Appellant was tried and convicted on an indictment which charges him with violation of §2322 Burns 1914, Acts 1913 p. 119. In prosecuting this appeal he assigns error in the overruling of his motion for a new trial and, under this assignment, first challenges the action of the circuit court in refusing to give his' requested instruction No. 13.
In order more readily to understand the purpose of this instruction it is important to note briefly certain circumstances connected with the trial of the cause. Oh the voir dire examination, the prosecuting attorney was permitted, over the objection of appellant, to interrogate each juror as to whether, in the event the jury should be allowed to hear evidence relating to bloodhounds, he would have any prejudice against receiving and considering evidence of that character?
Later, in his opening statement to the jury, one of the attorneys for the State was permitted, also over the objection of appellant, to comment at some length on the evidence which the prosecution, if permitted so to do, expected to introduce for the purpose of showing that a bloodhound had tracked appellant from the scene of the crime to his place of residence, and to prove the pedigree and training of the dog in question. During the trial, evidence of this character was offered by the State and, on objection, thereto by counsel for appellant, the jury was excused while the matter was argued. The court sustained appellant’s objection and the prosecution thereafter made no direct effort to prove the conduct of the bloodhound. On the subsequent examination of several witnesses, however, it developed incidentally that a dog was brought to the scene of the crime and accompanied its owner while the latter followed a trail of visible footprints leading in the direction of appellant’s house; also that appellant became *240excited and threatened to shoot the animal when it was brought onto his premises.
To meet the situation thus presented, counsel for appellant tendered and requested the court to give the following instruction: “In the opening statement of counsel for the State he was permitted to detail at length the evidence which the State proposed to prove if the court should hold such evidence admissible concerning the actions and conduct of a bloodhound upon which •the State relied for the conviction of the defendant. The court has held such evidence not to be admissible and has excluded the same from your consideration. You will therefore wholly disregard the statement of the attorney for the State upon this subject and determine this case upon the evidence introduced in connection with the presumptions referred to in other instructions. You will not consider as evidence of guilt any testimony as to the. conduct of such bloodhound which has incidentally crept into the case in proving other facts.” The court refused this instruction but gave another in which the jury was charged generally that in passing on the questions at issue it should “not give any consideration to the unsuccessful efforts of the attorneys on either side of the case to prove facts which the court has deemed to be improper in evidence.” The State now claims that the latter instruction includes the substance of instruction No. 13, above set out, and it was not error to refuse the more specific charge. Ginn v. State (1903), 161 Ind. 292, 293, 68 N. E. 294; Rains v. State (1894), 137 Ind. 83, 91, 36 N. E. 532.
*2411. 2. 3. *240It must be noted, however, that the ruling on the admission of bloodhound evidence as tending to prove guilt was made in the absence of the jury. It further appears that while evidence concerning the dog and its presence on appellant’s premises was thereafter prop'erly admitted for other purposes, there was nothing *241before the jury to indicate that such evidence was of restricted probative value. Under our Constitution, in a criminal case, the jury is the judge of the law as well as of the facts, but that provision is sub-. ject to the rule of procedure that, in both civil and criminal cases, the court determines the competency of evidence and the jury is bound by its decision thereon. It is well settled, also, that in cases where evidence is admissible for a restricted purpose only, the court should, on request, limit the jury to a consideration of such evidence for that purpose. Eppert v. Hall (1892), 138 Ind. 417, 420, 31 N. E. 74, 32 N. E. 713; Thistlewaite v. Thistlewaite (1892), 132 Ind. 355, 357, 31 N. E. 946. On the other hand, it must be borne in mind (1) that if the trial court improperly sustained appellant’s objection to the admission of bloodhound evidence, its refusal later to give instruction No. 13, was; at most, harmless error; and (2) that unless appellant’s requested instruction is correct as tendered, he may not complain of its refusal. American Motor Car Co. v. Robbins (1913), 181 Ind. 417, 420, 103 N. E. 641; Town of Newcastle v. Grubbs (1908), 171 Ind. 482, 500, 86 N. E. 757; Mosier v. Stoll (1889), 119 Ind. 244, 252, 20 N. E. 752.
