DISSOLVED MATTER CONTAINED IN RAIN-WATER 



151 



Table VI. — Shewing true average composition of rain-water 

 received in difterent seasons. Results expressed in parts per 

 million. 



Before concluding it is advisable to notice the work of other 

 investigators, that has come under our notice. 



First among the researches on the constituents of rain-water 

 stand those of Messrs. Lawes, Gilbert, and Warington, of 

 Rothamsted, Herts, England. 



Experiments have been carried on since 1853 and results are 

 published in the " Journal of the Royal Agricultural Society of 

 England," (Vol. XVII, 1881, p. 241, and Vol. XIX., 1883, p. 313.) 

 The summary of results contained in the latter paper are here 

 inserted for comparison. 



'■' (1) One hundred and fifty-two analyses of rain, snow, dew, and hoar 

 frost, representing the daily collections from June 22, 1881, to January 5, 

 1882, gave an average of 0.248 of nitrogen as ammonia per million of 

 water ; the extremes observed were 5.491 and .043, The variations are 

 dependent on the richness of the atmosphere in ammonia and on the 

 quantity of the ramfall, the smaller deposits containing the larger propor- 

 tion of ammonia. 



(2) Analyses during two years of fresh monthly mixtures of rain gave an 

 average of 0.316 of nitrogen as ammonia per million. Analyses of fifty 

 monthly mixtures, a considerable number of them old, gave a mean of 0.340. 

 The rain collected in the leaden gauge at Rothamsted generally tends to 

 increase in ammonia by keeping The rainfall of summer is generally richer 

 in ammonia than that of winter. 



(3) The nitrogen as ammonia annually supplied by rain per acre reckoned 

 from the daily determinations of six months is 2.374 lbs. ; from the 

 analysis during two years of fresh monthly mixtures 2.466 lbs; from the 

 analyses of fifty monthly mixtures, many of them old, 2.662 lbs. The 

 nitrogen as nitric acid is apparently, from Frankland and Ways' results, 



