United States v. General Foods Corp., 446 F. Supp. 740 (N.D.N.Y 1978)

*741 *742 *743 Paul V. French, U. S. Atty., Syracuse, N. Y., for plaintiff; Joseph R. Mathews, Asst. U.S. Atty., Syracuse, N. Y., Food and Drug Administration, Richard M. Cooper, Chief Counsel, Robert P. Bardy, Associate Chief Counsel for Enforcement, Michael R. Taylor, Asst. Chief Counsel for Enforcement, Rockville, Md., of counsel.

Burditt & Calkins, Chicago, Ill., for defendants; George M. Burditt, Arnold Kanter, Chicago, Ill., Bond, Schoeneck & King, James E. Wilber, John J. Dee, Syracuse, N. Y., General Foods Corp., Murray D. Sayer, House Counsel, White Plains, N. Y., of counsel.

MUNSON, District Judge.

This is an action commenced by the United States, through the Food and Drug Administration (F.D.A.), for violation of section 301 of the Food, Drug, and Cosmetic Act, 21 U.S.C. § 331^[1]; jurisdiction being alleged under section 302 of that Act, 21 U.S.C. § 332.^[2] The Government is primarily seeking to enjoin the shipment of frozen, french-style green beans^[3] processed and packaged by defendant General Foods Corporation at its Fulton, New York facility, between August 23, 1977 and September 10, 1977, inclusively.^[4] Also named as defendants are Ross Barzelay, President of General Foods, Edward R. Fencl, General Foods Business Manager for Fruits and Vegetables in the Food Products Division, David E. James, Director of Quality Assurance, Environmental Control and Occupations Safety for General Foods, and James W. Beno, Facility Manager for General Foods' Fulton, New York processing plant.

The Government claims that the defendants have violated the Food, Drug, and Cosmetic Act in two separate and distinct fashions. The Complaint alleges that the beans in issue are adulterated under section 402(a) (3) of the Act, 21 U.S.C. § 342(a) (3) [hereinafter referred to as (a) (3)], whereby a food is adulterated "if it consists in whole or in part of any filthy, putrid, or decomposed substance . . .." The Government also contends that the beans are adulterated within the meaning of section 402(a) (4), 21 U.S.C. § 342(a) (4) [hereinafter simply (a) (4)], which provides that a food is adulterated "if it has been prepared, packed, or held under insanitary conditions whereby it may have become contaminated with filth . . .."

The (a) (3) charge of adulteration stems from analyses performed upon five samples *744 of finished product, frozen french-style green beans, three of which were taken by F.D.A. inspectors during an inspection of the Fulton facility on August 23-25, 1977, the others having been confiscated during a September 9-10, 1977 inspection. Microscopic examinations of those samples revealed the presence in all five of Geotrichum mold, commonly known as machinery mold.

The (a) (4) violation relates to observations made by F.D.A. inspectors during the August and September inspections. In particular, the (a) (4) charge relates to a visible buildup of green, slime-like vegetable matter on belts and equipment contained within defendants' Fulton processing plant. Samples of that slime taken on two separate occasions revealed the presence of Geotrichum, or machinery mold. The (a) (4) portion of the Complaint also contains allegations of other insanitary plant conditions, including inspectors' observations of house and fruit flies within the Fulton facility.

The defendants have moved for dismissal of the Complaint, both under Rule 12(b) (1) of the Federal Rules of Civil Procedure, Title 28 U.S.C., claiming that this Court lacks subject matter jurisdiction and Rule 12(b) (6), claiming that the Complaint fails to state a claim upon which relief can be granted. The gravamen of the defendants' arguments in support of dismissal concerns the failure of the F.D.A. to establish defect action levels (DAL), or tolerances, with regard to Geotrichum mold, to serve as guidelines to the food processing industry. The defendants claim that, because the presence of Geotrichum in processed frozen green beans is inevitable and unavoidable to a certain degree, the F.D.A. must establish a DAL below which the presence of Geotrichum in frozen green beans will be tolerated by the F.D.A. under section 306 of the Act, 21 U.S.C. § 336. The failure of the F.D.A. to establish such a tolerance, the defendants claim, should result in the Government's not being able to establish good cause for granting the injunction, thereby depriving this Court of subject matter jurisdiction. 21 U.S.C. § 332(a). The defendants also contend that such failure to establish DALs for Geotrichum renders F.D.A. actions, such as this, for (a) (3) and (a) (4) violations, based upon the presence of Geotrichum, arbitrary and capricious, thereby requiring that this Court deny injunctive relief based upon the equitable "clean hands doctrine."^[5]

As previously stated, the green beans in issue are adulterated, under (a) (3), if they consist in whole or part of any filthy substance. The courts which have had occasion to interpret the word "filthy", as used in that section, have been unanimous in applying a common, everyday definition, rather than attempting to define it in a scientific or technical sense. See, e. g. United States v. Cassaro, Inc., 443 F.2d 153 (1st Cir. 1971); United States v. 44 Cases, Etc., 101 F. Supp. 658 (E.D.Ill.1951); United States v. Swift & Co., 53 F. Supp. 1018 (M.D. Ga.1943). Under such a definition, there can be no question, and the defendants do not assert otherwise, that Geotrichum mold fragments constitute filth. See United States v. Swift & Co., supra.

