A Conversation with Professor Andy Walker

A Conversation with Professor Andy Walker (in Long Form)1

          Andy, as you recall, the last time we spoke, I was very keen on the idea of growing grape vines from seedlings at our new property in San Juan Bautista.  I’d like to catch you up on my current thinking and ask a few questions, as this project is potentially fraught with a non-trivial amount of danger.2,3,4  It would seem that there are some clever things that one might do, and some not so clever ones as well.

          I thought I’d review some of my assumptions and hypotheses and also share with you what I am really hoping to achieve with the initiative.  The fundamental hypothesis/assumption is that a profoundly mixed (or mixed up) population of genetically distinctive individual plants (if the crosses are made thoughtfully) will yield a wine of far greater depth and complexity than a comparable one made from a relatively discreet number of clones or genotypes.5  Is this utterly far-fetched? I imagine that with this program, the particular varietal qualities of the grape will recede in prominence and potentially, the unique characteristics of the site, that is to say, its terroir, might begin to emerge.6  Now, this hypothesis is a bit tentative, for as we know, scrambling and re-expressing the genetic information of the source plant will generally result in a very different expression of characteristics – usually, though presumably not inevitably, less desirable than those found in the previous generation.  Put another way, might you gain more than you potentially lose in re-expressing all of this information, and might the gentle guidance of human intelligence vis-à-vis the inclusion or exclusion of particular individual seedlings in the mix tilt the balance of benefit to the right side of the equation?

          As to the issue of what I’m really trying to achieve here:  Firstly, the project needs to be fun, great fun,7 and the side-splitting amusement factor may of course have to do with whether we end up producing some eerily dramatic, soulful wines that taste unlike anything else in the world.  It would also be just wonderful if the human contribution to the experiment – the thoughtful establishing of criteria (even if they are a moving target) for inclusion or exclusion of vines with certain characteristics – was found to actually be helpful to the process, i.e. bringing greater brightness and definition to the wine.  Truly, the overall objective of the exercise is to produce a wine of great distinctiveness, an original wine, bringing something into the world that was not there before.  It is of secondary importance to me – it may in fact be utterly impractical – to identify the “best” new grape varieties, but rather, more important to create an experience for the consumer of a wine of breath-taking resonance and harmony.  It is my belief that the creation of an “original” wine benefits the world in many ways – enriches our experience as well as our imagination.  But there is still another consideration:  As amusing as this exercise might be for me, how might I make a real contribution to the world of wine? Might the creation of these new varieties actually yield a particular genotype of real utility for the future?  Are there any lessons learned in this exercise that might have application elsewhere?

      So here are some more really hard questions:

  1. My initial thought was to conceptualize some sort of idealized blend, based on assumptions about which varieties might do particularly well on the site, and then set about hybridizing these different grape varieties in some vaguely proportionate manner.  One obvious question: What do you need to know to decide which grape varieties might cross well with another?  My biggest fear is that I might begin with two noble grape varieties, and in hybridizing them, end up with something that is (in aggregate) absolutely wretched.8  So, for purposes of discussion, I’m beginning with the idea that grapes that have been planted in the same neighborhood, let’s say, the southern Rhône, might in fact have a reasonably good compatibility with one another.  Is this a fair assumption?  Is there any other way that I might look at criteria for hybridizing one variety with another?9

  3. So, say we’re going for something like a Rhône-ish blend, of perhaps 66% grenache, 15% mourvèdre, 10% syrah, 6% cinsault and various others from the hood.  Does it make sense to hybridize between varieties (grenache x mourvèdre) or hybridize within the variety (grenache x grenache)?  Are there any major incompatibilities among these grapes? Amongst the Rhône grapes are there some that make better pollinators, others better pollinated?  Are there more dominant characteristics expressed in the male or the female parent, or is this utterly random? Intuitively, for purposes of this project’s stated aims, it would seem to make more sense to hybridize between varieties, but maybe the world at large would be better served by hybridizing within the variety.  (Identification of a brilliant grenache or cinsault selection would seem to have some tangible benefit for vineyardists of the future.) Does that make any sense? Also, this is rather a biggie: How much do the seedlings of particular varieties resemble their parents?  Which varieties of grape vine offspring tend to remain truer to their patrimony?

