delivered the opinion of the court:
The appellant and appellee were, respectively, the democratic and republican candidates for supervisor at the annual town meeting held in the town of Salem, in Knox county, on April 7, 1908. The canvass of the votes cast showed that appellant had received 189 votes and appellee 184 votes, and appellant was declared elected. Appellee contested the election in the county court, which, after a hearing, declared the election null and void, and this appeal is prosecuted from that order.
The return of the judges shows that 378 ballots were cast, of which five were not voted for supervisor. On the hearing it was agreed that six of the ballots should not be counted for either candidate for supervisor, that 186 should be counted for appellee and 183 for appellant. The remaining three ballots, when produced at the hearing, were each marked in the square before the name of each candidate for supervisor. They could therefore be counted for neither, and if the count of the ballots ought to prevail, the appellee should have been declared elected. It is insisted, however, by the appellant that the evidence shows that the ballots were not so kept after the election as to make it certain that their integrity has been preserved or to entitle them to prevail over the return made by the judges. Un*332der the statute the ballots were admissible in evidence in any event, but their probative force depends upon the care with which they have been preserved. If it is shown by satisfactory evidence that they are in the identical condition at the trial in which they were when canvassed, they are the best evidence of the result of the election and the count of them will be conclusive. If, however, they have been so carelessly kept as to have been exposed to change, their effect as evidence is destroyed. It is not incumbent upon the candidate declared elected to show that the ballots offered to impeach his title have been changed. The contestant is the moving party, and the burden is upon him to show that the ballots are those cast at the election and that they are in the same condition as when cast. Unless their preservation is shown to have been such that there has been no reasonable opportunity for tampering with them they cannot overcome the returns. Jeter v. Headley, 186 Ill. 34; Bonney v. Finch, 180 id. 133; Eggers v. Fox, 177 id. 185; Beall v. Albert, 159 id. 127; Murphy v. Battle, 155 id. 182.
In this case the ballots, after being counted, were strung upon a wire, which was then wrapped around them and sealed with wax, in which the wire was imbedded. The bundle was then placed in a manila envelope, which was sealed by means only of the mucilage on the flap and was delivered to the town clerk, who the same evening placed it in a wooden cupboard about five feet wide, six feet high and fourteen inches deep, in the town hall, where the election was held. This cupboard was used for keeping township papers and had two doors, one of which was fastened by a lock and the other by two buttons- on the inside of the door. Between the two doors was an upright piece four inches wide. No further attention was paid to the ballots until eleven weeks later, when the town clerk took them out of the cupboard to take them to Galesburg to be used as evidence on the hearing of this case. It was then discovered that the flap of the envelope had sprung open and *333was half torn off. The ballots were in the envelope but there was nothing to interfere with taking them out. They were still on the wire, which was sealed, but the corners could readily be turned back so as to expose the circle at the head of the tickets and the squares before the names of appellant and appellee. Many of the ballots could easily have been marked in these squares or circles without disturbing them on the wire. The room in which this cupboard sat was the town hall, which was used for many public purposes, to which access was easy, to which several persons had keys and the doors of which were not infrequently unlocked. During the trial of the case one of the witnesses several times opened the door which was fastened with buttons and closed and fastened it again within two minutes by inserting an ordinary pocket knife between the door and the jamb and turning the buttons with it.
The three ballots which are marked in the square before the name of each candidate for supervisor have been certified to us. An inspection of them throws little light on the question, although they are so plainly marked for both candidates that it seems very improbable that with so small a ballot, containing so few names, they could have been honestly counted for either candidate. All of the judges testify that there was no ballot having a cross in the square opposite the name of each candidate for supervisor, and that these threeo ballots are not now marked the way they were when they were counted. In view of the manner in which the ballots were kept and of the fact that with little difficulty any person desiring to do so could obtain access to the ballots and readily mark enough of them to change the result of the election with little fear of detection, we are constrained to hold that the probative force of the ballots has been destroyed.
The appellee insists that the returns of the election are also discredited by the carelessness of the judges in the performance of their duties, and the court found that both the *334returns and the ballots were discredited as evidence in the case. There is no evidence impeaching the good faith of the judges or tending to show any misconduct, fraud or unfairness. The only evidence of mistake in the counting 'of the vote is that one ballot marked with a straight line instead of a cross in the square was counted for appellant and another similarly marked was counted for the appellee. These were mistakes of judgment as to the effect of a particular manner of marking the ballot. They indicate a lack of knowledge which is certainly not culpable, but do not show any carelessness or neglect of duty in canvassing the vote or declaring its result. They do not affect the result, they do not indicate any unfairness in the count and they do not impeach the fairness and credibility of the returns. The correction of the count in accordance with the evidence does not change the relative position of the parties but leaves appellant with his original majority of five.
In cases where the evidence shows a failure of duty both upon the part of the judges of election and the custodian of the ballots, so that neither the return of the judges nor the re-count of the ballots can be allowed to prevail over the other, the result must be determined from a consideration of both, with all the facts and circumstances surrounding the case. (Catron v. Craw, 164 Ill. 20; Dooley v. VanHohenstein, 170 id. 630; Caldwell v. McElvain, 184 id. 552; Collier v. Anlicker, 189 id. 34; Smith v. Reid, 223 id. 493.) In this case, however, there is nothing which impeaches the count of the ballots as shown by the returns except the re-count made in court. There is no evidence of carelessness, fraud or misconduct in the conduct of the election, in the counting of the vote, proclaiming the result or making the return. It is not until after these things were all done, properly and fairly so far as the evidence discloses, that any neglect of duty on the part of the judges appears. The ballots were sealed up and delivered to the *335town clerk,.the proper officer. The sealing appears to have been carelessly done, for apparently the mucilage did not hold and.the envelope readily came open. Canvas bags and sealing wax were not used but there is no circumstance which indicates any wrong intention, and, in spite of the insecure fastening, if the package had been safely kept as the statute requires, in a secure place, the integrity of the ballots could not have been questioned. When the returns and ballots left the hands of the judges no reason existed for discrediting either. The failure to safely keep the ballots has discredited them, but the evidence did not justify the court in disregarding the return. On the whole evidence we think the finding should be in favor of the count as shown by the return.
Appellee claims that five illegal votes were cast for appellant. One of them is the vote of Samuel Gullett. He was an old man, seventy-seven years old, and a resident and householder of Salem township until the death of his wife, in February, 1907. In the summer of 1907 he sold his house in Douglas, in Salem township, and afterward spent some time in Peoria, in Kansas and in the State of Washington, with various relatives. He was occasionally in.Douglas, when he stayed with George Beck, whose wife was his step-daughter, and there is no evidence of any intention on his part to change his residence from Douglas. He was a legal voter.
It is unnecessary to consider the legality of the other votes which are questioned. If all are rejected appellant still has a majority of one.
The judgment of the county court will be reversed and the cause remanded, with directions to enter a judgment in favor of appellant.
Reversed and remanded, with directions.