HUNTIA A Journal of Botanical History

Volume 16 Number 1 2017

Hunt Institute for Botanical Documentation Carnegie Mellon University

Pittsburgh The Hunt Institute for Botanical Documentation, a research division of Carnegie Mellon University, specializes in the history of and all aspects of science and serves the international scientific community through research and documentation. To this end, the Institute acquires and maintains authoritative collections of books, plant images, manuscripts, portraits and data files, and provides publications and other modes of information service. The Institute meets the reference needs of botanists, biologists, historians, conservationists, librarians, bibliographers and the public at large, especially those concerned with any aspect of the North American flora.

Huntia publishes articles on all aspects of the , including exploration, art, literature, biography, iconography and bibliography. The journal is published irregularly in one or more numbers per volume of approximately 200 pages by the Hunt Institute for Botanical Documentation. External contributions to Huntia are welcomed. Page charges have been eliminated. All manuscripts are subject to external peer review. Before submitting manuscripts for consideration, please review the “Guidelines for Contributors” on our Web site. Direct editorial correspondence to the Editor. Send books for announcement or review to the Book Reviews and Announcements Editor. All issues are available as PDFs on our Web site. Hunt Institute Associates may elect to receive Huntia as a benefit of membership; contact the Institute for more information.

Hunt Institute for Botanical Documentation Carnegie Mellon University 5th Floor, Hunt Library 4909 Frew Street Pittsburgh, PA 15213-3890 Telephone: 412-268-2434 Email: [email protected] Web site: http://www.huntbotanical.org

Editor and layout Scarlett T. Townsend Editor, Emeritus robert W. Kiger Book Reviews and Announcements Editor Charlotte A. Tancin Associate Editors Donald W. Brown Lugene B. Bruno T. D. Jacobsen J. Dustin Williams Photographer Frank A. Reynolds

Printed and bound by RR Donnelley, Hoechstetter Plant, Pittsburgh, Pennsylvania

© 2017 Hunt Institute for Botanical Documentation All Rights Reserved

ISSN 0073-4071 Contents Gouan and Guérin: Professor and student Roger L. Williams 5–15

Augustin Augier’s Botanical Tree: Transcripts and translations of two unknown sources Nils Petter Hellström, Gilles André and Marc Philippe 17–38

Candolle’s Law of Analogies, a savant as useful citizen Roger L. Williams 39–50

Edwin B. Payson, 1893–1927 Roger L. Williams 51–60

Book Reviews and Announcements 61–74

Huntia 16(1) 2017 Candolle’s Law of Analogies, a savant as useful citizen

Roger L. Williams

Abstract Antoine-Laurent suggested in 1774 who developed it; and to Antoine-Laurent de that conforming in their external characters Jussieu who subjected it to fixed laws” A( .-P. (the natural families) might also possess conforming de Candolle 1804, p. 6). medicinal virtues. Thus, once the natural order had been established, which he was at the moment preparing In the beginning, there was Linnaeus to publish, it followed that one should be able to (1707–1778; Fig. 2), remembered mainly for determine a plant’s medicinal virtues by its exterior his invention of the sexual system of plant signs. His disciple, Augustin-Pyramus de Candolle, classification and of the completion of binomial when revising Lamarck’s Flore Françoise, also studied for a medical degree and designed his medical dissertation plant nomenclature, still in use today. As to test that Jussiean prediction. He soon found he could all botanists in the 18th century received describe new natural families by segregating them from their botanical training as herbalists while those of Jussieu, if based upon medicinal distinctions. in medical school, it would be legitimate to He defined the phenomenon as the Law ofA nalogies. Some of the well-known Candollean plant families were wonder whether their medicinal knowledge of described in the dissertation, illustrating the rationale plants affected their later endeavors to develop behind his prolific later publications. At the end of the a natural system of plant classification. Candollean era, Jussieu’s 100 natural plant families had Linnaeus, for financial support, had been a been expanded to 200. practicing physician in his initial professional years. In the list of dissertations later directed Augustin-Pyramus de Candolle (1778– by Linnaeus, the Amoenitates Academicae, there 1841; Fig. 1) was a descendant of an old noble is only one direct connection with the materia Provençal family, which, becoming Protestant medica, Johan Lindwall’s dissertation no. 171, in the mid-16th century, expatriated to published in 1772: Observationes in Materiam Geneva. He moved to in 1798, then the Medicam, long after Linnaeus’ major works. preeminent center of botanical research, to Many others, however, included medical obtain the medical degree, which, in those references. Moreover, as Arthur Cain pointed days, was still the prerequisite for botanical out, Linnaeus was also a religious man and study. Botanists today are familiar with a reader of the Bible. He had been greatly Candolle’s abbreviation, DC; but the influence impressed by the Hebrew regulations for of medicine on his is little perceived. health, to be found in the book of Leviticus, His doctoral dissertation was dedicated to the references that included such topics as Tranquil founding botanists of the theory of natural mind, Insane love, Eating of permitted families: “To Tournefort who anticipated it; to meats, Sleep, Healthiness of exercise and who proved it; to Adanson Infectious air. His medical interest, in sum, was preventive medicine, not botanical 1701 South 17th Street, Laramie, WY 82070 classification (Cain 1993). USA 39 40 Huntia 16(1) 2017

