Wine collectors of a generation ago were perhaps not so aware of the necessity to store wines at constant temperatures and that a cellar should not, under any circumstances, have a temperature over 60 degrees Fahrenheit. In my capacity first with Christie’s, with Butterfield & Butterfield and finally with Zachys Auctions, I have had many opportunities to visit private cellars all over America and Europe. The variation in cellar conditions is remarkable. I have seen many perfect cellars, but none as fine as that of Dr. Benjamin Ichinose of Hillsborough, California, which will be discussed in detail. I have also seen many cellars, especially those in grand houses of older collectors that are simply appalling — wine stored in garages in Los Angeles (it might as well be in a Bedouin tent in the Sahara!) and wine in basements next to heaters. It is a sad tale of neglect and lack of knowledge of the importance of good storage for the survival and proper aging of wine.
The obvious question is, therefore, “If we know that poor cellars exist, how do you prevent wine from a poor cellar being sold at auction?” This is a very important query. The truth of the matter is that it is impossible for any auctioneer to send a representative to personally inspect every cellar. Some wines are taken in where the auction house may have only talked to the client by phone or corresponded by mail. In these cases, we certainly ask the client about his storage conditions and try to judge the cellar accordingly. Wines stored in a warm but not hot cellar may be perfectly good drinking but would not bring the premium price that serious wine collectors would offer for wine coming from cellars with unassailably fine storage conditions.
When a collection does come into any auction house, the staff is trained to spot the signs of abuse that always turn up in bottles that have been improperly stored. A cardinal sign is leakage from the cork, which is indicated by tell-tale stains of wine from under the capsule, down the bottle; sometimes extending to the label. A rather simple sign is labels that are faded and sometimes “greasy,” an indication that the bottles may have been in the sun or stored in the kitchen. Obviously, one must look at the color of the wine in the bottle. Though simple to accomplish with a clear bottle, with practice, color variations can be determined in red wines bottled in darker-colored glass by holding the neck over a bright light or good candle. One of the easiest wines to evaluate is Château d’Yquem. I know from long experience the usual color for Yquem of various vintages; for instance, for 1959 Yquem it should be a medium honey color. If this wine has been abused by heat or excess exposure to sunlight, it will change by oxidation to caramel and later even as dark as molasses. An overly dark bottle of 1959 Yquem will yield much less taste than will a properly stored bottle of appropriate color. The darker bottle will not have the proper sweetness and clear, fresh fruit, and honeyed flavors of the better stored example.
Another good indicator of abuse is the level of wine in the bottle. (The distance below the cork at which the level of the wine stands is the amount of “ullage” in the bottle.) It is well known that when wine is stored at a fairly high temperature, liquid escapes out of the bottle through and around the cork. Any level at or above what we call “bottom-neck” (in a Bordeaux-style bottle) is without much doubt an acceptable level. This is, however, no guarantee that this bottle is “perfect” or even drinkable! Even in a scaled case of wine stored in a perfect cellar, there will be natural variation in the levels of the bottles in that case after ten or twenty years of aging. One cause is that years ago wines were bottled by hand and thus there was human variation in the level of bottles at the time of corking. There is less problem with this today as most wineries, even in Europe, have modern automatic bottling machines. Another cause for variation in level is that corks differ greatly. Usually one or more corks in a case will be slightly less good and will allow more liquid to escape from the bottle during aging. In rare instances a cork can get a fold or crease or fissure down the length of the cork as it is compressed before insertion into the bottle. This almost imperceptible little channel down the length of the cork allows wine to run out. More importantly, it lets air into the bottle which in turn rapidly accelerates aging by exposing the wine to oxidation and thus causing deterioration of the wine. (Note that as the channel is very, very tiny it does not result in the entire contents of a bottle running out, but rather of a small amount running out before the channel is closed by solid matter in the wine plugging the channel.) If you carefully examine all your corks, you will ultimately come across one with a dark stained line all the way down one side and have proof of the way the wine went bad.
