THE QUEST FOR UNREDUCED GAMETES – PART 1: A HISTORICAL PERSPECTIVE
By: MaryAnn D. Pruden
When I first emailed members of the AHS Robin for suggestions of daylilies which produced unreduced gametes, I had no idea that I was about to embark on such an interesting and complex project. My initial purpose was to get some suggestions for daylilies that could cross with diploids in order to avoid converting my own daylilies with colchicine. I do not like working with toxic chemicals and I thought this would be a good way to expand the gene pool of my hybridizing efforts. As I began to research the registration data of these daylilies, more and more questions surfaced. This report is an accumulation of data obtained from researching registration data, various studies and articles, and gathering opinions and observations.
The limitations of this study are as follows:
1. Most daylilies are not produced under controlled conditions. Stray bee pollen and human error could be factors involved in incorrect parentage as registered.
2. Historical data was obtained from the AHS CD Checklist, 2003 edition. This is limited to only registered cultivars and does not include seedlings that may pertain to this article.
3. I am not a biology expert or geneticist.
To gain a better understanding of what factors could be involved here, we should start with some definitions and some simplified genetics. Basically there are three types of daylilies: diploid, triploid, and tetraploid. Only diploid and triploid species are found in nature. Tetraploids were introduced by inducing diploid daylilies with a chemical called colchicine. The definitions of these three types are as follows (from the AHS Dictionary).
Triploid: A plant with a triple set of chromosomes (in daylilies, 3n = 33). Triploids are usually infertile.
Triploid daylilies are generally thought to be infertile because the gametes do not divide in half—16.5 and 16.5, but rather in various numbers. Occasionally they will divide with either 11 (1n) or 22 (2n) chromosomes allowing them to cross with a diploid or tetraploid daylilies and produce viable offspring. A cross of a diploid with a tetraploid (or vice versa) may result in a pod forming; however, the presence of a “triploid block” in daylilies prevents most of these from reaching maturity, and the pods fall off.
If a diploid daylily is producing unreduced gametes (UGs), then instead of 11 chromosomes in the egg or pollen cell, there are 22. The gametes do not divide in half during meiosis as would normally occur. If this “unreduced gamete” having 22 chromosomes then meets with a tetraploid gamete having 22 chromosomes, the result will be a tetraploid daylily having 44 chromosomes. Unreduced gametes can occur on the pod and/or pollen side of a daylily, but they do not have to occur on both sides. Thus, a cultivar can produce unreduced pollen, or unreduced ovules, or both. UGs on the pod side would be evidenced by a diploid daylily being pollinated by tetraploid pollen and forming viable seeds which germinate and produce tetraploid daylilies.
Toru Arisumi, a noted Research Plant Geneticist, has approximated that unreduced gametes do occur naturally in daylilies. In controlled tests, Arisumi “obtained two tetraploid seedlings from over 1000 pollinations of diploid x tetraploid crosses…Assuming about 30 functional ovules per flower, and that every ovule was fertilized, the frequency of unreduced egg cells is 1 out of 15,000.” 1 This translates into an average frequency of about 1/500 crosses in the general daylily population. Arisumi further stated “This is only a rough estimate. The frequency could be higher because all tetraploid zygotes are not necessarily viable and we could have recovered only part of the unreduced egg cells”. 1 Basically this estimate is very conservative. Tetraploids rarely, if ever, produce 30 viable zygotes per pod, but that is the figure being used in the calculation. Also, his assumption does not take into account any possible increase in frequency among the specific cultivars shown to produce unreduced gametes as only 10-15 crosses were performed on each. The cultivars producing the unreduced gametes were BOUTONNIERE (Stout ’38) diploid X TETRA STARZYNSKI* (Straub ’49) tetraploid and SUMMER INTERLUDE (Hall ’55) diploid X Tetra HEARTS AFIRE* (a conversion of the diploid HEARTS AFIRE (Connell ’53)). A review of the offspring of BOUTONNIERE and SUMMER INTERLUDE did not reveal any further production of unreduced gametes as there were no registered tetraploids listed. (Note: this does not imply that these cultivars never produced another unreduced gamete. Historical data is limited to AHS records, and according to such, no tetraploid offspring are registered to these cultivars).
