Herbarium specimens

The study of a large number of specimens gathered by the Humboldt and Bonpland expedition and entries in the “Journal Botanique” compiled by the two explorers have convinced us that Aimé Bonpland played the more significant role in the botanical work of the expedition and that Humboldt merely contributed a limited number of collections and field notes (see also Sarton, 1943). Consequently, we believe that the botanical specimens emanating from this expedition should be attributed to “Bonpland and Humboldt”, and not “Humboldt and Bonpland” as is commonly done in the botanical literature. The major sets of the collections gathered by the two explorers are thought to be deposited in a limited number of herbaria (see Lanjouw & Stafleu, 1954, 1957; Stafleu & Cowan, 1976, 1979; Hiepko, 2006) of which the two most important sets always have been considered to be the ones deposited in the Muséum National d'Histoire Naturelle in Paris (P-Bonpl.) and in the Botanischer Garten und Botanisches Museum Berlin-Dahlem (B-W). These two collections are also the only ones that continue to be maintained as separate, historical herbaria within their respective institutions. As P-Bonpl. and B-W contain the most comprehensive sets of the original group of specimens gathered by the two explorers we have focused our investigations on them. Additional searches were made in P, P-JU, B, and LR. An electronic search was conducted for HAL. Similarly, the microfiche of MA-CAV was scanned for relevant collections. Given the heterogeneous nature of these different sets of specimens, the specific methodology employed for our investigation of each one is described below.

The Bonpland herbarium (P-Bonpl.), Paris

P-Bonpl. (Fig. 2A, 2B) was studied over the course of five visits from 2002 to 2009. All of the monocotyledon specimens from the Humboldt and Bonpland expedition that were deposited in this herbarium were inventoried and each specimen was compared to the description of the corresponding species in the first volume of Humboldt & al. (1816–1825), which treats the monocotyledons, or to the supplementary treatment of monocotyledons published in the seventh volume of Humboldt & al. (1816–1825). Monocotyledons collected in Venezuela were identified according to one or more of the following criteria: 1) the collection number on the label could be attributed unequivocally to the Venezuelan portion of the journey when compared to collection numbers cited in the “Journal Botanique”; 2) the specimen label lacked a collection number, but clearly indicated a locality in Venezuela; or 3) the specimen label lacked all useful data (e.g., collection number, date of collection, etc.) but the only specimen cited and associated with the species described in Humboldt & al. (1816–1825) had a Venezuelan locality. Specimens that failed to meet any of these criteria were considered to have been collected elsewhere. Critical data on the original Bonpland labels or the labels copied by Kunth (see below) were recorded.

Fig. 2.

Herbarium of the Muséum National d'Histoire Naturelle, Paris. A. General view of the main building of the herbarium ; B. Representative specimens corresponding to the monocotyledon collection of P-Bonpl.

The general phanerogamic (P) and Jussieu (P-JU) herbaria, Paris

Humboldt and Bonpland split the large collection of specimens they gathered in the Americas shortly after they returned to Europe. The portion of the collection that Bonpland kept for himself has traditionally been called Bonpland's “private” herbarium and the exact number of specimens involved remains unknown. Bonpland kept this private collection with him when he returned to South America in 1816 (Sarton, 1943; Lourteig, 1977). However, before Bonpland's departure for South America Kunth had requested that Bonpland provide him with these specimens and the “Journal Botanique”, but as discussed by Lack (2003, 2004a) “The problem at Le Havre” very nearly brought the publication of the botany of the expedition to a halt. In November 1816, Kunth failed to reclaim the specimens Bonpland decided to take back to South America as these specimens were already stowed in the ship on which Bonpland was about to depart from the port of Le Havre, but Kunth did recover the field notes. Kunth's lack of access to Bonpland's “private” herbarium greatly reduced the number of specimens available to him for his challenging task of describing the Bonpland and Humboldt collections.

In 1832, Bonpland sent his “private” herbarium from Argentina to France and it was integrated into the general phanerogamic herbarium (P) in Paris the following year. According to Lack (2009), Bonpland continued to retain an unknown number of specimens with him in South America, which after his death in 1858 also were conveyed to Paris and ultimately also intercalated into the general phanerogamic herbarium. Physically locating these specimens that were once part of Bonpland's private collection represents a Sisyphean task as today more than eight million specimens are deposited in the general phanerogamic herbarium. Consequently, our analysis of these specimens is based on searches initiated through the specimen database of the Muséum National d'Histoire Naturelle, Paris ( (http://coldb.mnhn.fr/).

According to our research, the possibility that specimens gathered by the expedition might be found in other historic herbaria (e.g., P-JU, P-LA) deposited in the Muséum National d'Histoire Naturelle was suggested by the close relationship Bonpland maintained with his botanical mentors. Antoine-Laurent de Jussieu (1748–1836), Jean Baptiste Antoine Pierre Monnet de Lamarck (1744–1829), René-Louiche Desfontaines (1750–1833), and André Thouin (1747–1824), all professors associated with the Muséum, appear to have contributed to Bonpland's botanical education during his medical studies in Paris from 1795 to 1797. Accordingly we focused our search for specimens gathered by the expedition in P-JU because it, as with the general herbarium, could be searched electronically.

The Willdenow herbarium (B-W), Berlin-Dahlem

More than anyone else the renowned Prussian botanist Carl Ludwig Willdenow (1756–1812) was responsible for introducing Humboldt to botany. Their on-going and close friendship encouraged Humboldt to share with his principal mentor an important number of specimens gathered during the expedition to the New World. According to Urban (1916), a set of ca. 3360 specimens collected by Bonpland and Humboldt in the Americas was sent to Willdenow and the specimens in this set are now kept separately in B-W (Fig. 3A, 3B). This collection was studied during a visit to Berlin by FWS and JS in April 2006. Since this herbarium has an unique physical arrangement that is not tied to a single publication, locating Bonpland and Humboldt specimens deposited in it was carried out via a complex procedure that can be summarized as follows: 1) all of the monocotyledon genera present in the herbarium were identified by making use of an alphabetical list of the herbarium's contents compiled by Hiepko (1972); 2) specimens of monocotyledon filed under the genera included in Hiepko's list were identified and studied in the microfiche edition of B-W (IDC, 1971) and a preliminary list of specimens that might have been collected by the Humboldt and Bonpland expedition was created; and 3) all the material identified in the second step was physically examined to determine whether or not a given specimen was in fact collected by Bonpland and Humboldt. It was possible to recognize specimens as emanating from the Humboldt and Bonpland expedition because Diederich Franz Leonhard von Schlechtendal (1794–1866), who curated the collection from 1819 to 1833, annotated the bottom right-hand corner of herbarium sheets with the word “Humboldt” and “W”, the “W” signifying Willdenow. Identification of the species in the collection was only possible through Schlechtendal's catalogue (also included in the microfiches of B-W), which assigns reference numbers to all species contained in the Willdenow collection (Fig. 4A, 4B [note the name “Humboldt” is associated with the species Cenchrus myosuroides Kunth and C. rigidus Willd., nom. illeg.]). These reference numbers were retained by Hiepko (1972) in his catalogue of the microfiches and used by us to locate specimens of concern. Additional information relating to Bonpland and Humboldt specimens was compiled through a search of the on-line database of B-W (Röpert, 2009).

Specimens of Venezuelan monocotyledon in B-W were identified according to the same criteria utilized in searching P-Bonpl. (see above). We are aware that according to Hiepko (1972) one cannot always trust the names of collectors written on labels by Schlechtendal. However, for practical reasons we decided that specimens with the name “Humboldt” written on their labels would be considered part of the original set of plants that the two explorers sent to Willdenow.

Fig. 3.

Herbarium of the Botanischer Garten und Botanisches Museum Berlin-Dahlem. A. General view of the museum building ; B. Representative specimens corresponding to the monocotyledon collection of B-W.

Fig. 4.

Extracts of D. F. L. von Schlechtendal's catalogue of B-W. The first column at left corresponds to the number under which the taxon is filed in the herbarium. A. List of Canna L. (Cannaceae) taxa in the herbarium ; note the name “Humboldt” is associated with Canna glauca L.; B. List of Cenchrus L. (Poaceae) taxa in the herbarium.

