COMMELINACE^E. 



MORPHOLOGICAL AND ANATOMICAL STUDIES OF THE VEGETATIVE ORGANS OF SOME 

 NORTH AND CENTRAL AMERICAN SPECIES. 



Hv Theodore Holm. 



(With Plates I- VIII.) 



Characteristic of the families that constitute the order Enantioolasta& is the atropous ovule." 

 The flowers are hypogynous; in some of the families the general monocotyledonous type with live 

 trimerous whorls completely developed may be readily recognized, while in others the flowers 

 arc so much reduced that the type is difficult to trace, as, for instance, in Restionacese and Cen- 

 trolepidacese. The habit of these plants is somewhat peculiar, and it seems as if a certain struc- 

 ture in regard to inflorescence, ramification of shoot, shape of leaves, etc., is prevalent in most 

 of these families. The Mayacacest show a habit unlike that of any other aquatic plants; the Xyri- 

 dacese with their cone-shaped inflorescences, the Eriocaulacese with their capitula, the Rapatean ;> 

 and the Restionacese represent types of very distinct and characteristic aspect. In the Commeli- 

 nacese, on the other hand, the structure of the flower exhibits a very pronounced variation from 

 actinomorphic (Tradescantia) to zygomorphic (Commelina); the foliage is also to some extent 

 different within certain genera ( Tradescantia), besides that the rhizomes exhibit several types of 

 growth characteristic of certain genera, or of species within the same genus. In other words, 

 the Comvielinacese do not possess a habit of their own or of so special and well-marked peculiarity 

 as the other families of the Enantioblastse; nevertheless the family seems to be a very natural one, 

 and, to use the words of Bentham and Hooker: "Ordo totus optime limitatus, nee cum ullo alio 

 generibus intermediis junctus." Several and very excellent monographs have been published on 

 some of these families. Masters has treated the Restionacese ; b Hieronymus the C'entrolepi- 

 </</<■< ,'iV Seubert the Xyridacese and Mayacacese.; d Bongard/ Kcernicke/ and Ruhland? the 

 Eriocaulacese; Hasskarl* and Clarke' the Commelinacese,. Besides these works a few papers 

 have also been published on the morphology and anatomy of these plants, for instance, by Cili;.-' 

 NlLSSON,* Poulsen.' Van Tieghem,'" and the author." 



" In a recently published paper, " Beitriige zur Morphologie der Commelineen " (Flora 1904, p. 512), Mr. J. Clark 

 stales that the atropous ovule is only characteristic of the genus Tradescantia as far as concerns the Commelinaeeie, 

 while this type of ovule occurs merely as an exception in "all the other genera of the order." Nevertheless the 

 ovules of Weldenia are certainly atropous, and we hope that some future investigator will undertake the study of this 

 particular point in respect to our native representatives. 



''In De Candolle: Monographic Phanerog. Vol. I. p. L'ls. 



= Abhdl. Naturf. Gesellsch. Halle. Vol.12. 1873. 



''In Martins: Flora I'.rasil. Vol.111. 1842-1871. Pars. I, pp. 211-227. 



< Mem. de l'Acad. imp. de St. Petersbourg. 1831 cet. 



/In Martius: Flora Brasil. Vol. III. 1842-1871. Pars. I, p. 273. 



PlnEngler: Das Pflanzenreich. 1903. 



h Commelinacese Indies. 1870. 



»In De Candolle: Monogr. Phanerog. Vol. III. 1881. 



-/Beitriige zur vergleichenden Anatomic der xerophilen Familie der Restiacess. Inauj;. diss. Leipzig. 1891. 



*Studien ueber die Xyrideen. Kgl. Sv. Vet. Akad. Hdlgr. Vol.24. 1892. 



' Anatomiske Studier over Eriocaulaceerne. Thesis. Kjobenhavn. 1888. 



Anatomiske Studier over Xiyra-Slaegtens vegetative organer. Vidensk. Medd. naturh. For. Kjobenhavn. 1891. 



Anatomiske Studier over Mayaca Aubl. Overs. Kgl. D. Vid. Forhdlgr. 1886. 



'" Structure de la racine et disposition des radicelles dans les ( V ntmlr/iiilrta, .lour/?*, Mayaaea el XyricUes. Journ. 

 de Botan. 1887. 



n Eriiji-ntdoii dernntjulare L.; an anatomical studv. Bot. Gaz. 1901. 



159 



