We must decide whether the Food and Drug Administration (FDA) acted lawfully in concluding that Indian shrimp contaminated with poisonous bacteria was "adulterated" within the meaning of the Food, Drug and Cosmetic Act (FDCA or Act). The district court answered yes. We affirm.

The Act permits the FDA to prohibit the importation of food that "appears" to be adulterated.1 In defining "adulterated," the Act focuses on the origin and dangers of "poisonous or deleterious substance(s)" in the food.2 When such substances are "added" to the food, the food is adulterated if the substance "may render it injurious to health."3 When, on the other hand, the substance is not added, the food is considered adulterated only if the substance would "ordinarily render it injurious to health."4

Beginning in late 1978, the FDA observed that a significant proportion of shrimp arriving from India contained salmonella. Without proper cooking and storage, food containing that bacteria can cause salmonellosis, a communicable disease that produces a number of serious bodily impairments and sometimes results in death.5 After officials of the Indian government acknowledged that insanitary processing facilities were responsible for the high rate of contamination of Indian shrimp, the FDA decided to sample all such shrimp arriving in the United States.6

In the spring of 1979, appellants offered two lots of raw, frozen Indian shrimp for import into the United States. The FDA sampled the shrimp and discovered salmonella in both lots. The importers each received a Notice of Detention and Hearing stating that the lots of shrimp would be refused admission because they "appear( ) to contain a poisonous deleterious substance (Salmonella species)."7 Appellants were afforded a hearing at which they tried to show that the shipments were not adulterated. After reviewing the evidence offered by the importers, the FDA issued a Notice of Refusal of Admission. The Notice said that the shrimp appeared to be adulterated and ordered the lots destroyed unless they were exported within ninety days. Appellants received permission to retain the shrimp in frozen storage pending the outcome of this litigation.

In their complaint in the district court, appellants contended that the import prohibition (1) was based on an incorrect interpretation of the Act's standard for "adulterated," (2) was unsupported by evidence in the record, and (3) reflected discriminatory treatment of shrimp in comparison to other foods. Based on a review of affidavits, expert testimony, and prior authority,8 the district court rejected all three arguments and granted summary judgment in favor of the FDA.

Appellants concede that the two lots of shrimp at issue contain salmonella. The FDA, in turn, acknowledges that salmonella is not "ordinarily injurious" since its dangers can be averted through proper cooking and storage. The only issue before us,9 then, is whether the FDA properly concluded that the shrimp met the more relaxed standard of adulteration for "added" substances. Specifically, we must evaluate the FDA's determinations that (A) salmonella was "added" to the Indian shrimp, and (B) that the bacteria "may render" the shrimp unhealthy.10

A. Is salmonella "added" to shrimp?

Appellants argue that this question was conclusively answered in the negative by our decision in American Public Health Association v. Butz ("APHA"), 511 F.2d 331 (D.C.Cir.1974). The plaintiffs in that case contended that inspection labels placed on domestic meat and poultry products by the Secretary of Agriculture were misleading since they failed to warn about the salmonella often found in those products. The court noted that inspection labels could be misleading under the Wholesome Meat Act if salmonella constituted adulteration within the meaning of that statute. Assuming that salmonella "may be inherent in the meat," the court commented: "(W)e think that the presence of salmonella in meat does not constitute adulteration within (the Wholesome Meat Acts') definition." Id. at 334. Appellants claim this remark amounted to a judicial determination that salmonella does not adulterate food.

The observation in APHA on which appellants rely consists of dictum uttered in a different legal and factual context. The issue presented by that case was whether the inspection labels were misleading, not whether the food was adulterated. In upholding the Secretary, the court relied on its view that consumers understand the importance of proper storage and cooking of meat and poultry as well as its conclusion that Congress did not intend "microscopic examination(s)" of domestic food products by the Department of Agriculture. Id. at 334-35. The court offered no evidentiary support for its additional suggestion that salmonella may not be "added" to meat. Moreover, the court's assumption about the inherent qualities of meat and poultry obviously do not bind the FDA's evaluation of shrimp.11 Finally, while the statutes involved in APHA define adulteration in terms similar to the FDCA,12 the government enjoys substantially greater discretion in preventing the import of apparently adulterated foods than it does in labelling domestic products.13

Accordingly, we must determine whether the FDA's conclusion that salmonella is an "added" substance in shrimp is consistent with the provisions of the FDCA. The parties urge different interpretations of the word "added." A regulation promulgated by the FDA in 1977 defines the word to include any substance that is not an inherent, natural constituent of the food.14 Appellants, on the other hand, argue that "added" describes only those substances which are present due to human intervention. Citing cases that have employed that definition,15 appellants contend that the FDA was required to demonstrate that human intervention introduced salmonella to the shrimp at issue here. The district court declined to resolve this definitional dispute. Instead, the court held that the FDA could have deemed the salmonella "added" under either standard.

