Since flowering and fruiting are rarely observed in most species of Lemnaceae, the following keys and descriptions are based primarily on vegetative characteristics. Minor traits which might seem insignificant in morphologically complex plants assume greater importance in the Lemnaceae. Ideally, it is best to observe living plants under a 30X dissecting microscope, preferably with substage lighting to view veins and the shape of budding pouches (dried herbarium specimens can be hydrated in water to obtain a resemblance of their former shape). For difficult species it is often necessary to grow them in containers to observe the development of diagnostic features such as shape, size, number of plants cohering, nervation, anthocyanin pigmentation and turions. Some species may exhibit considerable morphological variation, particularly when growing under less than optimal environmental conditions, making their precise vegetative identification very difficult.

Duckweeds are small aquatic herbs floating on or below the surface of quiet streams and ponds, often forming dense, homogeneous clonal populations. The plant body is not differentiated into a stem or leaf. It is reduced to a fleshy or thalluslike ovoid or flattened structure bearing one-several roots (without root hairs) on the underside, or rootless. The terms dorsal and ventral are often used in the literature for the upper and lower surfaces of the plant body floating in water. The terms adaxial and abaxial are typically used for leaves, referring to the surface adjacent to the leaf axil (adaxial) and the opposite surface away from the leaf axil (abaxial). Adaxial and abaxial also refer to the upper and lower sides of a leaf; however, the abaxial side is also the back or dorsal side. This terminology is especially appropriate for leaves arranged vertically on a stem. Since the plant body of a duckweed is not technically a leaf, the terms adaxial and abaxial are confusing for general descriptions. For duckweeds it is preferable to use upper and lower surface. [Thanks to Elena George of Humboldt State University for bringing this to my attention].

The plant body often has one-several layers of conspicuous air spaces (aerenchyma) and one-several veins (nerves). Daughter plants are produced in a budding pouch at the basal end or along the 2 lateral margins of parent plant, often remaining attached to parent plant by a short stipe. Some species produce rootless (or very short-rooted), starch-filled daughter plants, called turions that sink to the bottom and overwinter. Flowers are bisexual and usually protogynous, the androecium consisting of 1 or 2 stamens and the gynoecium consisting of a single pistil. The flowers are produced in a floral cavity on the dorsal surface (Wolffiella and Wolffia), or in a membranous, saclike spathe (utricular scale) within a lateral budding pouch (Spirodela, Landoltia and Lemna). Some authorities consider duckweed species to be monoecious with one or two staminate flowers (each consisting of a single stamen) and a pistillate flower (consisting of a single pistil). There is no corolla or calyx. The ovary is superior and unilocular with a short style and circular concave stigma. The stigma often secretes a fluid droplet at anthesis. The stamen has a short filament and unilocular or bilocular anther, transversely or apically dehiscent, bearing spinulose pollen grains. The fruit an indehiscent, bladderlike utricle containing one-several seeds with prominent operculum.

The traditional duckweed family (lemnaceae) contains 5 genera and at least 38 species. DNA studies indicate that duckweeds are best included within the Araceae. Duckweeds have a worldwide distribution, especially temperate and tropical regions. They are the smallest and structurally simplest of all angiosperms, with greatly reduced vascular tissue (tracheids) limited to the veins of plant body, filaments of stamens, and roots of some species. Duckweeds and associated microfauna are an important food source for certain waterfowl. They are potentially valuable for waste-water reclamation and one species, (Wolffia globosa (Roxb.) Hartog & Plas) known locally as "khai-nam," is eaten by people in S.E. Asia.

Two of the Wolffia species included in Landolt's 1986 Monograph of the Lemnaceae (Vol. 1) have each been split into two species (E. Landolt, 1994, Ber. Geobot. Inst. ETH, Stiftung Rubel 60). The justification for two additional Wolffia species is based on allozyme studies by D.J. Crawford, Columbus, Ohio (Crawford, D.J. & E. Landolt, 1995, Allozyme Diversity Among Species of Wolffia (Lemnaceae), Plant Systematics & Evolution 197: 59-70). South African populations of W. globosa (Roxb.) Hartog & Plas are now recognized as W. cylindracea Hegelm., an older name used in the literature since Hegelmaier (1868). The widespread Asian W. globosa (also from California and southern Florida) has been retained as W. globosa. Populations of W. angusta Landolt in Pakistan and India have been named W. neglecta Landolt. The Malaysian and Australian populations of W. angusta have been retained as W. angusta. In addition, a new species of Wolffiella from the Amazon Basin has been named W. caudata Landolt (E. Landolt, 1992, Ber. Geobot. Inst. ETH, Stiftung Rubel 58). The specific epithet for this latter curious species refers to the tail-like, tapering distal end of the plant body (See WAYNE'S WORD: Weird Duckweeds From Far Away Lands). Another new species of Lemna (L. yungensis) was also described by Landolt from vertical wet rocks of the Andean Yugas in Bolivia (E. Landolt, 1998, Bulletin of the Geobotanical Institute ETH 64). D.H. Les and D.J. Crawford (1999) have proposed the new genus Landoltia containing one species L. punctata, formerly Spirodela punctata. This species is morphologically intermediate between Lemna and Spirodela. According to Les & Crawford, it represents an isolated clade distinct from both Lemna and Spirodela. [Les, D.H. and D.J. Crawford. 1999. "Landoltia (Lemnaceae), A New Genus of Duckweeds." Novon 9: 530-533.] These revisions raise the total worldwide number of taxa in the Lemnaceae to 38 species in five genera.

