My orchestration of this early waltz by Ziehrer, dating from 1865. This probably needs some explanation as there are numerous postings of a version of this on line. I came across it when I was looking for some early Ziehrer to hear his orchestration at this period. The versions of this piece online do not say that it is an orchestration, although the original CD does. I got through the introduction with only a few doubts about its authenticity but even the first bars of the waltz made me pretty sure it wasn’t Ziehrer’s orchestration. It also didn’t sound like an 1860’s orchestration. This is not a criticism of the work as it is a very good orchestration, but it is not the approach I would take. The only slight issue I have is that I think it is too slow. I am not alone in this as one of the posters has speeded the waltz section up. I prefer that version to the original.

I think there are broadly three ways to approach an orchestration of the piano score of a lost waltz:

a) Do a thorough study of surviving contemporary orchestrations by the composer, if they survive, with a view to discovering patterns in his use of instruments, or studying scores by contemporary composers to determine the approximate style of the period if not;
b) Study scores and recordings that are readily available to get a reasonable idea of the composer’s work of the period and also that of his contemporaries;
c) Use your general experience to produce a working orchestration.

All of these approaches produce output that is better than a piece gathering dust on a library shelf. I am usually in the “b” group. With some composers, the Lanners in particular, I have seen a lot of scores but even with them I have not done a rigorous analysis of their methodology. I sometimes drift into the “c” camp too, although I try to resist this. (An example is the first eight bars of this piece which are probably a bit Wagnerian. I couldn’t resist the scrunchy harmonies in bars two and three. If I’m honest, the first eight bars of the other version are likely to be nearer to the original.)

I decided to have a go myself, with an 1860’s orchestra and an 1860’s mindset, if there is such a thing. The three big names of the period are Johann Strauss II, Josef Strauss and Philipp Fahrbach I. Their styles are completely different but they do have some common ground.

Some general guidelines are –

a) They don’t use much percussion. Percussion is used for effect rather than as a general accompaniment. “Cymbal on the first beat and then side drum on the other two beats” is rare at this period and I can’t think of any example in a quiet passage except of four bars in Johann’s Künstlerleben where it is definitely for effect.
b) Waltz tunes are almost invariably led by the first violins, perhaps with wind doubling. I can’t think of a single example of a main waltz theme played by a solo wind instrument. There must be some.
c) Avoid the oboe. I don’t know why but every waltz composer from Lanner on seems to hate the thing. The transcription of a Fahrbach waltz I did recently didn’t even have oboes in it at all. Johann uses it occasionally as a solo instrument in introductions (e.g. Flugschriften, Künstlerleben) but, even there, when the tune comes around again in the waltz it is given to strings.
d) Keep it light. Josef has a very powerful orchestra available and Fahrbach has very heavy brass but both use that power sparingly. Johann is even lighter. By the 1890’s you can throw the kitchen sink at it, but not in 1865.

Working from piano scores has its problems. They were cheaply and quickly produced and so there are errors and omissions. It’s sometimes not possible to determine whether it’s an error or omission or something deliberate. Consequently, we have “corrected” different things in our orchestrations. Also, the piano score is written for a good amateur pianist who has a limited number of fingers and can’t possibly play everything a full orchestra can play. Sometimes you have to guess what’s missing.

We have possibly both wasted our time as I have discovered an entry in the Austrian National Library catalogue that suggests they have Ziehrer’s original manuscript score but if that’s the case, why did the record company use an arrangement?

The other version is here - https://www.youtube.com/watch?v=JCSnRQbDvgY