240 Methods of Election. 



Case of Several Vacancies. 



Hitherto we have supposed that there is only one vacancy 

 to be filled. If there be more than one vacancy we have to 

 settle a most important question before we can consider what 

 method of election is to be adopted. This question is as 

 follows : — Is the majority of the electors to fill the whole of 

 the vacancies, or are the successful candidates supposed to 

 represent the different sections of the electoral body? The 

 first case is that of the selection by a board of governors of 

 officers to fill various offices. No question of representation 

 is involved, but simply the selection of those persons most 

 fit, in the opinion of the whole electoral body, to fill the dif- 

 ferent offices. The second case is that of the selection of re- 

 presentatives by a large electoral body. In the first case the 

 whole electoral body has to decide for itself once for all, and 

 the majority must rule. In the second case the electoral 

 body has to select representatives, who are to decide and act 

 for it in a variety of matters ; and in order that the decision 

 may be as far as possible in accordance with the views of the 

 electoral body, it is necessary that all the different sections 

 thereof should, as far as possible, be represented. 



In the first case there is only one method of arriving at 

 the correct result, and the method is to fill each vacancy 

 separately. Thus one person must be elected by the method 

 described above ; then by means of the same set of voting 

 papers we must proceed to a second election for the next 

 vacancy, and so on till all the vacancies are filled. After 

 each vacancy is filled we must of course suppose the name of 

 the successful candidate erased from all the voting papers. 



The second case — that of the selection of representatives 

 — has been considered by Hare, Andrae, and other writers. 

 It is not proposed here to discuss this question beyond point- 

 ing out that it follows from the principles which have been 

 established in this paper that the process of " elimination" 

 which lias been adopted by all the exponents of Hare's sys- 

 tem is not satisfactory. 



