There has not been any systematic study as to how the loan words of Indic origin got introduced into the Old Uigur language. In this paper I attempted to describe the differences between forms of Indie origin in Uigur documents and their counterparts in Sanskrit, to give a systematic description of the fact that the differences directly reflect the forms in the intermediary languages, although they can be supposed to have originally come from Indic languages other than Sanskrit.
For this purpose the following steps were taken. Firstly, I paid particular attention to the finals of the relevant forms. Although they show great differences from the Sanskrit counterparts (as basic forms for comparison), it was found that with a few exceptions the differences show a certain regularity with respect to the relation between the Uigur finals and the phonological characteristics of the Sanskrit finals, and between the Uigur finals and the meanings of these Uigur words. This regularity in turn can be shown to conform to the relation which Tocharian exhibits with the forms of Indie origin. It was also found that a small number of Uigur words of Indic origin which have not the 'canonical forms' have also deviant forms in Tocharian.
Secondly, except for the finals, there are also basic similarities with Tocharian forms. That is, forms which differ from their Sanskrit counterparts in Uigur are also found to show parallel differences from Sanskrit in Tocharian.
In both cases (word finals and non-finals), some Uigur forms do not accord with the Tocharian forms of the same word. They are variants due to the nature of the documents they appear in, and very few are incorporated into the Uigur language. These forms came through Sogdian or some other languages.
From the above procedure it was sufficiently demonstrated that the loan words of Indic origin came essentially through Tocharian. These loan words are later borrowed into Mongolian from Uigur directly, a fact which suggests to us the possibility of reconstructing forms, unattested in Tocharian and Uigur from Mongolian forms; e.x. Mong. bandab (name of a mountain, Skt. pāṇḍava)<Uig. *pandap～*pantap<Toch.*pāṇḍap.
In concluding, I would like to point out that from the research on the loan words of Indic origin we can see that the Uigur Buddhism in its early period was closely connected with Tocharian Buddhism.