In this case the curator's cooperation paid off. In about two weeks the amateur submitted his report: The ancient owner of the bone had died 8,639 +/- 450 years ago.

This figure compared so well with that previously established for the same specimen by Willard F. Libby, who developed the carbon dating method in collaboration with E. C. Anderson and J. R. Arnold, that Bird sent other samples to Highland Park. They were two pieces of cotton cloth from a Peruvian mummy of the Paracas Period. The amateur promptly obtained dates of 2,500 +/- 260 and 2,800 +/- 310 years for the two samples. Libby had measured the ages at 2,190 + 350 and 2,336 +/- 300, respectively. The Museum next sent the amateur undated charcoal samples from a Louisiana site. He got measurements of 2,685 +/- 210 and 2,339 +/- 200. These dates were later supported by studies of the material by professionals at the Lamont Geological Observatory of Columbia University.

The curator decided the amateur should be sent for and looked over. Very likely he had contrived some interesting short-cuts in the very rigorous procedure of carbon dating. When the amateur turned up at the Museum, Bird felt a momentary shock. He was only 15 years old! The boy, Fred Schatzman, had no short-cuts to offer. His samples had been dated the hard way.

That the youth had succeeded brilliantly in a field which defeats some veteran experimenters can be explained in part by the fact that he had learned well how to make a very little go a long way, including the contents of his wallet. The apparatus required for radiocarbon dating is expensive, even if you make most of it yourself, as Schatzman did But even more important is the necessity for economy and skill in handling the material. The substance being measured, a radioactive isotope of carbon, exists in trace amounts too small to be seen. Schatzman's original pound of ancient bone yielded less than an ounce of purified carbon in which the radiocarbon amounted to only one atom in some three trillion atoms of stable carbon. At this ratio 50,000 tons of charred bone would yield just about enough carbon 14 to serve as pigment for the ink consumed in printing one line of text in this magazine. Schatzman is able to measure the minute amount of radiocarbon contained in a fifth of an ounce of purified carbon.

What he is measuring is energy stored in atoms by cosmic rays that struck our atmosphere thousands of years ago. Atomic nuclei from outer space continuously plunge into the upper air at great speed. Such nuclei are called primary -cosmic rays. They collide with and shatter atoms in the air. The resulting debris, somewhat less energetic, flies off in all 2 directions and collides with still-other atoms. Eventually, after a series of these secondary encounters, the initial energy is widely distributed and the velocity of; the rebounding particles drops to a value which depends on the temperature of the gas. Some neutrons in this drifting debris are captured by nitrogen, which has a great affinity for "thermal" neutrons. When stable nitrogen 14 captures, a thermal neutron, it immediately ejects a proton and is transformed into an atom of carbon 14. This radioactive isotope of carbon has a half-life of 5,568 years.

Cosmic-ray physicists estimate that this process produces about; a billion billion carbon-14 atoms per second into our atmosphere. Balancing the rate formation against the rate of disintegration, Libby estimates that the earth's total stockpile of carbon 14 amounts to about 90 tons. This quantity does- not appear to have varied appreciably for many thousands of years.

Where is the radiocarbon stored? The newly formed atoms of radiocarbon in the atmosphere quickly combine with oxygen to form carbon dioxide. On diffusing to the surface, much of this gas is dissolved in the oceans and there reacts to form inorganic carbonates and bicarbonates. Some radiocarbon, both in the air and in water, is ingested byplants and bound in carbon compounds by the process of photosynthesis. Animals eat the plants and also acquire radioactivity. So long as the organisms live, they keep on taking in carbon 14; after death the processes of decay usually return the carbon to the air for another round trip—air to organism and back again. But not all organic material is consumed by decay. In the case of sea shells the carbon may be locked up for impressive periods in the form of calcium carbonate. Charred wood or bone can also resist decay for long periods. Even wood and similar organic substances can be preserved for many centuries if protected from the elements by a dry climate which discourages bacterial invasion.

Libby suggested that carbon 14 would disappear from such materials at precisely the rate at which it disintegrates. This would mean that an ounce of carbon extracted from a quantity of modern material would show just double the radioactivity of a like amount taken from a sample 5,568 years old. Libby's prediction was sustained by a series of elegant experiments concluded seven years ago.

