1867.] 



Atmospheric Tide at Melbourne. 



497 



In both cases, at the hours followuig, and at those preceding the meri- 

 dian passage, the minimum is decidedly at the 3rd hour, while the 

 maximun in the first series occurs at the 0th, and in the second at the 1st 

 hour, kt Singapore and St. Helena, both within the tropics, the lunar- 

 diurnal variation shows a maximum *at the 0th and a mhiimum at the 6th 

 hour. The discussions, based on observations made at Prague, and already 

 referred to above, exhibit a greater conformity in respect to the lunar 

 tides at Melbourne than any of the tropical stations. This conformity is 

 especially clearly expressed in the series for the hours following the 

 meridian passage" (which series seems to present in each respective case 

 the greatest reliability), and we observe that the minimum occurs at the 

 3rd and 4th, and the maximum at the Gth hour. But in turning Prof. 

 Kreil's labours in this direction to account, we mxust remember that they 

 refer to a period of only one year, and cannot be considered as present- 

 ing great guarantees for decisive results, especially when considering that 

 so high a latitude as 50° 8' N. would rather have required a longer period 

 of observation than is necessary to prove the existence and character of 

 the lunar atmospheric tides within tlie tropics. So very few discussions on 

 this topic being at our command, it is nevertheless of considerable interest 

 to compare the results for Prague with those at St. Helena, Singapore, and 

 Melbourne, as done in the following little Table : — 





Mean of three 

 jears at 

 Singapore 

 (+1° 19')- 



Mean of two 

 years at 

 St. Helena 

 (-15° 57')- 



Mean of five 



years at 

 Melbourne 

 (-37" 48')- 



Mean of one j 

 year at j 

 Prague 1 



( + 50° 8'). ! 



II 



in. 



in. 



in. 



in. 



h 





4-o'oo57o 



4-0-00365 



4-0-0C080 



o-ocooo 









-f -00475 



4- -00336 



4- -00121 



4- -00043 



I 



2 



+ -00330 



4- -00275 



4- -0C073 



4- •00080 



2 



3 



4- -00280 



4- -00158 



-00000 



4- '00039 



3 



4 



+ -00145 



4- -coiio 



4- -coc6i 



4- -00005 



4 



5 



+ -00035 



4- -00046 



4- -00045' 



4- -00032 



5 



6 



•ocooo 



•cocoo 



4- -00062 



4- -00078 



6 



Mean 



4- -C02621 



4- -C01843 



4- 0-00631 



4- "000396 



Mean. 



The decrease in extent of oscillation, as we recede from the equator, is 

 clearly illustrated by the mean values of this Table. 



Speaking of the extent of the oscillations, it is of im])ortance to add a 

 few facts relative to the amplitude, as resulting from the monthly curve?. 

 We have seen, in the course of this exposition, that the amplitude for the 

 semiannual periods from April to September, and from October to March, 

 is respectively 0"'006r)3 and 0"-00432, which result will be materially 

 altered in case we consider only tlie single months ; for inasmuch as the 

 sense of oscillation varies considerably in the single months, constituting a 

 semiannual period, chiefly during summer, the combination of the hourly 

 values of six months in one group must necessarily tend to diminish, or 



