The record made in this case is very voluminous, only a small part thereof being printed, and its examination is attended with correspond*48ing labor, in order to present the various subjects of complaint by the parties. In oral argument and in briefs cotinsel have aided us in eliminating the immaterial and non-essential matters put in controversy in the trial court, and our aim now is, to settle the substantial questions found in the record, the proper disposition of which will answer all the purposes of this review.
As in all cases, where several weeks have been consumed in the trial of an important case, irregularities, inaccuracies in rulings of the court, and even slight errors of judgment committed during a heated contest can be found, but if they have been cured by the court in later rulings, or if they are innocent of any controlling influence on the main issue, they should not be seriously regarded by a reviewing court. And so with the record before us.
The circuit court found but few errors in the record, and has stated them, and we must assume from the entry that they declined to reverse the judgment of the court of common pleas for any others assigned in that court. The state urges that the circuit court erred in finding error as it states on the record, and that its judgment should be reversed and that of the common pleas affirmed and enforced. On the other hand, Dickerson urges that the circuit court should have found error in a number of other particulars, and that if it' was mistaken as to the grounds sustained, there are many other grounds of reversal upon which the judgment may be predicated. These grounds, the defendant in error has pointed out *49to us, and they will be considered in their proper order.
We will first notice the grounds of reversal found by the circuit court. It concluded that the trial court erred in permitting witness Trego to testify as to what Mrs. Hughes (the deceased) told him. An examination of the evidence of Mr. Trego shows that for some time before the homicide he was a justice of the peace in that township, and that Dickerson went to the residence of Trego and inquired if Mrs. Hughes had been to see him, the justice, “for law.” Trego informed him that he was not at home when she was there, and Dickerson remarked: “You need not give her law on what she was after.” Mr. Trego did not testify to any conversation with Mrs. Hughes; for he had none, being from home when she called. We surmise that Dickerson had heard in some way that Mrs. Hughes would consult Justice Trego about his conduct towards her; or, that she had consulted him on the subject. Hence, the visit of Dickerson to ascertain what she had done. We see no impropriety in permitting the justice to tell of this visit and what Dickerson said while there. This was but an item of evidence which might be considered with other conduct and statements of the accused relating to any trouble or hostile feeling between them.
Again, the circuit court held that the common pleas erred in “permitting Elijah Thomas to testify that Mrs. Hughes told him about Dickerson hiding in the bushes.” The residence of' Mrs. Hughes was not far from that of Dickerson and his family, and Mr. Thomas was the father of *50Mrs. Hughes and lived with her and her husband. It is intimated' that some time prior to the homicide, Dickerson had annoyed Mrs. Hughes with improper solicitations, or attentions, which were resented by her. On the occasion referred to, the witness, Elijah Thomas, in answer to an inquiry, stated what Mrs. Hughes said about seeing Dickerson hiding in the bushes near the Hughes residence. But, as we read the record, the court promptly and distinctly ruled that part of the answer from the jury and gave mandatory instruction that such evidence must not be considered. On page 66 of the portion of record printed, the trial court said to witness Thomas: “What Kate (Mrs. Hughes) told you is stricken out, and also that 'Dickerson was down hid in the brier patch’; that is struck out — what the witness said about that, and the jury will disregard those statements that I have struck out.” The court intervened soon as practicable to prevent the incompetent answer being made, and after it was made, promptly exercised the right to strike it from the consideration of the jury. These two particular instances of alleged incompetent testimony are not well grounded and were not sufficient to justify the reversing judgment.
The lower court also found error in the latitude allowed the state in the cross-examination of defendant’s character witnesses. Many witnesses were examined and cross-examined on this subject, and a large part of the vast record is taken up with the questions and answers. We can only select some fair samples from the record. *51which will be sufficient to show the general field of controversy.
G. F. McConnel, perhaps the first character witness called by the accused, testified to an acquaintance with Dickerson of twenty-five years; that he had frequently met him, sometimes in public gatherings and places. Objections to several questions propounded to this witness were sustained, because of either the form of the question or a lack of sufficient knowledge to answer as to character of the accused. Efforts to qualify witness were pursued, and finally the following question was permitted and answered:
Q. “During the last fifteen years and prior to June 28, 1905, have you been at places, gatherings, or in company with Benjamin Dickerson, or not?” Objection overruled and exception. A. “I have.”
