— At the April term, 1910, of the criminal court of Jackson county, the defendant was tried for the murder, by poison, of Colonel Thomas H. Swope (spoken of hereafter as Col. Swope), found guilty of murder in the first degree, and sentenced to life imprisonment in the penitentiary. He complains that the trial court committed numerous errors in the course of the proceedings, and on his appeal to this .court asks that the judgment be reversed.
The case comes here with a voluminous record of more than four thousand typewritten pages, representing a trial that iasted more than a month.
*215The defendant is a physician, thirty-nine years old, resident in Kansas City. Col. Swope, the deceased, was a bachelor, eighty-two years old at the time of his death, which occurred October 3, 1909-. He possessed a fortune of about three and one-half millions. He resided at the home of Mrs. Logan Swope, widow of his brother, in Independence, Mo., a town about ten miles from Kansas City. The family of Mrs. Logan Swope consisted of her five children, Lucy Lee, Margaret, Stella and Sarah, daughters; Chrisman, a son, thirty-one years old; Col. Moss Hunton (hereafter spoken of as Col. Hunton), a bachelor cousin about sixty years old, and Col. Swope, the deceased. Mrs. Swope has two married children, a son, Tom, living on a farm near Independence, and Frances, wife of the defendant.
Col. Swope was somewhat infirm in health. He received a fall on September 4,1909, resulting in an injury not serious in character, but which confined him to the house. On September 13th, at the defendant’s suggestion, a trained nurse, Miss Kellar, took ¡charge of the case, and continued in charge as nurse, with de- » fendant as attending physician, until Col. Swope.’s death, which occurred October 3d following. On October 1st, two days before Col. Swope’s death, Col. Hunton was seized with a stroke of apoplexy at the supper table, and died that same night, while being attended by the defendant and Doctor Twyman, the family physician of Mrs. Swope. About two months later an epidemic of typhoid fever appeared in the Swope household, and brought to bed, between December 2d and December 18th, nine persons, namely, Chris-man Swope, Margaret, Lucy Lee, Sarah and Stella Swope; Leonora Copriclge, a colored servant; Georgia Compton, a seamstress; Nora Dickson, a visiting cousin, and Mildred Fox, a transient visitor. All of these recovered excepting Chrisman Swope, who died December 6, 1909.
*216It is claimed by the State that the defendant caused the death of Col. Hunton by excessive bleeding, while treating him for apoplexy; that he caused the deaths of Col. Swope and Chrisman Swope by poison; that he inoculated the Swope household with typhoid germs, and that he attempted the life of Margaret Swope-by injecting into her arm what he supposed were diptheria germs, and also by poison administered in the form of medicine. That the defendant did all these thing's feloniously, with the malicious intent to kill, and that he did them for the purpose of securing to his wife the bulk of the fortune of Col. Swope.
The State introduced evidence tending to show that defendant knew, from a conversation had with Col. Swope on September 12, 1909, that Col. Swope had made a will, by the terms of which he had provided liberally, by way of specific legacies, for his various nephews and nieces, including the Swope children, and had also bequeathed to them, in equal shares, his residxxary estate, estimated at one and one-half millions, but that he intended to make a new will and give this residuary portion to charity. The residuary legatees were ten in number. There was also evidence that the defendant knew that Col. Hunton was one of the executors of the existing will. Had Col. Swope made a new will, and, by its terms, devoted the residuary estate to charity, the result would have been to dixninish the portion of each niece and nephew, including Frances, wife of defendant, about $150,000. The death of any brother or sister of Frances, unmarried and childless, would enlarge her fortune by the axnount she would inherit from such brother or sister.
It is "the theory of the State that the defendaixt committed the various alleged crimes above-mexxtionecl in order to secure these results to his wife, and that one common motive, avarice, prompted them all. ''The State introduced testimoxxy tending to prove that on September 13, .1909, defendant purchased from his *217druggist four five-grain capsules of cyanide of potash, a deadly poison; that on the next day, September 14th, he purchased from the same druggist a box of five-grain Holadin ¡capsules, a digestive medicine; that on the morning of October 3d following, he handed to the nurse in charge of Col. Swope a capsule, with directions to give it to Col. Swope, stating to the nurse that it was a digestive medicine; that in about fifteen minutes after taking the aforesaid capsule, which the nurse gave him at 8:30 in the morning, Col. Swope had a convulsion, followed by coma which lasted until death ensued, at about seven o’clock that evening; that an autopsy was had of the body of Col. Swope on January 12th following, and that chemical tests of his organs made later disclosed the presence of nearly a grain of strychnine in the liver, a small amount, one two-hundredth of a grain, in the stomach, and also, in the stomach, “a very minute quantity” — spoken of as “a trace ’ ’— of cyanide. The evidence for the State tended further to prove that on November 12th the defendant secured from a brother physician a culture tube in which were planted typhoid germs, also a tube of what he supposed were diphtheria germs, but which were in fact pus germs; that defendant visited the Swope home on November 21st and November 25th, preceding the typhoid outbreak, December 2d; that on December 4th defendant purchased from the same druggist six five-grain capsules of cyanide of potash; that on Sunday, December 5th, he administered to Chrisman Swope, then seriously ill with typhoid fever, a capsule of what he claimed was Chrisman’s regular fever medicine; that shortly afterwards Chrisman had a convulsion similar in many respects to the one which Col. Swope had on October 3d, and which followed the taking of a capsule by Col. Swope, as above stated; that Chrisman recovered from this convulsion, • but grew worse, developing delirium, with rational intervals ; that on the next day, Monday, defendant handed *218to the nurse in charge a capsule which she gave to Chrisman, not followed by any marked effects; that Chrisman died at ten o ’clock at night, Monday, December 6th. Further, that on December 12th defendant gave Margaret Swope, who was ill with typhoid, a hypodermic injection of what he claimed was camphorated oil; that this was followed by inflammation about ' the point of puncture, and severe pain and swelling of the arm. That on December 18th, Margaret was given a capsule by the nurse, from the box of fever capsules, which, shortly before, had been in the hands of defendant, and that this was followed by a convulsion similar to Chrisman’s. Dr. Twyman, coming in, administered a hypodermic of morphia and nitro-glycerine, which was followed by vomiting within two hours from the time the capsule was given. That the contents of her stomach so ejected were saved in a bottle, and a chemical test .made thereof on February 4th following, the test indicating the presence of a trace of strychnine, but no cyanide. That an autopsy was had of the body of Chrisman Swope on December 30th following his death, and that chemical tests subsequently made indicated a trace of strychnine in his stomach, but no cyanide. The State further introduced evidence tending to prove that on the evening of December 18th, the defendant took from his pocket and threw away some capsules, whose fragments were found,- and that chemical tests thereof, made later, indicated cyanide as the contents.
A further and more detailed statement of portions of the evidence will become necessary in the opinion.
The defendant files 247 assignments of error. Numerically they deal chiefly with specific rulings on the admission and rejection of testimony, and on these points will require no detailed consideration.
The main points urged upon our attention are: 1. Failure to prove the charge in the indictment that the other deadly drugs and poisonous substances *219were unknown to the grand jury. 2. That the evidence is insufficient to sustain the verdict. 3. Error in admitting the evidence’ of the other alleged crimes. 4. Error in admitting expert testimony. 5. Error in the instructions given and refused. 6. Prejudicial remarks by counsel for the State to the jury. 7. Error in revoking defendant’s bond during the pendency of the trial.
I. The trial proceeded upon the following indictment :
“The grand jurors for the State of Missouri, in and for the body of the county of Jackson, duly impaneled and sworn, upon their oaths present and charge:
“That on the third day of October, 1909, at the county of Jackson and State of Missouri, one B. Clark Hyde, then and there wickedly contriving and intending one Thomas H. Swope to deprive of his life, and then and there feloniously contriving and intending him, the said Thomas H. Swope, willfully, feloniously, deliberately, premeditatedly, on purpose and of his malice aforethought, to kill and murder, did then and there willfully, feloniously, deliberately, premeditatedly, on purpose and of his malice aforethought, administer and give to, and caused to be administered and given to, him, the said Thomas H. Swope, a certain deadly drug* and poisonous substance, to-wit, strychnine, and other deadly drugs and poisonous substances to these grand jurors 'unknown, he, the said B. Clark Hyde, then and there well knowing the said strychnine and other poisonous substances to be deadly drugs and poisons, the said B. Clark Hyde then- and there wickedly, willfully and feloniously contriving and intending the said strychnine and other deadly drugs and poisonous substances to these jurors unknown, then and there to be taken and received into his body of him, the said Thomas H. Swope; and he, the said Thomas H. Swope, not knowing that the said *220strychnine and other deadly drugs and poisonous substances were in fact deadly poisons, did then and there actually take' and receive into his body the said strychnine and other deadly drugs and poisonous substances, to these grand jurors unknown, so willfully, feloniously, deliberately, premeditatedly, on purpose and of his malice aforethought, administered and given by him, the said B. Clark Hyde, to him, the said Thomas H. Swope, as aforesaid, and from the operation and effect thereof he, the said Thomas H. Swope, did then and there become mortally sick, and of said mortal sickness did suffer and languish, and, languishing, did live until the evening of said day, on which said third day of October, 1909', at said county and State, of the deadly drugs and poisons aforesaid, and of the mortal sickness caused thereby he, the said Thomas H. Swope, died. And so the grand jurors aforesaid, upon their oaths aforesaid, do say that he, the said B. Clark Hyde, him, the said Thomas H. Swope, on the third day of October, 1909, at the county of Jackson and State of Missouri, in the manner and by the means aforesaid, willfully, feloniously, deliberately, premeditatedly, on purpose and of his malice aforethought, did kill and murder, against the peace and dignity of the State.”
