Botanical Journal of the Linnean Society, 72: 115‑148. With 8
plates and 3 figures
Reproduced here with the
permission of the author
The floral anatomy of Victoria Schomb.
E. L. SCHNEIDER
Lindley (1830) established the taxon Ranales and was the first to place it at the base of the dicotyledons. Subsequent authors, notably Bessey (1897), accepted the 'primitiveness' of this taxon and as a result formulated the "Ranalian Theory". This theory postulates that the most primitive type of flowering plant (i.e. that from which others have been derived) is characterized by flowers which: are bisexual; have a perianth composed of many sepals and petals which are free; have many stamens and carpels which are also free and separate; and have these organs spirally arranged on an elongated floral axis. Bailey & Tupper (1918) demonstrated the evolutionary trend for the secondary xylem and thereby placed the 'Ranalian Theory' on a sound anatomical basis. Their concept of the trend rested upon the phyletic reduction in the length of the tracheary elements, and their placement of the Ranales (sensu Eames, 1961) was based upon its possession of the longest xylary elements among the dicotyledons. This xylary feature, associated with other primitive features such as monocolpate (anasulcate) pollen (Erdtman, 1954; Walker, 1974) and the presence of primitive sieve tubes (personal communication, Carlquist*), has brought wide acceptance of the Ranales as primitive angiosperms (e.g. Cronquist, 1968; Takhtajan, 1959, 1969).
As a corollary to the Ranalian Theory, it is believed that the Ranales should reveal primitive angiosperm features (e.g. Bailey & Swamy, 1951) which, if discovered and evaluated, may be of value in solving the 'abominable mystery' of Angiosperm descent.
Since thorough studies of all ranalian taxa have been urged (Bailey & Smith, 1942), this investigation represents one in a series dealing with the morphology of the Nymphaeaceae (sensu lato) initiated by Moseley (1958, 1961, 1965, 1971). The ultimate goal of these studies is threefold: (1) to clarify the interrelationships among the nine genera of the Nymphaeales (sensu lato), (2) to gain further insight into the nature and evolution of ranalian flowers and hence achieve a better understanding of the nature of the angiosperm flower, and (3) to further our understanding of floral anatomy in general, and that of the Nymphaeaceae specifically.
METHODS AND MATERIALS
The flowers and peduncles of Victoria amazonica, V. cruziana var. cruziana and V. cruziana var. trickeri† have been examined. Material has been collected exclusively from various botanic gardens, as indicated in Table 1, owing to the remote natural distribution of this genus.
Material collected for examination was fixed in 70% formalin‑acetic acid (Johansen, 1940). Preliminary dissection was usually necessary because of the large size and extensive aerenchyma of the flowers. The material was the dehydrated in a tertiary butyl alcohol series and embedded in Tissuemat (61'Q following standard techniques. Transverse and longitudinal serial sections of flowers at various developmental stages were made at 10 to 20 μm on a conventional rotary microtome. Sections were then stained in Harris's hematoxylin, safranin and fast green.
Mature whole flowers and fruits were also serially sectioned, both transversely and longitudinally, by the use of a commercial meat slicer, which allowed uniform sections less than 1 mm in thickness. Clearing and/or staining of this material was generally unnecessary, but when clearing was deemed necessary it was done either in 70% ' lactic acid or in 2% sodium hydroxide, washed in water, and then placed in 10% chloral hydrate. The cleared material was then stained with a solution of basic fuchsin in ammonia and stored in glycerin.
Voucher specimens of the material investigated have been filed in the herbarium at the University of California, Santa Barbara (UCSB).
CarIquist has recently observed sieve tube members in the secondary phloem of Austrobaileya. Prior to this discovery, it
was believed that this genus had only sieve cells (Bailey & Swamy, 1949;
† As indicated
by Cutter (1961) there is, in the available literature, disagreement about the
number of species of Victoria. Li (1955), Planchon (1850‑52, 53) and Conrad (in Bailey, 1939)
believed there to be two species or possibly three. Lemaire (1847) accepted
only one species, V. regina (V. amazonica), and considered other species as varieties. In this investigation, V. trickeri is considered as a variety of V.
cruziana of d'Orbigny (1840). V.
amazonica (Poeppig) Sowerby (V. regia Lindl., Euryale amazonica Poeppig) is considered
distinct from V.
than as a variety of it as held by Hooker (1847).