MARKELL, J., delivered the opinion of the Court.

This is an appeal from a judgment on the verdict of a jury in a workmen's compensation appeal, which reversed the decision of the State Industrial Accident Commission that the employee died on July 29, 1950 from a cause not connected with the accidental personal injury. The employer and insurer appeal from the judgment and from refusal of a motion for a directed verdict and a motion for judgment n.o.v.

On June 19, 1950 the employee was employed as a carpenter at work on construction of an apartment house in Salisbury. He slipped on a rock and hurt his back. He was forty-seven years old. On June 19th and 20th Dr. Philip A. Insley, of Salisbury, treated him at his office, on June 20th sent him to the Peninsula General Hospital, where he remained until June 24th, and saw him again at his office on June 26th and sent him back to work. He returned to the same work for a day and then went to Seaford, Delaware, to work for the duPont Company. He had trouble with his back and on July 25th saw Dr. Briele, of Salisbury, and was sent to the Peninsula General Hospital. The same day, Dr. William H. Fisher, an associate of Dr. Briele, saw him in the hospital. Dr. Fisher testified that "examination of him at that time revealed a typical picture for ruptured intervertebral disc * * *. We tried to keep him comfortable during the rest of that day and night with sedation. He required several doses of morphine for pain, the pain was so severe. We felt there was no doubt about what he had and he should be seen by a neuro-surgeon rather promptly to relieve him of the disc, because we didn't feel he would get any lasting relief other than continuously using morphine until he had operative relief of the ruptured disc. So Dr. Briele contacted Dr. Arnold [a neuro-surgeon] in Baltimore the morning of the 26th. * * * Dr. Arnold told him to send him up, which we did." At the argument counsel for appellants said that the evidence that the ruptured disc was the result of the accident on June 19th was rather thin, but conceded that it was sufficient to go to the jury. We shall, therefore, assume that, as the jury found, the accident caused the ruptured disc.

Dr. Fisher made a general physical examination of him to determine his general physical condition. There was nothing to indicate excessive use of alcohol. If he had been an excessive user of alcohol nothing in the examination would have revealed it.

On the morning of July 25th the pain suddenly became much worse. The next morning he was sent by ambulance from Peninsula General Hospital, at Salisbury, to University Hospital, at Baltimore. On July 28th at about 12:30 p.m., he was operated on by Dr. Arnold under a spinal anaesthetic. Dr. Arnold testified, "After a routine examination * * * — the usual routine for pre-operative evaluation, we felt that the operation should be done as soon as possible because of the severe pain and disability the patient had. * * * His general physical status was considered satisfactory for operation and, in fact, had it not been entirely satisfactory, we would have still gone ahead with the operation because of the severe pain the patient was having. The point is he was admitted as an emergency because of the severe pain, and the operation was facilitated as quickly as possible for that reason. * * * He was given the usual pre-operative medication for a spinal, and before the operation was started, the one thing that I clearly remember, while we were draping the patient and preparing the side for operation, the patient was more talkative and I called attention to this fact. We occasionally see that with patients who have had pre-operative medication, but that was unusual. He was much more talkative, and this continued through the operation. He talked in a confused fashion, but we thought nothing more about it for the moment, because it does sometimes happen from the medication. Well, the operation was done quickly and without event. The whole procedure took a little less than an hour. Nothing out of the ordinary was encountered, except the pathology was clear cut. He had a ruptured disc with severe nerve root compression. Nothing else was noticed during the procedure, except this continued talkativeness and mental confusion. I made rounds in the middle — the mid-afternoon, sometime between 3:30 and 4:30, I presume, and his general condition seemed quite satisfactory except that again he would go off into these temporary periods of mental confusion, seeing objects in the room. He was a little active. He didn't know where he was, and then he lapsed back into a reasonably clear mental state. His pulse, blood pressure and general physical condition at the moment was quite satisfactory, except for this mental state, and I discussed this with my associates, Dr. Crosby and Dr. Nichols, and wondered then if he might not be developing acute toxic hallucinosis. * * * After making rounds, I had to be out of the city the night and day following, so that was my last contact with the patient. He was under the direct supervision of Dr. Crosby [resident in neurological surgery] and Dr. Nichols [assistant resident] both experienced physicians."

