A jury in the District Court found the defendant Crowder guilty of second degree murder, robbery and carrying a dangerous weapon. The district judge sentenced him to concurrent terms of imprisonment of five to twenty years, five to fifteen years, and one year respectively. On this appeal he contends the district judge erred (1) in receiving in evidence a bullet which pursuant to a court order had been surgically removed from the defendant's forearm without his consent, and (2) in denying his request for an instruction on self-defense. Finding no error we affirm the conviction.

As we shall see the bullet that was surgically removed from Crowder's arm proved to be damaging evidence against him. He contends that the operation was an unreasonable intrusion into his body and therefore violated his Fourth Amendment rights. Considering this argument in the light of all the facts and circumstances, we reject it.

On the afternoon of December 18, 1970 Dr. James E. Bowman, a dentist, was murdered in his office. Death was caused by a gunshot wound entering his chest and coursing through his heart. A caliber .32 slug was recovered from his body, and another caliber .32 slug, which apparently had passed through his body, was found in his underwear. The police found a caliber .32 Smith & Wesson revolver on the ground, across the street from the doctor's office. The revolver contained four expended rounds and two live rounds. It had been kept in the doctor's office and was registered to his wife.

On December 23, 1970 the police arrested one Sandra Toomer, charged her with the murder, and she in turn implicated Crowder. She told the police that she and Crowder had gone to the doctor's office intending to rob him. When Crowder confronted the doctor with a toy pistol, she said, a scuffle ensued, she ran, and she then heard gun shots. Rejoining her, Crowder told her he had been shot in his arm and his left leg, but he thought he had killed the doctor. She observed Crowder's wounds.

Acting on the information given by Sandra Toomer the police arrested Crowder. They noted that his right wrist and left thigh were bandaged. They took him to D.C. General Hospital where x-rays disclosed a bullet lodged in his right forearm and another in his left thigh. The bullets appeared to be caliber .32 slugs. Crowder refused to be treated for his wounds.

As might be expected the police and the prosecutor thought it important to determine whether the slugs in Crowder's arm and leg had been fired from the Bowman revolver. Accordingly, on February 10, 1971 the United States Attorney presented to Chief Judge Curran of the District Court an application for an order authorizing the surgical removal of the bullet from Crowder's arm. The application was supported by an affidavit from Detective Richardson of the Homicide Unit, Metropolitan Police, narrating in substance the evidence that we have set out. In addition, the United States Attorney tendered an affidavit from Dr. Marcus Goumas, Senior Medical Officer at the District of Columbia jail where Crowder was incarcerated. Dr. Goumas affirmed that x-rays of Crowder's left thigh and right forearm revealed the presence of metallic foreign bodies resembling bullets. Dr. Goumas expressed the opinion

that it would be medically inadvisable to remove surgically the slug from Mr. Crowder's left thigh because such a procedure might cause the reduction of use or function of his left leg. The slug in his right forearm, however, is lying superficially under the skin. It is therefore my medical opinion, based upon reasonable medical certainty, that the surgical removal of the slug would not involve any harm or risk of injury to Mr. Crowder's arm or hand or the use thereof. The surgical removal of the slug would be considered as minor surgery. If authorized, it will be performed in accordance with accepted medical practices by one or more surgeons in an operating room at D.C. General Hospital under a local anesthetic. Mr. Crowder will probably be hospitalized for a few days in D.C. General Hospital after this surgery to guard against possible infection.

The United States Attorney's application came on for hearing before Chief Judge Curran on February 10. The defendant Crowder appeared with counsel and objected to the entry of the requested order. Dr. Goumas took the stand and in response to questions by Crowder's counsel testified that the bullet in Crowder's right arm "entered the lateral aspect of the . . . arm and then extended distally and it ended superficially, roughly about three inches or so from the point of entrance. It was a very superficial wound . . . ." The wound was "so superficial" said the Doctor, that removal of the bullet would not affect any radioulnar or medial nerves. "(I)t is a superficial lesion, just like removing a small sebaceous cyst; it would not produce any lasting defect that I can envision." He concluded that the operation could be done under a local anesthetic, and the risk involved would be similar to the risk in "crossing the street".

Chief Judge Curran found probable cause to believe that Crowder murdered Dr. Bowman and that "evidence or instrumentality of that offense" was located in Crowder's right forearm and left thigh. This finding is not challenged. Further, the judge found:

It is medically inadvisable to remove surgically the slug from James L. Crowder's left thigh because this procedure might cause the reduction of use or function of his left leg. However, the surgical removal of the slug in his right forearm which is lying superficially beneath the skin would not involve any harm or risk of life or injury to James L. Crowder's arm or hand or the use thereof.

