The identification of Sorbus species   

The identification of Rowans can be very difficult.

The best method I know is to use the taxonometric key developed by Dr McAllister. [1].

The problem I find with this is that it was developed for those familiar with botanical nomenclature, of which I'm not one, so I have to keep looking up the meanings of the terms used, (e.g. carpels, papillose, calyx lobes, styles, etc).

It is also based on a variety of traits, some of which do not occur together, (e.g. the appearance of the young leaves and of the fruit), so you have to take notes and wait.

I think identifications problems are also exacerbated by the tendency of the apomicts to form small groups of very similar plants [2].

The key occassionally uses actual sizes of parts of the plant for identification 'though Dr. McAllister warns the user of variations with age and growing conditions.

Another problem with Rowans is the description of berry colour. 'Yellow', 'Orange', 'White' and 'Pink' are probably OK but there is a lot of uncertainty about 'Red' which can mean rather different things to different people. I never refer to any berries as 'Red', using 'Scarlet' for the Sorbus subgenus and 'Magenta', where applicable, for the Albocarmesinae group.

However, don't be too put off! The tree you're trying to identify was probably bought from a nursery rather than collected in the wild. Whilst you can't always believe their nomenclature, (I've been sold five S. matsumurana; all of which were commixta!), you can check the given name against the key to see if it's correct and the species in general cultivation should be easy to identify.

But one further warning! Dr McAllister's key is for species and doesn't include any information on cultivars or hybrids, other than the naturally occuring ones of Sorbus sp. 'Ghose' and Sorbus 'Joseph Rock'.

I've produced a computer version of the key which can be used as an on-line, interactive app. or downloaded, (1.7Mb) as a free-standing application which has been scanned and found to be free of malicious items by Avira anti-virus. Do download and read the ReadMe(rk).txt file before installation. One problem with this package is that it doesn't contain data on any cutivars or new species and I don't feel competent to modify the key.

I'm also writting an app. that gives an approximate ID on the basis of berry, bud & leaf colour together with the leaf pair count & relative leaflet size. Its advantage is that also contains data on the cultivars we have and on any new species discovered. e.g. Sorbus keenanii

A trial version of this is available which can be downloaded, (2.6Mb) as an installation file which has been scanned and found to be free of malicious items by Avira anti-virus. Definitely download and read the ReadMe(id).txt file before installation.

There is also a work sheet file for gathering the data you require, available as either a Word document or an RTF file.