Appendix. Philosophy of animal classification by Georges Cuvier
p. 456-473
Note de l’éditeur
This appendix is a translation of Cuvier’s final chapter of volume 1 of the Histoire naturelle des poissons
Texte intégral

Georges Cuvier
Portrait from a painting by Mathieu-Ignace Van Brée (oil on canvas, 1798).
Cliché Bibliothèque centrale, MNHN
1The differences in external and internal organs appropriate for characterizing fishes are not only obvious but also numerous; in fact, there are few classes of animals in which it is as easy to recognize natural genera and families and to classify the species.1 Upon the slightest inspection, anyone can perceive the relationships that connect the herrings (harengs), for example, to the shads (aloses), anchovies (anchois), tarpons (mégalopes), ten-pounders (élopes), and wolf herrings (chirocentres) and those that connect the freshwater eels (anguilles) to the moray eels (murènes), swamp eels (synbranches), and snake eels (cécilies). No less striking are the similarities among the innumerable tribes of minnows (cyprins), those of the catfishes (silures), the salmon (salmones), and the mackerels (scombres) and their kin. But to arrange these genera and families in some sort of order, it would be necessary to fix on a small number of important characters, which would result in a few large divisions that, without breaking up natural relationships, would be precise enough to leave no doubt about the place of each species. And this is what no one has yet been able to do in a sufficiently detailed manner.
2For example, the numerous unique characters of the Chondropterygi, or fishes with cartilaginous skeletons (or to speak more precisely, those with a grained periosteum), are too obvious not to have been noticed by all methodical minds. Therefore, all ichthyologists have placed these fishes in an order apart, but nearly all of them have altered the correctness of this division by including fishes that are similar only in the softness of the skeleton. However, the latter fishes ought not to be thrown back indiscriminately into the crowd. There are indeed some, such as the anglerfish (baudroie) and the lumpfish (lump), that, except for this softness, are no different from ordinary fishes and cannot always be separated from them; but there are also those that have particular characters of integuments, teeth, and especially the disposition of the skeleton in the head. The pufferfishes (tétrodons), the porcupine fishes (diodons), the boxfishes (coffres), and even the triggerfishes (balistes) are of this number, as are the pipefishes and their allies (syngnathes), whose gills have distinctive characters of great importance. The remarkable external appearance of these genera has caused many naturalists to separate them from the others, but generally we have not been very fortunate in discovering their true characters.
3Artedi, for example, not only included them [tetraodontiform fishes, i.e., Balistes and Ostracion] with the anglerfishes (baudroies) and the lumpfishes (lumps) in the order Branchiostegi, but also based that whole order on a false supposition, that these fishes do not have rays in the branchial membrane,2 whereas they all have them, and Artedi himself described those of the lumpfish (lump).3
4Linnaeus, in the tenth edition of his Systema naturae, after placing the Chondropterygii with the reptiles, included, according to a scheme just as ill founded, the anglerfishes (baudroies) with the Chondropterygii. After including with Artedi’s Branchiostegi the elephantfishes (mormyres) and pipefishes (syngnathes) and giving them all the characters of lacking not only branchiostegal rays but also opercula, which for several of these fishes is contrary to the simplest observation, he united in his twelfth edition the Chondropterygii and the Branchiostegi into one order of reptiles (Amphibia nantes), based on the character, again entirely opposite to the truth, of possessing both gills and lungs.
5Gmelin reestablished Artedi’s two orders but still attributed to the Branchiostegi the absence of branchiostegal rays.4 Goüan characterized them only as having incomplete gills, a description that is vague and highly contestable in nearly all the genera.5 Pennant reunited them with the Chondropterygii under the common name of cartilaginous fishes,6 a denomination adopted by Lacepède,7 the inappropriateness of which we have already seen. In fact, it is not good in either a positive or a negative sense. It cannot in any way be said that the skeleton of the triggerfish (balistes) is cartilaginous, and among the number of fishes that Pennant and others following him have included among the bony forms there are some, such as the leptocephalus (leptocéphale), that have hardly a sign of a skeleton.
6And so, first, I have had to select from among these somewhat anomalous fishes those that deviate enough from the ordinary fish type to merit being separated from them, and then to find in them distinct characters capable of being clearly explained in words.
