Fecal analysis to determine the diet of grazing cattle is going high-tech.

Livestock fecal analysis can give producers a lot of information about what their cows are eating. But, the current method of analysis can be tedious, expensive and slow in getting results to the rancher. That could change, however, if Dean M. Anderson has his way.

Anderson, a research scientist at the USDA-ARS Jornada Experimental Range in Las Cruces, NM, is utilizing the talents of several space-age researchers in an attempt to improve accuracy and cut the time required to analyze livestock feces. He's discovered it can be done with fluorescent spectroscopy by using either lasers or multi-spectral fluorometers.

Why Is It Important? Accurate identification of the contents of livestock feces can give range managers a good basis on which to base nutritional management decisions. But, Anderson says if the new methods can be taken to commercial levels, that information could be provided the same day.

The technology also could be spun off for use on other kinds of samples. His cooperators have already proven, for example, that optical sensing could be used to detect pathogens in food.

Traditionally, detecting what livestock eat on the range has been labor intensive and time consuming. There are special difficulties with observing their food at various locations along the digestive tract, their eating habits or with sampling rumen contents. That's why fecal samples (the technique came from the field of wildlife biology) have proven so useful. But there are other problems associated with the digested food.

Some things are thoroughly chewed and digested, making them difficult to visually identify, especially in the feces, Anderson points out. With the commonly used micro-histological technique, ground material is observed under a microscope. It's time consuming and requires a technician trained in plant fragment identification for each geographic region and season, followed by statistical analysis of the data to determine the relative percentages of plants in the diet.

"From identification through statistical analysis, these type of data are difficult to obtain in real time," Anderson says. That means if a rancher sends samples to a lab, he may not get results until long after he needed the information to make a nutritional management decision.

Anderson has used fecal sampling in his research, and the leap to adapting this new technology came about accidentally in the fall of 1991. In discussing his work with Ron Pinnick, then a researcher at the Army Research Laboratory, White Sands Missile Range, Pinnick thought he could use laser technology to detect differences in plant species growing in the field as well as those found in feces.

After preliminary work, Pinnick and his physics colleagues at New Mexico State University found both pre- and post-digested plant materials would fluoresce (give off light) when suspended in a solvent that was then illuminated with a beam of laser energy. The fluorescence among the materials had distinguishing features that served as a "fingerprint" to identify it.

Anderson explains that fluorescence occurs when electrons in a molecule are kicked up to a higher energy state after being hit with energy from a laser or other light source. "Preliminary results show most plant materials produce fingerprints having visibly distinguishing features," he adds.

Fingerprinting Plants Though some plants give visually similar fingerprints that can't be differentiated with the naked eye, scientists at Sandia National Laboratories have developed sophisticated computer programs to untangle similarities among materials and even unravel the most complex mixtures of molecules from a fluorescing solution.

At Sandia, Gary Tisone, Perry Gray and John Wagner took some of Anderson's samples and began looking at them using multi-spectral fluorescent spectroscopy (see Figures 1 and 2). In their technique a xenon lamp is used to get the molecules to fluoresce.

"At this point we aren't certain which molecules are fluorescing," Anderson says. "However, with this technique the same molecule should fluoresce the same regardless of where in the world the technique is used."

Any plant material could be identified once an accurate reference library of plant species, a "foolproof" recipe of equipment settings and intelligent algorithms suitable for computer analysis are available. This would remove human bias and could make reliable data available in less than three minutes per sample.

There is a lot of work yet to be done, and a lot of data processing to be worked out. "A rancher in West Texas probably doesn't care a hoot about neural net processing algorithms," Anderson says of the technicalities and mathematics. But that rancher does give a hoot about having a quick, accurate and likely much less expensive means of finding out what his cattle are eating so he can manage proactively.

"The one obstacle is that research money has always been tight for developing new tools," Anderson says. He and his colleagues are looking for funding to continue multi-spectral fluoresence spectroscopy research on fecal samples, which so far has been conducted without dedicated funding, basically on weekends in workers' spare time.

Effects Could Be Far-Ranging The spinoffs could be far-ranging. One of the questions Anderson has had about cattle diet is determining what part may be heritable.

"We know that animals have variable diet preferences, just as humans do. Could some of these preferences be heritable?" The answer to that question could lead to the breeding of "designer" cattle with specific dietary preferences or avoidances.

Another spinoff would be in the wildlife field. Currently, feces sampling and species identification under the microscope is the primary means of determining wildlife diet, and a breakthrough in technology could help wildlife researchers greatly.

For further information, contact Anderson at 505/646-5190, by fax at 505/646-5889, or by e-mail at deanders@nmsu.edu