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

Let $A^n$ be the affine variety defined by $k^n$, $k$ is a field, and $U_i=\{(x_0, \ldots, x_n) \in \mathbb{P}^n \mid x_i \neq 0\}$. Define $\phi: U_i \to A^n$ by $$ \phi_i(x_0, \ldots, x_n) = (\frac{x_0}{x_i}, \ldots, \frac{x_{i-1}}{x_{i}}, \frac{x_{i+1}}{x_{i}}, \ldots, \frac{x_n}{x_i}). $$ How to show that $\phi_i(U_i \cap U_j)$ is a principal open set $D(f)$ in $A^n$ for some function $f$? Here $D(f)=\{x \in A^n \mid f(x) \neq 0 \}$ Thank you very much. I think that $$ U_i \cap U_j = \{(x_0, \ldots, x_n) \in \mathbb{P}^n \mid x_i \neq 0, x_j \neq 0 \}. $$ If $i=j$, then $$ \phi_i(U_i \cap U_j) = \phi_i(U_i) = \{(\frac{x_0}{x_i}, \ldots, \frac{x_{i-1}}{x_{i}}, \frac{x_{i+1}}{x_{i}}, \ldots, \frac{x_n}{x_i}) \mid x_0, \ldots, x_{i-1}, x_{i+1}, \ldots, x_n \in A \}. $$ How to show that $\phi_i(U_i)$ is some $D(f)$? Thank you very much.

share|cite|improve this question
Isn't it just the complement of the zero locus of the $j^{\text{th}}$ coordinate function? – Alex Youcis Sep 23 '13 at 11:02
You mean principal open set, where "principal" is an adjective meaning "main". A principle is a noun describing a moral point of view. – Georges Elencwajg Sep 23 '13 at 11:27
@GeorgesElencwajg, thank you very much. – LJR Sep 23 '13 at 11:29
Dear LJR, congratulations on your positive attitude. English is a foreign language for me and since I am very grateful when someone corrects me, I thought I could help others too. And, just to show that I'm not only a grammatical pedant, I have answered your question :-) – Georges Elencwajg Sep 23 '13 at 11:59
@GeorgesElencwajg: perhaps you know the topologist's mnemonic for remembering the distinction: "A principle bundle is one with moral fibre". A bit off-topic here, but maybe amusing enough to be forgiven. – user64687 Sep 23 '13 at 12:34
up vote 2 down vote accepted

I am answering this question because it hides a slight combinatorial problem.

a) The image $\phi_i(U_i)$ is all of $\mathbb A^n$, so can be written $D(1)$, which answers your last question.
b) The description of $\phi_i(U_i\cap U_j)$ for $i\neq j$ however must be split into two cases.
Call $(z_1,\cdots,z_n)$ the coordinates in $\mathbb A^n$. Then we have:
$\bullet$ $\phi_i(U_i\cap U_j)=D(z_{j+1})$ if $j\lt i$
$\bullet \bullet$ $\phi_i(U_i\cap U_j)=D(z_j)$ if $j\gt i$

Try it with $n=3, i=2, j=0$ and $j=3$.
In particular you will see that you can't have $\phi_i(U_i\cap U_j)=D(z_j)$ for $j=0$ since there is no $z_0$ !

Maybe I should have emphasized that $z_j=x_j/x_i$ for $j\gt i\geq 0$ and $z_j=x_{j-1}/x_i$ for $1\leq j\leq i$.

share|cite|improve this answer
thank you very much. Let $n=3, i=2, j=0$. Then $U_i \cap U_j = \{[x_0, x_1, x_2]: x_0 \neq 0, x_2 \neq 0\}$ and $\phi_i(U_i \cap U_j) = \{[x_0/ x_2, x_1/ x_2] : x_0 \neq 0\}$. It seems that $\phi_i(U_i \cap U_j) = D(z_0) \neq D(z_1) = D(z_{j+1})$. I am confused. – LJR Sep 23 '13 at 12:13
Of course, this is all just indexing. The point is that it's the principal open set associated to one of the coordinate functions. – Alex Youcis Sep 23 '13 at 12:17
Dear LJR, for $n=3$, remember that $\mathbb P^3$ has coordinates $[x_0,x_1,x_2,x_3]$. You forgot $x_3$. Also remember that $x_0/x_2=z_1$, not $z_0$ which, once again, doesn't exist!! – Georges Elencwajg Sep 23 '13 at 12:18
@GeorgesElencwajg, thank you very much. – LJR Sep 23 '13 at 12:21
Dear LJR, you're welcome! – Georges Elencwajg Sep 23 '13 at 12:29

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.