The duckweed family is well represented in western North America with nearly half of the world's species. The plant body of duckweeds is quite unlike other flowering plants because it does not have stems or leaves. It represents the ultimate in reduction of an entire vascular plant. The terms "frond" and "thallus" are sometimes used in the literature, but these terms are not appropriate because the plant body of duckweeds is not homologous to the fronds of ferns or the body of fungi and algae. Although the body of duckweeds does have paired guard cells and stomata on its upper surface and superficially resembles a leaf (particularly the flattened duckweeds Spirodela, Landoltia and Lemna), it is morphologically and embryonically completely different. In Spirodela, Landoltia and Lemna it is a flattened structure with slender, hairlike roots on the underside. Spirodela and Landoltia are unique among duckweeds because of a minute, membranous scalelike leaf (prophyllum) enveloping the dorsal and ventral surfaces of the basal end. In Spirodela polyrrhiza the prophyllum is visible on young plants (fugacious in older plants) and on overwintering turions. This basal portion and its connecting stalk correspond to a condensed shoot that has become greatly reduced through evolution. Landoltia has a reduced prophyllum that perishes in full grown plants. A prophyllum is lacking in Lemna, Wolffia and Wolffiella. The latter two genera have been reduced through evolution to minute, rootless spheres or flattened ribbons. Wolffia has a minute globose or ovoid body one mm long or less. In Wolffiella the thalluslike body is transparent and flattened, with the free ends often curved downward in the water.

Although all species of Lemna have a basal root sheath near the attachment node, two species in section Alatae (L. aequinoctialis and L. perpusilla) have a distinctive root sheath with 2 lateral wing-like appendages.

Elongated tracts of cells called nerves are present in Lemna, Landoltia and Spirodela. They originate at the node (point of root attachment) and extend through the plant body toward the distal (apical) region. A similar tract of elongated cells (called the costa) can be seen in the triagular budding pouch of Wolffiella. The position of the coasta in relation to the budding pouch is an important characteristic used to separate W. lingulata from W. oblonga. Tracts of elongated cells also extend through the center of the roots of Lemna, Landoltia and Spirodela. Nerves and tracts of elongated cells may serve to transport minerals and sugars, similar to the function of veins. In some species of Lemna, Landoltia and Spirodela, the elongated cells of nerves contain tracheids with ring-shaped or spiral-shaped thickenings in the walls (annular tracheids). These elongated cells are not called veins because the plant bodies of duckweeds are not homologous to leaves.

Different genes within the nucleus and cytoplasmic organelles (chloroplast and mitochondria) can be used to construct phylogenetic trees called cladograms. One gene in the nucleolus codes for the smaller subunit of the ribosome. The gene is called SSU rDNA or small subunit ribosomal DNA. Base sequences from this gene are sometimes used to compare taxa at the species level. Chloroplast DNA, including the protein-coding rbcL gene, is often used at the family level to show the relationships between genera and species within the family. Introns are also used to construct family trees. Introns are sections of messenger RNA that are removed prior to translation at the ribosome.

Most botanists consider the Lemnaceae to be closely related to the arum family (Araceae), and comparative chloroplast DNA studies have confirmed this taxonomic affinity (Duvall, et al. Annals of the Missouri Botanical Garden Vol. 80, 1993). In fact, several authorities have proposed some drastic and significant changes in the classification of many traditional angiosperm families, including the placement of all duckweeds in the Araceae rather than the Lemnaceae. [See: Angiosperm Phylogeny Group. 1998. "An Ordinal Classification For The Families Of Flowering Plants." Annals of the Missouri Botanical Garden 85: 531-553; Judd, W., C. Campbell, T. Kellogg and P. Stevens. 2002. Plant Systematics: A Phylogenetic Approach. Sinauer Associates, Inc., Sunderland, MA. Some of these proposed changes are summarized in an article by E. Dean in Fremontia 30 (2): 3-12, 2003. If accepted by the botanical community, the incorporation of these changes into botany textbooks, floras, checklists and herbarium collections will be a formidable task.

Computer-generated evolutionary trees or cladograms have been used to show the taxonomic relationships of duckweed species within the family. The cladograms are based on thousands of data characters, including morphology, anatomy, flavonoids, allozymes, and DNA sequences from chloroplast genes and introns. The branch (clade) length and position in the tree correspond to the number of character differences between taxa. The characters are numerically weighted according to their evolutionary importance. For example, a root would have a higher value than a papule. Cladograms are generated multiple times, and they don't always come out the same. The term "bootstrapping" refers to a cladogram or phylogenetic tree that comes out the same way out of a total number of times. For example, one thousand cladogram "trees" are generated and the same pattern comes out 900 times. This cladogram would have a bootstrap value of 90 percent. The following cladogram shows all the five genera and 38 species within the duckweed family (Lemnaceae). It was generated from DNA sequences of rbcL genes from all known members of the the family using the computer program PAUP:

Because of their degree of reduction, Landolt (1986) considers the two diminutive genera Wolffia and Wolffiella to be the most recently evolved offshoots in the phylogeny of this family. Wolffia has the fewest shared characters with the presumed ancestral Spirodela and is placed farthest away in an evolutionary tree (cladogram). The new genus Landoltia is morphologically intermediate between Lemna and Spirodela. According to D.H. Les & D.J. Crawford (Novon 9: 530-533, 1999), it represents an isolated clade distinct from both Lemna and Spirodela. DNA comparisons of all members of the Lemnaceae by Les, et al. (Systematic Botany 27 (2): 221-240, 2002) indicate that all five genera represent distinct clades. With the exception of Landoltia and a few changes in sections, the 38 taxa recognized in the study by Les et al. (2002) are remarkably consistent with those recognized as morphologically distinct by Landolt.

