A toothcomb is a dental specialization in which the lower incisors and canines combine to form a structure that sort of resembles a comb. In the diagram to the right you can see how the mandibular incisors project forward, alongside the canines (which actually sort of resemble incisors rather than canies) to create this structure.
The toothcomb is unique to strepsirrhine primates (lemurs, lorises, and galagos). The fossil record, in conjunction with molecular clock estimates, suggest that lemurs and lorises diverged in the Paleocene, over 60 million years ago. This tells us that the toothcomb has very deep evolutionary origins in the strepsirrhine primate lineage.
The only extant strepsirrhine that lacks a toothcomb is the ugliest of the lemurs, the Aye-Aye (Daubentonia madagascariensis). In the past this lead some researchers to suggest that the toothcomb may have evolved twice, once in the lemur clade, and then again among the lorises and galagos. A good old classic debate over homoplasy, parallel, and convergent evolution was brewing.
Then a primate anatomist by the name of Freiderun Ankel-Simons came along and demonstrated that if you look at the deciduous dentition (the baby teeth) of the Aye-Aye there's a very strong indication that they had a toothcomb earlier on in their ancestry and have since lost this specialization much more recently.
Today, the evolutionary debate surrounding the toothcomb surrounds questions of it's origin and function. Most strepsirrhines use the toothcomb for purposes of grooming, but some strepsirrhines, such as mouse lemurs (Microcebus), Fork-marked lemurs (Phaner), and the Needle-clawed bushbabies (Euoticus), use their toothcombs to scrape gums and resins from trees. The question now becomes: did the toothcomb evolve for grooming purposes and then it was co-opted for feeding OR did the toothcomb evolve for feeding purposes and then get co-opted for grooming? It's difficult to say either way. Among extant strepsirrhines the toothcomb is most commonly used for grooming, but just because a morphological feature might be used for a particular function in the present, it could have easily been evolved for an entirely different function in the past.
Aye-aye, there's the rub.