[© Botanischer Garten und Botanisches Museum Berlin-Dahlem]

The general herbarium (B), Berlin-Dahlem

B was studied by FWS and JS during their April 2006 visit. A number of Bonpland and Humboldt specimens, especially type specimens, were deposited in this herbarium in the 19th century, but many of them subsequently were destroyed by Allied bombing during World War II (Merrill, 1943; Grimé & Plowman, 1986; Hiepko, 1987). Some groups in the original Kunth herbarium (e.g., ferns, lobelias, etc.) survived this bombing (Lack, 2009). Fortunately, J. Francis Macbride (1892–1976) of the Field Museum of Natural History, Chicago, photographed type specimens in Berlin from 1929 to 1933 and these images are available electronically (  http://fieldmuseum.org/explore/our-collections/berlin-negatives). Using these images we were able to record data from specimens that are no longer extant.

Unlike B, B-W was evacuated to a bank vault at the beginning of World War II (Hiepko, 1987) and therefore was not in the museum during the Allied bombing. Consequently B-W survived World War II intact and the Bonpland and Humboldt specimens deposited in it can still be studied, while the Bonpland and Humboldt specimens that were deposited in the general herbarium (with the exceptions noted above) only can be studied now by the proxy of Macbride's images.

The herbarium (HAL), Halle

Although Lanjouw & Stafleu (1957) reported the presence of Bonpland and Humboldt collections in HAL, it was a poster presented by Tkach & al. (2009) that made us realize that this herbarium held a greater quantity of specimens collected by the two explorers than generally reported. We used an online database ( http://herbarium.univie.ac.at/database/index.php) to search for and analyze Bonpland and Humboldt specimens in this collection.

The Cavanilles Herbarium (MA-CAV), Madrid

Humboldt's correspondence with Willdenow (Moheit, 1993: 129 [letter nr. 41]) indicates that shortly after the arrival of Humboldt and Bonpland in the New World and later during the Mexican portion of their journey an unknown quantity of seed of interesting plants was shipped to the Spanish botanists Casimiro Gómez Ortega (1740–1818) and Antonio J. Cavanilles (1745–1804). Accordingly, the microfiche of MA-CAV (IDC, 1994) was systematically searched for specimens that might be associated with this type of transaction.

The “Bonpland” herbarium (LR), La Rochelle

Humboldt's correspondence with Willdenow (Moheit, 1993: 122 [letter nr. 41]) also suggested the presence of expedition collections in LR. Accordingly, FWS and JS visited this herbarium in November 2009. All monocotyledon specimens present in LR that was assembled by Michel-Simon Goujaud Bonpland (1770–1850), Aimé Bonpland's older brother, were examined in an effort to locate any specimens that could be attributed to the Humboldt and Bonpland expedition. In addition, letters received by Michel-Simon, which are now kept in the archives of the Muséum d'Histoire Naturelle, were examined. Finally, the Médiathèque Michel Crépeau, La Rochelle was visited by FWS and JS in January 2011 in an effort to locate original documents or correspondence relating to Aimé Bonpland or his brother.

The “Journal Botanique” of Bonpland and Humboldt

The “Journal Botanique” (Fig. 5, 6) was examined by FWS and JS on two visits (November 2006 and 2008) to the Bibliothèque Centrale, Muséum National d'Histoire Naturelle, Paris. Original entries in these manuscript field notes pertaining to Venezuelan monocotyledons were compared to descriptions of corresponding taxa in Humboldt & al. (1816–1825), as well as to the lists of specimens that were compiled on earlier visits by FWS and JS to P-Bonpl. and B-W. Additional information concerning the “Journal Botanique” was obtained from Lack (2004a, 2004b).

Bibliographic research

Data relating to Venezuelan monocotyledons also were abstracted from several of the principal botanical publications associated with the expedition, notably Humboldt & Bonpland (1808–1809) and Humboldt & al. (1816–1825). Secondary sources providing information on the dates and itineraries of the expedition, and the collections and their fate (e.g., Sandwith, 1925; McVaugh, 1955; Biermann & al., 1968; Stearn, 1968; Hiepko, 2006) also were studied. Additional information about Venezuelan monocotyledons was obtained from the published correspondence of Humboldt (Moheit, 1993) and his Venezuelan diaries (Faak, 2000).

Shipment and distribution of the Venezuelan specimens

Details as to how specimens collected by Bonpland and Humboldt in Venezuela were shipped to Europe can be gleaned from letters that Humboldt sent from Havana to Europe as well as from letters that Humboldt sent to Bonpland and Willdenow shortly after the South American explorers themselves returned home. Venezuelan specimens appear to have arrived in Europe by at least three different means (Fig. 7, this flow chart is based on original data and information extracted from Lanjouw & Stafleu, (1954), and Hiepko, (2006). First, after the explorers had completed the Venezuelan leg of their journey and before they continued on to other parts of South America specimens were sent in two different shipments from Havana to Berlin. The vast majority of specimens, however, appears to have remained in the custody of Humboldt and Bonpland throughout their five-year long journey and these specimens were part of the baggage they brought back with them to Europe. Finally, a set of specimens that Bonpland had kept in his possession was shipped by him in 1832 to France from Argentina where Bonpland had settled following his return to South America.

Fig. 5.

Pages from the “Journal Botanique” of Bonpland and Humboldt. A. Left-hand page lists plants sent to Cavanilles, the arrow points to an annotation by Humboldt ; right-hand page lists some plants gathered in the Canary Islands ; B. First pages detailing the first collections made in Venezuela.

[© Bibliothèque Centrale, Muséum National d'Histoire Naturelle, Paris]

Fig. 6.

A page from the “Journal Botanique” of Bonpland and Humboldt. Humboldt's handwriting can be recognized on the left-hand side of the page (arrow 1) ; Bonpland's handwriting can be recognized on the right-hand side of the page (arrow 2).

[© Bibliothèque Centrale, Muséum National d'Histoire Naturelle, Paris]

Fig. 7.

Main pathways by which monocotyledon specimens collected by Bonpland and Humboldt in Venezuela (1799–1800) were distributed to individuals and herbaria.

Indeed, in late February 1801 Humboldt stated in a letter to Willdenow (Moheit, 1993: 122 [letter nr. 41]) that the plants collected in Venezuela had been distributed in three sets; a first and apparently small format set (“Ein Herbarium in kleinem Format …”) that would accompany the two explorers throughout their journey; a second set, that Humboldt described as Bonpland's set, which had already been shipped to France; and a third set containing about 1600 species, sent to Willdenow on 21 February 1801 through the courtesy of John Fraser (1750–1811) from Charleston, South Carolina via London. Fraser was a Scottish nurseryman then engaged in his own botanical explorations of North America and who also had a nursery in Chelsea near London. In a letter dated 1 April 1801 (Moheit, 1993: 136 [letter nr. 43]) that Humboldt sent to his brother Wilhelm he also mentioned another set of collections that had been left in Havana, Cuba. We have been unable to trace this set and its fate remains unknown. Humboldt collections have been reported as being present in HAC (Holmgren & al., 1990; see also  http://sweetgum.nybg.org/ih/), but the basis of this report is unknown and this Cuban herbarium has never had Humboldt collections (L. Regalado Gabancho & P. Herrera, pers. comm.). Humboldt's decision to send the first specimens and seeds that Bonpland and he had collected in the New World to Willdenow and to Cavanilles can be attributed to the close relationship that Humboldt had established with these two prominent botanists prior to leaving Europe. Humboldt had become a close personal friend of the slightly older Willdenow after first meeting him 12 years earlier when he sought Willdenow's help in identifying a plant using Willdenow's “Flora Berolinensis” (Stearn, 1968; Hiepko, 2006). Cavanilles had been of great value in helping to organize the expedition and may have played a role in advancing the interests of the explorers before the Royal Court of Spain.

