142 On the Composition of Waters of Land-Drainage. 



We may learn from this comparison that whilst nitrate of 

 soda is more energetic and rapid in its action than ammonia- 

 salts, it is more liable to be washed out of the land than the 

 latter ; and that, in wet seasons, there is greater risk of losing the 

 benefits of a top-dressing of nitrate of soda than is the case when 

 sulphate of ammonia, guano, or other ammoniacal fertilizers are 

 employed for cereal crops. 



With regard to the distribution of rain, and the general 

 character of the season in 18GG, it maybe observed that January, 

 and the first half of February, were unusually warm, though in 

 January there was a heavy fall of snow, which, however, rapidly 

 thawed, and the whole period was very wet. March on the 

 whole was cold and dry. The beginning of April was cold and 

 rather wet. May was unusually cold, and there was a deficiency 

 of rain. June was changeable, and during this month a con- 

 siderable excess of rain fell. The beginning of July was cold 

 and wet ; then followed a week of hot and dry weather, but 

 from about the middle of the month to nearly the end of 

 September the weather was generally cold, with a good deal 

 of rain and wind in August, and an almost continuous and 

 excessive fall in September. October, however, was drier than 

 usual. A good deal of rain fell about the middle of November 

 and beginning of December, causing a full flow of the drains 

 at the time when the samples of drainage-waters were collected. 



Second Series of Drainage-Watek Experiments. 



The second series of drainage-waters (Table II.) was collected 

 on the 21st of May, 1867. All the drains except that of Plot 2, 

 manured every year with about 14 tons of farmyard-manure, 

 were freely running at the time of collection. Hardly any water 

 passed through the drain on Plot 2 at that time, and a sufficient 

 quantity for the purpose of analysis could not be obtained. 



In the first series of analyses neither potash nor soda were 

 determined separately, nor was an attempt made to weigh the 

 traces of phosphoric acid, which the qualitative examination has 

 shown to exist in water of land-drainage. 



In the second series of analyses both potash and soda were 

 separately determined in all the samples, as well as the amount 

 of carbonic acid combined with basic constituents ; and in three 

 instances the traces of phosphoric acid detected in the water 

 were likewise determined by the delicate molybdate of ammonia 

 process. 



None of the samples contained more than merely faint traces 

 of ammonia. It was not, therefore, considered necessary to 

 determine the amount. 



