The appellant, Medlock, was convicted in 1952 of the crimes of second-degree murder, assault and armed robbery, and assault and robbery. He was sentenced to imprisonment in the Wisconsin State Prison for terms of fourteen to twenty-five years, three to thirty years, and three to ten years to be served consecutively.

In 1959, Medlock sought executive clemency. On April 23d he was granted a commutation of his sentence. The document issued by the governor, in the portions pertinent to this appeal, provided:

"GREETINGS:

"WHEREAS, Before the Municipal Court for Milwaukee County on the tenth day of June, 1952, Governor Telford Medlock was convicted of the crimes of Second Degree Murder, Assault and Robbery (Armed), and Assault and Robbery, and thereupon sentenced to imprisonment in the Wisconsin State Prison for terms of fourteen to twenty-five years, three to thirty years, and three to ten years, all consecutive; and

"WHEREAS, Pardon is solicited in behalf of the said Governor Telford Medlock, and it appearing that he is deserving of executive clemency;

"NOW, THEREFORE, KNOW YE That in consideration of the premises, I, Gaylord A. Nelson, Governor of the State of Wisconsin, do by these presents commute the sentences of three to ten years and of three to thirty years to be served concurrently with the sentence of fourteen to twenty-five years."

It is the contention of the plaintiff-appellant that the effect of the commutation is to telescope all of the sentences into the maximum period of twenty-five years. The state department of public welfare concluded that the thirty-year sentence controls the maximum term of the sentence and that the only effect of the commutation was to make the sentences run concurrently rather than consecutively, thus reducing the possible maximum from sixty-five years to thirty years.

Medlock brought an action in the circuit court for Dane county for declaratory relief pursuant to sec. 269.56 of the Wisconsin statutes. The trial court permitted the testimony of a linguistics expert, Dr. Robert C. Schweik of Marquette University. Dr. Schweik, over the objection of the attorney general, was permitted to testify that the pardon document was ambiguous, and concluded that, in his opinion, the probable meaning was that the maximum term of imprisonment was limited to twenty-five years. The circuit judge suggested that Governor Nelson's pardon counsel be called to interpret the document. Both counsel objected to such testimony, agreeing that the document must be construed on its face. The court, nevertheless, did seek the advice of Governor Nelson's pardon counsel, who, in a letter, advised the court that the intent of the pardon counsel and of the governor was:

"that these sentences should be commuted to run concurrently rather than consecutively. . . . We intended . . . a minimum sentence of 14 years the longest of the three minimum sentences) and a maximum of 30 years (the longest of the maximum sentences). We did not intend that any significance be given to the sequence in which these sentences were set forth in the commutation itself."

The trial judge also placed in the record a carbon copy of a letter dated March 25, 1959, addressed to Governor Nelson, setting forth the pardon counsel's recommendation to the governor. That letter concluded with the statement:

". . . it is my recommendation that his sentences should be commuted by making all three sentences concurrent rather than consecutive."

The trial judge, in the absence of any other evidence of intent, concluded that he assumed that the governor followed the recommendation of his pardon counsel and made his decision that the only effect of the pardon was to change the three consecutive sentences to three sentences to be served concurrently, with the maximum set at thirty years. Judgment was entered to that effect. The plaintiff appeals.

HEFFERNAN, J.

We agree with the decision of the learned trial judge, but conclude that the pardon document may on its face be construed to provide for a maximum sentence of thirty years.

The appellant in his brief and in oral argument placed great emphasis on the fact that the sentences of three to ten years for assault and robbery, and three to thirty years for assault and armed robbery, were by the document of commutation directed to be served concurrently with the sentence of fourteen to twenty-five years for the second-degree murder. He contends that the effect of this is to "squeeze" the thirty-year term into twenty-five years, thus reducing the maximum. He bases this conclusion upon the premise that it was unnatural for the author of the pardon document to refer to the three sentences in that order unless he intended the fourteen to twenty-five-year term to define the maximum.

In interpreting this document, however, it must be remembered that the sentences as originally imposed were consecutive, and that the term for fourteen to twenty-five years commenced in 1952 and at the time of the commutation was still running. The appellant was in fact still confined by virtue of the first of the three consecutive sentences, the one for second-degree murder. The other two would commence only when the preceding sentence had been served. It is conceded by both appellant and respondent that the effect of executive clemency was to make all three of the sentences concurrent. The effect is as though the sentencing judge had originally ordered three concurrent sentences. Since the record shows that one of the three sentences was partially served, it was necessary that the author of the document make it clear that the other two sentences were to be concurrent with it; hence, the very natural phrase, under the circumstances, that "the sentences of three to ten years" and "three to thirty years" (neither of which had started to run) were "to be served concurrently with the sentence of fourteen to twenty-five years," which had been served in part. By virtue of this commutation directing the sentences to be served concurrently, there was in effect sentencing nunc pro tunc, having the effect of starting all of the terms simultaneously with the one that had commenced in 1952. The ordinary and well-recognized rule is therefore applicable.

"Where sentences imposed at different times or for different periods of time run concurrently, the sentences run together during the time that the periods overlap; and the new or longer term does not necessarily terminate at the same time as the prior or shorter term." 24B C.J.S., Criminal Law, p. 660, sec. 1996 (1).

