The mission of the association is to promote market development goals and government affairs goals for the benefit of Arkansas rice growers. The Arkansas Rice Growers Association is an organization of rice growers, for rice growers, run by rice growers.

Please comment!

Talk about your experiences with the current protocols. Different milling yields you have gotten with the same sample. Your lack of confidence in the Grain Man giving a “true” milling yield. How much you think a whiteness meter will give a more equitable comparison. Whatever your experiences are.


Grain Inspection, Packers and Stockyards Administration

7 CFR Part 868

United States Standards for Rough Rice, Brown Rice for Processing, and Milled Rice

AGENCY: Grain Inspection, Packers and Stockyards Administration, USDA. ACTION: Request for information.

SUMMARY: The United States Department of Agriculture’s (USDA) Grain Inspection, Packers, and Stockyards Administration (GIPSA) is seeking comment from the public regarding the United States (U.S.) Standards for Rough Rice, Brown Rice for Processing, and Milled Rice under the Agriculture Marketing Act of 1946 (AMA). To ensure that standards and official grading practices remain relevant, GIPSA invites interested parties to comment on whether the current rice standards and grading practices need to be changed.

DATES: We will consider comments we receive by March 21, 2016. ADDRESSES: You may submit written or electronic comments on this proposed
rule to:

Mail: Irene Omade, GIPSA, USDA,
STOP 3642, 1400 Independence Avenue SW., Room 2530–B, Washington, DC 20250–3604

Fax: (202) 690–2173
Internet: Go to http:// www.regulations.gov and follow the on-line instructions for submitting comments.

All comments will become a matter of public record and should be identified as ‘‘U.S. Standards for Rough Rice, Brown Rice for Processing, and Milled Rice request for information comments,’’ making reference to the date and page number of this issue of the Federal Register. All comments received become the property of the Federal government, are a part of the public record, and will generally be posted to www.regulations.gov without change. If you send an email comment directly to GIPSA without going through www.regulations.gov, or you submit a comment to GIPSA via fax, the originating email address or telephone number will be automatically captured and included as part of the comment that is placed in the public docket and made available on the Internet. Also, all personal identifying information (for example, name, address, etc.) voluntarily submitted by the commenter may be publicly accessible. Do not submit confidential business information or otherwise sensitive or protected information.
Electronic submissions should avoid the use of special characters, avoid any form of encryption, and be free of any defects or viruses, since these may prevent GIPSA from being able to read and understand, and thus consider your comment.

GIPSA will post a transcript or report summarizing each substantive oral comment that we receive. This would include comments made at any public meetings hosted by GIPSA during the comment period, unless GIPSA publicly announces otherwise.

All comments will also be available for public inspection at the above address during regular business hours (7 CFR 1.27(b)). Please call the GIPSA Management and Budget Services support staff (202) 720–8479 for an appointment to view the comments.

Beverly A. Whalen at GIPSA, USDA
10383 N. Ambassador Drive, Kansas City, MO, 64153
Telephone: (816) 659– 8410
Fax Number: (816) 872–1258
email: Beverly.A.Whalen@usda.gov

The Case For Adopting the PAZ Technology

The Case For Adopting the PAZ Technology For Sampling For Milling Yield:

We have done some comparisons ourselves and we have shared those results with FGIS. The PAZ represents improvements not just in the whitener but the way in which the brokens are separated and the Degree of Milling (DOM) is determined. We feel confident your experiments will reveal the PAZ to be a far superior method of determining potential milling yield.

I would like to review several issues we discussed during our conversations in Washington at FGIS and Kansas City at the USDA National Grain Center. Please understand I am not a trained scientist and will explain these issues in terms I (and my fellow farmers) understand.

PAZ Whitening vs Grain Man:

We have made the point several times that the Grain Man does not resemble modern rice milling equipment in any way. The PAZ uses the same components as modern milling equipment.

One dramatic difference is that each grain is whitened more or less individually in the PAZ. In the Grain Man the grains are milled all at once, together and in the same chamber. I have been told by people that understand the Grain Man that when you have a sample that does not mill highly, as the grains breaks it causes more grain to break, which lowers the overall milling yield. Evidence of this would be a widening of the difference in milling yield between PAZ samples and Grain Man samples milled to the same whiteness as the milling yield goes down.

That would only be one advantage of the way the PAZ whitening mechanism is superior to the Grain Man. Another is heat.

