A revision of Geostachys (Zingiberaceae: Alpinioideae) in Cambodia, Laos and Vietnam, including a new species, G. aristata Author Leong-Škorničková, Jana 0000-0003-4490-3490 The Herbarium, Singapore Botanic Gardens, 1, Cluny Road, 259569 Singapore & jana _ skornickova @ nparks. gov. sg; https: // orcid. org / 0000 - 0003 - 4490 - 3490 jana_skornickova@nparks.gov.sg Author Bình, Nguyén Qu ốc 0000-0002-6267-839X Vietnam National Museum of Nature, Vietnam Academy of Science and Technology, 18, Hoàng Qu ṍc Vi ệt Street, Cǻu Giǻy, 122300 Hà N ội, Vietnam & binhzing @ gmail. com; https: // orcid. org / 0000 - 0002 - 6267 - 839 X binhzing@gmail.com Author Đăng, Tr ần H ữu Becamex Institute of Research and Development, Becamex IDC Corp., 08, Hùng Vuõng Street, Hòa Phú Ward, Th ủ Dǻu M ột City, Bình Duõng Province, Vietnam & Southern Institute of Ecology, Institute of Applied Materials Science, Vietnam Academy of Science and Technology, 1 D, TL 29 Street, District 12, H ồ Chí Minh City, Vietnam Author Trư ờng, Lưu H ồng 0000-0002-7036-7081 Southern Institute of Ecology, Institute of Applied Materials Science, Vietnam Academy of Science and Technology, 1 D, TL 29 Street, District 12, H ồ Chí Minh City, Vietnam & hongtruongluu @ gmail. com; https: // orcid. org / 0000 - 0002 - 7036 - 7081 hongtruongluu@gmail.com Author Nuraliev, Maxim S. Joint Russian-Vietnamese Tropical Scientific and Technological Center, Nguy ễn Vãn Huyên street, Cǻu Giǻy, Hà N ội, Vietnam Department of Higher Plants, Biological Faculty, M. V. Lomonosov Moscow State University, 1, 12, Leninskie Gory, 119234 Moscow, Russia text Phytotaxa 2023 2023-03-07 585 4 245 260 http://dx.doi.org/10.11646/phytotaxa.585.4.1 journal article 57380 10.11646/phytotaxa.585.4.1 aa4b6a5e-18fa-4915-a746-95d1cb74d648 1179-3163 7703837 Geostachys pierreana Gagnepain (1906a: 147) ( Fig. 1 , 5 , 6 ) Type: — Country unknown [S. Vietnam, Cambodia or Thailand], 1865–1877, J.B.L. Pierre 5851 ( lectotype , designated here, P : P00686635 !; isolectotypes P : P00686634 ! and P032698 !; flowering) . Clump-forming herb up to 1 m tall. Rhizome supported by stilt roots, sheathed by dry brown scales. Leafy shoots with 6–13 leaves per pseudostem, bladeless at basal 1/3; bladeless sheaths 2–6; most basal (outermost) sheaths very short, soon becoming dry and brown, inner sheaths distally gradually longer, green with reddish tinge, abaxially glabrous, striate (more prominent in dry material); leaf sheaths similar to bladeless sheaths; ligule ca. 3 mm long, triangular with obtuse apex, green with reddish to brown tinge (darker at base), glabrous; petiole 5–10 mm long, shallowly canaliculate, mostly green, adaxially puberulent and abaxially glabrous to sparsely puberulent; leaf blade narrowly elliptic, 25–35 cm long, 4–6 cm wide, weakly plicate, adaxially green to dark green, somewhat shiny, glabrous, abaxially paler green, glabrous, base narrowly obtuse, apex acuminate, midrib adaxially sunken, pale green to ochraceous, glabrous, abaxially protruding, pale green, glabrous. Inflorescence radical; peduncle erect, covered by 5–7 sheathing bracts; sheathing bracts distichous, ovate to narrowly ovate, overlapping, basal ones shorter, distal ones significantly longer (sometimes developing blade vestiges at the apex, see Fig. 5D on the right), soon becoming brown and papery, striate, glabrous on both sides; thyrse lax, many-flowered; rachis 4–8 cm long, decurved, dull green with dark red tinge (turning dark red to maroon in fruiting stage), hirsute, shortly branched; bracts ovate, broadly ovate to semi-circular, 1–3 mm long, 2–3 mm wide, brownish-maroon at base, greenish to whitish distally, glabrous, caducous, each subtending cincinnus of 1–5 flowers on 4–15 mm long branch; branches secund, of the same colour as rachis, hirsute; bracteoles broadly ovate to broadly elliptic, spathaceous, enveloping the flower with the rest of the cincinnus, 13–30 mm long, 4–17 mm wide, pale pink-orange with greenish apex when young, soon scarious and pale brown to ochraceous, somewhat striate (visible in dry material), mostly glabrous, apex mucronate and puberulent. Flower gullet-type; pedicel negligible; calyx tubular, 12–15 mm long, apex with single lobe ending with prominent ciliate mucro, semi-translucent pale yellow with red tinge, mostly glabrous, apex puberulent; floral tube yellow; dorsal corolla lobe elliptic-oblong with cucullate apex, yellow with white semi-translucent veins, glabrous; lateral corolla lobes elliptic-oblong with obtuse apex, yellow with white semi-translucent veins, glabrous; labellum obovate, apex