Royal Botanic Gardens 
Kew, Richmond, Surrey

Introduction

During the past thirty years most new literature on the systematics of the N. Chilean genus Copiapoa, amounting to over 150 printed pages, has resulted from the activities of three men: P. C. Hutchison (1953), F. Ritter (1959, 1960, 1961, 1963, 1980) and C. Backeberg (1959, 1962).

Hutchison (l.c.) wrote detailed, well illustrated accounts of two species he had studied in the field between 1952 - 53. His work should be consulted by anyone desiring a basic knowledge of the ecology and gross morphology of Copiapoa. Hutchison’s discussion of geographical and general variability in C. cinerea gave clear guidelines to would-be describers of new Copiapoa species. However, his very rational views were not shared by the two other authors mentioned above.

Ritter, after spending many fruitful years in the field, has now described no less than 36 new species. His most recent work, cited above, contains useful descriptive and distributional information, but far too many ill-defined taxa. However, he does provide an arrangement of species into informal sections which, by and large, seems to make sense. I have followed his sectional groupings in the key to the distribution map (p. 60), but his book should be consulted for details of their circumscription.

Backeberg has done little to improve our understanding of Copiapoa. Volume 3 of ‘Die Cactaceae’ is particularly misleading, containing numerous errors of identification and various improper and unnecessary names. Ritter (1980) has already pointed out many of the faults in Backeberg’s writings on the genus, which are best consulted with due caution.

The present account is primarily intended to list the taxa that have been named to date, and summarize their most important data and references; the reader is referred to Ritter (1980) for their descriptions. An attempt has been made to group together the most narrowly defined of Ritter’s microspecies and indicate their probable synonymy. Only in the case of C. cinerea do we have a slightly better understanding of geographic and ecological variation, which permits the establishment of five varieties in this species (one requiring a new combination: C. cinerea var. haseltoniana). Otherwise considerable further field study - by someone with a synthetic approach to the problem - is necessary if a key is to be constructed and proper species-limits established.

The typification of old Copiapoa epithets

About 15 binomials were published prior to 1890 which have since been considered, by various authors, to belong in Copiapoa. Excepting the three species described by Philippi in 1860, it seems likely that most, if not all, of the names that can still be identified, were based on plants collected by the Englishman Thomas Bridges (1807 - 1865), who spent a considerable part of the period 1828 to c. 1851 working in Chile (cf. Pfeiffer, 1847; Johnston, 1928). Bridges made a number of plant-collecting trips in different parts of Chile, but a study of his itinerary suggests that only one of these, in the latter half of 1841, concerns us here.

The extent of our knowledge of Bridges’s travels between July and the end of 1841 has been recorded by Johnston (l.c.), who lists the following relevant localities in their probable chronological order: Port of Copiapó (Puerto Viejo), Copiapó, Chañarcillo, Totoral, Los Pozos , Vallenar, Freirina, Huasco (all in Prov. Atacama); Coquimbo, Valle Elqui, Vicuña, etc. (Prov. Coquimbo). Early the following year Bridges intended to return to Valparaiso, from where presumably he would have despatched any cacti he had found to interested persons in Europe: probably to Salm-Dyck, Schelhase, Pfeiffer and W. J. Hooker. Pfeiffer’s reference (l.c.) to Valparaiso as the locality of his E. columnaris is perhaps another example of a not infrequent mistake made by early describers of cacti: that of giving the port or town of despatch, rather than the actual place of wild origin.

Bridges’s only other recorded excursions to N. Chile seem both to have been in 1844, when he twice landed at the little port of Cobija, located between Antofagasta and Tocopilla, in Prov. Antofagasta (then part of Bolivia), before making his way inland. Only the recently discovered Copiapoa tocopillana is known from this part of Chile. Therefore, on the evidence available we should try to apply names such as Echinocactus bolivianus, E. bridgesii, E. columnaris (all of Pfeiffer), E. echinoides, E. malletianus (both Salm-Dyck) and E. streptocaulon Hook., to plants that grow in the Provinces of Atacama and Coquimbo, and not to those from Prov. Antofagasta as Britton & Rose, Backeberg, Lembcke and Ritter have done. Pfeiffer’s references to Bolivia are doubtless also in error, though whether Bridges or he is to blame we will probably never know.

A decade after Bridges, R. A. Philippi made a long journey through the Atacama Desert during 1853 - 54. Subsequently, he based three names on plants from Prov. Antofagasta which have since been taken up in Copiapoa. Two of these comprise the well-known C. cinerea and C. humilis (cf. Hutchison, 1953); the third is discussed below.

Of the older names as defined here, the following have been used recently and merit individual mention; none are known to have extant holotypes:

C.  marginata; Echinocactus marginatus Salm-Dyck (1845). The original description on its own is insufficient to permit reliable identification of this name, but its application can be clarified so long as E. columnaris Pfeiffer (1847), illustrated in Pfeiffer (1850), is considered to be the same. Its rediscovery by Ritter in 1956, at Morro Copiapo, SW. of Caldera (Prov. Atacama) may be within the area traversed by Bridges in 1841. In the past C. marginata has been used in a different sense by Britton & Rose and Ritter (see Checklist). C. streptocaulon based on E. streptocaulon Hook. (1851) clearly must be a synonym of C. marginata as applied here.