Amorphophallus titanum

Most authors now agree that duckweeds are an early offshoot from the aroid linkage (Araceae) and are represented in the fossil record since the late Cretaceous by the genus Limnobiophyllum. Although the latter genus has affinities with water lettuce (Pistia), the oldest fossils attributable to Pistia date back only to late Oligocene/early Miocene. Because of its morphological similarity, the aroid Pistia stratioides has been considered a close relative (cousin) of the Lemnaceae. Morphological analysis of the fossil paleocene aroid Limnobiophyllum scutatum by Stockey et al. (1997) indicates that Lemnaceae plus Pistia form a monophyletic group within the Araceae; however, more recent DNA cladistical analyses have different results. Phylogenetic studies by G.W. Rothwell et al. (2004) and L.I. Cabrera et al. (2008) indicate that Pistia and Lemnaceae belong to distantly related clades, suggesting at least two independent origins of the floating aquatic growth form within the arum family (Araceae).   Cladogram From Cabrera et al. (2008)   More Amorphophallus titanum Images

Cabrera, L.I., Salazar, G.A., Chase, M.W., Mayo, S.J., Bogner, J., and P. Dávila. 2008. "Phylogenetic Relationships of Aroids and Duckweeds (Araceae) Inferred From Coding and Noncoding Plastid DNA." American Journal of Botany 95 (9): 1153-1165. Rothwell, G.W., Van Atta, M.R., Ballard Jr., H.E. and R.A. Stockey. 2004. "Molecular Phylogenetic Relationships among Lemnaceae and Araceae Using the Chloroplast trnL-trnF Intergenic Spacer." Molecular Phylogenetics and Evolution 30: 378-385. Stockey, R. A., Hoffman, G.L., and G. W. Rothwell. 1997. "The Fossil Monocot Limnobiophyllum scutatum: Resolving the Phylogeny of Lemnaceae." American Journal of Botany 84 (3): 355-368.

Many traditional phylogenetic groupings of species within families and genera are not monophyletic and are inconsistent with modern cladistical analyses based on DNA. In other words, the groupings are paraphyletic or polyphyletic, and do not show all species within a group descending from a common ancestor. Monophyly is the natural evolutionary pattern in which all species are descended from a common ancestor. In order to have consistent computer-generated, monophyletic cladograms, it is sometimes necessary to change paraphyletic and polyphyletic groupings by moving species into different genera, and by moving genera into different families. Many of the taxonomic revisions in the Jepson Manual 2nd Edition (2012) were made in order to have consistent monophyletic groupings. This is why Spirodela punctata was placed in the genus Landoltia and why the Lemnaceae was placed in the familiy Araceae.

The cladogram (left) is from D.H. Les and D.J. Crawford (1999). It has high boot strap values and is based on molecular (rbcL) data from chloroplast DNA. It clearly shows that a grouping composed of 3 species of Spirodela is paraphyletic. This is why S. punctata was placed in the monotypic genus Landoltia.

In 1999, D.H. Les and D.J. Crawford proposed the new genus Landoltia containing one species L. punctata, formerly Spirodela punctata. This species is morphologically intermediate between Lemna and Spirodela. According to Les & Crawford, it represents an isolated clade distinct from both Lemna and Spirodela. Without this change, the genus Spirodela would be paraphyletc.

According to Professor Dr. Elias Landolt (personal communication, 2001), the creation of the new genus Landoltia is not necessary based on a purely morphological point of view; however, based on DNA and enzymatic studies, the change is warranted in order to form phylogenetically consistent taxa. The inclusion of a fifth genus Landoltia appears to be necessary based upon a comprehensive analysis of the Lemnaceae by D.H. Les, D.J. Crawford, E. Landolt, J.D. Gabel, and R.T. Kimball (2002). More that 4,700 characters were studied, including data from morphology and anatomy, flavonoids, allozymes, and DNA sequences from chloroplast genes (rbcL, matK) and introns (trnK, rpl16).

The Angiosperm Phylogeny Group (APG) has proposed some significant changes in the classification of many traditional angiosperm families, including the placement of all duckweeds in the Araceae rather than the Lemnaceae. Nomenclatural changes are cited under the APG II system (2003) and superceeded by APG III system (2009). These changes are based on computer-generated evolutionary trees or cladograms. Thousands of data characters have been used, including morphology, anatomy, flavonoids, allozymes, and DNA sequences from chloroplast genes and introns. The Jepson Manual Second Edition (2012) essentially follows the changes summarized in the following reference by W.T. Judd, et al. 2008. Since the genus Landoltia was proposed by D.H. Les and D.J. Cawford in 1999, several classic papers on the phylogeny of the duckweed subfamily (Lemnoideae) and other aroids (Araceae) have used the name Landoltia. In my opinion, the name Landoltia is warranted because it is consistent with the objectives of the Jepson Manual 2nd Edition (2012) based on phylogenetic studies using plastid DNA.

This was Meyer's original name based on the type specimen collected along the Essequibo River, Guyana, South America in 1818. Unfortunately, Meyer's original type specimen was lost.