The State offered evidence on the theory that Col. Swope was poisoned by a combination of strychnine and cyanide of potash. No other deadly drugs or poisonous substances were suggested. It is claimed by defendant that the grand jurors'knew that cyanide was the other deadly drug relied upon by the State, and hence should have charged it in the indictment, and that the evidence fails to sustain the allegation that the other deadly drugs were unknown.
The indictment was returned on March 5, 1910. While fte testimony shows that the grand jury had before them the evidence of the purchase of cyanide by defendant, and of the finding of capsules containing *221cyanide alleged to have been thrown away hy defendant, yet there was no expert report of finding cyanide in the stomach of Col. Swope until after March 24,1910. This assignment is, therefore, without merit,
II. We have carefully considered, hut cannot attempt to discuss all of the testimony concerning the symptoms attending the sickness and death of Col. Swope — the autopsy, the expert evidence of the chemical tests of his organs; the purchase of cyanide hy defendant, and the alleged administration of poison hy him to Col. Swope at the hands of the nurse, Miss Kellar. Doubtless the evidence showing the age of Col. Swope, about eighty-two; his infirm' condition during the previous year; the condition of his internal organs as revealed by the autopsy, and the symptoms of his last sickness, would, if that were all, justify the conclusion that he died from senile debility or uraemic poison, and would fail to establish the corpus delicti. In passing upon this question, we may not properly consider the strong array of experts for the defense, whose testimony completely negatives the poison theory. This conflict between equally credible witnesses is a question solely for the jury. If- we were to lay aside the testimony of one witness for the State, Dr. Vaughn, we would find it difficult to reach a satisfactory conclusion that there is sufficient proof to make a prima-facie case of death by poison.
According to the authorities, the symptoms, as detailed in the evidence, are not clearly and unmistakably those of strychnine poison, nor of cyanide. The opinions of experts for the State indicate this as well.
Dr. Hektoen, in answer to the hypothetical question based upon the symptoms, autopsy and chemical analysis, says: “In my opinion, death resulted from some convulsive and paralyzing poison or combination of poisons.” Dr. Vaughn says: “In my opinion, the man had convulsive poison administered.” Dr. Hall *222says: “From such symptoms, one conld only conclude that he was suffering from some poison.” According to the medical authorities, a fatal dose of strychnine should ordinarily kill within two hours. If the patient lives longer, it is .exceptional, and not to be expected in a man of Col. Swope’s age and physical condition. Cyanide should act. in even less time. Col. Swope lived more than ten hours after the administration of the alleged poison capsule. The State seeks to meet this proposition by the suggestion that cyanide and strychnine were mixed in the capsule, and that their counteracting effect delayed the deadly action. This theory is not established by the proof. No expert witness professed to know what the effect of such combination "would be. Dr. Haines said he would assume that such would be the effect. Dr. Vaughn said his reading would indicate such effect, as also experiments on animals. The experts for the defendant thought the effect would be to hasten death. No expert had known of an actual case. The expert testimony is of little force on this point.
The autopsy showed. Bright’s disease and other conditions which, in the absence of further knowledge, would justify, as stated by one of the State’s experts, a death certificate that death resulted from senile debility. It also appears from the medical authorities (Wharton & Stille’s Med. Jur. sec. 754), and from the testimony of the State’s experts as well, that (convulsions from strychnine poison are in some respects similar to those resulting from uraemia. ’
Dr. Vaughn, however, testifies that he found in Col. Swope’s liver, in February and March, following the death in October, nearly one grain of strychnine. The testimony is that half a grain may be a fatal dose. Dr. Haines also found one two-hundredth of a grain in one-half of the stomach. Of cyanide Doctors Haines and Vaughn, together, found in the stomach a slight trace.
*223In view of the testimony of the purchase of cyanide by defendant, unsatisfactorily explained; the testimony of the nurse that defendant, the attending physician, handed her a five-grain capsule, with a direction that she administer it to the deceased; that she did administer it to him; that within twenty minutes thereafter he had a convulsion which exhibited symptoms consistent, according to the State’s experts, with the presence in the body of a fatal dose of poison, .from which convulsion he did not recover, but died within ten hours; that expert chemists testified that, upon applying the usual and proper tests, they found, five months later, lodged in his organs a lethal amount of strychnine, about one grain, also a trace of icyanide, and that in their opinion'he was poisoned, we think that there was a case to go to the jury.
Dr. Vaughn, by himself, extracted and weighed the strychnine found in the liver — different portions at different times, and by different methods. Doctors Haines and Vaughn are eminent and experienced toxicologists. They were both employed on behalf of thó State. Dr. Haines took the liver of Col. Swope to Ann Arbor, and in conjunction with Dr. Vaughn there tested it up to the point where they became satisfied that strychnine was present; but they did not, together, make the tests to ascertain the quantity of strychnine in the liver. Before making any tests for quantity Dr. Haines returned to Chicago, leaving the liver with Dr. Vaughn,-who later, and by himself, tested for quantity in his laboratory at Ann Arbor. Why they did not together pursue the tests to the end does not appear. We do not mean to discredit the testimony of Dr. Vaughn as to the results of his tests for quantity, nor to hold that the testimony of one expert as not enough to sustain a verdict. But when two experts, are employed, and they are free to work out their tests together, it seems to us that they should do so, as a safeguard against possible error in making the necessarily *224very delicate chemical tests. It was not necessary that so vital a matter as the amount of strychnine in Ool. Swope’s liver should depend upon the results obtained by a single chemist in the privacy of his laboratory. This comment is given emphasis by the fact that experts for defendant were refused any part of the organs of Col. Swope, which they requested be given them for the purpose of making tests for poison, and by the further fact that an expert chemist for the defense testifies that the tests made by Dr. Vaughn to ascertain the quantity of strychnine were unsatisfactory and incomplete.
III. The question of the admission of testimony of other crimes alleged to have been committed by defendant is a serious one, and of first importance in this case. The testimony admitted concerning other crimes imputed to the defendant makes up the major part of the record, and, without regard to its probative value, is of a character to powerfully impress a jury against the defendant.
In objecting to this testimony, the defendant invokes the settled rule that, upon the trial of a defendant on a specific charge, evidence of other crimes committed by him is not admissible. This rule, firmly established in English and American jurisprudence, has been recognized and adhered to by this court in numerous cases. A man may be a notorious criminal, but this fact may not be shown to influence a jury in passing upon the question of his guilt or innocence of the particular offense for which he is on trial. A man may have committed many crimes, and still be innocent of the crime charged in the case on trial. To permit proof of other crimes would naturally predispose the minds of the jurors ag-ainst the defendant. One who commits one crime may be more likely to commit another; yet, logically, one crime does not prove another, nor tend to prove another, unless there is *225such a relation oetween them that proof of one tends to prove the other. Unless such a relation exists, it is illegal and manifestly unfair to require a man, who is charged with a specific crime in the indictment, to prepare a defense against other crimes that the State may attempt to prove against him, but which are not charged in the indictment.
Section 22 of our Bill of Rights declares that “in criminal prosecutions the accused shall have the right to demand the nature and cause of the accusation.” To this end it is said in Chitty’s Orim. Law, 169: “The charge must contain a certain description of the crime of which the defendant is accused, and a statement of the facts by which it is constituted, so as to identify the accusation, lest the grand jury should find a bill for one offense, and the defendant be put upon his trial in chief for another, without any authority.” [State v. Murphy, 141 Mo. 267.]
The rule, therefore, rests upon two grounds: First, the impropriety of inferring from the commission of one crime that the defendant is guilty of another, and, second, the' constitutional objection to compelling a defendant to meet charges of which the indictment gives no information. [People v. Shea, 147 N. Y. 78; Commonwealth v. Jackson, 132 Mass. 16; State v. Goetz, 34 Mo. 85.] The rule against admitting proof of extraneous crimes is subject, however, to certain qualifications or exceptions. In making proof against a defendant it is competent for the prosecution to put in evidence all relevant facts and circumstances which tend to establish any of the constitutive elements of the crime of which the defendant is accused in the case on trial, even if such facts and circumstanjces tend to prove that the defendant has committed other crimes.
The following classification of the exceptions to the rule has been generally adopted, and has received *226the approval of this court: ‘ ‘ Evidence of other crimes is competent to prove the specific crime charged when it tends to establish (1) motive, (2) intent, (3) absence of mistake or accident, (4) a common scheme or plan embracing the commission of two or more crimes so related to each other that proof of one tends to establish the other, (5) the identity of the persons '■■barged with the commission of the crime on trial.” [Wharton on Crim. Ev. (9 Ed.), sec. 48; Underhill on Ev., sec. 58; People v. Molineux, 168 N. Y. l. c. 293; State v. Bailey, 190 Mo. l. c. 280.]
