Sep 082016

One of the main problems with sampling largely from herbarium specimens, rather than from material that has been specifically collected for DNA work (rapidly dried in silica gel then maintained at low humidity), is that the quality of the DNA is unpredictable and usually rather poor. Therefore, despite starting out with 169 accessions and about 20 species of Marchantia, the actual successes, where we were able to get good quality DNA sequence data, were substantially lower. What we currently have is a slightly unbalanced data matrix, with 82 Marchantia accessions for rbcL, and 78 Marchantia accessions for psbA-trnH.

Reboulia hemisphaerica thallus, photographed by David Long (Long 34254)

Reboulia hemisphaerica thallus, photographed by David Long (Long 34254)

We also sequenced both rbcL and psbA-trnH from material of two accessions that we thought were Marchantia but where the sequences turned out to be Reboulia (from Texas) and Wiesnerella (from Bhutan). A quick check of the herbarium voucher specimens for both of these showed that they represented mixed collections of more than one complex thalloid species, for which the “wrong” plant parts had ended up in our silica dried tissue collection. Taking fortune from misfortune, both Reboulia and Wiesnerella form quite adequate outgroups for the phylogeny!

Wiesnerella denuda, photographed by David Long

Wiesnerella denuda thallus, photographed by David Long (Long 36267)

Out of the 20 species that we HAD hoped to sample, we ended up with only 12 named Marchantia species for rbcL (Marchantia polymorpha, M. paleacea, M. linearis, M. papillata, M. inflexa, M. emarginata, M. pinna, M. chenopoda, M. debilis, M. hartlessiana, M. quadrata and M. romanica), and 15 for psbA-trnH (Marchantia polymorpha, M. paleacea, M. linearis, M. papillata, M. inflexa, M. emarginata, M. pinna, M. chenopoda, M. debilis, M. globosa, M. pappeana, M. hartlessiana, M. subintegra, M. quadrata and M. romanica); we also had three Marchantia polymorpha subspecies (polymorpha, ruderalis and montivagans) and two Marchantia paleacea subspecies (paleacea and diptera).

That’s a little disappointing, representing, as it does, fewer than half of the 38 currently recognised species in the genus. However, we did also sequence a number of Marchantia accessions that had not been determined to species, and although many of them were good DNA matches to species that we had sampled, several are clearly different to everything else that we have included: one distinct lineage in Yunnan, China, another that occurs in Yunnan and Nepal, and a third in Indonesia and Malaysia. That’s balanced again by taxa that may not have been identified correctly; the psbA-trnH sequences from African material of M. debilis, M. globosa, M. pappeana and M. polymorpha, for example, are identical.

Intriguingly, in the “Preissia” clade, as well as M. romanica, there appear to be two lineages of Marchantia quadrata, one consisting of accessions from Denmark, Sweden and Sichuan, China, and the other with accessions from Svalbard, Norway and Utah, USA. These may tie in with subspecies quadrata (for the first lineage) and subspecies hyperborea (for the material from Svalbard and Utah), but the degree of genetic divergence is far higher than that found between many of the recognised species in Marchantia. It is a bit disconcerting, however, to notice that we have managed to overlook any Marchantia quadrata material from Scotland in our sampling!

The next step in the project, before it’s time to reveal any of the phylogenetic trees I’ve alluded to, is a phase of reciprocal illumination where we reconcile morphological information from the herbarium specimens with the information derived from the molecular sequence data. In other words, it’s time to double check our plant identifications, a part of the project that’s now in the capable hands of Dr David Long; the pile of Marchantia specimens is already on his desk!



Relevant posts

A rapid phylogeny of Marchantia, from the RBGE collections. I. Sampling

A rapid phylogeny of Marchantia, from the RBGE collections. II. Illuminating our sampling

Jun 302016

The genus Aitchisoniella contains a single species, A. himalayensis, which was described by Pakistani botanist Professor Shiv Ram Kashyap from plants that he collected in Mussoorie, Uttarakhand, India (1914). Subsequently (1929) he also found the species in Shimla and Kullu in Himachal Pradesh. Although reports of new locations followed (Kanwal, 1977, Pant et al., 1992, Bischler et al., 1994), near Nainital in Uttarakhand, the species was only known from the north-west Himalayas of India until 2010, when its range was extended by RBGE bryologist Dr David Long, on the ‘Kunming/Edinburgh Expedition to Sichuan’. David found Aitchisoniella at two localities in China, in Litang and Daocheng counties of south-west Sichuan Province.

Athalamia pinguis, Sichuan, photographed by David Long (Long # 40305)

Fig. 1. Athalamia pinguis in Sichuan, showing stalked sexual branches (carpocephala), photographed by David Long (Long # 40305)

Aitchisoniella looks quite different to other complex thalloid liverworts: it doesn’t have the stalked sexual branches (carpocephala) that are present in most complex thalloid species like Marchantia, or Athalamia (as can be seen in the photograph, Fig. 1). Instead, the female sex organs (archegonia), and therefore the sporophytes, grow on the lower (ventral) side of a short receptacle (as can be seen in the photograph, Fig. 2). The receptacle is part of the main thallus, with air chambers and air pores, and a groove on the underside of the thallus from which characteristicly complex-thalloid pegged rhizoids grow.