Under the terms of the statute, an (a) (3) adulteration violation is proven once the presence in the food of any quantity of filth is established. United States v. 900 Cases, Etc., Peaches, 390 F. Supp. 1006 (E.D. N.Y.1975); cf. Dean Rubber Manufacturing Company v. United States, 356 F.2d 161 (8th Cir. 1966); also cf. A. O. Andersen & Co. v. United States, 284 F. 542 (9th Cir. 1922). Moreover, to prove adulteration under (a) (3), it is not necessary for the Government to demonstrate that the food is injurious or unfit for consumption, United *745 States v. Cassaro, Inc., supra; United States v. 484 Bags, More or Less, 423 F.2d 839 (5th Cir. 1970); United States v. 449 Cases, Containing Tomato Paste, 212 F.2d 567 (2d Cir. 1954).

The problem which arises with regard to adulteration under (a) (3), as well as other portions of the Food, Drug, and Cosmetic Act, is brought into sharp focus by the facts of this case, together with the defendants' legal arguments in support of their motion to dismiss. There can be no question that an expert, armed with the proper equipment, could detect filth in virtually every food marketed. United States v. 484 Bags, More or Less, supra; United States v. 1,200 Cans, Pasteurized Whole Eggs, Etc., 339 F. Supp. 131 (N.D.Ga.1972); United States v. Capital City Foods, Inc., 345 F. Supp. 277 (D.N.D.1972). The F.D.A., in fact, seemingly recognizes this as being true, as is evidenced by an article, cited by the defendants, which appeared in the Federal Register on September 30, 1977. 42 Fed.Reg. 52,814 (1977). Consequently, strict enforcement, by the F.D.A., of section 402 of the Act would result in the banning of literally all processed foods. United States v. 484 Bags, More or Less, supra.

In order to ameliorate the hardships which would result from such a literal interpretation, Congress empowered the F.D.A. to use its discretion and decline prosecution for minor violations. 21 U.S.C. § 336; United States v. 484 Bags, More or Less, supra; United States v. 449 Cases, Containing Tomato Paste, supra. The defendants contend that the failure of the F.D.A. to set defect action levels, in order to alert the industry as to under what conditions an (a) (3) violation would be prosecuted, based upon the presence of Geotrichum mold, is unfair, and should result in a dismissal of the present case. In support of this argument, the defendants rely heavily upon language contained in Judge Frank's dissent in United States v. 449 Cases, Containing Tomato Paste, supra, and upon the Seventh Circuit Court of Appeals' opinion in United States v. 1,500 Cases More or Less, Etc., 236 F.2d 208 (7th Cir. 1956).

Despite the defendants' assertions to the contrary, the F.D.A. is not statutorily required to establish tolerances, or DALs, with respect to (a) (3) violations. United States v. Certified Grocers Co-Op., 546 F.2d 1308, 1311 fn. 5 (7th Cir. 1976); United States v. Hunter Pharmacy, Inc., 213 F. Supp. 323 (S.D.N.Y.1963); United States v. 233 Tins, More or Less, Etc., 175 F. Supp. 694 (W.D.Ark.1959). While Congress has apparently mandated that the F.D.A. establish tolerances for poisonous or deleterious substances which are required or unavoidable in the production of food, such regulations relating to adulteration under § 402(a) (2) of the Act, 21 U.S.C. § 342(a) (2), see 21 U.S.C. § 346, there is no like requirement relative to (a) (3) violations.^[6] That is not to say that the F.D.A. does not have the authority to enact such DALs with respect to (a) (3) violations, whether under 21 U.S.C. § 336, United States v. 484 Bags, More or Less, supra; cf. United States v. Goodman, 486 F.2d 847 (7th Cir. 1973), 21 U.S.C. § 341, United States v. 1,200 Cans, Pasteurized Whole Eggs, Etc., supra, or pursuant to 21 U.S.C. § 371(a). United States v. Nova Scotia Foods Products Corp., 568 F.2d 240 (2d Cir. 1977). But see United States v. 449 Cases, Containing Tomato Paste, supra.

The mere enactment of regulations establishing tolerances for filth in foods, below which levels the F.D.A. will not prosecute, does not always alleviate the court's task of deciding whether an (a) (3) violation has occurred. The courts are not bound by such DALs, and may well choose to apply a stricter standard than that set by the *746 F.D.A. in any particular instance. United States v. 484 Bags, More or Less, supra at 842; United States v. 449 Cases, Containing Tomato Paste, supra at 575; see also Dean Rubber Manufacturing Company v. United States, supra [dictum]. In fact, a court is not empowered (in theory) to overlook any (a) (3) violation, regardless of what tolerance, if any, has been set by the F.D.A. United States v. 449 Cases, Containing Tomato Paste, supra.

