Abstract. This work compares cometary and solar wind data with the purpose of determining the solar wind conditions associated with comet plasma tail disconnection events (DEs). The cometary data are from The International Halley Watch Atlas of Large-Scale Phenomena (Brandt et al. 1992a). A systematic visual analysis of the atlas images (Voelzke & Matsuura 1998) revealed, among other morphological structures, 47 DEs. The solar wind data are in situ measurements from IMP-8, ICE and PVO, which are used to construct the variation of solar wind speed, density and dynamic pressure during the analysed interval. Data from these same spacecraft plus Vega 1 were used to determine the times of current sheet crossings. These data were fitted to heliospheric current sheet curves (Hoeksema 1989) extrapolated from the corona into the heliosphere in order to determine the best-fit coronal source-surface radius for each Carrington rotation. This work presents the kinematic analysis for two onsets of DEs (1985 December 13.5 and 1986 February 22.2). The DE onset time, when the comet supposedly crossed a magnetic sector boundary, is determined assuming that the disconnected plasma moves away from the nucleus at constant velocity (Voelzke & Matsuura 1998). This is compared with the analysis of (Yi et al. 1994b) which determined the time of disconnection assuming constantly accelerated linear motion. The velocity varies broadly from one DE to another. In this paper, we report the results of an analysis of four DEs observed from 13 through 14 December 1985 and on 22 February 1986. From a kinematic analysis of the widefield photographic images we calculated the disconnection time of the 1985 December 13-14 DEs to be December 13.5 and of the 1986 February 22 to be February 22.2, which are in good agreement with Yi et al. (1994b). The solar wind conditions around comet P/Halley, inferred by corotation of IMP-8 data to the comet, were such Send offprint requests to: M.R. Voelzke, e-mail: mvoelzke@astro.uni-bonn.de that comet P/Halley had just crossed the interplanetary magnetic field (IMF) sector boundary; the solar wind density was, respectively, 15 cm −3 and 7 cm −3 ; the solar wind speed was, respectively, 370 km s −1 and 500 km s −1 . We conclude that our results corroborate the idea that DEs are associated with sector boundary crossings and that the magnetic reconnection plays an important role in the formation of DEs and can be considered as the triggering mechanism (Niedner & Brandt 1978;Brandt et al. 1999).