About Species and "Unique" Varieties

First, I should confess that I approach the whole question of taxonomy or classification with the bias of an evolutionary biologist. Specifically, I'm inclined to think of a "species" as one or more populations of something which interbreed with each other and do not or cannot interbreed with others. I.e., there is a geneflow barrier between this group and other groups. From an evolutionary perspective this is very important, of course, because it defines a "unit" of evolution ... anyone inside the group can potentially exchange genes with anyone else in the group to produce new offspring and so that "pool" of genes is that on which selection operates.

Now, the rest of biology doesn't operate this way ... and, to be truthful, some of my former colleagues in biological anthropology don't either. To much of biology "species" is simply one level in a hierarchy of classification with genus and family above it and subspecies and populations below it. No special assumptions are made about geneflow and quite often two plants or animals which are from different species are actually interfertile and thus crosses or intermediate forms are possible. This drives me nuts, but I can't change it, so I'm stuck with what other people do when operating in a realm like this.

Plants, if anything, are "worse" than animals in this regard, particularly cultivated ones since humans are constantly messing around with the reproduction process. At the extreme with grapes we have propagation without sexual reproduction by budding resulting in clones which are genetically identical. Of course, sometimes that which is called a clone is really a "selection" since more than one original plant is included in the propagation and thus the offspring are not genetically identical. (See this Swan Winery Newsletter [6] for a nice discussion of the various terms.)

So, we are starting off with a large number of grape "varieties" classified into multiple species and even genera ... and they are all interfertile ... and people have been mucking about crossbreeding them, probably for thousands of years, most of which with no records. I.e., we have very little idea what we have and how it got that way.

Until fairly recently, classifying grapes was mostly a combination of "ampelography" and tradition. Ampelography, relying on the physical characteristics of the vines and grapes, is necessarily imprecise, especially since different climate, soil, etc. can cause the same genes to manifest themselves differently. Not to mention, of course, that there is no magic formula to tell one whether a particular difference is significant or not. Tradition is naturally less precise, especially since a great many crosses were created at times when we didn't even have a modern understanding of genetics.

Mixed up in all of this is a lot of local color ... take a grape from one country and move it to another and there is a good chance that someone will decide to change its name to something that fits the local language. Plant that grape in Alsace and have many years of the land changing hands between the French and the Germans and this can get very colorful. And, of course, the whole history of Europe and many other wine growing regions is full of moving political and language boundaries.

Now, of course, we have The Answer -- DNA testing. Well, sort of an answer, anyway. Given two or more sets of samples from two or more possible varietals, one can decide that they either seem to be the same or they seem to be different. Now, it isn't really quite that simple, of course, since we don't really have any standards for how different things can be and still be considered the same. E.g., what happens if one starts using DNA testing on all the Pinot Noir "clones" ... might they end up being so different in some cases that one wouldn't call them both Pinot Noir?

And, how similar is similar enough to be considered identical? I think of Zinfandel, Primativo, and Crljenak Kaštelanski where it appears that Zinfandel and Crljenak Kaštelanski, while historically related with Crljenak Kaštelanski as the parent, have had over 100 years with no on-going geneflow, but lots of selection, human and natural, among the Zinfandel grapes. With most organisms, one would expect that to have produced a certain amount of genetic diversity such that one would want to both recognize the parentage, but also the uniqueness of the offspring.

We just have precious few rules established at this point and that is hardly surprising since we know from other organisms that a single mutation can result in no geneflow, lack of interfertility, true evolutionary divergence ... so how can we have a clear scale for deciding what is "different enough"?

On top of which, only a tiny percentage of known grape varieties have been tested so, mostly we just don't have the data. And, most of the testing is taking a small number of possibilities and comparing them to each other, without comparing them to the universe of other things that have also been tested. Important and interesting work, but scratching the surface of the problem.

Consequently, there is a lot in this database and the source materials from which it is drawn that is not clear, hard science. I am noting elsewhere a number of examples where our only source is tradition and that tradition is not clear so we are left with ambiguity. For someone "simply" trying to tally up their list of 100 "unique" varieties, this can be frustrating, but it is also the current state of the science, so we can't simply apply some simple rule and make it all clear.

Open Issues

In working with Vitis and other sources I have periodically encountered ambiguous or questionable relationships between cultivar and prime names. As I hope I have explained elsewhere, this is very much to be expected since much of the data being collected is anecdotal or traditional, not rigorously scientific. There is a huge amount of work to do to sort through all the sources, determine what is most reasonable, and, when necessary, to subject to DNA testing.

