The classifications by Pimentel and Orozco were widely accepted, and were given still wider currency by republication in standard works, such as the classic dictionary of the Nahuatl tongue by Rémi Siméon, in which is defined “La famille _Seri_, dans la Sonora, avec 3 idiomes: le _Seri_, le _Guaima_ et l’_Upanguaima_.”[191] In his ethnographic tableau of the nations and languages of Mexico, M. V. A. Malte-Brun followed Orozco almost literally, save that he emphasized the suggested Caribbean affiliation of the Seri, saying: [191] Dictionnaire de la Langue Nahuatl ou Mexicaine, rédigé d’après les Documents imprimés et Manuscrits les plus authentiques et précédé d’une Introduction; Paris, 1885, p. xviii. They make use of poisoned arrows, and when one studies their manners, their habits, their modes of life, one is tempted to find in them a strong affinity [grande affinité] with the Caribs of the continent and the islands.[192] [192] Tableau de la Distribution ethnographiques des Nations et des Langues au Mexique; Congrès International des Américanistes, Compte-rendu de la Seconde Session, tome II, 1878, p. 37. During the seventies Hubert Howe Bancroft was engaged in collecting material for his monumental series of works, and in arranging the ethnologic data for publication. Of the Seri he wrote: East of the Opata and Pima bajo, on the shores of the Gulf of California, and thence for some distance inland, and also on the island of Tiburon, the Ceri language with its dialects, the Guaymi and Tepoca, is spoken. Few of the words are known, and the excuse given by travelers for not taking vocabularies is, that it was too difficult to catch the sound. It is represented as extremely harsh and guttural in its pronunciation and well suited to the people who speak it, who are described as wild and fierce. It is, so far as known, not related to any of the Mexican linguistic families.[193] [193] The Works of Hubert Howe Bancroft, vol. III (The Native Races, vol. III, 1882, p. 704). The “east” in this quotation is obviously a misprint for west. The only vocabulary of this language which Bancroft was able to find was added (without reference to the aboriginal source); it comprised the eleven words collected by Lavandera and discussed by Ramirez in 1850.[194] [194] Ibid., p. 705.