Phagnalon graecum Boiss. & Heldr.

Montes-Moreno, Noemí, Garcia-Jacas, Núria, Nualart, Neus, Susanna, Alfonso & Sáez, Llorenç, 2018, Typification of plant names referable to Phagnalon (Compositae) with some taxonomic notes, Phytotaxa 360 (1), pp. 1-18 : 6

Phagnalon graecum Boiss. & Heldr. in Boissier (1849: 6)

≡ Conyza graeca (Boiss. & Heldr.) Nyman (1855: 16)

≡ Phagnalon rupestre var. graecum (Boiss. & Heldr.) Fiori in Fiori, Béguinot & Paoletti (1903 –1904: 284)

≡ Phagnalon rupestre subsp. graecum (Boiss. & Heldr.) Batt. in Battandier & Trabut (1889: 444)

Type:—[ GREECE.] Ad rupes circa Pylos, May 1844, Heldreich s.n. (lectotype: G barcode G00300631 [digital image!], designated here; isolectotypes: BM!, G barcode G00300630 [digital image!], P barcode P02829201 [digital image!], W number W0025366 !, W number W0025367 !) .

(Image available at: http://www.ville-ge.ch/musinfo/bd/cjb/chg/adetail.php?id=225124).

Notes:— Boissier (1849) cited several syntypes in the protologue, all of them from Greece: “Hab. in rupestribus regionis calidae totius Graeciae ad Patras, Nauplia (Boiss.), Atticâ ad Acropolim (Sprun.), Messeniâ ad Pylos (Heldr.) Cretâ (Sieb. sub Con. saxatili [“sexatili”]), Zacyntho (Marg.)”. We also considered as a potential lectotype the specimen W 0025369 because morphologically it agrees better than the G specimen with the taxonomic concept of subsp. graecum (see explanations below). However, the W specimen has no data locality and collection date in the labels and the only information about its origin comes from indirect sources [ L. Anton, pers. comm.; Stafleu & Cowan (1976 –1988)]. Therefore, we selected as the lectotype the specimen deposited at the G general herbarium that morphologically agrees with the original description and has data about locality and collection date. Specimens from the center and east of the Mediterranean basin have capitula with a greater number of rows of involucral bracts, the wider outermost bracts being triangular and acute, and a greater number of hermaphrodite florets per capitulum. These specimens were recognized by Boissier (1849) and Battandier & Trabut (1889) at specific or subspecific levels [ Ph. graecum Boiss. & Heldr. , Ph. rupestre subsp. graecum (Boiss. & Heldr.) Batt. ]. The study of specimens from the whole distribution area of Ph. rupestre reveals that these morphological discontinuities are constant and restricted to a delimited geographic area. Moreover, according to molecular data, Ph. graecum and Ph. rupestre form a highly supported monophyletic group ( Montes-Moreno et al. 2010). Therefore, we conclude that this combination of morphological traits allows to distinguish Ph. graecum as a subspecies of Ph. rupestre .