Authorship, dates of publication, Status and types of Spix and Agassiz's Brazilian fishes By M. Kottelat Abstract Spix & Agassiz's "Selecta genera et species piscium Brasiliensium" has been published in two fascicles, in June 1 829 and January 1831. The plates had been prepared in part under Spix's supervision and the text has been written by Agassiz, but Agassiz is the sole author for nomenclatural purposes. Some types are still extant (Table 3). Nomen- clatural problems related to names proposed in this book are d'iscussed: Hypopththalmus, Sorubim, Osteoglossum, Pacu and Anodus are authored by Cuvier, 1829, Pirarara Agassiz, 1829 has priority over Phractocephalus Agassiz, 1829, Osteoglossum vandellii Cuvier, 1829 has priority over O. bicirrhosum Cuvier, 1829, Pacu Agassiz, 1829 is a junior homonym but not a synonym oiPacu Cuvier, 1829, Pacu Cuvier, 1829 is an objective synonym of Curimata Bosc, 1817, Rhaphiodon Agassiz, 1829 has priority over Cynodon Agassiz, 1829, R. gibbus Agassiz, 1829 is type species oi Rhaphiodon. Introduction Louis^ Agassiz was born on 28 May 1807 at Motier (46°58' N 7°05' E) in the French-speaking part of Switzerland. Between 1817 and 1821 he studied at the Lycee de Bienne; from 1822 to 1824 at the Academie de Lausanne; from 1824 to 1826 at the University of Zürich (all in Switzerland); from 1826 to 1827 at the University of Heidelberg and from 1827 to 1830 at the University of München [Mu- nich], where he was awarded the degree of doctor of medicine in 1 830, having already obtained a doc- torofphilosophydegreeat Erlangen in 1829. From 1830 to 1831, he worked on fossil fishes in hisfa- ther's home in Concise (46°5 1 ' N 6°43 ' E) in Switzerland and from 1831 to 1 832 in Paris at the Jardin des Plantes under Cuvier. From 1 2 November 1 832 to November 1 84 1 he was prof essor at the College de Neuchätel (Switzerland), and from its foundation in October 1841 to March 1846 at the Academie de Neuchätel (the former name of the university). In 1 846, he left Neuchätel for Paris and London and arrived in Boston in October 1846. His work and its impact on new world natural sciences during bis life and long after his death on 14 December 1873 are well documented. The biographical details given here are taken from E. Agassiz (1885, 1887^) and Marcou (1896). Lurie (1960) provides additional data on Agassiz's life, but is of very limited use for his European period and is in places inaccurate for this period. In 1828, while Agassiz was still a Student, Carl Friedrich Philipp von Martins^''* (1794-1868) asked him to prepare a report on the fishes of Brazil. Martins had spent the years 1817—20 exploring this country with Johann Baptist von Spix (1 78 1 - 1 826), Spix being the zoologist and leader of the expedi- tion and Martins the botanist. They had already published several books on the results of this expedi- tion, but Spix died before the completion of the zoology (see Fittkau, 1983 a, b). Agassiz accepted the work and this resulted in the first monograph on Brazihan fishes, both marine and freshwater, since Marcgrave (1648); it contains descriptions of numerous new species and genera. The specimens coUected by Spix (except those subsequently given to Agassiz) were housed in Zoo- logische Staatssammlung München until their destruction during a British bombing raid on the night of 24-25 April 1944 (Terofal, 1983). The fish coUections had been packed and were in the entrance 69 of the buildlng of the Bayerische Akademie der Wissenschaften [Academy of Sciences of Bavaria]: they were to be moved to a safe place out of the town in the forthcoming days but the entire museum was destroyed and the whole fish coUection burned (Hoogmoed & Gruber, 1983: 321). At the request of C. Dufour and J.-P. Haenni, I undertook a revision of the fish coUections now housed in the Musee d'Histoire Naturelle de Neuchätel. As could be expected, types of several species described by Agassiz or received in exchange from contemporary scientists turned up (Kottelat, 1983, 1984). The coUection contains 83 jars of Brazilian fishes, some of them being types or putative types of the fishes described from the Spix coUection. These had been received by Agassiz from Mar- tins (E. Agassiz, 1887: 62, 72). The bulk of Agassiz's private coUection has been bought by the Musee d'Histoire NatureUe de Neuchätel in 1 834 (Marcou, 1 896 : 55 ; [Godet], 1 899 ; Godet, 1 907) and Agas- siz also left some specimens when he went to the New World. As a result of the discovery of these ty- pes, I became interested in the Status of some of the Brazilian fish names, although it soon turned out that several problems concerning the dates of publication, the authors and the nomenclatural Status of several names had not been definitively solved. Whitehead & Myers (1971) examined some of these problems; my conclusions Supplement, or occasionally disagree with, their opinions. Data which du- plicate those of Whitehead & Myers have been omitted, unless necessary for the argument. Material and Conventions I follow Whitehead & Myers (1971) in referring to Spix & Agassiz (1829-31) as "Brazilian fishes". The Inter- national Code of Zoological Nomenclature (1985 edition) is referred to as the Code. Documents referred to as AEN are "Archives de Louis Agassiz; Fonds de l'Institut de Geologie de l'Universite de Neuchätel, Suisse"; they have been catalogued by SuRDEZ (1974) and they can be consulted in the Archives de l'Etat, Chäteau, CH-2000 Neu- chätel, Svi^itzerland. Other abbreviations used are : MHNG Museum d'Histoire Naturelle, Geneve ; MHNN Musee d'Histoire Naturelle, Neuchätel ; SL Standard length ; TL total length ; ZSM Zoologische Staatssammlung München. Spix & Agassiz's Brazilian fishes Since a limited number of this book was printed and it has not been reprinted and is quite difficult to find (there is a microfiche edition), it is useful to give a description of its general appearance. It was distributed as an unbound series of sheets (28x36 cm), as foUows: a. Half title page: Selecta genera et species piscium Brasiliensium [Selected genera and species of Brazilian fishes]. This page forms a single sheet. b. Title page: Selecta / genera et species / piscium / quos / in itinere per Brasiliam / annis MDCCCXVIl- MDCCCXX / jussu et auspiciis / Maximiliani Josephi I. / Bavariae regis augustissimi peracto / collegit et pin- gendos curavit / Dr. J. B. de Spix, I ... I digessit, descripit et observationibus anatomicus illustravit / Dr. L. Agassiz, / praefatus est et edidit itineris socius / Dr. F. C. Ph. de Martius. / Monachii, / typis C. Wolf. / 1829 [Selected genera and species of fishes, as Dr. J. B. von Spix collected them and had them drawn during his jour- ney in Brazil in the years 1817—1820 by order and under protection of Maximilian Joseph 1., very august king of Bayern (Bavaria). Classification, descriptions and anatomical plates by Dr. L. Agassiz. Work prefaced and pubiished by travel companion Dr. F. C. Ph. von Martius. Printed in München (Munich) by C. Wolf. 1829]. This vi'ork was pubiished in München [in Latin Monaco] and not in [Principaute de] Monaco, as is occasionally given in some bibliographies. c. Dedication to Cuvier: Viro illustrissimo L. B. de Cuvier summo zoologo qui praeclaro ingenio totam naturam quae est quaeque fuit amplexus e terrae aequorique penetralibus uberiorem animalium historiam eduxit dispo- suit stabilivit hoc opus ab amico defuncto inchoatum D. D. D. L. Agassiz et C. F. Ph. de Martius [To the most illustrious Baron (L. B. = Le Baron) de Cuvier, the eminent zoologist who, thanks to his brilliant intelligence, embracing the vi^hole nature as it is and as it has been, extracted from the profoundness of the earth and the sea, arranged and consolidated an enriched history of animals, L. Agassiz and C. F. Ph. von Martius have dedicated this work which a defunct friend had begun]. Items b and c form a single sheet. 70 d. List of subscribers; 2 pages; 1 sheet. e. Preface, by Martius; page I-XVI; 4 sheets. f. Obituary of Spix, by Martius; pages I— II; 1 sheet. g. Conspectus (table of contents); pages 1 —4; 1 sheet. h. Explanations of anatomical plates; pages 5 — 6; 1 sheet. i. Text: 138 pages as follow: 20 sheets of 4 pages (1-80), 1 sheet of 2 pages (81-82), 11 sheets of 4 pages (83-136), 1 sheet of2 pages (137-138). j. 7 plates representing scenes of fisheries; lettered A— G. Plates A— D are signed "lith[ograph] v[on] Nep[omuk] v[on] Ott" and plates E— G are not signed. The copy in Neuchätel public library is uncoloured; the one in Basel university library is coloured. k. 84 hand-colour plates representing fishes, numbered 1 - 76, also 8 a, 1 3 a, 24 a, 50 a, 50 b, 56 a, 56 b, 69 a. 1. 6 anatomical plates, lettered A-F; plate B is signed "S[ebastian] Minsinger del W[ilhelm] Siegrist sc." m. One plate representing Spix. Dates of publication As Whitehead & Myers (1971) point out, an exact dating of the "Brazilian fishes" is essential because of conf Hcts with names in Cuviers's Regne animal (1829), which was published not later than 3 1 March 1829 (BoESEMAN, 1962). Document AEN 1 1 8/2.1 is a prospectus for the various books on Brazil pubhshed by Martius. This prospectus is in French on one side and German on the other. The French side bears the title "Ouvrage sur le Bresil public par Mr. le Dr. de Martius". Book 6 is given as: Spix et Agassiz, Pisces, pet. en fol. Prem. Cah. avec 53 planch. color. Pr. 129 Frcs. La fin de cet ouvrage va paraitre incesamment [sie] . . . [Spix and Agassiz, Pisces, small folio. First fascicle with 53 coloured plates, price 129 francs. The end of this work will appear at once . . .] On the reverse side is the same text in German, with the price 105 Fl. [Gulden (see Klimpert, 1896)]. This pros- pectus is not dated. It is clear that the book was pubhshed and distributed in more than one part. By chance, AEN 118/3 is an un- bound copy (unfortunately not complete) containing items a — c, and i (from page 1 to 1 14) still in the original wrap- per of the second fascicle, whose title is reproduced here as Figure 1. SPIX ET AGASSIZ PISCES BRASILIENSES , EOroiT MARTIUS. FASCICüLÜS DL ET ÜLTMÜS. CONTINET: Praefet. (4 fol. pag. HI-. XVI.) Meaior. Spixü, Conspect. et Tab. expKcat. (a fol. p. I. 11 et i— &) 'Text, (fo). 