells with low expected values, the variable Age was

transformed to an ordinal classification of three groups (birds 1, 2 and 3 or more years

13 old). Constructing the cross-tabulation in this manner (Table 2) reduced the number of

cells with low numbers of expected observations, and avoided the low marginal totals

(n=2) for both the 5-year-old and the 6-year-old birds. Number of Tail Spots was significantly related to Age (when female birds were grouped into the three ordinal

2 categories), for both χ (p<0.05) and Kendall’s τβ (p<0.05). It is likely that this

simplified cross-tabulation increased the statistical power to detect differences between 1-

yr-old birds vs. older birds, consistent with the previously mentioned concept that

developmental plumage changes are most dramatic between these groups.

While the significant outcome observed in Table 2 indicated the presence of an association of age and tail spot number, the effect was modest, though it did suggest that tail spots numbers are indicative of age. Although the significant relationship observed between Age and Tail Spots, in the ordinal regrouping of female birds, is likely attributable to the 1-yr-old versus older birds distinction. However, it remains possible that differences between 2-yr-olds versus 3+-yr-olds also contributes, at least marginally, to the significant relationship between these variables. In addition, these results may reflect the importance of the first year molt in most passerines, which is incomplete and produces a duller plumage than is characteristic of older birds. Thus, there is no reason to

expect continuing change in plumage (and hence tail spots) from year-to-year in older

birds. Overall, the results of this study suggested an association between tail spot number

and age in female birds.

Tail Spots and Age in Males. Because males typically exhibit little variation from

the five spot (ten total) pattern, tail spot number and age in male P. citrea were expected

14 to be unrelated. The present analysis showed this relationship was not significant either

2 when using linear regression (p=0.506) or when using cross tabulation χ (p=0.224) or τb

(p=0.236) repeating the ordinal age grouping used for females. The analyses of these data reject the hypothesis that tail spots are a reliable indicator of age in male

Prothonotary Warblers.

Age in Mated Pairs. Also as predicted, age of males significantly correlated with

the age of their mates (p<0.01) failing to reject the hypothesis that older males mate with older females and younger birds would likewise mate other younger birds (see Table 3).

Current opinion among observers of Protonotaria citrea is that they are not monogamous, even within season. However, this significant correlation does suggest a selection pressure towards similar aged mates among this species. This finding should be balanced with the understanding that there were only four one-year-old males and four one-year-old females in the sample used in this study.

Future research could include tracking of tail spot numbers in individual females over time. Repeated recapture of wild females is possible only when individual

Prothonotary Warblers return to the same nesting area annually (breeding at Dutch Gap only in consecutive seasons versus breeding periodically at other Prothonotary Warbler study sites such as Deep Bottom Park in Henrico Virginia or Presquile National Wildlife

Refuge in Warsaw, Virginia, for example). Individuals with breeding success in a territory have been shown to exhibit a high rate of breeding site fidelity. Successfully

15 raising two broods within a season yields a > 80% return rate for both males and females

the following seasons. (Hoover 2003).

Age and Weight: I hypothesized that weight in Prothonotary Warblers would decrease with age. The mean age of the Prothonotary Warblers in this study was 2.5 years

(male mean = 2.6 years, females mean = 2.4 years) (see Figures 2 and 3). There was no significant relationship between age and weight in males (r= 0.16, p= 0.28), (see Table

3). The correlation matrix in Table 3 shows a statistically significant negative

relationship between age and weight in females (r = -0.41; p<0.01), indicating weight

decreases with age. It is therefore likely that female birds undergo significant atrophy

with advancing age.

Wing and Tail Length: Rensch's rule, which in part describes differences in body

sizes and corresponding body parts, dictates that longer wings would likely run in

accordance with longer tails (Fairbairn 2007). The correlation matrix in Table 3 confirms

this in both genders (r = 0.68, p < 0.001 in males, r = 0.67,

p < 0.001 in females).

Weight and Tail Spots: I hypothesized that because tail spots are probably an

indication of fitness, and because heavier birds might be better at foraging, a correlation

might be found between the weight of a bird and the number of tail spots it possesses.