 Aitchisoniella himalayensis in Sichuan showing terminal sporophyte-bearing receptacles, from Long 39886

Fig. 2. Aitchisoniella himalayensis in Sichuan, showing terminal sporophyte-bearing receptacles, photographed by David Long (Long # 39886)

Originally, Aitchisoniella was thought to belong to the Exormothecaceae family of complex thalloid liverworts, along with Exormotheca and Stephensoniella. Earlier this year, we transferred all the plants in Exormothecaeae into another family, Corsiniaceae, where they joined Corsinia and Cronisia (Long et al. 2016a).

Complex thalloid phylogeny reconstructed by Villarreal, J.C., B.J. Crandall-Stotler, M.L. Hart, D.G. Long, L.L. Forrest. 2015. Divergence times and the evolution of morphological complexity in an early land plant lineage (Marchantiopsida) with a slow molecular rate. New Phytologist. DOI: 10.1111/nph.13716

Fig. 3. The phylogenetic position of Aitchisoniella, from the complex thalloid phylogeny reconstructed by Villarreal et al. (2015)

However, having fresh material of the plant meant that we were able to extract DNA from it, and add it into our molecular phylogeny for the group (Villarreal et al. 2015). The results were unexpected, with Aitchisoniella grouping with species from a different complex thalloid family, Cleveaceae (Fig. 3). The growth forms are quite different, as all the species in Cleveaceae have got stalked carpocephala. However, once we started to think more about the evolution of these plants, and look more objectively at differences and similarities, one thing struck us: the spores of Aitchisoniella (see Fig. 4) look more like the spores of plants in Cleveaceae (e.g. Athalamia, Fig. 5) than they do like spores of plants in Exormothecaceae (e.g. Fig. 6).

Spores of Aitchisoniella himalayensis. (A) Distal view; (B) proximal view; (C) lateral view; (D) detail of distal view. (A and C) from Long 39886. (B and D) from Long 40020. Scale bars: (A–C) = 5 μm, (D) = 2 μm.

Fig. 4. Spores of Aitchisoniella himalayensis from Long et al. 2016: (A) Distal view; (B) proximal view; (C) lateral view; (D) detail of distal view. (A & C) from Long 39886. (B & D) from Long 40020. Scale bars: (A–C) = 5 μm, (D) = 2 μm.

Athalamia hyalina spore images from M. P. Steinkamp and W. T. Doyle American Journal of Botany Vol. 68, No. 3 (Mar., 1981), pp. 395-401

Fig. 5. Spores of Athalamia hyalina (Cleveaceae) from Steinkamp & Doyle (1981): 1. distal view (x 1,100); 2. proximal view (x 1,100); 3. equatorial view (x 1,100); 4. close-up of pore (x 6,000); 5. distal face (x 3,000).


Exormotheca spores: E. bulbigena - A, distal view, B, proximal view. E. holstii - C, distal view, D, proximal view, E, distal view, F, proximal view. From Bornefeld et al., 1996.

Fig. 6. Spores of Exormotheca from Bornefeld et al. (1996): E. bulbigena – A, distal view, B, proximal view. E. holstii – C, distal view, D, proximal view, E, distal view, F, proximal view.

Characters like spore shape used to be considered to be quite neutral in bryophyte evolution, features that were not really acted on by natural selection. Following this theory, spore characters were thought to be indicative of true and ancient relationships, changing very little across huge amounts of time. We have since moved away from this view, with, for example, small changes in the shape and size of spores known to have drastic effects on their aerodynamics. However, within the complex thalloids, it does seem that characters like the presence or absence of carpocephalum branches are quite variable within families, while spore morphology can be indicative of deeper relationships.

As a result of this work, based on both molecular and morphological evidence, we have transferred the genus Aitchisoniella to the family Cleveaceae (Long et al. 2016b), where it joins the four genera accepted by Rubasinghe et al. (2011): Athalamia, Clevea, Peltolepis and Sauteria.



Bischler, H., Boisselier-Dubayle, M.C. & Pant, G. 1994. On Aitchisoniella Kash. (Marchantiales). Cryptogamie. Bryologie-Lichénologie, 15: 103–10.

Borenfeld, T., O.H. Volk & R. Wolf. 1996. Exormotheca bulbigena sp. nov. (Hepaticae, Marchantiales) and its relation to E. holstii in southern Africa. Bothalia 26,2: 159–165.

Kanwal, H.S. 1977. Marchantiales of district Naini Tal (Kumaun Hills) U.P., India. Revue Bryologique et Lichénologique, 43: 327–38.

Kashyap, S.R. 1914. Morphological and biological notes on new and little known West-Himalayan Liverworts. I. New Phytologist, 13: 206–26. doi: 10.1111/j.1469-8137.1914.tb05751.x.

Kashyap, S.R. 1929. Liverworts of the Western Himalayas and the Panjab Plain. Part 1. Lahore: The University of the Panjab.

Long, D.G., L.L. Forrest, J.C. Villarreal & B.J. Crandall-Stotler. 2016a. Taxonomic changes in Marchantiaceae, Corsiniaceae and Cleveaceae (Marchantiidae, Marchantiophyta). Phytotaxa, 252: 77–80.

Long, D.G., L.L. Forrest, J.C. Villarreal & B.J. Crandall-Stotler. 2016b. The genus Aitchisoniella Kashyap (Marchantiopsida, Cleveaceae) new to China, and its taxonomic placement. Journal of Bryology.

Pant, G., S.D. Tewari & S. Joshi. 1992. An assessment of vanishing rare bryophytes in Kumaun Himalaya – thalloid liverworts. Bryological Times, 68/69: 8–10.

