taxonID	type	description	language	source
03EB8743FFFABC23FFEB620AFEEEFB85.taxon	description	Notes — 1. Hooker & Thomson (1872) merely cited ‘ Malacca, Maingay ’ in the protologue of A. maingayi. Sinclair (1955) cited ‘ Maingay 34 (C., Kew) ’ as type material, but the number ‘ 34 ’ actually refers to the Kew distribution number rather than Maingay’s collection number. Regardless, Sinclair’s type statement constitutes a first-step lectotypification of the name. Turner (2009) cited ‘ Maingay 2617 [Kew distrib. no. 34] ’ as the holotype of A. maingayi, but it is impossible to ascertain that this specimen is the only specimen used by the authors. Turner (2011, 2012) likewise regarded ‘ Maingay 2617 ’ as the holotype of A. maingayi. Eventually, Turner (2018) designated Maingay 2617 (Kew distribution no. 34) as the second-step lectotype (Fig. 2 a, b), explicitly stating that it is a single specimen over two sheets (K 000381024, K 000381029). Both sheets bear the same collection label and one of the sheets (K 000381029) bears the annotation ‘ 2617 continued’ (Fig. 2 b). This is thus consistent with Art. 8.3 of the ICN (Turland et al. 2018), which states that “ a specimen may be mounted as more than one preparation, as long as the parts are clearly labelled as being part of that same specimen, or bear a single, original label in common ”. 2. There is some confusion between the type specimens of A. maingayi and A. pleurocarpus Maingay ex Hook. f. & Thomson, the latter differing from A. maingayi by its cuneate (vs decurrent) leaf base, larger number of flowers per inflores- cence, sparsely (vs densely) hairy petals and long-stipitate (vs short-stipitate) monocarps. Artabotrys pleurocarpus occurs in Peninsular Malaysia and Peninsular Thailand but not in Singapore (Chen & Eiadthong 2020). Both sheets of the lectotype of A. maingayi were originally identified as ‘ Artabotrys pleurocarpus ’ but subsequently re-labelled as ‘ A. maingayi Hf & T’. Turner (2018) cited BM 001014846 as the isolectotype of A. maingayi. This is erroneous, because BM 001014846 represents a sepa- rate gathering (Maingay 3261) and is actually the isolectotype of A. pleurocarpus. This confusion probably arose as the original material of A. maingayi and A. pleurocarpus were annotated with the same Kew distribution number (Kew distribution no. 34). Furthermore, BM 001014846 bears pencil markings of both names (A. maingayi and A. pleurocarpus), with a type label wrongly indicating it as the isotype of A. maingayi. However, the morphology of the specimen and the annotation ‘ 3261 ’ clearly indicate that it is actually the isolectotype of A. pleurocarpus. 3. A mixed gathering from L (Fig. 2 c) may contain an isolectotype of A. maingayi. This specimen bears a label that indicates Maingay’s name, the Kew distribution no. 34 and the names of both species (A. maingayi and A. pleurocarpus). It also has an annotation slip (not glued to the sheet) that states ‘ 2617 ’, ‘ Malacca’ and ‘ 1867 ’, corresponding to the collection number, type locality and year of collection, respectively. The specimen comprises three twigs, two detached leaves and a monocarp enclosed in a packet: the twig bearing a single flower on the right and the monocarp in the packet (outlined in red boxes in Fig. 2 c) constitute a probable isolectotype of A. maingayi, the twig bearing many flowers on the left and the two detached leaves (with cuneate leaf base) are clearly A. pleurocarpus, and the twig at the bottom cannot be identified as it lacks leaves and flowers. 4. Multiple gatherings of Haviland and Beccari were cited in the protologue of A. havilandii. Some of them (e. g., Beccari 381, Beccari 713, Haviland 1629) correspond to Ridley’s description but two of them (Beccari 554 and Haviland 3340) represent another species, Artabotrys roseus Boerl. Turner (2009) selected Haviland 1629 from K (Fig. 2 d) as the lectotype of A. havilandii and reduced A. havilandii to a synonym of A. maingayi as the types of the two names are conspecific and A. maingayi represents the earliest legitimate name. It should be noted that Haviland & Hose 1629 A from K (K 000691273) and Haviland & Hose 1629 E from L (L 0180468) are not types of A. havilandii. The former was collected on 13 November 1894 and the latter was collected on 26 October 1894, both of which are later than the date of collection of Haviland 1629 (9 September 1892).	en	Chen, J., Baldini, R. M. (2020): Flora of Singapore precursors, 19: Nomenclatural notes on Artabotrys (Annonaceae) and Magnolia (Magnoliaceae). Blumea 65 (3): 179-187, DOI: 10.3767/blumea.2020.65.03.01, URL: https://doi.org/10.3767/blumea.2020.65.03.01
