Reclassification of Costaceae

As a result of extensive study of the DNA makeup of plants in the family Costaceae, three new genera have been established, dividing the genus Costus into the following four genera:
  1. Cheilocostus - Asian Costus species - C. speciosus, C. lacerus, C. globosus, C. sopuensis
  2. Paracostus - former subgenus Paracostus - C. englerianus and C. paradoxus
  3. Chamaecostus - former neotropical members of subgenus Cadalvena - C. cuspidatus, C. subsessilus and others
  4. Costus - all the remaining African and neotropical species of Costus.
The genera Monocostus, Dimerocostus, and Tapeinochilos remain unchanged.

Okay Costus fans, before I start hearing moans and groans about "those people" who are always changing plant names I need to clearly explain that this is not some capricious whim of some egghead scientist. These changes in nomenclature are the end result of a long period of extensive study by the foremost world expert in Costaceae - Dr. Chelsea D. Specht.

Dr. Chelsea D. SpechtI have known Chelsea for several years, and I can tell you she has her feet firmly planted on the ground. She worked in the field in Bolivia for the World Wildlife Fund, then at the New York Botanical Garden and at the Smithsonian Institution before taking her current position as an Assistant Professor at the University of California, Berkeley. You can read more about her work at, or even visit her research lab at Berkeley at (Some things you will not read on those pages - her favorite plant in Costaceae is the beautiful yellow-flowering Monocostus uniflorus which really began her love of Costaceae. She also likes to snack on costus flowers.)

When Chelsea started studying the molecular phylogeny (DNA characteristics) of plants in Costaceae, she recognized that the traditional classifications based on floral morphology - the appearance of the plants - could not be fully supported scientifically. Through cladistic analysis, she found that there is a close association between the Asian Costus species, C. lacerus and C. globosus, with plants in the genus Tapeinochilos. Plants in the subgenus Cadalvena (C. cuspidatus and C. subsessilus) were in a group close to Dimerocostus and Monocostus. This early work was published in 2001 in Volume 21, No. 3 of the publication, Molecular Phylogenetics and Evolution - A Molecular Phylogeny of Costaceae (Zingiberales).

She continued her study of the molecular makeup of plants in Costaceae, with genetic sampling and cladistic analysis of the following sample species (including muliple forms of some species): Costus afer, C. allenii, C. amazonicus, C. arabicus, C. claviger, C. comosus, C. curcumoides, C. curvibracteatus, C. cuspidatus, C. deistelii, C. dinklagei, C. dubius, C. englerianus, C. erythrocoryne, C. erythrophyllus, C. fissiligulatus, C. gabonensis, C. globosus, C. aff. globosus, C. aff. globosus, C. guanaiensis var. guanaiensis, C. guanaiensis var. tramicus, C. lacerus, C. laevis, C. lanceolatus, C. lasius, C. lateriflorus, C. letestui, C. lucanusianus, C. aff. lucanusianus (yellow), C. malortieanus, C. maculatus, C. megalobractea, C. montanus, C. mosaicus, C. paradoxus, C. phaeotrichus, C. pictus, C. picus, C. "el wiskey", C. pulverulentus, C. pulverulentus, C. sp. nov., C. sp., C. sp. Colombia, C. speciosus, C. speciosus, C. spectabilis, C. stenophyllus, C. subsessilis, C. talbotii, C. tappenbeckianus, C. varzearum, C. villosissimus, C. woodsonii, Dimerocostus argenteus, D. strobilaceus subsp. gutierrezii, D. strobilaceus subsp. strobilaceus, Monocostus uniflorus, Tapeinochilos ananasse, T. dahlii, T. solomonensis, T. hollrungii, T. pubescens, T. queenslandiae - and of Siphonchilus decorus and Siphonochilus kirkii to establish a DNA base from the genus that has been determined to be basal to Zingiberaceae.

The following graphic shows how the groupings from her cladistic analysis came out:

Cladistic Analysis of Costaceae

Dr. Specht's study of Costaceae also included investigation of the evolutionary development of this plant family, and a determination of how the plants' characteristics evolved either separately or as a continuum. There were 95 separate morphological characters identified and analyzed. One of her primary conclusions was that the flower shape is NOT determinative of the plants' relationships - the tubular type (hummingbird pollination) actually evolved several times along different genetic lines, whereas the more open bee-pollinated structure evolved only once as an earlier form in these plants' evolution.

After much scientific review, the results of these studies were published by the American Society of Plant Taxonomists in the January-March 2006 issue (Volume 30, Number 1) of the prestigeous publication, Systematic Botany - entitled "Systematics and Evolution of the Tropical Monocot Family Costaceae (Zingiberales): A Multiple Dataset Approach". The reclassification of Costaceae genera was published by TAXON, Volume 55 (1) in February 2006. Chelsea has worked long and hard on this, and is continuing to conduct research on Costaceae and other members of the order Zingiberales at UC Berkeley.

You will find the reclassification has been adopted on this website and most file names of image files and notations within the text have been updated with the new names. As of 2013, seven years after this publication, I am happy to say that the new names have generally been accepted and adopted in the horticultural trade as well.

Copyright © 1999-2015 Dave Skinner