The set of events that are observed after the magnetopause crossing shows distinct evolution of the associated plasma parameters while maintaining a similar double structure in the leading and trailing parts. The plasma in the leading part of the first plasma transient at ~ 23:00 UT is indistinguishable from the magnetosheath plasma, while the trailing part is hotter and more diluted. The plasma transients observed at ~ 23:35 UT and at ~ 23:55 UT are progressively less dense, hotter, and move more slowly than the plasma feature observed close to magnetopause (at ~ 23:00 UT). Finally, the plasma transient observed at ~ 00:08 UT on February 16, 1996, is nearly stationary in the rest frame of the magnetospheric plasma and also consists of two parts. This monotonic change in plasma properties of the observed transients with distinct double structure is most easily seen on the Ti-Ni scatter plot (Fig. 3).
The first plasma transient at ~ 23:00 UT has a magnetic field profile far different from that seen in a typical FTE signature. The transient observed at ~ 23:35 UT shows many features that are found in transient (3) discussed above (at ~ 23:55 UT). The normal component of the magnetic field changes sign, and the magnetic field magnitude peaks in the region separating the two parts of this transient, as in the case of the event at ~ 23:55 UT. The plasma transient observed at ~ 00:08 UT on February 16, 1996 does not have the distinct magnetic structure of an FTE, and it is at rest relative to the surrounding magnetospheric plasma.
There is substantial evidence that the observed transients are associated with reconnection. We observe nearly anti-parallel magnetosheath and magnetospheric magnetic fields at the magnetopause on February 15, 1996, which is favorable for reconnection. Bipolar variations of the magnetic field normal component in the magnetosheath, (Fig. 1) accompanied by the leakage of magnetospheric ions (at ~ 22:14, 22:28, 22:38 and 22:44), are evidence of ongoing reconnection. Association of these transients with reconnection is confirmed by depletion of magnetospheric ions that is strongest in the first transient and weakest in the last one. Finally, the magnetic structure of two of the events has a typical FTE structure, which is indicative of sporadic reconnection.
There are similarities and differences between transients discussed in this paper and the most studied AMPTE-UKS FTE. While the AMPTE-UKS FTE showed the reversible variation of plasma parameters upon entering the FTE and exiting it, the transient discussed here has quite different plasma characteristics in the leading and trailing parts. We do not observe the central region with a magnetosheath population and with high b i. Unlike Smith and Owen  we did not find distinct D-shaped magnetosheath-like ion distributions within the events discussed in this paper. Thus the plasma transients observed by the Interball Tail Probe in February 15, 1996, including ones which have an FTE magnetic field signature, could not result from the passage of the spacecraft through an active reconnection region.
The plasma transient observed at ~00:08 UT on February 16, 1996, perhaps is similar to the class of "dead" FTEs [Klumpar et. al, 1990]. Thus, Interball Tail Probe gives a clear indication that different classes of FTEs exist. The fact that only two of the four transients have the distinct magnetic profile of an FTE suggests that these plasma transients, which are observed in the LLBL on the magnetospheric flanks, and FTEs represent two different classes of events.
The traveling vortices in the LLBL that are suggested by Sckopke et al.  could be one of the possible explanations for the observed sequence of events. Plasma transient seen at the magnetopause at ~ 23:00 UT, probably may be considered as LLBL adjacent to magnetopause. However, similar observations of plasma transients at the dawn flank of the magnetosphere on September 2, 1995 also do not show the LLBL just after magnetopause crossing [Vaisberg et al., 1997 a, b].
The progressive displacement of the plasma characteristics within the observed transients along the Ti-Ni scatter plot from values closer to the magnetosheath values to those found in the magnetosphere indicates the evolution of the plasma parcels. Similar evolution of the ion moments in plasma transients, observed in the LLBL at the dawn side, is considered to be evidence for plasma penetration from the magnetosheath into the magnetosphere and subsequent dissipation of these plasma clouds [Vaisberg et al., 1997 a, b].
The density, temperature and velocity profiles in the plasma transients discussed in this paper suggest that the leading part is the body of the transient, and the trailing part is, in a sense, the turbulent wake of it. This asymmetry is easily seen on the Ti-Ni scatter plot (Fig. 3). The leading and trailing parts of transients at ~ 23:35 UT and at ~ 23:55 UT are clearly separated from each other.
The absence of clear D-shaped ion distributions inside the observed plasma transients indicate that reconnection on the field lines threading the plasma transients ceased, and that they are no longer connected to the magnetosheath. We interpret these data in terms of magnetosheath plasma transfer and subsequent evolution of transferred plasma parcels in the magnetosphere. This interpretation is reinforced by the fact that while the plasma transients decrease in velocity and density, and increase in temperature, they maintain a basic double structure with respect to the plasma parameters, with a denser and cooler leading part and a less dense, hotter, and more turbulent trailing part. The substantial normal velocity component supports this explanation of plasma transients discussed as detached from the reconnection layer.
We present here a new class of transient events, neither active reconnection nor "dead" FTEs, but an intermediate case, Disconnected Magnetosheath Transfer Events (DMTEs), having distinct double plasma structure. The fact that only two of the four observed transients have the distinct magnetic profile of an FTE suggests that the magnetic field structure also evolves as the DMTE travels inside the magnetosphere in a way not yet fully understood. The evolution of the plasma and magnetic field inside DMTEs requires further study.