Bull. AIS 233: 73-77 (1979)
Currier McEwen, South Harpswell, Maine

Since the duration of bloom is of great significance in any flower, the capacity to rebloom is important — especially in irises, which usually bloom for only two to four weeks. I first became interested in rebloom in Siberian and Japanese irises in 1972. Prior to that, I grew two reblooming Siberians, MY LOVE and VIOLET REPEAT, but I had received them as gifts; I had not sought them for their reblooming trait.

In 1972 those two, plus several of my own Siberian and Japanese seedlings, put on a fine show after bloom had ended in the others, and I could not help but be impressed. That was the year I began crossing those cultivars showing the reblooming trait, and I have done so each year since. During 1977 and 1978 especially, I kept detailed records of the performance of the Siberian rebloomers in our garden. Thus far, I have made less careful observations on the Japanese rebloomers. Hence this article is based primarily on Siberians with only briefer comments regarding Japanese irises.


The time of rebloom in Siberian and Japanese irises, as I reported previously,1 differs strikingly from that in tall bearded irises. As is well known, the latter usually have a rest period of several months following first bloom until bloom occurs again. In contrast, the rest period in Siberian and Japanese irises usually is short, lasting only one to three weeks. In some, the new stalks for the second period of bloom may be showing small buds by the time the last of the first bloom fades.

On the basis of experience over the past two years, I believe there are two fairly distinct types of reblooming Siberians, those which rebloom only occasionally and those which, I believe, deserve the term "preferential rebloomers." Zurbrigg has used the terms "confirmed" and "proven" rebloomers in writing about bearded remontant irises2 but I think "preferential" is more descriptive in the case of Siberians. In them, cultivars of this type perform better at second than at first bloom.

The occasional rebloomers do not do so every year; they send up more — usually many more — stalks at first bloom than at second, and the stalks at first bloom carry many more branches and buds than do those at rebloom. Conversely, the preferential rebloomers send up more stalks at rebloom than at first, and the reblooming stalks are better branched and budded. Those which were mature enough to rebloom in 1977, did so also in 1978.

During 1978, 58 of my Siberian irises rebloomed. Of these, 28 met the criteria for preferential and 27 for occasional rebloomers. Another three showed some features of each. Some characteristics of the occasional and preferential ones are shown in the table that follows, in terms of average figures.

  Column 1 Column 2 Column 3 Column 4
  Average Stalk
Height in inches
Average Number
of branches
Average Number
of buds
Average Number
of stalks
Bloom Category 1st Bloom Rebloom 1st Bloom Rebloom 1st Bloom Rebloom 1st Bloom Rebloom
Occasional 29 35 (+6)* 0.3 0.5 (-.3) 3.5 2.5 (-l.) 5.1 l.8 (-3.3)
Preferential 22 34 (+12) 0.6 1.1 (+.5) 2.7 3.6 (+.9) 3.2 4.6 (+l.4)
*() = difference between first bloom and rebloom
- = decrease
+ = increase

As shown in column 1, the average height of the rebloom stalks was greater for both occasional and preferential rebloomers, than the height at first bloom. This is especially true of those in the preferential group and important because, in some of them, the stalks at first bloom have been too short, placing the flowers down in the foliage. This failure will, I trust, be corrected by selective breeding.

Column 2 shows the average number of branches. None of those categorized as occasional rebloomers had more branches at rebloom than at first, and some with a branch at first bloom had none at the second period. In contrast, no preferential rebloomer had fewer branches at second bloom; 25 of the 28 had more.

In column 3, the data for average number of buds were similar to those for branches, but the difference between occasional and preferential ones for the two periods was even more pronounced.

Adding the number of buds at first and second bloom, the total number of flowers in a year is only a little more for preferential than occasional rebloomers. However, since the preferential ones can be expected to rebloom much more consistently year after year, their total amount of bloom over a period of years should greatly exceed that of the occasional ones.

Most striking of all, as shown in column 4, were the data regarding the number of stalks. Since some clumps in both groups, especially among the occasional reblooming cultivars, were older and larger with more stalks, the numbers shown for the average number of stalks at first and second bloom between the two catagories, probably are not accurately representative. The differences in each category between first bloom and rebloom, however, are accurate.

The number of stalks at second bloom among the occasional rebloomers was small compared with the first bloom, rebloom in them was not very significant. Rebloom in the preferential group, on the other hand, was a very impressive and useful feature in the garden.

I believe that the yearly reblooming reliability of the preferential group will mark another impressive difference. Certainly the cultivars in the occasional category have been very spotty in their rebloom from year to year although some are more consistent than others.

I have had CAMBRIDGE and VIOLET FLARE for at least ten years and they have rebloomed, rather sparsely, only once. RUFFLED VELVET also has rebloomed only once in seven years, but that one time was impressive. DREAMING YELLOW has rebloomed fairly well in each of three clumps, in four of seven years.

In contrast, most of the preferential clumps which have been of blooming age and size for several years have rebloomed abundantly each year, although MY LOVE has not been completely consistent. However, many of those which I have categorized as preferential bloomed and rebloomed for the first time in 1978, and their future performance remains to be seen.


My experience to date is too limited to permit more than rather general comments about the genetics of the reblooming trait in these irises. However, a few observations can be cited. WELCOME RETURN, a colchicine induced, tetraploid preferential rebloomer from a cross of two other preferential ones (ON AND ON and MY LOVE), when crossed with another preferential, induced tetraploid closely related to it, has given only preferential rebloomers. The same was true when it was selfed. Of four "bee pod" seedlings from WELCOME RETURN, three were preferential and one had features of both groups.

Another cross—MY LOVE (preferential) by WHITE MAGNIFICENCE (occasional) — gave two rebloomers; one showed features of both categories, and the other was fully "occasional." I hope to have better data in another two years.

