Initially thought to be an “Old African” group, pangolin fossils have now been found in Europe, Asia, and North America. The most recent phylogenetic studies state that the evidence points to a European origin for the order, Pholidota (Gaudin et al. 2009).
The taxonomy of Patriomanis americanus and the pangolin family has been the subject of long-standing debate. Some researchers have argued for up to six genera (Pocock 1924) for the eight living species, and others (Emry 1970) for just two. Due to the confusion, many studies have simply kept all the species in a single genus, pending more systematic study. The most recent study (Gaudin et al 2009) , places the extant pangolins in three genera. On the other hand, there seems to be a general consensus that most species of fossil pangolins, including P. americanus, are from outgroups (sister taxa, not ancestral taxa) of the extant species. The family Patriomanidae was recognized in 1998 by Szalay and Schrenk. Gaudin et al., in their extensive 2009 study based on hundreds of measurements of living and fossil species, placed P. americanus as the closest relative of a North Asian fossil pangolin, Cryptomanis gobiensis. Before the discovery of this fossil in 2006, (Gaudin et al 2006), most paleontologists pointed out P. americanus’ striking similarities to the European fossil pangolins (Eomanis, Necromanis) and the modern pangolins (Manis). At any rate, there seems little doubt that all of the known species of pangolins are very similar in form and are closely related, though the overall placement of the group within mammalia is also very murky, partially due to the lack of illuminating ancestral characteristics in the fossil families.
Early researchers had a great tendency to group the pangolin order, Pholidota, as a sister taxon to the xenarthrans (anteaters, armadillos, and sloths) due to a handful of adaptations to eating termites in the former two, such as loss of teeth, elongated tongue, and narrowing of the jaw (e.g. Gaudin and Wibble 1999). Later, evidence based on careful measurements of important skeletal elements showed that this was not the case, and led researchers to caution against the use of lost traits to define relationships between taxa (Emry 2004). Emry (2004) cites the ease and frequency with which traits are lost in the evolutionary record, and suggests that these traits are simply an example of convergent evolution. The most recent comprehensive studies of mammalian intraordinal (between orders) relationships suggests that pangolins are most closely related to the carnivora (Beck et al 2006).