Spencer, C. J.
This is an action to recover damages for injuries sustained by appellee by reason of the fall of a heavy steel plate on his foot. At the time his injuries were received appellee was in the employ of appellant and, with other employes, was engaged in lifting the plate from the floor to its position in a certain machine known as a gas producer. This work was being accomplished by the use of a rope and pulley attached to an I-beam, about twenty feet above the floor, by means of a certain dog or clamp, of which a'sketch is reproduced herewith.
While the plate was being lifted, the dog or clamp slipped from the I-beam and thus allowed the plate to fall on appellee’s foot, producing the injuries complained of. The cause was tried by a jury which returned a verdict in favor of appellee and from a judgment on the verdict this appeal is taken.
*6551. 2. *654Appellant seeks to challenge a number of the instructions given to the jury but appellee earnestly in*655sists that no question is properly presented as to any of said instructions for the reason that the record does not show that before charging the jury the court indicated its disposition of such instructions as were tendered to it. Section 561 Burns 1914, Acts 1907 p. 652, expressly provides that the court shall indicate “before, instructing the jury, by a memorandum in writing at the close of the instructions so requested, the numbers of those given and of those refused.” (Our italics.) The context requires that the words “given” and “refused” should be construed in the future tense. When so construed, the meaning and direction of the statute is plain beyond doubt and unless it appears that its provisions have been substantially complied with, no question is properly presented as to the giving or refusal of instructions. Holcomb v. Norman (1909), 43 Ind. App. 506, 87 N. E. 1057; Petrie v. Ludwig (1908), 41 Ind. App. 310, 83 N. E. 770. At the close of the instructions tendered by appellant there is a memorandum, dated and signed by the trial judge, in which are indicated by number those of such instructions as were refused by him and those which were not refused. This memorandum, in substance,. complies with the provisions of the statute except that it does not show affirmatively the fact of its preparation before the jury was instructed. The better practice is to show by the memorandum itself that it was prepared in strict compliance with the statutory requirements and then to indicate by record entry that the court in fact gave such of the instructions tendered as it had previously indicated it would give and refused the others. In the case at bar this practice was not adopted but as, in the absence of an affirmative showing to the contrary, the presumption is that the trial court did its duty, and as the memo*656randum before us is in other respects sufficient, we must now assume that it was prepared at the proper time.
3. It further appears, however, that certain other instructions were tendered by appellee with a request that each of them be given to the jury and with a further request that the court indicate by a memorandum in writing which of such instructions would be given and which would be refused. From the original record in this cause it does not appear that a memorandum of any kind was endorsed by the trial judge on the instructions so tendered by appellee and there is nothing in the record to suggest that he did in fact comply with the requirement of the statute in that regard. As we have already determined the presence in the record of a memorandum which is in other respects sufficient will give rise to the presumption that it was prepared at the proper time and such presumption will prevail unless affirmatively rebutted. It is equally true that from the absence of such a memorandum it must be assumed that none was prepared, and its absence cannot be supplied by record entry or other endorsement which does not meet the requirements of the statute. In the case at bar, however, appellant has sought to bring into the record by writ of certiorari a memorandum signed by the trial judge showing that each of the instructions tendered by appellee was given to the jury. But the return to the writ also shows affirmatively that such memorandum was in fact prepared and signed after the jury had been instructed. This is not a compliance with the provisions of the statute and precludes our consideration of the instructions. As the instruc-' tions tendered by appellee are not properly in the record, no question is presented as to any of the instructions given or refused. Lake Erie, etc., R. Co.. *657v. Holland (1904), 162 Ind. 406, 416, 69 N. E. 138, 63 L. R. A. 948; Indianapolis Traction, etc., Co. v. Gillaspy (1914), 56 Ind. App. 332, 105 N. E. 242.
