Defendant has appealed from a judgment of the District Court entered upon a jury verdict finding the defendant guilty on six counts of an indictment charging violations of 26 U.S.C.A. (I.R.C.1939) 2553(a), 2554(a) and 21 U.S.C.A. 174, relating to the unlawful acquisition, quisition, concealment and sale of heroin. The errors relied on for reversal arise out of rulings on evidence and the failure of the District Court to grant defendant's motion for a mistrial.

The defendant contends that the District Court erred in permitting the government to introduce in evidence certain exhibits prepared by federal narcotic agents who on two occasions made purchases of heroin from the defendant. These exhibits are envelopes in which the heroin purchased from the defendant was placed and then transmitted to a government chemist for analysis. Various notations made by the narcotic agents concerning the purchases and the circumstances under which they were made appear on the envelopes.1

Defendant also complains that it was error to admit into evidence certain exhibits prepared by the government chemist who made analyses of the heroin purchased from the defendant. These exhibits are envelopes on which the chemist recorded the findings of his analysis.2 It was testified at the trial that the heroin could not be conclusively identified as such until subjected to chemical analysis.

The exhibits were prepared out of the defendant's presence and contain the conclusions of government witnesses as to matters which were in the jury's province to determine. They are clearly hearsay and inadmissible unless within an exception to the hearsay rule. The District Court admitted the exhibits into evidence on the theory that they qualified as business entries under the Shopkeeper's Rule, 28 U.S.C.A. 1732. The government also urges that they were admissible under the Government's Records Act, 28 U.S.C.A. 1733.

Although both parties have lumped together consideration of the exhibits made by the narcotic agents and the exhibits made by the government chemist, we have found it necessary to consider them separately. We think there can be no doubt that the exhibits or memoranda made by the chemist were admissible as having been made in the regular course of business and that it was the regular course of business to make such memoranda or record of the findings of the chemist's analyses of the substances purchased from the defendant. Cf. Wheeler v. United States, 93 U.S.App.D.C. 159, 211 F.2d 19, certiorari denied 347 U.S. 1019, 74 S. Ct. 876, 98 L. Ed. 1140; E. K. Hardison Seed Co. v. Jones, 6 Cir., 149 F.2d 252; and see Rennicke v. United States, 8 Cir., 207 F.2d 429. These exhibits also satisfy the underlying reason for this exception to the hearsay rule and the admission of this class of statements under section 1732-- the probability of their trustworthiness. See Palmer v. Hoffman, 318 U.S. 109, 113-114, 63 S. Ct. 477, 87 L. Ed. 645; 5 Wigmore, Evidence 1522 (3d ed., 1940).

The exhibits or memoranda made by the narcotic agents however satisfy none of the requirements of section 1732. The government has not shown, in the language of the statute, that the statements were recorded in the regular course of business and that it was the regular course of business to make such memoranda or records. And even if memoranda such as the ones in question are regularly prepared by law enforcement officers, they lack the necessary earmarks of reliability and trustworthiness. Their source and the nature and manner of their compilation unavoidably dictate that they are inadmissible under section 1732. They are also subject to the objection that such utility as they possess relates primarily to prosecution of suspected law breakers, and only incidentally to the systematic conduct of the police business. Cf. Palmer v. Hoffman, supra.

The rule contended for by the government could have far-reaching results not envisioned by the drafters of section 1732. In the instant case, the entrants or makers of the memoranda were available for cross-examination and testified as witnesses at the trial, but we can well imagine the situation where the entrant or maker was dead or unavailable and a conviction would be sought on the basis or such memoranda alone. Further, we are not aware of any dire consequences that will befall law enforcement officials by holding such memoranda inadmissible as might occur by the exclusion of the business entries of commercial enterprises. See Massachusetts Bonding & Ins. Co. v. Norwich Pharmacal Co., 2 Cir., 18 F.2d 934, 937. We do not suggest however that section 1732 is limited in its application to enterprises commercial in nature.

Nor do we think that the exhibits were admissible under the Government's Records Act, 28 U.S.C.A. 1733, 'Books or records of account or minutes of proceedings of any department or agency of the United States shall be admissible to prove the act, transaction or occurrence as a memorandum of which the same were made or kept.' On their face the memoranda in question do not come within the scope of this statute.

The government argues in the alternative that the error complained of is not of sufficient substance to warrant reversal. In support of this argument the government points to the fact that the persons who made the memoranda were present at the trial and were tendered for cross-examination, that the memoranda were merely cumulative of other evidence properly in the record and that there was overwhelming evidence properly received of the defendant's guilt. Unquestionably these factors point to harmless error. See Finnegan v. United States, 8 Cir., 204 F.2d 105, certiorari denied 346 U.S. 821, 74 S. Ct. 36, 98 L. Ed. 347; United States v. Rosenberg, 2 Cir., 195 F.2d 583, certiorari denied 344 U.S. 838, 73 S. Ct. 20, 97 L. Ed. 687; United States v. Spadafora, 7 Cir., 181 F.2d 957, certiorari denied 340 U.S. 897, 71 S. Ct. 234, 95 L. Ed. 650. And under Rule 52(a), Federal Rules of Criminal Procedure, 18 U.S.C.A., we must disregard harmless error. The error in this case however was compounded by the fact that the jury was permitted, over objection by the defendant, to have the exhibits in the jury room during its deliberations. The jury thus had before it a neat condensation of the government's whole case against the defendant. The government's witnesses in effect accompanied the jury into the jury room. In these circumstances we cannot say that the error did not influence the jury, to the defendant's detriment, or had but very slight effect.

