The genus Didymodon in its present concept (Zander, 1993) is a moderately large genus (ca 120 species) within the largest moss family (the Pottiaceae). The delimitation of the genus has been subject to considerable recent changes, following the revolutionary concepts of Saito (1975), in which emphasis was for the first time given to gametophyte characters.
The European members of the informal Didymodon rigidulus group comprise the taxa described as D. rigidulus Hedw., Grimmia andreaeoides Limpr. (=D. rigidulus subsp. andreaeoides (Limpr.) Wijk & Margad.), D. glaucus Ryan, Eucladium verbanum W.E. Nicholson & Dixon (=D. verbanus (W.E. Nicholson & Dixon) Loeske), Tortula acuta Brid. (=Didymodon acutus (Brid.) K. Saito), D. validus Limpr., Barbula icmadophila Schimp. ex Müll. Hal., and B. mamillosa Crundw. These taxa were subjected to a complex taxonomic treatment in the course of my PhD studies. The taxonomic methods involved herbarium study, statistical evaluation of quantitative characters, and isozyme analyses.
From the eight intensively studied taxa in the D. rigidulus group, three were discovered not to have a close relationship with D. rigidulus. They are D. glaucus, D. verbanus and Grimmia andreaeoides; the latter was found to be conspecific with the North American D. subandreaeoides (Kindb.) R.H. Zander. Both D. glaucus and D. subandreaeoides merit specific status. Their differences from D. rigidulus are well illustrated by a number of qualitative and quantitative characters, including species-specific zymogram patterns in some isozyme systems. D. verbanus is also specifically distinct from D. rigidulus, but its relationship to D. glaucus has not been fully elucidated. Both taxa are to a great extent distinct in some qualitative and quantitative characters, but they could merely be geographically segregated male and female populations of the same taxon. On the other hand, several other explanations are possible, and so it seems appropriate to distinguish D. verbanus as a species, until more information is available.
D. mamillosus was found to fall within the range of variation of D. rigidulus in all of its characters. D. validus is the taxon with the closest relationship to D. rigidulus, based on the shared qualitative characters (occasional presence of axillary gemmae and identical leaf costa anatomy), and forms transitional to D. rigidulus occur in the centre of its distribution. Plants which are not fully developed are also extremely difficult to distinguish from D. acutus.
D. acutus is regarded as specifically distinct from D. rigidulus. Its distribution is suspected not to extend substantially beyond the European subcontinent, and difficulties in distinguishing this taxon from D. rigidulus in America seem to be caused by the occurrence of other minor taxa of the D. rigidulus group, similar but not identical to D. acutus.
D. icmadophilus is also undoubtedly specifically distinct from D. rigidulus, as shown by the presence of species-specific patterns in several isozyme systems. Mixed stands of the two species have also been found. There are obvious sporophyte differences but fertile material of D. icmadophilus has been found very rarely. No stable distinguishing gametophyte characters could be found during my studies. Barbula abbreviatifolia was shown to be a synonym of D. icmadophilus.
Studies indicated the possible existence of another taxon, extremely closely related to D. rigidulus, in the western Mediterranean area, but the variability of this taxon has to be studied with respect to other Holarctic taxa close to D. cordatus Jur. and D. rigidulus.
Despite my studies, several taxonomic problems remain in both the D. rigidulus group and several other informal groups. Molecular studies are needed to clarify the status of D. validus with respect to D. rigidulus and D. acutus. The same applies to the distinction between D. icmadophilus and D. acutus; between D. verbanus and D. glaucus; between D. insulanus (De Not.) M.O. Hill, D. bistratosus Hébr. & R.B. Pierrot, and D. vinealis (Brid.) R.H. Zander; and between D. maximus (Syed & Crundw.) M.O. Hill and D. giganteus (Funck) Jur. One or two additional taxa may still remain to be described within the D. rigidulus group in the Mediterranean area, particularly with respect to D. cordatus. Several little-known taxa have to be newly evaluated, e.g. D. incrassatus (Lindb.) Broth., D. lamyanus (Schimp.) Thér., and D. tomaculosus (Blockeel) M.F.V. Corley. The types have to be located for several ‘forgotten’ taxa, such as D. barbulae Wibel ex Roem., D. barbuloides Lib. ex Marchal, D. camusii Husn., D. soaresii Luisier, and D. tenellus R. Hedw. ex Brid.; it is anticipated that they will be synonymised with other European taxa.
References
Saito K. 1975. A monograph of Japanese Pottiaceae (Musci). Journal of the Hattori Botanical Laboratory 39: 373-537.
Zander RH. 1993. Genera of Pottiaceae: mosses of harsh environments. Bulletin of the Buffalo Society of Natural Sciences 32: 1-378.
A dozen or so bryologists met on the island of Tenerife in February 2001 for a week in the field, organised ably by Roy and Hilary Perry. The meeting was most successful; a great variety of fascinating habitats was visited and we saw many interesting plants, including a good number of Macaronesian endemics. A slide presentation was given to illustrate some of the species seen and localities visited. An account of the meeting has been written for the Bulletin by Tom Blockeel.
