# Is the result always n+1?

I'm reading a book on algorithms by Kurt Mehlhorn and Peter Sanders. On page 2, the following Theorem is stated:

The addition of two n-digit integers requires exactly n primitive operations. The result is an n+1-digit integer.

Is the result always n+1. What if you have 2 2-digit integers like 11 and 22 or others like that?

-
At most $n+1$ digits (also, maybe the book considers $033$ as a $3$ digit number), and I don't think the last two tags are appropriate for this question – Belgi Aug 5 '12 at 4:32
To use the decimal system as an example, consider the case when you're adding $\underbrace{99\dots99}_{n\text{ times}}$ to itself... – J. M. Aug 5 '12 at 4:33

In fact, the text here makes this same point right before starting the section on addition: "We can artificially turn any n-digit integer into an m-digit integer for any m≥n by adding additional leading zeros. Concretely, “425” and “000425” represent the same integer. We shall use a and b for the two operands of an addition or multiplication and assume throughout this section that a and b are n-digit integers. The assumption that both operands have the same length simplifies the presentation without changing the key message of the chapter. We shall come back to this remark at the end of the chapter.We refer to the digits of a as an−1 to a0, with an−1 being the most significant digit (also called leading digit) and a0 being the least significant digit, and write a = ($a_{n−1} . . . a_0$). The leading digit may be zero."
To see why n+1 digits are sufficient here, if you are working in base b, then any n-digit number is less than $b^n$, so the sum of two such numbers is less than $2b^n$, which is less than or equal to $b^{n+1}$ if b is at least 2. Being less than $b^{n+1}$ means that n+1 digits are sufficient to represent it.