THE CLADOGRAMS

Sixteen cladograms are presented here to summarize the cladistic analyses. Cladograms 17 through 19 interpret diagrammatically trees of other workers at the suprageneric level. In the cladograms given here, dichotomous branching at a node is shown as “─┤” while multiple branching at a node is symbolized by a longer vertical line with several horizontal lines appended to the right. Thus all terminal taxa and subclades attached to the right of a single vertical line are branches, e.g. “G” is a multiple-branched node in the consensus trees of Cladograms 3 and 4, and so on. The outgroup is also shown as coming from a multiple-branched node in all cladograms because the algorithm, apparently, cannot evaluate the relative phylogenetic position of the outgroup and the first branch; for the purposes of this study, this can be ignored.

Cladograms 1 through 6 are based on the full data set, which includes information on ten genera not in the Pottiaceae: Polytrichum (Polytrichaceae); Bryobartramia and Encalypta (Encalyptaceae); Ceratodon and Ditrichum (Ditrichaceae); Diphyscium (Diphysciaceae); Grimmia (Grimmiaceae); Ptychomitrium (Ptychomitriaceae); Syrrhopodon (Calymperaceae); and Timmia (Timmiaceae). Cladograms 2–4 do the following: (1) compare the cladograms of genera relatively distant from the Pottiaceae, Polytrichum and Timmia, to those of putative sister groups of the Pottiaceae, and (2) identify the closest appropriate haplolepideous sister group to the Pottiaceae as those two (Ditrichum and Ptychomitrium) placing the relatively distant genera Polytrichum and Timmia closest to the base of the tree. Because Ditrichum is potentially a derivative of the Barbula lineage or at least may belong to the Merceyoideae (Cladogram 4), Ptychomitrium is probably the ideal sister for use as outgroup in analysis of the Pottiaceae.

All the listed non-pottiaceous genera were examined individually as outgroups; Cladograms 5 and 6, with Encalypta and Grimmia respectively as outgroups demonstrate that, like cladograms with Bryobartramia, Ceratodon, Ditrichum, Diphyscium, and Syrrhopodon (these not shown), sister groups that are apparently reduced and simplified in morphology create cladograms that place Polytrichum and Timmia high in the tree. Because the data set used is the same in Cladograms 1–6, the branching structure of Cladograms 2–6 would be identical if they had been able to be calculated exactly. To the extent that the actual heuristically derived trees of Cladograms 2–6 agree, this ideal branching pattern is approximated. Branching patterns of Cladograms 2–6 (as well as the other cladograms) that are well resolved support the classification indicated by lettered lineages in Cladogram 16.

Cladograms 7 through 10 include only pottiaceous genera in addition to a non-pottiaceous outgroup. Subclades that are similar throughout the 16 cladograms and which are also formally named in Cladogram 16 are marked with letters at ancestral nodes. Cladograms 11 and 12 analyze character state changes in two highly evolved subclades that comprise the same species and the same structure in both Cladograms 9 (Timmia as outgroup) and 10 (Ptychomitrium as outgroup); these are approximately the same as the Merceyoideae and Pottioideae of Cladogram 16 with the major exception that the Hyophileae is not a distinct subclade. Cladograms 13 through 16 involve only genera in the Pottiaceae, use Timmiella as functional outgroup, and summarize the best hypothesis for phylogenetic relationships and of a projected suprageneric classification.

Cladograms 1, 2, 7, and 8 have Polytrichum as outgroup, Timmia is outgroup in numbers 3 and 9, Ptychomitrium is outgroup in numbers 4 and 10, Encalypta is outgroup in number 5, and Grimmia is outgroup in number 6.

Cladograms 1–10 and 13–14 are strict consensus trees, summarizing the information from many equally parsimonious trees. Other cladograms are of single trees selected from sets of equally parsimonious trees (one set for each outgroup), usually through the use of the functional ingroup and functional outgroup (FIG/FOG) method of Watrous and Wheeler (1981) on various subclades that have multiple-branched nodes in the consensus tree.

Cladograms 1, 7 and 13 weight all characters alike (at weight 1). All other cladograms in this analysis have 22 reduction-related characters weighted at 1:15 (i.e. reduction-related characters are weighted at 1, non-reduction-related at 15). The reason for this particular level of weighting is that cladograms generated at successively higher weightings but below 1:15 are different from each other, but at 1:15 and higher weighting ratios, they are the same. Because reduction series are observed internal to each of many genera in the Pottiaceae, reduction series may be expected to be several and parallel within the family. These reduction series should be distinguished from each other on the basis of characters other than those generally associated with reduction intragenerically. The 1:15 weighting provides for this. Reduction-related characters cannot be eliminated because, although they may not be expected to be homologous globally, they may be important clues to relationships in particular lineages.

The first cladogram (Cladogram 1) clearly shows what are comparatively reduced taxa, such as Crossidium and Aschisma, clustered in the same lineage. To see if such taxa were associated by characters other than those of reduction, twenty-two characters were selected for the weighting above as either directly associated with reduction in size and complexity, or which are elaborations (e.g. costal ornamentation, capsule wall ornamentation) associated with reduction in the Pottiaceae. The characters each commonly contribute at least two additive steps in the clade. The apotypic states of these reduction-related characters, numbered according to the main character list above, are: 0 gregarious (as opposed to caespitose); 1 stem short; 11 leaf short; 21 leaf base not differentiated; 34 bulging pad of cells present on costa; 35 filamentous costal outgrowths present; 41 dorsal superficial laminal cells walls clearly thicker than the ventral; 52 monoicous; 55 seta short; 56 seta not twisted; 57 theca short; 58 theca spherical; 59 cleistocarpous; 60 exothecial cells bulging; 61 exothecial cells with lens-like medial thickenings; 62 capsule cleistocarpous and rupturing along weak transverse walls; 65 peristome absent; 66 teeth short when present; 68 teeth untwisted when present; 71 calyptra short; 72 spores large; and 74 calyptra mitrate.