|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
DIFFERENTIAL ALLOCATION IN THE FLAGFISH, Jordanellafloridae, IN
RESPONSE TO MALE CONDITION
MELISSA ANN NASUTI
A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE
UNIVERSITY OF FLORIDA
Melissa Ann Nasuti
For advice on project design and thesis preparation I would like to thank Colette St.
Mary and Rebecca Kimball. I would also like to show appreciation to Craig Nasuti, Luis
Bonachea, Jena, Chojnowski, Hope Klug, and Holly Kindsvater for aiding in the
collection of the study species.
TABLE OF CONTENTS
A C K N O W L E D G M E N T S ......... .................................................................................... iii
L IST O F T A B L E S .............................................................................................. v
LIST O F FIG U R E S .... .............................. ....................... ........ .. ............... vi
ABSTRACT ........ .............. ............. .. ...... .......... .......... vii
INTRODUCTION ............................... .................... .............
M E T H O D S ............................................................................ . 7
Stu dy Sp ecies ................................................ 7
Study Design...................................................... ..... ............... ...............
Experimental Manipulation of Male Condition........................................................8
Fem ale M ating Trials ............................................................ ... ... ....8
M measures of Reproductive Allocation .................................. ............ .................. 10
Color A analysis of M ales ......................................... ...... .. .... .. .............. .. 11
Lipid A analysis of M ales .................. ........................... .................... 12
B ehavioral Observations of M ales..................................... .......................... ......... 12
Statistical A nalysis................................................... 13
R E S U L T S ................................................................................14
Effects of Diet Manipulation on Male Body Size, Condition, Coloration &
B behavior .................. .......... .. ......... ..... .... ..............................14
Measures of Reproductive Allocation & Male Food Treatment .............................15
Measures of Reproductive Allocation & Male Traits..............................................17
Female M ate Choice ............................. ....................... ... ............... 19
Condition-Dependent Expression of M ale Traits...................................................21
D ISC U SSIO N ............... .................................... ...........................40
LIST OF REFEREN CE S ........................................ ........................... ............... 48
B IO G R A PH IC A L SK E TCH ..................................................................... ..................55
LIST OF TABLES
1 Description of nest and non-nest directed behaviors recorded for each male
fla g fish ............................... ......... ...... ..................... ................ 2 2
2 The effect of male food treatment on morphological traits.................................23
3 The effect of male food treatment on measures of courtship prior to spawning......24
4 The effect of male food treatment on female reproductive allocation.. ..................25
5 Results of analyses addressing variation in average egg area and larval length in
response to male morphological and behavioral traits.........................................26
6 Differrences between mated and unmated males in measures of male body size,
fat reserves, condition factor, and coloration ................................. ............... 27
7 Results of condition-dependent expression analyses. ............................................28
LIST OF FIGURES
1 The relationship between fresh weight and standard length is shown for both
high food and low food m ales. ........................................................................... 29
2 The relationship between dry weight and standard length is shown for both high
food and low food m ales. ........................................ ................................. 30
3 The relationship between fat weight and standard length is shown for both high
food and low food m ales. ...... ........................... ........................................31
4 Values of male condition are plotted against residuals from a univariate analysis
of co-variance. ....................................................................... 32
5 Average larval length (mm), calculated as the average total length of individual
larvae within a clutch is plotted against the average hatch date for that clutch.......33
6 Values of male condition are plotted against residuals from a univariate analysis
of co-variance. ....................................................................... 34
7 Individual egg areas were measured within a clutch and then averaged. These
values are plotted against residuals from a univariate analysis of co-variance........ 35
8 Average proportion of total observation time each male spent at their nest( + SE)
for mated and unmated males prior to receiving eggs. .........................................36
9 Average proportion of time spent at the nest fanning ( SE) for mated and
unmated males prior to receiving eggs .... ........... ........................................ 37
10 Frequency of chases ( SE) performed by mated and unmated males during their
recorded observation time prior to receiving eggs.. ........................................... 38
11 Relationship between fanning and fat reserves. Transformed values are show.......39
Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science
DIFFERENTIAL ALLOCATION IN THE FLAGFISH, Jordanellafloridae, IN
RESPONSE TO MALE CONDITION
Melissa Ann Nasuti
Chair: Colette St. Mary
Cochair: Rebecca Kimball
Major Department: Zoology
The Differential Allocation Hypothesis argues that females should invest more
resources into reproduction when paired with high quality mates as a result of greater
direct/and or indirect benefits received from reproduction. Evidence suggests that energy
reserves often influence male character elaboration and hence the benefits received from
mate choice. The current study examined the potential for differential allocation by
females in response to male condition in the flagfish, Jordanellafloridae. Male condition
was manipulated through diet and was expected to affect measures of male body size,
condition (weight/length3), fat reserves, color expression, and parental care behavior.
Females interacted sequentially with two males from the same or opposite feeding
treatment during a two week mating trial. Egg number, egg size, egg energy, hatching
success and larval length were recorded as measures of female reproductive allocation.
High food males were found to be heavier and longer, suggesting that males prioritize
additional nutrients to growth rather than fat stores. As expected, high food males showed
greater expression of blue and yellow coloration. Measures of egg number, egg size, egg
energy and hatching success were not found to significantly vary with measures of male
condition, coloration or behavior. However, males in low condition spent more time
fanning, an indicator of parental care. Counter to expectations, male condition had a
significant effect on larval length as males of a lower weight for a given length fathered
larger larvae. This may suggest that females allocate more resources to eggs when mating
with males that provide better parental care. These results are consistent with the
Differential Allocation Hypothesis suggesting that female flagfish potentially manipulate
reproductive investment in response to mate attractiveness.
One of the intriguing issues still facing evolutionary biologists today is to explain
the evolution of elaborate secondary sexual traits that often appear detrimental to the
survival of an individual. Darwin (1871) proposed that such traits do not evolve under
natural selection, but via sexual selection. He argued that the cost of production and
maintenance of these traits is offset by an increase in reproductive success gained through
the acquisition of mates. Typically, males are viewed as the competitive sex and use these
traits to compete over access to females or resources for mating. Females are seen as the
"choosier sex" and use these traits to discriminate amongst males (Andersson 1994).
Considerable evidence from studies of sexual selection have shown that female mate
choice is one of the processes which influences the evolution of male traits. Empirical
support also suggests that females can further influence trait evolution by varying the
amount of reproductive investment in their offspring in response to male trait expression
Studies in birds, amphibians, and more recently fish, indicate that females are able
to adjust the amount of parental effort (Burley 1988; Szentirmai et al. 2005), manipulate
the total number (Petrie & Williams 1993; Reyer et al. 1999; Parker 2003; Edvardsson &
Arnqvist 2005), size (Cunningham & Russell 2000; Kolm 2001; Kolm & Olsson 2003),
and sex ratio of eggs produced (Kempenaers et al. 1997; Sheldon et al. 1999), and control
egg quality through adjustment of testosterone and immune factors (Gil et al. 1999; Saino
et al. 2002a). Adjustments in reproductive investment in response to mate attractiveness
were first tested by Burley in 1986 by manipulating the perceived attractiveness of a bi-
parental care species of zebra finch, Taeniopygia guttata, with colored leg bands. Burley
referred to this concept as the Differential Allocation Hypothesis (DAH).
The DAH is based on the premise that the attractiveness of a mate influences the
reproductive value of the breeding attempt with that mate (Sheldon 2000). Differences in
male trait expression will lead to selection acting on females to invest more in current
reproduction when the investment is balanced by a greater benefit. Attractive high quality
mates may affect the payoffs received from reproduction by providing greater direct and/
or indirect benefits to the female and her offspring than unattractive low quality mates.
However, greater investment in current reproduction comes at a cost by compromising
the amount of available resources for future reproduction. Therefore, selection for
differential allocation will depend on the magnitude of perceived benefits received from
reproduction. Furthermore, the expected future reproductive lifespan of the female and
the expected quality of future partners will also influence the role of differential
allocation. Increased allocation is only expected when the current partner is more
attractive than expected future partners (Sheldon 2000).
While attempting to explain the evolution of female mating preferences, models of
sexual selection have clearly described the potential benefits females receive through
mating. Indicator models propose that male traits evolve to "indicate" or "honestly
signal" some aspect of male phenotypic and/or genotypic quality (Zahavi 1975; Kodric-
Brown & Brown 1984; Grafen 1990). Male traits become targets of female preference as
a result of the correlation between the direct and/or indirect benefits a female receives
and the expression of the male trait. Indirect benefits, commonly referred to as "good
genes" are expected to increase offspring performance through inheritance of paternal
genes for viability (Norris 1993; Petrie 1994), sexual attractiveness (Gwinner & Schwabl
2005), and parasite resistance (Hamilton & Zuk 1982; Moller 1990). Alternatively,
females may directly benefit from non-heritable resources that increase female survival
or fecundity. Direct benefits include better parental care (Petrie 1983; Norris 1990),
courtship feeding (Nisbet 1973; Wiggins & Morris 1986), nuptial gifts (Thornhill 1976),
or access to resources such as superior nest and oviposition sites (Holm 1973;
Campanella & Wolf 1974).