I have said that three cultivars seem to fall between the two categories. One, the child of MY LOVE and WHITE MAGNIFICENCE, mentioned above, put up twice as many stalks at first bloom as at second but those at second bloom were better branched and budded.

Two others, (the seedling from WELCOME RETURN by unknown noted above, and a third generation tetraploid going back to SNOWY EGRET twice and to WHITE SWIRL) both sent up about three times as many stalks at second as at first. The stalks at first bloom were better branched and budded. It is not surprising that there should be cultivars with these mixed features; but, on the whole, the two categories of occasional and preferential rebloomers seem to hold up fairly well.

One may question to what extent the capacity to rebloom is governed by genetic and physiologic factors. My experience leaves me in no doubt that the trait is chiefly genetically determined, but that physiologic, or cultural, factors are also important. Even a preferential rebloomer may not rebloom the first season after transplanting, or if crowded or unhealthy. The physiologic state of the plant is even more important in governing rebloom in cultivars which have the genetic potential for only occasional rebloom. My Siberians have now been growing for 12 years in beds that have never been fertilized. (I do not mean that I recommend using no fertilizer for Siberian irises! I intend to make myself do it next year) I suspect rebloom might be better if I pampered them more.


Perhaps it will be useful to list the named rebloomers I know. Most in our garden are merely under number and I will not list those.


Occasional Rebloomers: AUGURY, BLUE BURGEE, BUTTER AND SUGAR, CAMBRIDGE, DREAMING YELLOW, OUTSET, RUFFLED VELVET, SNOW QUEEN, SOOTHSAYER, TOKAY GRAPE, and VIOLET FLARE. Others, which I am told rebloom but which I have not as yet had an opportunity to observe, include BLUE RIDGE, ERIC THE RED, SILVER TIP and ZEST.

The comments thus far have dealt only with Siberians of the 28-chromosome group and tetraploids derived from them (subseries Siberiricae). I have not as yet observed cultivars of the 40-chromosome group (subseries Chrysographes) in detail, but have four that rebloom well: BLUE FORTY, ECHO II, MAUVE MOOD and PURPLE PRINCESS. Of these, rebloom on ECHO II and PURPLE PRINCESS in our garden has been especially abundant, but I do not have details as to how these cultivars compared in various features at first and second bloom.

Similarly, I do not have a detailed record of the performance of my Japanese iris rebloomers, but casual observations suggest that they are like the Siberians. Many appear to be of the occasional type whereas others, such as GARDEN CAPRICE and PURPLE PARASOL, are preferential rebloomers.


It is too early to make solid decisions about terminology regarding rebloom in Siberian irises, but I think there are at least two aspects that need further study leading to decisions. One has to do with what I have called the preferential and occasional categories in this article. Are the terms which I have described valid and suitable?

The other aspect is concerned with the broader question of what is rebloom in Siberians. Raymond Smith in his chapter on rebloom in The World of Irises3 has defined a rebloomer as "an iris that produces an extra period of full bloom each year." He continues that, "By full bloom is meant bloom of one or more increases from each rhizome that flowered during the immediately preceding regular period. This definition excludes those sorts whose stalks emerge serially over an extended season, as well as clones with only a fraction of the mature rhizomes blooming during one period and the remainder during the next …"

WHITE ENCORE, a reblooming Siberian variety by Dr. Currier McEwen

The Reblooming Iris Society, however, has currently accepted as a rebloomer any iris with a significant bloom period other than the regular spring bloom season.4 I must emphasize at once that I do not know the means by which the second period of bloom appears in Siberian and Japanese irises. In this article, I have used the term "rebloom" not in a botanical, but in a purely descriptive, sense to indicate a second or even third period of bloom following the regular period.

As mentioned in the beginning of this article, second bloom in Siberian and Japanese irises usually follows a rest period of only a few weeks. This is so different from the usual remontant behavior of bearded irises, that one faces the obvious question — whether it should be considered rebloom or extended bloom. In my own records, I have called rebloomers those cultivars which show no starting of new stalks when the last of the first period of bloom fades. Those in which new stalks have started, I have referred to as "extended bloomers." Perhaps this is splitting hairs.


As a final note, let me say that there is still very fresh in my mind the performance of WELCOME RETURN in 1978. It bloomed first in June, again abundantly in July and then sent up one more stalk in September. Last year one of its children did the same, as did VIOLET REPEAT. As I write these notes October 15, my Japanese iris PURPLE PARASOL opened its last bud this day, after blooming first in July then again abundantly in August and sending up this one stalk in late September.

Rebloom in these beardless irises is rewarding and, I am sure, will be steadily improved in reliability and abundance of rebloom as well as in quality of flower. Out of some 3,500 Siberian irises of blooming age and size in our garden in 1978, 58 or about 1.7% rebloomed. In contrast, 52% of seedlings from crosses of preferential parents rebloomed. All of these had been lined out as newly sprouted seedlings in the spring of 1976. They were young and had been planted only 6 to 8 inches apart and hence were crowded. I have no doubt that still more would prove to be rebloomers when older and given better growing conditions. The point I wish to make, of course, is that the reblooming trait is readily enhanced by selective breeding. Hence, one can look forward confidently to steady improvement.

  1. McEwen, C., "Reblooming Siberian Irises," The Siberian Iris, Vol. 4, No. 2, Fall 1975, pp 20-21.
  2. Zurbrigg, L., Personal Communication.
  3. Smith, R.C., "Rebloom," The World of Irises, Warburton, B. and Hamblen, M., The American Iris Society, 1978, Chapter 7, p. 136.
  4. Zurbrigg, L., "Definition of a Reblooming Iris," Reblooming Iris Recorder, No. 15, Fall 1978, p.8.