In bars 90 and 92 and in their written-out repetitions, the 1st R.H. chord has 4 versions in the sources. The table below presents their appearances in respective impressions of PE and in GE:
| b. 90 | b. 92 | b. 90/114 | b. 92/116 | |
|---|---|---|---|---|
| PE1 | – – – –  – – – – |
|||
| PE2 |  |
– –  – – |
||
| PE3 PE4 |
– – – – – – – – |
|||
| GE | – – – –  – – – – |
|||
The authenticity of all three versions of PE is questionable:
- In Chopin's pieces, combinations of parallel major and minor chords generally occur as an element of modulation and never create uncertainty as to the prevailing key, as is the case with the first version. Therefore, it probably resulted from a misinterpretation of the manuscript or – which, according to us, is most likely – from a revision of [A] consisting in adding a
in a place where Chopin omitted a 
, 'obvious' in the face of the intended repetition of the previous chord. - The second and third versions are a later revision, introduced in stages – in PE2 only in bar 90, while in PE3 (→PE4) in the remaining three. The d
1-g1 fourth appearing in PE2 almost certainly resulted from the action of starting work on the introduction of an a-c1-g1 chord, which, however, was interrupted halfway through, for unknown reasons – the superfluous accidentals were removed (the b note with a , absent in the target chord), but the stage of adding new elements was omitted (the a minim and the ledger line crossing the d1 minim, which is enough to turn it into c1 – one can see in PE3 that this is what was done).
It is difficult to say what led to such a change. Were so many mistakes – at least two – committed in PE1 while reading the supposed source of the text of the Waltz, that is [A]? Or maybe within the two years that probably separate PE2 from PE1 another source was found, considered more reliable? In the light of the alleged circumstances of the creation of the piece – see the description of PE1 – both hypotheses seem unlikely.
Taking into account the above observations, we consider the version of GE to be the most likely reconstruction of [A], performed on the basis of PE1, which we thus suggest in the main text.
Compare the passage in the sources »
category imprint: Differences between sources; Editorial revisions
issues: GE revisions, Errors in PE,
notation: Pitch