This is a petition for workmen’s compensation in which a decree was entered by the superior court awarding compensation to the petitioner for total incapacity and containing, among others, the following specific findings of fact: “Second: That the petitioner sustained a personal injury arising out of an accident, arising out of and in the course of her employment, when she received a blow upon her right breast, resulting in a cancer of the right breast,” and “Fourth: That the petitioner is totally incapacitated from work as a result of the personal injury, resulting in a cancer of her right breast, since the 26th day of March A. D. 1945 and that such total incapacity still exists.”

From that decree respondent has appealed to this court, but in its reasons of appeal it complains only of the above findings and not of the award of compensation, because *96it concedes that there is evidence in the record which supports another finding in the decree that petitioner is incapacitated by an injury to her back which is independent of the condition resulting from the cancer. The issues which respondent has raised by its appeal have been concisely summarized in its brief in this manner: “that the' Trial Justice erred in weighing the evidence concerning the cause of the cancer; that particular reference to the cause of the cancer or the fact of cancer does not belong in the decree because these are evidentiary facts leading to the ultimate fact of injury; and that these references in findings numbered two and four are prejudicial to the respondent.”

Respondent has briefed and argued those issues in that order under three points. Under its first point it argues that the trial justice in making the second and fourth findings erred in that he misconceived the nature of the question before him on the issue concerning causal relationship between the blow which petitioner received by the-accident and the condition of cancer which was found later. It is urged that whether such a relation exists is a medical question and must be determined by the court on competent medical evidence. Prescinding from that principle, respondent contends that the trial justice disregarded the medical evidence of the lack of a causal relation between trauma and cancer and adopted a theory under which he liberally interpreted the evidence in favor of the petitioner and found that the blow on her right breast which she had received by the accident was the cause of her subsequent cancerous condition.

Respondent especially points out certain statements made by the trial justice in his decision and contends that they illustrate the alleged errors of law which he committed. The following statement, it argues, shows his refusal to accept the medical evidence on causation of the cancerous condition: “Hence, the question of causal.relationship between an injury and a subsequent condition ought not to be *97determined or controlled by nice philosophical reasoning, nor by the reluctance and hesitancy of the scientist to accept as satisfactory á conclusion not demonstrated to be necessarily a fact to a scientific certainty.” This observation, respondent argues, shows that in reaching his conclusion that the blow on petitioner’s breast had caused the cancer the trial' justice disregarded the medical evidence before him on that question. Earlier in his decision the trial justice quite frankly stated: “The testimony before us indicates that there is no unanimity of opinion on the question. We are not-called upon to decide a medical question.” And further along he said: “In other cases involving trauma to the person followed by a subsequently appearing physical injury or condition without any further intervening explanatory occurrence, courts and juries have no difficulty in finding the one is the cause of the other.”

Looking into the trial justice’s mind as mirrored by the above expressions respondent urges that in all this'judicial reasoning three legal errors are discernible: “a misconception of the nature of the question, the failure to accept the scientific approach to the problem, and the acceptance of conjecture for proof in respect to causal relationship.” Moreover respondent contends that the trial justice further erred by assuming that the rule that the workmen’s compensation act should be construed liberally in order to effectuate its purpose could be applied to the exercise of his judicial function in weighing the evidence.

These contentions merited and they have received our careful consideration. The correction of errors of law is the sole jurisdiction of this court in the review of workmen’s compensation cases. Precluded as we are in such cases from reviewing the evidence and passing upon the credibility of the witnesses, we have always been concerned that the superior court should keep within the rules of law in exercising its exclusive final appellate jurisdiction in the finding of facts therein. Cranston Print Works v. Pascatore, 72 R. I. 471; Walsh-Kaiser Co. v. Kooharian, 72 R. I. 390; *98Antonelli v. Walsh-Kaiser Co., 72 R. I. 1; Foy v. Juilliard & Co., 63 R. 1.233; Lupoli v. Atlantic Tubing Co., 43 R. I. 299; Jillson v. Ross, 38 R. I. 145. And where we have been in doubt whether that has been done we have remanded the case to that court for clarification. Campbell v. Walsh-Kaiser Co., 72 R. I. 358. Respondent in the case at bar relies upon the latter case and argues that even though we should not agree with the contention made here that the trial justice has been guided by an erroneous rule of law in weighing the evidence we should nevertheless remand the case to the superior court for a new trial because there is greater room for doubt as to what rule that court actually followed in the instant case than there was in the Campbell case.

After carefully reading the transcript and paying particular attention to the content and character of all the medical testimony on the present state of medical knowledge pertaining to the causation of cancer of the breast by a single trauma, we examined the trial justice’s rescript wherein he discussed that evidence. Without isolating certain of his observations on that subject from their context but reading all that he said together as parts of a connected whole, we are of the opinion that his decision does not disclose that he misconceived the nature of the question before him or that he decided it on conjecture.

