IN THE TROPICS : III. 99 



lessened after the process of segregation had taken place, as 

 compared with that of the same character in a pure race. 



It will not be necessary to follow Correns through the whole 

 of his account of the details of wi xenia" in the case of various 

 characters. Suffice it to say that, so far as my experience goes. 

 that account is remarkably accurate, and that I have found 

 no case of disagreement in my own experiments. Thus, so far 

 as "xenia" itself is concerned, it will only be necessary to 

 describe a few cases in which races were examined which 

 Correns did not make use of, and these will afford differences 

 only of detail. 



Correns's results were uncomplicated by cases of coupling, 

 no such effect being observed in any of the characters men- 

 tioned above. 



Various cases were observed where a cross of the form A x a 

 differed in appearance from one of the forma x A, such cases 

 being also confined to characters of the endosperm. These 

 again are explained by reference to the greater quantity of 

 heritable substance presumably introduced on the female side 

 of the transaction. 



In a later paper Correns records a curious apparent excep- 

 tion to Mendel's law in the case of a cross between a blue sugar 

 corn and anon-blue pop corn. Here the hetero/ygote on self- 

 pollination yielded only 15-61 per cent, of sugary grains a tnong 

 a total of nearly 9,000. On pollination with the recessive 

 parent, however, there appeared nearly 50 per cent, of sugar 

 grains, so that the segregation of at least the female germ cell 

 was of the normal character. Correns suggests that in this case 

 the combinations A x A, A x a, a x A, a x a, do not all take 

 place with equal facility, owing to competition among the 

 pollen cells or to some other cause. 



CULTIVATION OF MAIZE IN GEY LOW 

 At the elevation of Peradeniya not only the native strain 

 commonly cultivated, but also several American sorts which 

 were specially introduced for the purpose of crossinj; with the 



