Tagetes dombeyi (Asteraceae, Tageteae), a new species from the Central Andes first collected in the 18th century

Abstract Tagetes dombeyi (Asteraceae, Tageteae), a new species from the Central Andes first collected in the 18th century. A new annual species of Tagetes from the Central Andes of Peru and Bolivia, T. dombeyi, is described and illustrated based on herbarium collections and photographs of the plants in their habitat. Tagetes dombeyi is closely related to Tagetes imbricata, Tagetes multiflora - both of which can occasionally be found mixed with the former on the same herbarium sheet - and Tagetes iltisiana. However, they can be distinguished by diagnostic characters of leaf, corolla, and pappus. A morphological description is provided along with an illustration, a range map, and a key to differentiating the Tagetes species that inhabit Peru and Bolivia.

Rodriguésia 74: e00162023.2023 Tagetes includes annual or perennial herbs, subshrubs, and shrubs, strongly aromatic due to the presence of secretory cavities in leaves and phyllaries; leaves opposite below, most of the upper ones alternate, with usually pinnately dissected blades; capitula radiate or subradiate, with phyllaries fused in one series; and a pappus of few scales in one or -more rarely -two series, each scale truncated or ending in a single long awn (rarely 3-5-awned) at the apex (Robinson 1973;Soule 1993;Schiavinato & Bartoli 2018).
As part of our integral taxonomic revision of Tagetes, in recent years we have been working in depth on a complex of South American species that are similar in appearance as they have an annual growth habit, their capitula are arranged in lax corymbiform capitulescences, and the involucres of phyllaries are partially purplish.Characters such as shape and color of leaf blades, number of phyllaries, color of corollas, and shape and size of pappus scales have proven useful for differentiating species, especially when working with early and/or depauperate herbarium specimens.An exhaustive morphological study of these diagnostic characters has allowed us to recircumscribe some very littleknown species (Schiavinato & Bartoli 2018, 2019a) and to describe a new species (Schiavinato & Bartoli 2019b).Continuing with our work in this complex, we have recently detected several herbarium specimens from the Central Andes of Peru and Bolivia that show a combination of the aforementioned diagnostic characters that does not match any of the known species of the genus.This evidence, supported by photographs of the plants in the wild, allows us to describe a species new to science.