22—35- pag. 83— «38.) Icones: Piscatio: 3 Tab. E. F. G. .* Pisces 36 Tab, 46— -50. 60 a. 5o b. 5i — 56. 56 a. 56 b. 57 — 69. 69 a. 70 — 76, AnatomesSTab, ) A. B. (duplex) ad Fase, i. .) A. B. (duple \ C. D. E. F. tmn Spixü Monacbi! Jaav^ tssi. Fig. 1. Title of the original wrapper of fascicle 2; document AEN 118/3. 71 The Statement "Fasciculus II, et ultimus" [second and last fascicle] clearly indicates that only two fascicles are in- volved. Therefore, the following parts must be considered as published on 31 January 1831 [Code, art. 21 (c) (I)]: d-h, i pp. 83-138, j pls E-G, k pls 46-76, 50a, 50b, 56a, 56b, 69a, 1, m. The Statement concerning anatomical plates means that all six plates were published with the second fascicle, but that plates A and B correspond to text in fascicle one; plate B is folded (duplex). In a letter to Agassiz dated 2 March 1831, Martius wrote that a copy had been sent to Cuvier eight days earlier (document in Houghton Library, Har- vard University, Cambridge, MA, bMS Am 1419 [480]). All the other parts were thus published in fascicle one, including the title page, which bears the date 1829. In a letter dated 29 December 1828 to his sister Cecile (E. Agassiz, 1887: 61), Agassiz informed her that the first volume was completed, that the printing had already begun and that 40 (50 in E. Agassiz, 1 886: 46) coloured plates were already finished. In a letter dated 14 February 1829 to his father (E. Agassiz, 1887: 74) he wrote that the book would appear soon and that the already completed part had been presented on November 1 828 to a meeting of Ger- man naturalists and physicians. It is not clear what this "presented" means but such a presentation does not render any binominal name available. In a letter dated 22 May 1829 to his brother (E. Agassiz, 1887: 82) Agassiz informed him that he had received his doctor degree at Erlangen on 23 or 24 April 1829. We know (E. Agassiz, 1887: 82) that Martius ("publisher" of the book) wanted Agassiz to get this degree so that this title could be printed on the title page (as indeed it was); this probably in order to insure better sales. In a letter to his father dated 4 July 1 829 (E. Agassiz, 1 887: 82) Agassiz wrote that he had sent him a copy of the first fascicle. His father received it on 31 August 1 829 (E. Agassiz, 1887: 87). This book took a long time to get from München to Concise, as it had been sent by Martius to Marie-Philippe Mercier (1781 — 1831), a botanist in Geneve [Geneva] (see Briquet, 1940), with herbarium specimens, and then sent to the Agassizs. My efforts to trace Mer- cier's correspondence in the Conservatoire Botanique, Geneve and Archives de l'Etat, Geneve, were fruitless. Cu- vier had already received his copy and written to Agassiz on 3 August 1829 (E. Agassiz, 1887: 86). As we can see from Agassiz's correspondence, he laid great value on this book, his first important work. Since there is no mention in his letter of 22 May 1 829 to his brother of it having been published already (E. Agassiz, 1887: 82), we may conclude that it was not yet published. Thus the date of publi- cation must have been after 22 May but before 4 July 1829. This agrees with Whitehead & Myers' (1971) argument; they favoured the last week in June 1829. It is clear, therefore, that Cuvier's Regne animal has about two to three months priority over the "Brazilian fishes". A review of the first fascicle of "Brazilian fishes" appeared in July 1829 issue of Isis (Anon. [Oken ?], 1829). Authorship and availability of names As the title page indicates, Spix collected the specimens and supervised the execution of the dra- wings, whilst Agassiz arranged the species in systematic order, described them and prepared the ana- tomical plates. However, it seems that several plates were drawn by Joseph Dinkel under the supervi- sionof Agassiz (E. Agassiz, 1887: 70), since Agassiz wrote in a letter to Cuvier (which he probably ne- ver sent) (E. Agassiz, 1887: 82) that several of Spix's plates were inaccurate and that they had been re- drawn (E. Agassiz, 1887: 80). Whitehead & Myers (1971) supposed that Spix had written a part of the text and that Agassiz com- pleted it. In fact, it appears that Agassiz wrote the whole of the text. Several manuscript drafts of the book still exist, written by Agassiz partly in French, German and Latin; there are also some annota- tions by Karl Friedrich Schimper (1803 — 1867), then a good friend of Agassiz and who studied with him. Document AEN 118/1.1 is the manuscript of the conspectus and AEN 1 1 7 is the manuscript of the text. The latter includes several drafts of the first fascicle, some text having been written on sket- ches for the anatomical plates (Fig. 2). Very often, the text is quite different from the final version (Fig. 3). Earlier drafts of the second fascicle are not present, but for the whole book a clean and final draft is present, again in Agassiz's band. It does not differ from the published text. Whatever Spix wrote, it can only have been rough notes (although he named the species for the plates). 72 **4J •f "^ «^ <^;^' *- i2oi4*^ » i Fig. 3 A-C. Differentdraftsof thesametext;docunient AEN 117. - A. Draft of the same text; document AEN 117. j&u^^^i-uiAÄ*^ •&^f<4U/>-^<>^ "-L /y-iz-f f~-^ *>yC-t ^-*'- >v*4< -♦i,.,-^»«. ^C> «^^ #1*.. ^j,^oei^. »-'«'^^--^«^^-v.j^- i^^v^^i^ua^; ^^^ G^~-^4^ / , ^ - /i^'uUic*^- pd^-hui^ i^f^ajl^^J^ f^y^^^^r.i— iU^' Üc^£< tS~ »i«^' iM^Ut^HK.- , '2^^^*;*— ^lAjeU^' ^■ÜOL.fufY' AA-^iJt*t' ^ji.- . Ü-J 9h^ I. l^i<: tLl*-(-K*^<**^^*^ie^,-^'»t^ iSc^^yti/ «rvj^^WV -J^aA-y:«*/ a^* «-«^>iS'. '-*■«/«• Fig. 3 B. Draft of the same text; document AEN 117. (/«/^ 75 ^^^^.»mCx^*.-^^ C^^^ff^ '^fJ^^-^ ^»^'^'^»^T^ ^^-f^^ 't^4,.iKt^ivm^ ^) — — —. - |<5fcA^' ,f«^S*^<«^. ^^rV^**^»^ . ^ £i..^ U^^ c*-.*^ S^^c^ ^.^^-^..-H..:,^ -ÜL^aM^ ' Ofy--f' #. y Fig. 3C. Draft of the same text; document AEN 117. 76 from time to time that the author names are not mentioned to indicate "ownership" but as a tool for retrieval of Bibliographie data]. When different names are used in the text and on the plates, two different interpretations of the Code seem possible; by chance, the result is the same: — first interpretation is that both names are available, the one on the plate by indication, the one in the text by description; — second interpretation is that names used in the text are available; names used on the plates are listed as Synonyms in the text and, as Synonyms, are only available if they have been treated as available names before 1961. In that case, the author and date is Agassiz, 1829. In both cases, they are simultaneous Synonyms and the one to be retained is the one selected by the first reviser. According to the English text of the glossary of the Code (p. 264), the first reviser is the first author to subsequently cite names, etc. [The French glossary (p. 290) is not very explicit on the subsequent character of the first reviser, but it seems to be implicitly recognized]. Such pairs of names appear in the first fascicle only. For specific names, the first subsequent reviser is Agassiz himself, in the Conspectus [table of contents] distributed with the second fascicle in 1831. There Agassiz cited both names and chose one of them (cited first) to have precedence over the other(s) (cited in brackets), in accordance vi^ith the requirements of the Code [art. 24 (b)]. As could be expected, the names retained by Agassiz are always those that he had introduced in the text, the only exception being Osteoglossum bicirrhosum (see below). The generic names are discussed in the next chapter. For some taxa, the author is given as "Cuvier, in litt.", presumably because Cuvier had seen Spix's plates (E. Agassiz, 1887: 80) and had senthiscommentsto Martins. Lurie (1960) mentioned thesecom- ments and Whitehead & Myers (1971) pubUshed them. From these comments, it is clear that Cuvier did not provide descriptions and that Agassiz's descriptions have been prepared without Cuvier's help, although Agassiz did pay attention to the proposed names (or to the suggested suppression of some barbarisms, that is names derived from Brazilian or Indian names [it is worth remembering that for the Ancients, Barbars vi^ere foreigners, not obligatorily sauvage and cruel, and that barbarism thus means "in a foreign language", or in the particular case "non Latin"]). These "Cuvier" names are avai- lable in zoology in as much as Agassiz (in "Brazilian fishes") published them. In terms of zoological nomenclature, Agassiz is the author of these names since Cuvier was not the author of both the na- mes and the conditions vi^hich make them available. Cuvier (1829) used some of Spix's manuscript names in his "Regne animal" and some of them should have priority over Agassiz's ones since this book appeared at the latest date on 31 March 1829 (Boeseman, 1962) while "Brazilian fishes" appeared at the earHest after 22 May 1829. Whitehead & Myers (1971) have shown that Spix's names as used by Cuvier are nomina nuda, since they are without description, or are simply not available, and the citation of an unpubHshed Illustration does not qualify as an indication. This is only partly correct; as shown below, Hypophthalmus, Sorubim, Osteoglos- sum, Anodus and Pacu are available from Cuvier (1829). As the dates of publication are known, one can also decide the priority of names appearing in Cuvier & Valenciennes's Histoire naturelle des poissons prior to January 1831 (see Bailey, 1951, for authorship and dates of publication of this work). Several of Spix and Martins' new zoological and botanical names are available (by description or in- dication) from the narrative of their expedition (Spix & Martius, 1823, 1828, 1831). I checked the three volumes and found that fishes are not involved : several are described, but none with a latinized name. An additional source of confusion is the custom at that time of indicating the author of the combi- nation and not of the name. Agassiz, like others, paid attention to authors of names whereas the cur- rent Code and practice concern names of taxa. For example, Prochilodus argenteus Agassiz (p. 63) does not mean that Agassiz is the author