Q. “From your own knowledge of Benjamin Dickerson, gained by means of associating and seeing him, as you have answered in your former answers, what, so far as your knowledge, is his general character prior to June 28, 1905, for peace and quietness?” A. “Well, it is good.”
Q. “Good as men in general?” A. “Yes, sir, I believe it is.”
The witness -was then taken on cross-examination as to the extent of the acquaintance, and the state asked witness about certain reports in the neighborhood concerning evil things that the defendant had done. Here is an instance:
Q. “You know A. L. Reed?” A. “Yes, sir.”
Q. “You heard that he gave away a bird dog to keep Ben from stealing the dog, did you not, *52prior to June 28th?” A. “I have no knowledge of ever hearing of it, of Al. Reed giving a dog to keep him from stealing it.”
Q. “You heard prior to June 28, 1905, that Ben Dickerson and his brother Josh were accused of killing some sheep and injuring some sheep of John Dickerson about twenty years ago, didn’t you?” A. “No, sir.”'
Q. “Didn’t you hear that Ben and Josh were both blamed with that prior to June 28th, and arrested, brought into town?” A. “No, sir.”
Q. “Did you hear that — did you ever hear prior to June 28 that Ben was accused along with his brother Josh, or accused along by himself, of putting arsenic or strychnine in a crock of milk and that when it was churned by John Dickerson and given to some hogs that it killed several hogs when they gave them the buttermilk?” A. “No, sir.”
These inquiries were followed up by others'of like nature, and all were objected to by the accused, and the court allowed them to go on as before. The witness was'also interrogated about hearing of the defendant injuring a sawmill of Frank Hardesty, or throwing away some of its tools. Other witnesses for defendant, who gave him a good character, were asked about reports of the misdeeds of defendant above referred to, and neighborhood rumors about Dickerson’s conduct. Some of these witnesses had heard of the suggested rumors of misconduct, and they were plied with questions as to whether they took them into consideration in making their answers to the general good character of the defendant for *53peace and quietness, and whether .such rumors would affect their testimony concerning his character. These rumors mentioned by counsel for the state were of various kinds and did not embrace in most instances the trait of character involved in a trial for homicide. We have not space for more extended quotation from the several cross-examinations of the so-called character witnesses.
It will be observed that the defense did not introduce. evidence of his general reputation for peace and a quiet disposition, and counsel for the state seems to think that that was the .only proper course of inquiry, and seeks to justify the cross-examinations by applying them to the course of inquiry which the defense should have adopted, namely, to establish general reputation.
This brings up the question: may the accused, under a charge of homicide, introduce evidence of his good character as to the trait involved in the charge ?
There is no doubt that counsel and even courts have sometimes forgetfully treated character and reputation as synonymous, and a want of a clear knowledge or recollection of the difference' between the two often leads to mistake in applying correct rules to regulate the admission of evidence of one or the other. Character of a person is that which he really is, rather than what he is reputed to be, and we think the accused is not confined to his reputation for a certain trait of character involved in the charge, but may, by those most intimate with him during a course of years, spread before the jury his real self, touching the *54quality of conduct involved in the issue. Such familiar and intimate acquaintance may enable his neighbors to read him as they would a familiar book.
If we look to the text-books we see the right of introducing proof of good character recognized. See Section 25, third volume of Greenleaf on Evidence. There it is said: “Upon the admissibility of evidence of character, whether of the prisoner or of the party on whom the crime is alleged to have been committed, there has been some fluctuation of opinion. Evidence of the prisoner’s good character was formerly held to be admissible, in favor-em vitae, in all cases of treason and felony; but this reason is now no longer given, the true question being, whether the character is in issue. T can not, in principle,’ said Mr. Justice Patterson, ‘make any distinction between evidence of facts and evidence of character. The latter is equally laid before the jury as the former, as being relevant to the question of guilty or not guilty. The object of laying it before the jury is to induce them to believe, from the improbability that a person of good character should have conducted himself as alleged,’ ” etc.
