38 BIOLOGICAL EFFECTS OF RADIATION 



slightly smaller height than that from which it was dropped, we do not 

 think of it as a different ball— we think of it as the same ball with a differ- 

 ent kinetic energy and a different momentum; and it may be expedient to 

 train ourselves to think in the same way of photons. A more serious 

 objection is that this image is necessarily inadequate, because it does not 

 account for the finer details of the process. I will speak of one of these, 

 which gives me opportunity to mention a feature of stationary states and 

 spectra not as yet introduced into this essay. 



To take the simplest case: it is easy to pick out, from among the many 

 stationary states of any atom, a group of three states which have peculiar 

 interrelations. I will denote the members of any such triad by *S, P, D, 

 and take the normal state as S. The peculiar interrelations are these: 

 in the emission spectrum of the gas constituted of these atoms appear the 

 lines PS and DP due to transitions from P to *S and transitions from D to 

 P, and in the absorption spectrum of the cool and rarefied gas appears 

 the line SP, while in the absorption spectrum of the gas when excited by 

 an electric discharge appears the line PD; but the line DS never appears 

 (or perhaps appears only under very abnormal conditions) in the emission 

 spectrum, and the line SD never appears in the absorption spectrum. 

 Direct transitions in both senses occur between S and P, and between 

 P and D; but direct transitions in either sense between S and D do not 

 occur— they are said to be "forbidden." The P state is a sort of neces- 

 sary stepping-stone between the ,S state and the D state; the atom cannot 

 pass between these two without making use of it (or some other Hke it) ; 

 it alone is accessible from both *S and D. 



Now it is a remarkable feature of the Raman effect, that when the 

 incident photons cause the atoms— or rather, the molecules— to make 

 transitions from the normal state to various excited states, the transitions 

 which occur are generally of the class SD\ This suggests that the 

 sentence in the section about scattering without change of frequency, 

 "One may say that the incident photon is absorbed, raises the atom to a 

 'virtual' state, and is instantaneously reemitted," may not be a mere 

 trick of words after all; for perhaps when scattering with change of 

 frequency occurs, it is necessary for the photon to put the atom momen- 

 tarily into a state which shares the character of the P state aforesaid, 

 in that it is a necessary stepping-stone from the initial normal state to 

 the final excited state. If a photon is totally absorbed, the atom goes 

 from the S state to the P state and stays there for the 10-« sec. or there- 

 abouts mentioned previously; but if only part of the energy of the photon 

 is to be absorbed, the atom goes from the >S state to a sort of "virtual P 

 state," and then without pause goes on to the D state. This picture, like 

 all pictures, is still inadequate; but at least it is more nearly adequate 

 than that of the photon merely bouncing off the molecule. (Note that 