On the basis of his findings, and citing Schmerber v. California, 384 U.S. 757, 86 S. Ct. 1826, 16 L. Ed. 2d 908 (1966), the judge ordered:

1. That the Superintendent of the District of Columbia General Hospital, or his authorized representative or representatives, shall remove from the right forearm of James L. Crowder the foreign matter disclosed by x-rays and positively believed to be a .32 caliber slug;

2. That such removal is to be done at the District of Columbia General Hospital with accepted medical procedures, with due regard given to the health and preservation of life of James L. Crowder;

3. That if at any time during the course of the removal procedures danger to the life of James L. Crowder develops, such removal procedures shall cease and such other steps as may be necessary shall be taken to protect the health and life of James L. Crowder; and

4. That after removal of the foreign matter, such matter shall be turned over to an authorized representative of the Metropolitan Police Department, who is to make a return to the Court in accordance with requirements of Rule 41 of the Federal Rules of Criminal Procedure.

5. The defendant shall not tamper with or disturb the wound in his right forearm, or remove, destroy or dispose of the bullet lodged therein.

A petition for a writ of prohibition against the execution of Chief Judge Curran's order was denied by this court on March 2, 1971. (McGowan and Leventhal, JJ.)

On April 9, 1971, at D. C. General Hospital, Dr. Henry H. Balch removed the bullet from Crowder's right arm. Thereafter a motion to suppress the bullet as evidence was filed by Crowder, a hearing on the motion was held before District Judge Waddy, and Dr. Balch appeared and testified. The record of the hearing discloses that Dr. Balch was a qualified general surgeon who had been practicing his specialty since 1948. He was chief medical officer in surgery at the District of Columbia General Hospital, Director of Georgetown University Surgical Division at the hospital, and special professor of surgery at Georgetown University Medical School. The surgery was done in the major operating room of the hospital. Another surgeon was present as the doctor's assistant, along with a registered nurse. Describing the operation Dr. Balch explained that the skin over the bullet "was carefully prepared with an antiseptic solution to make it clean, surgically clean. Sterile instruments were used. The surgeons were scrubbed and gowned and gloved and had masks on. . . . So the maximum precautions that we ordinarily take surgically were used." The bullet "was lying on the upper outer aspects (sic) of the right forearm under the skin, easily . . . seen and palpated, a firmness under the skin."

This area was injected with a local anesthetic which we call Xylocaine, and an incision of about 2.5 centimeters was made over this lump dividing the skin completely and exposing the fat; and then by gently squeezing on the outer side of this incision the foreign body, which, in fact, was a bullet,1 was extracted and handed to a police officer, Officer Shelley, who was standing watching the procedure.

The operation was then terminated by placing three or four stitches in the skin to close the skin, and a dressing was applied.

Q. From the time of the scrubbing to the time the stitches were put in, how long did that operation take?

A. A total of about ten minutes.

The area of skin made numb by the local anesthetic was only an inch or two in diameter. The bullet was found "immediately under the skin in the fat. . . . Not in any muscles" and "Not in any effective portion of any major nerves or veins." In the Doctor's opinion the risk to the patient was "negligible" although slightly more than there would be in drawing a sample of blood "because the degree of skin penetration is somewhat larger." Less than five cc.'s of blood were lost, compared to five or ten cc.'s required for a premarital examination, and from ten to thirty cc.'s for a blood study on a sick patient. Because the bullet was removed by court order, without the consent of Crowder

we bent over backwards to make sure from the surgical point of view that he had every protection. Therefore, we brought him in the hospital; we examined him in detail prior to the surgery to make sure that he had no underlying physical disability that would run contra to indicate surgery. (sic)

We tested his sensitivity to the Xylocaine because very rarely people are sensitive to local anesthetics. We tested him prior to the operation to establish he was not sensitive to that.

We then removed the bullet, and then we thought we would keep him around under close observation a number of days to make sure there were no complications that would develop; and that was the reason we did it that way.

Now, in other circumstances in another arrangement, we might have done this whole thing on an out-patient basis and never hospitalized the patient at all.

(In the case of a private patient) I would have done the surgery identically as described, and then I would have given him the note to come and see me on a future day to remove the stitches and I would have discharged him immediately from the hospital.

And if he would have said, "May I go to work?" I would say, "Yes, you can go on to work."

Q. You would have discharged him immediately after the surgery?

A. Correct. Send him on out. Correct.

Crowder was kept in the "care ward" for four or five days and then sent back to jail. There were no complications from the surgery.

The district judge overruled the motion to suppress the bullet.

