Basic Histogram Terminology

Most of us are familiar with charts and graphs. They offer us a visual representation of data. While the data may look like an unintelligible row of data, once we can see it in chart format, we can easily identify trends and sizes represented with lines and colors, bars, intervals and arrows and the like. Charts and graphs are a great way for us to comparatively view statistics.

‘Histos’ is a Greek word for something standing upright. The Empire State Building is a histos. So is this. Note the colored bars. Note the vertical numbering from 0 to 90. Note the horizontal labels (1^st Qtr, etc). This chart offers us a comparison. We know from the base labels that we are representing something through quarters (usually means the 4 quarters of a year) and that the color chart to the right matches the bars with the cardinal directions, but we would need to label the vertical axis (0 – 90) to get any meaning from this.

So a histogram is a graphical display of numerical data. A picture is worth a thousand words. Histograms help alpaca breeders evaluate each alpaca’s fiber scientifically. Below I am evaluating the fiber characteristics of two male Huacaya.

I am comparing the percentage of fibers on the vertical axis against the fiber diameter measured in microns. One male is in green, one in blue.

What I would like to see is more of both colored lines shifting to the finer (left side) side of the graph (finer = $$$$$ in general). I would also like to see the bars reaching higher than they are. Why? The farther to the left I move and the higher I move, the finer and more uniform the fleece will be. That can be worth big bucks.

The histogram below is Vicuna fiber. Note how this graph is shifted left and is elongated vertically. Vicunas are known to be very fine and very uniform. This is what fine and uniform look like on a histogram.

Note the symbols on the top left corner of the above diagram. AFD, SD, CV, CF, Curve, Min, Max, Staple. The green bars are a graphical representation of those measurements. To understand our alpaca’s fiber better, to measure its value against others in your herd, we need to understand these numbers – not in depth, just in high level. To be successful as a breeder, you better be able to evaluate your end product. Besides, potential customers will ask.

Take a deep breath and let’s dig in.

Getting a histogram test is simple. You basically cut a fiber sample (the size of a man’s thumb for the OFDA 2000 test and 2 square inches for the OFDA 100 test) from your alpaca, send it to the appropriate lab and your results are sent back within days or a few weeks either in the regular mail or by email.

Two examples of where to send them are online:

Yocum Mccoll - http://www.ymccoll.com/

Alpaca Consulting Services (ACS) USA - http://www.alpacaconsultingusa.com/

(Thanks to Ian Watt of ACS for permission to use samples of his histograms for this material).

AFD – average fiber diameter

(also referred to as MFD – mean fiber diameter)

For many breeders this seems to be the ‘be all end all’ statistic. We call it ‘micron madness.’ While it is a key statistic in alpaca fleece evaluation, it does not give the whole picture of an alpacas worth.

If the tested alpaca is ill, the fiber will be finer since most of the energy the alpaca has is being channeled to fight the illness.

If the alpaca was tested in the dead of winter, the test result will show up finer than normal since the alpaca is channeling most of its energy to keep warm.

If the alpaca has been over eating and has been on a high protein diet, this number may be very high and could be reduced if the alpaca is fed a balanced diet.

Note how the environment and nutrition play a big part in fiber testing. There may be some truth in the adage, ‘figures don’t lie, but liars figure.’ Most of us sample fiber in the spring, during shearing.

Also note that some tests of AFD are measured at only one point while others take measurements along the entire fiber. No alpaca has a perfectly uniform fiber diameter – it changes due to health, nutrition and environment, as well as from pregnancy, nursing and other factors.

Also note that, in general, a sample is taken from one spot on the alpaca’s blanket. The AFD therefore only reflects a sampling of one area – the side of the blanket. While there are other methods to sample fleece, this method seems to be the standard for now.

The following is a graphical representation of a fiber sample where the micron diameter is measured along the fiber and not from one single location. It is called the Average Fiber Profile.

Note the wide variation of fiber from less than 28 microns to nearly 33 microns. This is a sample of a year’s growth on a pregnant dam. The left side of the graph represents shearing day the year before and the right side represents the tip of the fiber sample, furthest from the body.

Note than accurate records of illness, pregnancy, delivery, nursing, and even stress from the spring/summer show circuit can be matched to the timeline of this graph. In addition to micron, this graph can give one an excellent record of health and nutrition.

So AFD is merely one factor necessary to assess an alpaca’s value and status.

