Supplement to

Measuring the Metabolism of Small Organisms

"With two stoppered test tubes, a capillary tube and a drop of oil you can count the oxygen molecules respired by any small cold-blooded creature. Here's how to watch a beetle breath, and measure the metabolism of many other fascinating forms of life."

by Shawn Carlson

This monograph provides supplemental information to "Measuring the Metabolism of Small Organisms" published in "The Amateur Scientist" Column of Scientific American magazine (December, 1995). You will need to read the Scientific American article before delving into this.

Warburg Apparatus

"The Amateur Scientist" article explained how to construct a simple device using two test tubes and a capillary tube to estimate the number of oxygen molecules consumed and the number of carbon dioxide molecules produced when a beetle breaths. Here's some important additional information that will help you conduct these experiments.

What Fluids Can You Use in the Capillary Tube?

There are two criteria that determine which liquids can be used for the plug. First, the fluid plug must not evaporate. Second, the plug should have as low a viscosity as possible. That way it will slide easily against the glass with little energy being lost to friction.

Lubricating oils work best. Vegetable oil and motor oil work well. General lubricants work well also. Try WD-40. You could also try liquid soap.


How To Center The Droplet In The Capillary Tube

First, thread the capillary tube through the rubber stoppers. Then dip the capillary tube into the fluid. Suck the fluid about 1/4 of the way up the tube. Place a test tube over the end you just dipped and gently grasp the test tube allowing heat from your hand to warm the air. As the air warms, it expands and will push the fluid up into the capillary tube.

Completely coat the inside of the capillary tube by letting the fluid run all the way to the other side of the capillary tube. By adjusting your grip on the test tube you can control how fast the plug moves. When it reaches the end, grip it very lightly so the plug moves very slowly. You want the plug to be just a few millimeters across so drive most of the fluid out through the end, while dabbing the end with a paper towel to catch it. When you've got about 5 millimeters of fluid left in the capillary tube, remove your hand and plunge the test tube into a room temperature water bath. Gently place a test tube over the other stopper. Warm this with your hand and the plug will move away from this end. When the plug is positioned in the center of the capillary tube remove the test tubes.

Move the plug slowly. If you try to position it too fast it will splinter into a number of fragments and you'll have to start all over again.

It's a bit tricky to get this right so you'll have to practice. Because of thermal inertia, the plug will move a bit even after you take your hand off the test tube and so you'll have to learn how to anticipate its motions. It took me about an hour to get the hang of it so don't get discouraged.


Tips-- Things to Do

• Keep lots of spare capillary tubes around. No matter how careful you are, you are going to break them.
  
• Try a number of different diameters of capillary tubes. Experiment. Different tubes will be suited for different conditions. It will take some practical hands-on experience to select the best tube size for your application.
  
• Shore up the inside of the plastic cup with a wooden disk that fits just below the capillary tube. Paint the disk black or cover it with felt. When you plunge the plastic cup into the water bath the water pressure will distort the cup a little and will place stress on the capillary tube. The disk reduces this stress and helps protect it against breakage. It also provides a dark background against which the plug can be clearly seen. (I made this modification after the article was in press, so this tip does not appear there.)
  
• Place both test tubes on the stoppers very carefully and at the same time. When the test tube seals over the stopper, the pressure inside will go up because the volume decreases slightly as the test tube is pushed tight. That change in pressure can be enough to propel the plug right out the other side. You can balance this by placing both test tubes on slowly and at the same time. In fact, with a little practice, you can do it without moving the plug at all.
  
• To remove a test tube from the stopper, pinch the stopper firmly to hold it place and gently twist the test tube as you pull it off. By holding the stopper with your fingers you are reducing the stress felt by the capillary tube.
  
• Tie a small fish weight at the end of the test tubes to compensate for the buoyant force experienced by the test tubes in the water bath. This will also reduce the stress on the capillary tube.
  
  The buoyant force is equal to the volume of the test tube in cubic meters times the density of water in kilograms per cubic meter times g (the acceleration due to gravity = 9.8 meters per square second). This force acts at the center of the test tube and produces a torque that tries to rotate the test tube upward and break the test tube off at the capillary tube.
  
