The Late Palaeozoic comprises the Carboniferous and Permian periods, and lasted from about 355 to 252 million years ago. It was brought to an end by one of the three major extinctions to have affected the Earth, at the end of the Permian, known as The Great Dying - or more technically, the P-Tr Extinction.
Elsewhere in the world, the Carbonifeous priod has left an important mark in the form of extensive sandstones, limestones and - as reflected in their name - coal measures, most notably in Europe and the USA. In New Zealand, by contrast, Carboniferous rocks are almost unknown. They are so sparse and localised that they are not even shown on the1:1 milion geological map of the country. The only outcrops of any note occur in south Canterbury (near Kākahu) and north Otago (near Dunback), but the Top of the South does have a small area of granodiorites, granites and gneiss that have been attributed to the Carbonifeous - though not of any interest to fossil hunters, of course.
Permian rocks, in contrast are extremely widespread. Two main sets of rocks developed during this period. The Torlesse group, which makes up the basement rocks of much of western South Island, mainly formed in deeper waters, and comprises greywackes that are somewhat lacking in fossils. The Murihiku supergroup formed in shallower water on the continental shelf and its outward slopes,and makes up much of the so-called eastern province. This comprises a number of smaller terranes, each of which formed in a somewhat different setting, where rather different rocks were laid down. In the Top of the South, they include the Brook Street, Caples, Dun Mountain and Maitai terranes. It is from these rocks that most of the fossil finds from the Permian derive.
Thge end of the Permian has moved around a bit over the years. Thjis may come as something of a surprise; we might expect an event like the Great Dying to be obvious in the geology. In the Nelson area, that's far from true. For many years, the boundary was positioned at the top of the Stephens Formation which crops out in the Pig Valley area, to the west of the Wairoa River, between Brightwater and Wakefield. Some years ago, however, isotope analysis of a sequence through the Little Ben Sandstone, in the upper Lea Valley, suggested that carbon levels suddenly fell in strata some way below the previously defined boundary.This was assumed to reflect the sudden loss of organic carbon in the world, during and immediately after the extinction event, and on that basis was considered to mark the real P-Kr boundary. Accordingly, the Triassic was extended several hundred metres downwards in the geological sequence, and several kilometres eastwards on the ground, to the new boundary within the Little Ben Sandstone. In many ways, this moves accords better with the fossil evidence, for Atomodesma - a common and diagnostic Permian bivalve - disappears within the Little Ben Sandstone, and afterwards the rocks remain generally lacking in fossils. There are, however, a few tantalising occurrences where Permian species, including Atomodesma, seem to pop up in what are now regarded as Triassic rocks. Some of these are explained as olistostromes: blocks of older rock that have become caught up in younger rocks, as they were deposited, either as a result of mass movement (e.g. landslides or slumping of material down the continental shelf) or during tectonic activity. Whether this explains all the anomalies, however, is uncertain.
Wherever the boundary is, Permian fossils can certainly be found, and in some formations they are abundant. Over 130 species of brachiopods, for example, have been reported in New Zealand, especially from the so called Productus Creek group (part of the Brook Street terrane). Finding fossils in the Top of the South area is nevertheless devilishly difficult and frustrating, and those that are found are often distorted and damaged and difficult to identify. A measly three taxa are thus shown here - and one of those is a trace fossil. But the hunt goes on!
Further Reading: Campbell H.G. 2019 Biostratigraphic age review of New Zealand’s Permian–Triassic central terranes. Chapter 3 in: Robertson, A. H. F. (editor) Paleozoic–Mesozoic geology of South Island, New Zealand: subduction-related processes adjacent to SE Gondwana. Geological Society, London, Memoirs49, pp. 31–41. MacKinnon, T.C.1983 Origin of the Torlesse terrane and coeval rocks, South Island, New Zealand. Geological Society of America Bulletin 94, 967-985. Waterhouse, J.B. 1964 Permian stratigraphy and faunas of New Zealand. New Zealand Geological Survey Bulletin 72, Wellington: New Zealand Department of Scientific and Industrial Research.