Dating by the carbon-14 method involves two major steps: one chemical, the other electronic. The chemical step is extraction of the carbon from the sample in a highly purified state. It may be reduced either to elemental carbon or to a gas or liquid compound, depending upon the type of instrument to be used in measuring its radioactivity. The earliest measurements were made with a Geiger counter of the ''screen-wall'' type The carbon was extracted as a solid and reduced to powder by grinding. It was then mixed with a little water as a slurry and painted on the inner wall of the counter tube. In another technique, developed later, the specimen is burned to carbon dioxide and the Geiger tube is charged with the gas. A newer method bubbles the carbon dioxide through u reagent and converts it into acetic acid or an aliphatic hydrocarbon such as hexane for measurement by a scintillation counter employing a liquid scintillation cell. The liquid scintillation counter has several advantages over the Geiger counter, including much greater sensitivity. It records nearly every radiocarbon disintegration.

In essence the radiocarbon dating method is simple. You extract the carbon from the specimen, place a known weight of it in the counter, measure the intensity of the radioactivity and compare this with the radioactivity of a sample of the same weight and of known age, usually recent. Calculations based on the difference of radiation intensity between the two give the age of the unknown specimen.

But the method turns out to be astonishingly complicated in practice. Most of the difficulty stems from contaminating radioactivity. The experimenter himself gives forth more radioactivity from his own carbon 14 than does the specimen he is trying to date. A hail of cosmic radiation from all directions continually penetrates the scintillation cell. These sources plus emanations from naturally occurring radioactive elements in the soil and air will add hundreds of counts per minute in the counter unless elaborate precautions are taken. Moreover, the spurious count may be increased to thousands by electrical disturbances in the circuits of the measuring apparatus. The photoelectric cell, for example, is designed around a cathode of an alkali metal, such as cesium, which ejects electrons when bombarded by atomic particles. It is also sensitive to heat. This source alone can give thousands of false counts per minute. In contrast, carbon-14 atoms in a gram of modern carbon disintegrate at the rate of only 15 per minute. Hence the accuracy of the dating method hinges on the effectiveness with which the background disturbances can be suppressed.

The following is young Schatzman's account of his experiments:

"When I decided to go in for radiocarbon dating, I had no apparatus. I was looking for a project to enter in a science fair at school. I liked mathematics, chemistry and electronics. Radiocarbon dating seemed a good bet because it involved all three. I soon learned that counters cost more than I could afford So I made friends with the staff of a local laboratory which had a good gas counter. They agreed to let me use the counter if I would do the chemical separations at home. Even this turned out to be expensive. So I first had to become an amateur glass blower.

"All the archaeological specimens I have come across have been heavily contaminated not only with other radioactive substances, such as radon, but usually with modern radiocarbon as well Chemical processing begins, therefore, with a careful examination of the specimen. Finely divided charcoal from a camp site may contain rootlets, bits of dried grass or mold and so on. This must be cleaned away, usually under a magnifying glass. If you detect evidence of recent decay or other changes which indicate contamination, the affected parts must be cut out and discarded. The cleaned material is then thoroughly washed.

"The sample is next tested in a hydrochloric acid solution. Bubbling indicates the presence of lime. Lime is apt to be present if the specimen has been recovered from a dig situated in a wet or humid region where limestone or chalk, leached away from underground deposits, has collected in the fissures of the specimen. If the slightest trace of bubbling is observed, the sample is treated for several hours in hydrochloric acid, washed until the acid has been fully removed and then dried. It is then stored in a clean, tightly capped glass jar.

"The extraction process begins with controlled burning of the specimen. A measured quantity of the material to be dated is placed in the center of a tube of high-silica glass, such as Vycor, capable of withstanding 2,000 degrees Fahrenheit. The ends of the tube are stuffed with glass wool and closed by stoppers fitted with entrance and exit tubing. The tube is then inserted into an electric furnace made by winding successive layers of Nichrome wire on concentric tubes of asbestos sheeting. The length of the heating unit overhangs the sample by an inch or so. After the furnace has reached a working temperature of about 1,100 degrees, oxygen is admitted to the sample at a pressure of slightly less than one atmosphere, and the combustion gases are passed through a bubbler [see Figure 2]. The carbon dioxide reacts with ammonium hydroxide in the bubbler to form ammonium carbonate.

