belong to the family. We argue that family
Lemnaceae deserves to continue as a unique
angiosperm family, for the sake of nomenclatural
stability and morphological clarity.
2.3 Phylogenetic Relationships
Because of their extreme morphological reduction, duckweeds traditionally were difficult to
place among other angiosperms. Fossils of
apparent duckweed relatives have supported a
link with the large family Araceae (Kvaček 1995;
Stockey et al. 1997, 2007, 2016; Coiffard and
Mohr 2018), and various morphological data also
support the current understanding of Lemnaceae
as closely related to Araceae (summarized by Les
et al. 2002). Ongoing molecular phylogenetic
studies (Cabrera et al. 2008; Cusimano et al.
2011; Henriquez et al. 2014) have begun to
portray a consistent phylogenetic position for
Lemnaceae, albeit based solely on plastid DNA
data.
Duckweeds and other Araceae belong to the
monocot order Arales Dumort., containing Araceae, Lemnaceae, and Orontiaceae (Les and
Tippery 2013). Arales are closely related to the
diverse order Alismatales Dumort., which comprises predominantly aquatic plants (e.g.,
Aponogetonaceae, Hydrocharitaceae, Potamogetonaceae), as well as family Tofieldiaceae,
a family of uncertain phylogenetic position (Les
and Tippery 2013). In an analysis that combined
extensive molecular data with numerous fossil
calibration points, Magallón et al. (2015) noted
that Arales diverged from alismatid families
approximately 128.9 Ma (i.e., Arales stem node).
In a prior study, Nauheimer et al. (2012) obtained
a similar stem node estimate (135 Ma), and also
provided a crown node estimate for the diversification of Arales at 121.7 Ma, as well as estimates for the stem (103.6 Ma) and crown
(73.4 Ma) ages of Lemnaceae.
We extended the divergence time estimates
from the study by Nauheimer et al. (2012) by
applying their crown age estimate to a duckweed
tree that was generated using plastid and nuclear
data (Tippery et al. 2015). We converted the tree
to be ultrametric using the chronos function of
the ape package in R, which employs the
penalized likelihood method of tree calibration
(Sanderson 2002; Kim and Sanderson 2008;
Paradis 2013; R Core Group 2018). Using this
strategy, we dated the crown node diversification
of Spirodela to 35.5 Ma, the stem node divergence of Landoltia to 56.8 Ma, as well as nodes
involving the more species-rich genus Lemna
(stem 54.4 Ma, crown 41.7). Because the phylogeny does not clearly differentiate the genera
Wolffia and Wolffiella, we only could date the
crown diversification ages of Wolffia sect. Wolffia
(8.0 Ma) and genus Wolffiella (15.6 Ma)
(Fig. 2.1).
Extensive molecular data have been obtained
for all duckweed species, and they largely support a single phylogenetic hypothesis. Early
studies (Crawford and Landolt 1995; Crawford
et al. 1996, 1997, 2005) used allozyme data to
evaluate interspecific boundaries, and these
upheld the taxonomy used by Landolt (1986).
More recently, Bog et al. (2010, 2013, 2015,
2018) also studied multiple individuals of closely
related species, using AFLP and plastid data, and
identified a number of species groups with
potential interbreeding or incomplete divergence
(see Taxonomy section above). Comprehensive
phylogenetic studies of Lemnaceae have
employed flavonoid biochemical data (Les et al.
1997), as well as DNA sequence data from
plastid (rbcL, matK/trnK, rpl16, rps16; Jordan
et al. 1996; Les et al. 2002; Martirosyan et al.
2009) and nuclear (internal transcribed spacer
(ITS); Tippery et al. 2015) gene regions.
Molecular data helped to justify establishing
Landoltia as a separate genus (Les and Crawford
1999), and they have provided important support
for modern sectional classifications of the larger
genera.
The genera Spirodela and Landoltia are easily
differentiated from each other and from other
duckweeds using morphological or molecular
data (Les et al. 1997, 2002; Les and Crawford
1999; Tippery et al. 2015). Among the larger
genera, Wolffiella consistently has received
strong support in phylogenetic analyses for being
monophyletic, and also for the monophyly and
24
N. P. Tippery and D. H. Les
Lemnaceae deserves to continue as a unique
angiosperm family, for the sake of nomenclatural
stability and morphological clarity.