  5. Which brings me to a very interesting project undertaken by an extremely bright young man, Sashi Moorman, down in Lompoc.  Sashi has collected a large number of pinot noir seeds (maybe 8000?), and has germinated them, and planted them out in a high density vineyard.  Obviously a number of them will not bear fruit10 and he will presumably have to discard them.  Sashi is imagining that somehow in this vast number of seedlings, he might be able to identify a particularly brilliant individual.  By brilliant, he is meaning a variety that may be slightly better adapted to his site – ripens at lower Brix, with better acidity, with more expressive pinot character, etc.  But might any of these offspring actually be truly pinot?  Perhaps these very gross parameters might be noted (or notable), but I am myself slightly dubious about the practical ability of making these determinations.  So, I think that Sashi imagines that he is doing one experiment – trying to identify a great pinot for his site (and maybe happily find a vine that might also have some other great endearing characteristic, like phylloxera resistance) but I think that in fact he may end up with a different experiment altogether.  I am anxiously waiting to find out what the wine made from the totality of his grapes might taste like, and hoping that even if few of them individually look or taste much like pinot noir, they might in aggregate somehow capture the Platonic nature of pinot-ness.  Any thoughts about that?

  7. So, I want to make a wine that captures a sense of this unique property in San Juan, and one element that I want to address is that of drought tolerance.  Since I’m hybridizing grapes here, might it make sense to consider adding other elements to the mix that might do that?  Is it totally crazy to consider adding some genetic material from vitis californica to this assemblage?11

  9. You had mentioned once that the best way to hybridize vines of a particular grape variety would be to cross a number of different clones of the same variety with one another rather than simply collect the seeds.  As we both know, going through the tedious process of castrating the male flowers of plants and going to the effort of pollinating them oneself is incredibly tedious. Can you tell me again why not simply collect the seeds from a grape, syrah, for example, and plant those out?  (Assuming that they being grown in a fairly sequestered area and you don’t have a lot of chenin blanc pollen floating around.)

  11. Which brings me to a rather geeky question:  If we are planting a mother block from which to collect pollen and also to pollinate, how far apart need the different grape varieties be from one another?  It would not be very rigorous (at all), but given the essential idea of the program, why not simply plant a small mixed block of grapes in randomized fashion (with a range of different clones of a variety, as discussed) in the proportions that one wishes and then simply collect the seeds from these grapes? – a lot less tedious than going through the whole hybridizing process. 

I am sure that I will have a million more questions for you as we get closer to really implementing this project.  You may well regret the encouragement that you have already expressed.  With very best wishes, Randall

  1. Andy is a Professor of Plant Science (Viticulture) at UC Davis, specializing in grape vine breeding, the logical person on this side of the planet with whom to have this discussion. []
  2. And of course, great possibilities of “success” (whatever that is), éclat, and the contribution of something of real value to the wine industry. []
  3. We are not even thinking about the utter riskiness of planting vines without the protection of a resistant rootstock.  A seedling is perforce a non-grafted vine, and hence vulnerable to phylloxera, though possibly far more resistant to other sorts of vine diseases. []
  4. My intuition tells me that something truly great might come of it, but likewise, it could easily turn into a very expensive obsession, a ruinous folly. []
  5. You had also mentioned to me that seedlings, if transplanted soon enough, exhibit a much higher degree of geotropism, as compared to rootings.  This factor alone may well be significant in creating a more eloquent expression of terroir, owing to a deeper rooting profile, and perhaps also conferring a greater degree of drought tolerance, which would be a very favorable outcome, indeed. []
  6. It would of course be extremely useful to have some other points of triangulation in helping to identify this terroir.  An adjacent vineyard planted more traditionally, i.e. from vegetative cuttings, to the self-same varieties would provide a good point of reference in illuminating the contribution of the difference in rooting habit, as well as the added dimension of extreme genetic diversity. []
  7. You had mentioned the great likelihood of plants grown from seedlings having the tendency to throw “suckers” (which need to be laboriously dug out with a shovel) essentially for the entire life of the vineyard.  This particularly tiresome phenomenon may well negate all of the countervailing fun features of the product, viz. the creation of a dizzying profusion of new grape varieties. []
  8. The obvious example of pinotage (pinot noir x cinsault) comes to mind. I do not consider myself a “varietalist,” that is to say, someone irrationally prejudiced against a particular grape variety, but in the instance of pinotage, I am quite hard pressed to find any real value in the grape. []
  9. This is in some sense a bit of a restatement of how does one begin to conceive of a blended wine.  In the New World it is particularly problematic if one is planting one’s vineyard from scratch.  You can opt for grapes with known affinities for one another – a Bordeaux blend or a Rhône blend (North or South), a Tuscan blend, but what if you really want to utterly break the mold and dare combine varieties from very disparate regions?  How do you insure that you are not creating utter chaos? The other significant part of the equation is that you must produce wines that you absolutely love to drink.  This problem is largely solved in the Old World, where young people grow up tasting the wines of their region and by the time they become winemakers, they already love those grapes.  Deeply. There were no vineyards in Beverly Hills when I was a lad, so I was not imprinted at an early age with a deep vitaceous cultural identity.  I just know that there are certain flavor components found in certain wines that just make me insane with – the flavor and aroma of citrus in whites, that of licorice and beetroot in reds.  I would be so utterly thrilled if the wines of my dreams would have these characteristics, and certainly I will contrive somehow to make that happen – but obviously, not by “trying.” []
  10. Just by the by, do you know of any relatively quick and dirty biochemical assay that would show whether a particular plant might be fruitful or not?  This would be incredibly useful in avoiding some the cost of planting non-productive plants. []
  11. If the aim of the project is to truly convey a sense of terroir, maybe adding an element of “indigenousness” is not totally crazy.  How one would have any sense at all as to how much to add to the mix is beyond me. []