C.H. Thompson placed this species in the genus Spirodela in 1898. Since the type specimen was lost, he based the new name on a specimen from the 1938-1842 Wilkes Expedition, labeled Orange Harbor, Tierra del Fuego. According to Landolt (1986), Thompson neotypified this species in his 1898 publication.

In 1999, D.H. Les and D.J. Crawford placed this species in the genus Landoltia based on DNA evidence.

Note: This is a complicated taxonomic subject involving many articles from the International Code of Nomenclature For Algae, Fungi, and Plants (Melbourne Code) 2011: Available on-line at: https://www.iapt-taxon.org/nomen/main.php.

An argument for replacing the names Landoltia punctata and Spirodela punctata with the previous name Spirodela oligorrhiza has been made by Daniel B. Ward (2011). In order to make sure we are referring to the same species, Ward has suggested calling this "Lesser Greater Duckweed" to avoid confusing it with the larger species of Spirodela (S. polyrrhiza & S. intermedia) called "Greater Duckweeds." In this article I will simply call it LG Duckweed instead of Lesser Greater Duckweed. Ward's proposal involves the re-neotypification of G.F.W.Meyer's 1818 type specimen named Lemna punctata which was apparently lost. Ward also proposed as the new type a different species that we know today as Spirodela intermedia.

• Ward, D.B. 2011. "Spirodela oligorrhiza (Lemnaceae) is the Correct Name for the Lesser Greater Duckweed." J. Bot. Res. Inst. Texas 5 (1): 197-203.

In July 2012, I received an e-mail message from Dr. Thomas Rosatti, editor of the revised Jepson Manual (2nd Edition), asking my opinion on Ward's retypification. Since C.H. Thompson already neotypified this species as Spirodela punctata in 1898, Ward's retypification should really be a "re-neotypification." Since I wrote the section on duckweeds (subfamily Lemnoideae), adopting Ward's re-neotypification would result in changes to several related species. In July 2012, I stated my opposition to Ward's proposal on my on-line Lemnoideae page on Wayne's Word. I also included a two-paragraph e-mail message from Dr. Elias Landolt, Zurich stating his opposition to the proposed re-neotypification (see below). This quotation can be verified on the Internet Archive Wayback Machine dated 8 September 2012.

Spirodela punctata (Meyer) Thompson was named by C.H. Thompson in 1898 based on a collection from the 1938-1842 Wilkes Expedition, labeled Orange Harbor, Tierra del Fuego. Whether this collection actually came from the tip of South America is debatable. The parenthetical author G.F.W. Meyer described this species earlier as Lemna punctata from a type specimen collected in Guyana, South America in 1818. Unfortunately, Meyer's original type specimen was lost. According to Ward (2011), LG Duckweed does not occur in the areas where these collections were made: The Tierra del Fuego collection was mislabeled and the Guyana collection was not LG Duckweed. Futhermore, he states that the only native Spirodela in South America is S. intermedia. Since Meyer's type specimen was lost, Ward re-neotypified the species as Lemna punctata G.F.W. Meyer and he designated S. intermedia as the type. Thompson's binomial is still Spirodela punctata (Meyer) Thompson; however, this no longer refers to LG Duckweed. It is now the correct binomial for the South American Spirodela intermedia. The correct name for LG Duckweed now becomes Spirodela oligorrhiza (Kurz) Hegelmaier, a name published by Hegelmeier in 1868. Hegelmeier apparently never saw the South American specimens discussed above, so his name is probably based on the true LG Duckweed.

Ward's 2011 neotypification will make Landoltia a synonym of Spirodela and no longer available for the intended LG Duckweed. The restoration of separate generic status for LG Duckweed now known as Spirodela oligorrhiza (Kurz) Hegelm. will require the creation of a new genus name. The binomial Spirodela punctata (Meyer) Thompson will now refer to the South American species known as Spirodela intermedia W. Koch. By neotypification the name Landoltia becomes a synonym of Spirodela intermedia.

According to E. Landolt (Personal Communication, 2012), the name change proposed by Ward is untenable. This quotation can be verified on the Internet Archive Wayback Machine dated 8 September 2012.

"I think this problem cannot be solved definitely. The main problem is the fact that it is not possible to decide which species Meyer was describing under the name of Lemna punctata. Certainly, it has to be a species of the genus Spirodela sensu lato because we don't know any other species within the Lemnoideae with pigment cells ("punctata"). The description of Meyer is very rudimentary. I could not find any herbarium specimen collected by Meyer. His description could concern Spirodela oligorrhiza, Spirodela intermedia or Spirodela polyrrhiza. I collected all of these species in northern South America. The description fits best for Spirodela oligorrhiza because it mentions 2-to 3 roots per frond. Most individuals of S. oligorrhiza in nature have 2 to 5 roots. S. polyrrhiza and S. intermedia mostly have more than 8 roots (up to 18). Only very rarely and only in very young fronds they show less then 5 roots. L. punctata was collected by Meyer in Guyana. On the other side, S. intermedia is known from the neighbouring state Surinam and surely is indigenous in the region. S. polyrrhiza and S. punctata might be introduced to South America. Today, S. punctata is frequent in the regions of Rio and Sao Paulo, in Venezuela, Colombia and Ecuador. I have collected S. polyrrhiza in Colombia and Ecuador. Even if S. punctata is introduced into South America it is not known at which year the introduction took place for the first time. It looks like S. punctata would be easily distributed by ship from harbour to harbour and from there by bird to places within a continent."

