138 ON A NEW CHARACTER IN: THE FRUIT OF QUERCUS. 
original position... When the ovules remain at the summit of the ovary 
above the seed, it is because they were pendulous at first ; when,at the 
base, it is because they were ascending when young... The imperfect 
state of herbaria has not enabled me to verify this as much as could 
be wished, but it is quite as it should be, and I have never found it 
otherwise. | 
— "This difference in the attachment appears at first important enough 
for generie or sectional characters, but when more closely examined, 
and attention is paid to what closely allied species have either kind: of 
ovules, the character is much weakened. The ovules originate laterally 
rom the re-entering, though incomplete, partitions which divide’ the 
ovary into three cells. They originate either near the base or near the 
summit of the ovary, or even at a certain appreciable distance from 
either. Their evolution is constantly semianatropal, the exostome being 
turned upwards, and this of itself proves that the superior ovules do not 
originate precisely in the superior angle of the cell. In the specimens of 
Quereus Suber which I have been able to: examine in different states of 
evolution, the ovules originate slightly above the base of the ovary, and 
the partitions are separated to the middle, as in Q: Robur, but the 
ovules being originally higher than in that species, they are found ulti- 
mately in'a spiral line round the mature seed, and the highest atro- 
phied ovale hardly extends to its middle line. If this evolution is con- 
stant, we have a specific character for Q. occidentalis and Suber, which 
have been so long confounded, and are so difficult to be distinguished, 
except by the duration of the fruit. Q: oveidentalis, to judge by a 
small number of acorns,* has its atrophied ovules decidedly inferior, 
like Q. Robur. Two Mexican species have afforded atrophied ovules 
above the base of, though still below the middle of the seed; and in 
some species with superior ovules, they are placed rather below the 
apex; hence the character is not absolutely clear. It will be used in 
the * Prodromus ’ to subdivide sections when combined with the dura- 
tion of the fruit. / 
The following is an epitome of the result to which I have been led 
after a more complete investigation than my predecessors. : 
The species of the genus Quercus form five very natural sections or 
subgenera, founded on the nature of the involucre or cupula, and con- 
b PROD H ii " fae fi 
Meier chatter ho da ase rino 
