Contributions to a Knowledge of the Naturalized Flora of New Zealand. No. 2.

Transactions and Proceedings of the Royal Society of New Zealand, Volume 77, 1948-49, Page 172

Contributions to a Knowledge of the Naturalized Flora of New Zealand. No. 2. By A. J. Healy, Public Works Department, Blenheim. [Received by the Editor, February 26, 1947; issued separately, April, 1948.] The bulk of the material contained in this paper was prepared while the writer was on the staff of the Botany Division, Plant Research Bureau, Wellington, and thanks are extended to Dr. H. H. Allan, Director of that institution for criticism and assistance. Thanks are due to Dr. W. R. B. Oliver, Dominion Museum, Wellington, for permission to examine certain critical specimens, to Miss B. E. G. Molesworth, Auckland Institute and Museum, Auckland, for the loan of specimens from the Cheeseman Herbarium, and to Mr. J. H. Lousley, Streatham Common, London, for the determination of Rumex sagittatus Thunb. In this paper, I have treated a number of species, the correct determination of which I have been unable to make, due to inadequacy of literature and lack of correctly identified overseas specimens; it was considered that the conclusions reached in respect of the various species would be of value to workers on the naturalized flora, since some time will elapse before the critical species can be examined by specialists in the different groups. The numbers quoted refer to herbarium sheet numbers in the herbarium of the Botany Division, Plant Research Bureau, Wellington, and unless otherwise specified, the specimens were collected by the writer. Ranunculaceae. Ran unculus bulbosus Linn. Originally recorded for New Zealand by J. F. Armstrong (1872:287) from Canterbury, the identity and distribution of this species appears to have been misunderstood by local botanists. Allan (1940:48) was the first to question the apparent wide distribution of the species, stating “Seen in only one locality by me.” I have examined a number of specimens in New Zealand herbaria under this name, and find that all belong to the closely allied R. sardous Crantz.; further, over a ten-year period I have collected Ranunculi from localities in both islands, but have not yet found any specimens which belong to R. bulbosus Linn. Until Armstrong's original specimens on which the record is based are found and proved to be correct, I consider that R. bulbosus Linn, should be regarded as a very doubtful member of the naturalized flora, and should perhaps even be removed therefrom. There are difficulties in the determination of the species, and I have found the bulbous character of the rootstock not to be entirely satisfactory, since New Zealand material of R. acer Linn. and R. sardous Crantz. often shows swollen rootstocks similar to those possessed by foreign specimens of R. bulbosus Linn. in local herbaria. Achene characters appear to offer an additional means of distinguishing

R. sardous Crantz. and R. bulbosus Linn., in the former the achene surfaces bear an almost peripheral ring of tubercles, whereas in the latter species the peripheral ring of tubercles is absent. Ranunculus (Ceratocephalus) sp. This plant was first gathered from Kurow, Bald Hill Flat, and other localities in Otago by D. Petrie about 1885, and the first published record of its occurrence that I have been able to trace is that by Kirk (1899: 20) as a naturalized species under the name R. falcatus Linn. Petrie must have originally considered this plant as indigenous, for his labels bear the name Ceratocephalus laniger nov. sp., but there is no evidence that he published the species. This plant has been treated as naturalized by Cheeseman (1906: 1063 and 1925: 1064) and Allan (1940: 50). I have examined Petrie's specimens from Bald Hill Flat and gatherings by subsequent botanists in the Otago region, and find that our plant is certainly not R. falcatus Linn., differing markedly from that species in growth habit, shape of the petals and achene conformation. To the present, I have been unable to match this plant with any species in the section Ceratocephalus of the genus Ranunculus. The problem is an interesting one, in that species of this particular section are stated to be endemic in Central Europe and Asia, and according to the Index Kewensis there are only two valid species; descriptions of these are available, but the local plant agrees with neither. Thus, it is possible that this plant could be indigenous to New Zealand and is an undescribed species, or, alternatively, that it is an alien species which has been described only recently. Argemone mexicana Linn. var. ochroleuca (Sweet.) Lindl. All the specimens of this species in the herbarium of the Plant Research Bureau belong to this variety, characterized by the flower colour and the distinct styles, the typical form of the species having the stigmas sessile. The material examined included: Henderson, Auckland, Long 17788; Rotonui, Te Puia Springs, F. W. J. Fox 5267; Wanganui, A. R. Dingwall 36271; Havelock, T. H. Munro 3885; Oamaru, H. H. Allan 7763. Cruciferae. Cardamine pratensis Linn. Originally recorded from Whangarei by Allan (1940: 67), I have found specimens in the Plant Research Bureau Herbarium (17295) collected from New Plymouth about the same time as the original collection from Whangarei, these having been omitted in making the record. The species has been found to be established in damp places along the Waihou River, near Matatoki 45938, and near the head of the Te Awaiti drainage canal 45939, both localities in the Hauraki Plains district. Lepidium sp. The species of this genus are difficult of determination, and the species naturalized in New Zealand have yet to be critically examined. To the present, eight species have been recorded for New Zealand, but the correct identity of certain of these is doubtful. This study has not produced a very clear picture of the naturalized species, due to the close relationships of the species as treated by Thellung in his monograph of the genus, and the lack of authentically named specimens for comparison purposes. The individual species are treated below.

L. bonariense Linn. Recorded by Allan (1935: 4) for Wellington, this plant is not uncommon on gravel ballast along the railway line between Wellington and the Summit, Rimutaka Range. L. campestre (Linn.) R. Br. Recorded by Thomson (1875: 372) from Otago. This species is correctly determined and the distribution is as given in local floras. L. draba Linn. Recorded by Smith (1904: 225) from Ashburton County. This species is correctly determined and the distribution is as given in local floras; it is thoroughly established in parts of Canterbury and Marlborough. L. heterophyllum (D.C.) Benth. Recorded by Kirk (1878a: 414) from Bendigo, Otago, as L. smithii Hook. This species has been confused with L. campestre (Linn.) R. Br., and it must be regarded as very uncommon to rare; I have seen only one specimen from the Canterbury district. L. neglectum Thell. Recorded from Central Otago by Allan (1940: 62). I have examined these specimens, and consider them to be more correctly placed under L. densiflorum Schrad. Working with Thellung's descriptions, there appears little difference between the two species, and this suggests that Thellung's species is identical with one of the varieties of L. densiflorum Schrad. L. ruderale Linn. Recorded by Kirk in Hooker's list of naturalized species (1867: 758), the plant to which the name was applied is widely distributed. Examination of specimens under this name in the Plant Research Bureau Herbarium indicated that these plants did not match the description of the species given by Thellung (1906: 135–39) in his monograph on the genus Lepidium. These specimens have, without exception, four rudimentary petals, varying from one-eighth to one-half the length of the sepals, four glands (“Honigdrüsen” of Thellung) about one-tenth the length of the sepals, and sepals broadly oblong, usually a dark purplish colour with long hairs on the outer surface. In his description of the typical variety of L. ruderale Linn., Thellung (1906: 137) states “…petala nulla…” and of a Siberian variety “…(in var. intercedente petalorum rudimenta interdum praesentia)…” and is thus definite as to the presence or absence of petals. He further states in a footnote: “Zeitweises Verkommen von Kornblättern [Kronblättern] wurde irrig angegben von:… Tatsächlich habe ich weder in der Natur noch in den Herbarien biem europäischen Typus der Art je Kronblattrudimente finden können; die Angabe solcher beruht also wohl auf der Verwechslung mit anderen Arten oder auch darauf dass die Honigdrusen für rudimentäre Petalen angesehen wurden (diese werden oft als gelblich geschildert).” Our specimens do not match the description of var. intercedens Thell., so it would appear that L. ruderale Linn. as delimited by Thellung is probably absent from New Zealand. It can be noted that Thellung (1906: 135) queried the validity of Kirk's record of the species in The Student's Flora of New Zealand, stating, “Aus dem australisch-polynesischen Gebiet sah ich noch kein echtes L. ruderale; es ist aber nicht ausgeschlossen, dass sich die Pflanze dort eingeschleppt findet.” There are several distinct plants included in our concept of the species, one of which appears to be close to

L. hyssopifolium Desv. Further collecting of plants of this complex is desirable, and will assist in the problem of clearing up the distribution of the various plants. L. sativum Linn. Recorded by Hooker (1864: 15), the species is occasional; it has appeared in linen-flax crops sown with imported seed during the recent war period. L. virginicum Linn. Recorded from Auckland and Whangarei by Allan (1940: 61), this species has been found in other localities: Edgecombe, Bay of Plenty, K. W. Allison 36685; Pongakawa Valley, Bay of Plenty, K. W. Allison 36691; Te Puke, 33777; Opotiki, N. Potts (?) 8310: 7780; Ongarue, King Country, 51049. The above specimens have the cotyledons accumbent, but there are a number of specimens from other localities with cotyledons oblique to incumbent, but identical with the above in all other respects, and which appear to belong to this species. Sinapis dissecta Lag. This species has appeared in linen-flax crops, Invercargill, A. Stuart 45185. Sisymbrium irio Linn. Recorded by Allan (1933: 38–39) from Auckland, the record is presumably based on specimens collected by H. Carse from a garden at Onehunga, Auckland, and represented by sheets numbered 5306, 5310 and 5311 in the herbarium of the Plant Research Bureau. These specimens are all S. orientale Linn., a species also recorded for the first time in the same paper by that author. It must be noted, however, that in the plate of drawings in Allan's paper, the representations of leaf and siliqua are correct for S. irio Linn., but they could obviously not have been made from the Auckland material, but were probably done from a specimen grown in New Zealand from Russian seed (15030). I have examined the Sisymbria in the main New Zealand herbaria, but have not seen any specimens of S. irio Linn., nor has it been noted in the field. Under these circumstances, I consider that this name should be removed from the naturalized flora. S. orientale Linn. The distribution given by Allan (1933:39–40 and 1940: 65) is for the North Island only, and it has now been collected from several South Island localities: Seddon, Marlborough; Kaituna, Banks Peninsula; Templeton, G. Proudfoot; Lyttelton, H. H. Allan 45184; Rangitata Island; Timaru, R. Mason 33749, 33945, 33946. Polygalaceae. Polygala myrtifolia Linn. Recorded from North Auckland and near Napier by Allan (1940: 69), this species has been found established under plantations at Tauranga 33795. Crassulaceae. Crassula colorata (Nees) Osten. Recorded by Healy (1946: 399) from near Blenheim, the species has been noted as established on stony river terraces, “Blairich Station,” Awatere Valley, Marlborough. C. decumbens Thunb. Cheeseman (1883: 282) recorded from Penrose, Auckland, a plant he provisionally determined as Tillaea (Bulliarda) trichotoma Ecklon et Zeyher, a determination accepted by Kirk (1899: 141), who used the name T. trichotoma Walp., a course adopted by subsequent workers on the naturalized flora. Specimens

from various North Island localities, including Cheeseman's gathering from Penrose, have been examined, and using Berger's treatment of the genus Crassula (1930: 388) these specimens work out to C. macrantha (Hook. f.) Diels et Pritzel, an Australian species. This Australian species and the South African C. decumbens Thunb. are closely allied, and local specimens worked through Australian and South African floras come to these species. Berger (1930: 388) differentiates the two species thus: C. decumbens Thunb.—“Karpelle vielsamig.” and C. macrantha (Hook.f.) Diels et Pritzel—“Karpelle 6-samig.”, implying that the number of seeds in the carpel of the former species is greater than six, but without specifying the number. New Zealand specimens agree with the description of C. macrantha (Hook, f.) Diels et Pritzel given by Ostenfeld (1918: 40–42) in his revision of the West Australian Crassulae, matching exactly in such important characters as number of seeds per carpel and shape and dimensions of nectary scales relative to petals, sepals and carpels. Neither Harvey (1861–62: 330) nor Schonland (1929: 161) give details as to the nectary scales or number of seeds per carpel in C. decumbens Thunb., although the latter says of the species “…ovaries multi-nucleate,” this in apposition to “…ovaries 1–4-ovulate” for C. langeburgensis Schonl. A critical examination of the Australian and South African species involved appears warranted, since it would appear that Berger did not critically examine, but rather incorporated the data of Ostenfeld and Schonland in his account.* While investigating the naturalized Crassulae, a series of specimens of the indigenous C. sieberiana (Schult.) Druce were examined, and it was found that in all the specimens four linear-clavate nectary scales were present in the flowers; this finding is contrary to the published descriptions of this species by New Zealand botanists and by Bentham (1864: 451). This raises a question as to the identity of this species, and it may be that it is not the true C. sieberiana (Schult.) Druce and is actually a plant without a name. Such a position obtained in West Australia, and Ostenfeld (1918: 43–45) separated out C. miriaemae Osten., a species with 5-merous flowers, from the complex. Caryophyllaceae. † Denotes species recorded for the first time for New Zealand.Moenchia erecta (Linn.) Gaertner-Meyer-Scherbius. Established in modified tussock grassland, Lowry Peaks Range (“Palmside,” “Ben Lomond,” “Hermitage,” “Mount Palm”); in pasture, stony flats, between Culverden-Waiau Road and the Lowry Peaks Range. † Denotes species recorded for the first time for New Zealand.Sagina nodosa (Linn.) Fenzl. This distinct species has been found growing in mats of Carmichaelia enysii Kirk, Naseby, Central Otago, G. Simpson 33869. Silene dichotoma Ehrh. This species was first recorded by Allan (1940: 77), based on specimens from Wellington. I have examined a series of specimens from Wellington and suburbs, and find that our plant does not belong to this species, differing in the inflorescence, shape and length of the petals, shape of the capsule, features of the seed, and in the styles being contained within the corolla. The correct identity of the species is not yet known, since Rohrbach's “Monographie der Gattung Silene” is not available in New Zealand, and William's revision of the genus (1896: 1–196) does not give a complete key to the species or descriptions of all the species.

Portulacaceae. Calandrinia caulescens H. B. K. The material assembled under this name in New Zealand herbaria consists of two distinct plants, which appear sufficiently different to belong to two species; the literature available on this genus is not adequate to determine this apparently new species. Further collections of Calandrinia from various localities are desirable, so that the distribution of the two types can be worked out. Montia minor Gmel. Cheeseman (1883: 278) recorded a Calandrinia sp., stating: “A small white-flowered species of this genus, which I have been unable to identify, has become plentiful in stony places by the South Road near Penrose, and thence to Onehunga. (South America?).” To the present, neither this record nor the species has been cited by subsequent workers, although Cheeseman's specimens have been available. I have examined a portion of the original Penrose specimens (3452), and find that the plant does not belong to Calandrinia, but is Montia minor Gmel., a species native to Central and South Europe, Central Asia and Oregon and California, United States of America. Other specimens examined: wet places, Fairburn, North Auckland, H. Carse 3451; garden weed, Palmerston North, N. R. Foy 45931; garden weed, Upper Hutt 45932; established in arable land, Spring Creek, near Blenheim. In the field, this species has the appearance of stunted Stellaria media Linn., this fact probably explaining why the plant has been collected so infrequently. Polygonaceae. Polygonum convolvulus Linn. var. subalatum Lej. et Court. In the first record of this variety by Healy (1944: 222), a typographical error occurs, and the varietal name is incorrectly published as var. subulatum Lej. et Court. Additional localities for this variety are: Nelson, H. H. Allan 1292; Christchurch 33836. P. lapathifolium Linn. Originally recorded by Smith (1904: 206, 222) from Ashburton County; Cheeseman (1906: 1085) says of the species: “Ashburton, W. W. Smith. I have not seen New Zealand specimens.” Later, Cheeseman (1925: 1061) admits the species in New Zealand, and gives the distribution thus: “North and South Islands: Ditches and roadsides, apparently increasing. In immense abundance in the lower part of the Awanui River Flats, H. Carse! Hauraki Plains, W. Godfrey! New Plymouth, W. W. Smith; vicinity of Ashburton, H. H. Allan.” Allan (1940: 87) also admits the species, stating “Infrequent in damp places in both Islands.” The specimens under this name in the Cheeseman Herbarium have been examined, and are representative of one locality only, namely, Kaiaka, North Auckland; there is an undetermined specimen from Ashburton, collected by W. W. Smith, this being typical P. hydropiper Linn. Carse's specimens from Kaiaka in the Cheeseman Herbarium do not match the descriptions of P. lapathifolium Linn., since instead of the dense, oblong-cylindrical racemes characteristic of this species, these specimens all possess long, lax, distant-flowered racemes, resembling those of P. hydropiper Linn. These specimens appear to be close to P. nodosum Pers., but the treatments of the genus available are not sufficiently full to make certain

identification. I have only seen one New Zealand specimen of P. lapathifolium Linn., from a linen-flax crop near Geraldine, R. Mason, January, 1943, 45933. It appears that L. lapathifolium Linn. has been confused with P. hydropiper Linn., and with the failure to meet the species in the field and the virtual absence of specimens in herbaria, it must be regarded as rare. † Denotes species recorded for the first time for New Zealand.Rumex brownii Campd. This Australian species has not been previously recorded from New Zealand, although it has been present in the country for many years. Examination of specimens of this genus in the Plant Research Bureau and Dominion Museum Herbaria show that this species has been confused with the indigenous R. flexuosus Sol. ex Forst. f. and the naturalized R. pulcher Linn., and has been placed in these species. The following key will serve to distinguish these three species: 1. Tubercle present on one or all fruit valves R. pulcher Tubercle absent from all fruit valves 2 2. Inflorescence flexuose: fruit valves with acuminate tip and marginal bristles hooked R. flexuosus Inflorescence not flexuose: fruit valves produced into a long hooked bristle, marginal bristles hooked R. brownii The following material has been examined: (a) Plant Research Bureau Herbarium—Mount Wellington lavafields, Auckland, D. Petrie, November, 1919 (as R. flexuosus?) 1910; roadside near Westshore, Napier, April, 1945 (as R. pulcher Linn.?) 33928; pasture, Dashwood Pass, Marlborough, May, 1942 (as R. pulcher Linn.?) 45251; Nelson, H. H. Allan, January, 1929 (as R. pulcher Linn.) 1291; Hae Hae te Moana Valley, South Canterbury, R. M. Laing, date not given (as R. brownii?) 1898. (b) Dominion Museum Herbarium—Whangarei, D. Petrie, February, 1895 (as R. flexuosus Sol.); Little Barrier Island, D. Petrie, December, 1896 (as R. flexuosus? Sol.); One Tree Hill, Auckland, D. Petrie, April, 1901 (as R. flexuosus? Sol.); Remuera, Auckland, D. Petrie, March, 1907 (as R. flexuosus? Sol.). I have noted recently that this species is thoroughly established on roadsides and in pasture, Dashwood and Weld Passes, Marlborough, and in coastal areas near Nelson. The species has been recorded by Laing (1915: 25) as introduced at Norfolk Island, and I have recently determined a specimen from the Kermadec Islands collected by Mr. J. H. Sorenson. Rumex sagittatus Thunb. In an earlier paper (Healy, 1944: 222), I expressed doubt as to the correct identity of the plant known in New Zealand as R. scutatus Linn. Mr. J. H. Lousley, who is working on the British Rumices, kindly examined New Zealand material for me, and has advised that the plant is really R. sagittatus Thunb. Mr. Lousley's notes (in litteris) deal with the distribution of the species in the Southern Hemisphere, and several extracts are given below: “It has been taken into cultivation as a yam on account of the tubers, and is also used in gardens as an ornamental plant because of the brightly coloured petals.” He mentions that there are specimens at Kew distributed from the Botanic Gardens, Brisbane, labelled “Subspontaneous, N. Sydney, C. T. White, no. 10333, 19/12/1935” and “…there is also material with a badly scribbled label which appears to read ‘Java—Lawang Botanic Garden’ which was received