(The account is also available in the Spring meeting 2001 post)
This paper reports progress in the three years since the 1998 BBS Loughborough meeting . It looks at aspects of Plagiochila in Britain, Europe and Macaronesia, but, as the title implies, it is also necessary to take account of the situation further afield. The topics covered include evidence relating to the presence in Macaronesia of P. bifaria (Sw.) Lindenb., P. dubia Lindenb. & Gottsche and P. spinulosa (Dicks.) Dumort., to the synonymy of P. killarniensis Pearson and P. bifaria, and to the systematic position of P. atlantica F. Rose. Much of the work has involved others (who are cited along the way) and I wish to acknowledge particularly close collaboration with Jochen Heinrichs (Göttingen), much facilitated by the tremendous benefits of electronic mail.
Madeira
A Plagiochila from Madeira mentioned at the 1998 Loughborough meeting (Drehwald 960277, initially identified as P. killarniensis) was found in two locations during a holiday in 1999, and has been determined as the Neotropical P. retrorsa Gottsche, along with material from Costa Rica that had been the subject of a chemical investigation . The discovery that the Southern Appalachian endemic P. sharpii H.L. Blomq. is synonymous extends the Central American and Macaronesian distribution.
A different Madeiran Plagiochila is likely to be mistaken for P. spinulosa, but has features (including more teeth, a smooth cuticle and a chemical profile with a large amount of 4-hydroxy-3´-methoxybibenzyl) that distinguish it. So far, no existing name has been found. This taxon may be found readily on the northern side of Madeira; it occurs less frequently than P. bifaria but apparently much more frequently than P. retrorsa.
Tenerife
During the BBS meeting organised by Roy Perry on Tenerife in February 2001, five Plagiochila taxa were observed, of which two were of particular interest. First, Gerard Dirkse demonstrated a taxon that had been determined as P. dubia. This taxon, from the Neotropics, was treated recently by and, following further work, has been synonymised with P. patula (Sw.) Lindenb. . The Neotropical taxon has leaf margins that are long-decurrent ventrally, whereas those of the taxon from Tenerife (and Madeira) are short-decurrent. We now find that the Macaronesian taxon agrees with the type of the North American P. virginica A. Evans; analysis of internal transcribed spacer (ITS) sequence variation of nuclear ribosomal (nr) DNA supports assignment to sect. Contiguae Carl .
The second Plagiochila has been regarded as P. spinulosa, but it has characteristics that distinguish it from that taxon as it is known in the British Isles: the odour of the crushed leaves is more spicy, the leaf insertion is more vertical, and the leaf cuticle is rougher. In addition, the shady, tending to dank, habitat where it was to be found, associated with species such as Trichomanes speciosum Willd. and Heteroscyphus denticulatus (Mitt.) Schiffn., was atypical of the haunts of P. spinulosa in the British Isles. The monoterpenes responsible for the odour are principally ocimenes (mainly alloocimene and neoalloocimene), compounds named after Ocimum basilicum L. (sweet basil). In contrast, the principal monoterpene in P. spinulosa is b -phellandrene . Ocimenes have not been reported previously from Plagiochila, but we have observed them in extracts of P. stricta Lindenb. from Costa Rica and Ecuador. The chemical profiles of these extracts were also similar to those from the Tenerife taxon with respect to the other components (including 9,10-dihydrophenanthrenes). Morphological comparison confirmed that the Tenerife taxon is P. stricta. Preliminary phylogenetic analysis of ITS sequence variation of nrDNA shows that examples of P. stricta from the Neotropics and Tenerife are closer to each other than they are to P. spinulosa .
In other systems, such as those of Mentha species, it is generally thought that a single enzyme is responsible for converting the precursor molecule (geranyl pyrophosphate) to a monoterpene product. If this scheme also applies to liverwort systems, then observation of a different monoterpene in P. stricta compared to P. spinulosa suggests the presence of a different enzyme, rather than switching off of part of a pathway so that one product builds up at the expense of another that would normally be formed further down the pathway.
Azores
At the time we were finishing work on P. retrorsa, we heard from Rosalina Gabriel of a Plagiochila that had been collected by Dias on Pico in 1992 and noticed by Sjögren in the herbarium AZU in 1999. Although the chemical profile was similar to that of P. retrorsa, the leaf cuticle was rough, and the plant matched the types of P. papillifolia Steph., P. deciduifolia Steph. and P. solmsii Steph., all collected by Herzog at Comarapa, Bolivia, in 1911; a further synonym is P. verruculosa R.M. Schust. . P. papillifolia can therefore be added to the Macaronesian flora.
Recently, the Azorean endemic P. allorgei Herzog & Perss. was found to be synonymous with the Neotropical P. longispina Lindenb. & Gottsche . It has been established that the blue colour of the stem tips of this species (at least in the Neotropics) is caused by the presence of oil bodies that contain 1,4-dimethylazulene, the blue compound that also occurs in the oil bodies of Calypogeia azurea Stotler & Crotz.