At trial a firearms identification expert testified that the bullet found in Crowder's arm, a caliber .32 Smith & Wesson long, and those recovered from Dr. Bowman's body and clothes were all fired from the Bowman pistol.

As the Supreme Court observed in Schmerber v. California, 384 U.S. 757, 768, 86 S. Ct. 1826, 1834, 16 L. Ed. 2d 908 (1966), "the Fourth Amendment's proper function is to constrain, not against all intrusions as such, but against intrusions which are not justified in the circumstances, or which are made in an improper manner." When this standard is applied to the facts we have outlined we think the conclusion is irresistable that the removal of the bullet from Crowder's arm was reasonable and proper.

In the first place, before acting to retrieve the bullet the prosecuting authorities presented the matter to Chief Judge Curran, a neutral and detached magistrate, for his decision. An adversary hearing was held at which Crowder appeared with counsel. The appellant concedes that probable cause was established. Chief Judge Curran's order was carefully drawn and hedged so as to protect Crowder's health and life; thus the order directed that only the bullet in Crowder's forearm be removed, leaving untouched the bullet in his thigh, and that every medical and surgical precaution be taken.

As the skilled and experienced surgeon who performed the operation testified "maximum precautions" were taken when the bullet was removed, "we bent over backwards". The bullet, which was small, close to the skin and easily felt, was extracted by gentle squeezing after an incision an inch long had been made. Less than five cc.'s of blood were lost, an amount smaller than may be taken in a premarital examination. The entire operation took ten minutes. In the opinion of the surgeon the risk was "negligible" and in fact there were no complications. Although Crowder was kept in the "care ward" for four or five days the procedure did not require such treatment. Had Crowder been a private patient he would have been discharged immediately and told he might go back to work; the care ward precaution was imposed only because the surgeon "bent over backwards", no doubt having in mind the possibility of future legal complications.

We do not find in these procedures, as the Supreme Court did in Rochin v. California, 342 U.S. 165, 172, 174, 72 S. Ct. 205, 209, 210, 96 L. Ed. 183 (1952), and the defendant does here, any conduct "so brutal and so offensive to human dignity" that it "shocks the conscience." Reasonableness of course is a matter of degree and we do not say that a court may authorize any challenged operation, no matter how major. The limits of reasonableness are well illustrated by the distinction drawn in Chief Judge Curran's order between the bullet in the arm and the one in the thigh; since removal of the bullet in the thigh might cause permanent injury to Crowder it was forbidden. In any event we are not here called upon to give general approval of surgical operations in search of evidence. We are concerned only with the procedures followed in this case. We think those procedures were reasonable and justified in the circumstances. We repeat and summarize the factors that lead us to this conclusion: (1) the evidence sought was relevant, could have been obtained in no other way, and there was probable cause to believe that the operation would produce it; (2) the operation was minor, was performed by a skilled surgeon, and every possible precaution was taken to guard against any surgical complications, so that the risk of permanent injury was minimal; (3) before the operation was performed the District Court held an adversary hearing at which the defendant appeared with counsel; (4) thereafter and before the operation was performed the defendant was afforded an opportunity for appellate review by this court.

If the defendant's argument is sound then no intrusion into a man's body that goes beyond a needle prick can ever be authorized by a court. We cannot agree that this is or ought to be the law. In our opinion the prosecuting authorities in this case made an intelligent and commendable effort to comply with the law; they resorted to "skillful and imaginative legal planning, bottomed upon cooperative utilization, rather than utter disregard, of judicial power, and designed to achieve legitimate ends . . . ." Adams v. United States, 130 U.S.App.D.C. 203, 208, 399 F.2d 574, 579 (1968), cert. denied, 393 U.S. 1067, 89 S. Ct. 722, 21 L. Ed. 2d 710 (1969).

The Denial of the Instruction of Self-Defense

Testifying in his own behalf Crowder gave the jury his version of the killing. He said he had gone to Dr. Bowman's office with Sandra Toomer and one Nick.2 Crowder sat in the waiting room while Sandra and Nick went into the office. Then, said Crowder:

They started to argue. Their voices got loud. So I went into the office.

Just as I was going into the office, Dr. Bowman was pulling a gun out of his pocket. So I grabbed the barrel of the gun and I twist the barrel of the gun this way and I got shot. That is where I got shot in the arm. And I twist the barrel of the gun down this way and I got shot in the leg and I hit the floor. I fell on the floor.

So by this time, Nick was close enough to him, Nick twist the gun out of his hand. By this time Sandra ran out the door and Nick shot him twice, and Nick ran out the door behind Sandra and I ran out behind Nick. (Emphasis supplied.)