Also in late February 1801 (Moheit, 1993: 125 [letter nr. 41]) Humboldt indicated that in a fortnight he would send Willdenow an additional set of 200 “very selected” specimens. These would be shipped directly to Hamburg. There is a list of specimens sent to Willdenow that is included on the inner front cover of the “Journal Botanique”, but we have been unable to establish whether or not this list enumerates specimens sent in the first or second shipment. These two shipments, which did not exceed 1800 specimens, were the only herbarium material collected by the explorers that was available to Willdenow until Humboldt returned from the New World (Hiepko, 2006). In the same volume of the “Journal Botanique” and before the description of the first collections, a similar list is provided of specimens sent to Cavanilles in Madrid. These too were shipped from Havana in February 1801 (Lack, 2004a). Humboldt and Bonpland evidently also wanted to send specimens to Bonpland's brother in La Rochelle under the care of the Franciscan monk Juan Gonzáles (fl. 1799–1801), who had accompanied the explorers in their Venezuelan travels. However, as stated in a letter (Moheit, 1993: 249 [letter nr. 113]) from Cumaná dated 22 August 1803, the colonial treasurer Manuel de Navarrete (1751–1819) informed Humboldt that neither Gonzales nor the specimens had survived a shipwreck off the coast of Spain.

Ultimately the largest set of plant specimens amassed by the two explorers remained with them throughout their journey and was brought by them back to Europe. We know from a letter dated 1 August 1804 (Moheit, 1993: 310 [letter nr. 160]) that Humboldt sent to his friend [Johann] Karl Freiesleben (1774–1846) that the prodigious amount of natural history material that Bonpland and Humboldt gathered, including the specimens collected in the Venezuelan portion of their journey, arrived in Europe packed in 30 boxes.

Representation of Venezuelan monocotyledons in several important herbaria

Hiepko (2006) presented a general overview of the complicated fate of the principal sets of Bonpland and Humboldt plant collections. He pointed out that when Kunth left Paris in 1829 there were four large sets of the expedition's plant collections of more than 5000 species. The main set (P-Bonpl.) was in the Muséum in Paris (Fig. 2A, 2B). A second set, retained by Bonpland, had been taken back to South America by Bonpland in 1816 (Fig. 7), but its precise size was unknown. A third set of ca. 3360 specimens was in B-W in Berlin (Fig. 3A, 3B), and it included both material sent from Cuba to Willdenow and material given to Willdenow by Humboldt after the expedition's return. The fourth and final set of ca. 3000 specimens was given by Humboldt to Kunth shortly before the former left Paris for Berlin.

In addition to these important sets, there are smaller sets, sometimes merely fragments of specimens removed from the principal sets, found now in many herbaria, including B, BM (grasses, at least), CGE, COL (ex P), E (cryptogams, at least), F, FI, G and G-DC (ca. 100 specimens; Burdet, 2008), H, HAL, K (fungi, at least; see Roberts, 2011), KIEL, L, LE (grass fragments ex B-W in the Trinius Herbarium, but specimens not otherwise in the South American collections; N. Imchanitzkaja, pers. comm.), LINN, LR, M (H. Esser, pers. comm.), MA (M. Velayos, pers. comm.), MA-CAV, MEDEL, MO (Juncus fragments ex P in Engelmann Herbarium), MPU (ex Dunal), NY, P, P-JU, PC, PH (Smith, 1962; Mears, 1981), S (specimens and type fragments ex P), and US (grass fragments and other specimens ex P) (Lanjouw & Stafleu, 1954, 1957; Stafleu & Cowan, 1979). The reports by Lanjouw & Stafleu (1954, 1957) of Bonpland and Humboldt specimens in W evidently are incorrect (Lack, 2004a) as no specimens are found there now and the basis of these reports is unknown. Furthermore, W. H. Lack (pers. comm.) does not believe that Bonpland orchid collections in Vienna, which were reported by Lanjouw & Stafleu (1954), ever existed and therefore were not among the collections destroyed during World War II.

We suspect that Bonpland and Humboldt specimens are in additional herbaria or collections as Wurdack (1971) noted that Bonpland had indicated in the “Journal Botanique” (Fig. 5A, 5B, 6) that specimens of Melastomataceae were given to Aylmer Bourke Lambert (1761–1842) in the summer of 1814 and to Robert Brown (1773–1858) in 1815. The Lambert herbarium was broken up and sold at auction and a lot with Bonpland material appears to have gone to G, but the contents of the lot were not recorded (Miller, 1970; see also Burdet, 2008). The Melastomataceae specimens given to Brown have not been traced, but should be sought in BM. A report by Solomon (1982) of a “Humboldt & Bonpland” specimen of Epilobium denticulatum Ruiz & Pav. (Onagraceae) from Ecuador in RJ [sic] has not been confirmed; the specimen probably is deposited in RB.

We provide information below on what we were able to learn in searching the main sets of specimens now in Paris and Berlin, and we present new information on several less wellknown sets assembled by the expedition that are now in Halle, Madrid, and La Rochelle.

The P-Bonpl. herbarium

On 18 December 1804, Humboldt wrote to the authorities of the Muséum in Paris and formally expressed his wish to deposit there a herbarium containing more than 6000 specimens packed in 45 boxes (Minguet, 1989). The Muséum currently holds ca. 3560 of these 6000 specimens in P-Bonpl., which it maintains as a separate historic herbarium (Fig. 2A, 2B, 8). The specimens contain a small label reading “Herb. Mus. Paris. Herbier Humboldt & Bonpland. Amérique Équatoriale” (Fig. 8). This collection is considered to represent the principal set of collections emanating from the Humboldt and Bonpland expedition (Hiepko, 2006). As explained below, the balance of the original deposit (ca. 3000 specimens) was offered by Humboldt to Kunth (Hiepko, 2006). The monocotyledon species present in the Bonpland Herbarium are arranged in exactly the same order as they appear in the first volume of Humboldt & al. (1816–1825) (Fig. 1). Additional taxa of monocotyledons published in the seventh volume (Humboldt & al., 1816–1825) were intermixed later in the same group of specimens.

Humboldt and Bonpland spent most of their time in Venezuela traveling together and as a consequence the collections in the Bonpland Herbarium should be attributed to both of them, but there is one notable exception. When they went from Cumaná to Caracas, a segment of their Venezuelan itinerary clearly described by Bonpland as the “Voyage de Cumaná à Caracas” (“Journal Botanique”), we know that they traveled separately. Humboldt sailed from Cumaná to the port of La Guaira and reached Caracas by the main road, whereas Bonpland, who did not like traveling by boat (Sandwith, 1925), left their ship well to the east of Caracas in the coastal region currently known as Barlovento (Miranda state) and reached the capital by land. Collections originating in these separate itineraries can be recognized from the localities on specimen labels.

The monocotyledon collections deposited in the Bonpland Herbarium (P-Bonpl.) were studied by Kunth and to a lesser extent by Willdenow. The collections do not seem generally to have been made available to other contemporary botanists for study. One exception, however, was the French botanist Louis Claude Marie Richard (1754–1821) who worked closely with Bonpland on the publication of Humboldt & Bonpland (1806–1823). Elsewhere Richard described Elodea orinocensis Rich. (Hydrocharitaceae) from a collection gathered during the Venezuelan portion of the Humboldt and Bonpland expedition and we surmise that his access to this and other expedition specimens was facilitated by his professional friendship with Bonpland.

For reasons that are unknown to us most of the original labels handwritten by Bonpland were replaced by Kunth and these replacement labels usually contain only the species name and very rarely the original number assigned to the collection and recorded in the “Journal Botanique” (Fig. 9C-F). Bonpland's labels are less frequently encountered than those of Kunth, but when found they contain more details such as the collection number, locality, date of collection (following the French Republican calendar) (Fig. 9A, 9B), and occasionally a common name. The infrequent inclusion of collection numbers on the replaced labels is from our current perspective a bit surprising, since this was the only way by which Kunth could unequivocally link a given specimen to a description in the “Journal Botanique” (Fig. 5A, 5B, 6). Admittedly, the citation of collection numbers was not a common practice in the 18th and early 19th centuries, but Kunth's decision to replace the original labels nonetheless is unfortunate. In early October 1814, one year before the publication of the first volume of Humboldt & al. (1816–1825), Bonpland commented upon this numbering practice in a letter sent to Humboldt (Hamy, 1906: 7 October 1814): “je ne vois aucune utilité à citer le numéro du Manuscrit. Nous ferions en cela ce qu'aucun voyageur et ce qu'aucun auteur n'ont fait et cette innovation serait en pure perte … Nous avons à cet égard à suivre ce qu'ont fait Desfontaines et les autres botanistes, c'est-à-dire nous devons garder nos manuscrits pour nous” [I do not see any utility in citing the manuscript numbers (making reference to the “Journal Botanique”). Doing so, we would do what no traveler and no author has ever done and this innovation would be a pure loss … We have followed in this respect what has been done by Desfontaines and other botanists, that is to say, we must keep our manuscripts for ourselves].

