302 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 51 



other hand, one other most remarkable point offers greater difficul- 

 ties. For we find that not only does the average curve for the anti- 

 cyclonic day show in its course a great similarity with the adiabats 

 of saturated air, but this also holds good for the individual curves, 

 some of which indeed agree altogether with such adiabats. 



For the average curve there is indeed not such close agreement, 

 but even for it at least the differential gradients follow almost 

 exactly the same law as does the adiabat of a saturated or ascend- 

 ing mass of air that left the ocean level with a temperature of i8°C. 



For the sake of comparison this adiabat is shown by dashes in 

 the diagram of annual means for various elements, in fig. 43 of the 

 third section of this memoir. We see at once that it needs only a 

 horizontal displacement of 8° toward the left in order to make this 

 adiabat coincide with the curve t m . 



This comparison is of great interest because it shows strikingly 

 how far the observed change of average temperature along the ver- 

 tical departs from the average of the adiabats of the dry stage and 

 the condensation stage, and how the temperature diminution with 

 altitude, as actually shown by observation, is not only in the lower, 

 but also in the median strata, much smaller than if it were exclu- 

 sively the consequence of the play of ascending and descending cur- 

 rents. 



The above-quoted cases in which the gradient of temperature 

 follows the type theoretically developed, point out the way already 

 indicated to unravel these peculiar relations. They occur always 

 at times of the day and year when the insolation is in excess or at 

 least begins to acquire the greatest importance. 



It is probable that this approach to the adiabat of the dry stage 

 or to unstable equilibrium would be much more frequently observed 

 if ascensions had been more frequently made in the midday hours, 

 whereas for reasons easily understood the later morning hours 

 must preferably be chosen. 4 



At any rate these cases present the proof of the fact that it truly 

 is the insolation and radiation processes at the earth's surface that 

 strongly influence the course of the temperature curve in the lower 

 strata. 



4 The most recent kite ascensions have shown that a temperature gradient 

 of over i° per 100 meters frequently occurs in the lower strata in the morning 

 hours in the warmer season of the year; e. g., compare R. Assmann and A. 

 Berson; Ergebnisse,"etc. (Results of 'the labors'at the Aeronautical Observ- 

 atory in the years 1900 to 1904. Publications of the Royal Meteorological 

 Institute, Berlin, 1902, 1904, 1905. (Note added 1905. W. v. B.) 



