In the paper

The Fishes of the Endau Drainage, Peninsular Malaysia with Descsriptions fo Two New Species of Catfishes (Teleostei: Akysidae, Bagridae)
Heok-Hee Ng and Heok-Hui Tan - Zoological Studies 38(3): 350-366 (1999)

the authors (currently both of the Dept of Biological Sciences, NUS) made the following remarks (pp 355):
What is currently recognised as the genus Puntius sensu lato is probably polyphyletic (Roberts 1989). Rainboth (1996) considers Systomus to be a valid genus distinct from Puntius sensu stricto in having a finely serrated (vs. smooth) dorsal-fin spine, 2-4 barbels (vs. 2 maxillary barbels), and less thatn 12 (vs. 12-20) gill rakers on the first gill arch. We tentatively follow Rainboth (op. cit.) in recognising Systomus as a valid genus.

also from Dr. Ng Heok-Hui (pers. comm.)
First the issue of Systomus vs. Puntius. Systomus was resurrected as a valid genus by Walter Rainboth, who used it for small barbs with a serrated dorsal-fin spine, 2 or 4 barbels, and less than 12 gill rakers on the first gill arch. In contrast, Puntius would have a smooth dorsal spine, only 2 (never 4) barbels, and more than 12 gill rakers on the first arch. So, by this definition, many of the barbs encountered in the aquarium trade are Systomus (S. tetrazona, S. binotatus, S. hexazona, S. partipentazona, S. johorensis, S. gemellus, and so on...).

Of course, this scheme is not adhered to by all workers, so some references (notably those of Kottelat) still refer to these as Puntius.