In the present case the jury were instructed that the' evidence of other transactions should be considered by them as bearing only on the question of intent and motive. We are then to consider whether the evidence introduced by the State of alleged extraneous crimes, as tending to prove that the defendant had a motive to kill Col. Swope, and that he intended to kill him, was properly received as coming within the above exceptions. ..... -,,,
Motive, in mu'rder, is the impulse or purpose that induces the murderer to kill his victim. Intent is the purpose to make the means adopted effective.
The State introduced evidence, over the defendant’s objections, concerning (a) the death of Col. ITunton; (b) the death of Chrisman Swope; (c) the alleged attempt to poison Margaret Svrope; (d) the alleged inoculation of Margaret Swope with pus germs, and the alleged inoculation of members of the Swope household with typhoid germs, on the theory that the defendant was responsible for all of these crimes, and that these various matters were related to the death of Col. Swope, in that they tended to show (1) the motive which induced the defendant to murder Col. Swope, and (2) the intent with which defendant administered to Swope an alleged fatal dose of poison.
We will discuss these matters in the above order,
(a). The State introduced testimony tending to *227prove the following facts concerning the death of Col. Hunton: On September' 12, 1909, Col. Swope told the defendant that he intended to make a new will shortly, and give to charity that portion of his fortune, about one and one-half millions, which under the terms of his then present will constituted his residuary estate. That Col. Hunton, on the evening of October 1st following, was stricken with apoplexy at the supper table. He was removed to his room, and soon became unconscious, in which state he remained until his death, which occurred that same evening. That Dr. Twyman (the family physician of Mrs. Swope) and the defendant, Immediately upon the seizure of Col. Hunton, were', summoned by telephone, and shortly thereafter arrived ; Dr. Twyman reaching the house first. A friend of Col. Swope, one Albert M. Ott, .a lawyer living in Independence, was also summoned. Mr. Ott was in the sick room until Hunton died. There were also present Üie nurse, Miss Kellar, and Mrs. Frances Hyde, who arrived with defendant. Dr. Twyman, the family physician, was an experienced practitioner, in vigorous health, and a little past middle life in age. He was a man of high character, and possessed the entire confidence of the Swope family. While waiting the arrival of the defendant, Dr. Twyman, in answer to an inquiry from Mr. Ott, said that in such a case bleeding was the usual practice, but said that as Dr. Hyde was a son-in-law and a physician, he would defer action until his arrival. In the meantime Dr. Twyman applied an ice pack to Col. Hunton’s bead. When defendant arrived he suggested bleeding. Dr. Twyman acquiesced, and the two doctors immediately proceeded to bleed the patient, Dr. Hyde making the incision in the arm with a lancet furnished by Dr. Twyman. A string was tied around the artery and held by the doctors. After a time Dr. Twyman, who evidently held the pulse, said to the defendant that the pulse was easing down, and further said, “Doctor, don’t you *228think we have got enough?” Defendant replied, “I don’t think we have relieved the pressure sufficiently;” and the bleeding continued. After a little while Dr. Twyman again suggested that they had blood enough. According to witness Ott, Dr.Twyman then said, “That is enough blood to take.” Mrs. Iiyde also said to her husband in effect that he had better do as Dr. Twyman suggested; whereupon the bleediug was stopped. The nurse says that the bleeding did not stop until Dr. Twyman suggested it a third time. It appears also that the doctors tied a bandage around the’ arm above the incision, to control the flow of blood, when Dr. Twyman first suggested that the bleeding stop. Dr. Hyde felt the pulse at the time he made the remark that they needed to take more blood. Mr. Ott says that the doctors felt the pulse alternately. The nurse testifies that at the suggestion of Dr. Twyman she took the slop jar, containing the blood which had been drawn, into the bath room, and measured the blood in a silver cup, and that by subsequent measurement of the cup she found that two quarts of blood had been taken from the patient. Col. Hunton died in ten or fifteen minutes after the bleeding stopped. Miss Kellar, the nurse, further testifies that defendant, an hour or two after Col. Hunton’s death, stated to her that Col. Swope would be making a new will in a few days, and requested her to try to influence Col. Swope to appoint him, defendant, an executor therein in place of Col. Hunton, which request she declined. Defendant asked her that evening, so she states, whether she had spoken to Col. Swope about this matter. She does not state whether she replied to this.
Two questions are presented by this testimony. 1. Do these facts bring the death of Col. Hunton within any of the above exceptions to the rule? 2. Is there substantial testimony that the defendant did feloniously cause the death of' Col. Hunton?
*229Upon the first proposition, the State claims that the defendant killed Col. Hunton from a double motive; that one motive was to supplant him as executor in Col. Swope’s new will; the other, to secure to the residuary estate, provided in the existing will, the legacy given in such will to Col. Hunton, so as to increase the share of Mrs. Hyde in such residuary estate by one-tenth of the legacy to Col. Hunton. If, as the State claims, defendant’s motive for killing Col. Swope was to prevent him from making a new will, and his motive for killing Col. Hunton was to displace him as executor in a new will to be made by Col. Swope, such motives are inconsistent and antagonistic, and in no sense could the one throw any light on the other. In such conflict of motives there could be no relation between the two crimes. The other motive ascribed to defendant, namely, a desire to secure Col. Hunton’s legacy to the residuary estate, is not inconsistent with his alleged motive for killing Col. Swope. • If defendant knew that Col. Hunton was a legatee under the existing will, a desire on his part to add this legacy to the residuary estate provided in that will, and of which his wife would receive one-tenth, would afford a motive to kill Col. Hunton consistent with his alleged motive-for killing Col. Swope. Both motives would contemplate that there was not to be a new will. There is, however, no evidence in the record that the defendant knew at the time of Hunton’s death that he was a legatee in the will of Col. Swope. As far as affirmative testimony goes, the State relies solely upon the testimony of Mrs. Logan Swope to the effect that at the time of the high water, -about a year before thé trial, she heard defendant say to Col. Hunton that the water threatened Col- Hunton’s building in the West Bottoms. It is assumed that the defendant by this remark referred to property specified in Col. Swope’s will, as a legacy to Col. Hunton. There is no evidence that the defendant knew of any property devised to Col. *230TIunton in the West Bottoms, or knew that he was a legatee in the will. Nor is there any testimony that the will gave him property in the West Bottoms. We, therefore, hold that the evidence of the death of Hun-ton was not within any exception to the rule excluding evidence of other crimes.
Upon the second proposition, we must hold that there is no substantial evidence that Col. Hunton’s death was caused by the bleeding. The record shows nothing inore than a difference of opinion between the attending physicians as to the amount of bleeding proper under the circumstances. There is no rule as to how much blood should be let in the case of apoplexy. The amount, depends upon the effect u produced, and this is determined by the. action of the pulse. It is a matter of judgment. No-reason is perceived why Dr. Twyman, the elder of the two and the family physician, should have acquiesced in the continued bleeding if it was clearly wrong. It appears that he yielded his judgment to his brother doctor on a doubtful point of proper treatment. It seems that the apoplectic attack was fatal in character. Dr. Twyman participated in the treatment from beginning to end, and it is inconceivable that, either through ignorance or carelessness, he should have allowed murder to be committed before his eyes. As to the quantity of blood drawn the evidence is not satisfactory. It was measured privately by the nurse by means of a silver cup, and we have only her conclusion as to the amount. Furthermore, it is not shown that the amount she gives — two quarts — was excessive under the circumstances. Whether is was or not, we can find no evidence of criminal intent.
(b). The objection to the evidence concerning the alleged poisoning -of Chrisman Swope raises the same questions discussed above in relation to Col. TIunton: (1) Does the evidence come within any exception to *231the rule? (2) Is there substantial testimony that the defendant poisoned Chrisman Swope?'
What we shall now say upon the first proposition will apply as well to the evidence concerning the alleged poisoning of Margaret Swope and to the evidence concerning the alleged inoculation with typhoid germs. This evidence was admitted upon the theory that it might be regarded by the jury as throwing light upon the motive which prompted the alleged poisoning of *Col._Swope, and also upon the question of intent; that is, whether the capsule was administered to Col. Swope by accident or mistake, or designedly, with the purpose to kill.
Intent and motive are often used to mean the same thing. There is, however, a well-defined distinction between them in their relation to crime. Motive, as we have said, is the impulse which moves a man to commit the criminal act. This impulse may arise from various causes, hatred, jealousy, avarice, etc. Intent is the purpose to make the means employed to commit the wrongful act effective to produce the desired result. Such intent involves knowledge of the character of the means used, and negatives accident, mistake and inadvertence. In some cases such knowledge may be inferred from the act itself. For illustration, if one stabs another to the heart with a dagger, it will be presumed that he knew that the dagger would kill. But poison might be administered in a capsule by mistake or by inadvertence, under the belief that the capsule contained an innocent medicine. Hence, when the charge is murder by poison, it is competent for the State, in order to show criminal intent, which is a constitutive element of the crime, to prove knowledge, and to negative accident and mistake. For this purpose the State is permitted to show that on other occasions the defendant used the same means in the same way and with the same effect. If the defendant administered a poisoned capsule to Col. Swope, the *232fact that on another occasion he administered similar capsules', with similar results, affords ground for an inference that be did so knowingly in each case.
