72 THE nerye-el?:ments of the embryonic lobster. 



must be taken not to injure or stretch the ganglia, and their continuity 

 with the ganglia of the thorax should be maintained. If the embryo, 

 tlius prepared, be placed with the dorsal surface uppermost in very 

 dilute methylene blue, satisfactory staining of many of the elements of 

 the abdomen will soon take place. 



The elements of the abdomen belong to types similar to those 

 described for the thorax. 



In each ganglion a pair of elements exists, taking origin in two 

 ganglion cells lying upon opposite sides. Each cell gives off a fibre, 

 which after decussation with its fellow, passes to the opposite side of 

 the ganglion, and gives off a branch to the neuropile. It then turns 

 forward and runs along the cord to the brain. In this way each of the 

 ganglia of the abdomen is placed in direct communication with the 

 brain. 



In the sixth abdominal ganglion two pairs of elements of this type 

 occur, thus pointing to the composite nature of the ganglion. 



A considerable number of motor elements, consisting of a cell in one 

 of the ganglia, and a fibre which passes out of the central nervous 

 system, have stained in the abdomen. These are of two kinds; first, 

 tliose in which the element is confined to a single ganglion, the fibre 

 passing out through one of the roots of the ganglion in which the cell 

 lies, and secondly, those in which the fibre leaves the central nervous 

 system by a nerve-root of a ganglion other than that in which the cell 

 lies. These will be described in detail in a later paper. 



Further observations have also been made on the sensory nerve 

 elements, which have their origin in cells lying outside the central 

 nervous system. These fibres, on entering a ganglion, make a charac- 

 teristic Y-shaped bifurcation, sending one branch forwards and the other 

 backwards along the cord. These branches have been traced for 

 considerably greater distances than was previously possible, the forward 

 one having been seen to pass through at least nine or ten ganglia. In 

 all probability, all these forward branches go in every case to the brain. 

 The backwardly directed branch has never been actually traced through 

 more than two or three ganglia, and no indication has been obtained as 

 to the locality or nature of its termination. 



A detailed account of the observations here recorded will be published 

 in the Quarterly Journal of Microseopical Science. 



