This case comes up on the plaintiff’s exceptions to the rulings and charge of the presiding Justice. It is an action on the case for injuries to the plaintiff’s person and property resulting from a collision between the plaintiff’s motor boat, 'Carolyn, an open launch 23 feet in length and 6 feet beam, and without any pilot house, and the defendant’s steamer, City of Rockland, a vessel of 274 feet in length, 38-J feet beam,, and of the burthen of 1696 gross tons:
The collision occurred at about 7.15 in the evening of September 21, 1909, on the Kennebec River abreast the southerly end of the City of Bath. At the time of the collision the night was dark, but good for seeing lights, and there was a moderate breeze from the southward.
The plaintiff and a friend were in his launch bound up the Kennebec River, having left the New Meadows River the same afternoon. The defendant’s steamer, the City of Rockland, was on one of her regular passages from Gardiner to Boston, and was proceeding down near the center of the river on her regular course.
*397It is conceded1 that the United States statutes, and the rules and regulations of navigation are to be interpreted in accordance with the decisions of the federal courts. Upon this rule of interpretation the discussion both of the plaintiff and defendant proceeds.
The plaintiff presents and argues several exceptions to the rulings and charge of the presiding Justice. During the progress of the trial the defendant introduced certain sailing rules and regulations established by the federal government for the purpose of showing that the plaintiff’s boat was not equipped and managed in accord^ anee with the requirements of law, and contended that a failure to comply with these provisions of law was, per se, an act of contributory negligence. It is unnecessary to insert a statement of all the rules which the defendant contended were violated', as that part of the charge of the presiding Justice to which exceptions was taken is based upon the assumption that one or all of these rules may or may not have been violated, and submitted to the jury the question whether such violation, if found, contributed to the accident. In submitting the contention of the defendant upon the question of contributory negligence in violating the rules and regulations, and defining the burden resting upon the plaintiff in rebutting it, the presiding Justice charged the jury as follows: “Now it is incumbent upon the plaintiff to prove by a preponderance of the testimony not only that the lack of these statutory requirements, if there was a lack, did not contribute towards the injury, but he must go a step further, and1 prove that it could not have contributed1 to- the injury. Not merely that you find that it did not, but that you must go further and find that it could not have contributed to the injury. That burden is upon the plaintiff. I think I make that clear.”
To that part of the charge above quoted stating “but he must go a step further, and prove that it could not have contributed to the injury. Not merely that you find that it did not, but that you must go further and find that i*t could not have contributed to the injury. That burden is upon the plaintiff. I think I make that clear,” the plaintiff objects and contends that it is a matter of exceptionable error.
While it is contended that the language of the presiding Justice in giving the above rule is in substantial accord with the language *398of the United States Courts in stating the rule, it nevertheless required the jury to find that the failure to comply with the rules and regulations not only did not contribute to the injury, and could not have done so in the case at bar, but was incapable of so contributing in any c^se that might arise.
An analysis of the phraseology claimed to be erroneous will, we think, clearly show that it was either meaningless or required a degree of proof that would, establish more than an absolute negative, a thing impossible, ist. The instruction imposed upon the plaintiff the positive duty of proving that his failure to observe the regulations did not contribute to his injury. 2nd. It imposed upon him the additional positive duty of showing it could have done so. The question was whether the plaintiff’s failure to comply with the federal regulations contributed to his injury. If it dlid, that was the end of his case. If it did not, then this phase of the defense disappeared. .The count said to the jury it was incumbent upon the plaintiff to show as a positive duty, that his failure did not contribute. That was the absolute end of the burden which could be imposed upon him. It was impossible for him ito do more. To establish that a thing absolutely does not exist is the end of negative proof. But the instruction did not stop here. It required that the plaintiff must prove not only an absolute negative, but in some way go further, and, as the instruction says, prove that his failure not only did not, but, as an additional positive duty, could not have contributed to the injury. At this juncture arises the perplexity to be solved by the jury. When they had taken this case under consideration and had concluded, as they might, that the failure of the plaintiff to observe these regulations had not in any degree contributed to the collision, it is obvious that, to their minds,.it was then incumbent upon them to do something more. But what more ? How could they go a step further? If they had already decided it did not contribute, and hence, as a corollary, could not contribute, then to their minds what more must the “could not” instruction have required? It seems obvious that lay minds might have concluded that this language required them to go beyond the case on trial and apply it to any similar case that might arise. This is the only interpretation that could give any meaning, by the context, to the “could not” phraseology. This language was at 'least capable of confusing *399the jury with reference to the excess of duty which the instruction imposed upon the plaintiff.