  To compensate, calculate the mass of the water displaced. It's the density of water (1 gram per cubic centimeter) times the number of cubic centimeters of the test tube. Find two fish weights that are about half this weight and attach each to a loop of thread. When you're lowering the test tubes into the water bath, hook the weights over the ends of the test tubes. Since these weights are about twice as far away from the cup, they only need to be half as heavy as the buoyant force to cancel it.
  
  

Limitations

The Warburg device can be used to study insect respiration, but the technique is limited.

You might think that a small capillary tube and small test tube give the best response because a small change in pressure can be most easily seen. You do want the smallest test tubes you can use, but the smaller the capillary tube is, the larger the friction/mass ratios will be in the plug. This friction will prevent the plug from reaching its true equilibrium position. I've conducted trials where the plug remained frozen for several minutes, and then, when the pressure difference built up sufficiently, it suddenly broke loose and blew out all the way through the capillary tube. Even if friction wasn't a problem, for some large insects the plug moves too fast for you to be able to conduct an accurate trial. The point is, you'll have to experiment to find the optimum capillary tube size to go along with your test tube size, choice of oil, physical conditions (like temperature) and the organism you're studying.

Although the friction in the tube is the biggest problem, the technique is also limited by one's ability to measure the distance the droplet travels, and control the time of each trial. It's quite an experimental challenge to control all these factors and get a respiratory quotient that's accurate to within 20 percent. I suspect that 10 percent is about as good as one can do with this method.

Remember, error analysis is very important for any experiment. Every quantity you measure has an associated error. You'll need to know how to handle these errors to pull meaningful results from your data. Make sure you have a good book on the topic by your side when you're struggling to understand what all this means. Get a good book on statistics-- Data Reduction and Error Analysis for the Physical Sciences by Philip R. Bevington, or Statistical Treatment of Experimental Data by Hugh D. Young will get you started. I also strongly recommend The Art of Science, by Joseph J. Carr (ISBN 1-878707-05-1) as an outstanding hands-on introduction to doing science, including simple data analysis. This is probably the best book for you to start with.

Electronic Pressure Sensor

Far more accurate and easy to build is a sensor with an electronic differential pressure transducer. The transducer senses tiny pressure differences between two test tubes and converts it into a voltage that is easily measured with a voltmeter. These sensors aren't too expensive. I picked mine up from Honeywell for just over $100 (#163PC01D36).

Keep in mind that the atmospheric pressure is approximately the equivalent of 32 feet of water. The pressure sensor's highest setting (output about 5 volts) is about five inches of water or just 1.3 percent of atmospheric pressure. A voltmeter can resolve a signal to within one millivolt and the pressure sensor is stable (that is, it's output voltage does not drift more than one millivolt.) That means that there are 5,000 different possible pressures that this device can distinguish between 0 and 0.013 atmospheres. So this device is sensitive to changes in pressure as small as three millionths of one atmosphere! That's an incredibly sensitive device, and it's only $100. (Honeywell recently told me that they are coming out with a new device all integrated on a single chip that will sell for about $40.) It registers pressure differences as small as 0.0003 percent of one atmosphere. (For more information about this sensor, call Honeywell at (800) 537-6945.) With this device, and a little care, you can take data that's good enough to be published in a professional science journal.

The input ports on the sensor are two small protruding cylinders. They are designed to fit inside standard 1/8th inch tubing. They are long enough to support the stoppers, test tubes, NaOH and insects, and they are separated enough so that the test tubes won't touch each other.

Using an electric hand drill, drill an undersized hole through the rubber stoppers. You want the stoppers to fit tightly about the ports. To insure an air-tight seal, put a dollop of aquarium cement to coat the cylinders before you slip over the stoppers.

As pointed out in "The Amateur Scientist" article, you'll need to calibrate this device. The simplest way to do that is to use a U-shaped manometer. The article has a good picture of one. Plot the voltage out vs. height difference. The result should be a straight line. Then, whenever you measure a voltage, you'll instantly know the pressure difference.