The harshness of this result is somewhat lessened by the application of a de minimis doctrine, whereby small quantities of filth can be overlooked by a court, especially in circumstances where no applicable DAL is in effect and there is no evidence that that quantity found is avoidable through the use of good manufacturing practice, taking into account the state of the industry. United States v. 1,500 Cases More or Less, Etc., supra at 215; United States v. 449 Cases, Containing Tomato Paste, supra; United States v. 900 Cases, Etc., Peaches, supra; United States v. Capital City Foods, Inc., supra at 278-279; United States v. 233 Tins, More or Less, Etc., supra at 702; see also United States v. 133 Cases of Tomato Paste, 22 F. Supp. 515 (E.D.Pa.1938), decided under the Section 7 of the old Food and Drug Act, formerly 21 U.S.C. § 8. In applying a de minimis standard, a court must, of necessity, rely heavily upon the testimony of expert witnesses well acquainted with both the food and foreign substance at issue, as well as the state of the industry. See, e. g., United States v. 1,500 Cases, More or Less, Etc., supra at 212. For this reason, it would seem that a motion for dismissal of an (a) (3) violation which is arguably de minimis should be denied, so that the Court can have the benefit of such witnesses' expertise in applying the de minimis doctrine.

As previously noted, the defendants argue that it is unfair to allow the F.D.A. to indiscriminately exercise its discretion under 21 U.S.C. § 336 in determining when or when not to prosecute for (a) (3) violations, in particular relating to Geotrichum mold, which they assert is an unavoidable component of processed green beans. This argument, one proferred in the context of the equitable clean hands doctrine, could easily be considered as a Constitutional vagueness argument, grounded in the Due Process clause of the Fourteenth Amendment. Cf. Boyce Motor Lines v. United States, 342 U.S. 337, 72 S. Ct. 329, 96 L. Ed. 367 (1952). The vagueness argument fails, however, by virtue of the fact that (a) (3) very clearly prohibits all filthy, putrid or decomposed substances contained in food, and is therefore far from vague; Dean Rubber Manufacturing Company v. United States, supra; see also United States v. Thriftimart, Inc., 429 F.2d 1006 (9th Cir. 1970), cert. den. 400 U.S. 926, 91 S. Ct. 188, 27 L. Ed. 2d 185 (1970), reh. den. 400 U.S. 1002, 91 S. Ct. 453, 27 L. Ed. 2d 454 (1971). The equitable argument based upon the unfairness of the F.D.A. charging (a) (3) violations, where no DALs exist to place the industry on notice as to what levels are acceptable, is somewhat more appealing, and has in fact been accepted, to varying degrees, by some. United States v. 449 Cases, Containing Tomato Paste, supra at 575 (Frank, J., dissenting); United States v. 1,500 Cases, More or Less, Etc., supra at 211. The Second Circuit Court of Appeals majority implicitly rejected this equitable argument in United States v. 449 Cases, Containing Tomato Paste, supra, a case in which Judge Frank issued a dissenting opinion explicitly accepting the position now espoused by the defendants. See United States v. 900 Cases, Etc., Peaches, supra at 1011. When weighed against the public health and welfare sought to be protected by Congress through enactment of the Food, Drug, and Cosmetic Act, United States v. Diapulse Corporation of America, 457 F.2d 25 (2d Cir. 1972), the unfairness to the industry is minimized and should not bar enforcement of the Act.

This Court therefore denies the defendants' motion to dismiss, finding that this Court does possess subject matter jurisdiction, by virtue of 21 U.S.C. § 332(a), and rejecting defendants' equitable clean hands argument. The Court will defer any consideration of the de minimus doctrine, as *747 applied in this case, until the evidence presented in the trial is weighed.

A hearing was commenced before this Court on December 19, 1977, upon the Government's request for preliminary and permanent injunctive relief. See Fed.R. Civ.P. Rule 65(a) (2). What follows herein may be considered the findings of fact of this Court. Fed.R.Civ.P. Rule 52(a).

Defendant General Foods Corporation is a Delaware corporation with its headquarters in White Plains, New York. Defendant Barzelay is the President of General Foods Corporation. Edward Fenct, also named as a defendant, is General Foods Corporation's Business Manager of Fruits and Vegetables. Defendant James is Director of Quality Assurance, Environmental Control and Occupational Safety for General Foods. James Beno, the last named defendant, is General Foods' Facility Manager for the Fulton, New York processing plant which is the subject of this action. All of the individual defendants exercise some degree of authority over the operation of the Fulton facility.

The parties have stipulated that all of the frozen, french-style green beans which were processed at defendants' Fulton facility between August 23, 1977, and September 10, 1977, inclusively, and which are currently within the custody or control of defendant General Foods, have either been introduced into, and subsequently received after movement in, interstate commerce by General Foods, or have been held in New York State for ultimate introduction into or delivery for introduction into interstate commerce.

On August 23, 1977, Food and Drug Administration (F.D.A.) Consumer Safety Office Investigator Kathleen Shanahan commenced a routine inspection of the General Foods Fulton, New York processing plant. Miss Shanahan, an F.D.A. employee since 1966, has held her current position since June, 1976, although as a "stride intern" from 1973-76, she conducted some F.D.A. inspections, and accompanied F.D.A. inspectors on others. The August inspection was only Miss Shanahan's second inspection of a frozen green bean processing facility, the other having been of General Foods' Avon, New York plant, which, at the time, was reprocessing frozen beans. Miss Shanahan has conducted approximately 125-150 inspections, 25-30 of which were of vegetable processing facilities, 5-10 of which, in turn, involved green beans.