If we unpacked the perfectly stored case of twenty year-old wine mentioned above, I would expect to find the following distribution of bottle levels:
Neck- to bottom-neck levels: 8 to 10 bottles
Top-shoulder levels: 2 to 4 bottles
High-shoulder levels: 1 bottle or less
Note that for wine sold in Burgundy- shaped bottles there is obviously no “shoulder” to its sloping side, therefore levels must be measured by noting the distance in inches below the cork that the level of the wine fills the bottle. This is somewhat more cumbersome and can be misleading. For one thing, Burgundy wines, in the past, were rarely filled as full as were Bordeaux bottles. While a level of two inches below the cork might be perfectly normal for a Burgundy wine, a similar ullage in a Bordeaux-shaped bottle might be suspect.
It should be clear that, while the level of wine in a bottle is a useful measure of potential wine quality, one should not be fanatical about this point. One must realize that a wider variation in ullage is acceptable than some might recognize. For instance, a “top-shoulder” level is, in my opinion, perfectly acceptable for Bordeaux or California red wines. The results of years of research into the subject of ullage are summarized in the bottle level chart (see above).
The trained eye evaluates the clues that allow an estimate to be made of the quality of bottles that are received, where less experienced staff members may not be at such an advantage. In case of doubt, auction houses insist on tasting samples to verify the quality. In many cases, the evidence of damage detected by inspection is thus confirmed. The damaged wine is either returned to the client or, at the client’s direction, discarded. If wines are found to have been stored in a natural cellar that has simply advanced their aging without seriously harming them, the catalogue will state the storage conditions and the estimates will be lower than usual so that buyers may take this into account.
One of the saddest tales I have to tell about wine storage concerns a cellar in the Los Angeles area. In about 1985, I fielded a call at the Butterfield & Butterfield wine department from a gentleman who had quite a large cellar. When I asked him how it had been stored, he stated that it had been in an air-conditioned environment. I subsequently went to his home to supervise the packing of the wine and it turned out that the wine had been stored since he purchased it in closets all around his home. It was not temperature-controlled storage at all, but wine stored in a home that was occasionally air-conditioned in the summer, when the owner was at home! I did not like the looks of the bottles I inspected (having seen some of the bad signs outlined above), and I told our client that all bets were off. However, the client stated that one way or the other he had to dispose of the wine under doctor’s orders, and so we should take it away and determine its fate later.
Thus the collection ended up in San Francisco. We immediately opened some sample bottles and these confirmed our worst fears. Wines such as 1966 Latour and Lafite, and 1975 Latour and Margaux, were flat, oxidized and lacked fruit, and, worst of all, often had a bitter aftertaste that is characteristic of the decay products of severe heat-abuse oxidation in wine. Even after the wines had settled from their trip up from Los Angeles for several weeks, the results were the same. We reported this to the client and I offered to have the wine shipped back to him, as it was clearly unsalable. The client with no further ado told us to dispose of the entire collection! Sadly, it was all thrown away. In this case, we have a really good example of what heat abuse can do, as, since we were going to throw all the wine away, we opened many sample bottles in hopes of finding at least some good bottles. While there was some variation and some bottles were “almost OK,” the bottom line was that the collection had been ruined, and, that amounted to a seventy-five to one hundred thousand dollar loss! When I asked the client why he hadn’t built a refrigerated cellar, he stated that he simply didn’t know that it was necessary nor did he know that he was damaging his wines by storing them in his home. I think this lack of knowledge was quite common in collectors who got their start in the fifties and earlier.
First, let me define the characteristics of a perfect cellar.
1. It is a constant temperature between 49 and 52 degrees.
2. It is absolutely dark, except when visiting the cellar.
3. It is absolutely still. That is, there is no vibration from trucks in the street above, the subway below or from a motor running the refrigeration.
