KKIOK, J.
Appellant was convicted of the crime of murder in the first degree, and sentenced to suffer death.
The information contains three counts. In the first count it is, in substance, alleged that on the 26th day of November, 1907, the appellant committed the crime of murder in the first degree by assaulting one Mary Yanee with the specific intent to take her life, and, with that intent and purpose, willfully, unlawfully, feloniously, deliberately, premedi-tatedly, and of his malice aforethought, with his fists, hands, and feet did strike, kick, beat, and bruise the said Mary Vance, and did then and there, and thereby, inflict upon the body of the said Mary Vance a mortal contusion, bruise, and wound, from which the said Mary Vance languished until the 8th day of December, 1907, when she died from the contusion, bruise, and wound aforesaid. The acts of appellant and the means used by him to produce the death of the said Mary Vance are alleged with much particularity, and the count contains a complete charge of murder in the first degree, and states a complete transaction. In the second count appellant is charged with having committed the crime of murder on the 27th day of November, 1907, by administering poison to one Mary Vance with the specific intent to take her life, and that said Mary Vance took said poison, and that by reason thereof she became mortally sick and languished until the 8th day of December, 1907, when she died from the effects of the poison so taken as aforesaid. In this count all the essential ingredients constituting murder in the first degree by administering poison as a means of death are alleged, and it is further alleged that appellant *8committed tbe crime by tbe means of said poison. In tbe third count all tbe allegations of tbe first count are set forth at length with tbe exception that it is not stated that tbe beating, bruising, and kicking resulted in death. In this count are also set forth in full all tbe allegations contained in tbe second count, excepting tbe statement that death was caused by the poisoning. After repeating tbe statements contained in tbe first and second counts as aforesaid, tbe cause of death is stated in tbe third count as follows: “That by means whereof, to wit, tbe striking, kicking, beating, and bruising of tbe said Mary Vance . . . and tbe drinking of tbe water and poison as aforesaid, tbe said Mary Vance became mortally sick and distempered in her body, and tbe said Mary Vance of tbe beating, kicking, and bruising aforesaid and of tbe poison aforesaid so* by her taken, drank, and swallowed as aforesaid, and of tbe mortal sickness and distemper occasioned thereby,” she languished from tbe 21th day of November, 1907, until tbe 8th day of December, 1907, when she “of tbe said mortal sickness occasioned by tbe said beating, kicking, bruising, and poison aforesaid died; and so tbe said Thomas Vance, tbe said Mary Vance, in tbe manner and form aforesaid, willfully, unlawfully, deliberately, premeditatedly, feloniously, and of bis malice aforethought did kill and murder.” After tbe state rested tbe appellant requested tbe court to require the state to elect on which count of tbe information tbe state would ask a conviction. To this request tbe district attorney replied: “Tbe state will elect to stand upon tbe count of tbe information which charges that death resulted from beating and poisoning, which, I understand, .is tbe third count in tbe information.” Tbe state having elected to stand on tbe third count, appellant interposed a further motion by whiph be requested tbe court to require tbe state to further elect on which charge in tbe third count, namely, tbe beating and bruising, or tbe administering of poison, tbe state would ask to go to tbe jury. This motion was opposed by tbe state and was overruled by tbe court, to which ruling tbe appellant duly excepted and assigns tbe ruling as error. In view that prac*9tically the same question arises upon another assignment relative to the giving of a certain instruction, which, in turn, involves some of the facts, we will first state what we deem to be the salient facts in the case.
The undisputed facts, as deduced from the state’s evidence, relative to the beating, kicking, and bruising are substantially as follows: Appellant and the deceased on the 26th day of November, 1907, and for more than twelve years prior thereto, sustained the relation of husband and wife. They lived together in Salt Lake City, with their children, four in number, ranging in age from two to twelve years. Appellant was apparently a poor man, and his wife and family were dependent upon his daily labor for support. Appellant was at work on the 26th day of November, 1907, and a little after the noon hour of that day came home for his midday meal. After all the family, including the deceased, had taken their places at the table and had been eating, appellant requested the deceased to refill his cup with coffee. The deceased proceeded to pour the coffee into appellant’s cup, and in doing so poured more coffee into the cup than he desired, whereupon he said: “Didn’t I tell you not to pour my cup so full of coffee ?” To this deceased answered that she did not hear appellant’s request. After some heated words, appellant picked up the cup of coffee and threw the coffee in the face of the deceased, and threw the cup against the wall and broke it. Appellant then picked up a chair, but the deceased admonished him not to strike her with it. He put down the chair, and said: “God damn you, I’ll kill you.” He then approached the deceased, and with his clenched fist struck her on or near the shoulder and knocked her down, and she fell against the kitchen stove, which was in the same room in which the family were then eating dinner. The deceased fell against the “middle part” of the stove with her left side. When this occurred, the children, or some of them, screamed, and the deceased got up from the kitchen floor and went through the door which led from the kitchen to the rear porch, and from there called to a neighbor, a Mrs. Wunderlich, who testified as follows: “She (the deceased) *10said: ‘Mrs. Wunderlich, go telephone for the police. He is hilling me.’ I said: ‘Who?’ She said: “Tom Vance, the brute which I call my husband.’ Then she went into the house again.” Mrs. Wunderlich was standing in her own back yard, which was adjoining the back yard and porch on which the deceased stood while she was -talking to Mrs. Wunderlich as above stated. Mrs. Wunderlich says that the deceased was but a short distance from the door leading from her kitchen to the rear porch, and that, while the foregoing statements were made by the deceased, she was about ten feet or such a matter from Mrs. Wunderlich. When the deceased returned to the kitchen, appellant resumed the attack upon her, and seized her with his hands “striking her and throwing her down on the floor.” After deceased fell, appellant kicked her, and she exclaimed, “Tom! you have killed me.” When appellant kicked her, she was lying on the kitchen floor with her back towards him, and the eldest daughter, who was present and saw and heard all that occurred, says that appellant “kicked her (deceased) on the thigh.” Within a few seconds after the kicking occurred, the deceased arose from the floor, sat down on a chair by the table at which appellant had resumed his seat just preceding her. Appellant finished his meal, and “had a smoke,” and in a short time left the house to go to his work. After appellant left the house, the eldest daughter, in describing the deceased’s condition, said: “She looked kind of pale. She was trembling.” About one hour and a half after the beating and kicking, the same daughter says the deceased “got herself and the baby ready and went to town.” The “baby” was about two years old, and the deceased carried him about two blocks to the street car line. At about five o’clock in the evening the deceased returned from town, and helped some in preparing the evening meal, but did not herself eat anything. After supper appellant in hearing the children rehearse some of their lessons found fault with the eldest daughter because she did not have her lesson, when the deceased spoke to him, and said, “I have had Lena helping me to-day. If you want to quarrel with anybody you can quarrel with me.” This *11is all that passed between tbe deceased and appellant that evening. Appellant and the deceased usually slept together in the front room, bnt on the night in question he slept in the front room with some of the children, while she, with the •other children, slept in what is designated as the “middle room.” The house, it seems, was divided into three rooms. The one next to the street was designated the front room, the next one the middle room, and the one to the rear where the trouble arose was the kitchen and dining room. Deceased :got np first the next morning and prepared breakfast for the family, including appellant, after which he left to go to his work. Nothing passed between them that morning. The ■deceased did not seem well this morning, and, after making ■some preparation for dinner, but before the dinner hour had arrived, she lay down on the bed in the middle room. Before lying down she filled an ordinary drinking glass or tumbler nearly full of water, which she took from the water bucket. The water had been taken from a flowing well in a neighbor’s yard, where the family obtained its water supply for domestic use. She set the tumbler with the water on the end of a sewing machine, which was just inside and near the door leading from the kitchen into the middle room at the head of the bed. That is, the machine was about three feet long and was standing between the door and the head of the bed on which the deceased was lying. Just before and about the time the deceased lay down on the bed, she looked pale, and, as the witness expressed it, was “trembling.” It seems that ■some time after the deceased went into the middle room to lie down on the bed Mrs. Wunderlich and Mrs. Amanda Yance Ward, a sister of the deceased, called upon her, and ■during that time, at the request of the deceased, made an ■examination of her body, and discovered a bruise four or five inches long and nearly as wide as the witness’ hand on the right side of the vagina. The color of- the bruised portion, the witness said, was “as black as it could get.” There were no other bruises or marks discovered on deceased’s body at that time. The two women left deceased’s home some time before the noon hour. A little after twelve o’clock noon, *12appellant came borne to dinner. Tbe eldest daughter, who, apart from the younger children, was the only person who saw what occurred, in referring to what appellant did, says: “He came home, washed his hands, and was talking to Albert and playing with him. Then he asked Florence where mama was. She said: 'She is in there on the bed lying down.’ He walked up to the door and put his hand on the door casing and looked in. Then turned around and came away.” Following this testimony the record shows the following questions and answers: “Q. Well, now, just describe his position when he had his hand on the door casing. You could see his hand all the time, could you ? A. Yes, sir. Q. Could you see the rest of his body ? A. I could see his side. Q. Do you know whether or not he was in that room during any other time that noon? A. No, sir. I don’t think he was. I don’t think he was in there at all.” After this appellant ate his dinner, and in about twenty minutes thereafter left to go to his work again. We remark here that the doctor who attended the deceased during her last illness from the 27 th day of November until the 8th day of December, 1907, in making an examination of her body on the evening of the 27th of November, and again some time thereafter, found external evidences of physical injury which he says was in the “crotch” at the place before mentioned, and that he at no time discovered any other bruises or marks of any consequence on her body. One other witness says that deceased was bruised on the abdomen, but the bruised part is not located. The doctor who made the autopsy testified that he found two bruises on deceased’s body. One he locates and describes as the other witnesses did, and the other he says was on the lower part of the abdomen on a direct line, from the hip to the pubis., and was about one and one-half inches long by about three-fourths of an inch wide. No other braises or marks were discovered on the body of the deceased. Both her attending physician and the county physician who made the autopsy agreed that there were no internal evidences of injury that was caused by or arose from the external bruises referred to, and that the deceased did not die from the shock *13wbicb in all probability resulted from tbe beating and kicking inflicted by appellant upon tbe body of tbe deceased. Both physicians, in substance, testified tbat death was not caused from either tbe bruises alone or the shock alone, but was caused by a co-operation or combination of tbe beating, kicking, and bruising and tbe effect of tbe poison.