7This examination has convinced me that it was wrong to remove from the great mass of ordinary fishes the anglerfishes (baudroies), lumpfishes (lumps), snipefishes (centrisques), elephantfishes (mormyres), and scabbardfishes (macrorhynques),8 which are essentially no different from ordinary fishes. But I have recognized that the pipefishes and their allies (syngnathes), having such singular form and constitution, could be distinguished by their gills shaped like tufts, hidden under an opercle having only a small opening near the nape to let water out; and that the porcupinefishes (diodons), pufferfishes (tétrodons), boxfishes (coffres), and triggerfishes (balistes), apart from whatever incompleteness there is in the skeleton, and the strangeness of their shape, have jaws and a generally complete skeleton in the head arranged slightly differently than in the ordinary fish; that the upper jaw and their palatal bones are articulated between themselves and with the vomer by means of immobile joints, which allow them much less freedom for opening and closing the mouth. Probably also related to this circumstance is the limited movement allowed their branchial apparatus by the skin that closely covers it and has kept some naturalists from observing that it is furnished with opercles and branchiostegal rays, as in all fishes.

Porcupinefishes
Genus Diodon. Engraving from Cuvier (Georges), “Sur les diodons, vulgairement orbes-épineux”, Mémoires du Museum d’Histoire naturelle, vol. 4, plate 6.
Cliché Bibliothèque centrale, MNHN

The skeleton of a triggerfish
Engraving from Cuvier’s Leçons d’anatomie comparée (1800- 1805), vol. 5, plate 5.
Cliché Bibliothèque centrale, MNHN
8But once these families are separated, there remain the nine-tenths of fishes among which the first observed distinction is that of fishes with soft fins, or with rays that are bifurcated and segmented, and fishes with spiny fins, of which some rays are small pointed bones without bifurcation or segmentation, or as Artedi has called the two types of fishes, the Malacopterygii and the Acanthopterygii. Unfortunately, this division is still very general, but even to apply it one must make exception for the first rays of the dorsal fin or the pectoral fins in certain minnows (cyprins) and catfishes (silures), which have strong, solid spines, although it is true that the spines in these genera are formed by the agglutination of a multitude of small joints, the traces of which can be seen.
9And there are other exceptions, or at least seemingly so, in certain fishes of the family of wrasses (labres) and of blennies (blennies), whose spines are so small or weak and few in number that they do not appear to have any; but except for these small irregularities, if this division does not lead far enough, at least it does not mislead and does not disunite fishes that nature has brought together.
10The same cannot be said of the distinctions that naturalists have sought to establish on other principles, or of the subdivisions that those who have adopted the overall division according to spines have attempted to introduce into the two branches. Thus, the general shape of the body and the absence of pelvic fins used by Ray, characters that for him took precedence over the presence or absence of spines, obliged him to group together the freshwater eel (anguille), the freshwater burbot (lote) and the goby (gobie), the pipefish (syngnathe), the swordfish (xiphias), and the mola or ocean sunfish (poisson lune).9
11Linnaeus, in his tenth edition, ignoring the distinction based on spines, was the first to conceive of dividing ordinary fishes into Apodes, Jugulares, Thoracici, and Abdominales according to whether they lacked pelvic fins or had the pelvics attached in front of the pectorals, under the pectorals, or farther behind [see Table 8]. Thus he considered himself obliged to put the swordfish (xiphias), the cutlassfish (trichiure), and the butterfish (stromatée) together with the freshwater eel (anguille) and the knifefish (gymnote), to place the codfishes (gades) between the weevers (vives) and blennies (blennies), the flounders (pleuronectes) between the dories (zeus) and butterflyfishes (chétodons), and the surgeonfishes or rabbitfishes (teuthis or amphacanthes) between the sheatfishes (silures) and the suckermouth armored catfishes (loricaires).
