This appeal is from a judgment holding that the number of write-in votes cast for appellant was insufficient to show that the voters generally had actual knowledge that an election for constable was being held. This was the controlling question in the opinion delivered on the first appeal, wherein most of the pertinent facts are stated. Lisle v. Schooler, Ky., 288 S.W.2d 652.
The appellees here, as the election commissioners of Clark County, were directed to canvass and certify the votes cast in the November 1955 general election for the office of constable in the First Magisterial District: They certified that 570 ballots were cast in the district in which appellant sought election, of which 567 were valid. There were 904 voters registered in the district. In the race for the office of Governor, 554 votes were cast.
From the certification, it appears that 17 voters wrote appellant’s name somewhere on the ballot'and 14 of these were counted as votes cast for him. His name was written at various places on the ballot; that is, 3 under the Democratic emblem, 1 under the Republican emblem, 1 under the Free Citizens emblem, 9 under the Constitutional Amendments, and 3 at unspecified places. No stamp was placed opposite appellant’s name on 3 of the ballots. These were not counted. See KRS 118.280(2). King v. McMahan, 179 Ky. 536, 200 S.W. 956.
A constable had never been elected in the first district, according to the record. Constables for the sixth and seventh districts were elected as the unopposed Democratic nominees at the November 1953 general election. The only contested race in that election was for the office of jailer. In the sixth district, the candidate for constable received 337 votes. The Democratic and Republican candidates for jailer received 420 and 135 votes, respectively, or a total of 555. In the seventh district, the candidate for constable received 916 votes. The same candidates for jailer received 1079 and 83 votes, respectively, or a total of 1162. At the August 1953 primary election, the two candidates for constable received 1 and 3 write-in votes, respectively. At the August 1951 primary election, a write-in candidate for constable received 10 votes, and in the following November election received 791 votes as the unopposed Democratic nominee for the seventh district.
The trial court held that the test directed by this Court did not warrant- a comparison *794of the vote received by appellant with that of other write-in candidates, “but must be of sufficient number in comparison with other duly nominated candidates to show that the people generally knew that an election for constable was being held.” It also held that in applying the test and considering the small percentage of votes received by appellant, less than 3 per centum, the voters did not generally know that an election for such office was being held. Appellant’s complaint was dismissed.
The ballots did not indicate that there was to be an election for the office of constable. No such election had ever been held in the magisterial district in question. At best, only an election could have been held to fill a vacancy for an unexpired term in the office.
Two very important privileges under a democratic form of government are involved. Here, there is a conflict between the right of the voters to cast their votes for a candidate for constable and the right of appellant to be elected to that office. It is of primary importance in the selection of officers by such process that the electorate shall have notice that such a selection is to be made; otherwise, the democratic process will fail. The success of the system is based upon an informed electorate, expressing the will of its majority. The want of notice that an election is to be held may well result in depriving sufficient electors of the opportunity to exercise their franchise to change the result of the election.
The appellant could have avoided any lack of notice of the election sought by him. He could have secured the placing of his name upon the ballot so that the electorate would have received notice of the election. Instead, he chose the method described in the first opinion. We concur in the trial court’s finding that in view of the small number of votes cast for appellant, when compared with the number received by duly nominated candidates in general elections, the electorate generally did not know that an election for the office of constable was to be held. The electors thus were denied the right to exercise their franchise. As between the appellant and the electorate, the appellant must suffer for want of notice of the election.
Judgment affirmed.