There appears an ever increasing need of studying not only the changes of values of climatic elements but also the changes of their annual variation. A convenient method of satisfying this need is the study of the principal singularities of the weather and of the singularities of individual climatic elements. Apart from purely synoptic analyses and the formal study of secular series of climatic elements, studies concerning the values of climatic elements and their dependence on the changes of factors of the atmospheric circulation as well as research into the geographical laws governing these changes are also most necessary. An analysis of the annual variation of precipitation in Brno has shown that this annual variation is very different in the successive distinct periods (1851/1875, 1876/1900, 1901/1925, 1926/1950) of the past one hundred years. An analysis of the annual variation in other European stations has confirmed the reality of these changes throughout Europe. We have analysed more fully the changes in the variation of autumn precipitation, and particular attention has been paid to the October precipitation in the Carpathian part of the Danube basin for the period 1901/50, from which the necessary synoptic and meteorological material is already available. In the annual variation of precipitation in this region there occurs a secondary October, or November, maximum. The principal cause of increased precipitation in October is the intensive precipitation of certain synoptic situations, which can be characterized (according to the catalogue of P. Hess and H. Brezowsky) as (GT) types TM, NW, E, SE+S, or (according to Fr. Rein an M. Konček's catalogue) as types Bc, Cc, Ec, SWc2 and NWc. Increased frequency of these situations occurs at the beginning and towards the end of the month of October. Diurnal totals of precipitation in these situations are so great that October registers a high monthly total, although most days in October have otherwise low diurnal amount of precipitation. In the diurnal variation of the October precipitation amounts a significant change has also taken place during the past fifty years. Thus, in the period 1901/1925, the maximum of precipitation occurs at the beginning of the month (about October 5th) and towards the end of the month there is little rainfall; in the period 1926/1950 the amount of precipitation is small at the beginning of the month and the maximum occurs towards the end (about October 25th). In both cases the minimum of precipitation occurs in the middle of the month (about October 15th). A change in the period referred to has also taken place in the variation of synoptic situations. The variation of precipitation is not much affected by the most frequent synoptic types W(GT) and HM(GT), but it agrees strikingly with the variation of otherwise infrequent situations, such as TM and NW. The last two situations should be regarded as the cause of the October precipitation singularities referred to. The variation of precipitation in Hungary and in the greater part of the Rumanian territory has similar characteristics as in Czechoslovakia. But there appears also a secondary maximum at the end of the month in the period 1926/1950. Southern and eastern Rumania has a different regime of precipitation. The above mentioned three distinct (in regard to precipitation) sections of October, which roughly coincide with three October decades, have been analysed according to the degree of similarity in precipitation. The synoptic basis of this method has been justified by M. Bouët, and in our study we have only modified the conditions for its climatological application. The maps of the degrees of similarity demonstrate that the region of moderate to considerable degree of similarity covers roughly the basin of the rivers Morava and Vah with Nitra and also north-~ astern and central Hungary. The remaining regions show a low degree of similarity or have an indifferent or even opposite regime of precipitation. In the above mentioned region there occurs roughly the same climatological fluctuation of the precipitation regime. Research of this type is important for a dynamic conception of the term of physiographical milieu in which a change of the precipitation regime affects changes in the hydrological regime with all consequences for biogeographical conditions. Besides its contribution to the theory of climate and to the rising science of climate forecast it is of great value for studies concerned with methods of climatological classification.