The discussion of EPDs in alpacas is still going strong and today I was copied on a piece written by Dr Sara Via at the University of Maryland. Its well worth reading as it clears up one or two misconceptions about how EPDs are derived and how they can be used.

"Concerning the issue of comparing the IAC animals to the national alpaca
herd-- we can't directly compare the average values of the fiber traits in the IAC herd to the averages in the national herd, because we don't necessarily have access to fiber measurements for a random sample of the national herd. (see below for why it is a definite no-no to directly compare EPDs between animals in the IAC herd and animals in another group for which EPDs have been obtained).

What we can do, however, is track how much progress has been made within the IAC herd itself, because we will know and record the average values for each of the fiber traits every year after shearing. Because EPDs are expressed as a predicted deviation of an individuals progeny trait values from the average in the population, it is necessary to update the average used in the analysis every few years (more frequently when establishing an EPD program). But, this updating aside, we will know where we started, and after a few years, we can determine if the averages of the fiber traits have improved (i.e., mean diameter thinner, crimp higher, etc) or not.

It may be a reasonable assumption to make that the averages in the IAC herd now are about the same as in the national herd, and if so, then a change in the average values of fleece traits in the IAC could be interpreted as a change from the status quo in the national herd (presuming it is not improving at the same rate as the IAC herd). But that is an interpretation issue that could be debated.

Without being too technical, let me just add that you can't compare EPDs themselves between animals in different populations because each individuals EPD for a given trait is expressed relative to the mean of its herd. So, a male in the IAC virtual herd with a fiber diameter EPD of -0.2 microns cannot be compared with a male from another herd with the same EPD value unless you know the mean values in the two herds.

It is entirely possible that the average values of the IAC herd for fiber diameter etc will become so good that it will be obvious that we have improved our herd relative to the "average" individual that one can see at shows. This is of course not a quantitative measure.

Finally, a breeding program using EPDs in the way that the IAC is doing is not motivated by the desire to breed a few really exceptional animals. The goal is to breed a whole herd that is consistently excellent. Of course there will be variation around the average values, but improving the herd as a whole is the goal."
Dr. Sara Via
Department of Biology and
Department of Entomology
University of Maryland
College Park, MD 20742
svia@umd.edu