Descendant Analysis

For many years, hybridizers have used pedigree information for planning their crosses -- both individual experiments for the coming year and long-range programs -- spending long winter nights curled up with Checklists. 

The most daunting task is searching for descendants of interesting progenitors, to find out whether they have performed as predicted and find advanced-generation descendants that would be promising candidates for a specific breeding program. If, for example, you want to identify modern iris that are descended from certain species (perhaps with a goal of obtaining specific traits in your own lines), descendant analysis is the appropriate tool.

When I changed from pollen-daubing [which I still admit can be lots of fun] to serious hybridizing, I was determined to learn just how accurate the folklore was concerning the small number of progenitors involved in the development of our modern arilbreds. So I went through the 1976 ASI Checklist and made a master list of every species and named cultivar that had registered aril or arilbred descendants – and noted whether it had been used as a pod parent, pollen parent, or both.

I found more than folklore had led me to expect, but then I started the descendant analysis. This involved reading through Checklist, pen in hand, to compile a list of descendants of each specific cultivar on my master list. Yes, I started this in the days before computerized databases and search capabilities – and it took years to compile. Some lines I've checked against every new R&I booklet released, and while I keep expecting to find that the older breeders are no longer used some of these pedigrees are so involved that the progenitors still show up.

I've posted some of these lists here. They were originally compiled for my own use, with no thought of publication, so the format and level of detail varies greatly.  For the complete list of cultivars introduced through 1975 that themselves already had registered offspring, go to the Gene Pool. 

There, you will find links to the individual pages that have been posted and the names of many others whose pages are still in queue. Although quite a few pages have been processed, there are clearly many more to come.... 

The major categories also have their own pages so that, for example, you can explore the C.G. White amphidiploidlike hybrids in depth without being distracted by other types.  

• The arils were originally treated separately, as three fertile familes: the diploid oncos, the diploid regelias, and the tetraploid regelias.  I soon realized, however, that the less-than-fully-fertile OGs and RCs were a significant conduit of aril chromosomes into the arilbred gene pool and thus, for the purpose of this analysis, started looking at all diploid arils as members of the same fertile family. 

• The Triploid Route has been the primary means of introducing addition aril genes into modern arilbreds.

• The C.G. White Cultivars introduced in the 1950s, which were counted as amphidiploidlike hybrids, are perceived by many as the beginning of modern fertile halfbreds.  They were certainly a major breakthrough, but not all contributed significantly to the modern gene pool so this section analyzes their impact in detail. 

• There are also a significant number of Bypassed Breeders among the halfbreds.  These appear to have been reliably fertile with the right mates, yet little used.  Often it was simply a matter of timing, but I do believe that some of these have unexplored potential that could still prove valuable today.

• The Three-Eighths Route shows one path to improved gardenability, widely used in the early days and still accepted under the quantum system but now mostly ignored.

• The list of Quarterbreds focuses on the few that have registered offspring.

If you're curious as to how this can be done without a computer – you must at least scan every entry in the Checklist and, whenever another descendant is found, add it to the list. By the time you've finished going through the books, you should have a complete list of first-generation descendants. Then it's time to go back through the books all over again, searching for the offspring of every variety on your first list, to compile a list of the second-generation descendants. Repeat the process using this new list to compile a list of third-generation descendants – and continue in this vein, again and again, until you've reached the end of the line, accumulated as many generations as you want, or have tired of the effort.

Yes, that's a lot of work - but the information that can be gleaned from descendant analysis helps identify desirable varieties to add to breeding stock, makes planning of crosses much more effective, and provides inspiration for experiments with new lines. You'll find that the time spent pouring over books is repaid many-fold in the seedling patch.

Recently, however, some of us have discovered that adding computer power to the process saves a great deal of time and effort with the paperwork. A many-generation pedigree that would take me at least 30 minutes to draw by hand can be flashed on the computer screen in the time it takes me to type the name of the iris, or printed out in less than a minute. The difference is even more striking when it comes to descendant analysis. The computer can produce a first-generation list in seconds instead of the many hours it takes to do this by hand. Multiple generation traces can easily be done in minutes instead of a series of evenings.

This is the tool of the future - what I use to analyze pedigrees, trace lines of descent, and devise my hybridizing plans. Playing "what if?" games on the computer doesn't substitute for making the actual crosses, but it certainly helps to focus the effort and I believe that it has saved years of effort in some of my lines.