Typification and Taxonomic Remarks on Names of Iris (Iridaceae) Associated with the Turkish Flora

In view of the forthcoming review of Turkish irises for Resimli Türkiye Florası (The Illustrated Flora of Turkey), nineteen names are typified or nomenclatural remarks are provided in the present report. Lectotypes are designated for Iris aschersonii, I. attica, I. bornmuelleri, I. purpureobractea (a taxonomic synonym of I. junonia), I. reticulata var. cyanea, I. reticulata var. sophenensis, I. suaveolens and I. taochia. The second-step lectotypification is made for I. histrio. Neotypes are designated for the names I. histrioides, I. junonia, I. masiae and I. reticulata var. histrioides; epitype, for the name I. reticulata var. sophenensis. For the previously typified names, I. bakeriana, I. musulmanica and I. reticulata, lectotypes are given. The lectotypes for I. histrio var. aintabensis, I. schachtii, and Xiphion danfordiae and the authorship for I. histrioides, are corrected. Images are provided for eight specimens selected as types that are not available online. Notes on distribution in Turkey are provided for all the accepted taxa.


Introduction
Iris L. is a taxonomically challenging and the largest genus in Iridaceae with ca. 280 species, distributed mostly in the Northern Hemisphere [1]. In Turkey, it is represented by rhizomatous or bulbous perennials growing in various habitats. Boissier [2] mentioned a few Iris species from Turkey for the first time. From a taxonomic point of view, the study of Turkish irises of the section Oncocyclus (Siemssen) Baker and subgenera Hermodactyloides Spach and Scorpiris Spach was initiated by Güner and Peşmen [3]. Mathew [4] was the first to carry out a complete taxonomic overview of the Iris species from Turkey. According to Mathew, in Turkey, the genus comprises 37 species. The recent Checklist of the Turkish flora by Güner [5] includes 50 species.
Recently, many taxonomic studies on plant species are oriented towards typification in order to establish choices of type or to identify cases where corrections are necessary to avoid nomenclatural disruption (e.g., [6]). During the revision of the genus Iris for the forthcoming Volume 3 of Resimli Türkiye Florasi (The Illustrated Flora of Turkey) by the second author, it became evident that most Turkish iris names required typification. Besides, some species names had been cited incorrectly or with incorrect authorship [7,8]. This report aims to summarize information concerning the previously typified names and to select the types of the remaining untypified names associated with the Turkish flora.

Materials and Methods
This work is based on a comprehensive study of the relevant literature, in addition to the protologues. Herbarium specimens deposited at AEF, ANK, B, BAK, BEI, BM, E, G, HUB, ISTE, JE, K, KFTA, LD, LE, M, MHA, NGBB, P, TBI, TGM, W, WAG, and WU (herbarium codes according to Thiers [9]) were examined. The lectotypes, epitypes, and neotypes are designated here following the recommendations of the International Code of Nomenclature for algae, fungi, and plants [10] (ICN). The taxa are arranged in alphabetical order of the names under which they were originally described. The accepted names are highlighted in bold. The relevant information indicated in the protologue ("Protologue citation") is provided here for all the names. Protologues and labels, originally composed in Russian and Turkish, were translated into English. Specimens were physically seen (!) unless indicated otherwise (i.e., [digital image!]). All specimens are cited in full, and most of them were assigned the barcode numbers following the herbarium acronyms. For the specimens deposited at AEF, HUB, ISTE, and TGM, inventory numbers are indicated. A conservative taxonomy of Iris is here used (e.g., [5,11,12]). For each accepted taxon, notes on distribution in Turkey are provided.

Typification of the Names
(1) Iris aschersonii Foster, Garden (London 1871-1927, 61(1): 288, 1902 ≡ I. grant-duffii var. aschersonii (Foster) Hayek, Ann. K. K. Naturhist. Hofmus. 28: 181, 1914  Notes-Iris aschersonii was described by Michael Foster from a dried plant and photographs received from Walter Erdmann Siehe [13]. Two images from the protologue refer to the original material of I aschersonii. The specimens from Siehe's exsiccatum, accompanied by labels with the printed notes "Flora Orientalis" and "Ed. W. Siehe, Mersina", may belong to the original material of I. aschersonii. As indicated on the labels attached to this exsiccatum, the plants were found in the Cilicia plain on limestone near Adana in 1898 and recognized as a new species by Foster in The Garden. Therefore, we believe that this exsiccatum and the dry plant indicated in the protologue refer to the same gathering, and are hence syntypes (see Art. 40, Note 1 of the ICN). As a consequence, the best-preserved specimen, i.e., JE00022414, is designated here as the lectotype. Iris aschersonii is endemic to Turkey, distributed in the Adana and Antalya provinces.