We are required, then, to determine whether evidence as to the conduct of bloodhounds used in trailing persons accused of crime is admissible under' any circumstances as tending to establish the guilt of a defendant. This question, although noticed in Stout v. State (1910), 174 Ind. 395, 398, 92 N. E. 161, Ann . Cas. 1912 D 37, has never been decided in this jurisdiction and the decisions thereon in other courts are not in harmony. Those cases which sustain the admissibility of such evidence do so for reasons which are thus stated *242in Commonwealth v. Hoffman (1913), 52 Pa. Super. Ct. 272, 277, where the subject is considered at some length and the authorities collected: “It is a mattér of common knowledge of which the courts are authorized to take notice, that many animals have a special innate propensity, or sagacity, which transcends the general experience of human beings. It is called a natural intuitive power; perception, or instructive quality; due to effect of habits in successive generations, or original intelligence, special to a. class, in which the choice of means suitable to each particular instance is unconsciously made.
“The migration of birds and fish; dread of animals of prey; invariable habits of many animals in their natural, or a domesticated state are so frequently and vividly brought to our notice, that the existence of these special endowments cannot be ignored. We know that certain breeds of dogs are invested with special traits and gifts, peculiar to their respective kind, — the pointer and setter take instinctively to hunting birds, the hound, to foxes, deer, and rabbits. While there is no breed of dogs that instinctively hunts mankind, yet we do know that dogs are capable of being trained, or acquire the habit of their own elective preference, to follow the tracks of human beings. It is a well-known fact that bloodhounds can be trained to follow or run the track of strangers. The gift or power or instinct being already inherent in the animal, he may be induced by special training to exercise it, under the persuasive influence and training of a skilled master. Being once accurately trained in this pursuit, we may presume that his exactness depends on the capacity bestowed upon him by nature, and developed by intelligent training. All are not equally unerring, and each may fail at times in being truthful. It is a novel feature of evidence in our jurisprudence, and is attended with some dan*243ger, against which courts must guard as the occasion arises, and when the circumstances are at all doubtful.
“We gather from the eminent Writers on this subject, that the theory or scientific fact as it may be, is, that the path of every human being, at every step from the cradle to the grave, is strewn with putrescent excretions from the body. This waste matter is in process of decomposition and is being resolved into its constituent elements, so that its power to make an impression on the olfactory nerves of a dog, or other animal, becomes fainter with lapse of time, and exposure to the elements; and when dissolution is complete, its characteristic scent is ended. The bloodhound, is endowed with an unusually keen scent, and has great ability' for differentiating smells. The methods of trailing are simple and well understood. There must be an intelligent and truthful starting point, which will make an impression that the dog is able to recognize and distinguish from all other impressions; personal attendants and surrounding conditions may modify his certainty; time, weather conditions and other circumstances must be considered in giving proper weight to the conclusion reached. In order to make such testimony competent, 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 be established that the dog in question is possessed of these qualities, and has been trained and tested in their exercise, in the tracking of human beings, by a person who has knowledge and experience in such a matter, and that the test was made by starting the dog at a point where the circumstances tend clearly to show that the alleged guilty party had been, at a time when ’his presence would instinctively be known to the dog. When so indicated, testimony as to the trailing by a bloodhound may be permitted to go to the jury for *244what it is worth, as one of the circumstances which may tend to connect the defendant with the crime with which he is accused.”
The above quotation, however, not only contains a statement of the reasons which have led some courts to hold such, evidence admissible but also indicates some of the valid objections which may properly be raised against it. It is conceded that all dogs, although carefully trained, “are not equally unerring, and each may fail at times in being truthful.” It is admitted, also, that time, weather conditions and similar circumstances may affect the accuracy of the animal’s work, and that the acts of personal attendants may modify, or destroy his certainty. These facts are equally matters of common knowledge.
A further objection is thus recognized in Pedigo v. Commonwealth (1898), 103 Ky. 41, 50, 44 S. W. 143, 145, 42 L. R. A. 432, 82 Am. St. 566. “It is well known that the exercise of a mysterious power not possessed by human beings begets in the minds of many people a superstitious awe, like that inspired by the bleeding of a corpse at the touch of the supposed murderer, and that they see in such an exhibition a direct interposition of divine providence in aid of human justice. The very name by which the animal is called has a direct tendency to enhance the impressiveness of the performance, and it would be dangerous in the extreme to permit the introduction of such testimony in a criminal case under conditions which did not fully justify its consideration as a circumstance tending to connect the accused with the crime.” The same objection finds similar recognition in the case of State v. Moore (1901), 129 N. C. 494, 500, 39 S. E. 626, 55 L. R. A. 96.
