170 EVOLUTION 



mean e.o. jj. nf t P^jllijL-fl T1 ^ t.Kp rlTyprfsi ^y^f* 

 Hit. nff , while the mean value of the character, 

 e .g^ the ay er age stature 1 r eni a.i r^ ^ 1 1 n oh an gf fL 

 This is "neriodic selection," whose reality 

 Weldon proved in the case of one of the snails : 

 it can be detected by the diminution in the 

 extent or range of variability. 



There are other extensions of the selec- 

 tion-idea. Thus in social insets 1i|ce a.nta. 

 where community competes with community, 

 wg_ graft an adumbration of the intersociet^ry 

 struggle and selection which 



ir> T^!^^ There is also some meas- 

 ure of "struggle of parts within the organ- 

 ism/* as Roux called it, i.e. between compet- 

 ing organs and tissues and cells; and where 

 there is rapid multiplication of elements and 

 discriminate destruction there must be a 

 definite selective process. There is also a 

 well-known struggle between potential ova, 

 clearly illustrated in Hydra and Tubularia, 

 reminding us of the struggle between sister 

 queen-bees. There is a kind of struggle 

 among the hundreds of spermatozoa in their 

 race towards the ovum. Allowing a margin 

 for chance, the most vigorous and sensitive 

 spermatozoa will tend to succeed and this 

 will be for the advantage of the species. We 

 are quaintly reminded of the race between 

 drones to overtake ttie queen-bee in her 



