1838.] 



GS a Marine Insfru?ne?if. 



tenths. The instruments having been in constant use for 3 or 4 years, a 

 slight concavity bad formed in the upper portion of the tube from the 

 trifling evaporation of the oil, allowance was always made for this in 

 every observation, and the due correction applied to the results. The 

 barometer, with which tbe Sympiesometer was compared, was a very 

 beautiful one, by Troughton ; its scale indicated to the hundredths of an 

 inch, and was divided in a manner worthy of its maker. Its attached 

 thermometer was always read off as a check to the other in the Sym- 

 piesometer, and thus as far as possible the chance of error removed. 

 Both instruments were placed in the poop cabin with a free exposure^ 

 the Sympiesometer attached to the bulk liead, which perhaps might 

 make a little difference in temperature, but nothing material, and the 

 barometer, as usual, suspended in jirabols. 



In the third and fourth columns of the register, are inserted the. lati- 

 tude and longitude of the ship, at noon of the day of observation, and 

 in the eighth, the direction of the wind is noted. The remarks on the 

 weather include its state during the whole 24 hours, as marked in the 

 -ship's log, which for the purpose of making these extracts was kindly 

 pla-'ed at my command. The range of the observations extends from 

 lat. 23° 57' S. to 20 39- N. and long. 26° 50- W. to 80° 50- E. so that, 

 in tracing isothermal lines on . the surface of the earth, a considerable 

 number of distinct points will thus be supplied, from the thermometric 

 columns, and though, to aid in the construction of each line, one ele- 

 ment only then is obtained, yet it will not be altogether useless. By a 

 reference to the register of the direction of the wind, it is found, that 

 during the whole time we were between the parallels of latitude 

 300 and 40^, running down about 90^ of longitude, out of 180 observa- 

 tions on the direction of the wind, upwards of f are westerly ; next to 

 these, the northerly prevailed most, and next again the easterly. Of 

 course these directions vary considerably, but under all circumstances 

 at this period of the year, westerly winds prevail in theselatitudes, and 

 are taken advantage of in running down the longitude. 

 ' Passing nov/ on to the main subject of this paper, I have first to re- 

 mark, in regard to the indications of the Sympiesometer, that it invari- 

 ably fell more on the approach of wind than of rain. It took little no- 

 tice of passing showers, though succeeding each other with considerable 

 rapidity, but the moment a squall seemed gathering, the level of the 

 oil began to be lowered, and if it continued gradually sinking, we 

 equally invariably had a strong breeze, and if it sunk very low a heavy 

 gale. An objection, which has some weight, has I believe been stated 

 to the Syropiesometerj arising from its being too delicate, and giving 