The power supply described in "The Amateur Scientist" article is the simplest one that will work. The power supply diagrammed there consists of a 12 AC to DC power adapter. The output is regulated at 12 volts with a 7812 integrated circuit chip. The 7812 chip is extremely common; you can pick one up at any Radio Shack for about a dollar. Most 12 volt adapters provide an unregulated voltage at a slightly higher voltage than specified on the case. That's because the designers anticipate that whatever device they are plugged into will have regulated the power at the specified voltage. For example, most 12 AD to DC adapters in fact provide about 13 volts out. The regulator chip then sets this to a stable 12 volts.

Using the power supply diagrammed in the article, the output will drift by about 10 millivolts. While this drift will not affect your experiments with a single large insect, it will limit your ability to measure the metabolism of small ones. So if you want to do delicate measurements on tiny insects, you'll need to improve on the power supply's performance. You'll need three capacitors; a 4800 microfarad, 0.1 microfarad, and 10 microfarad. Mount the 4800 capacitor first at the far left end of a circuit board. Make sure the positive end is connected to the positive terminal of the 12 volt adapter and the negative side is connected to the other (ground) terminal. Mount the 0.1 microfarad capacitor at least two inches to the right from the larger capacitor. Next, mount the 7812 chip just to the right of the small capacitor as close as you can easily do it. Finally, mount the 10 microfarad capacitor just to the right of the chip. Each element is connected directly to the next in a line; so the large capacitor is connected between the adapter and the small capacitor; the small capacitor leads to the 7812 chip; the chip leads to the 10 microfarad capacitor. The terminals from the 10 microfarad capacitor lead to the pressure sensor. The addition of these capacitors will stabilize the power supply and permit you to make more detailed measurements.

You'll have to hook the power and signal leads up to three tiny wires that protrude from the back of the sensor. I decided to poke the wires through a circuit board and then solder three solid copper wires to them and then feed the power and signals through these wires. Using a hot soldering iron and being careful to minimize the contact time between the iron and the wires, I was able to connect the wires without damaging the sensor. Be very careful! Let the device cool completely before going on to the next wire. At $110, you don't want to damage the sensor. Once the wires are attached you can connect the power and signal wires however you wish. Clip your digital voltmeter to the ground and output leads.

Next, prepare the test tubes and slip them over the stoppers. Submerge the test tubes in a glass of water to keep them at the same temperature. Now you're ready to take data.

Collecting the Insects

You'll want to test this device with the largest beetle you can find. Hunting around your garden or in a park isn't necessarily the most efficient way to find them. It's best if you set several insect traps and let them come to you. The simplest trap I know is to dig a hole deep enough to hold a mayonnaise jar. Push the dirt around the jar so that it comes just up the rim. The jar's mouth should be just a little lower than the surrounding dirt. By doing this you've created a pit that's lined with a glass jar. You'll want to place a shelter over the jar's mouth to shield the inside of the jar from the sun. Otherwise your quarry could succumb to heat exhaustion before you can collect it. I usually stake out a piece of cardboard like a lean-to tent to protect the jar. You can come up with your own method if you wish.

Place something inside that beetles like to eat. Old vegetables work well on some species. Avoid potatoes. They produce a powerful nerve toxin when damaged (as when they begin to rot) and this will injure or kill many species of insects. Cut everything up and wet it with water. You can also use raw or cooked meets. Remember, what you get will depend on what you bait the trap with. You'll have to leave the food out until it begins to spoil. Keep the bait moist so the bacteria can really go to town. That stench of decay, unpleasant to us, rings the dinner bell for insects. They will approach the jar, fall in, and will be unable to get back out. You can keep things fairly sanitary by placing the bait inside a smaller jar or beaker and placing this in the center of your trap. The insects will be attracted by the smell, but will be unable to actually reach the bait. You'll then be able to extract them without touching the bait.

Check your trap three times a day. It's especially important to check it early in the morning so you can rescue the night critters that may be more sensitive to sunlight and heat before they are affected by midday conditions.

Taking the Data

Once you begin your trials, you may be astonished at how quickly the readings on the voltmeter change. It's best to take data in a three-person team, with one person calling out the time, one calling out the data and the third recording the numbers. The time keeper watches a the second hand on a watch and calls out "time" every 10 seconds. The data reader then calls out the number on the voltmeter and the recorder records it. This one-person one-task scheme works quite well for long stretches of data taking.

