The plaintiff in error, hereinafter referred to as the defendant, was convicted of murder in the first degree, with no recommendation to mercy, and to the judgment of conviction he takes writ of error.

The deceased was a police officer of the city of West Palm Beach. The defendant resisted an effort of the police officer to place him under arrest, and during the encounter which ensued the police officer was fatally wounded with a bullet from his own pistol. The State claims that the bullet was fired by the defendant from a premeditated design to effect the death of the officer, while the defendant claims that the pistol was discharged by him accidentally during the struggle.

The first assignment of error questions the correctness of the order overruling defendant's motion to set aside the verdict and grant a new trial. The grounds of the motion are as follows:

"1. The verdict is contrary to the evidence.

"2. The verdict is contrary to the law.

"3. The verdict is contrary to the law and evidence. *Page 274

"4. The verdict is contrary to the instructions of the Court."

Under the first ground of the motion the defendant questions the sufficiency of the evidence to support the verdict. The verdict was rendered on March 5, 1924. On March 8, 1924, the defendant moved the court for an extension of time of fifteen days in which to file his motion for a new trial, which motion was granted. The motion was filed on March 18, 1924, more than four days after the rendition of the verdict. The motion for extension of time was apparently made under the supposed authority of Section 2811, Revised General Statutes, 1920, providing, in effect, that in civil cases the trial judge, upon cause shown, may by an order made within four days after the rendition of the verdict and during the same term, extend the time for the making and presentation of such motion, not to exceed fifteen days from the rendition of the verdict. That this section is distinctly confined to civil cases is plainly shown by its caption, as well as by the concluding sentence of the section, which is as follows: "The provisions of this section shall not apply to criminal cases." This sentence is Section 3 of the original Act, Chap. 5403, Acts of 1905.

In criminal cases, the motion for a new trial must be filed within four days after the verdict shall have been rendered, and during the same term. Section 2810, Revised General Statutes, 1920. The provisions of this section are mandatory. Baxley v. State, 72 Fla. 228, 72 South. Rep. 677. There is no authority for extending the time for filing a motion for a new trial in criminal cases. We have repeatedly held that in order for an appellate court upon writ of error to review the sufficiency of the evidence to sustain the verdict, a motion for a new trial upon the ground of the insufficiency of the evidence must be duly presented to and acted upon by the trial court. Baxley v. State, supra. This motion was made more than *Page 275 four days after the rendition of the verdict. The case therefore stands as though no motion for a new trial had been made at all. It is obviously not error for the trial judge to deny a motion which in legal contemplation has never been made.

The third assignment of error questions the correctness of the ruling of the trial court in admitting evidence offered by the State of certain experiments purporting to show the distance at which powder marks and burns would be left upon an object when fired at by the pistol which inflicted the mortal wound upon the deceased, using ammunition identical with that used in the fatal encounter. In its rebuttal testimony, the State offered the testimony of one R. C. McGriff, who is apparently a man of wide experience in the use of firearms and the result of their operation and discharge. This witness testified that on the day of his testimony, using the pistol with which the deceased was killed, and using also the same kind of ammunition that was in the magazine of the pistol when it fired the fatal shot, the same powder load, the same jacket and otherwise the same in every particular, he fired several shots into certain targets of paper and cloth, holding the muzzle of the pistol at varying distances from the targets. The targets themselves, four or more in number, were placed in evidence. The first and second were paper targets, fired upon, respectively, at a distance of six inches and twelve inches. The third and fourth were clean linen handkerchiefs, fired upon, respectively, at a distance of six inches and three inches. The bill of exceptions indicates that two other targets, either of paper or cloth, fired upon at distances of one inch and one foot, were also admitted in evidence. Using these targets, the witness pointed out to the jury the marks left upon them by such gun fire and stated which marks were bullet holes, which were powder marks and which were powder burns. The effect of the *Page 276 gun fire upon the targets, however, was evidenced by the targets themselves. The defendant vigorously objected to this testimony, as well as to the introduction in evidence of the targets themselves, upon the ground "that the material of the targets was not such that would enlighten the jury on any question involved in this case; that it was not shown that conditions were similar; that the material used is different from any material before the jury with reference to this homicide." These objections were overruled and this ruling is assigned as error. The undisputed testimony shows that the fatal bullet wound entered the head of the deceased about an inch above and to the rear of the right ear. No powder marks or burns were found upon the flesh of the deceased. The State's evidence tended to show that when the defendant fired the fatal shot, he was standing some distance away from the deceased and hence that the shooting was not accidentally done in the struggle, but with premeditation and after the defendant had disengaged himself from the deceased. The defendant testified, however, that "when he (the deceased) got his pistol I knocked it out of his hand and when it fell to the floor it shot, and me and him were scrambling for it, and I beat him to it, and I struck him aside the head with it, and it shot. He tried to catch me and he fell in the front room." It appears that the latter shot was the fatal one. Under these circumstances it became very material to determine the distance at which defendant stood from the deceased when he fired the fatal shot. If the homicide occurred as the defendant said it did, there might have been burns and powder marks left upon the deceased. If it occurred as the State contends there would probably have been no powder marks. These matters would have a most material effect in determining the degree of homicide.

