Portulaca psammotropha Hance (1851: 660)

Kokubugata, Goro, Kato, Hidetoshi, Iamonico, Duilio, Umemoto, Hana, Ito, Takuro, Nakamura, Koh, Murakami, Noriaki & Yokota, Masatsugu, 2015, Taxonomic reexamination of Portulaca boninensis (Portulacaceae) in the Bonin (Ogasawara) Islands of Japan using molecular and morphological data, Phytotaxa 217 (3), pp. 279-287 : 285-286

Portulaca psammotropha Hance (1851: 660) .

Lectotype (designated here):— TAIWAN. Pratas Island , April 1831, Wilford s.n. ( K!, image of the lectotype available at http://apps.kew. org/herbcat/getImage.do?imageBarcode= K000340154 ).

= Portulaca quadrifida L. var. formosana Hayata (1911: 37) ≡ Portulaca formosana (Hayata) Hayata (1917: 7) .

Type:— TAIWAN. Lanyu Island, 26 November 1899, Miyake s.n. (holotype TAI!, image available at http://tai2.ntu.edu.tw/specimen/ specimen.php?taiid=T00380).

= Portulaca insularis Hosokawa (1932: 229) .

Type:— TAIWAN. Liuchiuyu Island, 01 August 1930, Hosokawa 1628. (holotype TAI!, image available at http://tai2.ntu.edu.tw/specimen/ specimen.php?taiid=118821).

= Portulaca boninensis Tuyama (1939: 6) , syn. nov.

Type:— JAPAN. Bonin, Chichijima Island, 24 July 1905, Hattori s.n. (holotype TI!)

= Portulaca hainanensis Chun & How (1958: 8 − 9) .

Type:— CHINA. Hainan, June 1932, How 70899 (holotype SYS!).

Description:—Perennial herbs, 3 − 10 cm tall. Stems not articulated, diffuse, branched basally, ca. 1 mm thick, basal stems woody, prostrate, upper stems herbaceous, upright, green. Root fleshy, much branched. Leaves spirally arranged or alternate, subsessile, with axillary hairs; leaf blade 2–3 mm thick, oblong to obovate, 5–8 mm long, base obtuse, apex obtuse or rounded. Flowers solitary, about 10–15 mm in diameter. Sepals 2, ovate-deltate, about 2 mm long. Petals 5, obovate to narrowly obovate, lemon-colored, mostly without margins overlapping. Stamens 8–30. Ovary ovoid. Stigma usually trilobed or tetralobed rarely pentalobed; capsule glossy, 2–4 mm long, 2–3 mm wide.

Distribution:—North Philippine (Batan Islands), South China Sea (Hainan, Pratas Island), South Taiwan (Kenting, Liuciou, Lutao, Lanyu and the Penghu Islets), and Bonin ( Ogasawara ) Islands.

Ecology:—On coastal rocky slopes and sandy beaches, xeric, saline, and exposed to direct sunlight.

Typification:— Hance (1851: 660) provided a detailed diagnosis, the provenance (“ Habitat in insula corallina ˵ Prata Island˝… ”), and habitat (“ ad littora maris arenosas ”). There are two specimen at K (codes 000340153, and 000340154) collected by C. Wilford in “ Prata Island ” (= Pratas Island) as reported in the label. However, the specimen code 000340153 (image available at http://apps.kew.org/herbcat/getImage.do?imageBarcode=K000340153) bears a plant collected in April 1858. As a consequence, it can be considered a post-1851 addition to the collection, it is not part of the original material for the name Portulaca psammotropha , and cannot be elegible as lectotype (arts. 9.2, and 9.3 of ICN, McNeill et al. 2012). On the contrary, the other specimen bears a plant collected in 1831 so being useful for the lectotypification purposes. The morphological characteristics of the exsiccatum matches the Hance’s diagnosis, and it is here designated as the lectotype of the name Portulaca psammotropha .

Morphological notes:—A diagnostic key to Portulaca psammotropha , P.tuberosa , and P. okinawensis is presented below:

........................................................................................................................................................... P. okinawensis var. okinawensis

- Number of stamens less than 15; petals lemon-coloured, obovate to oblanceolate; stems bright green ............................................. .......................................................................................................................................................... P. okinawensis var. amamiensis

Biogeography notes:—Although the different climates, the floras of Bonin Islands (subtropics climate), and Ryukyus (subtropical climate) are considered similar from the phylogenetic point of view ( Hara 1959,Yamazaki 1970).Therefore, it was unexpected that the species identified was not the Ryukyu endemic P. okinawensis , but P. psammotropha , that is commonly found in the Bonin Islands. The Bonin Islands represent the northwestern limit of the distribution area of P. psammotropha which was geographcally isolated from Hainan Island of China, Taiwan, and the Philippines ( Chung et al. 2008). Since Portulaca is a tropical genus and P. tuberosa (the sister species of P. psammotropha ) is distributed in the Malesian and Pacific regions ( Puy et al. 1993), we can hypothesized that the Bonin populations of P. psammotropha were derived from ancestral southern populations from Taiwan or the Philippines.

The Bonin Islands were formed during the Paleocene (66 Ma to 56 Ma) and rose above sea level before the middle Pleistocene, about 0.8 Ma to 0.1 Ma ( Kaizuka 1977, Imaizumi & Tamura 1984). They are oceanic islands that have never been land-connected with other islands/landmasses. So, P. psammotropha must have migrated to the Bonin Islands oversea. The species has dry dehiscent fruits and its seeds are 0.6–0.7 mm in diameter ( Tuyama 1939, Chung et al. 2008). Given that seeds in Portulaca do not float, Geesink (1969) suggested three possible mechanisms for oversea dispersal: 1) attachment to drifting wood, 2) exo- and/or endozoochory, or 3) artificial dispersal. Artificial dispersal is unlikely because Minami-io-jima has not been inhabited and few naturalized plants have been found in the island ( Kato et al. 2008). On the other hand, seeds of some Portulaca species remained viable after floating in seawater for a few weeks ( Ridley 1930). P. psammotropha occurs in coastal rocky slopes in Taiwan ( Chung et al. 2008), and the Bonin Islands ( Fujita et al. 2008, G. Kokubugata, pers. observ.). Therefore, we consider sea-current dispersal a likely scenario to explain the occurrence of the species in the Bonin Islands. The Kuroshio Current washes Taiwan and the northern Philippines, flowing northward along the Ryukyu and Japanese archipelagos and periodically meanders from Japan proper toward the Bonin Islands. The Kuroshio Current is a likely medium, although the absence of the species in the Ryukyus needs to be explained by assuming extinction or haphazard results. Long dispersal with exo- and/or endozoochory by migratory birds is also possible. Further studies, especially phylogeographic/population genetic studies, on plants showing similar distribution patterns could be used to explore the migratory origin and route of Bonin populations.