222 ANNUAL. REPORT OF THE Off. Doc. 



caught. The relatively small amount of milk caught iu the pail is 

 very readily mixed, and the gill taken. The fine wire-mesh strainer 

 distributing the milk into a thousand streams serves to quite an ex- 

 tent to mix it. I do not advocate the automatic device unless the 

 fine wire mesh be used. 



Governor Hoard of Wisconsin states that the first device of that 

 kind used was placed in his creamery, and is yet in vogue. As used 

 by him a hole is punched in the bottom of the conductor running 

 from the weigh can to the vat at a point near the vat. The milk 

 when turned into the weigh can is mixed to quite an extent; the gate 

 is then lifted and it pours out in a rush and it mixes itself running 

 and tumbling over and over, and just as it nears the vat, a drop or so 

 from every pound of milk falls into the jar. The drip obtained as 

 far from the weigh can as possible. 



This method is not only theoretically accurate but has proved to 

 be practically correct in thousands of trials; and it has been found 

 to obviate a large part of the errors and annoyances of sampling. 

 The device has been tried over and over again as against extreme 

 care in sampling, and has proved, I think, correct in every case. It 

 may be mismanaged, but it more surely takes an accurate sample 

 than any other practicable method since the sample in part takes 

 itself, regardless of care or lack on the operative's part. 



APPARATUS. 



A law was passed at the last session of the Vermont legislature 

 which required, among other things, that the Babcock test apparatus 

 used in dividend-making be accurate. I have on this table six 

 • bottles. Three are good and three are bad. Can you tell which is 

 which? The manufacturer ''guaranteed" that all were accurate; yet 

 notwithstanding this guaranty some were excessively inaccurate. 

 Here is an accurate cream bottle. How do we know it is so? Not 

 because the manufacturer says so, but because, in accordance with 

 law, the Vermont station has found out whether it is accurately 

 graduated or not, and certified thereto, if correct, by grinding in- 

 delibly upon the neck of the bottle VtExSt. 



One creamery insisted that we send back all the bottles we found 

 to be incorrect. We did so. I doubt whether they were used, how- 

 ever, afterwards; for we ground indelibly upon six places on each 

 bottle the word BAD. 



We found that one out of every thirty pieces (three per cent.) of 

 the apparatus in use before the law was enacted was inaccurate, 

 some exceedingly so. All the apparatus that is being sold by the 

 Vermonc supply houses to-day is correct, because it is all submitted 

 to our inspectoi's and only the correct pieces shipped them. As it 



