SHINN, J.
Plaintiff sustained severe injury to one of his eyes, which was cut by a particle of glass from a bottle of root beer which burst as he removed it from a beverage container in the Sportsman’s Club, an establishment operated by defendant Faber. He recovered damages against Faber, who had defaulted, and Double Cola Ice and Bottling Company in an action tried by the court. The bottling company, hereinafter referred to as defendant, appeals. Responsive to the allegations of the complaint, the court made the following findings:
“That it is true that the defendant Double Cola Ice and Bottling Company so carelessly and negligently manufactured, made, bottled, kept, maintained, distributed, and sold said beverage known as Real Tang Root Beer and so carelessly and negligently filled, sealed, inspected, kept, maintained, and distributed the same in said glass bottles as to ' cause said beverage to become and it did become explosive in said bottles and unfit for the purpose for which the defendants Double Cola Ice and Bottling Company and Sam Faber intended them to be used and that it is true that the said beverage in said bottles became dangerous to human health, welfare and well being. It is true that the defendant Double Cola Ice and Bottling Company knew or in the exercise of ordinary care should have known that said beverage in said glass bottles was inherently and imminently dangerous to human health, welfare and well being and unsafe for human use and consumption.
- “That it is true that at all times herein mentioned the defendant Double Cola Ice and Bottling Company warranted to said Frank Weinberger and through said Frank Weinberger to said Sam Faber and to the general public who might purchase the beverage contained in said glass bottles that the contents of said bottles was fit, pure, and safe for human consumption as a beverage and that it contained nothing, and had no inherent qualities, and had not been manufactured,
made, bottled, filled, sealed, inspected, distributed and sold in a manner which would make its intended consumption from the bottles in which it was contained or the handling, opening, and drinking thereof dangerous to such persons, including the plaintiff herein, as would purchase and consume the same or that would cause it to explode, burst and blow apart.” It was found that this warranty was breached by defendant.
“That it is true that the defendant Double Cola Ice and Bottling Company so carelessly and negligently manufactured said beverage and so carelessly and negligently placed said beverage in said bottle and so carelessly and negligently closed said bottle against the escape of its contents and so carelessly and negligently sold said bottle of beverage and so carelessly and negligently inspected said bottle so as to cause said bottle of beverage to become and it did become an instrumentality imminently dangerous to human health. ’ ’
The attack here is upon the sufficiency of the evidence to prove negligence or that there was any warranty or breach of warranty upon the part of defendant. The conclusion that we have reached is that the evidence was wholly insufficient to support either of the questioned findings, and that upon the undisputed evidence in the case no facts were shown which would support a judgment against defendant for damages.
Preliminarily we should say that it was not and is not contended that the root beer was unwholesome, unfit to drink, or that it contained any deleterious or foreign substance. No reliance was or is placed upon alleged violation of the Health and Safety Code (div. 21, ch. 3, art. 2), or other statutory law with respect to the beverage itself. The argument is that the bottle was not strong enough to withstand the gas pressure.
Defendant bottled the root beer in question, and sold it to one Weinberger, an independent distributor, who bought beverages in case lots and distributed them to his customers in his own truck; defendant retained title to the bottles, which were to be returned; Weinberger delivered the bottle in question, with other bottles, to Faber, who placed them in a large container in his establishment, where they were kept on ice until removed for sale by the proprietor, his employees or his customers, who were allowed to help themselves from the container. Plaintiff was so helping himself
to a bottle of root beer after asking Faber’s permission, when it burst in his hand before it had been removed entirely from the container.
The root beer bottles had a capacity of ten ounces. They were purchased by defendant from one of two established and reputable bottle-making concerns. They were such bottles as are ordinarily used in the trade. The following facts were established by testimony which stands without dispute in the record: both of the bottle-making companies used approved and standard methods of manufacture, testing and inspection. The bottles before they left the hands of the manufacturer were tested for temper of the glass and for tensile strength and also for thermoshock, which consisted of immersing them alternately in hot and cold water. One of the companies tested each bottle under air and hydraulic pressure of not less than 250 pounds per square inch before it was released from the factory. There was evidence that the bottles of the other manufacturer would withstand a pressure of from 350 pounds to 560 pounds per square inch. When new bottles were received from the manufacturer they went into defendant’s soaking machine, where they were cleansed and sterilized. Before going to the soaker the bottles were inspected visually and those that were cracked or chipped were discarded. In the soaker they were again picked up and examined and those found to be defective were thrown away. They then passed to a filling machine on a conveyor chain in front of a light, where they were again inspected and defective bottles were removed. In defendant’s process the bottles were filled by machine and gas was injected and the bottles were capped by other machines. The bottles were then under fifty pounds pressure to the square inch and went to an agitator, where the pressure was increased by agitation about five pounds; thence they went to an accumulation table, where they were again inspected in front of a light for defects in the bottles or foreign matter in the fluid, thence to the stock room and to distributors, who received the merchandise at the back door of the stock room. The pressure exerted in the filling operation registered on a gauge. A government inspected volume tester tested the volume and checked the pressure in the bottles. These tests were made two or three times on each run of bottles. The ingredients of root beer were at the time in question oil of sassafras, natural oil of wintergreen U. S. P., oil of anise lead free, and oil of cloves U. S. P., caramel coloring, gum of acacia,
sugar and water. There is nothing in the ingredients which of itself would cause an increase in pressure. Bottles were returned to the bottler and used many times. Used bottles which had been returned to defendant went through the same operation as new bottles, and the percentage of breakage was somewhat less in the used bottles. If a bottle has even a hairline crack it will not retain pressure. The evidence was that the machinery of defendant was in good order, that its gauges were operating properly during the month of plaintiff’s accident and that the gauges coincided with tests made on various bottles during each run. The use of gauges was standard practice and all machine equipment was such as is commonly used in the trade and came from the largest manufacturers of such machinery in the United States. The methods and procedure of bottling used by defendant were standard in every respect and defendant used all of the safety devices and tests known to the trade.
In the Faber establishment the bottles were immersed in ice water surrounded by ice, and the undisputed evidence was that the pressure in a bottle in such a cooler would be reduced to twenty pounds per square inch and would not be appreciably increased by taking it in the hand and removing it from the cooler.
As evidence of defendant’s negligence plaintiff calls attention to certain admitted facts, namely, that as many as 60 to 100 bottles a day were broken in the bottling process; that a boy picked up four bottles at a time in the soaker; that there was evidence that from thirty-six to forty-eight bottles were broken daily in the cleaning process; that about a case of bottles a day were broken in passing from the soaker onto moving platforms and that as many as four cases of bottles a day broke when put under fifty pounds pressure; that 4,500 bottles passed before the eyes of two inspectors within the course of an hour; that only five men were employed in the bottling operation, and that bottles broke or exploded in the storeroom. The argument is that defendant could reasonably be found negligent in using a bottling process in which so many bottles were broken.
The weakness of plaintiff’s argument lies in the fact that it was incumbent upon him to prove negligence consisting of the failure to exercise the degree of care that was commensurate with the hazard attaching to the handling of such
carbonated beverages, and that if defendant did nse the machinery, methods and processes which were accepted as standard in the trade and which were as reliable and satisfactory as any others, it did not fail in the exercise of proper care.
In a case involving a question of negligence in the manufacture of a chair, the court in
Sheward
v.
Virtue,
(1942) 20 Cal. (2d) 410 [126 P. (2d) 345], at 414, said: “The appropriate standard of care applicable to the facts of the present case is expressed in
O’Rourke
v.
Day & Night Water Heater Co., Ltd., supra,
[31 Cal. App. (2d) 364, 366 (88 P. (2d) 191)], to the effect that if the defective condition of the part could have been disclosed by reasonable inspection and tests, and such inspection and tests had been omitted, the defendant has been negligent. In
Smith
v.
Peerless Glass Co., Inc.,
259 N. Y. 292 [181 N. E. 576], it was held that reasonable care consisted of making the inspections and tests during the course of manufacture and after the article was completed, which the manufacturer should recognize were reasonably necessary to secure the production of a safe article.” This rule is the one under which defendant’s conduct is to be examined for negligence.
Immediately after the accident plaintiff collected the fragments of the bottle and later his attorney sent them to a chemist experienced in metal and glass testing work, who placed the pieces together (some twelve in number, one small piece being missing) and who testified that from observation the bottle had the appearance of having exploded. The chemist suggested to the attorney that the bottle be photographed. The latter left without giving any instructions, the pieces were kept for some time and eventually were thrown away. When asked for his opinion as to the breaking of the bottle,. the witness answered, “The bottle burst, I don’t know whether it was an accumulation of gas or a flaw in the bottle or what it was.” He further testified that it appeared that the contents had been forced “from the within rather than from the without,” and that he had no opinion as to why the bottle burst but that it might have been due to a flaw in the bottle, and he further said, “Well, there are a lot of factors if the conditions are right, that' is, a flaw in the bottle or often a minute flaw even from the outside or inside and if the conditions were right, that is, such as temperature or some sudden jar or impact or anything that may cause that bottle to give way ... if you put that bottle out
in the sun and there was a flaw in it, the bottle might burst from the increase of volume of gas, and if it was a bottle that had a flaw in it of some type, if it was weak at some point, it might give out at the point of least resistance.” Another witness for plaintiff, who was defendant’s plant manager during the month of the accident, testified that there was occasional breakage in the storeroom of bottled beverages, and he stated, “. . . wherever you find glass there will be breakage”; that he had reports of breakage in stores and ice boxes and coolers outside of the plant; that he had seen bottles explode, and when asked whether they might explode some time after being shaken or struck, replied, “That is a matter of opinion and my opinion is if a bottle is mishandled at some particular time, you may call it a defect in the bottle, then on the taking of the bottle with the warm hand or mishandling it later on, the bottle may explode and the original cause may be traced back to some time when the bottle has taken some particular abuse and causing the defect. . . . “Q. Now you yourself, Mr. Beber, have you ever seen a bottle setting quietly without being touched, suddenly explode ? A. Yes. Q. And when was that? A. They will explode under certain climatic conditions. You take a bottle and just let it set in the sun and if it sets long enough, it will explode if there is any defect at all in the bottle, that is if it is a carbonated drink. Q. If you let the contents of that bottle be exposed to the direct rays of the sun-A. I couldn’t tell you how long, I have seen them pop, particularly Pepsi-Cola. These are stronger bottles than Pepsi-Cola, but I have seen the truck driving along and have seen them pop up, I have seen more Pepsi-Cola burst than any other kind of bottles. Q. At the time you saw this Pepsi-Cola bottle explode, bottles explode, they had been subjected to the direct rays of the sun? A. Yes and the moving of the truck. Q. And when there is heat and when there is motion, that will increase the pressure in the bottle ? A. Yes. Q. And if there is a defect, that has caused bottles to break? A. Yes.”
The word “explode” is frequently used in the testimony and the accident in question is spoken of by respondent as an explosion. This is true in the limited sense that the breaking of the bottle, from whatever cause, released the contents because of the pressure under which they were contained, but it is of course not true in the sense that the substances themselves were explosive or that an explosive force would be
generated within the bottle. The bottle could have burst only-through one or more of several causes: (1) the pressure under which the bottle was filled was greater than it could stand; (2) pressure thereafter increased to the bursting point; (3) if the pressure was a safe one in the beginning and was not increased, the bottle burst because of the manner in which it was subsequently handled, which could happen whether the bottle was with or without a flaw, that is to say, it might have been broken if it had been flawless, but would have been more easily broken if it had contained a flaw.
Plaintiff’s witness, the chemist, as we have said, did not testify that there was any flaw in the glass, and an assumption that there was such a flaw would be wholly unsupported by any direct evidence. As we have said, the undisputed evidence was that the bottle, when taken from the cooler, would be under only about twenty pounds pressure as against fifty pounds at the time it was bottled, and the hypothesis that it burst solely because of excessive pressure would be wholly false in the absence of evidence that other bottles filled at the same time had burst. That the bottle had sustained some sort of crack before plaintiff took it out of the cooler was not unlikely in view of plaintiff’s testimony that he did not strike it against anything in removing it. That the glass was not cracked all the way through is demonstrated by the fact that the pressure in the bottle had not escaped. Plaintiff’s testimony that he did not strike the bottle on anything must be taken as true and it would then appear that if the bottle was not itself defective it had been struck and cracked at some other time, but not severely enough to cause a complete fracture, although sufficiently to break when plaintiff took it from among the other bottles in the ice in the cooler. A finding that it had been cracked while in the possession of defendant would be based upon mere supposition and not upon any evidence of defendant’s negligence. Furthermore, defendant was not an insurer of the safety of the bottle and if it had sustained some injury while going through the bottling operation, that fact alone, if established, would not show negligence in the absence of some substantial evidence that the results of the injury to the bottle were such as might have been discovered by reasonable and careful inspection and that no careful inspection had been made, and there is of course no such evidence.
Plaintiff’s entire evidence amounts to this: that glass bot-
ties filled with carbonated beverages do break from a variety of known causes, and others which are unknown, both before, during and after the bottling process. Add to this the undisputed evidence that the bottles were carefully made, tested and inspected, and that defendant’s operations were conducted without negligence and the sum total is proof that there is an unavoidable hazard in the use of glass bottles as containers for carbonated beverages. This does not make out a case of negligence.
The further contention is made that negligence was to be inferred under the res ipsa loquitur doctrine and reliance is had upon authorities from other jurisdictions, some of which apply the doctrine to the bursting of bottles containing carbonated beverages. The appellate courts of our own state have not passed upon the question, but as the doctrine has been developed in California it is clearly inapplicable to the bursting of a bottle after it has passed from the bottler into the hands of others. In support of his argument plaintiff relies upon
Grant
v.
Graham Chero-Cola Bottling Co.,
(1918) 176 N. C. 256 [97 S. E. 27, 4 A. L. R 1090] ;
Payne
v.
Rome Coca-Cola Bottling Co.,
(1912) 10 Ga. App. 762 [73 S. E. 1087];
Coca-Cola Bottling Works
v.
Shelton,
(1926) 214 Ky. 118 [282 S. W. 778] ;
Macon Coca-Cola Bottling Co.
v.
Crane,
(1937) 55 Ga. App. 573 [190 S. E. 879] ;
Riecke
v.
Anheuser-Busch Brewing Assn.,
(1921) 206 Mo. App. 246 [227 S. W. 631];
Stolle
v.
Anheuser-Busch,
(1925) 307 Mo. 520 [271 S. W. 497, 39 A. L. R 1001]; and
Healey
v.
Trodd,
(1939) 122 N. J. L. 603 [7 A. (2d) 640].
There are two well considered eases which to our minds demonstrate the fallacious reasoning of the cases relied upon by plaintiff. These are
Slack
v.
Premier-Pabst Corporation,
(1939) 40 Del. 97 [5 A. (2d) 516], and
Stewart
v.
Crystal Coca-Cola Bottling Co.,
(1937) 50 Ariz. 60 [68 P. (2d) 952], In these eases the authorities are reviewed at length, many of those discussed being found in notes in 4 A. L. R. 1094; 8 A. L. R. 500; and 39 A. L. R. 1006. In the Slack case the reasoning in most of the cases relied upon by plaintiff herein was criticized and rejected and the court said (p. 103 [518]): “The rule which is supported by reason, considerations of justice, and by the decided weight of authority, denies application of the doctrine of
res ipsa loquitur,
as against a bottler, to the explosion of a bottle of carbonated or fermented beverage in the hands of a dealer.” And further (p. 104 [519]) :
“It is no more probable that the defendant’s negligence was the cause of the explosion and resulting injury, than was the negligence of others who had the management, supervision and control of the bottle after it had been delivered safely by the defendant. In such ease, the plaintiff must fail.”
The same conclusion was reached in
Stewart
v.
Crystal Coca-Cola Bottling Co., supra,
where the authorities sustaining that conclusion are reviewed and reasons are given which we regard as satisfactory for refusing to apply the doctrine of res ipsa loquitur to the explosion of a bottle of coca-cola after it had been handled by a dealer and a customer and had been drawn from and replaced in a box containing broken ice. In
Slack
v.
Premier-Pabst Corporation, supra,
it is pointed out that the courts which had endeavored to apply the res ipsa loquitur doctrine had done so in eases involving charges of specific acts of negligence under circumstances which rendered the doctrine inapplicable.
There is a further reason why we cannot accept the reasoning of the cases upon which plaintiff relies. While purporting to apply the res ipsa loquitur doctrine, they either do not go on the rule at all or confuse it with the general rules for proof of negligence. We except from this statement
Coca-Cola Bottling Works
v.
Shelton, supra,
282 S. W. 778, a Kentucky case, where twenty-seven bottles of coca-cola exploded in a single afternoon, because the exceptional facts of that case distinguish it from all others as well as from the case we are considering. In all but one of respondent’s cases the instrumentalities which caused the accidents were not at the time of the accident in the possession or under the control of the bottler or manufacturer. Viewing such a situation as one where the res ipsa loquitur doctrine would not ordinarily apply, the courts accepted either allegation or proof exonerating all others than the bottler or manufacturer from a charge of negligence and held them liable under what they considered to be the res ipsa loquitur doctrine. That this would be an unwarranted extension of that doctrine is readily apparent. In a case where plaintiff undertakes by his proof to eliminate all of the causes except the negligence of one defendant, he cannot then invoke the doctrine as against that defendant; that is to say, in a case where control and management of the instrumentality is divided between two or among several persons, plaintiff by proving want of negligence on the part of all but a single defendant cannot
then call into play the res ipsa loquitur rule which would have had no application in the first instance. It either applies on proof of the circumstances of the accident without plaintiff’s fault, or it does not apply at all.
(Sanders
v.
Nehi Bottling Co.,
(Tex. 1939) 30 F. Supp. 332;
Keller
v.
Cushman,
(1930) 104 Cal. App. 186 [285 Pac. 399] ; see, also, cases cited
supra.)
The definition of res ipsa loquitur as a rule of law, quoted in
Hill
v.
Pacific Gas & Electric Co.,
(1913) 22 Cal. App. 788, 790 [136 Pac. 492], from the opinion in
San Juan Light & Transit Co.
v.
Requena,
224 U. S. 89, 99 [32 S. Ct. 399, 56 L. Ed. 680], emphasizes that the rule applies only where the instrumentality at the time of the accident was under the
exclusive control
of the defendant, and that is the interpretation of the doctrine which has been applied by our courts without exception. Where there is a division of responsibility in the use or management of the instrument which causes the injury, and such injury might in equal likelihood have resulted from the separate act or acts of either one of two or more persons, the res ipsa loquitur doctrine cannot be invoked against any one of them.
(Speidel
v.
Lacer,
(1934) 2 Cal. App. (2d) 528 [38 P. (2d) 477];
White
v.
Spreckels,
(1909) 10 Cal. App. 287 [101 Pac. 920].) In the last mentioned case it was pointed out that the res ipsa loquitur doctrine follows the familiar principle stated in
Searles
v.
Manhattan R. Co.,
101 N. Y. 661 [5 N. E. 66] as follows: “Where, in an action to recover damages for injuries alleged to have been caused by defendant’s negligence, it appears that the injuries were occasioned by one of two causes, for one of which defendant is responsible, but not for the other, plaintiff must fail if the evidence does not show that the injury was the result of the former cause. If under the testimony it was just as probable that it was caused by the one as the other, he cannot recover. ’ ’ The res ipsa loquitur doctrine applies only when the facts proved by the plaintiff admit of the single inference that the accident would not have happened unless the defendant had been negligent, although it continues to apply for plaintiff’s benefit as against evidence produced by defendant that it may have happened without his negligence.
(Lejeune
v.
General Petroleum Corp.,
(1932) 128 Cal. App. 404 [18 P. (2d) 429]; see, also, cases collected in 19 Cal. Jur. 708 and 10-Tr. Supp.) When, therefore, it becomes incumbent upon plaintiff to establish
negligence other than by application of the doctrine of res ipsa loquitur, because the proof does not show exclusive control in the defendant, he must go all the way with his evidence and prove some act or omission on the part of the defendant or defendants upon whom he would fasten responsibility. A contrary rule would upset many settled rules of evidence in negligence cases. We do not find in the authorities a real conflict as to the true rule of res ipsa loquitur. The apparent conflict has resulted from the failure of some courts to take into account its limitations and to use care in its application.