See also Section 3039, Elliott on Evidence, where it is said: “It is the general rule that the defendant’s good character or reputation for peace and quiet is admissible in his favor. But the evidence must be confined, in general, to the trait involved in the crime charged,” etc. Also Section 2721 from the same authority. Further citations to text-books is not necessary, for this court has long ago considered and passed upon the ques*55tion. In Gandolfo v. The State, 11 Ohio St. 114, we have the following as' the third section of the syllabus:
“A defendant who, by the nature of the issue, is entitled to give evidence of his character for peace and quietness, is not limited to proving what people may have said of him, as to his being or not being a quiet and peaceable man, but is entitled to inquire as to his character from those acquainted with him, and they are authorized to speak from his general peaceable and quiet conduct, and from not having known or heard anything to the contrary.” See also Griffin v. The State, 14 Ohio St., 55, 56.
It seems from the nature and form of the questions propounded by the defendant to his character witnesses, that he followed the light of the above case, and the trial court should have recognized it much earlier in the progress of the trial.
But the cross-examinations were conducted as if these witnesses had been testifying to the general reputation of the accused, and they were asked concerning alleged rumors and reports in the community about this and that misconduct, or criminal act of the accused. This course of procedure was strenuously objected to at every step. Then what followed? The state having thrust these rumored offenses into the cross-examination to the extent of page after page, the accused or his counsel thought it necessary to refute the insinuations, and he was called to the stand and interrogated as to the truth or falsity of all of these various rumors. Thus was the record expanded with immaterial and incompetent testi*56mony, which could only tend to cloud the real issue and confuse the jury. The circuit-court correctly held against such lines of cross-examination, and that the error in this regard was ground of reversal.
It is urged by defendant in error that the circuit court should have found further grounds of reversal in addition to those stated in the entry, and he had his exceptions to not so finding entered of record. As one of.the additional grounds, it is said there was no proof of the venue of the crime, and what is claimed to be the only evidence on the subject is copied in one of the briefs for defendant in error. In answer to direct questions, the coronet: testified for the state that he found the dead body in a certain township and that the township is in Coshocton county. But he was not asked to state, nor did he say that Coshocton county is in the state of Ohio. We are not disposed to encourage the lax method of establishing the venue adopted in this case, but from the evidence set out and the main facts as to location and description contained in other parts of the record, we think the point is not well taken. It clearly appears from all the evidence that the criminal transaction occurred in the state of Ohio. The venue need not be proved in express terms, where the evidence is such in the state’s case, that no other inference can be reasonably drawn by the jury. Tinney v. State, 111 Ala., 74.
Another complaint of the trial court relates to the admission of evidence that about two weeks before the death of Mrs. Hughes, the house in *57which she was residing was burned, and that evidence was admitted tending to prove that it was set on fire by Dickerson. The fire occurred in the forenoon when all the family were absent. Mrs. Hughes had gone with one of her children to the house or spring of a neighbor to get water. Her father was in sight of the house and so, perhaps, was Mrs. Hughes. Their attention was called to the fact that their house was on fire. Dickerson was not seen' at or about the house, but one or more testified to seeing him coming from that direction, and looking back as he crossed a fence. The fact was also introduced that he did not go to the fire after its discovery and assist in trying to save the building and its contents, as did other neighbors. On the evidence in the record it would be difficult to convict Dickerson of the incendiary act, but the court admitted it on the theory, it is supposed, that the fire might have furnished a motive for committing the homicide.