At the close of the evidence defense counsel proposed an instruction on self-defense, expressed in some 640 words, divided into 10 paragraphs. The predicate of the instruction, expressed in the first paragraph, was "If you are satisfied beyond a reasonable doubt that the defendant Crowder fired the fatal shot . . . then the following self-defense instructions may be considered by you". The district judge refused to give the requested instruction or to charge on self-defense. The defendant says this was error. We disagree for several reasons.

1. The proposed instruction was complicated and confusing on its face, and if it had been given it would have confused the jury. For this reason alone the court was justified in rejecting it.

2. A more fundamental objection to any instruction on self-defense is that such a charge would have contradicted Crowder's testimony and repudiated the theory of his defense.

Crowder now says the evidence fairly raised the possibility that he fired the fatal shot in defending himself, but at trial he clearly and unequivocally testified that he did not shoot at all, that he was lying on the floor when the fatal shots were fired by Nick. If the jury believed him he should have been acquitted, not because he shot in self-defense, but because he did not shoot. Thus we think the issue of self-defense was not raised by Crowder at trial.

We are not impressed by the effort of counsel to conjure up a case of self-defense from the theory that Crowder perjured himself in his testimony, and that a shooting in self-defense may be inferred from various wisps of evidence. Our conclusion is fortified by Belton v. United States, 127 U.S.App.D.C. 201, 382 F.2d 150 (1967), upon which counsel relies. In that case the defendant testified that after a quarrel and scuffle with his wife he ran out of the house, leaving her uninjured. When he returned some time later he found she had been shot. The District Court denied the defendant's request for a manslaughter instruction. On appeal we said that the defendant's story

if believed by the jury, could lead only to acquittal since appellant claimed he was not even present when deceased was shot. Appellant's counsel argue to us, however, that the jury was not bound to treat his testimony as a unit, either believing or disbelieving it in its entirety, and that the jury was entitled to accept part of appellant's version that his wife assaulted him without provocation and that they struggled for possession of the gun while disbelieving his assertion that his wife was not shot while he was in the room.

127 U.S.App.D.C. at 206, 382 F.2d at 155. We held however that "the charge requested was objectionable as based on speculation without foundation in the evidence." Id. at 207, 382 F.2d at 156. In the present case we think the self-defense theory is also based upon impermissible speculation. The theory fragments the testimony in a "selective process . . . so attenuated as to strain credulity to the breaking point." Brooke v. United States, 128 U.S.App.D.C. 19, 23, 385 F.2d 279, 283 (1967). Granted that the items of proof in a criminal case may resemble the pieces of a jigsaw puzzle, in this case the pieces required for a showing of self-defense are missing and cannot be fabricated from the evidence.

3. Had the court granted the requested instruction on self-defense, Crowder might well contend now that he was fatally prejudiced by the plain implication of the charge that he had lied in his testimony. Such a statement, coming from the court and so at variance with Crowder's testimony, would have eviscerated his defense.

The judgment is affirmed.

I concur unreservedly in Judge Robb's opinion, and I add this word only to emphasize the degree to which, in my judgment, the Government's sensitivity to procedural orderliness and fair play contributed to the result we reach. Had it declined to invoke the authority of the judiciary in advance, relying instead upon after the fact justifications, we would have been presented with quite a different and palpably more difficult case. But because it proceeded as it did, appellant was, prior to the removal of the bullet and at the Government's insistence, afforded an evidentiary hearing before a United States District Judge in which he was represented by counsel, asserted his objections, and had the benefit of cross-examination of the Government's medical witness. Opportunity was further provided appellant, before the operation, to seek appellate scrutiny of the District Court's findings and authorizations.

This case is something of a sport on its facts, and frequent recurrence of anything like it is hardly to be expected. But any future prosecutor confronted with a similar problem would, in my view, be well advised to look to the procedural example set in this case.

I concur in Part II of the dissenting opinion written by Judge Robinson. As to the issue of unreasonable search, I do not feel that I can say in all cases a minor surgical procedure is prohibited by the Fourth Amendment. That is why I declined to vote for a prerogative writ when this matter was before Judge McGowan and myself as a motions panel. I am clear that a surgical procedure is unreasonable unless it is authorized by a judge after a hearing, for reasons touched on by Judge McGowan, and I am also clear that the prosecution must make a strong affirmative case that the surgical procedure is necessary in order to prevent a miscarriage of justice. No such showing was made here, in my view. The prosecution may have thought it convenient to extract the bullet as part of an investigation, but that is not enough. Here there was no proffer beyond an indication that extraction of the bullet would show Crowder was present at the scene of the crime. This surgery was not necessary to establish that fact.*

ROBINSON, Circuit Judge, with whom BAZELON, Chief Judge, and WRIGHT, Circuit Judge, join, dissenting:

In my view, the District Court erred both in admitting into evidence the bullet which had been surgically extracted from Crowder's forearm1 and in later denying his request for an instruction to the jury incorporating his theory of self-defense.2 Accordingly, I would reverse the conviction and remand the case for a new trial. My reasons follow.

