No abstract

The distribution and morphological characteristics of dopamine (DA) neurons in the midbrain of the squirrel monkey (Saimiri sciureus) were investigated by peroxidase-antiperoxidase (PAP) immunohistochemistry with a highly specific antiserum raised against DA-glutaraldehyde-lysyl-protein conjugate (donated by M. Geffard). Four contiguous areas contained DA-immunostained nerve cell bodies: (1) the substantia nigra, pars compacta (SNc), (2) the ventral tegmental area (VTA), (3) the retrorubral area (RRA), and (4) the periaqueductal gray (PAG). The SNc composed the vast majority of DA-immunostained neurons. Most of these neurons were relatively large (mean diameters: 35 x 15 micron) and varied in shape from fusiform to polygonal, but a few smaller (16 x 10.5 micron) globular cells were dispersed among them. The caudal two-thirds of the SNc was particularly rich in DA somata. Rostrally, these DA cells formed several distinct columns impinging deeply upon the underlying pars reticulata. Large oval sectors mostly devoid of immunoreactivity were delineated by these trabeculae. The long dendritic processes of DA neurons in the SNc were generally oriented in prominent dorsoventral bundles the ventralmost portion of which arborized diffusely along the dorsal surface of the cerebral peduncle. In the VTA, the DA neurons were regrouped in a triangular zone located dorsal to the interpeduncular nucleus, medial to the substantia nigra and ventral to the oculomotor nucleus. These DA cells were of medium size (19 x 10.5 micron), globular or fusiform, and usually showed one or two thick primary dendrites oriented dorsoventrally. The DA cells in the RRA lay in continuity with the most caudal DA-containing elements of the substantia nigra but could be distinguished by their smaller size (26 x 12 micron), shorter and more profusely branched dendrites, and darker immunostaining. These DA neurons were characteristically scattered among and medial to the fibers of the medial lemniscus, and a few could be observed as far caudally as the pedunculopontine nucleus. In the PAG, DA-immunostained neurons were seen in the rostral half of the mesencephalic central gray and predominated in its ventral half. These cells were of medium size (22.5 x 10 micron) and some of them were found in proximity to the ventricular lining. At caudal levels, the DA-positive cells in the PAG did not intermingle with dorsal raphe neurons.(ABSTRACT TRUNCATED AT 400 WORDS)

In this paper, a method is presented for the synthesis of 3‐RRR planar parallel mechanisms. The method uses a genetic algorithm while considering three different design criteria: the optimization of the mechanism workspace to approach a prescribed workspace, the maximization of the mechanism's dexterity, and the avoidance of singularities inside the mechanism workspace. It is shown that, for a given mechanism, some working modes do not have any corresponding singularity curves located inside the mechanism workspace. Furthermore, a case is presented where, for a given orientation range of the mechanism's end‐effector, there are no parallel singularities located inside the workspace. Finally, two methods are described and compared to deal with the nonuniform units of the mechanism's Jacobian matrix during the dexterity computation. © 2004 Wiley Periodicals, Inc.

No abstract

It is a generally well-known fact that the design of parallel mechanisms while optimizing performance is quite difficult. In this paper, a reliable synthesis method capable of optimally selecting the geometrical parameters of planar parallel mechanisms is presented. Three different architectures are considered and a genetic algorithm is used to perform the optimization. The performance of each mechanism is evaluated according to four different criteria: workspace, singular configurations, dexterity, and stiffness. In order to make the synthesis method as realistic as possible, mechanical constraints affecting the angular rotation of the 2-RPR and 3-RPR mechanisms' passive revolute joints are considered. Moreover, since the conventional methods for computing the dexterity and the stiffness index are not valid for the 3-RPR and 3-RRR mechanisms, an alternative computation method is used.