actually, there are many ways to build Arlequin files. However, it would be useful to know what kind of genetic data you have. A general recommendation I could give you is to try to format your data in the GenAlex (http://biology-assets.anu.edu.au/GenAlEx/Welcome.html) which is way easier to visualize your data. After generating your Arlequin data file, when you load it in Arlequin, never forget to check the structure settings tab. If you want to test differences between groups of populations, you have to set it there.
In pages 44-50 you have several examples of input files depending on your data. You can take one of the example files and just modify it with your data:
Thanks Wilson and Noemi, your guidelines were really helpful. I can now preper arliquin input file of various forms and run my analysis in several ways. Once again, thank you.