The entire monocotyledon collection, including species from all of the present-day countries visited during the journey of Humboldt and Bonpland, contains about 350 specimens and more or less the same number of species. There are 86 specimens of monocotyledon from Venezuela and these represent 86 species belonging to 57 genera and 19 families (Table 1). In almost all cases monocotyledon specimens deposited in P-Bonpl. are represented by a single sheet. In fact, we found only 14 species, half of them orchids, represented by more than one sheet. We suspect that the paucity of duplicates can be attributed to the fact that extra material probably was distributed by Willdenow, Kunth, and even Humboldt or given to Kunth.

Fig. 8.

Representative specimen of the monocotyledon collection deposited in P-Bonpl. The example is Heliconia psittacorum L. f. (Heliconiaceae) collected in the region of Caripe, northeastern Venezuela.

[© Herbarium, Muséum National d'Histoire Naturelle, Paris]

Fig. 9.

Examples of labels associated with the monocotyledon specimens at P-Bonpl. A–B. Original labels with Bonpland's handwriting (i.e., collector's number and locality) and subsequent determinations made by Kunth ; C–F. Replacement labels with only Kunth's handwriting.

[© Herbarium, Muséum National d'Histoire Naturelle, Paris]

Table 1.

Venezuelan specimens of monocotyledon deposited in P-Bonpl.

continued

continued

continued

continued

continued

The specimens now in P-Bonpl. must be considered the nomenclatural types (holotypes, isotypes, and syntypes) of names of species described in Humboldt & al. (1816–1825) as this is the collection that formed the basis for this publication. However, taxonomic problems arise when type material is not found in P-Bonpl. as there are only a few indications by Kunth that the material he studied corresponded to specimens deposited in Bonpland's “private” collection. We determined that 235 species of monocotyledon described in Humboldt & al. (1816–1825) are not represented now by specimens in P-Bonpl. Of these 235 species, no material whatsoever could be found for 203 species (Table 2). Thirty-two species, however, were represented by pen and ink line drawings with wash (grisailles), which in most cases were rendered by Pierre Jean François Turpin (1775–1840) (Table 3). Some of the grisailles contain slight corrections in pencil suggesting that they were preliminary versions of plates that were later engraved and included in Humboldt & al. (1816–1825). Approximately half (106) of the species for which type specimens are not found now in P-Bonpl. are Poaceae. This suggests that Humboldt allowed Kunth to remove grass specimens for the latter's treatment of the family in volume one of Kunth (1833) and its supplement (Kunth, 1835).

The whereabouts of material thought to be deposited in P-Bonpl. but not found there has always puzzled botanists interested in this historical collection. Hind & Jeffrey (2001) encountered a similar problem in their study of the Asteraceae that were described in Humboldt & al. (1816–1825). They analyzed the microfiche edition of P-Bonpl. (IDC, 1972), and were perplexed that many taxa had not been imaged and they had no explanation for this except that perhaps specimens were unavailable to be filmed, or alarmingly, that specimens had disappeared from this collection. We believe that there is strong evidence that the missing specimens did not disappear, but in fact were deposited in the general collections in Paris (P) or in B-W or B. Alternatively, it is possible that original collections, and not just duplicates or fragments, were acquired by the herbaria we listed above as minor depositories of Bonpland and Humboldt specimens. Yet another possibility, suggested by Lack (2004a), is that Bonpland did not number all of the herbarium specimens and that a complete set corresponding to the entries in the “Journal Botanique” never existed. Also, as noted by Lack (2004a), it seems reasonable to expect that herbarium specimens of succulent or bulky plants might not have been prepared as these plants present unusual challenges to the collector. Palms (Arecaceae), a family that incited great interest in the two explorers, typically require special handling and yield bulky specimens. Interestingly, 14 of the 24 palm species described by Kunth in 1816 in Humboldt & al. (1816–1825) are not found now in P-Bonpl.

We also know from an 1801 letter (Moheit, 1993: 126 [letter nr. 41]) from Humboldt to Willdenow that many specimens collected in Venezuela were destroyed by a combination of extremely high humidity and temperature. Of particular note is Humboldt's comment regarding the conservation of plant specimens: “Aber ach! mit Thränen eröffnen wir fast unsere Pflanzenkisten. Unsere Herbarien haben dasselbe Schicksal, über das bereits Sparman, Banks, Swartz und Jacquin geklagt. Die unermessliche Nässe des amerikan[ischen] Klimas, die Geilheit der Vegetation, in der es so schwer ist, alte, ausgewachsene Blätter zu finden, haben über 1/3 unserer Sammlung verdorben ... Ist man 3–4 Monate abwesend, so kennt man sein Herbar[ium] kaum wieder, von den 8 Exemplaren muss man 5 wegwerfen ....” [With tears in our eyes we are opening the boxes with plants. Our herbarium has suffered the same fate to which already Sparman [sic], Banks, Swartz and Jacquin complained. The immeasurable humidity of the American weather, responsible for the accelerated growth of the vegetation that makes impossible to find fully developed leaves have destroyed more than one third of our collection … one scarcely recognizes his herbarium after 3–4 months of absence, from 8 specimens 5 should be thrown away …]. Interestingly, almost four decades later Bonpland informed the director of the Muséum in Paris that some of his herbarium specimens were destroyed by similar environmental problems during the time he spent in Argentina and Paraguay (Hamy, 1906: 5 January 1837): “ … le temps m'a à peine permis de la parcourir et d'en séparer le papier en partie détruit par l'humidité et les plantes réduites en poussière” [… time hardly allowed me to go through it (referring to the collection sent to Paris) and remove the paper partially destroyed by the humidity and the plants (that were) reduced to dust].

The P and P-JU herbaria

Toward the end of December 1805 Humboldt wrote Bonpland (Hossard, 2004: 22): “I will send your plants as soon as we have finished sorting them out”. This seems a clear indication that not long after their arrival in Europe the two explorers had divided the specimens into two main sets, one permanently kept by Humboldt in Paris and the other consisting of an unknown number of specimens kept by Bonpland (referred to here as Bonpland's “private” set of collections). As can be inferred from correspondence between the two explorers (Hossard, 2004), soon after Humboldt sent the plants to Bonpland he realized the importance of this material for the preparation of Humboldt & al. (1816–1825) and he asked Bonpland to return it to Paris. Thus, in at least three letters sent to Bonpland in 1806 (Hossard, 2004: 26, 46, 52), Humboldt firmly pointed out the critical importance of receiving without delay Bonpland's private set, as well as an unknown number of specimens belonging to Humboldt's set of collections but still kept by Bonpland in Paris. Much later, when Bonpland left Europe in November 1816 with the intention of settling in Argentina (Sarton, 1943; Lourteig, 1977; Hossard, 2001) (Fig. 7) he took with him these collections (i.e., the “private” set) and attempted also to take the “Journal Botanique” (Hiepko, 2006) (Fig. 5A, 5B, 6). The field notes fortunately were reclaimed at the last moment by Kunth (Lack, 2004a), but the specimens went with Bonpland back to South America and were not returned to Europe until 1832 (Fig. 7). In 1837 Bonpland described the return of these specimens to Paris in a letter to the director of the Muséum (Hamy, 1906: 5 January 1837): “Ces plantes peuvent être divisées en deux parties. Dans la première je placerai les plantes doubles du voyage que j'ai fait avec M. de Humboldt et qui me sont tombées en partage; et dans la seconde, des plantes de tous les pays qui fesaient partie de mon herbier général” [These plants can be divided into two parts. In the first I will place the duplicates from the voyage I undertook with M. Humboldt and which were my share; and in the second, plants from all countries that constitute part of my general herbarium].

