This paper presents a mathematical model that predicts both the qualitative features and quantitative details of a class of perturbed and unperturbed large-amplitude, voluntary movements performed by primates at intermediate speed. The model is based on dynamic optimization theory, which is used to determine the movement that minimizes the rate of change of acceleration (jerk) of the limb. The model assumes that the goal of the movement is to make the smoothest possible movement, minimizing accelerative transients. The model also incorporates the concept of a "virtual position," determined by the active states of the muscles, which represents a mechanical consequence of neural commands to the muscles. This concept helps in separating the descriptions of movement organization and execution. The model is applied to a class of voluntary arm movements observed in monkeys, specifically large-amplitude, single-joint, elbow motions at intermediate speed. The model is derived from experimental observations of intact and deafened monkeys performing voluntary pointing movements. The model uses a differential equation to describe the system behavior, assuming the elbow joint has a constant center of rotation and the forearm is treated as a rigid body. The resultant torque is determined by the active states of the muscles spanning the joint and the angular position and velocity of the joint. The model is validated by comparing its predictions with experimental data, showing good qualitative and quantitative agreement with observed movements. The model is also extended to multi-joint movements by Flash and Hogan (1982). The paper discusses the implications of the model for neurophysiological theories, such as "final position control," and highlights the model's ability to predict the organization of voluntary movements. The model is shown to be a useful tool for understanding the planning and execution of movements, as well as for predicting the effects of perturbations on movement. The model's ability to generalize to more complex movements and its potential for further research are also discussed.This paper presents a mathematical model that predicts both the qualitative features and quantitative details of a class of perturbed and unperturbed large-amplitude, voluntary movements performed by primates at intermediate speed. The model is based on dynamic optimization theory, which is used to determine the movement that minimizes the rate of change of acceleration (jerk) of the limb. The model assumes that the goal of the movement is to make the smoothest possible movement, minimizing accelerative transients. The model also incorporates the concept of a "virtual position," determined by the active states of the muscles, which represents a mechanical consequence of neural commands to the muscles. This concept helps in separating the descriptions of movement organization and execution. The model is applied to a class of voluntary arm movements observed in monkeys, specifically large-amplitude, single-joint, elbow motions at intermediate speed. The model is derived from experimental observations of intact and deafened monkeys performing voluntary pointing movements. The model uses a differential equation to describe the system behavior, assuming the elbow joint has a constant center of rotation and the forearm is treated as a rigid body. The resultant torque is determined by the active states of the muscles spanning the joint and the angular position and velocity of the joint. The model is validated by comparing its predictions with experimental data, showing good qualitative and quantitative agreement with observed movements. The model is also extended to multi-joint movements by Flash and Hogan (1982). The paper discusses the implications of the model for neurophysiological theories, such as "final position control," and highlights the model's ability to predict the organization of voluntary movements. The model is shown to be a useful tool for understanding the planning and execution of movements, as well as for predicting the effects of perturbations on movement. The model's ability to generalize to more complex movements and its potential for further research are also discussed.