Rubasinghe, S.C.K., D.G. Long, R. Milne & L.L. Forrest. 2011. Realignment of the genera of Cleveaceae (Marchantiopsida, Marchantiidae). The Bryologist 114: 116-127.

Steinkamp, M.P. & W.J. Doyle. 1981. Spore wall ultrastucture in the liverwort Athalamia hyalina. American Journal of Botany 68: 395-401.


Villarreal, J.C., B.J. Crandall-Stotler, M.L. Hart, D.G. Long & L.L. Forrest. 2015. Divergence times and the evolution of morphological complexity in an early land plant lineage (Marchantiopsida) with a slow molecular rate. New Phytologist. 209: 1734–46, doi: 10.1111/nph.13716.


Long, D.G., L.L. Forrest, J.C. Villarreal, B.J. Crandall-Stotler. 2016. The genus Aitchisoniella Kashyap (Marchantiopsida, Cleveaceae) new to China, and its taxonomic placement. Journal of Bryology.

Apr 262016
Setting up PCRs in our laminar flow hood

Setting up Aneura PCRs in our laminar flow hood

Sitting in Edinburgh airport on a Monday morning, waiting for David Long to join me, checked in through to Trondheim via Copenhagen, I felt completely unprepared. The previous week had been a fluster of lab work and reading DNA sequences, trying to get everything ready in time – a stressful Friday evening, trying to copy all the Aneura files into my Dropbox and onto flash drives before the building shut down at 6pm, willing all the file transfers to go faster and faster… but in the end having to leave many of the images that I had planned to take with me behind in the office. Despite a relatively early start on the Monday, we had a 7 hour layover in Copenhagen,

The Lego shop, Copenhagen

The Lego shop, Copenhagen

time for a train ride into the city, lunch, and a meander through the downtown streets, so didn’t get to Trondheim until late. From the airport bus we could make out snow and birch trees, before getting off on a near-deserted icy street. A short walk to the Comfort Hotel Park, an easy check-in, and sleep.

Swedish bryologist Lars Söderström picked us up in the foyer at about 9am. The university is only 20 minutes or so walk away, but the icy pavements made that impractical, so we were taking the bus. Lars had our bus tickets on his phone, cheaper and easier than using cash on the bus. Once purchased, they’re good for an hour and a half, with a spinning bar that gets shorter over the time period until it eventually disappears and the ticket has gone. It was a short ride, across the river and uphill, through mostly painted wooden buildings. Ana Séneca, our Portugese team-Aneura colleague, met us on the bus.

The university building is modern and airy, with open atriums the height of the building, planted with dead bamboo. Ana and I made our way to the Herbarium, a windowless room filled with cupboards of bryophytes that had mostly been collected by herself and by Lars. This was the day that the two of us had put aside for compiling and analysing our Aneura data. I’d begun sending sequences over to Ana on the Friday, so the datasets were already joined together. We had sequences from just over 300 accessions of Aneura, mostly from the British Isles and Norway, but with representatives from Albania, Sweden, Iceland, Portugal, Belgium, Austria, Latvia, the Faroe Islands, China, Fiji, India, the Falklands, Reunion, South Africa, the US, Canada, Panama, Peru…

Building phylogenetic trees in the University Herbarium, Trondheim

Building phylogenetic trees for Aneura in the University Herbarium, Trondheim

We used PAUP to run some quick parsimony analyses, printing out multi-page phylogenetic trees for each of four gene regions that we had been sequencing.

Papering over the table with our Aneura data

Papering over the table with our Aneura data

Clades that were in common between all four trees were marked on using some provisional, and informal, clade names, and after a search for coloured crayons, Ana undertook the serious business of marking geography onto the trees. Although she tracked down a pack of 12 coloured crayons, that wasn’t quite enough to separate the regions we were interested in, so we ended up with a key that combined colour and symbols.

Back to basics - colouring in the trees

Back to basics – colouring in the trees

A little after 5pm, it was time to call it a day, roll up the trees we’d made, and head out into the cold and dark to catch the bus back into town; the four of us headed to the Microbrewery in town for beer and burgers, then a nightcap of whisky at the hotel before Lars and Ana caught the bus out to their home.

Lars, David and Ana pick their way across the least icy route to the Museum

Lars, David and Ana pick their way across the least icy route to the Museum

The next morning, Lars and Ana met us at the hotel again, but this time instead of a bus, we were walking to the Museum, only five minutes or so from where we were staying. The paths were icy, but the views across the river were beautiful in the sunlight. We signed in at the Museum, where our Norwegian friend and colleague, Kristian Hassel, was waiting. First we headed up to the Herbarium, with views out across the city, before going downstairs for coffee, and settling on a sofa in the library to roll out our trees and start the conversation – what are we going to do with this data?

View from the Herbarium, NTNU Museum

View from the Herbarium, NTNU Museum

Luckily, we all agreed on the next actions – we are going to give names to a set of new species, based on molecular characters. We won’t name things that have only been collected and sequenced once, but if there are 4 or more accessions that form a lineage, then they will get named. Because of the focus of our sampling, we will restrict the taxonomy to taxa that occur in Europe. We also have to deal with the species of Aneura that have already been described. Because we are planning to use DNA for taxonomy, then we need to also have sequence data for all the existing names in the genus, even those that were described before anyone know what DNA was. This can be done retrospectively, using epitypification.