Tbe evidence on tbe part of tbe state with regard to tbe poisoning is, in substance: Tbat appellant about a year prior to the' occurrences above narrated bad received some injury to bis band. Tbat at tbat time, upon a prescription, be obtained from a firm of druggists in Ogden, where be and bis family then lived, certain tablets wbicb were to be used on bis injured band by being dissolved in water and tbe band washed with tbe solution as an antiseptic. Tbat each tablet contained about 7.3 grains of bichloride of mercury, commonly called “corrosive sublimate,” wbicb is an intense poison. Tbat these tablets were kept in tbe front room where appellant slept tbe night after be bad quarreled with and beaten tbe deceased as we have narrated. Tbat both tbe deceased and tbe appellant knew tbat tbe tablets were in tbe trunk and they seemed to realize tbat they contained an intense poison, and tbe deceased bad at some time prior to tbe occurrence suggested tbat appellant burn or destroy tbe tablets, but be refused to do so, claiming tbat be might want them for some injury such as be bad bad before. This was tbe situation when tbe appellant went to tbe door of tbe room where tbe deceased was lying on tbe bed at noon of tbe 27th day of November, tbe day after the beating. In her dying declaration deceased stated tbat she was lying on tbe bed with her eyes closed during tbe dinner hour of tbe 27th; tbat on opening her eyes she saw tbe appellant leave her bed and go out of tbe middle room door; tbat some time after appellant bad left tbe bouse to go to bis work she drank some of tbe water out of tbe glass or tumbler we have mentioned and tbat immediately, or at least within a comparatively short time after drinking tbe water, she experienced great pain in her stomach and bowels and commenced vomiting. Deceased became alarmed and sent for her neighbor, Mrs. Wunderlich, *14and also for Mr». Amanda Vance Ward, the sister to whom we have already referred. Mrs. Wunderlich arrived at the home of the deceased about two or half past two o’clock that afternoon, and she says that when she arrived the deceased was suffering and “screaming with pain,” and that she was vomiting and purging from the bowels almost constantly. Mrs. Wunderlich says that she discovered a tumbler or glass about one-third or one-fourth full of water standing on the sewing machine near the head of the bed on which deceased was lying; that the witness saw nothing in the glass but water, which, as she said, contained “bubbles.” On cross-examination she explained this by stating that the water seemed stale. She emptied the water out doors in the back yard. We remark that the testimony of the state chemist and the physicians was to the effect that the tablets of mercury if dropped into the water whole would require all the way from 15 minutes to an hour or more to dissolve; that the tablets in question could, however, be crushed between the thumb and finger, and, if so crushed before depositing in the water, would dissolve in about five minutes, more or less; that there would be no sediment. of any consequence, but that there would be a sharp metallic taste to the water.
Dr. Kerr, who subsequently attended the deceased, was telephoned for by Mrs. Wunderlich after she arrived at the bedside of the deceased, but the doctor for some reason did not see her until about seven o’clock on the evening of the 2'Tth. We shall not refer to the evidence of the doctors with respect to the symptoms of the deceased, but shall content ourselves with the statement that the symptoms described by all the witnesses who saw the deceased on the afternoon and evening of the 27th. day of November and thereafter are conceded by the doctors who testified on the part of the state to describe the symptoms of mercurial poisoning. The doctors further testified that the autopsy disclosed that the deceased had received into her stomach some corrosive poison; that, while the symptoms might have been produced in part at least by some corrosive poison other than bichloride of mercury, yet, when all the symptoms exhibited by the de*15ceased are taken together with the evidences that were discovered at the autopsy in a portion of the stomach and in the intestinal canal of the deceased, that the controlling and predominant symptoms all pointed to the fact that the effects discovered were produced by mercurial poison. In this connection it is but just to state that the evidence disclosed that bichloride of mercury has a sharp metallic taste of which the deceased did not speak in her dying declaration, or at any other time. Moreover, another witness, Mrs. Wunderlich, said that during the afternoon of the 27th, while the deceased was vomiting, the witness discovered the odor of the oil of savin, which is classed among the corrosive poisons, and, among other things, is used to produce an abortion or mis-, carriage. No other witness, however, detected such an odor*, and, as the oil is quite volatile and for other reasons that it is not necessary to state, the jury were authorized in concluding that the witness, Mrs. Wunderlich, was mistaken. It was also the theory of the defense that the deceased attempted self-destruction. The theory is based upon the claim that the deceased and the appellant did not live happily together ; that she was averse to bearing more children both on account that they were poor and because she was at times at least not well. There was also some evidence, elicited on cross-examination of the eldest daughter and from others of the state’s witnesses, that deceased at times was despondent; that on some such occasions she had said that she were better dead, or that she wished she were dead; or that she would rather die than have more children. There was no evidence whatever that the deceased was pregnant or that she had any of the poisons above referred to at hand except the tablets we have referred to, but there was evidence that on the day of the beating she was menstruating and the autopsy disclosed nothing to indicate that the deceased had recently been pregnant. Moreover, the doctors testified that the ovaries of the deceased were normal.
When the doctors who testified for the state were asked to state the cause of death, they stated that death was caused by two causes co-operating together, namely, the effects of the *16beating, kicking, and bruising and of tbe corrosive poisoning. The doctors deduced this result from the following facts, namely, that a fatal dose of bichloride of mercury is all the way from less than one to three grains, and that death would ordinarily result from a fatal dose in from several hours to several days after the poison wTas taken into the system; that under the circumstances it was impossible to say how much in quantity the deceased took into her system by drinking the water from the tumbler, and, even if she had swallowed one-half or more of the seven grains contained in one of the tablets, it would be impossible to state how much she ejected by the violent and continued vomiting; that what can be said is, that an uncertain quantity of mercurial poison was taken internally by the deceased as was shown by both the symptoms before death and by the autopsy after death; that there were no internal evidences that the beating and bruising was sufficient to produce death by itself, but that the beating and bruising produced shock which lowered the vitality and diminished the resisting power of the deceased, and thus made the deceased’s system more vulnerable to the effects of the poison, and in that way it is quite natural that the deceased should languish from the 27th day of November to the 8th day of December, and then succumb to the combined or co-operating effects of the beating and bruising, and the poison. It was also made to appear that the state chemist, Mr. Harms, made a most thorough analysis of those parts of the internal organs which were submitted to him by the physicians who conducted the autopsy. He, however, discovered no mercurial or other corrosive poison. The absence of all such poison is however readily accounted for both by the state chemist' and the physicians, and the mere fact that such poison was not discovered in the internal organs of the deceased after death was by the physicians not deemed of great, much less of controlling, influence.
The second motion to elect was based upon the claim that the third count, like the two preceding ones, referred to and covered two separate and distinct transactions, and hence the state should be required to elect upon which one it would *17ask a conviction. The court denied the motion, and in our opinion the ruling was right, although, as appears from one of the court’s instructions to the jury, the ruling 1 was based upon an erroneous theory. The ruling of the court can be sustained only upon the theory that, although the third count referred to two separate and distinct transactions, yet, since it was alleged that it was the combined effect arising out of the two transactions that in co-operating together constituted the means or the cause of death, the third count in legal effect stated but one transaction which consisted in stating the effect produced by the two cooperating causes. It will be observed that in the first count the beating and bruising was charged as an independent means of death. This was followed by a second count which gave the means used by appellant in committing the murder by the administering of poison, and the poison was alleged as a . distinct and separate means of death. In these two counts it was not alleged, and was not intended to be, that the appellant caused the death of the deceased by the means of beating and bruising and by the administering of poison, as the pleader might well have done, but the charges in the two counts described two independent transactions, one of which occurred one day and the other upon another day, but each of which constituted a complete statement of the crime of murder, and as having been committed by separate and distinct means on different days. After the pleader had thus charged that the murder had been committed in two ways in the two counts, he added a third count in which he simply repeated the charges contained in the two counts in so far as the charges alleged the means used to produce death, and the third count was completed by charging that the death of the deceased was caused by the mortal sickness which was produced by two co-operating means or causes, namely, the beating and bruising which was inflicted upon, and the poison which was administered to, the deceased by appellant. Two co-operating causes were thus stated as the means used by the appellant to accomplish his purpose. The statement, in the *18form it was thus presented, was not tbe statement of the use of several or various means, but it was a statement that appellant resorted to but one mode or means to accomplish the death of the deceased, namely, by the effect of two co-operating causes which in combination, and not singly, produced or caused death. The pleader, having thus fairly and clearly stated that two co-operating causes when acting in combination produced death, could not, in fairness, under the rules of pleading, claim that his allegations, single and re2 strictive as they were, authorized a conviction for only one of the causes, but, under the restricted form of his allegations, he was required to prove that the two causes cooperated, and that death resulted from their combined effect. If the pleader had desired to rely upon the two causes separately as well as upon their combined effect, he easily could have done so by stating in one count that the means used by the appellant to kill were beating, kicking, bruising, and by administering poison, and by any other means which the pleader thought the evidence might show were used. The pleader could have gone further and stated that the deceased was killed by means unknown, and thus have proved any means upon the trial. Moreover, under our statute (section 4734, Comp. Laws 1907), the means could have been stated in the alternative in the same count.- Again, the means could have been stated as continuous; that is, that whatever means were used were applied and continued for a shorter or longer period of time, and that death was ultimately caused by the means so described. Where this form of pleading is used, the prosecutor need not establish anything more than that death resulted from any one or more of the means alleged, or that it was produced by a combination of two or more of the means described. This rule has so often been passed upon and enforced by the courts that it has become established beyond dispute. Among numerous other cases that might be cited we refer to the following well-considered eases: King v. State, 137 Ala. 47, 34 South. 683; State v. McDonald, 67 Mo. 17; Howard v. State, 34 Ark. 436; Smith v. Commonwealth, 21 Grat. (Va.) 811; Anderson v. United *19States, 170 U. S. 492-500, 18 Sup. Ct. 68-9, 42 L. Ed. 1116; Joy v. State, 14 Ind. 143; Merrick v. State, 63 Iud. 327; State v. Edmundson, 43 Tex. 163 ; Commonwealth v. Macloon, 101 Mass. 23, 100 Am. Dec. 89; State v. Fiester, 32 Or. 254, 50 Pac. 561; People v. Davis, 56 N. Y. 95; Jackson v. State, 39 Obio St. 37; State v. Hewes, 60 Kan. 765, 57 Pac. 959; Commonwealth v. Stafford (Mass.), 12 Cush. 619; Gonzales v. State, 5 Tex. App. 584; State v. Smith, 24 W. Va. 820-821. See, also, Wharton on Homicide (3d Ed.), section 563; Bish op’s Directions and Forms (2d Ed.), sections 20-21, 535. Bishop, in section 21, just referred to, in summarizing the benefits and advantages resulting from stating the means applied by the accused in the commission of crimes in one count, is very careful to state, however, that the statement must be strictly limited to one, and only one, transaction. He says:
“This method, namely, charging the offense, whatever it is, in one count, as committed in all the ways known to the law and not inevitably inconsistent with one another, within the prohable range of the proofs, and directing the jury that they may find a verdict of guilty on being satisfied of the truth of so much of the allegation as constitutes the offense, is abundantly sustained by the authorities; while it is practically superior, above all comparison, to the cumbersome indictment of many counts. Let it be borne in mind that what is thus to be set out is simply one transaction, which, and only which, is to be given in evidence to the jury. Thei charge^ therefore is homogeneous. The prisoner, the counsel, the court, the jury, all have before them the one thing and no more.” (Italics ours.)