Nearly .all of the decisions recognize the fact that such evidence is of a very dangerous character and that *245when received it is to be with caution, and as proof only of a circumstance pointing toward the guilt of the accused. In the case of Carter v. State (1913), 106 Miss. 507, 512, 64 South. 215, 50 L. R. A. (N. S.) 1112, it is expressly decided that such evidence, standing, alone and unsupported, is insufficient to convict; that there must be other and human testimony to sustain a finding of guilt.
It is not to be doubted that under some circumstances a proper use of bloodhounds may lead unerringly to the object sound, particularly when that object is the arrest and capture of known fugitives. This fact, as well as some of the objectións against the admission of their evidence in other cases, is recognized by Mr. Justice Guffey, in his well-reasoned dissenting opinion in the case of Pedigo v. Commonwealth, supra, where 'he says, at’page 52: . “If the dog in fact took up and followed the trail of a (known) fugitive and found him, or aided his pursuers to find him, the object was accomplished, and there could be no mistake as to whether he was the party sought or not. His guilt and right of capture having been theretofore established, and in fact being unquestioned. If the hound took the wrong trail and brought to bay the wrong party that fact would be ascertained so soon as the pursuers reached the party, and the utility of the hound in that regard then ceased. It is now proposed to use the hound not to capture a fugitive, but to ascertain or furnish evidence to convict some citizen of - crime. It seems to me * * * that neither the life nor liberty of a citizen should be taken away, or even jeopardized by the mere fact that some person testified that the hound was well trained to track human beings, etc., and that he had trailed the accused from the scene of the crime to the habitation of the accused, or until, he came upon the accused party. * * * There is too much danger of an innocent per*246son being convicted, or at least arrested and permanently disgraced by the admission of such testimony.” Similar expressions are embodied in the opinion of the Supreme Court of Illinois in the case of People v. Pfanschmidt (1914), 262 Ill. 411, 104 N. E. 804, 823, Ann. Cas. 1914 A 1171. After' reviewing a number of the decisions, the court says, at page 461: “We have reached the conclusion that testimony as to the trailing of either a man or an animal by a bloodhound should never be admitted in evidence in any case. A bloodhound may be used to track down a known fugitive from justice. If a dog, in fact, takes up and follows the trail of a known fugitive and finds him, or aids his pursuers to find him, there can be no mistake as to whether or not he is the party sought. His guilt or innocence of a given crime, however, should be established by other evidence. Neither court nor jury can have any means of knowing why the dog does this thing or another, in following in one direction instead of another; that must be left to his instinct without knowing upon what it is based. The information obtainable on this subject, scientific, legal or otherwise, is not of such a character as to furnish any satisfactory basis or reason for the admission of this class of evidence. We agree fully with the statement in Brott v. State, supra, that the ‘conclusions of the bloodhound are generally too unreliable to be accepted as evidence in either civil or criminal cases.’ ”
4. In our opinion the conclusion thus reached is well founded. When it is considered that the use of bloodhounds, even under the most favorable conditions, is attended with some degree of uncertainty, which may readily lead to the conviction or accusation of innocent persons, and that, at best, evidence as to their conduct in following a supposed trail is properly not of great probative value, it follows, as *247is suggested in Brott v. State (1903), 70 Neb. 395, 398, that both reason and instinct condemn such evidence, and courts should be too jealous of the life and liberty of human beings to permit its reception in a criminal case as proof of guilt.
5. Under the circumstances shovm by the record in this case it seems very probable that the jury may have attached undue weight to the showing that a bloodhound accompanied the parties who followed a supposed trail to appellant’s house and that he was harmed thereby through the refusal of an instruction to which he was entitled. This conclusion requires a reversal of the judgment below and renders unnecessary our consideration of certain other questions which are presented but which may not arise on a retrial of the case.
Judgment reversed, with instructions to sustain appellant’s motion for a new trial and for further proceedings not inconsistent herewith.