of argenteus, but that he considered himself as the author of a new combination of Pacu argenteus of Spix. The Status of all these names, as well as that of a few generic names of American and African fishes proposed by Agassiz in footnotes have been examined. New generic names proposed in "Brazilian 77 fishes" are listed on Table 1 ; the type species of each has been traced. All specific names used in the text are listed in Table 2 in order of their appearance in the text. Short comments on nomenclature are given as footnotes. Major nomenclatural problems encountered and/or solved are dealt with in the next chapter below. Table 1. New generic names in Spix & Agassiz's "Brazilian fishes" with their type species. Five of Spix's names available from Cuvier's Regne animal are also included. Abbreviations: S: Spix; A: Agassiz. When two different names have been used in the text and on the plates, the one retained in the conspectus is listed first. Genus AcanthicHS A, in S & A, 1829 Rhinelepis A, in S & A, 1829 Glanis A, in S & A, 1829^ Ceratorhynchus A, in S & A, 1829*' Cetopsis A, inS&A, 1829 Centrochir A, in S & A, 1829 Hypophthalmus Cuvier, 1829 Pirarara A, in S & A, 1829 Phractocephalus A, in S & A, 1829 Sorubim Cuvier, 1829 Platystoma A, in S & A, 1829 Osteoglossum Cuvier, 1829 Ischnosoma Cuvier, 1829 Glossodus A, in S & A, 1829' Anodus Cuvier, 1829 Pacu Cuvier, 1829 Prochilodus A, in S & A, 1829 PacH A, inS&A, 1829 Leporinus A, in S & A, 1829 Schizodon A, in S & A, 1829 Rhaphiodon A, in S & A, 1829 Cynodon A, in S & A, 1829 Xiphorhynchus A, in S & A, 1829*'' Salminus A, in S & A, 1829" Xiphostoma A, in S & A, 1829" Micropteryx A, in S & A, 1831'^ Cormger A, in S & A, 1831 Pachyurus A, in S & A, 1831 Type species A. hystrix A, in S & A, 1829 R.aspera A, in S & A, 1829 Silurus hagre Linnaeus, 1766 Silurus militaris Linnaeus, 1758 Silurus coecutiens Lichtenstein, 1819 Doras crocodili Humboldt, in Humboldt & Valenciennes, 1821^ H. edentatHS A, in S & A, 1829 Pirarara bicolor A, in S & A, 1829 Silurus lima Bloch, in Schneider, 1801 Osteoglossum vandellii Cuvier, 1829 G. forskalii A, in S & A, 1829 A. elongatus A, in S & A, 1829 Salmo edentulus Bloch, 1794 P. argenteus A, in S & A, 1829 L. novemfasciatus A, in S & A, 1829 S. fasciatus A, in S & A, 1829 R.gibbus A, inS&A, 1829 Salmo falcatus Bloch, 1794 Hydrocyon brevidens Cuvier, 1819 X. cuvieri A, in S & A, 1829 Caranx dumerili Risso, 1810 C. spinosus A, in S & A, 1831 P. squamipennis A, in S & A, 1831 Designation by monotypy by monotypy present designation original designation Bleeker, 1862: 16^ by monotypy Bleeker, 1862: 15 by monotypy Bleeker, 1862: 10 by monotypy by monotypy Eigenmann & Eigenmann, 1899 present designation Eigenmann, 1918:424 Eigenmann, 1910: 426'° || Eigenmann, 1910:425" Campos, 1945:472 by monotypy by monotypy Eigenmann, 1910:446'^ Jordan & Gilbert, 1883 by monotypy by monotypy Nomenclature Hypophthalmus Cuvier, 1829 Hypophthalmus is usually considered as authored by Agassiz, in Spix & Agassiz, 1829. However, Cuvier (1829: 293) described a group of catfishes: "Quelques uns se fönt remarquer par une tete deprimee; des yeux places tres bas sur ses cötes, et une adipeuse extremement petite; ils ressemblent beaucoup aux Schubes (4)." 78 Table 2. Names and autorship of species described in Spix & Agassiz's "Brazilian fishes". Abbreviations: A: Agassiz, S: Spix. When different new specific names are used in the text and on the plates, the one retained in the conspectus is listed first. If the same specific name is used with different generic names, the name used on the plate is hsted in brackets. Acanthicus hystrix A, in S & A, 1829 Rhinelepis aspera A, in S & A, 1829 Loricaria rostrata A, in S & A, 1829 HypoStoma etentaculatum A, in S & A, 1829 Cetopsis coecutiens Lichtenstein, 1819 Cetopsis [SÜHrtis] candiru S & A, 1829 Doras humboldti A, in S & A, 1829 Corydoras edentatas A, in S & A, 1829 Hypophthalmus edentatus A, in S & A, 1829 Hypophthalmus nuchalis A, in S & A, 1829 Pimelodus spixii A, in S & A, 1829 Pimelodus albidus A, in S & A, 1829 Pimelodus rigidus A, in S & A, 1829 Pimelodus pirinampu A, in S & A, 1829 Pimelodus ctenodus A, in S & A, 1829 Pirarara [Phractocephalus] bicolor A, in S & A, 1829 Sorubim infraoculare A, in S & A, 1829" Sorubim planiceps A, in S & A, 1829 Platy Stoma pirauaca A, in S & A, 1829 Platystoma spatula A, in S & A, 1829 Sorubim jandia A, in S & A, 1829 Platystoma corruscans A, in S & A, 1829 Sorubim carapary A, in S & A, 1829 Platystoma truncatum A, in S & A, 1829 Hetero brauch US sextentaculatus A, in S & A, 1829 Sudis pirarucu A, in S & A, 1829^' Erythrinus salvus A, in S & A, 1829 Erythrinus unitaeniatus A, in S & A, 1829^^ Erythrinus macrodon A, in S & A, 1829 Erythrinus trahira A, in S & A, 1829 Erythrinus microcephalus A, in S & A, 1829 Erythrinus brasiliensis A, in S & A, 1829 Osteoglossum vandellii Cuvier, 1 829 Osteoglossum bicirrhosum Cuvier, 1 829 Glossodus forskalii A, in S & A, 1829 Engraulis sericeus A, in S & A, 1 829 Engraulis bahiensis A, in S & A, 1829 Engraulis grossidens A, in S & A, 1829 Engraulis Janeiro A, in S & A, 1829 Engraulis tricolor A, in S & A, 1829 Engraulis piquitinga A, in S & A, 1829 Clupanodon aureus A, in S & A, 1829 Megalops thrissoides (Bloch, in Schneider, 1801) Pristigaster martii A, in S & A, 1829 Anodus elongatus A, in S & A, 1829 Anodus latior A, in S & A, 1829 Prochilodus [Pacu] argenteus A, in S & A, 1829 Prochilodus [Pacu] nigricans A, in S & A, 1829 Leporinus novemfasciatus A, in S & A, 1829^^* Schizodon [Curimata] fasciatus A, in S & A, 1829 Chalceus angulatus A, in S & A, 1829 Chalceus [Characinus] amazonicus A, in S & A, 1829 Tetragonopterus chalceus A, in S & A, 1829 Serrasalmo aureus A, in S & A, 1829 Serrasalmo nigricans A, in S & A, 1829 Myletes [Tetragonopterus] aureus A, in S & A, 1829 Myletes bidens A, in S & A, 1829 Raphiodon [Cynodon] gibbus A, in S & A, 1829 Raphiodon [Cynodon] vulpinus A, in S & A, 1829 Xiphostoma cuvieri A, in S & A, 1829 Saurus longirostris A, in S & A, 1829 Saurus intermedius A, in S & A, 1829 Saurus truncatus A, in S & A, 1829 Rhombus ocellatus A, in S & A, 1831^' Rhombus solaeformis A, in S & A, 1831 Solea brasiliensis A, in S & A, 1831 Monochir maculipennis A, in S & A, 1831 Plagusia brasiliensis A, in S & A, 1831 Gymnothorax rostratus A, in S & A, 1831 Gymnothorax ocellatus A, in S & A, 1831 Anarrhichas leopardus A, in S & A, 1831 Labrus crassus A, in S & A, 1831 Julis dimidiatus A, in S & A, 1831 Xyrichthys uniocellatus A, in S & A, 1831 Scarus frondosus A, in S & A, 1831 Cichla labrina A, in S & A, 1831 Cichla monoculus A, in S & A, 1831 Cybium maculatum (Mitchili, 1815) Micropteryx cosmopolita (Cuvier, 1829) Caranx latus A, in S & A, 1831 Caranx lepturus A, in S & A, 1831 Caranx macrophthalmus A, in S & A, 1831^^ Caranx punctatus A, in S & A, 1831 Argyreiosomus vomer (Linnaeus, 1758) Vomer brownii Cuvier, 1816 Coryphaena immaculata A, in S & A, 1831 Ephippus gigas Cuvier, 1829 Gerres linaetus (Humboldt, in Humboldt & Valenciennes, 1821)^ Pagrus argyrops (Linnaeus, 1758) Mesoprion uninottatus Cuvier, in Cuvier & Valenciennes, 1828 Mesoprion aurovittatus A, in S & A, 1831 Corniger spinosus A, in S & A, 1831 79 Uranoscopus occidentalis A, in S & A, 1831 Corvina [Sciaena] adusta A, in S & A, 1831 Pachyurus squamipennis A, in S & A, 1831 Lobotes ocellatus A, in S & A, 1831 Haemulon canna Cuvier, in Cuvier & Valenciennes, 1830 Haemulon schrankii A, in S & A, 1831 Batrachus punctatus A, in S & A, 1831 Mugil brasiliensis A, inS&A, 1831 Atherina taeniata A, in S & A, 1829^' Atherina macrophthalma A, in S & A, 1831 Alutera punctata A, in S & A, 1831 His footnote (4) reads: "Spix en fait son sous-genre Hypophthalmus, dont il a deux especesiHyp. edentatus, IX, Hyp. nuchalis, XVII." Hypophthalmus is clearly available from Cuvier's description. As we know that Cuvier had material of this genus (see comments in Whitehead & Myers, 1971 : 485), it is clear that he based his description on the specimens and is alone responsible for the conditions which make Hypophthalmus available. No specific names are available from Cuvier (1829). Spix & Agassiz (1829) were the first to use avai- lable specific names with Hypophthalmus: H. edentatus and H. nuchalis. Bleeker (1862: 15) designa- ted H. edentatus as the type species. Pirarara Agassiz, in Spix & Agassiz, 1829 Phractocephalus Agassiz, in Spix & Agassiz, 1829 Both Pirarara and Phractocephalus are available names. Phractocephalus has been commonly used, but Bleeker (1862: 11) clearly placed Phractocephalus as a synonym oi Pirarara. As Bleeker was the first author to mention both names and to place one in the synonymy of the other, he is the first revi- ser. Pirarara is the name which should be used instead oi Phractocephalus. Since Bleeker's (1862) ac- tion, the name Pirarara has not been used again, except in synonymies et chresonymies. Boeseman (1983) lists 4 references using Pirarara against 58 using Phractocephalus. The name Phractocephalus has been widely used and its replacement by the senior synonym Pirarara cannot be in the interest of the stability of nomenclature. Phractocephalus being an important food fish, conservation of this name seems justified and the case has been submitted to the International Commission on Zoological No- menclature; awaiting the ruling of the Commission, Phractocephalus is the name to use for this genus [Code art. 80(a)]. Sorubim Cuvier, 1829 Platystoma Agassiz, in Spix & Agassiz, 1829 Platy Stoma is described by Agassiz in the text of "Brazilian fishes" while Sorubim appears in com- bination with several specific names on the plates. Both are available, but Platystoma Agassiz cannot be valid as it is preoccupied by Platystoma Meigen, 1803 (p. 227), a genus of insects. Agassiz would be the author of Sorubim if that name had not already been established by Cuvier (1829: 293), who diagnosed a group of catfishes: "Parmis ceux ä six barbillons, les plus remarquables ont le museau deprime et large, autant et plus que le brechet [Esox](l)". His footnote (1) reads: "Sil. lima Bl. Sehn.; — 5/7. fasciatus, Bl. 366, et diverses especes nouvelles. Spix fait de cette division son genre Sorubim. " As a result, Sorubim is clearly available, Cuvier is the author and two nominal species are included. Bleeker (1862: 10) designated Silurus lima Bloch, in Schneider, 1801, as the type species. 