Dr. Arnold then summarized what happened after he left, as later told him by Dr. Crosby and Dr. Nichols and testified to by them. "The mental state continued to be confused and got worse through the night. He was quite restless, agitated, and began to see animals and small objects going through the room, and he was completely disoriented as to time and place. The usual treatment for this type of thing is to quiet the patient with sedatives such as paraldehyde and this was done. He was seen throughout the night at frequent intervals by one or both of these physicians, and his general condition, except for his mental state, was considered satisfactory until the early morning [about 4:30 a.m.], at which time he went into shock suddenly. His blood pressure dropped, there was a sweating, signs of shock, and he died very quickly [about 5:10 a.m.]. It was the opinion of those who saw him that he had a heart attack and died probably of a coronary occlusion. Now, in discussing the mental state of the patient and the cause for it * * * Dr. Crosby and Dr. Nichols discussed with some members of the family, or brought up the question as to whether Mr. Powell consumed unusual quantities of alcohol, because we were looking for a cause for the acute toxic hallucinosis, and with this particular type of mental blank, alcohol is the only one I know that will do it, and I am told some members of the family stated he was a chronic user of alcohol. In going over the entire record and evaluating the case from beginning to end, I can see no possible relationship between the operation and the man's death. In over eight hundred disc operations which I have done, there have been two deaths. Mr. Powell is one. The other one was a case of embolism which occurred on the fourth day. The patient was out of bed, making splendid progress and died with an embolism, but he was perfectly clear and rational all through. I have never seen this acute toxic mental state in relation to surgery except with Mr. Powell. So, in eight hundred cases, if there were a parallel along the line, we would certainly see it in eight hundred cases, if the disc surgery would produce it. Furthermore, it is not produced by embolism or thrombosis, or things which may be secondarily related to operative procedure of this sort, and I further believe, had we not operated on him, Mr. Powell would still have had the acute toxic state which is known as delirium termens, whether we operated or not. Of course, that is only an opinion which I can't prove, but I think the stress and strain of his confused mental state, with his agitation, in a man who has — rather, who was hypertensive, moderately arterio-sclerotic and gave a history of exertional dyspnea, it would seem reasonable. The confused mental state and the agitation and physical effort which he exerted following operation, in a man who had a deficient vascular system, it is reasonable to assume his death was that of coronary occlusion." An autopsy was requested, but the family would not permit it.

On cross-examination Dr. Arnold testified, "Q. You don't know who made the statements in regard to his use of alcohol? A. Well, the names I do not have, and I think you can clarify that issue more directly with the persons involved. I think we would get a clearer picture of it. As long as we seemingly, from a medical point of view, had found the reason which we accepted as being the cause of this acute mental state, the three of us felt without question that was what it was, and when we had corroborating evidence from some members of the family that he did use alcohol to excess, there seemed to be no reasonable doubt but that that was the answer, but the details I did not go into, as to who said what." Asked what are some of the causes of such hallucinations other than alcoholism, Dr. Arnold testified, "Coming on precipitately this way, I don't know of any other, frankly: * * * Now bromides is more of a gradual process. I have never seen it come on in such fashion. * * * I have never seen an acute toxic hallucinosis of this sort from morphine. * * * [After examining hospital records] He received pantopon, grains of six, scopolamine, and a grain and a half of nembutal at 11:30. That was the premedication before operation. Q. What, if any, of those drugs could cause toxic hallucination? A. None of them in the dose administered. * * * Q. In other words, you believe there is a difference in the delirium caused by one toxic from that caused by another toxic? A. Not entirely. They are very similar in aspects, but the general clinical cause is different, one from the other, the mode of onset. Q. Is it possible to determine from the type of hallucinations the man is suffering from which particular toxic was the cause of it? A. Well, one cannot, of course, say with absolute certainty because we have no way of measuring the cause. We have no way of doing laboratory tests that would be absolutely conclusive, but experience is hard to beat. Q. Well, isn't it true that the man could have suffered these hallucinations as a result of a particular sensitivity, we'll say, to morphine or bella donna or sodium pentathol or any one of those toxics that sometimes cause hallucinations? A. In the doses which he received, no." Dr. Arnold talked to him, examined him, before the operation. "Q. How long prior to the operation? A. I don't recall, either the day of admission or the following day. I have no note as to the exact time I saw Mr. Powell." He was given nembutal, one and a half grains, morphine derivative, grains of six, pantopon. "Q. And those are drugs that can also cause toxic hallucinations? A. No, sir, not in the dosage, no. * * * Q. He had morphine in Salisbury before he came up there? A. I don't know, but I wouldn't be surprised if he had. * * * A. Let me inform you we had no history of alcoholism until after he developed the hallucinosis, and then I instructed my associates to inquire about the use of alcohol, because that was what it looked like, because up to the time he developed his mental state, we had no knowledge of alcohol, and from the clarity of his mind, we had no reason to suspect it at all — but I can think of nothing else except alcoholic psychosis that would have produced this particular picture, going over the whole gamut, the type of onset and clinical course. * * * to take the whole running story of the development, how it came on and the progress through the night, the best explanation I know is chronic alcoholism." Apparently no one questioned Mr. Powell as to his drinking habits.