We have pointed out that the Grain Man heats the sample to 140 degrees. The effect of the heat from the Grain Man reduces milling yield and cannot be discounted. We measured the temperature of the grain and the temperature of the Grain Man machine while we were running our tests as well as the tests in KC. Both the temperature of the rice sample and the Grain Man were a consistent 140 degrees after one sample.

After the Grain Man has run several samples it must be stopped to cool down or the effects are very negative.

The PAZ barely heated the sample at all. The PAZ uses an aspirator to remove the bran from the milling chamber as the rice is milled. This means a cleaner environment for milling and a room temperature sample. 

Indented Cylinder vs Shaker Table.

FGIS claims you need a 1000 gram sample to be representative, yet they reduce the sample to 40 grams for the most important part of the test, determining whole grains. The indented cylinder uses the entire sample and very effectively separates the whole from the broken.

Description: Macintosh HD:Users:harvey:Desktop:Page 33 Comparision of the Zaccaria.PNG

This table represents 11 different samples that were hand picked for brokens. The same samples were run over the shaker table and indented cylinder and the results compared.

As you can see from the table the indented cylinder is a much more accurate way to separate whole from broken. I would also note the wide range of results with the shaker table.

FGIS has spent a lot of time working to get away from the hand picking with a scanner. We applaud that effort but it was found to be unreliable. However it seems that the machine capable of replacing the shaker table/hand pick is the indented cylinder.  This is exactly the same technology and hardware that all rice mills use to remove brokens from whole kernels.

Degree of Milling:

As we have studied the protocols we think that the biggest failure of the current system exists in determining the degree of milling (DOM). The current protocols have only three reference points, and as far as we can tell only use one of them. The one FGIS does use is a minimum, which means it is the bottom of an unlimited range. The lack of a DOM graduated scale and the consequences of that make the current protocols almost useless as a predictor of milling yield.

OVER MILLINGING: From our experiments and observations it seems that the Handbook is written in a format that highly overshoots hard milled so that they will be sure they have crossed the minimum threshold. This means that farmers samples are over milled and suffer from the lower reported milling yield as a result.

For example, during our Jan. 16 visit to the National Grain Center a sample was run using both the current protocol and the PAZ system. The Grain Man sample was tested in the Satake MBZI whiteness meter and registered a value of 48.3. We consider 40 well milled.

A common denominator should be evaluated to determine the threshold for hard milling, as some buyer contracts ask for a higher degree of whiteness than is expected from consumers and commonly found in the supermarket.

Description: Macintosh HD:Users:harvey:Desktop:Page 14 --Rule of thumb.PNG

Farmers are losing 10 pounds of whole grain on the over milling by the current protocol.

Since the protocol does not allow for any adjustment on the Grain Man, no remedy exists. There is not even a measure for degree of milling in the current protocols (besides the one data point).

We attempted to determine the whiteness value of the hard milled reference card but were unable to get a reading.   The reference cards we ordered were not the same as the ones in Kansas City and we were told they were no longer available.  

OVER MILLING 2: We purchased rice from several grocery stores and tested them for whiteness. These samples represented a broad range of rice mills. The two large Arkansas rice mills were well represented in these samples.

The whiteness of the samples ranged from 38 to 48 with the mean being 42. The current protocol mill samples are milling way beyond what the industry sells.

I understand FGIS is conducting tests in California to determine if this is true, but they are using the current standards for degree of milling. As far as I can tell, this is no test at all. The same research, done at the University of Arkansas, is discussed and attached later in this document.

OVER MILLING 3: The following table shows the direct relationship between the degree of milling and the whiteness of the rice.

Description: Macintosh HD:Users:harvey:Desktop:Page 7-Results of over milling.PNG

As whiteness increases, brokens increase and milling yield drops. I doubt anybody would disagree with that fact. Yet with this chart it is possible that none of the samples were hard milled. 

OVER MILLING 4: Referencing the current protocols, hard milled is 60 seconds in the Grain Man. There is no other accommodation for lesser milled samples.  There is no adjustment for milling time, only the prescribed amount of weight to apply pressure. As far as the farmer and the tester are concerned, 60 seconds is hard milled, not the relative lighter or darker of the reference cards. I have seen hundreds if not thousands of samples run. I have never seen a reference card used. That includes the tests I have witnessed in Kansas City.

WHITENESS: I have heard FGIS officials say that the whiteness meter is not a good judge of DOM. That would seem to contradict your reference cards, which reference lighter than or darker than.

