igoo-i.] Spanish Documents Relative to the Canary Islands. 41 



Montreal. Now that no doubt remains as to the language of the 

 Guanches of the Canaries being Celtic, a new interest is created in 

 the Peruvian problem. The Celtic connection of the Peruvians is not 

 a subject confined to the writer. In 1871, V. F. Lopez, published in 

 Paris and Monte Video, a book entitled, " Les Races aryennes du 

 Perou," in which very learnedly he contended that, in spite of post- 

 positions and other indications of Turanian syntax, the Quichua and 

 cognate Peruvian dialects pertain to the Indo-European family of 

 languages. M. V. Henry, a philologist of some note, made a laudatory, 

 but, at the same time, destructive, criticism of the volume in an article 

 read before the International Congress of Americanists, held at 

 Luxemburg, in 1877. As, however, he therein exhibited no acquain- 

 tance with the special features of Celtic speech, his decision is not to 

 be accepted as final. As to postpositions, he ought to have known 

 that they are as common in Sanscrit as prepositions, an indication that 

 the normal prepositional order of Indo-European language had, in 

 its case, been greatly modified by the intimate presence among its 

 speakers of a population making use of a postpositional Turanian 

 language. If it appear that the twin elements, Celtic and Iberic, of 

 the Canary Islands, migrated together to i\merica, finally reaching 

 Peru, and there insensibly fusing the divergent elements of their two 

 distinct forms of speech, the problem of a language which puts 

 Celtic vocables into Iberic grammatical order need no longer present 

 difficulty. It is granted that postpositions to nouns, the postposition 

 of the word governing the genitive, and similar peculiarities, are not 

 Celtic, but Iberic and Turanian. Moreover, there are many Turanian 

 vocables in Peruvian, and the ruling class in Peru, namely, the Incas, 

 was distinctively Turanian. The unanswerable fact, however, still 

 remains that the larger part of the Peruvian vocabularies are Celtic. 

 Before passing to the consideration of these, it may be stated that a 

 German author, Herr Frenzel, has recently contended for the Celtic 

 origin of the Peruvians and of the Aztecs of Mexico. 



The writer's Peruvian material has not been sufficiently abundant 

 to enable him, as he would gladly have done, to compare the Canary 

 Island words with it to any extent. An intimate acquaintance with 

 the obscurer forms of Peruvian speech would be necessary for such a 

 purpose. Ordinary vocabularies have little to tell of Eisteddfods, 

 rock-pools, piebald goats, stone implements, and problematical plants, 

 such as fill the Guanche lists. He has, therefore, been compelled to 

 compare the usual vocabulary for comparative purposes with corres- 

 ponding terms in the Celtic languages to the amount of some three 