* To legitimate its fruits, any search for evidence of crime must meet the stringent demands of Fourth Amendment reasonableness3 and Fifth Amendment due process of law.4 Only seldom have these constitutional prerequisites experienced application to searches extending into the human body. Nonetheless, the natural and logical starting point in this case is an inventory of the three precedents binding upon us in this relatively narrow area of the law.

The earliest is Rochin v. California,5 wherein the Supreme Court held that evidence wrested by state authorities from a suspect's stomach was unusable in a criminal prosecution because the seizure violated the due process guaranty. Police had broken into his bedroom and, seeing him ingest two capsules, had jumped on him and tried to extract them. That effort failing, the suspect had been driven to a hospital where on police orders the capsules were recovered by stomach pumping.6 The Court readily found that the techniques utilized "shock(ed) the conscience"7 and were "too close to the rack and the screw to permit of constitutional differentiation."8 Accordingly, a conviction of possession of morphine, found in the capsules, was reversed.

The Court distinguished Rochin, however, in two subsequent cases involving bodily intrusions to secure blood samples from persons suspected of drunk driving. In Breithaupt v. Abram,9 the Court sanctioned a physician's removal of a sample from an unconscious patient following an automobile accident. Since Breithaupt, like Rochin, preceded judicial integration of the Fourth and Fourteenth Amendments in state prosecutions,10 the Court was called upon to examine the incident only from the standpoint of due process. A divided Court concluded that the methodology employed neither "shock(ed) the conscience"11 nor "offend(ed) a sense of justice,"12 and therefore comported with due process.13

Nine years after Breithaupt, in Schmerber v. California14 the Court again sustained the receipt in evidence of a blood sample withdrawn, without a suspect's consent, to discern its alcoholic content. As in Breithaupt, the Court ruled that blood sampling for that purpose was not violative of due process,15 but shifted its primary focus to two other constitutional considerations. As to one, the Court, for reasons more elaborately discussed later,16 held that the Fourth Amendment's test of reasonableness had been met.17 As to the other, the Fifth Amendment's Self-Incrimination Clause,18 the Court dismissed the accused's argument on the ground that the sample was not of a testimonial or communicative nature.19

This comparatively brief resume of Supreme Court precedents reveals that the only specific judicial decisions binding upon us are that forcible removals of evidence from the body by stomach pumping are unconstitutional while extractions of blood samples, though not assented to, are not. Put another way, the only nonconsensual bodily penetration in the pursuit of evidence thus far sanctioned by the Court is blood sampling, and this for due process as well as for reasonable search. Neither Breithaupt nor Schmerber requires the holding in the case at bar, nor supports it to the extent it goes.

To so observe is not, however, to intimate a view that the surgical recovery of the bullet from Crowder's arm surpassed the limits of due process. The procedures invoked here were devoid of the brutality that loomed so large in Rochin;20 contrarily, not only did the Government shun physical force but it was scrupulously careful to afford Crowder an adversary hearing, with an ensuing appeal, on the propriety of the operation contemplated, and to arrange for maximum medical attention and precautions both during and after the surgery. In these circumstances, it is hardly arguable that the modus operandi "shock(ed) the conscience"21 or contravened civilized standards of decency,22 and I would not conclude that Fifth Amendment due process was implicated. But the further and certainly the decisive question in this case is whether the unconsented-to surgical excision of the bullet was a search of his person exceeding the scope of intrusions tolerable under the Fourth Amendment.

Quite apart from the Government's laudable adherence to the dictates of due process is the precept that no search reaching substantively beyond the Fourth Amendment can stand.23 And although its technology does not run afoul of due process as refined in Rochin and subsequent cases,24 a search is not valid unless it is one which the law deems reasonable.25 It is at this stage of the inquiry that I would conclude that the surgery leading to acquisition of the bullet lodged in Crowder's forearm was barred by the Fourth Amendment's proscription against unreasonable searches and seizures though not intercepted by the Fifth Amendment's ban on inhumane behavior.

