APPENDIX. 
Now the ^bove assumed position of Tombuctoo, falls exactly at the 
given distance from Morocco through Tatta; and 18 short of that from 
Fezzan (Mourzouk) ; but 61 beyond that from Tunis. The coincidence 
therefore of the three lines of distance from the Gambia, from Morocco 
^nd Fezzan, may be regarded as complete, since, in using the same rate 
nearly across the whole continent of Africa, from Cape Verd to Egypt, a 
difference of 18 miles only, arises. And hence, the public mind may well 
be satisfied at present respecting this important position. Whichsoever of 
the two determinations may be right, is of little consequence : but I hold it 
to be more prudent to adhere to that line which is the result of computation 
in detail, and corrected as above, than to the long lines given in the aggre- 
gate, and in which there is more risk of error. And thus I close the subject 
of the position of Tombuctoo; placing it in latitude 16° 30', longitude 
1° 33' east of Greenwich. 
Something, however, is proper to be said, concerning the rate of tra- 
velling adopted on the present occasion, as it differs materially from that 
allowed on long lines of distance, in the Proceedings of the Association, in 
1.790. 1 there allowed 16^ for a single day, but diminished the rate 
according to the length of the lines of distance. It would appear that the 
proportion of diminution allowed, although proper enough in countries, 
wh^re obstacles interpose to change the line of direction, is not applicable 
to that part of Africa, where the great Deserts are crossed in so straight a 
line, as hardly to increase the simple winding, arising on each day's course, 
in any considerable degree ; and that even the simple winding is less than 
elsewhere. And hence i6^, or more, arises on the camel routes, on long 
lines of distance, across the Desert ; and on the pilgrims' routes, where it 
appears the camels travel with light burthens. It was through misconcep- 
tion of this rate, that I placed Tombuctoo, so far to the north. I am igno- 
rant of the cause of M. D'Anville's error. 
The particulars of the geography between Silla and Tombuctoo, are 
fcbpied exactly from Mr. Park's map ; and require but little elucidation, 
jinr e, a large town, is two short journies below Silla : and Tombuctoo, 
twelve still lower down. It would appear that all the journies were con- 