Ana and Lars compare names in the Museum library

Ana and Lars compare names in the Museum library

When a species name is published, it is linked to something known as a ‘type’. Usually this is a physical specimen, botanically, a dried out plant sample, although historically, illustrations were also used. The specimens are particularly important, often placed in special red folders, and treated with great respect. Methods like DNA extraction, which involve physically destroying parts of the material, are frowned upon. Given that some of the material can be over a hundred years old, DNA methods can also have very low success.

Instead of trying to get hold of old plant types and grind them up, we intend to use an alternative, which is the designation of new good-quality plant material as ‘epitypes’ – explanatory types that have more characteristics than the original material had, and so allow a better understanding of the correct application of the plant name. The material that we will designate as epitypes will be from large collections, with associated DNA material, and will have been sequenced for the set of four DNA markers in our project.

Trondheim, by the river




Thursday saw us back in the University, continuing discussions about data handling, dealing with mundane tasks like tracking down specimen information and compiling tables of data. More excitingly, bringing together collection details for plants in different evolutionary groups in our trees started to reveal some biology behind our proposed new species, with different ones occurring in different habitats. Although our departure on the Friday morning can only be described as totally uncivilised, with a 6.30 am flight from Trondheim to Oslo, a short stopover then an arrival in Edinburgh at approximately breakfast time, at least we had the satisfaction that the story of Aneura is finally beginning to come together – and an agreement that the next time we meet, it will be somewhere a bit less frozen, like Portugal…

A land of snow and ice - Norway from the plane

A land of snow and ice – Norway from the plane



Nov 022015

Although the exact relationships between the earliest land plant lineages are not yet well resolved, there is consensus that liverworts are one of the most ancient land plant groups. We are also confident that liverworts are a “good” taxonomic group, united by morphological and structural similarities, as well as by analyses of molecular sequence data that show that the entire group shares a common origin.

Fig 1a, Villarreal et al. 2015: molecular branch lengths across the liverwort tree of life

Fig 1a, Villarreal et al. 2015: molecular branch lengths across the liverwort tree of life, based on sampling in Laenen et al. 2014.

However, as a group, liverworts encompass a lot of variation – for example, some have leaves, some don’t; some have specialised sexual branches, others don’t; some grow in water, some on leaf surfaces, some on rocks; some are green, some red or brown, some closer to black… A less familiar metric that also varies across the liverworts is the rate of evolution of different parts of their genomes. Strikingly, complex thalloid liverworts have slow rates of molecular evolution for both the plastid and mitochondrial organelles. (In the figure on the left, they are the lineage represented in brown.)

While this phenomenom has been commented on by various authors over the years (e.g. Forrest et al. 2006), hard data has been lacking. Given the amount of sampling that we have for the complex thalloids, we have finally been able to put a number on it: We have calculated the rate of molecular evolution for the plastid to be 2.63 x 10-10 (SD 4.6 x 10-12) substitutions per site per year, based on coding sequences from atpB, psbA, psbT-psbH, rpoC1, rbcL and rps4, and the mitochondrial rate to be 5.31 x 10-11 (SD 9.4 x 10-13) substitutions per site per year, based on coding sequences from nad1, nad5 and rps3. The absolute mean substitution rate for the nuclear large ribosomal subunit (26S) is faster, at 7.76 x 10-10 (SD 1.4 x 10-11) substitutions per site per year.

In contrast, for the sister group (most other liverworts, shown in green in the figure), the calculated plastid rate, based on rbcL, is over three times higher than that for the complex thalloids, at 9.0 x 10-10 (Feldberg et al. 2014).

Bainard et al. 2013, Fig 1: Parsimony reconstruction of nuclear genome size evolution in liverworts (blue = small, red = large)

Bainard et al. 2013, Fig 1: Parsimony reconstruction of nuclear genome size in liverworts (blue = small, red = large)

However, seeing that there is a difference is the easy part; finding an explanation that fits the pattern is quite something else. Why would orgenellar genes evolve more slowly in the complex thalloids than in Haplomitrium, Treubia, the simple thalloid clades or the leafy liverworts? Could it tie in with a phenomenom known as RNA editing that is common to most liverwort lineages, but lacking in the complex thalloids? Do complex thalloids have longer generation times than the other liverworts, or slower metabolisms? On average, complex thalloids also seem to have smaller nuclear genomes than other liverwort lineages, although this is a broad generalization, with comparatively few genomes measured.

While we came up with some speculations in our last paper, we certainly don’t yet have an answer!

New Phytol title

Oct 302015
Archegoniophores and antheridiophores of Marchantia; taken by Julia Bechteler

Archegoniophores and antheridiophores of Marchantia; photograph by Julia Bechteler

One of the earliest plastid genomes to be sequenced, in the late 1980s (Ohyama et al.), was that of Marchantia polymorpha, one of the commonest liverworts around town, and an increasingly widely used model organism for genetic research. The complete mitochondrial genome followed, with a key publication by Oda et al. in 1992. The research on both organellar genomes came from Kyoto University, Japan.

When we were generating a DNA sequence matrix for the complex thalloids, we also included GenBank sequences from the published genome sequences for both these organelles (mitochondrial loci nad1, nad5 and rps3 and plastid loci atpB, cpITS, psbA, psbT-psbH, rpoC1, rbcL and rps4), to see how they compared with sequences from the three subspecies of Marchantia polymorpha that are found in the United Kingdom (subspecies ruderalis is the weedy plant that is commonly found in paving cracks, flowerbeds and plant pots).