Appellants' contention that "added" refers only to substances introduced by humans was recently adopted by the Fifth Circuit in United States v. Anderson Seafoods, Inc., 622 F.2d 157 (1980). The court there concluded that the more relaxed standard for adulteration "should apply to all of a toxic substance present in a food when any of that substance is shown to have been introduced by man." Id. at 161. The court further held that a showing of human intervention could be based on general scientific knowledge about the origin of the additive in question. See id. at 162. Anderson Seafoods thus implicitly rejected the notion advanced by appellants here that the FDA must demonstrate human intervention with respect to the particular articles of food it wishes to regulate. We reject the argument explicitly. In light of the FDA's broad authority to prohibit import of any food that "appears" to be adulterated, the agency need not prove that substances present in a particular lot were introduced by man.16

We agree with the district court that the FDA's decision in this case satisfied appellants' definition of "added" as elaborated in Anderson Seafoods. The evidence offered by both parties indicated that salmonella in shrimp can result from human acts and can frequently be attributed to insanitary processing procedures.17 The FDA had ample reason to believe that the salmonella frequently observed in Indian shrimp in 1978 and 1979 resulted from insanitary conditions. Representatives of the Indian government acknowledged to the FDA that the salmonella contamination was a result of unclean handling.18 And several months before the administrative hearing in this case, FDA officials visited processing facilities in India and determined the nature of conditions there first hand. The district court described what the delegation saw:

(T)he delegation observed unscreened, fly-infested processing areas; inadequate icing of shrimp subjected to temperatures in excess of 90o F.; the placement of porous shrimp-holding pans on heavily-travelled floors; and the use of pitted and cracked work surfaces covered with the residue of fecal matter from previously processed frogs. At a shrimp landing area, FDA officials observed hundreds of people, mostly barefoot and dressed in soiled clothing, milling around the dock where the shrimp were unloaded; poor icing of shrimp; and the use of bamboo baskets, which are virtually impossible to sanitize, to collect shrimp.19

Together with the FDA's understanding of the cause of salmonella in shrimp generally, these observations provided more than enough support for the agency's conclusions that the bacteria had been at least partly introduced by man.

B. Did the FDA show that salmonella "may render" shrimp unhealthy?

Appellants claim the evidence does not support the FDA's determination that salmonella "may render (shrimp) injurious to health." This complaint is groundless. The record indicates that ingestion of even small amounts of the bacteria can cause serious cases of food poisoning in some people.20 Appellants do not dispute the medical evidence; indeed, their own expert witness confirmed the danger.21 Instead, they rely on (1) the fact that few, if any, cases of salmonellosis traced to shrimp have been reported and (2) their expectation that consumers will properly cook and store the shrimp and thus prevent contraction of the disease.

The FDA offered several reasonable explanations for the low number of reported cases of salmonellosis from shrimp: The sources of information on which appellants rely report only a small proportion of salmonella cases; reporting of "home episodes" is rare; specific cases may often be attributed to the wrong sources; and shrimp products are consumed in much smaller quantities than other foods which may contain salmonella.22 These explanations were supported by the record and adopted by the district court.23 Accordingly, we agree with the court that the "absence of documentation does not foreclose the (FDA's) discretion to determine that salmonella in shrimp may be injurious to the health of those who consume it."24

Nor are we comforted by appellants' sanguine assurances that consumers will properly cook and store the shrimp. The FDA has authority to ban contaminated articles from import notwithstanding promises that the deleterious condition will be corrected. See United States v. 52 Drums of Maple Syrup, 110 F.2d 914, 915 (2d Cir. 1940). Moreover, there was evidence in the record that many people either do not cook shrimp properly25 or, like the patrons of Japanese restaurants, eat it raw.26 Under these circumstances, the FDA was well within its authority in concluding that salmonella "may render" shrimp injurious to health.

Affirmed.