It is vital that the data reader be scrupulously honest as to what the number is. It's human nature to try and anticipate what the next number will be once one starts to get a feel for how the numbers change. It's possible that the anticipated number will appear on the screen between when the recorder says "time" and the reader reads the number. It's vital that the reader read the actual number when the recorder called "time". Guessing the correct number makes people feel good about themselves and so there is a strong psychological incentive to cheat just a bit and call out the expected number. (Oddly, this is true even when no one else knows that the reader is doing this mental guessing game.) This could skew the results! Remember, be scrupulously honest even if it means proving yourself wrong.

I strongly recommend that the recorder type the numbers straight into a spread-sheet program on a computer. This will make it easy to graph the data (assuming you have a program that graphs data) and see just how the insect is reacting over time. Believe me, you don't want to have to re-enter all this data. It's laborious, time consuming, and prone to entry errors.

You will want to record the data in two columns; a time column on the left and a voltage column just to the right of it. Since the data is taken every ten seconds, the data recorder need not record the time for each voltage reading. The time keeper should say "minute," followed by the number at the start of each minute. For example, "minute 37" for the start of the 37th minute past the hour. The recorder can then mark this position by typing "37" in the time column and then tabbing over to the data column and recording the voltage number called out by the reader. After the trial is over, it's a simple matter to go back and fill in the seconds. Most spread sheets have an option that will let you generate a sequence of numbers down a column. Once you've verified that you have no gaps in your data, you can use this feature to quickly fill in the time column.

If you only have two people in your team, then the data reader can double as time keeper. If you're all by yourself, you can still do it, but you will need to be very dedicated and very careful. It's very taxing to do all three jobs by yourself. You barely have enough time, and so, you'll be putting out 100 percent effort through out the entire trial. You can do it, but I strongly recommend you getting at least one other person to assist.

No matter what, every member of the data-taking team will have to be dedicated. If you're measuring the metabolism of a beetle, you'll need to take data without interruption for at least 45 minutes! It's very difficult to stay focused for that amount of time so be sure you're ready. Every member of the team should have a glass of juice near them with a drinking straw, and take a bathroom break before you start. And make sure everyone in your house knows what you're doing and agrees not to disturb you throughout the experiment. You can't stop and you won't be able to answer questions like "where are my house keys" without loosing valuable data. So be sure you will be absolutely undisturbed.

Note, if you should loose some data for some reason, DO NOT STOP THE TRIAL! A few missed points here and there won't destroy the value of your work. Make sure to record the time at which you resume taking data. So don't get upset if something happens to interrupt the flow of data. Just recover as best you can and keep going.

The best way to take data is to let your PC handle it. Then data taking won't take much of your time. You can set up the trial, come back 45 minutes later and start a new trial. Indeed, with an electronic pressure sensor and a computer taking your data, it will be much easier for you to make original contributions to knowledge. And a computer interface will be useful for virtually every experiment you will ever do! Believe me, one can change your life!

You'll need some software and hardware that allow you to interface the sensor with your computer. There are some inexpensive ways to do it, but they require some sophisticated knowledge of electronics, and are beyond the scope of this monograph. Commercial packages are available, but they tend to be pretty expensive-- costing $1,000 or more. My personal favorite is LabView because there is absolutely nothing it can't do, but it is expensive and can be tricky to use for the uninitiated. I intend to write an article about computer interfaces for "The Amateur Scientist," but it won't appear for a number of months.

General Don'ts

Here are a few things to look out for while conducting your trials:

Hypoxia. The organism's metabolism depends on the amount of oxygen available. If the oxygen content drops by more than 2 percent the organism may be affected. Oxygen makes up about 21 percent of the atmosphere, so don't let the pressure drop by more than about 0.4 percent during your trials.

Exposure to NaOH. While not too caustic, NaOH is caustic enough to warrant care in how you use it. A drop on your clothes will burn through in a day, and any in your eyes will cause you extreme pain and could send you to the hospital. So whenever using the NaOH make sure to use the proper caution. Wear old clothes (preferably a lab coat if you have one), rubber gloves, and a pair a safety goggles. Should you get any in your eyes, flush with water and call a medical hot-line for advice.

