Judge Hutcheson, speaking for this Court, recently observed:

Appellate review of the sufficiency of the evidence is never an easy matter, and the difficulty is compounded when a jury in a criminal case has reached a verdict contrary to both medical testimony and visceral reaction. Brock v. United States, 5 Cir. 1967, 387 F.2d 254, 257.

As in Brock, we are here confronted with the question whether "reasonable doubt must necessarily have existed in the minds of reasonable jurors regarding appellant's sanity". Answering this question in the affirmative, we reverse the conviction of Richard C. Nagell.

As Judge Hutcheson pointed out, "Each [case] must be decided upon its own facts, with careful attention to the weight of the evidence on each side." Brock v. United States, 387 F.2d at 258. This case, like Nagell himself, has had a long and stormy course. Nagell was first tried in 1964. The charge was then, and is now, entering a federally insured bank with intent to rob in violation of 18 U.S.C. § 2113(a). He was convicted as charged. Later, the court held a full evidentiary hearing on the defendant's motion for a new trial when Nagell revealed a number of facts previously unknown to his counsel. After denial of this motion, the defendant appealed to this Court. We reversed. Nagell v. United States, 5 Cir. 1966, 354 F.2d 441. A second trial followed. Again the jury found Nagell guilty. On this appeal, Nagell challenges the conviction on a number of grounds, one of them being the sufficiency of the evidence to support the jury's conclusion that he was sane — i. e. whether a reasonable man would have had no reasonable doubt as to his sanity. Since we conclude that reasonable doubt must have existed in the minds of reasonable jurors regarding Nagell's sanity, we do not reach the other issues presented here.

The facts of this case are extensively reviewed in this Court's earlier opinion.1 354 F.2d 441. Briefly, the facts underlying the charge are as follows:

The record shows that late in the afternoon of September 20, 1963, appellant went into the State National Bank of El Paso, Texas. He asked where travelers' checks could be obtained, and upon reaching the proper cage asked the teller, a young woman, for one hundred dollars worth of checks in ten dollar denominations. The teller moved to get them, whereupon Nagell said, "Lady, this is a real gun". She immediately ran, and appellant took several steps away from the cage, fired two shots into the wall at a height of about seven feet, not aiming at the teller, and ran out of the bank. He was followed by a police officer who happened to be in the bank at the time. He was, without difficulty, arrested at a time when he was about to leave in an automobile which he had left parked near the bank. 354 F.2d at 442.

On appeal from his first conviction, Nagell asserted that his sanity was not shown beyond a reasonable doubt, but we rejected this assertion without discussion. During the first trial four doctors were called to testify and all four, though suggesting that Nagell manifested some psychological abnormalities, testified that he had the capacity to distinguish right from wrong on September 20, 1963 the date of the alleged offense. A substantial and significant segment of Nagell's medical history came to light after that trial; it was developed in the hearing on his motion for a new trial, and we related it in our earlier opinion. 354 F.2d at 447.

In the second trial this new medical history and the diagnosis related thereto was unfolded through the testimony of Dr. Edwin A. Weinstein.2 With the newly discovered relevant history at hand, two psychiatrists, Drs. Bennett and Hernandez, who had testified at the first trial that Nagell could distinguish between right and wrong on the date in question reversed their testimony and without hesitation announced at the new trial that Nagell could not distinguish between right and wrong on that date, could not appreciate the nature of his actions, could not refrain from doing wrong. Both of these doctors are "certified" in neurology and psychiatry. Another psychiatrist, Dr. Alderete, on the basis of observation, testing, and a review of the medical history, testified that in committing the act for which he was being prosecuted Nagell did not think he was doing wrong, did not know the nature and quality of the act, and could not conform his conduct to the requirements of the law. Finally, two clinical psychologists, both of whom had done psychological testing of Nagell, testified that when Nagell went into the State National Bank he did not appreciate the implications of his act and was not in control of his conduct.

We recognize that "expert opinion as to insanity rises no higher than the reasons upon which it is based" and that "it is not binding upon the trier of the facts". Dusky v. United States, 8 Cir. 1961, 295 F.2d 743, 754; Breland v. United States, 5 Cir. 1967, 372 F.2d 629, 633; see Mims v. United States, 5 Cir. 1967, 375 F.2d 135, 143. "Description and explanation of the origin, development and manifestation of the alleged disease are the chief functions of the expert witness." Carter v. United States, 1956, 102 U.S.App.D.C. 227, 252 F.2d 608, 617; Fitts v. United States, 10 Cir. 1960, 284 F.2d 108, 113. Here the record is replete with expert testimony regarding Nagell's mental condition: "Mentally disturbed", the particular characterization being "chronic traumatic encephalopathy" — a disease of the brain caused by trauma. Its symptoms: paranoia suicidal preoccupations, "confabulations", tendency toward projection, impaired judgment, lack of contact with reality.

