54 Selim Lemström. 



If there is much dampness in the air, so intense a condensation of it takes 

 place, immediately after sunset, that a thin cloudy veil of niist is formed, 

 snflicient to restrain the radiation of heat from the ground. The nature of 

 this veil is denoted by the colour of the sky. After an otherwise clear day, 

 there appears at sunset, on the opposite side of the sky, S. or S.W., a greyish 

 lilac colour which, wlien the sun has disappeared, becomes more perceptible 

 and rises higher and higher, until, changing into a bluish-grey tint, it finally 

 spreads o ver ail the sky. 



This mist always arises after warm, suunner days and is usually a suffi- 

 cient protection against frost. The more the blue tone prédominâtes, the greater 

 is the danger of frost, and \ve hâve already proposed (page 2) that thèse co- 

 lours should be referred to certain types, such as blue, bluish-grey, greyish-hlue 

 and (jrey, and printed on coloured plates, with which the tlien prevailing co- 

 lour of the sky can be compared. We should then hâve the following scale 

 of colours, which during a calm night might be applied: - blue - threatening 

 frost, bhiish grey = possible frost, greyish blue = frost not probable, and grcy 

 — no danger of frost. 



The surest forecasts must, however, be made from the hygrometer, ther- 

 mometer and barometer, taken in connection with a careful study of the other 

 meteorological circumstances, but thèse failing, the scale of colours given above 

 may be used. It will, in some measure, even complète other meteorological 

 observations. 



The general, wide-spread frosts are always preceded by 2 or 3, (usually 

 3), cold windy days, and there is no doubt that for us the chief question is 

 to prevent their ravages, the injury done by the local frosts being slight in 

 comparison with the general famine which foUows in the footsteps of the first- 

 named. The latter kind of frost also very offen dépends on local circumstances 

 which may be alteied and improved by fore-thought and attention, and by a 

 close study of the warnings hère pointed out, they may be foreseen with accu- 

 racy. To the lirst description belongs the frost which we may term the most 

 dangerous, i. e., when the dew-point is below zéro. 



Although no hygrométrie measurements were taken, it was easy to déter- 

 mine, from the température, the state of the wind, and the small amount of 

 dew, that the dew-point on the nights of the 19* and 20*'' June was very 

 little above freezing-point, which had been foretold by the blue tone of the 

 sky, and this was the cause of a greater danger of frost. 



When the dew-point is below zéro, there is no fall of dew, neither is 

 there any sheltering condensation, so that the vegetation freezes before any 