The house did not belong to Mrs. Hughes, but belonged to her father, Elijah Thomas, although occupied by both families; and the question is, should the state be permitted to introduce the evidence tending to prove the arson on the issue joined on either count of the indictment? It is argued that the evidence was competent under the first count wherein it is charged thé killing was done with a blunt instrument and other violent means, and that the acts which produced death were accompanied with deliberate and premeditated malice and purpose to kill. It is not seriously claimed that such evidence was competent *58under the second or third counts, inasmuch as the killing is alleged to have occurred while perpetrating or attempting to perpetrate a rape. Was the evidence competent under the first count? If not, it could not be under either of the others. To justify its admission to show or reflect on the question of motive, it has been intimated that Dickerson knew that Mrs. Hughes saw him set the fire, or, that he was near by when it was discovered; or, that he in some way heard that she saw him in close proximity to the fire, and that in either case it became an object for him to destroy her evidence by destroying her. We find nothing in the record to establish by competent evidence, that Dickerson knew that Mrs. Hughes saw him in guilty connection with the firing of the house. Whatever evidence there is relates as forcibly to Thomas and several others who saw as much as she did, and there appears no good reason why the motive to suppress or destroy testimony should apply to her more particularly than to the others who knew fully as much. On a retrial, the state may be able to clear up this situation, but it is not satisfactory to us, as the record now stands.
The state can not give evidence to prove other crimes in support of the charge in the indictment, nor in chief, as reflecting on the character of the prisoner. In Barton v. The State, 18 Ohio, 221, it is held, that on the trial of one for larceny, it is error to admit evidence for the purpose of proving that just before the defendant committed the act for which he is being tried, he committed *59another larceny. The reasons for the holding have some weight on the question here.
To the same effect is Coble v. The State, 31 Ohio St., 100. True, the former assaults introduced in the last case had not been made upon the person who was the victim of the .assault under investigation; but they were claimed to be crimes of like character, but the court held their proof was inadmissible.
There are some offenses in which scienter or g'uilty knowledge may be material, such as passing forged paper or counterfeit money. In such case it is competent, for the purpose of proving guilty knowledge to show, that the person had in his possession other spurious money or paper, or had recently passed similar spurious money. In such case the former acts are pertinent to guilty knowledge, and are of the same class. But here the burning of the house and the killing of Mrs. Hughes are crimes not of the same class or category, and until a better preliminary foundation is furnished than appears in this record — one that connects the murder with the burning of the house as its motive — the evidence of the latter should not be admitted. There should be some logical connection between the two crimes.
But it is said that the jury did not convict on the first count, and therefore the evidence was not regarded by the jury as material, and the prisonr was not prejudiced. It is true the verdict of guilty was rendered on the third count, which is a charge of killing by certain acts of violence while attempting to perpetrate a rape. Nevertheless, the evidence was admitted generally, and *60the court in passing on its competency did not restrict its application to the first count, nor did he limit it in any instruction during the trial, or in the charge to the jury. It was admitted and remained in the case until the last.
The admission of this evidence on the foundation laid in this record was error for which the judgment might have been reversed.
It is further claimed that the judgment should have been reversed because of the admission of what counsel denominate “bloodhound testimony.” The day after the body of deceased was found, a Mr. Woodward brought from Dayton, Ohio, to the scene of the crime two dogs, and they were taken to the two saplings between which the neck of the deceased had been found compressed, and given a scent from them, and their conduct is described in. relation thereto; and then they were given the scent of a limb or branch that had been by some one broken from one of these saplings, as evidenced by fitting it to the stub from which it had been broken. The dog, or dogs, were to some extent controlled by Mr. Woodward ■ by means of a leash attached to the collar or harness on the dog. Immediately after the scent of the branch the dog started on his course, which was of a devious character, and finally went to the house of Dickerson, entered and went to a room’ upstairs. He was not at home, but was not far away. The conduct of the dogs from the giving of the scent to the entrance of the Dickerson home was submitted to the jury on his trial. It was admitted over his objection, and the very important question is made that the evidence was *61not competent. The history and experience of these animals is given by Mr. Woodward, who operated them on this occasion, and their conduct and acts are described by him and Mr. Wilson, who claimed to have been present. They are contradicted in material respects by a Mr. Daugherty, who also claimed to have been present on the course.
This class of evidence is somewhat new to the courts, and but few cases involving it have reached courts of last resort. It enters this court now for the first time. Other courts of final resort disagree upon the question, and we are obliged to settle it according to the best light of reason and authority at our command.