"The nature and quantum of rebuttal evidence sufficient to present a jury question is to some degree determined by the strength of the case for insanity." Brock v. United States, supra, 387 F.2d at 258, and authorities cited n. 10. Although in many of the cases when an appellate court has reversed a jury finding of sanity the Government had only introduced lay testimony as to the defendant's sanity, in some cases the Government had introduced psychiatric testimony on this issue. Isaac v. United States, 1960, 109 U.S.App.D.C. 34, 284 F.2d 168; United States v. Westerhausen, 7 Cir. 1960, 283 F.2d 844. We do not resolve this "battle by psychiatrist" quantitatively; that is why we reassert that "each [case] must be decided upon its own facts with careful attention to the weight of the evidence on each side." We acknowledge that "questions of the credibility and weight of expert opinion testimony are for the trier of facts," Mims v. United States, supra 375 F.2d at 140, but we also recognize that "we must reverse a criminal conviction when it is `clear to us that upon the evidence * * a reasonable mind must necessarily have had a reasonable doubt as to * * * guilt'." Hopkins v. United States, 1960, 107 U.S.App.D.C. 126, 275 F.2d 155, 157 n. 2 and cases cited therein. "The quantum and nature of proof the Government must offer to take the case to a jury varies in different situations and to some degree depends upon the quantum and nature of proof the defendant offers." Wright v. United States, 1957, 102 U.S. App.D.C. 36, 250 F.2d 4, 7; United States v. Westerhausen, 7 Cir. 1960, 283 F.2d 844, 852; see Hopkins v. United States, 1960, 107 U.S.App.D.C. 126, 275 F.2d 155, 157; Brown v. United States, 5 Cir. 1965, 351 F.2d 473, 474. We merely conclude here, as the Tenth Circuit did in McKenzie v. United States, 10 Cir. 1959, 266 F.2d 524, 528, that

We are constrained to the view in this case, in which the evidence of trained and disinterested psychiatrists, whose duty it was to determine the mental condition of the defendant, is so overwhelming as to his insanity, that if the burden of proving sanity beyond a reasonable doubt has any significance at all, it was not met by the meager evidence of the prosecution.

The Government in this case did introduce the testimony of two psychiatrists as to Nagell's sanity. Neither had special training in neurology, and while this is not crucial, it is an important consideration when the disorder involved is asserted to be neurological in origin. More important is the "inadequacy of the factual assumptions on which the opinions are] based". Mims v. United States, supra, 375 F.2d at 143. Dr. Schwartz, while asserting that Nagell had the capacity to tell right from wrong on the date in issue, admitted that he did not have a complete medical history of Nagell, did not check the veracity of the answers related to him by Nagell during interviews (many of which answers were later shown to have been pure confabulation), did not have pertinent medical reports or psychological reports when his diagnosis was made, and did not check relevant nurses notes on Nagell's behavior. Dr. Baker, also concluding that Nagell was sane when he entered the bank, acknowledge that he did not give any of the tests normally given before making a complete evaluation, did not have a "good diagnosis" of Nagell, and had not read the complete file of Nagell's medical history. Finally, Dr. Alderete, a defense witness, observed that he would be hard pressed to reach the diagnosis reached by the defense witnesses without the records that the Government's witnesses did not have in making their evaluation.

We feel constrained to emphasize the particular and peculiar facts of this case reinforcing phychiatric testimony that Nagell was symbolizing, projecting, dramatizing, and/or confabulating when he entered the bank. Cf. Brock v. United States, supra. Nagell demanded travelers' checks, not cash. He asked for a specific (and relatively small) amount. He said, "this is a real gun", not "this is a stick-up". He fired two shots into the wall, not at anyone and for no apparent reason. And he was not at all evasive when he left the bank.

The peculiar facts of this case also suggest error in the charge of the court below regarding the requisite intent under the statute that Nagell was charged with violating. A specific intent to rob is required under 18 U.S.C. § 2113.3 The trial court instructed the jury that it might infer intent from Nagell's conduct in the bank.4 Viewing the totality of Nagell's conduct, and in light of the strong evidence of his insanity at the time, we observe that it was error for the court not to instruct the jury in terms of the more restrictive specific intent required by the statute. See Clifton v. United States, 5 Cir. 1965, 341 F.2d 649, 650-651; Mann v. United States, 5 Cir. 1963, 319 F.2d 404, 409-10.5

Considering the facts of this case and the evidence in the record, we conclude that the evidence introduced by the Government is not sufficient to sustain the conviction. While in some of the cases in which the appellate courts have reversed convictions on this ground the cases were remanded to allow the Government an opportunity to strengthen its position,6 we feel that "no good purpose could be served in ordering a new trial" in this case. United States v. Westerhausen, 7 Cir. 1960, 283 F.2d 844; Argent v. United States, 5 Cir. 1963, 325 F.2d 162; see Hartford v. United States, 9 Cir. 1966, 362 F.2d 63; McKenzie v. United States, 10 Cir. 1959, 266 F.2d 524. The judgment of the conviction is reversed and the case is remanded to the district court with directions to vacate that judgment and to enter an order granting the defendant's motion for acquittal and a judgment acquitting the defendant.