If thus but oue trausactiou is referred to, the accused cauuot require the state to elect as between different means that it is alleged caused the death. The whole case is submitted to the jury, and it is for them to say from the evidence what the means were that were used and what one caused death, and, if any one caused it, the charge is sustained. The doctrine just stated is illustrated in the foregoing cases under various circumstances! If the cases are critically examined, it will be found that in all of them it is made apparent that only one transaction was intended to be charged and described. In other words, it is clear from the cases that *20"the pleader intended to state, and did state, that tbe means ¡alleged in tbe indictment or information were tbe means nsed by tbe accused to kill tbe deceased, and that the charge referred to and covered but one transaction. This is especially illustrated' in tbe cases of State v. Edmundson and Commonwealth v. Macloon, supra. In those cases tbe means used to produce death were applied during a period of several months, and yet tbe acts of tbe accused weTe treated in tbe information as a single transaction. This fact is also frequently referred to in most, if not all, of tbe other cases cited. Moreover, some of tbe cases also1 illustrate that different means may be stated in different counts, and, if it is clear that only one transaction is intended to be referred to, tbe prosecutor is not required to elect between counts, although different means are pleaded in each count. In tbe case at bar tbe pleader, however, adopted tbe method of stating not only different means in different counts, but be also referred to two separate and distinct transactions, one occurring and having been completed on tbe 26th day of November, and tbe other occurring and having been completed on tbe 2Ytb day of tbe «ame month. Tbe district attorney recognized this situation when, be admitted that tbe state should elect, but tbe difficulty is that in adding tbe third count and upon which the state elected to ask a conviction tbe pleader simply reiterated tbe statements contained in the first two counts, and left tbe matter, in so far as charging two transactions, one occurring on one day and tbe other upon tbe next, just as they were in tbe two counts, which, by tbe election, were eliminated from tbe case. Tbe state, therefore, surrendered nothing and tbe accused gained nothing by such an election. There is one, and only one, ground upon which tbe court’s ruling in denying appellant’s second motion.can be sustained, which is that tbe pleader in tbe third count stated but a single means as causing tbe death of the deceased, namely, tbe eo-operation of two distinct causes which produced tbe mortal sickness from which it is alleged tbe deceased languished and finally died. Tbe pleader bad a right to accuse appellant of having used one, two, or more means, or of causing death by applying *21two distinct means as co-operating canses, the joint effect of which produced death. As we have seen, the pleader could have availed himself of charging all these means by pleading in a certain way. He did not choose to do so, but specifically alleged that the appellant committed an assault upon the deceased and used certain means on one day, and that he committed another assault and applied certain other means on another day, and that the effect of the means so described co-operating together produced a mortal sickness which caused death. The effect of the allegations in the third count is that the two causes, one arising from the beating and bruising and the other from the administering of the poison, co-opera.ting together, produced death. This being so, the court was-right in not requiring the state to elect, since in legal effect there was but one cause of death alleged, namely, the joint effect of the beating and bruising co-operating with the effect produced by the poisoning. It is also clearly shown by the record that the trial proceeded upon this theory. The physicians testified that death did not result from either th& beating alone, or from the effects of the poison alone, but that it was caused by the co-operating influence of the effects of' both, and this thus constituted but a single, and the real,, cause of death. This, in legal effect, is just what was charged as the means of death in the third count. But, notwithstanding this, the court ruled and instructed the jury that they could find the appellant guilty if they believed from the' evidence beyond a reasonable doubt .that the deceased died from the effects of the beating and bruising and kicking alone, or from the effects of the poison alone, or from the combined effects of both. As we have seen, this charge might have been proper if the information had been different, but,, in view of the language contained in the third count of the-information, there was no distinct charge left upon which to base a finding that death was caused except from the eo operating causes, which in and of itself excluded the charge that either one of those causes produced, or could have produced, death. The court therefore erred in charging the jury as aforesaid.
*22But the court further erred in instructing the jury as stated above, because there was no evidence before the jury upon which they could find the appellant guilty of having committed murder by the means of beating, kicking, and bruising. The first count was eliminated from the case, and, as we have seen, the third count did not contain a sufficient statement that the beating, kicking, and bruising produced death. In the third count it was charged that death, was caused by two co-operating causes. This being the charge, it would have to he sustained by proof in order to convict. In referring to this point, the author of Wharton on Homicide (3d Ed.), section 563, at page 848, says: “But an indictment charging the death to have been occasioned by two co-operating causes, if the evidence fail to support one of the causes, is insufficient.” From this it further follows that, although there had been a sufficient charge left in the inform .(nation that death resulted from the beating and bruising, yet, there being no evidence in support of this charge, the instruction that a conviction could h® based thereon cannot be sustained. The Attorney General recognizes this difficulty, and, to overcome it, suggests in his brief that the charge of the court can be sustained upon the theory that the jury were authorized to disregard the testimony of the doctors with respect to the cause of death, and that they were authorized to find that the effects of the beating and bruising alone caused death, notwithstanding the testimony of the doctors to the contrary. If there had been no autopsy from which it was made to appear that there were no internal injuries or evidences attributable to the beating and bruising, and that death was not caused by the beating, the jury might, perhaps, have been justified in finding that the beating resulted in internal injuries from which deceased died. 'While it is true that ordinarily a jury, after considering and weighing purely opinion evidence if contrary 3 to their own knowledge and experience, may disregard such evidence entirely, but they may not do so where witnesses, although testifying as experts, testify to symptoms and conditions actually seen and observed by them. Where *23witnesses, although experts, describe to the jury the conditions as they have observed them, and the effects attributable to such conditions, and such evidence stands unimpeached, neither the court nor the jury have any more right to ignore such evidence than they would any other credible evidence in the case. Verdicts and findings must be based on some evidence, and, when the evidence fails, all fails. In view of the testimony of the doctors who testified for the state, there was therefore no evidence upon which the corirt could authorize the jury to find that death was caused from the effects of the beating and bruising alone. The instruction which directed the jury that they might so find was, therefore, without support in the evidence, and constitutes reversible error.
Counsel for appellant, however, insist that there is no evidence to support a finding that the accused administered poison to the deceased, and hence, in view of what we have already said upon the other branch, no evidence upon which the case can be submitted to a, jury upon any theory. In view of some facts that are disclosed in the application for a new trial, which it is not necessary to specifically set forth, the evidence on a subsequent trial may be somewhat different from what it was at the last one. While the evidence that the accused administered poison to the deceased is quite unsatisfactory and inconclusive, yet, in view that the evidence upon that question may be different upon a retrial of the case,, we are unwilling to direct the trial court in advance whether the case should be submitted to a jury or not on that branch of the case. Upon a new trial, with all the evidence before the court, it will no doubt be able to determine whether the case ought to go to the jury or not. For the reason, therefore, that a new trial is necessary, we refrain from a further discussion of the evidence upon that branch of the case.
At the trial, the state produced Vhat purported to be a dying declaration of the deceased, which was reduced to writing and duly signed and sworn to by her. The state also offered in evidence in connection with the written statement of the deceased oral declarations which it is claimed she made in extremis a short time after the written state*24ment was made. Appellant now urges that the court erred in admitting these oral statements, and insists that the court should have permitted nothing but the written statement to go to the jury. The law is that dying declarations, otherwise competent, may be either in writing or oral, 4 and where there are both written and oral declarations, if otherwise competent, eithér one or both may be given in evidence. (21 Cyc. 981; Wharton on Homicide [3d Ed.], section 644; State v. Schmidt, 73 Iowa, 469, 35 N. W. 590; State v. Carrington, 15 Utah, 480, 50 Pac. 526.) The court committed no error in admitting either the written or oral declaration referred to.
It is also asserted that the court erred in admitting in evidence what the deceased said to Mrs. Wunderlich at the time of the beating, when she asked Mrs. Wunderlich to call the police. It is insisted that the statement made by the deceased, which we have set forth in full in the statement of facts, was not admissible (1) because not part of the res gestae, and (2) because it was not made in the presence of appellant. The statement was made while appellant’s maltreatment of the deceased was still in progress. Deceased had just arisen from the kitchen floor after having been “knocked down” by appellant and left the kitchen, where appellant was, and, after going through the door of the kitchen onto- the rear porch, she, in calling to Mrs. Wunder-lich, made the statement complained of. It appears from the evidence that the kitchen door was not closed after the deceased had passed out, and appellant was in the act of closing it when the deceased returned to the kitchen, when he again seized her and threw her down on the kitchen floor, and kicked her, at which time she uttered the exclamation that appellant was killing her. It is thus made to-appear that what she said to- Mrs. Wunderlich was 5 in fact said while the acts of appellant were still in progress. The statement or exclamation made by the deceased was thus a part of the res gestae, and admissible upon that ground. Moreover, the evidence shows that the deceased was but a short distance from the kitchen door when *25she called to Mrs. Wunderlich; that she immediately returned to the kitchen, and thus frustrated appellant’s attempt to close the kitchen door against her. There is also evidence to the effect that the children, who were in the kitchen, heard the deceased, in speaking to Mrs. Wunderlich, “calling for the police,” that appel6 lant was nearer to the deceased than any of the children, and therefore he must have' heard all that the deceased said, and therefore what was said by her was, in effect, said in his presence. Under these circumstances, appellant has no cause for complaint and the court did not err in permitting the statement to go to the jury.
It is further asserted that the court erred in admitting in evidence the testimony of a certain witness who testified for the state on the preliminary hearing, and who was cross-examined by counsel for appellant at the hearing aforesaid. At the time the accused was tried in the district court, the witness in question was absent from the state of Utah, and was a resident of the state of Oregon, all of which was made to appear. By section 4670, Comp. Laws 1907, the testimony given at a preliminary hearing in a homicide case must be reduced to writing. The testimony may, however, be taken in shorthand in the form of questions and answers by a stenographer duly appointed by the magistrate before whom the preliminary hearing is held, and, when so taken, the stenographer is required to transcribe or extend his shorthand notes into longhand and certify that the transcript contains a correct statement of the testimony and proceedings in the case. The transcript, when certified to as aforesaid, is by the statute declared to be prima) facie a correct statement of the testimony. If the accused is held to answer to the district court, the stenographer is by the statute required within ten d^ys after the close of the hearing to transcribe his stenographic notes into longhand, attach his certificate thereto, and file the same with the clerk of the district court of the county in which the accused is held for trial; the stenographer in all eases being required to file his original shorthand notes with the clerk aforesaid. By subdivision 4 of section 4513 *26it is, in substance, provided tbat where tbe testimony of a witness taken at a preliminary bearing is taken in tbe presence of tbe accused, and tbe witness is cross-examined by bim, or be is given tbe opportunity to cross-examine, and tbe testimony is taken in tbe form of questions and answers, then, in case tbe witness is dead, insane, or cannot be found witbin tbe state, .bis testimony so taken may be read in evidence on tbe trial of tbe case by tbe state. But appellant objected to tbe reading of tbe testimony, first, upon tbe ground tbat no sufficient foundation was laid to admit it in evidence; second, tbat tbe admission of tbe testimony was in violation of tbe constitutional provision tbat tbe accused must be confronted by tbe witnesses against bim; third, tbat tbe stenographer’s certificate to tbe transcript was insufficient; and fourth, tbat tbe statute bad not been complied with, for tbe reason that tbe original stenographic notes were not filed with tbe clerk witbin tbe time fixed by section 4670, supra.