12Goüan, in combining the two methods and dividing each of Artedi’s branches according to Linnaeus’s four orders, avoids some unnatural groupings, yet he still places the swordfish (xiphias) and the cutlassfish (trichiure) far from the mackerels (scombres); he also commits positive errors in making the cuskeel (donzelle) and the catfish (silure) acanthopterygians and the butterfish (stromateus) a malacopterygian.10
13Lacepède takes up Pennant’s characters and divides fishes into cartilaginous and bony forms [see Table 10]. Each of these branches is subdivided, without regard for spines, according to the presence or absence of the opercle and the branchiostegal membrane. The final subdivisions are based on the relative positions of the pelvic and pectoral fins, a quite regular distribution that gives thirty-two a priori orders, but fifteen of them cannot be filled because the fishes relating to them have not been found in nature, and some seem to have been filled only because of the erroneous belief that the opercle or membrane is lacking in fishes that in fact have them, such as the elephantfishes (mormyres), moray eels (murènes), and swamp eels (synbranches).

An ocean sunfish, Mola mola
Engraving from Everhard Horne’s Leclures on Comparative Anatomy (1814-1828), vol. 6, plate 50.
Cliché Bibliothèque centrale, MNHN

Scombroid fishes
The classification of Duméril from his Zoologie analytique (1806).
Cliché Bibliothèque centrale, MNHN
14This classification, apart from the displacement of the anglerfishes (baudroies) and the lumpfishes (lumps), and the mixing of malacopterygians with acanthopterygians that also occurred in Linnaeus’s classification, would have the disadvantage of placing the moray eels (murènes) and the swamp eels (synbranches) far from the freshwater eels (anguilles) that resemble them so strongly, if, as regards this detail of its distribution, the classification were not founded, as we have just said, on characters that have no real existence. Nonetheless, Duméril has retained these orders in his own classification, which is basically that of Lacepède, subdivided according to body shape and other details, in order to bring together as much as possible, the natural families; but the introduction of characters taken from the pelvic fins prevents the achieving of this goal [see Table 11]. Thus one sees the anglerfishes (baudroies) with the triggerfishes (balistes) and chimaeras (chimères), and the codfishes (gades) with the weevers (vives) and the stargazers (uranoscopes). A single family [the péroptères] unites the snake eels (cécities and ophisures) and swamp eels (monoptères), which are [better placed next to the] freshwater eels (anguilles); the featherback (notoptère), which is really a herring (hareng); the cutlassfishes (trichiures), which are related to the mackerels (scombres), and so on.
15These causes have led Rafinesque and Risso to similar conclusions in the combinations they attempted to make of Pennant’s and Lacepède’s classifications, either between the methods or with natural families [compare Tables 12, 14, and 18]. I do not see that one has been any more fortunate in these attempts in Germany. Goldfuss, making no other changes to Linnaeus’s division than to combine the Jugulares with the Thoracici and the Branchiostegi with the Chondropterygii, has deprived himself of any way of arranging the families in the order of their similarities [see Table 17]. The lumpfishes (cycloptères) and the anglerfishes (baudroies) will never go, as he has placed them, between the lampreys (lamproies) and sharks (squales); one will never be able rationally to place, as he did, the cutlassfish (trichiure) with the freshwater eels (anguilles) and far from the scabbardfish (lepidopus), which resembles it at almost every point; the longfin herring (gnathobolus), which is a true herring (hareng), can never remain with the butterfish (stromateus), which is almost a butterflyfish (chétodon). The author himself [Cuvier] thought he was obliged to break his own rule for the swordfish (xiphias), which he leaves with the mackerels (scombres) among the Subbrachiens, although it is most assuredly a member of the Apodes.11

A stargazer
Ichthyscopus inermis. Illustration from Cuvier and Valenciennes, Histoire naturelle des poissons (1828-1849), vol. 3, plate 65.
Cliché Bibliothèque centrale, MNHN
16Oken had more freedom in arranging his families, because he gave to his grand orders — his fish fishes, amphibian fishes, bird fishes, and mammal fishes— almost indeterminate characteristics [see Tables 19-23]; and yet, because he used the position of the pelvic fins in his subdivisions, one sees him placing the herrings (clupées) between the mullets (muges) and the rabbitfishes (amphacanthes, buro), the codfishes (gades) near the sticklebacks (gasterostes), and the swordfish (xiphias) near the wolffishes (anarrhiques), while leaving the spiny eels (rhinchobdelles) and ribbonfishes (bogmares) in the same family as the freshwater eels (anguilles).