The instance is not one of a dog -testifying, but it is his .actions and conduct when given the start of a track, or more specially — the trail of a human being who, at a recent date was at the beginning of the trail, or as claimed here, was at the scene of the crime. It seems true, that the day before these dogs arrived, very many people, young and old, had trodden upon and over this scene, and some had experimented, with the pressure of the two saplings by placing their own necks where, as supposed, the neck of Mrs. Hughes had been found. This fact goes to the weight of such evidence rather than its competency, if it be otherwise competent.
Counsel for Dickerson not only urge that such evidence is not competent and violates constitutional guarantees in favor of one accused of crime; but also, if that point is not well taken, that a sufficient foundation or qualification had not been *62shown to make the conduct of the dogs admissible. As to the constitutional question, it is sufficient to say that the dogs were not zvitnesses whom the accused had a right to confront at the trial, in any different sense than the tracks of a man accused may be described as to form, size, or any other characteristic by which he may be identified. It is like the use of a yardstick in measuring goods. The witness is he who uses the stick to make the measure and it is a mere instrumentality. So runs the theory of the state.
The older text-books furnish us no light on this subject, for the reason that this class of evidence is of modern origin. In Section 1253 of Elliott on Evidence we have the views of the author stated as follows: “Although the subject does not fall strictly within the province of this chapter, yet the question of the admissibility of evidence of the actions of bloodhounds in following the track of a supposed criminal is not entirely foreign to the general subject under consideration, and this seems a convenient place in which to treat it. There is no certainty in such evidence. It is really the dog. that is the witness, and the evidence would seem to be hearsay in this view, ánd one court has vigorously maintained, in a very recent case, Brott v. State, [97 N. W. Rep. (Neb.), 593], that suqh evidence is not admissible. But other courts have agreed that it is admissible under, and only under, substantially the following conditions: Even when it is shown that the dog is of pure blood, and of a stock characterized by acuteness of scent and power of discrimination, it must also be shown by preliminary evi*63dence that the dog in question is possessed of these qualities, and has been trained or tested in their exercise in the tracking of human beings; and it must also appear that the dog so trained and tested was laid on the trail, whether visible or not, concerning which testimony has been admitted, at a point where the circumstances tend clearly to show that the guilty party had been, or upon a track which such circumstances indicate to have been made by him.” Citing Pedigo v. Commonwealth, 103 Ky., 41; Davis v. State, 35 So. (Fla), 76; Hodge v. State, 98 Ala., 10; State v. Hall, 1 Ohio Dec., 147; Simpson v. State, 111 Ala., 610.
We will give attention to the cases which recognize the class of evidence under consideration. The case of Pedigo v. Commonwealth, 103 Ky., 41, was decided in 1898 by a divided court, one of the seven judges composing the court dissenting in a very vigorous opinion. The majority of the court sanctioned the admissibility of testimony as to trailing by bloodhounds, but circumscribed its admissibility so that its use may be very limited. The proposition is that, “testimony as to trailing by bloodhounds of one charged with crime may be permitted to go to the jury for what it is worth as one of the circumstances which may tend to connect the defendant with the crime, when it is shown by some one having personal knowledge of the fact that the dog in question is of pure blood and of a stock characterized by acuteness of scent and power of discrimination, is itself possessed of these qualities and has been trained or tested in their exercise in the tracking of human beings, and that the dog so trained *64and tested was laid on the trail, whether visible or not, concerning which testimony has been admitted, at the point where the circumstances tend clearly to show that the guilty party had been, or upon a track which such circumstances indicate to have been made by him.” The court, on page 50, after stating the above rule, said: “When not so indicated the trial court should exclude the entire testimony in that regard from the jury.” The case sets out' the evidence upon which the commonwealth relied to show that the accused had been at the scene of the crime and at the point where the dog was given the scent, and the reasons why this species of dog is selected to trail a man, are fully stated in the majority opinion in that case, and we will not repeat them here.