Miss Shanahan began her inspection with a walk-through of the processing facility which, on that occasion, was processing only green beans. Inspector Shanahan noted that unprocessed, raw beans are initially placed in a hopper-like farm box outside the plant. (Exhibit 7a, representing a flow chart of the defendants' Fulton green bean operation, is appended to this Opinion). From the hopper, the beans are transported to a "cluster-buster", designed to separate the beans, then on to an air cleaner, and finally a washer. The aforementioned production stages take place in an outside production area. The beans are then transported to the inside production area, by conveyor belt, where they enter a stainless steel shaker which feeds the beans into 18 snippers, where the ends are removed. The beans are then conveyed, by a combination of belts and shakers, to a series of graders, or sieves, which sort them by diameter to determine whether they are processed as whole, cut, or french-style beans. All of the belts throughout the process are wet. The water used in the plant is Fulton city water and contains approximately one part per million chlorine.

In the course of her inspection, when reaching this point in production, Miss Shanahan decided to follow the french-style beans production line. She found that from the graders, those beans enter a blancher which cooks the beans slightly for five to six minutes at 212° F., as measured by Miss Shanahan. From the blancher, the beans are passed through a cooling tank, taken by belt to a shaker, and passed over one of three inspection belts, at which point human inspectors remove undesirable beans. *748 They are then fed, by shaker, into six french cutters. From the cutters, the beans are taken by water flume to the packaging area where they are placed in cardboard boxes. The beans, other than those to be preserved for later reprocessing, are then passed through a freezing tunnel where they are subjected to freon for fast freezing.

On the first day of the inspection, virtually all of the belts, in the preblanch area, be they metal mesh, plastic coated, or cleated, appeared to Miss Shanahan to be covered with a green, slimelike material, generally measuring between one-sixteenth and one-quarter of an inch in thickness. The slimelike material, which Miss Shanahan distinguished from vegetable residue, was grayish-white underneath the green surface. The slimy material smelled strongly of decaying vegetable matter, a smell which pervaded throughout the entire facility, particularly in the preblanch area.

Miss Shanahan returned on the next day, August 24, 1977, to find that no beans were being processed, and the cleaning crew was working in the preblanch area. The condition of the belts had not changed since the day before. Miss Shanahan collected eleven subsamples of the slimy material, ten of which were from the preblanch area, and all taken before production resumed. The samples were taken from the following locations: 1) belt, snipper to grader; 2) belt out of sieve # 2; 3) belt, sieve # 1 to second graders; 4) belt carrying # 3 and # 3 sieve beans out from sieves; 5) plate above belt carrying # 2 and # 3 sieve beans to elevating belt; 6) belt elevating beans to cutters; 7) cleated belt to french blancher; 8) inside cutter # 1; 9) belt, cooler to french bean cutter; 10) belt, cutters to inspection; and 11) belt, carrying whole beans from graders to pump into blancher. The parties have stipulated that the integrity of these samples, as well as others taken at a later date, was maintained from collection until testing.

As noted earlier, defendant's quality control manager for the Fulton plant, J. B. Beidleman, accompanied Miss Shanahan on the inspection, and collected duplicate samples of those taken by her. Mr. Beidleman was not called as a witness by either of the parties.

On August 24, 1977, Miss Shanahan also collected 24 subsamples of finished product manufactured late on August 23, 1977, and 24 finished product subsamples from the first hour of production on August 24, 1977.

During the August inspection, Miss Shanahan observed several flies, both of the fruit and house variety, saw preblanch beans being shovelled from the floor back onto the belts, and noticed that in general, the conditions post blanch were substantially better, from a sanitation standpoint, than those in the preblanch area.

Miss Shanahan returned to the Fulton facility for a second inspection in the late afternoon of Friday, September 9, 1977. That date was chosen because clean-up at that plant generally took place on Sunday, so that the inspection would come nearly a full week thereafter. Miss Shanahan was accompanied on this inspection by Perry Nichols, an F.D.A. trainee. Mr. Nichols did not testify in these proceedings. When the two arrived, the plant was not in operation, and had not been since 2:00 that morning. Production recommenced about 6:00 that evening.

During her inspection of the french line, Miss Shanahan again observed a one-sixteenth to one-quarter inch slime buildup on the belts, in approximately the same degree as before, as well as the odor of decaying vegetable matter in the preblanch area, flies, especially near the presnipper shaker, and open doors and windows. Miss Shanahan collected 25 subsamples, 23 of the slimelike material and 2 of water. Of the subsamples taken, all but three were from the preblanch area.^[7] On September 9, 1977, Miss Shanahan also collected eight subsamples of finished product manufactured between *749 August 23, 1977, the date of the first inspection, and that day. She returned the next day, September 10, 1977, and collected subsamples of finished product manufactured on September 9, and September 5. All subsamples of finished products collected were of frozen french-style green beans.

The subsamples collected in defendant's Fulton facility by Miss Shanahan were analyzed by F.D.A. chemist Jane Kaminski. The finished product subsamples were tested using a modified version of the filth analysis contained in Chapter 44 of the Association of Official Analytical Chemists Manual.^[8] After the net mass of the product is determined, the product is thawed by adding 250 millilitres (ml) of boiling water and gently stirring. The solution is drained for three minutes on a # 8 sieve, and the solid product is discarded. The sieve and container are then washed with 300 ml of water, and put through a # 16 sieve. The rinsings are then put through a # 230 sieve. The residue from that last sieve is put into a centrifuge tube (50 ml), and water is added until a volume of 10 ml is collected. To that volume is added crystal violate, to stain any filth, and 10 ml of a stabilizer solution. Four ½ ml portions of that 20 ml volume are separated out, placed under a 30 × magnification, and the number of Geotrichum mycelial fragments^[9] is counted in each. The total number of fragments in the four specimens is multiplied by 10, giving the number of mycelial fragments in the subsample. That number is divided by the net mass of the subsample, and multiplied by 500 grams, thereby giving the results in the form of a mycelial fragments of Geotrichum per 500 grams of product.

