Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

This is an exercise from Miles Reid, Undergraduate Algebraic Geometry. The proof has two parts, one of which I can do and one of which I can't. I could have some misunderstandings about notations here as well so anything people can point out will be helpful.

a) Let $k$ be an infinite field and $f\in k[X_1,...,X_n]$ be nonconstant. Prove $$V(f)=\{P\in \mathbb{A}^n_k|f(P)=0\}\neq \mathbb{A}^n_k.$$

$A^n_k$ here is the point set $k^n$.

My proof (which follows from a hint of his) is write $f=\sum a_i(X_1,...,X_{n-1})X_n^i$. Then $V(f)$ consists of two kinds of points

  1. $P=(p_1,...,p_{n-1},p_n=0)$
  2. $P=(p_1,...,p_{n-1},p_n\neq 0 )$ with $\sum a_i(p_1,...,p_{n-1})p_{n}^i=0$

In the first case we have $V(f)\neq \mathbb{A}^n_k$. Assume we have $P$ like in the second case and now write $f=\sum b_i(X_1,...,\hat{X}_{n-1},X_n)X_{n-1}^i$. Here again, either $P=(p_1,...,p_{n-1}=0,p_n\neq 0)$ and $V(f)\neq \mathbb{A}^n_k$ or $p_{n-1}\neq 0$, and by induction we should get to $V(f)\neq \mathbb{A}^n_k$.

b) Now let $k$ be algebraically closed. Let $a_m(X_1,...,X_{n-1})X_n^m$ be the leading term of $f$. Show that when $a_m\neq 0$, there is a finite set of points of $V(f)$ corresponding to every value $(X_1,...,X_{n-1})$. Thus, $V(f)$ infinite for $n \geq 2$.

So I have kinda no idea what to do. A given value of $(X_1,...,X_{n-1})$ corresponds to points in $V(f)$ $P=(p_1,...,p_{n-1},p_n)$ where $f(P)=0$, but I don't know anything about the value of $p_n$; if it vanishes then clearly there is a correspondence $(X_1,...,X_n)\to (p_1,...,p_{n-1},0)$ but I guess I need to show there are a finite number of $p_n$ where this is true and I don't know how.

share|cite|improve this question
For (b) given any fixed value $(X_1,\ldots,X_{n-1})$, then you're just looking at a polynomial in the variable $X_n$ no? – user38268 Aug 23 '13 at 23:21
up vote 3 down vote accepted

For the first part, fix $a_1,...,a_{n-1}\in k$ and consider the polynomial $g(x)=f(a_1,...,a_{n-1},x)$. Then $g$ is a polynomial in one variable so its solution set is finite. As $k$ is infinite, this means that there exists $a_n$ such that $g(a_n)\not= 0$. Then the point $(a_1,...,a_n)$ is in $\mathbb{A}_k^n$ but not in $V(f)$.

For the second part, note that for any $a_1,...,a_{n-1}\in k$ the polynomial $g(x)$ from above must have at least one solution. This is because $k$ is algebraically closed. Since all algebraically closed fields are infinite, there are infinitely many choices for the first $n-1$ coordinates.

share|cite|improve this answer
Dear jkramerm47, what about $f(x_1,x_2)=x_1\cdot x_2$ and $a_1=0$: do you think that the solution set of $g(x)=0\cdot x$ is finite and that you can find $a_2$ with $g(a_2)=0\cdot a_2\neq 0$ ? – Georges Elencwajg Aug 26 '13 at 18:31
He didn't say it explicitly in his answer, but part of the assumption for the proof is that $a_m\neq 0$ for highest $m$. – levitopher Aug 31 '13 at 5:08

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.