Duckweed Disclaimer
Home Keys To Genera List Of Species Search Lemna Spirodela Wolffia Wolffiella Wayne's Word
Mr. Wolffia's Duckweed Identifications Based On Morphology
© W.P. Armstrong 9 March 2011
Disclaimer Regarding Dried Herbarium Specimens:

In my opinion, it is practically impossible to identify some duckweed species from pressed herbarium specimens, even when they are hydrated. Ideally, you need field observation notes on fresh plants when they are collected using a 10x hand lens. This is particulary true of Lemna minor, L. turionifera and nongibbous L. gibba. For L. turionifera you need to see a row of minute papules along the midline of upper surface. Also if available, evidence of reddish anthocyanin on the lower surface (near root) and overwintering turions (usually formed in the fall). For L. minuta and L. valdiviana you need to see the extent of the midvein. Does it extend 3/4 of the distance between the node (point of root attachment) and apex of the plant body (L. valdiviana) or does it only extend 1/2 the distance (L. minuta). These latter traits are visible in hydrated herbarium specimens if they are pressed adequately.

Left: Wayne P. Armstrong (alias Mr. Wolffia) photographing living specimens of two minute duckweeds, the Asian Wolffia globosa and the Australian Wolffia angusta, through a Bausch & Lomb dissecting microscope (circa 1980).

I became interested in the duckweed family (Lemnaceae = Araceae) in 1980 after discovering several species of Wolffia (the world's smallest flowering plant) in Lake Hodges and the San Dieguito River (below the Lake Hodges Dam). Since that time I have written many articles on the subject and dedicated a web page to the Lemnaceae (originally on the Oregon State University server). Over the years I have had many requests to identify duckweed species for museums, botanical gardens and public agencies. I also wrote the section on Lemnaceae = Araceae for the Jepson Manual: Higher Plants of California (1993) and the 2011 revision of this flora. Many of my original duckweed determinations were made through the kind and generous assistance of Professor Dr. Elias Landolt of the Geobotanical Institute in Zurich, Switzerland. During the 1980s, Dr. Landolt sent me clones of all North American species plus many other Old World & New World species that I carefully grew at my home in San Marcos, California. Without Dr. Landolt's help, I could never have become as knowledgeable about the taxonomy of these remarkable aquatic flowering plants.

Duckweeds are very difficult to identify, particularly pressed, desiccated samples. Unlike most flowering plants, there are seldom any diagnostic reproductive features, such as flowers, fruits or seeds, and identification is based on relatively few vegetative traits that may be influenced by environmental factors. You just can't glance at a herbarium sheet and make quick IDs. Ideally, herbarium specimens should be soaked (hydrated) in a small container of water for several hours. This allows the tissues to soak up (imbibe) water so that some characteristics are discernible under a dissecting microscope. For example, minute papules are difficult to see unless the plant body is hydrated. It is preferable to examine them with background lighting or a white background in order to observe the number of veins, shape of budding pouch, and other diagnostic features. In the case of difficult intermediate duckweeds, such as Lemna minor and nongibbous Lemna gibba, it is preferable to grow live specimens and observe them over a period of several weeks. I also find it very useful to photograph them with a digital camera. Comparing the images can be very helpful in the identification of difficult species, especially when the original specimen has withered away. In fact, sometimes I find differences in the images that I didn't notice under the microscope.

Lighting From Above

Substage Lighting Only
Left Image: Dorsal view of six Wolffia species. 1. W. microscopica (India); 2. W. globosa; 3. W. columbiana; 4. W. brasiliensis; 5. W. borealis; 6. W. arrhiza (Germany). Species 2 through 6 are all known to occur in California.

Right Image: A comparison of Wolffia columbiana (lighter plants) with W. arrhiza (darker plants). Species with dark green dorsal surface, such as W. arrhiza, W. brasiliensis and W. borealis appear darker when viewed with substage illumination only. Transparent green species without dark green dorsal layer, such as W. columbiana and W. globosa appear lighter and more translucent when viewed with substage illumination only.

  Sony W-300 & Bausch & Lomb Microscope  

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. The original image color was inverted using Photoshop CS-1.

Because of the time involved in duckweed identification and space limitations at my home, I prefer to work on no more than 5-6 specimens at a time. As I stated above, dried samples need to be placed in small dishes of water for several hours. In fact, sometimes I soak them overnight to make sure all the tissues are completely hydrated. I may examine and re-examine difficult species for several days. This is when good photo images can really be useful. Because of the time involved, I am hesitant to take on the task of too many duckweed identifications at one time. Depending on the species, I may spend several hours on a single specimen, including photography and writing up my results.

The following two examples illustrate how difficult some duckweeds can be, and how taxonomists can disagree on their precise identification.

