Fulton 2.3

Posted on November 15, 2017

2.14.* A set V\subset\mathbb{A}^n(k) called a linear subvariety of \mathbb{A}^n(k) if V=V(F_1,\ldots,F_r) for some polynomials F_i of degree 1.

(a) Show that if T is an affine change of coordinates on \mathbb{A}^n the V^T is also a linear subvariety of \mathbb{A}^n.

Let T=(T_1,\ldots,T_n) be an affine change of coordinates, and V=V(F_1,\ldots,F_r) a linear subvariety of \mathbb{A}^n. Then I(V)=(F_1,\ldots,F_r), since F_1,\ldots,F_r are of degree 1. Note that F_1^T,\ldots,F_r^T are also degree 1. So I(V)^T=(F_1^T,\ldots,F_r^T). therefore V^T=V(I(V)^T)=V(F_1^T,\ldots,F_r^T).

(b) If V\ne\emptyset, show that there is an affine change of coordinates T of \mathbb{A}^n such that V^T=V(X_{m+1}\ldots,X_n). So V is a variety.

Suppose r=1, that is, V=V(F). Let F=\sum_{i=1}^na_iX_i+a_0. Let T&39; be an affine change of coordinates relabeling varables such that F^U=\sum_{i=d+1}^na_iX_i+a_0 with a_{d+1},\ldots,a_n\in k\setminus\{0\}. By Lemma 2.8 there exists an affine change of coordinates U on \mathbb{A}^{n-d}, defined by polynomials U_d,\ldots,U_n\in k[X_d,\ldots,X_n] such that (F^{T&39;})^{U}=X_n. Let T&39;&39; be the affine change of coordinates defined by T_i=X_i if i<d and T_i=U_i if i\ge d. Let I_d be the d\times d identity matrix and observe that T&39;&39; is represented by the following upper triangular matrix      T&39;&39; = \begin{bmatrix}             I_d & 0\\             0 & U         \end{bmatrix} Thus T&39;&39; is invertible and T=T&39;&39;\circ T&39; is an affine change of coordinates such that F^{T}=X_n. Thus V^{T}=V(X_n).

Let r\ge 1. Suppose for any F_1,\ldots,F_r\in k[X_1,\ldots,X_n] there exists affine change of coordinates T such that V(F_1,\ldots,F_r)^T=V(X_{m+1},\ldots,X_n) for some m\ge 0. Let F_1,\ldots,F_{r+1}\in k[X_1,\ldots,X_n] and V=V(F_1,\ldots,F_{r+1}). By the inductive hypothesis there exists affine change of coordinates T&39; such that V(F_1,\ldots,F_{r})^T=V(X_{m+1},\ldots,X_n). Therefore V^{T&39;}=V(F_{r+1}^{T&39;},X_{m+1},\ldots,X_n). If F_{r+1}^{T&39;}\in (X_{m+1},\ldots,X_n) we are done. Otherwise, let us select G\in k[X_1,\ldots,X_{m}] such that      F_{r+1}^{T&39;}+(X_{m+1},\ldots,X_n)=G+(X_{m+1},\ldots,X_n). Then V^{T&39;}=V(G,X_{m+1},\ldots,X_n). By the same method as the base case, we may construct an affine change of coordinates T&39;&39; on \mathbb{A}^n, such that G^{T&39;&39;}=X_{m} and X_i^{T&39;&39;}=X_i for all i>m. Thus if we define T=T&39;&39;\circ T&39;, then V^T=V(X_{m},\ldots,X_n).

(c) Show that the m appearing in part (b) is independent of the choice of T. Thus V is isomorphic to \mathbb{A}^m(k).

Suppose there exist change of coordinates T,U such that V^T=V(X_{m+1},\ldots,X_n) and V^U=V(X_{d+1},\ldots,X_n). Observe that T|_{V^T}:V^T\to V and U|_{V^U}:V^U\to V are each isomorphisms. So      k[X_1,\ldots,X_m]\cong\Gamma(V^T)\cong\Gamma(V)\cong\Gamma(V^U)\cong k[X_1,\ldots,X_d]. Therefore m=d. Moreover, \Gamma(\mathbb{A}^m)\cong k[X_1,\ldots,X_m]\cong\Gamma(V).

2.15.* Let P=(a_1,\ldots,a_n), Q=(b_1,\ldots,b_n) be distinct points in \mathbb{A}^n. The line through P and Q is defined to be L=\{ (a_1+t(b_1-a_1),\ldots,a_n+t(b_n-a_n)) \ | \ t\in k\}.

(a) Show that if L is the line through P,Q and T is an affine change of coordinates, then T(L) is the line through T(P),T(Q).