14 Responses to “A Conversation with Professor Andy Walker”

  1. DC says:

    Was this a conversation or just stream of consciousness?

  2. Randall, thanks for opening your letter files and letting the rest of us into the conversation. Your project is intriguing, fraught, and quite possibly genius. The multiplicity of crossings makes this a factorial exercise of epic proportions, and the near-impossibility of running a full control group means that only time, and a lot of labor, will tell.

    Mixed vineyards of course aren’t new, and in some cases are very old. The older vineyards of the Douro contain a mixture of grape vines, and many Port winemakers have no idea what they’ve got growing. I was speaking recently with a 5th-generation Port maker who validated this for me, saying, essentially, it’s a mix, and we like it that way, because it makes more interesting wine. One can only guess whether, over time, hybrid seeds produced in these vineyards found their way into soils, then were ignored long enough to produce fruit that influenced the flavor profile of the resulting wine.

    But your scheme makes that process intentional, not accidental, and relies on—well, mating, to be blunt. Since plants themselves are not so good at locomotion, and since hand-pollination would, as you note, be tedious, you might also want to attend to the vector of the mating process: the bee (and its brethren). Are particular grape varieties more attractive to pollinators? Are particular pollinators more desirable in some senses, whether more active or more wide-ranging or more targeted in their ministrations? In other words, can you use the pollinators’ evolved intelligence(s) to coax some specialized work, work that would be impossible for you to do on your own?

    This project strikes me as an intimate, exquisitely complex conversation with the landscape, a dialogue among man, plant, soil, rock, and bee. This is not the true definition of terroir, but it’s my current working definition. Let a thousand flowers bloom.

    • DC says:

      I agree with you on the scale of this endeavor. It is mind boggling that a accidental cross may be found and propagated in a lifetime. Many have spent there life trying to create superior crosses through intentional crossing and re crossing for desired traits.

      The only problem I saw in your post was that bees may be utilized. Unfortunately grapes are wind pollinated, independent of insects.

    • DC, thanks for your note. My understanding is that while grapes are perhaps mostly wind- or self-pollinated, they’re also insect-pollinated. Randall cares about cross-pollination, and my skim of the literature seems to indicate that insect-fertilized grapes of various V. vinifera cultivars respond differently to self- versus insect pollination, and the resulting fruit (and seed) have different qualities. So the suggestion here is that it might make sense to think through role these agents could play in his project.

  3. One of the real questions I have is what is the real advantage of going through the immensely tedious work of actually hybridizing the plants – removing the male flowers, hand-pollinating the plants, tying them up with bags to avoid the random wind-blown or insect borne bit of pollen. Instead might it make sense to design a sort of mother plot of interplanted varieties, let the wind and the insects do with them as they will, and collect seeds. This would be orders of magnitude simpler – still a ton of work, and maybe ultimately more trusting (if that is really the right word) in the intelligence of nature to make the crosses that she needs. My understanding is that virtually all grape-vines self-pollinate preferentially, but maybe the design of the plant architecture (one cane sitting directly above another) might improve the likelihood of a greater degree of hybridization.

    • DC says:

      A few things to consider about letting “the wind and the insects do with them as they will”. As you mentioned Vitis self pollinates readily. Also The timing of flower of each variety should be considered. Using multiple varieties that might not bloom at the same time would also exclude certain crosses from happening.
      One thought that crossed my mind would be to select which vines would be the “Mother” vines and remove the stamens on those clusters. Then allow “natural” fertilization and collect the resultant seeds. Not even half as tedious as collecting pollen and manually fertilizing. Block design would play a huge roll in available donors.


