As the calendar has turned to autumn and the hay harvest season continues in full swing across the country, forage analysis labs expect to be hit by a tidal wave of samples from clients. These customers are looking to gauge the quality and nutritional profile of the product they have produced fo r on-farm use or for resale, while others may be using the data to make a purchasing decision based upon the nutritional needs of their livestock.

To obtain the best nutritional profile of your hay, one of the most important points to remember is that the forage analysis process begins on the farm and not the moment the sample arrives at the lab.

Where does the analysis process begin on the farm? With sampling! It’s a step many (or even most) times we all take for granted, but it is the most critical part of the process and the one that is the main driver of the results returned to the customer. Always remember the golden rule: “An analysis is only as good as the sample submitted.”

How do you collect a “good” sample? The key to submitting a good sample is to collect several subsamples to form a composite. Remember, the 1-pound sample you submit for analysis is going to potentially represent several tons of forage. Thus, you want to be sure the sample represents a good cross-section of the forage, not just one bale. For example, if you sampled the worst bale in the lot, feeding recommendations based on this information would result in overfeeding and increased feed costs. A composite sample properly collected will always be more representative of your forage source.

Now that we’ve established that sampling is a vital driver in the forage analysis process, how is it performed properly? Let’s keep our focus on hay but realize that proper sampling is essential to all sample types submitted for analysis.

Hays of different types, cuttings or lots should be sampled separately. Using a Penn State Forage Sampler (or other suitable hay probe), core 12 to 20 bales selected at random. For square bales, sample through the center of the small square end (perpendicular to the slices). For round bales, sample on the curved side with a core perpendicular to the side. Combine all core samples and submit in a clean quart or liter-sized zip-close-style plastic bag.

So how does all this information relate to the title of this article? Well, the questions we receive at the lab quite often are as follows: Can bales from the same field really vary that much in composition? Isn’t it all the same exact hay since we took all the bales from the same place? The answer to the first question is a resounding yes and to the second, no. Let’s look at some data from a small internal study conducted at Equi-Analytical with the help of a customer who supports these answers.

To set the stage, here are some background details connected to this study.

• An 86-acre field consisting of an orchardgrass-brome mixture was cut in the morning as second cutting in the month of September, baled into small squares after proper field curing, with bales stored indoors by both the producer and buyer.
• 20 random bales were sampled by the customer with a hay probe.
• Cored samples were pulled from the center of the short end of each bale and placed into plastic zip-close-style bags, then submitted to the lab for analysis.
• Upon reception at the lab, all 20 samples were processed in their entirety.
• Samples analyzed by near-infrared reflectance spectroscopy (NIRS) using the Equi-Analytical 600 – Fast Track Package consisting of the following components: DM, DE, CP, ADF, aNDF, ESC, WSC, starch, NFC, Ca, P.
• Results compiled across the individual 20 bales and also against a composite sample pulled by the same customer a month prior from the same 20 bales in Table 1.

It’s evident from the results in Table 1 that the analyses vary quite significantly depending upon the individual bale. The shaded green Difference, Max – Min row in the table spotlights the entire picture and exemplifies the issue at hand. Considerable differences exist across the array of nutrients. Sampling only one or even just a few bales will bias the results and could potentially result in purchasing the wrong hay for your horse.

For example, in the case of a metabolically challenged horse, equine nutrition experts currently recommend that these horses should be fed hay with a water-soluble carbohydrates (WSC) and starch total of 10 or less. Therefore, if the single bale sampled was No. 8 where the WSC + starch = 10.1%, a horse owner might decide to use this hay when, from the average data, the hay being delivered is actually 15.5 + 1.2 = 16.7%, which is far too high without soaking prior to feeding. Even if your horse is not metabolically challenged, there are still ramifications. A ration designed for the other major components, such as protein and fiber, will also be impacted by the use of the results from only a single or few bales and result in over- or under-delivering these nutrients.

Also of note is the comparison of the average of the 20 individual bales to a composite sample collected a month earlier from those same 20 bales. One will note the overall good agreement between the average and the composite, and this demonstrates the client did a good job pulling the composite sample. The slight noise in ADF and NDF values is most likely the result of some within-bale variability. Yes, within-bale variability also exists, as even a single bale is not uniform in composition throughout as the baling process, field and environmental conditions all can affect the final bale structure. However, the purpose of this study was not designed to address that topic.

In summary, sampling is a key component of the forage analytical process. Following proper sampling protocols is the best way to help ensure good, representative samples are taken on the farm and submitted to the lab for analysis. Rations developed will then deliver nutrients in sufficient quantity and proportion.

Improper samples contribute to be the biggest error in the forage analysis process and result in over- or underfeeding nutrients and prove costly to the animals' nutrition and health and to your pocketbook. Following simple steps to minimize the chance of a poor sample starts the analysis process off on the right foot, and then it’s up to the lab to properly process and analyze the submitted sample.