Table 2.

Monocotyledon species cited in Humboldt & al. (1816–1825) for which corresponding specimens were not found in P-Bonpl.

continued

continued

continued

continued

continued

Table 3.

Monocotyledon species cited in Humboldt & al. (1816–1825) but only represented by grisailles in P-Bonpl.

Bonpland specimens arrived again in Paris in 1833 and on 21 July 1837 the Muséum agreed to incorporate all of them in P (Lack, 2003). These specimens and their duplicates in other herbaria can be recognized by a printed notation on their labels (viz., “Herbier donné par M. Bonpland en 1833” or “Herbier de l'Amérique équatoriale, donné par M. A. Bonpland”) (Fig. 10). Bonpland's set is important not only because some of the unicates contained in it were used by Kunth while editing Humboldt & al. (1816–1825), but also because this private set undoubtedly contains a large number of duplicates of P-Bonpl. as well as types associated with Humboldt & Bonpland (1808–1809) and Humboldt & Bonpland (1806–1823). Bonpland was deeply invested in these two publications and one would expect that many of the specimens he cited in these monographs were always kept in his private herbarium. It is possible, too, that types associated with those in Bonpland (1813), none of which have yet been located (Stafleu & Cowan, 1976), also once were contained in his private collection or, as will be explained below, are present in other historical herbaria such as that of P-JU.

The exact number of specimens and taxa once contained in Bonpland's private set of Bonpland and Humboldt collections, now fully integrated in P, remains unknown and physically locating these specimens is a monumental task. This is not only attributable to the large number of specimens involved, but also because relevant specimens might have had several name changes over the years as they were studied and curated. However, targeted efforts focused on specific families such as Asteraceae (Hind & Jeffrey, 2001) and Solanaceae (Granados Tochoy & al., 2007; S. Knapp, pers. comm.) have often shown promising results. Moreover, preliminary studies carried out in the palm (Arecaceae) family clearly demonstrated that some of the specimens used for the publication of Humboldt & Bonpland (1816–1825) are to be found in the general collection in Paris.

An electronic search of P-JU yielded 32 collections of different pteridophyte and angiosperm families linked to Bonpland. Among these collections, the specimens labeled with New World localities can definitely be associated with Aimé Bonpland, but for specimens labeled with European localities it is not clear whether they were gathered by Aimé or his brother Michel-Simon, who also was in close contact with Antoine-Laurent de Jussieu. In this historic herbarium there is a particularly good representation of specimens of the genus Tacsonia Juss. (Passifloraceae), which is not surprising as de Jussieu based several species (e.g., T. glaberrima Juss., T. manicata Juss., and T. tripartita Juss.) on specimens collected by Aimé Bonpland in present-day Colombia, Ecuador, Peru, and Venezuela.

We also are aware that Humboldt and Bonpland had sent by the end of January 1805 not only dried herbarium specimens to the Muséum National d'Histoire Naturelle, Paris, but also abundant seed (MS 617, Médiathèque Michel Crépeau, La Rochelle) (Fig. 11) that undoubtedly was given to the Jardin des Plantes for cultivation. There is always the possibility that specimens prepared from this cultivated material found their way into P.

The B-W herbarium

The retrieval, examination, and evaluation of monocotyledon specimens collected by Bonpland and Humboldt and now deposited in B-W (Fig. 3A, 3B, 7) is challenging and has inherent practical difficulties (see e.g. Hind & Jeffrey, 2001). The complicated methodology we adopted for our project, however, proved to be quite satisfactory for retrieving and analyzing Venezuelan monocotyledon specimens. We discovered 126 specimens of Venezuelan monocotyledon in this herbarium (Fig. 3B, 12) and they correspond to the same number of species arranged in 64 genera and 26 families (Table 4).

In addition to herbarium specimens, Humboldt and Bonpland sent seed to Willdenow and while we do not know the exact number of propagules, we do know from a letter Humboldt sent to Bonpland on 21 December 1805 (Hossard, 2004: 22) that as many as 75 “species” conveyed to Willdenow were successfully cultivated in Berlin and that Willdenow expected many more seeds to germinate.

Willdenow's set of Bonpland and Humboldt specimens contained not only duplicates of most of the collections deposited in P-Bonpl., but also some unique specimens not represented in the principal set. After Willdenow's death in 1812, his herbarium passed to the custody of his friend D. F. K. von Schlechtendal (1767–1842) (McVaugh, 1955) and eventually Schlechtendal's son, Diederich Franz Leonhard von Schlechtendal. The herbarium was purchased by the Botanischer Garten und Botanisches Museum Berlin-Dahlem and came into its possession in 1818. The younger Schlechtendal was appointed curator of this collection the following year and he initiated a complete rearrangement of B-W according to the Linnaean system (Schlechtendal, 1832). He produced a detailed catalogue of the Willdenow Herbarium, which L. Krug copied much later introducing mistakes (Hiepko, 1972) (Fig. 4A, 4B). In Schlechtendal's catalogue all species represented in the collection are assigned a reference number. With respect to the groups of monocotyledon studied, most specimens of the same family are physically united in only one place in the collection, but in some cases specimens of a family are found in two (e.g., Arecaceae, Cyperaceae, and Marantaceae), three (e.g., Araceae) or even four different places (e.g., Poaceae) (Table 4).

Fig. 10.

Representative specimen collected by Bonpland and Humboldt deposited in P. The example is Smilax cumanensis Willd. (Smilacaceae). Note Bonpland's handwriting on the original label (bottom left-hand corner of the specimen).

[© Herbarium, Muséum National d'Histoire Naturelle, Paris]

Fig. 11.

Partial text of a list of seed sent by Humboldt and Bonpland to the Muséum National d'Histoire Naturelle, Paris. The page shown is dated “9 pluv an 13” (i.e., 29 January 1805).

[© Médiathèque Michel Crépeau, La Rochelle]

Fig. 12.

Representative specimen collected by Bonpland and Humboldt deposited in B-W. The example is Epidendrum atropurpureum Willd. (Orchidaceae). Note Bonpland's handwriting on the original label. Arrows point to D. F. L. von Schlechtendal's annotations at the top right-hand corner of the sheet with the species name and at the bottom right-hand corner with the word “Humboldt” and letter “W” (for Willdenow).

[© Botanischer Garten und Botanisches Museum Berlin-Dahlem]

Table 4.

Venezuelan specimens of monocotyledon deposited in B-W.

continued

continued

continued

continued

continued

continued

continued

continued

continued

Bonpland and Humboldt specimens are in general well preserved and often retain the original label in Bonpland's hand (Fig. 12, 13A, 13B), although in some cases Humboldt added spelling corrections to the localities proposed by Bonpland. In many cases these labels contain the collection number, locality, and date of collection (following the French Republican calendar), and occasionally they include a common name. Willdenow was in the habit of noting on labels, which he subsequently affixed to folders, the geographical origin of the specimens within. He did this also with specimens from the Humboldt and Bonpland expedition. However, in some cases the locality written on the label affixed to the folder does not correspond to the locality cited in either Humboldt & al. (1816–1825) or the “Journal Botanique”. Two examples illustrate this situation. First, the label on the folder for Eriocaulon congestum Kunth (Eriocaulaceae) indicates “Cumaná”, but the label on the Bonpland and Humboldt specimen lacks a collection number and locality. Nevertheless, Kunth indicated “Amazonas” as the type locality when he described the species in Humboldt & al. (1816–1825). The genus Eriocaulon L. is not present in Cumaná or northern Venezuela. We believe therefore that the specimen was collected in the Venezuelan Guayana, but for whatever reasons Willdenow failed to copy the locality information correctly on the folder's label. A second example is Scirpus hirtus Poir. (Cyperaceae) where the label on the specimen clearly indicates the collection number (“755”) and locality (“Orinoco”). However, we believe that this collection number is incorrect since collection numbers for specimens gathered in this region of Venezuela (i.e., present-day Amazonas state) begin with number 821. Moreover, the same collection number (755) was given to a specimen of Schoenus spadiceus (Lam.) Vahl (Cyperaceae) collected in the mountains near Caracas and deposited in P-Bonpl.

