Catalogue of Brazilian Plants

A transcription of George Gardner’s manuscript

Click here for full transcript (PDF)

Introduction and explanation

Page 1 of Gardner’s manuscript copy of his Catalogue of Brazilian plants. Note: In binding the numbers on the leaf hand side of the page have largely been obliterated. Later pages are somewhat less complicated to read and transcribe!George Gardner’s Brazilian collections were made between July 1836, when he arrived in Rio de Janeiro, and April 1841, when he started his return to England (finally leaving Brazil in June). They number a little over 6,000 numbers. At the prompting of George Bentham, Gardner summarized his collections in the form of the Catalogue of Brazilian Plants: the unique manuscript copy of which is in the possession of the Archives collections of the Herbarium, Library, Art & Archives, Royal Botanic Gardens, Kew.

Gardner collected in fairly large sets of duplicates, with many numbers having 20 or more. By his own estimate, there were some 3,000 species and 60,000 specimens. Because of the purchase of Gardner’s working collection from his estate, the largest set is most probably in BM. Stafleu & Cowan (1976) stated that, under the entry for Henry Barron Fielding, there was ‘a very rich second set at OXF ... Henry Fielding subscribed in advance to his [Gardner’s] Brazilian collections and received the most complete set after that of BM.’; this is incorrect and appears to have ignored completely Sir William J Hooker’s set.

The ‘second set’ was in the possession of W. J. Hooker, to whom Gardner had explicitly stated that Hooker could select what he wanted, even unicates. It is only in his first paper that he notes that for some plants ‘it is to be understood that the plant was not found in sufficient plenty to allow of the specimens being distributed’ (Gardner, 1842a). A few later collections were also apparently of unicates. Both Hooker and Bentham were responsible for dividing up the sets, and making them available for sale via Pamplin’s of Soho – at £2 per 100 specimens. Other than Hooker’s set, K also has Bentham’s set, with its simple, preprinted labels, and a very small number of duplicates from Trinity College given some thirty years ago. In BR there are several collections that were purchased/presented to C. F. P. von Martius, clearly marked in the Elenchus of Martius’ herbarium. Large numbers of duplicates were evidently distributed from BM to W, and thence also to other institutions, including NY. The number of collections in NY is large, and with many taxa seven or more duplicates are known: the result of many bequests since the mid-1800s. Duplicates also exist in E, GH, LE, P, R and US, and many other institutions, and several once existed in B.

Also worth pointing out is that Gardner added a footnote to one of his ‘Contributions towards a Flora of Brazil, ...’ (Gardnder, 1847) which should be borne in mind when taxa he described from Goiás are typified. His footnote reads: ‘When my former papers on these two tribes were prepared, I could not lay my hands on my Goyaz collections, the bundles containing them having been lost during my removal from England to Ceylon. I have, however, been fortunate enought to obtain, by purchase, at the sale of Professor Graham’s Herbarium, his set of my plants from that Province, which, although an early one, I find to be deficient in nearly one hundred species of Compositae alone.’ Gardner’s Goiás collections were numbered from 2,996 through to 4,397, and according to the Catalogue were divided into three Lots, of 548, 524 and 310 species respectively. Since this accounts for between 1,382 (the total of the number of species) and 1,401 numbers (some missing, some with number splits, in the Catalogue), a not insignificant part of his collection in the BM is missing. Since he had clearly worked on his own set, and that of Hooker, when he was in London, lectotypes for taxa in the ‘former papers’ should be very carefully selected.

Harley (2000, pers. comm.) has indicated that Sandwith had two copies of determinations of Gardner’s collections, one at K the other at OXF (where Sandwith worked for a significant amount of time). A list of determinations of the Gardner collections in the Natural History Museum (BM) and the Royal Botanic Gardens, Kew was compiled by Hiern and transcribed, possibly by Sandwith, although it is unsigned. A PDF transcribed version of the Hiern determinations list is available.

A separate, numbered, list of ferns was published as one of his earlier papers (Gardner, 1842b).

The transcription

The Catalogue has been fully transcribed by me and, having become reasonably familiar with his handwriting, there should be few errors – only a few words defeated me. Any errors that are present I apologize for.


