SYMMETRY OF THE CENTROSPERM X. 295 
from the opposite cotyledons to the alternate higher leaves, and might be equally found 
in truly alternate-leaved types. 
PHYTOLACCA ESCULENTA I. (Curve 21.) 
A specimen of the seedling plant of this perennial herb was uprooted in September. 
Above the tuberous tap-root at the base of the stem were two large opposite buds flanked 
by a smaller bud on each side and subtended by a single small bud. The large buds I 
assume to be those on the axils of the fallen cotyledons. 
Above these succeeded the decurrent bases of two fallen leaves, decussate in reference 
to the cotyledons, but not inserted at the same level. The distance from the cotyledons 
to the lower of these was 12:5 (circa), from the latter to the upper leaf 3 (circa). The 
succeeding two leaves were approximately decussate in reference to these, but not 
inserted at the same level; the interfoliar distances were 175 and 7'5 (circa) respec- 
tively. The next pair of leaves, though showing a decussate saps were wie not 
separated by an angle of 180°; the interfoliar measurements were 24 5 and 23:5 € 
tively. In all the remaining leaves evidence of opposite phyllotaxis, as far as "E ; 
divergence is concerned, was lacking. The following comprise the complete series o 
EO ies, Le 7:5, 245, 995, 37°5, 54, 57°5, 0075, 77, 54, 43, 30°5, 27°5, 24, 175, 14. 
Measurements were stopped at this point because the internodes above vp Fra 
and incomplete in length, and a number of the later measurements given here un- 
doubtedly represent immature internodes. 
The resultant internode-curve is zigzag, 
subeurves. 
but ean be analysed into two very regular 
