202 THE CANADIAN NATURALIST. [June, 



party looking to Nature, the other to Art, for sufficient aGjricultu- 

 ral supplies of this element, in the form of ammonia. Yet both 

 sides must and do admit that each acre of soil receives from nature 

 an annual quantity of ammonia, greater or less as the seasons are 

 more or less propitious; part being supplied by the air, in the 

 manner already explained, part (as we may now fairly presume) 

 being generated within the soil itself, by some reaction analogous 

 to that observed by Schonbein. 



Thus much agreed on, both parties would probably be prepared 

 to admit, as a perfect or typical soil, for the growth of any given 

 rotation of maximum crops, one containing a duly proportioned 

 and available supply of all the cinereals requisite during such 

 rotation ; and on the other hand, receiving from nature, during 

 the same period, a quantity of volatile plant-food, nitrogenous, 

 carbonaceous, and aquatic, precisely corresponding to this cinereal 

 supply. Assuming, of course, the mechanical and physical con- 

 ditions of such a soil to be also typically perfect ; and assuming it, 

 further, to be worked durin^; a series of typical seasons ; it would 

 evidently require only typical manuring; i e., the exact restitu- 

 tion, during each rotation, of the cinereals withdrawn by the crops. 

 This is a proposition to which no one, at the present time, will 

 demur. But in reality, as we all know, these various classes of 

 typical conditions, mechanical, physical, chemical, and climatic, are 

 never simultaneously fulfilled. Each deviation from one or more 

 of them involves a corresponding deviation from typical manur- 

 ing. Hence arises a series of special agricultural cases, as manifold 

 as the changes on a set of bells ; and an accurate knowledge of 

 every condition, in each of any number of cases selected for com- 

 parison, is necessary for their correct interpretation. It is in the 

 midst of these complications that oversights take place, and diflfer- 

 ences creep in. Many of these are wholly irrespective of the nature 

 of the soil. Take for example, two experiments, otherwise (by 

 hypothesis) equal, but made in two different counties or districts, 

 one happening to enjoy, during the growth of the crop, a larger 

 number of hours of unintercepted sunshine than the other ; it is 

 obvious that, notwithstanding the assumed equality on all other 

 points, the results must differ more or less, and may differ very 

 notably, in the two cases. Agiin, assume, for argument's sake, 

 absolute equality in all the external conditions of plant-growth, 

 but a difference in the quality of the seed employed in two trials ; 