However, since encountering ambiguity is one of those things that can easily make one confused, I have created this page to provide a reference to issues which I have identified so that people who encounter them can have a place to check and verify that it is a known ambiguity. If others encounter new ones, as I am sure they will, I will happily add them here to create a more comprehensive resource. As ambiguities are resolved, they will get reported here.

Unless otherwise noted, the issues below relate to data in the Vitis database. The typical pattern to which they relate is that wine A as a cultivar name will point to both A (i.e., blank) and B as prime names and wine B as a cultivar name will point to both B (blank) and A. Thus, we have a circular reference. I should note that it is fairly common to have half of this pattern, i.e., that a given cultivar name can be either a prime name in its own right or another prime name. In some cases there are differences in berry color or country of origin to explain this pattern and sometimes there isn't.

I am not including here any patterns in which a cultivar name is not easily resolved to a prime name because there are just too many of those. Perhaps later I will consider creating some discussions on another page about how one might make such distinctions in specific cases, but for the purposes of the current list, I am only including cases where there is a circularity of some kind, i.e., the appearance of an error. I have discussed some of these with the people at Vitis and they have explained that the "error" is intentional since it best represents the current state of knowledge. I.e., it isn't known which is prime or exactly what classification is appropriate.

In the following, "points to" means that a cultivar name "points to" a prime name.

• Rabigato points to Rabo de Ovelha and vice versa. Neither points to itself as is typical.
• President has one entry for which every field other than the cultivar name is blank and another entry in which the prime name is blank, but species, berry color, and country of origin are provided. Vitis has clarified that the first one is actually President Carnot, a French V. vinefera from Oberlin, Colmar.
• Arinto de Bucelas points to Arintho and vice-versa.
• Gouveio points to Gouveio Real and vice-versa.
• Syrah is shown as pointing to both itself and to Durif, but it is Petite Syrah which is considered the same as Durif and this is known to not be the same grape as Syrah. This would appear to be an error, but it might simply reflect old data before this relationship was clarified.
• Alvarelhão points to itself and Brancellao points to itself and to Alvarelhão.
• Sangiovese points to Brunello di Montalcino among others and Brunello di Montalcino points back to Sangiovese.
• Cynthiana points to Norton and vice-versa.
• Pignul points to Pignolo and vice-versa.
• Boutignon points to Pardotte and vice-versa.
• Tihany Kek points to Szekszardi which points to itself and to Szerdi. Tihanyi Kek points to Szerdi. Possibly not ambiguous, but needs checking.
• Verdello points to Breval and vice-versa.
• Almeria points to Ohanes and vice-versa.
• Lungyan points to Ryugan and vice-versa.

Note, there are also cases where cultivar name A points to prime name B and cultivar name B points both to itself and to other prime names. The assumption is that A is a synonym for the version of B which is a prime name. Thus, this is not ambiguous.

Also, there are quite a few cases where the same cultivar name and prime name (often the same) is listed more than once with a different berry color, sometimes with a different country of origin and sometimes with the same country of origin. It is presumed that these are genuinely different grapes which are simply not provided with different names because of local tradition.

Wine Century Club Application

This spreadsheet is intended to help you track your total varieties for your Wine Century Club application.

To use this sheet, fill in lines below the title line. We have included a couple common varities to get you started, so delete any that you haven't tasted. You may copy these from the Muse Grape Varieties database [7] or enter them by hand from data obtained from the Vitis database [8] or any other source. If you cut and paste from the Muse database, you should use Paste Special and check only "Values" to avoid pasting in the validation constraints that are used in the Muse database. You may add new wines anywhere you would like below the titles, but don't leave any blank lines. Use the Sort By Cultivar Name and Sort By Prime Name buttons to put things back in order for you after you have created new entries. The Species, Berry Skin, and Country of Origin columns are not required, but can be useful in distinguishing some grapes. If you encounter grapes with the same prime name, but which are not the same grape, e.g., different berry color, modify one or both Prime Names so that they are unique. The notes column is for wine name, appellation, vintage, etc. and is optional. It can also be used to note when you are uncertain which of several prime names matches the grape you have tasted. Grape varieties that you've tried only in blends with other varieties are permitted. For Wine Century Club purposes, it is the count of Prime Names which count, i.e., the actual unique grapes, but we are including the Cultivar names because that is usually the name on the bottle and so search by Cultivar name can make it easier to find out whether or not you have tasted a grape before. E.g., while Shiraz and Syrah are both Syrah, having Shiraz in the list will tell you that you have tasted it already and figured out to which Prime name it corresponds. Use the Refresh Counts button to check your current status. You should be in Cultivar Name sort to do this, but that would be the normal way to keep the sheet.

If you have at least 100 varieties checked, mail or fax the application to the address at the top of the sheet and allow 3-4 weeks to receive your certificate and further information regarding your membership. Please note that the application is entirely on the honor system; should you lie, may the wrath of Bacchus curse your palate!

