545 



ed. I examine it with Dollond's large telescope. It is al- 

 ready so dark, that I cannot without some difficulty read 

 the divisions on the limb of the instrument. The ground is 

 no doubt getting colder ; but I still perceive air (an aerial 

 space) between the depressed horizon of the sea and the base 

 of the island. 



<e The fifth of September. During the morning twilight. 

 Therm. 21-3°. Hygr. 45-2°. The Sun's disk not yet visi- 

 ble, and already the whole of Picuita appears suspended in 

 the air. Twilight very feeble. (Tn what follows I have in- 

 serted after the degrees the hundred and twentieth parts, as 

 they appeared on the quadrant.) 



A 94° 62 

 B 95° 103-7 

 C 95° 111-2 

 D 95° 1159 

 E 95° 93 

 F 95° 1177 



iC Three in the afternoon. Therm, 23 2° R. Hyg. 362o 

 Deluc. Cyan. 22°. 



B 95° 105*3 

 C 95° 112-7 

 D 95° 1175 

 E 95° 93 

 F 95° 117-5 



st Six in the afternoon. The Sun is only four degrees 

 high. Therm. 22-8° R. Hyg. 36*5°. 

 A 94° 62-2 

 B 95° 104-5 

 C 95° 111 3 

 D 95° 116 

 E 95° 92'7 

 s F 95° 1163 



" During the setting of the Sun, the horizon is oscillating. 

 Picuita sinks, and is no longer suspended. The depression 

 VOL. IIL 2 N 



