828 



ECOLOGY 



petals and to others that petals are transformed stamens, neither view 

 having adequate support. When the calyx, corolla, and stamens are 

 inserted on the receptacle below the ovary, the flower is called hypogynous 

 (figs. 1137, 1138); when the corolla and stamens are inserted on the 

 calyx at the level of the ovary, the flower is called perigynous (fig. 1139); 

 and when the calyx appears to be inserted on the ovary, the flower is 

 called epigynous (fig. 1140). A determinate inflorescence is one in which 



1139 



Figs. 1138-1140. — Diagrams, showing the position of the floral organs in hypogy- 

 nous (1138), perigynous (1139), and epigynous (1140) flowers; in 1138 the calyx, corolla, 

 and stamens are attached to the receptacle; in 1139 the corolla and stamens are attached 

 to the calyx tube; in 1140 the other floral organs appear to be attached to the ovary. — 

 From Ganong. 



the terminal flower blossoms first, while an indeterminate inflorescence 

 is one in which the lateral flowers blossom first, so that a shoot may con- 

 tinue to bloom somewhat indefinitely (fig. 1141). Cymes are a represen- 

 tative form of determinate inflorescence (fig. 11 36), and common forms 

 of indeterminate inflorescences are spikes (fig. 1163), catkins (fig. 1161), 

 racemes (fig. 1199), corymbs (fig. 1173), umbels (figs. 1196, 1197), pan- 

 icles (fig. 1162), and heads (fig. 1193). 



The significance of differences in floral structure. — The floral diver- 

 gences heretofore noted are of great convenience in classification, because 

 they are relatively invariable, but they appear to have had little or no 

 significance in determining the success or failure of plants. It is believed, 

 for example, that the trend of plant evolution has been toward epigyny, 

 but there is practically no evidence that epigyny is more advantageous 

 than hypogyny. Monocliny or dicliny and the presence or absence of 

 a perianth may be of greater consequence, and they will be considered 

 later, but it appears that floral evolution has taken place in large part 

 without relation to r61e or to ecological advantage, especially in those 

 structures most used in classification. In many other respects, how- 



