Tulu generally follows Kanarese and has sƒvira and its contracted form sƒra.

     Sanskrit ought to have, either derived sahasra direct from the Kanarese sƒsira, or have formed it by prefixing ‘sa’ to the supposed S‡ras„ni original of the Hindi hazƒr, thousand, which seems to be a corruption of the Tamil ƒyiram.

     It is to be noted, that Tamil has no other alternate word for the numeral ‘thousand’, and Sanskrit is completely dumb as to the history of the word ‘sahasra’.

     As for Dr.P.S.Subrahmanya Sastri's translation of the first two parts of Tolkƒppiyam, which is also included in the Bibliography for the Dravidian Etymological Dictionary, it will be dealt with in the body of the work.

     Dr.Burrow has also committed some mistakes in his ‘Dravidian Studies VI entitled “The loss of initial c/s in South Dravidian”, and included in the ‘Collected Papers on Dravidian Linguistics’ by the same author published by the Centre of Advanced Study in Dravidian Linguistics, Annamalai University, in cyclostyled form.

     The preamble to the Studies reads:

     “A frequent phenomenon in Tamil and Malayalam loan-words from early Indo-Aryan is the dropping of an initial sibilant. More rarely examples occur in Telugu and Kanarese. Cases can be quoted for all the three Sanskrit sibilants, though the immediate source of the borrowings are forms with Prakrit.”

     The opening statement of the preamble is one-sided, and accounts for most of the mistakes committed in the course of the Studies. It takes only the aphetic omission and not the prosthetic addition of letters into consideraion. Further, the learned Professor is not familiar with the root-meaning of many Tamil words, and hence the glaring errors of deriving ‘ƒdai’ from ‘šƒ—a’, ‘„–i’ from ‘šr„–i’, ‘palƒ’ from ‘panaša’ and ‘šadai’ from ‘ja—ƒ’. It is not possible to explain the fallacy of all these derivations in detail here. Space forbids.