about the middle of the nineteenth century, in the eighties, and during the past few years, in which time more folktales have been published than in both the preceding periods. Fernan Caballero was the chief figure in the first period; the second was a period of organization of folklore societies and journal publication. At present important works are constantly appearing by such eminent contemporary folklorists as Cabal, Espinosa, Llano Roza de Ampudia, etc. Furthermore, one notices many translations of Spanish folktales into English, German and French.
I found no definition of a folktale which served as a practical rule by which material could readily be accepted or rejected. The rule finally adopted was to accept all texts that manifest the existence in Spanish oral tradition of material like or bearing close analogy to the body of material included in Aarne's index. To reproduce a tale exactly as told by the folk would mean to reproduce every detail of syntax, form and vocabulary in a text written in phonetic symbols, and with all the vicissitudes of colloquial peasant speech in matters of repetition, omission, illogical and contradictory statements. Such a text would be of interest to the anthropologist, the psychologist, the linguist and the literary historian. Unfortunately, however, I have never seen such a text in print. I assume that such a text is what is referred to as a text "of scientific value". Being unable to find such texts, I was forced to lower my requirements. Even though a tale be written in literary Spanish, English, German or French, if its motives are preserved essentially as they are known to exist in oral tradition, I accept it. If, on the other hand, literary tradition and logic seem to have modified not only the language but also the motives themselves, I reject the tale. Or if the source of the tale is not indicated, and it seems to derive from literary or foreign tradition, I reject it. By saturating myfself in Aarne's index and the standard, wellknown Spanish collections,
