One of my favourite mammal assemblages of all is Xenarthra, the ‘strange joint’ group that includes the remarkable and odd anteaters, sloths and armadillos…
Caption: a xenarthran montage. I’ve drawn a few xenarthrans over the years, but not as many as I should have by now. Here are a few illustrations I have to hand, but many more will be included in my textbook once I get to finishing it. Images: Darren Naish.
I’ve published a bit on those groups here, over the years (see the list of links below). I’ve just never published all that much. I aim to change that this year, and let’s start here, with assorted thoughts and observations on ARMADILLOS. Be sure to check the links for additional info.
Long-nosed armadillos. If you’ve heard of armadillos, the chances are high that you’re aware of the Nine-banded armadillo Dasypus novemcinctus, one of the best known, most intensely studied, and most geographically widespread of armadillo species. Until very recently, it was regarded as the only living long-nosed armadillo species that ranges into North America*. Incidentally, not all individuals have the eponymous nine bands: the actual number is variable across the range of the species. Some individuals have eight, others ten.
* The text here initially stated – erroneously – that the Nine-banded armadillo is the only living armadillo species that occurs in North America, something that’s very much incorrect in view of the presence of naked-tailed armadillos in Mexico. The ‘Until very recently’ there refers to the 2025 splitting of the D. novemcinctus complex, something I will come back to in a future article.
Caption: the long-nosed armadillos include very familiar species – like the Nine-banded armadillo shown at left – as well as very obscure ones that are still poorly known and mostly unphotographed in living condition. Hence we have species like this one (on the right: the holotype of the Yungas lesser long-nosed armadillo D. mazzai), mostly only known from taxiderm specimens like this. More on D. mazzai in another article! Images: Mwcolgan8, public domain (original here); Feijó et al. (2018), CC BY 4.0.
Relatively little known is that D. novemcinctus is only one of eight or nine extant species in the genus Dasypus, collectively termed the long-nosed armadillos. Excepting D. novemcinctus, all are limited to South America. While their taxonomy has been considered mostly settled over recent decades (a real contrast to the 19th century), a few late 20th century and 21st century revisions have resulted in a bit of confusion with respect to the more poorly known taxa (Vizcaíno 1995, Feijó & Cordeiro-Estrela 2014, 2016, Feijó et al. 2018).
One thing to note is that some of these species are – in real contrast to the Nine-banded armadillo – of conservation concern and officially classified as ‘vulnerable’ or ‘near threatened’. That goes for the Southern long-nosed armadillo D. hybridus and the Hairy long-nosed armadillo D. pilosus, though a case has been made by the IUCN (Superina & Abba 2014) that the latter should better be treated as ‘Data Deficient’ (this is one of the worst categories to be in, as it creates the impression that the animal is ok). The Southern long-nosed armadillo, by the way, might not be a distinct species, since one recent revision found it to be a subspecies of the Seven-banded armadillo D. septemcinctus (Feijó et al. 2018).
Caption: yes, this is a real animal of the modern age, not a CG reconstruction of a Mesozoic mammal, not a rhinogradentian (reference). There are very few photos of live specimens of the Hairy or Woolly long-snouted armadillo, and this is the one that gets used the most. It was taken by André Baertschi and this version was taken from the relevant IUCN page. Other images of the species can be seen at iNaturalist. Image: © André Baertschi / wildtropix.com.
The Hairy or Woolly long-nosed armadillo. D. pilosus is radically odd. Endemic to the montane cloud forests of the Peruvian Andes (specifically to the departments of San Martín, La Libertad, Pasco, Huánuco, Junín and Amazonas), it’s very poorly known: just six specimens were known to science between its description in 1856 and recent decades, and even today few individuals are available in collections (Castro et al. 2015). The original specimen was described by Leopold Fitzinger based on a taxiderm specimen purchased in London by the animal dealer and merchant Ludwig Parreyss in 1833 (Feijó et al. 2018). It had no locality data beyond merely ‘Peru’ and Fitzinger was so perplexed by it that he wondered whether it might be an intermediate between armadillos and anteaters (Feijó et al. 2018).
Caption: the Dasypus pilosus holotype, the mounted taxiderm specimen NMW ST 222 at the Naturhistorisches Museum, Vienna, Austria, as featured in Feijó et al.’s (2018) excellent and very thorough taxonomic review of long-nosed armadillos. It’s labelled with the name Praopus hirsutus, this being given to the species by Hermann Burmeister in 1862 since he was unaware of Fitzinger’s earlier publication. Check Feijó et al. (2018) for more data on this specimen and the taxonomic history of D. pilosus in general. Image: Feijó et al. (2018), CC BY 4.0.