The same rule, that the prisoner will be discharged at the expiration of the longest of concurrent terms, is recognized also by 15 Am. Jur., Criminal Law, page 123, sec. 465.

This court in the Application of McDonald (1922), 178 Wis. 167, 172, 189 N.W. 1029, held that when sentences are determined to be concurrent there was no authority for holding a prisoner after the date at which "the latest of such sentences as to the period of confinement thereunder had expired." When sentences are to run concurrently, the maximum term imposed is determined by the longest sentence.

While we agree with the decision of the trial court, it is apparent from a perusal of the record and of his opinion that he did not construe the pardon on its face, rather he concluded that it was ambiguous and sought extrinsic evidence to interpret it. It is proper that he should do so if he considered the document to be ambiguous. We cannot agree, however, with the evidence which he, on his own motion, and over the objection of both counsel, utilized to reach his conclusion. The trial judge sought to determine the intent of the governor by seeking the advice of Mr. Thomas W. Pierce, pardon counsel for Governor Nelson at the time of the Medlock pardon. It was on the basis of this information that Judge BARDWELL concluded:

"In the absence of any other showing of intent we assume that Governor Nelson was following the recommendation of his Pardon Counsel when he issued the pardon . . . ."

It is our conclusion that the intent of the pardon counsel, as evinced either by his memory of the occurrence or by a copy of the letter of recommendation he sent to the governor at the time, is not relevant. The intent to be determined is that of the governor, as declared in the document, and not that of a member of his staff. In addition, the documents relied upon by the judge are not of such a nature that they may be utilized in the proper exercise of a court's discretion to take judicial notice of notorious or undisputed facts or of the contents of public documents, or of documents in the possession of an official custodian. 31 C.J.S., Evidence, p. 967, sec. 36.

This document, a communication from an attorney member of the staff to the governor, is not in the usual course of government business published or otherwise made available to the general public. It is not a public record. The letter, at the time of its use by the trial judge, was in the hands of one not then a public officer.

Even though it may readily be conceded that the evidence noted by the court was, in this instance, reliable and trustworthy, it is apparent that to allow what amounts to testimonial evidence not under oath, not subject to cross-examination, and without even the safeguard of the authentication by the seal of an official custodian, might well lead to the erosion of some of the most venerable and necessary rules of evidence.

The testimony of the: pardon counsel was not properly before the court. If it had been, it would not only have been inadmissible, in view of the obligation of the court to construe the pardon from its face, but it would have been irrelevant as not being the intent of the official author of the pardon. We also conclude that, since the trial judge was mistaken in his conclusion that this evidence was permissible under the doctrine of judicial notice, it was improper to consider it. It is well recognized, moreover, that a judge may not supplement the record by seeking out evidence, unless such evidence may come into the case by judicial notice or by some other recognized and approved evidentiary process.

In the brief submitted by the respondent in this appeal, the attention of this court was invited to the Journal of the Senate for January 26, 1961. The legislative journals, are properly the subject of judicial notice, State v. P. Lorillard Co. (1923), 181 Wis. 347, 370, 193 N.W. 613; State v. Joe Must Go Club (1955), 270 Wis. 108, 111, 70 N.W.2d 681; and an executive message published therein may be used by a court in determining the legislative history of a particular act of the legislature. Brad A. Liddle, Jr., Statutory Construction — Legislative Intent — Use of Extrinsic Aids in Wisconsin, 1964 Wisconsin Law Review, 660, 668, 31 C.J.S., Evidence, page 989, sec. 41.

Sec. 6, art. V of the Wisconsin constitution defines the pardoning power of the governor. In part it directs that:

"He shall annually communicate to the legislature each case of reprieve, commutation or pardon granted, stating the name of the convict, the crime of which he was convicted, the sentence and its date, and the date of the commutation, pardon or reprieve, with his reasons for granting the same."

Pursuant to this constitutional direction, the governor's report of the Medlock commutation appears at page 108 of the Senate Journal for January 26, 1961. This is an official document of which we may properly take judicial notice. State v. Jelco (1957), 1 Wis.2d 630, 638b, 85 N.W.2d 487, 86 N.W.2d 428. Since we find it possible to construe the pardon on its face, it is not necessary to take judicial notice of this executive message. However, we note that the reading of this document substantiates the conclusion of the trial judge and the conclusion reached by this court.

The governor's message states:

". . . In view of his war record, and because the consecutive sentences totaling 20 to 65 years seemed severe, I commuted his sentences to run concurrently rather than consecutively on April 23, 1959, said commuted sentences to commence as of the date of the original sentences imposed by the Court."

It is clear from this that the governor's only purpose was to make the sentences concurrent. There is no intent expressed to reduce the length of any of the sentences, and the reading of the commutation itself makes it evident that no reduction of any one of the sentences was granted therein.

While we disagree with the ground upon which the trial judge reached his conclusion, it is apparent that the conclusion was correct.

"It is immaterial what ground is assigned by the trial court for its ruling if it is in fact correct." Wintersberger v. Pioneer Iron Metal Co. (1959), 6 Wis.2d 69, 73, 94 N.W.2d 136.

By the Court. — Judgment affirmed.