80 Osteoglossum Cuvier, 1829 Ischnosoma Cuvier, 1829 Osteoglossum vandellii Cuvier, 1829 Osteoglossum bicirrhosum Cuvier, 1829 Agassiz (p. 46) used the heading "Osteoglossum Vandelli. Ischnosoma Spix". This means that he considered Ischnosoma a junior synonym of Osteoglossum, presumably based on Cuvier's manuscript notes (see Whitehead & Myers, 1971 : 487) stating: "Ischnosoma bicirrhosum: Vandelli a decrit et represente ce poisson sous le nom d' Osteoglossum". From Valenciennes (in Cuvier & Valenciennes, 1847: 289, 294) it is clear that this Vandelli descrip- tion was in manuscript. Cuvier (1829: 328) actually published the first description of Osteoglossum and his work appeared a few months before "Brazilian fishes". Cuvier included a single species in his genus Osteoglossum. This is explicit by his Statement: "On en connait une espece assez grande du Bresil (Osteoglossum Vandellii, n., ou Ischnosoma bicirrhosum, Spix, XXV.)". [We know a single, quite large species from Brazil (Osteoglossum Vandellii, n. [nobis = mine], or Ischno- soma bicirrhosum, Spix, [plate] xxv)]. This means that Cuvier named the new species O. vandellii and considered that the /. bicirrhosum on Spix's unpublished plate was the same species. Thus Osteoglossum and O. vandellii are available names. This is particularly obvious in the case of the generic names as the heading of the chapter was "Les Osteoglosses", clearly showing that Cuvier wanted to use the name Osteoglossum. By analogy and considering the context and practices of that time, this also applies to the specific names. Cuvier (and later Cuvier & Valenciennes) often mentioned names existing only on labeis, in manuscripts or on unpublished drawings. In some ways, this is analoguous with our present practice of Hsting "mate- rial examined"; one of the "specimen" of O. vandellii was the unpublished Illustration labelled /. bi- cirrhosum by Spix. Ischnosoma and /. bicirrhosum are first published in synonymy and are available only if they have been treated as available before 1961 [Code art. 11 (e); see also example thereafter]. The next use of Ischnosoma and bicirrhosum by Agassiz (in Spix & Agassiz, 1 829) makes them available; but they must be considered as authored by Cuvier (1829) [Code, art. 50 (g)]. These four names being available from Cuvier (1829), we have two sets of simultaneous Synonyms. The decision of the first reviser is decisive [Code art. 24 (a), (b)]. For the specific names, the first reviser is Agassiz (in Spix & Agassiz, 1831 : conspectus) who retained O. vandellii. The combination Osteoglossum vandellii has never been used again for this fish. As it is an impor- tant food fish, conservation of the name seems justified. This request has been submitted to the Inter- national Commission on Zoological Nomenclature; awaiting the ruling of the Commission, O. bicirr- hosum is the name to use for this species [Code art. 80 (a)]. Ischnosoma too has not been used again. The first reviser is Günther (1868: 378) who retained Osteoglossum. Kanazawa (1966) considered Vandelli, in Cuvier (1829), to be the author of Osteoglossum. As there was no pubHshed description by VandeUi or indication that Cuvier's description in fact was by Van- delli, this is not acceptable. He considered Cuvier's Osteoglosum vandellii as a nomen nudum and Vandelli (in Spix & Agassiz, 1829) as author of O. bicirrhosum. If Agassiz mentioned Vandelli as the author of O. bicirrhosum, he did so on the basis of a misinterpretation of Cuvier's notes. As for other species, the various manuscript versions of "Brazilian fishes" in AEN shows that Agassiz prepared the description alone. Agassiz used Spix's specific name bicirrhosum and associated it with Vandelli be- cause Cuvier had written that Vandelli had described Ischnosoma bicirrhosum in the genus Osteoglos- sum (see notes in Whitehead & Myers, 1971). This does not mean that Vandelli intended to describe it as O. bicirrhosum; we know that Vandelli intended to name it O minus (see Valenciennes, in Cuvier & Valenciennes, 1847: 289, 294). As Vandelli is authorof neitherthenamenor thedescription, hecan- not be the author of O. bicirrhosum as postulated by Kanazawa (1966). Further, it is clear from Valenciennes (in Cuvier & Valenciennes, 1847: 289) and Kanazawa (1966) that both the material and description available to Lacepede, Cuvier, and later Valenciennes were Alexandre Rodriguez Ferreira's ( 1 756 — 1 8 1 5), which Vandelli sent to Lacepede (see also Myers, 1 964). If Vandelli simply sent them to Lacepede, I see no reason why he should be considered the author of anything and Ferreira author of nothing. According to Carvalho (1983), VandeUi stole the manu- script. Ferreira's manuscript is not the one later published by Miranda-Ribeiro (Ferreira, 1903) which describes Arapaima gigas. Pacu Cuvier, 1829 Pacu Agassiz, in Spix & Agassiz, 1829 Prochilodus Agassiz, in Spix & Agassiz, 1829 Both Prochilodus and Pacu would be available from "Brazilian fishes" had not Cuvier (1829: 309) already made Pacu available by a short description ended by a footnote: "Ce sont les Pacu Spix. Ses Anodus en different seulement par une bouche plus fendue". These short Statements made both Pacu and Anodus available. As there is no indication that Spix is responsible for any part of the description, Cuvier alone is the author. Cuvier's description appearing under the French heading "Les Curimates", Pacu cannot be considered a name first published in synonymy of Cunmata Bosc, 1817. No type species has been designated for Pacu Cuvier. The following nominal species were originally included: Salmo edentulus Bloch, 1794, Salmo unimaculatus Bloch, 1794, Salmo taeniurus Valencien- nes, in Humboldt & Valenciennes, 1817, Salmo curima Cuvier, 1829, Cunmata gilb er üQ^oj & Gai- mard, 1824 and Curimata cyprinoides Linnaeus, 1766. These species are currently placed in Curimata, Hemiodus and Semaprochilodus. Although Cuvier used Spix's Pacu, Pacu Cuvier and Pacu Agassiz are not the same. Pacu Agassiz originally included two species, P. argenteus and P. nigricans, both described by Agassiz, and is a syn- onym of Prochilodus Agassiz, in Spix & Agassiz, 1829. Eigenmann (1910) designated P. argenteus as the type species of Prochilodus. As Prochilodus was intended as a replacement name for Pacu Agassiz, P. argenteus is also type species of Pacu Agassiz. Pacu Agassiz and Prochilodus are simultaneous synonyms. The first revisers, Müller & Troschel (1844: 84), retained Pacu Agassiz and considered Prochilodus to be a junior synonym. Since Müller & Troschel (1845), Pacu Agassiz has not been used again as a valid name; it must be rejected as it is a junior homonym of Pacu Cuvier. Prochilodus has been placed on the Official List of Generic Names in Zoology (Opinion 772; see also Gery, 1963). In Order to avoid confusion, Pacu Cuvier is also better not used. As there has been no type species designation until now, I designate Salmo edentulus Bloch, 1794 as the type species of Pacu Cuvier, 1829. Since Salmo edentulus is the type species of Curimata Bosc, 1817 by designation under the ple- nary powers of the International Commission of Zoological Nomenclature (Opinion 772), Pacu Cu- vier is a junior objective synonym of Curimata. In addition, "pacu" is the vernacular name for completely different fishes (Colossoma Eigenmann & Kennedy, 1903, Mylossoma Eigenmann Sc Kennedy, 1903 and Myleus Müller & Troschel, 1844). Its use as a generic name could only lead to numerous confusions, the species of all the genera mentioned in this discussion being among the most important Amazonian food fishes. 82 Anodus Cuvier, 1829 As shown above (see discussion oi Pacu), Anodus is available from Cuvier (1829). Cuvier did not include species. His Anodus is obviously the same as described in "Brazilian fishes", which included the foUowing nominal species: Curimata amazonum Humboldt, in Humboldt & Valenciennes, 1817, C. f^e«?«TO5 Valenciennes, in Humboldt & Valenciennes, \'i\7 ,Salmo edentulus'^Xoch, \794,A. elon- gatus Agassiz, in Spix & Agassiz, 1829 and A. latior Agassiz, in Spix & Agassiz, 1829. Eigenmann & Eigenmann (1899) designated A. elongatus as the type species. Rhaphiodon Agassiz, in Spix & Agassiz, 1829 Cynodon Agassiz, in Spix & Agassiz, 1829 Rhaphiodon is described by Agassiz in the text while, Cynodon appears in combination with two specific names on the plates. Agassiz proposed Rhaphiodon as Cynodon was already used in botany, but this does not preclude its use in zoology [Code art. 1 (c)]. Both names are simultaneously available. The first revisers, Müller & Troschel (1844), retained Rhaphiodon and considered Cynodon to be a junior synonym. Two species were originally included: R. vulpinus Agassiz, in Spix & Agassiz, 1829 and R. gibbus Agassiz, in Spix & Agassiz, 1829. Eigenmann (1910) designated R. vulpinus as the type species oi Rhaphiodon and R. gibbus as the type species of Cynodon. Since Rhaphiodon is a [unneces- sary] replacement name for Cynodon, the type species of both must be the same [Code art. 67 (h)]. Eigenmann's action is not acceptable as he simultaneously selected two different species as type spe- cies of a Single genus. Campos (1945: 472) incorrectly placed Rhaphiodon as a synonym of Cynodon and considered C. gibbus to be the type species by original designation. Although his Statement is wrong, he is ipso facto author of the first vaHd type species designation [Code, art. 69 (a) (iv)]. Rhaphiodon gibbus is the type species of both Rhaphiodon and Cynodon. Campos (1945: 473) considered that R. gibbus and R. vulpinus were not congeneric and created Rhaphiodonichthys (type species : Rhaphiodon vulpinus) for the last named species. But he overlooked Hydropardus Reinhardt, 1849 (type species: H. rapax Reinhardt, 1849 = Rhaphiodon vulpinus) which has priority over Rhaphiodonichthys. Travassos (1946) reviewed the case of Rhaphiodon and Cynodon and concluded that Cynodon is a synonym of Rhaphiodon for various reasons (not all correct) and accepted Eigenmann's (1910) type species designation oiR. vulpinus as the