184 1. P. Косн. 
temperature in the lowest 200 m, we get a fall in temperature which 
is, however, smaller than the normal — adiabatic — fall of temperature. 
As far as the month of May is concerned, there are no deter- 
minations of the co-efficient of refraction, but judging from the above 
it is to be supposed that the value during this month is very low. 
June stands as the transition between the minimum of the month 
of May and the maximum of the month of July. 
In contradistinction to the maximum of July—August, which 
turned out to be particularly connected with the coasts, the mini- 
mum of the early summer must, from what has been 
stated, be supposed to be found everywhere over land in 
high-arctic regions. | 
As far as the period August 23rd— October 21st is concerned, 
the lines of vision over land show mean values of the co-efficient 
of refraction, whereas the lines of vision over water show markedly 
low values. 
The cause of the fall from the maximum of the summer is to. 
a certain extent common to all lines of vision, that is the cold 
which begins to make itself felt towards the end of August, and 
which makes the water freeze. In this manner a considerable 
amount of heat is discharged, which preferably serves to heat the 
very lowest strata in the air, and there occasions a strong fall in 
temperature in an upward direction. This state of affairs, as was to 
be expected, in particular makes itself felt in the three lines of 
vision above water with mean altitudes of 52m, 15m and 7m re- 
spectively, in that the co-efficients of refraction here assume the 
values : КУ — 0.059, kS — 0.175, КИ + 0.203. In the case of lines of 
vision across land, the freezing of the water cannot come to play 
such an important part, partly because the areas of water and those 
of snow filled with water are comparatively small, and partly be- 
cause the lines of vision here lie higher. Ås far as the lines of vision 
over land are concerned, another circumstance has to be taken into 
consideration, that is the fact that the temperature of the sea winds 
in September is not noticeably different from that of the land winds, 
whereby the special inversion, brought ahout by the sea wind, dis- 
appears. 
The variation of the co-efficient of refraction 
according to the time of day. 
In the preface to these remarks I have accounted for the casual 
character of the observations. The lack of system in the arrange- 
ment makes itself very much felt here where the object is to deter- 
mine the influence of the time of day on the co-efficient of refraction. 
