The genus Lipocarpha R. BROWN, its morphology and systematic position in the family Cyperaceae

Abstract

Lipocarpha has long been a controversial genus as to its systematic position within the family Cyperaceae. Up till now, the genus has been attributed to either Hypolytreae (=Mapanieae) or Scirpeae, largely depending upon the morphological interpretation of the structure of its inflorescence. As illustrated in Fig. 1, a spike of Lipocarpha (Fig. 1, C) bears on its rhachis many herbaceous scales in the appearance of a glume, each subtending a bisexual flower placed between a pair of small, hyaline scales usually termed "inner scales" as shown in Fig. 1, D-I. Most cyperologists including NEES VON ESENBECK (1834), KUNTH (1815, 1837), BOCKELER (1871), PAX (1886) and OHWI (1944) assumed the homology between the "inner scales" of Lipocarpha and the "squamellae" of Mapanioid genera, and as the matter of sequence placed Lipocarpha in the tribe Hypolytreae. This assumption is not convincing, because the structure of the "Fachel" (a cymule) in the Mapanioid genera amply differs from the above structure of Lipocarpha not only in the strictly unisexual nature of flowers but in the arrangement of the "squamellae" themselves. Furthermore, the leaves of Mapanioid genera featured with highly developed mechanical tissues in association with the chlorenchyma well differentiated into pallisade and spongy portions (KOYAMA, 1966; METCALFE, 1971) do not suggest any similarity at all to those of Lipocarpha, which is characterized by the radiate chlorenchyma and the very poorly developed mechanical tissues as shown in Fig. 3. C. B. CLARKE (1893, 1908) and KERN (1974) treated Lipocarpha as a member of the tribe Scirpeae, judging that the "inner scales" constitute the perianth elements and hence are homologous to the hypogynous scales in Fuirena. If so, then why are the "inner scales" always two even in the trimerous flowers of L. microcephala and of L. chinensis, and are dorsi-ventrally situated? The leaf anatomy of the tribe Scirpeae (METCALFE, 1971) does not exhibit any close affinity with that of Lipocarpha in lacking the Chlorocyperus-type anatomy in the former. In my previous treatments (KOYAMA, 1960, 1961) I associated Lipocarpha with the tribe Cypereae, yet I have not had an opportunity to mention the details of this morphological thoughts. I realized at that time that the structure of a Lipocarpha-spike is essentially the same as those of specialized Cyperus-allies like Kyllinga and Remirea as schematically explained in Fig. 2, A and B. In this discussion, the floral unit of Lipocarpha composed of two "inner scales" and a bisexual flower represents a true spikelet, in which these "inner scales" are glumes, the upper one of which is bearing an axillary bisexual flower. These 1-flowered spikelets are borne at axils of the subtending bracts, which look like glumes in the Lipocarpha-spike. In Kyllinga the reduced spikelets are still recognizable over the small subtending bracts, while those of Lipocarpha are completely hidden by the glume-like bracts. Recently, RAYNAL (1973) expressed an opinion which is in principle concurring with mine just mentioned above. He has recognized Lipocarpha as being a highly specialized, ultimate group of Cypereae evolved from a mariscus-type ancestor, a view largely based upon the Chlorocyperus-type leaf anatomy occurring in both Lipocarpha and Mariscus. In this evolutionary trend, the spikelets of Hemicarpha bearing a single glume (=inner scale) can be derived from a Lipocarpha-spikelet by reduction, and it further gives the way to the spikelets of Rikliella which no longer retain any glumes at all and thus a spike has become indistinguishable from a spikelet of Scirpus (Fig. 3, C) structurally. In spite of this strong similarity between the Lipocarpha-allies and the tribe Cypereae, the former group appears clearly distinguishable from the latter by the far more reduced

(View PDF for the rest of the abstract.)