Because, MP3 files usually add a bit of silence at the beginning and end of file.
If you used Feurio, you would have been told this when you tried that:
[i]"This is caused by the MP3-format:
The mp3-format is a kind of compression with some loss of data, i.e. not the whole samples are stored, but - expressed in a simple way - some kind of "frequency curves", i.e. the mp3-encoder analyses a certain range of data and then stores the "frequency curves" for this range.
If now e.g. a live recording is stored in parts - one part in the first file, the following part in the second file - and compressed to the mp3-format, the encoding process is resetted for the second file - so the frequence curves (the end of the first file and the beginning of the second) don't match exactly.
In addition most mp3-files are "frame orientated", i.e. the encoder sometimes adds empty samples to the end of a file to get a whole frame at the end.
To put it together: Unfortunately it is not possible to always reconstruct the original data 100%-ly out of two mp3-files, that contain parts of continuous data.
If you want to work with continouos data, you must save all tracks into ONE mp3-file or use the wave-format." [/i]
So, if you phisically divide the MP3 file in several files, you would have to remove those silences manually with an editor and then burn with Feurio and "Do not insert pauses between tracks" (Nero can't burn that kind of modified files correctly).
The best is to use a non-destructive editor like the Nero's "Limits, Split, indexes" tab in the properties of the track. If you do this a lot of times, I recommend Feurio's Track Editor and "Split Track (linked)", wich is faster, easier and you don't have to remember to set the pauses to zero.
And some versions of Nero fail when you save that compilation. So, do it just before burning.