British Isles (and Europe)
Evidence that P. norvegica H.H. Blom & Holten and P. porelloides (Torrey ex Nees) Lindenb. are very close genetically has been provided by Cronberg , who observed no difference in isozyme banding patterns between the two taxa; in contrast, P. asplenioides (L. emend. Taylor) Dumort. showed several differences. P. norvegica should be placed in sect. Plagiochila.
The plagiochilines (T and U) that we reported from Scottish material of P. carringtonii (Balfour) Grolle suggested a systematic position close to sect. Plagiochila. We have now observed the same two plagiochilines in extracts of subsp. lobuchensis (leg. D.G. Long) from Sikkim and Nepal; the close relationship of the two taxa is supported.
Three years ago there was no chemical evidence to support the synonymy of P. killarniensis and P. bifaria . Now, however, there is: a specimen from Ecuador (collected in 2001, Holz EC-01-416, GOET) that is close morphologically (teste Heinrichs) to the type of P. bifaria from Jamaica has a very similar chemical profile to that of a specimen (Rycroft 00051) from the type locality of P. killarniensis, above the Torc Cascade. The methoxyl region of the proton NMR spectra illustrates vividly that the ratios of three of the characteristic compounds (methyl everninate, killarniensolide and methyl 6-methoxy-2-methyl-3,4-methylenedioxybenzoate) in the two extracts are extremely similar to each other and to those published .
In 1975, Eustace Jones and Francis Rose presented an exhibit entitled ‘The mystery of Plagiochila ambagiosa solved’ at the BBS Conversazione in Reading; a long-standing taxonomic problem had indeed been solved, but the systematic mystery remained. Our chemical studies of P. atlantica showed that the chemical profile was dominated by plagiochiline C, and that a new bicyclogermacrenol derivative, that we named ‘atlanticol’, was present as a minor constituent . In connection with other work, we examined the extract of a specimen of P. aerea Taylor from Costa Rica and observed an extremely similar chemical profile; not only was plagiochiline C dominant but atlanticol, the compound hitherto unique to P. atlantica, was also present. P. aerea, the type species of Plagiochila sect. Bursatae Carl, is one of the few members of the Plagiochilaceae that are characterised by the presence of leaf surface waxes . The obvious investigation was undertaken, and waxes were duly observed on leaves of P. atlantica using scanning electron microscopy . Our results indicate that P. atlantica has close relatives in sect. Bursatae of the Neotropics.
Conclusion
Five taxa (P. papillifolia, P. retrorsa, P. stricta, P. virginica, and the Madeiran indet.) have been added to the recorded Plagiochila flora of Macaronesia since 1999, and one, P. dubia, has been deleted. The position with respect to P. spinulosa requires further investigation. There are now several species linking the Plagiochila floras of the Neotropics and Europe (including Macaronesia).
The recent check-list of hepatics for Europe and Macaronesia includes 14 Plagiochila species in seven sections, four being those of Carl . Ten of these species, in five sections, occur in the British Isles. Now, only one year on, the species tally is 16. Chemically, there is a division between those that contain 2,3-secoaromadendrane sesquiterpenoids and those that produce aromatic (in the chemical sense, i.e. benzenoid) compounds. The chemical division accommodates the morphologically-derived sections: sect. Plagiochila (P. asplenioides, P. britannica, P. porelloides; P. arctica and P. norvegica are included here but have not been studied chemically), sect. Carringtoniae Inoue (P. carringtonii), sect. Bursatae (P. atlantica), and sect. Contiguae (P. virginica) have 2,3-secoaromadendranes; sect. Arrectae Carl (P. bifaria, P. punctata, P. retrorsa, P. spinulosa, P. stricta), sect. Bidentes Carl (P. exigua (Taylor) Taylor and, tentatively, P. papillifolia), and sect. Glaucescentes (P. longispina) have aromatic compounds. Analysis of ITS sequence variation in nrDNA is proving to be a useful tool in understanding these relationships.
References
Anton H, Kraut L, Mues R, Morales Z MI. 1997. Phenanthrenes and bibenzyls from a Plagiochila species. Phytochemistry 46: 1069-1075.
Blomquist HL. 1940. Another new species of Plagiochila from the Southern Appalachian Mountains. Bryologist 43: 89-95.
Carl H. 1931. Die Arttypen und die systematische Gliederung der Gattung Plagiochila Dum. Annales Bryologici, Supplementary Volume 2: 1-170.
Cronberg N. 2000. No difference in isozyme banding patterns between Plagiochila porelloides and P. norvegica. Lindbergia 25: 17-19.
Grolle R, Long DG. 2000. An annotated check-list of the Hepaticae and Anthocerotae of Europe and Macaronesia. Journal of Bryology 22: 103-140.
Heinrichs J, Anton H, Gradstein SR, Mues R. 2000a. Systematics of Plagiochila sect. Glaucescentes Carl (Hepaticae) from tropical America: a morphological and chemotaxonomical approach. Plant Systematics and Evolution 220: 115-138.
Heinrichs J, Anton H, Gradstein SR, Mues R, Holz I. 2000b. Surface wax, a new taxonomic feature in Plagiochilaceae. Plant Systematics and Evolution 225: 225-233.
