ON THE METEOROLOGY OF PORT LOUIS. 109 



cussed for the whole period, Liit for such portions of it as have been deemed 

 sufficient for the object in view. 



The instruments are by Newman, J^egretti and Zambra, and Casella, and 

 have been compared with the Greenwich and Kew standards. The barometer, 

 the tube of which has an interior diameter of -564 inch, is, with the thermo- 

 meters, 30 feet above the sea-level. The rain-gauge and solar thermometers 

 are 40 feet above the ground, and the vane of Osier's anemometer 10 feet 

 above the highest point of the building. 



From 1852 to 1859 a sinular series of observations was taken by the 

 Koyal Engineers, in a tower about 400 yards west of the Observatory ; so 

 that the two series embrace a period of 14 years. I confine myself to the 

 second series (1860-66), taken under my own direction. The results are 

 given in 42 Tables, to which I beg to prefix a few remarks, intended to 

 direct attention to some of the more salient features. I begin with the 

 temperature. 



I. Tempekatuee. 



Diiinwl Variation. — The last line but one in Table I. exhibits the mean 

 temperature of the air at 3i a.m., 9^ a.m.. 3| p.m., and 9| p.m. The means 

 for these hours are 75°-50, 77°-59,' 78°-99, and 76°-36, respectively, which 

 gives a mean daUy temperature of 77°- 11. The last line shows the excess 

 or defect of the mean for each observation hour on the mean (77°-ll) of 

 the 10,220 observations taken during the whole period of seven years. 



As the intervals between the observation hours are considerable, it is neces- 

 sary, in order to obtain more complete information regarding the diurnal 

 march of the temjierature, to have recourse to the hourly observations taken 

 on the term-days. The results of these, for a period of four years, are pre- 

 sented in Table II., in which the last column but one gives the mean tem- 

 perature for each hour, commencing with 6 a.m. We perceive that there is 

 a single progression, having one ascending and one descending branch, the 

 temperature gradually increasing from 75°-55 at 6 a.m. to 79°-43 at 1 p.m., 

 and then decreasing till 6 a.m. This progression, it need scarcely be 

 remarked, is dependent on the earth's rotation on its axis with regard to the 

 sun. In the last column is presented the amount by which the mean for 

 each hour falls short of (— ), or exceeds (-|-), the mean for the 24 hours 

 (77°-14). We see that there are nine hours, namely, 10 a.m. to 6 p.m. 

 inclusive, during which the temperature is above the mean for the day, and 

 fifteen hom's during which it is below the mean. The range is 3°-88, The 

 greatest increase in any two hours takes place between 9 and 11 a.m., and 

 amounts to 1°-81, and the greatest faU in any two hours from 3 to 5 p.m., 

 and amounts to 1°. The mean temperature for the day occurs very nearly 

 at 9 A.M. and 7 p.m. 



On inspecting the other columns, which give the diurnal variation for 

 each month, it may be seen that, though the minimum generally occurs at 

 6 A.M., and the maximum at 1 p.m., the epochs of the turning-points vary a 

 little with the season. 



Comparing Tables I. and II., we find that the mean daUy temperature is 

 almost identical in both, being 77°-ll in the one, and 77°-14 in the other, 

 notwithstanding the fewness of the observations in the latter case. 



Greatest Diurnal Range. — Table III. shows the greatest range of tem- 

 perature, on any one day, in each month, obtained fi'om daily observations of 

 the maximum and minimum thermometers for five years (1862-1866). It 

 \\i)l be seen that the greatest range on any one day during that period was 



