Aposematic theory has historically predicted that predators should select for warning signals to converge on a single form, as a result of frequency‐dependent learning. However, widespread variation in warning signals is observed across closely related species, populations and, most problematically for evolutionary biologists, among individuals in the same population. Recent research has yielded an increased awareness of this diversity, challenging the paradigm of signal monomorphy in aposematic animals. Here we provide a comprehensive synthesis of these disparate lines of investigation, identifying within them three broad classes of explanation for variation in aposematic warning signals: genetic mechanisms, differences among predators and predator behaviour, and alternative selection pressures upon the signal. The mechanisms producing warning coloration are also important. Detailed studies of the genetic basis of warning signals in some species, most notably Heliconius butterflies, are beginning to shed light on the genetic architecture facilitating or limiting key processes such as the evolution and maintenance of polymorphisms, hybridisation, and speciation. Work on predator behaviour is changing our perception of the predator community as a single homogenous selective agent, emphasising the dynamic nature of predator–prey interactions. Predator variability in a range of factors (e.g. perceptual abilities, tolerance to chemical defences, and individual motivation), suggests that the role of predators is more complicated than previously appreciated. With complex selection regimes at work, polytypisms and polymorphisms may even occur in Müllerian mimicry systems. Meanwhile, phenotypes are often multifunctional, and thus subject to additional biotic and abiotic selection pressures. Some of these selective pressures, primarily sexual selection and thermoregulation, have received considerable attention, while others, such as disease risk and parental effects, offer promising avenues to explore. As well as reviewing the existing evidence from both empirical studies and theoretical modelling, we highlight hypotheses that could benefit from further investigation in aposematic species. Finally by collating known instances of variation in warning signals, we provide a valuable resource for understanding the taxonomic spread of diversity in aposematic signalling and with which to direct future research. A greater appreciation of the extent of variation in aposematic species, and of the selective pressures and constraints which contribute to this once‐paradoxical phenomenon, yields a new perspective for the field of aposematic signalling.

Averting attack by biting flies is increasingly regarded as the evolutionary driver of zebra stripes, although the precise mechanism by which stripes ameliorate attack by ectoparasites is unknown. We examined the behaviour of tabanids (horse flies) in the vicinity of captive plains zebras and uniformly coloured domestic horses living on a horse farm in Britain. Observations showed that fewer tabanids landed on zebras than on horses per unit time, although rates of tabanid circling around or briefly touching zebra and horse pelage did not differ. In an experiment in which horses sequentially wore cloth coats of different colours, those wearing a striped pattern suffered far lower rates of tabanid touching and landing on coats than the same horses wearing black or white, yet there were no differences in attack rates to their naked heads. In separate, detailed video analyses, tabanids approached zebras faster and failed to decelerate before contacting zebras, and proportionately more tabanids simply touched rather than landed on zebra pelage in comparison to horses. Taken together, these findings indicate that, up close, striped surfaces prevented flies from making a controlled landing but did not influence tabanid behaviour at a distance. To counteract flies, zebras swished their tails and ran away from fly nuisance whereas horses showed higher rates of skin twitching. As a consequence of zebras’ striping, very few tabanids successfully landed on zebras and, as a result of zebras’ changeable behaviour, few stayed a long time, or probed for blood.

The nighttime environment is being altered rapidly over large areas worldwide through introduction of artificial lighting, from streetlights and other sources. This is predicted to impact the visual ecology of many organisms, affecting both their intra- and interspecific interactions. Here, we show the effects of different artificial light sources on multiple aspects of hawkmoth visual ecology, including their perception of floral signals for pollination, the potential for intraspecific sexual signalling, and the effectiveness of their visual defences against avian predators. Light sources fall into three broad categories: some that prevent use of chromatic signals for these behaviours, others that more closely mimic natural lighting conditions, and, finally, types whose effects vary with light intensity and signal colour. We find that Phosphor Converted (PC) amber LED lighting – often suggested to be less harmful to nocturnal insects – falls into this third disruptive group, with unpredictable consequences for insect visual ecology depending on distance from the light source and the colour of the objects viewed. The diversity of impacts of artificial lighting on hawkmoth visual ecology alone argues for a nuanced approach to outdoor lighting in environmentally sensitive areas, employing intensities and spectra designed to limit those effects of most significant concern.

Many defended species use conspicuous visual warning signals to deter potential predators from attacking. Traditional theory holds that these signals should converge on similar forms, yet variation in visual traits and the levels of defensive chemicals is common, both within and between species. It is currently unclear how the strength of signals and potency of defences might be related: conflicting theories suggest that aposematic signals should be quantitatively honest, or, in contrast, that investment in one component should be prioritized over the other, while empirical tests have yielded contrasting results. Here, we advance this debate by examining the relationship between defensive chemicals and signal properties in a family of aposematic Lepidoptera, accounting for phylogenetic relationships and quantifying coloration from the perspective of relevant predators. We test for correlations between toxin levels and measures of wing colour across 14 species of day‐flying burnet and forester moths (Lepidoptera: Zygaenidae), protected by highly aversive cyanogenic glucosides, and find no clear evidence of quantitative signal honesty. Significant relationships between toxin levels and coloration vary between sexes and sampling years, and several trends run contrary to expectations for signal honesty. Although toxin concentration is positively correlated with increasing luminance contrast in forewing pattern in 1 year, higher toxin levels are also associated with paler and less chromatically salient markings, at least in females, in another year. Our study also serves to highlight important factors, including sex‐specific trends and seasonal variation, that should be accounted for in future work on signal honesty in aposematic species.

The distinctive black and red wing pattern of six‐spot burnet moths (Zygaena filipendulae, L.) is a classic example of aposematism, advertising their potent cyanide‐based defences. While such warning signals provide a qualitatively honest signal of unprofitability, the evidence for quantitative honesty, whereby variation in visual traits could provide accurate estimates of individual toxicity, is more equivocal. Combining measures of cyanogenic glucoside content and wing color from the perspective of avian predators, we investigate the relationship between coloration and defences in Z. filipendulae, to test signal honesty both within and across populations. There were no significant relationships between mean cyanogenic glucoside concentration and metrics of wing coloration across populations in males, yet in females higher cyanogenic glucoside levels were associated with smaller and lighter red forewing markings. Trends within populations were similarly inconsistent with quantitative honesty, and persistent differences between the sexes were apparent: larger females, carrying a greater total cyanogenic glucoside load, displayed larger but less conspicuous markings than smaller males, according to several color metrics. The overall high aversiveness of cyanogenic glucosides and fluctuations in color and toxin levels during an individual's lifetime may contribute to these results, highlighting generally important reasons why signal honesty should not always be expected in aposematic species.