20, 24 or 26 staves, no makers' marks recorded, 350 x 270 (r=?)

No detailed examination of the manuscript was possible

Sold at Sotheby's, 10 May 1984, lot 86. The bidder was anonymous, and the current whereabouts of the manuscript are unknown.

Facsimiles of the title page, and bb. 362–8 of the finale (black & white), and bb. 102–9 of the finale (colour) in the Sotheby's catalogue of the sale on 10 May 1984.

Hn in F 1–7, tpt in F 1–4, trb 1–3, btuba

Timp (1 player), [?trgl], cym, bd

[Harp], strings

This information is derived from the two pages reproduced in the Sotheby's catalogue.

See SMFS, 107–8 for a tabular summary of the gradual expansion of the instrumentation of the work in ACF1, AF2 and ACF2.

A four-movement version of the work, lacking Blumine, which presumably used ACF2 and the revision short scores (AR1, AR2) as its copy texts. The unavailability of this source is particularly frustrating as it leaves so many questions unanswered – not least, those concerning its own date and role in the early creative and performance history of the work. However, from the detailed description in the Sotheby's catalogue it appears that from the outset this copy was in four movements only, so must have post-dated the Weimar performance of the five-movement version in June 1894, and may have been prepared specifically for the first performance of the four-movement version on 16 March 1896 in Berlin. The date of the extensive revisions is uncertain: if this score was used for the Berlin performance one would expect two layers, one made during rehearsals and a second (probably more extensive) made after the performance.