Welcome to Environmental Inquiry: authentic scientific research for high school students  
Header component
For Teachers
For Students
Navigation menu separator
Current Location Arrow    Duckweed
   Lettuce Seeds
 Invasive Species
 Long-term Research
Navigation menu separator
 Assessing Toxic Risk
 Invasion Ecology
 Decay and Renewal
 Watershed Dynamics
Navigation menu separator
Navigation menu separator

Analyzing Duckweed Data
This page contains information on how to analyze your duckweed bioassay. It will help you make comparisons to the control, analyze trends, examine variability, and draw conclusions about the toxicity of your tested compound.

Comparison to the Control
The first thing to check is your control (the beakers that contain just spring water and fertilizer solution). The purpose of the control is to identify how well the duckweed will grow under uncontaminated conditions. You can expect the number of fronds to roughly double in the control beakers during the five-day incubation period. If your control plants did not grow much or do not look healthy, something may have gone wrong in your experiment. Perhaps the nutrient solution was too strong or too weak, or the plants were not healthy to begin with. Or maybe a problem developed with the environmental conditions. Did the solutions get too hot, too dry, or contaminated in some way?

Analysis of Trends
Looking at your graph, do you notice any trends? For example, does the toxicity of your test chemical appear to increase as the concentration increases, or does it stay the same from one concentration to the next? Are there any data that don’t seem to make sense? If so, make a note of these and try to think of any possible explanations for why they are different from your expectations.

A Look at Variability
The means for each treatment tell only part of the story. It is also useful to take a look at the individual data points (the number of fronds in each of the three beakers) to get an idea how much variability exists within each treatment. Try graphing individual data points for each treatment. The wider the spread between data points, the greater the variability within that treatment. The more variability there is within each treatment, the less confident you can be that one treatment is different from another, even if the means appear different on your bar graph.

Because of individual differences among organisms, you shouldn’t expect each plant to respond in exactly the same way. However, it is reasonable to expect that the groups of individuals in each treatment will follow predictable trends. Did replicate beakers have similar numbers of duckweed fronds at the end of the five-day growth period? If your data are more variable than you think is reasonable, you could look into the potential sources of this variability. For example, did the plants appear to be healthy at the beginning of your experiment, or were they already stressed? Were the serial dilutions carefully made according to directions? Did one person do all the counting of duckweed fronds, or did two or more people share this task? Based on your experience with this bioassay protocol, what ideas do you have for reducing variability caused by measurement techniques?

Estimating the TC50
The next step in your data analysis is to figure out how to answer the question: How toxic is the solution or sample to the type of organism you tested? In bioassays there are two ways to report results: LC50, the lethal concentration that kills 50% of the test organisms, and TC50, the toxic concentration that causes organisms to grow 50% as well as a control group. In duckweed bioassays, the plants don’t necessarily die— they may just grow more slowly than they would in a less toxic solution. So in this case use the TC50 to represent the concentration at which the duckweed in the treatment grow approximately half as well as those in the control group. Using your graph, you can estimate at what concentration the duckweed has grown roughly half as much as the plants in the control group. If none of your concentrations produce rates that are close to half those of the control, it makes sense to report the TC50 as a range rather than a single number. For example, you might have to say that the TC50 is greater than or less than all the concentrations you tested, or that it lies somewhere between two of your tested concentrations.

Drawing Conclusions about Toxicity
After you have estimated the TC50 for your experiment, you will be able to use this number to make a statement about the toxicity of the substance you were testing. Usually this statement will be something like:

The TC50 for chemical X and duckweed growth is in the range of __ to __.

If you have TC50 values for duckweed exposed to other chemicals, you can use these numbers to rank which chemicals are most toxic to duckweed. For example:

The TC50 for chemical X is a smaller number than the TC50 for chemical Y. This means that chemical X can affect duckweed growth at lower concentrations than chemical Y. Therefore, I conclude that chemical X is more toxic to duckweed growth than chemical Y.

It is important to remember that duckweed bioassays are not designed to help you reach conclusions about toxicity to humans because duckweed plants and humans are likely to respond very differently to chemical exposures. In order to use bioassays to predict toxicity to humans, you would need to use organisms such as laboratory rats that are known to provide a better model of human response to toxic chemicals.


Copyright 2009 Environmental Inquiry, Cornell University and Penn State University
Click here for a printer-friendly page.