"I can understand that Thompson choose a new type for Lemna punctata. The correctness of his decision is not disputed. I checked the neotype the collection of Wilkes from copies in four different Herbara. It is clearly the species which is now called "punctata". It is not important if the material was collected in Orange Harbor or somewhere else. Because it is not possible and will probably never be possible to decide the identity of Lemna punctata with certainity it is not advisable to change the correctly published neotype of Thompson. If we change the type of L. punctata again we will have a terrible chaos in nomenclature. Therefore I am not following the proposal of Ward."

My objection to Ward's proposed neotypification is based on two primary points. (1) He is re-neotypifying Meyer's lost type specimen with the name Lemna punctata; however, he is using Spirodela intermedia as the type. It is impossible to know with 100% certainty which species Meyer was describing under the name Lemna punctata back in 1818. It could have been the "LG Duckweed" that we know as Landoltia punctata (Spirodela punctata = Spirodela oligorrhiza), or it could have been another species of Spirodela such as S. intermedia. Why complicate this taxonomy based on speculation. (2) Cladistical analysis has clearly shown that Spirodela punctata belongs in a separate genus (Landoltia), otherwise the grouping of Spirodela with 3 species is paraphyletic. The trend in modern floras such as the Jepson Manual Second Edition (2012) is for consistent monophyletic groupings.

Ward's re-neotypification of Lemna punctata has been reviewed by J.H. Wiersema of the USDA Agricultural Research Service, National Germplasm Resources Laboratory, Beltsville, Maryland.:

• Wiersema, J. H. (2014), Application of the name Lemna punctata G. Mey., the type of Landoltia Les & D. J. Crawford. Plant Biology. doi: 10.1111/plb.12209.

Here are Dr. Wiersema's conclusions: "The re-neotypification of Lemna punctata G. Mey. by Ward (2011) is to be rejected on the grounds that it has neither been established unequivocally that the previously selected neotype differs taxonomically from the original concept of Meyer, nor that this neotype is in serious conflict with Meyer's protologue. S. punctata and Landoltia punctata, both based on Lemna punctata, remain the correct names in Spirodela or Landoltia for the widespread species sometimes known as S. oligorrhiza and the name S. intermedia remains correct for a related neotropical species."

See Another Taxonomic Controversy Regarding Incorrect Type Specimens

---

6. An Updated Key To The Duckweed Family

The following indented dichotomous key separates the duckweed family into five distinct genera:

A Key To The Genera Of Lemnaceae  1a. Plant body with 1 - several roots.           2a. Root one.....................................................................Lemna           2b. Roots 2 - 12.                 3a. Roots 7 - 12 (or more); plant 10 mm long.........Spirodela                 3b. Roots 2 - 3 (up to 5); plant 3 - 6 mm long..........Landoltia  1b. Plant body without roots.           4a. Plant body flattened; 3 - 10 mm long.......................Wolffiella           4b. Plant body globose-ovoid; 0.6 - 1.2 mm long.........Wolffia

Depending on the genus, daughter plants are produced vegetatively in 2 lateral, flattened, budding pouches (Spirodela, Landoltia & Lemna), a flattened, triangular budding pouch at the basal end (Wolffiella), or a funnel-shaped budding pouch at the basal end (Wolffia). Each plant produces up to a dozen daughter plants during its lifetime of 1-2 (or more) months. The daughter plants repeat the budding history of their clonal parents, resulting in exponential growth. It has been estimated that the Indian Wolffia microscopica (Griff.) Kurz may reproduce by budding every 30 hours under optimal growing conditions. At the end of 4 months this would result in about 1 nonillion plants (1 followed by 30 zeros) occupying a total volume roughly equivalent to the planet earth. This astronomical vegetative growth and the ability of some species to grow in stagnant, polluted water is why some duckweeds are well suited for water reclamation. Some species not only thrive on manure-rich water, but can be fed back to livestock, thus completing the recycling process. In addition, some species (such as Wolffia) are a potential source of food for humans because they contain about 40 percent protein (dry weight) and are equivalent to soybeans in their amino acid content (with high levels of all essential amino acids except methionine).

Although flowers are rarely observed in some species, all duckweeds bloom and reproduce sexually; however, some populations in small ponds may be clones of each other and not able to produce viable seeds. Since the flowers are typically protogynous with the stigma receptive before the anther is mature, the plants must be cross pollinated by genetically different individuals with mature pollen-bearing anthers in synchronization with the receptive stigmas. During the summer months, 2 stamens (androecium) and one pistil (gynoecium), all enclosed in a membranous saclike spathe, appear within budding pouches at the edge of the plant body in Spirodela, Landoltia and Lemna. In Wolffiella and Wolffia, a minute floral cavity develops on the upper side of the plant body containing a single stamen and pistil (not enclosed by a spathe). The tiny bisexual flowers have no sepals or petals, and are barely discernible without magnification. Because of the sweet (sugary) stigmatic secretions and spiny pollen grains (covered with minute protuberances), there is evidence that certain species may be pollinated by insects. In fact, Lemnaceae pollen has been detected on flies, aphids, mites, small spiders, and honey bees on the surface of dense duckweed layers. With floral sex organs projecting from the surface or lateral budding pouches, many duckweed species may be contact-pollinated as flowering individuals bump together or become piled up in windrows along the edges of ponds and lakes.

