52 Selim Lemsteöm. 



been formed eveiywhere on the clover, both in higher and lower parts of the 

 field; hence, the saine curions phenomenon, that I had previonsly observed in 

 late snmmer or in autumn. Lost in thonghts about tliis phenomenon, the 

 explanation of which 1 had not found yet, I returned to the honse and hère I 

 met one of the men of the estate, with whom I entered into conversation. The 

 sun had now risen some degrees above tlie horizon, and now I noticed, every 

 time I turned my back to it, that the shadow of my Jicad was surrounded by 

 a clear bright lialo, of over 15 cm in breadth, with decreasing intensity of 

 light ontside the halo, and faint signs of coloiu's. 



Not a little surprised at this phenomenon, in a completely clear and trans- 

 parent air, 1 turned to my companion and asked if lie could see this shining 

 halo round the shadow of my head. He answered in the negative, but declared 

 that he saw such a bright halo round the shadow of liis own head. During 

 this conversation \ve had approached a sloping hillside, so that our shadows 

 were considerably lengthened. ïhe heads of our shadows feil on a meadow 

 below and now the halo grew in breadth and intensity of light. 



Hère we evidently had to do with a refraction of light of the same kind 

 as the halos or circles round the sun, and the necessary conséquence of this 

 is that, on this occasion, a numbcr of infinitely small drops of water in liquid 

 or solid form, but completely transparent, was falling to the ground, surround- 

 ing us by such a curions rain \ The importance of this observation for the 

 acciu-acy of the opinion about over-cooled mist, I did, however, not see till 

 later on. 



From the above observations, in connection with general expérience it 

 may be concluded with great probability, that the frosts in autumn present 

 a process similar to that of summer night-frosts and can be counteracted as 

 to their ravages, in a similar way. In spite of the nights being longer, this 

 différence is accorded by tlie atmospheric circumstances prevailing then: the air 

 is lelatively damp, and the wind does not abate completely till some hours 

 before sunrise, so that the time of the autumn night during which plants are 

 in need of shelter, is not much longer than during a summer night. ïhe sky 

 not being clear ail night long, is a general occimence. 



The calculations and conclusions given above with respect to the summer 

 night may, therefore, with some modifications, necessitated by outward circum- 

 stances, be applied dming the autumnalfrosts too. 



' When mentioning this phenomenon to Herr A. Rindell, he said he had seen a simihir pheuo- 

 jnenon, during a journey in Tammela, eue afteruoon when the air was clearing up after rain. 