Unreduced gametes on the pollen side would be evidenced by pollen from a diploid fertilizing a tetraploid daylily and producing viable seeds which germinate and result in tetraploid offspring. UG pollen is characterized as larger than diploid pollen when viewed under a microscope. Both diploid (1n) and unreduced (2n) pollen are likely to be found on a daylily which is producing UGs. Arisumi did not obtain any tetraploids resulting from tetraploid x diploid crosses. He suggested that “the small amounts of unreduced pollen are more likely to get lost amongst thousands of normal pollen.”2 But once again, he was assuming that unreduced gametes occur at the same level among all cultivars.
Note: Unreduced gametes on the tetraploid level would produce a 4n gamete which would be unusable by diploids, triploids or tetraploids and would not produce any viable seed. Therefore, the main focus of this report will be on diploids.
A final term needed for this report is as follows:
Chimera: A plant with two genetically different tissues growing side by side. Chimeras may or may not be stable. Ploidy chimeras are common in plants treated with colchicine, part of the tissues being converted to tetraploid, other parts remaining diploid.
In the case of a chimera, the plant would actually have different cell layers with different ploidies. Depending on how fully the conversion took, a plant could produce diploid, tetraploid, or both kinds of gametes. I have read of instances where the pollen from a converted plant was examined and found to contain both diploid and tetraploid sized grains. A chimera cannot pass on this trait to its offspring. Chimeras are usually unstable and can revert back to diploid or even change back and forth over time.
Probably the most well known daylily that is said to produce unreduced gametes is ED MURRAY.
ED MURRAY (EM) is a registered diploid with both registered diploid and tetraploid offspring. Peter Mueller of Heidelberg, Germany analyzed EM’s root tips on two occasions. He presumed EM to be diploid; however, aneuploidy could not be completely ruled out because 21, 22, or 23 chromosomes were counted. It was difficult to identify the correct number because of the appearance of a larger chromosome or one that was overlaying another. The definition of an aneuploid per the AHS Dictionary is as follows:
Aneuploid: Having a chromosome count that is not an exact multiple of the haploid number, with either more or fewer than the normal number of chromosomes in the cell. Some daylilies converted with colchicines are suspected to be aneuploids.
Records of the original diploid ED MURRAY’s parentage do not suggest that it was the result of a conversion; however, hybridizer records are not always clear on this issue. A review of its pod parent, TIS MIDNIGHT, shows no aberrations. All offspring are registered diploid and appear to be just that. ED MURRAY’s pollen parent was listed as a “seedling” which could be just about anything. We will probably never know what it was. It is clear, however, that EM produces tetraploid offspring from both the pod and pollen side. This is evidenced by it having registered tetraploid offspring when used either as pod or pollen parent. Additionally, from reviewing AHS registration data, it also appears somewhat diploid pod fertile, albeit to a lesser extent, as there are far fewer diploid registered offspring. Despite the number of registered offspring, EM is notorious for having low fertility. Low fertility is associated with aneuploidy and other chromosome aberrations.
Several conversion attempts have been performed on EM. However, it is believed that the conversions were not successful and that tetraploid offspring produced by the so-called Tetra ED MURRAYs were just a result of EM’s usual fertility pattern. Many, if not all, of these conversions are now circulating as diploid ED MURRAY.
In a memo dated 7/26/99 from Peter Mueller to Nick Chase, Mueller writes,
“I've checked the pollen of Ed Murray, and compared it with the one of Blue Purple, a very pollen-fertile diploid. The grains are smaller, quite irregularly shaped, and of different sizes. Indeed, I also found very few quite huge and regularly shaped grains too, I guess less than 1 out of 300. These probably are unreduced and thus usable pollen cells.”
It is important to note that both the pollen analysis and the above referenced chromosome count performed by Mueller were done on the original diploid clone of ED MURRAY. Mueller’s pollen description differs from that seen elsewhere. For example, Patrick Stamile posted the following description of ED MURRAY’s pollen on the AHS Email Robin:
“I, too, have checked ED MURRAY's pollen and it looks just like some treated tets. Fractured pollen grains with other huge normally "tet" size pollen. The grains are large even for a tetraploid. I saw no diploid size pollen at all.”
Stamile believes he may have had one of the ED MURRAY’s that had a conversion attempt performed on it, although he cannot be certain. It appears that his belief is correct from a comparison of the two pollen descriptions above.
Looking at the diploid offspring of EM, we see that THORNBIRD is the only one that also has its own registered offspring—both diploid and tetraploid.
