152 
IOWA ACADEMY OF SCIENCE 
cycle incomplete. Britton (1) places the number of staminate flowers 
at 10 to 15, .which is too low for the material examined in this study. 
Danforth (3) and others have shown that certain of the Compositae 
have their flowers arranged in spirals. It is possible that in Iva the 
cyclic arrangement '-of lowers may have beeu derived from the spiral. 
The cyclic arrangement of the parts of the angiosperm flower has long 
been considered as derived from an ancestral spiral arrangement through 
the shortening of the floral axis. It appears that in the Compositae a 
similar transition has occurred with respect to the arrangements of 
flowers in the head, the cyclic being derived from the spiral through 
the shortening of the spike to form the capitulum. 
A floral bract subtends each dower in the head, . except the oute^ whorl 
of staminate dowers. The slender bracts of all other central dowers 
are short and stand erect in the interstices between the dowers., The 
bracts of the marginal pistillate dowsers, on the contrary,; are very, large, 
taper to a point and conform to the inner surface of the subtending 
involucral bracts, Britton ,(1) suggests that these constitute an inner 
whorl of involucral bracts- This study shows that they are intimately 
associated with the pistillate dowers (figs. 7, 8, 9, and 10) during their 
development and are morphologically similar to the floral bracts of the 
staminate dowers. Furthermore, if Warming’s (15) theory of the spicate 
origin of the capitulum is accepted, it would seem that these structures, 
subtending the pistillate dowers, should be considered doral bracts. 
The abortion of the doral bracts of the outer whorl of staminate 
dowers is probably due to their peculiar position. It is. evident that the 
excessive lateral development of both the involucral bracts and the bracts 
of the pistillate dowers would result in crowding and excessive protec- 
tion in this region. Knupp (7) believes that the development of the 
sepals of Myriophyllum was arrested through excessive protection. 
Warming (15) attributes the formation of pappus from the typical 
calyx to the pressure and crowding of dowers in the head. It is possible 
that these factors may have resulted in Iva in the complete abortion of 
the bracts of the outer whorl of staminate dowers. 
A study of the vascular anatomy of the head shows that the marginal 
dowers are most closely connected with the bundles of the stem. Each 
of the dve strands entering the head proceeds directly to an involucral 
bract. The pistillate dower is supplied by a branch from this bundle. 
Normally the dowers of each succeeding cycle receive their vascular 
supply through branches from the bundles of the next outer cycle. 
Whatever the determining factors in the arrangement of this system, 