In performing this modified AOAC analysis for filth on the samples of finished product collected by Investigator Shanahan, Miss Kaminski obtained the following results:

 Sample     Mfg.  No. Samples     Average
   No.      Date    Tested         Count             Range
XX-XX-XXX   8/23      6         55 frag/500 g  46-70 frag/500 g
XX-XX-XXX   8/24      6        107 frag/500 g  81/137 frag/500 g
XX-XXX-XXX  9/9       6        194 frag/500 g  108-368 frag/500 g
XX-XXX-XXX  9/5       6         32 frag/500 g  16-66 frag/500 g
XX-XXX-XXX  9/5       6        500 frag/500 g  133-2146 frag/500 g

Miss Kaminski also tested the environmental samples collected by Inspector Shanahan during the course of her August and September inspections. For those analyses, portions of the subsamples, which were contained in a formaldehyde solution designed to prevent growth, were placed on a covered slide and qualitatively examined under a 100 × microscope. The chemist then looked for the very characteristic features of Geotrichum, including a feathery texture, crossed walls, and series of at least three branches growing off the main trunk at 45° angles, with tapered growing ends. The presence of Geotrichum, or machinery mold, was detected in each of the 11 subsamples collected by Miss Shanahan on August 24, 1977, and in all but one, # 7,^[10] of the subsamples from September 9, 1977.

Added to the testimony of Miss Shanahan's observations, coupled with the results of tests performed on the samples collected by her, the Government points to two of General Foods' internal documents which seemingly acknowledge the presence of *750 some buildup of green, slime-like material on the Fulton bean processing equipment during the period in issue. The first, Exhibit 37, is J. B. Beidleman's report of the August 23-24, 1977 F.D.A. inspection. In that report, Mr. Beidleman referred to the 11 duplicate samples collected by him from the processing equipment at the same time as Miss Shanahan took hers as "green slime". The second document, Exhibit 38, is a report to J. W. Beno, the Fulton plant manager, of a plant quality assurance audit conducted by R. B. LaBelle, of General Foods' Corporate Quality Assurance Department, on August 31, 1977. In that report, Mr. LaBelle noted "the presence of green slime at various points in the vegetable processing area . . .."

The defendants called Margaret B. Moore, a group leader in the Fulton plant, as a witness to testify as to plant conditions in August and September of 1977. At that time, there were generally 40 line inspectors and 12 sanitation personnel working during each shift in connection with the bean processing. Mrs. Moore has never seen a one-quarter inch buildup of slime on the equipment, although she does notice occasional small buildups of slime in the plant. Mrs. Moore saw Miss Shanahan in the plant in September, but does not recall anything extraordinary about that night. Mrs. Moore noted that a putrid odor occasionally exists within the plant, generally when the drains become plugged.

The only other witness having direct knowledge of the actual conditions existing within the plant during the critical time was Richard L. Hazard, food inspector with the New York State Department of Agriculture and Markets. Mr. Hazard made a routine, unannounced inspection of the Fulton plant on August 18, 1977, at which time fresh and frozen green beans were, he thinks, being processed. The only deficiencies noted by Mr. Hazard involved flaking and peeling paint in the production area, and a one or two inch opening of an overhead door through which rodents or insects could enter. He does not recall seeing any slime buildup, smelling a putrid odor, or observing the presence of any flies. Samples of reprocessed, frozen beans taken by Mr. Hazard tested satisfactorily, although the only test requested by Mr. Hazard was for paint particles.

This Court finds, based upon the foregoing, that on August 23 and 24, 1977, and one again on September 9, 1977, there was some visible buildup, on food processing equipment at the defendants' Fulton facility, of a green, slime-like substance which, when qualitatively analyzed, revealed the presence of Geotrichum.

Geotrichum, commonly referred to within the food processing industry as machinery mold, also known as dairy mold, oospora and oidium, is one whose spores are generally isolated from soil, sewage, and surfaces of fruits and vegetables growing in the field. Geotrichum is not harmful to humans when consumed in food. In fact, it is intentionally introduced into several types of cheese (Limburger, Camemberti, and Brie) as a curing agent, and as such both live and dead Geotrichum is present in those cheeses as part of the finished products. Geotrichum would be killed when subjected to a blanch of 5-6 minutes at 212° F., such as that utilized by the defendants at Fulton, but the dead mycelium would remain unless otherwise removed. Some, though not all, of the mycelial fragments drop off during post blanch washing.

Geotrichum spores enter a plant such as that at Fulton in several manners, though primarily on the raw vegetables delivered for processing.^[11] The spores require nutrient, *751 moisture,^[12] and a proper temperature to grow into viable mycelium. The liquid exuded from a cut or damaged green bean provides an excellent nutrient for Geotrichum spores. Geotrichum, when present in any quantity in a finished food product constitutes filth.