4. All the bottles are on their sides with the corks wet.
5. It has a constant relatively high humidity, hopefully above 70 percent. Humidity is the factor that I am the least certain about. The role of humidity in the storage of wine is essentially limited to its effect on the corks. Dry cellars are death for wine; conversely many fine natural cellars, such as in Germany, actually drip water from the ceiling and are constantly wet. Normal home air conditioning extracts moisture from the air and the result is that moisture is also extracted from the end of the corks. This renders the outer part of the cork useless as it shrinks, losing its elasticity and reducing its ability to seal the bottle. The longer the cork the longer a fissure must be to admit air and spoil the wine. In a dry cellar, moisture is also sucked directly out through the cork, without a fissure being formed in the cork. When the cork is removed from a bottle that has been stored in a dry cellar, it will be hard, inelastic and often shriveled at the outer end. Conversely, corks from bottles stored in a wet cellar, even if they are a hundred years old, will be resilient to the touch and seemingly very youthful. (It is interesting to watch how quickly a really old cork will shrivel and dry after it has been removed from the bottle!) There is a problem, though, with storing bottles in a very wet cellar: The labels, literally, rot on the bottles! This makes such extremely humid cellar conditions less than ideal for a home cellar.
Given these characteristics, there are two ways to get a perfect cellar, you can buy one ready made or you can build one. Examples of ready made perfect cellars are the cellar at Glamis Castle in Scotland and cellars at some of the estates in Germany. These cellars would be entirely underground, with rooms in the foundations of a medieval castle (in the first case) or tunnels cut out of solid rock (as are many of the cellars in Germany). Because in both cases the cellars are located at a relatively high latitude, the ground temperature would be about 49 degrees, both summer and winter. In such a cellar, all the other four characteristics are automatically fulfilled. Presto! A perfect cellar! But, realistically, are cellars like this easy to duplicate in the United States? The answer is no and that leads us to our explanation of how to build a perfect cellar.
Dr. Benjamin Ichinose is one of the greatest wine collectors in America. He is also a man who is constantly thinking and constantly experimenting. I have been privileged to have known Dr. Ichinose and to have worked with him for over forty years. He started collecting wine in 1959 and applied his characteristic energy to the study of wine both in books and through tasting widely. He also met many of the professors from the Enology Department at the University of California at Davis. From the late Professor Jim Guyman, he learned of Dr. Guyman’s decades- long experiments with the aging of wine. Dr. Guyman was attempting to see whether wine could be “speeded along” to maturity if it was aged at a higher temperature. He told Dr. lchinose that he had found that heat did indeed speed the aging of wine, but that it damaged the wine by “cooking” it. This caused the delicate chemical components that comprise the fruity flavors and aromas in the wine to decompose to a greater or lesser degree depending on how hot and for how long the wine had been stored. In fact, wine stored at the coldest temperatures used in Dr. Guyman’s study, approximately 50 degrees Fahrenheit, were by far the best, although they were often not ready to drink when sampled. Ultimately, Dr. Guyman was able to leave wines for long periods at the different temperatures and thus finally discovered that the wines stored at the colder temperatures were found to be more lively and “better” by blind tasting panels. Not only that, but after much more careful work with his data, Dr. Guyman was convinced that the wines stored at the colder temperature not only aged slower, but ultimately reached a higher peak of quality than did wines aged quicker at higher temperatures. Dr. Ichinose was intrigued by this information and determined to build a “perfect” cellar as soon as he could. This opportunity presented itself when, in 1966, he bought a new house for his growing family in Hillsborough. Before moving into the house, many improvements were made, including the construction of a refrigerated underground wine cellar. The spacious cellar consists of three rooms, each one ideal for aging a certain type of wine. The first room one comes to when visiting the cellar is actually an anteroom completely below ground (there are five feet of earth on top of the cellar). This room is actually a hallway lined with wine racks and is at ground temperature, which is 58 degrees at this latitude. This area is used to store young fortified wines (that is, Port, Sherry, etc., that is younger than 1970). Dr. Ichinose feels that if these wines were stored any colder they would take longer than a lifetime to age to the point where they were at their peak. Periodically, wines are moved from this area into the cold room. Most recently, all the fortified wines from 1950 to 1970 were moved to the cold room to make room for younger wines.
Through a door is the second room, the first that is refrigerated. I make a point now to stress refrigeration, not air conditioning. (See the section, Refrigeration or Air Conditioning.) This room is the largest in the cellar and is filled with rows of racks like the stacks in a library radiating off a central open area. This room is kept at 54 degrees and is used to store younger red wines, which at this time means red wines younger than 1990, both California and French. Again, Dr. Ichinose feels that if these wines were stored any colder, many would simply not mature in his lifetime.
