(161 ) 
group minus the first gives us the increase due to the presence 
of the petiole or, in other words, the petdole factor. 
2 
this means we arrive at a value of approximately + 37°, 
which represents the tncrease inan interval as a result of the 
presence of the petiole within tt. 
Considering the character 
of the experiments the similarity in the values obtained from 
the different intervals is very striking. Thus we get 29° 
from the interval AB, 40° from BC, 41° from CE and 347° 
from ZA. 
It should be noticed that in the last case, that of 
ZA, where the distribution between the two groups is most 
even, we have the closest agreement with the average, as we 
should expect. 
Changes in the Length of the Leaflets. 
Parthenocissus guinguefolia.— A comparison of the lengths 
of the leaflets in the normal series with those in the operated 
series, is given in Table XVI, which is derived from Tables I 
and II, 
We find that each of the leafiets shows a considerable 
decrease and this decrease in each case is above the sum of 
the probable errors of the averages. 
Trifolium pratense. The changes in length as figured 
out from Tables III and IV are given in Table XVII. The 
general result is very similar to that obtained in Parthenocis- 
sus the percentage of decrease being nearly the same. 
Lupinus albus.— The changes in length as figured out 
from Tables V and VI are given in Table XVIII. 
If we 
TABLE XVI. 
PARTHENOCISSUS QUINQUEFOLIA. 
CHANG 
E IN LENGTH OF LEAFLETS, 
(MEASUREMENTS ARE IN MILLIMETERS.) 
| A | B c | D E 
Normal Series. 65.3 + 1.44] 80.0 + 2.93 eran 80.1 13.71 | 66.4 + 1.47 
Operated Series. 61.0+1.13| 74.64 1.15 | 79.3 + 1.26 60.6 + 1.00 
Difference = Change = —5.4 ar —58 
EP. EM. #257 | 4.08 42.86 | £2.47 
Per cent. of Change.| —6.6% | —6.1% —5.1% | 8.2%, 
Average Change. | —6.8% 