At the end of a long cross-examination Dr. Arnold said, "I don't want to have the court feel that one can, with absolute certainty, say from this hallucinosis this is alcohol, but certainly one has to go on the clinical picture; otherwise we would never make a diagnosis of anything, and when you see the same thing happening time and time again with the same pattern, the same agent — just as you look at a friend and say you know him, but you don't know his name. That is your picture of him, and that is my picture of delirium tremens, and if it is not due to alcohol, I don't know what it is due to, and you can read the medical books — and read your head off — and if you can duplicate this picture, I'll be obliged to you, and if this man was not a chronic alcoholic, I don't know what it is. I'm not trying to develop that he was an alcoholic. I will simply have to say he had hallucinosis, cause undetermined. Q. Based on the information you received about the man being an alcoholic, you made the diagnosis that the hallucinations were due to alcoholism, is that correct? A. That is my opinion, yes. [On re-direct examination] Q. You had also made that same diagnosis prior to receiving those statements? A. I wouldn't say diagnosis. I would say suspicion. It looked like that to me. I told the house staff to inquire from the family and see if you can support it, and the evidence seemingly was forthcoming and we dropped it then and there." Previously he had said, "But I wouldn't stress the point that one can be absolutely certain from looking at a patient with delirium that you could say this is alcohol or bromide. If we had the history in this man, for instance, that he had taken large quantities of bromide and got a blood bromide and found it highly elevated, you can say this is highly unusual for bromides, but here it is. But, from clinical experience, we say this looks like delirium tremens, and we go to the family and say, `Has he used alcohol to excess?' and they say yes, and that is the answer, as far as we are concerned, you see."

The commissioner asked Dr. Arnold, "Q. Now, doctor, would the trauma of this operation have any influence or effect upon the onset of the delirium tremens? A. That is purely a hypothetical thing. I don't know, I should not think so." Later, the commissioner questioned Dr. Arnold at length as to the effect, if any, of the trauma of an operation as a precipitating factor in delirium tremens. Dr. Arnold in substance answered, none, because "we don't consider the disc operation to be traumatic in the least of ways", a reason not easy to understand without medical knowledge or verbal or other explanation.

The "history" of alcoholism on which Dr. Arnold and Drs. Crosby and Nichols based their opinions was obtained by Dr. Crosby from a sister of Mr. Powell. The sister, however, testified and flatly denied that she had made the statements testified to by Dr. Crosby or that such statements were true and said she had told Dr. Crosby that her brother was not alcoholic. The sister and the widow testified that Mr. Powell was not and never had been alcoholic and never drank except "socially" and that rarely and temperately. Except the testimony of the widow and the sister that he drank "socially", there is no testimony that anyone ever saw him take a drink or saw any evidence of drink. His landlady, from whom he rented a room, testified that he told her he had drunk years ago but had stopped. Six acquaintances testified that they never saw any evidence of drink. The sister testified that, after the operation, "To me he [Mr. Powell] appeared to be very heavily drugged or under some narcotic." Dr. Crosby testified that when Dr. Arnold saw Mr. Powell about 4:30 he suggested that Drs. Crosby and Nichols should watch this man carefully and inquire for a history of alcoholism, and that evening he talked with the sister, who "admitted alcoholism on the part of the patient".