For male traits to "honestly" represent the fitness benefits gained by females,
theory suggests that male traits must be physiologically costly either to produce or
maintain (Andersson 1994). Indicator models of sexual selection, therefore predict the
expression of the trait to be condition-dependent (Grafen 1990). That is, the degree of
trait expression is positively correlated with physical condition. Males in better condition
signal their quality through greater sexual trait size or more vigorous displays. Males in
worse condition incur higher costs that are associated with the production and/or
maintenance of these traits and are unable to express the traits to the magnitude that
males in better condition can. Evidence for honest advertisement has been found across
taxa (Johnstone 1995).
In practice, researchers have used measures of energy reserves (Bolger & Connolly
1989), vigor (Kodric-Brown & Nicoletto 1993), and/or health (Milinski & Bakker 1990)
to estimate an individual's condition. Empirical support for the influence of condition
measures on the expression of male traits, such as breeding coloration and courtship
rates, has been shown in fish. Research of guppies reveals that several of the traits which
females use to discriminate amongst males are condition-dependent and function to
indicate viability (Nicoletto 1993). For example, male display rate, amount of orange
coloration, and ranks of ornamental complexity were all found to be positively correlated
with condition measures (Nicoletto 1993). Breeding coloration has also been shown to be
positively correlated with condition measures in sticklebacks (Milinski & Bakker 1990)
and in pupfish (Kodric-Brown & Nicoletto 1993).
Furthermore, condition measures have been successful in explaining male variation
in reproductive success as a result of the influence of male energy reserves on parental
care. In the sand goby, Pomatoschistus minutus, food-supplemented males with higher fat
reserves spent more time at their nest caring for their young. As a result, these males
mated sooner and received more eggs than un-supplemented males (Lindstrom 1998).
Diet manipulation in small mouth bass also resulted in an increase in reproductive
success as a result of food-supplemented males providing longer periods of parental care
after swim up of larvae (Ridgway & Shuter 1994). Similar patterns of condition-
dependent parental care are also seen in the bicolor damselfish, Stegastespartitus, as
females receive higher quality parental care from more vigorously courting males which
is a reflection of male fat reserves (Knapp & Kovach 1991; Knapp 1995).
These experimental results suggest that condition-dependent expression of male
secondary sexual traits is common in many fish systems. Furthermore, female
preferences for condition-dependent traits often lead to greater direct and/or indirect
benefits. According to the DAH, selection should favor females who invest more with
The objective of this study was to determine if female flagfish, differentially
allocate reproductive resources in response to male condition. Previous evidence in this
system suggests that females may discriminate amongst males based on condition. St.
Mary et al. (2001) showed condition values of flagfish males with low spawning success
to decline more drastically over the breeding season than more successful males.
Furthermore, evidence that females are evaluating males based on traits associated
with direct reproductive benefits makes flagfish an attractive system for the study of
differential allocation. Males provide exclusive parental care of eggs by defending
breeding territories from egg predators, oxygenating their developing eggs through
fanning, and cleaning their nests of debris. Parental care has been shown to be used in
mate attraction as a positive relationship was found between the numbers of eggs a male
received and initial fanning behavior (St. Mary et al. 2001). If the DAH holds, females
should allocate resources to more vigorously displaying males who may be in better
condition and are able to provide better parental care.
It is likely that parental care is a function of male energy reserves in this system.
Courtship takes place at the nest with males continually chasing and displaying towards
passing females. These activities confine males to their nests and may restrict foraging
opportunities to short periods of time interspersed between courting females and
aggressive interactions with neighboring males. Males may therefore be constrained by
food availability. This may affect their ability to provide parental care and subsequently
influence female resource allocation. Courtship displays and parental care are known to
be energetically costly in other fish species (Smith & Wootton 1999).
In this experiment, male condition was experimentally manipulated through diet by
assigning males to a high or low food treatment. This manipulation was expected to
affect measures of body size, fat reserves, color expression, and courtship/parental care
behaviors. Over a two week period, females interacted sequentially with two males from
equal or opposite feeding treatments. To evaluate differential allocation the total number
of eggs spawned with each male was recorded in addition to measures of hatching
success, egg energy, egg size, and larval length. Under the DAH, females were expected
to provide a larger number of eggs and/or higher quality eggs to males who have been
maintained on high food rations than low food rations.
Flagfish, Jordanellafloridae, are found in fresh, brackish and salt water systems of
central and southern Florida (Foster et al. 1969). Flagfish inhabit areas of dense
vegetation and live approximately one year (Boschung et al. 2000). Flagfish have a
promiscuous mating system. Their breeding season extends from late March to early
September (Boschung et al. 2000). Males defend breeding territories from male
competitors and egg predators. During courtship, males will repeatedly chase and circle
the female. Spawning occurs on submerged vegetation or leaf litter within the breeding
territory. During daylight hours, a female may spawn with a single male multiple times or
distribute her eggs among males (Mertz & Barlow 1966). Females will lay one egg at a
time, releasing approximately 20 or more eggs during a single spawning bout. Males
continually accept eggs from multiple females and provide parental care by fanning,
guarding, and cleaning their eggs (Mertz & Barlow 1966). Males continue to care for
their eggs until hatching occurs within 4-5 days of spawning at 250C (Smith 1973).
Flagfish show sexual dimorphism in both body size and coloration. Males are
larger in size and are the more colorful sex. Males possess horizontal red-orange stripes
on their dorsal and anal fins as well as along their side. A large dark spot surrounded by
yellow is located on their mid side and blue iridescence appears on their operculum.
During periods of courtship and nest guarding, males temporarily express dark coloration
over their entire body. Females are silvery gray in color and are characterized by a dark
spot located at the tail end of their dorsal fin (Page & Burr 1991).
This study was conducted at the University of Florida in Gainesville, Florida.
Experimental research was carried out over one breeding season from March to August
of 2005. Flagfish for this study were collected during the summer of 2005 from the Otter
Creek/Waccasassa River drainage system in northwest central Florida. Sexes were
housed separately in 152 liter holding tanks and maintained on a 14L:10D light cycle.
The room was kept at a minimum of 270 C.
Experimental Manipulation of Male Condition
Males were randomly assigned to one of two food treatments. In the holding tanks,
high food (HF) males were fed a diet of algae tabs, commercial flake food, and frozen
chironomid larvae in excess two times a day. Low food males (LF) received a diet of
algae tabs and were fed once every other day. Males were maintained on the diet for a
period of 14 to 42 days prior to interacting with a given female. Females were fed in the
same manner as high food males in order to maximize their potential for egg production.
To control the diet of both sexes during female mating trials, males and females were
separated by a clear acrylic partition and fed independently at the end of each day. Any
remaining food was removed from the tank before the start of the mating trial the
following morning using a fine mesh net.
Female Mating Trials
A single female and single male were placed in a 38 liter holding tank. A spawning
mat was set in the center of the tank and an artificial plant (Ludwigia species) was
positioned in the back of the tank. The spawning mat was constructed from a square
ceramic tile (10cm x 10cm) with green felt carpet attached to the top. Male flagfish have
been observed in previous research to readily accept these spawning mats as breeding
territories and typically all eggs are spawned on the mat. Clear monofilament fishing line
was used to separate the spawning mat into 20 square quadrants to aid in the task of
counting freshly spawned eggs.
Over a ten day period, each female was given the opportunity to interact with two
different males for five days each. Each male-female pair interacted from 0830 to 1630
hours, for a total of eight hours every day. At the end of each day a perforated clear
acrylic partition was placed in the center of the tank between the male and female to
prevent spawning overnight. Clear perforated partitions were used to facilitate visual and
chemical communication between the pair throughout the night. At the end of the first
five days the male was removed and replaced with a new male from the opposite or same
feeding treatment. Each female was randomly assigned to one of the following four
treatments; 1) HF/HF 2) LF/LF 3) HF/LF and 4) LF/ HF. For every ten day period, each
treatment was replicated four times. This design ensured that the treatments were evenly
distributed throughout the breeding season. Each male and female was used only once.
Prior to each five day mating trial, perforated clear acrylic partitions were used to
physically separate each male-female pair for a two day period. This period of
acclimation gave each male a sufficient amount of time to establish his breeding territory
over the spawning mat. Prior to interacting in the mating trial, the fish were weighed on
an electronic balance to the nearest 0. g and measured for standard length to the nearest
Measures of Reproductive Allocation
To assess differential allocation the following five parameters were measured with
each male 1) total number of eggs spawned with each male 2) average individual egg size
per clutch 3) average egg energy content per clutch 4) hatching success rate of larvae and
5) average larval length per clutch.
Spawning mats were checked for eggs four times a day at approximately 0900
hours, 1100 hours, 1300 hours and 1600 hours. At each time the total number of new
eggs was recorded. Freshly spawned eggs were differentiated from older eggs by viewing
the developmental stage under a dissecting scope and by noting the placement of eggs
within the monofilament grid. After the first spawning event, up to 30 eggs were removed
from the spawning mat. The remaining eggs were left on the mat to promote continued
spawning. If the removal of 30 eggs reduced the total clutch size by more than fifty
percent, eggs were removed on subsequent spawning events. Of these eggs, 15 were
immediately frozen for energetic analysis. The remaining 15 eggs were then placed in a
water filled plastic dish and digitally photographed under a dissecting scope.
