264 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1931 



Now the masses of the stars do not show anything like this 

 wide variation. In fact, we can say in a general way that all stars 

 are made of pretty nearly the same amount of matter. And if this 

 is the case, then matter must be very tightly packed in the small 

 stars and spread very thin in the larger stars. If we could dip up 

 with a teacup a sample of the stuff of which the companion of 

 Sirius is made, we should need a powerful derrick to lift it, for at 

 the surface of the earth it would weigh about 10 tons. It is hard 

 to see how matter can be so tightly packed as this. The other extreme 

 is no easier to understand, for an ordinary room full of the average 

 stuff of which Antares is built would weigh hardly more than an 

 ounce and the outer parts of the star must have matter even more 

 rarefied than in the best vacuum we can produce in our own labo- 

 ratories. It is hard to understand how a star built of such thin 

 stuff can be sufficiently opaque to show any definite shape at all, but 

 modern physics has answers ready for both these questions. 



The methods so far described have given us much information 

 about the characteristics of many stars — their distances, sizes, masses, 

 and superficial temperatures. But we are never satisfied. What 

 we should really like to do is to be able to dip a thermometer into the 

 stars and attach to them pressure gauges and read off their tempera- 

 tures and pressures directly. We should also like to attach speed- 

 ometers to stars to find out their velocities. Unfortunately we can 

 not do this. But, as a matter of fact, nature has provided us with 

 instruments which act as excellent thermometers, pressure gauges, 

 and speedometers, if only we can read them. They occur in great 

 numbers throughout the universe. They are atoms in the stars. 



Now it is true that atoms do not carry indicating dials on which 

 we can read off directly what we want to know about the stars; 

 but although very small they are nevertheless remarkable mecha- 

 nisms, extraordinarily sensitive to the conditions in which they find 

 themselves. All the light which comes to us from a star has started 

 from individual atoms, and so the light carries with it a record of the 

 physical conditions under which it was sent out. The atoms in the 

 stars have been broadcasting a description of their surroundings for 

 millions of years, but it is only recently that we have paid the least 

 attention. The spectroscope has made it possible to record these 

 atomic messages, but they are always in code. The study of atoms in 

 our laboratories is giving us the key to this code and is helping us 

 to read the information which is hidden in a stellar spectrum. 



We must not ask to see a correct and up-to-date picture of an atom. 

 That has been impossible since 1925, in which year the atom became 

 a ip function in higher mathematics and quite impossible to draw. It 

 is just as real as it ever was, however, and just as reliable. But for 



