318 THE MECHANICS OF THE EARTH'S ATMOSPHERE. 



when be reflects that all the hypotheses introduced into the computa- 

 tion (the neglect of the vertical motion, of the friction, and of the dif- 

 ference between a day and tbe interval between two luuar culmina- 

 tions) contribute to increase the computed tidal ebb and flow. With 

 reference to the last mentioned difference, I might further remark that 

 it is easily introduced into the computation. The ratio 3 to 1 between 

 the lunar tide and the solar tide as assumed by Laplace holds good 

 only so long as the value 4fc (in which the depth of the ocean or tbe 

 temperature of the air enters in the respective problems) is far from a 

 certain critical value which lies between 11.1 and 11.2. With 4ft=10 

 the ratio in question is 2.2 to 1, but with 4fc=ll.l the ratio becomes 1 

 to 5. 



I do not carry out the computation here, because it seems too hypo- 

 thetical to compare the atmosphere with a spherical shell of perfectly 

 definite temperature, and under this assumption then to consider this 

 semi-diurnal variation of pressure as a consequence of the solar attrac- 

 tion. Much more probable is it that it arises from a regular constitu- 

 ent of the semi-diurual variation of temperature. 



