Opinion of the Court
The accused was charged in two specifications with desertion and absence without leave, in violation of the Uniform Code of Military Justice, Articles 85 and 86, 10 USC §§ 885 and 886, respectively. He pleaded guilty to absence without leave as to both offenses and, after a trial on the merits, was found guilty of two unauthorized absences. He was sentenced to dishonorable discharge, total forfeitures, and confinement for one year, and intermediate appellate authorities have affirmed. Accused sought review here, and we granted his petition on the issue of whether his first absence, beginning on August 4, 1953, commenced “in time of war” within the meaning of Article 43 (a) of the Code, 10 USC § 843. That question became important because of its possible effect on the statute of limitations applicable to absence without leave.
On August 4, 1953, accused absented himself from his unit at Fort McClellan, Alabama. His absence was terminated on January 4, 1956, when he returned to that post and inquired concerning a discharge certificate, at which time he was taken back under military control. Sworn charges against him for this alleged desertion were received by the officer exercising summary court-martial jurisdiction on January 11, 1956. While awaiting trial on this charge, accused, on January 25, 1956, again absented himself, and remained absent until Jan*649uary 81, 1956, when he once more returned.
No question is raised as to the conviction for the latter period of absence, so we turn our attention to the oifense charged as desertion. As previously mentioned, the accused pleaded not guilty to that charge, but guilty to the lesser included oifense of absence without leave for a shorter period of time than that alleged. At the time of arraignment, the law officer instructed him concerning the meaning and effect of his plea, and advised him he need not so plead, but neither mention nor explanation was made to accused of the applicability of the statute of limitations to the lesser offense and his right to assert the statute as a bar to prosecution on the included offense. Not being fully advised, the accused adhered to his plea, and it was accepted by the law officer. The Government thereupon proved a prima facie case, and accused elected to become a witness in his own behalf. He admitted that he was absent from August 4, 1953, to January 4, 1956, but he offered the following explanation. While absent, he was arrested by civilian authorities “around the 1st of September, 1953,” and tried and convicted by them on November 11, 1953, for assault with intent to commit murder, for which he was sentenced to imprisonment for one year. During his incarceration, he was visited by agents of the Criminal Investigation Division, who informed him that he need not return to the Army upon his release, and that a discharge would be mailed to him. He therefore believed he had been discharged. Accused concluded his testimony by saying that he was released from prison on July 7, 1954, and subsequently, in an effort to avail himself of Veterans’ Administration medical care, proceeded to Fort McClellan to see about his discharge, at which time he was taken into military custody.
II
On appeal, defense counsel contends that this conviction cannot stand because the “time of war” contemplated by Article 43 (a), Uniform Code of Military Justice, supra, ended July 27, 1953 —several days before the commission of the offense in question; the two-year statute of limitations under Article 43 (c) therefore applies; more than two years intervened between the commission of the offense alleged and the time when charges were received by an officer exercising summary court-martial jurisdiction over the accused; the law officer did not inform accused of his right to assert the statute of limitations in bar of punishment for the lesser included offense as required by the Manual for Courts-Martial, United States, 1951, paragraph 68c, page 100, and accused was not independently aware of his right; and, therefore, the right to interpose the defense has not been waived and accused is entitled to a rehearing.
While certain circumstances which were not argued before us on appeal— and which require an affirmance of the findings and sentence — would make it possible for us to avoid answering the question we elected to hear, we believe that the orderly administration of military justice demands that we end uncertainty concerning the termination of the Korean wartime period. We, therefore, proceed to decide the original granted issue.
Article 43 (a) of the Code, supra, provides in pertinent part:
“A person charged with desertion or absence without leave in time of war . . . may be tried and punished. at any time without limitation.”
Article 43(c) of the Code, supra, sets out a two-year statute of limitations for absence without leave in time of peace in the following terms:
“Except as otherwise provided in this article, a person charged with any offense is not liable to be tried by court-martial ... if the offense was committed more than two years before the receipt of sworn charges and specifications by an officer exercising summary court-martial jurisdiction over the command. . . .”
Thus, in deciding whether the two-year period of limitation -bars pros*650ecution of accused’s offense, we are confronted with the problem of determining whether or not his absence, beginning August 4, 1953, occurred during time of war.