The experimental evidence tended to corroborate the State's witnesses in respect to the distance at which the *Page 277 pistol was held from the head of the deceased when the fatal shot was fired, at the same time refuting the defendant's assertion that the fatal shot was fired accidentally in the close proximity of the head of the deceased. In Lawrence v. State, 45 Fla. 42, 34 South. Rep. 87, this court cited with approval the rule as to the admissibility of experimental evidence of this nature as stated in 12 Am. Eng. Ency. of Law, 2nd ed., p. 406. The rule is there stated as follows: "Where a party seeks to show by evidence of experiments that an alleged result would or would not follow from the conditions proven, the experiments, evidence of which is sought to be introduced, must have been made under circumstances and conditions similar to those constituting, as it were, the premises from which the original event is alleged to have been the conclusion. Thus, where a material question was the distance between the muzzle of a gun and the body of the victim of an alleged murderous design, evidence was excluded of the result of experiments as to powder marks upon pasteboard targets by firing a gun at varying distances, upon the ground that the human body is fundamentally different in nature and texture from the substance upon which the experiments were made." This rule, in so far as it was applicable to the facts of the case, was adhered to in Spires v. State, 50 Fla. 121,39 South. Rep. 181, and in Hisler v. State, 52 Fla. 30,42 South. Rep. 692. In the latter case it was held that the similarity of circumstances and conditions go to the admissibility of the evidence, and not merely to its credibility, and that where such evidence is admitted over proper objections and it appears that the rule as to the similarity of circumstances and conditions attending the occurrence and the experiment does not appear to have been complied with in admitting the evidence, the error may cause a reversal of the judgment. Evidence of this kind is not favored by the courts and should be received *Page 278 with caution. It may be properly admitted only when the accuracy, precision and similarity of conditions and circumstances, so essential in such matters, has been preserved. It should only be admitted when it is obvious to the court from the nature of the experiments that the jury will be enlightened, rather than confused. The close similarity of conditions which would be requisite to the proper admission of such testimony was strongly intimated in Hisler v. State,supra, wherein this court said: "In many instances a slight change in the condition under which the experiment is made will so distort the result as to wholly destroy its value as evidence, and make it harmful rather than helpful." See also Lawrence v. State, 45 Fla. 42, 34 South. Rep. 87. Does the experimental evidence admitted in evidence here conform to this well-established rule? It should be borne in mind that the witness did not undertake to give in evidence his opinion as a duly qualified expert as to the distance at which the firing of a pistol of the kind and calibre that killed the deceased, and loaded with the same ammunition, would leave powder marks and burns upon human flesh. Testimony of this kind, by one properly qualified, is sometimes held admissible. Straight v. State,138 S.W. Rep. 742; Pollock v. State, 136 Wis. 136; Underhill Crim. Ev., Sec. 194; People v. Clark, 84 Cal. 573; 24 P. 313. In the case before us the targets themselves were put in evidence, accompanied by the testimony of the witness as to the various distances at which the shots were fired and the character and cause of the marks left upon the targets. The targets themselves constituted the evidence before the jury of the effect of the shots, at the distances named, with reference to the powder burns or marks. The experiments unquestionably comply with the rule with respect to the similarity of the weapon used as well as the ammunition. Differences, if any, in atmospheric conditions *Page 279 obtaining at the time of the homicide and at the time of the experiment might or might not affect the accuracy of the evidence. The evidence does not inform us upon this point. But is a sheet of paper or a linen handkerchief sufficiently similar to human flesh on the scalp in texture, substance, vulnerability and susceptibility to powder marks and burns to satisfy the rule and render the experiment accurate, faithful and helpful? The courts are not entirely harmonious in the answer to this question. In Thrawley v. State, 153 Ind. 375,55 N.E. Rep. 95, it was held in a trial for murder where the defendant claimed he shot the deceased in a hand-to-hand combat and in self-defense, that the result of experiments in shooting with defendant's revolver at blotting pads is admissible in evidence for the purpose of ascertaining the distance at which the shot would burn or powder-mark the scalp of the deceased, it having been shown that the cartridges used in the experiment were such as were used by the defendant at the time of the homicide. See also State v. Asbell, 57 Kan. 398,46 Pac. Rep. 770. In the latter case, however, the targets themselves were not introduced in evidence. The verbal testimony of the witness, based upon the experiments, was held to be competent. The doctrine announced by the Indiana case also finds support in State v. Jones, 41 Kan. 309, 21 Pac. Rep. 265; Boyd v. State, 14 Lea (Tenn.) 161; Fein v. Convent c. Association,67 Minn. 298, 69 N.W. Rep. 923, the two latter cases involving questions of civil liability. In many homicide cases where the mortal bullet first penetrated the clothing of the deceased, it is held that no error is committed in admitting evidence of an experiment made by shooting at a piece of the same clothing worn by the deceased when he was killed, or other material identical therewith, with the same pistol with which the deceased was killed, loaded with similar shells, for the purpose of showing the distance at which powder marks or *Page 280 burns would be left upon that particular material. Newkirk v. State, 106 A. 694; Pollock v. State, 136 Wis. 136;116 N.W. Rep 851; State v. Nowells, 135 Iowa 53, 109 N.W. Rep. 1016; Huestis v. Aetna Insurance Co., 131 Minn. 461,155 N.W. Rep. 643; People v. Fitzgerald, 138 Cal. 39, 70 Pac. Rep. 1014; Sullivan v. Commonwealth, 93 Penn. 284; State v. Asbell,supra.