Figure 1. Augustin-Pyramus de Candolle (1778–1841), Figure 2. Carolus Linnaeus (1707–1778), engraving photographic reproduction, 13 × 10 cm, after an by H. Meyer, 50.8 × 34.29 cm, after an engraving by oil painting, ca.1813, by Boilly, Hunt Institute for Ogburne with design by Bartolozzi after an original by Botanical Documentation Archives portrait no. 1. Magnus Hållman for Robert John Thornton (1768– 1837), New Illustration of the Sexual System of Carolus von Linnaeus … (London, 1807, 3 parts), Hunt Institute for Botanical Documentation Archives portrait no. 10. Having always recognized that his sexual methods and in defining genera had convinced system was artificial, if providing a key to plant him that there were no perfectly constant identification to be universally applicable, characters. That is, characters useful in Linnaeus regarded it to be only a substitute establishing one genus did not necessarily have for a natural method. His beginning of a the same collective value for another genus. He more natural classification, Methodi Naturalis could not, as a consequence, complete a natural Fragmenta, appeared in Philosophia Botanica in system based on all characters, seemingly 1751, 65 “natural orders” followed by a list of unable to depart from his initial belief that over 700 genera, based not upon reproductive definitive characters for natural orders could function, but upon manner of growth and be found only in reproductive systems: an appearance: habitus. essentialism (Larson 1967). Many such groupings are still recognized In 1759 Bernard de Jussieu (1699–1777; Fig. today. Orchideae, for example, included 3) planted the first demonstration of the natural , , , , Neottia, classifications in the royal garden at the Petit , Cypripedium, Epidendrum, Limodorum Trianon, dividing all plants into 15 classes and Arethusa, all in today. This based upon the only invariable characters he method could also lead to erroneous groups as had found, namely, the number or absence Linnaeus recognized. His experience with his of cotyledons; whether a plant was apetalous, Williams: Candolle’s Law of Analogies 41 monopetalous or polypetalous; the insertion of the stamens or corollas; and whether hypogynous, perigynous or epigynous, all distinctions still recognized today. The 15th class, Diclines irregulares, comprised conifers whose irregularity was being unisexual (B. de Jussieu 1789). As Bernard de Jussieu did not publish the description of his planting at the Petit Trianon, delayed until publication by his nephew in 1789, Dr. Louis Gérard’s (1733–1819) Flora Gallo-Provincialis in 1761, covering Provence alone, became the first published flora to employ a natural method. While it bore considerable resemblance to the planting of 1759, the conventional reference that Gérard simply borrowed Bernard de Jussieu’s plan is untenable. Both men worked not only independently but also were obviously concerned to perfect Linnaeus’ Methodi Naturalis Fragmenta. While their arrangement Figure 3. Bernard de Jussieu (1699–1777), engraving of plant families differed in significant ways, by Ph. Langlois after an original by A. Guilleminot, Hunt Institute for Botanical Documentation Archives their systematic concepts made them much portrait no. 8. closer to each other than either one was to (1727–1806) in 1763, discussed below. Gérard began with the acotyledons, proceeded to the classifying the medicinal virtues of plants. and concluded with the dicotyledons. In his The classes he cited were essentially those organization he did not recognize any groups of Pierre Chomel (1671–1740), the standard superior to the families, simply classifying herbal since its first publication in 1715, according to natural affinities as he perceived going through repeated augmented editions them. until 1803 (Chomel 1739). Adanson meant to It is evident that both Bernard de Jussieu illustrate the point that none of such classes, and Gérard grouped plants initially on the arranged by reputed medicinal virtues, was basis of cotyledon, presence and number. natural. He added that the defects of Chomel’s They both apparently held ovary position system, moreover, were three: (1) a given plant to be of greater significance than corolla may not be limited to a single virtue, but often structure in the further segregation of families. have two or three; (2) different parts of a plant An explanation of Bernard’s method had to may not all have the same virtues; (3) virtues wait for his nephew, Antoine-Laurent de associated with plants of a particular family Jussieu (1748–1836; Fig. 4), in 1774 and 1789 are not usually of equal strength throughout (Williams 1988). that family (Adanson 1763, 1:lxxv–lxxviii). In 1763 Michel Adanson, a student of The recommended solution to these defects Bernard de Jussieu, published a method for he published in a “Table raisonné des virtues 42 Huntia 16(1) 2017

