I think you want this.
When you say integration, I assume you mean finding the area bound by the function and the axis, between and .
Suppose that you made subdivisions where and . Also mark the midpoint of each subdivision so that .
Then the area of each rectangle formed by the subdivisons is , and the entire area can be approximated by
The more subdivisions we make, the closer this approximation gets to the exact area. Therefore
Now we need to examine the Mean Value Theorem.
for some .
We can rewrite this equation as
Notice that the LHS is of the exact form of the summand for your area. We can use the Mean Value Theorem to simplify the summand, since we are making the subdivisions infinitessimally small so that the midpoint is the only point in between.
since the function no longer depends on .
Therefore the area bounded by a function and the axis between and is , where is an antiderivative of .
My question would be "how are you defining 'antidifferentiation' and 'integration'"?
If you are using the usual definitions, that the "antiderivative" of a function f(x) is a function F(x) such that F'(x)= f(x) and "integration" is " = the area of the region bounded by the curve y= f(x), x= a, x= b. y= 0 (assuming f(x)> 0 for x between a and b)", then you are trying to prove the "Fundamental Theorem of Calculus".
Any text book on Calculus should include that. Wikipedia has a good discussion:
Fundamental theorem of calculus - Wikipedia, the free encyclopedia
What Prove It posted is a nice outline of that proof.