Even today, almost nothing is known about the ecology, behaviour or biology of this species. Some authors have speculated that it might be a specialized insectivore that eats ants and/or termites (in which case it’s a myrmecophage or termitophage) (Castro et al. 2015), the unusual anatomy of its skull, teeth and jaw joint indicating that it’s even more specialized for this lifestyle than other Dasypus species.
However, its existence at altitude (2600-3400m above sea level; Castro et al. 2015) casts doubt on this suggestion, since surveys demonstrate that insects are comparatively rare in such places. Feijó et al. (2018) noted that D. pilosus has a number of similarities (including an especially elongate snout and mandible and reduced teeth) with Asian montane rodents that eat soft-bodied invertebrates, in particular worms. Could, then, D. pilosus be a (mostly) worm-eating armadillo? It’s a really interesting idea and more data is needed.
Caption: very nice image of Hairy or Woolly long-nosed armadillo in life, by Jorge González and included in Castro et al. (2015). It is not well known – even among people who know mammals – that there are armadillos that look like this. Image: Jorge González, Castro et al. (2015).
And it really is hairy, or woolly; being properly covered in a pelt, this giving it a body which superficially recalls that of a shaggy rodent or giant shrew. If you know about armadillos, you’ll know that there are species – most notably the hairy armadillos or peludos (Chaetophractus) – where a reasonable amount of hair is present across the unarmoured underside, and also here and there across the carapace. However, the Hairy or Woolly long-nosed armadillo is a wholly different thing. Scutes are only visible on the top of its head and snout, ears, hands, feet and slender tail. The hairs grow through pores on the scutes (Castro et al. 2015), it being one of several armadillos which show that you can be totally armoured and sport a pelage at the same time.
The case for Cryptophractus. D. pilosus is, in fact, so odd that a person who knows armadillos might doubt, purely intuitively, whether it should be included within Dasypus. It’s not just different because of the hairiness but also because it has a notably longer, more slender skull, especially small teeth, a reduced mandible with a more limited range of motion than other long-nosed armadillos, a higher number of movable bands than other long-nosed armadillos (as many as 11), and unusual osteoderms that lack sulci and have a unique pattern of pores (Castro et al. 2015). Accordingly, a long-standing idea is that it should be given its own ‘subgenus’ within Dasypus, namely Cryptophractus.
Caption: the Hairy or Woolly long-nosed armadillo is a special armadillo. As shown in these images – all from Castro et al. (2015) – the hairs emerge from numerous pores in the scutes, which appear to match the numerous foramina (bony openings) in the underlying osteoderms. Osteoderms = bony structures embedded in the skin. Scutes = keratinous structures overlying osteoderms (albeit not always!). There’s a long-standing discussion on whether foramina in bony structures are at all relevant to the presence or otherwise of hairs. This is a case where they apparently are. Scale bars = 50 mm. Images: Castro et al. (2015).
Inspired by the distinct nature of this species (and by the lack of detailed study), Mariela Castro and colleagues published an analysis in 2015, their conclusion being that it is indeed ‘distinct enough’ from other long-nosed armadillos to warrant exclusion from Dasypus (Castro et al. 2015), this being consistent with their phylogenetic analysis. This, if right, has implications for a number of fossil armadillos also named as species of Dasypus, since their position outside the Cryptophractus + Dasypus clade requires that they need distinct ‘genus-level’ names.
Caption: Castro et al. (2015) found this phylogenetic arrangement for dasypodids and other armadillos via their analysis of 58 anatomical characters. Stegosimpsonia is a fossil animal from the Eocene while Peltephilus and Stegotherium are Miocene; Propraopus and Anadasypus are dasypodids from the Pleistocene-Holocene and Miocene, respectively. The topology here has the Hairy or Woolly long-nosed armadillo outside a clade that contains all other Dasypus species, this being deemed consistent with the idea that the species might be distinct enough to be recognised as the distinct genus Cryptophractus. Does molecular data support this view? Read on. Image: Castro et al. (2015).
The case against Cryptophractus. However, 2015 is a little while ago now. How has this proposal been received? Hautier et al. (2017) looked at shape variation across long-nosed armadillo skulls using morphometrics, and did find pilosus to be an outlier in several of their analyses. They mostly found, however, that it was not so extreme that it should be excluded from Dasypus; rather, it was ‘extreme’ in terms of the sort of shape variation they possess.
