202 ELECTRON-MICROSCOPIC STRUCTURE OF PROTOZOA 



but somewhat crumpled in a position of minimum tension — 

 neither contraction nor extension. The M-bands always appear 

 straight. Clearly, both accommodate to extreme changes in length 

 without visible folding. Randall and Jackson have considered at 

 some length the question of contractility in Stentor, and we shall 

 return to this matter in the next chapter. 



The adoral membranelles of Stentor (Text-fig. 15) are made up 

 individually of three rows of cilia (two rows in the buccal cavity) 

 with 20 to 25 cilia per row. Except for the presence of fine villous 

 extensions of the ciliary membrane there is no structure present 

 that would serve to bind the cilia together as a unit. From the 

 base of each kinetosome, numbers (usually about 10) of fine 

 fibrils, about 22 ra.fi in diameter and apparently tubular, extend 

 straight down toward the endoplasm. Root fibrils from the cilia 

 of each membranelle gradually converge into a bundle extending 

 into the endoplasm for as much as 20 /x. At their posterior 

 extremities the bundles bifurcate and join those from neighboring 

 membranelles, forming a zig-zag basal fiber. In addition, a thick 

 strand of fibrous material laterally joins fibers of adjacent mem- 

 branelles in the cortical region. Cross-sections of the membranelle 

 roots cut in the region where the fibrils are converging into a 

 bundle show a regular hexagonal pattern of fibrils linked by fine 

 (4 mfi) filaments — an arrangement in two dimensions remarkably 

 like the three-dimensional network in the infundibular fiber of 

 Campanella (p. 195). 



In view of the complexity of the total fibrillar system, and 

 particularly the length of the overlapping kinetodesmal fibrils, 

 Stentor *s capacity to recover from surgical insult is truly astonish- 

 ing. Tartar's (1960) recent results showing reconstitution of 

 normal patterns after thorough mincing of the ectoplasm prove 

 that cortical patches that have been totally disoriented have the 

 ability to realign themselves and fuse. This follows only after the 

 patches rotate to assume parallel or tandem, homopolar orienta- 

 tion, and implies in the latter case a healing or rapid regrowth of 

 the cut ends of kinetodesmal fibrils. What is perhaps equally 

 surprising is the persistence of a critical relationship between areas 

 of the body having wide stripes (widely spaced kineties) and those 

 with narrow stripes. The juxtaposition of abruptly contrasting 

 broad- and narrow-stripe zones normally determines the site of 