This hypothesis was rejected by statistical analyses (r= -0.25, p= 0.079 in males, r=

0.068, p= 0.613 in females) (see Table 3).

Unexpected correlations: As may be anticipated from the large number of correlations (45) computed for Table 3, coefficients for several pairs of variables were statistically significant, although no clear interpretation could be offered. A few of the parameters analyzed in Table 3 resulted in unexpected correlations, as the magnitude of

these correlations was weak. For example, wing length in mated pairs is negatively correlated (r = -0.27, p = 0.04). Wing length in females is negatively correlated to tail spots in males (r= -0.34, p= 0.01) and weight of females is weakly correlated to age in males (r =-0.33, p = 0.01).

Ornamentation and Sexual Selection: In animal species with classical sexual

dimorphism (ornamented males and drab females), ornamentation is used in sexual

selection by females choosing males (Darwin 1871, Blount et al. 2003, Pennisi 2003).

Some degree of mutual sexual selection is not uncommon in birds with mild to moderate

sexual dimorphism (Amundsen 2000, Jawor et al. 2002, Ketterson 2004, Wolf et al.

2006, Kristin et al. 2007). In the last decade more research has focused on avian species

such as Dark-Eyed Juncos (Junco hyemalis), Northern Cardinals (Cardinalis cardinalis),

and Bluethroats (Luscinia svecica svecica) to study female ornaments (Amundsen 2000,

Jawor et al. 2002, Wolf et al. 2004). Researchers found that in both Bluethroats and

Dark-Eyed Juncos females are selected by males and that in agonistic interactions in

Northern Cardinals, females would use their ornamentation to maintain possession of

territory (Amundsen 2000, Jawor et al. 2002, Wolf et al. 2004). Prothonotary Warbler

females exhibit levels of similar phenotypic variation as are found in those species.

17 Therefore, it is hypothesized in this study that variation in phenology among female

Protonotaria citrea results in males choosing females to some degree based on this visual information.

Because a male of any species of animal can create an almost unlimited supply of sperm, it is in his best interest to copulate with as many females as possible, thereby fathering the most offspring (Shuster and Wade 2003). A female has a limited lifetime supply of eggs; to prevent wasting ova on less fit males, it is in her best interest to be discriminating (Hapgood 1979). It is hypothesized this is why the males tend to be

‘fancier’ than females (Darwin 1871, Amundsen 2000). Female birds tend to incubate the eggs more often than males, and cryptic coloration disguises her and her eggs from potential predators (Darwin 1871, Chapman 1966, Amundsen, 2000). This is the case in

P. citrea: males do not incubate eggs, although they do feed the hatchlings so there is some level of paternal investment in the offspring.

Sexual selection is not the only selective pressure offered as an explanation for ornamentation in either male or female birds. Intrasexual competition is thought to influence plumage variations in addition to or instead of sexual selection (Amundsen

2000, Jawor et. al. 2002, Fowlie and Kruger 2003, Wolf et al. 2004).

Alternately, if ornamentation in females is a byproduct of shared genomes, females carry these traits simply as a result of being a member of the same species as stated in the correlated response hypothesis. Supporters of this hypothesis explain the ornament in question is not fully expressed in females to allow for cryptic coloration as protection from predation while she is incubating her eggs. Ornamentation in females has, until recently, been regarded as superfluous. For these reasons it has historically

18 been assumed that in species with classic sexual dimorphism (also known as male-biased dimorphism), where males are ornamented and females are drab, females select males for mating and males do not select females. In reverse sexual dimorphism (also known as female-biased sexual dimorphism), females are the more ornamented sex, and take on the roles traditionally held by the males in other species of birds, including mating displays, nest building, territory selection and defense (Chapman 1966, Heinsohn 1997, Vanner

2004, Thayer Birding Software 2006). Avian groups with different types of reverse dimporhism include phalaropes, raptors, and eclectus parrots, the latter of which display an extreme version of reverse sexual dichromatism (Darwin 1871, Chapman 1966,

Heinsohn et al. 1997, Amundsen 2000, Saetre 2000, Vanner 2004, Thayer Birding

Software 2006).

Ornamentation in male birds is understood by ornithologists to be a sign of fitness, a trait selected for by females during courtship rituals (Darwin 1871, Blount et al.

