6 BULLETIN 1281, U. S. DEPAETMENT OF AGRICULTURE 
oats, and old-process linseed meal, all of good quality. The hay 
was average quality alfalfa grown in Colorado, nicely cured, fairly 
uniform in color, and contained a normal proportion of leaves. 
The hay was cut into inch lengths by means of a silage cutter, with 
the blower removed to prevent the loss of the finer leaf particles. 
This cut hay was spread m a uniform layer on a barn floor, and care- 
fully shoveled over a few times, taking special pains to distribute the 
fine material with the coarse as evenly as possible. It was afterwards 
transferred to large sacks and placed in a storage bin. 
The hay was weighed out for 14 days at a time including the diges- 
tion period, and when possible for more than one animal at the same 
time. A sufficient quantity of the cut hay was carefully mixed on 
the loft floor, the quantity for each feed weighed in a metal receptacle, 
transferred to dust-proof cotton bags, and stored until needed. 
The sample for chemical analysis was prepared from small amounts 
withdrawn at each weighing. When the weighing of the rations was 
completed this sample material was chopped finer in a meat chopper, 
then carefully mixed and quartered. This process of mixing and 
reduction was repeated until a sample of the desired size was obtained. 
The sample was then taken to the laboratory for air-drying and 
grinding. The finely ground air-dry samples were kept in glass- 
stoppered jars which, as a precaution against moisture change, were 
sealed with soft wax. 
The grain mixture consisted of: 
Parts 
Wheat bran 300 
Ground oats 300 
Corn meal 300 
Linseed meal (old process) 100 
To each 1,000 parts of this mixture were added 2.22 parts of dairy 
salt. From this mixture the feeds were weighed out for the pre- 
liminary periods. For the digestion periods and for the five days 
preceding them, the feeds were weighed out from the separate com- 
ponents and 5 grams of fine dairy salt were added to each of the half- 
day rations. 
DIGESTION EXPERIMENTS 
The ration of each cow remained unchanged throughout the whole 
investigation. The feeding period for cow No. 631 was 143 days; for 
No. 615, 157 days; and for No. 579, 82 days. 
The spacing of the digestion experiments was fixed by the time of 
the respiration-calorimeter tests which could be carried on only bi- 
weekly. During these respiration-calorimeter tests the urine and 
feces could not be collected separately; therefore, in this experiment, 
the calorimeter test was followed by a 9-day period for the separate 
collection of the urine and feces. 
For this collection of excreta the animal was taken to the dairy 
barn where the excreta were collected by a man stationed behind the 
animals. 
WEIGHING AND SAMPLING EXCRETA 
Each voiding of urine was transferred to a 20-liter bottle, and of 
feces to a covered galvanized-iron box. At the end of each 24 hours 
the excreta were weighed. The urine was then well mixed and a 
sufficient quantity taken to the laboratory for the making up of the 
9-day composite sample as well as for certain determinations made 
on the fresh daily sample. The aliquot weighed out daily for the 