Tbe objection based upon tbe first ground stated above is untenable. Tbe evidence was ample to justify tbe court’s conclusion tbat tbe witness was a nonresident of and absent from this state at tbe time of trial. This was sufficient to 'admit tbe evidence under tbe rule adopted by this court in State v. King, 24 Utah, 482, 68 Pac. 418, 91 Am. St. Rep. 808. Tbe contention tbat appellant’s constitutional rights were disregarded by admitting tbe testimony of tbe absent witness is answered by this court adversely to bis contention in State v. King, supra. Nothing need be added to what is there said why tbe admission of tbe testimony is not in violation of appellant’s constitutional rights. Tbe subject of tbe admission of tbe testimony of witnesses taken at preliminary bearings who may be dead, insane, or absent from tbe state is also discussed and tbe cases upon tbe 7, 8 subject, both pro and con, under statutes and in tbe absence of statutes, are collated in notes to tbe cases of State v. Nelson, 1 Am. & Eng. Ann. Cas. 471, and State v. Spencer, 13 Am. & Eng. Ann. Cas. 973. Tbe court did not err in admitting tbe evidence upon this ground.
*27The third ground of objection, namely, that tbe certificate of the stenographer attached to the longhand transcript of his notes was insufficient, was, in our opinion, entirely overcome by the amendment of the certificate at the trial by the stenographer. We can see no force to the contention that the court had no power to permit the stenographer to amend a mere formal defect in his certificate. The statute does not prescribe a form of certificate, but simply requires that the stenographer shall certify that his longhand transcript is “a correct statement of such testimony and proceedings in the case.” The certificate in this case was “that the above and foregoing is a full, true, and correct transcript of all the proceedings therein” in the case. If by a technical interpretation of this certificate it can be said that the statute was not literally complied with in the first instance by omitting to state that the transcript contained a correct statement of the testimony, this defect was supplied by the stenographer at the trial by amending the cer- 9 tificate. We can conceive of no good reason, and counsel for appellant have suggested none, why such a formal defect of the certificate is not curable by amendment. It is in the nature of curing a defect in a return of process, or in. any other formal certificate, which, in' practice, is permitted as a matter of course. We think, as a matter of law, the formal defect was curable by amendment.
This brings us to the last ground of objection, namely, that the statute (section 4670) “had not been complied with by the filing of the original stenographer’s notes” within the time fixed by said section. It is not quite clear from the reading of the section whether the original stenographic notes are required to be filed within ten days or not. The statute provides that, if at the preliminary hearing the accused is held for trial, the stenographer shall, “within ten days after the close of such examination, transcribe his said shorthand notes into longhand, and certify and file the same with the clerk of the district court of the county in which the defendant shall have been examined, and shall in all cases file his original note with said clerk.” If we assume that the *28foregoing language implies that the original notes must themselves be filed within the time stated, the question still remains whether the time limit is of the essence and thus mandatory, or whether it is merely directory. We confess that we can see no good reason for holding that, unless the original stenographic notes are filed 10 within ten days, therefore the record of the proceedings had at a preliminary hearing is made unavailing for the purposes contemplated by the statute. It requires but slight reflection to convince the ordinary man that what is required by section 4610 is as much for the benefit of the accused as it is for the state. If the witnesses for the state alone testify at the preliminary hearing, and such is usually the case, especially in homicide cases, the accused obtains access to a full and reliable transcript of their testimony, and at the trial of his case he is thus afforded the opportunity to call the attention of the court and jury to any discrepancy in the testimony of any witness as given at the trial when compared with what he testified to at the preliminary hearing. In this very case the accused availed himself of this advantage, and it seemingly did not occur to his counsel that the transcript of the testimony could not be used for any purpose until the state desired to use it for the purpose of showing what the testimony of the absent witness was. It is quite true that it is possible that, under certain circumstances, a transcript or record may be admissible for one purpose, and yet not for all other purposes. Such, however, as we view the matter, is not the case here. It seems to us that, if the filing of the original stenographic notes within the ten days is essential for one purpose, it is equally essential for every purpose that the transcript of the testimony taken at the preliminary hearing may be used for. It would seem, therefore, that the provision fixing the time of filing the original stenographic notes is directory merely, and a failure to comply with that provision will not affect the right of either the state or the accused to use the transcript of the testimony taken at the preliminary hearing for the purposes contemplated by the statute, if it is otherwise *29•competent, or invulnerable to other objections. In the case of State v. Morgan, 27 Utah, 103, 74 Pac. 526, this court held that the provision with regard to the time of filing the notes aforesaid is directory merely. True, this was hold to be so •upon an objection other than the one now before the court, but, if that provision is directory for one purpose, it logically and necessarily must be held to be so for all purposes. Moreover, the filing of the original notes can be required but for one purpose, namely, to enable counsel for the state and the •accused to determine whether the longhand transcript of •the testimony and the proceedings corresponds with the original notes. Beyond this, they can perform no function whatever. It would seem, therefore, that, unless the statute in terms prevents the use of the transcript unless the original ■stenographic notes are filed, the courts cannot do so unless It is made to appear that the accused is in some way prejudiced, or is thereby prevented from having a fair and impartial trial. It is our opinion, therefore, that in view that the appellant made no claim of prejudice because the original notes were not filed within the ten days mentioned in the statute the court committed no error in permitting the notes to be filed at the time of trial, nor in permitting the transcript of the testimony of the absent witness taken at the preliminary hearing to be introduced in evidence.
The next assignment relates to^ the cross-examination of ■appellant. It is strenuously insisted by appellant’s counsel that the court erred in compelling appellant to answer cer--fcain questions propounded to him on cross-examination by the prosecuting attorney. Appellant exercised his constitutional and statutory right to testify in his own behalf. He took the stand, and, in answer to certain questions put to him, in substance, testified that he came home for the noon meal a little after noon on the 27th day of November; that he remained at home about thirty minutes, during which time he ate dinner with the children; that he went to the door •of the middle room in which his wife, at the time, was lying •on the bed, put his left hand on the door casing, looked into the room, saw the deceased lying on the- bed “apparently *30asleep;” that be did not go into tbe room; that be neither then, nor at any other time, put any poison of any kind, or any other substance of any kind, into the glass of water which was standing on the sewing machine, nor did he then, or any at any other time, place any poison of any kind in anything- or in any place where the deceased would be likely to get it; that he knew that there were tablets containing bichloride of mercury in the house somewhere, but he did not know that they were in the trunk in the front room where he slept the preceding night; that he obtained the tablets about a year prior to the time in question as an antiseptic medicine for his injured hand; that he obtained them from Bich & Osgood,, druggists in Ogden; and that he obtained them and used them under the directions of a doctor. The witness identified the envelope in which the tablets were put up by the druggists when he received them. After testifying as aforesaid, the-district attorney cross-examined the witness with regard to the foregoing statements, and then asked him whether he did not have a conversation with the deceased on the evening of the 27th of November, to which the witness answered: “Yes, passed a few words with her then.” The district attorney then asked: “What was said about poison, if anything?” Counsel for appellant objected to the question upon the ground that it was not proper cross-examination, and that the question assumed that poison was the subject of conversation. The court overruled the objection and required the witness to answer, and, after stating that there was nothing said about poison “at that time,” he said that afterwards, about nine o’clock that night, he had “a little conversation”' with the deceased upon the subject of poison. He was then asked whether on the following day, when the deceased was taken to the hospital for treatment, he did not say to her, “Mary, if you make any trouble for me, I’ll make trouble for you,” and he was compelled to answer the question over-counsel’s objection that it wás not proper cross-examination. After denying having made the statement, he was further-asked whether at that time he did not state: “This is a damned nice country. A woman can do anything she wants *31to and be protected, but a man can’t do anything.” The witness, over the same objection, was required to answer the question, which he did by denying the statement. He was further compelled to state on cross-examination whether there was any trouble with the mouth and gums of the deceased, and whether they were bleeding or not, and whether the deceased in referring to her condition did not say to him that she wanted him to see what he had done, and that she requested .him to examine the bruises upon her body as she was lying in bed on the night of the 27th. After denying these stater ments and occurrences, he was asked further questions upon the same and similar subjects, all of which he was compelled to answer over the objection aforesaid.
It is strenuously insisted by appellant’s counsel that the court erred in permitting the prosecuting attorney to extend the cross-examination as indicated. Counsel, in their brief, in referring to the conversations inquired into on cross-examination, say: “Thera is nothing in the conversation itself which was detrimental to the defendant, and yet any lawyer of experience would have immediately said, “The recitation of that conversation would injure your ease, not because that it will tend to prove you guilty, but because the jury will not believe you.’ ” The contention, therefore, is: (1) That the matters inquired into by the district attorney were not proper cross-examination; and (2) that, in view of the effect that the questions and answers had upon the jury the rulings of the court were very prejudicial, although the matters testified to by appellant in themselves did not possess any probative force either for or against him. In view of our statute, did the court err in requiring the appellant to answer the questions as part of the cross-examination?
Section 5012, Comp. Laws 1907, especially provides that the rules of evidence in civil eases apply to all criminal actions, except as otherwise provided by the Code. Section 5015 provides that, “if a defendant offers himself as a witnesss, he may be cross-examined by the counsel for the state the same as any other witness.” Section 5012 thus gives us the rule to be followed, and section 5015 in express *32terms provides that the accused, if he becomes a witness, must be treated on cross-examination the same 11 as any other witness. In view of the provisions of these sections, the test the court must keep in mind is: Would the particular question be proper cross-examination if the same were propounded to any other witness who had testified to the same facts that the accused has testified to ? If the question would be proper cross-examination if asked of any other witness it would likewise be if propounded to one on trial for a criminal offense, or vice versa. The rule is that as to whether the accused has made certain admissions, or has made statements of material facts against himself, and everything which may contradict, modify, explain, or make clearer, limit, or enlarge the meaning of the statements made by him while testifying with respect 12 to any subject of which he has testified, may be inquired into on cross-examination. The inquiry must, however, be limited to the subject-matters gone into by the witness in his testimony in chief. Sometimes, and under peculiar circumstances, it is a matter of some difficulty to determine the precise limits of the subject-matters gone into by the witness, and, for that reason, if for no other, courts sometimes differ with regard to what is and what is not proper cross-examination. Where the accused, as a witness, denies that he committed or was connected with the commission of the criminal act or acts constituting the offense for which he is being tried, the cross-examination ordinarily must be permitted to extend to the whole range of facts which in some way are related to the transaction constituting the offense. But where, as in the case at bar, the witness limits his statements to negativing or explaining 13 mere isolated facts, or merely states what occurred at a particular time and place, then what took place at such time and place ordinarily constitutes the subject-matter upon which the witness testified, and the cross-examination should be limited to that subject.
