SYMMETRY OF THE CENTROSPERM.E. 299 
ERCILLA spicata. (Curves 25m, 25 1.) 
The lateral shoot measured showed at its base a number of very short internodes, which 
were ignored. Above these the following were the successive measurements :— 
2:5, 2, 9, 9:5, 2°75, 9:5, 8, 7°5, 8°5, 9, 85, 12, 19, 18:5, 19:5, 8, 14:6, 17:5, 17, 18:5, 12, 
8:5, 14, 21, 19, 22, 23:5, 27, 21, 195, 29:5, 22, 14. 
A little above the last measurement the brancli-end was broken off. 
Branch A given off from the preceding gave the following measurements :— 
15:5, 8, 30°5, 3:5, 33, 6'5, 11, 20°5, 1:5, 11:5, 6, 6, 6. 
Above the last measurement the end of the shoot was broken off. 
251 
The irregular internode-curve of the relatively main axis when analysed does not yield 
two regular subeurves, and thus supplies no evidence of any primitive opposite phyllo- 
taxis (when analysed into three subeurves these are moderately regular) But the 
Branch A shows an irregular internode-eurve which yields two very different but almost 
regular subeurves, each of which is disturbed at only one point; the double maximum 
of the displacement-subeurve cannot, however, be assumed to be a sign of a transition 
from the normal (ascending-descending) curve to a descending curve, for in other cases 
this curve undergoes such a change later than the internode-subcurve. 
PORTULACCACE.E. 
This family includes both alternate-leaved and opposite-leaved species. I examined 
one of the former, namely :— 
CALANDRINIA PILOSIUSCULA. (Curves 261, r.) 
A specimen of this species was uprooted. The base of the main stem showed indistinct 
scars and, a little higher up, a cluster of branches. The lowest. two of these were 
approximately opposite, and the following pair seemed to be L EA ME in 
reference to them. The lowest two branches (Branches 1 and I’) were measured. — 
ended in a terminal flower, to the peduncle of which the continuing branch was * fused. 
The last measurement in each case was to the topmost leat beneath the terminal flower. 
The internode-curves are irregular and zigzag, but show strong likeness to each 
other, and this likeness is even more striking when comparison 1$ made between the 
first two-thirds of the two internode-subcurves and the two displacement-subcurves 
2y 
SECOND SERIES.—BOTANY, VOL. VII. 
