I 



f you were asked to construct a clock 

 you would probably throw up your 

 hands at the vision of tiny interlocking 

 gears, springs, wheels, and bearings. It 

 might not, however, be such a difficult 

 job. There are many kinds of clocks. 

 All that is required is a regularly spaced 

 sequence of events to mark the passage 

 of time. 



The Romans simply used dripping 

 water. Even today many housewives 

 measure the time required to cook an 

 egg by turning over a pinched glass tube 

 containing fine sand — a three-minute 



TABLE I 



Passage 

 of Time 







1 year 



2 years 



3 years 



4 years 



5 years 



6 years 



7 years 



Green 



Red 



6,400 



3,200 



1,600 



800 



400 



200 



100 



50 





 3,200 

 4,800 

 5,600 

 6,000 

 6,200 

 6,300 

 6,350 



And so on 



Red 

 Divided 

 by Green 







1 



3 



7 



15 



31 



63 



127 



If we know that this process goes on 



A 



World 



of TIME 



Edward J. Olsen 

 Curator, Mineralogy 



version of the ancient hourglass. The 

 picturesque sundial seen in many gar- 

 dens measures time by the movement of 

 the shadow of the fixed arm across the 

 face of the dial. It is, essentially, a clock 

 with no moving parts at all. Finally, 

 the experienced woodsman, prospector, 

 trapper, or farmer can usually tell you 

 the time of day within fifteen minutes 

 by simply seeing the position of the sun 

 in the sky. 



Now let us consider a less obvious kind 

 of clock. Suppose we had a large box 

 with 6,400 green marbles in it. Then 

 imagine that by some process in exactly 

 one year half of the marbles had turned 

 red. This leaves 3,200 green ones and 

 3,200 red ones. Suppose that in one 

 more year half of these remaining green 

 ones become red, leaving 1,600 green 

 and a total of 4,800 red. If the process 

 continues in this manner we may then 

 construct a table. 



with regular precision, we could look at 

 such a box, count the reds and greens 

 and then say how long the marbles had 

 been sitting there. For example, if we 

 found 6,200 red ones and 200 green ones 

 we could say that the process had been 

 going on for five years. In fact, we need 

 not necessarily go through the trouble 

 of counting all the marbles. The right- 

 hand column in the table shows the quo- 

 tient of reds divided by greens. Thus, 

 we need only take out a random sample 

 of a few hundred marbles and count the 

 reds and greens, divide the former by 

 the latter and, if our sample is average, 

 we should obtain a value close to 31 — a 

 time of five years. This process goes on 

 until the last green marble has shifted to 

 a red color. At that time the clock may 

 be considered to have "run down." 



What we have just described, in a 

 fairly simplified form, is the so-called 

 "atomic clock" upon which the much 



publicized methods of radioactive dat- 

 ing are based. Rather than by marbles 

 changing color, the actual atomic clock 

 operates by atoms changing to other 

 atoms. The time required for half the 

 population of atoms of one kind to 

 change to another kind is called the half- 

 life. 



Before going on let's look once more 

 at the box of marbles to clear up another 

 definition. Let us imagine that every 

 time a green marble converts to a red 

 one it gives off a loud clicking sound. 

 During the first year we would observe 

 3,200 clicks, or an average of around 62 

 per week. This is moderately noisy. 

 During the second year, however, there 

 would be only 1,600 clicks, with 31 per 

 week on the average. During the third 

 year there would be only 800 clicks, or 

 about 15 per week; and so on. Thus 

 the rate of noise-making would drop off 

 year by year until it finally stopped. At 

 any time during the life of this clock we 

 would have a definite noise level. This 

 we call the level of activity. In the case 

 of atoms this is called the level of radio- 

 activity. 



So far then we have two methods to 

 measure time. We might, as mentioned 

 before, count a sample of red and green 

 marbles and figure the time from that; 

 or we might simply count the number of 

 clicks per week, or per day, etc., and fig- 

 ure the time from the rate at which they 

 are being produced. In the first method 

 we need not necessarily know how many 

 green marbles were present in the begin- 

 ning since we are only measuring the 

 quotient of reds divided by greens, which 

 will be the same no matter how many 

 greens were there originally (if you don't 

 believe me you might give it a try, start- 

 ing with, say, 10,000 green ones). We 

 need to know only the half-life, which in 

 this example is one year. In the sec- 

 ond method, however, we have to know 

 the original population of greens in or- 

 der to correlate the level of activity with 

 the age of the system. 



i* rom what has been described already 



we can now say something about the 



efficiency, or value, of such atomic clocks. 



(Continued on page 6) 



JULY Page S 