03EB8743FFF8BC26FFEB6317FF62FD73.taxon	description	Notes — 1. The protologue of Aromadendron elegans does not mention any specimen, but states “ in sylvis circa viam Lebak Provinciae Bantam, necnon in montosis Salak et Gede ” (i. e., in the forest near Lebak Road, Bantam Province, and in the moun- tain regions of Salak and Gede). All the localities mentioned in the protologue are in West Java, Indonesia. Nooteboom (1987) made a first step lectotypification by citing ‘ Type: Blume 215 (L; iso BO), Java’. There are, however, two sheets of Blume 215 in L that are not cross-labelled and hence represent duplicates. According to Art. 9.17 of the ICN (Turland et al. 2018), a designation of a lectotype that is later found to refer to more than one specimen of a single gathering must be accepted, but may be further narrowed to a single one of these specimens by a subsequent lectotypification. Therefore, we here designated L 0038349 (Fig. 4 a) as the second-step lectotype. Nooteboom (2012) regarded L 0038349 as a holotype; however, the mention of multiple localities in the protologue indicates that multiple specimens were seen by Blume. In addition, Nooteboom (2012) cannot be considered to have selected a second-step lectotype as his type statement lacks the phrase ‘ designated here’. 2. Korthals (1850) merely cited a locality (Kassan, Sumatra) without mentioning any specimen in the protologue of Aromadendron glaucum. The first effective typification of Aromaden­ M. glauca var. sumatrana (Miq.) Dandy, he did not formally typify the former name as he has not seen its type. According to Nooteboom, Korthals also wrote the name M. oortii on another sheet that was later re-identified as Manglietia macklottii Korth. by Korthals himself. This specimen also bears a label with the locality ‘ Singalang’, corresponding to Korthals’ protologue of M. oortii. However, Nooteboom considers this specimen to be the type of Manglietia macklottii, which was also collected from Singalang, Sumatra (see discussion below under Magnolia macklottii). Nooteboom lectotypified M. oortii using the specimen L 0038346 (Fig. 4 c) that Dandy had rejected as being mislabelled, although Nooteboom noted “ a slight difference between the flowers [of this specimen] and their description by Korthals, which is very obscure anyhow ”. Since there is no major morphological conflict between Nooteboom’s lectotype and Korthals’ protologue, Nooteboom’s choice of lectotype has to be followed, according to Art. 9.19 of the ICN (Turland et al. 2018).	en	Chen, J., Baldini, R. M. (2020): Flora of Singapore precursors, 19: Nomenclatural notes on Artabotrys (Annonaceae) and Magnolia (Magnoliaceae). Blumea 65 (3): 179-187, DOI: 10.3767/blumea.2020.65.03.01, URL: https://doi.org/10.3767/blumea.2020.65.03.01
03EB8743FFFDBC27FFEB66A6FD06FC4E.taxon	description	2. When discussing Magnolia macklottii var. beccariana, Nooteboom (1987) cited the basionym as ‘ Michelia beccariana Agostini’ immediately following the proposed combination on page 348. Nooteboom (1987) provided a full reference of the basionym ‘ Agostini, Atti Com. Accad. Fisiocrit. Siena IX, 7 (1926) sep. 23 ’ on page 347, where Michelia beccariana is cited as a synonym of Magnolia macklottii. Although Nooteboom’s citation is erroneous (the series and volume number are erroneous, and the day and month of publication are provided in place of the page number), these errors do not prevent valid publication of the new combination according to Art. 41.6 of the ICN (Turland et al. 2018). Agostini (1926) cited a single gathering ‘ Becc. P. S. n. ° 116 ’ from ‘ monte Singalan’ in the protologue of Michelia beccariana. In the brief introduction to her paper, it was mentioned that she consulted the Webb’s Herbarium in FI, which is in the same room as Beccari’s Herbarium (FI-B). We located only a single specimen of Beccari PS 116 in FI-B (Fig. 5 b), which represents the holotype of Michelia beccariana. Nooteboom (2012) correctly stated that the holotype is deposited in FI. 3. Although King cited a single gathering ‘ Maingay No. 17 ’ in the protologue of Magnolia maingayi, several duplicates in various herbaria exist. These are considered as syntypes and