Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 80, 1952
This text is also available in PDF
(2 MB) Opens in new window
– 323 –

Further Notes on Pterostylis R. Brown and Thelymitra J. R. and G. Forster

[Read before the Auckland Institute October 24, 1951; received by the Editor November 22, 1951.]


This paper records the re-discovery of Pterostylis arcolata Petr. and includes an amended description and illustration from living material It records the Australian Pterostylis cycnocephala R. D. Fitzg. from New Zealand for the first time and illustrates this species and Pterostylis mutica R. Br. for comparison. An unusual form of Pterostylis mutica is also noted. The writer retracts his Pterostylis furcata var. typica and substitutes therefore Pterostylis furcata var. micromega (Hook. f.) Hh. Thelymitra pulchella Hook. f. is recorded for the first time from the South Island.


The writer is grateful to Mr. H. Talbot for living material of Pt. areolata, Pt. mutica and Pt. cycnocephala; to Miss A. Lush for the loan of specimens and drawings from the Dominion Museum; to Mr. J. Bruce Irwin for his excellent drawings of Pt. mutica and cycnocephala; and to Mr. K. W. Allison for spirit specimens of Th. pulchella from Herbert.

Pteroslylis areolata Petrie has been discovered in flower on Mt. Torlesse, in the Waimakariri Valley, Canterbury. This is only the third time since Kirk discovered the species, in 1890, that flowering specimens have been collected, and the following description is the first to be drawn up from living material.

Pterostylis areolata Petr. Trans. N. Z. Inst., 50, 1918: 210; Cheeseman, Manual N.Z. Flor., 2, 1925. 351; Hatch. Trans. R.S.N.Z., 77, 1949: 241, t25.

Amended description based on the holotype and on living material from Mt. Torlesse.

Up to 18 cm. high Leaves 1–6, up to 6 cm. long by 2 cm. broad, obovate through oblong to elliptic-lanceolate, acute. Floral bract foliaceous, elliptic, acute. Flower proportionately large, mclined to be squat and rounded, green with reddish striae, up to 4 cm high Dorsal sepal acuminate, a little longer than the broad petals, meuryed. Lateral sepals with a narrow sinus, the lobes acuminate, shorter than or hardly exceeding the galea. Labellum narrowlinear, flat, the tip recurved and varying from symmetrical to unevenly constricted. Column typical, stigma obovate. Column-wings with the upper lobes subulate, shorter than the anther. Lower lobes narrow-oblong, twisted, finely pubescent, hardly reaching the stigma. Anther connective suberect.

There are two types of labellum. One (Fig. F) unevenly constricted and characteristic of foliata and the australis-montana section of the australis group; the other (Fig. E) symmetrically flat and characteristic of the banksii-oliveri section of the australis group This occurrence of the foliata labellum adds proof to the writer's theory (expressed in a text-figure in Trans. R.S.N.Z.,

– 324 –

77, 1949: 239) that the oliveri section to which areolata belongs, is fairly close to the ancestral foliata stock.

There are also two types of leaf arrangement. One (Fig. A) with the lower 2–3 leaves arranged in subrosulate fashion round the base of the stem; the other (Fig. C) with the same leaves scattered up the stem. These two forms occur indiscriminately in any one colony without relation to the environment. The two types of labellum also occur at random in both the rosulate and cauline-leaved forms. The flower is identical in all cases, with a tendency for the lateral caudae to recurve as in banksii patens

Pt. areolata appears to have been originally derived from oliveri (which also occasionally produces cauline-leaved forms) with a subsequent admixture from banksii patens. This latter species often sports the foliata labellum, usually but not always by crossing with australis. There can, however, be no question of crossing in areolata. Whatever its ancestry it appears to be a valid species, occurs in isolation over a wide area, and breeds true to type.

The juvenile is a remarkably symmetrical rosette of 2–5 obovate, oblong or elliptic, petrolate leaves. The peduncle elongates with maturity, tending to scatter the leaves up the stem.

Note on the Mt. Torlesse locality: “Areolata occurs in a number of isolated colonies, over a distance of 5–6 miles with some considerable ridges from 3,000–4,000 feet high in between. The only other Pterostylid in the area is one colony of banksii patens. which occurs in association with areolata under manuka scrub at 1,600 feet. The remainder of the colonies occur in the open tussock Both labellum types and both leaf forms are indiscriminately mixed.”

