Human induced environmental change has led to a decline in biodiversity worldwide.
Freshwater ecosystems are heavily impacted by local land use changes and agricultural
practices, resulting in a wide range of pollutants entering surface waters. These pollutants
can produce a variety of physiological effects, but we know comparatively little about
how organisms respond behaviorally. Behavioral responses to environmental stressors
have been understudied, but behavioral responses can indicate the immediate ecological
consequences of pollutants on populations. Behavioral responses are often first in a
cascade of responses when organisms deal with environmental change. Behavior might
be valuable as a tool for determining the mechanisms of how populations change over
time, and could be used to understand how sensitive species are to environmental change.
We often observe changes in behavior during the presence of environmental change, but
less is known about whether these changes are plastic —giving an individual the ability
to temporarily respond to a stressor and return to baseline behaviors after stressors pass—
or irreversible. In addition, even less is known about the potential for these changes to be
adaptive evolutionary response. My dissertation focused on two key aspects of behavioral
variation: (1) the evolutionary and physiological mechanisms underlying mating
behaviors and (2) how elevated nitrates affect behavioral variation, and if behavioral
responses to nitrate are plastic responses, or evolutionary responses.
4
One behavior that plays a critical role in population growth and stability is female
mate preferences. When a female’s mating preferences is reduced, females mate more
randomly, and the adaptive benefits of optimal mate choice are reduced. In my first
chapter, I examined possibility that stressors could affect mating behaviors by
determining if cortisol was associated with variation in female mating preferences. I
performed this study using a system with established female mate preferences: in the
high-back pygmy swordtail (Xiphophorus multilineatus), females exhibit stronger
preferences for large courter males compared to small sneaker males. Our initial
predictions are based on the possibility that cortisol functions as a stress hormone,
although glucocorticoid hormones, such as cortisol, are not synonymous with stress.
Courtship in live-bearing fishes involves females interacting with male behaviors that are
considered harassment, suggesting that reproductive interactions with males may induce
stress in females. In support of these points, I found that cortisol reduces a female’s mate
preference for optimal (large courter) males. Larger females also had a stronger
preference for courter males, which supports previous work in this system on the
evolution of female mate preferences. We also found that females from a courter lineage
(i.e., courter fathers and grandfathers) had higher cortisol levels than females from a
sneaker lineage (i.e., sneaker fathers and grandfathers). Reduced mating discrimination
might therefore be a behavioral consequence of an environmental stressor, and I follow
up on this idea in Chapters 3 and 4 where I describe how behaviors (including mate
preferences) respond to an additional environmental stressor: nitrates.
5
For my second chapter, I took advantage of long-term breeding mesocosms
maintained by the Morris Fish Lab to describe how female lineage and social
environment impact variation in mating preferences. I found patterns in Chapter 1 where
females exhibited different preferences based on parental lineage, but I wanted to
determine how a female’s social environment also affects mating preferences. These
mesocosms allowed me to address these hypotheses together. I compared the mating
preferences of females from three distinct mesocosms that varied based on the frequency
of each male reproductive tactic in X. multilineatus. Two mesocosms had all courter
males, two had all sneaker males, and two were a 50/50 split. These mesocosms also
allowed me to describe the mating preferences of virgin females, which allows us to
understand how preferences differ based on sexual experience. I found that virgin
females from both courter and sneaker lineages did not differ in their mating preferences,
and that both types of females preferred large, courter males. After sexual experience,
females from a sneaker line exhibited much stronger preferences for courter males, and
females from a courter line did not change their mating preferences. I also discuss some
alternative hypotheses explaining these patterns, specifically about how a female’s
growth rate impacts her mating preferences.
While my first and second chapters focused on describing potential links between
glucocorticoids, mating behaviors, and a female’s reproductive environment, my third
and fourth chapters focus on behavioral responses to environmental pollutants.
Environmental pollutants are deposited into surface waters worldwide at an alarming rate,
and I was interested in describing how these chemicals and nutrients elicit behavioral
6
responses in freshwater fish. I focused on the behavioral responses to nitrates, and how
responses varied across populations and generations. Nitrate is a naturally occurring
nutrient essential for life, but at elevated concentrations causes toxicity in fishes. Nitrate
can oxidize hemoglobin lowering the blood oxygen carry capacity, leading to low levels
of oxygen in vital tissues as well as disruptions to steroid hormone synthesis. However,
we know comparatively little about how and if fishes respond behaviorally to sub-lethal
concentrations of nitrates. In my third chapter, using wild-caught green swordtail fish
(Xiphophorus hellerii) from agriculturally impaired populations to protected and low
human impact populations, I assessed how nitrate at two distinct concentrations (10 mg/L
and 100 mg/L) impacts aggression, boldness, and female mating preferences. I chose
these behaviors because each has been hypothesized to impact survival and reproduction
in live-bearing fish. Then in chapter four, we raised offspring from wild-caught adults
and exposed them to one of the aforementioned concentrations of nitrate (100 mg/L). We
assessed if adults from impaired sites exhibited different behavioral responses than adults
from protected sites, in both their baseline behaviors as well as their response to nitrate.
We assessed if these offspring —reared in a common garden— differed in their baseline
behaviors as well as their responses to nitrate.
In wild caught adults, I found that exposure to high (100 mg/L) concentrations of
nitrate increased aggression and decreased boldness, but low (10 mg/L) nitrate
concentrations decreased another measure of aggression, measured as the number of
lateral displays. I also determined that lab reared offspring were more aggressive and less
bold than their parents when exposed to nitrate, indicating that behavioral variation in
7
response to nitrate shows signatures of a plastic response. Adults from impaired sites
were less aggressive and less bold, while their offspring were more aggressive and less
bold. This result suggests that a reduction in boldness behavior when exposed to nitrates
may be a maternal effect or an adaptive evolutionary response. Further studies of
potential fitness benefits and the precise mechanism behind this change in behavior are
needed. Chapters three and four suggest that swordtail fish have both plastic and
irreversible responses to pollution exposure, some of which may be adaptive evolutionary
responses. Determining which suites of behaviors respond plastically in addition to which
may be adaptive represents a critical avenue of future research on the mechanisms
species use to respond to environmental stressors. Taken together, the chapters of my
dissertation broaden our understanding of how environmental factors induce behavioral
variation, and how behavioral variation mediates organismal responses to environmental
stressors.