Taxon 3: 231-232. 1954.
The type of Amaryllis bella-donna L.
J. E. Dandy (London) and F. R. Fosberg (Falls Church, Va.)

There has been complete disagreement among current interpretations of the typification of the genus Amaryllis L., beginning with Uphof's attempt to typify it in 1938 (3) and culminating in articles in Taxon by Dyer (8) and Traub (9). This led to the calling of a special meeting of those interested in the question on June 30, 1954, by the International Bureau of Plant Taxonomy in connection with the Nomenclature Section of the Paris Congress. This was attended by a small number of botanists and the matter was discussed in detail. It was decided that, after examination of the material in the Hortus Siccus Cliffortianus, the present authors should prepare a succinct statement of the case for the publication in Taxon.

It may be stated at the outset that the facts in the case seem fully in accord with the careful and complete discussion by Sealy (4). The reason for a further article is to re-emphasize the fact, unknown at first to Uphof and later disregarded by him and by Traub and Moldenke, that there is an actual specimen preserved at the British Museum (Natural History) in the Hortus Siccus Cliffortianus of the plant upon which Linnaeus must have based his definition of Amaryllis bella-donna; "Amaryllis spatha multiflora, corollis campanulatis aequalibus, genitalibus declinatus". This specimen unquestionably represents the South African plant commonly known as the Cape Belladonna and agrees perfectly with Linnaeus's definition of A. bella-donna.

It may be pointed out that Linnaeus's Hortus Cliffortianus is a work actually describing the plants growing in Clifford's garden and which when preserved are in Clifford's Herbarium (Hortus Siccus Cliffortianus), now in the British Museum, and that in this treatment the synonymy must be regarded as secondary. It is also clear that the Amryllis bella-donna of Species Plantarum must be typified by the Hortus Cliffortianus reference, since the definition of the species given in the Species Plantarum is taken verbatim from Hortus Cliffortianus.

Therefore, although there is no annotation by Linnaeus on the sheet in the Hortus Siccus Cliffortianus (BM) — such annotations are unusual in this herbarium — it seems clear enough that since a specimen exists there it must be regarded as the lectotype of the species. The dismissal of this as circumstantial evidence by Uphof, Traub and Moldenke would indicate that they are abandoning the type method for species described before types were designated as such by the authors of names. We cannot subscribe to this.

Their consistent refusal to consider this specimen after its existence had been pointed out by Sealy makes it look as though these authors were merely indulging in the natural, though scarcely scientific, human tendency to cling to an untenable position merely because it had been adopted. Clarity and ultimate stability will certainly be better served by due consideration of all new evidence, regardless of previously held opinions.

Readers are referred to Sealy's article for an extend and fully adequate discussion of the whole matter. The following conclusions seem inescapable:

1. The overlooking or ignoring of the existence and identity of this specimen by Uphof, Traub and Moldenke led them to typifying Amaryllis bella-donna to rely completely on the synonymy given by Linnaeus. This synonymy admittedly applies to the American plant known as Hippeastrum puniceum (Lam.) Urb., resulting in their application of the name A. bella-donna to the American plant.

2. The existence of an actual specimen in the Hortus Siccus Cliffortianus which must be regarded as the type makes Uphof's and Traub's entire lengthy discussions of and choice of a lectotype for this species unnecessary and irrelevant.

3. The existence of this specimen upon which the Hortus Cliffortianus definition must have been based and the fact that this is the African plant make it necessary to uphold the traditional application of the name Amaryllis bella-donna to the Cape Belladonna.

4. The fact that everyone agrees the Amaryllis L. must be typified by A. bella-donna L. makes it follow that the generic name must be applied to the African and not to the American genus.

Pertinent literature:

1. Linnaeus, C., Hortus Cliffortianus. Amstelasdami 1737 (p. 135).
2. Linnaeus, C., Species Plantarum. Holmiae 1753 (p. 293).
3. Uphof, J. C. T., The History of Nomenclature — Amaryllis (Linn.) Herb., and Hippeastrum (Herb.). Herbertia 5:100-109. 1938.
4. Sealy, J. R. Amaryllis and Hippeastrum. Bull. Misc. Inf., Kew, 1939:49-68. 1939.
5. Uphof, J. C. T., Critical Review of Sealy's "Amaryllis and Hippeastrum", Herbertia 6:163-166. 1940.
6. Uphof, J. C. T., Linnaeus confirms Dr. Hill's Identification of Amaryllis belladonna Linn. Herbertia 13:97-98. 1948.
7. Traub, H. P. and Moldenke, H. N., Amaryllidaceae: Trib Amarylleae, Stanford, Calif. 1949.
8. Dyer, R. A., The Cape Belladonna Lily. Taxon 3:72-74. 1954.
9. Traub, H. P., Typification of Amaryllis belladonna L., Taxon 3:102-111. 1954.

The blank specimen in the Clifford Herbarium agrees with the vague description in Hort. Cliff., but not with Mantissa Plantarum (1767). Linnaeus specifically mentioned the beard pads, which excludes the blank specimen and all other Cape Belladonnas.

The name Amaryllis belladonna was misapplied to the Cape plant long before the blank specimen became known, and only among botanists and gardeners associated with Kew. Starting with "Uphof's and Traub's entire lengthy discussions", I have added to the list of authors who acknowledged that Amaryllis belladonna was exclusively the American plant. The error originated at Kew, and can be traced through a series of errors and misapplications going back to Parkinson's Paradisi (1629). It continued, with interesting branches, through Rea, Morison, Plukenet, Ehret, P Miller, J Miller, Linn. fil. (while at Kew), L'Héritier, Hort. Kew. (1789), Candolle, Dryander (1804).

Ultimately, the confusion can be traced to Ferrari's Flora. Ferrari gave two or three names for some species, but was not entirely clear as to which name applied to which plant. He described Aldinus's (Hort. Farn.) Lilionarcissus rubeus indicus, which Plukenet thought was the same as Narcissus indicus liliaceus diluto colore purpurascens. Ehret accepted Plukenet's synonymy, but replaced Aldinus's phrase-name with Linnaeus's from Hort. Cliff. Subsequent writers assumed that Aldinus's plant was another Cape Belladonna, presumably because it was growing in Italy.

The blank specimen is remarkably similar to Amaryllis belladonna var. rubra, a garden variety bred by Charles Truffaut in the mid-19th century, presumably descended from a Brunsdonna. Ravenna (2003) had a piece of stem from the blank specimen C14 dated. The date is consistent with the plant introduced by Truffaut.

It is beyond absurd to take a 19th century garden variety derived from an intergeneric hybrid as a lectotype for a species described in the 18th century.