On cross-examination, Dr. Crosby, when asked how he differentiated between the type of hallucination brought on by alcohol and that brought on by some other toxic, said, "One of the best ways, I think, to differentiate between the two is to get a history of the use of toxics. Q. Not from observing the type of hallucinations? A. I didn't say that. One makes a clinical impression from the observations, but to make a pretty good diagnosis, presumptive diagnosis, one has to get a history of the use of a toxic agent, and I got a history of it. Now, one must believe the history one is given and use it, and that is all I did. I got a history of alcoholism and used it." He testified, "I don't qualify as an expert on that differential" [between hallucinations brought on by one type of toxic and those brought on by alcoholism].

Claimant offered only one medical witness, Dr. Harry M. Mattax, who has been in practice, general practice, about two years. He had read the other medical testimony and read from "an authoritative text" on scopolamine and atropine, "The behavior and mental symptoms may suggest an acute psychosis. Memory is disturbed, orientation faulty, hallucinations (especially visual) common, sensorium clouded, and mania or delirium not infrequent. The diagnosis of acute schizophrenic episodes or alcoholic delirium has been mistakenly made, and some patients have been committed to psychopathic institutions. * * * The syndrome described persists for many hours and may completely disappear only after several days have passed. Depression and circulatory collapse result only in cases of severe intoxication. The blood pressure declines, respirations become inadequate and death due to respiratory failure occurs after a period of paralysis and coma." He testified that in overdose alcohol is a toxic. "In overdosage any drug [including scopolamine] can be a toxic." He has observed between twenty and twenty-five cases of delirium tremens in his medical experience and about ten or twelve of other toxic hallucination. "Q. Now, based on your medical knowledge and your experience in the observation of these cases that you have already mentioned can you determine from observing the type of hallucinations from which a man is suffering as to whether or not the hallucinations are caused by alcohol, or scopolamine, or some other toxic? A. No, I don't think so." Mere negative opinion testimony from a physician of two years practice might be entitled to little or no weight against a positive opinion of a surgeon of Dr. Arnold's experience. On the other hand, the fact that Dr. Arnold has performed eight hundred disc operations indicates that he is highly expert in such operations, but he says he has "never seen this acute toxic state in relation to surgery except with Mr. Powell", and there is nothing to indicate that he is specially expert on delirium tremens — or hallucinosis. Apparently most physicians — including Dr. Arnold, Dr. Crosby, and Dr. Nichols — get their experience with delirium tremens principally during their interneship or other early hospital training. Moreover, Dr. Mattax is certainly qualified to point out, for what it may be worth, his opinion as to any lacunae in Dr. Arnold's reasoning.

Appellants contend ( a) that where the subject is one for experts or skilled witnesses alone and concerns a matter of science or specialized art or other matters of which a layman can have no knowledge, expert opinion evidence in which there is no conflict is conclusive, and ( b) that as all the testimony is that the operation did not cause death, it is not necessary for appellants to show that delirium tremens caused death or what did cause it. We need not approve or disapprove the first contention, as an abstract proposition. As applied by appellants in the instant case, on the facts and the evidence, neither contention can be maintained.