The final room is separated from the 54 degree room by a wall of thermopane glass with sliding doors in the center. The open central area of the cellar continues into the cold room, which is kept at 49 degrees. Again, rows of wine racks radiate off the center of the room. In this room are stored all the white wines and sparkling wines, and also all the older red and fortified wines. It is this room especially that makes Dr. Ichinose’s cellar so special. Time and time again I have tasted wines that in any other cellar would be over the hill, but when stored at 49 degrees are still remarkably youthful and absolutely delicious. My most recent experience was with some German wines, wines that are notoriously delicate as they are made from Riesling and other similarly delicate grapes. The resulting wines are low in alcohol (which is a preservative) and are easily damaged by poor storage. I recently tasted at the Ichinose’s a 1959 Ayler Herrenberg Natur (one of the lower grades of German wine, not in any way a special late-picked wine). It was a very light yellow color and tasted as if it were no more than ten years old, when, in fact, it was nearly 50 years old! This wine had been stored in Dr Ichinose’s cellar since it was purchased in the early ’60s and had been moved from Dr. Ichinose’s old house (which was also refrigerated) to this new cellar in 1966. What a difference this sort of cold storage makes in wine! There are many, many other examples that illustrate this point.
Another important factor in the storage of wine is humidity of the location where a cellar is being built. It is clear that a good cellar has a relatively high humidity, however it must be vapor sealed so the humidity inside does not fluctuate. In a dry climate, moisture can migrate out of the cellar to the drier outside air, thus lowering the interior humidity. Conversely, there is danger in a humid climate, such as that in Florida or Hawaii, of moisture being sucked into the cellar. Once inside, the moisture condenses when it hits the colder air in the cellar. This moisture can run out of lighting fixtures and it will also condense on all the bottles, eventually destroying the labels either by dissolving the paper or by mold growth. Also, the cooling unit will constantly labor in a cellar that sucks in moisture; it takes a great deal of cooling capacity to condense all this liquid. It is critical that a cellar be vapor sealed with plastic sheeting before the insulation is put in the walls and ceiling of the interior of the cellar. If you contemplate building a cellar, it would be good to consult with a refrigeration professional before the architect draws any plans.
When wine is stored in a very cold cellar, that is, one with a temperature between 49 and 52 degres Fahrenheit, it is my contention that the wines will ultimately reach a higher degree of perfection. The best way to explain this is to illustrate the theory with a graph (see illustration). There are three different aging possibilities presented in the graph:
A. Standard storage at or near the ambient temperature in a home or restaurant, from 70 to 75 degrees.
B. Ground temperature storage in a non-refrigerated cellar, from 60 to 65 degrees.
C. Premium storage in a refrigerated cellar, from 49 to 52 degrees.
First, to describe any one of the curves in graphs, you must understand the aging cycle of wine. All wine is at least drinkable when it is young, even if it is rough and unpleasant (not to mention that it is probably not at its peak!) Thus, the line showing how the quality improves with aging starts not at zero, but some way up the chart. This starting point will, of course, be different for different wines.
Some white wines (especially California white wines) might actually start off when first released by the winery at 80 to 90 percent of their potential quality, and the final 10 or 20 percent of quality might be reached after only 2 or 3 years further aging. The illustration given here is obviously for a big red wine such as a Bordeaux or California Cabernet Sauvignon. Thus the starting quality at release might be 30 percent of the ultimate potential quality. Then the line of the graph tends upwards over time as aging smooths out the wine and as it develops complexity. When it reaches its peak, it doesn’t immediately start to head downhill. In my view, wines often stay at the peak for long periods (say ten years) with no seeming degradation. Then the wine trails off slowly and really takes quite a long while to reach the point where it is undrinkable. Often a high-quality Bordeaux wine from a fine vintage will sit on the plateau of peak quality for about as long as it took to reach the peak before its final decline begins. Thus the plateau can often be far longer than ten years. This period on the plateau can in some cases be as long as 25 years.