in 1913—this was presumably cultivated, but may be the source from which the plants introduced into Australia and New Zealand were derived.” This plant is becoming more widely distributed in New Zealand, due to its use as a garden ornamental and its subsequent escape from cultivation. Phytolaccaceae. Phytolacca octandra Linn. Recorded previously from Marlborough in the South Island by Healy (1944: 222), its range is extended southward to Gough's Bay, Bank's Peninsula, 45940. Chenopodiaceae. Atriplex semibaccata R. Br. var. typica Aellen. This species has been recorded from New Zealand by Aellen in an overseas publication not readily available here, and has not yet been noted in local literature. Aellen (1938: 410–11) gives the record as follows: “Neu-Seeland: North Island: Wellington; alkaline soil; sea shore, sea level, 1909. H. H. Travers (Wien).” Chenopodium auricomiforme Aellen et Thell. This species has also been recorded in an overseas publication, and not cited in local works. Probst (1928: 56–57) gives the record: “…liegt von Neu-Seeland in Herb. Univers. Zurich (Aellen br.) v. kommt seit 1912 fast alljahrlich Derend….” Rosaceae. Cotoneaster microphylla Wall. Allan (1940: 287) records a species C. integerrima Med. as “Noted as escape in Wellington Province.” I have been unable to trace the specimens on which the record was based, but have examined specimens of Cotoneaster collected from the vicinity of Wellington late in 1940 and which, it must be presumed, are typical of the plant on which the record was founded. These specimens were under the name C. integerrima Med. in the Plant Research Bureau Herbarium, but they are actually typical C. microphylla Wall. In C. microphylla Wall, the leaves are small, obovate, to ½ in. long (usually ±⅓ in.), flowers solitary, the sepals tomentose on the outer surface, while in C. integerrima Med., the leaves are ovate, ¾–2 in. long, flowers in 2–4-flowered cymes, and sepals glabrous on outer surface. Specimens examined include: pasture, Karori, R. Mason 28361, 28364; Karori, V. D. Zotov 54287; established on hillsides between Mangaroa and Silverstream, 45925. Crataegus monogyna Jacq. Kirk (1899: 136) recorded C. oxycantha Linn. from the North Island, and subsequent botanists have retained the species in the naturalized flora. Examination of Crataegus in the field has shown that our plant is not true C. oxycantha Linn., but is another distinct species C. monogyna Jacq.; plants have been checked in most districts in both Islands, and there is no evidence that the true C. oxycantha Linn. occurs as a naturalized species in New Zealand. In C. monogyna Jacq. there is a single style—hairy at the base, and a single pyrene to each fruit, whereas in C. oxycantha Linn. there are 2 (–3) styles—glabrous at the base and 2 (–3) pyrenes to the fruit. The New Zealand material differs from the descriptions of the species in respect of the hairy flower pedicels, our plants usually having ± glabrous pedicels.

Mimoseae. † Denotes species recorded for the first time for New Zealand.Acacia verticillata (L'Herit.) Willd. This Australian species has been found as a hedge escape, established on roadsides between Te Puke and Pongakawa Valley, Bay of Plenty, 45923. Our specimens have phyllodes broader than in the typical form and appear to belong to var. latifolia Benth. Papilionaceae. Calycotome spinosa (Linn.) Link. The distribution given by Allan (1940: 120) is for the vicinity of Palmerston North; this plant has now been noted as established on roadsides, Bunnythorpe; Sanson; Stanway, near Feilding; Shannon (now eradicated). Lathyris pratensis Linn. Recorded from Ashburton County by Smith (1904: 218), its occurrence at Gore was noted by Healy (1944: 224); it has been found at Wainui Bay, Bank's Peninsula, Mrs. E. Brooker 51399, and is established along the railway line near Mangaroa, North Island, 33998. Melilotus albus Med. Allan (1940: 123) gives the distribution as Canterbury and Napier; it is established in waste places about Seddon, Marlborough; waste land, Thorndon, Wellington. Trifolium parviflorum Ehrh. Previously recorded from Feilding by Allan (1940: 291) and Marlborough by Healy (1944: 224), the species is now well established on stony land near the mouth of the Hutt River, Wellington, 45934, 45942. Trifolium suffocatum Linn. Previously recorded only from Blenheim [Boulder Bank, Wairau Bar] by Allan (1940: 291), the species has been found in depleted Danthonia pasture, Wither Hills Soil Conservation Reserve, Blenheim, 33639; roadside, Dashwood Pass, Marlborough, 36309; stony river terrace, Culverden, 33598. † Denotes species recorded for the first time for New Zealand.Trifolium tomentosum Linn. The present concept of T. resupinatum Linn. in New Zealand covers two distinct plants; there is the true T. resupinatum Linn, which constitutes the T. resupinatum Linn, var. typicum Fiori et Paol. of Hegi and other European botanists, and T. tomentosum Linn, which constitutes the T. resupinatum Linn. ssp. tomentosum (Linn.) Gibelli et Belli of European authors. There occur in New Zealand, therefore, a group of three species of Trifolium with inflated fruiting calyces which can be distinguished by the following key: 1.Stems creeping, rooting at the nodes; flower head with involucre of lobed bracts ± lengths of calyx tube T. fragiferum Stems ± prostrate, not creeping or rooting at nodes; involucre of lobed bracts absent 2 2.Fruiting calyx with two short recurved teeth ± hidden in woolly tomentum; peduncles shorter than leaves T. tomentosum Fruiting calyx conspicuously reticulate, with two long, divergent teeth; peduncle longer than leaves T. resupinatum The following specimens in the herbarium of the Plant Research Bureau give some concept of the distribution of the two species: T. resupinatum Linn.—Dargaville, H. H. Allan 18450; Mangonui, collector not stated, 3456; Whangarei, E. H. Arnold 33934; North Auckland, P. W. Smallfield 3507; Bucklands, Auckland, M. Hodgkins 8947; Kopu, near Thames, J. H. Hudson 45935; Ellesmere [Canterbury?], collector not stated, 802. T. tomentosum Linn.—