themselves, whose two extremes are the smallest herb and the tallest tree. By an insensible gradation, one will rise from one to the other by arranging in succession those whose affinity is marked by a greater number of relations. That order, which is that of Nature, interests not only natural philosophers, but offers a more actual utility. Reasoning, supported by experimentation, demonstrates that plants conforming in their external characters also enjoy the same properties (virtues). Once the natural order is given, it follows that one should be able to determine their virtues by exterior signs. An objective of such importance should merit occupying the attention of botanists who mean to combine the title of Savant to that of Useful Citizen (A.-L. de Jussieu 1774, p. 175).

He continued by describing the principal invariable characters that defined the new natural method as had been planted by Bernard de Jussieu at the Petit Trianon in 1759, but without mentioning Bernard’s name. He Figure 4. Antoine-Laurent de Jussieu (1748–1836), subsequently developed those principles into engraving by Meno Haas, 1798, 21.59 × 12.7 cm, Hunt Institute for Botanical Documentation Archives his own Genera Plantarum. The flowering portrait no. 2. plants were divided into 15 classes, subdivided into natural orders, that is, plant families, et usages des plantes,” arranged alphabetically describing a chain of being from the most and practically, all that he provided and original to the most advanced. Class XV, the without the rationale that Candolle meant to Diclines irregulares, contained unisexual families provide in 1804 (Adanson 1763, 2:621–640). he regarded either as unnatural or as a possible Antoine-Laurent de Jussieu, before advancement beyond bisexuality in Jussieu’s publishing his natural orders in 1789, had opinion (A.-L. de Jussieu 1789, p. lxii). described them before the Académie des It is curious that A.-P. de Candolle, a Sciences in 1774 as then being planted in the disciple of Jussieu, owed his professional Jardin du Roi in Paris: advancement to Jean-Baptiste Lamarck (1744–1829; Fig. 5), who had long resisted Among the sciences, which have the study of the possibility of a natural classification, as natural bodies as their objective, Botany, whose evident in the first edition of his Flore Françoise, extent is the most considerable, requires for that reason more method in the disposition 3 volumes, in 1778. A second edition had of the beings submitted for its examination. been issued in 1795, virtually unchanged. His The numerous authors who have successively decision to entrust a third edition to Candolle occupied themselves with that work amounted to a tacit recognition that it must demonstrate its necessity. The deficiency of their researches also demonstrates the difficulty be converted into a natural classification, and in finding the natural method. he properly assigned its royalties to Candolle. The plants spread over the globe must Candolle later praised Lamarck’s use of the appear to form a continual chain among Williams: Candolle’s Law of Analogies 43 dichotomous key as a useful artificial method to find the correct name of a plant, but added “to know the name of an object is not to know it. Only a natural classification can reach that goal” (A.-P. de Candolle 1813b, pp. 49–52). His revision of the flora began in 1802, and the first four volumes were completed and published in 1805; during which time he also completed the work for the medical doctorate (Stafleu 1967, pp. 64–65). Candolle’s medical dissertation was designed to test the Jussiean prediction that all plants in a natural family would be found to possess similar medicinal virtues. The further implication was that a plant that was alien in virtue from the family or genus in which it had been placed truly belonged in another family. Menyanthes L., for example, initially placed in Primulaceae, was surprising in its febrifuge properties. Etienne-Pierre Figure 5. Jean-Baptiste Lamarck (1744–1829), stipple Ventenat (1757–1808) had demonstrated engraving by Hopwood with design by S. Porter after that it belonged in the Gentianaceae where an original by Jacques Louis David for Robert John Thornton (1768–1837), New Illustration of the Sexual that virtue was characteristic. As Candolle System of Carolus von Linnaeus … (London, 1807, 3 noted, for additional exactitude, a botanist parts), Hunt Institute for Botanical Documentation must discover which parts of a plant actually Archives portrait no. 2. possess the alleged medicinal virtue. Jussieu had shown that in Euphorbia L. and several and calcareous terrains produced notable other plants, the endosperm is mild while differences in their chemical compositions the embryo is bitter and strongly purgative. and their nutritive qualities. That observation Within Umbelliferae, you do not compare the implied that the virtues of plants could be roots of the carrot with the leaves of hemlock. altered by differences in soils. A greater Rather, as the seeds of umbels all have vessels knowledge of the chemical composition of filled with an essential oil, it can be shown that plants was essential (A.-P. de Candolle 1804, the seeds are aromatic and stimulating in all pp. 29–32). of them (A.