By: Marily Young, roseynut[at]aol[dot]com, Consulting Rosarian
Within the past 5 years, the popularity of shrub roses in the United States has increased dramatically. Rosarians cite many reasons for this trend, including the increased disease resistance and hardiness which many hybridizers have introduced into the shrub class.
Whether the current interest in English Roses by David Austin helped to start this trend, or whether his roses have simply benefited by being at the right place at the right time, is subject to much debate. In either case, it is certainly true that there has been a corresponding huge increase in the popularity of the Austin roses over the same time period.
However, many rosarians have asked the question “What is an English Rose?” Are there any characteristics that define an English rose? Is there anything that would justify separating this group of roses into a separate horticultural classification? Furthermore, the Austin “English” roses remain a frustration to many of us in the United States because it seems impossible to predict the growth habits, hardiness, and disease-resistance of most of the varieties, when grown in different parts of the country. In an attempt to address some of these questions, I have, over the past several years, studied the genealogy of the Austin roses. I hoped that I could find some “predictive” quality which could be identified in their breeding.
To begin with, David Austin himself makes an attempt to organize the results of his efforts into eight different “strains”, based on the flower characteristics and growth habits. The eight strains used by Austin to categorize his roses are:
- “Old Rose”
- “Mary Rose”
- “Wife of Bath”
- “Gloire de Dijon”
- “The Squire”
Austin’s classification system, however, was based on the characteristics and habits of the roses when grown in the United Kingdom.
For example, he describes the “Heritage” strain, including varieties such as Heritage itself, Perdita, and Graham Thomas, which “tend to have short, rather shiny leaves, good bushy growth and usually a ‘tea rose’ fragrance”. Unfortunately, it seems that in most parts of the U.S., the growth habits of these three varieties are nothing alike! Perdita seems to stay rather compact, forming a relatively open bush. Heritage, on the other hand, grows to almost twice that size and forms an upright and dense shrub. Graham Thomas (although somewhat tender in cold areas) often grows very large, making long, slim canes 7′-9′ in length.
When the “blood lines” of these three roses are examined, some interesting differences appear. Although Heritage and Graham Thomas both go back to Duchesse de Montbello (Gallica) and Belle Isis (Gallica), Graham Thomas comes to us through the Chaucer/Charles Austin group, which includes the large-flowered modern climber, Aloha. Heritage is part of the Wife of Bath group, whose parentage includes the more moderate-growing hybrid tea Mme. Caroline Testout, and the Floribunda Ma Perkins. The climber heritage, versus the HT/Floribunda heritage, seems to be quite distinctive.
As another example, let’s look at some of the roses which Austin groups into his “Gloire de Dijon” strain, including Tamora, Jayne Austin, Evelyn, and Sweet Juliet. Jayne Austin and Evelyn both have Tamora as the pollen parent, which in turn has Conrad Ferdinand Meyer (HRg) and then Gloire de Dijon (Cl T). These three all have apricot/yellow blooms and a relatively compact, upright growth habit. But Sweet Juliet? All it seems to have in common is the color. It is a much more vigorous grower, often reaching 6 feet tall. This difference seems to be explained by the fact that Sweet Juliet does not have Gloire de Dijon on the pollen side at all, but instead expresses the characteristics of its seed parent, Graham Thomas.
As an alternative to a grouping based on visible characteristics, I believe it is much more useful and practical to evaluate Austin’s English roses based on the primary Old Garden Roses or modern roses from which they are descended. In my experience, most of the Austin roses behave similarly, in a given climate, to the “Eve” rose from which they are descended.
The groupings that I have begun to use in an attempt to predict how Austin roses will grow in my yard are the following:
The varieties in this group, like Graham Thomas, seem to express the influence from the pollen parent Aloha (LCl), even though they do include Duchesse de Montebello and Belle Isis, both Gallicas, in their backgrounds. Included here are many of the very vigorous growers, and many of Austin’s yellow and apricot roses. Abraham Darby, Golden Celebration, Graham Thomas, Evelyn, and The Pilgrim all belong here. It is also worth noting that Austin seems to be using Graham Thomas quite a bit in his most recent breeding, back-crossing it with roses from most of his other groups.
Although the seed parent of Lilian Austin is Aloha, the characteristics of the pollen parent, The Yeoman, seem to dominate this group. The Yeoman includes much Floribunda influence, as well as the Gallica Belle Isis, and the roses in this group tend to stay a more manageable size. In fact, these roses tend to conform more closely to the sizes that Austin claims for them in their home climate. Roses in this group include Bibi Maizoon, Warwick Castle, English Garden and Queen Nefertiti.
The Wife of Bath/Mary Rose
This group of English roses is, unfortunately, not as predictable as it seemed at first. Austin used The Wife of Bath extensively as a seed parent, crossing it with many different pollen parents. What I have found striking about this group is that with the exception of Gertrude Jekyll and Mary Rose, most of the resulting varieties have remained little-known. This group includes such roses as Dove, Mary Webb, Yellow Button, Country Living, Robbie Burns, and Cottage Rose. I think that, when considering any offspring of The Wife of Bath, one must look at the pollen parent to make a reasonable guess as to the growth habits.
The match of Dusky Maiden (Fl) with Tuscany(G) produced Chianti, one of Austin’s first English roses. One of the Chianti seedlings was The Knight, which was then crossed with Chateau de Clos Vougeot (HT) to produce The Squire and Prospero. These, in turn, have fathered a remarkable group of red and dark red roses which all seem to have very powerful damask fragrance. In this group are Fisherman’s Friend, Othello, The Prince, and The Dark Lady. Many of these roses look very similar, and have been mis-identified by growers, exhibitors and judges! Fortunately, most of the roses in this group also show the Gallica hardiness.
Last, but not least, is a very small group which originated in a cross between Ivory Fashion and an unnamed seedling. The result of that cross was The Friar. This rose, self-pollinated, produced Admired Miranda; with Iceberg as the pollen parent, it produced Perdita. Although I wish it were otherwise, it appears that Austin has stopped using these roses in his breeding.
Much of the above discussion has been based on my experiences growing Austin roses in USDA zone 5 (-20°F) and on my visits to other gardens in this area. I realize that most Austin roses grow significantly larger in the warmer parts of the U.S., but I think that the relative proportions remain the same. Unfortunately, size is virtually the only characteristic that I have found it possible to group based on genealogy. Fragrance and color are very predictable in the Chianti line, but less so otherwise. Most of the Austin roses seem to be cold-hardy, with the exception of Graham Thomas and some of its descendants. Most also appear to adapt well to heat. I do wonder, however, if those with a strong Gallica background (the Chianti group) need substantial chill-hours in order to bloom well. (I would appreciate hearing from those in the South and Southwest on this issue, since I have a surplus of chill-hours….)
Finally… should there be a separate taxonomic group for the Austin roses? I’m afraid I still don’t know. Austin’s breeding program has, so far, been very “tight” and has used a limited number of Gallicas, Floribundas, and modern climbers, with a few classic Hybrid Teas thrown into the mix. He has done many, many back-crosses among his own breeding lines. I propose that if Floribundas (basically a cross between Hybrid Teas and Polyanthas) constitute a separate class, then so do the Austin English roses. Whatever their class, they are some of the most interesting roses in my garden!
To download the pdf version of this article click here.