578 Sir E. Rutherford on Collision of 



In the same photograph the a particle shows another 

 sudden bend of' 10°'5. In this case, the range of the recoil 

 atom should be only about 1/800 of the a particle and 

 could not be distinguished on the photograph. 



Number of N atoms. 



In a previous paper we have calculated the number and 

 distribution of H atoms produced by a. particles on the 

 assumption that the nuclei may be regarded as point centres 

 of force repelling according to the law of the inverse square. 

 When these calculations are applied to the collision of 

 a particles with nitrogen or oxygen nuclei, the distribution 

 with velocity of the N and atoms is very similar to that 

 for H atoms. We should consequently expect on the simple 

 theory that the number of N and atoms should fall off 

 very rapidly between 7 and cm., and that the number of 

 short-range atoms should greatly preponderate. Quite the 

 contrary is observed in the experiments (fig. 1), where it is 

 seen that the number of scintillations fall off quite gradually 

 with range. 



There seems to be no doubt that the effects produced by 

 the collision of a particles with N and atoms are very 

 similar to those observed in hydrogen. The observations 

 only receive an explanation on the assumption that the N 

 and atoms like the H atoms are thrown forward mainly 

 in the direction of the a particles and, at any rate for swift 

 a particles, the velocities of the recoil atoms are nearly 

 uniform for a given velocity of the a particles. It should 

 be pointed out that the experiments with air and oxygen 

 differ in one respect from those with hydrogen. In the case 

 of air the a particles are completely absorbed in the column 

 of gas, while in the case of hydrogen the stopping power 

 was usually equivalent to less than 1 cm. of air. Con- 

 sequently in the air experiments, the scintillations observed 

 are due to N and atoms which are produced by a particles 

 of all ranges between 7 and cm., and thus have a wide 

 range of velocities. 



A number of experiments were made by the use of absorb- 

 ing-screens of aluminium and gold in order to determine the 

 number of N and atoms produced by a particles of different 

 range. The result as a whole showed that, for example, the 

 number produced in the first 3*5 cm. of the range of 

 the a particle from radium C was greater than in the last 

 3\5 cm., but accurate deductions were vitiated by the lack of 

 uniformity in thickness of the metal films. 



A number of concordant measurements were made to fix 