Now the real point of this graph. The three lines each illustrate the development of the same wine under the three storage possibilities mentioned above. For the “standard storage” at 70 to 75 degrees, the curve tends upward sharply as the wine is “forced” by the warm storage to age quickly. But due to the detrimental effects of warm storage, the wine never reaches its potential. In fact, although it is at its peak after about five years, it only reaches about 60 percent of its theoretical quality.
For the wine stored in the “non-refrigerated storage” at 60 to 65 degrees, the curve goes up more slowly and reaches the normal level of potential quality after about 13 years. This is the quality that one would expect if a bottle were tasted from the cellar of a Bordeaux château (note that cellars in Bordeaux châteaux are often at ground temperature) or a normal personal cellar. The wine will fade only imperceptibly for the next ten or so years and then will slowly go downhill. This period of slow trailing off in quality can go on for an almost indefinite period of time.
For wine stored in “premium storage” at from 49 to 52 degrees, the curve goes up even more slowly, and then as Dr. Guyman predicted, it goes past the theoretical peak point fo quality and eventually reaches a new peak that I believe can be up to 30 percent higher than the theoretical peak expected from normal storage! Thus, in this example, the new peak would be reached after about 25 years and the wine would stay at this peak for up to 20 years!
But you might ask for proof of this theory in tasting wines that have been aged in this way. We tasted several wines from the famous 1947 vintage some years ago with Dr. Ichinose. While these bottles were only in his cellar since 1965 (or for about 60 percent of their life), the effect of perfect cold storage was remarkable. Often connoisseurs complain that the wines of the 1947 vintage are cracking up and are past their prime. I disagree with this and feel that the good experience we had is typical of the beneficial effects of perfect storage. Among the wines tasted were the 1947 Latour, Mouton, Pétrus and Yquem. They all showed expremely well and were not cracking up at all.
Now that I have laid out the theory, I will throw in another wrinkle that was proposed by Dr. Ichinose. He states that since wine quality is affected strongly by aging conditions, wines from fine cellars can seem to be better than their price (or official rating) would indicate. Thus, a Third Growth Bordeaux stored in a “premium storage” cellar could have the same ultimate quality as a premier cru Bordeaux stored in a “standard storage” cellar! Thus, if you are storing wine in a perfect cellar, you could, potentially, save some money by getting the best out of lesser wines. You could almost pay for the cellar in this way!
It must be pointed out that “premium storage” may not be practical for restaurants where they cannot afford to keep wines for extended periods, or for the average collector who only has a modest quantity of wines of average quality. In both these cases, storage at ground temperature will yield perfectly acceptable results for short-term storage. It is for the real connoisseur who can afford the time and expense of perfect storage to reap the benefits of my theory!
The effects of the three storage temperature ranges can be summarized in the chart. For each temperature range, several factors are given. The first is the theoretical peak of quality. We have seen that this peak is expected for storage at 60 to 65 degrees and thus this is 1.0 P (or 100 percent of P, the theoretical peak of quality). The warmer cellar yields only .5 to .7 P (or 50 to 70 percent of the peak) and conversely, the perfect cellar yields 1.3 to 1.8 P.
In the column for maturing time, the warmest cellar is shown as 1.0 T (or the norm for maturing time). The colder cellars are correspondingly slower, at 50 percent to 500 percent slower!
For the abstract concept of ultimate quality, which is different from the theoretical peak, the differences are, in my opinion, even more pronounced. With the warmer storage nominally given 1.0 Q (or 100 percent of normal quality), the range for “standard storage” is from 1.5 to 3.0 Q and for “perfect storage” from 3.5 to 5.0 Q! I believe this startling increase in perceived quality will be obvious to anyone who tastes a wine that has spent its entire life in a perfect cellar. Wine such as the 1870 Lafite from the cellar at Glamis Castle pop into my mind. Here is a wine that consistently rates at 20 out of 20 points (or 21 if you believe in magic!) and yet it is now over one-hundred and forty years old. This is typical of a great wine stored in a great cellar.
