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 exercise $4.4$ part (c) of Hartshorne's book. Let $Y$ be the nodal cubic curve $y^{2}z=x^{2}(x+z)$ in $\mathbb{P}^{2}$. Show that the projection $f$ from the point $(0,0,1)$ to the line $z=0$ induces a birational map from $Y$ to $\mathbb{P}^{1}$.


Consider the open subset of $Y$ given by $Y \setminus V(z)$ , that is we set $z=1$.

Define $f: Y \setminus V(z) \rightarrow \mathbb{P}^{1} \setminus \{[1:1],[1:-1]$ by:

$f([x : y : z]) = [x: y]$

Now define $g: \mathbb{P}^{1} \setminus \{[1 : 1],[1 : -1]\} \rightarrow Y \setminus V(z)$ by:

$g([x : y]) = [(y^{2}-x^{2})x : (y^{2}-x^{2})y : x^{3}]$

Question: what if $x=0$? then we get $y=0$ so $[x :y : 1] = [ 0 : 0 : 1]$ but the point $[x^{4} : x^{3}y : x^{3}]$ is not defined at $x=y=0$.

share|cite|improve this question
$[ab:ac] = [b:c]$, right? – Hurkyl May 17 '12 at 17:13
You can simplify your formula for $f$ (assuming it's right). p.s. your formula would be ill-defined were it not for the fact the simplification is possible: your functions aren't homogeneous, and so the required identity $f[x:y:z]=f[ax:ay:az]$ isn't immediately evident. – Hurkyl May 17 '12 at 17:23
@Hurkl: just modified it, can you have a look please? – user31509 May 17 '12 at 17:44
It's been a while since I've look at at this fine detail -- but as I recall, it's good enough for them to be inverses on an open set, right? You can restrict the domains further to make it work. – Hurkyl May 18 '12 at 17:54
up vote 1 down vote accepted

Recall that the line in $\mathbb P^2$ joining $P=[a:b:c]$ to $P'=[a':b':c']$ is given parametrically by $[ua+u'a':ub+u'b':uc+u'c']$ where $[u:u'] \in \mathbb P^1$.

So, the line joining $P'=O=[0:0:1]\in Y$ to $P=[a:b:c]\neq O\in Y$ has parametric equation $[ua:ub:uc+u']$.
It intersects the line at infinity $z=0$ at the point $uc+u'=0$ i.e. at $[a:b:0]$.
So the required rational map $Y \cdots \to \mathbb P^1$ (not a morphism!) is $$\phi:U=Y\setminus \lbrace O\rbrace \to \mathbb P^1:[a:b:c]\mapsto [a:b] $$
The inverse rational map is the morphism ( you have actually already calculated this: your computation is correct) $$\psi=\phi^{-1}:\mathbb P^1\to Y: [c:d]\mapsto [c(d^2-c^2):d(d^2-c^2):c^3]$$
Notice that $\psi$ is the normalization of $Y$, a remark which enlightens the whole exercise.
In particular we see that the two points $[1:\pm1]\in \mathbb P^1$ are sent by $\psi$ to $O$ .
This is why the rational map $\phi$ we started with cannot be a morphim: a non-injective morphism like $\psi $ cannot be inverted as a morphism, only (sometimes!) as a rational map .

share|cite|improve this answer

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.