Camouflage, Warning Colouration: A Large-Scale Global Experiment On The Success Of Prey Defense Strategies
While warning colouration fared better in environments with low predation intensity, camouflage strategy helped when camouflaged prey species were less common and when light levels were low. Climatic seasonality had no effect on predation risk, and predation was not lower for warning-coloured prey at lower latitudes
A global experiment spread across 21 forests on six continents found that both camouflage and warning colouration were effective in keeping predators at bay under different ecological conditions.
Warning colouration or aposematism (a conspicuous signal to alert predators of its unsuitability) fared better when there was a high percentage of warning-coloured lepidopterans in the prey community, including toxic warning-coloured (aposematic) species, while the performance of the camouflage strategy (crypsis) was affected by more ecological variables, and affected more strongly by these variables. The camouflaged prey are more affected by changes to the predator community, prey community and light environment. Climatic seasonality had no effect on predation risk, and unlike what was predicted, predation was not lower in the case of warning-coloured prey at lower latitudes.
The study tried to uncover why different moths use one form of defence over the other. And it found that the answer was not simple — it is influenced by predator community, prey community and habitat. That explains why both camouflaged and warning-coloured moths are found in all 21 forests studied across the world. That said, the authors of a study published on September 25, 2025 in the journal Science hypothesise that camouflage, although widespread, “may be a less stable defense that is more vulnerable to ecological and anthropogenic change”. The hypothesis predicts that “predation outcomes should be more variable for individuals and populations pursuing camouflage strategies compared with warning coloration”. As a result, camouflage should be “gained and lost more frequently than warning coloration at macroevolutionary scales”.
How the experiment was done
The study used 15,018 paper “moth” artificial prey targets, with about 720 prey targets per location. The targets were printed with a bark-like brown colour for camouflage, typical orange-black warning coloration or turquoise-black warning colouration, which though conspicuous is not a typical warning colouration used by moths.
Dr. William Allen from the Department of Biosciences, Swansea University, Swansea, U.K. and the first and one of the corresponding authors explained in an email why the experiment limited to just eight days was sufficient to draw conclusions. He says: “We carried out the experiment over eight days because we were interested in differences between the different coloured prey in how quickly predators learned that the prey was good to consume. Prior work, including pilot work for this experiment, showed that predators are pretty quick to do this. If we ran the experiment for much longer, we would likely just find that in some places predators find all the different coloured prey very quickly, making trials on these days not so valuable. Our study simulated the scenario where novel prey phenotype aims to establish a population in a new predator community —the key scenario for the evolution of antipredator colour diversity.”
However, care was taken to make sure that the study was not conducted during the breeding season of forest birds and lizards, which are some of the prominent visually hunting predators of moths and other insects. The National Centre for Biological Sciences (NCBS) in Bengaluru carried out the study in India at two sites — Honey Valley in Kodagu (Coorg), Karnataka and Buxa in West Bengal — between September and November 2021. During the experiment at the two India sites, the artificial “moth” models were observed over eight days under natural conditions, and records were taken as to which models were attacked versus left alone by the community of predators that lived in the area.
Explaining why the study was not conducted during the breeding season, Dr. Krushnamegh Kunte from NCBS and a coauthor of the paper, says: “During the breeding season, older predators that are experienced would likely know which prey to eat and which ones to avoid, while young, inexperienced predators would lack that knowledge. A mixed population of experienced and inexperienced predators would lead to highly variable results that may not make much sense if we do not control for the experience factor. Well outside of the breeding season, we naturally control for this factor, as even young predators would have likely learnt by then.”
About why the study found that seasons did not have a role to play in predation, Dr. Allens says: “As much as was logistically possible, we standardised the season the experiment ran in each location to the peak of the breeding season. This is before uneducated fledgling birds that do not have knowledge of the relationship between warning colour and defence enter the predator population. Tropical locations, however, are often less seasonal than temperate locations, and do not have such a pronounced bird breeding season. This means that there are likely to be more uneducated predators in less seasonal locations, which could have changed predation patterns on the different colour treatments. We did not find evidence of this, however.”
How predation risk changes with prey/predator density
The study found that latitude was not linked to antipredator warning-colouration strategy. Though predation was slightly greater at low latitudes at the start of the experiment, it reduced as the experiment progressed. Overall, predation was not higher in locations at low latitudes.
When the number of predators was high, predation risk increased throughout the experiment. However, the increase was at a slower rate in the case of the paper “moths” with typical warning colouration compared with camouflage and atypical warning-colouration strategies. Though camouflaging strategy initially helped protect prey even when predator numbers were high, the success of camouflage strategy reduced when the number of prey using camouflage to hide from predators increased in number and as predators were presumably better at locating camouflaged prey.
Bright illumination lowers protective effect of camouflage
Also, low illumination initially helps improve camouflage, but the advantage reduces with time as predator performance improves. According to the authors, this supports the hypothesis that camouflage strategies decline in performance once predators learn what to search for in low-light environments. Under high illumination, camouflage had lower protective value while protective value was only slightly weaker in the case of prey that used typical warning-colouration strategy.
“Even if camouflage is good, most predators’ eyes have better resolution under better illumination. So, prey deploying camouflaging strategies can still be spotted more frequently with increased efforts under well-illuminated conditions,” says Dr. Kunte. “Typical warning-colouration strategy works best under bright light conditions as warning-colouration signals are conveyed more clearly to predators.”
Camouflage fitness of prey in Honey Valley forest
Viraj Nawge, a PhD student at NCBS and a coauthor of the paper says that in the Honey Valley site in Kodagu, India, which is a wet evergreen forest with low light conditions, tree bark was often were covered by greyish lichens and green algae. As a result, the brown paper “moths” were not always as camouflaged as they would be on drier barks. Even Buxa is a semi-evergreen forest and the brown crypsis is effective on some of the tree barks but not all, although low-light conditions in the understorey probably improve camouflage.
In the Honey Valley Forest where the paper moths are brown but the trees are covered with lichen and algae that are grey and green, respectively, can the prey be considered as adopting crypsis/camouflage to escape from predators? Can conclusions be drawn from poorly camouflaged prey in such a setting? “Yes, there was variation both between and within locations in how well camouflaged the cryptic paper “moths” were due to difference in the bark colour and lichen coverage of the tree community. This variation is something we were really interested in — how important is it that camouflage prey are highly cryptic? We measured this using the digital image analysis and investigated the degree of crypsis (in the paper we call it ‘salience’) in our statistical models. Importantly, in all locations the cryptic prey were less salient than both warning-coloured prey. Overall, we did not find that any of the three prey colour types were more or less successful — instead their relative success depended on ecological context. However, predators did get better at finding the more salient cryptic prey as the experiment progressed,” Dr. Iliana Medina from the School of Biosciences, the University of Melbourne, Australia, and the first and a corresponding author said in an email. “Overall, the camouflage prey [even when they did not blend very well with the background] were not more vulnerable than the warningly-coloured prey.”
Camouflage works best in areas with a high predation pressure. However, predation risk increased rapidly when camouflaged prey became more prevalent. Compared with greatly reduced protective value of camouflage when prey was found in abundance, the risk of predation was about 20% lower in the case of prey with typical warning colours even when they were more common. The reduced risk was seen at the start of the experiment.
“When there is plenty of prey with typical warning colours, predators will likely become more adventurous and risk attacking prey that might be dangerous or distasteful but to a lesser degree, especially when predators compete intensely for food,” says Dr. Kunte. “The results show how multiple ecologically relevant mechanisms determine which strategy is evolutionarily more advantageous under specific circumstances.”