*33We are of the opinion that all the questions asked by the district attorney on cross-examination, excepting those which referred to the beating and bruising and to the condition of the accused which was assumed was caused by the beating, were proper cross-examination under the rules we have just stated. The court therefore committed no error in requiring the appellant to answer the questions as above referred to. But we cannot see upon what theory the court’s rulings can be sustained by which appellant was re14 quired to answer the questions relating to the conversations and statements that were inquired into respecting the beating and bruising and deceased’s condition attributable thereto. Appellant had neither directly nor indirectly denied, nor in any way negatived his connection with the beating. He left the subject untouched in his examination in chief. The subject, therefore, was mot opened up1 for cross-examination. Nor can it be seriously contended that the latter questions were proper as affecting appellant’s credibility, or the weight that should be given to his testimony. The cross-examination, therefore, came within the rule announced in State v. Shockley, 29 Utah, 25, 80 Pac. 865, 110 Am. St. Rep. 639. Mr. Justice McCarty, in referring to the subject now under consideration, in that case, at page 48 of 29 Utah, at page 873 of 80 Pac., says: “Nor did it (the testimony of the defendant) directly or remotely refer to any fact or circumstance testified to by him, or that came within the range of his examination in chief.” Upon the other branch, as to whether the questions were asked for the purpose of affecting the weight to be given to his testimony, the Justice says: “But it is apparent that the questions complained of in this case were not asked, nor was the evidence sought to be elicited thereby, for any such purpose.” What was true in that case is true here, namely, that the questions with regard to the beating and the condition of the deceased attributed thereto were not proper for any purpose, and hence the court erred in requiring appellant to answer those questions. See, also, State v. Williams, 36 Utah, 273, 103 *34Pac. 250. While under tbe rule as adopted by some courts the defendant in a criminal proceeding, when he takes the stand as a witness, is held to waive his eonstitu-15 tional rights with regard to testifying against himself, yet both the better reason and the weight of authority are in favor of the rule we have adopted. In the recent case of People v. Smith, 9 Cal. App. 644, 99 Pac. 1111, numerous cases upon this subject, including other California cases, are collated. The last case referred to is squarely in point upon the question presented in this case. See, also, State v. Duncan, 38 Am. St. Rep. 894, and Evans v. O’Connor, 75 Am. St. Rep. 318, where, in notes, the cases pro and con upon this subject are reviewed. In California, as in this state, the subject of cross-examination is regulated by statute. While the phraseology of the California statute is somewhat different from ours, yet, when the two sections of our statute to which we have referred are considered together, there is in our opinion, in legal effect, no difference between the California statute and our own; and such, in effect, is the holding of the California court, as appears from the decisions to which we have referred.
We are aware that the case of State v. Shockley, supra, has been severely criticised by an eminent legal writer on evidence. In the fifth volume of Wigmore on Evidence, the distinguished author of that work, in a note to section 21, at page 10, in referring to the Shockley Case, says: “This is perhaps the most glaring example of our modern failures of justice to be found in the records of a decade. The defendant, who had in July, 1903, three times robbed street cars in Salt Lake City, was charged with the murder of two passengers in a fourth attempted robbery of a car in January, 1904. The defendant took the stand and confessed all the facts, endeavoring to make exculpation by declaring that he had only intended to ‘try to hit his arm.’ The verdict was reversed by the majority solely on two erroneous rulings of evidence — first, because the claim of witness’ privilege was required to be made by the defendant himself and not his counsel; and; secondly, because of improper cross-*35examination to past misconduct. Not only were tbe court’s rulings easily supportable on orthodox principles', but tbe Supreme Court majority opinion gave not even one word’s consideration to tbe question whether tbe alleged errors should have affected tbe verdict. On a perusal of tbe testimony of tbe defendant, full of self-justifying ethics of a reckless desperado, it is bard to say whether one is more aghast at tbe coldbloodedness of tbe robber in taking tbe lives of bis innocent victims, or tbe cold-bloodedness of tbe Supreme Court in mechanically grinding out a reversal without a regard to tbe demands of justice.” Referring to tbe same case again, at page 235, in tbe same volume, it is said: “Tbe decision makes confusion in tbe law, and helped to set free a confessed villian.” In view that neither tbe facts nor tbe reasons upon which the majority opinion in tbe Shockley Case is based are correctly reflected in the statement made by our distinguished critic, we are not only justified, but owe it as a duty to ourselves and to tbe profession generally, to correct tbe manifest errors into which tbe author by a misapprehension of tbe real facts that controlled tbe decision in that case has been betrayed. Referring to tbe statement that tbe decision was tbe means of setting free a “confessed villian,” it no doubt would require a little more evidence than tbe foregoing statement to convince tbe villian in question that be was set free. Tbe fact is that a legal conviction followed tbe reversal of tbe judgment, and tbe villian is now serving a life sentence at bard labor. We do not refer to this fact, however, for tbe purpose of showing that tbe criticism is for that reason not well founded. Wa refer to it for tbe sole purpose of showing that, where a critic is careless in bis statement of facts in one particular, be may be, and usually is, found to be so with regard to other particulars. A comparison of tbe real facts as they appeared in tbe Shockley Case with what they are either stated or assumed to be by Prof. Wigmore in bis criticism leaves no room to doubt the correctness of tbe foregoing observation. As a further illustration, we refer to tbe distinguished author’s statement that tbe defendant “bad in July, 1903, three *36times robbed street cars in Salt Lake City,” as wholly unjustified as a statement of fact. There was no evidence in the Shockley Case that the accused had committed the robberies attributed to' him prior to the one he was being tried for. On the contrary, the record shows that the prosecuting attorney merely asked the accused whether or not he had not committed such robberies, and, when the accused claimed his privilege and declined to answer the question, the district attorney, very improperly, plied the accused with further questions in which the attorney assumed and stated in the .presence of the jury that the witness declined to answer because a negative answer was impossible. It was in this connection that the question arose as to whether counsel or .accused could claim the privilege. By a series of questions adroitly put to the accused he was placed in such a. light before the jury that it would have been far better for him if he had either denied or admitted the assumed facts, and had not claimed his privilege. If such a course may be pursued when a witness claims his privilege, of what virtue is the much vaunted privilege ? Instead of shielding him from being compelled to give evidence against himself, the shield is turned into a sword, and the effect upon the jury is worse, if possible, than an admission of the crime which is assumed in the question would be. In this manner, therefore, and by this method, the accused is prevented from having a fair and impartial trial, since a jury whose passions are inflamed are utterly incapable of thenceforth judging his case with impartiality. But the distinguished author would, of course, not contend that, where a certain individual was on trial for an offense committed in January, evidence that he had committed some other offense in the preceding July would be competent upon the issue of guilt or innocence. As substantive proof no court would admit such evidence for that purpose, and no law writer would justify its admission. But the distinguished author contends that the error in the Shockley 'Case consists in' holding that the facts sought to be elicited hy the questions of the district attorney were proper upon the ground that they tended to affect the credibility of the *37accused as a witness. Tbis presents tbe crux of tbe Sbockley Case.
We venture tbe assertion tba,t no one can even cursorily read tbe majority opinion in tbat case and not bave tbé conviction forced upon bim tbat tbe court neither attempted to nor did follow any rule of cross-examination of a defendant in a criminal proceeding excepting tbe rule wbicb is supported by both reason and tbe weight of authority. Under our statute any witness, for the purpose of affecting bis credibility, may be asked whether or not be has previously been convicted of a felony. If be answers “Yes,” tbis ends tbe inquiry; and, if be answers “No,” tbe record of bis conviction is conclusive against bim. He must, however, answer tbe question, since to admit a former conviction cannot expose bim to further prosecution for tbat offense. But may a witness’ credibility also be affected or destroyed by mere insinuations or assumptions on tbe part of tbe cross-examiner tbat tbe witness has committed crimes generally, and, if tbe witness denies tbe assumptions or claims bis privilege, may be be further plied with questions by which it is further assumed tbat be denies because be dare not admit or claims tbe privilege because be cannot truly deny ? If such a course may be pursued, then no witness is safe from attack, and bis credibility may be seriously affected, if not destroyed, by mere insinuations and accusations without a scintilla of proof tbat be has committed any offense or has in any way been connected with its commission. Moreover, in asking generally whether crimes bave been committed by tbe witness, tbe prosecutor’s good faith may well be questioned, since it is well known tbat, in case a witness denies having been charged with or of having committed any crime of wbicb be has not been convicted, no proof with respect to such charge can be presented. From tbis tbe conclusion is inevitable tbat such questions, as a rule,' are asked for tbe sole purpose of prejudicing tbe defendant before tbe jury, and, if tbat is not tbe purpose, such nevertheless is tbe usual -effect. It was upon these grounds, and no other, tbat tbe majority of tbis court condemned tbe ques*38tions and rulings in the Shockley Case. If the position of our esteemed critic is correct, then any person by becoming a witness may be condemned of any crime without a hearing, and the presumption that all men are innocent until their guilt is established beyond a reasonable doubt is a mere fiction to be heeded only in the instructions to the jury, but may be entirely disregarded upon the trial of the case.
