In the midst of high technology, it usually comes as a surprise to realize that the simple ionic make-up of the fluid in the pulmonary airways still remains almost completely undefined. Historically, most attention has been dedicated to the volume and viscoelastic properties of pulmonary secretlons-esputurrr'-ratner than to their composition. Cystic fibrosis (CF), which was also named mucovisidosis, le," thick mucus" disease, is a highly visible and provocative example of this tradition. Mucus obstruction of airways has long been considered the most vicious agent of morbidity and mortality in CF, even though it is not clear that the mucus produced is qualitatively or quantitatively different from that produced in inflammatory and hypersecretory states of any other airway disease. In fact, if normal mucus could be defined, there is some evidence that viscoelastic properties in CF might even be" normal"(1). Nonetheless, the focus of therapy has been on thinning airway mucus, which began with adding water to the airways using mist tents-an approach that has been almost totally abandoned for more than a decade. More recently, however, two other approaches designed more specifically to thin mucus have been undertaken. Recent approval from the FDA of recombinant human DNase will allow widespread use of this enzyme. The strategy is clear enough in that most of the viscosity of sputum is due to the presence of naked DNA unleashed in the airway from bacteria, macrophages, and other cellular debris. Enzymatic digestion of these macromolecules effectively decreases mucus viscosity in vitroand is expected to enhance the clearance of airway secretions and inspissations in vivo, but whereas its use may be of benefit in some patients, it is clear that it is not a panacea, even though it is not yet clear whether its use will have a major impact on long-term disease progress (2). In another tact, amiloride, a K+-sparing diuretic that inhibits epithelial Na'channels, has been considered for its effectiveness in retaining water in the airways by blocking Na+ and, thereby, fluid absorption. The strategy here is that retaining salt and water in the airways should dilute the macromolecular component of airway surface fluid (ASF) and therefore thin the mucus, making it easier to clear (3). Again, therapeutic results have been less than anticipated. The fact that these attempts to improve the hygiene of the airways by thinning mucus have met with limited therapeutic success may suggest the presence of a much more fundamental abnormality-altered composition. It is disturbing to realize that although we know the ionic concentration of virtually every fluid in the body defined to a consistent millimolar or even submillimolar range, we have only a vague idea of the composition of the ASF, and we know virtually nothing of how or whether composition varies. Even its pH is uncertain. No doubt our lack of awareness of the composition of ASF is grounded in the fact that collection of sufficient volumes of uncontaminated and unaltered physiologic fluid for conventional analysis is all but impossible, simply because it exists as a very thin layer of liquid, probably much less that 50 11m in depth in a normal airway. But in CF at least, there is good fundamental reason to expect that the composition of the ASF is abnormal; that is, j3-adrenergically stimu-