DJVL8I0N OF COLLAR-CELLS OF CLATHIUNA COIUACEA. 639 



Fig. 25. — Late diaster-stage, with l)eginning reconstitution of tlie 

 daugliter-nuelei. Slightly ol)lique ; only one daugliter-flagellum to Vie 

 seen. 



Figs. 26-28, 28a. — Stages in the division of the cell-body, with recou- 

 stitntion of the daughter-nuclei. In all figures, except 28a, the remains 

 of the original collar can be seen clearly. In figs. 26 and 27 the remains 

 of the achromatic spindle can he seen lietween the two daughter-nuclei. 

 In fig. 28a the division is practically complete. 



Figs. 29, 30. — Pairs of young, recently divided collar-cells. In fig. 

 30 some of the adjacent cells ai-e drawn to show the way in which the 

 cells at this stage are raised up above the surrounding cells. 



Figs. 31-35. — Collar-cells showing enclosures of various kinds, some 

 of them perhaps of parasitic nature. In the cell shown in fig. 34 the 

 nucleus seems to have disappeared, but may have lieen cut off. 



Fig. 36. — Nuclei of resting collar-cells, magnified 3000 linear, a, h. 

 c, dark nuclei ; d, a light nucleus ; e, /, g. thin sections of nuclei showing 

 the reticular structure ; in y the karyosome does not come into the 

 section. 



Fig. 37. — Transverse sections of collar-cells in the region of the 

 collar, a passes through the base of the collar, and b just above this 

 level; both show the l^lepharoplast centrally. In c the collar is cut 

 transversely with the flagellum in the centre. 



PLATE 26. 



[All the figures are drawn from sections of material fixed with osmic 

 acid and stained with picro-carmine ; magnification throughout 2000 

 linear.] 



Fig. 38. — Two collar-cells, one of the normal resting type (on the 

 left), the other with the nucleus migrating towards the apex of the cell 

 preparatory to division. 



Figs. 39, 40. — Collar-cells showing the nucleus at the apex of the cell, 

 and the flagellum in process of retraction by means of a pseudopodium- 

 like process from the cell. 



Fig. 41. — On the left a normal resting cell; on the right a cell with 

 the nucleus at the apex and the flagellum completely I'etracted, but 

 represented by the still persistent pseudopodium-like process seen in 

 the two preceding figures. 



Figs. 42, 43. — Stages with the daughter-flagella growing out from the 

 poles- of the nuclear spindle, and with the collar beginning to collapse. 

 The achromatic elements, namely, spindle and centrosomes, ai-e not 

 stained and are not visible in the preparation, but the equatorial plate 