The second error complained of relates to an answer of a government witness to a question propounded by defense counsel on cross-examination. The trial judge ruled that the answer was not responsive and instructed the jury to disregard it but denied defendant's motion for a mistrial. Defendant contends that the answer was irreparably prejudicial and that no direction from the court, however strong, could eliminate the prejudice thereby created. We believe the error committed relating to the admission of the memoranda made by the narcotic agents is sufficiently prejudicial to warrant reversal and, since it is unlikely that the second alleged error will recur on a new trial or will occur under the same setting and circumstances, no useful purpose would be served by passing on it.

The judgment of the District Court is reversed and the cause remanded for further proceedings not inconsistent with this opinion.

SCHNACKENBERG, Circuit Judge (concurring).

The result reached by Judge Swaim is, in my opinion, right. I believe that, with one exception, the grounds stated are also right. I find myself unable to agree with the proposition that the memoranda made by the chemist were admissible under either 28 U.S.C.A. 1732 or 1733. I concede that the entries made by the chemist were admissible to prove the fact that he made analyses of the substances purchased from the defendant. However I cannot agree that under either 1732 or 1733 what the chemist ascertained in making these analyses is proved by the entries in question which in effect state that the substances analyzed were heroin. Inasmuch as the chemist who made the analyses and the entries in question failed to testify and submit himself to cross-examination, the defendant at the trial was helpless because he had no way to determine whether proper methods of analysis were used and were free from error in their execution. There has been no case cited to us so extending the effect of 1732 or 1733.

On the other hand, in Wheeler v. United States, 93 U.S.App.D.C. 159, 211 F.2d 19, certain slides were held to have been properly received in evidence where the prosecution proved by the testimony of a hospital interne that he had made smears on a slide and that the slide produced in court was of the type used for making smears at said hospital, that a slide shown him bore the letters of the patient's last name, and the date '8-11-51', although he could not say definitely that these were the slides in question. A bacteriologist then testified that these slides were received in the laboratory in accordance with established hospital procedures, that he made a microscopic examination of substances thereon, and he testified as to what that result was, and that the result of the examination was placed on the slides. Testimony of the interne and the bacteriologist demonstrated that the procedures followed were those regularly used by the hospital to make and keep a record of examinations of this kind. The court, 211 F.2d at page 22, said:

'Since Dr. Mangum was unable to say there had been sexual intercourse of the child about the time of the alleged assault, proof that she had been carnally assaulted rested substantially on the bacteriologist's testimony regarding the presence of sperm. That testimony, in turn, rested solely on the slides which were ultimately admitted in evidence over appellant's objection that they were not sufficiently identified by Dr. Mangum as those upon which he placed the smears taken from the child.'

The court distinguished the case from Novak v. District of Columbia, 82 U.S.App.D.C. 95, 160 F.2d 588.

That entries regularly made as a routine matter in laboratory notebooks cannot properly be held to prove themselves merely by identification by those who made the reports, is held in Teter v. Kearby, 169 F.2d 808, at page 816, 36 C.C.P.A., Patents, 706, where the court said:

'We do not find in the record that there is adequate evidence to prove that the type of fuel called for by the counts was produced. The tests necessary to show such production were not adequately proved. While there are data sheets in evidence bearing the signatures of those who conducted the tests, they were not called to testify; and we are unable to hold that there is any reasonable and certain testimony to establish that the process in issue here was carried out. The data sheets of themselves, not having been properly proved as to the accuracy of their contents, can not be held to be evidence of such contents.

'It is contended by appellant that the laboratory notebooks intended to support the testimony of appellant are admissible as evidence of performance of the acts recorded therein under the express provisions of the U.S. Code, title 28, Sec. 695, 28 U.S.C.A. 695.1 * * *

'We are of opinion that the code provision cited by appellant was never intended to embrace such records as are here before us. While it may be true that in the research department of appellant's company it may be a matter of ordinary business to file such reports, nevertheless the contents thereof can not be considered legally sufficient even though they are routine with respect to their contents. We do not think that the facts or operations related in those reports can be properly held to prove themselves merely by identification by those who made the reports. * * *'

To the same effect, see Hartzog v. United States, 4 Cir., 217 F.2d 706.

I can see no objection to applying 1732 to an entry of a fact which can be ascertained by mere observation by the person who makes the entry, such as the sale of a certain quantity of coal by a retail dealer, or the sale of a loaf of bread by a grocer. However, as to an entry made by a chemist indicating that he had made an analysis of a substance, while I believe that 1732 bestows admissibility on the entry as tending to prove that the chemist had made an analysis, I do not believe that the method of, or result reached in, the analysis as indicated by the entry, can be proved by mere production of the record of the entry. The testimony of the chemist is necessary. Such testimony was not presented by the government in the case at bar.