(  Notes-Iris attica was described by Pierre Edmond Boissier and Theodor Heinrich Hermann von Heldreich based on several gatherings collected by Boissier, Giacomo (or Jiacinto) Guicciardi-Barazetti, Theodor von Heldreich, and Wilhelm von Spruner in Greece [14]. The specimen from Boissier's own herbarium ( Figure 1) is designated here as lectotype because it matches the protologue and is the most informative one. It was collected near Mount Parnassus, central Greece, by Heldreich in 1852. Iris attica is close to I. suaveolens Boiss. & Reut., from which it differs by its longer perianth tube and bracteoles tightly sheathing the ovary and perianth tube. In Turkey, it occurs generally in western Anatolia (Balıkesir, Bilecik, Bolu, Çanakkale, Eskişehir, and Kütahya provinces) and has a fragmented distribution pattern (also see [4]). Notes-Iris bornmuelleri was described by Heinrich Carl Haussknecht [15] from plants collected by Joseph Bornmüller. According to Haussknecht [15], Bornmüller gathered plants in the Amasya Province, northern Turkey, in mid-March 1889. The protologue of I. bornmuelleri cited the habitat and geographical data. However, it did not provide exact citation of specimens. Two gatherings collected by Bornmüller in the Amasya Province in mid-March 1889 were found. The specimen JE00020029 in Haussknecht's own herbarium [16] matches the protologue of I. bornmuelleri and is designated as the lectotype. The specimens of the exsiccatum, annotated as "J. Bornmüller, Plantae Anatoliae Orientalis", can also be referred to the original material of I. bornmuelleri. As it follows from the content of the printed labels, the plants of this exsiccatum were collected by Bornmüller in mid-March 1889 in the area from which I. bornmuelleri was described by Haussknecht. Baker [17] and Mathew [18] considered I. bornmuelleri as a taxonomic synonym of I. danfordiae (Baker) Boiss., which is correct. Notes-Iris histrio was described by Heinrich Gustav Reichenbach, based on two cultivated plants and two specimens collected in Lebanon and belonging to the series "Herbier de Syrie" [19]. We were not able to locate the specimens examined by Reichenbach, but only duplicates of the specimens of the "Herbier de Syrie", i.e., the no. 93, collected by Gaillardot in "Sommet du Liban, entre les deux mamelons appelès Taumets et Djerzine, fiès de Saïda". We found these duplicates at G, K, and P. They are accompanied by labels with the printed note "No 93. HERBIER DE SERIE. 1855" and were initially identified as I. reticulata M.Bieb. In his work, Mathew [4] indicated "iso. K!" as the "type" of I. histrio. Thus, he actually designated the isosyntypes at K as the lectotype. However, we traced two isosyntypes at K (K000499055! & K000499058!). According to Art. 9.17 of the ICN, the type citation by Mathew [4] can be further narrowed to a single specimen by a second-step lectotypification. Therefore, K000499055!, which is more informative, is designated here as the lectotype of the name I. histrio. In Turkey, I. histrio is distributed in the Adana, Gaziantep, Hatay, Kahramanmaraş, Mersin, and Osmaniye provinces.  Notes-Iris histrioides was described by Gustavus Reuthe (Fox Hill Nursery, UK) from cultivated plants without indicating the collection locality [20]. The original material on which this name was based has not been found in the investigated herbaria. Therefore, a neotype may be selected according to the Art. 9.13 of the ICN. The neotype ( Figure 2) is a complete specimen collected in Amasya Province, Turkey. It clearly shows the features of the inner perianth segments, the markings aggregated towards the centre of the lamina, which is characteristic of I. histrioides. Iris histrioides is an endemic to Turkey, distributed in the Amasya, Gümüşhane, Rize, Samsun, Sinop, and Tokat provinces (also see [10,21,22]).  Notes-The name Iris junonia was validly published in a paper by Heinrich Wilhelm Schott and was ascribed there to Schott and Kotschy [23]. As follows from the protologue, the name was based on plants collected in the Cilicia region within the Taurus Mountains, southern Turkey, by Karl Georg Theodor Kotschy. According to Baytop [24], Kotschy botanized in the Taurus Mountains from June to early October 1853. Kotschy's and Schott's types are deposited at W, and some of the important material is in the Haynald Herbarium at BP [16,25]. Mathew [4] noted on the original material of I. junonia as follows: "holo. W -specimen not traced, perhaps destroyed". The original material of I. junonia at