1: \begin{abstract}

2: We consider the model of random trees introduced by Devroye

3: [\textit{SIAM J. Comput.} \textbf{28} (1999) 409--432]. The model

4: encompasses many important randomized algorithms and data structures.

5: The pieces of data (items) are stored in a randomized fashion in the

6: nodes of a~tree. The total path length (sum of depths of the items) is

7: a natural measure of the efficiency of the algorithm/data structure.

8: Using renewal theory, we prove convergence in distribution of the total

9: path length toward a distribution characterized uniquely by a fixed

10: point equation. Our result covers, using a unified approach, many data

11: structures such as binary search trees, $m$-ary search trees, quad

12: trees, median-of-$(2k+1)$ trees, and simplex trees.

13: %

14: % By using renewal theory we prove convergence in distribution of the

15: %total path length, i.e., the sum of all depths\footnote{the depths of

16: %what!! --N},

17: % for a large class of random split trees introduced by Devroye [SIAM J

18: %Comput \textbf{28}, 409--432, 1998]. Split trees are used as models

19: %for tree data structures or sorting algorithms, and the total path

20: %length represents a natural measure of their efficiency. Our result

21: %covers most split trees like binary search trees, $m$-ary search

22: %trees, quad trees, median-of-$(2k+1)$ trees and simplex trees. %The

23: %case of tries and digital search trees is not tread.

24: \end{abstract}