Distribution.: Endemic in the South Island of New Zealand. Localities detailed below.


1. Bealey, Waimakari Valley, Canterbury; 2,500 feet, about 1890; T. Kirk. One seeding and one juvenile specimen mounted with the holotype in the Dominion Museum, and figured in Trans. R.S.N.Z., 77, 1949: t25, Figs. B and C. The label, which is in Kirk's writing, identifies the specimens as Pt. micromega. Hook. f, across which Petrie has written “probably areolata.” They appear to belong to the cauline-leaved form of areolata, but without the flower this can only be a tentative opinion. They are certainly not micromega, which has not so far been found in the South Island.

2. Base of Shingle Peak. Awatere Valley, Marlborough; 3,000 feet, in shade; 12, 1917; L. Cockayne. One flowering specimen, the holotype in the Dominion Museum, from which, in the dried state Petrie described Pt. areolata This specimen also belongs to the cauline-leaved form, and is figured in Trans. R.S.N.Z., l.c.; Fig. A.

3 Poulter River, Waimakariri Valley, Canterbury; 2,500 feet, in grass; 12, 1919; A. Wall Two flowering specimens of the rosulate-leaved form in the Auckland Museum (Sheets No. 24659/60). Wall sent them to H. B. Matthews tentatively labelled “australis”. Matthews softened them up, leaving the flowers distorted, wrote “what is it?” across the label in pencil, and referred them to Cheeseman. The present writer shelved them in 1945 as a potential new species pending living material, but in the light of the Mt. Torlesse specimens they now prove to be areolata.

– 325 –

4. Bealey, Waimakariri Valley, Canterbury. In Nothofagus forest; 1, 1928; W. R. B. Oliver. One seeding specimen in the Dominion Museum. Identical with Kirk's original material and probably the cauline-leaved form of areolata.

5. Mt. Torlesse, Waimakariri Valley, Canterbury; 1,600–3,000 feet; 11, 1950–51; H. Talbot. Specimens of rosulate and cauline-leaved flowering plants and juveniles deposited in the Auckland Museum.

Pterostylis cycnocephala R. D. Fitzgerald Austr. Orch, 1, 1876; 2.

Distribution. Australia: N.S.W., Victoria, South Australia and Tasmania; nowhere common. New Zealand: Canterbury, Kowai River, Mt. Torlesse, 11, 1950–51; H. Talbot. New Zealand specimens in Herb. Hatch and in the Auckland Museum.

With the Pt areolata material, Talbot sent living specimens of what appeared at first glance to be two distinct forms of Pterostylis mutica R. Br. One of these later proved to be the Australian Pterostylis cycnocephala R. D. Fitzg., which is here recorded from New Zealand for the first time. These two species are superficially so very similar that in a revision of the whole genus the writer would be tempted to reduce cycnocephala to a variety of mutica The former has, however, been accepted in Australia for so very many years that it seems best to follow suit and treat it in the meantime as a valid species. The differences between the two, while inconspicuous, are quite constant, and are tabled below Fitzgerald himself, in the original description, was not quite sure in which taxonomic category the plant belonged, and was inclined to blame his indecision upon the lack of any definite taxonomic rules Actually the rules are no less vague at the present time, and the category in which a particular new species is placed is largely a matter of personal opinion Irwin's drawing of cycnocephala, reproduced here, corresponds exactly with Fitzgerald's original plate.

The particular form of mutica which Talbot sent from Springfield differs in minor characters from the form which occurs in the North Island (compare Irwin's drawing in this paper with Trans. R.S.N.Z., 77, 1949: 236; t19; f3k.) The floral morphology is the same in both jordanons, but the Springfield form is very much redder and more slender than that collected by the writer from the Waiourn Hills in 1944, differences more “obvious to the eye than capable of measurement or delimitation.”

Cycnocephala mutica
Robust. pale green, with large flowers up to 12 mm. high by 9 mm. broad. Slender, lax, reddish-brown to greenish-grev, with small flowers up to 5 mm high by 4 mm broad.
Galea and column curving considerably. Galea and column almost erect.
Lower lobes of the column-wings with densely crliate margins. Column-wings with only sparsely e [ unclear: ] hate margins.
Stigma rather narrow, midway on the column. Stigma almost cordate, high on the column.
Labellar appendage recurved and pointing outwards. Lebellar appendage more or less straight and pointing inwards towards the base of the column.
– 326 –

Pterostylis furcata Lindley.

