5
$\begingroup$

Consider a smooth plane curve $X\subset\mathbb{P}^2$ of degree $d$. We will say that $x\in X$ is an inflection point of order $s$ if the tangent line $T_xX$, of $X$ at $x\in X$, intersects $X$ in $x\in X$ with multiplicity at least $s$.

For instance, any point of $X$ in an inflection point of order $s = 2$, and the inflection points of order $s = 3$ are the flexes of $X$.

Does there exist a closed formula for the number $I(s)$ of inflection points of order $s$ of $X$. For instance, it is well known that $I(3) = 3d(d-2)$ since flexes are the intersection points of $C$ with its Hessian which is a curve of degree $3(d-2)$.

Furthemore, we must have $I(s) = 0$ for $s > d$ since otherwise $C$ would be reducible.

$\endgroup$
1
  • 1
    $\begingroup$ As pointed out already, the general curve will not have a hyperflex (ie an inflection point of order $\geq4$). The loci in the parameter space $|\mathcal{O}_{\mathbb{P}^2}(d)|$ which do are studied by considering the bundle of principal parts and computing its Chern classes. This is explained really well in the "3264" book by Eisenbud-Harris $\endgroup$ – Frank Feb 4 at 8:19
11
$\begingroup$

(The statements you quote are only true if you are working over a field of characteristic zero or $p > d$. I will continue to make that assumption)

The formula is not $I(3)=3d(d-2)$ but rather $\sum_{s>2} (s-2)I(s) = 3d(d-2)$ (changing notation slightly so $I(s)$ counts the points with contact exactly $s$ instead of at least $s$).

The typical curve (non-empty open set of the parameter space) does indeed satisfy $I(3)=3d(d-2)$ and $I(s)=0, s > 3$. I think you can expect many different possibilities under the above constraints but most likely not all of them. I don't this has been worked out in general. For $d=4$ all possibilities and the dimension of the space of curves that satisfy them has been worked out in the PhD thesis of Vermeulen. I have unfortunately misplaced my copy so I can't quote the exact result.

Edit: Found the thesis (Amsterdam 1983). If $d=4$ then $I(3)+2I(4)=24$ and $I(4)$ can take any value between $0$ and $12$, except for $10$ and $11$.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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