Amazonite

01 · The mineral

What is amazonite?

Amazonite is the green to blue-green variety of microcline, a potassium feldspar. It is a variety name rather than a separate mineral species. That distinction matters because an attractive green stone, or even a green feldspar-looking stone, is not automatically amazonite.

Amazonite’s colour is tied to structural defects and trace constituents in feldspar rather than to a simple resemblance to copper minerals such as malachite. For a rare national occurrence, appearance is a starting point. Mineral identity and provenance are the evidence that make the label useful.

02 · A famous wrong name

The 1906 “amazonite” that was probably Aotea

At the New Zealand International Exhibition of 1906–07, the Westland mineral court displayed a deep sea-green specimen from Jacob’s River, South Westland, and the official record later described it as amazonite. The label was real. The mineral identification almost certainly was not.

Only months earlier, a government expedition report had described attractive green stones from the Makawhio/Jacob’s River as “malachite” and recorded their local name as Aotea. Rodney Grapes later traced these historical descriptions and compared them with the known Makawhio material. His conclusion was that the green stone was Aotea — a rock whose colour is dominated by chromian muscovite, or fuchsite, commonly with blue kyanite, quartz and feldspar — rather than either malachite or amazonite.

The importance of the episode is not that early observers were careless. It is that a genuine specimen can preserve a wrong mineral name for generations when colour is allowed to stand in for analysis.

A real historical specimen can carry the wrong name for more than a century.

The Makawhio evidence trail

03 · Westland

The Greenstone River specimen

The strongest modern Westland specimen trail comes from the Greenstone River/Hokonui system. In 2022, geologist Rodney Grapes published a photograph captioned as green amazonite with “rusty” quartz from the Greenstone River immediately downstream of its confluence with French Creek.

The photograph is credited to John Caygill and was supplied to Grapes by Mary Trayes. For this project, Caygill has separately been confirmed as the finder of the specimen. That gives the story an unusually useful human chain: a local collector, a local geological network and a specialist mineralogist who brought the specimen into print.

What the publication does not provide is just as important. It does not give the collection date, present repository, accession number, dimensions or the analytical method used to identify the blue-green material as amazonite. Grapes’s attribution has expert weight; it is not the same thing as a public specimen-level analytical file.

The geology makes the occurrence plausible. French Creek lies within the Hohonu Batholith, where granitic rocks contain alkali feldspar and late-stage features. Grapes describes Westland amazonite as float derived from granite. The photographed piece was therefore transported material: the precise bedrock body that produced it has not been traced in the published record reviewed here.

04 · Geography as evidence

Greenstone Creek is not one simple old name

Historical place names around the Hohonu waterways are unusually treacherous. Modern Greenstone River/Hokonui is tied to the former name Greenstone or Big Hohonu River. But nineteenth-century writers did not use “Greenstone Creek” consistently.

An 1893 official report explicitly calls Greenstone Creek the Little Hohonu River in one section, while elsewhere in the same report it refers to “Little Hohonu and Greenstone Creeks” as separate streams. Julius Haast’s 1865 account also distinguishes Greenstone Creek proper from Big Hohonu, while an 1896 official report calls the latter the Big Hohonu or Greenstone River.

This matters because an old specimen label saying only “Greenstone Creek” cannot safely be dropped onto a modern map. The date, writer, mining claim and map context have to travel with the name. Eastern Hohonu is also a separate river, not another label for the Greenstone River specimen locality.

05 · Fiordland

Upper Lill Burn: the strongest official occurrence statement

Jocelyn Thornton’s 1985 Gemstones booklet provides the earliest modern New Zealand amazonite statement located in the project research. She wrote that blue-green microcline amazonite had “recently” been found in Fiordland in small pieces. She did not name the collector or stream.

Twenty-five years later, the GNS Science Geology of the Fiordland area QMAP made the locality and geological setting much more precise: amazonite occurs as float in the upper Lill Burn catchments and is described as derived from pegmatite dikes.

That is the most authoritative modern occurrence statement found in the accessible record. The word float still matters. It describes loose transported material rather than a specimen fixed in its source dike. The published interpretation points back to pegmatite dikes, but the exact source dike and its specimen-to-float match are not given in the text reviewed here.

06 · Open files

Three leads that could change the story

University of Otago. A reported specimen lead, sometimes described as OU 15575 and associated with a nineteenth-century “Greenstone Creek” claim, remains unverified. This needs the actual catalogue card and label before it can be treated as a New Zealand amazonite specimen. The caution is especially important because an 1897 University report proves its Mining Museum held imported American “Amazon stone”.

Eastern Hohonu / PETLAB. A reported 2025 sample attributed to “M. Falloon” could become a very important modern record if its database entry contains photographs and mineral-specific analysis. The target PETLAB record was not retrieved in the research used here, so this page does not invent an identifier, coordinate or analytical result.

Lake Brunner and Brown River. A photographed book page supplied to the project says amazonite is “reported around” Lake Brunner and Brown River near Greymouth. The book title, author and evidence chain have not yet been identified, so this remains a secondary lead rather than a confirmed locality.

07 · Identification

What would make a New Zealand specimen definitive?

For a common mineral, a field label can be enough for casual discussion. For a rare national occurrence with a history of misidentification, a better standard is useful. A robust amazonite record would connect the object, its locality and its mineral identity.