The Bonpland and Humboldt specimens in B-W were studied almost exclusively by Willdenow, but after his death in 1812 and especially after D. F. L. von Schlechtendal was appointed director of the herbarium in 1819, they were made accessible to many other botanists including Johann Jakob Römer (1763–1819) and Josef August Schultes (1773–1831), who apparently also had access to the botanical notes left by Willdenow (McVaugh, 1955). Schultes's son, Julius Hermann Schultes (1804–1840), Curt Polycarp Joachim Sprengel (1766–1833), Johann Heinrich Friedrich Link (1767–1851), and many others also were able to study these Bonpland and Humboldt collections. After Link was appointed director of the Berlin garden in 1815, he had access to B-W and he published (Link, 1820) at least seven new species of monocotyledon based on specimens collected by Bonpland and Humboldt. As will be described below, it was only much later that Kunth was able to study the Bonpland and Humboldt collections in B-W.

Several authors (e.g., Stafleu & Cowan, 1983; Rankin Rodríguez & Greuter, 2001; Lack, 2003) have suggested that it was D. F. L. von Schlechtendal who enabled Römer and Schultes, and others, to use and publish Willdenow's manuscript names and brief descriptions. However, more recently Hiepko (2006) has demonstrated convincingly that the publication of most of Willdenow's manuscript names was arranged instead by the elder Schlechtendal. In any case, this arrangement led to the multiplication of synonyms because of the independent but nearly simultaneous publication of taxa based on these collections (and manuscripts); Kunth in Humboldt & al. (1816–1825) and Römer & Schultes in the third volume of their Systema Vegetabilium (McVaugh, 1955). The nomenclatural consequences for over 250 names have been discussed elsewhere (McVaugh, 1955; Hiepko, 2006). Hence, while preparing the manuscript of Humboldt & al. (1816–1825), Kunth was unable to study Willdenow's set of the Bonpland and Humboldt collections, which contained not only duplicates, but also some unicates (Hiepko, 2006). Kunth only gained access to Willdenow's set when he was appointed assistant-director of the Botanischer Garten in 1829. Apart from Römer and Schultes, other taxonomists also studied the Bonpland and Humboldt collections and in some cases they too contributed to the multiplication of synonymous names. Just to cite one example, Kunth described Yucca acaulis Kunth (Asparagaceae) based on Bonpland and Humboldt collection number 633, which presumably was deposited in P-Bonpl., but the same collection was later described as Codonocrinum agavoides Schult. & Schult. f. (Asparagaceae) and based on a duplicate specimen deposited in B-W.

Fig. 13.

Example of labels associated with the Bonpland and Humboldt monocotyledon specimens at B-W. A. Bonpland label showing different handwritings (1) Bonpland, (2) Humboldt, (3) Willdenow, (4) D. F. L. von Schlechtendal ; B. Bonpland label showing different handwritings (1) Bonpland, (2) Humboldt, (3) D. F. L. von Schlechtendal.

[© Botanischer Garten und Botanisches Museum Berlin-Dahlem]

Three specimens in B-W have numbers, localities (i.e., Madrid, Aranjuez), and dates that correspond to the Spanish segment of the Humboldt and Bonpland expedition, the portion of the voyage before the two explorers left Europe. The specimens are determined as Allium candidissimum Cav. (nom. illeg.) (Amaryllidaceae), Milium confertum L. (Poaceae), and Cynosurus sp. nov. (ined.) (Poaceae) (Table 3), all taxa considered to be European. While it is tempting to interpret this as a case of mistaken numbering (the numbers otherwise correspond to plants gathered near Cumaná), we suspect that these specimens were numbered independently of the American ones. The analysis of Lack (2004a) of the “Journal Botanique” noted that the first of the seven volumes (MS 221) has “several short runs” of collection numbers, and this volume covers Humboldt and Bonpland's itinerary in France and Spain.

Finally, the active exchange program maintained by Willdenow with other contemporary botanists seems to be the principal reason why some of the specimens used by Willdenow as types of names that he published are not found now in B-W. Yet this might not be the case with respect to Bonpland and Humboldt specimens since we assume that Willdenow appreciated the novelty and value of these botanical collections made in what had been largely unexplored areas of the Americas and he would have been reluctant to part with them.

The B herbarium

The private herbarium of Kunth contained ca. 70,000 specimens, comprising about 54,500 species and including almost 3000 types of names of taxa described in Humboldt & al. (1816–1825) (Fig. 1), as well as many duplicates from the herbarium in Paris (Hiepko, 1987). There is little doubt that it was Humboldt himself who donated to Kunth the set of almost 3000 specimens from the American expedition shortly before Humboldt left Paris for Berlin. After Kunth's death in 1850 his herbarium was purchased by the Prussian state and added to B (Fig. 3A, 7, 14). Unfortunately most of these specimens were destroyed during the Allied bombing of Berlin in 1943, along with the greater part of B (Merrill, 1943; Grimé & Plowman, 1986; Hiepko, 1987).

As noted earlier, it is only through the photographs of J. Francis Macbride of the Field Museum of Natural History, Chicago that we have any visual record of many specimens that once were deposited in B. Studying these photographs we were able to identify nine specimens that could be attributed unequivocally to the set of collections that Humboldt gave to Kunth. To this group belong not only specimens collected by Bonpland and Humboldt that Kunth described almost 30 years after finishing Humboldt & al. (1816–1825) (e.g., Tradescantia cumanensis Kunth, Commelinaceae), but also Bonpland and Humboldt specimens (e.g., Cyperus oligostachyus Kunth, Cyperaceae, Eriocaulon tenue Kunth, and Juncus densiflorus Kunth, Juncaceae) not found in P-Bonpl. (Table 2). A well documented case of a Bonpland and Humboldt specimen described by Kunth long after he left Paris is that of Phaedranassa multiflora Kunth (Amaryllidaceae) (Leuenberger & Arroyo-Leuenberger, 2006). We believe that this confirms that Humboldt permitted Kunth to take with him to Berlin not only an important group of type specimens of taxa described in Humboldt & al. (1816–1825), but also undetermined specimens that required further study (Fig. 14).

The HAL herbarium

HAL proved to be a more important repository of Bonpland and Humboldt specimens than previously appreciated. There was no direct contact between the explorers and the curators of this herbarium. Instead, the specimens were acquired through D. F. L. von Schlechtendal, who held positions in both Berlin and Halle. From 1819 to 1833, Schlechtendal was in charge of the Berlin herbarium, including the large Willdenow collection (B and B-W). He left Berlin when he was appointed professor of botany and director of the botanical garden of the university in Halle where he served until his death in 1866. Presumably he brought these Bonpland and Humboldt collections with him to Halle (Fig. 7).

In an electronic search ( http://herbarium.univie.ac.at/database/index.php) of HAL we identified more than 50 Bonpland and Humboldt specimens that clearly represent duplicates of specimens deposited in B-W (Fig. 15). We do not know why Schlechtendal removed these duplicate collections from B-W, but suspect that as director of the Berlin herbarium he probably believed that their removal would not impact the value of the Willdenow collection, especially in those cases where enough material was clearly available.

Two collections in HAL associated with the species Scleria cyperina Kunth (Fig. 15) and S. nutans Kunth, respectively, clearly can be regarded as duplicates (i.e., isotypes) of corresponding specimens in B-W as the specimen pairs contain identical information on their labels (viz., “America Merid., Cumaná, Humboldt”). All specimens studied have original labels on which Schlechtendal wrote that the specimen was a duplicate from B-W and he copied the scanty information contained on the Willdenow specimens (Fig. 15). Hence, in most cases only the locality “America Meridionalis” and the additional note “Humboldt” was written by Schlechtendal; very rarely more detailed information on the locality (e.g., Cumaná) was provided. Numbers and dates of collections are completely lacking, which complicates the recognition of isotypes or designation of lectotypes. In any case, it is expected that eventually the number of Bonpland and Humboldt specimens discovered in the general herbarium of the Martin-Luther Universität will elevate this herbarium to fourth (in terms of specimen number) among the important depositories of Bonpland and Humboldt specimens. Inasmuch as Schlechtendal was interested in genera of Cyperaceae (e.g., Carex L.) and Poaceae (e.g., Panicum L., Paspalum L.) we suspect that eventually a large amount of Bonpland and Humboldt material will be located in these families in HAL.

