A simple and rapid (<60 s) nonstaining technique with 3% potassium hydroxide to determine Gram reactions was tested with 495 food-borne and waterborne bacteria and yeasts. In KOH, suspensions of gram-negative bacteria become viscous and string out. Gram-positive bacteria are not affected. There was 100% correlation between the KOH string test results and gram-positive and gram-negative strains. A simple, nonstaining method, requiring only one reagent to determine the Gram reaction, was first reported in 1938 by E. Ryu of Taiwan (see references 13 and 14 [also cited in references 6, 7, 11, and 19]). In spite of some more recent evaluations (2-4, 6-12, 19), the method still appears to be obscure and not widely known among food and water microbiologists. The potassium hydroxide string test has been tested against veterinary isolates (6, 7), bacterium and yeast isolates from brewery samples (11), marine isolates (3, 18), Listeria spp., (10), clinical anaerobes (2, 4, 8), and other clinical isolates (9). The purpose of this study was to verify the efficacy of the KOH string test for determining the Gram reactions of food-borne and waterborne bacteria. Test microorganisms included isolates from food, milk, and water and from stock cultures of microorganisms found in foods and water. Cultures were obtained from U.S. Army stock cultures, other government agencies (U.S. Department of Agriculture, Food and Drug Administration, and Centers for Disease Control and Prevention) food industries, hospitals, universities, and the American Type Culture Collection (Rockville, Md.) or were obtained directly from military foods analyzed. The KOH string test was performed by mixing a visible amount of growth from a colony or an agar slant in a loopful (3 mm) of 3% aqueous KOH on a glass slide. The KOHbacterium suspension was mixed continuously with a bacteriological loop in a 1-to 2-cm 2 area on the slide. If such a suspension gels or becomes viscous and strings out when the loop is lifted (positive KOH reaction), the isolate is gram negative. Gram-positive cells do not form a viscous gel or string out (negative KOH reaction). The cutoff time for negative reactions was 60 s. The string test was the best indication of a gram-negative isolate, and the result was best observed by raising the loop about 1 cm above the slide. Holding the slide at an angle against a dark background also aided the observation. Colonies from culture plates of food samples and from selective and differential media were transferred to Trypticase soy agar (Difco, Detroit, Mich.) or plate count agar (Difco) for both the KOH string test and Gram staining. Gram staining (5, 15, 17) was performed on 18-to 24-h cultures in parallel with the KOH string test. Identification of gram-negative isolates was confirmed by using, singly or in combination, the API 20E system (Analytab Products, Plainview, N.Y.), the Enterotube system (Roche Diagnostic Systems Inc., Montclair, N.J.), and the Micro-ID system (General Di

Spices purchased by the Army, Navy, Marines, and Air Force were tested for incidence and levels of Bacillus cereus. One hundred and ten processed spices, including bay leaves, red pepper, chili powder, cinnamon, garlic powder, mustard powder, and oregano were tested. Bacillus cereus was found in 53% of the spices and counts ranged from 50 to 8500 per gram. Eighty-nine percent (88/99) of the isolates tested were toxigenic in rabbits, by the vascular permeability assay, and toxigenic B. cereus was found in each kind of spice. These data have significant implications for food safety and sanitation and for fumigation of spices by gas, or irradiation.

The microbiology of spices procured by the Army, Navy, Marines and Air Force was determined. The incidence of Clostridium perfringens in the spices analyzed was 15%. The organism was found in 4 out of 7 types of spice and in 53% of the oregano samples. No other bacteria of public health significance were found. The microflora of spices from 10 different brands varied widely.

Chlor-Floc (CF) emergency water purification tablets were tested for bactericidal, virucidal, and cysticidal efficacy in water at temperatures ranging from 5 to 25°C. The minimal required log reduction was achieved for bacteria, Giardia muris, and rotavirus, but CF did not achieve the required log reduction of poliovirus at any of the temperatures or times investigated. The biocidal properties of the CF tablet were equivalent to if not greater than those of the Globaline iodine tablet, and the CF tablet was a more rapid cysticide under several potential use conditions. Therefore, it is a suitable substitute for iodine tablets for emergency purification of drinking water. Clarification of turbid waters was effective, but filtration through a cloth is necessary to prevent flocculated sediment from entering the canteen. The CF tablets met military requirements for emergency water purification and are safe and acceptable for use by the military.

An interlaboratory study was conducted to evaluate a method for the determination of campesterol, stigmasterol, and beta-sitosterol in saw palmetto raw materials and dietary supplements at levels &gt;1.00 mg/100 g based on a 23 g sample. Test samples were saponified at high temperature with ethanolic KOH solution. The unsaponifiable fraction containing phytosterols (campesterol, stigmasterol, and beta-sitosterol) was extracted with toluene. Phytosterols were derivatized to trimethylsilyl ethers and then quantified by gas chromatography with hydrogen flame ionization detection. Twelve blind duplicates, one of which was fortified, were successfully analyzed by 10 collaborators. Recoveries were obtained for the sample that was fortified. The results were 99.8, 111, and 111% for campesterol, stigmasterol, and beta-sitosterol, respectively. For repeatability, the relative standard deviation (RSDr) ranged from 3.93 to 17.3% for campesterol, 3.56 to 22.7% for stigmasterol, and 3.70 to 43.9% for beta-sitosterol. For reproducibility, the RSDR ranged from 7.97 to 22.6%, 0 to 26.7%, and 5.27 to 43.9% for campesterol, stigmasterol, and beta-sitosterol, respectively. Overall, the Study Director approved 5 materials with acceptable HorRat values for campesterol, stigmasterol, and beta-sitosterol ranging from 1.02 to 2.16.