Before I decide to post this question I did google it, and there are at least two similar questions on Academia with very good and informative suggestions. Since the situation and my concerns are slightly different than the existing ones, I hope it's ok to ask a similar question.
I was writing a paper in pure math (while the situation of the existing questions is that the author already submitted his paper and was asked to split the paper into two). Initially the things in my mind is to prove two theorems using the same method, so it does make sense to put the results in one paper. However, when I was working on the paper I found that I need to modify/extend/generalize the existing background setting which requires some work. Let's say the two theorems I have in mind are in field A and the background setting is in field B. Field A and B are different but closely related. I have papers in these two fields, but I would say I work on field A more often than in field B.
I am not sure if I should put all the results in one paper or in two separate papers. The reasons are
- according to the answers to the similar questions posted on Academia, if I put it in one paper it could be difficult to find referee(s). I would say there are less than 20 people specializing in both of these two fields.
- however, if I put it into two papers, I anticipate the motivation of the paper about modifying/extending/generalizing the existing background setting, although requires some work, may be underestimated (no matter how hard I try to promote it).
- based on the questions posted on Academia, if I put it into one paper then there is a good chance that I would be asked to separate the paper or even a rejection.
One interesting thing is no matter I write one or two papers, the number of suitable journals does not increase nor decrease drastically. But if I put it into two papers, I would possibly submit these papers to different journals.
I would not fully disagree if you think I worry too much, and tell me I should just go ahead in either way and see what will happen. Anyway, any suggestion is welcome.