type species oi Rhaphiodon (as stated above, this designation cannot be accepted and the next first reviser, i.e. Campos, 1945, is decisive). He also thought that R. gibbus and R. vulpinus were not congeneric; thus Rhaphiodonichthys Campos was considered to be a synonym oi Rhaphiodon and Travassos created Camposichthys (type species: R. gibbus). In summary, if two genera are accepted (see Howes, 1976; G£ry, 1986b), they are: a) Rhaphiodon Agassiz, in Spix & Agassiz, 1829 (type species: /?. gibbus Agassiz, in Spix & Agassiz, 1829, by subsequent designation by Campos, 1945: 472) Synonyms: Cynodon Agassiz, in Spix & Agassiz, 1829 (type species: R. gibbus Agassiz, in Spix & Agassiz, 1829, by subsequent designation by Campos, 1945: 472) Camposichthys Travassos, 1946 (type species: R. gibbus Agassiz, in Spix & Agassiz, 1829, by original designa- tion) b) Hydropardus Reinhardt, 1 849 (type species : H. rapax Reinhardt, 1 849 — Rhaphiodon vulpinus Agassiz, in Spix & Agassiz, 1829, by monotypy) synonym: Rhaphiodonichthys Campos, 1945 (type species: Rhaphiodon vulpinus Agassiz, in Spix & Agassiz, 1829, by original designation) 83 Table 3: Specimens of species described in "Brazilian fishes" still in ZSM, MHNN and MHNG, with Agassiz's size indications and potential type Status. All specimens in alcohol unless otherwise stated. New species are marked by asterisks (*). Species '''Cetopsis candiru '■'■'Hypohthalmus edentatus ''Sorubim planiceps '■'Sudispirarucu ''Erythrinns macrodon ''Engraulis tncolor '■'Clupanodon aureus '■'Schizodon fasciatus '•'Tetragonopterus chalceus '•'Myletes aureus ''■'Myletes hidens '■'Rhaphiodon vulpinus '•'Xiphostoma cuvieri '■'Saurus longirostrls ''■'Saurus truncatus '■'Solea hrasiliensis '■'Monochir maculipennis '■'Plagusia hrasiliensis '■'Gymnothorax ocellatus '■'Julis dimidiatus ''Scarus frondosus Material'" MHNN 735, 1 ex., 126 mm SL, 145 mm TL MHNN 737, 1 ex., 147 mm SL, 170 mm TL MHNN 736, 1 ex., 204 mm SL, 230 mm TL MHNG 210.5, 1 ex., 135 mm SL, 157 mm TL" MHNN 706, 2 ex., 193-209 mm SL, 240-245 mm TL' MHNN 811, 1 ex., 458 mm SL, 560 mm TL ZSM 26725, 3 tongue-bones'' MHNN 773, 1 ex., 226 mm SL, 280 mm TL'' MHNN 1 142, 6 ex., 55-62 mm SL, 63-73 mm TL MHNN 1159, 1 ex., 175 mm SL, 213 mm TL MHNN 781, 1 ex., 167 mm SL, 200 mm TL MHNN 785, 1 ex., 82 mm SL, 108 mm TL" MHNN 787, 1 ex., 1 15 mm SL, 140 mm TL MHNN 788, 2 ex., 119- 126 mm SL, 147-153 mm TL MHNN 789, 1 ex., 150 mm SL, 182 mm TL MHNN 2216, Idryex., 191 mm SL, 222 mm TL MHNN 822, 1 ex., 315 mm SL, 335 mm TL MHNN 823, 1 ex., 335 mm SL, 392 mm TL MHNN 793, 1 ex., 186 mm SL, 211 mm TL MHNN 795, 1 ex., 145 mm SL, 161 mm TL MHNN 804, 1 ex., 272 mm SL, 316 mm TL MHNN 685, 2 ex., 50- 109 mm SL, 78- 140 mm TL MHNN 691, 1 ex., 150 mm TL MHNN 1 193, 1 ex., 335 mm TL MHNN 563, 1 ex., 153 mm SL, 173 mm TL MHNN 581, 1 ex., 178 mm SL, 210 mm TL Agassiz's comments several specimens, 135, 162 and 217 mm' 2 ex., 271-325 mm a dry specimen, 677 mm 1 ex., 298 mm several specimens 2 ex., 217 mm 2 ex., 162-230 mm 1 ex., 102 mm 3 ex., 135-189 mm 2 ex., 135-189 mm Type Status syntypes potential syntypes'* potential syntype" potential holotype syntypes syntype potential syntype'*" potential holotype potential syntypes'** lectotype paralectotype''' one mutilated specimen, 292 mm" potential holotype 1 ex., 433 mm 2 ex., 189 and 230 mm 2 ex., 162 and 189 mm 2 ex., 325 mm 2 ex., 81-162 mm lex., 183mm'° 2 ex., 325-379 mm several specimens, 171 mm 5 ex., 135-325 mm potential holotype potential syntype'* syntype syntype syntypes potential syntype" syntype syntype syntype Species ''Cichla labrina '■'Cichla monoculus Micropteryx cosmopolita ''Caranx latus '■'Caranx punctatus Argyreiosus vomer Vomer brownii ''Coryphaena immaculata Mesoprion uninottatus ''Mesoprion aurovittatus ''Corvina adusta ''Batrachus punctatus ''Mugil brasiliensis Material'" MHNN 599, 1 ex., 120 mm SL, 145 mm TL MHNN 2188, 1 dry ex., 318 mm SL, 373 mm TL MHNN 339, 2 ex., 115-116mmSL, 127-132 mm TL MHNN 455, 1 ex., 180 mm SL, 223 mm TL MHNN 334, 2 ex., 125-135 mm SL, 141-153 mm TL MHNN 336, 1 ex., 141 mm SL, 185 mm TL MHNN 338, 1 ex., 1 10 mm SL, 140 mm TL MHNN 449, 1 ex., 357 mm SL, 450 mm TL MHNN 76, 2 ex., 150-170 mm SL, 185-210 mm TL MHNN 73, 1 ex., 122 mm SL, 142 mm TL MHNN 312, 1 ex., 133 mm SL, 168 mm TL MHNN 390, 1 ex., 187 mm SL, 225 mm TL MHNN 507, 4 ex., 151, 186, 190, 241 mm SL, 188, ,230, 300 mm TL Agassiz's comments 2 ex., 122-189 mm 1 ex., 352 mm several specimens, 135-162 mm 1 ex., 162 mm 2 ex., 149-162 mm several specimens, various sizes 1 ex., 176 mm 1 ex., 568 mm several specimens, 189-244 mm 2 ex., 217 mm 1 ex., 271 mm 2 ex., 271 mm 2 ex. Type Status lectotype'*''" potential syntype*^ potential syntype'* syntypes potential syntype'* potential syntype'* potential syntype'* potential syntype potential syntypes'* 84 Eigenmann (1909: 256) created the family-group name Cynodontinae (type genus Cynodon) for ac- commodation of the genera under review here. Travassos (1946: 136) replaced it by Rhaphiodontinae. This family-group name has been widely and variously used at subfamily (for example G£ry, 1977) er tribal level (Howes, 1976). Cynodontidae having been rejected in 1946, and Rhaphiodontinae having met "general accep- tance", it is to be retained [Code art. 40 (b)]. It must then take author and date as follows : Rhaphiodon- tinae Travassos, 1946 (1909) (Code Recommendation 40 A). Specimens All the labelled specimens of species described in "Brazilian fishes" and still extant in the Agassiz coUection in Neuchätel are listed in Table 3. The size given by Agassiz is indicated, as well as the po- tential type Status. %».^y4*. Uf^jl / ) — i-^^ <' %l ^.^ (^ _^_, Fig. 4. Upper part of first page of document AEN 115/2.5, entitled "Verzeichnis der Fische der zoologischen Samlung in München (mit Ausnahme der Dubletten). August 1829. Von L. Agassiz aufgestellt". The number of specimens is identified in the following way, as examplified by Raja clavata: 1 specimen in alcohol, two specimens and five half specimens dry. 85 Surprisingly, the Neuchätel collection contains specimens of many species for which Agassiz indi- cated that he had only one specimen. There could be several explanations for this : (a) Agassiz received this Single specimen; (b) Agassiz had it on loan and never returned it; (c) there was more than one spe- cimen in the collection. The first explanation seems very unlikely, the other two are possible, but I fa- vour the third one and propose the following scenario: Agassiz prepared descriptions of the species illustrated on Spix's plates. He added a few other spe- cies (those whose plate number include letters: 8a, 13a, etc.) selected from many not illustrated but present in Spix's collections. He had new plates prepared for them. The title of the book (Selecta ge- nera et species . . .) and document AEN 115/2.5, which is a list of the fishes in the zoological collection in München, hand-written by Agassiz on 22 August 1829 (Fig. 4), are evidence for this. The list is en- titled "Verzeichnis der Fische der Zoologischen Samlung in München (mit Ausnahme der Dubletten)" [inventory of the fishes in the zoological collection in Munich (with exception of the duplicates)]. It includes all the names of the fishes described in 1 829, most of those which would be described in 1831 and several species labelled as "spec. nov. bras." [new species from Brazil];theybelongedto the genera Torpedo, Raja (3 ssp.), Malthe, Gadus, Gobius (2 spp.), Sciaena, Muraena (2 spp., but the two new Gymnothorax described by Agassiz are not listed), Scorpaena, Sparus and Xyrichthys. Not only species but the actual specimens were selected to be illustrated and described. Thus, when Agassiz mentioned a given number of specimens, this should be understood as the number of speci- mens on which he based the description and not as the total number of specimens at band. When he received specimens it was sometimes material he had used for description and sometimes duplicates. As there was no type concept at that time, the specimens used for the descriptions did not received more attentive care than the others and a curator would have no reason to send one in preference of the other as gift or exchange (unless one was in a better condition). Only the specimens actually used for description, mentioned by Agassiz or used as modeis of illus- trations, can be considered as types. Most of the specimens in MHNN correspond to Agassiz's indi- cations and could be the material used for the descriptions. The rest are probably from the "Dublet- ten", the existence of which is shown by the München list mentioned above. Some of them are possibly the modeis used for the Illustration; the illustrations may also be composite, based on several speci- mens of a Single species. It is impossible to say exactly which specimens are types and which are not. A clue to the identity of the possible type specimens is the size given by Agassiz (not in all cases) in Zoll (inches). Several dif- ferent kinds of Zoll have been used in Germany; Agassiz indicated (p. 4) that he used the Parisian one, which was one twelfth of a foot (32.48 cm, see Klimpert, 1896). Unfortunately, Agassiz did not indi- cate anywhere what kind of length he used (Standard length, total length, fork length). Some MHNN specimens correspond to the lengths given by Agassiz in either Standard length or total length. Surely he did not use both! Generally, the length indicated by Agassiz corresponds to the total length of the specimen. I conclude that he consistently used total lengths. The size of the drawings is of limited use as the fishes are not reproduced to natural size and the magnification is not indicated; there is also a possibility that they are based on more than one specimen. Some might have been discarded after com- pletion of the illustration. Types used as modeis for the plates can be recognized only by resemblance. When the MHNN specimen is approximately the same length as the single specimen mentioned by Agassiz, I tentatively consider it to be a potential holotype. When it is a different length, I suggest that it might be the model or part of the modeis used for the drawing and that there was more than one spe- cimen. This