The geographic/temporal pattern of cases of tick-borne encephalitis (TBE) registered in the Czech Republic since 1970 was analysed to verify the surmise of a global warming effect. Using a geographic information system, over 8,700 notified places of infection were pin-pointed on a map and overlaid with a digital elevation model to estimate the vertical distribution of the cases. Series of yearly disease ceilings (assessed alternatively as the respective maximum altitude or mean altitudes of the upper 5 or 10 cases) were tested against the null hypothesis of random elevation course and analysed for correlation with concomitant factors (yearly TBE incidence rate, mean yearly temperature, population density of small rodents and roe deer). Statistical tests proved that the TBE ceiling has gradually moved upwards in the course of the last three decades. The average rate of ascension within this period was approx. 5.4 +/- 1.7 m yearly, which corresponds well with concurrent mean temperature rising of approx. 0.036 +/- 0.007 degrees C yearly, and the vertical temperature gradient of 0.0065 +/- 0.0004 degrees C m(-1). The TBE-ceiling estimates significantly correlated with TBE-incidence data and the mean yearly temperature recorded 1-2 years earlier. Although TBE incidence correlated with rodent population density that was observed 1-2 years earlier, the TBE ceiling does not seem to be influenced by rodent population dynamics nor did the population dynamics correlate with mean yearly temperatures. TBE incidence as well as mean altitudes of the upper 10 cases also correlated with official data on harvested roe deer. Overall, the fluctuations of TBE incidence and TBE ceiling proved to be synchronous processes that correspond with temperature changes. Although the dependence of TBE on temperature is not a direct one and various factors could be involved, an impact of climate warming on the vertical disease distribution in Central Europe is evident.