written language of India. But while Sanskrit was recovering its ancient supremacy, the Prākrits had exercised a lasting influence upon it in two respects. They had supplied its vocabulary with a number of new words, and had transformed into a stress accent the old musical accent which still prevailed after the days of Pāṇini.
In the oldest period of Prākrit, that of the Pāli Açoka inscriptions and the early Buddhistic and Jain literature, two main dialects, the Western and the Eastern, may be distinguished. Between the beginning of our era and about 1000 A.D., mediæval Prākrit, which is still synthetic in character, is divided into four chief dialects. In the west we find Apabhraṃça ("decadent") in the valley of the Indus, and Çaurasenī in the Doab, with Mathurā as its centre. Subdivisions of the latter were Gaurjarī (Gujaratī), Avantī (Western Rājputānī), and Mahārāshṭrī (Eastern Rājputānī). The Eastern Prākrit now appears as Māgadhī, the dialect of Magadha, now Behar, and Ardha-Māgadhī (Half-Māgadhī), with Benares as its centre. These mediæval Prākrits are important in connection with Sanskrit literature, as they are the vernaculars employed by the uneducated classes in the Sanskrit drama.
They are the sources of all the Aryan languages of modern India. From the Apabhraṃça are derived Sindhī, Western Panjābī, and Kashmīrī; from Çaurasenī come Eastern Panjabī and Hindī (the old Avantī) , as well as Gujaratī; while from the two forms of Māgadhī are descended Marāthī on the one hand, and the various dialects of Bengal on the other. These modern vernaculars, which began to develop from about 1000 A.D., are no longer inflexional languages, but are analytical like English, forming an interesting parallel in their develop-