See contact pollination in Lemna gibba Stigmatic Secretion Of Wolffiella lingulata

---

7. Identification Of Morphologically Similar Species

Lemna minuta vs. L. valdiviana

Since flowers and fruits are rarely observed, most taxonomic keys to the Lemnaceae are based on relatively few diagnostic vegetative characteristics that may vary under different environmental conditions. This often makes precise identification of some species difficult, or in some cases, practically impossible. All North American species have been separated by their flavonoid spot patterns using two-dimensional paper chromatography [see McClure & Alston (1966), Amer. J. Bot. 53: 849-860]. It should be noted that flavonoid chemistry is not always reliable for taxon distinction because chromatographic patterns may be influenced by environmental factors [see Ball, Beal & Flecker (1967), Brittonia 19: 273-279]. In addition, R. Scogin of RSA and J.L. Platt of OSU studied two-dimensional chromatography on clonal populations of Lemna minuta Kunth from San Diego County and came up with patterns identical with McClure & Alston's L. valdiviana Phil. According to Landolt (1987), the original clones of L. valdiviana studied by McClure & Alston may have actually been L. minuta. During the past century, the taxonomy of L. minuta Kunth has been complicated by different names used by different authors. Several of the synonyms commonly found in the literature include L. valdiviana var. minima Hegelm., L. minima Phil. ex Hegelm. and L. minuscula Herter. James L. Reveal (Taxon 19: 328-329, 1990) neotypified the oldest name L. minuta Kunth and cleared up some of the confusion and controversy about this widespread species.

The plant bodies (fronds) of Lemna valdiviana are often connected in clusters of four to seven, and the nerve ("vein") typically extends 3/4 of the distance from the node (point of root attachment) to the apex. The closely related L. minuta has one faint nerve that only extends about 1/2 the distance from the node to the apex. When growing in full sunlight, plant bodies of L. minuta are often only 2 mm long and are connected clusters of two. One of the most difficult duckweeds to identify in the field is the growth form of Lemna minuta found in shady habitats. The plant bodies are often connected in clonal clusters of four and are slightly longer than typical L. minuta growing in full sunlight. The shade form of L. minuta can be separated from L. valdiviana by the extent of the nerve. The obscure nerve of L. minuta only extends about 1/2 the distance from the node to apex.

Veins (Nerves) and Air Spaces

Dorsal view of Lemna validiviana with backlighting, showing extent of nerve in relation to node (point of root attachment) and apex of plant body. The single nerve extends beyond the midpoint to about 3/4 of the distance between the node and apex. The nerve clearly extends beyond the region of air spaces (aerenchyma tissue). These characteristics rule out L. minuta, at least the typical form that grows in full sunlight. In L. minuta, the nerve rarely extends beyond the aerenchyma tissue and only extends about half the distance from the node to apex. These may seem like relatively minor morphological differences, but DNA sequencing studies clearly separate these two closely-related species.

General shape and extent of nerve in Lemna valdiviana compared with L. minuta. Plants of L. valdiviana are connected in clonal clusters of four to seven, while in L. minuta the plants are typically connected in two's. Each daughter plant is connected by a short stalk (stipe).

Note: Sometimes placing difficult species in an observation dish and examining them over several days can be helpful. Digital images can also bring out subtle differences. The following duckweeds were photographed through a dissecting microscope using a Sony digital camera with backlighting:

Three duckweeds from Pinnacles National Monument in central California. A. Lemna minor: Three veins arising from point of root attachment (n), without dorsal row of papules and reddish anthocyanin on ventral side (as in L. turionifera) and without winged root sheath (as in L. aequinoctialis). B. Lemna valdiviana: One faint vein extending more than 3/4 distance from root node (n) to apex (red arrow), plant body very thin and transparent throughout and floating on or just below water surface (slipping under plant bodies of L. minor and L. minuta in an observation dish). C. Lemna minuta: One vein extending less than 2/3 distance from root node (n) to apex, vein not extending beyond region of larger air spaces (red arrow), plant body slightly thicker in middle (not as uniformly thin and transparent as L. valdiviana), small size (only 1-2 mm long) or larger when growing in shade, floating on water surface (not submersed as in L.valdiviana). Photo taken with substage illumination.

Ventral view of Lemna valdiviana showing a single vein that extends 3/4 of the distance between the node (point of root attachment) and apex of the plant body. According to Landolt, this is one of the most reliable characteristics to separate it from L. minuta because of the variability of these two species under different growing conditions. This specimen was placed on a microscope slide with cover slip and photographed through a Bausch & Lomb microscope with a Sony W-300. The aerenchyma tissue shows up better when all the water under the cover slip has allowed to dry. The image was inverted to a "negative" with PhotoShop to show the extent of the vein.

Image Of Typical Lemna valdiviana Growing In The Shade Image Of Typical Lemna minuta Growing In Full Sunlight

Using Dorsal Row Of Papules To Separate Lemna turionifera From L. minor

Another difficult group of duckweeds is Lemna turionifera and L. minor. L. turionifera has three main veins and is superficially similar to L. minor and nongibbous L. gibba. It differs from L. minor and L. gibba in having a row of 3-7 minute papules along the midline of the dorsal surface. It also differs from L. minor by developing reddish anthocyanin on its underside, starting in the region around the root. What really sets this species apart from other duckweeds is the presence of rootless, overwintering turions in the fall months. These are referred to as "winter buds" in the Jepson Manual of California Plants (1996). L. turionifera appears to be more common than L. minor in San Diego County. It generally replaces L. gibba in the higher elevations. Unfortunately, reddish anthocyanin and turions are not always present, so you must rely on the row of papules along the midline of dorsal surface. This can be difficult to see, especially on dried herbarium specimens. Ideally, herbarium specimens should include field notes on the presence of a dorsal row of papules and reddish anthocyanin on the ventral surface. With some practice, these traits can be observed with a hand lens.

