The analysed specimen is a holocrystalline medium-grained olivine basalt containing 5–10% of olivine in idiomorphic to subidiomorphic phenocrysts up to 1 mm. long. The olivine is largely altered to greenish-yellow chlorite. The feldspar is labradorite in lath-shaped crystals averaging 0°2 mm. in length. A few small phenocrysts of colourless augite are also present, and the groundmass consists of a fine-grained aggregate of skeletal crystals of ilmenite and magnetite and granules of augite. Thin sections of other specimens of these rocks show similar mineralogical composition, although some are hypocrystalline, with crystals of olivine, augite, and feldspar set in a matrix of brown glass, and others are amygdaloidal, with the amygdules completely or partly filled with calcite, with which analcite also occurs occasionally.

The centre of eruption of these agglomerates seems to have been near Pahau River, because in the outlier of Tertiary rocks in the Dove-Mandamus area they are represented by fine-grained red-brown calcareous tuffs. As igneous material of any kind is unknown in the Tertiary rocks of the Waikari-Weka Pass district, the country on the south side of the Hurunui River was rapidly examined during the course of this survey in order to determine how far this tuffaceous horizon could be followed to the south. The tuff content gradually decreases, and about a mile beyond the river, behind the Waitohi Downs homestead, the formation passes into a calcareous sandstone lithologically resembling the Weka Pass Stone.

Interbedded with the agglomerates in the Pahau River area are lenses of limestone, up to 60 ft. thick in places, but thinning rapidly both in the direction of the dip and of the strike. These limestones are lithologically very similar to the limestone M5 overlying this series, and the number of limestone bands in this very disturbed area caused some difficulty in interpretation during the early part of the field work. They were evidently deposited during lulls in the volcanic activity. Another interesting rock in this series is a curious conglomerate with pebbles of agglomerate set in a limestone matrix.

In places these agglomerates show cavities filled with well-crystallised zeolites; natrolite, apophyllite, and analcite have so far been identified (Mason. 1946) and thorough collecting may yield further species.

In the Pahau area it is difficult to be sure of the thickness of these agglomerates, as they are repeated by folding and possibly also by faulting, but it is at least 150 ft. and is probably about 200–300 ft.; in the Dove-Mandamus area, where tuffs alone occur, the thickness is some 60 ft.

These rocks undoubtedly originated from the submarine eruption of basaltic material, as shown by the nature of the rocks themselves and the occurrence of marine fossils and limestone lenses within them. This formation is apparently conformable with the rocks above and below; the contact with the rocks of M3 is a sharp one, but there is no discrepancy in strike and dip, nor is there any sign of erosion between them, the incoming of volcanic matter marking the junction of the two formations.

Macrofossils are rare in these beds, as might be expected from the nature of the rocks, Fossils have been found at several places in the