In holding that offenses committed during the Korean conflict were perpetrated during a time of war within the meaning of the language of the Code, we reasoned that war does not result only from a formal declaration by Congress, but can exist in fact, without a declaration, and that its existence is to be determined by the realities of the situation as distinguished from legalistic niceties. United States v Bancroft, 3 USCMA 3, 11 CMR 3; United States v Gann, 3 USCMA 12, 11 CMR 12; United States v Ayers, 4 USCMA 220, 15 CMR 220; United States v Taylor, 4 USCMA 232, 15 CMR 232. The factors which, considered collectively, compelled our finding of a “time of war,” are set out in Bancroft, supra, and its allied cases. Of crucial importance in all of the cases, however, was the existence of armed hostilities against an organized enemy. In Bancroft, noting among other things the presence of Americans on Korean battlefields and the casualties involved, we said (3 USCMA 3, 6) :
“• • • Moreover, we believe that battle conditions, where many lives depend upon the proper performance of hazardous duty by each and every individual, require that peacetime sentences with regard to military offenses be discarded and the more severe wartime sentences be invoked. It would indeed be an insult to the efforts of those servicemen who are daily risking their lives in defense of democratic principles to hold that peacetime conditions prevail. Conceding that other courts in certain civilian cases have held that a formal declaration of war is a condition precedent to a state of war, the reasons which are influential there are not persuasive here. For our purposes we need not get into the refinements of those cases which interpret the terms of a contract nor decide whether we are engaged in a de facto or de jure war. Practical considerations are more important . . . when The President, as Commander-in-Chief, ordered members of the armed services into the conflict, he involved this country in hostilities to such an extent that a state of war existed; and that Congress, when it used the phrase ‘in time of war’ in the military Code, intended the phrase to apply to that state regardless of whether it was initiated or continued with or without a formal declaration.”
Similarly, we said in Gann, supra (3 USCMA 12, 13) :
“. . . The factors which make certain offenses — of which the present is one — more heinous and reprehensible in time of war depend upon upon the existence in fact of substantial armed hostilities — regardless of whether those hostilities have been formally declared to constitute ‘war’ by action of the Executive and Congress.”
In Ayers, we also applied a “yardstick of practicality,” and approved the previous holdings in Bancroft and Gann in these words (4 USCMA 220, 221) :
“. . . The hostilities existing at that time in Korea must be equated to a state of war for purposes of military criminal law.”
And in Taylor, we used that same pragmatic approach, cited with approval our decision in Ayers, and said (4 USCMA 232, 237) :
“In construing Article 43(a) — which removed all limitations on the prosecution of certain offenses committed ‘in time of war’ — we determined that an absence without leave transpiring during the period of actual hostilities in Korea might lawfully be punished at any time.”
Manifestly, if, in the absence of formal Executive or Legislative action, a collection of factors compels a conclusion that a state of war is being maintained, when those factors or a substantial part thereof no longer exist, a contrary deduction may be required. It is true, as we indicated in United States v Sanders, 7 USCMA 21, 21 CMR 147, that various official actions in other governmental departments were con*651sidered to be circumstances touching on the ultimate issue before us, but they were only straws in the wind which we used to support our holding in that case, and were not essential to the decision. There we held that the Korean war had ended by June 4, 1955, at least, but we did not hold that it had not terminated before that date.
In the instant ease, we see a vast change from the conditions found in Bancroft and its compan- ions. On July 27, 1953, an armistice was effectuated between the United Nations Command, of which our forces are a part, and the enemy command. Korean Armistice Agreement, Senate Document No. 74, 83d Congress, 1st Session, page 98 (1953); 99 Cong Record 9858. Pursuant to its terms “. . . a complete cessation of all hostilities in Korea by all armed forces” was accomplished. Armed combat ended and battlefield conditions ceased; there was no more shooting and there were no more battle casualties; a Demarcation Line and Demilitarized Zone were established; and war prisoners were repatriated. Assessing the situation existing as of that time, we note the absence of many of those factors upon which we predicated our previous holdings that we were at war. Thereafter, it was no longer necessary to provide the logistical support essential to maintain combat operations in terms of both quality and quantity. American troops no longer had to be rotated to equalize combat duties. Patrolling, while no doubt still intensive, no longer was aimed at penetration into enemy territory. A buffer zone was established to prevent just that sort of activity. The kind of medical support required in a combat situation no longer had to be furnished. Last, and perhaps most important, the whole theory upon which the presence of our troops in Korea was premised changed with the signing of the Armistice. We were no longer there to repel aggression; thereafter our mission was to maintain a peaceful status quo. Our military situation was that of-immediate readiness, not armed conflict.
