248 THE EVOLUTION THEORY 



and how necessary, therefore, the continual re-combination of the 

 ids of the germ-plasm by means of amphimixis must be. We 

 understand why bi-sexual reproduction was only abandoned in 

 one generation, and that the one in which parthenogenesis was of 

 considerable advantage. But such transformations must have come 

 about with extreme slowness, since they were the result of climatic 

 changes which only come about very gradually. We thus come 

 again to the same conclusion to which we were led by our study 

 of vestigial organs in Man, that numerous species which appear 

 -to be at a standstill are continually working towards their own 

 improvement. But for this amphimixis is essential ; consequently 

 the descendants which have arisen through amphimixis, and whose 

 ancestors have arisen in the same way, have an advantage over those 

 of parthenogenetic origin. On the whole, at least, this must be so ; 

 in special cases it may be otherwise, namely, when the advantage 

 offered by parthenogenesis in respect to the maintenance of the 

 species preponderates over the advantage which amphimixis implies 

 as regards possibilities of transformation. 



As far as we have seen from the case of the gall-wasps, the 

 absence of amphimixis in every second generation implies no disad- 

 vantage in regard to the capability for transformation which the 

 species exhibits. As to whether any disadvantage would ensue if the 

 number of parthenogenetic generations in the life-cycle were greater 

 we can only guess, since no case is known which enables us to decide 

 this point, 2^to or con, with any certainty. The heterogony of the 

 plant-lice, the Aphides, and their relatives might be cited as against 

 the probability, for in this case a long series of parthenogenetic 

 generations often alternates with a single bi-sexual one, but the 

 difference in structure is not so great in this case, although it does 

 exist, and moreover we can quite well assume that the adaptation to 

 parthenogenesis was effected at the beginning of heterogony, when it 

 still consisted of a cycle of only two generations, and that further 

 virgin generations were interpolated subsequently. 



This assumption is supported by the fact that in some species of 

 our indigenous Ostracods, in Cypris vidua and Candona candens, in 

 contrast to the Daphnids, several bi-sexual generations alternate with 

 one parthenogenetic generation. But in this case again there is no 

 difference whatever in the structure of the two generations, the 

 parthenogenetic generation being distinguished from the bi-sexual 

 generation simply by the absence of males. 



The alternation of generations in the plant-lice is particularly 

 instructive, because it emphatically indicates how much Nature is 