The defendant now contends that there is no question of intent in the case. His counsel now concede that if the defendant gave Col. Swope a deadly'dose of poison, the criminal intent may he inferred from the act. The defense denied the act. It is true that the evidence in the record raises no .question of intent. ' It may also he conceded that if the defendant knowingly administered deadly poison, the intent may he inferred from the act, and no further proof is required. But .it must be remembered that in a criminal case the defendant files no written pleading setting up his grounds of defense. His plea of not guilty puts in issue every element of the charge, including intent. It would have been consistent with his plea had the defendant claimed in his defense that there was accident or mistake, both of which are involved in intent, in the giving of the alleged poisoned capsule to Col. Swope. It is a part of the State’s case to show criminal intent, and in doing so to negative accident or mistake, by any proof competent for that purpose. The State was not required to rest upon the assumption that the proof would show that the intent accompanied the act, nor to hold hack, for rebuttal, evidence showing intent. The State could not anticipate the theory of the defense. ■ ■ -1 ■, r
In the case of Trogdon v. Commonwealth, 31 Gratt. 862, quoted with approval in State v. Myers, 82 Mo. l. c. 569, in answer to the suggestion of counsel for the prisoner that the jury would infer the intent from the act, and therefore evidence of collateral facts was unnecessary and improper, the court said: ‘ ‘ It is impossible for the court to foresee what may he developed in the progress of the trial. When evidence is offered of other transactions to show the guilty intent of the accused, is the court to say the intent is *233already conclusively proved, and the evidence is therefore irrelevant? What would be thought of a judge who would thus prejudge the case and invade the province of the jury?” This court further, in the above Myers ease, said: “It has ben held by the highest authority in this class of cases [fraudulent trick], that even the admission of the accused that the act was done with criminal intent cannot preclude the State from proving it by any other competent testimony, for the jury are the sole judges of the evidence.” We, therefore, hold that it was competent for the State to introduce evidence on the question of intent in its case in chief.
As tending to prove the absence of mistake or accident, it is competent to prove that on other occasions the defendant used the same means, with the same effect. An accident or mistake 'may occur, but every time such act is repeated the likelihood of accident or mistake diminishes. A doctor may give a poisoned capsule once to a patient by mistake, but such mistake is not likely to happen twice, still less three times. But the evidence of the other acts relied upon must show them to be identical with the act in question, otherwise they have no probative value so far as this question of mistake" or accident is concerned. On this point Underhill on Orim. Evidence, sec. 89, says: “Suppose the question is, was a given act, either by the accused or by some other person, intentional or accidental? Here it is relevant to prove that the person whose intention is in question had performed acts of a precisely similar nature before or after the act the intention of which is in question. And if it be found that he has performed many such acts, we have the best grounds for drawing the conclusion that the act, in the present instance, is intentional and not accidental.”
In the case of Goersen v. Com., 99 Pa. St. 388, cited by the State, the defendant was on trial for poi*234soning Ms wife. The State proved that an autopsy performed upon the body of his mother-in-law, who had died about two weeks earlier, revealed in her stomach a deadly dose of the same poison which was found in the stomach of the wife. Experts testified that they both died from the effects of the same Mnd of poison which was found in a deadly quantity in the stomach of each. There was also evidence that the defendant, a physician,, had attended upon both. The evidence of the death of the mother-in-law by poison was admitted on various grounds; one being to rebut the theory that the death of the wife “was the result of accident or suicide, or of the negligent or ignorant use or administering of arsenic.” In that case the autopsy showed that the poisons were identical in character, and subsequently the same in amount, and without question produced death in each case. So in the case of Commonwealth v. Snell, 189 Mass. 12, cited hy the State, it was held competent, on the question whether the deceased was drowned by accident or by the act of the prisoner, to prove that the prisoner had suggested that another person be murdered by drowning in identically the same manner as was the deceased.
On this point of the identity of the acts, the evidence in this case fails. The jury must have found that there was administered to Col. Swope deadly poison contamed in a capsule administered by the defendant. The act in that case was giving to Swope a capsule containing at least a deadly dose of strychnine, and also of cyanide. That the capsule did contain these deadly drugs was demonstrated, it is claimed, by finding in the body of Swope a lethal quantity of strychnine and a trace cyanide. The State’s case depended upon this demonstration. To show that this capsule was not given to Col. Swope by mistake or accident, the State undertook to prove that on two other occasions the defendant gave similar capsules of poison to his patients, with similar effects. The proof *235must show that the capsules were similar in character and effect, otherwise these other acts have no bearing on the question of accident or mistake. That the capsule given Col. Swope was poisoned was proved, according to the State’s contention, by the symptoms and by finding the poisoned contents of the capsule in the body after death. Therefore, if the capsules were similar, we must expect the chemical analysis of the organs of Chrisman Swope and the chemical analysis of the contents of Margaret’s stomach to show poisons precisely similar in character and effect to those found in the body of Col. Swope, and the cases should also show identical symptoms: The evidence shows that the same experts who found the poison in the organs of Col. Swope also examined the organs of Chrisman Swope, who died two months after Col. Swope, and also examined the contents of Margaret’s stomach which were vomited within two hours after the alleged administration to her by defendant of the poisoned capsule. In neither case did they find any cyanide, and in both cases found merely a trace of strychnine, which might be accounted for by the hypodermic injections of strychnine which had been given to each. The testimony is that strychnine introduced into -the body hypodermically will penetrate to all the organs, also that “a trace” of strychnine is a quantity less than one six-hundred-and-fortieth of a grain, and that it may be as little as one sixty-thousandth of a grain and still respond to the color test. The amount of each hypodermic injection was one-sixtieth of a grain. A comparison of the symptoms in the cases does not show that in so far as the convulsions had common features they necessarily resulted from the same causes. In neither case do the experts, in giving their opinions upon their symptoms, identify the poisons. In each case they go no further than to say that the convulsions resulted from “some convulsive poison.” *236How many different poisons come within this classification we are not advised.
Ur. Vanghn was asked, upon a hypothetical question which purported to cover the history of the case and the symptoms, what in his opinion caused the convulsion which preceded the death of Chrisman by about thirty hours. He replied: “In my opinion the patient had some convulsive poison. ’ ’ ■ There is nothing in his answer to identify the poison with that which he says killed Col. Swope, and which also was some convulsive poison or poisons.” Dr. Hektoen testifies, in answer to ‘a similar question: “Well, in the first place, this person had typhoid; in the second place, it seems to me, he suffered — it presents symptoms of some convulsive or paralyzing poison or combination of poisons.” This answer not only fails to identify the poison with that given to Col. Swope, but indicates that typhoid was the important factor. In fact, the evidence is clear that Chrisman did not die from the effects of poison. No expert expresses an opinion that he died of poison. He lived thirty hours after its alleged administration. He had a fatal- temperature which resulted from typhoid and which was not caused by poison. Dr. Hektoen testified that the autopsy showed what might have been a fatal stage of typhoid. So far as the case of Margaret is concerned, on this question of intent, she did not die, and no poison, except a trace of strychnine, was found in her stomach contents, ejected within two hours after the alleged administration to her of a poisoned capsule. Nor are the convulsions in either case said to be wholly characteristic of any poison named. Nor are they entirely similar. It therefore follows that, if defendant did administer poison capsules to Chrisman and Margaret, it is not proved that they were “precisely similar” to that which he is alleged to have administered to Col. Swope. If they were different, then the fact that the defendant gave poison of some kind to Chrisman *?or Margaret would not tend to show that he knew the nature of the capsule which he is charged with giving to Col. Swope.
Was this evidence competent to show motive? This, question presents considerations other than those involved in the discussion of intent. If it can be shown that defendant attempted the death of Chrisman, Margaret and others of the Swope heirs, either by poison of any kind or by inoculation with disease germs, will such proof tend to show a motive for the alleged killing of Col. Swope? If it will, it is competent. If it will not throw light on the question of motive, it is inadmissible.
Every conscious act has a motive behind it. This motive may or may not appear. If it should be demonstrated that defendant murdered Col. Swope, no question of motive need arise. Failure to show motive in such a case would not acquit. [State v. David, 131 Mo. 381.] But where the defendant denies the act, the question of motive becomes important. Absence of motive in such case tends toward innocence. Presence of motive tends toward guilt. Motives are usually hidden, and must be found by considering all circumstances and facts which tend to reveal their existence. If the prior or subsequent conduct of the defendant is so' related to the act in the present instance, namely, the alleged poisoning of Col. Swope, as to supply a motive for the murder of Col. Swope, then proof of such extraneous conduct is competent, even if it tends to prove that the defendant committed another crime. The authorities to sustain this proposition are numerous and harmonious. [State v. Dettmer, 124 Mo. 426; State v. Spaugh, 200 Mo. 571; State v. Collins, 181 Mo. l. c. 261; Goersen v. Commonwealth, 99 Pa. St. 388; Pierson v. People, 79 N. Y. 424; Commonwealth v. Robinson, 146 Mass. 571; Hawes v. State, 88 Ala. 37; Zoldoske v. State, 62 Wis. 580.] The case of Shaffner v. Commonwealth, 72 Pa. St. 60, cited by defendant, sustains the doctrine announced in the foregoing cases, although it holds that the facts in- that case did not show such a relation between the crimes' as to make evidence of one competent to prove the other.