Heinrichs J, Anton H, Holz I, Gradstein SR. 2000c. On the blue stem colour in Plagiochila longispina Lindenb. & Gottsche (Plagiochilaceae). Cryptogamie, Bryologie 21: 109-111.
Heinrichs J, Gradstein SR. 2000. A revision of Plagiochila sect. Crispatae and sect. Hypnoides (Hepaticae) in the Neotropics. I. Plagiochila disticha, P. montagnei and P. raddiana. Nova Hedwigia 70: 161-184.
Heinrichs J, Grolle R, Drehwald U. 1998. The conspecificity of Plagiochila killarniensis Pearson and P. bifaria (Sw.) Lindenb. (Hepaticae). Journal of Bryology 20: 495-497.
Heinrichs J, Pröschold T, Renker C, Groth H, Rycroft DS. In press-a. Plagiochila virginica A. Evans rather than P. dubia Lindenb. & Gottsche occurs in Macaronesia; placement in sect. Contiguae Carl is supported by ITS sequences of nuclear ribosomal DNA. Plant Systematics and Evolution.
Heinrichs J, Rycroft DS. 2001. Leaf surface waxes and lipophilic secondary metabolites place the endemic European liverwort Plagiochila atlantica F. Rose in the Neotropical Plagiochila sect. Bursatae Carl. Cryptogamie, Bryologie 22: 95-103.
Heinrichs J, Rycroft DS, Groth H, Cole WJ. In press-b. Morphological and phytochemical studies of Plagiochila papillifolia Steph., a Neotropical liverwort new to Europe. Journal of Bryology.
Rycroft DS. 1999. A chemist’s view of liverworts: NMR fingerprinting and chemotype classification of British Plagiochilae. Bulletin of the British Bryological Society 72: 50-54.
Rycroft DS, Cole WJ. 1998. Atlanticol, an epoxybicyclogermacrenol from the liverwort Plagiochila atlantica F. Rose. Phytochemistry 49: 1641-1644.
Rycroft DS, Cole WJ, Aslam N, Lamont YM, Gabriel R. 1999a. Killarniensolide, methyl orsellinates and 9,10-dihydrophenanthrenes from the liverwort Plagiochila killarniensis from Scotland and the Azores. Phytochemistry 50: 1167-1173.
Rycroft DS, Cole WJ, Heinrichs J, Groth H, Renker C, Pröschold T. In preparation. Phytochemical, molecular and morphological evidence for the occurrence of the Neotropical liverwort Plagiochila stricta in the Canary Islands, new to Macaronesia.
Rycroft DS, Cole WJ, Lamont YM. 1999b. Plagiochilines T and U, 2,3-secoaromadendranes from the liverwort Plagiochila carringtonii from Scotland. Phytochemistry 51: 663-667.
Rycroft DS, Heinrichs J, Cole WJ, Anton H. 2001. A phytochemical and morphological study of the liverwort Plagiochila retrorsa Gottsche, new to Europe. Journal of Bryology 23: 23-34.
The Sematophyllaceae contain three British species (one discovered only recently), all of which are treated in the British Isles as belonging to the genus Sematophyllum. This paper focuses on the significance of the two species known since bryological research started in the British Isles (S. demissum and S. micans), and where they fit within the family.
Although the Sematophyllaceae are predominantly found in the tropics, where they can be very common, there are a number of essentially temperate species, which on the whole – as with the British plants – are rather rare. It is a family that poses a lot of taxonomic problems, particularly in the area of genus and family delimitations, and the relation of the genera in Sematophyllaceae s.l. with the rest of the Hypnales. Molecular studies indicate that diversification in the Hypnales occurred comparatively rapidly, probably associated with the spread of trees in the early Tertiary; because of the speed of change, there is very little morphological or molecular evidence indicating how taxa are related. In addition, the majority of the extant species are probably very recent and actively evolving, so there has been little time for distinctive characters to appear and be differentiated.
The level of diversity in mosses measured by the number of described species is significantly skewed in the tropics, where many species were described at a time when the tropics were seen as an area of boundless fecundity and diversity, and almost every new collection was described as a new species. Taxonomic work in the tropics has been constantly hindered by the sheer volume of work in examining hundreds of type specimens to find the few that are good species. Ireland (1992) boldly reduced the Latin American Isopterygium species from 92 to eight in his revision of the genus, and suggested there might be further reductions to come. Similar reduction might occur when the 64 African species of Sematophyllum are revised; 59 of them are endemic, and 40 occur in only one country, many of them known from only one collection. O’Shea (1997) estimated that the number of moss taxa recognised in sub-Saharan Africa might reduce by 57%.