- THORNBIRD (Ford ’86) diploid
- Parentage: (LITTLE GRAPETTE X ED MURRAY)
- Offspring: 3 Tetraploid/1 Diploid
- LITTLE GRAPETTE (Williamson ’70) diploid
- Parentage: LAVENDER DOLL (Williamson ’68) X SDLG
- Offspring: 64 Diploid
- 1 Tetraploid
THORNBIRD is the cross of two diploids. Yet, we see a scenario where a registered diploid is producing both diploid and tetraploid offspring. So, is THORNBIRD really a diploid? Or could it be triploid or tetraploid?
THORNBIRD’s pollen has been examined on at least two occasions and was noted to be tetraploid in size with the even larger EM type grains scattered throughout. If THORNBIRD is fully tetraploid, then diploid LITTLE GRAPETTE could not be the pod parent, unless LITTLE GRAPETTE is also producing unreduced gametes.
To muddy the water further, ABOUT TIME* is another registered tetraploid offspring of LITTLE GRAPETTE.
Again, this scenario could only occur if LITTLE GRAPETTE produces UGs. However, Chase tested LITTLE GRAPETTE for unreduced gametes and was unable to set any seed with tetraploid pollen.
Other scenarios: ABOUT TIME* could be triploid or possibly diploid if stray bee pollen was involved. There is even still a chance that it could be tetraploid if the pod parent was not LITTLE GRAPETTE, as there is a tetraploid cultivar named LITTLE GRAPE* (Durio ’80) which is a cross of SDLG X Tetra LITTLE GRAPETTE*. Thus, we have more variables added. There has been speculation that LITTLE GRAPE* could be the pod parent of both THORNBIRD and ABOUT TIME instead of LITTLE GRAPETTE. The name similarities would make this an easy error to make in the field and would easily account for both ABOUT TIME* and THORNBIRD being fully tetraploid.
Another possibility is that LITTLE GRAPETTE acted just as a normal diploid supplying 1n while ED MURRAY supplied 3n via the “giant pollen” producing 4n (tetraploid) THORNBIRD. In this scenario, suggested by Maurice Dow, ED MURRAY would not be producing unreduced gametes, but would be a “meiotic mutant” producing pollens of various ploidies. This would account for the extreme large size of the pollen grains (described as larger than tetraploid) and why either crosses with diploids or tetraploids result in tetraploid offspring. This occurrence has been documented in other plant species.
Next we will examine THORNBIRD’s diploid registered offspring:
NORMAN LEE HENNEL (Ford ’91) diploid
Parentage: RECENT VINTAGE (Lambert ’72) X THORNBIRD
Offspring: (1) tetraploid: CRACK OF DOOM* (Hanson ’01) tetraploid
Although, NORMAN LEE HENNEL was registered as a diploid, I could not find any reports of its use in diploid breeding. There are numerous accounts of it being tetraploid fertile both ways and the pollen was checked on at least two occasions and found to be tetraploid.
Let’s now take a look back at EM, THORNBIRD and NORMAN LEE HENNEL. All three cross with tetraploids as either pod or pollen parent producing viable tetraploid offspring. From the number of registered offspring, it appears that all three are much more fertile when used with tetraploids. EM is a confirmed diploid (genetically tested) and produces mostly small irregular sized pollen with some additional gigantic pollen. Although THORNBIRD and NORMAN LEE HENNEL also have been shown to produce these gigantic pollen grains, their remaining pollen is tetraploid in size. Because of this, it appears that both THORNBIRD and NORMAN LEE HENNEL are, in fact, tetraploids but which also inherited the ability to produce the gigantic pollen from their parent, ED MURRAY.
The remaining three diploid registered offspring of ED MURRAY are: JUPITER LIGHT HOUSE (Warrell ’02), MURRAY MIDNIGHT (Powell L.K. ’88), and BANGLADESH (Smith W.H. ’79).
- JUPITER LIGHT HOUSE (Warrell ’02) diploid
- Parentage: NIGHT WINGS* (Williams, J. ’85) X (BLACK CAT (Wild ’80) x ED MURRAY)
- Offspring: None
According to Dottie Warrell, JUPITER LIGHT HOUSE’s parentage is a combination of the above three cultivars crossed, selfed, and sibbed several times. There is not an accurate record of the combinations as Warrell is a self-described ”backyard hybridizer”. Although there are not any offspring to examine and Warrell did not use JUPITER LIGHT HOUSE in any breeding, it is very probable that JUPITER LIGHT HOUSE is actually tetraploid due to its parentage.
