This is a longish quote from a paper on the subject. It should give a hint at the main problem with figuring out exactly what happened...--CV

**II. ON THE DIRECTION OF THE TCB FLIGHT

The first investigators of the TCB (L.A.Kulik, 1939; E.L.Krinov,1949; and I.S.Astapovich,1951) who analyzed evidence of the object's flight in the area of the Angara River had no doubts that it had been moving generally from the south to the north. However, there were three options for a more precise trajectory: a southern trajectory, proposed by L.A.Kulik, a southeastern one proposed by E.L. Krinov, and the southwestern path proposed by I.S.Astapovich. (See Sytinskaya, N.N., 1955, and Levin, B.Yu., 1954 as well). By the early 1960s it was Krinov's trajectory, namely 135ø, that was considered most realistic.

Later, however, a "corridor" of axis symmetrical deviation of the vectors of the felled forest from the dominating radial pattern was revealed. This deviation was interpreted as the track of the ballistic wave ( Boyarkina, A.P.et al., 1964; Fast V.G., 1967; Zotkin, I.T., 1966; Zolotov, A.V., 1969). The direction of the "corridor," wich was initially estimated as 111ø,E from N (114ø, east of the true meridian) (Fast, V.G., 1967), was later found to be 95ø, E from N (99ø, east of the true meridian) (Fast,V.G. et al., 1976). During this period of time a number of elderly residents of the area who had lived in the upper reaches of the Nizhnaya Tunguska in 1908 were questioned. No eyewitnesses from this area had been questioned in the 1920s and 1930s. This resulted in the conclusion that the object had moved from the ESE to the WNW (Konenkin, V.G., 1967), i.e. by the path coinciding with that of the trajectory of the TCB. This coincidence caused revision of the notion of the TCB trajectory, so since the year 1965 the ESE-WNW option has cominated the literature. For several years it was assumed to be a final result.

But later the publicaton of a catalog of eyewitness accounts made analysis of the whole event possible. (Vasilyev, N.V. et al., 1981) Two fundamental facts were established:

1. "Images" of the bolide observed in the area of the Angara Rriver and observed in the area of the Nizhnyaya Tunguska River are quite different, and the evidence seems to indicate that they belong to different objects (Demin, D.V. et al., 1984).

2. The trajectory, calculated on the basis of Angara eyewitness accounts, differs considerably from that determined by analyzing the vector structure of the felled forest area and the radiant burn area. Indeed, evidence of the Angara eyewitheness, including the report of a district officer, suggests that the bolide flew "high in the sky," which is hardly consistent with the path 99ø east of the true meridian. On the contrary, the data obtained on the Nizhnaya Tunguska River, even though agreeing with the shape the area of devestation, is in contradiction with the Angara observations.

Another complication is that the Nizhnaya Tunguska data suggest, virtually unambiguously, that the bolide's flight occurred in the afternoon, whereas the data from the Angara River identify it as an early morning flight. Attempts to resolve the conflict between both sets of data present considerable problems.

In the search for an exit from this maze, more then one approach has been tried. Some researchers practically ignored eyewitness accounts as unreliable subjective material. This approach could be agreed with to a certain extent if only it were a matter of a few inconsistent accounts, whereas there were many hundreds of independent accounts.
**

http://www.galisteo.com/tunguska/docs/tmpt.html