ImageProPlus (Version 4.5.1) software was used to measure the total area (mm2) of each
individual egg within a clutch. To reduce error, three area measurements were recorded
and an average of these measurements was used in calculating the average egg area for
the entire clutch. The eggs were then placed in a 100 ml water filled glass rearing
chamber. Each rearing chamber was supplied with an air stone and maintained at room
temperature of 250C. Hatching success was measured as the proportion of fertilized eggs
hatched by day seven. Once hatched, larvae were digitally photographed by the same
methods used for the eggs. Three measurements of total length (mm) were taken for each
larvae and an average of these measurements was used in calculating the average larval
length for the entire clutch.
To measure the average egg energy content per clutch a dichromate oxidation
method was used following the procedure of McEdward and Coulter (1987). Three eggs
were randomly chosen from the 15 egg sample. Each egg was incubated for 15 minutes at
1100 C in lml of 70% phosphoric acid and then oxidized in 2 ml of 0.30% potassium
dichromate for an additional 15 minutes and incubated at the same temperature. Samples
were diluted with 3.5ml distilled water and energy was measured with a
spectrophotometer (440nm) by comparison of the sample to glucose standards ranging
from 0 to 4 joules. An average of the three measurements was calculated to estimate the
egg energy content per clutch.
Color Analysis of Males
To record variation in color expression between males, digital photographs of each
male were taken with an Olympus C 2500L camera. To ensure the development of
nuptial coloration, photographs were taken at the end of the first day of each five day
mating trial in order to allow males significant interaction time with females. Each male
was hand-netted and placed in a clear water filled acrylic box (9cm x 3cm x 6cm) within
a standardized photo-box. The small box prevented the movement of the fish, exposing
the lateral side of the male to the camera. Each photo included a red, yellow, and blue
color standard. For each photograph we first sampled the three color standards recording
the hue, saturation, and intensity of each. This allowed for the adjustment of variation in
light intensity between images when recording color measurements. Red was defined as
having hue values ranging from 0 to 40 and saturation values ranging from 30 to 100;
yellow was defined as hue values from 40 to 55 and saturation values from 70 to 100; and
blue was defined as hue values from 100 to 160 and saturation values from 25 to 100. An
image analysis software tool, SigmaScanPro (Version 5.0.0), was used to measure the
mean intensity of red, yellow and blue coloration for each male. The total area of each
color on the fish's body was also measured and a proportion was calculated from the area
of the entire fish.
Lipid Analysis of Males
A sample of males from each food treatment were randomly selected and
euthenized in MS-222 at the end of the five day female mating trial. Males were frozen
and then later dried in an oven at 600 C for 24 hours. Upon removal, the dried males were
weighed on an electronic balance to the nearest 0.0001g. To assess differences in male
condition as a result of diet manipulation, lipids were extracted with petroleum ether
(Knapp 1995). Males were continually placed in ether until a constant dry weight was
achieved. The difference in weight was then calculated and the percentage of fat weight/
dry weight was used as a measure of condition in data analyses. Fresh weight/length3 was
also used as a second condition measure.
Behavioral Observations of Males
Males were videotaped using a VHS recorder for ten minutes on days one and two
of the five day mating trial between 1200 hours and 1600 hours. Flagfish are more active
in the afternoon (personal observation). This provided an estimate of a male's
reproductive behavior. To evaluate parental care, males were also taped for an additional
ten minutes in the presence of their first bout of eggs. Male behavior was analyzed using
an event recorder program written for this purpose. Male position was recorded as 1) at
the nest or 2) away from the nest. Measured male behaviors included 1) nest fanning 2)
nest cleaning 3) following 4) chasing and 5) spawning (see Table 1 for a more detailed
descriptions of these behaviors). Chasing and cleaning were recorded as counts since the
duration of these behaviors is short. Duration was recorded for fanning, following, and
spawning. The percentage of total observation time spent at the nest, away from the nest,
following, and spawning was calculated for each male. Frequencies for count behaviors
were calculated as the number of times a behavior was performed divided by the total
observation time. Fanning and cleaning are behaviors that occur only while the male is at
the nest. Therefore the percentage of time fanning and the frequency of cleaning were
calculated from the total time spent at the nest rather than the total ten minute observation
For all analyses, only data from females (and their mates) that spawned in at least
one week of the two week mating trial was considered. Parametric statistical analyses
were utilized when the variables met the requirements of these methods. For all variables
normality was tested using the one-sample Kolmogorov-Smirnov Test. If the variables
did not meet normality requirements, the appropriate transformations were made.
Variables expressed as proportions were arc sin square-root transformed. These included
color proportions as well as the proportion of measured fat weight/dry weight (g). The
numbers of eggs a female spawned were square-root transformed. Male and female fresh
weights (g) and standard lengths (cm) were log transformed. For general linear models,
variables with p-values greater than 0.15 were removed in a step wise fashion until the p-
values that remained were less than this set criterion. Nonparametric statistics were used
for behavior data since these variables did not satisfy normality assumptions when
transformed. All statistical analyses were performed using SPSS Version 12.
Effects of Diet Manipulation on Male Body Size, Condition, Coloration & Behavior
Male food treatment was found to have a significant effect on male body size and
coloration, though no differences were found for measures of male condition, fat
reserves, or behavior. High food males on average were significantly heavier and longer
than low food males (Table 2). However when controlling for length, condition did not
differ. To evaluate the effect of food treatment on male condition, a multivariate analysis
of variance was performed on fresh weight, dry weight, and fat weight with male
standard length as a covariate. All weight variables increased with standard length for
both food groups as expected (Figuresl-3 MANCOVA Fresh Weight F1, 124 = 997.61 P =
0.00; Dry Weight F1, 124 = 473.47. P = 0.00; Fat Weight F1, 124 = 98.78 P = 0.00). Male
food treatment had no significant effect (MANCOVA F3, 122 = 1.33 P = 0.27 Wilk's k =
0.97). Furthermore, weight measures did not increase more sharply with body length for
high food males than low food males (MANCOVA Food Treatment* Standard Length F3,
122 = 1.36 P = 0.26 Wilk's X = 0.97). Thus, while diet manipulation did have a significant
effect on the weight and standard length of high food males, their body composition
remained stable. Comparisons of % fat weight/dry weight showed no significant
differences in fat reserves between male food treatments (Table 2).
High food males expressed a significantly larger proportion of blue coloration as
well as more intense yellow coloration than low food males (Table 2). No differences
were seen between male food treatments for the proportion of red and yellow coloration
or the intensities of red and blue coloration (Table 2).
Although high food males were expected to display more vigorously, no difference
was found in the amount of time spent in courtship behavior between male food
treatments (Table 3). Furthermore, the probability that high food males performed each
measured behavior during courtship was not significantly different from that of low food
males (Table 3). Behavioral differences were also expected post-spawning. Male food
treatment did not influence the amount of time spent in each behavior once eggs were
received (Kruskal-Wallis Test At Nest X2 = 0.78 P = 0.34; Fanning At Nest X2 = 1.41 P =
0.24; Chase Frequency X2 = 0.85 P = 0.34, Following X2 = 0.37 P = 0.54; Cleaning
Frequency X2 = 0.11 P = 0.75).
Measures of Reproductive Allocation & Male Food Treatment
Females were expected to alter their reproductive investment in response to male
food treatment by allocating a larger number of eggs to high food males. These males
were also expected to receive clutches with larger more energetic eggs, a higher hatching
success rate and larger larvae. No such patterns were found.
To determine female egg allocation patterns with respect to male food treatment, a
repeated measures analysis of variance was used to investigate statistical differences in
egg numbers between weeks one and two of the mating trial. Two separate analyses were
completed, one in which females were paired with males from the same food treatment
(LF/LF & HF/HF) and a second in which females were matched with males from
separate food treatments (LF/HF & HF/LF). The first analysis determined if females on
average spawned more eggs when two high food males were received. The second
analysis determined if the order of male presentation affected female egg allocation.
Females paired with high food males during both weeks one and two, did not
allocate more eggs than females who received only low food males. Female treatment
had no significant effect on the average number of eggs spawned over weeks one and two
of the mating trial (LF/LF & HF/HF F1, 36 = 1.92 P = 0.18). Furthermore, the difference in
eggs spawned between weeks one and two did not vary with female treatment. There was
no significant interaction between week and female treatment when considering females
matched with males from the same food treatment (LF/LF & HF/HF Week*Treatment F1,
36 = 0.00 P = 0.95). The order in which high food males were presented had no affect on
patterns of egg allocation given a non-significant interaction between week and treatment
when considering females matched with males from separate food treatments (HF/LF &
LF/HF Week*Treatment F1, 50= 0.71 P = 0.40).