Now to the bad news. For cost of storage, of course, there is practically no cost to store wine that is at 70 to 75 degrees. Usually a corner can be found that doesn’t require renting additional space and certainly there is no electricity bill for natural storage. This is given as 0.0 C (or zero cost). The bad news is that it costs money to store wine properly. When the cost of building a cellar, then refrigerating it and finally paying the recurring electricity bills is factored out to a monthly charge, it can be a big number. Thus, “standard storage” can be from 1.0 to 2.0 C and “perfect storage” can be truly frightening at 8.0 to 10.0 C! It costs money to obtain the sort of cellar that will squeeze the last iota of quality out of your wines.
The next question that might be asked is whether the fact that I have stored my wine perfectly will be rewarded with higher prices if I ever need to sell my wine. The answer is that theoretically this should be true. Auction houses try to indicate in their catalogues when an exceptional cellar is on the block, but I am not sure that the premium at this time is sufficient to repay the extra cost of perfect storage. This article will hopefully be a start in the education process that is necessary to have connoisseurs be aware of the critical importance o f storage conditions on the ultimate quality of wine. It is my belief that the prices reached at auction should be as noted in the chart, that is, wines stored in the warm cellar should reach the low estimate in an auctioneers catalogue, wines stored in “standard storage” should bring between the minimum and the maximum estimate in the catalogue, and wines stored in “perfect storage” should bring from 100 percent to 700 percent of the maximum estimate in the catalogue. It might seem outrageous that I would state that a wine could bring up to seven times as much as a less well-stored counterpart, and yet, if you look at the magnums of 1870 Lafite from Glamis Castle, these magnums in perfect condition sell now for about S25,000 each! This is several times what a typical normally-stored magnum of 1870 Lafite would bring.
Now for one of the most interesting factors elucidated in the chart — the quantity of wine offered for sale from perfect cellars. The truth is that there is very little. I know of very few cellars in the U.S. with the cold storage conditions that prevail in Dr. Ichinose’s cellar. Thus, I feel that less than ten percent of wine offered at auction comes from “perfect cellars.” This tiny quantity of wine represents the best examples of any given wine that can be purchased, but it is nearly impossible to obtain wine of this quality!
One of the most common errors that is made in building a wine cellar is to confuse refrigeration and air conditioning. Any commercial space that needs to be cooled, be it a flour warehouse, lettuce storage facility or wine wholesalers warehouse, will be cooled with refrigeration. Of course, refrigeration can also be used to make warehouses downright cold, If not freezing. Examples would be meat aging warehouses, which are normally at about 39 degrees and caviar storage refrigerators, which are normally only one or two degrees above freezing. If all this is done with refrigeration, then why have air conditioning?
An air conditioner is a form of refrigeration unit, except that it is designed to cool spaces only a few degrees. Almost the exclusive use of air conditioning is in cooling homes and offices. Compared to a refrigeration unit the coils that circulate the cold gas are quite small and very cold. I am told by a refrigeration engineer that the problem with using air conditioning in a commercial situation, such as a vegetable warehouse, .is that because the coils are small and very cold, they suck moisture out of the air. This has a dessicating effect on the air conditioned space and in the case of the vegetable warehouse, there would be noticeable drying out of the produce. Note that if a wine cellar is air conditioned, it is the wine that will suffer.
An excellent source of food refrigeration for both temperature and humidity control is suppliers to poultry and egg dealers. This type of refrigeration uses very large coils that are on the inside of the space to be cooled. Tubes carrying the cooling agent pass through a wall of the space to an outside area, even if the tubes must run some distance through intervening rooms. Once these tubes have passed through the outside wall, they are connected to the refrigeration pump. It is this pump that actually “cools” the freon gas (or other cooling agent). A refrigeration unit is a “heat pump”that pumps heat out of the space to be cooled and carries it to the outside air where the refrigeration pump releases it to be carried off in the air. By nature, refrigeration units do not cause moisture loss inside the cooled space, and thus are ideal for cooling a wine cellar.
My advice is that you should refrigerate, and under no circumstances air condition, your cellar!
This article, which was originally published in Rarities, Volume 1, No 4, August 1992, has been revised and updated.