Left: Ventral view of Lemna turionifera showing blotches of reddish anthocyanin, especially in the root region. L. minor is typically not suffused with red anthocyanin. Nongibbous L. gibba generally lacks the dorsal row of papules and often develops anthocyanin on its upper side. Right: Dorsal view of L. turionifera showing midline row of minute papules. L. minor typically does not have a distinct row of papules, although it may have minute apical and/or nodal papules.

A. Lemna turionifera from Moose Lake, Minnesota. The plant body has a distinct midline row of dorsal papules and is suffused with reddish anthocyanin. B. Lemna minor (apparently) from Clearwater Lake. It does not have midline row of dorsal papules and does not have reddish anthocyanin. Some plants identified as L. minor had a minute apical papule.

When identifying duckweed species (especially Lemna, Landoltia and Spirodela), it is very important to view the plant bodies with backlighting (substage illumination) in order to see the number and the extent of the nerves. With a good 10x hand lens this can be accomplished by holding the plant body up against the bright sky. Backlighting is also crucial in order to see the tract of elongated cells (costa) in the budding pouch of Wolffiella. The position of the costa within the triangular budding pouch is very important in order to distinguish between W. lingulata and W. oblonga.

In these times of high technology, as botanical research moves toward a molecular emphasis, it is very important to have specimens verified by a taxonomist. It is also imperative to have carefully prepared voucher specimens on file in a nationally recognized herbarium. Modern molecular techniques, such as DNA sequencing, may lead to a better understanding of these fascinating species.

Although some duckweed species superficially resemble each other, they may have significantly different biochemical patterns, such as an entirely different photoperiodism in response to day length (hours of darkness). During the hours of daylight the protein leaf pigment called phytochrome 730 (P-730) is formed. During the hours of darkness P-730 is slowly converted into phytochrome 660 (P-660). In short-day plants P-730 inhibits flowering. Short-day plants typically need about 15 hours of darkness to convert all the P-730 present at sundown into P-660. In these plants, P-660 stimulates the release of the essential flower stimulant "florigen" which induces flowering. The P-660 pigment is very sensitive to specific wavelengths of light, and a flash of light during the 15 hours of darkness can instantaneously convert all the P-660 back into P-730. Lemna aequinoctialis is clearly a short-day plant because it requires 16 hours of darkness (8 hours of light) to flower. The closely related L. perpusilla is also a short-day species that exhibits maximum flowering with 13-18 hours of darkness, and no flowering with 9 hours of darkness (15 hours of light). These species will generally not bloom during the longest days of summer or in a pond next to a bright street light.

Long-day plants require 15 hours of daylight and 9 hours of darkness in order to flower. In these plants P-730 stimulates the release of florigen and subsequent flowering. If the nights are long enough to convert all the P-730 into P-660, no florigen will be released and flowering will not occur. Lemna gibba is a long day plant that flowers with 9 hours of darkness. This species typically flowers during the longest days of summer. It will generally not flower with 12 hours of darkness, such as at the equator or during the vernal equinox, because the nights are too long. The physiology of these long-day and short-day species of duckweeds can definitely affect their range and potential for flowering and seed production.

Exactly how some duckweed species are dispersed and how they survive intermittent streams and ponds that dry up during summer is an enigma. Being carried from pond to pond on the feet of water fowl (tucked neatly under the ducks' bodies during flight), probably explains the distribution of some species. In the southeastern United States there are records of wolffia plant bodies being carried by a tornado, and they have even been reported in the water of melted hailstones! Some species have been carried by rivers and streams, and in the shipment of fish and aquarium cultures. Professor Dr. Elias Landolt (1997) discusses some of the ways duckweeds survive dry conditions (Bulletin of the Geobotanical Institute ETH, Stiftung Rubel 63). Seeds of all Lemnaceae investigated so far tolerate drying for at least a few months to several years; however, seeds are rarely produced by clonal populations of some species. Although vegetative plant bodies are unable to withstand desiccation for more than a few hours, they may survive days (or weeks) embedded in wet mud and debris. According to Dan Richards (The Distributional Ecology Of Duckweeds (Lemnaceae) In Local Populations Of Northern California, MA Thesis, Humboldt State University, 1989), vegetative plants of two species survived up to six hours of desiccation (out of water). The two species tested by Richards (1989), Lemna minor and Landoltia punctata, had a much higher survival percentage when they were in large clumps compared to individually dried plants. Richard's experiments clearly show that these species could easily be carried short distances by migratory water fowl. Species that do not readily form seeds can also survive weeks or months of drought as turions, especially if the turions are imbedded in mud, silt and debris. This is especially true of the minute turions of Wolffia species. According to Landolt (1997), the South African Wolffia cylindracea may survive seasonally dry ponds for at least 16 months if the minute turions are firmly imbedded in clayey soil.