Marchantia phylogeny based on Villarreal et al. 2015, Fig. 2

Marchantia phylogeny based on Villarreal et al. 2015, Fig. 2

The results of the analyses (Villarreal et al. 2015) were rather unexpected: the plant from GenBank didn’t cluster with our Marchantia polymorpha accessions, but instead with a Marchantia paleacea accession from Mexico. Admittedly, both species form a clade, but there’s a convincing amount of genetic distance between them.

Marchantia polymorpha ventral scales, photograph by Des Callaghan Licensed under CC BY-SA 4.0 via Wikimedia Commons -

Marchantia polymorpha ventral scales; photograph by Des Callaghan. Licensed under CC BY-SA 4.0 via Wikimedia Commons


Helene Bischler-Causse (1989, 1993) separated Marchantia into sections based on morphological characters of the plants, placing Marchantia polymorpha in section Marchantia, and Marchantia paleacea alone in section Paleaceae, due in part to differences in the way the scales on the underside of the thallus are arranged, as well as spore morphology. Now it seems that the organelle genome sequences that have, for the last 1/4 century, been thought to be from Marchantia polymorpha, are in fact from this lesser known (but also widespread) species, Marchantia paleacea.

(Kijak et al.’s research at Mickiewicz University in Poland shows a similar picture; although their study is still unpublished, a poster from the group is available to download here.)

New Phytol title

Oct 292015
Complex thalloid phylogeny from Forrest et al. 2006

Complex thalloid phylogeny from Forrest et al. 2006

Rather a while ago, back in 2003, we started working on a phylogeny of the complex thalloid liverworts at the Royal Botanic Garden Edinburgh (as a Molecular Phylogenetics Project). We were beautifully placed to conduct the work, with taxonomist Dr David Long on the staff, and all the necessary molecular facilities in place. Plants were found, identified, extracted and sequenced, and a matrix started to develop for a set of five loci: the nuclear large ribosomal subunit, the mitochondrial gene nad5, and plastid loci rbcL, rps4 and psbA. The data were analysed in conjunction with sequences generated by Christine Davies, then a PhD student at Duke University, and by myself, as a postdoc, at Southern Illinois University, were presented at the International Botanical Congress in Vienna in 2005, and were published as part of a phylogeny of liverworts, in The Bryologist in 2006.

That same year, we made plans to expand the study, to tie in with a Tree of Life project on liverworts that was coordinated through Prof. Jon Shaw‘s lab at Duke University. Our aim was to sequence more plants for more loci (adding in mitochondrial loci nad1 and rps3 and plastid loci atpB, cpITS, psbT-psbN-psbH and rpoC1), to contribute to the larger project, but also to produce a stand-alone paper on the evolution of the complex thalloids. And this is where the continental drift, or the glaciers, come in. They are both things that can move faster than a scientific paper can be completed…

Fig. 2 inset: a. Riccia cavernosa (Des Callaghan), b. Cyathodium (Zhang Li), c. REboulia hemisphaerica (Zhang Li), d. Plagiochasma (Zhang Li), e. Lunularia cruciata (Des Callaghan), f. Marchantia (Zhang Li)

Fig. 2 inset: a. Riccia cavernosa (Des Callaghan), b. Cyathodium (Zhang Li), c. Reboulia hemisphaerica (Zhang Li), d. Plagiochasma (Zhang Li), e. Lunularia cruciata (Des Callaghan), f. Marchantia (Zhang Li) – from Villarreal et al. 2015, New Phytologist, Fig. 2

We did have data: I even talked about it at conferences – in 2007, at the sixth biennial conference of the Systematics Association in Edinburgh (Forrest, L.L., D.G. Long, A. Clark, M.L. Hollingsworth, Complex thalloids can be simple – unravelling the evolutionary history of the Marchantiopsida), and again at the Botany 2010 meeting in Rhode Island (Forrest, L.L., D.G. Long, M.L. Hollingsworth, J.C. Villarreal A., A. Clark, J. Tosh, P. Hollingsworth, Into the Tropics: a brief history of complex thalloid liverworts). However, there were holdups. Partly, we were waiting until we had complete genus-level sampling for the group, and a small number of very small genera were proving problematic to obtain. Indeed, we only got hold of adequate DNA for Monocarpus and Austroriella (both from Australia) in 2009, and Cronisia (Brasil), Aitchisoniella (China) and Stephensoniella (India) in 2011.

David Long and Juan Carlos Villarreal hunt for liverworts at Balerno Moss (2007) - photograph by Laura Forrest

David Long and Juan Carlos Villarreal hunt for liverworts at Balerno Moss (2007) – photograph by Laura Forrest

However, the main stumbling block to getting a phylogeny for the complex thalloid liverworts published was staff time, and in 2014 we were fortunate to obtain 6 months of funding from the Sibbald Trust to employ someone to complete the data sampling, conduct the analyses, and spearhead the paper-writing. Panamanian bryologist Dr Juan Carlos Villarreal joined us in January this year, from Munich Botanical Garden, where he had been working with Prof. Susanne Renner. And just yesterday, Wednesday the 28th October 2015, after 12 years in progress, our stand-alone paper on the phylogeny of the complex thalloids was finally published (Early View), in the journal New Phytologist.

