"Class tree" view functionality is similar to that of "Class list". However, "Class list" is noticeably faster than "Class tree" and is preferred for memory distribution analysis. Use "Class tree" only if your goal is memory distribution between packages.
This view is a tool for examining how memory is distributed among instances of different classes grouped by packages.
Together with the "Biggest objects" and "Class list" the "Class tree" provides a comprehensive picture of memory consumption.
Information for each tree node is shown in four columns: "Name", "Objects" (number of objects), "Shallow Size" and "Retained Size" (for details please see Shallow and retained sizes).
A tree node can represent a package, a class or an instance field.
Package node 
shows objects that are instances of all classes (without subclasses)
of the package including subpackages.
Class node 
shows all instances of the class (not including instances of subclasses).
Field node shows all objects retained by this field and only by this field, in all instances of the class. In other words, the shown objects are those that will be freed, should the field be nulled in all the class instances. This helps to understand whether a particular field is the reason for memory retention.