One of the most common questions received at this site is how to control population explosions or "blooms" of duckweeds in which ponds, lakes and reservoirs become covered with a thick green layer of Lemna, Spirodela, Landoltia and Wolffia. Lemnaceae blooms typically occur in waters rich in nutrients, especially phosphorus and/or nitrogen. The nutrients originate from pollution from excessive use of fertilizers or possibly by an imbalance in the populations of fish or water fowl resulting in excessive nitrogenous waste products in the water. The recirculation of nitrogen and phosphorus from the cycle of growth and decomposition of duckweeds may also contribute to the high levels of these elements. Destroying the duckweed layer with herbicides does not solve the problem of excess nutrients in the water. In addition, the chemical herbicides may be toxic to the animal life, either directly or through biological magnification. Because of the exponential growth rate of Lemnaceae, herbicides must be used repeatedly (perhaps several times a year). Ideally, it is best to eliminate the influx of concentrated nitrates and phosphates into the water and avoid the use of concentrated fertilizers.

The manual or mechanical removal of the duckweed cover can also remove a lot of the nitrogen and phosphorus nutrients. The duckweed mats can be composted and used as "green manure." They can also be fed to livestock, rabbits, poultry and fish. It has been estimated that 10 acres of duckweeds could theoretically supply 60 percent of the nutritional needs of 100 dairy cows, the manure of which could be recycled to provide fertilizer for the thriving duckweeds. According to R.M. Harvey and J.L. Fox, 1973 ("Nutrient Removal Using Lemna minor," J. Water Poll. Control Fed. 45: 1928-1938), one hectare of water area is sufficient to raise 4000-7000 chickens and ducks during a vegetation period. And according to E. Rejmankova, 1981. ("On The Production Ecology of Duckweeds," Intern. Workshop on Aquatic Macrophytes, Illmitz, Austria), one hectare of Lemnaceae cover is sufficient to produce protein for 480 ducks during the warm season. The utilization of duckweeds as food for animals is summarized by E. Landolt and R. Kandeler, pages 382-389 in Veroff. Geobot. Inst. ETH, Stiftung Rubel 95 "The Family of Lemnaceae: A Monographic Study" Vol. 2, 1987. An extensive bibliography of Lemnaceae is also given on pages 414-580. The following 3 classic papers discuss duckweed use in aquaculture:

Stopping the inflow of nutrients and the repetitive removal of the duckweed layer will greatly reduce the growth of duckweeds. Since water fowl and most fish feed on the duckweeds, they can help control the exponential population growth of these plants. In addition, Lemnaceae have a positive effect in eutrophic water because they remove ammonia which is toxic to fish in high concentrations.

In general, Lemnaceae are very sensitive to herbicides. In fact, duckweeds are often used to test the toxicity of herbicides and to detect the presence of herbicides in water. According to Professor Dr. E. Landolt (pages 161-170 in Veroff. Geobot. Inst. ETH, Stiftung Rubel 95 "The Family of Lemnaceae: A Monographic Study" Vol. 2, 1987), heterocyclic compounds (e.g. 6-methylpurin), urea derivatives, and quaternary ammonium compounds (e.g. diquat and paraquat) are the most toxic substances for Lemnaceae. Some algicides, including PH 40:62 are extremely toxic to some species of Lemna. Some of these products are available from agricultural supply companies depending on federal, state or local regulations. They should be used with extreme caution and under very careful supervision. It would be advisable to consult with your city or county weed/mosquito abatement department before attempting any large herbicidal control project.

Biological control using ducks, fish, turtles and crustaceans (water shrimp, crayfish, ostracods, freshwater prawns, daphnia, amphipods, etc.) may also help to control duckweed populations. There are a number of species of freshwater fish that eat duckweeds to supplement their diets, including grass carp (Ctenopharyngodon idella), channel catfish (Ictalurus punctatus), common carp (Cyprinus carpio), common mullet (Mugil cephalis), goldfish (Carassius auratus), and Tilapia (Sarotherodon), including S. mossambicus, S. hornorum, and S. nilotica. Duckweeds are also eaten by Pacu (Colossoma bidens), a freshwater fish native to the Amazon River. Some of these fish species may be available through aquafarm distributors or local county and state agencies. One aquaculture company in southern California was raising tilapia for local seafood restaurants.

More Information About Duckweeds For Wastewater Treatment: LEMNA Corporation 1408 Northland Drive Suite 310 St. Paul, Minnesota 55120, USA Phone: (612) 688-0836 FAX: (612) 688-8813 12. The WAYNE'S WORD® Copyright Policy All photo images & illustrations are copyright protected by encrypted watermarks and/or archival original 35 mm color transparencies. The name Wayne's Word and hippo logo are registered with the U.S. Patent & Trademark Office. We have received hundreds of requests from various organizations, agencies and individuals to use photos and information from Wayne's Word on their web sites and in published newsletters, compact discs, books, etc. All the articles, photos and illustrations in Wayne's Word are copyright protected and may not be used in other on-line, CD or printed publications without our expressed written permission. This includes the display of our images and illustrations on other web sites. For anyone interested in this material, you may place hyperlinks (but no frames) to anchor tags on our pages from your web sites. Send An E-Mail Message To Professor Armstrong: Due To The Overwhelming Number Of Messages Sent To WAYNE'S WORD (And Due To A Staff Of Only One Human), Replies May Not Be Forthcoming.

Return To The Lemnaceae Home Page Go To Keys To The Genera Of Lemnaceae Go To The List Of Species Of Lemnaceae Go To List Of .GIF Images Of Lemnaceae Go To The WAYNE'S WORD Home Page Go Back To LIFE SCIENCE HOME PAGE