MURRAY MIDNIGHT AND BANGLADESH are both crosses of a diploid with ED MURRAY pollen. As EM appears to cross much more favorably with tetraploids, the ability of EM as pollen parent on these crosses could be questioned. It could be possible that ED MURRAY is not even the true pollen parent in these crosses if stray bee pollen were involved. It would then appear that ED MURRAY crosses only produce tetraploids. Furthermore, the registrations for these cultivars could be incorrect and they could actually be tetraploids. A pollen and/or root tip chromosome inspection could easily confirm their ploidy.
The last EM case that will be examined is the following:
- IMPLAUSIBILITY* (Chase ’97) tetraploid
- Parentage: H. fulva ‘Europa’ (triploid) X ED MURRAY
- Offspring: None registered.
In a letter to the Nostalgia Robin dated August 31, 1999, Chase wrote “I also checked out my tetraploid IMPLAUSIBILITY* (H. fulva 'Europa' X ED MURRAY, you may recall) and found that it also has the large, unreduced-gamete pollen grains, so this summer I tried the pollen on diploids and got lots of pods. I actually set pods with good-looking seeds on about half the diploids I used, and there seems to be no way to predict which diploids will respond and which will not.” The ability to produce seed in this example would not be due to IMPLAUSIBILITY* producing unreduced gametes since a tetraploid unreduced gamete (4n) could not cross with a diploid gamete (1n). At this point, it is impossible to determine what the above seed production (resulting from diploid crosses) really means. Non-viable seed can be formed in cross ploidy pollination, but the true test occurs only if the seed germinates. I have been unable to reach Chase for further information regarding this.
Assuming that IMPLAUSIBILITY* is tetraploid and also crosses with diploids, we could ascertain that this ability was inherited from ED MURRAY, as evidenced by the gigantic pollen grains. However, the following cultivars bring that inference into question.
- ROSA FULVULTRA (Powell W. ’94) diploid
- Parentage: [HYPERION (Mead ’24) x TRIPLE THREAT (Wild ’62)] X H. fulva Europa
- Note: Parentage per Bill Powell
- Offspring: None
ROSA FULVULTRA is a registered diploid from the cross of a diploid seedling with the triploid H. fulva ‘Europa’. As a result, ROSA FULVULTRA could be diploid, triploid or possibly even tetraploid if unreduced gametes are involved. Statistically, the most likely outcome would be diploid, just as it is registered. However, according to Bill Powell, ROSA FULVULTRA is “sinfully fertile with diploids” and produces pods full of seeds when crossed with tetraploids. Not all the seeds from the tetraploid crosses germinate, but some do and produce viable offspring.” Based on this statement, we can rule out tetraploidy as it is impossible for a tetraploid (4n) to produce a diploid (2n). So, could ROSA FULVULTRA be triploid, with better than average fertility? Or a diploid that produces unreduced gametes. The answer is not clear without pollen analysis and/or genetic testing although we will see a similar situation in the following example.
- GARNET ROBE (Milliken G. ’48) diploid
- Parentage: Unknown
- Offspring: 7 Diploid
- POLLY PINK (Saxton ’80) diploid (noted as “ploploid” on checklist)
- Parentage: GARNET ROBE X FAIR ANNET* (Peck ’67)
- Offspring: None
GARNET ROBE is registered as a diploid and has been noted to have low fertility. Arisumi chromosome tested the root tips of GARNET ROBE and found that it is actually triploid. He also noted that GARNET ROBE was a better than average seed setter for a triploid. He set four pods from ten pollinations and ended up with nine seedlings. Eight seedlings were tetraploid and one was triploid. In comparison, he had made over 1000 crosses with H. fulva L. and ‘KWANSO’ and did not produce any seeds. So it appears that GARNET ROBE has very low fertility with diploids, better fertility with tetraploids, and excellent fertility considering it is a triploid.
Crossing a triploid with a tetraploid can produce only another triploid, or a tetraploid. POLLY PINK is a GARNET ROBE offspring and therefore, must be either triploid or tetraploid but it cannot be diploid as it is registered. According to a website update to an article entitled Triploids Are Fertile from the Spring 1994 Daylily Journal, Nick Chase wrote, “POLLY PINK will set pods with tetraploid pollen (and the seeds germinate), and its pollen sets pods on both diploids and tetraploids, and the seeds of both types of crosses will germinate.” Based on this statement, it is very probable that POLLY PINK is actually triploid.
