Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 103 (42), 15511-6

The Deepest Divergences in Land Plants Inferred From Phylogenomic Evidence


The Deepest Divergences in Land Plants Inferred From Phylogenomic Evidence

Yin-Long Qiu et al. Proc Natl Acad Sci U S A.


Phylogenetic relationships among the four major lineages of land plants (liverworts, mosses, hornworts, and vascular plants) remain vigorously contested; their resolution is essential to our understanding of the origin and early evolution of land plants. We analyzed three different complementary data sets: a multigene supermatrix, a genomic structural character matrix, and a chloroplast genome sequence matrix, using maximum likelihood, maximum parsimony, and compatibility methods. Analyses of all three data sets strongly supported liverworts as the sister to all other land plants, and analyses of the multigene and chloroplast genome matrices provided moderate to strong support for hornworts as the sister to vascular plants. These results highlight the important roles of liverworts and hornworts in two major events of plant evolution: the water-to-land transition and the change from a haploid gametophyte generation-dominant life cycle in bryophytes to a diploid sporophyte generation-dominant life cycle in vascular plants. This study also demonstrates the importance of using a multifaceted approach to resolve difficult nodes in the tree of life. In particular, it is shown here that densely sampled taxon trees built with multiple genes provide an indispensable test of taxon-sparse trees inferred from genome sequences.

Conflict of interest statement

The authors declare no conflict of interest.


Fig. 1.
Fig. 1.
The 50% majority-rule BS consensus trees from ML and MP analyses of the multigene supermatrix. The numbers above the branches are ML BS values >50%; those below are MP BS values >50%. For nodes where ML and MP analyses differ in topology, only the ML topology and BS values are shown. The BS values depicting the backbone relationships in land plants are shown in boldface.
Fig. 2.
Fig. 2.
One of the 18 shortest trees from MP analysis of the intron matrix [tree length, 43 steps (DELTRAN optimization); consistency index, 0.65; retention index, 0.80]. The numbers above the branches indicate branch lengths; those below are BS values >50%. Asterisks represent the nodes that collapse in the strict consensus of all shortest trees. Ch, charophytes; Lv, liverworts; Ms, mosses; Hw, hornworts; VP, vascular plants.
Fig. 3.
Fig. 3.
Results of ML and MP BS analyses of the eight cp-genome matrices. (A) The BS consensus tree from ML analysis of the all codon position matrices, with all taxa presented. (B–D) Schematic diagrams of the BS consensus trees from ML analyses of the other seven matrices, with only major land plant lineages indicated. In all trees, ML and MP BS values are presented above and below branches, respectively. In BD, the BS values for analysis of each matrix, when different, are separated by slashes, and the one BS value of 100 indicates identical values in different analyses. a, Amborella alone was sister to all other angiosperms with 100/77 MP BS support; b, Vitis was sister to all other eudicots with MP 100/50 BS support; c and d, Marchantia was sister to all other land plants with 100/45 and 100/99 MP BS support, respectively; e and f, Psilotum and Adiantum formed a clade sister to all other land plants with 100/63 and 100/55 MP BS support, respectively; and further for f, the clade of bryophytes and lycophytes was sister to other vascular plants; g, Huperzia was sister to Anthoceros plus Selaginella. Ch, charophytes; Lv, liverworts; Ms, mosses; Hw, hornworts.

Similar articles

See all similar articles

Cited by 144 PubMed Central articles

See all "Cited by" articles

Publication types

Associated data