1895.] 
F. A. Shillingford —On the Kusi River. 
Memoir in the Philosophical Transactions (Yol. LXXI, p. 87) as well 
as in his letters that, the Coosy had at no distant date flowed past 
the station of Purneah and joined the Ganges 45 miles further down 
than its present junction. ” Then he quotes the extract from Bucha¬ 
nan Hamilton given above, about its being highly probable that the 
Kusi went away eastward to join the Brahmaputra, and adds, “ Indeed 
an attentive study of the successive changes that have taken place 
renders this almost certain, and it is probable that the Oorasagar is the 
mouth by which the combined waters of the Coosy, the Mahanuddee 
and the Atree were originally discharged into the Assam river, ” i.e., 
the present Brahmaputra. 
JDr. W. W. Hunter in his statistical account of Bengal 1 writes : 
Dr. W. Hunter’s ** The Kusi is remarkable for the rapidity of 
views. its stream, the dangerous and uncertain nature 
of its bed, and chiefly for its constant westerly movement; ” and fur¬ 
ther on he explains this westerly movement by saying, “ Within 
historical times, that is, from the beginning of the eighteenth century, 
we have evidence that this river (the Kusi) passed below the town of 
Purneah and then due south to the Ganges. It has since then worked 
across 25 miles of country till at present it forms the western boundary 
of the District * * * where the original bed of the Kosi was, it is 
impossible to state. ” Then referring to Dr. Buchanan Hamilton’s 
theory of the eastward channel of the Kusi quoted above, he says, 
“Dr. Buchanan’s suggestion of the union with the Brahmaputra seems 
less improbable than other parts of his theory. The course of the 
Brahmaputra appears in early days to have run far to the East of 
Karatoya a large ^he district Maimansingh. The Kusi also 
river in early Hindu in its eastward course would first meet the 
times. Karatoya, then a vast river having the 
Atrai and Tista for its affluents. In my account of the district of 
Bogra, I have mentioned the very great importance of this river 
during early Hindu times, both on account of its great volume and 
its sanctity, and I have stated that it marks an ethnical frontier 
clearly defined to the present day. If we assume that the Kusi and 
the Mahananda formerly joined the Karatoya, we have at once an ex¬ 
planation of the great size the latter river once undoubtedly had, and 
we shall also be able to account for the process by which the great 
sandy plain was built up between the Barendra of Rajshahi and the 
Madhupur jungle of Maimansingh, through which the Brahmaputra 
made its way at the beginning of this century. The ethnical frontier 
1 Statistical account of Bengal, Purneah, 1877. 
