25 



morning or mix it with the morning's milk. The morning's 

 milk will be distinctly better in quality although less in 

 quantity if milking is done three times a day. 



(2) The Alnwick and other tests show that the milk of a 

 few cows may very frequently be under the standard, even 

 when the intervals between milking are nearly even. It is 

 evidently desirable that when a sample is taken from the milk 

 of a few cows, and found to be suspicious, that a prosecution 

 should not be decided on till further investigations are made. 



(3) Appeal to the cow. This is not by any means a safe 

 test. "When the mixed milk of six cows was tested at Seaton 

 Delaval there were two cases in which the variation in amount 

 of fat was over 1 per cent, in the afternoon's milk within 

 twenty-four hours. The Offerton and Broomhaugh results 

 show that the variations from fortnight to fortnight may be 

 considerable. 



(4) Solids not fat These were under the standard at 

 Broomhaugh during July and August, and were low in the 

 autumn at Newton Bigg. They may, therefore^ be under the 

 standard in the autumn, and this might be helped by a number 

 of cows calving at that time, as the Cockle Park tests indicate 

 that these are higher for some time after calving. 



(5) Effects of feeding Although considerable changes 

 have been made in the feeding of the different herds while the 

 tests were in progress, none of these have materially affected 

 the quantity or quality of the milk. There is no doubt also 

 that improper or careless feeding upsets the cows, and in 

 consequence interferes with the quantity and quality of the 

 milk. It is probable that feeding experiments on cows from 

 which the production of large amounts of milk are desired will 

 give most useful results. Some time ago a herd of cows, which 

 was producing milk rather poor in fat, was found to be receiv- 

 ing more food than they were likely to be able to assimilate. 



