On the Rain and Drainage -Waters at Rothamsted. 267 
Chlorides reach their maximum in the rain collected at the sea- 
coast. In the rain from Valentia, on the west coast of Ireland, 
Dr. A. Smith found 47'35 parts of chlorine per million. 
Few quantitative determinations have been made of the amount 
of chlorides furnished to the soil by rain. A long series of 
observations on this subject has been carried out at the Royal 
Agricultural College, Cirencester. The determinations were 
commenced in 1870 by Professor Church, and have since been 
continued by Professor Prevost. Through the kindness of 
these gentlemen we are able to refer to the results. The mean 
rainfall from October to March during ten years has been 16'801 
inches, containing an average of 5' 11 parts of chlorine per 
million of water. The mean rainfall from April to September 
has been 16*770 inches, containing an average of 3'46 parts of 
chlorine per million. For the whole year the mean rainfall has 
been 33"571 inches, the chlorine amounting to 4"28 parts per 
million, equal to 53'66 lbs. of pure common salt per acre. It 
will be observed that the amount of chlorine is, on an average, 
greatest in the winter months ; this, however, is largely due to 
the result obtained in one winter of the series. Cirencester is 
about 35 miles from the Bristol Channel ; the amount of 
chlorides in the rain is thus naturally higher than at Rothamsted. 
The amount of ammonia and nitric acid supplied to the soil 
in rain in the course of a year has been determined in various 
parts of the continent of Europe. The results of twenty-two 
determinations, each extending over a whole year, and made at 
eleven different stations, will be found in Table XVIII. (p. 268). 
The large amount of nitric acid found in many of these rain- 
waters is remarkable. It would appear also, from the results at 
Regenwalde and Montsouris, that the quantity of nitric acid may 
vary extremely from year to year in the same place. Are we 
to assume that the large quantities of nitric acid indicated have 
been produced by the union of atmospheric nitrogen and oxygen 
under electrical influences, or chiefly through the oxidation of the 
ammonia of the air by means of ozone or peroxide of hydrogen? 
The latter alternative seems the most probable. 
It is seen that the numerous widely varying determinations, 
some made in the vicinity of towns, give a mean of 10*23 lbs. 
of combined nitrogen annually supplied per acre by rain, with a 
mean rainfall of 27'03 inches. The two years' determinations 
of both the ammonia and the nitric acid in the rain at Rotham- 
sted give, as has been seen, 6"58 lbs. and 8'00 lbs. of combined 
nitrogen as the annual supply per acre from the same source. 
Making all allowance for far inland open country positions 
on the one hand, and for proximity to towns on the other, the 
very small amounts of combined nitrogen so supplied per acre 