If it can be shown that the defendant poisoned or attempted to poison Chrisman Swope, does this fact tend to show a motive fpr killing Col. Swope? Is there such a relation between the two transactions that the killing of Chrisman indicates the motive for killing Col. Swope? Chrisman was unmarried and childless; he was to receive a liberal legacy under the will of Col. Swope; upon his death, his sister Prances, wife of the defendant, would inherit one-seventh of his estate, including what he should receive from Col. Swope. ' As matters stood on October 3,1909', the death of Col. Swope alone would not give defendant any benefit from Chrisman’s estate. Nor would the death of Chrisman alone, Col. Swope being still alive, benefit defendant so far as concerned Chrisman’s legacy under Col. Swope’s will. Laying aside for the present the direct benefit that would come to defendant, through his wife, from the death of Col. Swope, and considering her only as an heir of Chrisman, the benefit which defendant would derive from his death supplies a motive to remove Chrisman, whose life stood between defendant and the money that Mrs. Hyde would inherit .as his heir; but another life stood between defendant and this money, namely, the life of Col. Swope, upon whose death the amount of Chris-man’s estate depended. Therefore, the defendant had the same motive to remove Col. Swope that he had to remove Chrisman; namely,'to get Chrisman’s money. The killing of Chrisman in order to get his money revealed a motive for killing Col. Swope, whose death to this end was also necessary. Therefore, these transactions were related; and the alleged killing of Chris-man throws light, from this point of view, on the mo*239tive for killing Col. Swope. This is no less true because of the fact that defendant would, also benefit, through his wife, from the death of Col. Swope, even if he did not also remove Chrisman. The State should not be limited in proving the extent of the benefit which defendant would derive from Col. Swope’s death. Doubtless it may be said that the fact that Mrs. Hyde was a legatee under the will supplies* a motive, but this fact would not preclude the State from showing, if it could, that defendant had the additional purpose to get into his possession the money that would come to his wife through the death of Chrisman and the other Swope heirs. What is said above applies as well to Margaret and the other Swope legatees. We must, therefore, hold that the other alleged crimes now under consideration are so related to the alleged murder of Col. Swope -as to make proof of them competent.
Regarding the question whether there is substantial evidence that the defendant poisoned Chrisman Swope, we are constrained to hold that the evidence does not authorize the submission of that question to a jury. We have already pointed out the failure to prove poison in his organs by the chemical tests. This fact is most persuasive, if not conclusive, to show that he did not die from the effects of poison. [State v. Nesenhener, 164 Mo. 461.] The testimony for the State shows that he had a fatal case of typhoid fever, with a temperature of 107 4-5, which, the experts say, means death. The evidence further shows that poison does not increase the temperature. The autopsy showed typhoid fever in wha,t the State’s expert, Dr. Hektoen, who conducted the post-mortem, admits might have been a fatal stage. The lower lobes of both lungs were consolidated. Chrisman died Monday night, December 6th, at about 10 o’clock. It is in evidence that on the preceding Sunday, at 2:45 p. m., the defendant gave him. a capsule, after which he Had a convulsion presenting symptoms similar in many *240respects to those presented in the convulsion of Col. Swojje. Aside from the inference which the State seeks to draw from this convulsion, there is no evidence that this capsule contained poison. The patient was taking fever capsules, also antiseptic capsules. Defendant was alone with the patient when he administered the capsules, but stated immediately thereafter that he had given it. He remarked to the nurse that he had given Chrisman his capsule. The patient recovered from this convulsion, became for a time rational, and was rational again part of next day. Subsequently he became delirious from the fever. He lived until ten o ’clock Monday night; more than thirty hours -after the capsule was given him. .The history of the case subsequently to the giving of the capsule is inconsistent with the poison theory. It is attempted to impute a poisoned capsule to the defendant, given about 4 o ’clock p. m., Monday, by the nurse. She says that the box of fever capsules was kept on the dresser; that at that time defendant said it was time for Chris-man to take his medicine; that he walked towards the dresser and then handed her a capsule which she gave to the patient, who afterwards became restless. There is nothing in the recital of this episode, nor in •the condition of the patient afterwards, to give it any probative force. Chrisman died December 6th. Counsel for the State point to the fact that on December 4th, two days preceding, the defendant purchased six five-grain capsules of cyanide; that on the next day, December 5th, he gave a capsule of some kind to Chris-man. The claim is made that this sequence of events shows that the cyanide was purchased with intent to poison Chrisman, and was given to him with that intent. It is fair to assume that the defendant, if he intended to kill, would give a deadly dose. We would expect, then, according to the logic of this theory, that Chrisman would die within an hour or two after the capsule was given, and that an autopsy would reveal *241enough cyanide in his organs to at least suggest a fatal dose. The chain of evidence breaks upon this point. Chrisman did not die from the effects of poison, and not a trace of cyanide was revealed by the autopsy. In all cases of murder by poison it is considered important, if not necessary, to trace poison into the possession of the accused, or at least show that he had access to it. The State recognizes this, and had built up a strong case showing the repeated purchases of cyanide by defendant, also that defendant, to escape detection, attempted to destroy the cyanide capsules which it is alleged he took with him to.the Swope home. The State urges that 'the purchase of cyanide by defendant on December 4th, two days prior to the death of Chrisman, is of great significance in that it shows preparation to do what the State charges the defendant did do on the 5th of December, namely, administer poison to Chrisman, and that he gave to Chrisman, on December 5th, the poison he bought on December 4th. If he did this, then he gave cyanide to Chrisman, and it would appear in his organs upon a chemical test. We may not infer that cyanide was given, when the chemical tests show that it was not given. The State is forced to stand upon testimony which tends to prove that a capsule loaded with cyanide discharged a trace of strychnine only. The inference which the State seeks to draw from the fact that defendant purchased cyanide on December 4th, and that on December 5th he gave to Chrisman a capsule, which was followed by a convulsion, namely, that this capsule contained cyanide poison, is not supported by either the symptoms or the autopsy. Evidently not by the symptoms, because the experts did not undertake to say that the symptoms indicated cyanide poison. Clearly not by the autopsy, because the-autopsy showed no cyanide. Nor is there any substantial evidence to support a theory that cyanide might have been given in fatal quantity, and yet *242be so completely eliminated as to leave no trace in the bodily organs. The defendant is not shown to have purchased, or to have had in his possession, any poison other than cyanide. We cannot infer the use of other poisons from the purchase of cyanide. As to poison by strychnine, it will hardly be contended that, in the absence of proof that defendant had strychnine in his possession, in the absence of proof by experts that the symptoms indicated strychnine, in the presence of proof that death resulted from typhoid fever, a case of murder should be submitted to the jury upon evidence of a mere trace of strychnine which may be as little as one sixty-thousandth *of a grain, and which may be accounted for by the hypodermic injections. If it be claimed that the character of the convulsions' which followed the administration of capsules to Chris-man and Col. Swope indicates that the convulsions in both cases proceeded from the same cause (poison), we must reaffirm the doctrine of the Nesenhener case, supra, which holds that mere symptoms of poisoning, not confirmed by an autopsy, cannot, in the absence of affirmative proof that poison was administered, support a conviction for murder. In this connection, we notice that the State’s printed argument says that “in Col. Swope the symptoms of strychnine poison predominated over those of cyanide poisoning; in Chrisman’s case the symptoms of cyanide poisoning predominated over those of strychnine;” thus affirming the theory that Chrisman was poisoned with cyanide, notwithstanding the absence of proof, as pointed out above, to sustain that theory. In our opinion, there is no substantial evidence, sufficient to go to the jury, that the defendant administered any poison to Chrisman Swope.
(c). In the case of Margaret Swope, it is claimed that the defendant substituted a poison capsule for a fever capsule while handling the box containing* the latter. Margaret was sick with typhoid fever. Drt *243Twyman was the physician in charge. Nurse Houlehan was in attendance. The defendant was also treating these typhoid patients in the house of Mrs. Logan Swope. On the morning of December 18th, the nurse says, defendant, who had been absent some days on a trip to New York, came into Margaret’s room, examined the nurse’s'chart, and then, picking up the box of fever capsules on the mantel, said, “Miss Houlehan, you are still giving these capsules, are you?” to which she replied, “Yes.” The nurse then left the room, the defendant remaining. The nurse testifies that on her return to the room, half an hour later, she gave Margaret a capsule out of the box aforesaid, and that shortly thereafter Margaret had a convulsion similar to Chrisman’s. Within two hours Margaret vomited, and the contents of her stomach so ejected were preserved in a stoppered bottle. Subsequent chemical tests showed no poison except a trace of strychnine. It is not claimed that defendant gave Margaret a capsule. The poison theory rests upon the supposition that defendant, while'handling the box, placed therein a poison capsule which the nurse afterwards gave to the patient, and rests also upon the fact that Margaret had a convulsion shortly after taking the capsule. The alleged handling of the box by the defendant and his question to the nurse were natural and proper, and in themselves innocent. Whát we have said about the convulsion which Chrisman had applied here. Dr. Vaughn, in answer to a hypothetical question describing her symptoms and convulsions, said: “In my opinion, the individuals had received some convulsive poison.” Dr. Vaughn made a chemical analysis of the vomit, and found only a trace of strychnine. He says, The amount was “very small.” No cyanjde was found. There is, therefore, no proof that the defendant gave Margaret any poison, and no proof that poison, beyond a trace, was found in the contents of her stomach ejected within two hours after the al*244leged giving of the capsule. We regard the suggestion that the defendant, while handling the box containing fever capsules, deposited therein a poison capsule, and that the nurse happened to take out this particular capsule and give it to Margaret, as not meriting serious consideration. It is mere conjecture. What we have said above regarding the symptoms and chemical tests in Chrisman’s case applies here as well. It may be noted here there is no one witness who was present when all of the several patients suffered convulsions.
