The genus Massalongia (lichenised ascomycetae) in the Southern Hemisphere

Abstract The species of Massalongia recorded and described from the Southern Hemisphere are revised and it is shown that only one is present; M. patagonica which is widespread, with populations in Australia and New Zealand that differ from the South American populations, but at present best regarded as part of the variation of that species. Records from this hemisphere of all other species placed in the genus are incorrect. The type species, M. carnosa, is restricted to the Northern Hemisphere. Two species, M. antarctica and M. novozelandica cannot be identified precisely due to lack of sufficient type material and with the types as the only collections known of these, but none belongs in Massalongia according to available data. Massalongia griseolobata (from Gough Isl.) is shown here to belong in the Pannariaceae and is part of the parmelielloid clade. M. intricata (from South Georgia) and M. olechiana (from South Shetland) have both recently been correctly transferred to the genus Steinera in the Arctomiaceae.


Introduction
The genus Massalongia was described by Körber (1855), based on the species Lichen carnosus described by J. Dickson in 1790 on material collected in Scotland, but later often called Pannaria muscorum (Ach.) Duby, an illegitimate, superfluous name. This reflects the difficulties which the early lichenologists had in classifying the species. Mo-lecular studies (Wedin et al. 2007;Muggia et al. 2011) have shown that Massalongia does not belong in the Pannariaceae, but is best placed in a family of its own, Massalongiaceae, in the Peltigerales. There is, however, only one major study of the species and their variation, made by Henssen (1963) covering North America. She accepts two species; the widespread, variable M. carnosa and the nearly crustose, microphylline, local Californian endemic M. microphyllizans (Hasse ex Nyl.) Henssen. Jørgensen (2000), whilst revising the Pannariaceae, had studied the type of Massalongia fauriei (Hue) Zahlbr. and found the poor type (the only material existing) to belong in Fuscopannaria leucophaea s.lat., now transferred to the genus Vahliella (Jørgensen 2008;Wedin et al. 2010). An additional Asian species has been recorded from the Philippines based on Rehm (1916). This is, however, based on a misunderstanding of Massalongiella imperatae Rehm., which is a non-lichenised ascomycete as originally described. In Europe, Harmand recognised a variety which Gyelnik (1940), in his notorious treatment of the Pannariaceae in Rabenhorst's Kryptogamenflora, raised to species rank as Massalongia meizospora, only representing a form with somewhat longer, 3-septate spores. In addition, he established a new Massalongia rabenhorstiana, the type of which has disappeared. It is most certainly only a synonym of M. carnosa, still the only species on the Northern Hemisphere in addition to M. microphyllizans, a species in need of a phylogenetic study.
The situation in the Southern Hemisphere is different, though it took a long time before any species in the genus was recognised. Zahlbruckner (1917) was the first when he recorded Massalongia carnosa from the Falkland Islands, followed by Lamb (1958) who recorded it from Patagonia. Later, it was mentioned from several regions in the Southern Hemisphere (Smith and Corner 1973;Lindsay 1974;Galloway 1985;Redon 1985;Jørgensen 1986;Jørgensen and Elix 1988;Øvstedal and Smith 2001). In addition, several new species were described from the Southern Hemisphere; M. antarctica Dodge (from the Antarctic Peninsula, Dodge 1971), M. novozelandica Dodge (from subantarctic New Zealand, Dodge 1971), M. griseolobata Øvstedal (from Gough Isl., Øvstedal and Gremmen 2010), M. intricata Øvstedal (from South Georgia, Øvstedal and Smith 2001) and M. olechiana Alstrup & Søchting (from South Shetland, Alstrup and Søchting 2011).
During fieldwork in Chile in 2015, one of the authors (A.E.) discovered a strange Pannaria-like lichen which, on closer inspection, proved to be a Massalongia with some differences from M. carnosa, as known by us from Norway. However, since this is a variable species, we felt that a more detailed study, including molecular screening, would be useful. This being done and the distinction of this material proven, we found it necessary to check on the surprisingly high number of species of Massalongia described from the Southern Hemisphere. This proved to be time-consuming and complicated, since it was difficult to get hold of suitable material and, when molecularly checked, often not giving clear results and involving quite unrelated lichen families. Fortunately, Ertz et al. (2017) solved some of our problems and, eventually, this new species was named by Kitaura and Lorenz in Liu et al. (2018). However, our project contains more data than their work includes and we present these here in an attempt to give full clarification of the taxonomic situation for the genus in the Southern Hemisphere. Some additional phylogenetic data were also added on the genus in the Northern Hemisphere.

