THE INDETERMINATE FORM OF ANTIIOTAXY 



201 



at length to a uiiiforni level, those of successively later (leveIo])ment 

 brought to successively higher ])oints, or left at successively lower 

 le\-els. This fact demonstrates that the cyniose or descending nature 

 of an inflorescence cannot be determined by noting the relative heights 

 of the flowers themselves, but only by noting the order of their 

 development. 



JS2 



MJ. 



J^SJ. J86 



Fig. 576. Scapo.sc 1-flowored peduncle of tulip. 577. Corymb of Crataegus. 578. Head of Cepha- 

 lantll^ls. 579. Umbel of ^Lscifpias. 580. Secund raceme of fiicucw/^o. 581. A descending inflorescence. 

 582. Ordinary raceme. 583. A spike. 584. Compound cyme of Saponaria. 585. Globular spadix 

 enclosed in spathe of Spathyema. 586. Cylindrical spadix of Acorus. 



The Indeterminate Form of Anthotaxy. — In the second series, the first 

 flower to develop is structurally the lowest of the cluster, the succession 

 being upward. Ascending or Acropetal (Figs. 582 and 583). If the 

 successive branches develop less rapidly than their predecessors, 

 the result is again a flat-topped inflorescence, with the development 

 from the outside to center, or Centripetal (Figs. 577 and 579). The 

 branches and flowers may be separated on obvious peduncles and 

 pedicels, or these may be not apparent, the flowers being sessile. In 

 accordance with the characters above explained, we obtain the following 

 simple forms of anthotaxy: 