Standard Deviation (SD)

Technically, this number measures the number of fibers within one standard deviation of the mean (arithmetic average). Yeah, right.

Imagine it is your job to find the average income of families living on two specific streets. On street one there are 20 families and they all happen to make $20,000 per year. The average income is easy to compute – it’s $20,000 per year. The average is very representative of each family – in fact it is a perfect match for each family. Street two as only two houses. The family in the first house is very wealthy. Their household income is one million dollars per year. The family breadwinners on the second block are out of work and have been for over a year. When we average the income on this block we get $1,000,000 plus $0 and divide by 2. This gives us $500,000. The out of work family laugh when they hear this average income of the block. So do the millionaire family. The average in this case is not very indicative of either household. This is because there is such a small sample (2 houses) and a big difference in income. We say the second average is highly deviated – or skewed. It doesn’t really give us a reliable indicator of the facts.

This happens often with statistics. We hear it a lot in the media when we sample voter satisfaction. They say things like ‘this sample has a plus or minus 3% error).

If we apply this to micron measurements, what if your alpaca’s fiber is 19 microns for each fiber? Then an AFD of 19 means something. But if ½ your alpaca’s fiber is 10 microns and the other ½ measures 40 microns, the AFD is 25. This is not a good representation of an average since you will feel the 40 microns and not the 10.

The last example shows a high deviation. Statisticians always include a standard deviation number with the average. The lower the SD, the less the deviation and the more representative of uniformity of measurement around the average.

From experience, an SD of 5 or less is an excellent number. It means their is a high degree of uniformity in the fleece. So an AFD of 19 with an SD of 4.9 is great. But an AFD of 15 with an SD of 23 is not so hot. Why? Poor uniformity.

From our histograms we can see that the less spread out the green bars are, the better – high uniformity. The more to the left the peaks are and the higher they are, the better – very fine. All this can be understood by an experienced histogram reader in a flash. This is much better than reading line after line of numbers.

Note – when comparing one alpacas SD to another, it is only valid within the same micros range. If the alpacas have wildly different micron counts, comparing SD is like comparing apples to oranges.

So how do we compare two alpacas with different AFDs?

Coefficient of Variation. (CV)

CV is defined as a formula to assess uniformity within a tested fiber staple. It is a statistic which relates SD and AFD displayed as a percentage. The lower the number, the better uniformity the sample will have. A CV around 20 is excellent. Often times a cria will have a high CV and as it ages (primary and secondary fibers mature). One can compare different alpacas with different AFDs using the CV to say ‘based on these samples, alpaca one and two are both uniform or not.’

Comfort Factor - CF

One of the statistics in your histogram report will be the percentage of fibers greater than 30 microns. It is at this 30 micron number where sensitive skin begins to react to these robust fibers ‘pricking the skin.’ At 30 microns a fabric will begin to annoy its wearer. Experience shows that if the amount of fibers at or above 30 microns is greater than 5% of the entire fleece, this is not a soft feeling fleece (or yarn). So if your histogram comes back with Fibers > 30 = 1%, this is a great number. The Comfort Factor (CF) is measured by subtracting the % of fibers > 30 from 100. In our example above, the CF is 100 – 1 or 99%. Pure luxury.

Curvature

This statistic is picking up steam as a valuable measurement. The curvature number relates to amplitude (wave height) of Huacaya crimp. A high curvature number indicates a higher frequency crimp and low numbers reveal a lower frequency crimp with higher amplitude. What some call ‘zipper crimp will have a high curvature number and the style called ‘French Fry crimp’ will have a lower curvature number.

Fiber length (stretched and unstretched) & Fiber Growth Rate

Fiber Length and growth rate metrics can help you determine several things. Fast growing fiber (if fine, uniform, etc) means you will have more product to sell at shearing time. There is also evidence that fast growing fiber is in general brighter than slow growing fiber since the fiber scales in fast growing fiber will be longer and smoother and thus reflecting available light better. The greater the difference in length between an unstretched fiber and a stretched one, the deeper the crimp the fiber most likely has. So an alpaca that looks to have short fiber may have deeper crimped fiber and indeed be carrying more fleece weight than an alpaca whose straight or less deeply crimped fiber makes it look ‘blown up and dense.’ These are critical statistics to keep track of.

Note; Not all of the above metrics are available from all vendors.