Females were not limited in their ability to allocate eggs between weeks. That is
females that spawn in week one are not more or less likely to spawn in week two. Egg
number did not differ between week one and two of the experiment across female
treatments (LF/LF & HF/HF F1, 36 = 0.026 P = 0.872; HF/LF & LF/HF F1, 50 = 0.261 P =
0.612). Given the lack of significant week effects, differences in egg allocation between
high food and low food males were directly analyzed in a third repeated measures
analysis. In this case, week was ignored, and differences between high food eggs and
corresponding low food eggs were compared. Within females there was no significant
difference in the number of eggs spawned with a high food versus a low food male (F1, 50
= 0.15 P = 0.70). Hence, no apparent patterns were seen in female spawning between
male food treatments.
Considering clutches from both weeks one and two, male treatment did not affect
hatching success (Table 4). Furthermore, when considering only the HF/LF treatment,
hatching success did not differ between male food treatments (Paired T-test t = -0.30, df=
7 P = 0.78). Low sample size prevented this same analysis with the LF/HF female
treatment. Overall, hatching success rate was high for both male food treatments.
Across all female treatments, there was a significant difference in egg size between
weeks for females that spawned both weeks of the mating trial (Paired T-test t = 2.99 df=
13 P = 0.01). Females spawned larger eggs in week one (1.15mm2 0.03 SE) compared
to week two (1.08mm2 0.02 SE) suggesting that females are limited in producing larger
eggs as spawning continues. Given this result, eggs produced during the second week of
the mating trial were removed from further analyses of egg size, egg energy and larval
length if the female spawned both weeks. Male food treatment did not affect average
individual egg size, average egg energy or average larval length (Table 4).
Measures of Reproductive Allocation & Male Traits
Additional analyses were utilized to investigate the association between allocation
variables and male morphological and/or behavioral traits. Variation in egg number, size,
and energy among females can not be attributed to morphological and/or behavioral
aspects of the sire. Larval length shows an exception.
The relationship between differences in egg numbers between week one and two of
the experiment with differences in male morphology and behavior between mates was
evaluated. A linear relationship was expected, with females allocating more eggs to males
with greater trait expression. Differences in egg numbers did not co-vary with differences
in male body size, condition, fat reserves and coloration (ANCOVA Weight F1, 40= 0.02
P = 0.90; Standard Length F1, 75 = 1.20 P = 0.28; Weight/Length3 F1,74 = 1.16 P = 0.29; %
Fat F1,45 = 0.13 P = 0.72; % Blue F1,43 = 0.08 P = 0.77; Intensity Blue F1,76 = 1.37 P =
0.25; % Red F1,41 = 0.04 P = 0.84; Intensity Red F1,42 = 0.05 P = 0.83; % Yellow F1,73
0.02 P = 0.89; Intensity Yellow F1,77= 2.20 P = 0.14). Nor were any correlations seen
with variation in male behavior (At Nest Spearman's p = 0.02 P = 0.85; Fanning At Nest
Spearman's p = 0.03 P =0.79; Cleaning Frequency Spearman's p = -0.02 P = 0.90;
Chasing Frequency Spearman's p = 0.03 P = 0.80; Following Spearman's p = -0.08 P =
To determine if females respond to differences in male morphology when
allocating eggs of varying size and energy, a univariate analysis of covariance was used.
Measures of color and male and female condition were used as covariates with male
treatment as a fixed factor. Male food treatment did not affect egg size or energy as
expected from previous results (ANCOVA Egg Size F1, 40 = 0.18 P = 0.68; Egg energy
F1, 34 = 1.75 P = 0.20). Male and female condition and blue and red coloration were not
found to co-vary with egg size or energy, nor were any significant correlations found
between egg size and energy and measures of yellow coloration or aspects of male
courtship behavior prior to spawning (Table 5). Male condition and red coloration did not
co-vary with egg size however marginal p-values were found. Therefore, residuals from
the ANCOVA model with proportion red as a covariate were regressed against male
condition. The linear regression was non-significant (R2 = 0.05 F1, 60 = 3.25 P = 0.08)
however a negative relationship between the variables are seen (Figure 4).
Similar analyses to those conducted on egg size and energy were utilized for larval
length. Measures of average hatch date, color, and male and female condition were used
as covariates with male treatment as a fixed factor. Average hatch date had a significant
positive influence on average larval length (ANCOVA F1,52 = 16.96 P = 0.00; Figure 5
Linear Regression R2 = 0.22 F1,79 = 22.63 P = 0.00). There was no effect of male food
treatment (ANCOVA F1, 48 = 0.16 P = 0.69) or female condition on average larval length
(Table 5). However, male condition showed significant effects on average larval length
(ANCOVA F1, 52 = 9.74, P = 0.00.) Residuals from an analysis of variance with average
hatch date as a covariate and average larval length as the response were regressed against
male condition. A negative relationship was found between average larval length and
male condition (Figure 6 Linear Regression R2 = 0.15 F1,65 = 11.64 P = 0.00) similar to
that found with egg size analyses. These results suggest that males of a lower weight for a
given length father larger larvae and received eggs of a greater size. Blue and red
coloration were not found to co-vary with larval length nor were any significant
correlations found between the residuals from the above analysis and measures of yellow
coloration or aspects of male courtship behavior prior to spawning (Table 5).
When hatch date is taken into consideration, egg area has a significant effect on
larval length (ANCOVA Average Hatch Date F1,67 = 11.50 P = 0.00; Average Egg Area
F1, 67= 5.06 P = 0.03). Residuals from an analysis of variance of larval length with
average hatch date as a covariate were regressed against average egg area. This
regression demonstrates a positive relationship between average larval length and average
egg area (Figure 7 Linear Regression R2 = 0.07 F1, 68 = 4.89 P = 0.03). Therefore, egg
size is predictive of larval length.
Female Mate Choice
High food males were expected to obtain more mates given the prediction that
these males would be in better condition. Females did not favor mating with one male
food treatment over the other. In week one of the mating trial, high food males were
successful in spawning 70% of the time in comparison to low food males with 60%
During week one, females were not more likely to mate with high food males than low
food males (Chi-square X2 = 1.09 P > 0.1). The same results are found when considering
week two. The probability of mating with high food verses low food males in week two
was not significantly different (Chi-square X2 0.29, v =1, P > 0.5). High food males had
a spawning success rate of 73% while low food males showed similar success with 78%.
On average, mated males were not significantly heavier or longer than unmated
males, nor were they in better condition, had greater fat reserves or expressed more
coloration (Table 6). Results of behavior analyses do indicate however, that female
flagfish prefer to mate with males who exhibit signs of parental care during courtship.
Prior to receiving eggs, mated males on average spent more time defending and fanning
their nests (Figures 8-9 Kruskal-Wallis Test At Nest X2 = 7.29 P = 0.01; Fanning At Nest
X = 5.08 P = 0.02). In addition, mated males chased females more frequently (Figure 10
Kruskal-Wallis Test X2 = 7.19, P = 0.01) than unmated males. Furthermore, the
probability that mated males performed these behaviors was greater than that of unmated
males (Chi-Square At Nest X2 = 4.18 P < 0.05; Fanning At Nest X2 =12.18 P < 0.001;
Chase Frequency X2 =7.59 P < 0.5). The proportion of time spent following a female and
the frequency of cleaning events did not differ between groups (Kruskal-Wallis Test
Following X2 = 0.73 P = 0.39; Cleaning Frequency X2 = 1.93 P = 0.16) nor did the
probability with which these behaviors were performed (Chi-Square Following X2 =0.79
P > 0.05; Cleaning Frequency X2 =2.35 P > 0.5). These results are consistent with
previous studies which demonstrate that female flagfish show preferences for increased
parental care during courtship (St. Mary & Lindstrom in prep; Hale & St. Mary in prep.).
The amount of time males spend fanning during courtship provides females with an
a priori expectation of the quality of parental care their offspring will receive. An
analysis of covariance, shows a significant influence of egg number (F1,60 = 23.76 P =
0.000) and time spent fanning pre-spawning (F1,60 = 9.20 P = 0.00) on fanning at the nest
post-spawning. A regression of the residuals from an analysis of covariance with fanning
post-spawning as the dependent variable and egg number as a covariate, indicates a
significant relationship between fanning pre-spawning and fanning-post spawning (R2
0.13 F1,61 = 9.45 P = 0.00), suggesting that mates who fan during courtship are also
likely to exhibit the behavior after spawning.
Condition-Dependent Expression of Male Traits
Diet manipulation was expected to generate differences in male condition. As a
result, males were expected to show condition-dependent expression of potentially costly
traits such as courtship behavior and coloration. Of the behaviors measured, fanning the
nest and chasing the female possibly require the greatest amount of energy from the male.
