Thea megacarpa Elmer, Leafl. Philipp. Bot. 5: 1842. (1913)

Zhao, Dongwei, 2022, Typification of six names in Camellia (Theaceae), PhytoKeys 201, pp. 15-22 : 15

5. Thea megacarpa Elmer, Leafl. Philipp. Bot. 5: 1842. (1913)

Lectotype.

(designated here): Philippines. Palawan: Puerto Princesa (Mt. Pulgar), March 1911, Elmer 12822 (E00504323!; its image is available at http://data.rbge.org.uk/herb/E00504323).

Notes.

A single collection, Elmer 12822, was cited in the protologue ( Elmer 1913: 1843) without indicating where the specimens were conserved. Ten duplicates of Elmer 12822 housed at various Herbaria A (00025101), BM, E (E00504323), G (G00354856), K, MO (705490), NY (00385756), P (P04511437), U (U 0226169) and US (00113904) were found, so they are the syntypes of T. megacarpa (Art. 40 Note 1 of the ICN). The citation of Ming (2000: 228), "Type:... A.D.E. Elmer 12822 (K, E, BM, P)", did not validate the lectotypification because the single herbarium in which the type was deposited was not specified (Art. 9.22 of the ICN). The specimen at E (E00504323), which bears immature fruit and seeds, is selected as lectotype.

Cohen-Stuart (1916: 68) transferred Thea megacarpa into Camellia . Sealy (1958: 142) treated it as a heterotypic synonym of C. lanceolata . Chang and Ren (1991: 68) thought that Elmer 12822 "much differed from" C. lanceolata because the former bore "free filaments and [a] thicker pericarp", whereas the latter bore "totally united filaments" and a "thinner pericarp". However, flowers, including filaments, were absent in all the specimens of Elmer 12822 examined above and there was no description of flower parts in the protologue ( Elmer 1913: 1842-1843). Later, Ming (2000: 228) recognised the plant as a subspecies of C. furfuracea (Merr.) Cohen-Stuart.

However, as a native and the single representative of Camellia in the Malay Archipelago, C. lanceolata holds a specific phylogenetic position ( Zhao et al. 2022). The plants under the broad circumscription of this taxon show a continuous variation in the size and shape of the morphological characters. For instance, the length of the leaf blade can vary from 2 cm (e.g. Beaman 8977 at K) to 13 cm (e.g. the lectotype of T. megacarpa , Elmer 12822 at E), but the elements of flower and fruit, such as the filament tube, the hairy ovary and the furfuraceous surface of the pericarp, are generally similar amongst them. Since there is no clear correlation between morphological variation and geographic distribution and molecular phylogenetic analysis of the plants is absent, I provisionally agree with the broad definition of C. lanceolata and place T. lanceolata var. stenophylla and T. megacarpa in its synonymy.