Errors of the Oriental Tables. 85 
the intervening space, is reduced to 7 geographical miles of direct 
distance. He first assumed the latitude of 42° 45’ N. as true, then 
infers the daily rate of direct travelling from this assumed latitude, 
and thence finds the assumed latitude to be true. Now, allowing 
_ the direct distance to be 300 geographical miles, and allowing one- 
half more to be added for inflection, the daily rate of travelling will, 
be 10 geographical miles of road measure between Cashmere and 
Cashgar for a caravan, a rate abundantly sufficient for a route 
through a country the most mountainous imaginable. 
- We find from Elphinston, that 51 days are occupied in the route 
from Cashmere to Yarkund, by Ladauk. Now Yarkund is but 4° 
to the N. and 2° 30’ to the E. of Cashmere ; so that the direct dis-. 
tance is very little more than that between Cashmere and Cashgar. 
Had Rennel allowed 60 cosses to a degree of horizontal distance, 
as he has very properly done in the map of the upper course of the 
Gogra, instead of 42 cosses to a degree, as in the plains of Hindoo- 
stan, he would uot have erred so egregiously in his latitude of Cash- 
gar and censured D’Anville for placing it in 40° N. lat. He has, 
in fact, committed the same error, in adjusting the quantum of 
space between Cashmere and Cashgar, which he charged on Father 
Tieffentaller in the construction of the map of the Gogra. 
Having praised the tables of Nasroddin, Abulfeda, and Ulugh 
Beigh, and regretted that our maps should be so incorrect, whilst 
possessing these tables in our language, as well. as the life of Ti- 
. moor by Sherefeddin, Rennel proceeds to say, that in discussing the 
position of Cashgar, he has entirely laid out of the question the 
Chinese and Tartarian geography of Du Halde. It will be asked 
why has he done so. To this query, Rennel gives the following 
notable answer. Because it does not agree with the Oriental Ta- 
bles above mentioned, and because of the errors committed in the 
wpper course of the Ganges in the map of the Lamas. Unfortu- 
nately for the justice of this severe censure, the latitude and longi- 
tude of the source of the Ganges in their map, are more correct 
than his own, although the Lamas confounded the sources of the 
Indus and Sutlej with those of the Ganges, as has been shown in 
our defence of the Lamas’ map, in the ‘ Critical Researches on Phi- 
lology and Geography,” Dissert. 2d. It must be remembered, that 
if the positions in Western Chinese Tartary, Lesser Bucharia, and 
‘Tibet, are laid down, not from observation, but from routes and 
itineraries, so are those (longitudes at least) laid down in the Ori- 
ental Tables. In this respect, the Tartarian geography of Du Halde 
and these tables are alike. I see no reason why the computed, not 
observed, latitudes and longitudes of the Oriental Tables, should 
be preferred to the computed ones of the Lamas and Tartars. It is 
astronomical observation alone, that can determine the point, and 
not opinion. Itis astonishing that Rennel should have paid no atten- 
tion, whilst determining the comparative merits of both, to the 11th 
volume of the Histoire Generale de la Chine, published at. Paris 
in 1780, where, in the appendix, there is a copious list of longi- 





