This doctoral thesis investigates the influence of nonverbal communication on human-robot object handover. Handing objects to one another is an everyday activity where two individuals cooperatively interact. Such close interactions incorporate a lot of nonverbal communication in order to create alignment in space and time. Understanding and transferring communication cues to robots becomes more and more important as e.g. service robots are expected to closely interact with humans in the near future. Their tasks often include delivering and taking objects. Thus, handover scenarios play an important role in human-robot interaction. A lot of work in this field of research focuses on speed, accuracy, and predictability of the robot’s movement during object handover. Still, robots need to be enabled to closely interact with naive users and not only experts. In this work I present how nonverbal communication can be implemented in robots to facilitate smooth handovers. I conducted a study on people with different levels of experience exchanging objects with a humanoid robot. It became clear that especially users with only little experience in regard to interaction with robots rely heavily on the communication cues they are used to on the basis of former interactions with humans. I added different gestures with the second arm, not directly involved in the transfer, to analyze the influence on synchronization, predictability, and human acceptance. Handing an object has a special movement trajectory itself which has not only the purpose of bringing the object or hand to the position of exchange but also of socially signalizing the intention to exchange an object. Another common type of nonverbal communication is gaze. It allows guessing the focus of attention of an interaction partner and thus helps to predict the next action. In order to evaluate handover interaction performance between human and robot, I applied the developed concepts to the humanoid robot Meka M1. By adding the humanoid robot head named Floka Head to the system, I created the Floka humanoid, to implement gaze strategies that aim to increase predictability and user comfort. This thesis contributes to the field of human-robot object handover by presenting study outcomes and concepts along with an implementation of improved software modules resulting in a fully functional object handing humanoid robot from perception and prediction capabilities to behaviors enhanced and improved by features of nonverbal communication.