In Schmerber, the Court carefully delineated the criteria undergirding its holding that entries into the body so minor and frequent as the insertion of a needle to secure a blood sample are not unreasonable searches. Blood tests have become "a commonplace":26

The blood test procedure has become routine in our everyday life. It is a ritual for those going into the military service as well as those applying for marriage licenses. Many colleges require such tests before permitting entrance and literally millions of us have voluntarily gone through the same, though a longer, routine in becoming blood donors.27

Then, too, "for most people the procedure involves virtually no risk, trauma, or pain."28 And "(e)xtraction of blood samples for testing is a highly effective means of determining the degree to which a person is under the influence of alcohol,"29 a matter to which the Court had earlier spoken in Breithaupt:

As against the right of an individual that his person be held inviolable, even against so slight an intrusion as is involved in applying a blood test of the kind to which millions of Americans submit as a matter of course nearly every day, must we set the interests of society in the scientific determination of intoxication, one of the great causes of the mortal hazards of the road. And the more so since the test likewise may establish innocence, thus affording protection against the treachery of judgment based on one or more of the senses. Furthermore, since our criminal law is to no small extent justified by the assumption of deterrence, the individual's right to immunity from such invasion of the body as is involved in a properly safeguarded blood test is far outweighed by the value of its deterrent effect due to public realization that the issue of driving while under the influence of alcohol can often by this method be taken out of the confusion of conflicting contentions.30

The Schmerber criteria, I submit, do not support the surgery performed in this case. Use of a scalpel to make a cut an inch long and a quarter-inch deep into the subcutaneous fat of the arm is not "a commonplace."31 Unlike the blood tests and perhaps a few other very minor medical procedures which most Americans regularly and unhesitatingly undergo, surgical incisions entail preoperative anesthesia, post-operative pain killers, stitches and permanent scars. They involve much more in the way of physical and mental discomfort and anxiety. They pose significantly higher risks of infection and post-surgery complications, and an immensely greater personal affront.32

Moreover, in weighing individual privacy against public needs for evidence of crime, surgery balances the scales much more closely. The measurement of alcoholic levels by blood testing is perhaps the most reliable method of identifying or exonerating the unfortunately large number of people daily suspected of driving while intoxicated. It may itself be a worthwhile deterrent of more drunk driving than we have. By contrast, the comparative rarity of surgical explorations for evidence of crimes casts doubt on their need for purposes of detection or deterrence.33 Surely in the instant case the bullet retrieved from Crowder's arm was far less crucial in the endeavor to bring him to book than blood tests generally are in convicting drunk drivers. Here the evidentiary significance of the bullet lay solely in its tendency to show that Crowder was present at the scene of the crime a fact never in dispute at Crowder's trial.

Even more fundamentally, the constitutionally indispensible balancing of the societal interest in effective law enforcement and the individual interest in personal integrity assumes an entirely different dimension in surgery cases.34 One's body simply cannot be equated with his car, his clothing, or even his home as a repository of evidence. By the same token, a surgical entry cannot be treated as just another police search.35 To me it seems incontrovertible that, with its marked intensifications of risk, pain, scarring and indignity, a surgical invasion of the body cannot be likened to the needle puncture which Schmerber condones, or justified constitutionally by uncritical comparisons with it. As the Schmerber Court treading on new ground as we are here,36 was careful to state:

It bears repeating . . . that we reach (our) judgment only on the facts of the present record. The integrity of an individual's person is a cherished value of our society. That we today hold that the Constitution does not forbid the States minor intrusions into an individual's body under stringently limited conditions in no way indicates that it permits more substantial intrusions, or intrusions under other conditions.37

I am unable to view calmly this court's action in sanctioning new and far greater bodily exploration than the ubiquitous needle-insertion permitted in Schmerber. By extending that decision to all instances of so-called "minor surgery," the court paves the way for judicial approval of any evidentiary expedition which on its own facts might seem to fall within that ill-defined limit. There are, I believe, grave dangers inseparably incidental to the precedent the court thus sets.

In the first place, the truth of the matter is that judges are not equipped to pass on the constitutional reasonableness of bodily intrusions with the facility and competence with which they deal with other kinds of searches. To be sure, the Fourth Amendment's mandate that searches be reasonable38 allows flexibility in judgments as to whether particular searches survive that test.39 But, as we all know, there are numerous precedents and principles guiding inquiries into and rulings on the reasonableness of governmental encroachments on the privacy of homes, cars, papers and other personal effects. On the other hand, few cases have treated the concept of reasonableness in the context of physical trespasses beneath the human skin.40 Even when the medical phenomena may appear clear, lack of judicial experience and know-how in this unaccustomed area could quite easily translate into unsound decisions.