The specimens
Specimens of Massalongia were obtained from various herbaria for phylogenetic analyses, see Tables 1 and 2. In addition, we microscopically studied material from the following herbaria: BAA, BG, BM, C, CANB, CANL, CHR, F, FH, H, MSC, NY, TROM and UPS. A total of 130 ascospores from collections from both hemispheres were drawn in detail and measured for comparison.
PCR reactions consisted of 1× GeneAmp PCR Buffer II (Applied Biosystems), 2.5 µM MgCl 2 (Applied Biosystems), 20 µM dNTPs (Promega), 0.4 µM of each primer, 0.03 U AmpliTaq DNA Polymerase (Applied Biosystems), 2-5.0 µl of genomic DNA extract and distilled water to a total volume of 25 µl. PCR reactions were performed on a C1000 Touch thermal cycler (Bio-Rad Laboratories), with the following temperatures; initial denaturation at 94 °C for 4 min, followed by a 62-56 °C touchdown annealing for the first 6 cycles, ending with 30 cycles at 56 °C for 30 sec, polymerisation at 72 °C for 1 min 45 sec and a final elongation at 72 °C for 10 min.
Direct sequencing of PCR products was run with the PCR primers using a Big-Dye Terminator Cycle Sequencing kit (Applied Biosystems) on an ABI Prism 3700XL DNA analyser (Applied Biosystems) at the DNA Sequencing Facility (UiB), Norway. Sequences were assembled and edited using Geneious v.11.0.2 (Kearse et al. 2012).
Newly generated sequences with GenBank accession numbers are listed in Tables 1, 2, together with sequences downloaded from GenBank.

Phylogenetic analyses
To align the sequences, MAFFT v7.309 (Katoh et al. 2002;Katoh and Standley 2013) implemented in Geneious v.11.0.2 (Kearse et al. 2012) was used with default settings, followed by manual adjustments. Suitable substitution models for the separate datasets were identified using MrAIC v.1.4.6 (Nylander 2004). Two different datasets were analysed; one broad analysis of Massalongiaceae and Pannariaceae to test whether the included species is part of Massalongia (Table 1) and a second analysis for species delimitation within Massalongia (Table 2). For the broader dataset, mtSSU and RPB1 were concatenated, using Lecidea fuscoatra as outgroup and for the species delimitation in Massalongia, mtSSU, LSU, ITS and RPB1 was concatenated using Polychidium muscicola as outgroup.
Separate analyses of all genes and concatenated datasets were run as Bayesian MCMC searches using MrBayes v.3.2.1 (Ronquist and Huelsenbeck 2003) with default options; substitution model GTR+G+I, 10 million generations starting with a random tree, four simultaneous chains and using the default temperature of 0.2. Every 1000 th tree was saved. Phylogenetic trees were visualised using Geneious v. 11.0.2 (Kearse et al. 2012).

Phylogeny
The two resulting concatenated datasets consisted for the broad analysis of Massalongiaceae and Pannariaceae of 63 taxa with 1435 characters, whereas for the species delimitation in Massalongia, of 21 taxa with 2983 characters (details in Table 3).
The resulting phylogenetic consensus tree from the broad analysis of Massalongiaceae and Pannariaceae are given in Fig. 1. Both Massalongia carnosa and M. patagonica are with high support a part of the Massalongiaceae, together with Polychidium muscicola and Leptochidium albociliatum. M. griseolobata is a part of the Pannariaceae, in the "Parmelielloid" clade 1 from Ekman et al. (2014) with high support. Within this clade, M. griseolobata is a part of a supported group with no internal resolution, including Degeliella rosulata, Degeliella versicolor, Leioderma, Erioderma and Joergensenia.
The resulting phylogenetic consensus tree from the species delimitation analysis of Massalongia is given in Fig. 2. M. patagonica from the Southern Hemisphere and M. carnosa from the Northern Hemisphere are nicely separated in two sister groups with high support.
The samples of M. patagonica from New Zealand are grouped in a separate subclade from the rest of the samples from Australia, Chile and Argentina. The phylogenetic tree indicates a high genetic variance within M. patagonica throughout the Southern Hemisphere, but further studies are necessary to evaluate these differences.
The samples from the Northern Hemisphere make a monophyletic clade with little variation between the samples and a sample from Sweden is practically identical to those analysed from Alaska and Greenland.
Chemistry. All reactions negative, no lichen substances detected by TLC.
Habitat and distribution. This is a species of wet to dry rock surfaces or boulders, usually growing in between mosses or on plant remains. It has a widely scattered distribution in South America, ranging from the temperate forests of south-central Chile, including the Juan Fernandez Islands and Patagonia, with two widely separated