The case quoted from followed the reasoning of Hodge v. State, 98 Ala., 11, except that the preliminary qualification of the dog in the latter case is somewhat less stringent. The court states the rule of qualification as follows: “When a dog trained to follow human tracks is, a short time after the commission of a homicide, put upon the trail of a person suspected of the crime, and follows it to the house of the defendant, this fact may be given in evidence, along with the evidence of several witnesses that .the tracks found at the scene of the homicide, were followed and identified along the precise route traced by the dog up to the house of the defendant, where he was shown to have been that night after the killing, as a circumstance, along with all the other evidence, tending to connect the defendant with the crime.” The court says in the opinion the evidence showed *65that the dog thus trained was within a very short time after the homicide put upon the tracks of the person toward whom all the circumstances strongly pointed as the guilty agent, and that the dog, as if following these tracks or trailing, went to the house of the defendant.
In the year 1903 this question was before the supreme court of Florida in Davis v. The State, 35 Southern Reporter, 76. The third paragraph of the syllabus, which was prepared by the court, reads: “Testimony as to the action of dogs in following the trail of a supposed criminal from the scene of a crime is admissible in evidence, provided such preliminary proof be given of the qualities and training of the‘dogs as to show that reliance may reasonably be placed upon their accuracy in following the trail of a human being.” In the opinion-of the court, by Cockrell, J., it is said, “there should first be testimony from some person who has personal knowledge of the fact that the dog used has an acuteness of scent and power of discrimination which have been tested in the tracking of human beings. The intelligence, training and purity of breed, are all proper matters for consideration in determining the admissibility of such evidence, as is also the behavior of the dog in following the track pointed out. In the record before us there is no proof of the breed of the dogs, and while there is proof that they had been trained for six months, there is no proof that they were trained in the tracking of human beings. It is questionable whether this is sufficient.”
But as the case had to be reversed on other *66grounds, the admission of the evidence referred to was not made a ground of reversal.
In 1904 a similar question was before the court of criminal appeals of Texas, in the case of Parker v. The State, 46 Texas Crim. Reports, 461. The rule there laid down is differently worded, but not much different in substance. It is: “Where the track assumed to be that of the murderer, and which the circumstances in evidence tend to show was his track, was pointed out to a dog of the hound species, who was trained in trailing tracks of human beings, and the dog trailed this track from where it was pointed out to him to the residence of the defendant, some mile and a half, and the course of his pursuit of the track was followed by witnesses who testified in the case, and showed that the dog followed this track which they saw upon the ground and described, such character of testimony was held to be admissible.” The facts entering into the preliminary proof in the latter case vary from those appearing in either of the former cases, but the court was passing on the sufficiency of the preliminary proof in that case, rather than endeavoring to formulate a general rule. It is argued that there are -cases holding a contrary doctrine, namely, State v. Moore, 129 N. C., 494; McClurg v. Brenton, 123 Ia., 368; and Brott v. State, 97 N. W. Rep. (Neb.), 593. But an examination of two of those cases, and particularly the facts upon which they rest, dispels the claimed antagonism. In State v. Moore, supra, the third branch of the syllabus is, that “the evidence in this case of the trailing by a bloodhound should not have been *67admitted.” But what was the evidence in this case? The conduct of the dog itself showed that his trailing was unreliable and that he did not possess the live and acute instinct so necessary to reliably trail human beings. After reviewing the evidence upon the character of the dog and his conduct, the court sums up on page 502, as follows: “In this case there is no evidence to connect the circumstance of the baying of the two defendants, or either of them, with the .making of the tracks at the time the larceny was committed; nor is there any evidence that the dog scented any that were made by either of the defendants; nor is there any way to ascertain that fact. The evidence failing to become a circumstance to connect the defendants with the crime, * * * there was error in admitting it.” It seems clear that the court so ruled because • the defective preliminary proof of qualification of the dog and his conduct on the trail and at its close, lead to the opinion that the evidence of the trailing should not have been admitted.