C.  echinoides; Echinocactus echinoides Salm-Dyck (1845). Ritter (1980) has rejected this name, and Britton & Rose (1922) remark ‘we know the plant only from descriptions and illustrations’. However, the fine illustration in Pfeiffer (1850) and our knowledge of Bridges’s activities leads me to suggest an identity for E. echinoides. One of the localities visited by Bridges was Totoral, and Ritter has described C. dura from east of this tiny settlement, which matches the descriptions of Salm-Dyck and Pfeiffer, and the latter’s illustration, fairly well. Backeberg (1959), who was followed by Lembcke, has applied C. echinoides to a plant growing a little to the south of the town of Antofagasta, but, as already discussed, we have no evidence to suggest that Bridges visited this area.

C.  boliviana; Echinocactus bolivianus Pfeiffer (1847). Known only from its original brief description, and in any case reduced to synonymy under E. echinoides in 1850 by Pfeiffer himself. However, Ritter, rather unwisely, has resurrected this name and used it for the plant from near Antofagasta mentioned above. A name of no value.

C.  bridgesii; Echinocactus bridgesii Pfeiffer (1847), with tab. Ritter has applied this name to a very distinctive plant found near the northern limit of the region traversed by Bridges. However, while the remarkable cephalium-like development of wool at the stem apex, mentioned by Pfeiffer, is matched in the Ritter plant, it is otherwise rather tall-growing, and seldom globose or conical as Pfeiffer’s description and figure requires. For the time being it seems best to allow this questionable identification to stand, since there is no other name currently available for what is an unmistakable and beautiful species.

C. cinerascens; Echinocactus cinerascens Salm-Dyck (1845); ?E. copiapensis Pfeiffer (1847). Of the old names accepted here E. cinerascens is the most uncertain as to type, since we have only Salm-Dyck’s description and cited locality of ‘Copiapo’ to help in its typification. Ritter (1980) uses it for a plant growing north of Chañaral, which is somewhat north of the region we know Bridges to have visited, though within Dept. Copiapo. Despite some misgivings, I am following Ritter’s identification because there is no real discrepancy between Salm-Dyck’s description and Ritter’s plant, nor is there an alternative name for it (save for the equally poorly typified C. applanata Backeb.).

C. malletiana; Echinocactus malletianus Salm-Dyck (1845). Although there is no illustration to clarify the application of this name, the original description could well relate to one of the varieties of C. cinerea, such as var. dealbata which grows within the area traversed by Bridges. E. malletiana is best treated as an inadequately typified name and ignored, otherwise the well-known C. cinerea (E. cinereus Philippi, 1860) will be at risk of being lost in its synonymy.

E. conglomeratus Philippi (1860) has been applied to the distinctive C. solaris (Ritter, 1961) by Lembcke in KuaS 17: 29 - 30 (1966) and in Aloe 17: 13(1979). Philippi’s type locality is precise, and correct for C. solaris, while his epithet is also very appropriate. However, in the absence of a type specimen or illustration, only the original description can typify the name, and here it is clear that Philippi must have confused his materials, for the description bears no relation to the plant at his locality. Lembcke’s use of C. conglomerata is thus illegitimate, since his circumscription effectively excludes Philippi’s type, as determined by the latter’s original description. E. conglomerata as to description cannot be positively identified, at least while the holotype remains unknown.

C. cupreata; Echinocactus cupreatus Poselger ex Hildmann (1885); and 
C. coquimbana; E. coquimbana Karw. ex Ruempler (1885). 
Both known only from very brief descriptions. The former suggests a form of C. fiedlerana (E. fiedleranus Schumann, 1903) collected by Knize (no. 21, as ‘C. cuprea’), but is too poorly typified to upset the younger Schumann epithet, while the latter could well apply to either of two plants recorded by Ritter in the vicinity of Coquimbo and the Rio Elqui. Both names are best abandoned.

C. pepiniana Backeb. (1935); Backeb. (1959). Backeberg’s first publication of this name was a combination for Echinocactus pepinianus Lemaire ex Foerster (1846) which, as a bare name, must be considered a combination for Cereus pepinianus Lemaire ex Salm-Dyck in Allg. Gartenz. 13:354 (1845) (?E. pepinianus Lemaire, Cat. Cels. 1845, nom. nud.). Britton & Rose (Cact. 2: 137. 1920) refer Salm-Dyck’s plant to the synonymy of Trichocereus chiloensis (Colla) B. & R., but it is, however, a very doubtful name. In 1959 Backeberg published 'Copiapoa pepiniana’ again, this time basing it on Echinocactus pepinianus sensu Schumann (Gesamtb. Kakt. 420. 1898) while stating ‘non Lemaire’ [ex Foerster]. Schumann’s plant might be a true Copiapoa (however, cf. Ritter, 1980, p. 1106 - 7), but more important is that C. pepiniana Backeb. (1959) cannot be used, since it is a homonym of C. pepiniana Backeb. (1935), being based on a different type as indicated by Backeberg, and therefore illegitimate.

Acknowledgements

I would like to thank Herr J. D. Supthut, Director of the Staedtische Sukkulenten-Sammlung, Zürich and Herr U. Eggli for their help and encouragement while I studied at the above establishment during February and March this year. I am also very grateful to Herr Mächler (Switzerland) and Mr. G. Charles (UK) for allowing me to examine their fine collections of Copiapoa. The last named has prepared the notes on the cultivation of the genus appended to this article, and together with Herr A. Wirth (Zürich), has kindly supplied the illustrations which accompany it.

Bibliography

Checklist     Cultivation     Map