-P. de Candolle 1804, pp. 25–28). Candolle then turned to an examination The next step in gaining medical precision, of the general properties of the natural Candolle noted, would be to determine to families as then known. He soon found that what degree the nature of the terrain affects he could describe additional natural families the virtues of plants. Although the manner by segregating them from those of Jussieu, if in which the soil influences vegetation was as based upon medicinal distinctions: yet imperfectly known, Theodore de Saussure (1767–1845) had recently shown that the Liliaceae Juss. Lily Family effects were more significant than previously Candolle noted that the lack of uniformity suspected. Plant species raised on granitic in the Liliaceae derives from the inclusion 44 Huntia 16(1) 2017 of the genus Colchicum L., autumn crocus. proposed, Cleomaceae DC (A.-P de Candolle Whereas the bulbs of the Liliaceae may 1804, p. 110). [He did not publish it later.] contain several distinct principles that can be determined chemically, Colchicum bulbs Simaroubaceae DC. Quassia Family contain a single poisonous virtue [colchicine] Both Quassia simarouba L. and Quassia amara prescribed as a sedative for gout and arthritis. L. are stomatics and tonics for dysentery. [This Consequently, Candolle segregated it as new family was segregated from Rutaceae Colchincaceae DC (Lamarck and Candolle Juss., a family omitted by Candolle because 1805, 3:192). His segregation did not stand of its multitude of conflicting medications (A.-P de Candolle 1804, p. 57). traditionally recommended.] The bitter bark of quassias was prescribed only for dysentery Globulariaceae DC. Globularia Family (A.-P. de Candolle 1804, p. 118). All the plants in Globularia L. have remarkably bitter stalks and leaves. They are DC. Stonecrop or Orpine Family strong purges but give tone to the stomach and Sedum acre L., the biting yellow stonecrop, intestines. [This new family was segregated is used to wash the gums of scurvy victims in from Scrophulariaceae Juss., whose plants were a concoction of its leaves. erectus DC. believed to be useful against scrofula (Lamarck is a calming and refreshing plant that can be and Candolle 1805, 3:427).] Given the lack of used externally on inflammations and burns analogy in Scrophulariaceae, he recommended (A.-P. de Candolle 1804, p. 123). [This new additional study of all its genera to verify their family name replaced Sempervivaceae Juss., medicinal virtues (A.-P. de Candolle 1804, the type genus becoming Crassula L.] pp. 76, 81). Grossulariaceae DC. Gooseberry Family Papaveraceae Juss. Poppy Family Grossularia L. and Ribes L. have distinctive The white sap is the critical analogous factor fruits that are sedative and refreshing (A.-P. in poppies. Argemone mexicana L. is used as a de Candolle 1804, p. 123). [This new family sedative in America (A.-P. de Candolle 1804, was segregated from Saxifragaceae Juss. on p. 107.) [He did not mention Papaver L.] the basis of its fruits. If he were influenced by their ovary position, inferior, superior in Fumariaceae DC. Fumitory Family Saxifragaceae, he did not say. Published in There is no white milky sap in this new Lamarck and Candolle (1805, 4(2)[=5]:405).] family, differing from Papaveraceae in that In his conclusion to the 53 families he character and in all its general medicinal described in his dissertation, Candolle properties. Fumitories act as aperitifs and explained the method he used in testing his promote sweating (A.-P de Candolle 1804, Law of Analogies. By assigning each natural p. 110). family surveyed for medicinal properties a number from 0 to 4 (zero meaning the Capparidaceae Juss. Caper Family properties were either nil or unknown); and The properties in this family are similar a plus or minus sign to indicate whether the to those in the Cruciferae, which they also concordance between structure and property approach in structure. Capers are stimulants, was strong or contrary to the Law of Analogies, aperitifs and anti-scorbutic, but Cleome he developed a list of 108 plant families, each icosandra L., from Ceylon, if applied to the skin, with a numerical value. The results, while produces inflammation. Thus, a new family is far from proving a perfect analogy between Williams: Candolle’s Law of Analogies 45 form and property, nevertheless suggested to the contemporary prescriptions of herbal Candolle that the Law of Analogies had some remedies. Anyone seeking an understanding validity in 87 of them but did not apply at all of the efficacy of herbal medications in the in only 7 of them. 18th century must recognize that, beyond the He was evidently emboldened by his absence of experimental tests of their efficacy, confidence that some of the natural families, as an inquiry complicated by the known fact recognized in 1804, would require segregation that apothecaries and herbalists, who sold the in the future; namely, that some had conflicting medications to physicians, had a long history medicinal virtues, thus were insufficiently of dubious practices. natural. He concluded with two general The corporation of apothecaries in propositions: (1) Inside every genus, the same had been in business since the 13th century, parts or their related juices feature similar not only selling medications and herbs but also medicinal properties; (2) Inside every natural enjoying a monopoly on sugar, the ingredient family, the same parts or their related juices of many medications. Peasant women, illiterate feature analogous properties. (He called them and unlearned, gathered the herbage for the two laws.) While he added eight possible similarly unlearned apothecaries, conspiring reasons for perceived exceptions to those two against the longevity of their contemporaries. laws, he was sufficiently confident in their The tag, folkloric medicine, has become a validity to believe that future research in prettifying euphemism (Christian 1904, pp. medicine, botany and chemistry would resolve 39–44). There is also reason to believe that such seeming exceptions and bring them into some folk healers may have been deviants, conformity with the Law of Analogies (A.