4. 5. Appellant further contends that the verdict of the jury is not sustained by'the evidence. As opposed to this contention appellee asserts that the evidenee, if in the record, fully sustains the verdict but insists that, in fact, the evidence is not properly before this court for the reason that in a number of instances various witnesses “indicated” or “illustrated” certain phases of their testimony and that the nature of such demonstrations does not clearly appear from the record as presented here. Conceding that the record suggests with some frequency that witnesses accompanied their testimony with a demonstration, the meaning in each instance is apparent and, although such practice has its dangers, we cannot agree that the record before us is insufficient in that respect. The most serious objection urged by appellant to the sufficiency of the evidence is that the verdict is based on inferenees drawn from other inferences and not from facts proved. The theory of appellant’s complaint is that his injuries are the proximate result of" a defective condition in the dog or clamp which was attached to the I-beam. The only proof of such a condition is that some of the threads on the bolt “A” were battered and worn and that there was a “shoulder” on said bolt which, appellee contends, prevented the nut “B” from being turned to its proper position. Construing the evidence most favorably to appellee the existence of broken threads and a “bend” or “shoulder” on said bolt may be taken as a proved fact. It must then be inferred, there being no direct evidence on the subject, that such defect would prevent the clamp from being ' *658drawn tight on the beam and properly adjusted thereto. It is necessary to draw this inference as there is a failure in the direct evidence to show that said defects in the bolt were in such a place thereon that they could have prevented the proper attachment of the clamp to an I-beam four inches wide, as in this case. In other words, the defects which were proved to have existed were not such in and of themselves as might cause the clamp to fall without reference to other conditions. A bolt may have a shoulder or broken threads which would not in a ay degree render its use unsafe if there is no bearing on the defective parts and there is nothing in the record before us to show that it was possible for the defects in the bolt in question to have prevented a proper attachment of the clamp to the I-beam. If a fact, it was easily susceptible of direct proof, but from the record in this cause it must be drawn as an inference. From this inference the further inference must be drawn that the clamp did fall because the condition of the bolt prevented its being properly adjusted to the beam. Under the evidence in this case there is but one other inference possible which will explain the falling of the clamp, viz., that it was not fastened to the beam in a careful manner, but as appellee himself attached the clamp to the beam such an inference would be fatal to his case. The two inferences above suggested are necessary to connect the proved facts that the bolt was defective and that the clamp fell, and one of these inferences rests squarely on the other. In brief, to sustain this verdict it is necessary to infer two things; (1) that because the bolt had a shoulder and broken threads on it, such defects were so placed and were of such a character as would prevent the clamp from being properly attached to the I-beam, and (2) that the clamp fell because the defects in *659the bolt did prevent its proper attachment to the beam. The latter inference alone would be- sufficient to sustain the verdict if the proved defects in the bolt were such that the very fact of their existence would lead directly to an inference that they did cause the clamp to fall. But here there is an intermediate element necessary to be proved, ■ viz., that it was possible for the broken threads and the shoulder on the bolt to prevent its safe use for the work at hand. This fact, if it be a fact, rests wholly in inference and without it, the second inference stated above cannot be drawn. Since the final inference drawn by the jury depends on the existence of a condition which is also a matter of conjecture and not a proved fact, such inference is not sustained by the evidence. Evansville Metal Bed Co. v. Loge (1908), 42 Ind. App. 461, 467, 85 N. E. 979.
Appellee relies on the case of L. T. Dickason Coal Co. v. Liddil (1911), 49 Ind. App. 40, 94 N. E. 411, to sustain this contention that the inferences drawn by the jury were properly drawn from facts proved. In that case decedent, an employe of the coal company, had been seen standing with one foot on the bumper of a tramcar and the other foot on the tail chain, driving a mule. When his body was later found under the wheels of the car, the tail chain was unhooked from the drawbar and the mule had moved away some distance. The Appellate Court said: “The character of the defects shown to exist in the hooks was such as to justify the jury in inferring that the tail chain became disconnected from the drawbar of the car by reason of such defects. If this happened while Raymond Liddil was standing in the position in which he was last seen, the jury might infer that he was thrown to the track in front of the ear, and that the momentum of the car caused it to run' upon him and cause his death. *660The jury evidently drew this inference, and under the facts, proved, we can not say that it was not justified in doing so. The jury has’ a right to consider any fact as proved that can be rightly and reasonably inferred from the facts proved in the case.” That decision is in harmony with the rule here stated. Conceding that the jury in the Liddil case drew more than one inference, each was independent of the others and was based directly on facts proved to have existed. The negligence charged was the use of a hook on the drawbar of the car which was not bent back sufficiently but formed a right angle to the car; also of a hook on the tail chain which was similarly defective. There, as suggested by the opinion of the court, the defects in the appliances furnished were of such a character as to lead directly to the only inference necessary to be drawn to connect such defects with the death of decedent. The other inferences drawn served only to explain the position of his body when found. We cannot agree that this case supports appellee’s contention and hold that the verdict in the case at bar is not sustained byjbhe evidence.
Other questions presented may not arise on a new trial of the issues, and, therefore, need -not be considered. Judgment reversed, with instructions to sustain appellant’s motion for a new trial and for further proceedings in accordance with this opinion.
Lairy, J., dissents.
Note. — Reported in 109 N. E. 960. See under (1) 3 C. J. 154, 38 Cyc 1770; (2) 4 C. J. 766, 3 Cyc 303; (4) 4 C. J. 535; 3 Cyc 166; (5) 16 Cyc 1051.