Measures of male condition and fat reserves were unable to explain the variation in the
frequency of chases a male performed during courtship (Table 7). However, a negative
relationship between fanning effort and fat reserves was found when considering only
those males who spent time fanning their nests (Figure 11 Linear Regression % Fat
Weight F 1, 37 = 3.96 P = 0.05). This relationship was not seen however when considering
an additional measure of condition (Linear Regression Weight/Length3 F1, 51 = 0.29 P =
0.60). Measures of male condition and fat reserves were not predictive of blue and red
color expression, nor were they correlated with yellow color expression (Table 7)
Table 1. Description of nest and non-nest directed behaviors recorded for each male
At the Nest
Away From Nest
Located within 6cm of vertical
distance above the nest with part
of the body over the nest.
Located away from the nest.
Nest Directed Behavior
Body is angled downward toward
nest while performing a rapid
A bite at the nest.
Synchronous movement of fish
towards nests. Fish are paired side
by side in close proximity, releasing
eggs as they make direct contact
with the nest.
Non Nest Directed Behavior
Following of female.
Rapid swimming directed towards
Table 2.The effect of male food treatment on morphological traits.
High Food Low Food Test P-
Average ( SE) Average ( SE) Statistic Value
Measures of Body Size
Fresh Weight (g) 1.97 (0.09) 1.69 (0.08) 5.727 0.02*
Standard Length (cm) 3.64 (0.06) 3.46 (0.05) 4.685 0.03*
% Fat Weight / Dry 0.19 (0.01) 0.21 (0.01) 0.521 0.47*
% Body Area Blue 8.56 (0.00) 5.06 (0.00) 6.40 0.01 *
% Body Area Red 0.04 (0.00) 0.05 (0.00) 0.28 0.60 *
% Body Area Yellow 0.00 (0.00) 0.00 (0.00) 0.74 0.39 t
Intensity Blue 3.54 (0.14) 3.50 (0.16) 0.03 0.86*
Intensity Red 2.36 (0.05) 2.29 (0.05) 0.43 0.51 *
Intensity Yellow 1.33 (0.10) 1.02 (0.11) 5.07 0.02 t
*ANOVA (two-tailed) with F Test Statistic
t Kruskal- Wallis Test with X2 Test Statistic
Table 3. The effect of male food treatment on measures of courtship prior to spawning.
Statistical analyses were used to determine differences between high and low
food males for 1) the % of observation time a behavior was performed
(Kruskal-Wallis Test) 2) the frequency with which a behavior was performed
(Kruskal-Wallis Test) and 3) the probability that a behavior was performed
Test Statistic P- Value
Test Statistic P- Value
Fanning At Nest
Table 4. The effect of male food treatment on female reproductive allocation. Measures
include hatching success, average egg size, average larval length, and average
egg energetic content.
% Hatched Larvae
Egg Area (mm2)
Larval Length (mm)
Egg Energy (J)
Average ( SE)
Average ( SE)
*ANOVA (two-tailed) with F Test Statistic
t Kruskal- Wallis Test with X2 Test Statistic
Table 5. Results of analyses addressing variation in average egg area and larval length in
response to male morphological and behavioral traits.
Egg Larval Egg
Area (mm) Length (mm) Energetics (J)
Test P Test P Test P
Statistic Value Statistic Value Statistic Value
Male Weight/ 3.14
FemaleWeight / 0.28
% Body Area Blue 1.79
% Body Area Red 3.28
% Body Area 0.24
Intensity Blue 0.22
Intensity Red 0.10
Intensity Yellow 0.10
At Nest 0.03
Fanning At Nest -0.25
* ANCOVA with F Test Statistic
8 Spearman's Rank Nonparametric
0.00* 0.59 0.45
0.69* 2.97 0.09
Correlation (two-tailed) with Correlation Coefficient
Table 6. Differences between mated and unmated males in measures of male body size,
fat reserves, condition factor, and coloration.
Mated Unmated Test P- Value
Average ( SE) Average ( SE) Statistic
Measures of Body Size
Fresh Weight (g) 1.81 (0.08) 1.89 (0.11) 0.77 0.38*
Standard Length (cm) 3.54 (0.05) 3.58 (0.07) 0.25 0.62*
% Fat Weight / Dry 1.87 0.17*
Weight 0.19 (0.00) 0.21 (0.00)
Weight/ Length3 0.04 (0.00) 0.04 (0.00) 1.78 0.18*
% Body Area Blue 0.04 (0.02) 0.07 (0.03) 0.00 0.99*
% Body Area Red 0.04 (0.01) 0.05 (0.02) 1.31 0.25t
% Body Area Yellow 0.00 (0.00) 0.00 (0.00) 1.56 0.21*
Intensity Blue 3.52 (0.12) 3.51 (0.20) 0.00 0.96*
Intensity Red 2.32 (0.07) 2.34 (0.07) 0.06 0.81t
Intensity Yellow 1.24 (0.14) 1.32 (0.14) 0.51 0.47*
*ANOVA (two-tailed) with F Test Statistic
t Kruskal- Wallis Test with X2 Test Statistic
Table 7. Results of condition-dependent expression analyses.
Weight / Length3 % Fat Weight / Dry Weight
Test Statistic P-Value Test Statistic P-Value
Measures of Coloration
% Body Area Blue 0.07 0.79 0.08 0.78
% Body Area Red 3.01 0.08 0.06 0.80
% Body Area Yellow -0.07 0.38 -0.01 0.90O
Intensity Blue 0.15 0.70 0.08 0.30
Intensity Red 0.15 0.70 0.85 0.36
Intensity Yellow -0.07 0.37 -0.09 0.36
Measures of Behavior
Fanning At Nest 0.29 0.60 3.96 0.05
Chasing Frequency 0.80 0.38 2.38 0.13
% Spearman's Rank Nonparametric Correlation (two-tailed) with Correlation Coefficient
"Linear Regression with F Test Statistic
O High Food
A Low Food
I I I I I
2.00 3.00 4.00 5.00 6.00
Standard Length (cm)
Figure 1. The relationship between fresh weight and standard length is shown for both
high food and low food males.
0 High Food
A Low Food
I I I I I
2.00 3.00 4.00 5.00 6.00
Standard Length (cm)
Figure 2. The relationship between dry weight and standard length is shown for both high
food and low food males.
0 High Food
A Low Food
Mt A'- AO 0
I I I I I
2.00 3.00 4.00 5.00 6.00
Standard Length (cm)
Figure 3. The relationship between fat weight and standard length is shown for both high
food and low food males.
R2 = 0.05
0 0 0 -
0% 0 00 0
0 @ 00 0
0 0 0
0 0o o
Male Condition (Fresh Weight (g) / Standard Length (cm))3
Figure 4. Values of male condition are plotted against residuals from a univariate analysis
of co-variance in which average egg area was used as the dependent variable
and % red coloration of the sire as a covariate .
R2 = 0.22
I I I I I I I
3.00 3.50 4.00 4.50 5.00 5.50 6.00
Average Hatch Date (Days)
Figure 5. Average larval length (mm), calculated as the average total length of individual
larvae within a clutch is plotted against the average hatch date for that clutch.
Average hatch date was calculated as the sum of the # of individual larvae
times their respective hatch dates divided by the total number of hatched
0.75- R2 = 0.15
E 0 0
> o o
-, 0 0 0
a, o o8; o
(D 0.25- o 0
(S0 0 0
> o 0 0O0
< -0.25- o
0.03 0.04 0.04 0.04 0.05
Male Condition (Fresh Weight (g) / Standard Length (cm) )3
Figure 6. Values of male condition are plotted against residuals from a univariate analysis
of co-variance in which average larval length was used as the dependent
variable and average hatch date as a covariate.
0 0 0
o' o o
Average Egg Area (mm2)
Figure 7. Individual egg areas were measured within a clutch and then averaged. These
values are plotted against residuals from a univariate analysis of co- variance
in which average larval length was used as the dependent variable and average
hatch date as a covariate.
Figure 8. Average proportion of total observation time each male spent at their nest(
SE) for mated and unmated males prior to receiving eggs.
Figure 9. Average proportion of time spent at the nest fanning (+ SE) for mated and
unmated males prior to receiving eggs.
Figure 10. Frequency of chases ( SE) performed by mated and unmated males during
their recorded observation time prior to receiving eggs. Frequency of chases
was calculated as the number of chases performed divided by the total
Figure 10. Frequency of chases (+ SE) performed by mated and unmated males during
their recorded observation time prior to receiving eggs. Frequency of chases
was calculated as the number of chases performed divided by the total
0.70- R2 = 0.10
S0.50- 0 o o
uo ^ o o o o
4- o o 0 o
0 0 0
0.00 0.10 0.20 0.30 0.40 0.50 0.60
% Fat Weight (g) / Dry Weight (g)
Figure 11. Relationship between fanning and fat reserves. Transformed values are show.
The current study attempted to influence male condition through diet. This
manipulation was expected to affect measures of body size as well as fat reserves. Male
food treatment had a significant effect on male weight and standard length, as also
previously found when a comparable HF/LF diet was implemented (Klug & St. Mary
2005). However, diet manipulation did not influence measures of condition. High food
males maintained their body composition, as weight did not increase more sharply with
body length. Additional nutrients were utilized for growth as opposed to accumulating in
stored fat reserves. This may suggest that body size plays a valuable role in the flagfish
Males may prioritize energy allocation to increased size with the aim of acquiring
and defending breeding territories essential for reproductive success. In nature flagfish
often aggregate and reproductive sites often occur in close proximity (R Hale personal
observation). Hence, high density conditions are likely. Larger males will be more
effective at excluding potential competitors and thus able to spawn more without
disruption, resulting in selection for males to invest in growth. Additionally, female
preference for larger dominant males may further enhance these selective pressures.
