The condemnation of several parcels of the poultry medications “Myconox” and “Ferro-Lac” generated this appeal. Pursuant to libel filed under 21 U.S.C. § 301 et seq., the products were seized in May, 1966. Subsequently, the appellant[*1128] Naremco, Inc., filed claim and answer. The cases were consolidated. Upon a jury verdict there was judgment that the seized products be destroyed as not in compliance with the Federal Food, Drug, and Cosmetic Act. We affirm.

It was alleged that the medicated feed Ferro-Lac was a food as defined in 21 U.S.C. § 321(f) ^[1] which was adulterated within the meaning of 21 U.S.C. § 342(a) (2) (C) ^[2] in that Ferro-Lac contained a combination of methylrosaniline chloride, sodium porpionate and sodium phthalylsulfacetamide. Further, it was asserted that Myconox contained a combination of methylrosaniline chloride and sodium propionate which were unsafe food additives within the meaning of 21 U.S.C. § 348 ^[3] since the intended use of the chemical combinations was not in conformity with an exemption or regulation in effect pursuant to 21 U.S.C. § 348.

It was additionally alleged that both Ferro-Lac and Myconox were misbrand-ed when introduced into and while in interstate commerce in violation of 21 U. S.C. § 352(a) ^[4] in that the labeling represented that Ferro-Lac was effective in the treatment of enteric infections associated with non-specific disease in poultry, and that Myconox was effective in the treatment of mycotic infections and that these statements are false and misleading since neither is effective for the stated purpose.

Finally, there was an allegation that Ferro-Lac and Myconox were drugs, as defined by 21 U.S.C. § 321(g), and “new drugs” within the meaning of 21 U.S.C. § 321 (p). In the terms of the statute a new drug is

“Any drug * * * the composition of which is such that such drug is not generally recognized, among experts qualified by scientific training and experience to evaluate the safety and effectiveness of drugs, as safe and effective for use under the conditions prescribed, recommended, or suggested in the labeling thereof * *

The jury found that Myconox was misbranded, was a new drug, and was a food additive. As to Ferro-Lac the jury found in favor of the government only on the food additive charge, which alone is sufficient to support condemnation.

Predictably, the evidence as to the qualities of the constituent drugs had to be “expert opinion”. A Food and Drug Administration veterinarian testified that he fed directed dosages of Ferro-Lac to certain chickens. At the end of this experiment the chickens were killed and certain parts were forwarded to a government chemist, who conducted tissue or chemical residue tests. Although these studies were limited to only one of the ingredients of Ferro-Lac, the chemist did find sulfanilamide, a breakdown product of the ingredient phthalyl-sulfacetamide, in tissues of chickens[*1129] which had been fed Ferro-Lac at the dosage level of twenty pounds per ton.

Having followed up the identification of these chemical residues, Dr. James Dollahite, Texas A. & M. veterinarian and toxicologist, testified that there was no existing scientific information concerning the safety of the combination of chemicals in Ferro-Lac, nor was there any scientific information concerning the safety of sodium phthalylsulfaceta-mide. Consequently he could not say how the three chemicals would act in combination or whether they would be safe.

On the food additive question testimony bearing on the general recognition for safety among experts came from Dr. Maurice Cover of the University of Delaware, Dr. Walter Gross of VPI, and Dr. Benjamin M. Pomeroy. Each testified that he kept abreast of veterinary pathology through professional meetings, colloquia, and constant review of the literature. In substance their testimony was that the particular combinations of chemicals at issue were not generally recognized as having been shown to be safe, since the available scientific literature was silent on the ingredients in Ferro-Lac and Myconox. Substantially identical testimony relevant to the new drug issue was elicited from both Drs. Cover and Pomeroy. In sum, these experts averred that Myconox was not generally recognized among qualified experts as safe and effective for its recommended use; that there was a total absence of published information on the subject.

Two persons, Dr. S. W. Coleman, a commercial poultryman, and Dr. D. C. O’Meara, a University of Maine professor, testified on the effectiveness (i. e., misbranding) of Myconox. Resolution of this question turned on the meaning of “mycosis”. The claimant insisted that the term referred only to digestive difficulties while the government’s expert witnesses unanimously agreed that mycosis meant any fungus infection. At any rate, Dr. Coleman testified that he abandoned Myconox treatment for digestive symptoms when his birds developed respiratory symptoms as well. Dr. O’Meara, on the other hand, stated that five tests he had run using Myconox to treat the mycotic infection of aspergil-losis had been successful.

The claimant’s evidence was largely devoted to the claimed actual effectiveness of Ferro-Lac and Myconox. No evidence was presented on either the food additive or new drug issues. All but two of the claimant's witnesses testified as lay witnesses that mycosis was understood in the poultry industry to be a fungus infection of the digestive tract and that Myconox was effective in treating it. Claimant’s chief expert witness, Dr. Emil Lorz, testified that the ingredients of both Myconox and Ferro-Lac were no more harmful in combination than they were singly.