We think the language of the instruction fails to express the idea intended. It is too apparent for discussion that if the plaintiff had shown that his failure to observe the navigation rules did not contribute to his injury, that he was entitled to recover upon this aspect of the case; he might, however, be unable to prove by direct testimony that his failure did not contribute. But this would not' end his rights. He could still show by the facts, circumstances and probabilities that his failure could not have contributed and, hence, did not.
The misleading feature of the instruction is found in the phrase “must go a step further.” It required that “could not” must be proven in addition to “did not.” We think the instruction should have been, that it was incumbent upon the plaintiff to establish: (i) that his failure to comply with the rules and regulations did not cause nor in any degree contribute to the collision; (2) that he might do this either by direct testimony or by any other facts and circumstances tending to prove that such failure could not have ■contributed; (3) and that, if he had shown that such failure could not have contributed, he had then produced sufficient evidence to rebut the presumption that it did contribute.
As before remarked, it was contended that the language of the instruction under consideration was in phraseology and effect substantially that employed by the federal courts in defining, what we may term, for the sake of brevity, the “did not” and “could not” rules. But we are unable to so construe it. We think a fair interpretation to be given the language employed by tbe federal courts in the various cases involving a consideration of these rules is in harmony with the foregoing conclusion as to what the instruction in the case at bar should have been.
The “could not” rule was first promulgated1 ini the case of the Pennsylvania, 19 Wall, 125. The facts in this case arose out of a collision in a very dense fog between a sailing bark and a large steamer. The bark was under way at a slow rate of speed, ringing a bell as a fog signal. The steamer was going about 7 knots an hour. In an action of the bark against the steamer it was held that the bark was guilty of contributory negligence in violating an Act *400of Congress providing that “sailing vessels when under way should use a fog horn and when not under way a bell,” inasmuch as it was unable to rebut the presumption that its failure to blow a fog horn was a contributory cause of the collision.”
That part of the opinion laying down the rule is as follows: “But when, as in this case, a ship at the time of a collision is in actual violation of a statutory rule intended to prevent collisions, it is no more than a reasonable presumption that the fault, if not the sole cause, was at least contributory cause of the disaster. In such a case the burden rests upon the ship of showing not merely that her fault might not have -been one of the causes, or that it probably was not but that it could not have been. Such a rule is necessary to enforce obedience to the mandate of the statute. In the case of the Femham, the Lords of the Privy Council said, Tt is of the greatest possible importance, having regard to the admiralty regulations, and to the necessity of enforcing obedience to them, to lay down this rule; that if it is proved that any vessel has not shown lights, the .burden lies upon her to show that her non-compliance with the regulations was not the cause of the collision/ In some cases it is possible to show this with entire certainty. In others it cannot be. The evidence in the present case leaves it uncertain whether if a fog horn had been blown on the bark, it would not have been heard sooner than the bell was heard, and thus earlier warning have been given to the steamer—seasonable warning to-have enabled her to keep out of the way.”
The Pennsylvania obviously adopts the English rule laid down in the Fenham. Law Reports, 3 Privy Council Appeals, 212. But the English rule does not refer either directly or by inference to the “could not” doctrine, as claimed by the defendant. It is stated in this direct language: “The burden lies upon her to show that her non-compliance with the regulations was not the cause of the collision.” Following this rule the Pennsylvania case nowhere declares that it must show that the ship’s fault did not contribute, and then further,, as a matter of independent proof, that it could not have contributed. The reasoning of the court negatives any such conclusion. Having by the process of elimination, declared that it was not enough to show, “not merely that her fault might not have been one of the causes, or that it probably was not, and, the *401ship having failed to establish that her fault did not contribute, the court required her to prove that her fault “could not have contributed” and, arguendo, did not. The Pennsylvania having adopted the rule of proof laid down in the Fenham, undoubtedly intended to declare that the “could not” rule in the one case and the “did not” rule in the other were identical in meaning and that proof of “could not” was equivalent to proof of “did not.”