S. micans has been seen as a bit of an oddment for some time. In America it has traditionally been placed in Hygrohypnum because of the leaf shape, semi-aquatic habitat and somewhat inflated leaf cells, whereas in Europe it has been put in Sematophyllum because of the inflated, coloured alar cells. In fact it has a poor resemblance to either genus, as it differs from Hygrohypnum in its lack of a central strand, decurrent leaves, short-cylindric capsules with round-pored stomata (Hygrohypnum has oval-pored stomata) and short-rostrate opercula, and from Sematophyllum in its atypical alar cells, decurrent leaves and quite different sporophyte. Buck (1997) created a new genus for it: Schofieldiella, which he placed in the Hylocomiaceae. Although it doesn’t look like a Hylocomium, Buck listed the following similarities: relatively strongly toothed, decurrent branch leaves, relatively strong double costa, short-cylindric capsules, and round-pored stomata. He aligned it to the genus Leptocladiella in the Hylocomiaceae, which he thought showed striking similarities, including the fact that it has species as small as S. micans, although S. micans lacks the sympodial growth form. Tan & Jia (1999) took a different approach and opted for the genus Hageniella in the Sematophyllaceae. They maintained that the differences from more typical sematophylloid plants is within a permissible range; the only problem is the decurrent leaves, but they point out that these are only slightly more decurrent than the type species of Hageniella. The conclusion from this is not entirely clear. It may be that we have found the right genus, but there may be more debate about whether the genus is in the correct family. S. micans remains an interesting but puzzling plant.
S. demissum also presents taxonomic difficulties. H.N. Dixon, apart from being the author of three editions of the Student’s Handbook of British Mosses and being the first BBS President in 1923, was also one of the most prolific authors of new species this century. Although he never went to the tropics, from the first few years of the 20th century he took an interest in what were called ‘exotic’ mosses. I don’t think anyone has calculated how many species he described before his death in 1944, but it was well over a thousand, and they were from all over the world. One of his interests was the Sematophyllaceae, and he spent some time looking at a group in Sematophyllum called the ‘caespitosum group’. Species in this group are most easily characterised by their leaf shape and the very short cells towards the leaf apex. Dixon published his findings in 1920, and made 58 species synonyms of S. caespitosum (now known as S. subpinnatum) on the basis that it was a very plastic taxon that grew in a wide variety of habitats and the variation was environmentally determined. In many ways this was quite revolutionary, as the tradition had been that anything slightly different was described as a new species (and Dixon himself followed this practice in much of his work). The interest in this for us is that S. demissum (and S. adnatum, recently found naturalised in Italy, via imported American trees) share similarities with this group, although they were not dealt with by Dixon, presumably because they weren’t tropical. Dixon spent two pages trying to define his concept of S. caespitosum, but he found it extremely difficult to characterise. He certainly gave up on the idea of defining varieties, as the variation was so great in all directions, with so little correlation between characters. A look at American moss Floras will indicate the similarity of S. caespitosum/subpinnatum and S. demissum. Buck (1998) considered that European S. demissum might be the same as S. adnatum, although American S. demissum is not the same and should be called S. carolinianum. Looking at collections elsewhere in the range of S. demissum, the plants in Hong Kong and Japan look similar to ours, but so does the Japanese S. robustulum. Whilst we are normally quite sure about how our species are delimited in Britain, here we have found a weakness that needs to be tackled, so perhaps this throws it back to bryologists in the BBS. What is S. demissum? Is it a good species, or is it a synonym of something else, e.g. is it the same taxon as S. adnatum? There are plenty of challenges left in the Sematophyllaceae – even in Europe.
Buck WR. 1997. Schofieldiella (Hylocomiaceae), a new genus for an old species. Journal of the Hattori Botanical Laboratory 82: 39-46.
Buck WR. 1998. Pleurocarpous mosses of the West Indies. Memoirs of the New York Botanical Garden 82: 1-400.
Dixon HN. 1920. Rhaphidostegium caespitosum (Sw.) and its affinities. Journal of Botany 58: 81-89.
Ireland RR. 1992. The moss genus Isopterygium (Hypnaceae) in Latin America. Tropical Bryology 6: 111-132.
O’Shea BJ. 1997. The mosses of sub-Saharan Africa. 1. A review of taxonomic progress. Journal of Bryology 19: 509-513.
Tan BC, Jia Y. 1999. A preliminary revision of Chinese Sematophyllaceae. Journal of the Hattori Botanical Laboratory 86: 1-70.
Historically, Wiltshire has been the least well-known bryologically of all the counties of southern England, and the first records were not published until 1891. South Wiltshire comprises the southern half of the county, the northern boundary of VC 8 being the Kennet and Avon Canal. There are some significant differences from the present administrative boundary.
Large parts of VC 8 are on chalk, some of which is covered with Clay-with-Flints. Other important geological formations include the Upper Greensand (which gives rise to relatively acid soils), Jurassic limestones in the west, and Tertiaries in the south-east with acid sands and gravels.
Wiltshire is less wooded than most other counties in southern England, but there are extensive areas of woodland in the west on some large estates. Some of the more interesting bryophytes in ancient woodlands (i.e. woodlands since at least 1600 AD) include Hylocomium brevirostre, Rhytidiadelphus loreus, Herzogiella seligeri, Plagiothecium latebricola and Trichocolea tomentella.
