36 Proportional Representation. 



in this colony, the successful party had a majority of 9 in 

 the 27 single electorates where contests took place, whilst 

 the voting powers of the two parties in those electorates 

 were approximately in the proportion of 19 to 17. The 

 representatives returned by the larger electorates were very 

 evenly divided between the two parties ; but, in the absence 

 of an analysis of the voting in each electorate similar to 

 that supplied by the scrutineers at Sandhurst and at West 

 Melbourne, it is quite impossible to estimate accurately the 

 relative voting powers of the two parties. So far as any 

 conclusion can be drawn from the published returns, it would 

 appear that the voting powers of the two parties were very 

 evenly balanced. Hence we may infer that the state of 

 representation obtained in the large electorates was much 

 more perfect, taken on the average, than that obtained in 

 the single electorates. There is, however, no reason why 

 this should be so. In fact, the single electorates might have 

 been expected to give the better average result ; for if a 

 given number of representatives have to be elected, the 

 smaller the number of electorates, and the larger the number 

 of representatives returned by each, the greater would the 

 anomalies of representation probably be. In particular, if 

 there were only one electorate, it is highly probable that all 

 the representatives might be returned by one party. 



We may conclude, then, that if the preferential vote were 

 applied to the present electorates of this colony the repre- 

 sentation would be much more perfect than it can be under 

 the present plan ; that it would be still more perfect if the 

 electorates were enlarged, and their number decreased, and 

 that if each electorate returned at least five representa- 

 tives we should have a very fair approximation to propor- 

 tional representation. By limiting the number of repre- 

 sentatives returned by each electorate to five, six, or seven, 

 we should not have any difficulty in filling any vacancies 

 which might arise from time to time ; whereas, if the whole 

 colony were thrown into one electorate, some totally new 

 principle would have to be adopted for dealing with such 

 cases. 



The methods hitherto given by Mr. Hare and others for 

 dealing with the contingent votes are open to certain 

 objections. These may be briefly stated a.s follows : 



(1.) The result of the election depends to a certain extent 

 upon chance. 



