In our recorded performance the bass line of the obbligato harpsichord part was reinforced by a viola da gamba. Reinforcement, albeit with a cello, was normally only usual in the case of works with basso continuo in order to bring out the outer parts (bass and solo upper part), with the fill-in chords of the harpsichord (organ, long archlute etc.) receding into the background. As regards ohbligato harpsichord parts (as for instance in the slow movements of the Triple Concerto or the 5th Brandenburg Concerto) Bach expressly waived reinforcement of this kind. However, one of the seven important sources (the only one with autograph pages in the harpsichord part) refers to the composer’s wish for reinforcement of the bass line with Viola da Gamba, “if so desired” (“...per Viola da Gamba se piace”). This remark is so unusual as regards style and makes performance so much more difficult that we cannot assume that it was entered by a copyist. One can more likely conclude that it was a very modern idea on Bach’s part, his intention being to make the three-part movement - which dominates in these sonates - more colourful and outstanding by brightening up one of the two harpsichord parts which otherwise would sound the same: firstly violin, secondly harpsidiord and thirdly viola da gamba/harpsichord. Of all the stringed instruments in the bass position, the viola da garnba is most suitable for this purpose because it sounds much more gentle than the cello and its tone fuses mudm more easily with that of the harpsichord. For many movements in which the equality of the parts is not so pronounced, where the bass of the harpsichord has more of a continuo character, the bass-reinforced sound pattern is in any case the “norm.” Bach's idea, if we accept that the note stems from the composer, already points to the early Haydn and Mozart piano trios, where the cello has a similar function to the viola da gamba in the present case.
Sonata II, 2nd movement, bars 12 and 104, the e” is already played on the second crotchet and tied to the next bar (according to three of the six sources concerning the violin part); bar 23, last quaver, instead of b', e" (tied over) is played for analogous reasons and on the basis of the J. Ch. Altnikol copy, the rnain source of the New Bach Edition; for the same reasons in bar 41 the last three semi-quaver notes are played as b” e” and a#”. Bar 92, on the basis of two sources the chord is played as e' b' g#”. Sonata III, 1st movement, bar 17: the rhythmic structure of the sources varies considerably here; a “certain approximation”, a quasi improvisando, is evidently intended. 4th movement, where they are not the motif, but are performing an accompanying function, the quaver notes in the bass are adapted to the triplets of the violin part (bars 35 to 77), as is also noted in some of the sources. 3rd movement, the dotted quavers and semi-quavers of the violin part and of the bass are adapted to the triplet motion.
Sonata VI, movement in bar 32 (harpsichord) and 54 (violin) is played in accordance with the main sources:

Bar 79, printing error: the fourth note of the violin part should be a semiquaver.
Naturally the note values, as Well as the interpretation of the arpeggios, are performed not on the basis of the music pattern but according to the practical performance convention of that period. The articulation was supplemented according to the various sources and the practical performance principles of the Bach era.