-P. sharing in some degree delusions of grandeur de Candolle 1804, pp. 145–147). and persecution, attracting them to the At that moment, however, his renown irregular practice of medicine and accounting depended not upon the reputation of his for why quacks are always with us (Ramsey medical dissertation but upon the popularity 1988, pp. 292–293). of the third edition of the Flore Françoise, four Before Candolle’s arrival in , volumes published in 1805, the substantial Dr. Paul-Joseph Barthez (1734–1806) had royalties directed to him by Lamarck. That been the major figure in medical botany. He popularity led him to stand for election to became a professor of medicine in 1761, and the Institut in 1806. The seat went instead to he replaced Antoine Gouan (1733–1821) as a highly reputable field botanist, François- principal botanist in 1773 when the latter’s Joseph Palisot de Beauvois (1752–1820), greatly eyesight failed. Barthez’s primary interests irritating Candolle. That accounts for his were plant physiology and the medicinal decision to turn his back on Paris and to go properties of plants. He was known for to Montpellier, where he was appointed to be his lessons emphasizing the irritability and professor of botany on the faculty of medicine sensitivity revealed by the leaves of certain on 15 January 1808 and installed on 18 April plants and by the sexual organs of many others. 1808 (Stafleu 1967, pp. 64–65). As a physiologist, he also taught that it was The appointment was especially appropriate unknown how drugs operated on the body; for a botanist concerned about the medicinal and, worse, there was little real knowledge virtues of plants, as the reputable medical about which drugs worked and which did not. and botanical institutions in Montpellier Consequently, he called for the entire medical had recently become focused on doubt about community to build what we would call a 46 Huntia 16(1) 2017 database that would make medicine something anti-epileptic plants as vague and imprecise; more than a game of chance. suggesting that lack of precision had always For reasons unknown, Barthez was plagued been a boon to charlatanism, providing by the jealousy of the chancellor, Imbert, learned verbiage for practitioners “who had who obstructed his attempts to improve the not received the first principles of a liberal botanical garden, neglected by a series of his education” (Roques 1807–1808, 1:v–vii). predecessors. He was glad for the opportunity Candolle, in sum, found a helpful colleague to become physician to Louis-Philippe, duc at Montpellier in 1808, as well as a benevolent d’Orléans. With the Revolution in 1789, he patron in Chaptal, who provided 17,000 removed to Narbonne to establish a medical francs between 1808 and 1813 to restore the practice. When the faculties of medicine and integrity of the . Candolle science at Montpellier were reorganized in began immediately to have the garden of the 1794, his name was not on the list of professors. botanical school replanted according to the In 1801 Jean-Antoine Chaptal, the minister of natural families, replacing Gouan’s Linnaean the interior, who had held a chair in chemistry order. He persuaded the city to purchase the at Montpellier, rectified the omission, but adjoining private garden, a portion of which he Barthez never again taught there (Martins dedicated to the Ecole forestière, removing the 1854, pp. 41–42; Brockliss and Jones 1997, trees from the Ecole botanique and planting pp. 437–441). them according to the natural order. He His mission, however, had been pursued also established a Conservatoire suitable for at Montpellier by Dr. Joseph Roques (1772– collections and engaged an artist, Toussaint- 1850). Roques set out to provide an alternative François Nodé-Véran (1773–1852), to make to the works of Chomel (1715) and Buchoz publishable drawings of the local plants. (1770), whose books had been standard in Part of Candolle’s time was consumed in the 18th century and were grossly inadequate teaching and directing doctoral dissertations. (Regnault 1774; Chomel 1739). Their works Of the seven dissertations he directed, only had been widely copied in France or quoted two were published (Biria 1811; Duval 1813). in numerous 18th-century floras by authors One can discern from their titles that Candolle ignorant of medicine, proving to Roques meant to give his students direct experience that the subject could be covered successfully with the imprecision and doubtful reliability only by physicians actively engaged in clinical of the materia medica of prior generations and to medicine. encourage additional research on the analogy While Roques was familiar with the between natural families and their medicinal methods of both Linnaeus and Jussieu, clearly virtues (Martins 1854, p. 52): respecting them, he chose an alphabetical Biria, Histoire naturelle et médicale des Renonculus, arrangement of plants by popular name, published 1811 aiming his book at both medical and popular Dufresne, Histoire naturelle et médicale des readers, noting thereafter each plant’s place Valérianées, 1811 in the systems of Tournefort, Linnaeus and Elmiger, Histoire naturelle et médicale des Jussieu. He rejected all plants traditionally Digitales, 1812 recommended, which his experience had not Dunal, Histoire naturelle et médicale des Solanum validated, adding others either ignored or little et genres voisins, published 1813 appreciated. Finally, he condemned traditional Viguier, Histoire naturelle des Pavots, 1814 categories such as cephalic, hepatic and Roubier, Dissertation sur l’Aloé-Pitte, 1816 Williams: Candolle’s Law of Analogies 47

