Math 8205
Spring 1998

Solutions to Exercises 3 and 6

Last updated Thursday, June 18, 1998.

Contents

Solution to Exercise 3a  By Exercise 6.4 in the text, every line in  Q = G(1,3) subset P5  is of the form  L\Sigma p,H,
where  p  is a point in  P3  and  H  is a plane in  P3.  Therefore, we obtain a map of sets \varphi: W -> F1(Qsubset G(1,5)  by sending  (p,H) --> \Sigma p,H .
So, the main point is to show that it is a regular map, i.e. that the Plücker coordinates of  L \in G(1,5)  are given, at least locally,
as polynomials in the coordinates of  p  and  H.  Now, in  W subset P3\times(P3)* we have an equality of points:
(p,H) = ([x0,..., x3],[y0,..., y3]).
and the inclusion  \in H  is expressed by the following equation:
x0y0 + ... + x3y3 = 0.
Since regularity is a local property, we may assume that  p  is in the open subset  x0 \neq 0.  Then  [y0,..., y3\neq [1,0,0,0],
so we may assume without loss of generality that  H \in (P3)*  lies in the open subset  y3 \neq 0.
To find the Plücker coordinates of   L \in G(1,5), it is enough to find the coordinates of two points of  L,
or equivalently, the Plücker coordinates of two lines, say \Lambda1  and \Lambda2,  which lie in  H  and pass through  p.
To specify \Lambda1  and \Lambda2,  it suffice sto find two points of  H  such that  p  doesn't lie on the line joining them.
One possibility is to choose the points  [0, y3, 0,-y1]   and  [0, 0, y3, -y2]  respectively.
This gives us the Plücker coordinates of \Lambda1  and \Lambda2  as the  2  by  2  minors of the matrices :
2 by 4  matrix and another  2 by 4  matrix
respectively.  These minors obviously are polynomials in the homogeneous coordinates of  p  and  H.
To finish, we use these minors as the entries of a  2  by  6  matrix whose  2  by  2  minors will be the Plücker coordinates of  L.
Those, again, will certainly be polynomials in the homogeneous coordinates of  p  and  H.

Back to the exercises.

Solution to Exercise 3b
 It's fairly clear that our map  W -> F1(Q)  is bijective.  Therefore we only need to find  dim(W).
We consider the map \pi: W ->  (P3)*  induced by the projection of P3\times(P3)*  to the second fiber.
Clearly, \pi  is surjective.  Since the inverse image of  H \in (P3)*  is (isomorphic to) the plane  H subset P3,
every fiber of \pi  is of dimension = 2.  We conclude that  dim(W) = 3 + 2 = 5.

Back to the exercises.

Solution to Exercise 6a   The tangential variety  TX = Tan(X)  has dimension \leq 2r, where  r = dim(X).
Also recall that the differential (d\pi)p : T p(X) -> T _{\pi(p)(P2r - 1) is not injective if and only if  L\captX,p  is nonempty,
where  tX,p  is the embedded tangent space of  X  at  p.

To check this last point, consider the closed subset  subset X\timesP2r+1,  consisting of all pairs  (p,z)  such that  z \in tX,p .  Choosing  so that  L\capTX  is finite and  L\capV0  is empty, we conclude that there are at most finitely many points  p  such that  (d\pi)p  is not injective. 

Back to the exercises.

Solution to Exercise 6b   The secant variety  Sec(X)  has dimension \leq 2r + 1, where  r = dim(X).

Let  W0 subset (X\timesX - \Delta)\timesP2r+1  be the set of triples  (p,q,z)  such that  z \in <p,q>  (= the line joining  p  and  q),
and let  be the closure of  W0  in  X\timesX\timesP2r+1.  By considering the projection to  X\timesX - \Delta,  we see that  dim(W) = dim(W0 ) = 2r + 1.
Now, let \psiY -> Sec(X)  be induced by projection of the product to  P2r+1.   Then a general fiber of \psi  is finite.
Let  W1 subset W  be the set of points where the local fiber dimension is \geq 1.  Clearly,  dim(W1) < dim(W) = 2r + 1.
We consider the restriction of \psi  to  W1 .  The key point that  dim(\psi(W1)) < dim(W1).
(Just note that a general fiber of the restriction has dimension \geq 1.)  It follows that  dim(\psi(W1)) \leq 2r - 1.
Therefore, if  L  is a general line in  P2r+1,   the intersection   L\cap\psi(W1)  is empty, and the double point set of \pi L  is  of dimension = 1.

Back to the exercises.

 

(wavy edve design)

Comments and inquiries to: roberts@math.umn.edu

Back to my homepage: http://www.math.umn.edu/~roberts