judgment; Moses ben Nahman from the right, which symbolises tenderness and mercy. On literature the pretended "Arab influence," if it exist at all, is nowise comparable to that of the Spanish Jews, who can boast that Judah ben Samuel the Levite lives as one of Dante's masters. Judah ranks among the great immortals of the world, and no Arab is fit to loosen the thong of his sandal. But it might very well befall a second-rate man, favoured by fortune and occasion, to head a literary revolution. It was not the case in Spain. The innumerable Spanish-Arab poets, vulgarised by the industry of Schack and interpreted by the genius of Valera, are not merely incomprehensible to us here and now; they were enigmas to most contemporary Arabs, who were necessarily ignorant of what was, to all purposes, a dead language—the elaborate technical vocabulary of Arabic verse. If their own countrymen failed to understand these poets, it would be surprising had their stilted artifice filtered into Castilian. It is unscientific, and almost unreasonable, to assume that what baffles the greatest Arabists of to-day was plain to a wandering mummer a thousand, or even six hundred, years ago. There is, however, a widespread belief that the metrical form of the Castilian _romance_ (a simple lyrico-narrative poem in octosyllabic assonants) derives from Arabic models. This theory is as untenable as that which attributed Provençal rhythms to Arab singers. No less erroneous is the idea that the entire assonantic system is an Arab invention. Not only are assonants common to all Romance languages; they exist in Latin hymns composed centuries before Muhammad's birth, and therefore long before any Arab reached Europe. It is significant that no Arabist believes the legend of the "Arab influence"; for Arabists are not more given than other specialists to belittling the importance of their subject.