¶1concurring and dissenting:
¶2I respectfully differ with the majority in that I would find the evidence insufficient to prove beyond a reasonable doubt that appellant caused Mr. Markiw’s death, and would therefore reverse appellant’s conviction for second-degree murder. I would affirm the sentence for conspiracy and remand for resentencing on the robbery conviction, since the sentence for robbery merged with the life sentence for second-degree murder.
¶3*31The test for sufficiency of the evidence is whether, viewing the evidence in the light most favorable to the Commonwealth and drawing all proper inferences favorable to the Commonwealth, the trier of fact could reasonably have determined all elements of the crime to have been established beyond a reasonable doubt. Commonwealth v. Keb-litis, 500 Pa. 321, 456 A.2d 149 (1983).
¶4In a criminal homicide case, an essential element the Commonwealth must prove beyond a reasonable doubt is that the defendant’s criminal acts caused the victim’s death. Commonwealth v. Hicks, 466 Pa. 499, 353 A.2d 803 (1976); Commonwealth v. Stoltzfus, 462 Pa. 43, 337 A.2d 873 (1975). It is not sufficient that death probably resulted from a criminal agency. Commonwealth v. Radford, 428 Pa. 279, 236 A.2d 802 (1968).
¶5Furthermore, the Commonwealth must prove a direct causal relationship between the defendant’s acts and the victim’s death; tort concepts of “proximate cause” have no place in a prosecution for criminal homicide. Commonwealth v. Root, 403 Pa. 571, 170 A.2d 310 (1961).
¶6I believe the evidence did not establish beyond a reasonable doubt a direct causal relationship between appellant’s admittedly reprehensible acts and Mr. Markiw’s unfortunate death. Witnesses testified that shortly after the assault Mr. Markiw appeared “ashen faced,” and was perspiring and holding his chest, where he had been hit by one of the robbers. However, he was still fully conscious and ambulatory when, owing to a policeman’s solicitude, he was driven to Temple University Hospital. A conspicuous hiatus in the Commonwealth’s evidence follows, for there is nothing in the record concerning Mr. Markiw’s state from the time he entered the hospital until he died five days later. No doctor testified that Markiw exhibited symptoms of a heart attack when he arrived at the hospital; no medical witness described Markiw’s course of treatment or the process of his physical deterioration. The Commonwealth attempted to fill this gap in the evidence with the testimony of the medical examiner who performed an autopsy on the *32body. This witness, Dr. Carpenter, stated that Mr. Markiw had suffered from severe heart disease, and was at the time of the attack “an already compromised human being.” He further testified that the cause of death was a heart attack occurring four to eight days before death. In response to a hypothetical question involving the facts of the robbery, the witness stated his opinion to a reasonable degree of medical certainty that the incident had induced stress so severe as to result in Markiw’s death.
¶7This case is virtually indistinguishable from Commonwealth v. Embry, 441 Pa. 183, 272 A.2d 178 (1971), where a seventy-one-year-old woman with a long history of coronary artery disease and prior heart attacks became the victim of a purse-snatching and died of a heart attack. The expert witness who had performed the autopsy opined to a reasonable degree of medical certainty that the victim’s struggle with the robbers produced stress which caused the heart attack, but testified he was not convinced of this beyond a reasonable doubt. The Supreme Court held there was insufficient evidence of causation: “when only one witness is presented by the Commonwealth to establish causation and that witness cannot do so beyond a reasonable doubt, a necessary element of the proof of that crime is missing.” Id.,441 Pa. at 186, 272 A.2d at 179. Accord, Radford, supra.
¶8It is true, as the majority points out, that an expert medical witness need not state his opinion about the cause of death in terms of “beyond a reasonable doubt.” Commonwealth v. Webb, 449 Pa. 490, 296 A.2d 734 (1972). An expert witness need only entertain a reasonable degree of medical certainty for his conclusions. Commonwealth v. Floyd, 499 Pa. 316, 453 A.2d 326 (1982); cf. Commonwealth v. Richardson, 307 Pa.Super. 191, 452 A.2d 1379 (1982) (citing Embry for this principle). However, it remains a legal question for the court whether the Commonwealth’s evidence is sufficient to warrant a finding of causal connection, Webb supra, and it remains the Commonwealth’s burden to prove this causal connection beyond a reasonable *33doubt. Commonwealth v. Williams, 476 Pa. 557, 383 A.2d 503 (1978).