When Geotrichum begins to grow, it generally takes 4-8 hours to become visible. As it grows, the mold generally appears gray or white in color, although in a food packaging plant its surface would generally take on the color of the juices flowing from whatever food was being processed at the time. The mold accumulation generally has a slimy texture.

Geotrichum is virtually unavoidable in a food processing plant, particularly one dealing with green beans. The mold can, however, be controlled through the use of proper routine cleaning supplemented by preventative measures, such as high pressure spray of belts and equipment using a chlorine solution. With proper cleaning, visible mold, together with the putrid odor often accompanying it, can be avoided. The presence of analytical Geotrichum, however, especially in a green bean processing plant, cannot be avoided.

General Foods does not recognize Geotrichum as an indicator of plant sanitation because the mold grows so quickly and because in a neutral pH atmosphere, such as on green beans, bacteria grows to a greater extent than molds or yeasts. Rather, it relies upon a standard plate count as the best indicator of sanitation, setting, as a standard limit, 100,000 living bacteria per gram. They also test for coliform, with a limit of 500 per gram. E. Coli is sometimes checked, and occasionally coagulase positive staphylococci with a limit of 100 per gram. Results of such tests performed by General Foods on August 23 and 24, 1977 and September 5 and 8, 1977, were all well within these established limits, which are in turn well below the average within the industry.

Just as General Foods does not recognize Geotrichum as an indicator of plant sanitation, the Government's experts do not rely wholly upon plate count as such an indicator, because most of the viable microorganisms in the product are killed during blanch, and would therefore not be detected by the standard plate count method. Thus, while these plate counts are adequate indicators of blanch effectiveness, they fail to detect the presence of nonviable filth, such as dead Geotrichum.

The Court believes the answer to lie somewhere between the two positions. Standard plate count is certainly an indicator of plant sanitation. So too is the visible accumulation of machinery mold, within the processing facility, particularly on surfaces and equipment within which the product will come into contact. Likewise, the quantity of Geotrichum found in the finished product has some bearing upon plant sanitation. The mere presence of analytical Geotrichum in the facility, however, is not necessarily an indication of insanitary conditions.

The presence of a few flies within a processing plant such as the defendants' is generally unavoidable. An accumulation of fruit flies, however, can be an indication of the presence of decaying matter.

The practice of shovelling beans off the floor back onto the conveyor belts is not desirable, but does not constitute a violation of Current Good Manufacturing Practices, at least when confined, as here, to the preblanch area.

The only Defect Action Level (DAL) established with respect to Geotrichum, or slime mold, is 20% for crushed and canned pineapple, using a Howard Mold Count method of tabulation. Based upon knowing the volume of the fields counted in the Howard Mold Count, and assuming the presence of one Geotrichum filament in each *752 positive field, defendants' experts calculate a 20% Howard Mold Count to be the equivalent of 200,000 mycelial fragments per 500 grams. However, Dr. Eisenberg, a Government witness, demonstrated by empirical analysis, based upon Howard Mold Counts and Cichowicz-Eisenberg tests on identical samples of canned pineapple, shows widely varied results, with an average Howard Mold Count of roughly 19% being the equivalent of approximately 1662 fragments per 500 grams, on an average. Of the 12 samples used in making the comparison, the Howard Mold Counts ranged from 16% to 23%, and the corresponding Geotrichum counts fell between 127 and 6,843 mycelial fragments per 500 grams.

In March of 1974, the F.D.A. conducted a HACCP (anagram for Hazard Analysis and Critical Control Points) inspection of the Fulton facilities. Following that rigorous, five-day check of equipment and procedures, the F.D.A. determined that the plant was in compliance with their standards. At the time of the inspection, the Fulton plant was processing mushrooms, and repackaging preprocessed, individually quick frozen (IQF) green beans.

The test most often applied in (a) (4) cases, such as this, is to determine whether or not the food in issue was processed under conditions whereby there is a reasonable possibility that filth contamination could have occurred. Berger v. United States, 200 F.2d 818 (8th Cir. 1952); United States v. 1,200 Cans, Pasteurized Whole Eggs, Etc., supra. Because the purpose of (a) (4) is to prevent contamination, or nip it in the bud, actual contamination of the finished product need not be shown. United States v. H. B. Gregory Co., 502 F.2d 700 (7th Cir. 1974), cert. den. 422 U.S. 1007, 95 S. Ct. 2629, 45 L. Ed. 2d 670 (1975).

While actual finished product contamination need not be shown, evidence of it is certainly relevant to an alleged (a) (4) violation. United States v. 44 Cases, Etc., supra. Also relevant is evidence of observations made during inspections of the facilities in issue, provided they are conducted at or about the same time that the product alleged to be adulterated under (a) (4) is manufactured. Compare Berger v. United States, supra; United States v. 1,200 Cans, Pasteurized Whole Eggs, Etc., supra; and United States v. 44 Cases, Etc., supra; with United States v. 1,500 Cases, More or Less, Etc., supra. This will depend, of course, on the particular circumstances involved, including how removed in time from the dates of production the observations are, whether the nature of the deficiencies are such as could quickly be remedied, and whether they are of an inherently fluctuating nature. United States v. 1,200 Cans Pasteurized Whole Eggs, Etc., supra; Federal Rules of Evidence Rules 401 & 402, Title 28 U.S.C.; also cf. Strauss v. Douglas Aircraft Co., 404 F.2d 1152 (2d Cir. 1968); Russell, Poling & Co. v. Conners Standard Marine Corp., 252 F.2d 167 (2d Cir. 1958); Shanahan v. Southern Pacific Co., 188 F.2d 564 (9th Cir. 1951); Lever Brothers Co. v. Atlas Assurance Co., 131 F.2d 770 (7th Cir. 1942).

