Appellants, two New York City narcotics detectives, were convicted on trial to the jury in the United States District Court for the Southern District of New York, Milton Pollack, Judge, of themselves conspiring to engage and engaging in narcotics traffic in violation of 21 U.S.C. 173, 174 and 26 U.S.C. 4705(a) and 7237(b). We find error in permitting the use on trial of the results of scientific tests made for such use when timely disclosure of the tests had not been made, and reverse and remand for new trial.

The two defendants were New York City narcotics detectives. They were charged with five counts of narcotics violations arising out of events on May 19 and May 31, 1967. They allegedly retained some of the cocaine they seized in a previous raid (on one Toni Troy) and made arrangements with a double agent informer, Gonzalves, to sell it for them. Gonzalves was wired with a transmitter by inspectors of the United States Treasury Service when he met with defendants and carried out the alleged conversations with them. Gonzalves gave the narcotics he got from the defendants on the dates in question to the inspectors (and on another date, August 12, 1967, he retained some of the narcotics for himself, thus committing essentially the same offense, except one step deeper in the intrigue). Count One was conspiracy under 21 U.S.C. 173, 174, 26 U.S.C. 7237(b); Counts Two to Five were substantive counts of selling without an order form and possession under the above sections and 26 U.S.C. 4705(a).

January 19, 1968 defendants moved for discovery among other things of scientific tests such as fingerprints, handwriting and voice analysis, but not limited thereto. The government opposed this discovery as a fishing expedition. The government, however, agreed to make available reports on narcotics drugs. On February 16, 1968 Judge Wyatt denied the broad discovery sought but did permit inspection of the chemical analysis of the drugs in the indictment. On March 13, 1968 the government served a bill of particulars to the effect that they would produce reports of chemical analyses of narcotic drugs mentioned in the indictment on 72 hours' notice. In June, 1968 the seized drugs from the Troy raid and the drugs returned by Gonzalves were sent to Washington for tests, including neutron activation tests which tended to show that the drugs all came from the same original batch. The government did not inform the defendants of this test. They only became aware of it at the trial after the testimony of the prosecution's first witness, when the government produced its exhibits. Defendants objected to the neutron activation evidence upon Brunelle's testimony (Brunelle was the government expert on this technique) six days later, on December 16, 1968. The objection was overruled. Apparently no request for the chemical analyses under the government's bill of particulars was ever made. After the government rested its case, the defense made a motion for a month's continuance to carry out its own version of the neutron tests. This motion was denied.

Appellants' first arguments concern the use of the neutron activation evidence. They claim that the technique is new and uncertain and too complex at this stage for the jury to consider it. They say that its flashiness would command undue jury respect in the light of its scientific unreliability. They argue that, therefore, it should have been withheld from the jury. (The neutron tests had tended to show that the cocaine came from the same batch.) In United States v. Wolfson, 297 F. Supp. 881 (S.D.N.Y.1968) the trial court sitting without a jury admitted such tests into evidence but refused to put them in the balance when deciding the findings of fact. He did this because of the newness and the credible testimony of the defense that there had not been much experience with these tests. However, he considered there was sufficient evidence without the tests to find the facts the government sought to establish, so there was no need to press the issue. The government here claims that the defense expert of great repute (Dr. Jervis) in his testimony did not question the reliability of the neutron activation process but rather only attacked the procedure used by Brunelle and alleged that more had to be done to insure the reliability of his tests. If the test was indeed unreliable at this stage, then it should not have been given to the jury. But no reversal is required here on that ground; a strong showing of unreliability must be made in the trial court. Here the government points out that Dr. Jervis affirmed the reliability of the process itself. This is, of course, a typical question for the jury to decide-- the probity of the government's expert evidence as attacked in its application by the defense's expert.

Appellants also contend, and with reason, that when the government took these new tests after the discovery attempt in January and February, 1968 the government had a positive duty to disclose the results or at least the fact that they had taken them. This is so especially in light of the fact that the government had opposed discovery on the grounds that the request was not particular enough, and now the government alone had knowledge of the particular tests it had taken. Indeed, it is important that the defense be given a chance to research the techniques and results of scientific tests taken by the government. And Rule 16(g) Fed.R.Crim.P. indicates that the prosecution had a continuing duty under the February, 1968 order to come forward with new scientific tests it made. See 8 Moore Fed. Practice 2d Ed. 16.05(3). The government argues that the tests in that February order were in regard to voice analysis, etc., and not chemical analysis of the drugs and hence did not comprehend the neutron analysis. This seems unduly technical and not in accord with the spirit of the rules.

The government had quite substantial evidence from which the jury might well have found that the narcotics turned over by Gonzalves and those delivered by Kelly and Imp to the Police Department came from the same batch. It chose to bolster this already quite strong case by a concededly new and, to any trier, quite dramatic demonstration of a method of determining trace elements in a substance by bombardment of the substance in an atomic pile and identification of the trace elements by measurement of the half life of each element over a period of time. While the newness of the test is not itself reason for depriving the jury of its results, and the opportunity to weigh conflicting claims as to its reliability, fairness requires that adequate notice be given the defense to check the findings and conclusions of the government's experts. In the light of the testimony of Dr. Jervis, an acknowledged expert in the field, that some three weeks at least were required for testing and evaluation, the failure to disclose the tests well in advance of trial faced the trial judge with an impossible choice, to sanction a month's interruption of a jury trial or deprive the defense of a fair opportunity to meet this part of the government's evidence.

The determination not to abort the trial, which had already consumed much time and expense, is understandable in the light of the strength of the government's case, but we think in the circumstances here the other choice was the proper one. The course of the government smacks too much of a trial by ambush, in violation of the spirit of the rules. A new trial is required, with a fair opportunity for the defense to run its own neutron activation tests of the material to determine the atomic similarity or dissimilarity of the trace elements in the samples.1

Reversed and remanded for new trial.

On Petition to Supplement the Record and for Rehearing.

PER CURIAM.

The United States moves to supplement the record and for rehearing on the ground that the court's opinion was founded on a misapprehension of the facts as to notice of government tests involved. We have considered the material submitted and conclude that although defendants may have learned of the test a day earlier than our opinion indicated this makes no material difference in the factors considered.

The petition to supplement the record is granted. The petition for rehearing is denied.