Some argument is advanced that, following our holding in Sanders, supra, we must require more than an armistice in order to find the war to have ended. A careful reading of that opinion will disclose that the Presidential Proclamation and Executive Order cited therein were relied on by us as indicative of an independent assessment of the situation by others, and an Executive determination that at that time a state of war did not exist, for which reason wartime benefits were ended. While it may be repetitious, we again say that nothing said there forbids a holding here that the war ended prior to the time critical in Sanders. There we left the precise time undetermined, but we have concluded it is better by far to fix a day certain for the Services than to reach the same result by a course of ad hoc decision. We, therefore, hold that the Armistice in Korea on July 27, 1953, ended those actual hostilities essential to a finding of “time of war,” and hence that for the purposes of military law the state of war terminated on that date.
Ill
Having held that accused’s offense did not occur during time of war, we are faced with the task of deciding the effect of the applicable two-year statute of limitations on his case. The accused absented himself on August 4, 1953, and it was not until January 11, 1956, well over two years later, that sworn charges against him were received by an officer exercising summary court-martial jurisdiction — a period which on its face would bar trial and punishment for absence without leave under Article 43 (c) of the Code. Manual for Courts-Martial, supra, paragraph 68c, page 101, provides:
“If it appears from the charges that the statute has run against an offense or (in the case of a continuing offense) a part of the offense charged, the court will bring the matter to the attention of the accused and advise him of his right to assert the statute unless it otherwise affirmatively appears that the accused is aware of his rights in the premises. See 53h. This action should, as a rule, be taken at the time of arraignment. If the *652accused pleads guilty to a lesser included offense against -which the statute of limitations has apparently run, the court will advise the accused of his right to interpose the statute in bar of trial and punishment as to that offense. See also 74Ji.”
At trial, accused pleaded guilty to such a lesser included offense against which the statute had apparently run. The law officer made no mention of the matter, and it does not appear from the record that accused was otherwise aware of his rights. Thus appellate defense counsel urge us to hold that the law officer committed reversible error by failing to advise accused of his right to interpose the statute as a defense. We believe, however, that counsel have overlooked or failed to argue one important facet of the question.
Article 43(d) of the Code, supra, provides:
“Periods in which the accused was absent from territory in which the United States has the authority to apprehend him, or in the custody of civil authorities, or in the hands of the enemy, shall be excluded in computing the period of limitation prescribed in this article.”
The accused himself testified that while absent without leave he was arrested by civilian authorities “around the 1st of September, 1953” and that he was not released from prison until July 7, 1954. Accused, then, was in civilian custody for approximately ten months during his twenty-nine-month absence, and the statute of limitations was accordingly tolled during that period. Consequently, for the purposes of Article 43 (c) in this case, the statute of limitations had run for only nineteen months at the time sworn charges were received by the officer exercising summary court-martial jurisdiction. Such being the state of the record before us, prosecution or punishment for the offense of absence without leave was not barred.
It is said, however, that this rule may not be invoked because of the provisions of the Manual for Courts- Martial, supra, paragraph 68c, page 101, which are as follows:
“The burden is not on the defense to show that neither absence from the territory in which the United States has authority to apprehend him nor other impediment prevents the accused from claiming exemption under Article 43. For example, if it appears from the charges in a peacetime desertion ease that more than three years have elapsed between the date of the commission of the offense and the date when sworn charges and specifications were received by an officer exercising summary court-martial jurisdiction over the command which includes the accused, the motion should be sustained unless the prosecution shows by a preponderance of evidence that the statute does not apply because of periods which, under the provisions of Article 43<f, are to be excluded in computing the three years.” ;
Thus, we are told that there was no evidence to toll the statute at the time accused entered his plea and was entitled to advice on the statute of limitations ; that had he been advised properly he would not have testified and convicted himself; and that because the burden is on the Government to show by a preponderance of evidence that the statute was tolled, it should not now be permitted to take advantage of accused’s testimony to supply the evidence, for the testimony was compelled by the law officer’s failure properly to discharge his duties. We find little merit in this argument. !