To refute allegations of an (a) (4) violation, a defendant may introduce evidence of routine customs, or business habits, involving established cleanup, sanitation, and maintenance procedures, provided they fall within Rule 406 of the federal Rules of Evidence, Title 28 U.S.C. cf. Spartan Grain & Mill Company v. Ayers, 517 F.2d 214 (5th Cir. 1975); Hambrice v. F. W. Woolworth Co., 290 F.2d 557 (5th Cir. 1961).

The relevant evidence presented bearing upon the (a) (4) violation boils down to: 1) observations by Miss Shanahan of an accumulation of a green, slime-like material on belts and equipment, especially in the preblanch area, samples of which tested positive for the presence of Geotrichum; 2) testimony of Mrs. Moore that neither she, nor any of the girls working under her supervision, would allow such a buildup to *753 accumulate without shutting down production; 3) Mrs. Moore does not recall ever seeing such a buildup of slime on the equipment; 4) Mr. Hazard, a state inspector, failed to observe any buildup of slime during his inspection of August 18, 1977, some five days prior to that of Miss Shanahan; 5) the plate counts resulting from defendants' tests performed on the dates in question showed remarkably low levels of living bacteria, coliform, and E. Coli; 6) when producing green beans, General Foods uses 40 line inspectors and a twelve man sanitation crew, with a foreman for every shift; 7) General Foods maintains more stringent sanitation requirements for the post blanch area than the preblanch areas; and 8) General Foods concedes that it does not rely upon Geotrichum as an indicator of sanitation.^[13] The defendants urge the relevancy of the fact that, following a rigorous 1974 HACCP inspection of the Fulton facility, the F.D.A. determined defendants' facilities and procedures to be in compliance.

The Court does not consider the 1974 HACCP inspection to be relevant to the (a) (4) allegations, for the simple reason that from 1974 to 1977 one would expect the conditions at the Fulton facility, and the procedures utilized there, to change, very drastically. What is more, the food processing industry is in a constant state of flux, so that what might have been acceptable in 1974 to the F.D.A. as current good manufacturing practice might, in 1977, be totally unacceptable. Evidence of the state inspection conducted in August of 1977 is, however, very relevant to the (a) (4) charge, it being very close in time to the period involved.

The analytical presence of Geotrichum in samples of both finished product and residue removed from defendants' equipment is not particularly startling, in light of the overwhelming amount of testimony, from experts called by both sides, that Geotrichum is unavoidable in a green bean processing facility. Moreover, the value of those test results as indicators of plant insanitation is outweighed by the extremely low plate counts recovered by the defendants in August and September of 1977.

Based upon the evidence presented, this Court does not believe the Government has shown, by a fair preponderance of the evidence, United States v. 233 Tins, More or Less, Etc., supra, that the frozen french-style green beans in issue were processed under such conditions as to present a reasonable possibility that they could have become contaminated with filth.

The nature of an (a) (3) violation has been explained in the portion of this opinion relating to the defendants' pretrial motion to dismiss. To briefly recapitulate, the literal language of (a) (3), which proscribes filth in food, in any amount, has been interpreted as being subject to exception, both by virtue of section 306 of the Act, 21 U.S.C. § 336, which gives the F.D.A. considerable discretion in deciding when to bring actions, and the judicially-evolved de minimis rule, which allows a court to overlook a small amount of harmless filth where there is no evidence of that amount of filth being avoidable within the industry.^[14]

Of the samples of finished product tested by Jane Kaminski, the highest count for Geotrichum found in any one subsample was 2146 mycelial fragments per 500 grams, *754 or just over one pound. The averages found in each of the five finished product samples were 55, 107, 194, 32, and 500 fragments per 500 grams, with the maximums found in each being 70, 137, 368, 66, and 2146 fragments per 500 grams, respectively. By way of comparison, the 20% Howard Mold Count defect action level established for crushed and canned pineapple corresponds to 200,000 fragments per 500 grams, using Dr. Olson's computations. According to Dr. Eisenberg's empirical data, an average crushed pineapple sample with a 19% Howard Mold Count, represents an average of 1662 fragments per 500 grams (with a very wide range of results).^[15] According to Dr. Olson's calculations, based upon a number of assumptions concerning the average size and mass of a mycelium of Geotrichum, 2146 fragments in a 500 gram specimen represents approximately .05 parts per million on a mass-to-mass ratio.

It must be remembered that the Geotrichum inoculum (spores) come into the processing plant primarily on the raw beans, and are provided an ideal environment for growth by virtue of the liquid exuded when the beans are snipped. Thus, we are dealing with contamination inherent in the raw product itself, rather than filth being introduced into the food from the environment during processing.