slightly bilobed, yellow with semitranslucent veins, adaxially with short glandular hair, abaxially glabrous; lateral staminodes irregularly obovate with round apex, nearly entirely connate to labellum, yellow with semi-translucent veins; stamen yellow; filament dorso-ventrally compressed, pale yellow with glandular hair; anther connective yellow, with glandular hair; crest 2–3 mm long, bluntly obscurely three-lobed, yellow; anther thecae dehiscing through the entire length; ovary trilocular with central placentation, cylindrical, greenish or with reddish tinge, glabrous; epigynous glands two; style filiform, cream white; stigma capitulate, cream white; ostiole transverse, forward-facing, ciliate. Fruit (immature) ellipsoid hesperidium, ca. 10 mm long, ca. 6 mm in diam., red, glabrous, with persistent calyx; seeds unknown. FIGURE 5. Geostachys pierreana . A. Leafy shoot (from above). B. Basal part of plant with inflorescences. C. Detail of ligule. D. Inflorescences. E. Flower with labellum removed (staminodes remain attached to floral tube). F. Cincinnus with one open flower (side view). G. Flower (front view). All from Netopilová & Faltová s.n . Photo by M. Netopilová. Eponymy: —This species is named after a French botanist Jean Baptiste Louis Pierre ( 23 October 1833 – 30 October 1905 ) who made the type collection. Distribution & IUCN preliminary assessment: —Currently known only from Bokor National Park, Cambodia . Geostachys pierreana is currently assessed as Data Deficient on IUCN Red List ( Olander 2019b ). Although the only known location in Bokor National Park is under legal protection, there is no information about the number or size of populations. The Popokvil Waterfall is accessible to public, and the effect of tourism on the vegetation surrounding this area needs to be assessed first for an evidence-based assessment. We therefore suggest continuing to treat this species as Data Deficient (DD). FIGURE 6. Geostachys pierreana . A. Leafy shoot (view from below). B. Pseudostem with ligules and leaf bases. C. Basal part of plant with inflorescences. D. Cross section of cincinnus showing total of five flowers/fruits. E. Stamen (adaxial and abaxial view). F. Inflorescence, top view. All from Luu s.n . Photo by H.T. Luu . Ecology & phenology: —Growing in broadleaved evergreen forest, including its edges and slightly disturbed areas, at elevations of ca. 1000 m a.s.l. Flowering has been observed in April and June. Additional specimens examined:— CAMBODIA . Kampot province : [ Bokor National Park ], Popokvil , April [year not given, inferred as 1967?], P. Dy Phon 1157 ( P : P02203163 , flowering); Bokor National Park , ca. 1.5 km east of Popokvil waterfall , ca. 970 m a.s.l. , 1 June 2017 , M. Netopilová & I. Faltová s.n. (mounted over 2 sheets, Accession Nos. 02602KBFR1 and 02526KBFR6 , herbarium of the Department of Botany and Plant Physiology of the Faculty of Agrobiology, Food and Natural Resources of the Czech University of Life Sciences Prague, Czech Republic, flowering) . Other field records:— CAMBODIA . Kampot province : Bokor National Park , 5 June 2014 , H.T. Luu s.n. (flowering, photo uploaded to LE : LE01093259 ; see also Fig. 6 ) . Nomenclatural notes: — Gagnepain (1906a) included a single gathering, Pierre 5851 , in the protologue of G. pierreana . From the title of Gagnepain’s paper it is evident that the specimens examined by him were located at P. Although the three available sheets at P are currently labelled as type (sensu holotype ) and two isotypes, with this perception repeated by Saensouk & Saensouk (2021) , they have to be considered syntypes . All three syntypes are of good quality, each including leafy shoot and one or two inflorescences. One of the sheets (P00686635) also has several flower buds and one neatly pressed flower in an enclosed envelope, which was most likely used in the preparation of pencil drawing attached to the sheet P032698, and subsequently reproduced in Fig. 10 in Gagnepain (1908) . We designate this particular specimen (barcode P00686635) as the lectotype ; the remaining two sheets (barcodes P00686634 and P032698) are isolectotypes. Taxonomic notes: — Larsen (1986) stated that G. pierreana is described from Vietnam , whereas Newman et al. (2007) assumed it to be collected in Laos . Saensouk & Saensouk (2021) produced three conflicting statements within a single paper that the type locality is in Laos (as stated in the figure legend on p. 3059), in Thailand , Chantaburi (as stated in species citation on p. 3064) or in Cambodia (as implied in the notes p. 3065). In spite of the previous guesses, the original collection locality of G. pierreana remains unknown as already clearly noted by Gagnepain (1906a) in the protologue: ‘INDO-CHINE.