is a wide Interpretation of Recommendation 72B of the Code. In my opinion, definitive recog- nition as a type should be decided by competent specialists for the various groups. As no holotype were formally designated, and as it is possible that there were more than one specimen, revisers should follow Recommendation 73F of the Code and designate a lectotype rather than assuming a holotype. Material of 34 of the 92 species described in "Brazilian fishes" is still extant. Of the 50 species descri- bed in 1 829 we still have 14 (28 %) while we have 19 (45 %) of the 42 described in 1 831 . This might be explained by the fact that the first part of the book was written in München while the second part was prepared in Concise (E. Agassiz, 1887) to where Agassiz took material from München. The Agassiz coUection in Neuchätel also includes specimens of several species of small size not Hs- ted in "Brazihan fishes" and labelled as coming from Brazil; their origin is unknown. As very few col- lections from Brazil reached Europe at that time, it is possible that they also belonged to the Spix col- lection and were received from München. While in Neuchätel, Agassiz arranged exchanges with several scientists and institutions. Among the copies of Agassiz's letters (document AEN 6/1), is a copy of a letter to Frangois-Jules Pictet (known as Pictet-de-la-Rive, 1809 — 1872) in Geneve to whom Agassiz proposed an exchange of specimens of fishes "des grandes rivieres du Bresil". Two Brazilian fishes received from Agassiz in 1836 appear in the registers of the Museum d'Histoire Naturelle, Geneve. One (Myletes bidens, MHNG 210.1) can- not be found (it apparently disappeared in the period 1920 — 1940) and the other (Cetopsis candiru) is shown on Table 1. It is possible that other specimens were received by other institutions, but I could trace none. Sven KuUander (in litt.) called my attention on Kaup's (1860) description o{ Hoplarchus pentacan- thus and H. planifrons which where based on specimens in the München coUection, and which prob- ably were from the Spix & Martins expedition. Acknowledgments I would like to thank the followings: Christophe Dufour and Jean-Paul Haenni, MHNN, for proposing that I revise their fish collections, for providing funding and for useful discussion, J. Courvoisier, AEN, for kindly giving me access to Agassiz's archives; Marinus Boeseman, Jacques Gery, L. B. Holthuis, Isaäc Isbrücker, Sven O. KuUan- der, Volker Mahnert, G. F. Mees, R. Melville, Han Nijssen, Fritz Terofal, Richard P. Vari, Alwyne Wheeler and Peter J. P. Whitehead for commenting on various versions of the manuscript or parts of it; H. Bürdet (Conserva- toire Botanique de Geneve) for assisting me in trying to find old correspondence between Mercier and Martius; An- toinette Kottelat-Kloetzli and M. Pointet for typing the various versions of the manuscript; and Jean-Louis Rais (Musee Jurassien, Delemont) for Latin translations. The illustrations were prepared with the help of Christophe Dufour. Notes ' Following Swiss custom, he only used one of his Christian names, Jean Louis Rodolphe. ^ The French translation by Auguste Mayor (1887) of E. Agassiz (1885) has been used throughout this paper. Mayor (1815 — 1 904) was a cousin of Louis Agassiz. ^ This is the name which appears on most books. In "Brazilian fishes" it appears as Friedrich Carl Philipp von Mar- tius. "* As there is great inconsistency and numerous mistakes in the use of nobiliary particles m nomenclatorial htera- ture, the French, German and Latin rules are briefly summarized: French: The particle de is used only to join the name to the Christian name, or its initials, a title, etc. (example: Le Baron de Cuvier). It is omitted if the name is used alone (example: Cuvier a decrit. . .), except: 1°) in front of monosyllabic names or bisyllabic names ending with a silent e; 2°) in front of names beginning with a vowel or a silent h. The particle d' is always used (see Grevisse, 1980, for details, exceptions and particular cases). In the authorship of a name, the particle de has thus to be omitted. Latin: The particle de is used only to join the name to the Christian name, title, etc. (example: Johannes de Spix). It is omitted if the name is used alone and in the authorship of a name (example: Pirarara Spix and not Pi- rarara de Spix). This rule is consistently used by Agassiz in "Brazilian fishes". German: The use of the particle von seems not very consistent and there are regional and time variations. But at the beginning of last Century, it was the same as the rule for Latin given above. For example Anon [Oken ?] (1829) wrote: "Martius hat vorher die Abbildungen an Cuvier geschickt" and not "von Martius hat vorher. . ." I would recommend that these rules be followed in order to avoid confusion or barbarisms (like a "J. B. de von Spix" which already appeared at several opportunities). 87 ^ Agassiz proposed Glanis with a short diagnosis as a replacement name for the "barbarian" Bagres of Cuvier. Cu- vier's Bagres is not a latinized name and Bailey & STEWART (1983) showed that the correct name is Bagre Clo- quet, 1816 and the type species is Silurus bagre Linnaeus, 1 766 by absolute tautonymy , a point of view followed by the International Commission on Zoological Nomenclature (OPINION 1403). Cuvier's Bagres being not an available name, Glams has not obligatorily the same type species. Agassiz did not include any nominal species and this name has not been used for these fishes until now. In order to avoid any fur- ther problem, I designate Silurus bagre Linnaeus, 1766 as its type species. Glanis is thus an objective junior syn- onym of Bagre Cloquet, 1816, as was intended by Agassiz. As first proposed, Glanis was not in the nominative singular, but it is available in the nominative singular [Code art. ll(g)(i)]. Glanis Gronovius, 1854 (p. 135, type species G. imberbis Gronovius, 1854, by monotypy) and Glams Agas- siz, 1856 (type species G. aristotelis Agassiz, 1856, by monotypy) are junior homonyms of Glanis Agassiz, in Spix & Agassiz, 1829. ^ Günther (1864: 191) stated that Ceratorhynchus had been proposed by Agassiz (1833-43) in his "Recherches sur les poissons fossiles" (see Jeannet, 1928, 1929, for the exact dates of pubHcation of the various parts). Gün- ther did not give page references. I checked the whole series without finding any trace of this name. As first proposed, Ceratorhynchus was not in the nominative singular, but it is available in the nominative sin- gular [Code art. 11 (g) (i)]. ^ Originally included species: S. coecutiens and C. candiru Agassiz, in Spix & Agassiz, 1829. ^ The dates of publication of the ichthyological results of Humboldt & Bonpland's travel are often given as 1811 or 1816. Sherborn & Woodward (1901) traced them and concluded that they were published in 1821. Gery (1976: 51), without giving the füll reference, stated that Valenciennes, in the "Histoire naturelle des poissons", wrote that the "Recherches sur les poissons fluviatiles" appeared in 1817. In fact, this paper appeared in two fas- cicles, nos. 11 (pp. 145-176) and 12 (pp. 177-244). Fascicle 10 appeared in 1817 (Sherborn & WoODWARD, 1901; the Basel University Library still has an original wrapper with the date 1817). Fascicle 12 appeared during or after 1821, as on p. 193, Valenciennes wrote: "Letravailque j'aifaitsurlegenrepoecilie ... etait imprime lors- que nous avons re^u le Journal de l'Academie des Sciences de Philadelphie dans lequel Mr. Lesueur — " The work on poeciliids is in fascicle 1 1 . Lesueur's paper appeared in 1 821 . It thus seems that fascicles 1 1 and 12 were not published together . Fascicle 1 1 may have been published as early as 1 8 1 7, while fascicle 1 2 appeared in or after 1821. Doras crocodili is described in fascicle 12. Fascicle 12 also included plates 45 — 48a, 48b, 49, 51, that is all those illustrating fishes. These conclusions somewhat disagree with those of Gery (1986a). ' Valenciennes (in Cuvier & Valenciennes, 1 847: 323) explains that this name was based on Cuvier's notes where the name "glossodonte" appears (see Whitehead & Myers, 1971 : 487). '° Originally included species: L. novemfasciatus and Salmo friderici Bloch, 1794. " Originally included species: 5. fasciatus and S. unimaculatus Bloch, 1794. '^ Agassiz, in afootnote (p. 76), dWiAtd Hydrocyon Cuvier, 1819 into three groups o{ yN]\ic\i Xiphorhynchus'is one, but this name is preoccupied in Aves (SWAINSON, 1827: 440). It has been replaced by Xiphorhamphus Müller & Troschel, 1844 which is also preoccupied in Aves (Blyth, 1843: 929). But this name appears as Xiphorhamphus on p. 947 and as Xiphoramphus on p. 929. The first spelling has generally been used by subsequent authors but I did not find any definitive first reviser action. If the first spelling is retained, then Xiphorhamphus is not availa- ble for the fishes under review. Acestrorhynchus Eigenmann, 1903 would then be the next available name. The type species of these three genera is Salmo falcatus Bloch, 1795. '^ This is Agassiz's third group of Hydrocyon. There is no description, but there is one included species and it is thus an available name. ''' Preoccupied in Hemiptera (KiRBY & SPENCE, 1829: 683). The next available name is Boulengerella Eigenmann, 1903, whose type species is X. lateristriga Boulenger, 1895. Whitley (1951) proposed Spixostoma as a replace- ment name for Xiphostoma. '^ Originally included species: X. cuvieri and Hydrocyon lucius Cuvier, 1819. '^ A replacement name for Seriola Cuvier, 1817. Species originally included in Seriola: Caranx dumerili, Scomber fasciatus Bloch, 1797, S. speciosus Fors[s]kal, 1775. '^ Cuvier (see Whitehead & Myers, 1971) saw the plate of C. edentatus and wrote that it was the Doras ^7^ger des- cribed by Valenciennes (in Humboldt & Valenciennes, 1821 : 284). Agassiz (p. 15) acknowledged this remark by Cuvier, but stated how this species (named D. humboldti in the text) differs from D. niger. Valenciennes (in Cuvier 8c Valenciennes, 1840: 291) considered Agassiz' species a synonym of D. niger, and used the names D. humboldti and C. edentatus without distinction. His synonymy has been adopted by all subsequent authors. '* Preoccupied by P. albidus