14 ARKIV FÖR ZOOLOGI. BAND 8. N:0 10. 



the triangle points towards the centre of the plate, but does 

 not reach so far, ending with a truncate angle at a short 

 distance from the centre. The triangles covered above by 

 the homogeneous part of the plate, are not of the same 

 thickness. They are deepest or thickest along the side of 

 the hexagon diminishing in thickness towards the centre of 

 the plate. Each triangle is made up of bundles of fibres 

 having three main directions. Some, the greatest in number, 

 run at right angles to the side of the hexagon inwards (to- 

 wards the centre). Others have a direction quite perpendicular 

 to the former, but on the same plan. Other bundles run 

 vertically from the under surface of the homogeneous por- 

 tion of the plate, crossing the other bundles at right angles. 

 Where they cross the bundles from the side of the hexagon, 

 the latter diverge a little from each other forming meshes 

 through which the vertical bundles pass. The bundles of the 

 same direction anastomose plentifully with each other and 

 run very close to each other forming almost a homogeneous 

 mäss. By making very thin slides through the plates I hasre 

 been able to follow the arrangement of the bundles just 

 described. Figs. E and F give a diagram of the direction 

 of the bundles. In order to make this clear only a few 

 bundles are taken into consideration. The anastomoses be- 

 tween the bundles obtain by this means a direction which 

 does not quite correspond to the natural arrangement. They 

 are, of course, rather short and run in the same direction 

 as the main branches. In the figures they appear long and 

 arranged more or less perpendiculary to the main bundles, 

 which is not the case. The »triangles» are separated from 

 each other by narrow »lines» which converge towards the 

 centre of the plate. I am not able to state with perfect 

 certainty the direction of the bundles of fibres in them, but 

 my preparations seem to speak for the arrangement shown 

 in the diagrams (Fig. E and F). The bundles which go from 

 the sides of the hexagon at right angles towards the centre 

 of the plate seem to make a slight bend when they »pass 

 över» the side of the triangle continuing for some distance 

 in the above mentioned lines. These »lines», the limits be- 

 tween the triangles, are broad near the centre of the plate, 

 the number of bundles forming them being much greater 

 there. Towards the angles of the hexagon the lines become 