— No 5851 [ Pierre ] sans désignation de localité, ni date’. In Flore Generale de L’IndoChine , the distribution is given, with no further explanation, as “Cochinchine” ( Gagnepain 1908: 102 ). There are three sheets of Pierre 5851 in P. They all have standard label common to herbarium collections of L. Pierre, with pre-printed locality given as Indochina; however, there is remark on all of them ‘etiquette perdue’ indicating that the original label was lost. The date of the collection is accordingly given as 1865–1877, i.e., the entire 12-year long period when Pierre was based in Saigon (now Ho Chi Minh City ) as the Director of the Jardin Botanique et Zoologique. It is known that during this period Pierre made many expeditions not only in southern Vietnam , but also in neighbouring Cambodia and even Thailand (former Siam ) ( Gagnepain 1906b ). The numbering of the specimens is of no help to deduce the location or date as it appears that final numbers on Pierre’s standard labels were applied as the collections were sorted/worked through, and these differ from his provisional field numbers. We assume this based on examination of several sheets of collection numbers close to 5851 (i.e. 5848, 5852, 5854), which all belong to Zingiberaceae , but all were collected at different locations and in different years. There is no sheet of Geostachys from Vietnam present in any herbaria we have examined, which would be in good match with the type of G. pierreana . The only specimens which we consider to be in good match with the type is collection by Dy Phon 1157 and collection and photographs made by Netopilová & Faltová s.n ., both from Bokor National Park, Cambodia . These two collections match the type and the protologue in all major aspects, such as the shape and indumentum of leaf blades and ligules, short petioles, length and decurvation of the rachis and secund arrangement of the flower-bearing branches, number of flowers in cincinnus, as well as more specific details such as hirsute rachis and flower-bearing branches, shape and size of bracts, spathaceous glabrous bracteoles with mucronate puberulent apices and calyx with a single lobe ending in a sharply mucronate apex. From other labelled collections, it is known that Pierre visited and made collections in the same area ( Kampot city and surrounding hills). Thus with all the evidence listed above, the most likely origin of the type of G. pierreana is Cambodia . The above description, although based on the protologue and the three original sheets, is therefore also supplemented with observations from Dy Phon 1157 and Netopilová & Faltová s.n. The indication of G. pierreana by Phạm (2003) as present in Vietnam originated from the uncertainty regarding the type collection, as follows from his reference to Gagnepain. The same is probably true for Nguyễn (2005). Nguyễn (2017) listed several additional Vietnamese collections under G. pierreana whose identifications remained preliminary as they were based solely on the herbarium material that does not show important features of floral structure. For two collections from Ninh BÌnh province , Averyanov et al. HAL 1638 & HAL 1639 , photographic material appeared to be available, which allowed to identify them as Wurfbainia sp. (Zingiberaceae) and Tropidia angulosa (Lindl.) Blume (Orchidaceae) , respectively. The other collections listed by Nguyễn (2017), all from Lâm Đồng province ( Hà Tuḗ 458 , Nhan 785 , Tṵ 82 , Biên 1228 ), are re-identified here as G. annamensis (see above). Saensouk & Saensouk (2021) have reported G. pierreana as a new record for Thailand , based on a single collection made by Nai Noe 77 (BK, K, P: P00599363) in Chanthaburi province . Although the specimen has been previously annotated as Geostachys cf. pierreana on the sheet deposited in P, it is clearly visible that the inflorescence is erect, with cincinni/flower-bearing branches evenly distributed around the rachis and thus fails to fit G. pierreana , which has decurved inflorescences with secund arrangement of flower-bearing branches. Although determination to the species level with absolute certainty is challenging on the herbarium material, the collection Nai Noe 77 matches well G. tratensis , a species described from a location distanced only about 80 km (air distance) from it. We therefore conclude that there is currently no evidence, that G. pierreana occurs in Thailand .