Reversing.

Appellant at the time of his injury was operating a service station in Louisville, and in connection therewith *Page 837 sold soft drinks from an ice cooler. These drinks included R. C. Cola which had been purchased from appellee. On the evening of November 4, 1942, when Gordon was straightening up and counting the bottles in the box, one of the cola bottles exploded severing tendons in the thumb of his left hand, and otherwise injuring that member. He filed suit seeking to recover for the consequent pain and suffering, loss of time, medical bills and partial permanent injury.

The original petition alleged that appellee in the manufacture of the contents, and the use, handling and filling of the bottle, did so in such negligent manner as to cause it to later explode. By amendment it was specifically charged that at the time of its being filled the bottle was defective, which fact was known, or by the exercise of ordinary care should have been known, to appellee. Answer was denial and plea of contributory negligence; it was agreed that the particular bottle had been supplied by appellee. The case was tried on the sole issue of defective container, and upon completion of appellant's proof the court directed the jury to find for appellee, and the case comes here on the question of the propriety of the court's ruling.

The court, in a written memorandum, concluded from the testimony that there was a proven defect, in the container, which may have induced the explosion, but that the proof lacked the "most essential evidence to make the defendant liable, and that is, he must have proved not only the defective condition and the explosion, but we must have in proof the fact that the explosion was caused by that defective condition."

Gordon's testimony was in substance that he was counting the bottles in the ice box, and setting up straight those which were lying down. He had set up about eight or ten bottles and reached to pick up another and it exploded, causing the injury. A boy who was working for Gordon picked out the pieces of broken glass and placed them in a box; these pieces were exhibited. The cap on the bottle was still in place after the alleged explosion. On cross-examination Gorden said the bottle had been placed in the cooler about fifteen hours before the explosion. No R. C. Cola had been in the cooler on the previous day. He did not know how many people had served themselves with drinks during the day. The custom was for patrons to select the drink, uncap the bottle and pay, and *Page 838 there were numerous customers. He did not recall whether the box had been iced that day or not. He knew that shaking the container would increase the gas pressure if the bottle was warm. Customers did not usually "root around in the water" for a bottle, but would pick them up by the top. He was sure that in picking them up he did not strike two bottles together. He saw no break in the bottle when he picked it up; the bottle which he was handling was the only one to explode that day.

Thomas Jenkins, the helper, said that he was near Gordon at the time and heard the explosion and saw that his hand was cut. On cross-examination he did not know what caused it to break; Gordon had set up a number of bottles before reaching the one which exploded. This witness said the box had been iced that morning. The broken bottle was turned over to Ed Sullivan, an estimator and engineer, familiar with the use of a micrometer; the one he used would measure accurately to one-thousandth of an inch, and he made five or six readings. He found variations in the thickness of the glass on different sides of the bottle to be about .1, and the maximum thickness to be about .2 inch at the thickest part of the bottle. Sullivan demonstrated to the jury his method of making the measurements, reading at different points showing the variations.

On cross-examination witness said he was not an expert on bottles, and did not know how thick the glass should be; that insofar as he knew the thickness of this particular bottle might have been sufficient to hold the contents. He said the bottle was split almost from the top to the bottom, except the neck; that the bottle was thicker as you measured from the middle toward the bottom; he was not qualified to say that the bottle was defective, and could not say what caused the explosion.

Appellant introduced, as on cross, F.M. Frazier, secretary of the appellee company, who had served with the company for about six years. He testified in substance that the company purchased bottles from two manufacturing companies. Some bottles are returned by customers and refilled, but before being first filled or refilled all are subjected to four different washings, sprayed and brushed. When the bottles come from the washing machine they are twice inspected for the purpose of detecting defects, or presence of foreign substances, but chiefly for defects such as chipped necks, or cracked *Page 839 bottles. The modus operandi is to place the bottles on a moving conveyor which carries the bottles past the inspector. A strong light is thrown against the bottle, which permits the inspector to see the whole bottle "over it, under it and through it in every way." The method of testing was the best that witness thought had been devised. If any defect was found the bottle was cast aside, but if none found it went to the filling machine to receive the carbonated contents. The gas used creates a ten pound pressure, which might rise to 50 pounds if "extremely agitated." This witness said that the thickness of the sides of the bottle was uniform. The R. C. bottles were the same thickness from top to bottom, and the bottle was straight until reaching the neck, and it then curved in sufficiently to hold the cap, the bottom somewhat thicker than the sides. The bottle weighed sixteen ounces, and held twelve fluid ounces of beverage. This witness was asked: "It is impossible for a bottle that is thinner on one side than the other to be filled by your product?" And he replied: "I don't know whether I could say 'impossible' but it is highly improbable." Mr. Frazier in describing the bottles used said the bottom was thickest, "but the sides and neck are the same thickness." He was then asked: "If one side of the bottle is thinner than the other side, is that a defective bottle? Ans. Yes, that is a defective bottle," but added that a bottle thus defective would likely have been broken in the filling machine.. Again, Mr. Frazier was asked whether there was any inspection made at any time that would indicate whether or not "these bottles" are uniform in thickness, and he frankly said: "Yes, sir, that could be detected by use of the light projecting mechanism, and by its use any defect in the walls of the bottle should be discovered."

