Macquarie University
Browse

Variation in mimicry accuracy in ant-mimicking spiders: quantification, evolution, trade-offs and predator perception

Download (8.9 MB)
thesis
posted on 2024-07-26, 02:49 authored by Michael Kelly

The aim of my thesis was to investigate why mimicry in ant-mimicking spiders is so variable, with some mimics closely resembling their ant models while others only possessing a vague, or poor, resemblance to ants. The phenomenon of inaccurate mimicry has been observed in many mimicry systems and the evolution and persistence of mimetic inaccuracy is intensely debated. I addressed this overall question in four data chapters that looked at 1) how mimetic traits are identified, quantified and measured, 2) how these traits evolve over evolutionary time, 3) how mimetic trait expression impacts other critical functions, and 4) how predators perceive and respond to mimetic traits.

In Chapter 1, I reviewed two broad approaches to assessing mimetic accuracy, receiver responses and trait measurements. I collected and described five different methods – three of which entailed various ways of measuring traits (linear morphometrics, geometric morphometrics and biometric trait quantification) and two that used receiver responses (machine learning and human assessment) to assess mimic accuracy. By evaluating and comparing the efficacy and cost of each method I was able to provide practical recommendations on the suitability of each method in the context of the specific research question. This work also provided helpful scripts that are freely available for future mimicry research. Chapter 1 has been published in Animal Behaviour.

In Chapter 2, I tested the idea that inaccurate mimicry represents a transitional stage moving from a non-mimetic phenotype towards an accurate mimetic phenotype (perfecting hypothesis). Focusing on two spider groups, the jumping spider tribe Myrmarachnini (Salticidae) and the sac spider sub-family Castianeirinae (Corinnidae), I measured eight morphological traits and one colour trait in 321 species. I then sequenced ultraconserved elements from those species to reconstruct the molecular phylogeny of the Myrmarachnini and the Castianeirinae. My results suggest that ant mimicry evolved via gradual processes (rather than saltatorial leaps) and in some groups there is support for the perfecting hypothesis, where mimics have evolved towards greater accuracy. Overall, however the inferred global optimum for ant mimics was an inaccurate state with accurate mimicry being highly labile at the macro-evolutionary scale. Of the ten traits measured, prosoma (or cephalothorax) constrictions, which emulate the head and thorax of ants, were among the most stable traits and generally was not lost once this trait evolved.

In Chapter 3, I tested the selection trade-offs hypothesis; that accurate mimetic traits lead to life-history trade-offs, which result in a selective optimum of inaccurate mimicry. Focusing on prosomal constrictions as a significant trait identified in Chapter 2, I tested if they have a negative impact on the central nervous system (CNS) or the venom glands that are located in the spiders’ prosoma. In order to quantify the size of the CNS and venom glands in a pair of closely related jumping spiders – the ant-mimicking Myrmarachne smaragdina (tribe Myrmarachnini) and the non-mimetic Astia hariola (tribe Astiini) – I implemented micro-CT imaging. Successful scans were analysed using segmentation software to quantify the volumes of the CNS, structures within, and the venom glands. As predicted, the ant mimic, with the highly constricted prosoma, had a relatively smaller CNS and venom glands but seemed to be compensating for any potentially negative impacts on cognition through a relatively larger arcuate body that is functionally associated with neural integration. These results suggest that impacts on cognitive and foraging functions might limit the extent of prosomal constrictions, resulting in less accurate mimics.

In Chapter 4, I used predator trials to investigate how predators respond to the absence or presence of prosomal constrictions. Two mutually exclusive hypotheses (the salient trait and the satyric mimicry hypotheses) predict different predator responses. If prosomal constrictions are the most salient information for misidentification (i.e., classifying a spider as an ant), predators are expected to differentially attack prey images with and without constrictions. Alternatively, if the presence of an ant-like trait together with remnants of spider traits create an ambiguous (satyric) signal, then predators are expected to take longer to assess the prey image before attack. Using the jumping spider Servaea incana as predator, I presented animated images of ants, spiders, spiders with prosomal constrictions and controls. While the predator was highly motivated to attack these images, they showed no attack preference or attack delay based on body shape. Therefore, we found no support for either hypothesis.

In conclusion, my thesis results support the conceptual shift toward considering inaccurate mimicry as the norm rather than the exception. Instead, research attention should shift towards explaining why accurate mimics seem to escape prevailing opposing selection processes that maintain seemingly sub-optimal inaccurate mimicry states. I advocate that this shift in research focus, together with more consistent quantification of mimetic traits and the simultaneous testing of multiple alternative hypotheses, will create exciting opportunities to examine the fascinating phenomenon of mimicry.

History

Table of Contents

Thesis introduction -- Chapter 1 - Measuring mimicry: methods for quantifying visual similarity -- Chapter 2 - Dense sampling phylogenomics reveals highly dynamic evolution of Batesian mimicry accuracy in two clades of myrmecomorphic spiders -- Chapter 3 - Do ant mimics have smaller brains? A trade-off between central nervous system volume and body shape mimicry in myrmecomorphic spiders -- Chapter 4 - Do morphological modifications in ant-mimicking spiders reduce predation risk? -- Thesis conclusions -- Appendix A - Ethics approval letter -- Appendix B - Stabilized morphological evolution of spiders despite mosaic changes in foraging ecology

Awarding Institution

Macquarie University

Degree Type

Thesis PhD

Degree

Doctor of Philosophy

Department, Centre or School

School of Natural Sciences

Year of Award

2023

Principal Supervisor

Mariella Herberstein

Additional Supervisor 1

Jonas Wolff

Rights

Copyright: The Author Copyright disclaimer: https://www.mq.edu.au/copyright-disclaimer

Language

English

Extent

177 pages

Usage metrics

    Macquarie University Theses

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC