(Opinion op the Court by

Commissioner Stanley

Reversing.

The appellee, Woodruff Development Company, engaged the appellant', Kentucky Glycerine Company, to shoot an oil well with 30 quarts of nitroglycerine for the agreed price of $128. Appellee filed this suit charging that as a result of the negligence of appellant’s agent in performing the work, and in advising its agent that the material had exploded when it had not, the company’s tools had been lost in the well, and it had expended $1,223 in extracting them, for which sum it asked judgment.

The appellant denied negligence, averred that the work had been properly done, and entered a counterclaim for its contract price. A verdict was returned for the development company, and this appeal follows.

*326 The well was an old one, having been shot twice or perhaps three times before, but had been drilled deeper to a depth of 1,122 feet. Pursuant to instructions of the Development Company, the charge of - nitroglycerine was anchored 6 feet above the bottom of the well. The plan used was to explode this torpedo by dropping on it a jack squib consisting of one stick of dynamite with cap and fuse compacted with dirt in a soft metallic shell 3% inches in diameter and 4 feet long. When this squib was dropped, there was an explosion in the well, which was heard some distance away and jarred the surface of the earth at a point at least 60 or 70 feet from the well where appellee’s manager, Haynes, was standing, and who testified to this fact. The shooter and other witnesses testifying for the appellant were positive that the whole charge had been exploded according to the intent, and appellee’s agent, a man of considerable experience, was of the same opinion. He signed a receipt for the material and acknowledgment of the completion of the work. Soon thereafter operations were begun to drill through a bridge of sand which had formed within, and to clean out the well. After about an hour’s drilling, through 5 to 10 feet of sand, there was a terrific explosion below, the effect of which is thus described by appellee’s manager: “When the shot went off the temper screw nearly blew off the beam and went back and came back into the engine and just missed me. I had hold of the clamp at the time, and the stem blowed out of the hole and the burnt gas fumes was so black you couldn’t see through them and I turned to the engine and started to pull the line out and I had to leave, there was so much gas and fumes there. ’ ’

The tools in the well were blown up into the well, and the end of the line was crushed. Afterward something like 20 or 25 feet of oil sand was drilled out from the top of the tools, which were found about 130 feet from the bottom of the well, with the line of cable turned back alongside the tools. After considerable work the tools were recovered, and it was found that the bit had been unscrewed from the stem, but not broken.

It is the theory of the appellee that the dyiamite squib lodged in one of the old shot holes and discharged, producing the first explosion, which threw a bridge across the well above the torpedo, and when the drill went through this bridge it struck the torpedo and discharged *327 it, thereby producing the second explosion and the damage. It is claimed that a tail should have been tied to the squib, which would have kept it straight and directed it to the bottom of the well on top of the torpedo. And also claimed that, had a cord or soft rope been laid across the mouth of the well, it would have been puffed up a little and then sucked into the well, if the squib alone exploded, but blown away if the torpedo exploded; thus demonstrating what had actually occurred below.

It is the theory of the appellant that the squib and torpedo exploded simultaneously, and that the bridge of sand formed and confined the gases, which it is shown follow the shooting of a well, and for which condition the shooter is in no way responsible. It is developed in the evidence that such risks are not assumed by the shooter, and in this case the ticket or acknowledgment signed by the appellee’s manager bears the notation: “All torpedoes put in well at risk of well owners. ’ ’

While the plaintiff introduced evidence to show that the precautions suggested were often taken, the preponderance of the evidence is that reasonable care was exercised, and that the tying of a tail to the squib and the placing of the cord across the mouth of the well are seldom used and have no effect when there is considerable fluid in the well; and it is shown that this well at the time had perhaps 90 or 100 feet of oil and water in it. The other negligence claimed is that of the shooter advising appellee’s manager on the ground that the torpedo had exploded, on which assurance he proceeded to resume drilling. It is established that experienced drillers and shooters can from the sounds easily distinguish the explosion of the torpedo from that of the squib alone; and it is pretty well demonstrated that the explosion of one stick of dynamite that far below the surface of the earth does not affect the top of the ground as it was jarred at this time. All of these men were of the opinion at the time that the torpedo had exploded, and appellee’s agents appear to have possessed as much knowledge in this regard as did appellant’s employee. It is further shown that there was no gas emitted when the "well was shot, and that the generation or release of such substance usually follows that operation. This lends credence to appellee’s theory that the bridge of sand then formed confined those gases until released by the drill. Moreover, appellee’s manager testified his drill had gone to within 6 feet of the bottom of the well as it was before *328 it was shot, which obviously it could not have done had the torpedo been unexploded.

In applying the law to the facts, we have the elementary principle that negligence is never presumed, and that it must be established by positive proof or proof of facts or circumstances from which it can be reasonably inferred. The evidence produced by the plaintiff was not of that character, for as is frankly admitted by its manager, “A person never knows what goes on under the ground,” and that one cannot state definitely what he cannot see. Prom the circumstances proven there is no more reason to infer that appellee’s theory as to what occurred is the correct one than there is to infer that appellant’s conjectures are well founded. If there is any difference, it is more favorable to the appellant’s surmises. Under such conditions no recovery can be had, and it is the duty of the trial court to direct a verdict for the defendant.

In Gregory’s Adm’x v. Director General, 195 Ky. 289, 242 S. W. 573, 376—to cite one of many cases—this rule is discussed and this pertinent extract from a former opinion is given:

“Neither courts nor juries are authorized to indulge in speculation or guesswork as to the cause of accidents; there must be some tangible evidence from which it may be fairly said what brought about the accident. It has long been the rule in this state that no recovery can be had in such cases where the evidence is so unsatisfactory as to require surmise or speculation as to how the injury occurred, and that there will be no presumption of negligence.”

The same rule generally applies in determining whether or not the damage is the proximate result of the negligence when it has been established. Idem. Sutton’s Adm’r v. L. & N. R. R. Co., 168 Ky. 81, 181 S. W. 938. It may not be reasoned in this case from the results that negligence was the proximate cause; in other words, from effect to cause. The rule of res ipsa loquitur—which is but the expression of an exception to the general rule that negligence must be affirmatively proved, and arises from the doctrine of probabilities—coupled with the evidence of the audible and visible effects of the shooting of the well, cannot be held sufficient to authorize a verdict for the plaintiff. That rule was also sought to be invoked in Carter Oil Co. v. Independent Torpedo Co., 107 Okl. *329 209, 232 P. 419, which is quite similar to this case. The destruction of an oil well was charged to have been due to negligence in shooting it when a torpedo of nitroglycerine exploded while being lowered in the well. The doctrine in connection with those particular circumstances is considered at length, and its application demonstrated not to be proper, it being pointed out that where the instrumentalities producing an accident are not exclusively under the control of the one charged with negligence, the res ipsa rule is inapplicable. See, also, Thompson on Negligence, sec. 7635; Louisville & N. R. Co. v. Mink, 168 Ky. 394, 182 S. W. 188; Stephens v. Kitchen Lumber Co., 222 Ky. 736, 2 S. W. (2d) 374.

Archer’s Oil & Gas, c. 37, treats of the subject of shooting wells, and upon the authority of Davidson v. Humes, 188 Pa. 335, 41 A. 649; Zahniser et al., trading as East End Oil Co. et al. Pa. Torpedo Co., 190 Pa. 350, 42 A. 707, declares:

“Where the owner of an oil well employs another to shoot the well, and does not require a guaranty that no damages shall result therefrom, if the well is injured by such shooting, in order to sustain an action for damages it devolves upon the owner to prove negligence in the shooting.”

There was no plea or evidence of any guaranty in this case. On the contrary, it appears to have been part of the contract that the shooting of the well was to be done at the owner’s risk. In all three analogous cases cited—and we are aware of no others directly in point— it was held that no recovery on the part of the owner of the well under similar circumstances was warranted under the law. And this court is of the opinion that the jury should have been instructed to find for the defendant. It.becomes unnecessary, therefore, to consider other points raised in brief.

The judgment is reversed for consistent proceedings.