n8 Appendix. 



pearance of emery, and containing a varying amount of fibrous 

 material. 



From Mr. Williams's letter the following information is 

 gained in regard to the circumstances existing at the time of the 

 occurrence : 



" The wind for two or three days previous to the 5th of 

 February had been blowing strongly from the west, but during 

 the day the weather had been pleasant, and at the time the snow 

 fell there was scarcely any wind what there was being from the 

 south-west. The ground was already covered with snow, from a 

 previous fall. Between the hours of four and six p. M. the large 

 flakes of light snow were noticed to present a strange appearance, 

 coloring the snow as they fell. This continued until it had 

 reached a depth of three or four inches, when the wind came 

 strong, blowing quite a gale from the south-west. The snow then 

 came down clear again." 



Some of this last fall of snow was gathered, as was stated, 

 and evaporated carefully in a suitable place, yielding the sedi- 

 ment which is had for analysis. These are the circumstances 

 attending the fall of snow at Saugatuck. 



The sediment received from Holland resembles very much 

 that from the former place, both in color and in structure, and 

 the condition of the weather and wind at the time was nearly the 

 same. During the fall of this snow at Holland, the atmosphere 

 was noticed to be sensibly warmer; at one locality the thermom- 

 eter rising eight degrees during the half hour the snow was 

 falling, falling again after it had ceased and the clear snow had 

 again commenced. The wind was very light during the fall, and 

 from the south-west. Not only at Holland and Saugatuck was 

 this phenomenon observed, but according to both of these 

 informants it extended over a wide range of territory. In Mich- 

 igan it was noticed from about five miles south of Grand Haveij, 

 in Ottawa County, to Ganges, south about thirty miles, in Alle- 

 gan County. Inland it was noticed at Allegan and other places. 

 It must, therefore, have fallen over an extent of from three to 