No cases are cited which attempt, by way of argument, to distinguish the identity in meaning of these two rules, nor does counsel in his brief undertake to differentiate them. In fact, only two cases are cited which employ the language of the instruction complained of. In the Livingstone, 87 Fed., 778, the court say: “The libellant must prove that it did not and could not have caused or contributed to cause the accident.” An examination of this case discloses that this phraseology was a mere academic statement of the law and had no application whatever to the decision of the case in which it was cited. After stating the rule the court immediately discarded it, as a precedent, as the very next sentence unequivocally declares the principle for which we contend. It reads: “The fact •being established that there was no look-out, no red light, no range light and no checking of speed, the libellant must be condemned unless it appears that these omissions did not produce or contribute to produce the collision!’ The latter rule, which is indicated by the Italics, was the one upon which the cáse was decided. The former was a mere abstract statement of law which was completely ignored by the court in its application to the facts. This case cites The Pennsylvania which we have shown affords no precedent for the phraseology used. In the 'Lansdown, 105, Fed., 436, the court in stating the law used this language: “It is incumbent upon her to show in case of collision or other disaster that the violation of the statute not only did not but could not have contributed to the collision.” This was also an abstract statement of law. It had no application whatever to the facts. The could not rule was not involved or referred to by the court in discussing the law and the facts. As a precedent, the case cites first the Fenham and then the Pennsylvania. But it already appears that the “could not” rule is not alluded to in the Fenham at all, and was not wárranted 'by any phraseology used in the Pennsylvania. Moreover both cases *402state an incompatible rule of law. It is then manifest that these two cases, which the presiding Justice used as precedents for the language employed in his charge to the jury, and upon which, without a careful examination, he had a right to rely, were erroneous interpretations of what might appear, upon a casual reading, to be the language of the rule laid down in the Pennsylvania, obviously the parent of these inapt expressions of the law. It is unnecessary to analyze the other cases in the defendant’s brief in support of his contention upon this branch of the case, as none of them appears to go to the extent of sustaining that part of the charge complained of in the case at -bar.
But notwithstanding the exceptionable error in the charge, we think under the statement of facts in the exceptions it becomes harmless error. It is admitted in the bill of exceptions that the plaintiff, it being dark, did not carry a stern white light, that is, did not have range lights, which would, or at least might, have enabled the pilot on the steamship to determine the course in which the plaintiff’s vessel was proceeding, and to take the steps necessary to avoid a collision, either by giving a warning whistle, stopping, or going out of the course. From the cases already cited, the following conclusions may be gathered:
1. That the presumption is that a violation of the navigation rules contributed to the disaster.
2. • That the presumption is the strongest—made so as a matter of policy—but yet rebuttable.
3. That the burden is on the vessel in fault to rebut the presumption.
4. That the presumption may be rebutted by showing in any way that the fault actually did not contribute.
5. That the presumption is not rebutted so long as the proof is doubtful. It must exclude every other hypothesis. And this, we are of the opinion, is what is meant by the “could not” rule.
Therefore, in the case at bar the “could not” rule, as above defined would seem applicable. The case is in the same class as those from which citations have been made. This being so, we are unable to see why the case should not fall precisely in the line of the Pennsylvania and other cases cited, where it was held that it was impossible to determine whether as a matter of fact the fault *403contributed to the disaster or not, and therefore that the presumption against the vessel was not rebutted. The plaintiff did not show range lights, as the rules require. The fact being admitted, upon the decisions cited the plaintiff cannot recover in any event, because it is impossible for him to show that the omission did not contribute, therefore the exceptions should not be sustained, and the case sent back for a useless trial.
Exceptions overruled.
Bird and Haley, JJ., concur in the result.