DAIRY FARMING DAIRYING. 1021 



"(3) Three per cent iucroaso iu the total dry matter eaten in the form of Athis 

 gluten meal prodnced 15 per cent more milk, IG per cent more total solids, and lit per 

 cent more butter fat. 



"(4) There appeared to he a slight tendency toward richer milk when the Atlas 

 gluten meal was fed. The ratio of fat to solids-uot-fat wna lowered as :in elVect of 

 feeding this meal, it resembling the other richer glutens in this respect.' The 

 change, however, is not large enough to have much practical signilicanco. 



"(5) One hundi-ed pounds of dry matter in the form of Atlas gluten meal, in the 

 place of an equal amount of dry matter in the forui of equal parts of corn meal and 

 bran, increased the yield of milk and total solids an eighth, and tiiat of flid Itnttor 

 fat a sixth. 



"(6) Assuming the correctness of the coefficient, the digestible dry matter in Atlas 

 gluten meal proved far more efhcient as a milk and V)utter prodiu^er iu tliese tests 

 than did that of corn meal and bran. 



"(7) [Atlas meal] seems to be a safe form of concentrated feed. Wo have not had 

 trouble in the matter of garget as with some regular gluten meals. [At $16 per ton] 

 it is unquestionably the cheapest protein now upon the market." 



Cotton- seed feed i^s.corn meal and hran. — A comparison oftbese mate- 

 rials on 4 cows dnring 5 periods of 4 weeks each ^ave the following- 

 results : 



"(1) One per cent increase in the total dry matter eaten (but 3 per cent decrease 

 in that eaten in the cotton-seed bran), produced 3 per cent less milk and total solids, 

 and 4 per cent less butter fat. 



"(2) Two per cent decrease in the total dry matter eateu (but 1 per cent increase 

 in that eateu in corn meal and bran), produce<l 2 per cent luore milk, autl 1 i>er cent 

 more total solids and butter fat. 



" (3) One per cent increase in the total dry matter eaten (but 2 per cent decrease 

 in that eaten in cotton-seed l)ran), jjrodnced 2 per cent less milk, 3 per cent less total 

 solids, and 4 per cent less butter fat. 



"(1) There was no material change iu the quality of the milk i»roduce<l by the 

 two rations. 



"(.5) One hundred i)Ounds of dry matter iu the form of cotton-seed bran ])r(>duced 

 less milk, total solids, and butter fat, than did an equal weight of dry matter as 

 corn meal and bran. The ditlerences, however, were not great. 



"(6) Cotton-seed bran may be assumed to be about half digestible. Tiiis would 

 make the digestible dry matter of cotton-seed bran much superior to that of equal 

 weights of corn meal and bran. 



"(7) Cotton-seed bran (or feed) does not appear from these results, which are 

 similar to those obtained elsewhere- to be worth the price asked as a milk-making 

 food.'' 



Exjyerimeiital error ill feeilin;/ U'st.s.— To study tliis 5 cows were fed 

 a constant ration of Lay, mixed grain, and sila.uc duriiif;- the period 

 covered by the above experiments. 



"A pound of total dry matter produced no more milk and butter at erne time than 

 at anotlier, lactation stages being e(inalized. . . . 



"Apparently if the aniuials fur feeding tests are carefully selected, antl a sufficient 

 number are used, the 'experimental error' may be nearly disregarded. If but two 

 or three animals are used, it will hardly bo safe to dogmatically assert that tluctna- 

 tious of the product are of necessity due to changes iu the character of the feeding, 

 unless they exceed 4 or 5 per cent of the larger product." 



I See Bulletin -IS of the station (E. S. R., 8, p. 972). 

 n'ennsylvauia Sta. Rpt. ISDl, p. 44 (E. S. R., 7, p. 1)85). 



