Warning colouration on a diet of ants and termites

Everyone knows that dendrobatid frogs, which are classic examples of warning colouration, tend to eat ants as a staple (https://en.wikipedia.org/wiki/Poison_dart_frog and https://www.researchgate.net/publication/232710571_Myrmecophagy_and_alkaloid_sequestration_in_amphibians_a_study_on_Ameerega_picta_Dendrobatidae_and_Elachistocleis_sp_Microhylidae_frogs).

Everyone also knows that there are various mammals, birds, reptiles and amphibians around the world that tend to be myrmecophagous (https://en.wikipedia.org/wiki/Myrmecophagy), i.e. to have specialised diets of various combinations of ants and termites.

But who knows whether there is any general association between warning colouration and myrmecophagy in these vertebrates?

There are two related reasons why myrmecophagous vertebrates can be expected to be particularly defensive against predators.

Firstly, the noxious substances of the social insects, once eaten, can potentially be sequestered by the myrmecophages themselves as a basis for their own chemical defence.

Secondly, social insects - particularly in the form of workers and soldiers as opposed to eggs or alates - tend to be poor foods. This is because:

  • ants consist largely of exoskeleton,
  • termites themselves tend to eat poor foods (fibrous detritus),
  • much grit and frass tends to be ingested while eating these insects, and
  • the chemical defences contained by the ants (and certain termites such as nasutitermitines) are metabolically costly to detoxify.

Given their energy-poor diet, myrmecophagies tend to have both small brains (relative to body size) and a limited metabolic capacity to flee or fight. They thus tend to be more vulnerable to predation than are like-size herbivores in the same animal communities.

So, how do mymecophages generally evade predation, and how does adaptive colouration relate to this beyond dendrobatid frogs?

The answer:

There is some degree of convergence among frogs on different continents, in the association between myrmecophagy and conspicuous colouration (e.g. Phrynomantis spp. in Africa and Notaden bennettii in Australia). However, this does not extend to myrmecophages beyond frogs; there is no syndrome of warning colouration and toxicity/noxiousness shared among the various ant- and/or termite-eating vertebrates. Conversely, among skunks - which epitomise warning colouration - even the most insectivorous species tend not to eat ants or termites (https://en.wikipedia.org/wiki/Hog-nosed_skunk#Feeding_habits and https://en.wikipedia.org/wiki/Striped_polecat#Diet).

Examples: the aardvark (https://en.wikipedia.org/wiki/Aardvark) survives mainly by virtue of strictly nocturnal activity and refuge in deep burrows by day. Pangolins, myrmecophagous armadillos and the short-beaked echidna are armoured, with this echidna and one species of pangolin also relying partly on unusual digging abilities for refuge. The numbat (https://en.wikipedia.org/wiki/Numbat) has extreme camouflage.

Among the various myrmecophagous reptiles and amphibians, some have bizarre defences (e.g. https://en.wikipedia.org/wiki/Thorny_devil and https://en.wikipedia.org/wiki/Horned_lizard) and some are toxic/noxious, but few have warning colouration (e.g. see https://www.reddit.com/r/Awwducational/comments/nbhx75/this_friendly_frisbee_is_the_twocolored_oval_frog/). Conversely, most of the aposematic species of amphibians do not qualify as myrmecophagous (https://en.wikipedia.org/wiki/Poisonous_amphibian and https://en.wikipedia.org/wiki/Corroboree_frog#Diet).

This brings us to certain anteaters of central and South America, which are among the few species providing some association between myrmecophagy and warning colouration. I refer to the two species of tamanduas (please see my latest Post) and the giant anteater (https://zslpublications.onlinelibrary.wiley.com/doi/abs/10.1111/j.1469-7998.1985.tb03544.x and https://en.wikipedia.org/wiki/Giant_anteater) .

In both tamanduas and their large relative (http://www.takepart.com/article/2014/07/29/oh-great-now-giant-anteaters-are-killing-people and https://www.istockphoto.com/photo/giant-anteater-walking-on-grass-gm1020098972-274095467 and https://outschool.com/classes/introduction-to-anteaters-bxMHWjIy and https://www.theanimalfacts.com/mammals/giant-anteater/), the colouration has some dark/pale contrast - particularly on the shoulders - which seems inconsistent with camouflage (https://www.inaturalist.org/observations/7614630 and https://www.inaturalist.org/observations/41560892). The patterns are not as graphic as in skunks but are, depending on the situation, conspicuous enough to suggest self-advertisement rather than blending into the surroundings (e.g. https://www.inaturalist.org/observations/21586823).