17After studying fishes for almost forty years, not according to the authors but the fishes themselves, their skeletons and internal organs, and after dissecting several hundred species, I am convinced of the necessity of never mixing an acanthopterygian with fishes of other families; I have learned that the acanthopterygians, which make up three-fourths of all known fishes, are also the type that nature has been the most careful with, the type that she has kept most nearly the same in all the variations in detail to which she has submitted it.
18All other characters should be used only in subordination to this one and without ever contradicting it, but the extreme constancy of the general plan and the predominant influence of this regulating character has made it very difficult to give to fishes in which it exists precise and perceptible applications of subordinate characters; therefore, the different families of acanthopterygians thus overlap one another, so that it is not known where one begins and the other ends. For example, the Percidae (famille des perches), which is distinguished from the Sciaenidae (sciènes) by its palatal teeth, is composed of a rather large but otherwise entirely natural group, part possessing teeth and part not.

The skipjack, Katsuwonus pelamis
After Cuvier and Valenciennes, Histoire naturelle des poissons (1828-1849), vol. 8, plate 214.
Cliché Bibliothèque centrale, MNHN

The yellow perch, Perca flavescens
After Cuvier and Valenciennes, Histoire naturelle des poissons (1828-1849), vol. 2, plate 9.
Cliché Bibliothèque centrale, MNHN
19The same happens in the otherwise well-characterized family of fishes with protected cheeks [mailed-cheek or scorpaeniform fishes]: most of its genera are related to the perches (perches), and the rest to the drums (sciènes) because of the palatal teeth.
20There are observable overlappings of some genera of the Sciaenidae (famille des sciènes) with those of the Chaetodontidae (chétodons) on account of the scales that more or less cover their vertical fins; on the other hand, one is obliged to bring together the Sparidae (famille des spares) and several genera of the Sciaenidae (sciènes) that have no trace of these scales.
21Overlappings no less noticeable combine certain genera of porgies (spares), such as [members of the centracanthid genus] Spicara (picarels) and the mojarras (gerres), with other genera such as the ponyfishes (équules), which cannot be separated from the dory (zeus), which leads to the Scombridae (famille des scombres), which in turn blends by such imperceptible nuances into those fishes that are extremely elongate and narrow, which we call ribbonfishes (taenioïdes), that it is almost impossible to say where one can draw the line separating the two.

The raccoon butterflyfish, Chætodon lunula
After Cuvier and Valenciennes, Histoire naturelle des poissons (1828-1849), vol. 7, plate 173.
Cliché Bibliothèque centrale, MNHN
22What recourse remains for naturalists desiring to describe organisms according to their true relationships except to acknowledge that the acanthopterygian fishes — which make up the old genera of perches (perches), drums (sciènes), porgies (spares), butterflyfishes (chétodons), dories (zeus), and mackerels (scombres), including the bandfishes (cépoles) and other ribbon-shaped fishes— compose, despite their innumerable species, one single natural family, in which nuances can be defined and incipient groups or slight differences can be perceived but in which it is impossible to determine perfectly clear boundaries that do not overlap each other at any point?
23It is not entirely the same with the anglerfishes (baudroies), the toadfishes (batrachus), the gobies (gobies), the blennies (blennies), and especially the wrasses (labres). Their characters are precise enough and, though in part internal, easy enough to assign and discern: in the first of these groups [the anglerfishes], the small restricted gill opening and the pectoral fins, with the base lengthened in the shape of an arm; in the second group [the toadfishes], similar pectoral fins joined to pelvic fins that consist of three rays; the flexible dorsal-fin spines in the third and fourth groups [the gobies and blennies]; the fleshy lips of the fifth [the wrasses]; the total absence of cecal appendages in nearly all these groups. While the characters mentioned above separate them from the other acanthopterygians, the last character relates them to the catfishes (silures) and minnows (cyprins), families that begin the order of malacopterygians; the malacopterygians themselves, as we have said, are related to the acanthopterygians by the spiny form of some of their rays.