I have heard that lipid content is the new Holy Grail for DOM. Siebenmorgen defines DOM as the extent of bran removal from brown rice. I dont have any problem with that definition. I will point out however, that farmers are paid for whiteness and not level of bran removal. As contradicting as that sounds, if that is the definition FGIS wants to put forward, thats OK.

Siebenmorgen (study attached) also states The importance of measuring the DOM of laboratory-milled rice cannot be overstated for allowing equitable comparisons of milling yields.

Also; The Current FGIS approach to establish a DOM level of milled rice is to use a classification scale. Based on visual grading, primarily of milled rice color.

Since lipid content is not a practical way to test for bran removal, color is what we are stuck with.  How better to measure light and dark than with a whiteness meter?

How well can a whiteness meter predict lipid content?

According to Delwich, McKenzie and Webb, (study attached), milling degree is highly correlated to whiteness (r = -0.980).  That seems pretty accurate to me.

A study by Puri and Dhillon (attached) states A study by (37) reported that whiteness of highly milled kernels increases linearly with increased amount of bran removal. Whiteness of the milled kernels can be measured using commercial whiteness meters.


Any assessment of the protocols that includes only the current reference cards will mean very little to me or my fellow farmers. Why wouldnt the assessment of current protocols include whatever tools are available?

I have attached several studies to this letter. All of these studies report the importance of whiteness to the value of rice or the correlation of bran removal to whiteness, or the correlation of bran removal to lipid content. Most state these factors can be measured or predicted to a very accurate DOM. 

Over-Aggressiveness of the Grain Man:

We took identical samples and milled them with a Grain Man and a PAZ then looked at them under a microscope. There was a marked difference between the Grain Man kernels and the PAZ kernels. 

The above picture is the variety Wells after the PAZ. You can see the surface is fairly uniform and the nose is still completely intact. There are also no cracks or fissures in the kernel.

The above is the variety Wells after the Grain Man. The surface seems fairly smooth but the inner kernels show deep fissures and cracks. You can see the excessive wear (with the tip missing) on the nose. These kernels were likely very close to being brokens.

Description: Macintosh HD:Users:harvey:Desktop:Page 29-XL745CL-PAZ.PNG

This picture is of a hybrid variety after PAZ. The hybrid varieties are very difficult to mill.

Description: Macintosh HD:Users:harvey:Desktop:Page 30 -- XL-745CL-McGill.PNG

This is a picture of the same rice run through a Grain Man. As you can see from the pictures the hybrid variety kernels (even though they are whole) are cracked up.

Losing Your Nose:

It has been our observation that the Grain Man tends to break off the nose of the kernel, where the PAZ leaves it intact. This is something we pointed out to FGIS while we were in Kansas City in the same sample test they ran. You could easily see there were more whole kernels without noses in the Grain Man sample than in the PAZ sample. Even thought without the nose it still is a whole kernel, the weight of the kernel is 10% of the weight of the grain. That is a significant loss.  

Rice Mills Opinion of the Current Protocols:

When an industry commissions a study to make a point and they find out something unrelated they dont want to know, the report usually gets buried. That is exactly what happened with the USA Rice Federation. 

Description: Macintosh HD:Users:harvey:Desktop:PAge 35 --Industry's own study.PNG

This study was never released because It makes us look bad. as one USA Rice Federation official was reported to say. We only got the study because the University of Arkansas had a copy and we filed a request for it under Arkansass FOI law.  

Commercial Milling Equipment Produces Better Milling Yields:

Description: ::Desktop:5 Whiteness relationship with brokens and weigh loss (6).jpg

This study indicates the rice milled samples were 6.6% higher in head rice yields than the same samples with the Grain Man system.


It should come as no surprise to anyone that the PAZ is a superior machine than the Grain Man.  It should come as no surprise the PAZ is a much better predictor of potential milling yield. The rest of the world uses this technology. Our trading partners have complained about rice grading issues for years, and it is not simply because USDA does not do all the testing. 

I wish I could say it was because the protocols are so subjective, but from what I see the protocols do not work very hard to predict rice milling yield. It produces a set of numbers, but nobody in the farming or milling industry has much faith in them.

I can tell you with full certainty there is no confidence in the current rice sampling protocols.  When you get results all over the board from the same sample for years that happens.

ARGA has invested a significant amount of time and capital to prepare a compelling case for change in the way paddy rice milling yield is determined. 


We would like to thank the guys at FGIS in Washington for your patience with us. We certainly appreciate the work you are doing to try and help solve our problems. It makes the hard work we do in our profession easier when our government will not only listen but help us.

Thank you,

Harvey Howington