Also introduced into evidence as claimant’s Exhibit 1 was a letter written April 13, 1961, by Mr. George P. Lar-rick, who was at that time the Commissioner of Food and Drugs. The most significant portions of that letter were as follows:

“The ‘Ferro-Lac’ and ‘TSC’ products sponsored by Mr. Scott are not new drugs and are not subject to the certification regulation; they fall into the category of drugs which may be distributed entirely on the responsibility of the firm.”

The claimant made appropriate motions for a directed verdict and judgment N.O.V. at trial and here asserts that there was insufficient evidence to support the verdict for the government. Our question, then, is whether there was substantial evidence to support the verdict. Hogan v. United States, 5 Cir., 1963, 325 F.2d 276; United States v. Simmons, 5 Cir., 1965, 346 F.2d 213, 218; Helene Curtis Industries, Inc. v. Pruitt, 5 Cir., 1965, 385 F.2d 841, 850; Boeing v. Shipman, 5 Cir., 1969, 411 F. 2d 365.

The thrust of claimant’s argument is that the statutory definition of food additive in 21 U.S.C. § 321 (s) contains four elements: The substance must be (1) “not generally recognized, among ex[*1130] perts qualified by scientific training and experience, (2) to evaluate its safety, (3) as having been adequately shown through scientific procedures * * * (4) to be safe under the conditions of its intended use * * *.”

A. Ferro-Lac — Food Additive

Since no witness was asked a question which embodied each of the four statutory elements, appellant argues that the probative expert testimony of the government cannot amount to the required substantiality. For example the testimony of Dr. Cover shows the following exchange:

“Q: Will you state whether you have an opinion as to whether the combination of these drugs [1] is generally recognized among experts qualified by scientific training and experience [3] as having been adequately shown through scientific procedures [4] to be safe for use as an aid in reducing non-specific enteritis in chickens and turkeys ?

“A: There is no such information that it is safe.”

The fact that the qualification [2] to evaluate its safety was omitted, appellant argues, means that the class of experts is not sufficiently delimited. Having in mind the reminder of the United States Supreme Court in United States v. Dotterweich, 320 U.S. 277, 64 S.Ct. 134, 88 L.Ed. 48, rehearing denied 320 U.S. 815, 64 S.Ct. 367, 88 L.Ed. 492 (1943), that the Food and Drug Act is a working instrument of government and not a collection of English words, we do not find appellant’s argument to be decisive. The government’s questions very clearly inquired into the safety of the product in the opinion of qualified experts. Thus, component number two of the statute is clearly implied.

B. Myconox — Food Additive

Likewise, claimant asserts that elements [1] and [2] in the statutory definition are absent in this interrogation of Dr. Cover:

“Q: Based on your knowledge and experience, and in your review of scientific literature and the colleagues you have consulted with, do you have an opinion as to whether the combination of sodium propionate and methyl-rosaniline chloride [3] has adequately been shown through scientific procedures [4] to be safe for use by adding it to the feed of broilers and market turkeys for the control of mycosis? “A: I know of no such information.”

Attack on this question as reflecting only the personal view of Dr. Cover is unavailing. The absence of scientific knowledge on the part of an expert and his colleagues is sufficient to show lack of general recognition of safety. United States v. Article of Drug etc., (N.D.Ga., 1968) , 294 F.Supp. 1307, affirmed per curiam 415 F.2d 390 (decided July 25, 1969) . As in (A) element number two may be implied.

C. Myconox — A New Drug

Alleged discrepancies between the “new drug” statutory definition and the questions posed to witnesses are once again the main complaint, and do not differ in kind from those discussed above. Here claimant also attacks the testimony of the government’s expert witnesses because they were not qualified in areas required by the statute.

The government’s witnesses are all veterinarians specializing in avian diseases. Even assuming that the witnesses are not qualified to judge the safety of the products, they are qualified to testify as to whether there is general recognition of the safety of the products. This is true because the absence of literature establishing safety is proof that a product is not generally recognized. United States v. Article of Drug, etc., supra, 294 F.Supp. at 1307, 1311. In addition, no objection was made to the competency of the witnesses in the court below. The expert qualification of a witness is a question for the trial judge, whose decision is conclusive unless clearly erroneous as a matter of[*1131] law. St. Joe Paper Company v. United States, 5 Cir., 1946, 155 F.2d 93.

D. Misbranding of Myconox

Misbranding of Myconox turns on whether mycosis refers to fungal infections in general or encompasses only digestive disorders. Conceding a conflict in the evidence, there was, at most, a question of fact for the jury.

Claimant moved to limit evidence concerning tissue residue tests on the ingredient sulfanilamide in Ferro-Lac until the government showed that the feeding of the combination of ingredients was less safe than the residues which would have resulted from feeding the ingredients individually. The trial court denied this motion and also denied an instruction to the jury to the effect that either of the substances in Ferro-Lac or My-conox could not be food additives unless the ingestion of the ingredients in combination was less safe than the individual ingredients.

