162 AGRICULTURE IN THE TROPICS [Ff. Ill 



many well-considered schemes for agricultural improvement 

 also. Instead of fixing upon a definite system, and adhering 

 to it steadfastly for long periods till it has had a chance of 

 showing results, we treat it only too often like children treat 

 the plants in their gardens, digging them up at frequent 

 intervals to see how they are getting on, and soon throwing 

 them away because they have not grown unnaturally quickly 

 to suit their wishes. 



When we take a survey of native agriculture as a whole, 

 it is easy to see that there are many points in which im- 

 provement is possible. Such are (1) the variety of products 

 cultivated; (2) the kinds or varieties of particular crops cultivated; 

 (3) the methods of cultivation ; (4) the cleanliness of cultivation 

 and freedom from weeds and disease ; (5) the preparation of 

 the produce for sale ; (6) the breed and quality of the stock ; 

 and so on. There cannot be the least doubt that all these are 

 capable of vast improvement, though one still at times hears 

 the contrary stated with regard to at least (3). Speaking 

 broadly, native agriculture is wasteful and inefficient, and 

 urgently needs improvement. But, and this is a point of the 

 most vital and essential nature, we must go slowly, and be 

 quite sure what we are doing, or we shall do more harm than 

 good. The proper course is to find out first of all the actual 

 facts of native knowledge in all agricultural matters, and then 

 to work from these or, in other words, to apply the methods 

 of pure scientific research and induction. The science of 

 agriculture is yet largely in an embryonic condition, and its 

 generalisations are almost entirely based (as in the case of 

 other biological sciences) upon work done in the totally different 

 climates of Europe and North America. To apply these directly 

 to tropical conditions is often to court failure. We must begin 

 again, using the results of temperate-zone work as a guide, and 

 collect facts patiently, group them, make inductions from them, 

 and test these again, until we have built up a sound science of 

 tropical agriculture. Scientific agriculture in the north has now 

 overtaken the empirical knowledge of generations of farmers in 

 many departments, and explained it, using the generalisations 

 thus obtained to deduce further rules of action, but this is by 



