VARIATIONS IN SHAFT STRUCTURE — SPERM TAILS 45 



to end part of the way along the principal-piece. These nine outer 

 fibres are all of similar size in human sperm, though not in some 

 other animals (see below), and are thought to be built up of many 

 filaments about 75 A thick. In the neck region the anterior 

 extremities of the coarse fibres are segmented into a series of 

 transverse bands, in each of which all nine fibre blocks have been 

 fused to form a ring. This curious segmented structure surrounds 

 the base of the axial bundle and the centrioles. The two outer 

 fibres nearest to the plane in which the centrals lie are the shortest, 

 and end before the others posteriorly. There is uncertainty about 

 the origin of these outer fibres, though similar structures are 

 found in the flagella of Pyrsonympha (p. 39) as well as in inverte- 

 brate sperm (p. 41). 



Both of the tail sheaths of the human sperm may appear spiral, 

 but in neither case is the structure a simple helix. The modified 

 mitochondria of the fully developed sperm are moulded to fit 

 closely around the fibril axis, and may be arranged in a spiral 

 manner without the coalescence into a spiral that has frequently 

 been described. The sheath which surrounds the principal-piece 

 is built up of a thick spiral fibre which frequently splits, diverges 

 and rejoins. Each gyre of this fibre is markedly thickened in the 

 plane of the two central fibrils, so that the sheath has an oval 

 rather than a circular outline. Fawcett (1958a) and Cleland and 

 Rothschild (1959) have found that in various mammalian sperm 

 tails the successive gyres of the spiral are linked longitudinally 

 along the two lines of thickenings, but Anberg did not find this. 

 In fact, Fawcett believed that the sheath was formed from two 

 rows of semicircular ribs running between two longitudinal strands 

 and surrounding the fibril bundle. For much of the length of the 

 principal-piece, the spiral sheath is in contact with the axial fibrils 

 by means of projections from the thickened regions which extend 

 towards the nearest peripheral fibrils, and are therefore near to 

 the plane through the central fibrils. The spiral fibre decreases in 

 thickness posteriorly and there lacks connexion with the axial 

 fibrils. 



Cleland and Rothschild (1959) have worked out the complex 

 structure of the principal-piece of the bandicoot (Perameles — a 

 marsupial) sperm tail. The *' lozenge -shaped " (1 '73 x 1 '45 ju,) 

 cross -sectional outline of the tail shows a curious asymmetry of 