In the current study, variation in egg number among males was not related to male
size. Nor were mated males found to be of larger size. However, these results may be
influenced by the single male presentation. Female preferences in other species have been
shown to change in response to their social environment (Kvarnemo et al. 1995; Forsgren
et al. 1996; Kangas & Lindstrom 2001). St. Mary and Lindstrom (unpublished data)
suggest a similar response in flagfish as male traits associated with competitive ability,
such as body size and aggressiveness (frequency of chases towards males) became more
important in mating and egg allocation decisions as the level of male-male competition
increased from the absence of competition (single male and single female) to moderate
(two males and two females) and high (four males and one female) levels.
Males were predicted to show condition-dependent expression of the traits
potentially used in mate choice. Measures of color expression were expected to be
positively correlated with a male's physical condition. Of the three colors measured, red
and yellow were specifically expected to be honest indicators of condition as these have
been demonstrated to be carotenoid based in several fish systems (Kodric-Brown 1998).
Since carotenoids cannot be synthesized, they must be acquired through the diet (Olson &
Owens 1998). As sources are scarce in nature, the intensity of carotenoid based
coloration should reflect the content of a male's diet and his overall foraging ability
(Kodric-Brown 1989). In addition, carotenoids function to support the immune system.
Recruitment of carotenoids to non-recoverable forms in the dead tissue of scales may
compromise a male's overall health, increasing his susceptibility to pathogens or
parasites (Folstad & Karter 1992). Only males in superior condition should be able to
meet this expense. Carotenoid expression should therefore provide a dependable basis for
females to choose males in good condition (Nicoletto 1993).
In the current study, measures of male condition were unable to explain the
variation in color that existed between males. However, diet manipulation was successful
in influencing aspects of color expression. Diet influenced the proportion of the body
covered with blue coloration as well as the intensity of yellow coloration, as high food
males showed greater expression in each case. While, reds, oranges, and yellows are
generally thought to represent carotenoid expression, pale purples, greens and blues can
be produced when carotenoids are bound to proteins (Olson & Owens 1998.) Algae are
the primary water-based sources of carotenoids for fish, in addition to aquatic
invertebrates (Olson & Owens 1998). High food males were fed these sources in
abundance, suggesting that blue pigments may be carotenoid based. Alternatively, blue
coloration has been found to be mediated by the expansion and contraction of
melanophores in other fish systems (Kodric-Brown 1996). Measures of red coloration
were not affected by diet manipulation, suggesting that these pigments may originate
from other sources. Red, yellow, and orange coloration are also produced by pteridine
pigments derived from purines which are synthesized from carbohydrates and proteins
(Hurst 1980). Pteridines have been found to contribute to sexual coloration in poeciliid
fishes as orange spots of male guppies were found to contain red pteridine pigments in
addition to carotenoids (Grether et al. 2001).
In light of these results, it seems that color expression may well be diet dependent.
However, it is not a simple reflection of fat reserves for instance. For example, previous
research in flagfish has found a significant effect of condition (measured as the residuals
from a regression of log weight to log length) on variance in male color expression in
response to temperature and salinity treatments (St. Mary et al. 2001), suggesting that
environmental stresses play a role in color expression. Furthermore, Johnstone (1995)
indicates that color expression is typically influenced by both measures of nutritional
status and parasite load. Parasites have been found to impede the uptake of carotenoids
from the gut, limiting the development of color expression in fish (Milinski & Bakker
1990; Houde & Torrio 1992). This measure would be of interest to investigate in future
research on male color expression in flagfish.
Measures of parental care and courtship behavior were also expected to show
condition-dependent expression. Courtship has often been shown to be expensive
(Vehrencamp et al. 1989; Hoglund et al. 1992) requiring considerable energy reserves to
be maintained over time. Thus, a positive relationship was expected between condition
and the frequency and/or time spent performing energetically expensive behaviors such
as fanning and chasing. No such relationship was found with regard to chasing. However,
counter to our expectations fanning was negatively correlated with fat reserves. This loss
in condition could be due to energy expenditure lost through displaying. Alternatively,
this may suggest that males in poor condition increased their signaling effort at the
expense of future mating as has been demonstrated in sticklebacks (Candolin 1999).
Furthermore, males in low condition may be able to adjust the amount of energy
they invest into courtship depending on the social context. For example, Candolin (2000)
demonstrated that male sticklebacks increased red coloration in the presence of females
and the absence of males when the cost of cheating was low in terms of harassment by
dominant males. Male condition may play a more important role in flagfish under
different social contexts as male-male competition may ensure honest signaling as also
suggested in sand gobies (Svensson & Forsgren 2003).
Females were expected to allocate reproductive resources in response to diet -
dependent differences among males. However, high and low food males were not found
to differ significantly in many of the traits measured. Regardless, females were expected
to use variation in morphological characters among males in egg allocation decisions. No
such patterns were found. The lack of significant results regarding female egg allocation
could be explained by the occurrence of filial cannibalism as male flagfish are known to
consume their eggs (Klug & St. Mary 2005). However, spawning mats were checked at a
minimum of four times a day to prevent filial cannibalism from influencing egg counts.
In fish, egg size has been found to be positively associated with larval size and
hence offspring survival (Ware 1975; Marsh 1986; Pepin 1991; Einum & Fleming 1999).
This pattern is also supported from results of the current study where larval length
increased with egg area. Thus egg size may be one valuable way in which females benefit
from increased investment. Counter to our expectations, males of a lower weight for a
given length fathered larger larvae. Although, the current study found no significant
effect of male condition on egg size, a negative pattern was apparent. These results
suggest that females may allocate larger eggs with more resources to males in poor
Granted, male condition had no influence on egg energy, differential investment
can not be excluded as resource variation may exist that was not detectable by the general
methods used in the current study. For example, the direct transfer of proteins
(hormones), immune factors, and mRNA to eggs and larvae has been demonstrated in
oviparous fishes (Heath & Blouw 1998) and in some cases has been correlated with
increased larval size and survival (Ayson & Lam 1993). Yolk may be an additional
resource to measure directly in future research as initial yolk volume was found to be
correlated with larval length at hatching in capelin, Mallotus villosus (Chambers et al.
1989). Furthermore, Kolm & Olsson (2003) propose that female Banggai cardinalfish,
Pterapogon kauderni, re-direct resources into eggs extremely close to spawning,
suggesting that females may have more control over egg investment than previously
Initially, one may interpret these unexpected results as the female's adaptive
response to low quality males. Females mating with unattractive males have been found
to increase maternal investment to enhance offspring health as a result of a decrease in
viability associated with low quality fathers (Bluhm & Gowaty 2004). For example, in
collared flycatchers, females have been found to compensate for low quality parental care
of young sires by increasing yolk testosterone concentrations which boosts begging
activity (Michl et al. 2004) and in barn swallows, females manipulated egg carotenoid
levels in response to unattractive males whose offspring may have greater exposure to
parasites (Saino et al. 2002b).
However in the current study, males with low fat reserves spent more time fanning
their nests during courtship, suggesting that females allocate larger eggs and hence larger
larvae to those males who are likely to provide better quality parental care and hence
possibly greater survivorship of offspring. The proportion of time spent fanning the nest
during courtship and when eggs were received was highly correlated, suggesting that pre-
spawning behavior indicates the quality of parental care a female may expect post-
spawning. Alternatively, paternal effects can not be ruled out as a source of these
unexpected results. Genetic differences among males may have influenced larval length
as has been demonstrated in other studies which contribute offspring size and growth
rates to paternal effects (Reynolds & Gross 1992).
Consistent with the "good parent process" of sexual selection which argues that
parental care functions as a cue in mate choice (Hoelzer 1989), female flagfish also
responded to variation in fanning during courtship in addition to nest defense when
selecting mates. The direct benefits associated with parental care behaviors have been
investigated in the flagfish as nest defense has been shown to increase egg survivorship
(Klug et al. 2005) and fanning is hypothesized to reduce fungal infection (St. Mary et al.
2001; 2004) in addition to aiding in the replacement of oxygen. Hence female flagfish
select mates that provide better parental care and offer increased offspring survivorship as
has been demonstrated in other fish species (Forsgren 1997; Ostlund, & Ahnesjo 1998)
In summary, the DAH suggests that the attractiveness of a mate influences the
value of the breeding attempt. Females should allocate more resources when mating with
an attractive male as the potential for direct and/or indirect benefits received from
reproduction are greater. Measures of attractiveness such as body size, condition, fat
reserves, and color expression did not influence female mating and egg allocation
decisions. However, male condition had a significant affect on larval length, suggesting
that females produce eggs that produce larger larvae when mating with males in lower
condition. Counter to our expectations, males with low energy reserves fanned more pre
and post spawning suggesting that females respond to variation in parental care by
investing more in reproduction.