The professor’s answer to this, however, seems to be that, if the admission of the evidence was supportable upon what he is pleased to call “orthodox principles,” it makes no> difference whether it is desired for a legitimate or an illegitimate purpose. Moreover, as we construe the professor’s criticism, he, in effect, holds that, when one accused of crime in effect confesses the acts constituting the offense, no error, however gross, can be prejudicial in his case, and hence any court that under such circumstances will reverse a judgment of conviction is quite as cold-blooded as the criminal, and hence like he should be condemned. While we think the doctrine that mere technical errors should not in any case work a reversal is well founded and always should be observed, yet we cannot, without qualification, subscribe to the further doctrine that prejudicial error is legally impossible against one who while testifying in his own behalf admits the facts constituting the offense charged against him and for which he is on trial, and who has no explanation to offer except some matters which constitute no defense. Even in such a case the minds of the jurors may by improper questions and insinuations on the part of the prosecutor bo so inflamed against the accused that a fair and impartial trial at their hands is as impossible as it would be to reflect anything but a distorted image from a convex or concave mirror. According to our distinguished critic, if the evidence sought to be elicited is proper according to some “orthodox principles,” all else is immaterial. In this connection it may be said that it may well be that a conviction is in accordance with “orthodox principles,” and thus without a flaw in the state of Illinois, and yet not without error in the state of Utah. Courts must have due regard for local statutes which *39may affect both procedure and substantive law. As we bave seen, tbe two sections to wbieb we bave referred clearly restrict tbe scope of tbe cross-examination of one accused of crime, yet Prof. Wigmore intimates that these sections were ignored in tbe Shockley Case because tbe examination there was not permitted to be conducted under the orthodox rule that, when tbe accused becomes a witness, be waives tbe constitutional injunction that be shall not be compelled to testify against himself. Tbe language of our statute is so clear that no construction is permissible. In civil cases, when tbe cross-examination is carried beyond its legitimate scope, tbe cross-examiner makes tbe witness bis own, and, under tbe provisions of .the foregoing statute, when this occurs in a criminal case, tbe accused is in legal effect compelled to testify against himself, and bis constitutional rights are thus ignored. To prevent this result is tbe manifest purpose of our statute. If this be not its purpose, it has none. But there is still another local statute which plays a very important part in homicide cases where tbe charge is murder in tbe first degree. Section 4162, which fixes tbe penalty for murder in tbe first degree, provides: “Every person guilty of murder in tbe first degree shall suffer death, or, upon tbe recommendation of tbe jury, may be imprisoned at bard labor in tbe state prison for life, in tbe discretion of tbe court.” Tbe legal effect of this section is that it conditionally provides for two penalties for murder in tbe first degree. In Calton v. Utah, 130 U. S. 83, 9 Sup. Ct. 435, 32 L. Ed. 870, tbe Supreme Court of tbe United States, in construing tbe foregoing provision, held that it is tbe duty of tbe trial court in every case where tbe charge is first degree murder to instruct tbe jury that it is their province to recommend that tbe punish; ment be imprisonment for life, that any accused person has a right to bave tbe jury so instructed, and that tbe right is a substantial one, which cannot be ignored. In this state, therefore, upon a plea of not guilty, although tbe accused confesses tbe facts constituting first degree murder at tbe trial, be still has tbe right to bave tbe jury recommend that tbe punishment be other than death. If, as tbe Supreme *40Court of tbe United States bolds, this is a substantial right, then it follows that the accused has the right to have the recommendation leased upon legal evidence which is limited to the crime for which he is on trial. If the minds of the jurors may be influenced by showing that the accused was guilty of other offenses for which the penalty amounts to no more than a term in the state prison, then the jury may easily be induced to permit the death penalty to be inflicted because lesser crimes have been insinuated but not proved against the accused, and not because the jury conscientiously believe and are satisfied beyond a reasonable doubt that the. accused should be permitted to suffer death for the sole reason that he committed the particular offense for he was tried. The accused is entitled to the unbiased judgment of the jury upon this question, and the court may, in case he thinks the accused should suffer death, disregard the recommendation.
But it would seem from Prof. Wigmore’s criticism that one of the principal offenses committed by the majority of this court in reversing the judgment in the Shockley Case was that the court did not make clear wherein the defendant was prejudiced by the rulings of the trial court. Here again Prof. Wigmore seems quite insincere. In his criticism it is assumed that the defendant was guilty of at least three other crimes which were committed by him in the preceding year. This assumption of fact is based entirely upon the mere insinuations and accusations of the prosecuting attorney, and the refusal of the defendant to answer the accusations upon the ground of privilege. If any proof of the fact were necessary that such insinuations and assertions of crime are prejudicial to the accused, Prof. Wigmore, wittingly or unwittingly,' has furnished it. If mere insinuations or accusations are sufficient proof of guilt for an eminent writer on the law and rules of evidence, why should they not be sufficient for ordinary laymen who sit as jurors ? If the mere fact that a court feels impelled to disagree with the theories of the eminent critic and is unwilling to permit mere insinuations to stand as proof will induce him to enter upon a most fervid and intemperate criticism of the court, what will the *41effect of sucb insinuations be upon the minds of ordinary men who abhor crime and who sit as jurors in the case? They, like Prof. Wigmore, will condemn the accused, regardless of proof or consequences.
But, entirely apart from the foregoing observations, Prof. Wigmore’s criticisms by reason of their intemperate character are valueless. His criticisms are of the kind which engender heat, not light; they irritate, but impart no information. They appeal to the passions rather than invite cool reflection, and in the long run will result in betraying courts into committing errors rather than to help them in avoiding them. It may be that our criminal procedure is out of date, and that by reason of the present methods of administering our criminal laws some of those who commit crimes may escape. Courts, however, can neither change the laws nor the rales of evidence. In their daily task of dealing with concrete cases they must apply tíb-th substantive law and procedure so as to reflect justice. Courts must have some regard for consequences in particular cases, and for that purpose the rales must receive a practical application. Critics like Prof. Wigmore revel in mere abstract theories. Their theories may or may not be practical in view of all the facts and circumstances of the particular case, and because the court finds it impossible to follow a particular theory to the same extent and under different circumstances in all cases the court must as a matter of course be condemned by the critics. But, whether our criminal laws and procedure be changed or not, the underlying principle that no one may be convicted without a fair and impartial trial by an impartial jury will always remain. Whatever the future may develop, one thing is certain — that no court that is' worthy of the respect and confidence of the public will affirm a judgment of conviction where it is clearly made to appear that the conviction is against law and is the result of prejudice, and unfair means used at the trial. To reverse a judgment of conviction without sufficient reasons may be'a grievous blunder and a wrong against society, but to affirm a conviction which is the result of unfair means and prejudice *42engendered at the trial is more than a wrong — it would be a crime.
Lest we be again misunderstood, we repeat that we have not held, and do not now hold, that under our statute the credibility of one on trial for a crime when a witness may not be attacked in the same manner as any other witness. What we do hold is that while a court may seriously err in its rulings in so far as it affects the rights of a mere witness without necessarily influencing the jury but that such interference with one on trial for a crime may, and ordinarily does, influence the judgment of jurors, and when it is manifest from the record that such has been the case, the verdict and judgment cannot be permitted to stand. While in the case at bar the court erred in permitting the cross-examination as herein indicated, yet, in view that the judgment must be reversed upon other grounds, it is not necessary to pass upon the question whether this error was of such a nature that in view of the whole record we would reverse the case upon that ground alone.
The contention that the court erred in refusing to grant a new trial for the reasons urged by counsel, in view of the conclusions reached, needs no further consideration. While one or two of the grounds argued by counsel, when considered in connection with other phases of the ease, are more or less serious, yet, in view that the case must be retried, and as the alleged errors cannot arise in the same way on a retrial, it is not necessary for us to either discuss or pass upon them. All other assignments have been carefully considered, and we are of the opinion that the court has committed no error in regard to them.
The judgment is reversed, with directions to the district court to grant a new trial, and to proceed with the case in accordance with the views herein expressed.
McOARTY, L, concurs.
STRAUP, C. J.
I concur in the result. The information contained three counts. The first charged a homicide by beating and bruis*43ing. Tbe second by poisoning. The third by the combined and co-operating canses of both. When the state rested, the court, on the defendant’s motion, required it to elect. It elected on the third count. By such action the first and second counts were withdrawn. The case was submitted to the jury on the third count only. The defendant requested the court to charge that, before the defendant could properly be convicted, the jury must find that both the poisoning and the beating combined to cause the death of the deceased, and as in the third count alleged, and that it was not sufficient that a portion of the jury believed that the death was caused by the beating only, and a portion by poisoning only. The request was refused. The court charged that the defendant could be found guilty of murder if the jury believed the death w'as caused “by either one, or a combination of both” beating and poisoning, as in the third count alleged.
There is no allegation in the third count that the death was caused by beating alone. Neither is there an allegation that it was caused by poisoning alone. The allegations in the third count in this regard are: That by the beating on the 26th day of November and the administering of the poison on the 27th day of November the deceased “became mortally sick and distempered in her body,'” and languished until the 8th day of December, when she died. When the court told the jury that they might find the defendant guilty of causing the death by beating alone, the jury had the right to assume that there was sufficient evidence to justify such a finding. So did they when the court told them that they might find him guilty of causing the death by poisoning alone. We have no means of determining whether the verdict was founded upon one or the other cause, or upon both. If it was founded upon the cause of beating alone, as the court told the jury they might find, the verdict lacks support in the evidence. As pointed out by Mr. Justice Feicic, there is not sufficient evidence to support a' verdict on that cause alone. Neither from the testimony of the witnesses, nor from the character of the discolorations or bruises on the deceased’s body, nor from any external or internal conditions of her body, *44as testified to by all tbe witnesses in the case, was the jury justified in such a finding. To the contrary, ©very physician in the case, both on the part of the state and the defendant, testified that the effects of the beating did not cause the death, and that there were no external nor internal conditions of the body indicating such a cause.
Nor in my judgment was there sufficient evidence to justify a finding that the death was caused by mercurial poison administered by the defendant- — the particular kind of poisoning alleged — and the only - poison claimed by the state that the defendant administered to the deceased. The conviction of the defendant on this cause rests alone upon circumstantial evidence. The administering of the poison by the defendant, and the death from such a cause, may, of course, be shown by circumstantial evidence. As has often been said by the courts, and the text-writers, it is not to be expected that witnesses should be called to state that they saw1 the deadly poison administered by the prisoner or mixed up by him openly before them, nor is it essential to show that a quantity of poison sufficient to cause death was found in the body, if other sufficient facts lead to the conclusion that the death was occasioned by poisoning. However, in determining the question of sufficiency of the evidence, when a conviction rests alone upon circumstantial evidence, proper regard must be had to the elementary rules of evidence governing that kind of evidence and the principles upon which they are founded. Circumstantial evidence, frequently termed “indirect evidence,” consists in reasoning from facts which are proved to establish such as are conjectured or asserted to exist. In the case of direct evidence the facts apply directly to the factum probandum. Circumstantial evidence is proof of a minor fact, which, by indirection, logically or naturally demonstrates the factum pn'obcmdum. 3 Enc. Ev. 63. That is, from the proof of certain facts in a given case may be inferred other connected facts which usually and reasonably follow according to the common experience of mankind. Hence, one of the most essential rules of circumstantial evidence, recognized by the courts and text-writers generally, is *45that facts claimed as tbe basis of a legal inference must be proved, and connected with the factum probandum. That is, the inferences, by which it is sought to establish a fact conjectured or asserted to exist, must be drawn from proven facts, not from mere inferences. Another well-recognized rule is that, before it can be said that circumstances prove anything, they must agree with and support the hypothesis sought to be proved; and in a criminal case they must not only concur and be consistent with and point to the defendant’s guilt, but they must also be inconsistent with any other reasonable conclusion or hypothesis, including that of innocence. Hence it has many times been said by the courts that “circumstantial evidence in a criminal case is of no value if the circumstances are consistent with either the hypothesis of innocence or the hypothesis of guilt; nor is it enough that the hypothesis of guilt will account for all the facts proven. Much less does it afford a just ground for conviction that, unless a verdict of guilty is returned, the evidence in the case will leave the crime shrouded in mystery.” (3 Ency. Ev., 92.)
