Selenicereus plumieri (Rol.-Goss.) Hoxey & Gdaniec, 2021

Hoxey, Paul & Gdaniec, Andrew, 2021, An evaluation of the Cactaceae of Barbados, Grenada, Saint Vincent, and The Grenadines, Phytotaxa 483 (1), pp. 25-74 : 58-68

9. Selenicereus plumieri (Rol.-Goss.) Hoxey & Gdaniec comb. nov. ≡ Cereus plumieri Roland-Gosselin (1908: 668) ≡ Hylocereus plumieri (Rol.-Goss.) Lourteig (1991: 406).

Ind. Loc.:— The Grenadine islands chiefly on that island called Bequia.

Lectotype (designated by Korotkova et al. 2017: 29):—[icon] Cactus repens Tab. CXCIX, fig. 2 in Burman (1755 –1760) ( Fig. 39 ).

Description:—Plants terrestrial on rocks or epiphytic on trees. Stems long (> 1 m) branching, 2–3 cm wide, with aerial roots, scrambling over rocks or supported by other vegetation, sometime hanging when not supported. Ribs 3 (occasionally 4), crenate, areoles in an elevated position. Areoles round, 3 mm diameter, white felt, spacing 2 cm. Spines 4, dark brown, spreading, thicker at the base to 10 mm long. Flowers unknown. Fruit, nearly globular 75 mm long by 65 mm wide, smooth, b urgundy red with greenish speckling and greenish lines below the scales. Small scales 5–10 mm long, 4–5 mm wide at the base, 25 mm spacing. Greenish colour, smaller and more closely packed towards the base of the fruit. Larger and yellows-pink towards the upper part of the fruit. Flower remains not persistent. Spines 4–7, light brown with darker tip, 1 cm long, thin and straight. Rind 3 mm thick enclosing pure white pulp with many black shiny seeds.

Distribution ( Fig. 2H ):— Selenicereus plumieri is only known from the islands of Saint Vincent and Bequia with certainly. We tentatively include here plants from Grenada but lack knowledge of fertile material (herbarium and living material). We also include here plants found on Barbados based on herbarium material. The species is therefore endemic to the study area.

Discussion:— Plumier (1689 –1697) reported three similar-looking plants which today we include in the genus Selenicereus following Korotkova et al. (2017) but note that these Plumier plants were previously placed in the genus Hylocereus ( Hunt et al 2006). However, Korotkova et al. (2017), which we follow here, has shown that this genus fits within Selenicereus .

The first of Plumier’s (1689 –1697) species was named as “ Melocactus repens trigonus, flore albo fructu coccineo” which was illustrated (vol. 3 t. 14) using a plant from the island of St. Croix (in the US Virgin Islands today). This illustration has been designated by Howard (1989: 404) as the lectotype of the name Cereus trigonus Haworth (1812: 181) (≡ Selenicereus trigonus ). The illustration shows a 3 angled stem with 8 fairly short spines per areole, as well as two fruits with large scales, each scale about one third the length of the fruit. Korotkova et al. (2017: 29) consider Cereus trigonus as a good species although Bauer (2003) previously proposed it should be a synonym of Selenicereus triangularis ( Linnaeus 1753: 468) Hunt (2017: 35) .