McClurg v. Brenton, supra, is a case where an action was commenced to recover damages for an unlawful search of plaintiff’s premises. The search was made without warrant, for the discovery of alleged stolen property, but by an officer of the law. The court say in branch three of the syllabus: “Evidence in an action for wrongful search as to conduct of hounds used to track the thief is only admissible on the question of malice, in mitigation of damages.” And in the fourth clause it is said: “In an action for an unlawful search aided by the use of hounds, evidence as to *68the breeding and training of the dogs, letters endorsing their usefulness and stories concerning their ability, are inadmissible.” Why? The court says on page 371-2: “The mere fact that a man is an officer, whether of high or low degree, gives him no more right than is possessed by the ordinary private citizen to break in upon the privacy of a home and subject its occupants to the indignity of a search for the evidence of crime, without a legal warrant procured for that purpose. No amount of incriminating evidence, whatever its spurce, will supply the place of such warrant. At the closed door of the home, be it palace or hovel, even bloodhounds must wait till the law, by authoritative process, bids it open. * * *” In another part of the opinion it is said: “But in an action for an unlawful search, it is no defense whatever to say that plaintiff was a thief, or did in fact have stolen property on the premises. The fact may be admitted but the right of action remains.” It is quite easy to see why the court held against the evidence offered.
Brott v. State, supra, as we read the case, is against the 'admission of the conduct of 'bloodhounds, • and perhaps under any circumstances. According to the opinion in that case, the court, as then constituted, had no faith.in such evidence and regarded its admission as dangerous to the rights of an accused party. It must be admitted that some reasons advanced for the conclusion reached are forcible and calculated to excite great caution, if not entire distrust, although they are couched in language of unsparing sarcasm. But we think that from a comparison of the views *69expressed by the different courts from whom we have quoted there may be deduced a rule which, until shown untrustworthy, may be followed in cases where this class of evidence is offered. It is apparent that before the acts and conduct of the dog. can be. shown, a proper preliminary foundation must be laid, and to establish such foundation it must be shown that the particular dog used was trained and tested in tracking human beings, and by experience had been found reliable in such cases; that the dog so trained was laid on the trail, whether it was visible or invisible, at a point where the circumstances tended clearly to show that the guilty party had been, or upon a track which the circumstances indicated to have been made by him. In addition to this the reliability of the dog must be proved by a person or persons having personal knowledge thereof. This foundation may be strengthened by proof of pedigree, .purity of blood, or the exalted standing of his breed in the performance of such peculiar work.
The charge of the court upon this subject seems to have been full and fair, and is not seriously complained of.
After all, it is the human testimony that makes the trailing done by the animal competent, and its actions are described by human testimony, just as it would describe the operations of a piece of intricate machinery. When the above foundation, or one similar in effect and efficiency is laid, the acts of the animal may be described. But the court, when such evidence is admitted, should explain its purpose and limitations to the *70jury, in order that they may not be unduly impressed thereby.
There was some evidence in this case as to the training of the dog and of his acuteness, and tending to prove that Dickerson had been seen in the bushes close to where the body was found; of tracks leading from the scene which were followed by witnesses contemporaneously with the trailing by the dog. We do not think it necessary or proper to comment upon its sufficiency now, but content ourselves with submitting the best rule we can formulate, which should govern when the case .is retried.
The circuit court was of the opinion that requests 5 and 10 made by the state and given to the jury should not have been given. If the objection to them consists in their being mere abstract propositions of law, it may be well taken, for such is their character; but giving one or more abstract propositions is not always error. The balance of the charge may cure the fault.
The court was less fortunate in the use of language we find in the general charge on page 438 of the printed part of the recor.d.
“In a civilized state like this, it is absolutely essential to the preservation of social order that the law should be enforced, and especially in cases where acts of violence have been done. The laws must be obeyed, violators of the law must be punished; and you as jurors would be faithless to your trust if you should return a verdict of acquittal in this case when the facts demand a conviction of the prisoner.
“It is equally important that innocence should *71not be punished. You were impanneled, not for vengeance, but to subserve the ends of public justice; and you would be disloyal to your obligations if you should find the prisoner guilty when the evidence required his acquittal.”
The wording of these statements may have been a mere inadvertence, but coming from the court so near the close of the charge, it is more than probable that they had a prejudicial influence.
Other errors in the record have been pressed, but we do not regard them as material.
The judgment of the circuit court is
Affirmed.
Shauck, C. J., Summers and Davis, JJ., concur.