W was destroyed by fire during World War II in 1943 (C. Bräuchler, pers. comm.), and no original material is now present at BP (I. Rácz, pers. comm.). Consequently, neotypification is required according to the Art. 19.11 of the ICN. For this purpose, we selected a specimen ( Figure 3) collected in the Cilician Taurus Mountains, the region of Turkey from which the species had been described. Iris junonia is an endemic to Turkey, frequently found along the Taurus Mountains and in montane areas of western Anatolia in the Adana, Antalya, Bilecik, Burdur, Denizli, Isparta, Karaman, Kayseri, Kocaeli, Konya, Mersin, Mugla, Nigde, and Sakarya provinces. Currently, I. junonia is an accepted species, close to I. germanica. Notes-Iris masiae was described by Foster [13] based on a cultivated plant originated presumably from Asia Minor. Foster stated that I. masiae is a purple-flowered relative of I. aschersonii and I. grant-duffii Baker. Dykes reported [26] that this species was discovered by Paul Sintenis in South Anatolia, Turkey, in 1888, and then introduced into cultivation by Max Leichtlin as follows: "an undetermined species gathered by Sintenis in the course of his journey in the East in 1888 on the steppes near the village of Süverek, on the lower slopes of the Karadja Dagh, a mountain which lies in the district of Diarbekr [Diyarbakır Province] in northern Mesopotamia". We found duplicates of the same gathering made by Sintenis near Siverek, which is closer geographically to Diyarbakır, at G, JE, K, LD, P, and WU. According to the results of our research, no original material for I. masiae is known to exist. For this reason, K001291625! is here designated as neotype since it is the most informative. All the above-mentioned I. masiae isoneotypes are accompanied by labels with the printed note "P. Sintenis: Iter orientale 1888. Kurdistania. det. Dr. O. Stapf", and the rest of the information was apparently handwritten by Stapf. In Turkey, I. masiae is distributed in the Diyarbakır, Gaziantep,Şanlıurfa, andŞırnak provinces. It comes nearest to I. aschersonii, from which it differs chiefly in the shape of the perianth segments [26]. Notes-Iris purpureobractea was described by Mathew and Turhan Baytop from plants collected on Mount Honaz, Denizli Province, southwestern Turkey [27]. The authors cited one gathering "ISTE 25725" and the cultivated plants ("ISTE 29712") raised from the rhizomes of the original collection as "Type" of I. purpureobractea. As it follows from the protologue, the holotype was deposited at K and the isotype at ISTE. However, it remains unclear what part (a single specimen) of the type material was indicated by the authors as holotype. In 2014, when one of the co-authors of the present report (E.V.B.) visited Kew, the original material of I. purpureobractea was represented by at least two specimens cited above. One of them, mounted on a sheet bearing the number "ISTE 25725", cited in the protologue, is currently absent from the Kew collection (A. Haigh, pers. comm.). Hence, the second specimen ( Figure 4) is designated here as lectotype. In the present study, I. purpureobractea is considered a synonym of I. junonia due to their morphological similarities. For example, in the natural populations of I. purpureobractea, yellow (as in I. junonia), blue, and purple forms are normally represented. Furthermore, not all individuals have purple bracts, which is a characteristic trait distinguishing I. purpureobractea from I. junonia. It was assumed [28,29] that I. junonia is intermediate between I. pallida Lam. and I. germanica L. or a hybrid of I. pallida and I. variegata L. [12]. A suggestion was also made that it has an ancient origin and is probably a form of I. germanica [12,30]. Notes-Iris reticulata var. cyanea was described by Eduard August von Regel from a cultivated plant without indicating the collection locality [31]. The protologue is accompanied by a colour drawing which is designated here as lectotype. This drawing is sufficiently detailed to provide a precise application of the name. The most characteristic diagnostic feature of I. histrio is that the lamina of the outer perianth segments is sparsely, however conspicuously blue blotched against the whitish background all over its surface. This diagnostic feature is clearly visible in the illustration published in the protologue of I. reticulata var. cyanea. Regel [31] pointed out that this variety is very similar to I. histrio. In our opinion, the two taxa are identical. Notes-Iris reticulata