58 Mr Lavvson on the Temperature and Atmosphere 



was such, as to prevent my undertaking again the hourly observations 

 even for a few days. 



The errors of one of the thermometers employed in these observations, 

 have been determined with a good deal of care, nearly in the manner 

 recommended by Professor Forbes, in a paper embracing this subject, 

 in the Philosophical Transactions, and those of the others were obtained 

 by comparing them with this one. The errors of the register thermome- 

 ters are, I believe, known every where to be about the tenth of a degree 

 Fahrenheit, and those of the others certainly considerably within that 

 quantity. All the observations have been corrected for the errors of the 

 corresponding instruments. 



DI0RNAI- CURVE OFTEMPERATDRE AT BARBADOES DURING J UNE, JULY, AND AUGUST 1841. 



The following observations were made at St Ann's, Barbadoes, which 

 lies on the SW. side of the island, in lat. 13° 4' N., and long. 59° 37' 

 W., and is about a mile and a half to the southward of Bridgetown. 

 St Ann's is situated on a pretty extensive flat, resting on one side on 

 the sea-shore. It is elevated about 30 feet above the level of the sea, 

 and is perfectly open to it from the E.S.E. to the XW. To the north- 

 ward and eastward the land rises gradually by a succession of flats to 

 a moderate elevation, but does not seem to offer any obstacle to the free 

 course of the trade winds. The distance from St Ann's to the wind- 

 ward side of the island, between E. and N.N.E., varies from 11 to 13 

 miles. 



The thermometers were placed on the windward side of a building, in- 

 side jealousies, which were closed so as to prevent the action of reflected 

 light or heat, though a current of air at all times passed freely through. 

 Outside the jealousies was a wide verandah, perfectly open to windward, 

 which effectually screened the instruments from the direct action of the 

 sun. The room was occupied as a sleeping apartment, but the jealousies 

 to leeward were constantly open, and arrangements were made to admit 

 of the thermometers being exposed to a constant current of air from 

 without, while the communication of heat from within was as much di- 

 minished as possible. There was never a fire in the room, and lights 

 were employed in it only a few minutes at a time, to read off* the tern- 



