CUTTER, J.

This is a claim by one Messersmith's widow for workmen's compensation benefits for herself and a minor child. The single member's findings and decision were adopted by the reviewing board. In the Superior Court, a final decree was entered (a) stating that the "deceased . . . did not receive an injury arising out of and in the course of his employment and that his death was not causally related to any injury . . . arising out of or in the" course of his employment, and (b) dismissing the widow's claim. The widow has appealed.

Messersmith was in South Carolina on a trip for his employer, New England Tape Company (the company). He collapsed and died while playing golf on February 28, 1954. The death certificate gave as the cause of death "probable coronary thrombosis." Messersmith worked for the company as an engineer and was "key salesman" with respect to the exploitation of a new process being developed by the company in cooperation with Owens-Corning Fiberglas Company (Fiberglas) in South Carolina. He travelled a great deal on business at all times and it became "imperative that he go . . . to . . . South Carolina to assist" in the project, in which "the stakes were high."

Messersmith had been confined to a hospital for about a month in 1951 with heart disease and had an underlying arteriosclerotic condition. On February 23, 1954, at the direction of one Brown, majority stockholder of the company and the deceased's immediate superior, the deceased went by train to South Carolina and met Brown. They drove together on February 24 two hundred fifty miles to the Fiberglas laboratory where they spent the next two days, Thursday and Friday. On Saturday they were asked to play golf with the work manager and an engineer, which they did. They were invited to play again on Sunday (February 28) but did not play in the morning because of rain. In the afternoon Brown and Messersmith did play golf with a third person. While playing Messersmith had a heart attack and died. The single member found (a) that the game on February 28 "was a means . . . of . . . cementing good relationship between the two companies"; (b) that "this . . . game was an incident of" the employment; (c) that the employee had an arteriosclerotic heart condition prior to February 28; and (d) that the decedent "suffered an injury arising out of and in the course of his employment on February 28, 1954; that the physical activity and nervous tension of the work . . . resulted in the acute [heart] failure . . . and caused . . . death on the same day."

In addition to the foregoing facts found by him, the single member improperly recited certain testimony, without clearly making findings about it as he should have done (see Craddock's Case, 310 Mass. 116, 125; Judkins's Case, 315 Mass. 226, 227), viz: (a) that all that Brown and the deceased talked about in South Carolina was their work project; and (b) that the widow, several days prior to Messersmith's departure for South Carolina, observed that "he showed signs of fatigue, was gray in color and his lips were bluish." The single member also referred to expert testimony of one Dr. Rattigan "that this man's death from coronary thrombosis on February 28, 1954, was causally related to a variety of factors relating to his employment by reason probably of acceleration of a preexisting disease process to a point that death occurred on that day."

1. The single member appears to have rested his conclusion that the fatal golf game was "an incident of" Messersmith's employment upon a subsidiary finding that the game was played with one of the Fiberglas employees. This subsidiary finding is not warranted by the evidence. The record indicates that plans to play golf with Fiberglas employees on Sunday morning were abandoned because of rain, that these Fiberglas employees had made other arrangements for the afternoon, and that Brown and the deceased merely met the Fiberglas people at the club and then "picked up a third individual and started off." No evidence gives basis for the inference that the third individual was connected with Fiberglas. We need not determine whether other considerations mentioned in Moore's Case, 330 Mass. 1, 4-5, existed which would permit a finding that the game was "an incident of" the employment, for no other basis was mentioned by the single member.

2. The single member's conclusion "that the physical activity and nervous tension of the work in which . . . [Messersmith] was engaged resulted in the acute failure of his heart" rests on no adequate subsidiary findings by the single member. The single member's relevant findings (as opposed to mere recitals of testimony), summarized above, by themselves do not warrant any inferences that Messersmith was under "nervous tension" or that his work contributed to his death. It is particularly significant that there were no findings about the widow's observations of Messersmith's fatigue and color prior to his departure for South Carolina which were an essential part of the basis for Dr. Rattigan's answers to the hypothetical questions, later mentioned. In Judkins's Case, 315 Mass. 226, 227, it was pointed out that there should be "such specific and definite findings upon the evidence reported as will enable this court to determine with reasonable certainty whether correct rules of law have been applied."

3. Both (a) the circumstance that the board's decision may rest upon an unwarranted subsidiary finding, and (b) the inadequacy of the findings, would lead us to remand the case to the Industrial Accident Board for clarification. We, however, also examine the record to see whether such remanding is unnecessary because "the evidence, including all rational inferences," was insufficient in law to sustain the widow's burden of establishing that the deceased's death arose out of and in the course of his employment. Judkins's Case, 315 Mass. 226, 228-231. Roney's Case, 316 Mass. 732, 739-740.

4. The following facts, in addition to those already stated, could have been found on the evidence. Messersmith, in addition to considerable travelling, worked on occasion sixteen hours a day, despite his record of heart disease. Upon his return on February 19, 1954, to Worcester from a trip to New York, his wife "knew he had had a heart attack; she had seen him have many of them; he was gray; his lips were purple in color; he was drawn and tired. He didn't want anything to eat." Between February 19 and February 23 when he left for South Carolina "he was very tired and stayed in his room a great deal; he worked on papers . . . [and] was very withdrawn." His wife "asked him not to take the trip . . . but he said . . . that it was very important to his company and when the trip was over he would be able to rest." On February 23 "he was pale with a gray look; he was very slow and didn't have his usual sparkle."

