Phylogenetic trees represent theoretical evolutionary relationships among various species. Mathematically they can be described as weighted binary trees and the leaves represent the taxa being compared. One major problem in mathematical biology is the reconstruction of these trees. We already know that trees on the leaf set X can be uniquely constructed from splits, which are bipartitions of X. The question I explore in this thesis is whether reconstruction of a tree is possible from subsplits, or partial split information. The major result of this work is a constructive algorithm which allows us to determine whether a given set of subsplits will realize a tree and, if so, what the tree looks like.