TITLE: A series of articles should be published all in the same journal, or in different journals?
QUESTION [37 upvotes]: There is a series of papers, which build on each other, to be published. The series starts with purely mathematical articles, and ends with articles oriented to some applications. What would be the best approach:

Submit all to the same journal. If so, what kind of journal would be best, a purely mathematical one, one of applications, or something between?
Submit them to different journals, possible ranging as well from pure mathematics to applied mathematics.
Try to make two series, one purely mathematical, and the other oriented towards applications, and send them to two journals.
Other solution.

What advantages and disadvantages has each solution?

REPLY [3 votes]: An old question, but an interesting example is Robertson & Seymour’s monumental series of papers Graph Minors I, Graph Minors II, Graph Minors III and so on up to Graph Minors XXIII.
All published in Journal of Combinatorial Theory, Series B EXCEPT for Graph Minors II, which was in Journal of Algorithms.
I guess we can count that as a vote for the “all in one journal” option.