24The families of the malacopterygians offer more differences and traits that are easier to recognize, and several of them are as natural as if they were subject to fixed limits, so much does each one of them, while clearly separate from the others, keep internally a great resemblance in details. This fixedness is so perceptible that most of the natural families we shall establish for this part of the class have already been discerned by Artedi and presented under the names of genera. His catfishes (silures), minnows (cyprins), salmons (saumons), herrings (clupées), and pikes (brochets) can remain together: it is not even inappropriate to distribute them according to the presence and position of the pelvic fins, for this character, slight as it is, does not vary in any of them. But I have found it impossible to preserve the distinction of the Jugulares, the Thoracici, and the Abdominales in the terms used by Linnaeus.12 In fact it is of little importance whether the pelvic fins are situated outside, slightly forward, a little behind the pectoral, or exactly below it; the important circumstance, based on the very structure of the fish, is knowing whether the pelvis is attached to the bones of the shoulder or simply suspended in the flesh of the belly. Thus I have coined the term “Subbrachien” to designate fishes of the first category, regardless of the location of their pelvic fins, which depends only upon the greater or shorter length of the bones of the pelvis; and I have left the name Abdominales for those of the second category. The Apodes are naturally found to be malacopterygians without pelvic fins.

A bagrid catfish
Genus Rita. Engraving from Cuvier and Valenciennes, Histoire naturelle des poissons (1828-1849), vol. 15, plate 429.
Cliché Bibliothèque centrale, MNHN
25We shall thus begin this natural history of fishes with the acanthopterygians, which actually constitute scarcely more than one immense family. We shall place after them the diverse families of the malacopterygians in the order in which they appear to us to be the most closely related to the acanthopterygians; but I would not wish it to be believed that they are related along only one line and in a single series. Although the abdominal malacopterygians can be arranged thus, and can even begin with those among them having spiny rays, neither the Apodes nor the Subbrachiens are to come next after them.
26The codfishes (gades), for example, are related as closely as any member of the Abdominales to certain acanthopterygians, and there would be no reason to place them after the Abdominales if one wishes to indicate their rank in nature. If we speak in terms of a series, it is only because facts put into books can be expressed only one after the other.
27The same observation applies to the other fishes, those having a fixed upper jaw, those with puff-shaped gills, and especially to the large and important family of Chondropterygii, with which we shall end this natural history.13
28The vanity of these systems that tend to arrange organisms in a straight line is easily seen in the Chondropterygii. Several of these genera —for example, those of the rays (raies) and sharks (squales)— are egregiously complex in some of their sense organs and in their organs of generation and are more developed in some of their parts than are birds. Other genera, which one reaches by obvious stages —for example, the lampreys (lamproies) and the ammocoetes (ammocètes)— are on the other hand so simplified that one might be authorized to consider them an overlap of fishes with segmented worms (vers articulés); most certainly the ammocoetes (ammocètes) at least do not have a skeleton, and their whole muscular apparatus has only tendinous or membranous supports.
29Therefore, let it not be thought that, because we shall be placing one genus or family before another, we actually consider it superior or more nearly perfect in the system of organisms. Only he can make that claim who pursues the wildly fanciful project of ranking organisms in a straight line, and this is an approach we renounced long ago. The more progress we make in our study of nature, the more we are convinced that this is one of the most erroneous ideas that has ever been held in natural history, and the more we have recognized the need to consider each organism or each group of organisms by itself, according to the role it plays on account of its characteristic qualities and its organization, and not to exclude any of its relationships, any of the ties that bind it to organisms nearest to and farthest from it.
30Once the naturalist adopts this point of view, the difficulties vanish and things fall into place for him. Our present systematic classifications see only the nearest relationships: they only wish to place one organism between two others, and they are constantly at fault. The true classification sees each organism in the midst of the others; it shows all the radiations or lineages by which organisms are linked more or less closely in this immense network that makes up organized nature; and only the true classification gives us great and true ideas of nature, worthy of this nature and its author —but ten or twenty radiating lineages would not often suffice to explain these innumerable relationships.
31Therefore, we inform our readers, once and for all, that it is in the descriptions themselves, which we shall give, that one must seek the proper idea of the degrees of organization, and not in the positions we are obliged to assign to the species.14 And yet we are far from claiming that relationships do not exist, that no classification is possible, and that one must not form groups of species and define them.