The evidence is overwhelmingly supportive of defendants' assertion that Geotrichum is an unavoidable, albeit controllable, problem in the green bean processing industry. There is little or no evidence whatsoever to show that the levels of Geotrichum discovered in defendants' finished product were avoidable through the use of current good manufacturing practices, as exist within the industry. Finally, it is uncontroverted that Geotrichum, at the levels with which we are dealing, is not harmful when consumed by humans. This Court therefore finds that, while the Geotrichum found in samples of defendants' frozen, french-style green beans constitutes filth, the amount is de minimis, and does not therefore constitute an (a) (3) violation.^[16]

The frozen french-style green beans in issue are not adulterated either under section 402(a) (3) of the Act, 21 U.S.C. § 342(a) (3), or § 402(a) (4), 21 U.S.C. § 342(a) (4). The Government is therefore not entitled to injunctive relief pursuant to section 302 of the Act, 21 U.S.C. § 332.

The Clerk shall therefore enter a judgment in favor of the defendants in all respects.

*756

*758

[1] § 331. Prohibited acts

The following acts and the causing thereof are prohibited:

(a) The introduction or delivery for introduction into interstate commerce of any food, drug, device, or cosmetic that is adulterated or misbranded.

* * * * * *

(c) The receipt in interstate commerce of any food, drug, device, or cosmetic that is adulterated or misbranded, and the delivery or proffered delivery thereof for pay or otherwise.

. . .

[2] § 332. Injunction proceedings Jurisdiction of courts

(a) The district courts of the United States and the United States courts of the Territories shall have jurisdiction, for cause shown, and subject to the provisions of section 381 (relating to notice to opposite party) of Title 28, to restrain violations of section 331 of this title, except paragraphs (h)-(j) of said section. . . .

[3] The defendants do not challenge the Government's assertion that the green beans in issue constitute food under Section 201 of the Act, 21 U.S.C. § 321.

[4] The beans in issue, according to the trial testimony of defendant David E. James, General Foods' Director of Quality Assurance, Environmental Control and Occupational Safety, total approximately 1,836,000 pounds, being valued at $1,294,000 at General Foods' price to the trade (Trial Transcript p. 602). Mr. James also noted that french-style green beans are in short supply this year. (Id.)

[5] The DAL argument pertains more to the (a) (3) violation than the (a) (4) claim. It is difficult to conceive of a way for the F.D.A. to prescribe tolerances for Geotrichum as relates to conditions in the processing plant. The regulations pertaining to (a) (4) are of necessity somewhat broad, aimed at the problem of sanitation in general, rather than any specific type of contaminant. See 21 C.F.R. Part 110. In light of the disposition of this motion, however, the Court need not decide whether the argument is equally applicable to (a) (4) violations.

[6] The Seventh Circuit Court of Appeals in United States v. 1,500 Cases, More or Less, Etc., supra, expressed disbelief of any assertion that Congress directed the F.D.A. to establish DALs for poisonous or deleterious substances, implying that those substances are tolerable in certain small quantities, and yet apparently determined filth in any amount to be intolerable. United States v. 1,500 Cans, More or Less, Etc., supra at 211. There is considerable merit to this argument. The fact remains, however, that the language of § 346 very clearly applies only with respect to poisonous or deleterious substances.

[7] A diagram, exerpted from Exhibit 9, showing the location of each of the 25 subsamples taken, is appended to this opinion.

[8] This is often referred to as the Cichowicz-Eisenberg method, after Stanley M. Cichowicz and William V. Eisenberg, both of whom testified as Government witnesses in this case.

[9] A mycelial fragment of Geotrichum is generally defined as three or more thread-like strands joined at one end of a branched filament.

[10] Subsample # 7 was collected from water dripping from a pipe in the area of the pre-snipper shaker.

[11] It has been estimated that as many as one billion living organisms can be found in a 500 gram sample of raw product delivered to a food processing plant.

[12] It should be noted that the summer of 1977 was unusually rainy for this region.

[13] While it is true that the Government failed to call as a witness Perry Nichols, the F.D.A. trainee who accompanied Miss Shanahan on the second inspection, it is equally true that the defense did not choose to call J. Beidleman, their employee who accompanied Miss Shanahan on both inspections, and who took duplicate samples of the "green slime" described by Miss Shanahan. This Court does not consider this to be relevant.

[14] A good example of an application of the de minimis rule can be found in United States v. Capital City Foods, Inc., supra. In that case, samples of butter tested revealed the presence of 28 insect fragments in 4125 grams, or about 3 fragments per pound.

[15] Caution should be observed in weighing the results of such a comparison, inasmuch as experts have testified that an acidic medium, such as tomato or pineapple, will favor growth of molds and yeasts over bacteria, whereas materials with a more basic pH, such as green beans, will generally favor growth of bacteria. Consequently, it could be argued that a 20% Howard Mold Count for slime mold would be unacceptable for green beans, despite being allowable for pineapple.

[16] In reaching this conclusion, this Court does not place any reliance whatsoever upon the failure of the F.D.A. to establish a Defect Action Level for Geotrichum in green beans. Such a DAL would certainly be of benefit to the industry, and the public at large. This Court rejects the notion, however, that the F.D.A. is required to set such a standard, and that F.D.A. enforcement of (a) (3) without such a DAL is unfair.