The Bunyaruguru volcanic field, referred to in some publications as Kichwamba, lies along the eastern side of the rift valley south of Lake George and extends over an area of 33x7 km. The field consists principally of tuffs that extend over the rift valley floor and partly onto higher ground to the east, together with rare occurrences of lava. Numerous explosion craters have been blown through the tuffs, and there is one isolated crater, Kasenyi (Holmes and Harwood, 1932), that lies to the northwest of the main field. In the northern part of the field the tuffs grade into Pleistocene to Recent sands, silts and clays, and to the south the tuffs overlie Precambrian schist, quartzite and shale (Reece, 1953). A borehole passed through about 100 m of tuff, but it is likely that these are very variable in thickness. The tuffs are generally fine-grained and highly calcareous. In thin section they reveal sub-angular to rounded fragments up to 0.5 mm across in a semi-opaque matrix, making up to 80% of the rock, which is largely calcite. The fragments include quartz, muscovite, biotite, olivine, plagioclase, ilmenite and volcanic glass. There are also oolitic structures in the tuff the centres of which are usually calcite. Fragments as much as several centimetres across, which are usually angular, include quartzite, shale, granite, gneiss, schist, amphibolite, dolerite and biotite crystals together with occasional pieces of lava. From the distribution of these fragments Reece (1953) was able to map the basement geology beneath the volcanic rocks. There are also mineral fragments and xenoliths of a more deep-seated origin including augite, biotite and olivine and the rock types pyroxenite, biotite pyroxenite, glimmerite and biotite peridotite (Combe and Holmes, 1945). Lloyd and Bailey (1975) and Lloyd (1981) have described ultramafic nodules from several localities of Bunyaruguru that are essentially clinopyroxenites with abundant mica and amphibole which are interpreted to be the products of mantle metasomatism. Analyses of clinopyroxene and mica in pyroxenite xenoliths from three craters are given and discussed by Lloyd et al. (1999). Occasional large bombs with associated sag structures are to be found. Reece (1953) considered that the tuffs were deposited in water. On the inside wall of the Kigezi crater large masses of a banded magnesian limestone outcrop, and outcrops of limestone have been found at four other craters. Papers describing these occurrences are reviewed by Reece (1953) who points out that limestone is not known in the Karagwe-Ankolean basement rocks. A carbonatitic origin is a possibility. Lava flows have been identified at four craters, and a further occurrence may be a volcanic neck, which is truncated by a crater, while blocks of lava occur within the tuffs at a number of localities. Holmes in a number of papers (e.g. Combe and Holmes, 1945) has classified the volcanic rocks of Bunyaruguru into the following major types: ugandite (augite + leucite); mafurite (augite + kalsilite); katungite (melilite + leucite glass + leucite); kalsilite-katungite (melilite + kalsilite). All four types also contain abundant olivine, perovskite and opaque phases, sometimes with biotite and glass. Depending on the mineralogy various intermediate varieties are distinguished (Holmes, 1945, Table 1). Lava which forms a flow on top of the tuff cone Kazimiro is a melilite ugandite (Holmes, 1945). An extensive area of rock in the northern wall of the large Chamengo crater surrounded by tuffs may be a volcanic neck, and specimens described by Holmes (1942) proved to be mafurite, leucite mafurite and kalsilite ugandite. A lava flow among the tuffs of the Mafura craters, and blocks among the overlying tuffs, are the type locality for mafurite, with some samples from the flow proving to be leucite mafurite (Holmes, 1942). In the mafurite Holmes discovered the mineral which Bannister and Hey (1942) named kalsilite. Northwest of Nyondo (Nyungu) crater, near the foot of the slope marking the edge of the rift depression, three flows, the upper and lower Kabirenge and the Lyakauli, are exposed (Combe and Holmes, 1945). These proved to be mafurites and ugandites with intermediate varieties and some types containing melilite. Reece (1953) later discovered another flow near Kabiswa which was identified as a ugandite. Detailed petrography, not only of the lavas but also of the associated tuffs and the blocks they contain, mineralogy and chemical analyses will be found in the papers given above, in Holmes and Harwood (1932) and in others by Holmes cited therein. Many of the specimens collected by Combe and described by Holmes have subsequently been subjected to further detailed study. For instance, Bell and Powell (1969) investigated strontium isotopes, Mitchell and Bell (1976) report whole rock REE data, Higazy (1954) trace element data and Bell and Doyle (1971) K and Rb abundances. Among the numerous xenoliths in the tuffs of the Bunyaruguru field are basement granites, examples of which from the Kariya crater have been transformed metasomatically to leucite, and in some cases leucitite and olivine leucitite, as described in detail by Holmes (1945). Samples from two glassy olivine melilitite bombs from the Chabachu crater have been investigated in one atmosphere melting experiments by Lloyd (1985), while Arima and Edgar (1983) carried out high pressure studies on a biotite mafurite from Nyungu crater, the mineral chemistry of this rock being described by Edgar (1979).