It was shown by the testimony on the part of the state that in a trunk kept in the house, and to which the defendant and the deceased had equal access, were kept tablets of bichloride of mercury, and a box or bottle of some kind of medicine or drug used to produce miscarriages or abortions. The mercury tablets were procured by the defendant something more than a year before the alleged homicide upon a prescription for an injury to his' hand. The box or bottle was procured, just when is not shown, by the deceased from or through the advice of a lady relative. After the death of the deceased, this box or bottle was taken from the trunk by a sister of the deceased, Amanda Ward, who was a witness for the state and a very hostile witness against the defendant. She gave it to the physician who attended the deceased in her last illness, and who was also a witness for the state. The physician testified that he returned it to Mrs. Ward. She testified that he did not. Neither claimed to have possession of it, or any knowledge of where it was, at the time of the trial. The nature of the contents of the box or bottle were not disclosed, except, *46as testified to by Mrs. Ward, that it contained some drug or medicine obtained by, or given to, the deceased to produce abortions or miscarriages. The deceased on different occasions asserted to her sister and to others that because of her ill health, the defendant’s cruelty towards her, and their poverty she would not give birth to more children. At moments of despondency she also threatened self-destruction. The deceased for some time had been in ill health. On the 26th day of November the defendant, without cause, became angered, and cruelly beat and kicked her, the details of which are referred to by Mr. Justice Erice. She afterwards went downtown shopping. The defendant went to his work. That night she, with some of the children, slept in one room, and the defendant, with others, in another room where the trunk was. The next morning she arose and got breakfast. The defendant ate his breakfast and went to work. About ten or eleven o’clock of the forenoon of that day the deceased felt so ill that she sent for her sister. When she arrived, she found her preparing the noonday meal. As testified to by her sister, the deceased was then menstruating, and was sick and weak, and looked “very delicate and pale.” She advised her to go to bed and lie down. About fifteen or twenty minutes before twelve o’clock she went into a room adjoining the kitchen, and lay down on the bed. Her sister, after making some examination of her, departed. Before lying down, the deceased obtained a glass of water and placed it on the sewing machine near the bed. The defendant returned from his work some time after twelve o’clock for his noonday meal. He entered the kitchen, where his eldest daughter was finishing the preparation of the meal. After talking and playing with some of the children, he asked where the deceased was. He Was told that she was lying down in the adjoining room. According to his testimony, and that of his daughter, he stepped to the door leading to the room where the deceased was, placed his hand on the door casing, looked in, and, thinking the deceased was asleep-, walked back, sat down, and ate his lunch. According to the oral dying statement of the deceased, she was lying on the bed with her eyes closed, and, *47on opening them, she saw the defendant leaving her bedside and walking ont of the room. After he ate his lnnch, he went back to work. In her oral dying statement she further stated that, after the defendant left the room, she took a drink from the glass of water near the bed, and that afterwards “she was taken much worse and continued to be worse from that time on.” At about two o’clock she began vomiting, and purging blood, strings of blood, and greenish looking matter, and complained of much pain. The daughter sent for a neighbor, Mrs. Wunderlich, and also for the deceased’s sister, Mrs. Ward. Mrs. Wunderlich, a witness for the state, testified that, when the deceased vomited, she noticed, among other things, greenish looking matter, and smelled the odor of oil of savin. It was shown that oil of savin has a strong and peculiar odor, and, when taken in sufficient quantities, will produce miscarriages or abortions, and that it is an irritant poison. Mrs. Wunderlich also testified that in giving the deceased water to drink she, thinking that the water in the glass near the bed was stale, threw it away, left the glass to bo washed, and got fresh water for her; that the glass emptied by her was about one^quarter full; that she noticed no sediment in the glass and nothing unusual about the water, except it looked stale. After the deceased’s sister arrived a physician was telephoned for at about 2:30 o’clock. He sent some medicine — what kind is not shown — and visited the deceased at about seven or seven thirty in the evening. He found her “in a good deal of distress, was nauseated, was vomiting some time, and was purging from the bowels, complained of pain.” At noon of the next day the deceased was removed to the hospital, where she died on the 8th day of December. An autopsy the next day and a chemical analysis of different parts of the body a few days thereafter were made, but no trace of mercury was found. The membranes and tissues of the stomach and intestines were found congested and inflamed, which conditions, as testified to by the physicians and the chemist, were indicative of the effects of an irritant poison, and that bichloride of mercury w'as such a poison. But they also testified that oil of savin, and other drugs used *48to produce abortions and miscarriages, were also an irritant poison, and, if taken in sufficient quantities, would produce the same conditions found in tbe stomach and intestines, and would occasion the same symptoms of vomiting and purging as were manifested by the deceased. It was further made to appear, and as stated by Wharton & Stille in their work on Medical Jurisprudence, .that oil of savin “has a strong, peculiar and heavy odor, and a nauseous, resinous, and bitter taste. . . . When administered in large doses, signs of irritant poisoning are manifested, such as heat at the stomach, epigastric and abdominal pains, vomiting of greenish matters, abundant and often bloody stools, flow of saliva, and, in fact, thé well-known signs of gastro-intestinal irritation,” and that “after death there are found, in general, undoubted evidences of inflammation of the stomach and intestines,” and that the strong odor is quite distinct in the vomit and urine. True, the autopsy did not disclose any condition of pregnancy, but the physicians testified that the deceased might have been ten or fifteen days in pregnancy, and all indications thereof removed by an accelerated menstrual flow.
Upon this evidence it is contended by the state that it is demonstrated that the deceased died of mercurial poisoning, and that the poison was administered to her by the defendant at about noon on the 27th day of November, when he stood at the doorway of, or went into, the room where the deceased was lying on the bed, and by then and there dropping a bichloride of mercury tablet in the glass of water standing on the sewing machine near her bed. It is conceded by the state that, if the defendant did not then place bichloride of mercury in the glass of water, there is no evidence to show that he administered any poison to the deceased. It is also conceded that he did not then go into the room or to the trunk where the mercurial tablets were, but it is conjectured by the state that he took some of them from the trunk when he arose that morning and before he went to work, or the night before, and had them about his person when at noon he returned from work for his noonday meal. To convict the defendant of causing the deceased’s death by poisoning, the state was there*49fore required to prove that the deceased died of mercurial poisoning, and tba.t such a poison was administered to her by him. There being no trace of mercury found in her body, the fact of death caused by mercurial poisoning was sought, and is claimed, to be established by other proven facts— the inflamed and congested condition of the membranes and tissues of the stomach and intestines, and the symptoms of vomiting and purging and of pain manifested and complained of by the deceased. But upon the evidence adduced by the state such proven facts equally well point to the cause of death by an irritant poisoning of a drug used to produce abortion. The proven facts of the state equally well point to two hypotheses, one consistent with the defendant’s guilt, the other with his innocence, in which case the circumstantial evidence of the fact sought or claimed to be established is, if not valueless, insufficient to support it.
Even though it should be assumed that the cause of death by mercurial poisoning was sufficiently established, yet it was also necessary to show that the defendant administered the poison. Again, that fact was also sought, and is claimed, to be established by other proven facts. The proven facts which it is claimed by indirection logically or naturally demonstrate the factum probandum — the administering of the poison by the defendant — are: Bichloride of mercury tablets were found in the trunk to which the deceased' and the defendant had equal access, and which were there for more than a year, and conceded not to have been procured for any wrongful or unlawful purpose; the deceased, sick, very delicate and pale, was lying on the bed with her eyes closed; the defendant, when he returned from his work, entered the kitchen, inquired where the deceased was, and, on being told that she was lying down in the adjoining room, went to the doorway of or into her room, looked at her, walked away, sat down in the kitchen, ate his lunch, and went back to work; after the defendant was at the doorway or in the’room, the deceased drank from the glass of water procured by herself, whereupon she became worse, and at about two o’clock began *50vomiting and purging, and complained of pain. There is no logical nor' natural connection between the proven facts and the factum probandum, the dropping of the mercurial tablets in the glass of water by the defendant while he was at the doorway or in the deceased’s room. This is not a case where a person without ailment or complaints of pain, or in an ordinarily normal condition, drinks or eats something, and is suddenly or shortly thereafter made sick and caused to suffer pain. The deceased, when she lay down on the bed, and before it is claimed that the defendant placed the poisonous tablets in the glass, was from some cause, according to all the evidence, a very sick woman, and was in a weak, pale, and trembling condition. Her symptoms of pain, of vomiting and purging, are equally attributable to causes other than that of her drinking water out of the glass after the defendant was at the door of, or left, her room. The defendant’s going to the doorway, or in the deceased’s room under the circumstances disclosed is not an unusual nor a suspicious circumstance. It is so usual and natural that it does not even support nor agree with the hypothesis of guilt, and much less is not inconsistent but wholly consistent with innocence.
The facts that the defendant took bichloride of mercury tablets from the trunk, and that the deceased drank mercurial or other poison from the glass themselves rest, not upon any proven facts, but upon mere inferences. That is the fact that the deceased drank poison from the glass of water is itself only an inference deduced from the facts that after she drank from the glass she became worse, began vomiting and purging, and had pain. Yet from such inferences it is sought to infer the further fact, the factum probandum, the placing of the tablets in the glass of water by the defendant, and thus we have an inference deduced, not from a proven fact, but from a mere inference, and have an inference upon an inference.