Whether or when evidence is "conclusive" depends upon the nature of the facts and the evidence in each case. The court may disregard testimony which is obviously impossible, without any testimony, expert or non-expert, to the contrary. "If a witness who can see testifies that he looked and did not see an object, which if he had looked he must have seen — such testimony is unworthy of consideration." Baltimore Traction Co. v. Helms, 84 Md. 515, 526, 36 A. 119, 121, 36 L.R.A. 215, and many subsequent cases. In some cases we have held that, without expert testimony, the evidence was legally insufficient to establish the relation of cause and effect between an accident and an injury. Bethlehem Steel Co. v. Ziegenfuss, 187 Md. 283, 294, 49 A.2d 793, a workmen's compensation case; Superior Transfer Co. v. Halstead, 189 Md. 536, 539, 56 A.2d 706, a tort case. In the cases cited, however, we said that it is not necessary that there should be medical testimony as to the cause of the injury in order to take the case to the jury, 187 Md. at page 294, 49 A.2d 793, and that in workmen's compensation cases, under some circumstances, non-medical proof that a disease or an injured condition followed an accident is sufficient to take to the jury, the question whether it resulted from the accident, 189 Md. at page 539, 56 A.2d 706, and we cited cases to this effect in which all the medical testimony was that there was no such causal relation. Uncontradicted opinion evidence is not necessarily conclusive upon a jury. The Conqueror, 166 U.S. 110, 131, 17 S.Ct. 510, 41 L.Ed. 937; Sartor v. Arkansas Natural Gas Corporation, 321 U.S. 620, 627-628, 64 S.Ct. 724, 88 L.Ed. 967. In paternity cases the Supreme Court of California has sustained jury verdicts which blood tests indicated to be impossible. Arais v. Kalensnikoff, 10 Cal.2d 428, 74 P.2d 1043, 115 A.L.R. 163; Berry v. Chaplin, 74 Cal.App.2d 652, 169 P.2d 442 (review refused by Supreme Court of California). In Shanks v. State, 185 Md. 437, 442, 45 A.2d 85, 163 A.L.R. 931, Arais v. Kalensnikoff was cited as holding that expert evidence was not conclusive. We do not think that citation or the citation of both these blood test cases in the instant case should be taken as indicating that we would or would not follow them on the precise question presented in those cases.

Unless, in the circumstances, facts shown by uncontradicted evidence amount to "uncontraverted or undisputed facts", a jury may disbelieve "uncontradicted evidence". Dunstan v. Bethlehem Steel Co., 187 Md. 571, 578, 51 A.2d 288, 291. In the instant case the facts are not undisputed or the evidence uncontradicted. Essentially all appellants' medical testimony is contradicted in that it rests upon a "history", said to have been obtained from the sister, which the sister flatly contradicts. If the jury believed the sister, as they apparently did, the basis for all appellants' medical testimony is destroyed. Baltimore Ohio R. Co. v. Brooks, 158 Md. 149, 148 A. 276.

Nor can any distinction be made in this respect between medical testimony that delirium tremens caused death and that the operation did not. "The law requires proof of probable, not merely possible facts, including causal relations. Reasoning post hoc, propter hoc is a recognized logical fallacy, a non sequitur. But sequence of events, plus proof of possible causal relation, may amount to proof of probable causal relation, in the absence of evidence of any other equally probable cause. Baltimore City Pass. Ry. Co. v. Kemp, 61 Md. 74; State of Maryland, to Use of Goralski v. General Stevedoring Co., D.C., 213 F. 51, 61, 62, affirmed Joseph R. Foard Co. v. State of Maryland, 4 Cir., 219 F. 827, 830-832." Charlton Brothers Co. v. Garrettson, 188 Md. 85, 94, 51 A.2d 642, 646. The undisputed fact is that, within seventeen hours after the operation, death followed without other apparent cause than the operation (including all the incidents that precede, accompany or follow an operation). To prove that death could not have resulted from the operation, it was necessary to show at least that it could have resulted from some other cause. The jury could not find that death did not occur and could not be expected to find that it resulted from a miracle. Appellants' undertook to prove that death could and did result from delirium tremens. If the jury believed the sister's testimony, death could not and did not result from delirium tremens.

