TWENTY-FIVE YEARS OF CHEMICAL INVESTIGATION'. II3 



place having been found to be some twenty per cent, richer in 

 fat than that at Cape Town. 



There are numerous other points that could have been 

 brought forward in illustration of the idea with which this 

 paper has been drawn up. In its opening paragraphs I sug- 

 gested that the most valuable results are seldom, if ever, seized 

 upon by eagle-swoops, but are the outcome of steady progress 

 along a seemingly monotonous way; and I have just pointed, 

 without entering into details, to the thousands of milk analyses 

 which have been performed at Cape Town and Grahamstown, 

 and from which we have built up a considerable amount of 

 knowledge of the average composition of milk in various parts 

 of the Cape Province. Only the other day, at an Agricultural 

 Congress, complaint was made by a Transvaal representative 

 that there had been no similar milk investigation in that Province. 

 I trust that ere long I shall have the opportunity of rendering 

 such a complaint out of date, but let me press the point that we 

 can get the information zuc seek only by carrying out a large 

 number of analyses zvith monotonous insistence, and the sooner 

 we begin the sooner shall we arrive at a sufficient accumulation 

 of data to enable us to express definite conclusions. 



That there are other aspects of research and investigation 

 of course I am perfectly well aware. Research in pure chemistry 

 is by no means encouraged as it should be, in spite of the dazzling 

 results that sometimes, as it were by purest accident, seem to 

 follow from investigations begun without any such expectations 

 — for instance, when William Perkin, endeavouring to synthesise 

 quinine, happened to discover mauve, and so opened the door 

 to the vast and wonderful coal-tar colour industry. There is 

 an opposite danger in this country, and it is the danger to which 

 I have alluded — the danger of being, in spite of the names by 

 which we call our operations, ultra-practical in our so-called 

 research, and so missing the fundamental units by which the 

 really valuable investigation that we need will he built up. About 

 eighteen months ago the editor of the United States Experiment 

 Station Record deplored the lack of a treatise on horticulture 

 in its scientific aspects. 



"What a help [he said] such a book would be to both the teacher 

 and the investigator ! It would give the status of science in horticulture 

 in such a way as to furnish a starting-point for original and productive 

 investigation, and something to build on to. 



I do not know whether the analogy is grasped, but 

 much of what some have been calling our " routine 

 work " is, with us, supplying precisely such a want as that 

 deplored in the sentence just quoted. Collated in the way 

 that we have been doing and hope to do in a greater de- 

 gree — for we have in hand a great deal of uncollated data — 

 these many thousands of analyses, by their cumulative and pro- 

 gressive effect, will form a broad foundation on which to build 

 such a structure of investigation as the country needs. Here, 

 as in other connections, '" manv a mickle makes a muckle." Do 



