by Kevin C. Vaughn*, PhD
Louisiana Irises and Comments on Existing Cultivars
ABSTRACT: The herbicides trifluralin (TREFlAN) and oryzalin (SURFLAN) were selected as chemical compounds likely to increase ploidy levels in Louisiana irises. The compounds, while not as .toxic to animals, affect the development of plant tubulin protein in the same way as colchicine. Two batches of seedlings in two years have received the treatments. The first group (with some seedlings set aside as a control) was treated for 4-24 hours with herbicide solutions of 2-10 ppm. Seedlings were then thoroughly rinsed and planted into a sterile, soil-less mix. Club-shaped swelling was noted on the roots, and the treated seedlings showed stomate size approximately 50 percent larger than the control group. Of 12 treated seedlings, o~e appears to be a sectorial chimera, but 11 appear to be tetraploids. All but one of the seedlings in this group should bloom during Spring 1992. A second group of seedlings from diploid crosses was treated this year, and the same high conversion rate was noted. Attempts to produce tetraploids by tetraploid X tetraploid crosses, however, has proved disappointing. The article concludes with comments on existing tetraploid cultivars. 
Colchicine has been utilized for many years to convert diploid plants into tetraploid forms. It is an alkaloid that occurs naturally in the autumn crocus, Colchicum autumnale, where it probably acts as a feeding deterrent. Colchicine acts on the protein tubulin, the protein constituent of the cellular structure known as micro tubules. Microtubules are formed in the cytoplasm as polymers of tubulin protein. Colchicine blocks this polymerization reaction so that micro tubules are not formed from the tubulin subunits. Microtubules have many functions in the cell, one of them being the movement of chromosomes during mitosis. Before attempting cell division, the chromosomes undergo a duplication from a diploid level to a tetraploid level. Thus, because colchicinereated cells are unable to complete a cell division, the cell remains at the tetraploid level. After this conversion, the colchicine is washed from the tissue and the treated tissues now remain at the tetraploid level. Although there have been many examples of successful conversions utilizing this drug, there are several problems associated with the conversion of diploids to tetraploids with colchicine. The most important of these is the relative insensitivity of plant cells to colchicine. Solutions of 0.1-1.0% are routinely used for conversions. Such a high level frequently leads to secondary effects on the plant unrelated to the effects on micro tubules. Additionally, colchicine is effective on humans at considerably lower concentrations and handling such a solution without proper laboratory safety equipment is potentially very hazardous. From some experiments performed the last few years it appears there may be a safe and more effective alternative to colchicine for inducing tetraploidy in Louisiana irises. Many of the commonly-used herbicides also affect tubulin protein in much the same way as colchicine, but affect only plant and not animal tubu1in. Two of these are the herbicides trifluralin and oryzalin (under trade names TREFLAN and SURFLAN). Although neither of these compounds has been utilized for such conversions, it seemed as though either should prove to be an effective and safe alternative to colchicine. I have successfully utilized these herbicides on two different batches of seedlings from two different years, both with positive results. In the first batch, I divided the seedlings into a control (untreated) group and then treated the remainder with these herbicides (2-10 ppm) from 4-24 hours. Seedlings were simply placed on filter paper in petri dishes and then covered with a dilute herbicide solution and placed in the dark. After this treatment, the root tips of the treated seedlings were swollen into a clubshaped structure rather than the nicely-tapered roots found in the control group. This swelling is characteristic of plants that are affected by colchicine as well as these herbicides. The seedlings were then washed in running water and planted into a sterile, soil-less mix (3 parts peat:1 part vermiculite:1 part perlite) and grown under constant illumination at 72° F. The control group grew much more quickly than the treated seedlings but after two months of growth even the treated seedlings had made substantial growth. The treated seedlings were distinctly thicker than the untreated group and the size of the stomates was approximately 50% greater than those of the control group. These are both good indicators that the plants are tetraploids. Of the 12 treated seedlings, one appears to be a sectorial chimera (stomates of two different sizes) but the other 11 appear to be full conversions, at least in the epidermal tissue that gives rise to the guard cells. This is certainly an incredible rate of success considering the relatively low percentage of conversions normally obtained with colchicine and the stunted growth that occurs in these plants. These treated plants were transferred to the garden in spring. All of these have made good growth this season and all except one should bloom next spring. Hopefully, with BLACK GAMECOCK in the pedigree, these plants should perform well in all climates (or at least be a test of Joe Mertzweiller's theory that the tetraploids may be naturally more cold-sensitive). 
This season, I had many more Louisianas in the patch including some of the new Australians as well as a good collection of Mary Dunn's and Ben Hager's cultivars and of course a good collection of the tetraploids. Many diploid crosses were made and the first of these have been treated with the herbicide. This first batch shows the same high conversion rate as in the previous season. I was reluctant to write about this as a technique until I had at least been able to repeat it. 