The nature of the case, and of the physicians' reasoning, makes it impossible to show conclusively either that death resulted from delirium tremens, or that it could not result from the operation. Neither could be deduced from mathematical or other general principles or from facts undisputed or established by laboratory tests. Like the jury, the physicians could only draw inferences from empirical facts, including the disputed "history", and meagre scientific facts. Dr. Arnold make his diagnosis of delirium tremens after obtaining the "history"; the same opinion before receipt of the "history" he says was not a "diagnosis", but only a "suspicion". If he had had a history of bromides — and a blood test for bromides — he could have said "this is highly unusual for bromides, but here it is". Dr. Arnold says "experience is hard to beat", but he had "never seen this acute toxic mental state in relation to surgery" before this case. Dr. Crosby says one of the best ways to differentiate between hallucination from alcohol and from some other toxic "is to get a history of the use of toxics", and adds that he does not "qualify as an expert on that differentiation". Dr. Arnold tried to differentiate as to the onset of the hallucinations but later said he would not "stress the point that one can be absolutely certain from looking at a patient with delirium that you could say this is alcohol or bromide". At the time of the operation Dr. Arnold evidently thought the patient's state was due to medication. Dr. Mattax quoted "an authoritative text" to the effect that hallucinosis from scopolamine has been mistakenly diagnosed as "alcoholic delirium", and expressed his own opinion that he could not from observing delirium determine whether a man is suffering from the one or the other.

Dr. Arnold's testimony is also based on other facts not undisputed and not conclusively proved. Three times, when asked whether drugs given Mr. Powell could have caused his hallucinations, Dr. Arnold said, not in the doses given. The sister says, inexpertly, "To me he appeared to be very heavily drugged or under some narcotic". Dr. Fisher's testimony indicates that at Salisbury he had been somewhat "heavily drugged". It does not appear that Dr. Arnold knew what quantities of drugs he had received at Salisbury. As to the drugs given at Baltimore, Dr. Arnold had only the hospital records. To show impossibility of drug hallucinosis in an unprecedented case, the chance of error in a hospital record — or a hospital dose — cannot be excluded as itself impossible.

Even if death did result from delirium tremens, the evidence is legally sufficient to show that it resulted from the accident. In McCahill v. New York Transportation Co., 201 N.Y. 221, 94 N.E. 616, 48 L.R.A., N.S., 131, a tort case in which the deceased was thrown from a taxicab and died in the hospital of delirium tremens, recovery was allowed on the ground that delirium tremens was caused ( i.e., precipitated) by the accident. A hospital physician testified, "I should say with reasonable certainty the injury precipitated his attack of delirium tremens, and understand I mean precipitated, not induced". 201 N.Y. 223. In the instant case Dr. Arnold testified, "In general, delirium tremens develops in a patient who is a chronic user of alcohol, or one of prolonged sprees, and then the alcohol is withdrawn, and that precipitates the symptoms, and that is why it comes on so suddenly. It is from a clear mental state to one of confusion, and the dosage has to do with the personality of the individual under discussion. Q. Now, if he were a user of alcohol to such an extent that the withdrawal of it would bring on delirium tremens, and the alcohol was withheld from him during the period he was in the hospital in Salisbury, would not he have suffered delirium then? A. Not necessarily. One cannot predict the onset of symptoms." In other words, the attack would not necessarily be precipitated while he was in Salisbury, but might have occurred — and did — after he was removed to Baltimore. In several states in workmen's compensation cases awards for death resulting from accident, by precipitating an attack of delirium tremens, have been sustained. Robbins v. Frohlich, 278 App. Div. 729, 103 N.Y.S.2d 189; Sullivan v. Industrial Engineering Co., 173 App. Div. 65, 158 N YS. 970; Carroll v. Knickerbocker Ice Co., 169 App. Div. 450, 155 N.Y.S. 1; Ramlow v. Moon Lake Ice Co., 192 Mich. 505, 158 N.W. 1027, L.R.A. 1916 F, 955; Valeri v. Village of Hibbing, 169 Minn. 241, 211 N.W. 8, 60 A.L.R. 1296; Judge Coleman so held under the Longshoremen's and Harbor Workers' Compensation Act. Lumbermen's Mut. Casualty Co. v. Branhan, D.C. 48 F. Supp. 141.