It is clear that plaintiff’s accident may have resulted from some cause other than any negligence on the part of defendant. It appeared from the testimony of the distributor Weinberger that the bottle which' exploded in plaintiff’s hand may have been on his truck for a week. There was no evidence that the bottle had been carefully handled during that time or during the process of its delivery to defendant Faber, nor by the latter or his employees when it was placed in the cooler. In addition to all of this, there is the admitted fact that Faber’s customers were allowed to help themselves from the cooler, and it does not appear how frequently that was done nor the manner in which it was done. The law does not impose liability upon one party in favor of another upon suspicion or mere supposition. Plaintiff had the opportunity and the only opportunity to prove, if it were a fact, that the glass contained a flaw, and his evidence, as we have seen, failed completely in that respect. The res ipsa loquitur doctrine cannot be applied so as to raise an inference that the bottle was cracked while in defendant’s possession, to the exclusion of the inference that it may have been cracked thereafter. As we have stated, it retained its pressure up to the time when it was handled by plaintiff, and it is at least as reasonable to infer that it was cracked after it left defendant’s warehouse as to infer that it was cracked before. If it was carefully bottled and handled by defendant and was cracked thereafter, defendant would not be guilty of negligence and the res ipsa loquitur doctrine would cease to be applicable from the time the bottle passed from defendant’s exclusive control.
A second ground for the decision of the trial court was a finding that the contents of the bottle were warranted to be fit, pure and safe for human consumption and contained no inherent qualities that would cause it to explode, that “the
contents of said bottled beverage did cause said bottle to burst and break into pieces” and that the warranty was thereby breached. The warranty sought to be invoked against defendant, and the sole one upon which plaintiff relies with reference to the contents, is that of a manufacturer, under section 1735 of the Civil Code, namely, that the goods were fit for the purpose for which they were sold. We find it unnecessary to go into the question of warranty further than to say that the evidence was insufficient to show any breach of warranty of the contents. Upon that point nothing need be added to what we have already said concerning the root beer itself. So far as disclosed by the evidence, it was not different from the general run of that beverage. It was alleged in the complaint that defendants also warranted the bottle. This allegation was denied, and the finding, as we have seen, limited the warranty to the contents of the bottle.
It will be recalled that defendant Double Cola Company did not part with title to the bottles. But it is now contended that defendant let the bottle and in so doing assumed the obligation to put it into a condition fit for the purpose for which it was let, which obligation is imposed by section 1955 of the Civil Code. The complaint did not allege the letting of the bottle by appellant or the other defendants, nor do the findings establish that there was a contract of letting, and necessarily it was not found that there was a breach of any such contract. But if we assume that the case was tried upon the theory that the allegations of the complaint were broad enough to charge a warranty by appellant as owner and letter of the bottle, and a breach of the warranty, the evidence was insufficient to sustain a recovery under that theory as well as under the theory that the contents of the bottle were explosive. All of the direct evidence and the reasonable inferences to be drawn therefrom are against the contention that the contents were explosive or otherwise unfit for use either because of their inherent qualities or excessive pressure in the bottle. There was no evidence that the bottle left the possession of defendant in an unsafe condition unless the bursting of the bottle be accepted as sufficient. That evidence, we have held, is not sufficient to establish negligence on the part of the bottler and it is also insufficient to establish a breach of warranty. In fact, we think it was more reasonable to infer that the bottle burst because of the manner in which it was handled after leaving
defendant’s possession, than because of its handling by defendant, during which it did not burst. What we have said along this line upon the issue of negligence is equally applicable to the alleged breach of warranty. The principal case relied upon by respondent on this point is
Naumann
v.
Wehle Brewing Co.,
(1940) 127 Conn. 44 [15 A. (2d) 181], in which a judgment for plaintiff for breach of warranty was affirmed against a bottler of ale. The trial court had found that there was a defect in the bottle, and this finding was sustained by reasoning which eliminated negligence in the manufacture of the ale, any blow which the bottle received after it left the hands of the bottler, leaving what the court considered to be the sole remaining inference, namely, that the bottle must have been defective. The court then expressly eliminated the doctrine of res ipsa loquitur because the bottle was not under the exclusive management and control of the defendant, and sustained the finding upon the assumption that the breaking of the bottle of ale was proof that the bottle was not of merchantable quality when it left the hands of the manufacturer. The effect of the decision is to make the bottler of carbonated beverages an insurer against the bursting of glass containers. We think this is not good law.
The judgment is reversed.
Schauer, P. J., and Wood (Parker), J., concurred.