¶9Here the only thing about Mr. Markiw’s death established beyond a reasonable doubt was heart attack as the cause. It was shown, and undisputed, that Markiw had suffered from severe heart disease. The autopsy revealed scarring of the heart, presumably indicating that the heart attack leading to Markiw’s death had not been his first. However, the record is absolutely silent as to Markiw’s condition or treatment either immediately before the assault or after his admission to the hospital. Therefore, Dr. Carpenter’s professional opinion that Markiw’s fatal heart attack occurred between four and eight days prior to death is of little use in establishing the assault as the cause of the heart attack. Taking this opinion into account, the evidence would support the inference that Markiw’s heart attack arose from natural causes before or after the robbery just as easily as it would support the inference that it arose contemporaneously with the robbery and because of it. There is, indeed, nothing in the record but conjecture to rule out the inference that this “already compromised human being” suffered his heart attack prior to and independently of the mugging. At most, the evidence establishes that the assault “probably” caused the heart attack; to conclude that the assault in fact was the cause, the fact finder would have to engage in speculation. Nevertheless, on this evidence Dr. Carpenter was allowed to superimpose his own opinion that the assault had caused the seizure. Whereas Dr. Carpenter’s opinion that death was due to a heart attack was based directly on his own observations, his opinion as to the cause of the heart attack was based on a hypothetical version of the facts of the robbery supplied by the district attorney. No matter that Dr. Carpenter expressed this latter opinion in terms of a “reasonable medical certainty”; in my view this was nothing more than the doctor substituting his own speculation and conjecture for that of the fact finder.
*34In a homicide case, where the life or liberty of a citizen is at stake, and where the guilt of the accused must be established beyond a reasonable doubt, the causal connection between the death of the decedent and the unlawful acts of the respondent cannot be supported on mere conjecture and speculation.
¶10Fine v. State, 193 Tenn. 422, 428, 246 S.W.2d 70, 72 (1952) (quoting State v. Rounds, 104 Vt. 442, 457, 160 A. 249, 254 (1932)).
Where it is equally possible and probable that the death resulted from one cause as from another cause and the defendant is not responsible for one of the causes, then any determination of the cause of death can only be speculative and. conjectural and the evidence will be held insufficient to support a guilty verdict.
¶11Rucker v. State, 174 Tenn. 569, 573, 129 S.W.2d 208, 210 (1939).
¶12My learned colleague Judge Stephen J. McEwen, Jr., cogently makes the point that it is no excuse to criminal homicide that the victim’s life was fragile even before the defendant’s criminal assault extinguished it. I don’t dispute this; it is a settled principle that an assailant must take his victim as he finds him. As stated in Commonwealth v. Hicks, 483 Pa. 305, 313, 396 A.2d 1183, 1187 (1979),
An accused may not escape criminal liability on the ground that, prior to the criminal act, his victim was not in perfect health, or the blow he inflicted was not mortal, or the immediate cause of death. If his blow started the chain of causation which led to the death, he is guilty of homicide. Commonwealth v. Hicks, 466 Pa. 499, 505, 353 A.2d 803, 805 (1976).
¶13However, the fact that the victim had poor health does not dispense with the need to establish beyond a reasonable doubt a “chain of causation” leading from the defendant’s crime to the victim’s death. “In a death as distinguished from a ‘killing,’ the prosecution must prove an absence of natural causes. That is, the first step is to prove the victim *35died because he was killed.” Welch v. State, 45 Ala.App. 657, 235 So.2d 906, 907 (1970).
¶14A comparison of various cases in which heart attack deaths were prosecuted as homicides will serve to illustrate what is and what is not an adequate “chain of causation” to establish that the victim was killed. In Commonwealth v. Hicks, 483 Pa. 305, 396 A.2d 1183 (1979), the victim collapsed and fell unconscious immediately after wrestling strenuously with Hicks, who in a craze had broken into the victim’s house and accosted him. The victim never regained consciousness, and died of “acute cardiac failure due to a pre-existing arteriosclerotic heart disease which was aggravated by the severe stress of his struggle with [Hicks].” 483 Pa. at 309, 396 A.2d at 1185 (emphasis added). The Supreme Court rejected a challenge to the proof of causation, saying it was “clear” the defendant’s attack had cause the victim’s cardiac failure. Id.,483 Pa. at 313, 396 A.2d at 1186. See also State v. McKeiver, 89 N.J.Super. 52, 213 A.2d 320 (1965) (evidence sufficient where victim dropped dead in bar during armed holdup).
