378 THE AMERICAN NATURALIST. [Vor. XXXIV. 
VI. INFLORESCENCE, AND CHANCES OF MATURING FRUIT. 
The flowering stem of Z. versicolor bears a terminal cluster 
of three flowers and generally one or more lateral clusters of 
two flowers ! each, the latter springing from the axils of alter- 
nating bracts. Fig. 4 represents a stem which bore the typical 
complement of flowers. For convenience of reference I desig- 
nate these clusters by the Roman numerals applied to them 
in the figure. Cluster / is developed in all 
(100 per cent); cluster //, in about 60 per 
cent of the stems; cluster ///, in about 10 
per cent of them; and cluster /V is rarely 
developed; I found it but once during the 
season.? The small letters indicate the 
several flowers in the clusters and also 
the order of their opening. The clusters 
are nearly isochronous, but there is some 
irregularity. Flower / a is generally the 
first to open. The few flowers which I 
Fic. 4.—Diagram of the Marked for observation opened in the 
ee morning before nine o’clock, and closed by 
withering during the afternoon of the day following. Flowers 
a, 6, and ¢ follow each other in regular succession, each being 
! The statement of Goodale (Wild Flowers of North America, p. 32, 1882) that 
"the flowers may be single or in clusters of two, rarely more” is certainly not 
true for the vicinity of Lake Forest; I found but a single clump to which it would 
at all apply — a clump of much-dwarfed plants growing on a very dry hillock in a 
pasture near Fort Sheridan. This singular clump bore twenty flowering stems, of 
which five had cluster Z only, with but two flowers (a and 4) in that cluster; 
twelve had clusters Z and ZZ developed, with the second flower (4) developed in 
Z7 but twice; and three with clusters Z, ZZ, and 7/7 developed, and only one flower 
(a) present in cluster ZZZ. Elsewhere I found three examples of cluster / in 
which a fourth flower (7) was developed, and two in which the third flower (c) 
was not developed. The usual four bracts were developed in all; but the termi- 
nal bract, usually sterile, was fertile in three, and the third bract, usually fertile, 
was sterile in the other two. I also found a single cluster ZZ containing three 
flowers. 
? These percentages w were derived from a count of 214 stems — the same ones 
other localities near Lake Forest yielded slightly different results, as follows: 
cluster Z developed in all (100 per cent); cluster ZZ developed in 87 (67.5 per 
cent); cluster 7/7 developed in 21 (16.3 per cent). 
