The Genomics of Yellow Bats

In terms of species, bats are the second-largest order of mammals, after the rodents. New species are identified all the time, due in part to the relative difficulty of closely examining night-flying mammals, many of which sleep in hard-to-access caves. The current total stands at over a thousand, representing over 20% of all known mammal species. 

Within this huge group, there is, perhaps, rather more diversity than many people realise. While bats have probably not received the same level of attention as some other mammal groups, scientists have nonetheless long attempted to disentangle the relationships between all these subgroups. (Also, when I say they have received less attention, there's a mammal-centric bias here; it's probably still a lot better than, say lizards, let alone millipedes or the like). 

Euplerids: Falanoucs and Fanalokas

Fanaloka

The relationships of the native Madagascan carnivores were uncertain before the advent of genetic testing at the end of the 20th century. The majority were assumed to be mongooses, since that is very much what they looked like, but three species didn't quite fit the pattern. 

Two of these were, so far as anyone could tell from their physical appearance and habits, civets. They were often placed in their own subfamily, reflecting their distance from other civets, but nonetheless, they were thought to belong among the viverrids. Following genetic studies in 2003, however, it became clear that they belonged in the same group as (i.e. shared a unique common ancestor with) the Madagascan "mongooses". Since some of them clearly weren't mongooses, and since they had diverged from the real ones so long ago, the subfamily was split off and promoted to full family level, now including both the mongoose-like and civet-like species.

Eocene (Pt 4): Ancient Beasts of the European Archipelago

Propalaeotherium

The story of the evolution of the horse is one of the most commonly cited examples of evolutionary trends, often illustrated by a series of increasingly horse-like animals with an ever-reducing number of toes. The animal typically shown at the start of that series is, depending partly on the age of the picture, either Hyracotherium or Eohippus.

Hyracotherium was long regarded as the earliest known member of the horse family. In recent decades, it has become apparent that it wasn't really a horse, in the sense that modern horses don't descend from it or its relatives, and today we call the family it belonged to the palaeotheres. The North American Eohippus, on the other hand, despite long being thought to be identical to Hyracotherium, probably is a horse. The confusion between the two means that it's often difficult to tell which is being referred to in older sources.