Moreover, the factors espoused by the court as criteria measuring the legitimacy of surgical searches41 do not offer much practical guidance to judges summoned to assess the import of medical evidence on which authority for such searches will be sought. Expectably in at least some instances the surgery will be much more extensive than here but still short of that presenting more than a "minimal" "risk of permanent injury" thus remaining classifiable as "minor" surgery. What is the judge to do if the evidence indicates that nonetheless the surgery will be performed on the face rather than the arm, or will necessitate an incision four inches by one inch rather than one inch by one-quarter inch, or will in any event be temporarily accompanied by excruciating pain, violent bodily reaction or prostrating disability, or permanently by nondisabling disfigurement, either of body or mind? Unless and until judges can be steered reliably through dilemmas of this magnitude, I think it essential to adhere closely to the Schmerber criteria, and to confine the stamp of our approval to those surgical procedures which meet their narrow and exacting demands.

Equally distressing is the complete absence of safeguards insuring the accuracy of medical evidence bearing on the hazards of proposed operations, upon which judges must necessarily rely. While there is virtually no dispute as to the safety of blood sampling, surgical procedures are frequently subjects of divergent medical opinions as to their complexity and the seriousness of the risks involved.42 Furthermore, despite the judge's best effort to conduct a full and fair adversary hearing on the medical characteristics of a surgical search, it cannot realistically be assumed that access to medical experts is equal as between the prosecution and the defense. So, not only may the degree of genuine conflict in medical opinion not be uncovered but, where conflicts do emerge, the judge's lack of medical expertise may well make it difficult for him to resolve them. The consequences of his decision may be both serious and irreversible, and where the health of an accused hangs in the balance, government should not be permitted to chance a misjudgment as to the effects of a particular operation.

These are among the perils inherent in approval of any extra-Schmerber penetration into the human body. Intrusions beyond the prick of a needle in a blood sampling introduce more than a difference merely in degree; they inexorably involve some differences and portend vast differences in kind. They invite too, a new round of police activity in bodily searches for evidence. There was, I thought, a bright line between Schmerber- and Crowder- type medical procedures, but now that the operation on Crowder is sustained on the loose criteria announced by the court, there is no clear cut barrier to further inroads on bodily integrity short of predictable permanent injury.

Governmental invasion of the citizen's body, even to serve an important governmental interest, poses questions of the most profound character imaginable. It challenges the very foundations of the relation binding the individual and the state. Thus far its constitutional permissibility has been extended only to commonplace, relatively innocuous and indisputably safe intrusions. If these limits are in any instance to be exceeded, the bare minimum is that new limits be established carefully, soundly and precisely. For, as the Supreme Court, when holding violative of the Fourth Amendment another threat to personal privacy, has warned,

(t)hough the proceeding in question is divested of many of the aggravating incidents of actual search and seizure, yet . . . it contains their substance and essence, and effects their substantial purpose. It may be that it is the obnoxious thing in its mildest and least repulsive form; but illegitimate and unconstitutional practices get their first footing in that way, namely, by silent approaches and slight deviations from legal modes of procedure. This can only be obviated by adhering to the rule that constitutional provisions for the security of person and property should be liberally construed. A close and literal construction deprives them of half their efficacy, and leads to gradual depreciation of the right, as if it consisted more in sound than in substance. It is the duty of courts to be watchful for the constitutional rights of the citizen, and against any stealthy encroachments thereon. Their motto should be obsta principiis.43

No one can confidently predict how far today's decision actually goes. While it professes to stop short of a declaration "that a court may authorize any challenged operation, no matter how major," it clearly sanctions any judicially approved surgical operation which may be thought to be "minor." It furnishes no standard by which necessary differentiations between "major" and "minor" undertakings are to be achieved, nor any assistance to judges who must function in the no man's land in which they are to be made. In sum, it leaves to ad hoc determination, simply on the elastic and imprecise scale of reasonableness, the extent to which the arm of government may reach inside the human body. That, I fear, starts us down a slippery slope on which there can be no stopping.44IV

It has long been settled that an accused is entitled to have his jury instructed on self-defense if the issue is fairly presented by the evidence.45 Before the District Court charged the jury in the case at bar, Crowder's counsel requested self-defense instructions on the second-degree murder count.46 The court refused to so instruct and my colleagues in the majority uphold its action in that regard. I must respectfully disagree.