First, it is to be noted that accused was charged with desertion, not absence without leave. The three-year statute of limitations which Article 43(b) of the Code provides for peacetime desertion clearly had not run. Had accused elected not to testify, an act he now protests, he in all likelihood would have been convicted of desertion — an unhappy prospect indeed, for the statute did not bar punishment for that offense. Now that the trial is over and prosecution on the greater offense barred, it is easy for the accused to say that he would not have testified had he been advised of the applicability of the statute. For us to accept that contention, would require that we indulge in the-' *653sheerest kind of speculation. The fact remains that accused benefited immeasurably by going on the stand' and, if we are to wander in conjectural fields, we cannot believe that his defense would have been altered one whit had he known of the statute or its significance.
Second, the fact is that accused testified. Let us suppose for the purpose of argument that the law officer erred in failing to advise him concerning the statute of limitations at the time of arraignment. Should a motion interposing the statute as a bar to prosecution stand on a different footing from other motions which may be predicated on the evidence? We think not, and we believe an appropriate analogy may be drawn here to the motion for a finding of not guilty. If the latter motion is made at the close of the Government’s case, and improperly denied, and thereafter the accused goes ahead and by his evidence cures the deficiency in the Government’s case, the authorities uniformly agree that he cannot take advantage of the erroneous ruling. The Manual for Courts-Martial, supra, paragraph 71a, page 109, sets forth this rule in these terms:
“The court on motion of the defense may enter a finding of not guilty as to one or more offenses charged after the evidence on either side is closed if the evidence is insufficient to sustain a conviction of such offense or offenses. If a motion for a finding of not guilty at the close of the evidence offered by the prosecution is not granted, the defense may offer evidence without having reserved the right to do so. But if all the evidence in the record, whether adduced by the defense or the prosecution, or both, is sufficient to sustain a conviction, such conviction need not be set aside upon review merely because the court erred in denying such motion for a finding of not guilty at the time it was made.”
Military law on this topic accords with the practice followed by Federal courts. In Leyer v United States, 183 Fed 102 (CA2d Cir) (1910), holding that if a conviction is sustainable by the whole record, accused cannot, after curing the defect, complain if the court erroneously denied his motion when made, the court said:
“. . . If the whole record indicates that a verdict of guilty was justified, it is immaterial that evidence essential to conviction was voluntarily introduced by defendant himself. There is no force in the contention that the denial of the motion to direct acquittal at the close of the case for the prosecution ‘would in effect shift the burden of proof, and the defendant would be compelled to go forward and prove his innocence before the prosecution had succeeded in proving his guilt.’ Defendant was not compelled to go forward. If the prosecution failed to make out its case, he could quite safely rest upon his exception, knowing that, even if the jury should find a verdict against him on such incomplete proof, it would be promptly set aside.”
The same result was approved after the adoption of Rule 29 of the Federal Rules of Criminal Procedure, 18 USC, in United States v Goldstein, 168 F2d 666, 670 (CA2d Cir) (1948), in these words:
“. . . But no more now than before can he take advantage of an error if he makes that error harmless. In other words, if the evidence is short as the prosecution leaves it, he may take advantage of that. But if he amplifies the record on the facts in attempting to make a case for acquittal he must assume the risk of having the prosecution’s case bolstered in the process.”
Hence, even had the matter been raised at the trial level, we would consider the entire record in determining whether the statute of limitations was tolled. Here, however, the question was raised for the first time on appeal, and we are not inclined to adopt a view which, in substance, asserts that only that portion of a record favorable to an accused is before us. His testimony on the merits, when not limited at trial, is a part of the record on appeal for all purposes, and here it supplies the deficiency for he established an exception to the running of the statute. We con-*654elude, therefore, that the statute of limitations had not run against the offense found by the court-martial.
Because no two Judges of the Court agree upon an affirmance of the findings and sentence, they are ordered reversed. As will appear from the Chief Judge’s opinion, he concludes the issue of an honest and reasonable mistake of fact was raised reasonably by the evidence, while the author Judge reaches a diametrically opposite conclusion. However, under the principles set out in the concurring and dissenting opinion of the former, an absence without authority covering a period from August 4, 1953, to July 7, 1954, may be sustained. To dispose of the case, and for that reason alone, I join him in the following disposition. The findings and sentence are reversed, and the record is returned to The Judge Advocate General of the Army for reference to a board of review, who may either grant a rehearing or affirm a finding of unauthorized absence for the period set out above and reassess sentence in the light of its findings.