1. Confusion Between Lemna minor and L. turionifera

Hydrated Lemna samples from Minnesota: A. Lemna minor? There are no distinct papules along midline of the upper surface. In addition there are no turions and no reddish anthocyanin on the underside of plant body. Most of the herbarium specimens I examined did not show discernible dorsal papules, unless these are all young L. turionifera (collected in June) that do not have well-developed papules. B. Lemna turionifera. The upper surface has a row of very minute papules that are barely discernible on the lower plant body. This sample also shows a slight color difference. According to E. Landolt (1986) Veroff. Geobot. Inst. ETH, Stiftung Rubel 71 "The Family of Lemnaceae: A Monographic Study" (Vol. 1), L. minor is "rather frequent" and L. turionifera is "frequent" in Minnesota. By far, the majority of Lemna samples in Minnesota verified by Dr. Landolt are L. turionifera.

2. Confusion Between Nongibbous Lemna gibba and L. minor

Lemna (SD79566) collected by Reid Moran and Robert F. Thorne at La Encantada in the Sierra San Pedro Mártir, Baja California (19 Aug. 1967). It was originally labeled L. minor and later determined by E. Landolt to be L. gibba. This collection apparently has intermediate forms that resemble L. minor and nongibbous L. gibba. The lack of a row of papules along the midline of dorsal side probably rules out L. turionifera, although the latter species would be expected at this elevation. The air spaces seem too small for typical L. gibba; however, the pattern and shape of air spaces resembles L. gibba. Nongibbous forms of the latter species have air spaces similar in size to L. minor. In addition, Lemna minor sometimes has a slightly more elongate plant body with a length/width ratio greater than 3/2. Plants of L. gibba often have a more orbicular body with an asymmetrical apex. The upper and lower surface may be suffused with reddish anthocyanin, a color trait that is typically not shared by L. minor. The problem is that all of these morphological characteristics, including size, shape, gibbosity and pigmentation, are influenced by environmental conditions. Lemna gibba often grows in high nutrient ponds, including stagnant water with rich, black mud. In San Diego County, I typically find this species in the coastal areas. Without air spaces that are greater than 0.3 mm in diameter or another trait such as reddish anthocyanin, I'm afraid that I am unable to separate borderline duckweeds intermediate between L. minor and nongibbous L. gibba with 100% certainty. The following quote was taken from page 362 of "Morphological Differentiation and Geographical Distribution of the Lemna gibba-Lemna minor Group" by Elias Landolt (Aquatic Botany 1: 345-363, 1975). "... we have to admit, even for specialists, it is often very difficult to determine plants belonging to the group of L. gibba-L. minor. As in other critical species groups, we are able to recognize a given species only if the plants are in a suitable developmental stage and if the range of variation is verified experimentally." This is precisely why I prefer to work with live specimens that I can grow and observe carefully over a period of several weeks.

  Herbarium Specimens From The San Diego Museum of Naural History  

Another example concerns a herbarium specimen labeled L. turionifera that I recently identified as Lemna minor for the Minnesota Department of Natural Resources. I was somewhat reluctant to question another botanist's identification, especially considering the very few observable traits that separate these very similar species. In fact, Dr. Landolt originally classified these species in California as Lemna minor I and Lemna minor II in his 1957 publication "Physiologische und okologische Untersuchungen an Lemnaceen," Ber. Schweiz. Bot. Ges. 67: 271-410. According to Landolt (1986) Veroff. Geobot. Inst. ETH, Stiftung Rubel 71 "The Family of Lemnaceae: A Monographic Study" (Vol. 1), L. turionifera is "frequent" and L. minor is "rather frequent" in Minnesota. He did not report L. gibba from Minnesota, so the problematic nongibbous L. gibba can probably be ruled out. I did not observe distinct papules along midline of the upper surface. In addition there were no turions and no reddish anthocyanin on the underside of the plant body. According to Occam's razor (principle of parsimony), when there are several tentative explanations (hypotheses), the one that makes the fewest new assumptions is probably the best explanation. In other words, do not generate a hypothesis (or plant identification) any more complex than is demanded by the data. Since I did not have compelling evidence to support L. turionifera, I must conclude that this is L. minor, even though L. turionifera is probably more common in colder regions of the northern United States.

  Herbarium Specimens From The Minnesota Department of Natural Resources  

Eventually it may be possible to identify duckweed species using a simple DNA test kit. With the exception of Landoltia and a few changes in several sections, the original 38 taxa recognized in the phylogenetic study by Les et al. (2002) are remarkably consistent with those recognized as morphologically distinct by Landolt (1986).

Left: A cladogram of the subfamily Lemnoideae (Lemnaceae) based on the chloroplast gene rbcL. Five genera and 38 species are shown. According to the cladogram, the ancestral genus is Spirodela and the genusWolffia is placed farthest away because it has the fewest shared characters with Spirodela. Spirodela, Landoltia and Lemna are more closely related, while Wolffia and Wolffiella have more characters in common. With the exception of one new genus Landoltia and a few changes within sections of the subfamily, most of the results are consistent with previous studies based solely on morphological characteristics made by meticulous botanists. Cladogram modified from Les, D.H., Crawford, D.J., Landolt, E., Gabel, J.D. and R.T. Kimball. 2002. "Phylogeny and Systematics of Lemnaceae, the Duckweed Family." Systematic Botany 27 (2): 221-240.

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