Moving parts

In continuing just what everyone was expecting, I brought into the score various sketches.  At the end of measure 122, I slotted in (what is now) mm. 135-145, and immediately after, (what is now) mm. 149-160.  Didn't like it.  It felt like an entire mess.

There was another sketch in my MS., though, of a marimba passage with a pattern, a sort of mosaic with tremolo notes separated by single attacks.  This passage I did not mean as the marimba solo passage, but was one part, and I had in mind adding the clarinet to it.

So, I interposed the 'tremolo mosaic', plus a clarinet part I improvised, at mm. 123-134;  and then a variant of this for mm. 146-148.  Completely pleased with the result, which (curious to observe) sounds as though all these pieces were designed to flow like this.

After all this, I had three measures of clarinet alone, essentially three sustained notes, and then a quick semiquaver sort-of-descending-arpeggio.

At this point, I had this strange feeling of completely owning all the piece up to m. 160, and then wanting to throw out the three mm. of clarinet alone;  thinking that I would toss that, and start afresh.

There was no hurry, though, so I let the matter rest.

As the neurons cooled, I was again aware of the valuable distinction between baby and bathwater . . . and came to feel that the only thing which bothered me about those final three measures, was the rapidity of that sort-of-descending-arpeggio.  So I feel that all that wants doing, is to alter the rhythmic profile of the sort-of-descending-arpeggio, and that the character of this altered version will pretty much drive the ensuing section.

Success here has resulted from a combination of the freedom to feel dissatisfaction, plus some patience, plus questioning just what it might be, exactly, that I am dissatisfied over.