This scenario has many similarities to that of ROSA FULVULTRA. Both POLLY PINK and ROSA FULVULTRA cross readily with diploids and tetraploids. In his correspondence with me, Bill Powell stated that H. “fulva triploid pollen will set the occasional seed on diploid plants and the progeny from them seem to be able to go both ways”. I have also read that this sometimes occurs with crosses of H. citrina and other daylily species.
These last four cultivars studied raise many questions. Could the ability of ROSA FULVULTRA and IMPLAUSIBILITY to seemingly cross both ways be related to the presence of the species in their immediate ancestry? Is this a recessive trait that has been bred out of many modern cultivars? Or, could these cultivars be triploids with greater than average fertility, similar to GARNET ROBE? Lastly, could they be diploids and produce unreduced gametes? At this point, without further testing, we cannot determine whether these abilities are caused by unreduced gametes, recessive traits, fertile triploidy or some other phenomenon altogether.
Moving on, LIGHTS OF DETROIT* (Weston ’82) seems to be the second most well known daylily with the possibility of unreduced gametes. It is a registered tetraploid from a cross of BOYD HORTON X FABULOUS PRIZE*.
- LIGHTS OF DETROIT* (Weston ’82) tetraploid
- Parentage: BOYD HORTON X FABULOUS PRIZE*
- Offspring: 6 diploid (pod parent for all)/1 tetraploid
- Parentage Information:
- BOYD HORTON (Blakely ’73) diploid
- Parentage: Unknown
- Offspring: 1 Tetraploid
- FABULOUS PRIZE* (Brown ’74) tetraploid
- Parentage: Sdlg X Induced (PRESIDENT GILES x EDGAR BROWN)
- Offspring: 45 (all registered as tetraploid.
According to Judith Weston, LIGHTS OF DETROIT* (LOD) was registered as a tetraploid, but it crosses both ways with both diploids and tetraploids. Judith Weston had the pollen genetically tested by the University of Michigan. It was “guessed” to be genetically tetraploid but the pollen sample was poor (had aged too much) by the time it was tested and so the results were not certain.
The problem with the unreduced gamete theory in this case has been that LOD was thought to be tetraploid. It is genetically impossible for a tetraploid to produce a diploid so LOD cannot be tetraploid as it has six diploid offspring. At this point, we must discount the conclusion of the University of Michigan because the pollen sample was of poor quality and the results were basically a guess. Additionally, LOD’s pollen has been visually examined on several occasions and found to be small diploid in size. Between the pollen inspection and the fact that it produces diploids, we can eliminate any possibility that LOD is tetraploid. Therefore, LOD’s pollen parent cannot be FABULOUS PRIZE* and must be stray diploid “bee” pollen.
A diploid can only produce a tetraploid if it produces (2n) pollen from unreduced gametes or some other mutancy. LOD also has one tetraploid registered offspring.
- GRAND MARNIER SOUFFLE* (Weston ’96) tetraploid
- Parentage: Not Available per AHS records
- LOD parentage twice as pod parent and twice as pollen parent.
- Offspring: 5 Tetraploid
One possible explanation for this is that LIGHTS OF DETROIT is diploid and produces a low level of unreduced gametes. However, it should also be noted that on several occasions LOD pollen was examined under a microscope and not found to contain any UG pollen. Its pollen was noted as looking even smaller than “normal” for diploid pollen. Additionally, Oscie Whatley used 50 plants of LOD and pollinated 25 with only diploid pollen and the other 25 with only tetraploid pollen. He found LOD to be very fertile with diploids. From the tetraploid crosses he obtained only three seeds, two were not viable and the third resulted in a diploid offspring.
How can we account for these findings? It has been shown in other plant species that environmental factors, such as high temperatures, can affect the ability to produce unreduced gametes. Could LIGHTS OF DETROIT only produce unreduced gametes if environmental conditions are favorable?
Another speculation is that LOD produced unreduced gametes on one occasion due to some meiotic malfunction. Arisumi estimated that unreduced gametes occur in 1/15,000 ovules. Maybe this was the one—not probable—yet still possible. Unfortunately, we just don’t know the answers to these questions.
One of LOD’s registered offspring, SUBSTANTIAL EVIDENCE (Norris ’03) is reported to also cross with both diploids and tetraploids.
- SUBSTANTIAL EVIDENCE (Norris ’03) diploid
- Parentage: [{LIGHTS OF DETROIT* x WHEN I DREAM (Yancey ’79)} x (LIGHTS OF DETROIT* X WHEN I DREAM)} X SPECIAL INVITATION (Santa Lucia ’95)
- Offspring: None Registered
Norris states that SUBSTANTIAL EVIDENCE crosses both ways readily with diploids, but also crosses as a pollen parent with tetraploids. The resulting tetraploid seed pods produce mostly one (at the most two) seed(s) per pod. According to two different sources, the flat form of the LOD heritage is readily apparent in the resulting tetraploid offspring.
