C is derived from both A and B - so before the constructor for C can be executed, the constructors for A and B must both be completed. If that wasn't the case, then the C constructor could not rely on anything that its base classes provided.
For example, suppose A contained a list of items which is created and filed from a database in its constructor. All instances of A can rely on the list being complete because the constructor has to be finished before the instance is available to the rest of the code. But C is also an A - it derives from it in the same way that "a Ford" is also "A Car" - so it may well want to access that list. When you construct an instance, the base class constructors are automatically called before the derived classes to ensure that everything is ready for action when the derived constructor starts.
For example, change your code slightly:
class A
{
public:
A() { cout << "A's constructor called" << endl; }
};
class B: public A
{
public:
B() { cout << "B's constructor called" << endl; }
};
class C: public B
{
public:
C() { cout << "C's constructor called" << endl; }
};
int main()
{
C c;
return 0;
}
and you will get:
A's constructor called
B's constructor called
C's constructor called
Because the base class constructors are all completed before the derived ones are executed.