Sampling error
Sampling error causes more lab analysis variation than any other single factor. That’s right – the variation you’re seeing between or within labs could very well be a problem stemming from your probe, your method or how you handle the sample. No one likes to admit they might be the problem, but at least consider refining your protocol. The challenge is that your sample (less than 1 gram) must represent tons, or the majority of the lot. It’s worth getting the sampling protocol right.
Hay sampling protocols
1. Identify a single lot of hay (a single cut, from a single field, baled within 48 hours). Do not mix samples from fields or cuts – it just complicates life and increases sampling error.
2. Sample close to the point of sale or feeding; dry matter (DM) percentage can change. Use “as received DM” only for tonnage, not for quality.
3. Choose a sharp, well-designed probe. (A 3/8- to 3/4-inch probe works best; a shaft too narrow does not represent true leaf-to-stem ratio and a shaft too wide gives too large a sample.) Never take “grab samples” – they’re a waste of time. Do not use slanted tips – these push stems aside and don’t take a true sample.
4. Take enough cores – a minimum of 20 cores per lot, more for variable or larger lots, and greater than 30 cores for GMO detection. (Variation is a part of biology – just look at the number of dog breeds or the average height or color of human beings; you wouldn’t just take three of either and create your average. The same is true for hay samples. Set 20 cores as the minimum to take per lot.)
5. Sample at random. Do not eliminate bottom bales, etc. (unless they are discarded from the lot); walk around and sample systematically.
6. Use the proper technique – sink the probe in 90 degrees from the butt end of the bale, between the ties, 12 to 18 inches deep.
7. Take the right amount – not too big, not too small – about 1/2 pound or 250 grams. Larger samples are poorer from the whole-process point of view; labs will not often grind large samples, which defeats the purpose of taking larger samples.
8. Handle the samples correctly: Seal samples in a Ziploc or similar container; keep the sample cool and away from excess heat (like a pickup windshield); and get the sample to the lab as quickly as possible.
9. Choose a qualified lab. Start by finding a certified lab, and ask important questions such as what the internal quality control protocols are, or whether the lab grinds the entire sample.
10. Never split the samples (to send to different labs) without grinding it first. Ask for ground samples back from the labs if you want to resend to a different lab. When cross-checking labs, don’t compare unground samples. Remember, good labs will always agree to work with you on standardization.
Questions to ask the lab
Before you send off your hay sample, select a lab keeping these things in mind:
- Is the lab certified by the National Forage Testing Association (NFTA)?
- What are the quality control steps? (Ask them.)
- What procedures do they use to obtain the estimate? (If they’re not using AOAC [Association of Official Analytical Chemists] standard procedures, ask why universities, etc., use them.)
- Does the lab grind their own samples? If not, ask why (every step in the process means another opportunity for error). A lab shouldn’t be ashamed to tell you how they handle the sample and report on whether or how they train staff to handle samples.
Remember, while variation in and among labs cannot be eliminated, it can be reduced to a reasonable level. But it’s got to start with you.
Do you have Supreme hay? Prove it. 
-
Lynn Jaynes
- Managing Editor
- Progressive Forage
- Email Lynn Jaynes
PHOTO 1: You might be surprised to learn what some folks use for hay probes when sampling hay – everything from sawed-off golf clubs to hand drills and hydraulics. There's more to sampling hay than sticking a tube into a bale, and your probe choice can influence the accuracy of the sample. This probe display was used at the Alfalfa Hay Quality Workshop in Reno, Nevada.
PHOTO 2: Jody Gale from the University of Utah demonstrates placement and angle of taking samples with a hay probe at the Alfalfa Hay Quality Workshop in Reno, Nevada. Photos by Lynn Jaynes.