Fig. 14.

Monocotyledon specimen collected by Bonpland and Humboldt deposited in B. The specimen corresponds to Amaryllis nervosa Kunth, hom. illeg. (Amaryllidaceae) collected in present-day Aragua state of Venezuela. Note Kunth's handwriting on the label (bottom left-hand corner of the specimen).

[© Botanischer Garten und Botanisches Museum Berlin-Dahlem]

Fig. 15.

Monocotyledon specimen collected by Bonpland and Humboldt deposited in HAL. The specimen is an isotype of Scleria cyperina Kunth (Cyperaceae) collected in present-day Sucre state of Venezuela. Note D. F. C. von Schlechtendal's handwriting on the label (bottom left-hand corner of the specimen).

[© Herbarium, Martin-Luther Universität, Halle-Wittenberg]

The Cavanilles Herbarium (MA-CAV)

The first page of the “Journal Botanique” (MS 1332) includes a list of 39 collections that Humboldt and Bonpland sent from Havana to Antonio José Cavanilles in Madrid (Fig. 6A) and correspondence between Humboldt and Willdenow (Moheit, 1993: 129 [letter nr. 41]), indicates that seed was sent in 1800 to Casimiro Gómez Ortega, director of the Real Jardín Botánico, Madrid, and probably also to Cavanilles, the clergyman and botanist who eventually succeeded Ortega as director (Fig. 7). Our research demonstrates that Cavanilles cultivated and later described new species from seed collected near Cumaná by Bonpland and Humboldt. We expected that we would find type collections vouchering the names of these new species in MA-CAV and following an examination of the microfiche of this herbarium (IDC, 1994) we were able to locate the types of two Asteraceae names: Tagetes peduncularis Cav. (Fig. 5A) and T. verticillata Lag. & Rodr. Labels on both specimens indicate that these plants originated in Cumaná and that they had flowered in the Real Jardín Botánico, Madrid in 1801. These label data agree with information published in the relevant protologues.

When Humboldt wrote Cavanilles from Mexico on 22 April 1803 (Moheit, 1993: 227 [letter nr. 103]), he indicated that an unspecified number of herbarium specimens gathered by the expedition were ready to be sent to Madrid. A search of the microfiche of the MA-CAV failed to yield any specimen that could be tied to such a shipment. Perhaps Humboldt learned that Cavanilles was in poor health and this prompted him to send the promised specimens elsewhere. In any case, no monocotyledon specimens, cultivated or otherwise, that could be linked to the Humboldt and Bonpland expedition were located in the Cavanilles Herbarium.

The LR herbarium

The specimens associated with the so called “Herbier Bonpland” (LR) in La Rochelle traditionally have been attributed to Michel-Simon Goujaud Bonpland, older brother of Aimé Bonpland. This herbarium was described by Rallet (1970) and Stauffer & Stauffer (2010). A letter dated 18 brumaire an X (i.e., 9 November 1800) (Hamy, 1906) that Aimé sent to his brother from Cumaná indicates that three boxes with plants collected by the expedition were to be sent to the Muséum in La Rochelle. If these specimens arrived safely we would expect them to be in the “Herbier Bonpland”. We studied a majority of the specimens (all of the monocotyledons and ca. 90 per cent of the dicotyledons) in this herbarium with the hope of finding collections that could be attributed to this shipment from Venezuela to France. The “Herbier Bonpland”, which is kept separate from other collections in the Muséum in La Rochelle, consists of 36 bundles each with 60–70 specimens for a total of 2000–2500 specimens. Few of the specimens are mounted, which means that we were obliged to handle them with extreme care to avoid creating confusion. The 36 bundles are arranged according to the Linnaean system and consist of seven bundles of monocotyledons, 28 bundles of dicotyledons, and one bundle of cryptogamic plants (algae, lichens, and mosses) and fungi. The monocotyledon portion of the collection appears to be better preserved than the dicotyledon one, as the latter shows evidence of what appears to be insect damage in Asteraceae and Brassicaceae, especially. In contrast, the Poaceae are very well preserved and surprisingly diverse taxonomically.

Few of the labels for specimens in LR have locality data. Those that do indicate that collections were made in France (e.g., Barèges, bois de Boulogne), French gardens (e.g., H.F., or “Hortus Fontainebleau”), Spain (e.g., Aranjuez, Barcelone, Madrid, Tenerife, “Hispania”), or other countries (St.-Thomas, Portugal, Suisse (Vaud), Tunisie). Michel-Simon actively corresponded with many contemporary botanists, whose names can be found on specimen labels, including Bellanger, Bonafosse, Cavanilles, de Candolle, Desfontaines, Delisle, Lorenti, Née, Pourret, Richard, Thibault, Thouin, Violet, Verdier, and Zimmermann. There are very few dates on the labels. In fact, only two were found: 1792 associated with the locality “Cévennes” and 1843 with “Barèges”.

It should be pointed out that doubts were raised by Jolinon & Pignal (1998) with respect to whether or not LR was distinct from the “Herbier Dessalines d'Orbigny”, even though Rallet (1970) did not express any doubt about their distinctiveness. The formation of the “Herbier Dessalines d'Orbigny”, also deposited in the Muséum in La Rochelle, has been attributed to Charles-Marie Dessalines d'Orbigny (1770–1856), father of the famous explorer Alcide Charles Victor Marie Dessalines d'Orbigny (1802–1857). It consists of 13 bundles stored adjacent to the “Herbier Bonpland”. We found no internal evidence that would allow one to unequivocally distinguish or separate the two herbaria. In many respects (e.g., the format of the specimens, the type of information recorded on labels, etc.) the two collections resemble each other and more critical study is needed to establish whether or not they were formed independently of each other.

Nor were we able to clearly establish a difference between the specimens collected by Michel-Simon and the ones that might have been collected by his brother Aimé. Moreover, a comparison of the handwriting of the two brothers does not allow us to distinguish specimens that either one labeled or annotated. Many of the collections from Spain match localities visited by Aimé shortly before he and Humboldt left Europe (e.g., Aranjuez, Madrid) and at least one locality (i.e., Tenerife) was visited by the explorers en route to the Spanish colonies in America. In fact, Michel-Simon wrote to André Thouin on 15 January 1799 (folder 68/11, archives, Bibliothèque, Muséum d'Histoire Naturelle, La Rochelle) and told the Parisian horticulturist that he had received some plants collected in Spain by his brother. This alone suggests that an important number of plants from Spain were indeed collected by Aimé shortly before he and Humboldt departed Europe.

Despite many complications and much confusion we were able to locate in LR two specimens that can be attributed unequivocally to the Humboldt and Bonpland expedition. The first is a lichenized fungus bearing the name Gymnoderma, but recently identified as Dictyonema glabratum (Spreng.) D. Hawksw. by P. Roberts (pers. comm.), and its label indicates that it was sent from Caracas by Aimé Bonpland. Curiously, among the Humboldt fungi deposited in K there are none from Venezuela (Roberts, 2011). The second specimen is a Mutisia L. f. (Asteraceae) labeled “hab. in Monte Quindio”. Monte Quinidio is a locality Humboldt and Bonpland visited during the Colombian portion of their journey. We discovered no monocotyledon specimens in the herbarium of the Muséum in La Rochelle that could be connected unequivocally to the Humboldt and Bonpland expedition.

The Médiathèque Michel Crépeau, La Rochelle preserves 53 documents associated with Aimé Bonpland (all arranged in their folder cote ‘MS 617'). The documents are a mixture of lists of seed that Bonpland distributed to European botanical gardens including Berlin, Montpellier, and Paris (Fig. 11), and correspondence he received from Humboldt, José Antonio Pavón (1754–1840), Augustin Pyramus de Candolle (1778–1841), and others. Also present in the Médiathèque are copies of letters that Aimé sent to the Empress Josephine (1763–1814) and several documents associated with his administrative duties in the Jardin de Malmaison. It is also worth mentioning an interesting group of botanical notes, most of them related to Arecaceae, Melastomataceae, and the genus Cinchona L. (Rubiaceae). Although none of these documents provided critical information for our research on Venezuelan monocotyledons they certainly represent an important source of data relating to the botanical work of Aimé Bonpland.