¶15With the immediate and clear chain of causation in Hicks, compare that in Stevens v. United States, 249 A.2d 514 (D.C.App.1969). There the decedent was afflicted with “developing” arteriosclerosis. He was injured in an automobile accident caused by the defendant, and hospitalized. Five weeks later he died in the hospital of a heart attack. There was expert medical testimony that the debility caused by the accident could have precipitated the heart attack from which the decedent died. The court found the proof of causation insufficient, noting that the Government had produced no testimony from an attending physician or expert who had knowledge of the decedent’s condition while in the hospital. The defendant’s conviction for negligent homicide was reversed.
¶16Certainly the present case bears a closer resemblance to Stevensthan to Hicks or McKeiver, in that Mr. Markiw did not die immediately, but spent a period of time in the hospital before succumbing. As in Stevens, the decedent’s *36condition during the period of hospitalization went totally unexplained on the record. The victim in Stevenslingered five weeks before dying, whereas in this case the period was five days. However, the length of time between incident and death is only one factor to consider in deciding the overarching question: whether the prosecution has sufficiently established a chain of causation.
¶17In at least one case, our Supreme Court has found the evidence insufficient to prove causation where death was immediate. Commonwealth v. Embry, supra. See also Hubbard v. Commonwealth, 304 Ky. 818, 202 S.W.2d 634 (1947). In Hubbard, the defendant was arrested for drunkenness, but resisted the decedent’s efforts to lead him to jail. The would-be jailer, Dyche, struggled with Hubbard until Dyche fell to the ground. As others subdued the defendant, Dyche, who suffered from a heart condition, sat on the ground with his hand on his chest. Rather than continue to sit and rest, however, Dyche got down on the ground and “rolled and tumbled” until he died about half an hour later. Despite opinion testimony from three physicians that the scuffle and its attendant stress precipitated Dyche’s death, the Court of Appeals of Kentucky held the evidence insufficient to support Hubbard’s homicide conviction:
In the present case the misdemeanor of the defendant must be regarded as too remote — not in time, to be sure, but as the cause. The failure of the man’s diseased heart was the cause. Livingston v. Commonwealth, 14 Grat., Va., 592.... It is, at least, speculative to say that the act of the defendant in this case was sufficiently proximate to impose criminal responsibility upon him for the unfortunate death. We are of opinion, therefore, that the court should have directed an acquittal.
¶18304 Ky. at 822, 202 S.W.2d at 637.
¶19Regardless of the time span involved, the principle is clear that the Commonwealth must establish a direct and *37unbroken chain of causation between the defendant’s criminal acts and the victim’s death. See, e.g., Commonwealth v. Roberson, 485 Pa. 586, 403 A.2d 544 (1979) (beating to head leading to fatal pneumonia); Commonwealth v. Green, 477 Pa. 170, 383 A.2d 877 (1978) (gunshot wound required removal of kidney; victim died in hospital of vomit sucked into lungs); Commonwealth v. Stafford, 451 Pa. 95, 301 A.2d 600 (1973) (victim beaten in head died ten days later after brain surgery); Commonwealth v. Odom, 448 Pa. 474, 295 A.2d 331 (1972) (victim died of head injuries sixty-seven days after beating); Commonwealth v. Jefferson, 445 Pa. 1, 281 A.2d 852 (1971) (stabbing led to blood clots causing death); Commonwealth v. Cartagena, 272 Pa.Super. 485, 416 A.2d 560 (1979) (stabbing leading to paralysis and death).
¶20Here there is an important link missing from the chain: proof that appellant’s acts started the process leading to death. Dr. Carpenter’s opinion in response to a hypothetical question that the robbery caused the heart attack may establish a possible, even a probable, causal relationship between the two. However, this opinion evidence was not sufficient to establish a causal relationship in this particular case beyond a reasonable doubt. Given the paucity of evidence upon which the opinion was based, I would hold it insufficient regardless of the degree of certainty with which the expert expressed it. Cf. Commonwealth v. Rawls, 270 Pa.Super. 432, 411 A.2d 796 (1979) (defendant punched victim in head; victim died of subdural hemorrhage; evidence sufficient despite medical examiner’s inability to say with certainty that fist punch in fact caused hemorrhage). Although it is of course possible that appellant Cotton caused Markiw’s death by heart attack, I believe the evidence was too tenuous and speculative to establish this causal connection, or to visit upon Cotton the consequences of murder.