New Phytol title

Sep 102015

As far as liverworts go, Monocarpus is a rather strange plant. It’s very small, in itself not that unusual for a bryophyte, but rather problematic if you need to understand what it actually looks like, as a good handlens, and preferably a binocular microscope, are required. It’s also not easily found. It’s known from sites in Australia, and South Africa – but not from the sort of sites that liverworts are generally found in. The short-lived plants grow on seasonally wet salt-pans and gypsum soils. Unfortunately Australia has recently been suffering from a series of droughts, and ephemeral salt-pan plants have just not been much in evidence. Herbarium collections of the plant are quite rare and fragile, but what we do know of its morphology doesn’t really fit in well with any other known liverworts – it has a unique combination of characters. Added to that, various people who have worked on it in the past have come to different conclusions about where it fits in the evolutionary tree of plant life.

Monocarpus sphaerocarpus line drawing from Carr 1956, reproduced with permission from CSIRO PUBLISHING

Monocarpus sphaerocarpus line drawing from Carr 1956, reproduced with permission from CSIRO PUBLISHING

The first known collection of the plant was by Mrs Masie Carr, in 1955, by a roadside in north-west Victoria, Australia. From this her husband, Dr Denis J. Carr, described the new genus (and species), Monocarpus sphaerocarpus, in 1956. His publication notably includes an illustration that, while lacking in beauty, is nonetheless one of the most useful diagrams that has yet been produced for understanding the plant’s structure. Carr also managed to grow some spores, and, based on the morphology of the sporelings, considered his new genus to be related to Sphaerocarpos (hence the similar species name).

Complex thalloid liverwort Sphaerocarpos texanus in the UK, photographed by Dr David Long

Potential Monocarpus ally Sphaerocarpos texanus in the UK, photographed by Dr David Long

However, in 1961 Dr Johannes Proskauer published two papers on Monocarpus. In the first, in the journal Taxon, he explained why he considered that the name of the plant was not validly published, and instead it should be known as Carrpos, while in the second he examined in detail the morphology of several sporelings he was able to grow, from spores that he had been sent by Carr. From these, he concluded that the spores that Carr had observed were probably from ferns rather than from Monocarpus, and thus Carr’s conclusion about the phylogenetic relationship of Monocarpus was probably wrong.

Proskauer instead considered that Monocarpus belonged “along a neotenous series between Clevea and Riccia”. Essentially, this means that he thought Monocarpus to belong to quite an advanced group of liverworts, but to only have juvenile, rather than adult, characters.

Riccia sorocarpa, photographed by Dr David Long

Potential Monocarpus ally Riccia sorocarpa in the UK, photographed by Dr David Long

Several molecular phylogenies have been produced for liverworts, using changes in DNA sequences to build a picture of the relationships between lineages (e.g. Forrest et al. 2006; Laenen et al. 2014). While these have allowed us to answer many other questions about liverwort evolution, for example showing that what a plant looks like is not always the best indicator of what it is related to (as in the placement of Pleurozia, which has leaves, with the Metzgeriales, a lineage of liverworts that lack leaves), a lack of recent collections for Monocarpus mean that it has not been possible to include it in any of these analyses, as good quality DNA has not been available.

Thus we have yet to confirm whether it was Carr or Proskauer who had the correct placement for Monocarpus.

(Proskauer’s taxonomic name change was not taken up, and a paper by Arthur Bullock, again in Taxon, explains why Monocarpus remains the valid name for the plant.)


Monocarpus heading



Bullock AA (1961) Our Corpus and Carrpos – a reply. Taxon 10, 240-242

Carr DJ (1956) Contributions to Australian Bryology. I. The structure, development, and systematic affinities of Monocarpus sphaerocarpus gen. et sp. nov. (Marchantiales). Australian Journal of Botany 4, 175-191.

Forrest LL, Davis EC, Long DG, Crandall-Stotler BJ, Clark A, Hollingsworth ML (2006) Unraveling the evolutionary history of the liverworts (Marchantiophyta): Multiple taxa, genomes and analyses. Bryologist 109, 303–334.

Laenen B, Shaw B, Schneider H, Goffinet B, Paradis E, Désamoré A, Heinrich J, Villarreal JC, Gradstein R, McDaniel S, Long DG, Forrest LL, Hollingsworth ML, Crandall-Stotler BJ, Davis EC, Engel J, Von Konrat M, Patino J, Vanderpoorten A, Shaw AJ (2014) Extant diversity of bryophytes emerged from successive post-Mesozoic diversification bursts. Nature Communications 5, 5134.

Proskauer J (1961) On Carrpos I. Phytomorphology 11, 359-378.

Proskauer J (1961) Our Corpus and Carrpos. Taxon 10, 155-156.


On Monocarpus –

Finding Monocarpus, in the Herbarium –

Finding Monocarpus, in the field –

Lost before found: Was there more than one species in Monocarpus? –

Jul 292015
The Genus Inga

The Genus Inga, T.D. Pennington, 1996

About 300 species of Inga (Leguminosae: Mimosoideae: Ingeae) grow in lowland and montane rain forest throughout the humid tropical zone, from Mexico to Uruguay. Most species diversity is in the Andean foothills of Peru, Ecuador, Colombia and southern Central America, where it occupies a variety of habitats up to 3000m. The genus was monographed in 1996 by RBG Kew‘s Terry Pennington, providing a robust starting point for molecular studies at, and below, the species level. However, obtaining a robust species phylogeny of the group has proved problematic. Phylogenies based on nuclear ITS and plastid trnL-F sequence data produced at RBGE in 2001 (Richardson et. al. Science) were based on only 106 and 16 potentially informative characters respectively, and provided very little resolution for relationships between species of Inga. Even using six potential plastid DNA barcode markers, rpoC1, rpoB, rbcL, matK, trnH-psbA and psbK-psbI (Hollingsworth et al. 2009; a seventh tested locus, atpF-atpH, only amplified in 32% of the samples), we were unable to resolve species relationships in the genus. Partially overlapping with the set of plastid DNA regions used at RBGE, Kursar et al. (2009) also sequenced six plastid DNA regions, rpoC1, trnH-psbA, trnL-F, trnD-T, rps16 and ndhF-rpl32, again without obtaining a well-resolved phylogeny for Inga. Likewise for Kyle Dexter (Dexter et al. 2010), using a combination of nuclear ITS and plastid trnD-trnT sequence data.