(d). It is claimed by the State that defendant, under pretense of giving Margaret Swope a hypodermic injection of camphorated oil, did in fact inject into her arm pus germs; also that he inoculated the members of the Swope household with typhoid germs by planting* such germs in the water cooler used by the family. These charges are based upon evidence tending to prove the following facts: On November 1, 1909, following the death of Col. Swope, which occurred October 3, 1909, the defendant called upon Dr. Stewart, a bacteriologist in Kansas City, and told him that he intended to fit up a little laboratory in connection with his office and do some work in the study of germs, and asked Dr. Stewart’s advice as to how he should go about it. Dr. Stewart suggested that defendant procure some culture tubes from Parke-Davis, wholesale druggist, also an incubator, and further said that he would assist defendant by instructions. Defendant procured the tubes as suggested, and on November 10th brought six tubes to Dr. Stewart, who planted in them germs as follows: One with typhoid; one with what he believed were diphtheria germs; three with pus, and one harmless. These tubes, two days later, were taken by defendant to his office. On November 25th, Thanksgiving Day, the defendant and his wife, living then in Kansas City, went to the Swope home in Independence for midday dinner, returning to Kansas City in the afternoon or evening of that day'. *245There is also testimony that defendant and his wife dined at the Swopes’ on November 21st, the Sunday preceding Thanksgiving. The evidence for the State as to whether they were there on the 21st is unsatisfactory. Two witnesses, members of the household, say that defendant dined there on the 21st. Mrs. Swope thinks he did, but cannot remember. Mrs. Cleary, a witness for the defense, testifies that the Hydes dined at her home in Kansas City on the 21st. On November 28th some members of the Swope family became indisposed, and by December 2d an epidemic of typhoid developed in the Swope home, as follows: Chrisman Swope, December 2; Margaret, December 2; Leonora Copridge, a colored servant, December 2; Georgia Compton, a seamstress, December 1; Miss Dickson, a visitor, December 4; Sarah Swope, December 10 ; Stella Swope, December 11; Mildred Pox, a visitor who spent the night of December 3d in the house, December 10; and Lucy Lee Swope, who returned on December 17th from New York, December 21. The defendant met Lucy Lee in New York, on her return from Europe, and accompanied her home. While on the train he gave her a. drink from a folding cup which he carried, and which he filled with water from the cooler in the car. -Aside from Mildred Pox and Lucy Lee, the other typhoid victims dined with the defendant on November 25th, also on November 21st, if he was there on the latter date, and all drank from the cooler which stood in the pantry and was filled with cistern water. On October 2d, preceding the ' foregoing events, the defendant and his wife brought with them from Kansas City to the Swope home bottled spring water of which they drank while in the house. In October, also, defendant suggested to Mrs. Swope that she should use distilled water. On one occasion in November Mrs. Hyde, wife of defendant, was heard to say that she would not drink the cistern water — that she would as soon drink poison.The *246Swope home was a large country mansion, located in the midst of fourteen acres. There was a cabin adjacent, occupied by colored servants, one of whom had typhoid fever there in January or February, 1900'. Excretions from her body were thrown into an ordinary country privy which stood forty or fifty feet from the house, and which was partially open in the rear, exposing the excretions which fell upon the ground. About four or five feet from this privy was an open cess-pool, into which liquids from the house were drained. The house was supplied with bath rooms and plumbing. On the premises were stables, cOws and horses. Milk crocks were kept on a bench outside. A colored man milked the cows and strained the milk. The privy was moved, the offal thrown into the cesspool, and the cess-pool covered over about December 1st. After the typhoid outbreak the Swope premises were examined, tests were made of the water, milk, and the earth where the privy had been, without discovering typhoid germs.
On December 14th, while defendant was in New York, Dr. Stewart went to defendant’s office and took away the typhoid culture tube which he had planted for defendant. He found that about one-half the germs had been removed. On December 27th, Dr. Stewart again visited defendant’s office, when the tubes were examined by them both, and Dr. Stewart found that the diphtheria tube, which defendant held up to the light for his inspection, had also been “ disturbed,” meaning that a portion of the germs had been removed. The defendant exhibited no reluctance to making this examination. Dr. Stewart, with defendant’s consent, took away tire diphtheria tube, and later discovered that it contained pus germs and not diphtheria, as he had supposed. It appears also that from time to time, in November and December, the defendant, through his druggist, bought from ParkeDavis'a large number, about two hundred, culture *247tubes. These tubes are of glass, and contain media appropriate for the culture of germs, but, as sold, contain no germs. The live germs for planting in these tubes are obtained from a bacteriologist. Dr. Stewart testifies that he charged nothing for the germs; that he gave them to any doctor who asked for them; that doctors got them for microscopic study of germs. It may be further remarked that the defendant testified that prior to his visit to Dr. Stewart, in November, he had negotiated with the agent of his office building for additional space for a laboratory, giving the name of the agent. This testimony was not rebutted. It does not appear that defendant did install a laboratory. He says that he was waiting the return' of some doctor who was to work with him. On. December 12th defendant administered to Margaret Swope, who was ill with typhoid, a hypodermic injection of what he claimed was camphorated' oil. The arm which had received the injection became sore, swollen and exceedingly painful. There was an area of inflammation about the puncture of the size of a dollar. ' The effects lasted for several days. The nurse testifies that she did not notice any odor of camphorated oil in the room afterwards, and did not know of any camphorated oil in the house..
Based upon the foregoing facts, the State advances the theory that defendant poisoned the water cooler in the Swope home with typhoid germs on November 21st, and again on November 25th, thus causing the typhoid epidemic; that he injected pus germs in the arm of Margaret, believing them to.be diphtheria germs; that he did these things pursuant to a plan formed by him, before the death of Col. Swope, to kill the various Swope heirs in order to enrich his -wife, who would inherit from-them; and that the doing of these things illustrates his'motive in killing Col. Swope.
If defendant attempted to kill the Swope heirs by inoculation with poison germs, this attempt .bore the *248same relation to the death of Col. .Swope as did the alleged attempt to poison Chrisman Swope, discussed above, and evidence of such attempt is competent in this case. We are unable to find in this record any substantial evidence to support the State’s theory on this branch of the case. There is no proof that the defendant did plant typhoid germs on the Swope premises. 'The State limits its claim to the water cooler as the repository of the germs. There is no proof that any germs came from the cooler. That they did come from the cooler is mere conjecture. There is no proof that defendant carried any germs to the Swope home in any manner whatever. Epidemics of this character are usually mysterious and inexplicable as to their origin. Prom the evidence for the State it is more rational to conjecture, if we must hazard a guess as to the cause, that the outbreak of typhoid in question resulted from germs which survived the sickness of the colored servant, some months earlier. The possession of germs in culture tubes-by defendant, as shown by the St.ate, is consistent with innocence. It is perhaps unusual, but some doctors study germs. D'r. Stewart proves this when he says that he gives the germs to any doctor who asks for them. If a doctor does study germ's, and if he experiments with them, naturally he will use them, and will remove them from the tubes in which they are kept. This would be necessary for microscopic study. Whatever may have been his purpose in buying so large a number of culture media tubes, he certainly did not need them to carry germs to the Swope mansion. We may further conjecture that, as they cost but a trifle, he, in the ignorance and enthusiasm of a beginner, overstocked himself. As to the bottled water incident, can it be said to be a suspicious circumstance that a doctor uses himself, and advises others to use, bottled water in preference to cistern water? This water was taken to the Swope home on October 1st, one month before the defendant *249suggested culture tubes to Dr. Stewart. The intimation. that the defendant put typhoid germs in the cup from which Lucy Lee drank on the train we consider unworthy of comment.
The State’s brief emphasizes in italics a remark imputed to the defendant’s wife, made in November, to the effect that she would not drink the cistern water, and would as soon drink poison. This remark is referred to darkly as being an unconscious betrayal of the then purpose in the mind of the defendant to, later, poison the water in the cooler. Inasmuch as there is no claim, certainly no proof, that Mrs. Hyde shared her husband’s alleged evil designs, this remark, if it had any value, must be regarded as merely an expression of her opinion that the cistern water was dangerous. We think it proper to remark in this connection that the State, in its brief as well as in the oral argument has thrown out a suggestion that Mrs. Hyde knew more than the State cares to directly charge against her. There is nothing in the record to justify this. Furthermore, Mr., Paxton, the family lawyer of the Swopes, and as such active in this prosecution, testified that he had known Frances Hyde all her life,- and that he knew her to be ‘‘an honest, capable, pure, good woman. ’ ’
There is no evidence to disprove defendant’s statement at the time that he had given. Margaret an injection of camphorated oil. The'soreness that'followed, was severe and unusual. The treatment may have been unnecessary, but this comes far short of proving felonious intent.
