Please note that this is thread is 6 years old. I did however head over and read the link which I assume must be your blog. You have some VERY incorrect information there.

This is 100% incorrect. While it is likely that your select will return the rows in order of the primary key there is absolutely no guarantee that this will happen or that this behavior will continue in future releases. There is one and ONLY one way to ensure the order of your data, add an ORDER BY clause to your query.

In addition to incorrectly stating that a primary key has anything to do with the order of retrieval you also are making an assumption that somebody trying to learn this could be very confused by. It is obvious that you are making the assumption that your primary key is also your clustered index. This is the default for a primary key but is NOT required. The ordering behavior you are describing is actually based on the clustered index and not the primary key.

Need some proof that the ordering is not going to work? Check out this article. http://blogs.msdn.com/b/conor_cunningham_msft/archive/2008/08/27/no-seatbelt-expecting-order-without-order-by.aspx

It will walk you through this fallacy of primary key (clustered index) ordering and show you exactly how to prove that it is inaccurate.