A second species which Plumier (1689 –1697) describes and illustrates (vol. 3 t. 16) was named “ Melocactus alius trigonus repens fructu e violaceo coccineo”. This illustration was copied and reproduced in Burman (1755 –1760: pl. 199. fig. 2). This later illustration was selected as the lectotype of Cereus plumieri by Korotkova et al. (2017: 29). Korotkova et al. (2017: 29) accepted Cereus plumieri as synonym of Selenicereus trigonus . The illustration shows a 3-angled stem similar to S. trigonus but with 4 fairly short spines per areole which resembles S. trigonus . However, the fruit is significantly different with much smaller scales and with about 6 thin spines emerging from behind the scales. The location for this plant is stated by Plumier (1689 –1697) as “the Grenadine islands chiefly on that island called Bequia”. Plumier (1689 –1697) clearly considered S. trigonus and S. plumieri as distinct species, but Plumier’s illustration has been questioned with regard to the spines on the fruit ( Mottram 2002: 89). Remarkably for a plant first recorded over 300 years ago, we can find no references to its re-discovery in habitat or observations based on contemporary material. We explored the island of Bequia and encountered S. plumieri at two localities. The first one was in the southern end of the island, where we found specimens scrambling over rocks and growing supported by the branches of small trees. Unfortunately, neither fruits nor flowers were observed, but the 3-angled stems with 4 spines per areole exactly match Plumier’s plate (see Figs. 40–41 ). The second population was found near the northern end of the island at the top of coastal cliffs and extending a little inside the adjacent forest, predominately scrambling through the branches of trees but with a few terrestrial plants growing on rocks. Here we found a ripe fruit precisely matching the Plumier’s illustration, with spines in the axils of the small scales ( Figs. 42–43 ). We consider the fruit sufficiently distinct from both S. triangularis and S. trigonus to accept Plumier’s taxon as a good species. Further studies on Caribbean Selenicereus are however required to better understand the taxonomy of the various species.

A population found at Saint Vincent (near the airport, on rocks or supported in trees) shows morphology of stems and spines which matches those of the Bequia plants. However, characters regarding fruit (brighter pink colour, spineless, with more elongated shape and thicker rind) are clearly different. Note that the plants were at an advanced stage of fruiting and it is possible that these features relate to this. On the contrary, Saint Vincent population could be a different taxon. Further investigations are necessary and, presently, we consider this plant to belong to S. plumieri .

Along the north coast of Saint Vincent (near the village of Fancy), we visited the type locality of Selenicereus innesii . Adjacent to the road, which is in turn close to the coast, we found some very thin-stemmed plants with weak spination growing on rocky cliffs. The plants were growing in poor light under quite dense tree cover. The stems were often hanging vertically down the rock faces ( Fig. 44 ). The stems had three or four sides but were so pumped up with water, ribs were not present and the stems had a triangular or rectangular cross section ( Fig. 45 ). They also had a profusion of aerial roots growing from the stems. We believe that these plants are referable to H. plumieri as seen elsewhere on Saint Vincent and Bequia, but growing in low light conditions. We speculate that these are the plants Innes collected, but he did not appreciate that they are the same as the Selenicereus found elsewhere on the island. In cultivation at Holly Gate, Innes’s nursery in west Sussex, UK, the material probably grew to look more like the typical plants, but meanwhile the label was mixed up with another thin-stemmed plant which was then used as the type of S. innesii . Kimnach (2011) relays the story that in April 2011, following his participation in the International Organization for Succulent Plant Study (IOS) Congress at Monaco, he had visited Holly Gate nursery after Innes died and recognized an old unlabelled plant as what he had described as Selenicereus innesii . The proprietor at that time, Terry Hewitt, found a duplicate plant that did have a label; it read ‘ Aporocactus × Weberocereus’. This is very likely to be the origin of the material described as S. innesii . This should not be considered a good species, but a human-made hybrid of cultivated origin.

We failed to find Selenicereus in the Grenadines other than on Bequia. However, Selenicereus is present on Grenada on the basis of a specimen deposited at the Herbarium A (barcode A01875293) collected at Grenada in March 28, 1977 by R. A. Howard (collection number 18291). This specimen consists of a single sterile stem with three ribs and adventitious roots. The spines, 2–4 per areole, are acicular, without thickening at the base and approximately 5 mm long. Interestingly this stem has small podaria or leaflets below the areoles and this may indicate that the stem is very fresh and possibly displaying juvenile characters. Nevertheless, these were not observed on any of the material from Bequia and Saint Vincent, including young stems. We encountered Selenicereus in one locality in Grenada, north of St. George’s growing on an inaccessible cliff with Cereus repandus . Unfortunately, we could not approach closely and the material was sterile ( Fig. 46 ). We are tentatively placing this taxon under Selenicereus plumieri but an examination of fertile material, especially the fruit, is required to confirm this identification.