it is unclear what material was used by King when preparing his description. It should also be noted that the no. 17 refers to the Kew distribution number rather than Maingay’s collection number. By indicating ‘ Type: Maingay 17 (?; iso L) ’, Nooteboom (1987) is quite explicitly excluding the L specimen from the possibility of it being the holotype. This seems to make sense as Maingay’s types are largely kept in K. Moreover, the L specimen consists of only a few detached leaves, a few seeds and fragments of petals, whereas the K specimen (mounted over two sheets) has intact twigs with leaves, flowers and fruits, consistent with King’s description. Therefore, the K specimen (Fig. 5 c, d) is here selected as lectotype. Nooteboom (2012) incorrectly indicated that the holotype is deposited in CAL.	en	Chen, J., Baldini, R. M. (2020): Flora of Singapore precursors, 19: Nomenclatural notes on Artabotrys (Annonaceae) and Magnolia (Magnoliaceae). Blumea 65 (3): 179-187, DOI: 10.3767/blumea.2020.65.03.01, URL: https://doi.org/10.3767/blumea.2020.65.03.01
03EB8743FFFCBC28FFEB67FDFF29FEA7.taxon	description	2. When Keng (1978) transferred Talauma candollei Blume (Blume 1823) to Magnolia, he made a new combination ‘ Magnolia decandolli ’ that was intended to be a replacement name in order to avoid generating a later homonym of Magnolia candollei Link (Link 1831). However, Magnolia candollei and ‘ Magnolia decandolli ’ are to be considered orthographic vari- ants (commemorating Augustin Pyramus de Candolle) and thus homonyms. An even earlier homonym exists, i. e., Magnolia decandollei Savi (Savi 1819), which makes Keng’s name il- legitimate and also Nooteboom’s combination which is based on Keng’s name. 3. Agostini (1926) cited a single gathering (Beccari PB 2102) collected in 1866 in her protologue of Talauma kutcinensis. Agostini spelt the specific epithet as ‘ kutcinensis ’ whereas Nooteboom (1987, 1988) spelt the specific epithet as ‘ kuteinensis ’. ‘ Kutcin’ is probably an orthographical variant of Kuching, an administrative division in Sarawak, whereas ‘ Kutei’ is probably an orthographic variant of Kutai, a historical region in East Kalimantan. Because Beccari travelled largely within Kuching in 1866 (Van Steenis-Kruseman 1950), the original spelling of the specific epithet ‘ kutcinensis ’ should be used. Agostini (1926) mentioned that she consulted the Webb’s Herbarium in FI, where we located a sole specimen of Beccari PB 2102. Thus, the FI-B specimen (Fig. 6 b) is the holotype of Talauma kutcinensis.	en	Chen, J., Baldini, R. M. (2020): Flora of Singapore precursors, 19: Nomenclatural notes on Artabotrys (Annonaceae) and Magnolia (Magnoliaceae). Blumea 65 (3): 179-187, DOI: 10.3767/blumea.2020.65.03.01, URL: https://doi.org/10.3767/blumea.2020.65.03.01
03EB8743FFF3BC28FFEB65D4FDD0FAF6.taxon	description	Notes — 1. Miquel mentioned “ Sumatra orient. in prov. Palembang, prope Muara-enim (T.) ” in his protologue of Talauma villosa. It is clearly indicated in the footnotes that ‘ T. ’ refers to Teijsmann. However, there is neither indication of a particular herbarium in which Teijsmann’s specimen was deposited nor evidence that only a single specimen is used for Miquel’s description. The type statement “ Type: Teijsmann HB 3690 (L; iso BO) ” provided in Nooteboom (1987) constitutes an effective lectotypification; thus, the L specimen (Fig. 7 a) is the lectotype of Talauma villosa. Nooteboom (2012) erroneously stated that the holotype is deposited in L. 2. Hooker & Thomson (1872) merely cited “ Eastern Peninsula, Griffith ” in the protologue of Talauma lanigera. We located only a single specimen in K (K 001292330) that was collected by Griffith, labelled as Talauma lanigera, and which has morphological features corresponding to the protologue (e. g., presence of immature, densely woolly carpels and absence of ripe fruits). Thus, K 001292330 (Fig. 7 b) is the holotype of Talauma lanigera. Although Nooteboom (1987, 2012) indicated “ Type: Griffith 65 ”, the ‘ 65 ’ actually refers to the Kew distribution number rather than Griffith’s collection number. Nooteboom (2012) correctly stated that the holotype is deposited in K.	en	Chen, J., Baldini, R. M. (2020): Flora of Singapore precursors, 19: Nomenclatural notes on Artabotrys (Annonaceae) and Magnolia (Magnoliaceae). Blumea 65 (3): 179-187, DOI: 10.3767/blumea.2020.65.03.01, URL: https://doi.org/10.3767/blumea.2020.65.03.01
