PEODUCING MILK IN WESTERN WASHINGTON. 3 
watch in hand he noted and recorded the exact minute each labor 
operation connected with the dairy was begun and ended. The labor 
operations during the next morning were recorded in the same man- 
ner, to complete the 24-hour period. 
The field agent also noted the feeds that were being fed on the day 
of his visit, recording the kind, quantity, cost, and description of 
each and comparing them with the quantity recorded by the cow 
tester in the cow-testing-association books. 
The quantity of milk sold and receipts each month were obtained. 
In addition the whole milk used by the proprietor and his help or 
fed to calves was measured or weighed and used as a basis for deter- 
mining the quantity kept on the farm during the month. 
Each dairyman kept an itemized account of expenses that were 
incurred between the monthly visits, and these items were recorded 
by the field agent. A monthly record was also kept of the purchase 
or sale of cows, calves, hides, outside bull service, and other miscel- 
laneous information relating to the herd. The condition and method 
of handling the manure were noted and reported each month. 
When all the labor operations about the dairy had been completed 
for the day at the first farm, the field agent drove to the second farm 
in time to observe the labor operations connected with the evening 
chores. This program was followed until Saturday afternoon, when 
he returned to his headquarters and finished his reports for the week's 
work. The same program was followed each week in the month, and 
each farm was visited for a full day every 30 days throughout the 
two years. 
COMPARISON OF WINTER AND SUMMER RESULTS. 
Since the winter and summer seasons have a marked influence on 
the principal factors entering into the cost of producing milk, the 
results have been computed separately for those periods. The months 
from November to April, inclusive, represent the winter season and 
from May to October the summer season. This division of time was 
based directly on the change in methods of herd management made 
in November and May. 
The various tables found in this bulletin are based upon figures ob- 
tained during two years of study, and the weighted average of the 
two-year records was used whenever it would more accurately express 
the result. The weighted average was obtained by giving each item 
a weight in the average according to its relative importance. 
DESCRIPTION OF HERDS. 
During the first year records were obtained on 17 herds, having an 
average size of 31.3 cows, with an average annual production of 7,369 
pounds of 3.74 per cent milk per cow. During the second year, 18 
herds, 15 of which had been in the first year's work, had an average 