"The remaining gases, mainly carbon monoxide, are then passed to a second furnace identical with the first. In this step the Vycor tube is packed with copper oxide, which usually consists of short lengths of fine blackened wire. When heated to about 1,250 degrees, the carbon monoxide reacts with the copper oxide to give carbon dioxide and copper. The carbon dioxide is then absorbed in a second ammonium hydroxide bubbler. At the end of the run the bubbler solution contains not only ammonium carbonate but also the oxides of nitrogen, sulfur and assorted 'crud' of incomplete combustion together with radon.

"Purification is accomplished in two succeeding steps. The ammonium carbonate is transferred from the bubblers to a flask and heated on an electric mantle. A solution of calcium chloride is then brought to a boil and added. Calcium carbonate precipitates rapidly. The precipitate is next washed completely free of ammonia, the salts and the radon.

"The carbon dioxide is then reconstituted by transferring the precipitate to a flask and adding hydrochloric acid. The liberated carbon dioxide is led through a liquid air trap for removing the water and then through a 'U' tube packed with a drying agent.

"If a gas-type counter is being used, the dried carbon dioxide can be stored in flasks at this stage to await subsequent measurement. If elementary carbon is desired, however, the oxygen must be removed from the gas. The reduction is made in a furnace much like that used for burning the specimen, except that an iron tube is substituted for the glass. This is packed with three or four ounces of magnesium turnings and about a thirtieth of an ounce of powdered cadmium. After an inlet tube has been sealed into the iron tube, the system is evacuated and tested for leaks. Carbon dioxide at a pressure approaching one atmosphere is then introduced into the furnace, and the entrance end of the iron tube is heated to about 1,200 degrees with a torch. The magnesium (with the cadmium serving as a cataIyst) burns intensely, burning off the oxygen from the carbon dioxide. Carbon, magnesium oxide and unburned magnesium are deposited on the tube The reaction is self-sustaining, and its intensity is controlled by the rate at which carbon dioxide is admitted. The flow of gas must be carefully regulated or the reaction will melt the iron tube. When reduction is complete and the tube cools, the deposited carbon, etc., is scraped off, moistened with a little water and treated with concentrated hydrochloric acid. After standing overnight the contents are brought to a boil and then filtered. The residue consists of acid-soaked carbon. This is washed in boiling water, thoroughly rinsed and finally dried on a hot plate. After drying it is again treated for two hours with hydrochloric acid, then washed and dried as before and transferred to an airtight storage bottle for subsequent measurement.

"I follow the simpler method, working with carbon dioxide rather than carbon The dried carbon dioxide is bubbled through a three-necked flask containing methyl magnesium iodide, with which the gas reacts to form acetic acid as shown [see Figure 3]. The gas enters through one neck, a Stirring rod to agitate the fluid is introduced through u second neck, and unreacted gas escapes through the third neck. The escape neck is fitted with a drying tube to prevent entry of water vapor from the air into the flask. If the chemical steps have been taken with care, the specimen, now in the form of clear acetic acid, is free of all radioactive substances except the original carbon 14. It is bottled promptly to prevent contamination.

"The scintillation cell of my counter consists of a pipe nipple about an inch long and two inches in diameter. The ends are closed by glass windows held in place by a pair of internally threaded bezels. The inner wall of the cell is coated with a special white paint impervious to the scintillation fluid. The fluid itself consists of toluene with four tenths of 1 per cent of diphenylaxazole and 20 parts per million of diphenyloxazolylbenzene, the latter chemical serving to shift the color of the scintillations to the portion of the spectrum to which the cathodes of the phototubes are most sensitive. The cell is filled through a small entrance tube in the side [see drawing in Figure 4]. The scintillation fluid has the property of fluorescing when excited by the emission of radioactive substances as well as by light and other radiations.

"The scintillation cell is shielded from the background by a tube of lead three inches thick. How to lay hands on that much lead had me stumped for a while. Then I learned that a manufacturer of toothpaste tubes lives in our neighborhood. I introduced myself, and when he learned about the project he not only supplied the lead but cast the shield for me. The shield reduces the effect o background radiation by about 80 per cent. The rest, and 'noise' from the electrical system, must be suppressed electronically.

