#pragma GCC optimize("Ofast,unroll-loops")
//#pragma GCC target("avx2,fma,bmi,bmi2,sse4.2")
#include<bits/stdc++.h>
#define all(x) (x).begin() , (x).end()
#define pll pair<long long, long long>
#define fi first
#define se second
#define bit(i,j) ((j >> i) & 1)
#define lowbit(x) (x & (-x))
#define sigma main
using namespace std;
const long long inf = 1e18+1;
const int mod = 998244353;
const int MAXN = 2e5+100;
#define int long long
template<class T , class U>
struct segment_tree{
T t[4 * MAXN];
U d[4 * MAXN];
const T Ti = 0;
const U Ui = 0;
T combine(T a, T b) {
return a + b;
}
void apply(T &a, U b, int l , int r){
a += b;
}
void lazy(U &a , U b){
a += b;
}
segment_tree(){
for(int i = 0 ; i < 4 * MAXN ; i++) d[i] = Ui;
}
void build(int id , int l , int r){
if(l == r) {
t[id] = Ti; return;
}
int mid = l + r >> 1;
build(id<<1 , l , mid ) ; build(id<<1|1 , mid+1 , r );
t[id] = combine(t[id<<1] , t[id<<1|1]);
}
void push(int id , int l , int r){
if(l == r) return ;
lazy(d[id<<1] , d[id]); lazy(d[id<<1|1] , d[id]);
int mid = l + r >> 1;
apply(t[id<<1] , d[id] , l , mid) ; apply(t[id<<1|1] , d[id] , mid+1 , r);
d[id] = Ui;
}
void update(int id , int l , int r , int x , int y , U val){
push(id, l , r);
if(l > y or r < x) return ;
if(l >= x and r <= y){
lazy(d[id] , val);
apply(t[id] , val , l , r);
return;
}
int mid = l + r >> 1;
update(id<<1 , l , mid , x ,y , val) ; update(id<<1|1 , mid+1 , r , x , y , val);
t[id] = combine(t[id<<1] , t[id<<1|1]);
}
T get(int id , int l , int r , int x , int y){
push(id , l , r);
if(l > y or r < x) return Ti;
if(l >= x and r <= y){
return t[id];
}
int mid = l + r >> 1;
return combine(get(id<<1 , l , mid , x , y) , get(id<<1|1 , mid+1 ,r , x , y));
}
};
segment_tree<int , int> T;
vector<int> adj[MAXN];
int nxt[MAXN] , par[MAXN] , h[MAXN];
int dfs(int u , int p){
int sz = 1 , mx = 0;
for(auto v : adj[u]){
if(v == p) continue;
par[v] = u;
h[v] = h[u] + 1;
int c = dfs(v , u);
sz += c;
if(c > mx){
nxt[u] = v; mx = c;
}
}
return sz;
}
int in[MAXN] , out[MAXN] , head[MAXN];
int Time = 0;
void hld(int u, int h){
head[u] = h;
in[u] = ++Time;
if(nxt[u] != 0){
hld(nxt[u] , h);
}
for(auto v : adj[u]){
if(v == par[u] or v == nxt[u]) continue;
hld(v , v);
}
out[u] = Time;
}
int lca(int u , int v){
while(head[u] != head[v]){
if(in[head[u]] > in[head[v]]){
u = par[head[u]];
}else{
v = par[head[v]];
}
}
if(h[u] > h[v]) return v;
return u;
}
int n , m , k;
void upd(int u , int v){
while(head[u] != head[v]){
T.update(1 , 1 , n , in[head[u]] , in[u] , 1);
u = par[head[u]];
}
T.update(1 , 1 , n , in[v] , in[u] , 1);
T.update(1 , 1 , n , in[v] , in[v] , -1);
}
int32_t sigma(){
//freopen("COLOREDBALLS.inp", "r", stdin);
//freopen("COLOREDBALLS.out", "w", stdout);
ios_base::sync_with_stdio(0); cin.tie(0);
int k; cin >> n >> m >> k;
vector<pll> E(n);
for(int i = 1; i < n ; i++){
int a , b; cin >> a >> b;
adj[a].push_back(b);
adj[b].push_back(a);
E[i] = {a , b};
}
dfs(1 , 1); hld(1 , 1);
for(int i = 1 ;i <= m ; i++){
int s; cin >> s;
vector<int> v;
while(s--){
int x; cin >> x; v.push_back(x);
}
sort(all(v) , [&](int a , int b){
return in[a] < in[b];
});
for(int i = 0 ; i < v.size() ; i++){
int j = (i + 1) % v.size();
int p = lca(v[i] , v[j]);
upd(v[i] , p); upd(v[j] , p);
}
}
vector<int> ans;
for(int i = 1 ; i < n ; i++){
int u = E[i].fi , v = E[i].se;
if(in[u] < in[v]) swap(u , v);
if(T.get(1 , 1 , n , in[u] , in[u]) >= 2 * k) ans.push_back(i);
}
cout << ans.size() << "\n";
for(auto x : ans) cout << x << " ";
cout << endl;
return 0;
}
