Appeal from a decision and award of the Workmen’s Compensation Board. Decedent was employed as an electrician. In *772the course of employment on February 8,1952, he struck his head on a steam pipe while walking in a cellar. His injury was then diagnosed as a contusion of the right temporoparietal area and an injury to the right eardrum. He was treated medically but did not lose time from work. X rays were negative for fracture. On the night of February 29 he visited a friend; they went to a tavern and decedent came home at 4:00 a.m. March 1 with a stranger and went to the cellar of his house. There were beer and whisky glasses and beer bottles found on a table there. Claimant, the decedent’s wife, saw him at 6:30 a.m. in the kitchen, apparently attempting to adjust the gas refrigerator. At 9:30 a.m. claimant found decedent lying on a bed in a room adjoining the kitchen from which the door was open. There was an open valve on the gas pipe which itself appeared to have been sawed off. Decedent was dead. Laboratory tests of the body following autopsy showed carbon monoxide in the blood in “very large amounts ” to the extent of “ 65% saturation ” and disclosed presence of 2-plus alcohol in the brain. Examination of the skull on autopsy revealed a fracture of the skull and brain contusions. There is adequate medical opinion in the record that the fracture of the skull could reasonably be attributed to the accident. The autopsy report showed that the primary cause of death was asphyxiation due to illuminating gas; and contributing causes, fracture of the skull -and contusion of the brain. A review of the medical opinion in this record, however, makes it clear that there is no proof that the fracture or brain injury had any direct physical association with the death due to gas asphyxiation. One of the physicians called by claimant theorized that it might be conceivable that decedent could withstand more illuminating gas if he had not had the fracture of the skull. He added: “ but this is conjecture ”. As to the direct effect of the fracture he said: “ I would consider that it would not be likely that he would die at that particular time had he not had the misfortune to be exposed to the gas.” The theory on which decision rests, in the findings of the board, is that as a result of the injury to the brain he suffered headaches and pain which “ caused [him] to change his drinking habits and change his regulatory mechanisms which controlled the adjustment to circulation and respiration to toxic factors, so dulling his perceptive senses to permit absorption of a lethal dose of carbon monoxide ”. We read this to mean that the fracture led decedent to drink to the extent he did not realize that gas was escaping in large quantities. We think the medical testimony in support of this attenuated theory rests on the merest speculation and does not provide substantial evidence in support of the award. An example of the speculation of the medical witnesses at a vital point in developing this theory is the testimony of Dr. Arthur Shapiro who testified: “ We don’t have enough psychological history to make the exact diagnosis but at any rate he took to drink, came home apparently drunk as judged from the smell of the brain when removed at autopsy, undertook the repair of his refrigerator and died as the result of a serious error in judgment involved in this action, that is, he left the gas turned on and was asphyxiated.” Award reversed and the claim dismissed, with costs to appellants against the Workmen’s Compensation Board. Foster, P. J., Bergan, Coon, ITalpem and Gibson, JJ., concur.