  4. Peter F May says:

    You ask “what is the real advantage of going through the immensely tedious work of actually hybridizing the plants”. Because it is the only way you control which two varieties are the parent of the eventual new vine, and by selecting those varieties you can attempt to combine the best attributes of them both.

    This is what plant (and animal) breeders have been doing for centuries, and also what viticultural centres have been and are doing now. In Cornel University’s research station at Geneva, New York they have been doing this with success in combining native American vine hardiness for upper NY state conditions with the preferable wine features of vitis vinifera.

    Geneva are using technology to identify specific genes with desired attributes in parent vines. Even so of the many thousands of resulting seeds that are planted many will produce new varieties that are sterile, grow poorly or are otherwise useless.

    Your random breeding idea may well throw up another successful variety: as you know Pinot Noir has been the parent of many of today’s top varieties through (probably) random wild crossing but those resulting successful varieties had to be identified and selected for propagation by humans.

    Peter F May
    Author of
    PINOTAGE: Behind the Legends of South Africa’s Own Wine

  5. Peter F May says:

    PS: On another site you say “Pinotage strikes a somewhat deeply seated resonant fear. There you had two perfectly wonderful grape varieties – Pinot Noir and Cinsault and the resultant cross is something that is significantly less pleasant than either one of them.”

    Cinsaut can make a pleasant wine when properly made – but that can be said about most varieties.

    Deeply seated fear? I wonder how many Pinotages you have had to make your statement and whether you have had properly made aged Pinotage.

    Problem with new varieties – and you’ll find this if you are successful – is that it takes time to learn how to best handle a variety and make the best of it. You and other new world wine makers have been using the old world as a template of what those variety should taste like. But a brand new variety cannot point to others and has to forge its own direction.

    Pinot Noir is not an easy grape to make great wines and there are many poor examples but that is a spur to winemakers to try harder. Pinotage is also not an easy grape from which to make a good wine and there have been too many bad examples but does that mean a variety is worthless because not everyone can make good wine from it?

    With just 50 Pinotage vintages they’re still learning but I think they are making better Pinotage than ever. And what about the California Pinotages from Fort Ross, Vino Con Brio, J Vineyards, Sutter Ridge and others?

    As you are a winemaker who likes a challenge, I challenge you to make a great Pinotage – there are grapes growing and available in California you can buy.

    Peter F May
    Author of
    PINOTAGE: Behind the Legends of South Africa’s Own Wine

  6. Peter, Agreed that that was perhaps an utterly unfair shot to take at pinotage, but I have yet in candor to have experienced a revelatory one. I do love a challenge, and would agree with the observation that even the humblest and apparently least prepossessing grape variety can under the appropriate circumstances produce a wine that is absolutely extraordinary. (Palomino negro aka Mission,) in the Canary Islands to cite but one example.

    I’m not sure if I made clear the real aim of the experiment with seeds/hybridization. It is not to attempt to identify a “superior” new variety or even a particular genotype that is perhaps better adapted to a given site than one similar to it. It is rather to create a great deal of genetic diversity within a certain defined family or clan of grapes, with the belief/hypothesis that this extreme genetic diversity considered in aggregate (if one can figure out the most appropriate way to devise criteria of inclusion or exclusion) will perhaps yield an utterly unique wine and perhaps something like a vrai vin de terroir. In any event, I believe that grapes grown from seedlings – just because of their rooting habit – will do something different from conventionally propagated ones. But how to turn this into a wine that people will actually enjoy is another question.

  7. GayRoulette says:

    Wow neat! This is a really great site! I am wondering if anyone else has come across something
    exactly the same in the past? Keep up the great work!

  8. van rooinek says:

    Stumbled on this old thread, had to comment –

    (1) Munson, famous Texas grape breeder, estimated that only 1 in 1000 grape seedlings were as good or better than their parents. Apparently winegrapes are quite unique combinations of sugar and flavor. It’s like expecting the daughter of a world class supermodel to also be a supermodel… likely she’ll be cuter than average, but she probably won’t be as good as her mom. Don’t let me dissuade you from, er, crosspollinating with a world class supermodel if you are given the opportunity of course.

    (2) Many, perhaps all, of our classic winegrapes began as accidental seedlings that someone recognized as valuable. Cabernet Sauvignon appears to have begun ~300 years ago as an acccidental cross of Cabernet Franc and Sauvignon Blanc. But, perhaps legions of such crosses happened in the centuries that these 2 vines grew next to each other and just this one was found good enough to be preserved in its own right. Likewise, about 14 different French grapes have been shown to be medieval crosses between Pinot Noir (which is ~2000 years old), and an old variety called Gouais Blanc. These 2 varieties were grown in close proximity for centuries; they must have crosspollinated millions of times, producing uncountable thousands of accidental seedlings… and only 14 were worth propagating? Scary odds… but then again, one of those 14 was Chardonnay, so you might get super lucky.