In the case before us there is no evidence that the effect of the gun fire on the paper and cloth targets would be essentially similar to, or would even approximate, the effect of such gun fire upon the human flesh in respect to resulting burns and powder marks. Upon this most essential feature of the evidence we are left entirely to conjecture. Under the facts and circumstances of this case, and in the absence of qualified and credible evidence that the effect of pistol shots upon human flesh and upon paper or cloth targets would be essentially similar in respect to resulting powder marks and burns, we can not assume that there is sufficient similarity between human flesh and paper or cloth, in texture, substance, vulnerability or susceptibility, to render such targets either helpful or enlightening as evidence. Nor can we assume that the effect of pistol fire upon human flesh and upon paper or cloth targets would be essentially similar, in respect to resulting powder burns or marks, when the requisite supporting proof is lacking. One of the earliest cases supporting this view is State v. Justus, 11 Or. 178, 8 Pac. Rep. 337, 50 Am. Rep. 470. In that case the essential features of the homicide were closely similar to the case at bar. It was proposed to show only the results produced by the firing of a gun near a pasteboard target, from which it was claimed the jury was qualified and authorized to infer that a similar phenomena would be produced by "near" gun-shot wounds on the human body and that as a result of such inference the targets would serve to illustrate the gun-shot wound *Page 281 from which the deceased died, and thereby establish the point in issue. It was held by the Oregon court that there was no such similarity between the pasteboard targets and the human flesh, in respect to the effect of powder marks and burns, as would entitle the pasteboard targets to be received in evidence, in the absence of supporting testimony that the effect would be the same on each. The following soundly reasoned and frequently followed cases are some of those supporting the rule last announced: People v. Solani, 6 Cal. App. 103,91 Pac. Rep. 654; Gibbons v. Terri, 5 Okla. Crim. 212,115 Pac. Rep. 129; People v. Fiori, 108 N.Y. Sup. 416; Martin v. State, 71 S.W. Rep. 281; Reagan v. State, 208 S.W. Rep. 523; Commonwealth v. Piper, 120 Mass. 188; Tesney v. State, 77 Ala. 33; State v. Fletcher, 24 Or. 295,33 Pac. Rep. 575; McAlpine v. Fidelity Guaranty Co., 134 Minn. 192,158 N.W. Rep. 967; the latter a case of civil liability. Many other cases to like effect may be readily found in the reports. See also Wharton's Crim. Ev., Vol. 1, page 610; 12 Am. Eng. Ency. Law, 2nd ed., 406. The authorities last referred to constitute the safe weight of authority and the rule as therein announced conforms more closely to the previously announced views of this court upon the subject of experimental evidence. The admission of the targets, in the absence of the requisite supporting testimony, was harmful and prejudicial error.

The bill of exceptions discloses that the cap worn by the deceased at the time of the fatal encounter was introduced in evidence, but there is nothing in the record to indicate whether or not the fatal bullet penetrated the officer's cap before entering the scalp. If it did, however, this circumstance would not alter our view of the matter, in the absence of the requisite supporting testimony as to similarity of texture, color and substance between the officer's cap and the targets. In Reagan v. State, supra, a white paper *Page 282 target was held inadmissible, for the purpose under consideration, where the evidence showed that the deceased wore dark clothes at the time he was shot, and there was no evidence that the effect of powder on cloth would be similar to that on paper.

For the reasons stated, the judgment must be reversed and the defendant awarded a new trial.

Judgment reversed.

WEST, C. J., AND WHITFIELD, TERRELL AND BROWN, J. J., concur.