Success in one endeavor does not confer success in others, however. I made hundreds of crosses between the tetraploids in my garden, and although there were a number of seed pods not one seed was obtained. All were enormous puff pods. I have not abandoned this approach, however, as there are some fine tetraploid cultivars with which to cross despite the rather narrow genetic base. Some of these cultivars are not well-known or distributed and all of them are most garden-worthy. The remainder of my remarks will be directed toward these tetraploid cultivars. 
The Durio cultivars are all derived from the two original tetraploid cultivars, PROFESSOR CLAUDE and PROFESSOR IKE. Despite this narrow base, these are very nice plants and distinct from their parents. SAUTERNE performed well in my own patch as well as in Bois D'Arc gardens near Houma, Louisiana. It is a clean light lemon-yellow with an infusion of lavender in the standards - but the landscape effect is bright lemon. The substance on this cultivar is amazing and the shape is very distinctive also: upright standards and short-shanked falls that flare near horizontal. Despite the heavy substance, SAUTERNE opened perfectly even on cooler mornings. It is a pity such an outstanding cultivar, as well as such a color break in the tetraploids, did not even receive an Honorable Mentionl DECOY is a light to medium rosy-purple that gives a garden effect of orchid pink. It has exceptionally wide petals and an attractive, almost flat form. My plant had four bud placements and bloomed over a long time. WINE COOLER is deeper shade of purple, although distinctively brighter and redder than the Professors. This cultivar has the best branching of any of the tetraploids for me, and the branches are separated well enough from the scape that each blossom is displayed very well. BAYOU ROUGE is the most vigorous tetraploid cultivar for me and is also exceptionally floriferous. It is a medium red from the rose side and is just slightly bitoned as it grows for me. RAGIN' CAJUN is distinct from the Professors in that each of the petals is edged a creamy yellow. This edge seems to enliven the whole blossom. The plant registered as WELCOME CHANGE does not appear to be the plant that is being distributed under this name but, whatever this cultivar is, it is a tetraploid and it is nice, a much brighter purple than the Professors. Substance and branching also appeared to be better in my garden. Perhaps the name could officially be transferred to this cultivar. 
Joe Mertzweiller's plants are well-known, but I would like to comment on several of his seedlings that are guested here as well as one cultivar that I thought most outstanding. PROFESSOR JIM was my pick among the tetraploid cultivars this season. A bright but dark red, it opened perfectly and has exceptional substance and branching. Another from the same cross was quite different than JIM and had wide branching like WINE COOLER but not as many buds as JIM. The dark, somber red coloring and the small signal elicited many favorable comments from garden visitors. A seedling with the garden nickname of "Barbara 11" was awesome in my garden despite being received as a rhizome the first week of December last year. It is a very clear lemon-yellow bitone with the greatest color intensity around the signal. Like SAUTERNE, there is a slight infusion of lavender that is obvious only on close inspection. The stalk is tall (42 inches) with 5 bud placements. The foliage and growth on this plant are perfect. Beautiful blue-green foliage and an average of a dozen increase on each rhizome; it is an example of what a perfect garden plant should be. I hope Joe officially registers this plant. 
Two other tetraploid cultivars in my garden that also performed admirably were COORABELL (Raabe '89) and SAMURAI WISH (Chenoweth '88). COORA-BELL is a dark purple self of flat form with all of the petals decorated with a bright golden signal. It had good branching and the added good trait that stalks on side rhizomes bloomed later than those from the main rhizome, making for a very long bloom season. SAMURAI WISH was much different than I expected from its registered description, but it matched plants seen elsewhere. It is a rosy-purple blend with a distinct. edging of lemon yellow and is a very flat blossom of good substance. Both of these plants are nice additions to the tetraploids. 
Considering how recently tetraploid Louisianas have appeared on the scene and the problems with their fertility, there really has been exceptional progress. Although purples tend to dominate, there are tetraploids available in every color except white and pink. Genes are present in the available tetraploids for both these colors and seedlings with these colors are currently under evaluation. Foliage on the tetraploids is in general superior even to the best diploids, making them better landscape plants. Conversion of the best diploid material should be a great boost to this program and Farron Campbell's attempt to add choice species material, chiefly Iris hexagona and I. brevicaulis,which could expand the hardiness and garden-adaptability of these plants. The 
future is bright!
*Kevin C. Vaughn, PhD has served since 1980 as a plant physiologist with the USDA in Stoneville, Mississippi working on herbicide resistance, herbicide modes of action and weed biology. Besides his work with Louisiana irises, he has produced more than 30 hosta introductions and 50 introductions of sempervivums. He has introduced six iris cultivars (four won HM's) and will introduce six cultivars of tradescantia in 1992. He is scientific chairman of the American Hosta Society and was cited as USDA Scientist of the Year in 1986. 
Editor's Note:  This article first appeared in the SLI Newsletter in the December 1991 edition.