One reason relationships within Inga may be difficult to reconstruct using DNA sequence data is that all the species are thought to have arisen in the last 2-10 million years, which means that they are considered ‘recent’ in evolutionary terms (Richardson et. al. Science, Lavin 2006). This leaves little time for molecular changes between DNA sequences to build up, making it difficult to work out how the plants are related to one another.

Lissy Coley collecting Inga in Ecuador

Phyllis Coley collecting Inga in Ecuador

Because Inga is important economically, forms a significant part of tropical biomass and biodiversity, and plays many ecological roles, including containing a wide armoury of antiherbivore chemicals (Kursar et al. 2009), RBGE and colleagues (including Graham Stone and Kyle Dexter at the University of Edinburgh, and Thomas Kusar and Phyllis Coley at the University of Utah) are continuing to research the genus, with ongoing projects aiming to generate enough DNA information to finally generate a well-resolved phylogeny for the group.


Kyle Dexter collecting Inga samples

Kyle Dexter collecting Inga samples

Transcriptome datasets have been generated for three species of Inga, and these were used to identify regions of DNA that are variable between the sampled species. These identified DNA regions were then used to generate a set of 276 target regions for which a MYbaits library was designed (Nicholls et al. submitted). This has now been successfully tested using DNA extracted from 45 silica-dried Inga collections (Nicholls et al. submitted).

Digitized herbarium specimen of Inga umbellifera, collected by Lawrence in 1932

Digitized herbarium specimen of Inga umbellifera, collected by Lawrance in 1932

Inga umbellifera - young leaves, photographed by Phyllis Coley

Inga umbellifera – young leaves, photographed by Thomas Kursar

One of the species used to generate the transcriptomes, and for which 19 individuals have been sampled by Nicholls et al. using the MYbaits protocol, is Inga umbellifera; widespread across Amazonia it is well represented in the collections at RBGE, both as herbarium sheets and silica-dried plant tissue for DNA extraction. Because of its availability and because we already have data available for comparisions, we chose this species to test the utility of next generation sequencing of hybrid baits on DNA extractions from herbarium specimens, dating from 2009 back to 1835.


James A. Nicholls, R. Toby Pennington, Erik J.M. Koenen, Colin E. Hughes, Jack Hearn, Lynsey Bunnefeld, Kyle G. Dexter, Graham N. Stone & Catherine A. Kidner. 2015. Using targeted enrichment of nuclear genes to increase phylogenetic resolution in the neotropical rain forest genus Inga (Leguminosae: Mimosoideae). Frontiers in Plant Science 6: 710. doi: 10.3389/fpls.2015.00710


Capturing Genes from Herbaria. I. What it’s all about.

Capturing Genes from Herbaria. II. Inga.

Capturing Genes from Herbaria. III. The samples.

Capturing Genes from Herbaria. IV. DNA.

Capturing Genes from Herbaria. V. Fragmenting the DNA.

Capturing Genes from Herbaria. VI. Size Selection.

Capturing Genes from Herbaria. VII. Comparisons.

Capturing Genes from Herbaria. VIII. Amplification.

Capturing Genes from Herbaria. IX. Hybrid capture.

Capturing Genes from Herbaria. X. An update.

Capturing Genes from Herbaria. XI. Some metagenomics of a herbarium specimen.

Figuring out your Tree

 Bryology, Other News, Point of Interest, Science  Comments Off on Figuring out your Tree
Jul 262013

Part 1: The Very Basics

Hornwort cladogram and alignment

The backbone cladogram was created as a simple tree file in MacClade, exported to FigTree, then final manipulations done in Illustrator, including the addition of part of the original datamatrix alignment which was exported as a jpg from Geneious

The analyses are finally over, you can fill in those blanks in the results section, and really start dealing with all those hypotheses you set up in your introduction. The end of the talk/ thesis/ paper is in sight… but wait… you have now arrived in that no-man’s-land where all the technical support you’ve been relying on dries up, and people start hiding under their desks when they see you coming – you have got to that point where you’ve got to prepare a Nice Looking Tree Figure. Your dropbox (or pin drive, or laptop) contains some Nexus or Newick files which summarize your tree topology and associated information, like branch lengths or support values. But what now?

It’s astonishing how hard this bit is, how few answers you can get, and how differently everyone you can pin down to question goes about it. For the simple truth is that there is no one answer, just as there is no one piece of software.