The Catalogue has been rebound, and what was originally five registers, or sections, is bound in the same five parts in this volume. Gardner numbered each register and the position of the beginning of each of these is indicated in the text.

A small, final, sixth section was added by Gardner to include ‘Medicinal and Economic plants of the north of Brazil’. This was never published by Gardner, but we intend to publish a paper on the plants mentioned. The numbers in this final section refer to Gardner’s collections in the Catalogue; several species are unnumbered.

The Catalogue is laid out in three columns: number, taxon, locality. Occasional notes (common names, frequency, etc.) were largely placed against the locality and I have adopted this throughout, even though Gardner varied this in his earlier collections. The list is divided in several places, the breaks headed giving an indication of the broad location of the plants in the following part of the list, e.g. ‘Rio de Janeiro’, ‘Organ Mountains’, ‘Province of Piauhy’, etc. Any entries for which there is no information (e.g. family, genus, locality) I have inserted two em-rules ‘––’, rather than leaving the entry blank.

Later in the list there is a clear indication of the division of plants from the state of Goiás into three lots. From Gardner’s correspondence it is clear that these represent lots of plants that Gardner provided for sale to interested parties.

Medicinal and Economic Plants

At the end of the Catalogue, Gardner presents a list of the economic plants he encountered on his travels, together with some of their uses. These data are currently the subject of research by William Milliken and Nicholas Hind at Kew, the intention being to publish a paper based on updated identifications and known properties of the species concerned. A transcription of this section is available here.

Corrections and Additions

Gardner’s spelling of family, generic and species names is sometimes eratic. In most instances it is obvious to which taxa he was referring to, and the odd family names, which are quite variable in their spelling, are of little consequence. Some examples include:

  • Apocyniaceae = Apocynaceae
    Cinchoniaceae = Rubiaceae
    Gramineaceae = Gramineae
    Melastomaceae = Melastomataeae
    Orchideae = Orchidaceae

Some variations in the spelling of generic names have been highlighted by me with ‘[sic!]’ where the reader should be aware of a problem; I have not bothered substituting the correct names.

I have made no attempt to correct the authorship of any of the names in Gardner’s list; many are quite erroneous.

Gardner corrected and altered many of the entries himself and I have used a score-through method to show such changes; there are, however, other additions. These appear to be largely in Bentham’s handwriting and often appear as straight entries after Gardner’s. There is little difference in some of the handwriting so it is often very difficult to tell them apart; doubtless some have been wrongly attributed. In many instances they are against Leguminosae. I have placed all such additions in square brackets ‘[ ]’.

Other problems

The list does contain a number of problematical areas, not least duplicated numbering and ‘add-ins’ to the numbering system. Missing numbers are present, but relatively few; some large blocks are omitted, however. Gardner renumbered only a few series; the earlier error he noted in the body of the Catalogue and then corrected all of the affected numbers.

I have also noticed whilst curating material in the Compositae that there are some split numbers which are not accounted for in the Catalogue. Researchers with ‘xxxx bis’ might not be able to locate all material.


  • Gardner, G. (1842a). Contributions towards a Flora of Brazil, being an enumeration of a series of collections of plants, made in various parts of Brazil, from 1836 to 1841; with brief descriptions of the new genera and species. London J. Bot. 1: 158–193.
  • Gardner, G. (1842b). List of ferns from the neighbourhood of Rio de Janeiro, collected in 1836. London J. Bot. 1: 545–548.
  • Gardner, G. (1847). Contributions towards a Flora of Brazil, being the characters of several new species of Compositae, belonging to the tribes Vernoniaceae aad [sic!] Eupatorieaceae, from the Province of Goyaz; by George Gardner, Esq., F.L.S. Superintendent of the Royal Botanic Gardens, Ceylon. London J. Bot. 6: 417–449.
  • Stafleu, F. A. & Cowan, R. S. (1976). In: Taxonomic Literature: A selective guide to botanical publications and collections with dates, commentaries and types. Ed. 2. Volume 1: A–G. Reg. Veg. 94: 829–830 [Fielding]; 915–916 [Gardner].

D.J.N. Hind, Royal Botanic Gardens, Kew, August 2012