Lesueur, 1819 (p. 148). Agassiz was aware of this (see his synonymy) and proposed P. spixii as a replacement name. '^ Considered a synonym of Silurus lima Bloch, in Schneider, 1801, by Agassiz (p. 24). ^° The mention in Agassiz's synonymy "Pisces nomine Pirayapeaui inscriptus" is not a binominal name but a name (Pirayapeau) borne by the dry mounted specimen which was in München collection. ^^ Agassiz considered it as a synonym of Sudis gigas. The later name is usually credited to Cuvier, but it first appea- red in a German translation of CuviER (1816; see WhiTEHEAD, 1967b, and CowAN, 1969, for the exact dates of publication of the first edition of Cuvier's Regne animal) by SCHINZ (1822: 305). Cuvier (1816: 5, pl. 10) used only the French name "vastre geant" but did not use any Latin name. The German edition was updated and com- pleted by Schinz who introduced the Latin name for the "vastre geant". He is thus author of the name which must be given as Sudis gigas Schinz, in Cuvier, 1822. ^^ Spelt E. 1-taematHS on plate 19. ^^ It is possible that E. trahira is an unjustified emendation oi Synodus tareira Bloch, in Schneider, 1801 (p. 397), but it is as likely based on a different transcription of the same vernacular name. I consider E. trahira a distinct and available name. ^^ Spelt L. 9-fasciatum on plate 37. ^^ WhitEHEAD & Myers (1971) wrongly stated that R. podas Delaroche was placed in synonymy of this species. In fact, R. podas as well as R. pantherinus Rüppell are compared with R. ocellatus. ^^ Caranx macrophthalrnns Agassiz is a junior homonym of C. macrophthalmus Rüppell, 1830 (see DoR, 1984: 131, 334 for dating of the various parts of this work). Agassiz (in Spix & Agassiz, 1831: conspectus p. 3) was aware of it; the text was probably already printed (but not yet distributed) when he wrote the conspectus. Agas- siz found that Rüppell's macrophthalmus has smaller eyes than his, and thus retained macrophthalmus for his species, renaming Rüppell's macrophthalmus as C. ruppellii. This is obviously not correct in vue of our present Code and C. macrophthalmus Agassiz is a junior homonym of C. macrophthalmus Rüppell. Caranx ruppellii Agassiz is not proposed in the nominative form (. ..Ruppellianum Carangem macroph- thalmum nomine Carangis RuPPELLll a nostro distinguemus) but is nevertheless available in the nominative Sin- gular [Code art. 11 (g)(i), 11 (h) (ii)]. Günther (1860) described C. ra/7/'e//z7for specimens described by RüPPEL (1830) as C. petaurista I. Geoffroy St. Hilaire, 1829 which he assumed were misidentified. rüppelli is an incorrect original spelling and must be corrected as rueppellii [Code art. 32 (d) (i) (2)] which is not a junior homonym of ruppellii Agassiz as they differ by one letter [Code art. 57(f)]. ^^ Appears as C. adusta in the heading of the text and as "Sciaena adusta in Tab. nostra" in the synonymy. The later name is on plate 70. This means that Agassiz directed the preparation of plate 70 and that he changed his opinion on the generic position after the plate had been engraved. ^^ Agassiz (p. 130) indicated „Haemulon canna Cuvier in litt." Cuvier's description of this species (in CuviER & Valenciennes, 1830) appeared before Agassiz's one and has priority . Günther (1859:310) indicated that Agas- siz's species is not the same as Cuvier's. ^' The Illustration appeared in 1829 while the text was distributed in 1831 only. ^° Except where noted, all these specimens have been identified by Agassiz himself under the names that the jars still bear. ^' It is not clear if Agassiz meant he had three specimens. In other cases (Megalops thrissoides, Myletes aureus), he clearly indicated that he had three specimens. I think that he had more than three and that he either indicated the lengths of a few of them or that several had approximately the same length(s). ^^ This specimen has been illustrated by Mahnert (1985: 12, fig. [4]). ^^ The 240 mm TL specimen has broken caudal fin lobes. ^'' Specimens whose sizes do not correspond with those mentioned by Agassiz may be syntypes (Interpretation of the Code. Recommendation 72 B; see text for explanation). ^^ ZSM 26725 are 3 tongue-bones recently received from the Botanische Staatssammlung München; they were part of the material brought back by Spix and Martius. The label reads : "Tongue bones of Pirarucu. They are used in Estado de Para to rasp Pasta Guarana. An Illustration by Martius is in herbarium with Paulinia sorbilis. " These tongues and their use are mentioned by SPIX & Martius (1831 : 1 098). ^^ Labelled Macrodon microlepis. ^^ This specimen, in a very poor State, has been examined by Jacques Gery who concluded that it might be the type of T. chalceus, but that the State of the specimen prohibites an accurate Identification. The specimen was labelled "Salminus sp., coli. Agassiz, Bresil". Its original label had evidently been lost and later replaced (by Godet ?). ^^ Lectotype designated by Gery (1986a). Agassiz indicated that he had a single mutilated specimen. MÜLLER & Troschel (1845: 19, footnoie) claimed to have examined oneofthetypesin München. This supports my hypothesis that Agassiz occasionally selected the described specimen(s) among a larger series. The MHNN specimen corresponds particularly well to Spix's plate. The "mutilation" described by Agassiz is the usual peculiar shape of the caudal fin. Agassiz indicated that he had a single specimen 6". 9". This might possibly mean 162x244 mm but would then completely disagree with the description and plate 50 which indicate a specimen approximately three times lon- ger than wide (187x63 mm). Possibly Agassiz meant 6".9"; that would be 183 mm. Lectotype designated by Ploeg (1986: 69) who assumed that the specimen had been collected by Martins. This is obviously not correct; Spix and Martins were together when they travelled in the Tocantins drainage (see itine- rary in TiEFENBACHER, 1983, and, of course, Spix & Martius, 1823-31 !). See Kullander (1986: 57) for a discussion of this specimen. References Anon. [Oken ?]. 1829. [Review of:] Selecta genera et species piscium, quos in itinere per Brasiliam ...[byj. B. de Spix & L. Agassiz, 1829]. - Isis (Oken) 1829: 715-716 Agassiz, E. C. 1885. Louis Agassiz, his life and correspondence. — McMillan, London, 2 vols, 794 pp. 1886. Louis Agassiz's Leben und Briefwechsel, Autorisirte deutsche Ausgabe von C. Mettenius. — Reimer, Berlin, 448 pp. 1887. Louis Agassiz. Sa vie et sa correspondence. — Mayor, Neuchätel, 619pp. Agassiz, L. 1833 — 43. Recherches sur les poissons fossiles. — Agassiz, Neuchätel, 6 vols 1856. [On the Glanis of Aristotle]. - Proc. Amer. Acad. Sei. Arts, Sei. 3: 325-333 Bailey, R. M. 1951. The authorship of the names proposed in Cuvier and Valenciennes' "Histoire Naturelle des Poissons". - Copeia 1951: 249-251 Bailey, R. M. & D. J. Stewart. 1983. Bagrus Bosc, 1816 (Pisces, Siluriformes): proposal to place on the Official List. Z. N. (S.) 2371. - Bull. Zool. Nom. 40: 167-172 Bleeker, P. 1862. Atlas ichthyologique des Indes Orientales Neerlandaises. 2. Siluroides, Chacoides et Hetero- branches. — Müller, Amsterdam, 112 pp., pls 49 — 101 Bloch, M. E. 1794. Naturgeschichte der ausländischen Fische. Vol. 8. — Morinosche Kunsthandlung, Berlin Blyth, E. 1843. Mr. Blyth's monthly report for December meeting, 1842, with addenda subsequently appended. - J. Asiat. Soc. Bengal 12(2): 925-1 011, 4 pls BOESEMAN, M. 1962. Triodon macropterus versus Triodon bursarius: an attempt to establish the correct name and authorship. - Zool. Meded. 38: 77-85 1983. Some remarks on the South American pimelodid catfish usually known by the name oi Phractocephalus hemiliopterus (Bloch & Schneider) (Pimelodidae, Siluriformes). — Zool. Meded. 57(12): 105 — 114, pls 1—2 Bosc, L. A. G. 1817. Nouveau dictionnaire d'histoire naturelle, etc. Vol. 9. — Paris Briquet, J. 1940. Biographies des botanistes ä Geneve de 1500 ä 1931. - Ber. Schweiz. Bot. Ges. 50a: x + 494 pp., Ipl- Campos, A. A. 1945. Sobre os caracidos do Rio Mogi-Gua^u (Estado de Sao Paulo). — Arq. Zool. Est. Sao Paulo 4:431-484, 3 pls Carvalho, J. C. de M. 1983. Viagem filosöfica pelas Capitanias do Grao Paraä, Rio Negro, Mato Grosso e Cuiabä (1783 — 1793). Uma sintese no seu bicentenärio. — Museu Paraense E. Goeldi, Para, 80 pp. COWAN, C. F. 1969. Cuvier's Regne animal, first edition. - J. Soc. Biblphy Nat. Hist. 5: 219 Cuvier, G. 1816. Le regne animal distribue d'apres son Organisation pour servir de base ä l'histoire naturelle des animaux et d'introduction ä l'anatomie comparee. — Deterville, Paris, 2: xviii + 532 pp. 1822. Das Thierreich eingetheilt nach dem Bau der Thiere, als Grundlage ihrer Naturgeschichte und der ver- gleichende Anatomie. Aus dem Französischen frey übersezt und mit viele Zusätzenversehen von H. R. Schinz. 2. Reptilien, Fische, Weichthiere, Ringelwürmer. - Cotta, Stuttgart & Tübingen, xvi -I- 835 pp. 1 829. Le regne animal, distribue d'apres son Organisation pour servir de base ä l'histoire naturelle des animaux et d'introduction ä l'anatomie comparee. Deuxieme edition. — Deterville, Paris, 2: xv -1-406 pp. Cuvier, G. & A. Valenciennes. 1830. Histoire naturelle des poissons. - Levrault, Paris, 5: xxviii -1-499 pp., pls 100-137 1840. Histoire naturelle des poissons. - Bertrand, Paris, 15: xxxi 4-540 pp., pls 421-455 1847. Histoire naturelle des poissons. - Bertrand, Paris, 19: xix-l- 544 pp., pls 520-556 90 DOR, M. 1984. Checklist of the fishes of the Red Sea. - Israel Academy of Sciences and Humanities, Jerusalem, 437 pp. Eigenmann, C. H. 1903. New genera of South American freshwater fishes, and new names for old genera. - Smithson.Misc. Coli. 45(1 431): 144-148 1909. The freshwater fishes of Patagonia and an examination of the Archiplata-Archhelenis theory. — Rep. Princeton Univ. Exped. Patagonia 1896-1899 3: 225-374, pls 30-37 1910. Catalogueof the freshwater fishes of tropical and southtemperate America. - Rep. Princeton Univ. Ex- ped. Patagonia 1896-1899 3: 375-51 1 Eigenmann, C. H. & R. S. Eigenmann. 