MINUTE STRUCTURE. 567 



As regards Malcvpterurus, it has been already pointed out that 

 there is no obvious columnar arrangement in the organ, but that the 

 connective tissue septa divide it into a very large number of irregularly 

 shaped compartments. 1 Cross-sections made in different planes show 

 that these compartments are more or less lozenge shaped, the acute angle 

 of the lozenge being towards the surface and deep layers of the skin 

 respectively, whilst the obtuse angles are cephalic and caudal in position. 

 These lozenge shaped compartments are dovetailed into their neighbours, 

 and thus practically constitute a large number of columns parallel with 

 the spinal axis. The compartments may be seen with a magnifying lens 

 in a fish 30 cms. in length ; they are filled with the usual transparent 

 jelly-like substance, and contain discs slung athwart them. The proto- 

 plasmic layer of each disc lies near the walls which form the caudal 

 boundary of each compartment, and the whole organ contains a very 

 large number of such layers, arranged in series from head to tail one 

 behind the other. The nerve fibres enter the compartments at the 

 lateral acute angles, and, running along close to the wall, pass finally to 

 a central stalk projecting from the caudal surface of each protoplasmic 

 disc. Each disc is shaped like a peltate leaf with the central stalk pro- 

 jecting from its caudal surface ; the cephalic surface is smooth and has 

 a slight depression opposite the emergence of the stalk from the other 

 caudal surface. The nerve fibre which enters the stalk was formerly 

 supposed to be traceable through this on to the cephalic surface, thus 

 piercing the plate by means of the stalk. Sections show that under 

 the cephalic surface the protoplasm has a punctiform appearance, and 

 this was thought to indicate the section, either real or optical, of the 

 finest divisions of axis cylinder twigs. The observations of Ballowitz 

 do not confirm this view, for the nerve fibre ends in the base of the 

 stalk in free branches with terminal knobs. The actual disc thus 

 contains no nerves at all. The nerve which supplies the organ is 

 probably unique as regards its structure. It consists on each side 

 of one axis cylinder, the process of one gigantic nerve cell, situated 

 in the grey matter, lateral and somewhat dorsal to the central canal, 

 at a level between the exit of the first and second spinal nerve 

 roots. The two cells, one on each side of the cord, were as much 

 as i mm. in cross-section, in a specimen of the fish 25 cms. long. They 

 are further characterised by conspicuous nuclei, fibrillar and cell pro- 

 cesses. The processes are extremely numerous, and those of one cell 

 often join another, but it appears that they do not join the processes 

 either of the other electrical nerve cell or of any neighbouring nerve 

 cells. The cell itself is traversed by special blood vessels, an arrange- 

 ment which secures an adequate blood supply. The single axis cylinder 

 process of each cell, large in its origin, soon dwindles, and coursing 

 along the inner surface of the organ, accompanied by blood vessels, 

 gives off' at regular intervals branches for the supply of the organ. It 

 is surrounded by an extensive connective tissue sheath, consisting of 

 numerous lamina?, prolongations of which extend along the subdivisions 

 as far as the angles of the organ compartments. The number of the 

 organ compartments is very large, there being little short of two 

 millions in each lateral half of the entire organ. Since these are all 



1 For structure of organ in Malcvpterurus, see Bilharz, " Elektr. Org. d. Zitterwels," 

 Leipzig, 1857 ; Fritsch, loc. tit.; Ballowitz, Anat. Ann., Jena, 1898, Bel. xv. S. 85 ; ibid., 

 "Das elektrische organ des Afrikanischen Zitterwelses," Jena, 1899. 