Northern Wairoa, North Auckland, D. H. McKenzie 1240; Mount Victoria, Wellington, E. O. C. Hyde 5466; Seatoun, Wellington, H. H. Allan 19042; Thorndon, Wellington, 35540; Picton, 36283; Boulder Bank, Wairau Bar, Blenheim, D. V. Perano 9061; Mason River, near Waiau, 33595; Stanton River, near Waiau, 33594. Vicia cracca Linn. Previously known from the Wellington Province and North-Eastern South Island, the species occurs in Southland; waste land near wharf, Invercargill, 33920; coastal waste land, Riverton, 33931. Umbelliferae. † Denotes species recorded for the first time in New Zealand.Aegopodium podagraria Linn. This Eurasian species appeared in a garden, Parewanui, Bulls, R. O. Dalrymple 45941. According to Mr. Dalrymple, the plant was apparently introduced amongst horticultural material imported from Japan about 1937. The stout, creeping root system has made this plant an aggressive weed in this instance. † Denotes species recorded for the first time in New Zealand.Coriandrum sativum Linn. This species was cultivated for medicinal purposes during the early part of the recent war period, and has persisted for several years in waste land after the project was abandoned: Wellington, 35131. CompositaeM. Aster subulatus Michx. Two species of Aster has been recorded from New Zealand, namely, A. imbricatus Walp. by Kirk (1896: 503–4) from Wellington, and A. subulatus Michx. by Carse (1916: 242) from Mangonui County. Dealing first with A. imbricatus Walp., the species is cited as A. imbricatus Linn, by Allan (1940: 294); now, this is a valid South African species (according to Index Kewensis), and Kirk gathered his specimens off ballast that had been deposited by the ship Silverstream, which had taken in this material from Buenos Aires. Kirk in the same paper records a South African grass, Bromus vestitus Thunb. from this ballast and notes that this is naturalized in the Argentine, but there is no mention that Aster imbricatus Linn. is naturalized there also, so that it is very unlikely that it could be a South African species. The species A. subulatus Michx. recorded by Carse is North American in origin, and has been treated as such by subsequent workers. Comparison of Kirk's original specimens of A. imbricatus Walp. with available specimens of A. subulatus Michx. showed that the two sets of plants were identical, so that there is only one species involved. I have been unable to match the specimens with descriptions of A. subulatus Michx., but they agree well with A. squamatus (Spreng.) Hieron., a South American species, and pending confirmation of this determination from Dr. A. L. Cabrera La Plata, Argentina, tentatively determine the species under this name. There is evidence that indicates the determination is feasible, for according to Cabrera (1941: 69–71) in his synonymity of A. squamatus (Spreng.) Hieron. there is the citation “? A. imbricatus Walp. Repertorium, 11 (1843), p. 574.” Therefore, the original record is that by Kirk (1896: 503–4), and Carse's record of A. subulatus Michx. constitutes a new distributional

record of A. squamatus (Spreng.) Hier., while the name A. subulatus Michx. must be removed from the naturalized flora. It is worth mention that the species has recently appeared near Kaiwarra, Wellington, 45937, having not been found in this original locality since Kirk's original gathering late last century. Bidens pilosa Linn. Recorded originally by Richard (1832: 235) without exact locality, the distribution has been given by all later botanists as the northern portion of the North Island. This would appear correct were it not that J. F. Armstrong (1872: 288) recorded “B. pilosus Linn.?” from Canterbury, a record either missed or disregarded by all later workers. Reference to Sherff's monograph of the genus Bidens (1937: 419–24) indicated that Armstrong may have been justified in reporting the species from Canterbury, since Sherff states in his list of material examined “…idem [Le Guillon], Akaroa, Bank's Peninsula, Middle Ist., New Zealand, 1841,” the specimen being deposited in the Paris Museum. He mentions further specimens of the same species from the south of the North Island—“H. H. Travers, introduced, swampy coast places, Wellington, North Isl., New Zealand, October, 1908,” these specimens being in the herbarium of the Bureau of Science of the Philippine Islands, Manila. In this monograph, Sherff gives B. aurantiacus Colenso (1895: 388), based on specimens from Te Kawakawa, East Cape, as a synonym of B. pilosa Linn. var. minor (Blume) Sherff. Carduus pycnocephalus Jacq. This species has been found in pasture, Te Puke, Bay of Plenty, 33818; waste land, Upper Hutt, 35585: it has been noted as established about Nelson and suburbs, and is increasing about Blenheim and on roadsides between Blenheim and Seddon. Senecio elegans Linn. Previously recorded only from the North Island by Allan (1940: 159), this species has appeared in coastal areas, Timaru, R. Mason 45914; Invercargill, 33919. Hieraceum pilosella Linn. Recorded previously from Canterbury and the Mackenzie Country by Allan (1940: 180) and Hawke's Bay by Healy (1944: 227), it has been found established in modified tussock grassland, “Upton Fells” Station, Medway River, Awatere County. Soliva anthemidifolia (Juss.) R. Br. This species was recorded by Cheeseman (1883: 285) from “Alluvial flats of the Northern Wairoa River, near Dargaville and Mangawhare,” and its most recent distribution is that given by Allan (1940: 152) as “Occasional in waste places in North Auckland and near Wellington.” While checking the material placed under this species in the Plant Research Bureau Herbarium, it was found that with the exception of one specimen, none would match descriptions of the species; the specimen that actually was S. anthemidifolia (Juss.) R. Br. was one of Cheeseman's original gatherings, and examination of specimens in the main herbaria show that the species has not been gathered since. The plant that has been placed under that name in New Zealand is a very distinct species, which I have tentatively determined with the Chilean † Denotes species recorded for the first time in New Zealand. S. valdiviana Phil.

