It was, of course, the best news of our lives. And, with both of us being scientists, it wasn't long after sharing a celebratory cup of hot chocolate that we started thinking about the opportunities for discovery that Michelle's pregnancy afforded. I wanted to find out how much entropy our growing baby will add to the universe by the time it is born. I even devised a simple experiment; it called for soaking Michelle in an insulated vat of tepid water repeatedly throughout her pregnancy, each time measuring how long it took the heat from her body to warm it. Alas, Michelle has made it quite clear to me that this fundamental number will have to remain a mystery (and a great amateur science project for another expectant couple to take on).

But my wife was happy to try other experiments with less demanding protocols. Just a few weeks ago I pieced together an electronic stethoscope that can detect all kinds of sounds produced inside the human body. Initially, I hoped to record the baby's heartbeats and movements. But the apparatus worked better than I had anticipated. Michelle and I have now also recorded a myriad of sounds produced by our own hearts, lungs and gastrointestinal tracts--and a few truly odd gurgles that don't seem to be emanating from any particular organ. You, too, may want to listen in on your own body or to record the internal sounds of your favorite cat or dog. Amateur scientists with an interest in marine creatures may want to adapt the apparatus for use underwater as a hydrophone. In each case, an ordinary tape recorder will serve to archive the sounds.

The device combines 19th-century and modern technologies. For more than 150 years, doctors have relied on a trick of geometry, not electronic circuitry, to amplify sounds within the body. Nearly all the sound energy that enters a stethoscope's relatively large chest piece is channeled into a hollow tube, then directed through a headset and finally deposited onto the doctor's eardrums. Focusing sound in this way increases the intensity of the sound by roughly the same ratio as the area of the chest piece to the inside opening of the tube.

You can use the same technique to make a serviceable stethoscope quite easily using any small funnel. Just place the mouth of the funnel against a friend's chest. When you press your ear over the neck of the funnel (something you should do very gently to avoid injuring your eardrum), you will hear your friend's heart and lungs quite clearly. A small length of Tygon tubing, with one end pushed over the neck of the funnel and the other end delicately tucked just inside your ear canal, will let you listen to the noises created within your own body.

Once amplified by a funnel, these sounds can be captured with a small microphone and processed electronically. Today quality microphone transducers cost next to nothing, and sophisticated systems can be built from scratch for less than $20. (You'll need an electret-type condenser element, a low-noise op-amp, some shielded speaker wire and a few garden-variety resistors and capacitors. Die-hard do-it-yourselfers can consult the Society for Amateur Scientists's Web page for details.) But it's much easier simply to purchase a small lavalier microphone (also called a "tie clip" microphone). The Optimus omnidirectional microphone (Radio Shack catalogue number 33-3013), for example, costs less than $25, and it outperformed all but my most extravagant creations. The unit comes ready to be plugged into any conventional tape recorder that is compatible with an 1/8-inch plug. If your tape player has a different size jack, you'll also need to buy an adapter.

You can secure the microphone inside the funnel using a scrap of foam rubber or similar material. Begin by threading the microphone through the neck of the funnel, as depicted in the illustration on the opposite page. I used a short, thin strip of antistatic foam (Radio Shack catalogue number 276-2400) to hold it in place. Wrap the strip around the microphone a few times. Then secure this package into the neck of the funnel so that the microphone rests just at the apex of the cone.

To test the system, press the open end of the funnel firmly against your chest and switch on your recorder. If you're using a stereo tape deck, make sure to turn the volume on your stereo amplifier all the way down. If you try to record heart sounds and to listen to them through speakers at the same time, your entire neighborhood could be treated to an earsplitting sample of audio feedback. I first attempted to avoid this problem by listening through headphones. Big mistake. The microphone was so sensitive it picked up the faint sounds leaking from the headphones and fed them back into the amplifier. The result was an extremely painful high-frequency blast emitted directly into my ears, which abruptly ended the experiment.

Many tape recorders have a volume indicator that shows the amplitude of the signal being recorded. If yours does not have this feature, you'll have to set the overall amplification by adjusting the volume control, recording for a few seconds and then listening to how the newly recorded track sounds. Repeat the procedure until the signal is as loud as possible without being distorted.

Unfortunately, your microphone will not just register the sought-after body sounds; it will also pick up whatever extraneous noises may be polluting your local acoustic environment. To forestall problems, use a simple RC circuit as a low-pass filter to block any signal with a frequency greater than about 800 cycles per second [see diagram]. The filter does not affect most body sounds, but it will help screen out chirping birds, honking horns and young neighbors' stereos. Although a single resistor-capacitor pair works, chaining two such pairs together, as shown, eliminates more noise, especially near the cutoff frequency. Make sure you use a shielded cable and that the electronics are housed in an all-metal and well-grounded project box.

My choice of 800 cycles per second for the cutoff frequency is completely arbitrary. Depending on your application, you may get better results by using a different limit. The cutoff frequency (in cycles per second) for any simple RC filter will just be the reciprocal of the product of the resistance (in ohms), the capacitance (in farads) and 2p (6.28).

Michelle and I have been regularly recording our baby's heartbeats since early July. We have noticed the sound getting steadily louder over the past few months and expect soon to observe the slowing of heart rate that happens as a baby develops. (In the fourth month of pregnancy, a baby's heart will beat typically at about 160 beats per minute; by the ninth month it normally drops below 140 beats per minute.)

Taking time out to listen in on our baby's internal doings has given us a special closeness with our unborn child. The emotion is not unlike that experienced by many scientists, professional and amateur alike, who develop a profound sense of intimacy with whatever they are examining. Often it is this personal connection that pushes such scientists onward in the pursuit of understanding. The motivation to undertake a program of careful observation is, of course, particularly strong when the subject is your own baby girl. (Body sounds don't reveal gender, but a routine ultrasound did.) Baby Katherine Joanne is due November 4.

For information about other activities for amateur scientists, contact the Society for Amateur Scientists, 5600, Post Road, #114-341, East Greenwich, RI 02818 or call 1-401-823-7800.

Images: Bryan Christie

The Society for Amateur Scientists (SAS) is a nonprofit research and educational organization dedicated to helping people enrich their lives by following their passion to take part in scientific adventures of all kinds.