Preface

Electronic publishing arrived just in time to save the Official Methods of Analysis of AOAC INTERNATIONAL from its own success. Each of the two volumes of the previous 17th Edition has grown to an almost unmanageable size. Now, for the first time, the results of 121 years of review and approval of collaboratively studied methods are captured on the Internet. The older methods are still there, but their use is probably confined to teaching “agricultural chemistry.” The properties of methods do not deteriorate with age and the “classical” analytes still need to be determined. But the change in emphasis with the regulatory winds is obvious: Microbiology and nutrition have blossomed as regulatory emphasis has shifted from economics to safety and health. New drug and food additive approval, based on preclearance of manufacturing operations and continuous quality control as well as safety and efficacy, have reduced the need for regulatory control through market sampling and analysis.

Most notable has been the shift from classical stoichiometric chemistry, based on the balance and buret, to calibration chemistry, based on an instrumental comparison of a response of an analyte with that of a standard. This shift was initiated by the remarkable separation powers of chromatography allowing an analyte to be separated from interfering components before being measured by the instrument. Chromatography moved analytical chemistry from the realm of gram quantities into microgram quantities but not without unrecognized sampling problems. The sensitivity, stability, and speed of modern electronics permit the performance of analytical work automatically, from the measurement of the test portion, through detection, amplification, and interpretation of the signal, to the printing of the analytical report.

The analytical problem has shifted from measurement to control. Much of the analytical operation has moved from an operator to a black box in a computer. Changes in physical properties, such as light intensities or ion conductances, are measured automatically and converted into analytical reports continuously, changing the laboratory into an automated factory. But the facility for automated performance allows the responsibility for reliability to easily shift from the analyst to the instrument. This is also true of the blind application of computer programs with no review of the applicability of the program to the problem. The computer has the ability not only to extract hidden information from a jungle of background, but also to formulate spurious peaks that it has been programmed to guess ought to be there.

AOAC initiated the procedure of validation of methods through interlaboratory studies. These studies produce results from a single sample of method performance in the hands of an assumed random sample of laboratories. Unfortunately, time and expense rarely permit performing additional studies. Therefore, the initial studies usually stand as the sole published evidence of satisfactory interlaboratory performance. AOAC members are investigating surrogates for this necessary, but lengthy and costly procedure.

Sixty-eight new methods have been added to this edition, predominantly microbiological or chromatographic in nature, all of them subjected to the rigors of an interlaboratory study. Many of these methods incorporate internal controls to ensure that the reactions are proceeding as intended. Most appealing is the introduction of system suitability specifications into chromatographic systems that permit flexibility without sacrificing reliability. For over a century, the guiding principle in the application of standard methods has been to follow instructions to the letter to obtain results equivalent to those originally obtained. But the competition for improvements in systems advanced the science of separation and detection so rapidly that suitability specifications for introducing flexibility without sacrificing performance had to be invented.

Internal controls require that the methods meet repeatability performance specifications. An appreciable fraction of the new microbiological methods are screening tests involving preassembled immunoassays kits. Relatively quickly, these kits separate laboratory samples that can be discarded as negative from those that presumably contain pathogenic organisms, requiring the application of confirmatory tests. These kits also invariably contain the requirement for accompanying positive and negative controls that provide concurrent assurance of proper performance.

This edition joins the universal movement toward the use of the international system (SI) of units, many of which are unfamiliar to U.S. scientists. During a transition period, both the common system as well as the SI system will be given. Note that the term “normality” is being replaced by “moles per liter.” Another editorial change being introduced is to move away from the tendency to designate anything being worked on as the “sample.” Instead, the sequence of “laboratory sample” —> “test sample” —> “test portion” is being introduced. This vocabulary is being used by the International Union Of Pure and Applied Chemistry (IUPAC) and the International Organization for Standardization (ISO), which does not permit the unmodified term “sample” to be used in conjunction with subsequent chemical operations.

An important feature of the 18th Edition is the international source of many of the methods, with many countries and international organizations contributing their expertise to method standardization. It is also gratifying to see the introduction of quality control features into the methods, which provide the analyst with guides to proper performance. On the other hand, the ease with which results are obtained from computers also permits the introduction of unanticipated errors, detected only by the unreasonableness of the results. In the absence of a blueprint of what is to be expected, gross errors may be made. The introduction of quality assurance principles into the laboratory may assist in minimizing such occurrences.

Numerous individuals, volunteer scientists, and professional staff have contributed enthusiastically to this century-old program of method validation. The analytical community is grateful for their continued valuable efforts.

—William Horwitz, Editor