There is much arable land on the chalk, as well as unimproved grassland. The latter is particularly extensive on the Ministry of Defence lands, the largest being Salisbury Plain Training Area which occupies about one-fifth of the vice-county. Generally, the chalk grassland is not very interesting bryologically as the grass is too long for bryophytes, but some closely grazed areas have characteristic mosses, such as Weissia species (including W. sterilis), Rhodobryum roseum and Thuidium abietinum subsp. hystricosum, while bare chalky patches may have Microbryum rectum, M. curvicolle, M. floerkeanum, Acaulon muticum and Ephemerum recurvifolium.
The south-east part of VC 8 in or near the New Forest (much now being in Hampshire) has wet and dry heathland with Dicranum spurium, Splachnum ampullaceum, Calliergon stramineum, Hypnum imponens, Campylopus brevipilus, Cladopodiella fluitans, C. francisci, Cephalozia macrostachya, Ptilidium ciliare and many other bryophytes of interest, including fifteen Sphagnum species.
The main bryologists in VC 8 in the first half of the 20th century were C.P. Hurst in the north-east and the Dunston brothers in the south-west. After 1950, much useful recording was done by Jean Paton, Joan Appleyard, Ted Wallace and Francis Rose. The BBS had its spring meeting in 1989 based at Salisbury, and it became clear that much of VC 8 was still poorly known bryologically. A relatively intensive survey was carried out over a period of eight years until 1999. Distribution maps were prepared on a 10-km square basis with the intention of producing a bryophyte flora of the vice-county.
Papers describing research using phylogenetic methods and molecular data are now widespread in many of the bryological journals, but how relevant are these methods, and the results obtained using them, to the bryologist in the field? It is of course impossible (at present) to use DNA to identify a plant in the field, just as it is impossible for most people to use scanning electron microscopy to look at spore walls to identify a plant. However, just as scanning electron microscopy permits an understanding of relationships that can be relevant to the bryologist in the field, so do phylogenetics and molecular data.
We all know taxa – species and genera – that are rather fickle and cannot be pinned down to a genus or a family, so that circumscriptions keep changing and taxa keep hopping about and turning up in different places in different classifications. One example is the plant that I learnt as Isopterygium elegans (Brid.) Lindb., previously Plagiothecium elegans Brid., and now Pseudotaxiphyllum elegans (Brid.) Z. Iwats. The new generic name has been largely accepted in the literature (e.g. Blockeel & Long, 1998), but the family placement has varied. The species has been transferred around in the Plagiotheciaceae, where it is still placed by the taxonomic database at Missouri Botanic Garden (http://mobot.mobot.org/W3T/Search/most.html), but is listed under the Hypnaceae by Buck & Goffinet (2000). In another example, Hypnum lindbergii Mitt. was transferred to Calliergonella in the Amblystegiaceae by Lars Hedenäs in 1990, but this move has not been widely accepted. On a much larger scale, molecular sequence data have shown that the orders Hypnales and Leucodontales (=Isobryales) are not natural – different groups of leucodontalean mosses appear to have evolved independently from the Hypnales. This may be no surprise to many – there are a number of taxa that have been transferred between families or from the Hypnales to the Leucodontales and back again by different authors. Despite accounting for perhaps half of the species of extant mosses worldwide, there is very little hierarchical structure in the pleurocarps when compared to the acrocarpous taxa. Mosses in the Leucodontales are characterised by epiphytic habit and reduced peristomes, features that have frequently been shown to be correlated, and that may reflect repeated parallel evolution from terrestrial plants with complete peristomes.
The first two examples given above both relate to changes based on morphology, and reflect re-assessments of the relationships of small groups of species. Do these changes actually make sense? Are they likely to be modified in the future? Is there any way of determining which changes are reasonable and which don’t really help? Where do these plants really belong? And on the larger scale, are there sub-groups within the pleurocarpous mosses that can be recognised as orders, or at any other level? It is these kinds of question that phylogenetics and molecular data may help us resolve.
Both new kinds of data and new ways of analysing data are involved. A very large proportion of the work of recognising and describing species, genera and other taxa has been carried out using morphological and ecological data, with progressive refinements, over at least the last 2000 years. But intractable problems remain, some of which we are not even aware of, that cannot be solved with traditional methods. Morphological characters in mosses are notoriously plastic. For example, costa length and form, cell size and shape, and peristome morphology, may be so variable that they provide little useful information, and arguments have raged about the relative importance of the sporophyte and the gametophyte for understanding higher level relationships. However, many morphological characters may be highly informative, including both traditional characters and ‘new’ characters, such as axillary hairs, pseudoparaphyllia, and the presence or absence of a hyalodermis on the stem. Character analysis, which basically consists of careful scrutiny to determine whether a morphological feature is positionally and structurally ‘the same’ as that in another plant, is a necessary preliminary step. The disadvantages of using morphological data are the variability and plasticity of the characters, the long period of training and experience necessary to understand what is seen down the microscope, and the rather few informative characters found relative to the input of effort (usually no more than one to two characters per taxon). One advantage is that morphological characters may reflect the processes of evolution in a stop/start fashion, with more changes happening at periods of great evolutionary change (the periods we are most interested in reconstructing) and fewer in the long quiet periods between.