Taking up the government’s concession that no attack was being made on the ingredients individually, the claimant stresses the immateriality of the presence of sulfanilamide unless the government shows that the residue would not have been found but for its combination methylrosaniline. chloride and sodium propionate.

Claimant’s arguments misconceive both the language of 21 U.S.C. § 321 (s) ^[5] and the government’s method of meeting its burden of proof. It does not matter under the Act if Ferro-Lac is in fact safe or whether the residues would be different if the individual ingredients were tested. The sole criterion for identifying a food additive is whether a substance which may become a component of or affect the characteristics of any food be not generally recognized among qualified experts as having been shown to be safe, 21 U.S.C. § 321 (s).

When this evidence was introduced the government had already shown that Ferro-Lac was not generally recognized as safe. The tissue residue evidence was directed to showing that sodium phthaly-sulfacetamide became a component of food for man (in line with the statutory definition of a food additive). If it is true, as the claimant alleges, that the introduction of this evidence was a scare tactic, the proper corrective device would have been a request for a precautionary jury instruction.

The use made of Commissioner Larrick’s letter at trial constitutes claimant’s next argument for reversal. The jury was instructed that they were to use the Commissioner’s forgiveness of the prior clearance requirement, 21 U.S. C. § 348(a) (2) only with respect to Ferro-Lac. Claimant challenges this instruction on the ground that Myconox is the same formulation as Ferro-Lac Feedmix with additional amounts of two of the active ingredients. The government supports the instruction on the theory that the change in formulation demonstrates that Myconox is a different drug.

The courts have been reluctant to give too narrow a construction to this statute since it touches public health. AMP, Inc. v. Gardner, 2 Cir., 1968, 389 F.2d 825, 830, cert. denied 393 U.S. 825, 89 S.Ct. 86, 21 L.Ed.2d 95. Thus it is sound policy to say that adding different amounts of two active ingredients might make the resulting formulation a “new drug” under 21 U.S.C. § 321 (p). In United States v. Articles of Drug Labeled in part Quick-o-ver, (D.Md., 1967), 274 F.Supp. 443, the Court made indi[*1132] vidual findings on the new drug issue as to each of the four variations and found that three were new drugs because one ingredient was dropped and another ingredient increased. Where the government can show that differing quantities of the active ingredients are present, an inference may arise that the drug is new. Then the opposing party can rebut the inference by showing that there is no material difference in the two formulations as far as general recognition for safety by experts is concerned. Here the claimant failed to make such a showing.

There is no legitimate basis in the record for the grant of a new trial based on newly discovered evidence. At trial Dr. James Dollahite, a government witness, testified that one-third of the persons taking sulfanilamide develop a sensitivity to any of the sulfa drugs and that further use of sulfa drugs in such a person might produce death. In conclusion, Dr. Dollahite claimed that a bird with enteritis (for which Ferro-L'ae might be prescribed) would absorb more of the drug than a normal bird. In support of the motion claimant proffered an affidavit by Dr. Omer E. Hagebush challenging Dr. Dollahite’s scientific conclusions.

In order to overturn a denial of a new trial we must find an abuse of discretion. Stilwell v. Travelers Insurance Company, 5 Cir., 1964, 327 F.2d 931, Newly discovered evidence must be evidence in existence of which a party was excusably ignorant, discovered after trial. In addition facts implying reasonable diligence must be provided by the movant. The evidence must be material, and not cumulative or impeaching, and it must be such as to require a different result. Chemical Delinting Company v. Jackson, 5 Cir., 1951, 193 F.2d 123.

On the due diligence ground claimant must fail. The assertion that Dr. Hagebush’s affidavit is backed by thirty years of scientific research puts to rest any notion that the evidence is newly discovered. Also, in interrogatories claimant admitted to making residue tests on the constituents of its products. Thus it cannot claim that it had no reason to anticipate that the government might make such tests and present testimony about the results.

This evidence is merely impeaching in nature and does not constitute newly discovered evidence. United States Fidelity and Guaranty Company v. Lawrenson, 4 Cir., 1964, 334 F.2d 464, cert. denied 379 U.S. 869, 85 S.Ct. 141, 13 L.Ed.2d 71. F. W. Woolworth v. Seckinger, 5 Cir., 1942, 125 F.2d 97. Finally, the evidence is not such as would require a different result upon retrial, Tracy v. Terminal Railroad Association of St. Louis, 8 Cir., 1948, 170 F.2d 635, since disagreement among experts as to the general recognition of a drug as being safe only creates a fact question for the jury, United States v. Articles of Drugs Labeled in part Quick-o-ver, sup ra, at 448.

Counsel for the claimant has vigorously presented every reasonable possible basis for a reversal in this case. Nevertheless, we perceive no error and the judgment of the District Court must be

Affirmed.