Breeding

by Don Smith
Summer 1997

Part I
Japanese tree peonies (Paeonia suffruticosa) are among the most beautiful of the flowering plants known to man. Unfortunately, they are much less popular than they deserve to be, especially in the U.S. I believe that part of the reason for this is due to their lack of fragrance. People expect beautiful flowers to smell beautiful and are usually disappointed when they do not. Fortunately, this is a problem that can now be remedied. We are in a position to breed fragrance into the tree peonies in the same way that yellow flower color was bred-in, through the use of Paeonia lutea and its hybrid progeny. Therefore, someone interested in hybridizing tree peonies should consider trying to develop a race of fragrant tree peonies as a long term goal.This goal should not be that difficult to accomplish because much of the ground work has already been accomplished by others.

Paeonia lutea has a pleasant lemon-like fragrance (also described by some as lily-like) that is sometimes passed on to its hybrid offspring. Crossing Paeonia lutea with Paeonia suffruticosa (Paeonia lutea x Paeonia suffruticosa) has resulted in a fair number of fragrant hybrid tree peonies, many of which are currently available. For example, the Daphnis advanced generation hybrid, Paeonia 'Leda' (lutea group) (Paeonia suffruticosa 'Kokamon' x Paeonia 'BC2' [lutea group]), has a very strong, wonderful, rose-like fragrance that is among the best in the peony world. Unfortunately, Paeonia 'Leda' (lutea group) is not especially fertile when crossed with other hybrid tree peonies and particularly when back-crossed with Paeonia suffruticosa. lt has yet to produce a single viable seed for me used either way. However, Zlatana Draskovich has reported some success using this plant as both a pollen and pod parent (see APS Bulletin No. 295, Sept. 1995). In addition, a few other fragrant tree peonies have also shown useful levels of fertility. By interbreeding these fragrant varieties and back crossing them with Paeonia suffruticosa, we can begin to produce a new race of fragrant hybrid tree peonies that approach the Japanese varieties in both flower size and quality. A list of some fragrant varieties that may have usable fertility is given in Table 1 below. All of these plants are commercially available.

Table 1.
Fragrant tree peonies that have been reported to be useful as parents.

Part II
Success in breeding fragrant tree peonies would be greatly enhanced if genes for fragrance were present on both sides of a given cross. To achieve this in back-crosses with Paeonia suffruticosa varieties it would be helpful to have on hand a list of fragrant Japanese (and Chinese) tree peonies. However, whereas lists of fragrant herbaceous peonies are extremely common, similar lists of fragrant tree peonies vareties are rather hard to come by.

Fortunately, several commercial sources of Chinese tree peonies have included information on fragrance in their catalog descriptions of listed varieties. From these recent catalogs I have compiled a list of fragrant Chinese tree peonies, broken-down into three catagories:

Unfortunately, I have no information on the usefulness of any of these varieties as parents. However, in some cases I have been able to indicate (based on photographs) which varieties have stamens and / or carpels. Many, if not most, of these varieties are probably fertile to some degree and all are commercially available from one or more U.S. sources. These fragrant tree peony varieties are listed below:

To my disappointment, I was not able to come up with a similar list of fragrant Japanese tree peony varieties. Fortunately however, one of our group, Theresa Griesbach, sent me a list of fragrant Japanese tree peony varieties obtained from a book by Michael Haworth-Boothe:

This is the first list of fragrant Japanese tree peony varieties that I have seen. Those who are aware of other fragrant varieties are encouraged to write so that I can try to build a "complete" list of all fragrant Paeonia suffruticosa varieties for future use. I do not know how many of these varieties are available commercially or which ones have proved to be useful parents, but once again, most probably have useful fertility. One of the varieties listed, Paeonia suffruticosa 'Koka-Mon', is the seed parent of the extremely fragrant (rose-scented) Daphnis hybrid, Paeonia 'Leda' (lutea group). And Paeonia 'Leda' (lutea group), when used as a seed parent with Paeonia suffruticosa 'Shima Daijin', produced the very fragrant new cultivar Paeonia 'Marilyne Jones' (lutea group) (APS Bulletin No. 295, Sept. 1995). These results suggest the possibility that heavy rose fragrance in tree peonies might be transmitted through the female parent by the mechanism of maternal inheritance. Paeonia 'Marilyn Jones' (lutea group) will probably not be available for some time to come, but it does appear to be fertile both ways and will likely prove to be much more fertile than its pod parent Paeonia 'Leda' (lutea group).

An analysis of the list of registered herbaceous varieties in the book "Peonies, Outdoors and In" indicates that of the 1289 peonies registered up to 1959 there were 237 (18.4 %) varieties which were listed as fragrant to some degree (from very to slightly). Of these fragrant peonies, only 53 were listed as having a pleasant (sweet or rose) fragrance. These amount to only about 4 % of the total.

By comparison, of the approximately 170 hybrid tree peonies registered since Professor Louis Henry produced the first lutea hybrid in 1897, at least 36 (21 %) are reported to be fragrant. How many of these 36 could be described as having a strong, pleasant fragrance is uncertain due to a lack of specific information. However, it is interesting to note that, of these three dozen fragrant varieties, more than 80 % are advanced generation hybrids.

By dividing the hybrid tree peonies (lutea hybrids) into two groups:

we can make an interesting observation concerning fragrance. In the advanced generation group, nearly 50 % of the hybrids are reported as being fragrant, whereas only about 6% of the F1s are listed as having a fragrance. Therefore, advanced generation hybrids are at least 8 times more likely to be fragrant than their F1 counterparts. It is important to note that this result occurs without any deleberate attempt to breed for fragrance. It seems clear that these results could be greatly improved with a conscious effort to breed for fragrance by using only fragrant varieties as parents.

In Part I of this article, I listed a number of fragrant hybrid tree peonies that had demonstrated useful fertility as either pod or pollen parents. In Table 2, I list 6 additional fragrant Lutea hybrids that have also shown some fertility. They are all F1 hybrids and all are commercially available.

Table 2.
Additional fragrant hybrid tree peonies that have been reported to be useful parents.