Firstly – what trees are you going to need to show? As a rule, multiple pages of trees are a Very Bad Thing. If your journal of choice still produces hard copies, imagine the editor sitting in their office wielding a huge red pen, as printed pages have an associated cost. Consider your results and your discussion. What is the absolute minimum number of tree figures that you require to clearly illustrate your findings?

cysA cladogram

This cladogram was exported from PAUP as a pdf, and heavily edited in Illustrator to add support measures, terminal taxon states and taxonomic information

Given the choice of presenting either a cladogram (where all the branches end at the same distance from the edge of the page) or a phylogram (where the branch lengths are representative of how much sequence variation occurs along them) I always favour the latter, which provides more information about the underlying DNA sequences. However, branch annotations and character mapping on these can be difficult, given that most phylograms contain at least some very short branches. In such cases there is an argument for presenting both. It is also a mistake to try to squeeze too much information onto any one tree – taxonomy plus geography plus character reconstructions plus support values cannot all be easily summarized in the same diagram.

Once you know which figures you are going to require for your paper/ talk/ thesis – and making figures is a laborious job, so it really IS best to settle this beforehand rather than wasting time on figures that turn out to be surplus to requirements – then comes the time to convert that Nexus/Newick file into a stunning visual. The basic output of a programme like PAUP* or Mesquite is not generally publishable as is. In order to be useful, informative and attractive, trees require annotations. Scientifically, things like bootstraps and jackknives and posterior probability distributions need to be added on, in order that your reader/audience can tell which relationships in your study receive statistical support, and thus are most likely to be robust to the addition of further data. It is never worth spending large amounts of ink pontificating about unsupported relationships, as adding more accessions or sequences from other loci often alter unsupported branching patterns.

Depending on the size and complexity of your tree, support values can either be annotated on in their entirity (best for a smaller or straightforward tree), or using a simple graphic measure like altering branch thickness to reflect levels of support (more suitable for a large or complicated tree, where other information is also being mapped).

cysA phylogram

These phylograms represent two character mappings from Mesquite. Illustrator was used to reverse one tree, place both together, remove and replace taxon names and add a scale

But ‘How do I do it?’ is one of those awful questions with 100 equally valid answers. Many phylogenetics programmes will allow you to export a pdf directly, although it can be wise to instead import tree files that have been produced from phylogentic programmes like PAUP or GARLI into FigTree , a nifty (and free) bit of software that allows many visual manipulations (trees can be rerooted, lineages removed, branch order changed, font altered and various colouration schemes applied) followed by exporting of images in graphical formats like pdfs and jpgs.

Personally I tend to import my trees (as graphics) into Adobe Illustrator, which then allows addition of text and alterations of individual branch thicknesses and colours. One slight headache is that font does not export nicely– it tends to look pixelated – and in my published trees I have laboriously deleted and overwritten the exported font with the taxon names that I want to publish. This is not something that you want to do multiple times, but people do not always want to publish the exact names that they have used in their analyses anyway, and it is unwise to spend a lot of time preparing a figure in which you cannot go back and alter taxon names at a later date (like when the reviewer points out your embarrassing typo). If at that stage you have to go all the way back to PAUP, edit a name in a Nexus or Newick file, reexport the tree graphic, and redo all the annotations, you are not going to be very happy, while if all you have to do is alter text in an Illustrator text window, you are in a far better situation.

However, if there is one small repetitive thing that you need to alter right across your tree topology, editing it in the Newick file makes sense – for example a ‘search and replace’ on ‘F.’ for ‘Fossombronia’ if you have used abbreviated genus names that will neither be clear in the publication, nor be acceptable for TreeBase when you come to put your tree files into a public database.

As far as taxon names go, simplicity and clarity are key. If your entire phylogeny is of one genus, then abbreviating the genus name to an initial reduces text without compromising information content. If you have multiple genera, however, genus initials are frankly annoying. Long complicated accession codes are also irritating – if you routinely use something like an institutional DNA number (like ours, e.g. EDNA13-000137) I would at a bare minimum drop the ‘EDNA’ that will appear in each taxon name, and may drop the entire number too. However, each terminal in the tree MUST be able to be compared to the relevant information in your voucher table, so if you have two samples from the same taxon, these must be distinguished either by voucher information, by institutional accession, or by locality information that appears both in the name on the tree figure AND in the voucher table.

If you talk about lineages in the text, don’t expect your reader to work out what they are for themselves – annotate! There are a number of ways to do this, including bars down the side of the tree [insert example]or dropping coloured or shaded boxes underneath the tree (and here using some sort of colour gradation in the boxes can prevent the tree topology from getting confused). Microsoft Powerpoint is an easy and widely available software programme that allows this sort of annotation where more expensive options like Illustrator are not available.

Liverwort genome size reconstructions

Character mapping on this phylogram was performed in Mesquite, then the image file was opened in Illustrator, all the taxon names and key information were removed and rewritten, and annotations were added down the right hand side of the tree

Colour is a powerful tool – tree branches can be coloured to reflect taxonomy. This can be an effective visual for picking out paraphyletic lineages that are scattered across the tree. Colour/shade is also appropriate for simple character mapping, particularly using parsimony. While it is easiest for either/or states, like gene presence or absence (where the reconstructions should also show up in grayscale, the way most people will print out copies of your paper), cold to warm colour gradations can also be used, for example, for small-to-large genome size characters. One of the simplest packages for character state reconstructions is Mesquite, and graphical files created in Mesquite can be exported and tidied up using Adobe Photoshop and Illustrator.

Hornwort backbone phylogeny

This backbone cladogram phylogeny was generated as a Nexus file, opened in FigTree, and the image then imported to Illustrator and Photoshop for manipulation