The claim that defendant attempted to poison Sarah Swope, by handing to Stella a capsule with a request that she ask the nurse to give it to Sarah, and the further claim that he inoculated them both with typhoid by giving them candy, are neither one supported by any evidence worthy of consideration.' According to the expert testimony in this case, the percentage of *250fatalities in typhoid fever does not exceed two per cent —one in fifty. The defendant, knowing this fact as a physician, would not be likely to select typhoid fever as a direct instrument of murder. To meet this situation, the State advances the theory or suggestion that “it would serve as a mask and cover for deaths that might be otherwise produced” (State’s brief); in other words, afford an opportunity to poison the typhoid patient. This theory of the State concerning germ inoculation is built upon a series of inferences upon inferences, without any substantial underlying basis' of fact.
^We are of opinion that none of the testimony of other alleged crimes should have been given to the jury. Having been admitted, it should have been withdrawn from their consideration. We also think that the better practice would be that the court should, as a preliminary matter, when the State proposes to offer evidence of other crimes, either hear the evidence or satisfy itself as to its character and scope by inquiry of the prosecuting attorney, and determine whether there is sufficient evidence of the other alleged crime to justify its submission to a jury. We do not mean to say that the evidence upon such preliminary hearing must prove beyond a reasonable doubt that the other crime had been committed, but that there should appear substantial evidence sufficient to take a case to a jury. A satisfactory precedent is- found in the case of Commonwealth v. Robinson, 146 Mass. 571, where such preliminary hearing was had. As to the degree of proof on such hearing, the court said (l. c. 581): “Where, in a case like the present, the admissibility of testimony depends upon the determination of some prior fact by the court, there is no rule of law that, in order to render the testimony admissible, such prior fact must be established by a weight of evidence which will amount to a demonstration, and shut out all doubt or question of its existence. It is only necessary that there should *251be so much evidence as to make it proper to submit the vhole evidence to the jury.”
In a ease like this, involving a large amount of testimony concerning other crimes which would occupy days in presentation, it would be impracticable to give a preliminary hearing to all the details. In such case the court may properly be guided by the offer of proof and by such testimony as can be conveniently presented; enough to satisfy the court that the evidence is relevant and of sufficient weight to authorize its submission to the jury. The great danger that evidence of other crimes, even if it fails to establish them, and even if it is by an instruction withdrawn from the jury, will prejudice the jury against the defendant and obscure their judgment upon the real issues before them, suggests the propriety of determining in advance of its introduction that such testimony is competent, y
IV. The defendant complains that the experts for the State were allowed to testify that in th.eir opinion Col. Swope died from the effects of poison, also that Chrisman and Margaret had been poisoned, and that the experts were thus allowed to usurp the functions of the jury, whose province it was to decide these questions.
There was propounded to Dr. Hektoen, expert witness for the State, a hypothetical question detailing the symptoms and history of Col. Swope’s ease, ending thus: “State to the jury what in your opinion that man was suffering from and died from?” To this the witness, over defendant’s objection, answered: “In my opinion, death resulted from some convulsive and paralyzing poison or combination of poisons.” This may serve as stating the situation as to all the hypothetical questions and answers on behalf of the State. The objection applies to them all upon the same ground.
*252There are many cases where the expert may give an opinion upon the very point in issue, as, for instance, where insanity is the only issue before the jury; also where there can be but one conclusion from conceded facts, as death from wounds. But where the cause of the existing conditions is in dispute, and where the jury must determine which of the causes claimed by the respective parties is the right one, an expert should only be allowed to state that, in his opinion, a certain cause might produce the condition — not that it did produce it. The latter is for the jury. Another expert may testify that some other cause might produce the existing condition. In cases of such contrariety of opinions, it is for the jury to say, considering the opinions together with all the evidence in the case, which opinion, if either, they will accept.
Such is the situation here. The question whether the corpus delicti is established is hotly contested. The theory of the State is that Col. Swope died from poison administered by the defendant. The theory of the defense is that he died from uraemic poison, or from old age. Each theory is strongly supported by the opinions of experts employed by the respective, parties. It is competent for one expert to say that, under the evidence, death might be due to poison, and for another to say that it might be due to uraemia. Both opinions might be correct, and the jury must judge.
These views are in harmony with the recent decisions of this court. In the case of Taylor v. Railroad, 185 Mo. 239, where the contention was whether plaintiff’s present condition resulted from the injury for which she was suing, or from other and natural causes, the expert was asked this question: “To what did you and do you now attribute her condition?” This was objected to as calling for a conclusion. The objection was overruled, and the doctor replied: “To her injury at the time she was hurt.” This court, speaking through Marshall, J., held this question and answer to *253be error. Judge Marshall says: “It is too clear to admit of serious argument that it was error to permit the experts to testify as above set out. For by so doing- the court permitted the experts to decide the whole case, to adjudge the sole defense of the defendant against it, and to draw a conclusion of fact which it was the province of the jury to decide. ’ ’ He says further (l. c. 256) that it would have been proper “to ask them whether or not in their opinion such injuries might, could or would result in paralysis. The experts having thus given.an opinion, it would have been for the jury to find the fact as to whether in this particular case the paralysis was caused as the plaintiff’s experts said it might have been caused, or whether it was the result of other causes, as the defendant’s experts testified might be the case.” This case is apr proved in Glasgow v. Railroad, 191 Mo. l. c. 364, and in Roscoe v. Railroad, 202 Mo. l. c. 595.
In the case of Wood v. Railroad, 181 Mo. 433, cited by thé State, there are some observations which would seem to sustain the contention that an expert may give his opinion upon the ultimate fact. What is there said is entirely appropriate to certain cases, and such no doubt the court had in mind. The question-to the expert, however, passed upon in that case was entirely proper, and within the rule laid down by Judge Marshall in Taylor v. Railroad, supra, and the propriety of that question is all that is decided in Wood v. Railroad.
We think the hypothetical questions were improper, in that they called for a conclusion on facts to be determined by the jury. We think them otherwise unobjectionable. The facts assumed were based upon the testimony. If the questions did not include all the facts testified to, that was a matter for cross-examination.
*254V. We cannot extend this opinion to the extent which would be required to notice all of the assignments of error in overruling objections to testimony. We will notice some of them.
The testimony of the nurse about a consultation between the nurses and Dr. Twyman, not in the presence of the defendant, was improperly admitted. It conveyed to the jury the impression that the nurses and Dr. Twyman had condemned the defendant. In other instances the nurses were permitted to get their suspicions before the jury. So with regard to the reasons given by Dr. Stewart for his visits to the defendant’s office, when he took away the culture tube. This testimony communicated to the jury the suspicions of Dr. Stewart, giving to such suspicions the practical effect of evidence.
The autopsy was performed by several doctors, among whom was Dr. Twyman, who died before the trial. One of these doctors was asked on the stand this question: “Was there any disagreement among the consultants then at the autopsy as to the conclusions arrived at!” The answer was, “I heard none.” This question and answer involved the opinion of Dr. Twyman’ as well as- others. They present the vice of hearsay testimony, where there is no opportunity for cross-examination. Statements of various parties, not made in the presence of defendant, were erroneously admitted; also evidence of orders given by Dr. Twyman to destroy medicines. This testimony, in effect, gave the jury the opinion of Dr. Twyman. The nurse, Miss Van Nuys, was asked this question: “Would you have remained in the house if the defendant had?” Answer, “No, sir, I would not.” This was obviously improper. It placed before the jury the opinion of the nurse that the defendant was guilty of improper conduct. It was also improper, for the same reason, to allow this witness to testify that she carried her medicines in her shirt waist all the time Dr. Hyde was in the *255house. It is unnecessary to particularize further. All hearsay testimony, and all evidence tending to bring out the suspicions and opinions of the nurses and others upon the character or conduct of the defendant, should, upon principles too familiar to require further elucidation, have been excluded.
VI. Objection is made to remarks and comments by the court in the presence of the jury. When a trial judge is conscientiously striving to do his duty, and is engaged in a long-drawn-out, arduous and hotly contested trial, it would be captious and unfair to criticise adversely every utterance of doubtful propriety. He is obliged to act and speak without time for calm deliberation, and is often fretted beyond the limits of patience by persistent counsel. The record in' this case indicates' a single purpose on the part of the trial judge — to permit a full and fair investigation of the facts and the law in this case. Indeed, we think he was too patient of interruption and argument, thus protracting the trial beyond reasonable limits of time, and creating a record exceedingly lengthy and difficult to read and digest. Some of the remarks of the court, we must notice and disapprove. Concerning a witness on the stand, Margaret Swope, the court remarked: “When she said she did not remember, that is as much as you can ask the witness. Objection sustained. The witness says she does not remember. That is as far as you can go with an honest witness.” The foregoing is objectionable, as being a comment on the credibility of the witness — a matter peculiarly within the province of the jury. This particular remark would not constitute reversible error, because not excepted to. Again, by the court: “The objection is overruled for .the reason that the witness has shown no lack of recollection, but has testified directly to the same state of affairs.” This statement of the court was objected *256to, and properly so. [Rose v. Kansas City, 125 Mo. App. l. c. 235.]
VII. It is assigned as error that the court permitted improper remarks, in argument to the jury, by counsel for the State. We fail to find any such, excepting the statement by the prosecuting attorney that he had the name and genealogy of every dog that was killed at that time, referring to the claim of defendant that he purchased cyanide to kill dogs. This remark was manifestly improper.
