This page contains photos and description of a species, form or cultivar of Costaceae.

Gingersrus Database Taxon ID 3225

Costus spicatus

OLD NAME: Costus spicatus

NEW NAME: Costus spicatus


FULL SCIENTIFIC NAME: Costus spicatus (Jacq.) Sw.

STATUS : accepted

CONTINENT: Neotropical

FIELD OBSERVATIONS:(If field observations are available, you can click on the link to open in a new window.)

PHOTOS:(If photos are available, you can click on the link to open in a new window.)

- Alpinia spicata Jacq. (1760) - Costus cylindricus Jacq. (1809) - Amomum petiolatum Lam. (1783) - Costus conicus Stokes (1812) - Costus micranthus L.F.Gagnep. (1903) - Costus arabicus auct. non L.: Aubl. (1775)

Cosus spicatus is an accepted species of Costus first described in 1760 as Alpinia spicata by Nicolai Josephi Jacquin in his Selectarum stirpium Americanarum published in 1763. Dr. Maas has selected as the type an illustration in Table 1 of that publication (photo top left). It was based on his collection in Martinique that was later destroyed in the Vienna Herbarium in 1943.

Costus spicatus is found only in the northern Windward Islands, southern Leeward Islands and Dominican Republic. It is not known from mainland Central or South America. Records of this species in Mexico are probably in error. In 2017 I went to the island of Dominica where I made several observations of this species, as shown at the Inaturalist link above. It flowers in the rainy season.

It is described as a plant growing to 1-3 meters tall with a truncate ligule 2-10 mm long. It forms a terminal inflorescence comprised of greenish-red non-appendaged bracts (red in the covered parts, green near the apex) and tubular flowers with yellow to pink corolla and yellow labellum. The middle lobe of the labellum is reflexed with its margin irreugularly lobulate to crenulate, distinguishing it from several similar looking species.

Paul Maas in his 1972 monograph stated, "The name C. spicatus has often been misapplied to specimens belonging to C. scaber, C. laevis, C. pictus and C. spiralis. It is closely related to C. scaber and C. spiralis but can be distinguished by its greenish bracts." Maas compares those three species in a chart, reproduced in a table on a separate page at .

A preliminary phylogeny was completed by Eugenio Valderrama and his associates in the Chelsea Specht Lab at Cornell University and was published in the journal Frontiers in Plant Science in September 2022. Using a plant grown at the Cornell greenhouse that originated with the same Smithsonian collection as my R3028, the species was in a clade close to Costus woodsonii.

In an earlier, preliminary phylogeny, DNA was extracted from two samples of this species: the John Kress collection from Dominica (same as my R3028) and another collection I made, also from Dominica. Strangely, these produced two widely separated clades in a partial phylogeny tree. The Kress collection (listed there with my number R3028 but received from a different source) was shown in a clade with a sample of Costus woodsonii from Costa Rica. My collection R3385 showed up in a separate lineage close to Costus scaber and Costus vargasii. It's whole genome was sequenced. Eugenio Valderrama sent me the following explanation:

The C. spicatus R3385 behaved like a potential hybrid in some of the analysis Jacob made with the accessions for which we have whole genome data, that were obtained with a different sequencing approach. I am inclined to keep only R3028 until we can check if there are more reasons to think that some introgression could have happened in the R3385 population.

I also have some doubts about the sampling done at Cornell for Costus spicatus. They listed my accession number (R3038) which was my collection from the Smithsonian that originated with a John Kress collection in Dominica. But in discussions with their lab I learned that they actually took a sample from an untagged, unaccessioned plant that was growing in the Cornell greenhouse along with a few other Costus species. How they knew to take the sample from the right plant, I do not know. Chelsea Specht has assured me that the sample used was from the same Smithsonian source. I am hoping that a comparison can be done with the real R3028 to confirm this, because the Cornell sample was used for the whole genome sequencing.

Based on information at the Botanic Gardens Conservation International (BGCI) this species is in ex situ cultivation at 20 botanical gardens world wide. Some of these records may be for Costus scaberwhich is sometimes confused with C. spicatus. I have found this species to be relatively easy to grow in my garden and greenhouses at Le Jardín Ombrage.

ACCESSIONS:Click links (if any) to see details of individual collections. R3028- R3385-

Copyright © 1999-2023 - Dave Skinner,
All text and images on this page are copyrighted. They may be downloaded and/or printed for personal use and scientific research, but may not be published in paper or electronic media without my express consent. They absolutely may not be used commercially in any way whatsoever without my express consent. To inquire about permissions, contact

Photos (if available) of Taxon ID 3225
Click on a thumbnail to load full sized image.
Use your "back" button to come back to this page.

Costus spicatus type specimen - Click to see full sized image
Photo# 530 Accession# R0
Costus spicatus type specimen

Costus spicatus, photo from Skinner R3385,  Dominica - Click to see full sized image
Photo# 307 Accession# R3385
Costus spicatus, photo from Skinner R3385, Dominica