Introduction

The geographical distribution of most species of Grimmia Hedw. (Grimmiaceae, Bryophyta) in the New World is well known due to field work and taxonomic revisions at continental and global scales (^{Greven, 1999}, ²⁰⁰³; ^{Muñoz, 1999}; ^{Muñoz and Pando, 2000}; ^{Delgadillo, 2015}). However, despite this upgraded information, there are still questions on the origin of the discontinuous range of various species. Most of the 26 species recognized for the Neotropics are unknown from the area between central Mexico and northern South America (e.g., Grimmia donniana Sm. and G. fuscolutea Hook.). This geographical discontinuity is even larger for those species that occur in northern Mexico and northern/southern South America (G. anodon Bruch & Schimp., G. laevigata (Brid.) Brid., and G. pulvinata (Hedw.) Sm.). Their apparent absence in southern Mexico and Central America may be due to a deficient collection record or to the lack of suitable habitats. At least six species are known from Central America (^{Muñoz and Allen, 2002}; ^{Delgadillo, 2015}).

Models of potential distribution for Grimmia indicated that some species were expected to occur in southern Mexico (^{Delgadillo et al., 2012}), but only G. longirostris Hook., G. pilifera P. Beauv., and G. trichophylla Grev. have been recorded for this area. The study of more than 900 specimens from Mexico (^{Delgadillo, 2015}) showed that the Neovolcanic Belt is the southernmost limit for most Mexican species of Grimmia, suggesting that the genus and its alpine species reached the Neovolcanic Belt area by migration from North America (^{Delgadillo, 1987}). Hypothetically, the discovery of other species in the highlands of Oaxaca, and in central or southern Chiapas would suggest an alternate southern origin of Grimmia in Mexico. This contribution reports on recent field exploration in Oaxaca and Chiapas, updates the geographical distribution of the genus and the species in southern Mexico, and evaluates the hypothesis of its geographical derivation.

Materials and Methods

Field work was conducted in the highlands of Oaxaca and Chiapas in 2013, 2015 and 2018. Emphasis was placed on the search of Grimmia to validate previous models (^{Delgadillo et al., 2012}) and offer support to an alternate hypothesis regarding a southern origin of Mexican species. Records of the specimens collected were added to a database originally prepared for a revision of the genus in the Neotropics (Delgadillo, 2015). Localities visited for Grimmia are given in Table 1 along with a summary of identified specimens; the latter were deposited in the Bryophyte Collection at MEXU.

Identification of specimens was made using the keys and descriptions of publications cited in the introduction.

Results

According to our Grimmia database, the species are frequent in alpine grasslands, coniferous or Quercus forests. It is a group of high elevation elements representing a broad spectrum of environmental conditions present in xerophytic scrubby vegetation, dry pinyon pine forests, pine-oak-juniper woodlands, tropical deciduous forests, and grasslands. In the northern states Grimmia is found at elevations of 500-600 m a.s.l., while in central Mexico some species may reach 4600 m. Samples of Grimmia from Oaxaca and Chiapas were obtained at 2950-3270 m a.s.l., but the altitudinal range for the genus there may be as low as 1944 m a.s.l. Exploration in Pinus forests yielded G. longirostris, G. pilifera, and G. trichophylla in the state of Oaxaca. The last one is the only species of the genus known from Chiapas. The specimens listed in Table 1 and those cited by Delgadillo and Cárdenas (¹⁹⁸⁹) from the same area represent the first records of the genus for Chiapas. All of them were collected in a disturbed Pinus forest on top of the Cerro Mozotal.

While the pine forest of the summit of Cerro Mozotal seems the habitat where Grimmia would be expected to occur, the accompanying moss flora is usually part of other plant communities. Braunia imberbis (Sm.) N.J. Dalton & D.G. Long, Polytrichastrum tenellum (Müll. Hal.) G.L. Sm., and Racomitrium subsecundum (Hook. & Grev. ex Harv.) Mitt. & Wilson are alpine/subalpine species; Caribaeohypnum polypterum (Mitt.) Ando & Higuchi, Dendropogonella rufescens (Schimp.) E. Britton, and Prionodon spp. are examples from cloud forests (Table 2).

Discussion

The number of species of Grimmia detected in southern Mexico is surprisingly low, particularly in the states of Oaxaca and Chiapas that exhibit topographic elevations and types of vegetation that seem regular habitats for several species elsewhere. According to their known altitudinal range in Mexico and their potential distribution (^{Delgadillo et al., 2012}), at least eight species would be expected in the area, but field work revealed only three of them (Table 1). Grimmia elongata Kaulf., G. mexicana Greven, and G. ovalis (Hedw.) Lindb. have been reported from northwestern Guatemala; G. longirostris and G. trichophylla have a broader range in Central America. There are no records of G. fuscolutea, G. pulla Cardot, and G. torquata Drumm. All three species form spores or gemmae that would enable them to spread into available habitats, but their consistent absence in southern Mexico suggests that other environmental or historical variables control their establishment.

The peculiar habitat of G. trichophylla in Chiapas is of interest; high altitude pine forests are the usual vegetation for this taxon. However, on the Cerro Mozotal it dwells with other alpine or subalpine mosses and species from deciduous or tropical derivation. Graham (¹⁹⁷³) has shown that members of the arborescent temperate element migrated into Mexico since the Eocene and moved progressively southwards; Abies Mill., Picea A. Dietr., Alnus Mill., Fagus L., Juglans L., Liquidambar L., and others reached southern Mexico by the mid-Miocene. This fact may explain the presence of temperate mosses in Chiapas, i.e., species that moved along with the vascular plant flora; the tropical taxa perhaps derive from montane forests from lower elevations. The presence of the alpine and subalpine mosses (Table 2), including some species of Grimmia, may be related to high elevation habitats in Chiapas that may have become available by the late Tertiary (cf. ^{López Ramos, 1983}) and were filled by new arrivals.

Field and herbarium work resulted in more than 900 Mexican specimens for study (^{Delgadillo, 2015}). These specimens and bibliographic information show that most species of Grimmia are unknown from southern Mexico and Central America. The lower altitude of the Central American mountains, relative to those in the Neovolcanic Belt and the northern Andes, would not facilitate the spread of Grimmia from the north or from the south. However, since the distribution of most species of that genus does not extend southwards beyond the Neovolcanic Belt one must conclude that the original hypothesis of a northern origin in Mexico is the most suitable explanation. The broad continental ranges of G. longirostris and G. trichophylla do not give definite support to a northern or southern origin, but the northern option is preferred because the Isthmus of Panama is considered a barrier, not a corridor, for high montane elements (^{Simpson and Neff, 1985}). According to these authors, even a 1200 m lowering of the vegetational belts would not provide continuous cool high forest areas between North and South America. Present distribution of these two species is perhaps associated with lower mountain habitats unlike other species in the Mexican highlands. Undoubtedly, additional geological, geographic, and floristic information is required to understand the history of the area; a sample of our incomplete knowledge is illustrated by the new records for the Chiapas moss flora as given in Table 2.