Caption: Hautier et al. (2017) used principle components analysis to analyse the degree of shape variation present in the long-nosed armadillo species. In the plots shown here – analysing the shapes of mandibles – the Hairy or Woolly long-nosed armadillo is an outlier, being mostly outside the main cluster occupied by most other Dasypus species (it’s the red circles at lower left in A and B, and lower left in the left part of the main cluster in B). It’s not the only species that was found to be an outlier though. Image: Hautier et al. (2017).
What seems to be the death knell for the ‘distinct Cryptophractus’ hypothesis came from the theatre of molecular systematics. Gibb et al. (2016) found a totally contradictory phylogenetic arrangement to Castro et al. (2015), since they recovered pilosus within the middle of the long-nosed armadillo radiation (not as its outgroup), and as a young lineage (not much older than Late Pliocene) whose “molecular divergence does not seem to match its morphological distinctiveness”. They were very much aware of Castro et al.’s study in saying this, since they took time to comment on it, noting that its anatomy-based conclusions were (according to their results) less reliable than their molecular ones.
Long-nosed armadillos, the book! One of the first things I checked when receiving Richard Webb and Jeff Blincow’s excellent A Field Guide to the Larger Mammals of South America (Webb & Blincow 2024) was the long-nosed armadillo section. Turns out that they give these animals excellent coverage, with full pages given to all species, four of which are depicted via paintings since no good photos exist! And there in the middle of this section is the Hairy long-nosed armadillo, smack-bang within Dasypus, and also featured in a location that doesn’t – contra Castro et al. (2015) – make it phylogenetically distinct relative to the others. I guess they’d seen Gibb et al. (2016) and followed it.
Caption: Richard Webb and Jeff Blincow’s A Field Guide to the Larger Mammals of South America is an extremely impressive book and I strongly recommend it. It features whole pages on most species, even tremendously obscure and recently named ones, and that includes all the obscure long-nosed armadillos. But good photos of some of those don’t exist, so they had to resort to the inclusion of artwork. Get the book here. Images: Darren Naish.
For all its obscurity, Dasypus pilosus is not – believe it or not – as obscure as other members of the long-nosed armadillo group. I was planning to talk about those (or, some of them, anyway) in this article… but it’ll have to wait. In fact, I had hoped in this article to talk about armadillo diversity in general. Alas, things expanded and time, as ever, is against me. I will be coming back to this fascinating group in time.
ADDENDUM: after I finished this article and prepared to get it published, I had a look to see if other bloggers have covered the species before. And turns out that Darin Croft – at his excellent blog The Rafting Monkey – published an article on Dasypus pilosus back in 2017. It essentially presages everything I’ve covered here, showing that great minds think alike, I guess. Check out that article for another take on this species.
For previous Tet Zoo articles on armadillos and othr xenarthrans, see…
Five things you didn’t know about armadillos, June 2007
What was that skull? (on glyptodonts), May 2008
I, Priodontes, the tatuasu, September 2008
Predation and Corpse-Eating in Armadillos, September 2010
Refs - -
Castro, M. C., Ciancio, M. R., Pacheco, V., Salas-Gismondi, R. M., Bostelmann, J. E. & Carlini, A. A. 2015. Reassessment of the hairy long-nosed armadillo “Dasypus” pilosus (Xenarthra, Dasypodidae) and revalidation of the genus Cryptophractus Fitzinger, 1856. Zootaxa 3947, 30-48.
Feijó, A. & Cordeiro-Estrela, P. 2014. The correct name of the endemic Dasypus (Cingulata: Dasypodidae) from northwestern Argentina. Zootaxa 3887, 88-94.
Feijó, A. & Cordeiro-Estrela, P. 2016. Taxonomic revision of the Dasypus kappleri complex, with revalidations of Dasypus pastasae (Thomas, 1901) and Dasypus beniensis Lönnberg, 1942 (Cingulata, Dasypodidae). Zootaxa 4170, 271-297.
Superina, M. & Abba, A.M. 2014. Dasypus pilosus. The IUCN Red List of Threatened Species 2014: e.T6291A47441122. Accessed on 30 April 2025.
Vizcaíno, S. F. 1995. Identificación específica de las "mulitas", género Dasypus L. (Mammalia Dasypodidae), del noroeste argentino. Mastozoologia Neotropical 2, 5-13.
Webb, R. & Blincow, J. 2024. A Field Guide to the Larger Mammals of South America. Princeton University Press, Oxford.