Discussion
The result of the phylogenetic analyses of Massalongia (Fig. 2) show that Massalongia carnosa and M. patagonica are located in different supported clades, as separate species as also described by Kitaura and Lorenz in Liu et al. (2018), M. patagonica being restricted to the Southern Hemisphere, whereas M. carnosa occurs only in the Northern Hemisphere. The clade with M. carnosa includes one circumarctic and circumboreal species, with low genetic diversity, whereas M. patagonica is more variable and shows a geographic pattern within this species. The material from New Zealand groups in a distinct branch within the M. patagonica clade and is superficially much more similar to the material of M. carnosa, but has extra short ascospores measured in two samples from New Zealand, all spores were shorter than 23 µm. This could be a result of the preference for moist, mossy habitats (Galloway 2007) as opposed to the drier, often exposed habitats in Chile. The material from Australia and New Zealand is, therefore, best classified as part of the M. patagonica complex.
That species is also found as far west as the Juan Fernandez Islands and is also possibly present on the Antarctic Peninsula and the Bouvet Island, but the material examined was sparse, sterile and too old for molecular studies.
Still, M. patagonica is not morphologically easily distinguished from M. carnosa; the two species have different spores, although there is an overlap zone in both length and degree of septation. Both species have a gross morphology showing high variation, probably due to habitat modifications, depending on light exposure, competition, moisture and water availability.
Chilean material of M. patagonica tends to have thicker, narrower and clearly radiating lobes than most material of M. carnosa. However, in cases where habitat information is available, they appear to be dry, but exposed to nutrient supplies by spring flooding (the Río Colorado collection), wind-transported saline lake dust (the Morro Chico collection) or seashore spray (the Tierra del Fuego collection). The New Zealand material, on the other hand, treated as M. carnosa, is cited as widespread and from moist habitats by Galloway (2007).
This detailed phylogenetic signal within M. patagonica is the result of a long history of evolution and isolation in austral areas, although shorter than the split-up between M. carnosa and M. patagonica. There is a record of M. carnosa from Mt. Kinabalu on Borneo (Sipman 1993) which could have indicated a migration route between a northern and a southern distribution area of the genus; however, a check of the material deposited at herbarium B revealed that it instead represents a sterile, richly squamulose specimen of a Parmeliella species. Future studies should investigate phylogeographic relationships between the three accepted species and the molecular distances between M. patagonica in New Zealand, Australia and South America/West Antarctica.
The examination of all relevant material from the Southern Hemisphere, shows the following, treated alphabetically according to the epithet: Massalongia antarctica Dodge is a species only known from the type specimen from Lambda Island at the tip of the Antarctic Peninsula (Siple 380c-2, FH!). The type specimen is minute and sterile and consists of two different species, none of which belongs in Massalongia. The one fitting best with the description has a crustose, hemigelatinous thallus in accordance with species of the Arctomiaceae. There are no apothecia present in the collection and the description of the apothecia, given by Dodge (1968), is at variance with characters of Massalongia, indicating a species of the Arctomiaceae, most probably in Arctomia. There is, however, no known species with such a distinctly crustose thallus. More material is needed to identify this taxon more exactly. The sample also contains squamules with a trebouxioid photobiont and this is possibly Pertusaria corallifera Vain. as pointed out by Castello and Nimis (1995).
Massalongia griseolobata Øvstedal is a species only known from the type specimen (from Gough Isl., coll. Gremmen 2006-91, BG!). Even if only incipient apothecia were found, we do not hesitate to place this species in the Pannariaceae, based on morphology and the original description of the asci. They are recorded to have apically blue in tholus in iodine with a weak ring-structure. (Massalongia has sheet-like structures, Jørgensen 2007). The molecular study confirms this (Fig. 1). The species groups in the parmelielloid clade (Clade 1) in the tree by Ekman et al. (2014), with Degeliella, Leiderma, Erioderma and Joergensiana. This is an unresolved group of subantarctic taxa (Jørgensen and Andersen 2015) in need of further studies.
Massalongia intricata Øvstedal was correctly transferred to the genus Steinera by Ertz et al. (2017). S. intricata has a semi-gelatinous thallus producing apothecia on special lobules, just as species in the Arctomiaceae.
Massalongia novozelandica Dodge was recorded by Galloway (2007), but the holotype (the only material) has not been possible to obtain. However, the original descrip-tion of the spores being brownish at maturity with disappearing septae (clearly pseudoseptae) is at variance with characters found in Massalongia. We agree with Galloway that this is probably a parasite growing on the thallus of a species in the Pannariceae.
Massalongia patagonica Kitaura and Lorenz, the recently described species (Fig. 3), belongs in the genus and, according to our phylogenetic tree (Figs 1, 2), prove to be distinct from M. carnosa, the latter being restricted to the Northern Hemisphere. The two species are morphologically variable due to the ecological conditions, but have different spores (Fig. 4), usually shorter than 25 µm in M. patagonica, but variable both in length and number of septae in both species. Chilean material of M. patagonica tends to have thicker, narrower and clearly radiating lobes than most material of M. carnosa.

Conclusion
From these facts, we conclude that there is only one, widespread species of Massalongia in the Southern Hemisphere, M. patagonica, though the populations in Australia and New Zealand differ somewhat molecularly, but more data is necessary to decide their taxonomic status. M. patagonica has a wider distribution than indicated in the original paper, also southwards and westwards. Previous records of several species in the Southern Hemisphere proved incorrect, most of them belonging in other genera.
The type species M. carnosa is restricted to the Northern Hemisphere, where it is widespread and variable, but without distinct molecular groupings requiring taxonomic recognition. There is also a local endemic, M. microphyllizans in California.