2003, Pennisi 2003). These flamboyant traits come at a cost of nutritional resources and are commonly expressed in the tail length or coloration (Burtt 1986, Blount 2003, Pennisi

2003, Kristin et al. 2007). Risk of predation increases if the plumage is so bright as to attract predators (Fitzpatrick 2000). Long tail feathers increase drag and slow flight that would otherwise allow an individual to evade predation or make migration impossible

(Burtt 1986, Fitzpatrick 2000). Risk would obviously vary from species to species depending on the type and extravagance of the ornament (Fitzpatrick 2000). The peacock is the classic example of structural and colorful ornamentation so extravagant as to be considered cumbersome; peahen plumage is comparatively drab as females are ornamented but not to the same extreme (Darwin 1871).

19 The correlations of the number of tail spots within mated pairs are noteworthy.

Looking at the box and whisker plot in Figure 7, it appears that the 5-spot males

apparently avoid the 5 spot females and that this is the primary source of the negative

slope to the regression. It may be that the older males, given the ability to be selective,

have some tendency to avoid 5-spot females. Perhaps those females pose an increased

risk of attracting predators which more than offsets any individual fitness advantage that the 5 spots may be signaling. Again, further study is needed before any conclusive statements can be made about these mated pairs.

Avian ornamentations such as tail spots, wing bars, bright coloration, or elaborate tail feathers are indicators of fitness in bird species and are used in courtship displays and

communication (Darwin 1871, Chapman 1966, Hanssen et al. 2006). Communication via

tail spot flashing is well documented in birds (Wible 1967, Bruce 1972, Johnsen et al.

1996, Mumme 2002). Prothonotaries flash their tail spots during grooming activities,

agonistic interactions, and flight take off and landing. The Slate-Throated Warbler

(Myioborus miniatus) uses its tail spots to flush prey out from under foliage. The Slate-

Throated Warbler then catches its prey in flight (Mumme 2002). However, for all of

their uses in signaling and communication, a lack of melanin in tail feathers compromises the integrity of the feather structure (Van Tyne and Berger 1971, Burtt 1986). Melanin provides resistance to abrasion and breakages, suggesting white tail spots have a cost

(Van Tyne and Berger 1971, Burtt 1986). This may provide a partial explanation for why

spotting is limited in younger females, allowing them to mature and gain life experience

before they develop the burden of tail spots.

20 In this context, it is interesting to note the observed significant correlation (r = -

35; p<0.01) for Tail Spot Number in mated birds (Table 3). These data, standing alone,

cannot establish that tail spots are used by the birds in selecting a mate. However, the

presence of the significant correlation at least requires the consideration of this

possibility. It is tempting to speculate that the number of tail spots on a given male is

related to the number on his mate, because of the presence of a third variable, for

example age. It was shown (Figure 6) that ages of mated pairs are positively correlated.

However, the effects of age cannot account for the significant tail spot correlation within

mated pairs, because of the negative correlation (r = -35) of the Tail Spot variables. In

addition, it was previously argued that the linkage between age and tail spots was of only

moderate strength in females (Figure 4, Table 2), and was probably negligible in males

(Figure 5). Figure 7 shows the linear regression analysis for Tail Spot Number in mated

male and female pairs. A significant (p<0.01) regression equation was obtained: Female

Tail Spots = 6.10 – (0.514 * Male Tail Spots). This result is of considerable interest, and

may relate importantly to the process of mate selection.

Referring to Table 4 provides correlation of tail spot numbers between birds of

mated pairs. The cross-tabulation of the tail spot numbers of Table 4 makes clear the

dominant pattern of 5-spot males paired with 3-spot females. However, the table also

shows exceptions to this pattern. For example, 22/23 (96%) of the 3-spot females paired

with 5-spot males, but only 10/17 (59%) of the 5-spot females were mated with the 5-spot

males. In addition, 4/5 (80%) of 3-spot males were paired with 5-spot females which

also runs counter to the dominant patterns seen in the 5-spot males.

