XIT.] CONJUGATION AND SEXUAL REPRODUCTION I33 



no difference between the maternal and paternal nuclear rods, 

 but leads to a halving of their number in such a manner 

 that the most varied combinations can arise ; so that if ^ + ^, 

 and c -\- d represent four rods, there will be found in the 

 mature germ-cell not only the paternal group a -^ b and the 

 maternal c -^ d, but also the combination a -^ c and b -\- d or 

 a + d and b + c, that is combinations of any paternal with any 

 maternal element. 



Now it is clear that only very few distinct combinations 

 can be brought about in this way, — in the above-mentioned 

 case of four rods, only six combinations. But if, as actually 

 happens, each of the rods is doubled before their number 

 is halved, there are a greater number of possible combinations, 

 viz. in the above case, ten. Hence an individual of such a 

 species could produce ten kinds of eggs or spermatozoa with 

 differing hereditary tendencies. At the fertilization of one 

 of these eggs by a spermatozoon of another individual of the 

 same species, two different idants would meet each other. 

 Each parent produces ten different kinds of germ-cells, hence 

 as many different children can proceed from such a union, 

 as there are possible combinations between the ten kinds of 

 spermatozoa of the father and the ten kinds of ova of the 

 mother, namely ten times ten or a hundred. I therefore believe 

 that the significance of the longitudinal splitting of the idants, 

 and the consequent doubling of their number, is an increase in 

 the number of possible combinations. 



It may be doubted whether the increase which is thus 

 rendered possible is sufficient to explain certain phenomena 

 of heredity. So far as our knowledge extends, it has never 

 happened that two children of the same family born succes- 

 sively have had that resemblance to each other which is 

 familiar in the case of identical twins. Precisely similar germ- 

 plasm never seems to be twice formed in the unions of the 

 same parents ; it only occurs in those exceptional cases in 

 which a fertilized ovum produces two children, when the 

 germ-plasm which gives to both of them proceeds from a 

 single Qgg and a single spermatozoon. Now a hundred different 

 combinations of germ-plasms can occur under the given con- 

 ditions, while a human pair can scarcely produce more than 

 thirty children : but if only ten were born, one of the hundred 