In the face of this proof and the undenied proof of Sullivan that micrometer readings showed a decided variation in the thickness of the walls of the bottle, and the entire absence of proof which shows no more than a bare possibility that the explosion might have been caused otherwise, there appears no reasonable cause for the explosion other than a defective container.

A review of our cases dealing with explosions of bottles containing carbonated beverages throws some light on the present case, though none so far as we observe, present similar facts. These cases are Stone v. Van *Page 840 Noy R. N. Co., 153 Ky. 240, 154 S.W. 1092; Coca-Cola Bottling Works v. Shelton, 214 Ky. 118, 282 S.W. 778; Loebig's Guardian v. Coca-Cola Bottling Co., 259 Ky. 124, 81 S.W.2d 910. In the Stone case there was an explosion of a single bottle and we concluded that the defendant could not be held liable for the explosion, saying (153 Ky. 240, 154 S.W. 1095):

"In the absence of any direct testimony, or any fact or circumstance from which it could reasonably be concluded that the Merchants' Company knowingly used defective bottles, it cannot be held liable on this ground, even though the bottle which injured plaintiff was defective, which the proof in this case utterly fails to show."

In the Shelton case we distinguished the Stone case, because the evidence showed that on the afternoon Shelton was injured by the explosion of one bottle, twenty-five or more others had exploded; that by reason of the numerous explosions under similar circumstances (214 Ky. 118, 282 S.W. 780), "it may be inferred that they were defective or were improperly charged," matters upon which no evidence was had. In the Loebig case we went into the field of the doctrine of res ipsa loquitur. There was a single explosion, and we held the doctrine not to apply, holding the facts more analogous to those in the Stone case. We said (259 Ky. 124, 81 S.W.2d 911): "Suppose the bottle was defective. A conclusion of negligence could not arise without some showing that the defect could have been discovered by the exercise of ordinary care. * * * it is not shown that these tests * * * would have disclosed the complained-of defect." Here we have proof of the condition of a bottle, and that the method of inspection could or should have developed the defect.

In the case of Macres v. Coca-Cola Bottling Co., 290 Mich. 567,287 N.W. 922, 928, there was an explosion causing injury, under circumstances quite like those here presented. The company was held to judgment. A reading of the case manifests that the court applied the doctrine of res ipsa, because the evidence "pointed only to one direction that a standard bottle was overcharged." In a concurring opinion the court quoted the rule in the Restatement of Torts (2) sec 395: "A manufacturer who fails to exercise reasonable care in the manufacture of a chattel which, unless carefully made, he should recognize as involving an unreasonable risk of causing substantial *Page 841 bodily harm to those who lawfully use it for a purpose for which it is manufactured and to those to whom the supplier should expect to be in the vicinity of its probable use, is subject to liability for bodily harm caused to them by its lawful use in a manner * * * for which it is manufactured."

In the recent case of Seale v. Coca-Cola Bottling Works,297 Ky. 450, 179 S.W.2d 598 (foreign substance in bottle), we held that if a plaintiff introduces the best evidence of which the case is susceptible at the time of trial, and it is reasonable to infer the ultimate fact to be proven from the evidence introduced, such showing will be deemed sufficient to sustain a finding that the ultimate fact has been established. The conclusions announced and the reasoning indulged in that case fully justify the opinion that appellant's proof was sufficient to carry the case to the jury, hence the judgment is reversed with directions to grant a new trial consistent herewith.