Just about any annual that is constant in its characters and goes by a cultivar name (as opposed to a 'strain' or 'series'). Think of any of the old-fashioned bean or tomato varieties.

One point: the concept of cultivar seems to have been muddied by the legalistic approach of authorities responsible for granting plant patents / plant variety rights. The impression I get is that they insist that a name given to a patented plant is NOT a cultivar name, and whenever it's used (e.g. on a plant label) it must be accompanied by some little superscript symbol in similar style to a trade mark.

In the case of woody plants there has been a tendency to apply the rule, different clone therefore different cultivar. In practice this is hard to maintain, because, if given 2 plants of identical appearance, you could only claim they were the same clone if you could trace the propagation of each back to a common plant -- either that, or match their DNA. And this cuts the other way too: think of all the Camellia cultivars that have originated as branch sports, often several from the one parent cultivar which are then all part of the one clone AND in all likelihood have identical DNA (at least as far as that can be identified with present techniques), even though they may be differently colored and/or have different petal conformations.

No doubt there is a definition of cultivar in the current International Code for Nomenclature of Cultivated Plants. Unfortunately this is a rare document, not easy to access. It's supposedly available from the International Society for Horticultural Science (headquarters in Belgium) for 55 euros, a large price for a slim paperback -- though good news, it's available online! -- you can download it as 13 articles for a fee of only 10 euros per article! Order it at http://www.actahort.org/books/647/. Or Amazon.com have listed a single secondhand copy at $100, which raises the suspicion that the printed version is no longer available new.

Contrast this with the International Code of Botanical Nomenclature, available FREE online at http://www.bgbm.fu-berlin.de/iapt/nomenclature/code/SaintLouis/0000St.Luistitle.htm

The term cultivar is short for "cultivated variety." It is any variant found in gardens as opposed to taxonomic varieties found in the wild as natural, self-sustaining populations.

Horticultural recognizes varieties, too. Bud sports would also be "variants found in gardens." I'm not looking for the definition of "cultivar" - I think I'm pretty familiar with that - but I was asking about ones that aren't vegetatively reproduced, which seem to be pretty common to me. I know that highly inbred plants can be cultivars and I guess that's what TonyfromOz was referring to with annuals that would not, almost my definition, be propagated vegetatively.

Yes, Serenoa is quite mistaken -- cultivar is not an everyday English word but a term deliberately coined to fill a particular need, even though its derivation is transparent. It came into use around 1950 I think, though may have been invented a bit earlier, and its application was clarified for the horticultural world in the first edition of the International Code for Nomenclature of Cultivated Plants.

My understanding of cultivar is that it requires both a nomenclatural element and a taxonomic element. To explain more clearly, a species can exist and be recognised regardless of whether it is given a name. Newly recognised species often float around for a long while in herbariums with just the designation 'sp. nov.' (= species nova). But a cultivar must exist as a a name, though of course that name must be consistently applied to a taxononomic entity. This is because, for every named cultivar (in some plant groups) there may be hundreds or even thousands of variations that are discarded and never make it to the status of cultivar. Can you call all of the seedlings of a particular cross made by a rose breeder cultivars? No, it's only the one or two that are kept and named.

I don't think Serenoa is mistaken, just oversimplifying a bit. "Cultivar" is indeed a coined term, but it was coined from "cultivated variety", in part to distinguish it from the botanical rank of variety. Many people use the terms "variety" and "hybrid" when they really should say "cultivar". A cultivar can be a hybrid or it can be a mutation (sport) of a species or of another cultivar. "Variety" is a botanical rank and like "species" requires a Latin description and designation of a type specimen, deposited in a herbarium, to be validly published.

A cultivar is (1) cultivated, (2) somehow different and distinguishable from other plants of its species or genus, (3) able to be propagated, and (4) named, with the name placed in single quotation marks and not Latinized (some older Latinized cultivar names are grandfathered in).

A blue-flowered morning glory is not a cultivar. A blue-flowered morning glory that is named 'Heavenly Blue' IS a cultivar--and this is one that is an inbred seed strain that comes true from seed. Other cultivars are created as F1 hybrids between inbred lines--this is how many cultivars of annual flowers and vegetables offered in seed catalogs are created. The cultivar is produced by making the same cross between the same parents, over and over, with predictable results. And still other cultivars, usually perennial herbs, shrubs, and trees, are clonally propagated--i.e, by cuttings, divisions, tissue culture, grafting, or other vegetative means. Unfortunately, many such plants are then propagated by seed, with the seedlings also given the cultivar name. This has led to the deterioration of many fine cultivars. (The blame partly lies with the plant breeders, who often release new cultivars without any detailed description published anywhere, or any indication of how the plant should be propagated.)