DAH has important implications for the field of sexual selection and should be
studied further as it potentially poses a problem for studies that support "good genes"
models of sexual selection. If maternal effects have not been accounted for, evidence for
increased survival of offspring fathered by attractive males may be attributable to
differential female investment rather than paternal genetic effects. Evidence for
differential allocation will improve our understanding of all factors which affect offspring
fitness. Furthermore, if conditions offspring experience during development influences
the expression of sexually selected characters in adulthood, differential allocation could
have a strong influence on the direction of sexual selected traits within a given species
(Qvarnstrom & Price 2001).
Despite support for differential allocation in a variety of taxa, experimental tests
of the hypothesis in fish is lacking besides those of Kolm 2001, Kolm & Olsson 2003 and
the current study. Given the prevalence of male parental care in fishes (Gross & Sargent
1985) and the importance of perceived benefits in differential allocation theory, future
research should aim to expand the number offish species in which this topic is
LIST OF REFERENCES
Andersson, M. 1994. Sexual Selection. Princeton, New Jersey: Princeton University
Ayson, F.G., & Lam, T.J. 1993. Thyroxine injections of female rabbitfish (Siganus
guttatus) broodstock; Changes in thyroid hormonal levels in plasma, eggs, and
yolk-sac larvae, and its effect on larval growth and survival. Aquaculture, 109, 83-
Bluhm, C.K., & Gowaty, P. 2004. Reproductive compensation for offspring viability
deficits by female mallards, Anasplatyrhynchos. Animal Behaviour, 68, 985-992.
Burley, N. 1986. Sexual selection for aesthetic traits in species with biparental care.
American Naturalist, 127, 415-445.
Burley, N. 1988. The differential-allocation hypothesis: An experimental test. American
Naturalist, 132, 611-628.
Bolger, T., & Connolly, P.L. 1989. The selection of suitable indices for the measurement
and analysis of fish condition. Journal ofFish Biology, 26, 171-182.
Boschung, H.T. Jr., Williams, J.D., Gotshall, D.W., Caldwell, M.C., & Caldwell, D.K.
2000. The Audubon Society Field Guide to North American Fishes, Whales and
Dolphins. Chanticleer, New York.
Candolin, U. 1999. The relationship between signal quality and physical condition: is
sexual signalling honest in the three-spined stickleback? Animal Behaviour, 58,
Candolin, U. 2000. Male-male competition ensures honest signaling of male parental
ability in the three spined stickleback (Gasterosteus aculeatus). Behavioral
Ecology and Sociobiology, 49, 57-61.
Campanella, P.J., & Wolf, L.L. 1974. Temporal leks as a mating system in a temperate
zone dragonfly (Odonata: Anisoptera). I. Plathemis lydia (Drury). Behaviour, 51,
Chambers, R.C., Leggett, W.C., & Brown, J.A. 1989. Egg size, female effects, and the
correlations between early life-history traits of capelin, Mallotus villosus: An
appraisal at the individual level. Fish Bulletin, 87, 515-523.
Cunningham, E.J.A., & Russell, A.F. 2000. Egg investment is influenced by male
attractiveness in the mallard. Nature, 404, 74-76.
Darwin, C. 1871. The Descent ofMan and Selection in Relation to Sex. Murray, London.
Einum, S., & Fleming, I.A. 1999. Maternal effects of egg size in brown trout (Salmo
trutta): Norms of reaction to environmental quality. Proceedings of the Royal
Society of London Series B, 266, 2095-2100.
Edvardsson, M., & Arnqvist, G. 2005. The effects of copulatory courtship on differential
allocation in the red flour beetle, Tribolium castaneum. Journal ofInsect Behavior,
Folstad, I., & Karter, A. J. 1992. Parasites, bright males, and the immunocompetence
handicap. American Naturalist, 139, 603-622.
Forsgren, E. 1997. Female sand gobies prefer good fathers over dominant males.
Proceedings of the Royal Society of London Series B, 264, 1283-1286.
Forsgren, E., Kvarnemo, C., & Lindstrom, K. 1996. Mode of sexual selection determined
by resource abundance in two sand goby populations. Evolution, 50, 646-654.
Foster, N.R., Cairns, J. Jr., & Kaesler, R.L. 1969. The flagfish, Jordanellafloridae, as a
laboratory animal for behavioral bioassay studies. Proceedings of the Academy of
Natural Sciences Philadelphia, 121, 129-152.
Gil, D., Graves, J., Hazon, N., & Wells, A. 1999. Male attractiveness and differential
testosterone investment in zebra finch eggs. Science, 286, 126-128.
Grafen, A. 1990. Biological signals as handicaps. Journal of Theoretical Biology, 144,
Grether, G.F., Hudon, J., & Endler, J.A. 2001. Carotenoid scarcity, synthetic pteridine
pigments and the evolution of sexual coloration in guppies (Poecilia reticulata).
Proceedings of the Royal Society ofLondon, Series B, 268, 1245-1253.
Gross, M.R., & Sargent, R.C. 1985. The evolution of male and female parental care in
fishes. American Zoologist, 25, 807-822.
Gwinner, H. & Schwabl, H. 2005. Evidence for sexy sons in European starlings (Sturnus
vulgaris). Behavioral Ecology and Sociobiology, 58, 375-382.
Hamilton,W.D., & Zuk, M. 1982. Heritable true fitness and bright birds: A role for
parasites? Science, 218, 384-386.
Heath, D.D., & Blouw, D.M. 1998. Are maternal effects in fish adaptive or merely
physiological side effects? In Maternal Effects as Adaptations (Mousseau, T.A., &
Fox, C.W. eds) pp 178-201. New York: Oxford University Press.
Hoelzer, G.A. 1989. The good parent process of sexual selection. AnimalBehaviour, 38,
Hoglund, J., Kdlds, J.A., & Fiske, P. 1992. The costs of secondary sexual characters in
the lekking great snipe. (Gallinago media). Behavioral Ecology and Sociobiology,
Holm, C.H. 1973. Breeding sex ratios, territoriality, and reproductive success in the red-
winged blackbird (Agelaiusphoeniceus). Ecology, 54, 356-365.
Houde, A.E., & Torio, A.J. 1992. Effect of parasitic infection on male colour pattern and
female choice in guppies. Behavioral Ecology, 3, 346-351.
Hurst, D.T. 1980. An Introduction to the Chemistry and Biochemistry ofPyrimidines,
Purines and Pteridines. New York: John Wiley.
Johnstone, R.A. 1995. Sexual selection: Honest advertisement and the handicap principle:
Reviewing the evidence. Biological Reviews, 70, 1- 65.
Kangas, N., & Lindstrom, K. 2001. Male interactions and female mate choice in the sand
goby, Pomatoschistus minutus. Animal Behaviour, 61, 425-430.
Kempenaers, B., Verheyen, G.R., & Dhondt, A.A. 1997. Extra-pair paternity in the blue
tit (Parus caeruleus): Female choice, male characteristics, and offspring quality.
Behavioral Ecology, 8, 481-492.
Klug, H., & St.Mary, C.M. 2005. Reproductive fitness consequences of filial cannibalism
in the flagfish, Jordanellafloridae. Animal Behaviour, 70, 685-691.
Klug, H., Chin, A., & St. Mary, C.M. 2005. The net effects of guarding on egg
survivorship in the flagfish, Jordanellafloridae. Animal Behaviour, 69, 661-668.
Knapp, R.A. 1995. Influence of energy reserves on the expression of a secondary sexual
trait in male bicolor damselfish, Stegastes partitus. Bulletin ofMarine Science, 57,
Knapp, R.A., & Kovach, J.T. 1991. Courtship as an honest indicator of male parental
quality in the bicolor damselfish, Stegastespartitus. Behavioral Ecology, 2, 295-
Kodric-Brown, A. 1989. Dietary carotenoids and male mating success in the guppy; and
environmental component to female choice. Behavioral Ecology and Sociobiology
Kodric-Brown, A. 1996. Role of male-male competition and female choice in the
development of breeding coloration in pupfish (Cyprinodonpecosensis).
Behavioral Ecology, 7, 431-437.
Kodric-Brown, A. 1998. Sexual dichromatism and temporary color changes in the
reproduction of fishes. American Zoology, 38, 70-81.
Kodric-Brown, A., & Brown, J.J. 1984. Truth in advertising: The kinds of traits favored
by sexual selection. American Naturalist, 124, 309-323.
Kodric-Brown, A. & Nicoletto, P.F. 1993. The relationship between physical condition
and social status in pupfish, Cyprinodonpecosensis. Animal Behaviour, 46, 1234-
Kolm, N. 2001. Females produce larger eggs for large males in a paternal mouth
brooding fish. Proceedings of the Royal Society of London Series B, 268, 2229-
Kolm, N., & Olsson, J. 2003. Differential investment in the Banggai cardinalfish: Can
females adjust egg size close to egg maturation to match the attractiveness of a new
partner? Journal ofFish Biology, 63, 144-151.