21 Although the numbers of 3-spot males are low (n=5) and must be interpreted with

caution, it is conceivable that these males tended to pair with 5-spot females, which may

be available to them due to rejection by 5-spot males, as discussed above. While the χ2  was not significant, this may be due to the large number of cells with low expected values, although this does not alter the importance of the significant correlation. Overall, these tail spot correlations are intriguing because they may reflect complex use of ornamentation in mate selection, perhaps involving participation of both sexes in the selection process.

In summary, this study obtained novel results, which may relate to the function of tail spots as a signaling device in mate selection. Specifically, the avoidance of 5-spot females by older males is an intriguing finding, possibly suggesting the avoidance of predation risk or of feather wear due to abrasion (Burtt 1986). Long-term studies targeting variations in phenology in both male and female birds are needed to clarify the natural history of Protonotaria citrea in Virginia. Due to its specific breeding habitat needs, this Wood Warbler is susceptible to declines in population due to drought and habitat changes from human activities such as logging and agricultural endeavors in the bird’s lowland habitats (Cartwright 1998, Petit 1999). It is my hope that this study contributed to the existing body of knowledge of Prothonotary Warbler natural history.

Such information may contribute to the conservation of this and other species as natural habitats are continually challenged by human activity.

Literature Cited

Amundsen, T. 2000. Female Ornaments: Genetically Correlated or Sexually Selected?

Pages 133-154 in Y. Epsmark, T. Amundsen and G. Rosenqvuist, eds. Animal Signals:

Signaling and Signal Design in Animal Communication. Tapir Academic Press,

Trondheim, Norway.

Baird, J. 1967. Some Courtship Displays of the Golden-Winged Warbler. Wilson

Bulletin.79:3, 301-306.

Blake, C. H. 1966. Warbler Tail Spots. Eastern Bird Banding Association News. 29: 2,

54-55.

Blem, C. R. and Blem, L. B. 1992. Prothonotary Warblers Nesting in Nest Boxes: Clutch

Size and Timing in Virginia. The Raven. 63: 1, 15-19.

Blem, C. R. and Blem, L. B. 1993. Composition and Microclimate of Prothonotary

Warbler Nests. The Auk 111(1):197-200.

Blem, C. R. and Blem. L. B. 1999. Relationships of Clutch Size and Hatching Success to

Age of Female Prothonotary Warblers. Wilson Bulletin. 11: 4, 577-581.

Blount, J.D., Metcalfe, N.B., Birkhead T. R., and R.F. Surai. 2003. Carotenoid

Modulation of Immune Function in Zebra Finches. Science. 300: 125-127.

Bruce, J. A. 1972. Tail flashing display in the Whip-poor-will. The Auk. 682.

23 Burtt Jr., E. H. An Analysis of Physical, Physiological, and Optical Aspects of Avian

Coloration with Emphasis on Wood-Warblers. The American Ornithologist’s Union.

1986. pp Washington, D.C.

Cartwright, L. 1997. Prothonotary Warbler Nest Box Program in Northern Virginia.

Sialia. 19: 2, 43-48.

Cartwright, L. 1998. Prothonotary Warbler Nest Box Program in Northern Virginia, 1997

Results. Sialia. 20:1, 8-9.

Chapman, F. M. 1966. Handbook of Birds of Eastern North America. Pages 93, 95-96.

Dover Publications Inc. New York.

Darwin, C. The Origin of Species by Means of Natural Selection and The Descent of

Man and Selection in Relation to Sex, 1859 and 1871. pp. 591-592. The Modern Library,

New York.

Ficken, M.S. and Ficken, R. W. 1968. Courtship of Blue-Winged Warblers, Golden-

Winged Warblers, and their Hybrids. Wilson Bulletin. 80: 2, 161-172.

Field, Andy. 2000. Discovering Statistics Using SPSS for Windows. pp. 47-48, 87-97.

Sage Publications Ltd. London.

Fitzpatrick, S. 2000. A Signalling Tail. Pages 121-130 in Y. Epsmark, T. Amundsen and

G. Rosenqvuist, eds. Animal Signals: Signaling and Signal Design in Animal

Communication. Tapir Academic Press, Trondheim, Norway.

Fowlie, M. K. and Kruger, O. 2003. The evolution of plumage polymorphism in birds of

prey and owls: the apostatic selection hypothesis revisited. Journal of Evolutionary

Biology. 18: 577-583.