For identification purposes, a key to the three species naturalized is given below. 1. Achene broadly winged, with two lateral lobes S. sessilis Achene with small wings or wingless, lateral lobes absent 2 2. Flower heads large, only near base of plant; achene woolly on each side of beak, surface transversely rugose S. anthemidifolia Flower heads small, distributed along branches; achene glabrous, not transversely rugose S. valdiviana ? Tolpis umbellata Bertol. This species is established along a railway line, Helvetia, near Pukekohe, 51041; roadsides and waste land near Rotoehu State Forest, Pongakawa Valley, K. W. Allison 36693; waste land, Te Puke, 33969; Kelburn, Wellington, E. H. Atkinson, 5705. Polemoniaceae. Collomia grandiflora, Dougl. This species was recorded by Healy (1944:227) from Cromwell, Otago, and indicated as a new record for New Zealand; this latter was incorrect, in that an earlier record by Kirk (1878a:415) had been overlooked. Scrophulariaceae. Bartsia latifolia (Linn.) Sibth. et Smith. Previously noted from Hawke's Bay by Allan (1940:300), this species has been found at Otorohanga, Miss R. L. Johnson 48463. Verbenaceae. Verbena bonariensis Linn. This species has now been found in the South Island: near the Delta Military Camp, Omaka, Blenheim, E. Reid 45915. V. venosa Gill. et Hook. Hilgendorf (1926:237) recorded this species as “Prominent in North Auckland,” but I have been unable to trace the specimens on which the record was based; from his statement, it is probable that he was actually referring to V. bonariensis Linn., which occurs there. Recently, a plant has been found in a number of North Island localities agreeing in all respects with descriptions of this species, and which is increasing in quantity; roadside near Upper Hutt, 45917; pasture, Havelock North, V. D. Zotov 33742; North Taranaki, A. T. Guddopp 45920; roadside, Haywards, Hutt Valley, J. H. Warcup 45921; along railway line, Kaitoke, 45916. Amaryllidaceae. Allium triquetrum Linn. Recorded previously from the North Island, this species has been found as a garden weed at Waitata Bay, Pelorus Sound, and about Blenheim and Nelson. Iridaceae. Crocosmia aurea Planch. Specimens of a persistent garden escape were determined at Kew in 1930 at Crocosmia aurea Planch., and recorded by Allan (1935: 3), and all subsequent gatherings of this plant have been placed under this name. I have been unable to match our specimens with descriptions of this species, and, in fact, have been, unable to work the plant down satisfactorily to a genus. The style branches in our material are distinctly bifid, the ovules many

(usually more than 15), the outer spathe valve usually emarginate, and the scape lacks wings, characters not typical of this species. It may be that New Zealand material belongs to the inter-generic hybrid Tritonia crocosmiaefolia (Leh.) Nichols, a hybrid between Tritonia pottsii Baker and Crocosmia aurea Planch., but I have been unable to obtain a good description of this plant. This garden hybrid has been recorded as a garden escape at Fiji by Smith (1943: 535) and at Hawaii by Fagerland and Mitchell (1944: 37). Apart from the question of identity, this species is a persistent garden escape in many localities, and has very definite claims to naturalization. Allan (1940: 305) gives the distribution only for the North Island, but it must now be extended to the South Island: Picton, 44724; Motueka, 44741; Takamatua Bay, Akaroa Harbour, 48796; Timaru, R. Mason 51375; Redan Valley, near Wyndham, Southland, 45926. Juncaceae. Juncus supinus Moench. Recorded by Allan (1940: 308) from pakihi land near Westport, the species has been recently found in the North Island; swampy land, near Kaitoke, 45968; near the Summit Railway Station, Wairarapa Railway Line, 45969. Cyperaceae. Carex divisa Huds. Recorded by Kirk (1875: 508), without locality, under the name C. chlorantha Brown and by the same botanist (1878b: xlii) with the locality as Waitemata, Auckland; Allan (1940: 221) gives the distribution as “Frequent on margins of brackish pools and estuaries about Auckland”; the species has been established on margin of the estuary at Havelock, Marlborough, for some years, but flowering material has not been obtainable. C. vulpinoidea Michx. Recorded by Allan (1940: 221) from Manakau Harbour, Auckland, the species has been found established in swampy land east of Te Puke, Bay of Plenty, 33807; it has also been noted near Tauranga. Gramineae. Phalaris paradoxa Linn. Recorded from Lake Ellesmere, Canterbury, by Healy (1946: 403), this has now been collected from waste land, Upper Hutt, 33790. Stipa neesiana Trin. et Rupr. Recorded by Allan (1940: 313) from near Auckland, it is now established in pasture and on a roadside near Blind River, Seddon, Marlborough. Bibliography. Aellen, P., 1938. Revision der Australischen und Neuseelandischen Chenopodiaceen. 1. Theleophyton, Atriplex, Morrisiella, Blackiella, Seniella, Pachypharynx. Botanische Jahrbucher, Band Ixviii, Heft ⅘, pp. 410–11. Allan, H. H., 1933. Notes on Recently Observed Exotic Weeds. (3) The Genus Sisymbrium. N.Z. Journ. Agric., vol. xlvii, pp. 38–40. —— 1935. Additions to the Alien Flora of New Zealand. Trans. Roy. Soc. N.Z., vol. lxv, pp. 3–4. —— 1940. A Handbook of the Naturalized Flora of New Zealand. N.Z. Dept. S. and I. R., Bull. No. 83. Armstrong, J. F., 1872. On the Naturalized Plants of the Province of Canterbury. Trans. N.Z. Inst., vol. 4, pp. 287–88. Bentham, G., 1864. Flora Australiensis, vol. 2, p. 461.

