VERMONT DAIRYMEN'S ASSOCIATION. 83 



for him to manipulate the sample. So can Brown tamper 

 with samples. Yet if the men are sincere and anxious to 

 know the truth, there are ways in which they can insure accu- 

 racy. Some little time ago the experiment station put out 

 a four page bulletin, — reprinted at the end of this article, — 

 giving - methods of sampling milk and cream. This has 

 been printed in poster form and is to be sent in the spring to 

 every creamery in the state of Vermont with the request that 

 it be posted near the weigh-can. We give three schemes for 

 sampling whereby the patron who desires to check the testing 

 work of the creamery may do so; first, the creamery sample 

 may be halved, second, the creamery man may be required to 

 take duplicate samples, and, third, a patron may take his sam- 

 ple for himself. Neither of these methods of sampling will in- 

 sure absolute accuracy. Errors of omission or commission, of 

 ignorance or intent, may be made. If the creamery samples be 

 halved, if Brown is asked to furnish half of it that it may be 

 sent to the station, it is located, prior to halving, in the control 

 of one of the interested parties, the creamery man, and if he 

 is inclined he may tamper with the sample instead of with 

 the result. If the second method is used, if every time 

 Brown's operative puts a gill of milk into his sample jar he 

 puts one into the jar which the patron holds, the objection 

 may be urged that the sample is in the hands of the other inter- 

 ested party, the patron. If the dairyman takes his own sam- 

 ple at home, he may be ill informed as to necessary precau- 

 tions in sampling, or careless, or, indeed, intentionally deceit- 

 ful, and the sample be not truly representative. In short 

 there is no way in which the station can be certain that the 

 samples sent it are correctly taken. Hence we are careful in 

 our reports to those sending us samples to disclaim all re- 

 sponsibility as to the accuracy of sample-taking. I think, 

 however, that this bulletin, which is to be sent to be posted 

 in every creamery and cheese factory in the state, which is to 

 be mailed by thousands throughout the state to our mailing 

 list, and which concludes this article, will do something to 

 make the samples that come to us more uniform and trust- 

 worthy. 



HI. Variation in Tests within the same Herd, 



Why is it that Smith's milk, taken to Robinson's factory 

 or creamery, varies one month with another? Why does it not 

 tes evenly? 



Several of the reasons cited under the first head obtain here. 



LACTATION CHANGES. 



The change in lactation of the cows is one important reason 