24 Hanke, J.E. and Reitsch, A. G. 1986. Business Forecasting 2nd Edition. Pages 119-129.

Allyn and Bacon Inc. Newton, Massachussetts.

Hanssen, S.A., Folstad I., and K. E. Erikstad. 2006. White plumage reflects individual quality in female eiders. Animal Behaviour. 71: 337-343

Hapgood, F. 1979. Why Males Exist. p. 123. Willam Morrow and Company.

Heinsohn, R., Legge, S., and S. Barry. 1997. Extreme Bias in Sex Allocation in Eclectus

Parrots. Proceedings: Biological Sciences. 264: 1386, 1325-1329.

Hoover, J. P. 2003. Decision rules for site fidelity in a migratory bird, the Prothonotary

Warbler. Ecology 84:416-430.

Jawor, J.M., Gray, N., Beall, S., and R. Breitwisch. 2002. Multiple ornaments correlate with aspects of condition and behavior in female northern cardinals, Cardinalis cardinalis. Animal Behavior. 67: 875-882.

Jeffrey, J. J., Fancy, S. G., Lindsey, G. D., Banko, P. C., . Pratt, T. K. and Jacobi, J. D.

1993. Sex and Age Identification in Palila. Journal of Field Ornithology. 64:4, 490-499.

Johnsen,T. S., Hengeveld, J.D., Blank, J.L., Yasukawa, K., and V. Nolan, Jr. 1996.

Epaulet brightness and condition in female Red-Winged Blackbirds. The Auk. 113:2,

356-362.

Kristin, A., Valera, F., Hoi, C. and Hoi, H. 2007. Do melanin-based tail patterns predict individual quality and sex in Lesser Grey Shrikes Lanius minor ? Journal of Ornithology.

148:1, 1-8.

Morton, E. S. 1989. Female Hooded Warbler Plumage does not become more Male-like with age. Wilson Bulletin. 101:3, 460-462.

25 Mumme, R. L. 2002. Scare tactics in a Neotropical Warbler: White tail feathers enhance

flush-pursuit foraging performance in the Slate-Throated Redstart (Myioborus miniatus).

The Auk. 119:4, 1024–1035

Pennisi, E. 2003. Colorful Males Flaunt Their Health. Science. 300: 29, 31.

Petit, L. 1999. The Birds of North America, Vol. 11, #409. Philadelphia, PA: American

Ornithologist’s Union; Academy of Natural Sciences.

Podlesak, D. W. and Blem, C. R. 2001. Factors Associated with Growth of Nestling

Prothonotary Warblers. The Wilson Bulletin. 113: 3, 263-372.

Pyle, Peter. 1997. The Identification Guide to North American Birds: Part 1. Slate Creek

Press, Bolinas, California.

Sawyer, Laurence. 1993. Prothonotary Warbler Nest Stub. Sialia. 15: 1, 17-19.

Saetre, G. P. 2000. Sexual Signals and Speciation. Pages 237-257 in Y. Epsmark, T.

Amundsen and G. Rosenqvuist, eds. Animal Signals: Signaling and Signal Design in

Animal Communication. Tapir Academic Press, Trondheim, Norway.

Schorre, B. 1998. The Wood Warblers. p. 95. University of Texas Press.

Shuster, S. M. and Wade, M.J. 2003. Mating Systems and Strategies. Pages124-127.

Princeton University Press.

Thayer Birding Software’s Guide to Birds of North America, v3.5. 2006. Cornell Lab of

Ornithology.

Vanner, M. 2004. The Encyclopedia of North American Birds. Barnes and Noble Books, pp. 88-89, 148-149.

26 Van Tyne, J. and Berger A.1971 Fundamentals of Ornithology. Page 101. Dover

Publications, Inc. New York.

Wible, M. 1967. Wing and Tail Flashing of Painted Redstart. Wilson Bulletin. 79: 2, 246.

Wolf, W.L. Casto, J.M., Nolan, V., Ketterson, E.D. 2004. Female ornamentation and male mate choice in dark-eyed juncos. Animal Behavior. 67: 93-102.

