182 EXPERIMENTAL FARMS 



2 GEORGE V. r A. 1912 



nutritive value, making it highly desirable to purchase on guaranteed analysis as to 

 protein and oil-content. From the results tabulated, it will be observed the range 

 in protein is from 31-3 to 37-7 per cent, and in oil from 8-4 to 11-9 per cent. 



Considerable care and scrutiny should be exercised in purchasing cotton seed 

 cakes and cotton seed meals, since great variations in composition exist among the 

 many brands on the market, not a few of which have been found of very inferior 

 quality. These inferior grades are characterized by the presence of excessive amounts 

 of hull discernible as dark, coarse, hard fragments, or by containing cotton fibre, very 

 easily detected from the woolly appearance of the meal or of the cake when freshly 

 broken. Much hull or husk is most undesirable, as it means a meal not only low in 

 protein and oil but one of impaired digestibility and possessing marked astringency. 

 Good cake or meal from thoroughly decorticated (hulled) seed should not contain 

 more than seven per cent of fibre, with a protein-content of about 40 per cent and 

 8 to 10 per cent of oil. A few years ago, 15 per cent of oil was not uncommon. Unde- 

 corticated cake, made from the unhulled seed, may not contain more than 25 per cent 

 protein and perhaps 5 per cent of oil, with an excessive amount, 20 to 30 per cent, 

 of fibre. 



The presence of cotton fibre arises from imperfect ' ginning ' the fine downy 

 layer immediately covering the seed being somewhat difficult to remove. It is a most 

 objectionable feature, not merely in lowering the amounts of the important nutrients 

 present but rendering the meal undesirable generally, and quite unsuitable as a feed- 

 ing material for young stock. 



Of the samples recently examined the greater number are of very fair quality, 

 though the data make it evident that a guarantee as to protein, oil and fibre is neces- 

 sary even when purchasing meal or cake of high grade. Thus, for example, comparing 

 Nos. 8465 and 8836, the former possesses but 1-7 per cent protein and 1 per cent oil 

 less than the latter. We notice, however, a very decided difference in fibre which most 

 markedly emphasizes the superior feeding value of the latter sample. 



Samples of undecorticated meals, Nos. 8875 and 8939 are distinctly inferior grades. 

 Their protein-content is practically but half that in good brands, and their fibre is 

 excessively high, due to the presence of both hulls and cotton. Though careful in- 

 spection would have revealed their inferior character, it would have been impossible 

 to classify or rank properly such samples without the analytical data and hence we 

 again urge purchasers to insist on a certified or guaranteed analysis, carefully com- 

 paring the figures of competing brands with the standards here outlined. A very 

 considerable saving may frequently be effected, not necessarily by buying the cheapest 

 brand in the market, but one presenting the greatest food value for the money. It is 

 the analysis that furnishes the very best basis for comparison, and in this case we 

 must not only look for high protein and oil but also see that the fibre-content is not 

 excessive, remembering that a high percentage not merely lowers the content of the 

 more valuable nutrients, but at the same time depresses their nutritive value. 



Burxet — (Potcrium Canadense). 



This forage plant has been used in Newfoundland as sheep pasturage, for which 

 according to our correspondent, Mr. Chas. Dodd, The Log Cabin, Nfid., it has a good 

 reputation. He writes ' this plant forms the bulk of my sheep pasture and I should 

 be glad to know its true feeding value. I believe it is the plant that causes our 

 cariboo to fatten so quickly. It gives a big crop from the middle or end of May till 

 late in the autumn.' 



As received, the sample consisted of stems with leaves attached, quite green, 

 fresh and succulent. The plants had been cut close to the ground and were about 18 

 inches in length. When air-dried they were found to be brittle, but not tough, and 

 apparently not unusually fibrous. Our analysis afforded the following data: — 