32Buffon was entirely correct when he asserted that absolute characters and clear distinctions between genera do not always exist, that there is no way of lining them up without coercion in our methodical framework; but this great man went too far when he rejected all relationships, when he objected to any ordering based on the similarities of organisms.15
33These relationships are so real, our mind is drawn to them by a propensity so necessary, that ordinary folk as well as naturalists have always had their genera.
34We shall therefore bring together what nature brings together, without forcing organisms into our groupings if nature has not placed them there; and we shall not hesitate —after describing all the species that may be arranged into a well-defined genus and all the genera that may compose a well-circumscribed family— to omit one or more isolated species or one or more genera not connected to the others in a natural way. We prefer a frank recognition of these sorts of so-called irregularities rather than to introduce errors by placing these anomalous species and genera in a series that does not embrace their characters.
35Our list of fishes, based on these principles, can be distributed into families approximately as indicated in the following table [see Table 24]. Unable to assign to each family one unequivocal and exclusive character, we indicate them for the moment by names derived from the most widely known genus of each, the genus that may be regarded as the type from which it is easiest to get an idea of the family. At the beginning of each family will be found a more extensive list of its characters, as well as the combinations according to which we are subdividing the family and that lead us to the different genera that compose the family.
Notes de bas de page
1 [As a basis for a comparative anatomy of all fishes, Cuvier ( “Livre deuxième. Idée générale de la nature et de l’organisation des poissons,” in Cuvier (Georges) & Valenciennes (Achille), Histoire naturelle des poissons, Paris; Strasbourg: F. G. Levrault, 1828, vol. 1, pp. 288-551.) provides an anatomical description of the European perch (Perca fluviatilis Linnaeus, 1758), “Idée générale de la nature et de l’organisation des poissons” (General idea of the nature and organization of fishes) that includes chapters on external morphology, osteology, myology, nervous system, sense organs, organs of digestion (from mastication to excretion, including the circulatory and respiratory systems), and reproductive system.]
2 At least this is how one explains the definition Artedi [1738] gives for the Branchiostegi [in his Genera piscium]: “branchiis ossibus destitutis” [opposite p. 1]; “branchiis, uti praecedentes, nulla ossicula gerunt” (p. 85).
3 [Artedi 1738], Genera piscium (p. 61): “Membrana Branchiostega ossicula sex gracilia et teretia utrinque continet.”
4 [Johann Friedrich Gmelin (see chap. 10, note 15), in the thirteenth edition of Linnaeus’s Systema naturae, 1789, vol. 1, pt. 3, pp. 1126-1516 (for a full citation of Gmelin’s edition of the Systema naturae, see Linnaeus, 1788-1793).]
5 [Goüan (see chap. 8, note 39), in his Historia piscium, 1770.]
6 [Pennant (see chap. 8, note 38), in his British Zoology, vol. 3, 1769.]
7 [Poissons cartilagineux, Lacepède; see, e.g., his première table méthodique, in the Histoire naturelle des poissons, 1798-1803, vol. 1, a large foldout preceding p. 1.]
8 The Macrorhinque argenté of Lacepède (1798-1803, vol. 2, p. 76) or Cheval marin argenté (Syngnathe argenté) of Bonnaterre (1788, p. 32), is none other than a scabbardfish (lépidope) incompletely described by Osbeck (1771, p. 107).
9 [Ray (see chapter 6, note 1), in his Synopsis methodica piscium, 1713.]
10 [Goüan (see chap. 8, note 39), Historia piscium, 1770.]
11 [Despite the absence of the pelvic girdle and fins in the swordfish (Xiphias), Cuvier places this fish together with the billfishes (Tétraptures, Makaira, Voiliers) in a tribe of their own ( “Tribu espadons, ou Scombéroïdes à museau en forme de dard ou d’épée”), among other scombroids; see Cuvier (Georges), “Tribu espadons, ou Scombéroïdes a museau en forme de dard ou d’épée,” in Cuvier (Georges) & Valenciennes (Achille), Histoire naturelle des poissons, Paris; Strasbourg: F. G. Levrault, 1831, vol. 8, p. 255.]
12 Those who have combined the Thoracici and the Jugulares have done so only on the basis of my Règne animal [see Cuvier (Georges), Le Règne animal distribué d’après son organisation, pour servir de base à l’histoire naturelle des animaux et d’introduction à l’anatomie comparée, vol. 2: Les Reptiles, les poissons, les mollusques et les annélides, Paris: Deterville, 1816, XIX + [1] + 653 p.]
