The intimate relation between vertical wind, humidity and crystal number density in cirrus is well documented in several case studies. Here we present a statistical analysis of these parameters from observations conducted within the international project INCA (Interhemispheric differences in cirrus properties from anthropogenic emissions). The data comprises of 10 flights from the Southern Hemisphere midlatitudes and 9 flights from the Northern Hemisphere midlatitudes. The results show that the data can be used to develop a phenomenological model driven by gravity waves to emulate the life cycle of cirrus clouds. Because different phases of a clouds life cycle occur on different timescales (formation, development and dissipation), the probability to observe a particular combination of relative humidity and crystal number density in-situ depend on how often this combination occurs and how long time each event exists. Our data points to a very clear maximum in observations that range only a factor 2 or 3 in crystal number density and approximately 20% in relative humidity units. We may then talk about a typical cirrus in terms of the probability to find a particular combination of relative humidity and crystal number density. In case of our midlatitude data the cirrus clouds are close to ice saturation with a crystal number density of about 2 cm ^-3. Data is exploited further to make conclusions about the relative humidity at the point of cloud formation. This analysis suggests that at temperatures below 235K nucleation occur typically around 120 to 130% (over ice) in the Northern Hemisphere but is shifted towards higher humidities in the Southern Hemisphere.