If there are questions or problems using the sheet, please contact me at thomas [at] A-Muse-In-The-Cellar.com

Working with Vitis and other sources

The Vitis database has a great deal of appeal for someone interested in tracking grape varieties and their synonyms both because it is massively more comprehensive than any other source I have found and because it is searchable, i.e., one can put in a name or a partial name and it will come up with some options, which can often make it easily to find the "real" information on what name means what grape. Naturally, there are some complications which make it only part of the solution that one would really want, i.e., instant, unambiguous, and definitive answers to everything!

The first problem is the data itself. While it is far, far more comprehensive, that doesn't mean that it has everything. It is maintained by a staff that is trying to do a lot more than they have resources for ... which seems typical these days ... so there is a lot of things they would like to do which just haven't gotten done yet. They are trying and I marvel at what they have accomplished, but one can find varieties not listed there that do seem to be valid grape names used somewhere and, of course, there are the problems of the data itself.

The big issues here is that we have only a soupçon of science, as in systematic DNA studies or the equivalent, and a whole lot of tradition, local nomenclature, anecdotal evidence, and the like. I.e., what the Vitis folks have done is an incredible job of assembling a huge amount of data about what is known ... but there is a lot we don't really know or where the data is sketchy or incomplete.

Among other things, this can lead to some search results which are confusing until one understands what one is seeing.

Vitis provides "Cultivar" name, "Prime" Name, Species, Berry Color, and Country of Origin. Cultivar name is the name by which the grape is known in general usage. Prime name is what grape that corresponds to. I.e., Shiraz is what the Australians (and now some other people) call Syrah. So, Shiraz is a Cultivar name for which the Prime Name is Syrah. Syrah is also a Cultivar name for which the Prime Name is Syrah. From the Wine Century Club perspective, "unique grape" corresponds to Prime Name, so for Shiraz and Syrah, there is only one unique grape, Syrah, the Prime name for both.

In a Vitis search, when the Prime name shows up blank, it means that the Cultivar name is the Prime name. I.e., searching on Syrah will have a blank in Prime name.

But, of course, things aren't quite that simple, especially since so much of the data is anecdotal. Consequently, one will run into a number of cases where one doesn't receive a simple definitive answer, including many where a definitive answer is not really possible.

For starters, all these names are simply what people call something and people are notoriously non-scientific about how they come up with such categories. So, there in addition to people having a number of different names for what is really the same thing, they also regularly give the same name to things that are different. One of my classic examples of this outside of wine is "farro", a term that can mean several different species of wheat depending on where it is used, at least two of which are in common use in Italy. The same is true of "spelt". To buy the kind of wheat one wants, one really needs the Linnaen species name to be sure what one is getting.

For this problem, i.e, the bottle says X and Vitis says A, B, or C, there really isn't any way to resolve the ambiguity except by going back to the source. The distributor might know, the vintner might know, and the grower (if different) might know ... or none of them might know. One can reach a surprising number of these people via the web these days ... although certainly not all ... and may of them are very helpful. E.g., I recently had a Portugese maker explain to me that Maria Gomes, what he called a grape in one of his wines, a name which might be either Rabigato or Fernão Pires, was in fact Fernão Pires because the call the same grape Maria Gomes in the north of Portugal and Fernão Pires in the south. Now, this is still anecdote, not science, but it is about as good as it gets. And, sometimes one can't find a contact or they don't know, so one can only guess.

Another problem which one can encounter, and which can be quite confusing until one figures out what is going on, is that sometimes Vitis knows the data is ambiguous, i.e., they know that they don't have any scientific basis on which to decide what is Prime. E.g., in one place they may have a local name A which tradition tells them is the same as some other name B, but elsewhere B may be the local name and tradition says it is the same as A. Without a scientific study, Vitis has no way to resolve this, so they take the simple expedient of showing both versions. I.e., if you search on A, it will say that it is A or B, but if you search on B, it will also say that it is B or A. I am creating a record of these when I find them which I invite everyone to add to. There is some hope that bringing these to Vitis' attention will cause more exploration, but one has to recognize that really solving the problem means DNA studies and someone has to fund those.

Another factor which may confuse some people is that, when one has two or more grapes which are known to be identical, the name used for the Prime is somewhat arbitrary ... they are identical, after all. From a WCC point of view, this doesn't really matter since any name counted as prime is what one wants to tally and it doesn't matter what it is. But, there are times when it can be disconcerting based on ones expectations.