2.3 Phylogenetic Relationships
Because of their extreme morphological reduction, duckweeds traditionally were difficult to
place among other angiosperms. Fossils of
apparent duckweed relatives have supported a
link with the large family Araceae (Kvaček 1995;
Stockey et al. 1997, 2007, 2016; Coiffard and
Mohr 2018), and various morphological data also
support the current understanding of Lemnaceae
as closely related to Araceae (summarized by Les
et al. 2002). Ongoing molecular phylogenetic
studies (Cabrera et al. 2008; Cusimano et al.
2011; Henriquez et al. 2014) have begun to
portray a consistent phylogenetic position for
Lemnaceae, albeit based solely on plastid DNA
data.
Duckweeds and other Araceae belong to the
monocot order Arales Dumort., containing Araceae, Lemnaceae, and Orontiaceae (Les and
Tippery 2013). Arales are closely related to the
diverse order Alismatales Dumort., which comprises predominantly aquatic plants (e.g.,
Aponogetonaceae, Hydrocharitaceae, Potamogetonaceae), as well as family Tofieldiaceae,
a family of uncertain phylogenetic position (Les
and Tippery 2013). In an analysis that combined
extensive molecular data with numerous fossil
calibration points, Magallón et al. (2015) noted
that Arales diverged from alismatid families
approximately 128.9 Ma (i.e., Arales stem node).
In a prior study, Nauheimer et al. (2012) obtained
a similar stem node estimate (135 Ma), and also
provided a crown node estimate for the diversification of Arales at 121.7 Ma, as well as estimates for the stem (103.6 Ma) and crown
(73.4 Ma) ages of Lemnaceae.
We extended the divergence time estimates
from the study by Nauheimer et al. (2012) by
applying their crown age estimate to a duckweed
tree that was generated using plastid and nuclear
data (Tippery et al. 2015). We converted the tree
to be ultrametric using the chronos function of
the ape package in R, which employs the
penalized likelihood method of tree calibration
(Sanderson 2002; Kim and Sanderson 2008;
Paradis 2013; R Core Group 2018). Using this
strategy, we dated the crown node diversification
of Spirodela to 35.5 Ma, the stem node divergence of Landoltia to 56.8 Ma, as well as nodes
involving the more species-rich genus Lemna
(stem 54.4 Ma, crown 41.7). Because the phylogeny does not clearly differentiate the genera
Wolffia and Wolffiella, we only could date the
crown diversification ages of Wolffia sect. Wolffia
(8.0 Ma) and genus Wolffiella (15.6 Ma)
(Fig. 2.1).
Extensive molecular data have been obtained
for all duckweed species, and they largely support a single phylogenetic hypothesis. Early
studies (Crawford and Landolt 1995; Crawford
et al. 1996, 1997, 2005) used allozyme data to
evaluate interspecific boundaries, and these
upheld the taxonomy used by Landolt (1986).
More recently, Bog et al. (2010, 2013, 2015,
2018) also studied multiple individuals of closely
related species, using AFLP and plastid data, and
identified a number of species groups with
potential interbreeding or incomplete divergence
(see Taxonomy section above). Comprehensive
phylogenetic studies of Lemnaceae have
employed flavonoid biochemical data (Les et al.
1997), as well as DNA sequence data from
plastid (rbcL, matK/trnK, rpl16, rps16; Jordan
et al. 1996; Les et al. 2002; Martirosyan et al.
2009) and nuclear (internal transcribed spacer
(ITS); Tippery et al. 2015) gene regions.
Molecular data helped to justify establishing
Landoltia as a separate genus (Les and Crawford
1999), and they have provided important support
for modern sectional classifications of the larger
genera.
The genera Spirodela and Landoltia are easily
differentiated from each other and from other
duckweeds using morphological or molecular
data (Les et al. 1997, 2002; Les and Crawford
1999; Tippery et al. 2015). Among the larger
genera, Wolffiella consistently has received
strong support in phylogenetic analyses for being
monophyletic, and also for the monophyly and
24
N. P. Tippery and D. H. Les