Geographical Distribution. -Throughout the Holarctic, Indo-Malayan, Australasian, African and Neotropical regions, from Arctic regions of Eurasia south to the Cape of Good Hope, and Tasmania; east to Fiji and other islands of the Pacific; Madagascar; and in the New World from Arctic regions including Greenland south to Tierra del Fuego, and including the Galapagos Islands. Number of Genera. -I have examined and retained one hundred and eight)'-six genera, divided among twelve subfamilies. At least six, and perhaps more, valid genera are not represented in the British Museum.(In the first volume, containing all other Rodents, one hundred and fortyseven genera were retained, and four besides these are not represented in London.)The Family Muridae belongs according to the present classification to the Superfamily Muroidae, a group containing also four families, Muscardinidae, Lophiomyidae, Rhizomyidae, and Spalacidae, which have been dealt with in the first volume.A key to these families has already been given. Notes on the essential characters of the family Muridae have also been given in Volume I. For reference purposes, these are repeated here.Char.\cters.-Zvgomasseteric structure primitively (Deomys), nearly as in Sicistinae (Dipodidae), except that the infraorbital foramen is more generalized, less enlarged, and not conspicuously wider below than above.In all remaining genera the zygomatic plate is broadened and tilted upwards to a greater or lesser degree; masseter lateralis extends its line of I -Living Rodents -II iWith reftrcnct to the cheekteeth formula of Muridae, I must note that according to some authors, the three cheekteeth represent P.4, M.i, and M.2, and it is suggested that M.3 is suppressed. There seems to be considerable evidence in favour of this view'; but for convenience throughout this work I adopt the customary notation. THE RATS FROM MADAGASCAR I'here are six genera that I have examined, containing seven well-marked species in all, from Madagascar. They are excessively difficult to classify. Most authors have referred them to a subfamily Nesomyinae, whose sole character seems to be "Habitat in Madagascar," or dumped them all in the subfamily Cricetinae. It does not seem to me possible to take either course.The sole character given by Tullberg for the group which is unusual among the Muridae he examined is the fact that the tongue possesses three papillae circumvallatae, but he only examined two genera, .Gymnuromys and Eliitrus. But this character, although in most other Muridae he examined was reduced to one, was present in Cricetomys (three); while two were present in Myospalax and Tachyoryctes. It is not a character which one can use throughout the Order; and is apparently variable in closely allied genera in other groups (for instance, Petaurista, o, Pteromys, 3, in the Pteromys group of Sciuridae). It is probably merely a primitive character which may be met with in any group.The Rats of Madagascar may be arranged as follows, in key form.Skull specialized, Microtine in aspect, with zygo...