I do not suggest that Crowder would have had a self-defense claim if Dr. Bowman had been killed in the course of commission of a felony in which Crowder participated. The jury, however, acquitted him on the felony-murder count of the indictment on which he was tried, and convicted instead on separate counts of second-degree murder, robbery and carrying a dangerous weapon. The alleged robbery harvested only that weapon Dr. Bowman's pistol and since the evidence showed without contradiction that the pistol was carried away only after Dr. Bowman was shot, and did not suggest that a robbery of the pistol was in progress when the shooting occurred, Crowder was at liberty to assert self-defense as to the homicide itself if there is appreciable evidence to support that claim.47 The instructions sought were addressed solely and specifically to the second-degree murder count, and I think the District Court erred in eliminating self-defense from the jury's consideration of that charge.

On this appeal, the court says first that "(t)he proposed instruction was complicated and confusing on its face, and if it had been given it would have confused the jury." I am unable to condemn defense counsel's proposals on that basis. But even if the court's characterization is correct, the matter should not be closed at that point. Of course, if the requested instructions were really what the majority says they were, the District Court properly refused to give them in the form in which they were tendered.48 But Crowder's counsel made clear what he wanted, and I see nothing that would have inhibited a rephrasing to eliminate any undue complexity or confusion, or a rendition of self-defense instructions framed by the District Court. Either course would have filled the gap which that court left when it declined to give any self-defense instruction at all.

The majority also argues that "(a) more fundamental objection to any instruction on self-defense is that such a charge would have contradicted Crowder's testimony and repudiated the theory of his defense."49 But, as we have heretofore had occasion to state, "a defendant is entitled to an instruction on any issue fairly raised by the evidence, whether or not consistent with the defendant's testimony or the defense trial theory."50 The jury is free to pick and choose from the evidence; it may believe part of the accused's testimony, disbelieve other portions and accept less than all of the Government's case.51 If there is some evidence of self-defense, the accused is entitled to an instruction on that score, even though the jury may be required to infer a state of facts ascertained by crediting part of his testimony and that of prosecution witnesses on other points in dispute.52

Of course, despite the jury's broad authority to choose from the evidence offered, "the selective process must not be so attenuated as to strain credulity to the breaking point."53 The factors which determine whether a requested instruction is supported by evidence or whether the request must be denied because it lacks an adequate foundation and thus invites speculation by the jury are "whether there must be extensive picking and discarding of evidence, whether the ultimate finding necessitates fragmentation of testimony in such degree as to distort it, and whether facts not supported by proof must be supplied."54 And the jury may reconstruct events which can reasonably be deduced from the evidence although not testified to directly.55 In Belton v. United States,56 we said that

. . . an accused is entitled to an instruction . . . if there is "any evidence fairly tending to bear upon the issue . . . " however weak, and . . . the court may not intrude on the province of the jury which may find credibility in testimony that the judge may consider completely overborne by the "simply overwhelming" evidence of the prosecutor.57

In the instant case, Crowder testified that he entertained no thought of robbery when he and Ms. Toomer went to Dr. Bowman's office; that he did not possess a toy pistol at the time; and that he did not know why Ms. Toomer had brought him there. He also testified that he entered Dr. Bowman's inner office only after hearing a commotion within; that he tussled with Dr. Bowman in an attempt to seize the gun, but only after he saw Dr. Bowman removing it from his pocket, and only in the belief that he was about to be shot. It is conceded that Crowder was, indeed, shot twice during the struggle for the pistol and that the bullet which killed Dr. Bowman was fired at an extremely close range. This assortment of evidence left room, then, for an inference that the fatal shot occurred during the encounter, and not after it had fully run its course.

This, in my view, constitutes evidence "fairly tending to bear"58 on Crowder's theory of self-defense. A jury could rationally interpret the events thus portrayed as proof that Crowder, lacking any preconceived intent to rob or injure and drawn into a scuffle during which he already had been shot twice, reasonably felt a necessity to shoot in self-defense.59 In sum, the jury might have accepted the Government's evidence that Crowder fired the pistol, even though Crowder claimed that one "Nick" had done so, but disbelieved the Government's version of Crowder's motivation for the shooting. This is precisely the type of picking and choosing from the evidence which our decisions authorize.60 I am mindful that the weight of the evidence supporting the self-defense instructions sought was weakened by the fact that Crowder did not testify affirmatively that he shot in self-defense, but that was a matter for the jury.

In close cases, especially those involving requested instructions pertaining to defense theories, trial courts should lean toward ample leeway for the jury to determine innocence or guilt in light of all legal considerations bearing on the issue. This is particularly true when, as here, the court has been apprised by defense counsel specifically of the reconstruction of events underpinning the desired instruction.61 And a refusal to give an instruction on a defense theory for which there is sufficient evidentiary support constitutes reversible error.62 I submit that, accordingly, Crowder should be granted a new trial for this reason as well as for the improper admission into evidence of the bullet extracted from his arm.