Don't expose your test specimens to NaOH either. You must be certain that your test subject can not possibly come in contact with the NaOH. I pack the NaOH very loosely into the bottom of the test tube and place a small dab of paper towel just at the base. The NaOH absorbs moisture from the air and dissolves somewhat. The paper towel is insurance that this material won't leak up into the rest of the test tube. Next, I place in a couple of layers of plastic wire mesh. They isolate the insect and yet allow air to pass to the NaOH freely. Avoid using a large piece of cotton wadding or anything else that might interfere with carbon dioxide molecules that leave the organism from reaching the NaOH.

Suggested Experiments

Here are some ideas for experiments you can do:

Stability of the Respiratory Quotient
Measure the respiratory quotient for a single beetle 30 times. Produce a histogram. Now, measure the respiratory quotient of 30 different beetles and plot a histogram of the result. How do these histograms compare? You'll need to use a statistical test to see if they are really different. Looking at it by eye won't do. You'll want to use the Student's T Test to see if the curves are gaussian ("bell-shaped"). Then see if their means and standard deviations are similar.

Respiratory Quotient Over Time
Does the respiratory quotient change over a beetle's life? To test this you'll want to raise a colony of beetles and track their respiratory quotients over their life time. You can measure the respiratory quotient of all the beetles together to get an average of the colony. What kind of variations do you find? Do you find any evidence of a biological clock that regulates the respiratory quotient?

Respiratory Quotients of Different Species
Compare the respiratory quotients of a number of different species of beetles taken under similar conditions. Can you find general trends in the data? Do large beetles have larger or smaller respiratory quotients than small beetles for instance?

Respiratory Quotient Vs Oxygen Content
Monitor the respiratory quotient of a beetle keeping NaOH in the test tube to absorb the carbon dioxide. Measure the respiratory quotient as the oxygen level drops. What changes do you see?

Respiratory Quotient Vs Time Of Anesthetized Beetles
Drop a little dry ice into a flask with water. The misty white gas that comes off is carbon dioxide. Direct this into the container holding the insects and flood out all the regular air. Measure how the respiratory quotient and oxygen consumption returns to normal after the insects are removed from the container. Run the trials for different exposure times. Also, measure the oxygen consumption and carbon dioxide production separately. How is the recovery time related to exposure time? To temperature? To the insects' size? Do repeated exposures affect the insects' recovery time? Try exposing the insect to other agents. You might experiment with a non-lethal dose of an insect poison.

Oxygen Consumption Vs Temperature
If you place a fish aquarium heater in the water bath and record the temperature with an accurate thermometer you can do all sorts of explorations into how the organism's metabolism is affected as it's temperature rises. Alternatively, by slowly cooling the water--adding a few fragments of ice every so often to the water bath, you can observe the metabolism as the temperature falls. It's easiest to control the temperature if you insulate the water bath. So, if you're using the electronic pressure sensor, perform your experiment inside three nested Styrofoam cups.

How is the respiratory quotient dependent on the insect's diet?
See if your insects can be coerced into changing their diets. Try to get them to eat a diet of pure sugar. Does their metabolism change? If you can't induce them to change their diets, try starving them. Does the respiratory quotient change over time? If so, you will be able to see how the insect draws on different energy stores in its body as a response to starvation. By monitoring the oxygen consumed you'll be able to see how it adjusts it's rate of energy consumption when stressed in this manner.

Other Interesting Organisms
Remember, don't limit yourself to insects. You can use this apparatus to measure the metabolism of any small oxygen consuming organism. Seeds, Molds and Fungus are very interesting kinds of life! We urge you to explore their properties as well.


Whatever experiments you do, please be very careful to be systematic and thorough. Keep a detailed lab book that describes exactly what you did and what you found. When it comes to experimental science, you should always live by the maxim "If it isn't in the log book, it didn't happen." Organize your lab book so that a knowledgeable stranger could read it and understand exactly what you did. Highlight your results.

Please let us know what you find! We'll share the information with insect experts who will advise you about how to develop your work to make original contributions to knowledge. While these experiments are fun, they can also benefit humanity if you are willing to share your results with others. Sharing your discoveries, no matter how obscure they may seem, is a wonderful way to contribute to the betterment of the world..