1 889. A revision of the edentulous genera of Curimatinae. — Ann. N. Y. Acad. Sei. 4: 409-440 Ferreira, A. R. 1903. Memoria sobre o peixe Pira-Urucü de que ja se remetteram, dous da Villa de Santarem para Real Gabinete de Historia Natural e agora se remettem mais cinco desta Villa de Barcellos, os quaes vao in- cluidos nos cincos caixoens, que constituem parte da sexta remessa do Rio Negro. — Arch. Mus. Nac. Rio de Janeiro 12: 155-158 FiTTKAU, E. J. [ed.] 1983 a. Festschrift zu Ehren von Dr. Johann Baptist Ritter von Spix. - Spixiana, Suppl. 9: 441 pp. 1983 b. Johann Baptist Ritter von Spix. Sein Leben und sein wissenschaftliches Werk. — Spixiana, Suppl. 9: 11-18 Gery, J. 1963. Curimata Walbaum, 1792 (Pisces): proposition ä l'inscription de ce genre ä ITndex Officiel des Noms Rejetes. Z. N. (S.) 1 590. - Bull. Zool. Nom. 20: 390-394 1976. Les genres de Serrasalmidae (Pisces, Characoidei). — Bull. Zool. Mus. Univ. Amst. 5: 47-54 1977. Characoids of the world. - TFH, Neptune, N. J., 672 pp., figs 1986 a. Notes de characologie neotropicale. 1 . Progres dans la systematique des genres Colossoma et Piaractus. - Rev. Fr. Aquariol. 12 (1985 [1986]): 97-102, 6 figs 1986b. Notes de characologie neotropicale. 2. Progres recents dans la connaissance des cynodontines (chara- cides). - Rev. Fr. Aquariol. 13: 61-68, 6 figs [GoDET, P.] 1899. Le Musee d'Histoire Naturelle. - Attinger, Neuchätel, 62 pp. GODET, P. 1907. Le Prof. Louis Agassiz et le Musee d'Histoire Naturelle de Neuchätel. - Bull. Soc. Neuchätel. Sei. Nat. 34: 288-294 Gr£visse, M. 1980. Le bon usage. Grammaire fran^aise avec des remarques sur la langue fran^aise d'aujourd'hui. 1 r""^ edition. - Klett, Stuttgart & Duculot, Paris-Gembloux, xiii + 1 519 pp. Gronow, L. T. 1854. Catalogue of fish eolleeted and deseribed by Laurence Theodor Gronow, now in the British Museum. — British Museum, London, 196 pp. Günther, A. 1859. Catalogue of the Acanthopterygian fishes in the collection of the British Museum. - British Museum, London, 1 : xxxii + 524 pp. 1860. Catalogue of the Acanthopterygian fishes in the collection of the British Museum. — British Museum, London, 2: xxi + 548 pp. 1 864. Catalogue of the fishes in the British Museum. — British Museum, London, 5 : xxii + 455 pp. 1868. Catalogue of the fishes in the British Museum. — British Museum, London, 7: xx + 512 pp. Hoogmoed, M. S. & U. Gruber. 1983. Spix and Wagler type specimens of reptiles and amphibians in the natural history musea in Munich (Germany) and Leiden (The Netherlands). - Spixiana, Suppl. 9: 319—415 HowES, G. J. 1976. The cranial musculature and taxonomy of charaeoid fishes of the tribes Cynodontini and Cha- racini. - Bull. Brit. Mus. Nat. Hist., Zool. 29: 201-248 Humboldt, A. de & A. Bonpland. 1813 — 32. Voyage aux regions equinoxiales du nouveau continent, fait en 1799, 1800, 1801, 1802, 1803 et 1804. Deuxieme partie. Recueil d'observations de Zoologie et d'anatomie comparees faites dans l'Ocean Atiantique, dans l'interieur du nouveau continent et dans la Mer du Sud pendant les annees 1799, 1800, 1801, 1802, 1803 et 1804. - Schoell, Paris, 2: 352 pp., pls 31-51 Humboldt, A. de & A. Valenciennes. 1817—21. Recherches sur les poissons fluviatiles de l'Amerique equino- xiale. pp. 145-216, pls 45-51 in Humboldt & Bonpland, 1813-32 Jeannet, A. 1928. Les poissons fossiles originaux conserves ä l'Institut de Geologie de l'Universite de Neuchätel. - Bull. Soc. Neuchätel. Sei. Nat. 52 (1927 [1928]): 102-124 1929. Additions et rectifications ä la note intitulee : les poissons fossiles originaux conserves ä l'Institut de Geo- logie de l'Universite de Neuchätel. - Bull. Soc. Neuchätel. Sei. Nat. 53 (1928 [1929]): 197-199 Jordan, D. S. & C. H. Gilbert. 1883. Synopsis of the fishes of North America. - Bull. U. S. Natn. Mus. 16 (1882 [1883]): Ivi -M 018 pp. 91 KanazawA, R. H. 1966. The fishes of the genus Osteoglossum with a description of a new species from the Rio Ne- gro. — Ichthyologica 1966: \6\-l72, 4 figs Kaup, J.J. \860. Hoplarchus, neues Genus der Familie Labridae. — Arch. Naturgesch. 26(1): 128—133 KiRBY, W. & W. Spence. 1829. An introduction to entomology, or elements of the natural history of insects. 5th ed. - Longman, London, 3: 731 pp. [not seen] Klimpert, R. 1896. Lexikon der Münzen, Masse, Gewichte, Zählarten und Zeitgröße aller Länder der Erde. Zweite, vielfach verbesserte und vermehrte Auflage. — Khmpert, Bremen, vüi + 429 pp. [1972 fac-simile re- print, Akad. Druck Verlag Anst., Graz] KOTTELAT, M. 1983. Quelques poissons dans des bocaux pp. 41—44, 3 figs. In Dufour, C.; Haenni, J.-P. [eds.]. Louis Agassiz, naturaliste romantique, et les premieres coUections du musee. Musee d'Histoire Naturelle, Neuchätel 1984. Cataloguedes types du Musee d'Histoire Naturelle de Neuchätel. I. Pisces. — Bull. Soc. Neuchätel. Sei. Nat. 107: 143-153 KULLANDER, S. O. 1986. Cichlid fishes of the Amazon River drainage of Peru. — Swedish Museum of Natural Hi- story, Stockholm, 431 pp. Lesueur, C. A. 1819. Notice de quelques poissons decouverts dans les lacs du Haut-Canada, durant l'ete de 1816. - Mem. Mus. Hist. Nat. Paris 5: 148-161, 2 pls 1 821 . Description of a new genus and of several new species of freshwater fish, indigenous of the United States. - J. Acad. Nat. Sei. Philad., N. S. 2: 2-8 Lichtenstein, M. H. C. 1819. Ueber einige neue Arten von Fischen aus der Gattung Silurus. — Zool. Mag. (Wie- dermann) 1 (3): 57—63. [not seen] Linnaeus, C. 1 758. Systema naturae per regna tria naturae, secundum classes, ordines, genera, species, cum charac- teribus, differentiis, synonymis, locis. Tomus I. Editio decima, reformata. — Laurentius Salvius, Holmiae [Stockholm], 824 pp. LURIE, E. 1960. Louis Agassiz, a life in science. — Univ. Press, Chicago, 449 pp. Mahnert, V. 1985. Les candirus ou le danger des baignades exotiques. — Musees de Geneve 253: 8 — 12 Marcgrave, G. 1648. Rerum naturahum historiae. Liber quartus qui agit le piscibus brasiliae. pp. 142 — 182 in Pi- sonis, G. & G. Marcgravi de Liebstad. Historiae naturalis brasiliae. — Hack & Elzevir, Amsterdam & Batavia, 300 pp. Marcou, J. 1896. Life, letters and works of Louis Agassiz. - McMillan, New York & London, 2 vols., 302 + 318pp. Meigen, J. W. 1803. Versuch einerneuen Gattungs Eintheilung der europäischen Zweiflügligen Insekten. — Mag. Insektenk. (lüiger) 2: 259-281 MiTCHiLL, S. L. 1815. The fishes of New York, described and arranged. - Trans. Lit. Phil. Soc. New York 1 (5): 355-492, 6 pls Müller, J. & F. Troschel. 1844. Synopsis generum et specierum familiae characinorum (prodromus descriptionis novorum generum et specierum). — Arch. Naturgesch. 10(1): 81—99 1845. Horae ichthyologiae. Beschreibung und Abbildungen neuer Fische. Die Familie Characinen. — Veit, Berlin, 1-2: 40 pp., 11 pls Myers, G.S. 1964. Abrief sketchof the history ofichthyology in America to the year 1850. - Copeia 1964: 33-41 Opinion 772. 1966. Curimata Walbaum, 1792 (Pisces): rejected as a generic name and placed on the Official Index. - Bull. Zool. Nom. 23: 41-45 Opinion 1403. 1986. Ä^grasBosc, 1816(Osteichthyes,Siluriformes): conserved. - Bull. Zool. Nom. 43: 233-234 Ploeg, A. 1986. The cichlid genus Crenicichla from theTocantins River, State ofParä, Brazil, with descriptions of fournew species (Pisces, Perciformes, Cichlidae). — Beaufortia 36: 57—80 Reinhardt, J. 1849. Nye sydamerikauske ferkrandspishe. — Vid. Medd. Naturh. Foren. Kjöbenhawn 1(3-5): 29-57 RüPPELL, E. 1828 — 30. Atlas zu der Reise im nördlichen Afrika von Eduard Rüppell. Zoologie 4. Fische des rothen Meeres. - Frankfurt am Main, 141 -f- 3 pp., 35 pls Rüssel, P. 1803. Descriptions and figures of two-hundred fishes collected at Vizagapatnam on the coast of Coro- mandel. — Bullmer, London, 2 vols Schneider, J. G. 1801. M. E. Blochii, [. ..] systema ichthyologiae iconibus ex illustratum. Post obitum auctoris opus inchoatum absolvit, correxit, interpolavit Jo. Gottlob Schneider, Saxo. Berlin, 1 x + 584 pp., 1 1 pls Sherborn, C. D. & B. B. Woodward. 1901. The dates of Humboldt and Bonpland's "Voyage". - J. Bot. 39: 202-205 92 Spix, J. B. & L. Agassiz. 1829 — 31. Selecta genera et species piscium quos in itinere per Brasiliam annis MDCCCXVII— MDCCCXX jussu et auspicii Maximiliani Josephi I. Bavariae regis augustissimi peracto col- legit et pingendos curavit Dr. J. B. de Spix [. . .]. — C. Wolf, Monaco [München], i + 2 + xvi + ii + 6 + 138 pp., 98pls Spix, J. B. & C. F. P. Martius. 1 823. Reise in Brasilien auf Befehl Sr. Majestät Maximilian Joseph I. Königs von Bai- ern in den Jahren 1817 bis 1820 gemacht und beschrieben. — Lindauer, München, l:vi+ X +412 pp. 1828. Reisein Brasihen auf Befehl Sr. Majestät MaximiHan Joseph I. Königs von Baiern in den Jahren 181 7 bis 1820 gemacht. — Lentner, München, 2: viii + pp. 413 — 884 1831. Reisein Brasilien auf Befehl Sr. Majestät MaximiHan Joseph I. Königs von Baiern in den Jahren 181 7 bis 1820 gemacht. - Martius & Fleischer, München & Leipzig, 3: viii + pp. 885-1 388 + 40 pp. SuRDEZ, M. 1974. Catalogue des archives Louis Agassiz (1807-1873). - Bull. Soc. Neuchätel. Sei. Nat. 97: 1 -202, plsl-3 SWAINSON, W. 1827. A Synopsis of the birds discovered in Mexico by W. Bullock, F. L. S. and H. S., and Mr. Wil- liam Bullock, jun. - Phil. Mag. Ann. Chem. Math. Astr. Nat. Hist. Gen. Sei. 1(6): 433-442 Terofal, F. 1983. Die Fischausbeute der Brasilien-Expedition 1817-1820 von J. B. v. Spix und C. F. Ph. v. Mar- tius. — Spixiana Suppl. 9: 313—317 Tiefenbacher, L. 1983. Die Brasihenexpedition von J. B. v. Spix und C. F. Ph. v. Martius in den Jahren 1817 bis 1820. - Spixiana, Suppl. 9: 35-42 Tra VASSOS, H. 1946. Contribuigöes para o conhecimento da familia Characidae Gill, 1893. III. Discussäo sobre es generös "Cynodon" Spix, 1829 e "Rhaphiodon" Agassiz, 1829, com novo nome de subfamilia (Classe Acti- noptergii — Ordem Ortareophygii). — Sum. Bras. Biol. 1: 129 — 141 Whitehead, P. J. P. 1967a. The clupeoid fishes described by Lacepede, Cuvier & Valenciennes. - Bull. Br. Mus. Nat. Hist., ZooL, Suppl. 2: 180pp., 11 pls 1 967 b. The dating of the 1 st edition of Cuvier's Le Regne Animal distribue d'apres son Organisation. — J. Soc. Biblphy. Nat. Hist. 4: 300-301 Whitehead, P. J. P. & G. S. Myers. 1971. Problems of nomenclature and dating of Spix and Agassiz's Brazilian fishes (1829-1831). - J. Soc. Biblphy. Nat. Hist. 5: 478-497 Whitley, G. P. 1951. Studies in ichthyology. 15. - Rec. Austr. Mus. 22: 389-408 Maurice Kottelat Zoologische Staatssammlung Münchhausenstr. 21 D-8000 München 60 BRD 93