After the 1951 attack, which was not his first, Messersmith had talked with his doctor and they decided that, "although he might increase his life expectancy by not returning" to work with the company, he should do so but "alter his schedule; . . . rest from each day's work before he undertook the next day's work, and . . . from one trip before he took the next trip."

Dr. Rattigan was asked a hypothetical question which stated the earlier testimony very fully. His opinion that Messersmith's death "was causally related to a variety of factors relating to his employment," he explained in part as follows, "This man's activities that particular day [February 28] did hasten his death." From his history of myocardial infarction "three years prior to this . . . it must be assumed that he had a severe degree of circulatory coronary artery disease" which still permitted him to continue his usual "activities which . . . were not especially conducive to rest and relaxation." Nine days "before his death . . . he was observed to have signs which to [the] witness clearly indicated . . . progressive depletion of his coronary reserve." Yet "under pressure . . . of his job and in a very fatigued state he did undertake to go on a trip which involved considerable mental tension and while engaged in a game of golf, again in witness's opinion not entirely divorced from tension, had a heart attack and died." In Dr. Rattigan's opinion "many of the factors related to his work, clearly spell . . . tension which we would advise patients with coronary artery disease to avoid." Nine days before his death, "he was in such a state that any doctor" knowing his medical history "would have recommended complete bed rest, if not complete hospitalization." The "activities of that last day were just enough to accelerate the already depleted process within the coronary system so as to bring about death before it normally would have occurred."

On cross-examination, Dr. Rattigan stated that his "theory on causal relationship is not entirely contingent on excess tension. . . . [I]t contributes very substantially but not to the exclusion of other factors." Contributing to Messersmith's death was the "fact that [for] nine days . . . prior to the last day, he had travelled and worked rather intensively . . . in a condition in which . . . every moment that he spent outside of complete bed rest or hospital care, placed him in jeopardy of exactly what happened on February 28."

The witness conceded that as to the period after February 23 there was no testimony of the grayness which had indicated "grave circulatory trouble on the 19th and again on the 23rd." He testified, however, (1) that once you have the information assumed in the hypothetical question as to the "man's condition as observed by his wife February 19 I could tell you that this man was headed straight for another type of coronary catastrophe that would prove fatal as it actually did," and that Messersmith "was in a precarious condition which could have gone either way, depending . . . on what he did after February 19 . . . I think that had the man been hospitalized, or had he rested, he would not have died nine days later." This opinion Dr. Rattigan described as "better than a guess," for although "[h]e could not give 100% assurance that he would not have died then . . . the probability is that he would not have died."

Other doctors with excellent qualifications took a different position. The single member could reasonably have concluded that in various respects Dr. Rattigan was neither convincing nor clear and that he inadequately explained important assumed facts at conflict with his theory. Nevertheless, we cannot say, as a matter of law, that the single member was bound so to conclude. The extent to which cross-examination weakened the weight of the testimony was for the trier of the facts, as was the circumstance that Dr. Rattigan at times in his testimony placed substantial but not exclusive weight on the events of the final Sunday (which on this record were not incident to the work) as a cause of death.

This is not like the expert testimony, admittedly based upon an "assumption of excitement and emotional factors," which in Brown v. United States Fid. Guar. Co. 336 Mass. 609, 613-614, was revealed by cross-examination to be "gratuitous." Here Messersmith, after the indications of fatigue and grayness shown on February 19 and 23, undertook, wholly apart from the final Sunday golf game, substantial work activities which could reasonably be regarded as seriously hazardous for him. We cannot say, as a matter of law, that Dr. Rattigan had no reasonable basis (a) for his opinion that Messersmith's condition on those days indicated "rather grave circulatory difficulty as of the moment the observation was made" or (b) for concluding (see Cormier's Case, 337 Mass. 714, 717) that Messersmith's substantial work activities thereafter would impose excessive tension on one with Messersmith's heart condition.

5. Where a preexisting heart disease of an employee is accelerated by the employment, there may be found to be an injury arising out of and in the course of employment. See McMurray's Case, 331 Mass. 29, 32-33, and cases cited; Baruffaldi v. Contributory Retirement Appeal Bd. 337 Mass. 495, 501. In Brzozowski's Case, 328 Mass. 113, 115-116, there is recognition of the difficulty of drawing the line between "gradual degeneration of . . . [an employee's] cardiac organs . . . [and death] accelerated by a strain or exertion attributable to his work." See also the somewhat comparable situation in Murphy's Case, 328 Mass. 301, 303-305, where a work induced heart condition, not the cause of death, so weakened an employee as to make his death from cancer occur sooner than it otherwise would have. We cannot say as a matter of law, whatever we might have done if charged with finding the facts, that upon the evidence it was not permissible to find that the work exertions after the signs of danger on February 19 contributed substantially to death more rapidly than if rest or hospitalization precautions had then been taken. Accordingly, this is not an instance where (cf. Judkins's Case, 315 Mass. 226, 228, 230; Roney's Case, 316 Mass. 732, 739) we can refrain from remanding the case to the reviewing board for proper subsidiary findings and for reexamination of its conclusions. See Ratigan's Case, 338 Mass. 712, 718-719.

6. Various exceptions to the admission of evidence have been argued with extreme brevity and without citation of authority. It is sufficient to say that each of them has been examined and no prejudicial error is disclosed.

7. The decree below is reversed. The case is to be remanded to the board for clarification of its findings and decision and for such other proceedings consistent with this opinion as the board may deem appropriate.

So ordered.