Material and Methods
The description of the new species was based on 29 herbarium specimens housed at LIL, LP, MA, NY, P, and US (Thiers, continuously updated), which were cross-checked against the type specimens of all accepted species of Tagetes.When physical specimens were unavailable, digital images were obtained from GBIF.org (2023), JSTOR (2023), or requested from the curators.The morphological study was complemented with photographic records of the new species available at iNaturalist (Lindqvist 2018).
General terminology for morphological and anatomical structures follows Beentje (2010).Terminology for leaf architecture follows Ellis et al. (2009), and for pappus morphology, it follows Roque & Bautista (2008) and Beentje (2010).When comparing the new species with its congeners, leaf blade measurements were taken in all cases from leaves located in the middle zone of the stems, excluding the reduced or bractiform upper ones to avoid distortions.The flowering and fruiting period of T. dombeyi was determined by cross-checking the phenological stages of the plants against the information available on the labels, considering only those specimens in which the reported collection date corresponds to a period of no more than two months.
The range map of the new species was made with QGIS.
Flowering and/or fruiting specimens of Tagetes dombeyi were collected from March to May, at the end of summer and the first half of autumn in the Southern Hemisphere, coinciding with the other annual Tagetes that inhabit that area (Soule 1996).Paratype J. Dombey s.n.(P 00673391 [ex Herb. Jussieu 9021]) has a label annotation that reads "floret novembri" (flowering in November), which is probably a mistake; it may as well be a specimen cultivated in the Northern Hemisphere from Peruvian seeds collected by Dombey.
Dried specimens of T. dombeyi might be easily confused with other annual Tagetes with lax terminal corymbiform capitulescences that occur in the Central Andes of Peru and Bolivia such as T. iltisiana, T. imbricata, and T. multiflora.The distinguishing character states of these taxa are summarized in Table 1.Tagetes dombeyi may be also confused with early specimens of Tagetes terniflora Kunth in Humboldt, Bonpland & Kunth (1818: 154), but they can be easily differentiated because T. terniflora is the only one in the genus having dimorphic capitula (i.e., multiflorous and uniflorous capitula arranged in the same capitulescence).
Unfortunately, one or more of the diagnostic characters of T. dombeyi cited in the diagnosis are not easily observable in most of the specimens studied for this work due to inadequate herborization.To avoid ambiguity, then, we believe that the most appropriate specimen to be selected as the holotype of the new species is J. Dombey s.n.(P 02140932), which is complete and well-preserved (even having been collected more than 200 years ago) and adequately fits the description of the new species.A duplicate of this specimen is also kept at P (P 02140933).Despite the fact that both specimens only show "Pérou" on labels, information about the geographical distribution, habitat, and phenology of T. dombeyi can be obtained from paratypes.The year of collection of the holotype and isotype may be around 1778-1781 according to the itinerary of the Botanical Expedition that Dombey was part of (Álvarez López 1957; González Bueno & Rodríguez Nozal 2003).Specimen J. Dombey s.n.(P 00673391 [ex Herb.Jussieu 9021]) shows a label with a different collection locality ("habitat Lima juxta aquas").Therefore, it cannot be considered a duplicate of the holotype and is treated here as a paratype of T. dombeyi.
Some collections from the 18th and 19th centuries have mixed materials.Specimen H. Ruiz & J. Pavón s.n.housed at MA shows one plant belonging to T. dombeyi (MA 816617) and the other three to T. multiflora (MA 816616) mounted on the same sheet.This mixture of material was opportunely detected by Rosario E. Delgado-Montaño in 1995, although she identified MA 816617 as T. gracilis.On the other hand, sheet P 02509422 of L. Savatier 539 shows only one plant belonging to T. dombeyi (mounted on the bottom left) and the rest to T. multiflora, but at least two other sheets of the same gathering (P 02509423 and P 02509424) have only T. multiflora material.
Schultz (1866) listed G. Mandon 66 as Tagetes pusilla Kunth in Humboldt, Bonpland & Kunth (1818: 152) -currently a synonym of Tagetes filifolia Lagasca y Segura (1816: 28) (Soule 1993;Schiavinato et al. 2017) -and G. Mandon 65 and G. Mandon 69 as T. multiflora, but they are recognized here (at least in part) as T. dombeyi.These gatherings are made up of a large number of sheets with plants collected from various sites near Sorata in Bolivia.All the plants mounted on the sheets of Mandon 65 and Mandon 69 examined for this work belong to T. dombeyi.On the other hand, many sheets of Mandon 66 (which come from two different sites, "Lancha de Cochipata" and "via ad Chulaya") show a mixture between the new species and T. imbricata; furthermore, at least one sheet of Mandon 66 (P 04286316) with only T. imbricata and no T. dombeyi was found.Since Mandon's collections come from the same area and the information on the labels is ambiguous, it would not be unexpected that unseen duplicates of Mandon 65 and/or Mandon 69 also show mixed material.Kuntze (1898) and Robinson (2014) reported Tagetes daucoides Schrader (1833: 5) from Bolivia based on specimen O. Kuntze (NY 03227914), but it differs taxonomically from the type material of this name and is here recognized as T. dombeyi.As was indicated by Soule (1993), T. daucoides is a dubious name that may refer to a cultivated taxon of hybrid origin and has never been found in the wild; there is no evidence of its presence in Bolivia (Soule 1993;Hind 2011).

Figure 4
Figure 4 -a-d.Tagetes dombeyi in the wild (Peru, Department of Lima, Huarochirí Province) -a.habit; b-c.capitula; d. habitat (plants of T. dombeyi growing along the edge of the road [marked with an arrow]).Scale bars = 1 cm.All photographs by Annika Lindqvist.

Figure 5 -
Figure 5 -Range map of Tagetes dombeyi (stars).Created using the Free and Open Source QGIS.

Table 1 -
Morphological differences between Tagetes dombeyi and related taxa.