The Student Room Group

A question about analysis...

In this pdf document:

http://math.berkeley.edu/~scanez/courses/math104/fall11/homework/hw7-solns.pdf

For the proof of quesiton 2, the text says "First, let r in Q. Since the irrationals are dense in R, there exists a sequence (yn)(y_n) of irrational numbers converging to r. Since the sequence f(yn)=0f(y_n) = 0 does not converge to f(r)0f(r) \not=0, f is not
continuous at r."

I'm a bit confused about this. Becuase the irrationals are dense, we know that there infinetely many irrationals around r...but how does that there there is a "sequence" that "converges" to r?

Also how do we know that f(yn)0f(y_n) \not=0?

Thank you indvance.
Original post by Artus
In this pdf document:

http://math.berkeley.edu/~scanez/courses/math104/fall11/homework/hw7-solns.pdf

For the proof of quesiton 2, the text says "First, let r in Q. Since the irrationals are dense in R, there exists a sequence (yn)(y_n) of irrational numbers converging to r. Since the sequence f(yn)=0f(y_n) = 0 does not converge to f(r)0f(r) \not=0, f is not
continuous at r."

I'm a bit confused about this. Becuase the irrationals are dense, we know that there infinetely many irrationals around r...but how does that there there is a "sequence" that "converges" to r?

Well, if you take say 12 \frac{1}{2} . You know that for each n there are infinitely many rationals inside (121n,12+1n) (\frac{1}{2} - \frac{1}{n}, \frac{1}{2}+\frac{1}{n}) . If you pick one of these rationals for each n, then the sequence of rationals that you pick will converge to 1/2. Actually, you can do the same for any rational number, not just 1/2 and that gives you that any rational number has a sequence of irrationals that tend towards it.



Also how do we know that f(yn)0f(y_n) \not=0?

Thank you indvance.

Where does it say that? If you mean f(r)0 f(r) \neq 0 then this is because r is rational and 1q0 \frac{1}{q} \neq 0 for any qZ q \in \mathbb{Z}
Reply 2
Given that the irrationals are dense, can you show that given any epsilon > 0, there is an irrational less than epsilon away from r? And once you have showed this, can you use it to show that you can construct a sequence tending to r?
Reply 3
Original post by IrrationalNumber
Well, if you take say 12 \frac{1}{2} . You know that for each n there are infinitely many rationals inside (121n,12+1n) (\frac{1}{2} - \frac{1}{n}, \frac{1}{2}+\frac{1}{n}) . If you pick one of these rationals for each n, then the sequence of rationals that you pick will converge to 1/2. Actually, you can do the same for any rational number, not just 1/2 and that gives you that any rational number has a sequence of irrationals that tend towards it.


Where does it say that? If you mean f(r)0 f(r) \neq 0 then this is because r is rational and 1q0 \frac{1}{q} \neq 0 for any qZ q \in \mathbb{Z}


Thanks a lot :smile:

Quick Reply

Latest