13 [Cuvier intended to describe all known fishes in the Histoire naturelle des poissons, but a number of important groups were left out, including the Chondropterygii or chondrichthyan fishes. Before he died in 1832 he left instructions for Valenciennes (see chap. 13, note 10) to continue the work, leaving to him all the appropriate notes, manuscripts, and library necessary for the job (Monod (Théodore), “Achille Valenciennes et l’Histoire naturelle des poissons,” Mémoires de l’Institut Français d’Afrique Noire, vol. 68, 1963, p. 25). But, as pointed out by a number of authors (e.g., Günther (Albert Charles Lewis Gotthilf), An introduction to the study of fishes, Edinburgh: Adam & Charles Black, 1880, p. 18; Jordan (David Starr), “The history of ichthyology, an address by David Starr Jordan,” Proceedings of the American Association for the Advancement of Science, vol. 51, 1902, p. 440; Monod (Théodore), “Achille Valenciennes…,” art. cit., p. 25), Valenciennes did not complete the work as originally conceived. For reasons unknown, publication ceased after volume 22, the last volume appearing in December 1849 (Bailey (Reeve M.), “The authorship and names proposed in Cuvier and Valenciennes’ Histoire naturelle des poissons,” Copeia, 1951 no. 3, pp. 249-251), even though Valenciennes lived for another sixteen years. In addition to groups then unknown to science —deep-sea fishes, fishes of the Antarctic, Lake Baikal, the Great Lakes of Africa, and so on— there are some obvious omissions, including the “Ganoides,” the Chondrichthyes, Anguilliformes, Gadiformes, Pleuronectiformes, and Tetraodontiformes. Substantial evidence that additional volumes were planned has survived. Valenciennes himself seems to have considered the twenty-two volumes a first series and believed that a second series dealing with the taxa not yet covered was to follow (Valenciennes (Achille), “Avertissement,” in Cuvier (Georges) & Valenciennes (Achille) (eds), Histoire naturelle des poissons, Paris; Strasbourg: Levrault, 1849, vol. 22, p. vi; Monod (Théodore), “Achille Valenciennes…,” art. cit., p. 25). For more on this topic, see Pietsch (Theodore W.), “The manuscript materials for the Histoire Naturelle des Poissons, 1828-1849: Sources for understanding the fishes described by Cuvier and Valenciennes,” Archives of Natural History, vol. 12, no. 1, 1985, pp. 59-61; Bauchot (Marie-Louise), Daget (Jacques) & Bauchot (Roland), “L’ichtyologie en France au début du xixe siècle: L’Histoire Naturelle des Poissons de Cuvier et Valenciennes,” Bulletin du Muséum national d’Histoire naturelle, 4e série, section A, Zoologie, Biologie et Écologie animales, Paris, suppl. no. 1, 1990, pp. 7-17].
14 I [Cuvier] make this observation because one author [whom we have not been able to identify] thought himself clever in noticing that the lampreys (lamproies) are not near the reptiles, and that consequently I had been wrong to place them immediately after this class in my Règne animal [Cuvier, 1816, vol. 2]; but because this time I shall place them at the other end of the class of fishes, one should not conclude that I wish to place them after all the others.
15 [Buffon’s ideas about plant and animal relationships are developed in the first discourse (Premier discours) of vol. 1 of his Histoire naturelle, générale et particulière, 1749-1804.]
Le texte seul est utilisable sous licence Licence OpenEdition Books. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.
Michel-Eugène Chevreul
Un savant, des couleurs !
Georges Roque, Bernard Bodo et Françoise Viénot (dir.)
1997
Le Muséum au premier siècle de son histoire
Claude Blanckaert, Claudine Cohen, Pietro Corsi et al. (dir.)
1997
Le Jardin d’utopie
L’Histoire naturelle en France de l’Ancien Régime à la Révolution
Emma C. Spary Claude Dabbak (trad.)
2005
Dans l’épaisseur du temps
Archéologues et géologues inventent la préhistoire
Arnaud Hurel et Noël Coye (dir.)
2011