Polylepis sect. Sericeaee T. Boza & M. Kessler,, 2022

Boza Espinoza, Tatiana Erika & Kessler, Michael, 2022, A monograph of the genus Polylepis (Rosaceae), PhytoKeys 203, pp. 1-274 : 1

Polylepis sect. Sericeaee T. Boza & M. Kessler, sect. nov.

Diagnosis.

Trees or shrubs; lower leaflet surfaces with sericeous, lanate or villous hairs; fruits with a variable number and placement of flattened, thin or short spines, densely sericeous or villous.

Type.

Polylepis sericea Wedd.

Notes.

The sectional epithet Sericeae is a plural adjective agreeing in gender with Polylepis . This section, already defined as a group by Simpson (1986), includes species that usually have sericeous hairs on the lower leaflet surfaces and/or stipule sheaths and whose leaves usually contain many pairs of leaflets. Furthermore, fruits in this section are turbinate with a variable number of slender spines. However, not all species share these traits. For instance, P. frontinensis and P. multijuga have villous hairs on the lower leaflet surfaces, but P. frontinensis has sericeous hairs on the stipule sheaths and P. multijuga has many leaflets pairs (5-7). In the same way, P. lanuginosa and P. serrata have lanate hairs, but both have many lateral leaflets pairs. Polylepis multijuga , P. ochreata , P. pauta and P. serrata are the species with most leaflet pairs (4-7), whereas P. rodolfovasquezii just has one pair. Moreover, P. pepei and P. rodolfovasquezii have turbinate, but slightly twisted fruits with short spines. This variation may reflect that this section is likely not monophyletic, but rather has been proposed to represent a basal grade within the genus ( Schmidt-Lebuhn et al. 2006a).

The majority of species in this section are morphologically clearly distinct. Probably the two most similar species are P. pepei and P. rodolfovasquezii , which only differ in a few, partly overlapping characters. They might be treated at subspecies level, but as detailed in the Introduction, we decided not to accept infraspecific taxa because of the difficulty of deciding at which level to discriminate between species- and subspecies-level differentiation. Table 4 provides an overview of the arrangement of the taxa by different authors.

Within section Polylepis Sericeae , we recognize four subsections, based on their morphological distinctness, as follows: subsection Polylepis Lanuginosae Lanuginosae (two species) with lanate or villous lower leaflet surfaces and densely villous fruits; subsection Polylepis Pauta Pauta (three species) with 4-6 lateral leaflet pairs and lanate or sericeous lower leaflet surfaces; subsection Polylepis Sericeae Sericeae (eight species) with sericeous lower leaflet surfaces (except P. frontinensis ) and fruits with flattened spines; and subsection Polylepis Pepea Pepea (two species) with 1-2 lateral leaflet pairs, emarginate leaflet apices and densely sericeous, slightly twisted fruits with short spines.

Climatic niches in Polylepis sect. Sericeae

Many species of this section differ markedly in their climatic niches (Figs 12 and 13 ). For example, Polylepis albicans and P. sericea from subsect. Polylepis Sericeae Sericeae and P. pepei and P. rodolfovasquezii from subsect. Polylepis Pepea Pepea grow under the coldest conditions (about 5.5 °C Mean Annual Temperature - MAT), whereas other species, such P. multijuga (10.0 °C) from subsect. Polylepis Lanuginosae Lanuginosae and P. serrata (9.6 °C) from subsect. Polylepis Pauta Pauta , grow under noticeably higher temperatures. These differences of 4-5 °C correspond to 800-1000 m in elevation. Regarding Mean Annual Precipitation (MAP), P. frontinensis grows under the most humid conditions (mean of 2048 mm MAP), followed by P. canoi (1572 mm). In contrast, P. albicans and P. humboldtii on average receive only 744 mm per year, whereas P. lanuginosa (847 mm), P. pauta (941 mm) and P. multijuga (986 mm) also grow in relatively dry conditions.

Focussing on the individual subsections, the two species in subsect. Polylepis Lanuginosae Lanuginosae show minor ecological differences and replace each other geographically. In subsect. Polylepis Pauta Pauta , P. longipilosa and P. pauta from northern Ecuador have quite similar climatic niches and replace each other geographically, whereas P. serrata from Peru grows under substantially higher temperatures and higher precipitation. The two very similar species of subsect. Polylepis Pepea Pepea have identical niches and complementary geographical distributions. These species clearly form a vicariant species pair, suggesting allopatric speciation after geographical isolation. Finally, in subsection Polylepis Sericeae Sericeae , there are major differences among almost all species, with only P. albicans and P. humboldtii having similar climatic niches, but these are geographically well separated. Indeed, all species of this subsection are geographical vicariants, except for P. argentea and P. canoi , which broadly overlap geographically, but have quite different niches, with P. argentea growing under colder and drier and P. canoi under warmer and more humid conditions. These marked ecological differences between species show that they are evolutionarily and ecologically independent lineages and support their treatment as separate species.

Figures

Figure 12. Box plots showing the climatic niches of the species of the subsections Lanuginosae, Pauta and Pepea in relation to Mean Annual Temperature (MAT) (A) and Mean Annual Precipitation (MAP) (B). The ends of each box represent the upper and lower quartiles and the median is indicated with a bold line inside the box; the whisker lines extend to the highest and lowest observations, except when observations are higher or lower than the interquartile range (i. e. outliers), in which case they are indicated by a dot. Box plots that share the same lowercase letters within each subsection are not significantly different at p = 0.05. Vertical lines represent subsectional divisions.

Figure 13. Box plots showing the climatic niches of the species of subsection Polylepis Sericeae Sericeae in relation to MAT (A) and MAP (B). See Fig. 12 for details on data presentation.