var. histrioides was described by George Fergusson Wilson from cultivated plants without indicating the collection locality [32]. In a brief description, Wilson specified that Leichtlin put these plants into commerce. Foster [22] (pp. 8-9) made the necessary explanations concerning the nomenclature history of this name. He noted that the plants had been sent to him before 1892 by Mary Wright from the American Mission in Amasia, and, since then, Leichtlin has obtained a large supply of such plants. Based on the fact that many specimens strongly resembled I. histrio in colour, Foster proposed to refer to it as "Iris reticulata var. histrioides". Mathew [4] noted that the type specimen of the name had not been found. Our attempts to find the original material for this name in the framework of this study have not been successful either. The specimen designated here as a neotype for I. reticulata var. histrioides refers to Bornmüller's exsiccatum "Pl. exs. Anatoliae orientalis a. 1889" collected in the Amasya Province, northern Turkey, where he carried out explorations in 1889 [24,33]. In his work, Bornmüller [33] noted that the iris he found was characterized by a beautiful colour, which varied a lot, and, therefore, could easily compete with that of I. reticulata ("sie zeichnet sich durch einen wundershönen Farbenschmuck, der viel variirt, aus, der sich leicht mit dem der iris messen kann"). The notes on the labels indicate that the colour of flowers in I. reticulata var. histrioides was azure (intense blue), which is a characteristic feature of I. histrioides.  Notes-Iris reticulata var. sophenensis was described and introduced into cultivation by Foster based on the plants from the bulbs sent by Mary E. Barnum from the American Mission at Harput (Elazıg Province, Turkey) in October 1884 [34]. These bulbs were gathered by Barnum in Sophene, the ancient name of the area around Harput. We found a specimen represented by a cultivated plant (K000499061!), which is the original material for the name. This sheet is accompanied by an envelope with dissected flowers and a colour illustration of the flower organs, together with Foster's letter to J. Baker. However, it lacks a whole plant. Therefore, in order to avoid any ambiguity in the interpretation of the lectotype, an epitype is here designated ( Figure 5). The selected epitype, is recently collected material from a locality geographically close to the type one, representing the same taxon as that to which the name was applied [22,28,35]. It provides additional characters that will contribute to the taxon identification. In the present study, this plant is treated as endemic to the Mardin Province, Turkey. Notes-Iris suaveolens was described by Boissier and George François Reuter from plants cultivated at the Botanical Garden of Geneva, Switzerland [36]. These plants were raised from the rhizomes collected by Boissier near Constant , a (formerly known as Küstendje), a city in the Northern Dobruja region, Romania. The specimen designated here as lectotype (G00164600) bears two labels containing Reuter's handwritten [37] notes with a Latin description, which is consistent with the information provided in the protologue of I. suaveolens. The specimens from G are dated before May 1854, when I. suaveolens was published [38], and can also be considered the original material of this name. Iris suaveolens frequently grows in western and northern parts of Turkey, and differs from I. attica by its stem with 1-2 terminal flowers, equal in length, navicular, sharply keeled bracts and bracteoles, and ovary not tightly sheathed by bracteoles [4,5,39]. In Turkey, I. suaveolens is distributed in the Amasya, Ankara, Aydın, Balıkesir, Bilecik, Bursa, Çanakkale, Eskişehir, Istanbul,İzmir, Karabük, Kastamonu, Kırklareli, Kocaeli, Konya, Kütahya, Manisa, and Samsun provinces.  Notes-Iris taochia was described by Alexander Alfonsovich Grossheim from cultivated plants that originated from Oltinsky Okrug of Kars Oblast, the Caucasus Viceroyalty of the Russian Empire in 1878-1917, which is part of today's Erzurum Province, Turkey [40]. No specimen or gathering is indicated in the protologue. Grossheim ascribed the name I. taochia to Jurij (Georg) Woronow, a Russian botanist and plant collector. Since 1914, Woronow had worked at the Caucasus Museum (Tbilisi, Georgia) and, between 1921-1925, at the Tiflis Botanical Garden, except for a short period when he was engaged by the Botanical Museum of Academy of Sciences in St. Petersburg [41]. According to this