221924) 
HARRIS—THE INFLORESCENCE OF MANFREDA VIRGINICA 415 
entirely from the vicinity of Cliff Cave. The 1000 stalks taken 
in 1906 for countings of flowers and fruits only and the entire 
series of 1270 inflorescences taken in 1907 are therefore directly 
comparable as samples of the same species from the same locality 
for 2 years. 
III. PRESENTATION AND ANALYSIS OF DATA 
1. TYPE AND VARIATION IN DIFFERENT HABITATS 
In this section we have to consider the frequency distributions 
of number of flowers and fruits per inflorescence, and the number 
of seeds matured per locule and per fruit. 
In doing this we shall hope to replace the impression of chaotic 
disorder which must be the result of mere inspection in the field 
by a definite mental picture of the orderliness which prevails in 
the frequencies of the numbers of flowers and fruits, in the position 
of insertion of the fruits, and in the number of seeds per locule and 
per fruit. 
The frequency distribution of number of flowers per inflores- 
cence in the 4 series in which the number of seeds was determined 
is shown in ает! For the 3 other series, in which flowers and 
fruits only were counted, the reader must consult table rr. 
. Because of the wide range of variation in number of flowers per 
inflorescence, very irregular graphs are obtained if the ungrouped 
frequencies are plotted. Grouping in classes of 3-units range and 
reducing to percentage frequencies we have the distributions for 
the two habitats represented in fig. 1 for 1906 and in fig. 2 for 
1907. In 1906 the results for Meramec Highlands (N = 300) 
and Jefferson Barracks (N = 1250) are not in good agreement. 
The distribution for Meramec Highlands is particularly irregular. , 
but this is doubtless due ір large part to the fact that this series 
comprises only 300 inflorescences—less than one quarter the 
. Dumber from Jefferson Barracks. In 1907 the: distributions 
. Tepresent 605 inflorescences for Meramec Highlands and 1270 for 
_ Jefferson Barracks, and are in much closer agreement. 
‘The frequencies of inflorescences with varying numbers of flowers are alone 
tequired here. The total fruits and total seeds which are also given will be used 
for the determination of correlations in a subsequent section. 
