14 ILLINOIS BIOLOGICAL MONOGRAPHS [104 



gradations from the more or less membranous to the thick, fleshy or even 

 coriaceous types. In many species there are present in the calyx, corolla 

 and capsular parts large pellucid or semi-pellucid glandular-appearing cells. 

 The cells of the flowers in some species (C. indecora and C. coryli) are lens 

 shaped giving the flowers a verrucose or papillate appearance and in others 

 (C. pentagona pubescens, C. gracillima saccharata, etc.) these papulations 

 are somewhat longer and give the flowers a pubescent appearance. 



The calyx is gamosepalous in the majority of the species and always 

 persistent. A small number, however, have the calyx segments entirely 

 free. The shape, texture, degree of separation, and overlapping of the 

 segments are characters that are useful for specific differentiation. 



The corolla is always gamopetalous, and usually tubular or campanu- 

 late frequently becoming urceolate as the fruit matures. The length of 

 the corolla lobes in comparison with the length of the tube; the shape of 

 the tube and of the lobes; the position assumed by the lobes, that is, 

 whether upright, spreading or reflexed; the texture of the corolla and the 

 shape of the cells causing papillate forms, and the presence or absence of 

 horn-like projections on the dorsal surface of the lobes are characters which 

 aid in the separation of the species. In certain species the shape of the 

 corolla is such that upon the maturation of the capsule the withered 

 corolla is left at its base (C. pentagona) ; in others, because of its narrowness, 

 it is torn loose from the base and carried either about the capsule (C. 

 calif ornica) or at its apex like a hood (C. cephalanthi). 



Opposite the stamens and alternating with the lobes of the corolla 

 are found a set of scale-like appendages in all but two of the North American 

 species. The morphology and function of these organs are somewhat 

 questionable. Babington (1844) thought that they were inserted opposite 

 the corolla lobes and had become joined opposite the stamens. He con- 

 sidered also that, as they alternate with the stamens, they should be 

 considered as an inner whorl of modified stamens. 



Engelmann says (1859): "The most peculiar organs of the flower are 

 the epistamineal scales, which are found in most of the species. The 

 simplest form of that organ (in C. inflexa, C. chlorocarpa, etc.) exhibits a 

 few teeth or lobes laterally adhering to the lower (attached) part of the 

 filament. These lobes, in other species, expand into membranes, forming 

 two lateral wings to the filament, crenulate or fringed at the tip and outside; 

 then these wings partly united at their upper end, thus forming a single 

 bifid scale; finally they unite entirely, forming an oblong, ovate, spatulate 

 or truncate, more or less crenate or fimbriate scale. Towards the base 

 the scales are always 'adnate in the middle,' or, properly speaking, attached 

 to both sides of the adnate filament. Their bases usually connect with 

 one another, forming inverted arches. . . . These scales are evidently 



