Splitting word across lines

[MarcoP]

possibly the VMs could be inspected with regards to which words can be compiled across line breaks

It is unlikely that words were split in the VMS. Any splitting would force the average length of line first words and line last words to be smaller. But in fact last words are average and first words are longer.

It's tricky. When You are not allowed to view links. Register or Login to view., I found something different: last words do not vary much, while first words tend to be longer. My guess is that this is because English (like most natural languages) tends to alternate Short Long Short Long. Short words can easily fit at the end of a line, so Long words are more likely to occur at line-start.

But researchers among whom Bowern pointed out that Voynichese has auto-correlated word-lengths, like: Short Short Short Long Long Long, so things might be different. 

Anyway, I agree with the conclusion that Voynichese words do not seem to be split across lines. I vaguely remember I tried merging words across lines and I didn't find a particularly high rate of actual words.