Berger, A., 1930. Crassulaceae. Die Natüralischen Pflanzenfamilien, Band 18a, p. 388. Cabrera, A. L., 1941. Compuestas Bonaerenses. Revista del Museo de la Plata (Nueva serie), Tomo IV, Sección Botánica, pp. 69–71. Carse, H., 1916. Some Further Additions to the Flora of Mangonui County. Trans. N.Z. Inst., vol. xvliii, p. 242. Cheeseman, T. F., 1883. The Naturalized Plants of the Auckland Provincial District. Trans. N.Z. Inst., vol. 15, pp. 275–78. —— 1906. Manual of the New Zealand Flora. Edition 1. —— 1925. Manual of the New Zealand Flora. Edition 2. Colenso, W., 1895. Phanerograms: A Description of a few more Newly-discovered, Indigenous Plants: being a Further Contribution towards making known the Botany of New Zealand. Trans N.Z. Inst., vol. 27, p. 388. Fagerland, G. O., and Mitchell, A. L., 1944. A Check-list of the Plants, Hawaii National Park. Nat. Hist. Bull., No. 9, p. 37. Hawaii Nat. Park., U.S. Dept. of the Interior, National Park Service. Harvey, W. H., in Harvey, W. H., and Sonder, O. W., 1861–62. Flora Capensis vol. 2, p. 330. Healy, A. J., 1944. Some Additions to the Naturalized Flora of New Zealand. Trans. Roy. Soc. N.Z., Vol. 74, pp. 220–30. —— 1946. Contributions to a Knowledge of the Naturalized Flora of New Zealand. Trans. Roy. Soc. N.Z., vol. 75, pp. 399–403. Hilgendorf, F. W., 1926. Weeds of New Zealand. Hooker, J. D., 1864–67. Handbook of the New Zealand Flora. Kirk, T., in Hooker, J. D., 1867. Handbook of the New Zealand Flora, Part II, p. 758. —— 1875. On the Occurrence of Lepilaena preissii Muell. and Carex chlorantha Br. in New Zealand. Trans. N.Z. Inst., vol. 7, p. 508. —— 1878a. Contributions to the Botany of Otago. Trans. N.Z. Inst., vol. 10, pp. 414–15. —— 1878b. An Enumeration of Recent Additions to the N.Z. Flora, with Critical and Geographical Notes. Trans. N.Z. Inst., Appendix, vol. 10, p. xlii. —— 1896. On the Products of a Ballast-heap. Trans. N.Z. Inst., vol. 28, pp. 503–4. —— 1899. The Students' Flora of New Zealand. Laing, R. M., 1915. A Revised List of the Norfolk Island Flora, with some Notes on the Species. Trans. N.Z. Inst., vol. 47, p. 25. Ostenfeld, C. H., 1918. Contributions to the West Australian Botany. Part II. Dansk. Botanisk Arkiv., Bind 2, nu. 8, pp. 40–45. Probst, R., 1928. Dritter Beitrag zur Adventiv-flora von Solothurn und Umgebung. Mitteilungen der Naturforschenden Gesellschaft Solothurn Achtes Heft (XX Bericht), pp. 56–57. Richard, A., 1832. Essai d'une Flore de la Nouvelle Zélande. Schonland, S., 1929. Materials for a Critical Revision of Crassulaceae. Trans. Roy. Soc. S. Afr., vol. 17, p. 161. Sherff, E. E., 1937. The Genus Bidens. Part II. Field Museum of Natural History. Bot. Series, vol. 16, publication 389, pp. 419–24. Smith, A. C., 1943. Studies in Pacific Island Plants—III. New and Noteworthy Flowering Plants from Fiji. Bull. Torrey Bot. Club, vol. 70, p. 535. Smith, W. W., 1904. Plants Naturalized in the County of Ashburton. Trans. N.Z. Inst., vol. 36, pp. 206–25. Thellung, A., 1906. Denks. Schweiz. Gesell. Naturwiss., 41, abh. 1, 135–39. Thomson, G. M., 1875. On Some Naturalized Plants of Otago. Trans. N.Z. Inst. vol. 7, p. 377. Williams, F. N., 1896. A Revision of the Genus Silene Linn. Journ. Linn. Soc. Bot. (London), vol. 32, pp. 1–196.