I cannot, yield assent to the proposition contended for by the state that “as to the poisoning it was necessary on the part of the state to satisfy the jury of' two things: First, that the defendant had possession of, or access to the poison; sec-*51on cl, that he had the opportunity to administer the poison to Mary Vance,” the deceased. Nor could I yield assent, if added to these two, were the third, that the deceased died of mercurial poisoning. That is to say,' it is not sufficient to sustain a charge of murder by poisoning by the mere proof of facts that the deceased died of poisoning, that the defendant under no suspicious circumstances merely had poison in his possession, or access to the poison, and that he had the opportunity to administer it to the deceased. Bichloride of mercury, carbolic acid, and other antiseptics are not unusual articles found in many households for antiseptic or mechanical purposes. If the contention of the state is sound, should the death of one member of a family having such poisons about the house be caused by such hind of poisoning, every other member of the family who had possession of, or access to, the poison, and had the opportunity to administer it to the deceased, might properly be found guilty, of murder. Courts do not, and should not, sustain convictions on such evidence alone. We have been referred to no case supporting such a contention of the state. Neither the case of Zoldoske v. State, 82 Wis. 580, 52 N. W. 778, nor the ease of Commonwealth v. Danz, 211 Pa. 507, 60 Atl. 1070, cited by the state, supports it. Those cases are much stronger in their facts than is the case in hand. In the first it was shown by direct evidence that the candy containing the poison and which was eaten by the deceased was given to her by the defendant; in the other, that the defendant put the poison in the deceased’s cup of coffee. So are the cases of the State v. Van Tassel, 103 Iowa, 6, 72 N. W. 497, where there was also some direct evidence that the defendant substituted a poisonous drug for the medicine left for the deceased by a physician, State v. Best, 111 N. C. 638, 15 S. E. 930, where it was shown by direct evidence that flour, bread, and dough from which the deceased had eaten was in the possession of the defendant and contained poison, and that it was administered to her hy him, and in the case of Speights v. State, 41 Tex. Cr. R. 923, 54 S. W. 595, where the defendant’ administered poisonous water to the deceased out of a dipper. A person having bichloride of *52mercury tablets about tbe house, conceded not to have been procured and kept under any suspicious circumstances, nor lor any wrongful or unlawful purpose, is one thing. One mixing a deadly poison with candy, bread, water, or other substance, or substituting a deadly poison for medicine left for a patient by a physician, and giving it, or causing it to be given to another, or who gives to another candy, bread, water, or other substance, containing a deadly poison, to be eaten or drank, is quite another thing. Having bichloride of mercury tablets about the house is not an unusual thing. Mixing poison with bread, candy, or other substance to be eaten, or drank, is not only a very unusual thing, but is also inconsistent with innocence. Besides, in the cases referred to, there were in some of them other facts and circumstances and in others admissions, confessions, and conduct of the suspected parties which were not only consistent with their guilt, but which were also wholly inconsistent with their innocence. The case here on the facts is no stronger than the cases of State v. Bertoch, 112 Iowa, 195, 83 N. W. 967; State v. Nesenhener, 164 Mo. 461, 65 S. W. 230, and Pitts v. State, 43 Miss. 472, where the evidence was held insufficient to support a conviction.
The deceased made two written dying statements, one written by her sister, the other written by an assistant county attorney, and signed by the deceased in his presence, and in the presence of a deputy sheriff. In neither statement did the deceased make any reference to any fact or circumstance on the subject of poisoning. Both statements were confined alone to the transaction of the defendant’s striking and kicking her on the 26th day of November, and to the things then said and done. After she had signed the statement in the presence of the attorney and sheriff, and after they had departed, the latter called the former’s attention to the fact that the deceased in her statement had said nothing on the subject of poisoning, and suggested that they return and ask her about it. They did so. . In response to questions asked her on such subject she made the oral statement heretofore referred to. I, of course, think the evidence of such state*53ment was properly admitted. But the fact that no reference to any circumstance of poisoning was made by her in the written statements is significant. Still more important is the fact that, when she was specifically interrogated on that subject, the statement made by her in that regard was of so little consequence. All that was said by her on that subject was that she saw the defendant leaving her bedside, and thereafter she took a drink from the glass of water, whereupon she “was taken much worse, and continued to be worse from that time on.” Whether she took a sip, swallow, or what quantity of water was drunk by her from the glass is not made to appear. It was shown by the physicians, and as stated by Wharton & Stille on Medical Jurisprudence, that “immediately after swallowing a poisonous dose of bichloride of mercury there is experienced a strong metallic taste, and a painful spasmodic contraction in the throat.” And, as also made to appear, such taste is very pronounced and unusual. It is highly improbable that the deceased could have drunk a sufficient quantity of mercurially poisoned water to produce the claimed corrosive and poisonous effects and symptoms without noticing or detecting the metallic taste, or some unnatural or unusual taste. Yet no word is said by her to the effect that she noticed anything unusual or unnatural about the taste, of the water drank by her, nor is there anything said by her that after she drank it she experienced or felt any pain, or any discomfort, except that she “was taken much worse.”
There is considerable evidence to show that the defendant treated the deceased with extreme cruelty. But I cannot say that he shall be shot to death because he is not a good man, or because he was a wife beater. The state cannot legally demand the life of the vilest human creature except upon sufficient proof of his guilt of murder. Such proof I think is here lacking. While the state sufficiently demonstrated that the defendant struck and kicked the deceased, it just as clearly demonstrated by all the evidence in its behalf, including the testimony of its physicians, that the effects of the beating and kicking did not and could not have caused the death. *54Tbe evident theory advanced by its learned physicians was that the beating and kicldng did not cause the death. So, too, did they testify that the poisoning alone did not cause it. When asked what did cause it, they asserted that it was caused by the combined and co-operating causes of both the kicking and the poisoning. When asked how the two could, and probably did, combine and co-operate so as to cause the death, they asserted that the beating and the kicldng of the deceased so lowered her vitality that her system could not resist the corrosive and poisonous effects of the deadly poison, and, in effect stated that, if she had not been kicked and beaten and her vitality lowered, the poisoning probably would not have killed her. As well assert that, because of his reduced or lowered vitality, the victim could not resist the penetration of the bullet as it was shot from the gun into his body ñor the destructive effects of the muscles and tissues, as it plowed its way through a vital organ. Eire burns. It bums the tissues and the cells of a strong as well as of a weak body; and they alike must also yield to the corrosive and distructive effects of such a deadly poison as bichloride of mercury. I think this is a case of too much conjecture and not enough evidence. I am of the opinion that there ought to be a retrial.
In view' of what has been said concerning the Shockley Case, I feel obliged to make but this observation. In that case the principal assigned errors related to the refusal of the court to submit to the jury the issue of self-defense, the rulings made against the defendant subjecting him to a cross-examination respecting past criminal transactions which were separate and distinct from that for which he was on trial, and the holding that immunity from answering such incrimi-native questions could only be claimed by the defendant himself, and not through his counsel. The majority members of the court held that there was no evidence to justify the submission of the case to the jury on the issue of self-defense, and that, therefore, the trial court committed no error in its refusal to submit the case to the jury on such theory. The dissenting member of the court was of a different opinion. *55He held that reversible and prejudicial error was committed by the trial court in such refusal, and on that ground concurred in the judgment of reversal and in remanding the case for a new trial. The majority members of the court held that reversible and prejudicial error was committed by the trial court in the rulings respecting the cross-examination of the defendant, and in not permitting him to claim immunity, through his counsel, from answering ineriminative questions relating to past criminal transactions which were wholly separate and distinct from that for which he was on trial. Because of such erroneous rulings the majority members reversed the judgment, and remanded the case for a new trial. On such holding on such ground the dissenting member did not concur. I thought then, and I think now, that the conclusion reached by the majority members was right. Our reasons therefor are stated in the prevailing opinion. The legal principles involved are very simple. They are: (1) When a defendant is a witness, immunity from answering ineriminative questions relating to criminal transactions or crimes wholly separate and distinct from that for which he is on trial may be claimed through his counsel; and (2) while a conviction of prior and other offenses may be shown to affect credibility, or impeach the witness, a. mere arrest, or specific acts, or past conduct, tending to show' the commission of other offenses, may not be shown.
These principles are elementary, and are supported by the clear weight of authority. They were violated by the rulings referred to. No criticism is made of our conclusion holding them erroneous. The criticism made of the case is that we “gave not even one word’s consideration to the question whether the alleged errors should have affected the verdict.” The criticism does not fairly reflect the decision. We did not reverse the judgment regardless of the question of prejudice. In the prevailing opinion are stated our reasons why we thought the committed errors were prejudicial and harmful and affected the verdict. We there stated that, under our statute, “every person guilty of murder in. the first degree shall suffer death, or, upon recommendation of the jury, may *56be imprisoned at bard labor in the state prison for life, in tbe discretion of the court,” and tbat it was held by the Supreme Court of the United States to be reversible error in the failure of the court to charge the jury the substance and effect of the statute. The verdict rendered by the jury in the Shockley Case was “Guilty of murder in the first degree,” without any recommendation. Upon the court’s refusal to grant the motion for a new trial, it became the imperative duty of the court to pronounce upon that verdict a judgment inflicting the death penalty, which was done. In the prevailing opinion, and in speaking of the prejudicial effect of the erroneous rulings of the court referred to-, we said: “Whatever deposition, if any, there may have been on the part of the jurors to make the said recommendation, may have been entirely overcome and removed by reason of these rulings of the court.” That is to say, though the jurors upon all the evidence in the case were satisfied beyond a reasonable doubt that the defendant was guilty of murder in the first degree as charged in the information, and though upon all the evidence there was no substantial conflict on such question, yet, by reason of the improper cross-examination, the jury may have been improperly led to the conclusion that the defendant was also guilty of other separate and distinct criminal transactions and crimes of which there was no evidence, but which were so vividly and persistently referred to by the improper incriminative questions on cross-examination, and for that reason the jury may have refused to make the- recommendation that the defendant be imprisoned at hard labor for life. That is the view we took of it, and for such reasons we held the erroneous rulings prejudicial and said so. Some may not agree with us on that, but a reading of the record in that case clearly shows that the improper cross-examination strongly tended to produce, and in all probability did produce, just such effect. After the court had overruled all and repeated objections interposed by the defendant’s counsel to the incriminative questions, and held with the contention of the district attorney that the defendant must himself claim the privilege, and after the defendant had himself claimed *57it, and bad declined to answer tbe questions which in themselves were pregnant and teeming with inquiries respecting ineriminative conduct and criminal transactions, the court not only permitted the district attorney, over objections, to further inquire of the defendant, and to compel him to answer whether he so declined on the ground that it tended to incriminate him, but the court itself turned to the defendant and demanded of him: “Do you claim that privilege and decline to answer the question on the ground that it tends to incriminate you and subject you to punishment for a felony?” And when the defendant thereupon answered, “I decline to answer the question upon that ground,” the district attorney, still not satisfied that the privilege and immunity had yet been sufficiently claimed, immediately asked the defendant, “Upon the ground it tends to incriminate you ?” And so each time, on three or four occasions, when such ineriminative questions were asked, such proceeding was substantially repeated; and, after the defendant had declined to answer such questions which were in themselves replete and freighted with inquiries respecting incriminating conduct and criminal transactions, the defendant was further .compelled to answer, to the effect, that he did so decline because to answer the questions would incriminate him and subject him to the punishment for a felony. The baleful effect produced by such proceedings would be no worse had the court turned to the defendant, and said: “If you are guilty of the criminal conduct and offenses so clearly suggested by the questions propounded to you, you need not answer, but, if you are not guilty of them, you must answer. Now with this admonition, do you, or do you not, decline to answer ?” The substance of all these matters referred to appears in the prevailing opinion. The criticism therefore made, that we did not consider the question of prejudicial effect of the erroneous rulings, is not founded .on truth. Neither the gi’ounds upon which we reached the conclusion that the rulings were erroneous nor upon which we held them prejudicial have, as yet, been criticised. The complaint made that we granted a new trial to a bad man, 'and did not hold any kind *58of a conviction of such, a man a proper and. legal conviction, is not worthy of notice.