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

These are problems from Miles Reid's book "Undergraduate algebraic geometry"

Let $k$ be an algebraically closed field.

Question 1.

Let $I= (xy,xz,yz) \subset k[x,y,z]$. I want to find $Z(I)$ ok it is clear that $I$ is the union of the three coordinate axes, now I want to prove rigorously each coordinate axis is closed and irreducible. Can we simply say:

For example the z-axis is homeomorphic to $\mathbb{A}^{1}$. But any homeomorphism preserves closed sets and irreducible sets so since $\mathbb{A}^{1}$ is algebraic (i.e closed) and irreducible we are done.

Question 2:

Let $I=(x^{2}+y^{2}+z^{2},xy+xz+yz)$. Identify $Z(I)$ and $I(Z(I))$.

I'm confused with this one because here in case the field is $\mathbb{R}$ then we get $x=y=z$ however $\mathbb{R}$ is not algebraically closed so how do we proceed?

Now I was thinking in using the identity $(x+y+z)^{2}=x^{2}+y^{2}+z^{2}+2(xy+xz+yz)$ from here we get that $x+y+z=0$. So it seems $Z(I)$ is the set of all points in the plane $x+y+z=0$. Is this wrong? But what if the underlying field has characteristic equal $2$?

Finally how do we find $I(Z(I))$?

share|cite|improve this question
Dear user6495, it is not true that for $k=\mathbb R$ you get $x=y=z$ as the set of zeros $Z(I)$ in Question 2 since $x^2+y^2+z^2$ takes the value $3a^2$ instead of $0$ at $(a,a,a)$. – Georges Elencwajg Feb 14 '12 at 19:10
up vote 4 down vote accepted

Question 1: The $z$-axis is closed and irreducible because it is the locus of zeros $Z(x,y )$ of the prime ideal $(x,y)$.

Question 2: No, the locus of zeros is not the plane $x+y+z=0$.
Your identity $(x+y+z)^{2}=x^{2}+y^{2}+z^{2}+2(xy+xz+yz)$ proves that you can write $I=(x^{2}+y^{2}+z^{2},xy+xz+yz)=((x+y+z)^{2}-2(xy+xz+yz),xy+xz+yz)$
Since the set of zeros of a polynomial is the same as the set of zeros of its square, you see that $Z(I)=Z(I')$ with $I'=(x+y+z,xy+xz+yz)=(x+y+z,xy+(x+y)z)$

Hence $Z(I)=Z(I')$ is the intersection of the plane $x+y+z=0$ with the quadratic cylinder$Q$ given by $x^2+xy+y^2=0$ .
That cylinder is a single plane (with multiplicity 2) or the union of two planes according as $char(k)=3$ or $char (k)\neq3$ .
So $Z(I)$ is a single (double) line if $char(k)=3$ or the union of two lines if $char (k)\neq3$.

The Nullstellensatz tells you that $I(Z(I))=\sqrt I$ where $I= ((x+y+z)^{2},xy+xz+yz)$.
$\bullet $ If $char(k)\neq 3$, then $\sqrt I=(x+y+z ,\: x^2+y^2+xy) =I'$
$\bullet \bullet$ If $char(k)= 3$, then $ I=((x+y+z)^2 ,\:(x-y)^2)$ [because $x^2+y^2+xy=(x-y)^2$]
so that $\sqrt I=(x+y+z, x-y)$

share|cite|improve this answer
thanks, is it possible that you can please explain how did you obtain the locus of zeros? – user6495 Feb 14 '12 at 1:23
Dear user6495, I have expanded the calculation of the locus of zeros $Z(I)$ by showing that it is equal to the locus of zeros $Z(I')$ of a bigger ideal $I'$, which turns out (in characteristic not 3) to be the radical of $I$, that is $I'=\sqrt I$. I have also added an explicit description of $I(Z(I)))$ in characteristic $3$. – Georges Elencwajg Feb 14 '12 at 20:08

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.