166 BULLETIN OF THE 



of the meshes of tlie brain itself. Passing inward from this external cap- 

 sule and directly continuous with its elements, fibres penetrate the brain 

 in every direction, either in definite strands, or in a delicate network 

 surrounding the ganglionic cells (Fig. 25). The fibres which make up 

 these meshes are finer than most of those composing the capsule itself, 

 and recall strongly the finer elements of the cutis. With these they 

 also agree in the possession of minute elongated nuclei, although the 

 clear nuclei previously mentioned are by no means rare. These fibres 

 surround each ganglionic cell with a definite covering (Plate III. Fig. 32) 

 of interlacing elements from which others pass off tangentially to neigh- 

 boring cells. ((?/. Eohde, '87, Taf. IV. Fig. 44-68.) 



Ganglionic Cells. — None of the many previous writers on Sipunculus 

 have considered the histological elements of the central nervous sj'stem 

 more than cursorily, so that a more extended description of these may 

 be of interest, especially for comparison with the recent exact deter- 

 mination of these elements in many other groups of worms. All the 

 ganglionic cells which were so situated as to admit of a positive answer to 

 the question of their polarity were unipolar, though by no means always 

 unifilar. Such cells as were accurately determined wei'e usually periph- 

 eral, since the mass of other fibres and the confusion of many cells make 

 an accurate determination in the case of those cells which are located in 

 the centre of the nervous mass often impossible. I am inclined to think 

 that in the latter region there are multipolar cells, although the demon- 

 stration of these was not wholly satisfactory. The cells are uniformly 

 without any proper cell membrane. Each lies enveloped in a covering 

 of delicate connective-tissue fibres (Plate III. Fig. 32), which accom- 

 panies the fibrous processes in the form of a delicate sheath (neuroglia). 

 These enveloping fibres are a part of the meshwork which has already 

 been described as arising from its external capsule, and penetrating 

 through the whole brain. 



Of all the ganglionic cells, the smallest (Plate III. Figs. 24, 25, cl. gn. 

 I.), which usually appear simply as nuclei measuring 6 by 4 /a (Fig. 30), 

 are the most abundant. They are highly refractive, and show a great 

 affinity for coloring matter. There is a nuclear membrane which is 

 stained deeply, as are also the numerous (4 to 10) nucleoli ; between the 

 latter many minute chromatine gi'anules are distinguishable with a 

 high power. These nuclei seem somewhat irregular in shape, varying 

 from circular to oval. This variation I regard as due to the direction of 

 the plane of section, and consider the true form as oval. In most cases 

 it is impossible to find even a trace of a cell body, and I was at first 