Hughes (1750: 186) describes a plant from Barbados he calls “ Cereus scandens minor” which is important to note. The fruit is described as oval shaped as large as a turkey egg, a dark-purple colour, with scales of a triangular shape (size not indicated) and the rind studded with very small sharp spines. This a very good likeness to the fruit seen at the type locality on Bequia. We did not find this species on Barbados but a search should be undertaken to try and determine if the species is still to be found there. Meanwhile, based on the herbarium material and documentary evidence available, we consider Selenicereus plumieri a native species of Barbados.

Gooding et al. (1965: 302) considers Hylocereus trigonus native to Barbados, with a description of 4–8 spines that are, each one, 4–7 mm long and he noted a specimen Murphy s.n. (UCWI 5533) which unfortunately we have not been able to see. However, a specimen at NY (barcode NY 679821) collected in 1924 by I. Robinson, morphologically matches the Gooding et al.’s description having 4–8 spines up to 10 mm long. The sheet is annotated as “ Hylocereus trigonus or H. napoleonis (note spines on fruit) Det. by Myron Kimnach Date: 1985”. Unfortunately, the fruit is not visible in the online image (see http://sweetgum.nybg.org/images3/3332/256/00679821.jpg) but Kimnach’s comment is especially significant as the spiny fruit strongly suggests this is Selenicereus plumieri and not S. trigonus . The stems and spination are also consistent with S. plumieri .

Bauer (2003: 42) designated an interpretative epitype for Cereus plumieri (≡ Selenicereus plumieri ) which is based on material gathered in Barbados ( US 1367210). In light of material now being found on Bequia that matches the type illustration, this epitype is scarcely necessary. In our opinion, now we accept this species as distinct from S. trigonus the lectotype illustration is an accurate representation of the species.

Specimens examined:— BARBADOS. 1924, I. Robinson s.n. (NY) . GRENADA. 28 March 1977, Howard 18291 (A) .

Figures

FIGURE 2. Distribution of cacti on Barbados, Grenada, Saint Vincent and the Grenadine islands based on observations by the authors. A. Acanthocereus tetragonus. B. Cereus repandus. C. Mammillaria mammillaris. D. Melocactus broadwayi. E. Melocactus intortus. F. Opuntia dillenii. G. Pilosocereus curtisii H. Selenicereus plumieri I. Opuntia dillenii (Barbados). J. Pilosocereus barbadensis (Barbados).

FIGURE 39. Reproduction of the lectotype illustration of Cereus plumieri (≡Selenicereus plumieri) (Burman 1755–1760: pl. 199. fig. 2). Fig. 2 is the stem with a single fruit on the left-hand side.

FIGURE 40. Stems of Selenicereus plumieri (Hoxey 1661.02) supported in trees, southern Bequia.

FIGURE 41. Stems of Selenicereus plumieri (Hoxey 1661.02), southern Bequia.

FIGURE 42. Selenicereus plumieri (Hoxey 1664.01) in fruit, northern Bequia.

FIGURE 43. Fruit of Selenicereus plumieri (Hoxey 1664.01), northern Bequia.

FIGURE 44. Selenicereus plumieri (Hoxey 1656.01), north coast of Saint Vincent (type locality of Selenicereus innesii).

FIGURE 45. Thin stems of Selenicereus plumieri (Hoxey 1656.01) with many aerial roots, north coast of Saint Vincent.

FIGURE 46. Thin stems of the Selenicereus found on Grenada, tentatively referred to S. plumieri here but without knowledge of the fruits (Hoxey 1698.03).