My "favorite" example of this is Zinfandel, Primitivo, and Crljenak Kaštelanski. There was a point where people thought Primativo in Italy was the source of American Zinfandel, but the history didn't work very well and more recently it was decided that the Croatian grape Crljenak Kaštelanski was the actual ancestor of both. Now, I will confess to a certain skepticism here since essentially 100 years of no gene flow between Croatia and the USA makes me wonder if Zinfandel doesn't deserve to be recognized as its own variety, but aside from this personal passion it seems that the "Prime" should either be Zinfandel, as the far most widely grown, or Crljenak Kaštelanski, as the apparent ancestor of the other two. But, in Vitis, it is Primitivo which appears as the Prime name. As I say, it doesn't really matter among things which are identical, so one just has to accept it and move on.

One of the practical difficulties of doing searches with Vitis is that many grape names, e.g., Crljenak Kaštelanski, include characters which are outside the "Latin" base set which is the foundation of computer databases. This is a wildly complex area. E.g., there are character pairs which appear identical in Finnish and Swedish, but which sort in different places depending on the language. Not to mention, of course, that not only do many sources not preserve all the diacritical marks which are important parts of many languages ... not to mention that those of use who use English, often have no idea how to produce these marks from our keyboard.

Vitis' response to this problem is to render everything upper case and without diacritical marks ... a good solution except for the "gotcha". The "gotcha" is that there are conventions in various languages such that a letter with a diacritical mark is turned into a diphthong for computer purposes. E.g., ä might be rendered as ae. Now, this is fine for those used to these substitutions ... driven by the limitations of earlier (and still current) computer systems ... but it can be confusing to those who are not familiar with the substitutions because a search on "cäd" which one enters as "cad" will return no results, but a search onf "caed" will.

One of the really useful aspects of Vitis that helps with this is the "wildcard" character "%". Every search has an implied wildcard at the end so entering "Cabernet" will show both Cabernet Franc and Cabernet Sauvignon (and 46 other variations). But, one can also insert a wildcard elsewhere in the string. E.g., if one puts in "%Sauvignon" one gets back 51 grapes which have Sauvignon somewhere in the name. This is particularly useful with a two part name in which the second part is fairly distinctive and the first part has some problematic international character. One can also put this wildcard in the middle of a name, i.e., the lack of results in the paragraph above can be cured by searching on "c%d". One may well get more than one wishes, but at least one finds it and then knows how it is spelled in the Vitis database.

And, of course, I should note that spelling is another one of those interesting variations. As you do your searches, especially if you enter partial names, you are likely to find multiple spellings of what seem to be the same name. Sometimes they map to the same grape and sometimes they don't! Spain and Italy are particularly "good" about having the same name with either a masculine or feminine ending. Once in a while this will make a smidgen of sense, e.g., the masculine ending is on a red wine grape and the feminine ending on a white wine grape, but frankly, smidgen is the best one manages and more often then not, it seems quite arbitrary.

While I have focused on Vitis here a lot, for what I think are fairly apparent reasons, there are lots of other sources. The Hawkins page mentioned above being one of the obvious ones, but ultimately one can find out a lot just by googling on the grape name. The usual googling rules apply, i.e., a grape name which is also used for a lot of other things can produce a large number of irrelevant hits, so adding the word grape to the search can sometimes help. And, of course, one has to have all of the usual cautions about any internet searches ... how many of what one finds are actually unbiased, authoritative sources. In addition to there being many public forums in which people are simply using names they have heard (sometimes from the winemaker, but then most of them are not scientists), there is also all of the usual issues with anything in the plant domain. E.g., Vendor X markets a grape under a trademark name and who knows what it is really (including the vendor). As long as the vendor can get away with claiming something distinctive, e.g., cold hardiness, people will buy the product based on the name.

I should also note that there have been a number of people in the history of grape growing who have produced a large number of hybrids. This has been very useful since there are many examples of now being able to grow grapes in climates where they could not be grown previously, not to mention many examples in which the hybrid was superior to the parents. The Vitis database has quite comprehensive listings of these grapes, typically with number and/or letter designations, but, of course, many of these have never been marketed and thus may never be in any wine that one could ever buy.

Still, they exist and you may run into one. Something like 40 years ago, my father decided he would like to grow some grapes along side the house ... in mid Illinois. Being the scientist he was, he did his research and found a source of one of the Seyve-Villard hybrids which was cold hardy enough to be expected to survive in Illinois and he planted a few. There weren't a lot of vintages before something went wrong and the vines died, but they were some pretty interesting wines. At this remove, I have no way of knowing which Seyve-Villard hybrid it was, but I do know it was one of them. By the way, searching Vitis on "Seyve-Villard" produces one hit, but searching on "Seyve%Villard" produces 155, so there are several tricks to be learned.