information, we found Woronow's specimens at TGM. All the selected specimens are accompanied by a printed label with the note "Notae criticae. G. Woronow", on which Woronow handwrote "Iris taochia m. October [19]23". While preparing his work [40], Grossheim was in Tbilisi [42] and used the TGM collection. Hence, the TGM specimens, in our opinion, can be considered the original material of the name I. taochia. One of them ( Figure 6) is designated here as lectotype because it matches the protologue and is the most informative one. Iris taochia is a narrow endemic species to Erzurum Province, Turkey. In addition to Oltu District, it has been recorded from the vicinity of Tortum [4,39]. Notes-Iris bakeriana was described by Foster from the plants that had been raised from the bulbs sent by Frank Gates of the American Mission in Mardin in 1887 which flowered in cultivation in February-March 1889 [44]. The bulbs were gathered in south-eastern Turkey. In citing "Typus . . . : Armenia, Rev. G.F. Gates, cult. K!", Wendelbo & Mathew [43] designated the specimen in K as the lectotype of I. bakeriana satisfying the requirements of Art. 7.11 of the ICN. At K, we found a single specimen (K000499060!) apparently related to the illustration from the protologue. In agreement with Mathew [18,21], we also accept I. bakeriana. In Turkey, it occurs only in the Batman and Mardin provinces. Iris bakeriana is very closely allied to I. reticulata, however differs chiefly in the nearly cylindrical, not tetragonal (a diagnostic character of I. reticulata), leaves [44]. Notes-Iris histrio var. aintabensis was described by George Percival Baker without indicating the collection locality [45]. According to the protologue, this variety was exhibited at a Royal Horticultural Society show. Baker travelled to numerous mountain areas where he collected plants and subsequently introduced them into British gardens [46]. The epithet "aintabensis" is most likely derived from "Aīntāb" or "Antep", the former (Ottoman) name for the modern city of Gaziantep located in the western part of the Southeastern Anatolia Region. This probably explains why Mathew [4,18,21] specified that the bulbs of I. histrio var. aintabensis had been collected near Gaziantep. Mathew [18] indicated that the holotype of I. histrio var. aintabensis was deposited at K as follows: "Type: S Turkey: a plant cultivated by G.P. Baker, exhibited at a Royal Horticultural Society Show, 10 Feb 1931, from bulbs collected near Gaziantep (holo. K!)". We found a specimen from a cultivated plant in K which matches the information from Mathew's work (Figure 7). This specimen was originally identified as "Iris histrio var. aintabensis". It is possible that only a single specimen ever existed (which, in this case, would be the holotype), however, this cannot be established for certain because the name I. histrio var. aintabensis was published without a holotype. Therefore, a lectotype designation is required (Art. 9.3 of the ICN), and besides, the term "holotype", used by Mathew [18], should be corrected to "lectotype" according to the Art. 9.10 of the ICN. Despite that Mathew [4,21] assigned this plant the taxonomic rank of variety, Crespo et al. [47] and Rukšāns [48] consider it a separate species. According to the protologue [45], the colours of the flowers in I. histrio var. aintabensis are characterized as Cambridge Blue with golden markings on the outer perianth segments. In our opinion, this name is a taxonomic synonym of I. histrio, a widespread species from southern Turkey. Adil Güner has studied I. histrio in natural populations over many years and has come to the conclusion that there are no morphological differences between I. histrio var. aintabensis and the autonymic variety, as the colour pattern of the blade is very variable within the species.

Conclusions
In Turkey, Iris s.l. is one of the richest genera in terms of number of species. The present contribution is a part of the full taxonomic revision of the genus for the Turkish flora carried out by A. Güner for Resimli Türkiye Florasi (The Illustrated Flora of Turkey) and a part of E.V. Boltenkov's continuing research on the Iris taxonomy, aiming to update the nomenclature and extend the existing systematic knowledge. We have summarized the previously published typifications for some names and designated the lectotypes/neotypes for the thirteen untypified names. Furthermore, a study of the original material and field observations on I. purpureobractea have shown it to be a synonym of I. junonia.