Zyla, J. and Donovan, M. 1993. Prothonotary Warblers Use Nest Box Inside Capital

Beltway. Sialia. 15: 1, 20-21.

Appendix A

27 28

Figure 1. Tail spot variation in Warbler species. A. Examples of Eastern Warbler tail spots, modified from Blake (1966). B. Variations of tail spot patterns in female P. citrea. C. Male Prothonatary Warbler exhibiting tail spots. Photo courtesy of Robert S. Mulvihill. 29 Descriptive Statistics: Female Birds

AGE N: 59 40 Mean : 2.6 30

Median : 2.0 20

1.12 Frequency S.D. 10 Min : 1

Frequency 0 654321 Max: 6 AGE1 Age (years) RETRICE 30 N: 59

Mean : 43.9 20 Median : 44.0 10

S.D. 1.65 Frequency

Min : 38.0 Frequency 0 474645444342414039 Max: 47.0 RETRICE1Retrice (mm)

WING 40 N: 59 30 Mean : 67.3 Median : 67.0 20 S.D. 2.06 10

Frequency

Min : 62.0 Frequency 0 74727068666462 Max: 74.0 WING1 Wing chord (mm) WEIGHT 16 N: 57 14 12 Mean : 15.5 10 Median : 15.5 8 6

Frequency 4 S.D. 0.95 14.5 15.5 16.5 17.5 2

Min : 13.5 Frequency 0 1717161615151414

Max: 17.0 NEWWGHT1Weight (grams)

TAIL SPOTS 30 N: 59

Mean : 3.7 20 Median : 2.0 10

S.D. 1.06 Frequency

Min : 1 Frequency 0 54321

Max: 5 TSPOT1Number of Tail Spots

Figure 2. Descriptive statistics for female birds. For histograms, weight was rounded to nearest gram, and length measures (retrice, wing) grouped into 2 mm bins. 30 Descriptive Statistics: Male Birds

AGE 40 N: 59 30 Mean : 2.4 20 Median : 2.0

10 S.D. 0.67 Frequency

Min : 1 Frequency 0 1 2 3 4

Max: 4 AGE2 Age (years)

RETRICE 30 N: 55 Mean : 46.4 20 Median : 47.0 10

S.D. 2.12 Frequency

Min : 42.0 Frequency 0 525048464442

Max: 52.0 RETRICE2Retrice (mm)

WING 30 N: 55

Mean : 71.8 20 Median : 72.0 10

S.D. 2.37 Frequency

Min : 65.0 Frequency 0 706866 72 74 76 78

Max: 77.0 WING2 Wing chord (mm)

WEIGHT 30 N: 52 Mean : 14.3 20 Median : 14.0 10

S.D. 0.84 Frequency

Frequency 0 Min : 13.0 13 14 18171615 Max: 17.5 NEWWGHT2Weight (grams)

TAIL SPOTS 50 N: 55 40 Mean : 4.7 30 Median : 5.0 20 S.D. 0.72 Frequency 10

Min : 2 Frequency 0 5432

Max: 5 NumberTSPOT2 of Tail Spots

Figure 3. Summary descriptive statistics for male birds in study. Conventions as in Figure 2. 31

3 Age

0 1 2 3 45 Number of tail spots

Figure 4. Box and whiskers plot showing the relationship of Age to the Number of Tail Spots in female birds. In this and similar figures to follow, the box extends from the 25th percentile to the 75th percentile, and the bold horizontal line depicts the median. The whiskers extend above and below the box to show the highest and lowest values. The dashed line shows the least-squares line of best fit when Age was regressed on Number of Tail Spots in female birds. The resulting regression equation [Age= 2.01 + ( 0.158 * Tailspots) ] was not significant. pt nml id.Tersligrgeso oe g .5+(.8 * (0.081 + 2.05 = Age [ Tail model of significant. regression Number not resulting on was The regressed Tailspots)] line was dashed 2 birds. Age The when had male fit bird in datapoint. best male Spots that of one for line Only whiskers least-squares of the 4. lack shows Figure the in in as resulting are spots, Conventions tail birds. male in Spots Tail 5. Figure Age 0 1 2 3 4 o n hsespo hwn h eainhpo g oNme of Number to Age of relationship the showing plot whiskers and Box 1 2 ubro alspots tail of Number 3 45 32 33

