SPERM TAIL STRUCTURE AND MOVEMENT MECHANISM 153 



In so doing, we adhere to the system of numbering adopted for cilia 

 and flagella by Bradfield (1955), and Afzelius (1961), as follows. A 

 plane perpendicular to the line joining the midpoints of the two 

 central fibrils divides the axial filament complex into equal halves 

 with four fibrils on each side and one unpaired fibril lying in the 

 plane of bilateral symmetry. It is customary to call the unpaired 

 fibril one and to number the others in clockwise sequence. In tfie 

 mammalian sperm there is a coarse outer fiber for each doublet in 

 the axial filament complex, and these are given corresponding num- 

 bers. It is numbers 1, 5, and 6 of the outer row that are usually the 

 largest (Figs. 5-7). If the plane of bending of the sperm tail is per- 

 pendicular to the line (XY in Fig. 7) through the central pair of fi- 

 brils, as appears to be the case for cilia (Fawcett and Porter, 1954) 

 then the three largest fibers of the outer row are in the most favor- 

 able position to contribute effectively to movement in that plane. 



The nine outer fibers are thickest in the first part of the midpiece 

 but become thinner farther along the tail and terminate at different 

 levels. Their length varies considerably from one species to another. 

 In man and monkey they rapidly diminish in thickness in the mid- 

 piece and end in the proximal part of the principal piece. In rat, 

 mouse, and guinea pig, on the other hand, they taper more grad- 

 ually and extend throughout the greater part of the principal piece. 

 A systematic study of transverse sections of the sperm tails of these 

 species (Fawcett, 1961; Telkka et id. 1961) reveals that the nine fibers 

 are of different lengths and that they terminate in pairs in a regular 

 sequence along the length of the tail. Fibers 3 and 8, which are rela- 

 tively small in diameter at the outset and are least favorably located 

 to contribute to the bending movements, end first in the upper part 

 of the principal piece. The remaining fibers terminate at points 

 farther along this segment in order of increasing size. Four and 7 

 drop out, then 2 and 9, while 1, 5, and 6, which are largest at the 

 outset, continue nearest to the distal end of the principal piece. The 

 observation that those three fibers, which are best situated to con- 

 tribute to bending in the presumed plane of undulation of the tail, 

 are both the thickest and the longest favors the belief that the dense 

 peripheral fibers of the tail are contractile elements. 



Further support for this supposition has come from studies of 

 Cleland and Rothschild (1959) on the bandicoot spermatozoon. The 



