12 INAUGURAL ADDRESS. 



In the meantime let us go back to the movements of molecules 

 and ions in the medium in which they find themselves — pure space 

 in the case of gases, space largely crowded by solvent molecules 

 in the case of any dissolved substance. I have already spoken of 

 chemical changes (the production of new matter from old) as 

 resulting in general from the collision of such molecules or ions 

 with one another ; but as this idea underlies a whole field of 

 experimental chemistry which has been much exploited in recent 

 years, let us examine it a little more closely. 



The production of new matter from old takes time, but some 

 chemical changes are extremely rapid, while others are extremely 

 slow ; and others, again, proceed at such moderate rates as to 

 make it possible for us to observe them quantitatively throughout 

 their course. We may define the rate of anv action as the number 

 of molecules of the new substance produced in unit space in unit 

 time. Now this velocity must depend on the number of effective 

 collisions in unit space in unit time in any case where the formation 

 of a new molecule occurs only as the result of such a collision, and 

 the number of effective collisions must be determined by three 

 factors. These are — (1) the kinds of molecule concerned (that is, 

 the specific characters of the original substances) ; (2) the number 

 of each kind present in unit space (that is, the concentration of 

 that substance) ; and (3) the velocities with which they are moving. 

 Now, when we study the rate of any particular chemical change, 

 the first of these factors is invariable, while if we work at a fixed 

 temperature we render the third factor also constant ; and thus 

 we may readily study the influence of varying concentrations on 

 the rate of progress of the action. Now the character of this 

 influence, or the form of the equation which expresses it, depends 

 on the exact mechanism of the change, as is obvious if one considers 

 the question from the point of view of the chances of an effective 

 collision occurring at any given moment. If a collision of twO' 

 different kinds of molecule is necessary, the velocity of the action 

 is always proportional to the concentration of each of two different 

 substances ; but if the action depends on the collision of two 

 similar molecules, its velocity will vary as the square of the con- 

 centration of a single substance. Such actions are known as 

 bimolecular, and may be distinguished as dissimilar and similar 

 bimolecular actions. But we have also unimolecular actions, where 

 the velocity is proportional to the first power of the concentration 

 of a single substance ; and in such cases we may assume that no 

 actual collision is necessary, but that the molecules concerned 