Nor can it be said, in our opinion, that he failed to accept the scientific approach to the question if by that expression respondent means that he did not give due consideration to all the medical evidence. Of course it is obvious that he did .not accept the opinions of three of the medical witnesses who testified that a single trauma cannot cause cancer. And it is equally obvious that he based his decision on the opinions of two other medical witnesses, who testified to the contrary, and also on certain other uncontradicted evidence that petitioner was in good health before the accident; that she never experienced any pain in the region of her right shoulder until after the accident; that *99a large can of orange juice which she was carrying under her right arm when she fell down the stairs “banged” her breast; and that about seven weeks later a lump was discovered there'which upon removal by surgery was found to be malignant.

We are of the opinion that the trial justice weighed all the evidence and found, notwithstanding the majority medical testimony, that it preponderated in favor of the petitioner’s contention that the blow on her breast caused the cancer. Someone else weighing the same evidence might have considerable difficulty in reaching that conclusion in view of such majority medical opinion. But, as we have so often said, that is of no consequence on the review of workmen’s compensation cases in this court, because in such cases we do not weigh the evidence.

When the trial justice said that he was not called upon to decide a medical question we understand him to mean that he was not required to decide as a matter of medical science whether a cancer of the breast could result from a single blow. His duty on the contrary was one that lay within the broader province of juridical science whose norms of fact finding from other proved facts are less exacting. The law is not and in the nature of things can never be an exact science. The judicial process by means of which the law is administered is an instrument for the attainment of substantial justice and not necessarily perfect justice. We think that the trial justice was saying no more than this when he said that the question of causal relationship was not to be controlled by “nice philosophical reasoning” or by the rules of “scientific certainty.”

We are, however, somewhat concerned with his reference to the rule of liberal construction which is applicable in cases arising under the workmen’s compensation act. He says in his rescript that “it is more in- keeping with the Act and the purpose intended to be effectuated by it to resolve the question of causal relationship by a broad and liberal view of that relationship.” What he meant by that *100statement is perhaps not as clear as it might be. Respondent argues that it means a liberal view of the evidence may be taken and that the trial justice, therefore, did not consider himself restricted to the usual rule of judicial proof in civil cases that the party having the affirmative of an issue must prove it by a fair preponderance of the evidence. We do not think his statement submits to that construction. It is certainly nothing like the statement which caused the difficulty in Campbell v. Walsh-Kaiser Co., supra. There the trial justice said: “We are bound to resolve any doubts in the whole testimony in the case in favor of the employee.” That was the enunciation of a rule for weighing the evidence which was directly at variance with the well-established rule.

We are of the opinion that the trial justice in the instant case was discussing the legal principles of proof of causation that were applicable in judicial tribunals and contrasting them with the stricter rules “of the scientist who requires scientific proof to a certainty.” It was in the course of such discussion that he made the statement which we quoted in the next preceding paragraph and about which the respondent complains. On the whole, after carefully studying the complete discussion of the trial justice we are of the opinion that he did not adopt a novel and incorrect rule for weighing the evidence on the question of causation. Rather we think his rescript, taken in its entirety, shows that he weighed the evidence with proper regard for the usual and correct rule of law. Therefore, no doubt exists here which would require remanding the case to the superior court for clarification as was necessary in Campbell v. Walsh-Kaiser Co., supra.

The final question is whether the second and fourth findings making particular reference to “cancer of her right breast” should be stricken from the decree. Respondent contends that only ultimate facts properly belong in the decree and that the trial justice’s “findings that the petitioner had a cancer which was caused by the trauma is *101evidentiary rather than ultimate in character.” This court has held that the provision of the statute prescribing the incorporation of findings of fact in the decree did not intend “that said decree shall include a statement of the evidence or the findings of probative facts from which conclusions are to be drawn as to the issuable facts.” Jillson v. Ross, supra, at page 148. And it has also held that a decree containing no findings of fact is erroneous. Dodge v. Barstow Stove Co., 40 R. I. 191. But no case has been cited by respondent and we are aware of none in which this court has held a decree erroneous in part because it contained a finding of fact which was not ultimate in character. It is true that in General Scrap Iron, Inc., v. LaPorte, 68 R. I. 98, 100, upon which respondent relies, we said: “It is not necessary that every finding of the trial justice on all the facts raised by the evidence be contained in the decree.” But this is scarcely to be considered as authority for the proposition that the decree is necessarily erroneous if such facts are incorporated therein. The requirement insisted upon by this court is solely that all the ultimate facts must be contained in the decree.

However, respondent raises the issue that it is prejudiced by the incorporation of the two findings of fact in question because the entry of a decree containing such findings will render the issue of causation of petitioner’s cancer of the breast res judicata. We agree that it will, but in our opinion respondent suffers no prejudice by reason of. that fact. It has had its full day in court for the trial of that issue and it actually presented to the superior court its evidence tending to disprove such causation. In the circumstances it cannot now claim prejudice by reason of such findings appearing in the final decree. Moreover, we are of the opinion that in view of the manner in which this case was tried in the superior court the findings in question are not mere evidentiary matters but are in reality ultimate facts.

The respondent’s appeal is denied and dismissed, the *102decree appealed from is affirmed, and the cause is remanded to the superior court for further proceedings.