1892.] NATURAL SCIENCES OF PHILADELPHIA. 369 



pistils of an Iris. These were recurved as in the annexed diagram, 



U which shows a cross section of the three. 

 From this it would appear that the ovarium 

 is made up of a number of carpellary 

 1,.^,^^ leaves of which the midribs form the axis. 



In the normal flower the compound ovarv 

 is usuallv from 12- to 24-celled, according 

 to Gray's Genera; in this case but three 

 Transverse section of^taioid carpels wcre attempted. In this genus the 



carpels, replacing the normal gyn- ^• ^ . .^ • n ^ 



(xcmm mNymphceaodoyuia ^lue bctwecn the various floral organs is 



very finely drawn. Petals, stamens, aud the carpellary system with 

 its pistils run closely into each other. Hence the numbers in each 

 class of organs easily vary. It would not be wholly unexpected to 

 meet with cases in this or allied genera where the flowers would 

 have the pistils wholly aborted ; that is, the plants might produce 

 wholly staminate flowers. 



On the Sexual Characters of Rhus. 



Exact botany suffers much from the want of care in the applica- 

 tion of terms, especially illustrated in the use of the words herma- 

 phrodite, dioecious, polygamous, and jwlygamo-dioecious, by differ- 

 ent authors in connection with the genus Rhus. There is a section 

 described as hermaphrodite, in which the mist tree of gardens, 

 Rhus cotinoides, is placed. But I have shown that this species is truly 

 dioecious.^ Chapman, in his " Flora of the Southern United States," 

 divides the North American sjiecies into two sections, one " Flowers 

 polygamous," the other " Flowers ditiecious." He j^laces Rhus copal- 

 Una in the former section, leaving the dioecious section to the poison 

 Rhus, R. venenata., and R. Toxicodendron. Don, in the " Gardener's 

 Dictionary," places Rhus copallina in the dioecious section, with the 

 poison vines of Chapman, while Dr. Darlington, in " Flora Cestrica," 

 styles all the species " polygamo-dioecious." As these terms are 

 employed in the sectional characters, their use is perplexing to the 

 student. After a careful examination, extending over some years, 

 I have found no case in which an isolated plant j^roduces seeds. 

 The plants are all truly dioecious, and the terms hermaphrodite and 

 polygamous applied to any Rhus are misleading, and should be 

 abandoned. Often, isolated plants will be found in which the car- 



^ Proceedings of the American Association for the Advancement of Science, 

 Vol. XXII, pp. 73-75. 



25 



