Since sense-annotated corpora serve as gold standards for the development, training, and evaluation of word sense disambiguation (WSD) systems, their availability is a necessary prerequisite for WSD. The purpose of the present paper is to describe the manual annotation of a selected set of lemmas in the TüBa-D/Z treebank [9, 21] with senses from the German wordnet GermaNet [7, 8]. With the sense annotation for a selected set of 109 words (30 nouns and 79 verbs) occurring together 17 910 times in the TüBa-D/Z, the treebank currently represents the largest manually sense-annotated corpus available for GermaNet. This paper describes the annotation process, presents statistics, analyzes inter-annotator agreement (and disagreement), and documents the technical integration of the sense annotations into the most recent release 9.1 of the TüBa-D/Z. The publication of this paper is accompanied by making the described sense annotations available to the research community.