Let T_i=\sum_{j=1}^nc_{i,j}X_i+c_{i,0}. Note that T_i(P)=\sum_{j=1}^n c_{i,j}a_j+c_{i,0} and T_i(Q)=\sum_{j=1}^nc_{i,j}b_j+c_{i,0}. Let M be the line between T(P) and T(Q). Let t\in k. Let      O=\left(\sum_{j=1}^nc_{1,j}(a_j+t(b_j-a_j)+c_{1,0},\ldots,         \sum_{j=1}^nc_{n,j}(a_j+t(b_j-a_j)+c_{n,0}\right)\in M Let O&39;=(a_1+t(b_1-a_1),\ldots,a_n+t(b_n-a_n))\in L. Observe T(O&39;)=O and thus T is a bijection between L and T(L)=M.

(b) Show that a line is a linear subvariety of dimension 1, and that a linear subvariety of dimesnion 1 is the line through any two of its points.

Let L be the line between P and Q defined above. Observe that (X_1,\ldots,X_n)\in L if and only if there exists t\in k such that X_i=a_i+t(b_i-a_i) every i. Since P,Q are distinct points, there must exist i\in\{1,\ldots,n\} such that a_i\ne b_i. Therefore we may solve for t and find t=\frac{X_i}{b_i-a_i}- \frac{a_i}{b_i-a_i}. So (X_1,\ldots,X_n)\in L if and only if F_j=(b_i-a_i)(X_j-a_j)-(b_j-a_j)(X_i-a_i)=0 for all j\ne i. Therefore L=V(\{F_j \ | \ j\in\{1,\ldots,n\}\setminus\{i\}\}). Thus L is a linear subvariety of \mathbb{A}^n. Observe that \{F_j \ | \ j\in\{1,\ldots,n\}\setminus\{i\}\} is a minimal generating set (Definition 2.10) for I(L). Therefore, by Lemma 2.11, L has dimension 1.

Suppose V is a linear subvariety of dimension 1. Then there exists an affine change of coordinates T such that V^T=V(X_2,\ldots,X_n). Note that V^T is the line between P=(0,0,\ldots,0) and Q=(1,0,\ldots,0). Thus, by part (a), T(V^T)=V is the line between T(P) and T(Q).

(c) Show that in \mathbb{A}^2, a line is the same thing as a hyperplane.

By Lemma 2.11, in \mathbb{A}^2 a variety has dimension 1 if and only if it is a hyperplane. Thus, by part (b), a line is the same thing as a hyperplane.

(d) Let P,P&39;\in\mathbb{A}^2, L_1,L_2 two distinct lines through P, L_1&39;,L_2&39; distinct lines through P&39;. Show that there is an affine change of coordinates T of \mathbb{A}^2 such that T(P)=P&39; and T(L_i)=L_i&39;, i=1,2.

Let U_P,U_{P&39;} be the translations sending P,P&39; to the origin. Let v_1,v_2 be unit vectors along U_P(L_1),U_P(L_2). Similarly let u_1,u_2 be unit vectors along U_{P&39;}(L_1&39;),U_{P&39;}(L_2&39;). Note that \{v_1,v_2\} and \{u_1,u_2\} are each a basis for \mathbb{A}^2. Let T&39; be the linear map defined by v_i\mapsto u_i. Then T=U_{P&39;}^{-1}\circ T&39;\circ U_P is an affine change of coordinates such that T(P)=P&39; and T(L_i)=L_i&39;, i=1,2.

2.16. Give \mathbb{A}^n(\mathbb{C})=\mathbb{C}^n the usual topology.

(a) Show that \mathbb{C}^n\setminus S is path-connected for any finite set S.

Let P,Q\in\mathbb{C}^n\setminus S be distinct points. Let R be the set of all points on lines between P or Q and a point in S, lines in \mathbb{R}^{2n}, not dimension 1 linear subvarieties of \mathbb{A}^n(\mathbb{C}). Since S is finite, R\ne\mathbb{C}^n. Therefore we may pick a point O\in\mathbb{C}^n\setminus R. Observe that neither the line segment from P to O nor the line segment from O to Q contain a point in S. Thus there is a path from P to Q in \mathbb{C}^n\setminus S.

(b) Let V be an algebraic set in \mathbb{A}^n(\mathbb{C}). Show that \mathbb{A}^n(\mathbb{C})\setminus V is path connected.

If V=\mathbb{A}^n(\mathbb{C}) then \mathbb{A}^n(\mathbb{C})\setminus V=\emptyset and is therefore path connected. Suppose V\subsetneq\mathbb{A}^n(\mathbb{C}). Let P,Q\in\mathbb{A}^n(\mathbb{C})\setminus V, P\ne Q, and let L be the line between P and Q, in the sense of linear subvariety of dimension 1. By Lemma 2.12, V\cap L is finite. Moreover, L is isomorphic to \mathbb{A}^1(\mathbb{C}). Note that polynomial maps are continuous with respect to the topology on \mathbb{C}^n, so L is in fact homeomorphic to \mathbb{A}^1(C). Therefore L\setminus V is homeomorphic to \mathbb{C} with a finite set removed. Thus L\setminus V is path connected by part (a). Therefore \mathbb{A}^n(\mathbb{C})\setminus V is path connected.