(136) 
4. The method of taking measurements. 
5. The tabulation of data. 
The Choice of Leaves. —Three species of plants were em- 
ployed in the experiments: Parthenocissus quinguefolia (the 
Virginia creeper), Zrifolium pratense (the red clover) and 
Lupinus albus (the white lupine). Each species will be 
taken up separately. 
In the case of Parthenocissus, 44 normal and 67 operated 
leaves were used. A few vines were taken from outdoors 
and planted in the propagating house. They were allowed 
to run along a wire near the glass roof. Leaves were chosen 
from various portions of the stem in such a way that the 
distribution was as nearly as possible the same in the two 
series. Only leaves with five leaflets were used. 
In Trifolium, 31 normal and 84% operated leaves were 
used. Several clumps of the clover were transplanted from 
outdoors to the propagating house and, here also, leaves were 
taken from various regions of the plant and the two series were 
made as nearly alike as possible. Only leaves with three 
leaflets were used. 
In Lupinus, 16 normal and 53 operated leaves were used. 
The plants were raised from the seed and the order in which 
the leaves appeared was noted. Several plants were neces- 
sary for each series, each of which, therefore, had leaves 
from all the different positions. Only the first four or five 
leaves were used because it was desired to have the ages as 
nearly alike as possible and also because in many cases these 
were the only ones with five leaflets, those further up on the 
stem having six or seven. 
Taking into consideration the number of cases in each 
group it seems improbable that there is any considerable error 
due to original differences between the two series of one 
species. 
The Operation. — For the sake of clearness in description 
it is necessary before proceeding further to give the general 
topography of the leaves used. F. 7 is a diagram of a 