Venezuelan monocotyledons and the “Journal Botanique” of Bonpland and Humboldt

The field notes of Bonpland and Humboldt, also known as the “Journal Botanique”, are deposited in the Bibliothèque Centrale, Muséum National d'Histoire Naturelle, Paris. The format and contents of these manuscripts were analyzed by Lack (2004a, 2004b). The “Journal Botanique” consists of seven bound volumes that include descriptions of 4528 collections made by Bonpland and Humboldt during their American journey (Fig. 5A, 5B, 6). Bonpland was clearly in charge of compiling all the information, but complete descriptions or comments on specific taxa (e.g., Arecaceae, Bromeliaceae, Orchidaceae) were occasionally added by Humboldt (Fig. 6). Entries were made in black ink. Although descriptions are in Latin and French, notes in Spanish are scattered throughout the text and mostly concern local uses, common names, etc.

Detailed information on the locality, often later complemented by Humboldt himself, and date of collection always precedes a group of entries. The entries are numbered sequentially throughout the work, although occasionally plants that were collected in one region received numbers corresponding to a later stage of the journey. Provisional scientific names were proposed and used, first by Bonpland and later added by Willdenow and Kunth to whom the “Journal Botanique” passed. As stated by Hiepko (2006), Kunth's use of many names proposed by Willdenow in the “Journal Botanique”, for which Kunth failed to give Willdenow credit, was a source of conflict with other German botanists who were aware of this situation.

The descriptions in the “Journal Botanique” are typically extensive (Fig. 6) but in a relatively few cases (e.g., numbers 423–475, which correspond to pteridophytes collected in eastern Venezuela), only a collection number and a provisional name (no description) were entered. Given the detailed nature of collection localities published in HUMBOLDT & al. (1816–1825), it seems clear that Kunth made full use of the information contained in the “Journal Botanique”, perhaps amended or even edited by Humboldt himself. Very few drawings are included in the “Journal Botanique”, but the few present were mostly done by Humboldt, who was deeply impressed by the morphology of orchid flowers and who chose to make very detailed sketches of some of them. As indicated by Humboldt in his diary of the journey (Faak, 2000), specimens were not always described in the “Journal Botanique” when they were first encountered in the field, but sometimes descriptions were added much later (e.g., 200 plants collected in the regions of Cocollar, Caripe, and Catuaro were described out of chronological sequence).

Three of the seven volumes comprising the “Journal Botanique” (MS 1332, 1333, and 1334) include notes pertaining to Venezuelan plants. One volume (MS 1332) contains descriptions of 690 plants; numbers 1–16 describe plants collected in the Canary Islands, some of them from cultivation in the botanical garden of Orotava (Fig. 5A), 17–572 plants collected from “Province Nouvelle Andalousie au Cumaná”, 573 a plant collected in “Province Caracas au Venezuela”, and 574–690 plants mostly collected in localities in eastern Venezuela (e.g., Bordones, Cumaná, and Caripe) (Fig. 5B) except for a handful collected during the journey from Cumaná to Caracas. This is the only volume of the “Journal Botanique” that terminates with a synoptic classification of all the plants cited in the volume. This classification (following Tournefort's system) includes also the number of collection (from page 184), and an alphabetic index to all taxa cited (from page 192). We assume that Bonpland did not continue producing this type of synoptic classification for the rest of the volumes as it entailed an extremely large amount of work. Those interested in more details on the plants collected by Bonpland and Humboldt in Cumaná can get additional information on 78 species that Humboldt described for this region in his field diaries (Tagebuch I, Bl. 37V-41R, cited by Faak, 2000).

A second volume (MS 1333) includes numbers 691–1215 that concern collections made in north-central Venezuela (e.g., Caracas, Aragua, and Lago Valencia), the llanos (e.g., La Villa), present-day Amazonas state (e.g., Atures, Yavita, Rio Negro, Casiquiare, and La Esmeralda), and the lower Orinoco (e.g., Angostura, now Bolívar state). On the inner side of the front cover of the manuscript, Humboldt included a detailed list of all the palms observed to date in their travels, proposing scientific and recording common names.

A third volume (MS 1334) covers numbers 1216–1591 and includes the last collections made in Venezuela (i.e., Cumaná, Bordones) and the first collections made during Humboldt and Bonpland's initial visit to Cuba. The lack of specific localities associated with these collections makes it difficult to clearly determine which collections were the last ones made in Venezuela and which were the first made in Cuba. The last collection unequivocally referring to a Venezuelan locality (i.e., Bordones) is number 1235 and the first collection clearly indicating a Cuban locality (i.e., Havana) is number 1254. Numbers 1236–1253 apparently were assigned to a series of plants that were collected in several central and southern Venezuelan localities, but for reasons unknown only described in a very late stage of the journey. This is the case for number 1244, which was assigned to a specimen of Lecythis ollaria L. (Lecythidaceae) collected in the Venezuelan llanos (i.e., Villa de Cura, Calabozo), a locality that Humboldt and Bonpland had visited long before their arrival in Cuba. As stated by Humboldt in his field diaries, the two explorers had to wait for two months for the ship that would carry them to Cuba and they used this time to arrange the large number of specimens they had collected in Venezuela (Faak, 2000).

Humboldt wrote Willdenow from Havana on 21 February 1801 (Moheit, 1993: 122 [letter nr. 41]) and pointed out that a copy of the manuscripts associated with the Venezuelan plants, specifically two volumes containing descriptions of 1400 species, had been made and was being shipped via the French vice-consul in Havana to Michel-Simon Goujaud Bonpland in La Rochelle. The survival of these manuscript notes was a major preoccupation of Humboldt and Bonpland, who not only copied “in triplo”, at least, the notes gathered in the early stages of their travels ( Faak, 2000), but also preserved them in a “waterproof” chest during the stormy passage from Barcelona (Venezuela) to Havana. We were unable to find any copies of these manuscripts and the one ostensibly deposited in La Rochelle could not be found in the archives of the Muséum where most botanical documents owned by the Bonpland brothers are thought to have been deposited. Nor were copies found in the Médiath èque Michel Crépeau, La Rochelle and their absence in the catalogue of historical documents of this institution (Musset, 1889) suggests that they may never have been deposited there. The fate of these copies of the “Journal Botanique” remains a mystery; however, one cannot completely exclude the possibility that a copy survives in one of the private collections consulted by Hamy (1906), all of which contained ample amounts of correspondence by Aimé Bonpland.

Curious, we also began to explore the extent to which Kunth made use of the descriptions produced by Bonpland. Thus for a few species of Venezuelan monocotyledon we compared the description proposed by Bonpland in the “Journal Botanique ” to the one published by Kunth in Humboldt & al. (1816–1825 ). In the cases of Tillandsia trichoides Kunth (Bromeliaceae) (Bonpland 347) and Schoenus tenuifolius Kunth (Cyperaceae) (Bonpland 542) the descriptions proposed by Bonpland in the “Journal Botanique” consisted of only one line whereas the descriptions published by Kunth in Humboldt & al. (1816–1825) were much more detailed. In contrast, the description of Smilax cumanensis Willd. (Smilacaceae) (Bonpland 285) (Fig. 10) prepared in the field by Bonpland is much more complete than the one provided by Kunth in Humboldt & al. (1816–1825). However, as this species had already been described by Willdenow in 1806 Kunth probably felt no need to elaborate on its description and chose to concentrate on describing unnamed taxa. In the case of Aiphanes praga Kunth (Arecaceae) (Bonpland 341) the amount of information presented by Bonpland and Kunth is more or less equal. These preliminary comparisons reveal no obvious pattern as to how the field note information was utilized and we are left to conclude that ascription of authorship for plant species described from collections made during the expedition probably should be assigned on a case by case basis.