If LIGHTS OF DETROIT is able to produce UGs, the double shot of it as parent in SUBSTANTIAL EVIDENCE’s background would increase the likelihood that SUBSTANTIAL EVIDENCE could also produce UGs.
According to an article in the Daylily Journal, Vol 45, No. 4, 1990 by Dr. Joseph Halinar, “The production of unreduced gametes is under genetic control. Thus, we can cross parents that produce unreduced gametes and greatly increase the chances that the resulting progenies will also produce unreduced gametes. Crossing a diploid that produces unreduced gametes with a diploid that does not will most likely result in progenies that do not produce unreduced gametes. However, self-pollinating or sib mating these seedlings should result in about 25% of them producing unreduced gametes.”
Only one seed per pod is usually produced when SUBSTANTIAL EVIDENCE’s pollen is used on tetraploids. This is exactly what would be expected in a case of unreduced gametes. Only the small amount of 2n pollen mixed with the 1n pollen would generate viable seed, resulting in very few seeds per pod. Of course, this same explanation could be used for stray tetraploid pollen contamination among the diploid SUBSTANTIAL EVIDENCE pollen resulting in the tetraploid offspring. This could also be the case with LOD offspring, GRAND MARNIER SOUFFLE*.
Since LOD must be diploid, SUBSTANTIAL EVIDENCE was not produced from an unreduced gamete cross. This does not mean that it is impossible for LOD and/or SUBSTANTIAL EVIDENCE to produce UGs. Just that they were not involved in this particular cross. As Arisumi estimated, the frequency of unreduced egg cells is 1/15,000 in the general population.
Unfortunately, there are just too many unanswered questions in this case to make any determinations about which scenario above (if any) are correct.
CONCLUSION:
While I had hoped to gain some definitive answers from this study, seemingly only more questions were raised. In my opinion, the ability to cross both ways with diploids and tetraploids appears to be genetic in some cultivars. Whether this ability is due to unreduced gametes, recessive traits, triploidy, “meiotic mutancy”, or some other phenomenon remains to be seen. Due to all the unknowns involved in these cases, we may never know with certainty what is occurring without testing to determine ploidy as well as genetic testing of the gametes. Unfortunately, even with genetic testing, we may never know for sure. Any pollen or ovule sample tested is just a “snapshot in time” and may or may not reflect what can occur in a particular cultivar.
At the very least, it was shown that it is possible to cross diploids and tetraploids and get viable offspring (even if only occasionally). Cross-ploidy hybridizing and hybridizing with species could conceivably open up a whole new world of possibilities in daylilies. All that is required is patience.
This summer, controlled reciprocal crosses to determine the frequency of unreduced gametes among specific cultivars will be conducted. Controls involve stigma protection to prevent pollination from unknown sources, as well as measures to prevent pollen contamination prior to its use. In addition, there are tentative plans to chromosome test the pollen of ED MURRAY. The results of these experiments will be made available when they are known.
A special thanks to all who have helped me with this project, especially Maurice Dow, Rebecca Hutchins, Brian Mahieu, Melanie Mason, and Richard Norris.
Footnotes/References:
1. Arisumi, Toru. Experiments In Breeding For Triploid Daylilies. The Daylily Journal. Volume 24, No. 2, June 1970, pp. 33-37.
2. As above.
3. Chase, Nick. Triploids Are Fertile. Spring 1994 Daylily Journal.
4. Arisumi, Toru. Triploidy As A Cause Of Low Fertility In “Garnet Robe”. The Daylily Journal. Volume 25, No. 1, pp. 32-35.
5. Halinar, Dr. Joseph. Polyploidy and Unreduced Gametes. The Daylily Journal, Vol 45, No. 4, 1990.
6. Chase, Nick. Daylilies Online. Website at http://nick.assumption.edu/Daylilies/about.html.
7. Archives of the AHS Email Robin at http://maelstrom.stjohns.edu/archives/daylily.html.
APPENDIX A
Below are the names of additional registered cultivars that were sent to me by Robin members. I have included any pertinent information that I could find relating to unreduced gametes with these cultivars.
Click on the link below to view some interesting pictures of pollen taken at high magnification.