Age of Female 2

0 1234 Age of Male

Figure 6. Box and whiskers plot showing the relationship of Female Age to Male Age within mated pairs. Conventions are as in Figure 4. The dashed line shows the least-squares line of best fit when Female Age was regressed on Male Age. The resulting regression model [ Female Age = 1.02 + (0.662* Male Age)] was statistically significant (p<0.01). 34

Number of Tail Spots in Female 0 12345

Number of Tail Spots in Male

Figure 7. Box and whiskers plot showing the association of Number of Tail Spots in mated pairs. Conventions are as in Figure 4. The dashed line shows the least- squares line of best fit when Number of Tail Spots in females was regressed on Number of Tail Spots in male birds. The resulting regression model [ Female Tail Spots = 6.10 - (0.514* Male Tail Spots)] was statistically significant (p<0.01). 35

Table 1. Cross-tabulation of Age and Number of Tail Spots in female birds. 36

Table 2. Cross-tabulation of Age and Number of Tail Spots in female birds, with the variable Age re-grouped into three ordinal categories: 1-yr-olds, 2-yr-olds, and 3+-yr-olds. 37

Table 3. Table of correlation coefficients between all pairs of study variables. Each table entry shows the value of the Pearson Product-Moment Correlation Coefficient, with the significance probability below (italics ). Eight pairings exhibited significant correlation (outlined in boxes). *p<0.05; **p<0.01; ***p<0.001. Probability values less than 0.0005 are shown as 0.000 (SPSS output).

Tailspot Weight Wing Retrice Age Tailspot Weight Wing Retrice Number Chord Length Number Chord Length

Age 0.110 -0.172 0.080 0.009 0.397 ** 0.150 -0.414 ** 0.019 0.130 0.424 0.224 0.562 0.949 0.002 0.257 0.001 0.887 0.325

Retrice -0.226 0.220 -0.186 0.010 -0.038 0.083 0.052 0.671 *** Length 0.097 0.116 0.173 0.943 0.778 0.533 0.700 0.000

Wing -0.343 * 0.172 -0.273 * -0.057 0.074 0.215 0.096 Chord 0.010 0.223 0.044 0.681 0.579 0.102 0.478

Weight -0.123 0.013 -0.117 0.041 -0.337 * 0.068 0.377 0.927 0.400 0.770 0.011 0.613

Tailspot -0.350 ** 0.210 0.049 0.129 0.111 Number 0.009 0.134 0.723 0.349 0.408

Age 0.093 0.156 -0.010 0.047 0.506 0.275 0.943 0.735

Retrice 0.252 -0.202 0.684 *** Length 0.067 0.156 0.000

Wing 0.170 -0.179 Chord 0.220 0.210

Weight -0.248 0.079 38

Table 4. Cross-tabulation of Number of Tail Spots in males and in females of mated pairs.

Number of Tail Spots 23 4 5Total 1 00 0 1 1 2 00 1 3 4 3 01 022 23 Number of 4 00 1 9 10 Tail Spots 5 14 210 17 Total 15 44555 c2 (12)= 13.90; p=0.307 tb = -0.304; p=0.027 r = -0.350; p=0.009

VITA

Terry Lee Smith was born in Mountain View, California in 1970. She graduated from Denbigh High School in 1988 and from Christopher Newport University in 1995.

During the summer of 1994 Terry studied tropical ecology at the Monteverde Institute of

Tropical Ecology in Costa Rica, where she did an independent study on the use of

olfaction in Artibeus toltecus, a species of fruit bat. She spent the three years following

undergraduate study working as a licensed veterinary technician (LVT) in Portland,

Oregon. She has over 13 years of experience in the veterinary field and has worked with

a wide variety of wild and domestic animals. Terry entered Virginia Commonwealth

University’s graduate program in Biology in 2002 and taught general and human biology

while she pursued her advanced degree. She is currently employed as a Laboratory

Specialist at VCU's Department of Anatomy and Neurobiology; she is part of a team that

investigates the effects of traumatic brain injury on electrophysiology.