    (3) It doesn’t look as if Dr Walker ever answered. But if he does, I suspect that he’ll tell you not to waste your time planting anything that isn’t resistant to Pierce’s Disease. Keep watching for the Foundation Plant Services to release his new PD resistant vines in the next few years, and start your homegrown crossing experiments with those. If you can’t wait (I can’t), look to old PD resistant hybrids such as Lenoir (very hard to get in Calif), or the ornamental “Roger’s Red” which is an edible but feral cross between a winegrape and the PD-tolerant native Vitis californica. .

    • So sorry to be almost three years in responding to your note, but have just seen your comments for the first time. It is certain that Munson is right about the progeny of self-crosses being virtually always less interesting (at least prone to a lot more genetic defects) than their parents – think Hapsburgs, collies, etc. I don’t wish to speak for Andy Walker, but he might gently suggest that when grapes are grown even in extreme proximity to one another, the odds of them pollinating each other are extremely rare, and that the vast majority of these “cross-pollinated” seedlings were undoubtedly self-crosses. He might also take issue (or not) with the suggestion that the creation of Cabernet sauvignon (or even Chardonnay) was a felicitous accidental cross. I am not quite sure what evidence he is willing to put forth, but I know that he believes that monks were deliberately making crosses with particular intention. (It does seem a little bit tricky given the lack of tools to perform the delicate surgery of the removal of the male parts.) But, my contention is that maybe there really is no such thing as “superior” grape variety, that many (with the possible exception of “Burger”) grape varieties have the potential to produce superb wine is there is a particularly congruent fit to their site. Further, I am postulating that perhaps a collection of genetically “inferior” – no, that’s too harsh, let’s call it genetically “unproven” grapes might in fact, as a suite, produce a wine of unrivaled complexity even compared to the same site planted to one or two “superior” varieties or clones. You certainly bring up an interesting idea about introducing the PD resistant vines into the mix. I’ll catch up with him on what sort of material he has managed to create. Obviously, tasting the resultant wine grown from grapes in Davis creates a certain imaginative challenge, but it’s an interesting place to start. Thanks for your comments.

Leave a Reply

* Required
* Required, Private




Where Art Meets Science

If you are an artisan vintner who wants to reach new customers, the solution is a simple as C, B, E.

At CBExpo, you’ll take a scientific approach to reaching new customers and converting them to ambassadors for your label. That approach begins with learning proven business strategies and post-production secrets from industry luminaries…

Including Bonny Doon Vineyard’s Randall Grahm, who will be a closing speaker at the Expo’s General Session on May 20th, 2016.

Date: May 18-20th, 2016

Location: The Oakland Convention Center

For more information about this event, click here.



CIA Sommelier Summit

Join us for the second annual CIA Sommelier Summit! Building upon the momentum of last year’s inaugural summit, this three-day conference once again brings together the most dynamic group of wine professionals in one place. Don’t miss this great chance to network with peers and industry leaders all while deeply expanding your knowledge of the fast-paced and ever-changing beverage industry

Bonny Doon Vineyard’s Randall Grahm will be presenting at the Tuesday, April 25th (10-11:15 a.m.) Breakout Session:

Natural Wines: WTF? (What’s the fuss?)

Moderator: Christie Dufault, Associate Professor – Wine & Beverage Studies, CIA at Greystone

Presenters: Randall Grahm, Winemaker & Owner, Bonny Doon Vineyard and Abe Schoener, Winemaker & Owner, The Scholium Project

Is it in the yeast or the ideology? Organic, biodynamic, vegan, no added/low sulfite, foottrodden or simply wine made with minimal intervention by a brave, well-meaning purist? What does it all mean and what do we think about it? Are natural wines good for the earth, good for you, and good tasting? Join Vintner Hall of Fame member Randall Grahm, who seeks to make “naturally soulful, original wine,” as we determine if this current movement is new hype or new hope.

Date: April 24-26

Time:  CIA at Greystone, 2555 Main Street, St. Helena, CA 94574

For the complete summit schedule and to register, please visit the CIA Sommelier website here.




Our crowdfunding campaign "Popelouchum Vineyard: 10,000 Grapes for a New Vine" is now over, but you can still buy perks, or donate to the cause.