Colladon, Histoire naturelle et médicale des Casses, I. Vascular plants or plants with cotyledons. 1816. 1. Endogens: Dicotyledons. a. With calyx and corolla: His first of six summer botanical field trips in Thalamifloralae (polypetalous hypogynous), France produced fifty plates, drawings of new Calycifloralae (polypetalous perigynous), plants previously not described or not drawn Corollifloralae (gamopetalous). (A.-P. de Candolle 1808). His subsequent b. Monochlamydeous plants (with a single floral envelope). catalogue of the Jardin Botanique contained 2. Endogens: Monocotyledons. 5,500 species, of which 300 were new or a. Phanerogams (true monocotyledons), poorly known (A.-P. de Candolle 1813a; b. Cryptogams (vascular cryptogams including Martins 1854, pp. 48–53). Naiadeae). The assertion made by Martins that II. Cellular plants or Acotyledons. Candolle’s residence in Montpellier was a a. Plants with leaves (Muscineae), period of reflection, a notion based upon his b. Plants without leaves (Thallophytae). publication of botanical theory in 1813, has This is the organizing system later featured in obscured his primary preoccupation with a Regni Vegetabilis Systema Naturale (1818–1821) complex of duties as a professor of medicine and subsequently in the Prodromus Systematis and botany. The Théorie Élémentaire he Naturalis Regni Vegetabilis he began publishing published in 1813 was largely a recapitulation in 1824. of those morphological relations deemed Candolle soon recognized what he called essential in establishing the principal characters two stumbling blocks preventing general for classification, those that remain unaltered acceptance of his natural system: the first, from even in instances of abnormal or monstrous those who persisted in the study of organized plant growth: the unalterable character life forms, either using analogies that were too of the cotyledons and the position of the abstract and general to be valid or metaphysical reproductive organs, as contrasted with the ideas that were uncertain and the second, relative variability of all other plant organs, from those who only saw isolated facts in the which were thus potentially artificial. structure of living forms and became simple Although in this respect he was a disciple of describers. He put the poet Goethe in the first Jussieu, as in his Law of Analogies, Candolle’s category for having emphasized the symmetry application of the principle of symmetry in the structure of plants while ignoring other (rapport) led him to reject in Théorie Élémentaire characters. In the second category, without Jussieu’s decision to limit the number of natural mentioning names, he cited several French families to 100 and to avoid putting more than naturalists who, even after the example of 100 genera in a single family, thus to avoid Jussieu in 1759 and Adanson in 1763, sought making natural classification too complicated. to establish groups or natural families on For Candolle, rapport alone determined rank. the basis of a simple intuitive knowledge of Future families and genera could be of an living beings, obviously Buffon after 1749 indeterminate size. Accordingly, he was also and Lamarck in 1778 (A.-P. de Candolle 1827, ready to recognize even monotypic families 1:vi–viii). (A.-P. de Candolle 1813b, pp. 78–79, 84). The An unrelated but significant stumbling main divisions of Candolle’s system follow block in the study and precise description below (A.-P. de Candolle 1813b, p. 138): of plants, not mentioned by Candolle, was 48 Huntia 16(1) 2017 cited by his son, Alphonse de Candolle bases, moreover, beginning about 1818 or (1806–1893), who undertook the completion 1820, permitted the dissection of plants with of the Prodromus after the first seven volumes both hands for the first time, accounting for after his father’s death: namely, the structure the great differences in floral dissections by of microscopes. Early microscopes have mostly authors between 1818 and 1840. Alphonse disappeared as they were personal rather than de Candolle believed that Karl Sigismund institutional property and fell to heirs for Kunth (1788–1850) was among the first to use disposal. Much of what is known about them a mounted lens, enabling him to work with has depended upon their display in engravings greater precision on the specimens Humboldt from the period. and Bonpland had brought back from Latin Use of concave and convex lenses was America after 1812, which were often of poor known since the 13th century, but their quality (A. de Candolle 1880, pp. 347–348). assembly into a magnifying instrument Thus, when Augustin-Pyramus de Candolle seems to date from Galileo early in the 17th decided to move to Geneva in 1816 to accept century. The theory for developing compound a chair of natural history created for him, he microscopes to obtain larger images was soon soon benefited from the improved microscopy known, but the technical capacity to produce to spend the rest of his life toward the superior lenses from ordinary glass lagged. completion of his natural system of botanical Those lenses produced such aberrations classification. His departure from Montpellier that the compound microscopes produced was greatly lamented by all who had been an image inferior to that given by simple his associates, but the more specialized loupes. Those microscopes were more objets professional opportunity in Geneva, even d’art than workable instruments, and even for a reduced salary, proved to be irresistible. Leeuwenhoek made his observations with His former student, Michel-Félix Dunal loupes. Improvements were made between (1789–1856), replaced him in Montpellier as 1780 and 1800, especially the adoption professor of botany. of achromatic lens. These were made by The growing size of Candolle’s personal combining lenses from different glasses having herbarium reflected the upsurge in plant different focal powers. The light emerging collections from all parts of the world after from such a combination forms an image the Napoleonic era. He started his collection virtually free from unwanted colors, but it did in 1794, and it was augmented after 1841 by not solve the problem of spherical aberration Alphonse, and in 1879 it comprised 291,965 (Crestois 1953, p. 71; Morton 1981, p. 366). specimens. About three-quarters of the Alphonse de Candolle attributed that species described by Augustin were based further improvement of the microscope to on specimens in his herbarium, according Giovanni Battista Amici (1786–1863) in 1823, to Stafleu, as were the greatest part of those astronomer and optician, who directed the published in the first volumes of the Prodromus. observatory in Florence. Having recognized The great size of the herbarium, if obviously that the good operation of a microscope beneficial, slowed his work on the Prodromus. depended upon the size of the aperture, Amici Herbarium records showed the sources of used a highly curved plano-convex front lenses duplicate specimens received from other to reduce substantially spherical aberrations. collectors, and their extent indicated a vast The new instruments were manufactured global correspondence (A. de Candolle 1880, in England. The use of lenses mounted on pp. 401–402; Stafleu 1967, p. 64). Williams: Candolle’s Law of Analogies 49

