io8 EXPERIMENTAL EVOLUTION LECT. 



are some external causes in operation which are 

 not yet detected. M. C. Cornevin thus summarizes 

 the proportion of males to 100 females in the 

 following species : 



Horses 101 males against 100 females. 



Oxen 104-6 



Sheep 115-4 



Hogs 104-9 . 



Turkeys 120 



Guinea-fowl 102 



Common fowl 101 ,, 



Duck 115 



That external influences do play an important part 

 in the determination of sex is shown by numerous 

 facts. Spallanzani, Bernardi, and Autenrieth have 

 shown that female plants of hemp when mutilated 

 bear male flowers, and M tiller has in some cases seen 

 male plants bear female flowers. M tiller has also ob- 

 served female plants of Zea Mays bearing male flowers 

 when nutrition was deficient. Hoffmann has noticed 

 that in Lychnis, Spinacia, and Rumex the proportion of 

 sexes varies according to the greater or less interval 

 between each individual plant ; and Cornu, Giard, and 

 Magnin have shown that in Lychnis vespertinci, under 

 the influence of parasitic " rust " ( Ustilago antherarum\ 

 female flowers bear stamens. Prantl has seen similar 

 facts among Cryptogams. While the (seeds/ of ferns 