Kvarnemo, C., Forsgren, E., & Magnhagen, C. 1995. Effects of sex ratio on intra- and
inter sexual behavior in sand gobies. Animal Behaviour, 50, 1455-1461.
Lindstrom, K. 1998. Energetic constraints on mating performance in the sand goby.
Behavioral Ecology, 9, 297-300.
Marsh, E. 1986. Effects of egg size on offspring fitness and maternal fecundity in the
orange-throat darter, Eihevi,\ t,uai spectabile (Pisces: Percidae). Copeia, 18-30.
Mertz, J.C., & Barlow, G.W. 1966. On the reproductive behavior of Jordanellafloridae
(Pisces: Cyprinodontidae) with special reference to a quantitative analysis of
parental fanning. Zeitschrift fir Tierpsychologie. 23, 537-554.
McEdward, L.R., & Coulter, L.K. 1987. Egg volume and energetic content are not
correlated among sibling offspring of starfish: Implications for life-history theory.
Evolution, 41, 914-917.
Michl, G., Toirk, J., Peczely, P., Garamszegi, L.Z., & Schwabl, H. 2004. Female collared
flycatchers adjust yolk testosterone to male age, but not to attractiveness.
Behavioral Ecology, 16, 383-388.
Milinski, M., & Bakker, C.M. 1990. Female sticklebacks use male coloration in mate
choice and hence avoid parasitized males. Nature, 344, 330-333.
Moller, A.P. 1990. Effects of a haematophagous mite on the barn swallow (Hirundo
Rustica): A test of the Hamilton and Zuk hypothesis. Evolution, 44, 771-784.
Nicoletto, P.F. 1993. Female sexual response to condition-dependent ornaments in the
guppy, Poecilia reticulata. Animal Behaviour, 46, 441-450.
Nisbet, I.C.T. 1973. Courtship-feeding, egg-size, and breeding success in common terns.
Nature, 241, 141-142.
Norris, K. 1990. Female choice and the quality of parental care in the great tit, Parus
major. Behavioral Ecology and Sociobiology, 27, 275-281.
Norris, K. 1993. Heritable variation in a plumage indicator of viability in male great tits,
Parus major. Nature, 362, 537-539.
Olson, V.A., & Owens, I.P.F. 1998. Costly sexual signals: Are carotenoids rare, risky or
required? Trends in Evolution and Ecology, 13, 510-514.
Ostlund, S., & Ahnesjo, I. 1998. Female fifteen-spined sticklebacks prefer better fathers.
Animal Behaviour, 56, 1177-1183.
Page, L. M., & Burr, B.M. 1991. A Field Guide to Freshwater Fishes: North America
North of Mexico. New York: Houghton Mifflin Company.
Parker, T.H. 2003. Genetic benefits of mate choice separated from differential maternal
investment in red junglefowl (Gallus gallus). Evolution, 57, 2157-2165.
Petrie, M. 1983. Female moorhens compete for small fat males. Science, 220, 413- 415.
Petrie, M. 1994. Improved growth and survival of offspring of peacocks with more
elaborate trains. Nature, 371, 598-599.
Petrie, M., & Williams, A. 1993. Peahens lay more eggs for peacocks with larger trains.
Proceedings of the Royal Society of London Series B, 251, 127-131.
Pepin, P. 1991. Effects of temperature and size on development, mortality, and survival
rates of the pelagic early life-history of marine fish. Canadian Journal ofFisheries
andAquatic Science, 48, 503-518.
Qvarnstrom, A., & Price, T.D. 2001. Maternal effects, paternal effects and sexual
selection. Trends in Ecology and Evolution, 16, 95-100.
Reyer, H.U., Frei, G., & Som, C. 1999. Cryptic female choice: frogs reduce clutch size
when amplexed by undesired males. Proceedings of the Royal Society of London
Series B, 266, 2101-2107.
Reynolds, J.D., & Gross, M.R. 1992. Female mate preference enhances offspring growth
and reproduction in a fish, Poecilia reticulata. Proceedings of the Royal Society of
London Series B, 250, 57-62.
Ridgway, M.S., & Shuter, B.J. 1994. The effects of supplemental food on reproduction in
parental male smallmouth bass. Environmental Biology ofFishes, 39, 201-207.
Saino, N., Paola Ferrari, R., Martinelli, R., Romano, M., Rubolini, D., & Moller, A.P.
2002a. Early maternal effects mediated by immunity depend on sexual
ornamentation of the male partner. Proceedings of the Royal Society of London
Series B, 269, 1005-1009.
Saino, N., Bertacche, V., Paola Ferrari, R., Martinelli, R., Pape Moller, A., & Stradi, R.
2002b. Carotenoid concentration in barn swallow eggs is influenced by laying
order, maternal infection and paternal ornamentation. Proceedings of the Royal
Society of London Series B, 269, 1729-1733.
Sheldon, B.C., 2000. Differential allocation; Tests, mechanisms, and implications. Trends
in Ecology and Evolution, 15, 397-401.
Sheldon, B.C., Andersson, S., Griffith, S.C., Ornborg, J., & Sendecka, J. 1999.
Ultraviolet colour variation influences blue tit sex ratios. Nature, 874-877.
Smith, W.E. 1973. A Cypriodontid fish, Jordanellafloridae, as a laboratory animal for
rapid chronic bioassays. Journal of the Fisheries Research Board of Canada, 30,
Smith, C., & Wootton, R. J. 1999. Parental energy expenditure of the male three-spined
stickleback. Journal ofFish Biology, 54, 1132-1136.
St. Mary, C.M., Noureddine, C.G., & Lindstrom, K. 2001. Environmental effects on male
reproductive success and parental care in the Florida flagfish Jordanellafloridae.
Ethology, 107, 1035-1052.
St. Mary, C.M., Gordon, E., & Hale, R.E. 2004. Environmental effects on egg
development and hatching success in Jordanellafloridae, a species with parental
care. Journal ofFish Biology, 65, 1-9.
Svensson, O., & Forsgren, E. 2003. Male mating success in relation to food availability in
the common goby. Journal ofFish Biology, 62, 1217-1221.
Szentirmai, I., Komdeur, J., & Szekely, T. 2005. What makes a nest-building male
successful? Male behavior and female care in penduline tits. Behavioral Ecology,
Thomhill, R. 1976. Sexual selection and nuptial feeding behaviour in Bittacus apicalis
(Insecta, Mecoptera). American Naturalist, 110, 529-548.
Vehrencamp, S.L., Bradbury, J.W., & Gibson, R.M. 1989. The energetic cost of display
in male sage grouse. Animal Behaviour, 38, 885-896.
Ware, D.M. 1975. Relation between egg size, growth, and natural mortality of larval fish.
Journal of the Fisheries Research Board of Canada, 32, 33-41.
Wiggins, D.A., & Morris, R.D. 1986. Criteria for female choice of mates: Courtship
feeding and parental care in the common tern. American Naturalist, 128, 126-129.
Zahavi, A. 1975. Mate selection- a selection for a handicap. Journal of Theoretical
Biology, 53, 205-214.
In June of 2003 I received a Bachelor of Science degree from Ohio State
University. As an undergraduate I was able to focus on three major research projects in
various disciplines of zoology. These projects occurred in the areas of neuroethology,
pharmacology, aquatic ecology, and mating systems.
I began at the Rothenbuhler Honey Bee Lab, assisting in research aimed at
understanding the role of nitric oxide in the feeding behavior of the adult sphinx moth,
Manduca sexta. In April of 2002, I presented this research at the 24th Annual Meeting for
the Association of Chemoreception Sciences in Sarasota, Florida. In May of 2002, I
continued to present this research at several undergraduate research symposiums. As a
result, I received the Arts and Sciences Award for Excellence in Scholarship and was
given the opportunity to travel abroad in November of 2002 to attend the University of
Sao Paolo's Symposium of Undergraduate Research in Brazil.
Following this experience, I worked as a graduate student assistant at the Aquatic
Ecology Lab. I gained invaluable field experience working in Ohio's reservoirs, aiding
in research that sought to determine the impact of stocked Saugeye on the abundance of
their primary prey, the Gizzard Shad. I became familiar with techniques used to collect
larval and adult fish, in addition to methods used for ageing fish.
Throughout my undergraduate years I volunteered in the lab of Jerry Downhower.
This research focused on the costs of compromised female choice and the consequences
of estrogen-like compounds on mating behavior in Japanese killifish, Oryzias latipes. It
was from this experience that I became interested in the study of sexual selection. I
decided to pursue my interests further by attending graduate school at the University of
Florida and studying the mating behavior of flagfish, Jordanellafloridae, in the lab of
Colette St. Mary.
As an undergraduate I was grateful to be given the opportunity to participate as a
research assistant in several laboratories. As a graduate student with my own research
program I have extended this opportunity to students at the University of Florida by
mentoring two undergraduates. Each student aided in the collection of data. Furthermore,
over the past eight semesters I have also interacted with the student body by teaching
laboratory courses in introductory biology, evolution and ecology. I now hope to pursue
a career in science education.