Candolle’s initial work in Geneva References included 12 natural families, beginning Adanson, M. 1763. Familles des Plantes. 2 parts. with Ranunculaceae and concluding with Paris: Chez Vincent. (Reprint, 1966, New York, Cruciferae (A.-P. de Candolle 1818–1821). J. Cramer, Lehre.) The work was then diverted into a larger Biria, J.-A.-J. 1811. Histoire Naturelle et Médicale des Renoncules, Précédée de Quelques Observations scheme for which he became celebrated: sur la Famille des Renonculacées. Montpellier: Prodromus Systematis Naturalis Regni Vegetabilis, Jean Martel. beginning with Resedaceae. He was able to Brockliss, L. and C. Jones. 1997. The Medical World complete seven volumes by 1839 before his of Early Modern France. Oxford: Clarendon Press. Cain, A. J. 1993. Linnaeus’s Ordines naturales. Arch. health failed, and he died in 1841. The work Nat. Hist. 20: 413. was then pursued by Alphonse de Candolle Candolle, A. de. 1880. La Phytographie, ou l’Art de for an additional ten volumes, before finishing Décrire les Végétaux Considérés sous Différents in 1873. Only the dicotyledons and the Points de Vue. Paris: G. Masson. Candolle, A.-P. de. Year XII [1804]. Essai sur les gymnosperms were included: 200 natural plant Propriétés Médical des Plantes, Comparées avec families, exactly double the number published Leurs Formes Extérieures et Leur Classification by Jussieu in 1789 (A.-P. de Candolle and A. de Naturelle. Paris: Méquingnon. Candolle 1824–1873). Candolle, A.-P. de. 1808. Icones Plantarum Galliae. Paris: H. Agasse. Because Candolle criticized the use of Candolle, A.-P. de. 1813a. Catalogus Plantarum Horti false analogies in 1827, had he ever come Botanici Monpeliensis. Montpellier: J. Martel. to suspect the legitimacy of his own Law of Candolle, A.-P. de. 1813b. Théorie Élémentaire de la Analogies? Not only had that law seemed Botanique. Paris: Déterville. Candolle, A.-P. de. 1818–1821. Regni Vegetabilis consistent with his conception of symmetry, Systema Naturale. 2 vols. Paris: Treuttel & Würtz. but also he retained in the Prodromus the five Candolle, A.-P. de. 1827. Organographie Végétale, ou dicotyledous families he had segregated in Description Raisonnée des Organes des Plantes. his medical dissertation in 1804. The sixth 2 vols. Paris: Chez Déterville. Candolle, A.-P. de and A. de Candolle. 1824–1873. family, Colchicaceae, a , Prodromus Systematis Naturalis Regni Vegetabilis. could not have been recognized, as Alphonse 17 vols. Paris: Treuttel & Würtz (for vols. 1–7, de Candolle stopped short of the monocots, 1824–1839); Paris: Masson (for vols. 8–17, but Bernard de Jussieu had earlier separated 1844–1873). Chomel, P.-J.-B. 1739. Abrégé de l’Histoire des Colchicum from Lilium in Ordines Naturales in Plantes Usuelles, ed. 5. 3 vols. Paris: Jacques Ludovici XV Horto Trianonensi Dispositi, anno Clousier. 1759, where and when this story began (A.-L Christian, A. 1904. Etudes sur le Paris d’Autrefois: de Jussieu 1789, pp. lxiii–lxiv). Les Médecins, l’Université. Paris: Roustan & Champion. Candolle’s Mémoires et Souvenirs were not Crestois, P. 1953. L’Enseignement de la Botanique published until 1862 by Alphonse in Geneva. au Jardin Royal des Plantes de Paris. Cahors: An insightful review of those recollections A. Coueslant. by Asa Gray (1810–1888) the following year, Duval, M.-F. 1813. Histoire Naturelle et Médicale du Solanum et Genres Voisins. Paris, Strasbourg, in the American Journal of Science and Arts, Montpellier. contained a hint of omissions that would make Gray, A. 1969. Augustin-Pyramus de Candolle. In: an inclusive biography difficult. Even with the A. Gray. 1969. Scientific Papers of Asa Gray. 2 vols. addition of numerous eulogies and obituary New York: Kraus. Vol. 2. Pp. 289–309. (Reprint, orig. ed. 1889, Boston and New York, Houghton notes, a final biography remains to be written Mifflin and Co.) (Stafleu 1967, p. 65; Gray 1969). Jussieu, A.-L. de. 1774. Exposition d’un nouvel order des plantes adopté dans les démonstrations du 50 Huntia 16(1) 2017

