Performance considerations and perhaps a comparision of feature list between different collections will help highly to make better apps. Why no C# syntax ? Need a little more explanation too.

public class Node<T>
     public List<Node<T>> Children = new List<Node<T>>();
     public T Value;
end class

Use it like this:

Node<string> root = new Node<string>() { Value = "I am the root node!" };

Node<string> child = new Node<string>() { Value = "I am a child of the root node!" }; 
root.Children.Add(child);

Node<string> sibling = new Node<string>() { Value = "I am a sibling of child!" };
root.Children.Add(sibling);

Node<string> grandchild = new Node<string>() { Value = "I am a grandchild of the root node!"};
child.Children.Add(grandchild);

A Tree class would be useful.

It would be useful a Tree class

But I guess, they are more or less same as the structure name conveys (List, Hashtable etc. ) in theory. But definitely I have come across significant variations due to the implementation differences. So the performance indicators (O notation) should be made available for various operations.

Would be nice to get performance and O notation differences here.