WN 37, Lento con gran espressione

In the main text we adopt the notation of CK, actually equivalent to CJ, yet more natural to interpret. It is difficult to say whether the absence of the indication in CB is an oversight or – which seems to be more likely – a purposeful omission (the indication may be regarded as superfluous after  in b. 33). The version of EL may be considered alternative in relation to our interpretation of CK – it was the beginning of sempre that was deemed reliable, whereas in our transcription we kept the placement of .  

The additional notes in the R.H. part present in b. 35-36 and 39-40 in CK – c²-d², echoing an authentic ending of the melody in b. 32-33 – are most probably Kolberg's inauthentic addition, since they are absent both in A1 and CJ, based on [A2], just like CK. CB also confirms the lack of authenticity thereof – Balakirev did not see them in the copied manuscript that was either CK before the notes were added or another copy of Kolberg in which they were simply absent. These motifs, in an already elaborate form, appear in EL in b. 35-37 and – moved an octave lower – in b. 39-41. 

These 9 bars were written by Chopin on the bottom stave only both in A1 and [A2] (→CJ); CB also applied such notation. At the same time, all sources – including CK and EL, in which a few/several dozen notes and rests were added on the top stave – convey the general idea of division into hands expressed through the direction of the stems and arrangement of the rests. In turn, as far as the details are concerned, there are many differences between them, which, however, are of no practical meaning; some of them are simply mistakes or simplifications related to the notation of repeated figures. Due to this reason, we do not discuss them in detail, leaving a possible analysis to the reader. The version suggested in the main text is based on CJ; we only complement the notation of b. 37-38, in which some of the L.H. rests were overlooked.

In the main text we stick to the notation of CJ, since Kolberg could have moved the indication dim. so that it does not contradict the continuation of the  dynamics, valid half a bar ago. Such a literal interpretation of these indications would also explain the omission of dim. in EL.

The fact that b. 42-43 are marked in CJ and CK as repetition of b. 40-41 generates uncertainty as to the placement of the indications sempre (più) piano and rallent. written in b. 39-41. According to us, the dynamic indication that begins in b. 39 must be continued right away in b. 40-41. In turn, it is not so certain in the case of rallent., which in the notation of CJ seems to be related to adagio in b. 44. Due to this reason, we suggest two possible interpretations for rallent. It may also be the reason for the difference between the sources based on Kolberg's copies – in CB rallent. is in b. 43, whereas in EL – in b. 41.

In the main text we give the literal interpretation of CJ:

• we consider sempre più piano to be a more likely interpretation of [A2] than sempre piano – the latter is a repetition of the indication from b. 34-35, which seems to disturb the continuity of the diminuendo that appears between them. Moreover, it is easier to envision an oversight of one out of two similar words (più and piano) than an addition of a word that is absent in [A2];
• an earlier appearance of rallent. offers greater possibilities to shape it in a deft manner.