Jardin Royal. Hist. Acad. Roy. Sci. Mém. Math. Morton, A. G. 1981. History of Botanical Science. Phys. (Paris). Année 1774 (13 April): 175. London: Academic Press. Jussieu, A.-L. de. 1789. Genera Plantarum Secundum Ramsey, M. 1988. Professional and Popular Medicine Ordines Naturales Disposita, juxta Methodum in in France, 1770–1830: The Social World of Horto Regis Parisiensi Exaratum, Anno M. DCC. Medical Practice. Cambridge: Cambridge LXXIV. Paris: Herissant & Barrois. University Press. Jussieu, B. de. 1789. Ordines Naturales in Ludovici Regnault, N.-F. 1774. La Botanique Mise à la Porte de XV Horto Trianonensi Dispositi, Anno 1759. Tout le Monde ou Collection des Plantes d’Usage In: A.-L. de Jussieu. 1789. Genera Plantarum dans la Médecine, dans les Alimens et dans les Secundum Ordines Naturales Disposita, juxta Arts. 3 vols. Paris: Chez l’Auteur. Methodum in Horto Regis Parisiensi Exaratum, Roques, J. 1807–1808. Plantes Usuelles, Indigenès et Anno M. DCC. LXXIV. Paris: Herissant & Exotique, avec la Description de Leurs Charactères Barrois. Pp. lxiii–lxxi. Distinctifs et de Leurs Propriétés Médicales. 2 vols. Lamarck, J.-B. and A.-P. de Candolle. 1805. Flore Paris: Chez l’Auteur. Françoise, ed. 3. 4 vols. in 5. Paris: Desray. Stafleu, F. A. 1967. Adanson, Labillardière, De Larson, J. L. 1967. Linnaeus and the Natural Method. Candolle. Lehre: J. Cramer. Isis 58(193): 313–320. Williams, R. L. 1988. Gérard and Jaume: Two Martins, C.-F. 1854. Le Jardin des Plantes de neglected figures in the history of Jussiaean Montpellier. Montpellier: Chez Boehm. classification. Taxon 37(1): 9–10.