CYCADACEAE UPDATED (1999)
 D. J. DE LAUBENFELS

INTRODUCTION
The most recent revision of the genus Cycas was more than sixty years ago by Schuster (l932) where he listed eight species with numerous subspecies and varieties in an almost incomprehensible confusion of categories.  Six years earlier Pilger reported “approximately 15” species but named only 13.  As a matter of fact, over forty species had already been described, at least half of which are perfectly valid.  Subsequently more than two dozen additional species have been offered, few of which are really new.  Several regional monographs [Thailand (Smitinand, l971, 1972), China (Cheng & Fu, 1978), the Philippines (Amoroso, 1986), and Queensland (Hill, 1992)] have lately added to our knowledge of the genus, but these have not been coordinated with areas outside of each treatment.  Jones (1993) undertook to describe every possible species of cycad from a horticultural point of view and achieved 44 in the genus Cycas.  He did not attempt any analysis and more than one third of his descriptions are duplicates and he missed a few.  The genus therefore remains very much in need of revision. 

The genus was originally described by Linnaeus (1737) based on a variety of material, much of it cultivated but including references to India, Amboina, and Japan.  Later, separate species were distinguished for Japan (Thunberg, 1783), Cochin China (DeLaureiro, 1790), Queensland (Brown, 1810), Bengal (Hamilton, 1825), and Madagascar [Desfontaines (ex Gaudichaud), 1826].  Confusion began with Roxburgh (1832) who described two species based on cultivated material in Calcutta, said to be from the East Indies.  The female specimen attributed to Ambon (specimen in BM) does not correspond to known cycads from there, instead it appears to be silvestris, common further west.  The associated male specimens which he described correspond best to javana, also from further west.  He actually described “two out of three” male specimens but did not describe the third.  Clearly, these “Ambon” cultivars represent a mixture.  At any rate, Roxburgh identified this material with circinalis.  The material attributed to the Moluccas (also in BM) appears to derive from nearby Orissa Province and Roxburgh erected the species sphaerica for it.  It does not help that he then opined that it better fits the Todda pana of Rheede, upon which circinalis is based, than does the Ambon material, which he had identified as circinalis.  The result of all this is to establish the erroneous idea that circinalis is represented from India through the East Indies. 

In a series of papers, Miquel tried to characterize the genus Cycas.  Thus (1839), among other things, he erected rumphii and celebica for the cycads described from Ambon and Celebes before Linnaeus by Rumphius, adding in 1840 circinalis var. angustifolia (India), rumphii var. timorensis (Timor), madagascarensis (Madagascar), and glauca (cult.).  Later (1842) he described circinalis var. javana from female material that came from Java and (1843) equated Roxburgh’s circinalis with his rumphii.  He assigned sphaerica to circinalis (1851).  Eventually (1868), he put both celebica and circinalis var. javana into rumphii, whose description he now altered to correspond to the Java material, a regrettable move because more than one species is clearly involved.  However, he again allowed sphaerica as a distinct species and he asserted that the identification of many cultivars as circinalis was inappropriate.  Miquel’s work suggested that rumphii is quite widespread, more so than is true.

The revision of the genus by Schuster further confounded the understanding of the genus.  By use of subspecies, varieties, and forms, he assembled a large part of the genus into circinalis.  Several other species were put into rumphii.  This has further promoted a broad species concept in the genus so that all sorts of things have been treated as circinalis, various others as rumphii, but his species subunits are seldom if ever used.  No wonder that recent authors express reservations about the identity of many cultivars when provenance is not definitely known.

DIVISION OF CYCAS INTO TWO GENERA
Only recently a group of strikingly different species of Cycas in southeastern Asia has become known in detail.  Unlike most other species put in the genus Cycas, these plants arise from a bulbous underground base.  Such bases occur in other cycad genera in other families along with emergent trunks but in these genera the emergent trunks are an unmodified continuation of the underground part, or the bulbous bases are not underground at all.  In the species of Cycas in question the underground bases often eventually do produce an aerial stem more slender than and not at all a continuation of the underground part. 

Schuster (1932) was the first to put the species of Cycas with bulbous trunks into a separate section which he named Indosinensis, including the two species then known to have this character.  Smitinand (1971) distinguished a section Stangerioides, based on the presence of a subterranean trunk, to accomodate C. micholitzii, leaving out C. siamensis because of its well-developed aerial trunk.  However, the latter has a gigantic, mostly or entirely subterranean bulbous base and just like the former only eventually produces a slender aerial trunk.  Two other species with underground bulbous bases had been described before Schuster introduced his section, but their underground base was not reported.  With the discovery of several more such species, the new genus Epicycas de Laub. (1998) was created to recognize this distinctive group of species. 

CYCADACEAE
Palm-like plants with or without underground bulbs, aerial trunks (rarely absent) with a large pith and generally armored with the bases of older leaves. Leaves large and pinnately divided, petiole usually with thorns corresponding to the pinnules above. Pinnules numerous, linear but more or less lanceolate, acuminate when about 15 mm wide or more, apiculate, straight or curved either basally in most species or apically in others, flat to revolute margined, with a single usually prominent midvein, more or less narrowed at the base (particularly in the lower part of the leaf), sessile, decurrent, and entire, after drying the base more or less placed in a groove on the rachis usually somewhat above the center of either side so that the lower side of the rachis is more prominent. Cataphylls in clusters alternating with clusters of leaves or of fertile structures, lanceolate, about 5-10 cm long and 2 mm wide at the base,sometimes pungent, covered like the fertile structures and the emerging leaves with a more or less dense tomentum usually of an orange-brown color.  Pollen cones terminal on a short, 2--13 cm, scaly stalk about 2 cm in diameter, ovoid or cylindrical, usually tapering, compact but elongating to shed the pollen, provided with numerous spirally placed microsporophylls, the growing point continuing by means of a lateral bud at the base of the pollen cone.  Microsporophylls numerous, spirally placed, wedge-shaped, about 3 cm long but variable depending on the placement within the cone, with numerous clusters (sori) of pollen sacs on the abapical side where there is a weak left-right division, pollen sac clusters extending from near the base nearly to the widest part, a sterile apical portion then narrowing gradually to abruptly to an apical spine or acumen, the whole apex or just the acumen turned upwards towards the apex of the cone with the amount of turning and sometimes the length of the apical part depending on the placement on the cone, basally to apically.  Megasporophylls clustered like the leaves, rather loosely to somewhat compact and eventually displaced by the continued growth of the apex, each sporophyll with a stalk about one cm wide or sometimes less, the stalk of quite variable length with the lower longer, bearing towards the apex two lateral rows of more or less opposite ovules, the apical part beyond the ovules expanded laterally gradually to abruptly to a broad flat sterile structure triangular to long-lanceolate and usually bearing lateral teeth which may be 1--2 mm long or in other species as much as 3 cm long, the longer teeth resembling pinnules, the teeth in most cases increasing in length from the base with the longer sometimes reaching their greatest length well below the acumen or all may be of the same size, ending in an acute long to short terminal acumen or a spike whose base may bear a few reduced teeth.  Seeds globular, longer than wide and wider than thick, usually glabrous but in two species with tomentum, with a fleshy outer layer and an inner stony layer.  Two hypogeal cotyledons.

KEY TO THE GENERA

Click here to return to the Outline

This site is copyrighted © 1998-2006, Palm & Cycad Societies of Florida, Inc.
For questions or comments, e-mail the webmaster.
Internet hosting provided by Zone 10, Inc.