In Bramble v. Shields, 146 Md. 494, 506, 127 A. 44, 48, after reviewing tort cases involving questions of proximate cause, the court said, "But in cases arising under the Workmen's Compensation Law, the question of negligence is excluded, and with it the rule as to reasonable and probable consequence would also be excluded, if such rule were applicable here. The only test is, Did the accidental injury arise out of and in the course of employment? If it did, it makes no difference whether it was a normal or abnormal occurrence; and so with `any such disease or infection as may naturally result therefrom'. It can make no difference whether the results are usual or unusual, if there is a direct causal connection between the injury and the disease, so that the disease is directly attributable to the injury." In Baber v. Knipp Sons, 164 Md. 55, 67, 163 A. 862, 866, it was said, "It is apparent that medical testimony is not required in those cases where by other evidence facts are shown which fairly and logically tend to prove that the accident was the efficient cause of the condition complained of. It would seem to be established in this state, in workmen's compensation cases, that `proximate cause' means that the result could have been caused by the accident, and that there has not intervened, between the accident and the result, any other efficient cause." The sentence last quoted is a broader — or looser — definition of "proximate cause" than is usual, at least in other than workmen's compensation cases, unless it be taken as limited by or to the preceding sentence. In Baltimore Towage etc. Co. v. Shenton, 175 Md. 30, 199 A. 806, an award was sustained for the death of a truck driver who, while standing, in the course of his employment, on a pier near his truck, wavered (from fainting or other unknown cause) and fell into the water, and was drowned. This court, through Chief Judge Bond, quoted, and after reviewing conflicting authorities in different jurisdictions, followed, a statement by Collins, M.R. in Wicks v. Dowell Co., Ltd. [1905] 2 K.B. 225, 229 (in which Wicks, while unloading coal from a ship and standing by the open hatchway through which the coal was being raised, was seized with an epileptic fit, and fell into the hold, to his injury), "When we get rid of the confusion caused by the fit and the confusion involved in not dissociating the injury and its actual physical cause from the more remote cause, that is to say, from the fit, the difficulty arising from the words `out of the employment' is removed. How does it come about in the present case that the accident arose out of the employment? Because by the conditions of his employment the workman was bound to stand on the edge of what I may style a precipice, and if in that position he was seized with a fit he would almost necessarily fall over. If that is so, the accident was caused by his necessary proximity to the precipice, for the fall was brought about by the necessity of his standing in that position." 175 Md. 33-34, 199 A. 807. See also Dunham v. Clare, 2 K.B. 292, 295-296.

The workmen's compensation cases in which, in words or in effect, "proximate cause" was given a broad definition seem to indicate that when, without medical testimony, there is shown a direct sequence from an accident to the alleged consequences, e.g., accident, operation, hallucinosis (from drugs or alcohol) and death, all apparently set in motion by the accident and culminating in death or disability, causal relation may be inferred from such sequence in the absence of affirmative evidence of an independent intervening cause. In the instant case it may be that, without any medical evidence at all, the fact that death followed the operation within seventeen hours would have been sufficient to justify an inference of causal relation between the accident and death. We need not express an opinion on this point, as we find that with all the medical testimony in the case, the court would not have been justified in directing a verdict for appellants.

Appellants stress the statutory presumption in favor of the commission's decision. "The statutory burden of proof that the commission's decision was incorrect is a burden of persuasion, which may be sustained by satisfying the jury from the same evidence on which the commission made its decision. No additional evidence is necessary. Moore v. Clarke, 171 Md. 39, 45-46, 187 A. 887, 107 A.L.R. 924; Williams Construction Co. v. Bohlen, 189 Md. 576, 580, 56 A.2d 694." Brooks v. Bethlehem Steel Co., 199 Md. 29, 34, 85 A.2d 471, 473. Whatever incongruity there may be in review by a jury of the presumptively correct decision of an administrative body supposed to be "informed by experience", only the legislature can correct. Apparently only in Maryland, Ohio, Oregon, and Oklahoma are such administrative decisions reviewable by a jury. Vanderbilt, Minimum Standards of Judicial Administration, p. 618. In occupational disease cases there is no such review in Maryland. In the instant case, for the reasons we have stated, we find ample evidence to support the verdict of the jury reversing the decision of the commission.

The court charged the jury that if the hallucinations were not the result of drugs administered at the time of the operation and could have been caused only from delirium tremens from an alcoholic background, then there is no causal connection between the injury and the death. From what we have said, this direction was too favorable to appellants. This is an error of which appellants cannot complain.

Judgment affirmed, with costs.