Shortly before his death W. H. Nicholls wrote to me from Melbourne, pointing out an error I had made in my classification of the New Zealand forms of Pterostylis furcata Ldl. I cannot do better than quote his letter.

“… the reprint of your paper on Pterostylis. I have only one adverse comment to make—your plate … and the specimens … constitute the first decent material and the first accurate illustrations I have seen of the New Zealand forms of Pt. furcata. You are right up to a point—there can be no doubt that the New Zealand plant is furcata—but you are not correct in assuming the large form to be var. typica. There appear to be at least three forms of this plant. The typical variety, both taxonomically (the original furcata was discovered in Tasmania by Ronald Gunn over a hundred years ago), and in your theory phylogenetically, is confined to Tasmania. Another form occurs in South Australia, Victoria, and on Mt. Kosciusko, in New South Wales. It is this latter form that you and Rupp considered identical with the large New Zealand plant. (Proc. Linn. Soc. N.S.W., 70, 1945: 59.) If you will compare the enclosed illustration (taken from Vict. Natr. 65, 1949: 256, t) which I drew up from living specimens of the Tasmanian plant, with your own illustrations and specimens, you will see distinct variations in flower size, the nature of the stem bracts, the stigma and labellum … I would suggest resurrecting Hooker's micromega as a variety, and bracketing it provisionally with the Australian mainland form. The third form is, of course, your var. linearis.”

H. M. R. Rupp's reaction to my treatment of furcata was much the same, although he was not prepared to be definite on the exact nature of the Tasmanian form, since he had not seen living specimens. He was, however, quite definite that the large New Zealand jordanon was identical with the Eastern Australian plant.

With my own and the Auckland Museum material, I had a fairly comprehensive range of specimens, photographs and drawings of furcata from most of the recorded Australian localities, and I subsequently checked and confirmed Nicholls' letter in detail. The Tasmanian and the large New Zealand forms of this species are as distinct from each other as are the various forms of Pt. trullifolia Hook. f., and in my opinion warrant varietal rank. The differences are tabled below:

Tasmania New Zealand
Flower proportionately small for the height of the plant. Flower rather large.
Stem bracts short, distant. Stem bracts long, often but not invariably overlapping.
Stigma oblong-obovate, slightly constructed at the centre. Stigma elliptic.
Labellum narrow acuminate, not twisted. Labellum linear-oblong, the upper third symmetrically twisted.

Art. 17 of the International Rules states in part: “No one may change a name … without serious motives, based … on more profound knowledge of facts…” I think this last phrase covers the case. The Tasmanian plant is var. typica, and the valid names of the two New Zealand jordanons then become—

(i) Pterostylis furcata Ldl. var. micromega (Hook. f.) Hh. no. comb. syn. Pt. furcata var. typica Hh. Trans. R.S.N.Z., 77, 1949; 242 which see for descriptions, illustrations and additional synonomy.

Picture icon

A—Pteroslylis areolata, natural size Subrosulate leaved form.
C—Cauline leaved form.
D—Lateral sepals.
E, F—The two labellum forms.
H—Column from side.

Picture icon

Pt. cycnocephala
A—Plants in situ.
B—Column and labellum from side.
C—Column from front.
D—Column from side.
E, F—Labellum from front.
G—Labellum from side.

Picture icon

Pt. mutica
A—Plants in situ.
B—Column from front.
C—Column and Labellum from side.
D—Labellum from front.
E—Labellum from side.

– 327 –
Picture icon

A—Pterostylis furcata var. micromega (New Zealand).

Picture icon

1—Pterostylis furcata (Tasmania).

(ii) Pterostylis furcata Ldl. var. linearis Hh. Trans. R.S.N.Z., l.c.

Thelymitra pulchella Hook. f. has not previously been recorded further south than about the middle of the North Island (Waitaanga).* It has now turned up near Herbert, about 60 miles north of Dunedin, an extension of range almost as peculiar as that of Th. venosa var. cyanea which occurs at Kaikohe and in Stewart Island and not between.

Th. pulchella Hook. f. Herbert, 1,000 feet, in grass; 12/1/1952; K. W. Allison.

[Footnote] * Th. pulchella has since been recorded from Wellington—The Puffer and Kaitoke. Wgtn. Bot. Soc. Bulletin, 25.1951.10, Druce & Irwin.