154 PROVISIONAL METHODS FOR ANALYSIS OF FOODS. 



Page 56, 10. Under determination of volatile and nonvolatile ether extracts I 

 think he consumes too much time. He extracts with a continuous apparatus for 

 twenty hours. /. A. Wesener. 



Comment by Mr. Winton. The method for volatile and nonvolatile ether extract 

 takes time, but I have found no satisfactory short cuts. Furthermore, our proposed 

 standards are based on this method. 



Page 56, 11. Alcohol extracts from pepper a large amount (10 to 12 per cent) of 

 matter which is easily volatile from its alcoholic solution, but is not volatile at the 

 same temperature by dry heating. (See Bui. 10, Inland Eevenue Laboratory, p, 18. ) 

 A. McGill 



Comment by Mr. Winton. A matter worthy of future study. 



Page 57, 12. The method we are using and one we much prefer to the copper- 

 reducing method for sugars is to conduct the determination similarly to the copper- 

 plating method until the copper suboxid has been dissolved in nitric acid. Then 

 proceed as follows: Add sulphuric acid to displace the excess of nitric acid and boil 

 until all fumes of nitric dioxid have disappeared. Then neutralize with sodium car- 

 bonate. Redissolve the precipitated copper in acetic acid, boil to expel excess of 

 that acid, cool, and add enough potassium iodid to change all of the copper to 

 cuprous iodid, with the liberation of free iodin. The free iodin thus liberated is 

 then titrated with deci-normal ammonium thiosulphate, using starch as indicator. 

 This method in our laboratory is both rapid and accurate. /. A. Wesener. 



Comment by Mr. Winton.^The association gives choice of several methods for 

 determining of reduced copper, any of which may be employed. Personally,. I 

 prefer weighing either the CuO or Cu 2 O in a Gooch crucible. 



Page 57, 12. In following Allihn's method, you direct that the solution, after 

 addition of reducing solution, be heated merely to boiling, whereas in the A. 0. A. C. 

 version the solution is boiled two minutes. W. D. Bigelow. 



Comment by Mr. Winton. The table used in connection with the Allihn method is 

 based on Allihn's experiments with different quantities of pure dextrose, heating in 

 each case until boiling begins again. So long as we use Allihn's table, we should 

 follow his instructions. The writer, in a long series of experiments 11 with both pure 

 starch and dextrose, found that the original Allihn method gave accurate results, 

 whereas longer boiling brought the results too high. The German food analysts 

 use the original Allihn method. 



Page 57, 12. We have found it better to make the copper solution very slightly 

 acid. We include 1 cc of strong sulphuric acid in the 500 cc of copper sulphate 

 solution. A. G. Woodman. 



Comment by Mr. Winton. A subject for study by the association. 



Page 58, 14. Several inquiries have been received regarding the details of the 

 method prescribed for the determination of crude fiber. Paper always gives a clear 

 filtrate and nearly filters rapidly, whereas linen does not always retain the fine 

 material and Gooch crucibles sometimes clog, rendering filtration impossible. A 

 possible disadvantage of a weighed paper is that it may lose weight on treatment 

 with soda, but this loss can be determined by blank experiment and a correction 

 introduced. Of course a Gooch crucible is more convenient, provided it does not 

 clog, but to make its use obligatory renders the method impossible for many sam- 

 ples. A. L. Winton. 



Page 58, 14. In the determination of crude fiber, the centrifuge will be found a 



Conn. Expt. Sta. Kept., 1897, p. 128; Jour. Anal. Chem., 1888, p. 129. 