Molecular data are derived from several different sources – sequences of DNA bases in nuclear, chloroplast, mitochondrial or ribosomal genes, amino acids in proteins, and structural elements in DNA, such as insertions and deletions (indels) and secondary folding. Chloroplast sequences are most often used by botanists, partly because their use reduces problems from non-plant contaminants. There is now a very large amount of data available, especially for the gene rbcL (which codes for production of the essential photosynthesis enzyme ribulose biphosphate carboxylase), allowing extensive comparative studies to be made. Molecular data are also subject to character analysis, in the form of sequence alignment, with each position in the DNA chain representing a character. One of the advantages of molecular data is that they are not directly affected by the ecological pressures that affect morphology. In addition, molecular data can be easily obtained in a suitable lab, and provide much larger amounts of information – frequently more than ten informative characters per taxon. This makes computer analysis much easier and faster; up to a point, the more comprehensive the sampling of taxa, and the more informative characters per taxon, the easier the analysis. One of the principal disadvantages of molecular data is that they appear to evolve at a more or less constant rate in any given lineage, so that ‘fast’ evolutionary events may be missed, or overwritten by later changes.
Phylogenetic analysis, or more specifically, cladistics, is based on principles laid down by Willi Hennig in the 1960s (Kitching et al., 1998), many of which codify good taxonomic practice that would have been used subconsciously by the better earlier systematists. One example is the nature of the information supplied by different characters. These are referred to as autapomorphies, synapomorphies, plesiomorphies and homoplasies. An autapomorphy is a character state that is unique to a single species (or to a single group), and which allows us to identify the species, but tells us nothing about the relationship of that species to others. For example, Rhytidium rugosum has rugose leaves, which allows us to identify it with ease, but this doesn’t tell us what it is related to. This can also apply at the level of a genus or other group – if a genus has a certain character state, but nothing else does, this tells us nothing about the relationship of the genus to other genera. However, if all descendants of a common ancestor share a character state which does not occur in any other taxa, it does tell us about the membership of the group, and is then termed a synapomorphy. But at a different level, if you are looking at relationships within a subset of the group, the character is not informative. Possession of an articulated peristome is a synapomorphy for the arthrodontous mosses, but within the group this character is plesiomorphic, and, for example, tells us nothing about the relationship of Bryum to Hypnum. (Other details of peristome structure might, however, provide useful information to address this problem.) Finally, character states are referred to as homoplasious if the apparent similarities are superficial and do not reflect a shared evolutionary origin. There are different ways in which this might happen – convergent or parallel evolution, reversal, or misinterpretation on the part of the observer. A character state that is homoplasious will provide incorrect or misleading information about relationships. For example, erect capsules are found in a wide range of taxa, but we don’t think that all species with erect capsules are more closely related to each other than they are to species with inclined capsules.
A large amount of data is used in most phylogenetic studies, and analysis is carried out using various computer programs; PAUP and Hennig 86 have been widely used, but there are many others. The programs use logical or evolutionary models to build and assess cladograms, which are approximations of phylogenetic (or evolutionary) trees. There are two basic types: maximum parsimony (which endeavours to find the cladogram that requires the smallest number of evolutionary changes to explain the data at hand) and maximum likelihood (which estimates the likelihood of finding the data at hand for a given cladogram). Both have advantages and disadvantages, and there is an enormous literature discussing the theoretical and methodological issues. New techniques are being developed all the time, e.g. one of the methods currently being actively explored is Bayesian inference. The end result is a cladogram (often referred to as a tree), or more often a series of cladograms which vary to a greater or lesser extent. Each cladogram represents a different possible arrangement of the data, and they are often treated as estimates of the actual course of evolution. Often such groups of cladograms are summarised as consensus trees, with a ‘strict’ consensus, for example, showing only the groups that are found in all the different cladograms. Jacknife and bootstrap values, and decay indices, are also often shown, providing different assessments of the support for individual groups. The cladograms are hypotheses, our current best estimates of the relationships of the organisms, and as such can never be regarded as ‘true’, but they do provide a tool to further explore and understand relationships.
So how do the results of this work affect the bryologist in the field? Much of the work is still very preliminary, but two different kinds of result can be expected. One is clarification of the relationships of individual problematic taxa, and the second is clarification of the relationships of larger groups, so that families and orders become more natural. There have already been some interesting rearrangements of taxa that previously were very poorly understood, although most of these have concerned species or genera that do not occur in the British Isles. For example, several odd taxa that lack peristomes have been transferred to groups very far from their original placements, but with which they share various other morphological features. One of these is Oedipodium griffithianum, usually placed in the Splachnales, but shown by molecular data to belong near the base of the moss lineage near Andreaea, and therefore primitively lacking a peristome (Newton et al., 2000). The position of Hypnum lindbergii has not yet been clarified, although most molecular studies place it distinct from the Hypnaceae, either near the Hylocomiaceae (Buck et al., 2000) or, supporting Hedenäs (1990), with members of the Amblystegiaceae (unpublished data).