VIII. A witness for the State, Tom Swope, had testified that he saw defendant throw away something, and stamp it in the snow on the night of December 1.8th; that he immediately afterwards picked up at the place some broken capsules which he exhibited to the nurse, Miss Van Nuys. On the stand, Miss Van Nuys testified that Tom Swope showed some capsules to her that night; that she smelled them and recognized the odor as that of cyanide, with which she was familiar. After she had so testified, the druggist, Brecklein, was permitted to put cyanide upon his fingers, to moisten them, and then Miss Van Nuys was allowed to smell his fingers thus prepared, all in-the presence of the jury. She was then allowed to testify that the odor from Brecklein’s fingers was the same odor she had smelled on the capsules aforesaid on December 18th. If this demonstration had been had to determine whether the substance on Brecklein’s fingers was cyanide, and if Miss Van Nuys were expert on the subject, it might have been admissible. If she was expert as to cyanide odor, no such demonstration was needed. She could simply state, as she did, that she knew the odor of cyanide, and recognized it on -the capsules. If she was not expert, then this demonstration -could only be on the theory that, while she did not know the nature of the odor which she smelled in December, she remembered and could identify it by comparison with tlm *257present odor of cyanide. This method of refreshing' the recollection of a witness is certainly novel, and, so far as we know, without precedent. No case of the kind is brought to our attention. We think, in view of the fact that Miss Yan Nuys had qualified as an expert on the odor of cyanide, that her testimony as such should have been taken for what it was worth, without this demonstration. In any event, we think this theatrical display was improper.
IX. The trial of this case began April 11, 1910, and lasted until May 13, the verdict of the jury being returned May 16th. The defendant was at large on a $50,000 bail bond. On the 27th of April, being the 16th day of the trial, the court made the following order, after the jury had retired from the court room: “The' court orders that the order admitting said defendant to bail, heretofore made herein, be, and the same is hereby, set aside andlheld for naught: and the court further orders that the said defendant, B. Clark Hyde, be remanded to the custody of-the marshal of Jackson county without bail.” In connection with this order the court made the following statement: “In view of the testimony that has been thus far given in this trial, the court is constrained to say that it amounts to a presumption that under the law deprives the defendant of the right to go on bond, and he is hereby and for that reason remanded to the custody of the marshal.” Thereupon the defendant, through his counsel, excepted to the action of the court m remanding him to jail, on the ground that such action would have the effect of prejudicing his rights before the jury, being construed as a declaration on the part of the court against his interest, and leading the jury to believe that he is guilty of the offense charged; and the court was asked to rescind the order for the above reason; to which the court replied: “The court states *258that the defendant will not be locked up with the jury in the same place with them, and knowledge cannot go to the jury from anything that the court can prevent.” The defendant again excepted.
The motion for a new trial urg’ed this action of the court as a ground of error. Upon the hearing of the motion there was' filed in court the affidavits of eleven of the jurors to the effect that they did not know until after the trial was over that the bond had been revoked, and consequently were not influenced in their verdict. It is stated by counsel for the State in their brief that the twelfth juror-wrote a letter to the effect that he had decided the case upon the evidence, but did not state whether he knew that the bond had been revoked.
An affidavit was offered by the defense, and rejected by the court, of one Frank O’Reilly, to the effect that one of the jurors, Mr. Clay Pool, on the day the jury were discharged, told him that on the evening Dr. Hyde was sent to jail he, the said juror, read the headlines in the Kansas City Post stating that Dr. Hyde was sent to jail by Judge Latshaw; that this paper was folded in such a manner that tJie above information could be read, and that it was placed between the jamb and the door leading into the jurors’ dining room.
The defense urge this assignment of error upon the ground that the action of the court in remanding the defendant to the custody of the marshal was prejudicial to the defendant in that it tended to impress the jury with the belief that the court considered that the Sate was making a strong case against the defendant.
• The recognizance in this case was conditioned that the defendant would personally appear before the court on the first day of the next term, and from day to day during said-term, and from term to term, and not depart the court without leave thereof. In a sense, the *259effect of the bail bond was to place the defendant in the custody of his bondsmen, but he was in custodia legis.- “A man’s bail are looked upon as his gaolers of his own choosing, and the person bailed is, in the eye of the law, for many purposes esteemed to be as much in the prison of the ¡court by which he is bailed as if he were in the actual custody of the proper gaoler.” [2 Hawkins, Pleas of the Crown, p. 140.] It is said in 1 Hale, 325: “Yet the law is all one if he be under bail, for he is in custodia still, for the bail are, in law, his keepers.” Wharton, in his work on Criminal Pleading and Practice, says: “The principal is supposed to be in the bail’s constant custody, and the latter being the'former’s jailor, may at any time surrender him to the custody of the law.” [Sec. 62.] Again, in the same section, speaking of the power to arrest outside the jurisdiction, he says: “The bail represents the court from which his authority emanates. Where the court has no power to arrest, the bail has no power to arrest.” The power to surrender his principal to the court at any time is given to the surety by our statutes, section 5230', Revised Statutes, 1909. The recognizance is not a contract by which the defendant secures an unrestricted right to be at large; nor does it deprive the court of its inherent right to deal with the person of the prisoner.
The defendant in this case does not contend that the court had ho power to order him into custody while the trial was in progress. The contention is that, under the circumstances, the order was an abuse of discretion, and prejudicial to the interests of the defendant.
It must be conceded that, under some circumstances, it might become the duty of the court to order the defendant into actual custody. If, for mstance, it should clearly appear to the court that the defendant intended immediate flight. Furthermore, if such order is made, it must be assumed, in the absence of *260a contrary showing, that the court acted in good faith and upon sufficient grounds. In this case, however, the court stated-its reasons for the order, and we must determine whether they are sufficient to justify this very unusual proceeding. It is not uncommon to order the defendant into custody at the beginning of the trial. In some States this course is authorized by statute. The case of People v. Williams, 59 Cal. 676, involves such a statute, also the case of State v. Baker, 125 N. W. (Iowa), 660; -yet in this latter case, the exercise of the power • was held to he discretionary, and was justified because it appeared that the court had grounds to believe that the defendant was feigning sickness, and designed to avoid a trial altogether. In the case of Hull v. Reilly, 87 Mich. 497, cited by the State, the defendant was ordered into custody, as here, while the trial was in progress. No reason is given. The Supreme Court refused to mandamus the trial court to take hail. There is no opinion filed. Evidently the preliminary writ was refused. There may have been a showing of good cause for the order by the trial court, or the Supreme Court may have assumed, in the absence of a showing, that the tidal court had good cause.
The record before us shows that the defendant had been let to hail on a $50,000 bond. There is no claim that this bond was inadequate in amount or weak in security. The defendant was in regular attendance at court, and had made no effort to delay a trial which was had within five weeks after the indictment was returned into court. He was supported by eminent counsel, and had apparently ample pecuniary means. His home and business were in Kansas City. That he could give a $50,000 bond indicates that he had friends. The case had obtained such widespread notice that flight and concealment would have been extremely difficult. The indictment charged murder in the first degree by poison. In such a case the court *261must have assumed, when bail was allowed, that tbe defendant, if convicted, would almost certainly be found guilty of first-degree murder. In ordering tbe defendant into custody, tbe court gave as its only reason an intimation that, to the mind of the court, the case was developing strongly against the defendant,- and that the evidence already given showed that the presumption of guilt was great. No testimony for the defense had been heard. In the opinion of the trial judge, on the motion for a new trial, preserved in the record, appears this statement, in explanation of the order of commitment: “The prosecuting attorney had lost the grand jury notes, and-if the court had lost the defendant, the State would have lost the case without bringing the defendant to trial. ” We will not pause -to criticise phraseology, which might be construed to indicate solicitude on the part of the court lest the State should lose its case, but proceed to say that we are unable to perceive any relation between the loss of the grand jury notes and danger that the defendant might abscond. If the loss of the grand jury notes weakened the State’s case, this fact would hardly present an additional inducement to flight by the defendant. Ye are not prepared to hold that the mere fact that the case for the State, as it proceeded, raised a strong presumption of guilt in the mind of the court,' justifies, in the light of all the circumstances of this case, an order revoking the bond, and committing the defendant to custody. Furthermore, it appears from the record that to the knowledge of the court the defendant had, by depositions taken in -civil causes, become familiar with the evidence for the State, and knew when the trial began substantially all that was later developed from witnesses on the stand. That he should, with his knowledge, appear for trial, certainly indicated that he intended to put in his defense. The trial court, in its opinion on the motion for a new trial, after stating that the defendant had secured this advance knowl*262edge of the State’s case, says : “In view of the foregoing unusual opportunities and sources of information, it is, to say the least, very doubtful whether counsel of such knowledge and ability as defendant’s counsel were, were taken unawares or surprised at the evidence introduced in support of the indictment in this case.”
We must hold, under the circumstances, that the order complained of was improper.
X. The instructions for the State were in proper form and of proper substance, except number 14, which, according to the views herein expressed, should not have been given. Instructions numbered 5-, 7, 8, 9, 10, 11 and 12, withdrawing the evidence of other crimes, should have been, given.
XL We deem it unnecessary to notice other assignments of error.
The judgment is reversed and the cause remanded.
Kennish, P. J., and Broom, J., concur.