In only one of these two genera is there any hint that the warning colouration is based on noxious substances sequestered from the diet.

Different warnings seem to apply in the two genera, Myrmecophaga vs Tamandua, if we assume that warning colouration tends to hint at defensive capabilities that are unapparent (e.g. toxic/noxious substances), rather than to emphasise those that are easily displayed as such (e.g. teeth, which anteaters lack completely).

In the giant anteater, the lethality of the fore claws (https://www.researchgate.net/figure/Giant-anteater-claws-which-are-very-elongated-in-the-second-and-third-digits-Photo-by_fig2_263971704) is not obvious, partly because the animal knuckle-walks with the claws folded (see https://vimeo.com/108411348). The colouration thus seems to warn 'Beware my strength in using my earth-raking claws as daggers on you'.

For their part, tamanduas display the claws and muscular shoulders in their characteristic defensive pose, but what is warned of by the colouration seems instead to be a somewhat skunk-like capacity to produce noxious substances from the anal glands.

This raises questions for future research: do tamanduas make their defensive secretions out of substances sequestered from their diet, or not? And are the defensive secretions of tamanduas different from those of pangolins, which lack any warning colouration whatsoever?

Posted on October 27, 2021 09:00 PM by milewski milewski

Comments

The striped possum https://animaldiversity.org/accounts/Dactylopsila_trivirgata/#3a6c4eaa16ac905e6670a91d3009d270 is an ambivalent case of warning colouration, because its black-and-white striping seems to function partly for camouflage and its foul odour has yet to be studied. Furthermore, it does not qualify as myrmecophagous despite much of its diet consisting of ants and termites.

Posted by milewski over 2 years ago

It is noteworthy that the dark/pale contrast in the giant anteater is precocial, being particularly clear in the infant: https://www.reddit.com/r/Awwducational/comments/3jcxkd/zoo_boise_is_happy_to_announce_the_birth_of_a/.

Posted by milewski over 2 years ago

One of the few African genera of frogs that combines myrmecophagy with aposematism is Phrynomantis, e.g. https://amphibiaweb.org/species/2084#:~:text=This%20frog%20prefers%20to%20walk,During%20the%20mating%20season%2C%20P and https://calphotos.berkeley.edu/cgi/img_query?enlarge=0000+0000+0513+2294 and https://www.inaturalist.org/observations/37503465. Although Phrynomantis has converged with dendrobatid frogs in eating ants and having warning colouration, another genus of frogs widespread in Africa, namely Hemisus (e.g. see https://amphibiaweb.org/species/1512) is myrmecophagous without warning colouration.

Posted by milewski over 2 years ago

The Australian frog Notaden bennettii (see https://www.jungledragon.com/specie/2754/crucifix_toad.html) is said to eat mainly ants and termites. It is the most colourful species in its genus and qualifies as aposematic.

Posted by milewski over 2 years ago

Thanks for this article @milewski . Very interesting. I wonder if African frogs like phrynomantis eat worker ants part of the time but eat the larger more nutritious winged ants as they emerge from holes in the ground. I need to get some observations of them eating ants to get the species they prey on. I guess people are also myrmecophagous here in Botswana !

Posted by botswanabugs over 2 years ago

I wonder if there are ant species avoided by phrynomantis !

Posted by botswanabugs over 2 years ago

@botswanabugs I agree that not enough attention has been paid to the distinction between the eating of alates and the eating of workers/soldiers/larvae/nymphs/eggs. For example, the Australian frog genus Notaden (https://en.wikipedia.org/wiki/Notaden) seems to spend most of its life dormant underground, and is active on the surface only after rain. It seems plausible that the ants and termites eaten by Notaden are mainly alates. If so, this is trophically different from the year-round eating of workers and soldiers, which are far poorer food than alates are. In the case of Phrynomantis (and various other microhylids on other continents) my impression is that it is mainly workers and soldiers that are eaten, but it would be good to confirm this. One clue may be the size of the mouths. Microhylids are partly characterised by small mouths, consistent with the eating of small items (https://en.wikipedia.org/wiki/Microhylidae). Phrynomantis compensates for the smallness of its mouth by having a particularly mobile tongue (https://homes.cs.washington.edu/~diorio/MURI2003/Publications/2003-2004pubs/Inside%20JEB2003.pdf), which seems analogous with Myrmecophagidae and pangolins with their small mouths and extremely mobile tongues. The mouths of Notaden spp. seem relatively large (http://animal.memozee.com/view.php?tid=3&did=25526), although not as large as in various other frogs and toads.

Posted by milewski over 2 years ago

Add a Comment

Sign In or Sign Up to add comments