Although some orders and families are well circumscribed and useful, others don’t really help much in placing individual species. The result is that many people learn each genus or species as an individual entity. This takes more time and effort than being able to quickly place any species in the higher levels of the hierarchy, where there are fewer possibilities to consider, and fewer chances of going wrong somewhere in the key. If molecular and phylogenetic studies result in the development of more natural classifications, this should eventually make identification and understanding of genera and species more straightforward.
References
Blockeel TL, Long DG. 1998. A check-list and census catalogue of British and Irish bryophytes. Cardiff: British Bryological Society.
Buck WR, Goffinet B. 2000. Morphology and classification of mosses. In: Shaw AJ, Goffinet B., eds. Bryophyte biology. Cambridge: Cambridge University Press, 71-123.
Buck WR, Goffinet B, Shaw AJ. 2000. Testing morphological concepts of orders of pleurocarpous mosses (Bryophyta) using phylogenetic reconstructions based on trnL-trnF and rps4 sequences. Molecular Phylogenetics and Evolution 16: 180-198.
Hedenäs L. 1990. Taxonomic and nomenclatural notes on the genera Calliergonella and Breidleria. Lindbergia 16: 161-168.
Kitching IJ, Forey PL, Humphries CJ, Williams DM. 1998. Cladistics. The theory and practice of parsimony analysis. Oxford: Oxford University Press.
Newton AE, Cox CJ, Duckett JG, Wheeler JA, Goffinet B, Hedderson TAJ, Mishler BD. 2000. Evolution of the major moss lineages: phylogenetic analyses based on multiple gene sequences and morphology. Bryologist 103: 187-211.
The group met in the large Forestry Commission car park below the Wyndcliff quarry (ST59I). After an initial get-together we moved off to the first site close to the car park. Several tiny limestone pebbles were shown to support Seligeria campylopoda, a S. recurvata look-alike listed in the Red Data Book as Data Deficient. There was discussion about the conservation of S. campylopoda, the need to include it in the Monmouthshire local Biodiversity Action Plan, and the importance of keeping the pebbles free from fly-tipped rubbish. At the time, this was one of only two known colonies at the species’ only known extant British site. However, Jonathan Sleath discovered two further small populations on pebbles below the car park.
After we had all enjoyed this tiny rarity, things continued to be ‘data deficient’ as we tried to relocate Ditrichum flexicaule s.s., collected from the Wyndcliff 110 years ago by Binstead and Shoolbred. Various candidates were collected from the quarry floor and taken home for later inspection; unfortunately, all proved to be Dicranella varia. Back in the car park, further interest was provided by some Hypnum lindbergii found by Mark Pool, a new record for the Wye valley, and by some ephemerals, including Dicranella schreberiana and Bryum klinggraeffii, demonstrated on a bank by Mark Hill.
Our descent into the Wye valley below the car park, an area of woodland known as Cave Wood, was accompanied by mutterings about how dry the area was. The leader assured people that there were a few more humid areas and, sure enough, we soon found Trichocolea tomentella on rocks in a small stream. Sharing the rocks were plenty of Fissidens rivularis c.fr., an unusually common species in the Wye valley, Riccardia chamedryfolia and Chiloscyphus polyanthos. In our eagerness to reach the stream we had walked past the best humidity indicator of all – Lophocolea fragrans c.fr. was found by Jean Paton and Tom Blockeel in abundance on a large plane tree by a ruined cottage some 50 metres up the path.
With lunch beckoning we split into several small groups to look for calcicoles. Most groups noted Porella arboris-vitae growing in small quantity on low crags with P. platyphylla, Neckera crispa and Anomodon viticulosus, as well as Taxiphyllum wissgrillii and Eurhynchium crassinervium on small rocks. Sam Bosanquet showed a few people Fissidens gracilifolius on fine limestone scree, and Nick Hodgetts found some Plagiochila britannica. A couple of crags plastered with Marchesinia mackaii were located and, while searching one of these, Graham Motley found a Cololejeunea that was assumed in the field to be C. rossettiana; a little was collected and it proved to be C. calcarea, another new species for the site.
After lunch a slightly smaller group moved to Great Barnet’s Woods (ST59C), a mile west of Chepstow. Sam had found Thuidium recognitum here in early 2000 and hoped to assess the size of the population. Unfortunately, finding the limestone pavement on which the Thuidium grew proved difficult enough and the scarce species was not relocated. Despite this, the group remained in good humour and found several new species for the site. Most notable were Jean’s female Pellia neesiana, growing on a trackside with male P. endiviifolia, and John Blackburn’s Leucobryum juniperoideum on a tree stump. As well as a number of calcicoles, this area of poorly-developed pavement held various locally rare mosses, including Hylocomium brevirostre, at one of its very few Monmouthshire sites, and Rhytidiadelphus loreus.
The chance of seeing unusual species, such as Seligeria campylopoda, Fissidens rivularis and Lophocolea fragrans, merely provided a background to bryologising in the field with many friendly BBS members. I hope everyone had an enjoyable time on their brief visit to Monmouthshire.
Sam Bosanquet
Cardiff