It is now over 100 years since Kayser (Archiv für Laryngol Rhinol 1895; 3: 101-120) first reported in the scientific literature that the human nasal passages exhibit spontaneous changes in unilateral nasal airway resistance, yet our understanding of this unusual phenomenon is still very confused. Spontaneous, reciprocal changes in unilateral nasal resistance are often referred to as a "nasal cycle" and although this term is now commonly used to describe spontaneous changes in nasal resistance in man and animals, there is little evidence for any true periodicity. A major problem in increasing our knowledge and understanding of the so-called "nasal cycle" is that most studies have relied on simple descriptions of the changes in nasal resistance and have not developed any numerical parameters to quantify the changes in resistance over time. This lack of definition of what actually constitutes a nasal cycle has meant that the literature of the present day generally accepts the views put forward by Heetderks (Am J Med Sci 1927; 174; 231-244) and Stoksted (Acta Otolaryngol (Stockh) 1953; Suppl 109: 159-175) that around 80% of the healthy population exhibit a regular cycle. In order to define the characteristics of the spontaneous changes in nasal airway resistance we have used numerical measures of reciprocity and also developed a measurement of the division of airflow between the nasal passages over time. With these two parameters it is possible to describe the nature of the spontaneous changes in airflow in numerical terms and to define what exactly constitutes a nasal cycle. Fifty-two volunteers underwent hourly measurement of unilateral nasal airflow for 8 h. For each volunteer, two values were derived from the graph of unilateral nasal airflows against time; the correlation coefficient between unilateral airflows (r) and the airflow distribution ratio between the two nasal airways (ADR). The spread of different types of airflow pattern (nasal cycle) throughout the population was illustrated by plotting r against ADR for each subject. A nasal cycle was defined as having an r value between -0.6 and -1.0, and an ADR value between 0.7 and 1.0. Only 21% (11 of the 52 volunteers) exhibited airflow patterns that could be defined as a nasal cycle in these terms. This finding contradicts the generally accepted, but undefined, view that around 80% of the population exhibit a regular nasal cycle. The numerical definition of a nasal cycle in terms of both reciprocity and airflow distribution, as described in this paper may help to clarify our understanding of this interesting phenomenon and allow rhinologists to describe the spontaneous changes in nasal airflow in more exact terms than have been used previously in the literature.