"The scintillation cell is sandwiched between a pair of photoelectric cells the photomultiplier type, consisting of a photocathode plus a self-contained amplifier. Each scintillation triggers a substantial pulse of current, which appears at the output terminal of each photomultiplier. Electrons dislodged from the cathode by heat cause similar pulses. These are minimized by refrigerating the entire pickup assembly–the scintillation cell, the photomultipliers, the preamplifiers and the lead shield. I talked my mother into locating the family's food freezer' in my corner of the basement. The compartment assigned to me holds the pickup assembly at 20 degrees below zero centigrade.

"A number of tricks have been developed for reducing background electronically. One takes advantage of the fact that scintillations created by the disintegration of carbon-14 atoms are fairly uniform in brightness. Flashes touched off by other forms of radiation vary in intensity over a wide range. Only a few of them coincide in intensity with the flashes of the carbon 14. The electrical pulses that appear at the output terminals of the photomultipliers vary in proportion. A circuit has been designed which favors the pulse size of carbon 14. It employs a set of vacuum tubes controlled by the signals from an auxiliary circuit. The size of each pulse is measured as it enters the device and is the routed into a branch circuit called u delay line–a kind of electrical blind alley. In the meantime the charge on the grids of the vacuum tubes is being adjusted in accordance with the measurement so that the tubes will conduct or not conduct according to the size of the pulse. The control action is timed for completion just prior to the instant the pulse returns from its side trip. Only those pulses meeting the specification get through. A significant number initiated by sources other than carbon 14 pass muster, however.

"Most of the remaining background is blocked out by a screening device of a somewhat different sort. The scintillation cell is watched by a pair of photomultipliers. After amplification, the current pulses from the two photomultipliers, screened by the previous device to uniform size, are presented to a 'coincidence' circuit. Essentially this is a pair of electric 'check valves,' or diodes, comprising the input circuit of an amplifier designed to ignore pulses of current below a predetermined minimum size. The pulse from a single photomultiplier is not energetic enough to activate the amplifier. Therefore the circuit is activated only when a pair of pulses from the photomultipliers arrive simultaneously. Thus only pulses generated by flashes in the scintillation counter, where the pair of watching photomultipliers is triggered simultaneously, gain admittance to the electromechanical counter.

"By these various methods I have reduced the spurious counts in my instrument to about 25 per minute. My equipment contains some 80 vacuum tubes and cost somewhat more than $1,000.

"In dating an archaeological specimen the counting equipment is first calibrated. A measured quantity of standard fluid is mixed with the scintillation fluid and put into the cell. The cell, the shield, the photomultipliers and the pre-amplifier assembly are refrigerated overnight. I then operate the counter for 48 hours, take the accumulated reading, reduce it to counts per minute and compute the efficiency. Having established this, I remove the cell, clean it scrupulously and proceed to measure the activity of the actual sample. The difference between the count of modern carbon and that of the specimen provides the basis for computing the unknown date. If the calibration step can be omitted, as is the case when you are working with a series of specimens, you can date an average of about one specimen per week if everything goes well. It rarely does. Without the help and encouragement of a lot of friends in science I would have made slow headway in what has now become a fascinating hobby. In particular I am indebted to J. R. Arnold, who has advised me from the beginning, and to Junius Bird, who supplied the specimens with which I started."

Sydney Chapman, president of the Special Committee for the International Geophysical Year, who last month outlined the program of auroral observation for amateurs, writes that manuals for the project are being prepared. Information concerning various observing techniques will be made available through this department at an early date. Chapman adds the names of a number of regional auroral reporters to the list published last month. The new names are: for Central America: Julian Adem, Instituto de Geofisica, U.N.A.M., Torre de Ciencias, Ber Piso Ciudad Universitaria, D. F., Mexico; for South America: O. Schneider, Paseo Colon 817, Buenos Aires, Argentina; for Great Britain: J. Paton, University, Drummond St., Edinburgh 8, Scotland; for Japan: M. Hurahata, Tokyo Astronomical Observatory, Mitaka, Japan; for New Zealand: I. L. Thomsen, Carter Observatory, Wellington, New Zealand; for India: A. P. Mitra, N.P.L. Building, New Delhi, 12, India; for Germany: G. Lange-Hesse, MaxPlanck Institut, Lindau uber Northeim, Germany.

Bibliography

ELECTRONICS: EXPERIMENTAL TECHNIQUES. William C. Elmore and Matthew Sands. McGraw-Hill Book Company, Inc., 1944.

RADIOCARBON DATING. Willard F. Libby. The University of Chicago Press, 1955.

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.