How The South African Guidelines Put a Knife in the Bee Industry

By Joe Traynor

A slightly different version of this article appeared in the October 2006 BEE CULTURE

The South African Guidelines (for getting seedless mandarins)

At least 10 rows of a buffer cultivar OR at least 1640 feet of open ground between 2 cross- pollinating cultivars.

Author’s note and disclosure: I sent a final draft of this article to David Krause, President of Paramount Citrus, requesting that he advise me of any factual errors so that I could correct them before submitting the article to Bee Culture. Mr. Krause replied: “We disagree with your report, which contains many factual errors. We have no further comment.” I went through the final draft again, line by line, and made some changes, guessing at what the errors might be. To the best of my knowledge, this article is factual, but the reader is forewarned that it may still contain some, even “many” errors.

For those that missed it, the big news out of California this spring was the April 19 letter a number of beekeepers received from Paramount Citrus’ attorney Andrew Asch requesting they keep their bees at least 2 miles from certain Paramount mandarin plantings and threatening legal action if the bees were not removed by April 21st. The “trespassing” (Asch’s word) bees were causing seeds, and seeds greatly reduce the market value of mandarins.

Needless to say, this letter (hereinafter referred to as The Letter) raised a firestorm of protest from affected beekeepers and sent a ripple effect through the entire US bee industry; if beekeepers can be forced to move because their bees are “trespassing”, every beekeeper in the US is affected. Some beekeepers removed their bees after receiving The Letter, some stayed. As of August, negotiations are taking place between Paramount and the California State Beekeepers Association (CSBA) but as of this writing it appears that the matter will be litigated and that the only real winners will be the attorneys.

How did such a standoff come to pass? What was the genesis of what some have called a Range War? The answer is The South African Guidelines. Around 1998, seedless mandarins, esp. Clementine mandarins, from Morocco and Spain (and some from South Africa) became increasingly popular in the US because of their intense flavor and easy peeling. California citrus growers took notice and took a crash course in growing mandarins. The largest grower to step to the plate with mandarins was Sun Pacific who planted over 1,000 acres of mandarins in the Maricopa area (aka the Maricopa Flats or the Wheeler Ridge-Maricopa area) in the southwest corner of Kern County.

Most of the popular new mandarin varieties (and all of those causing problems for Paramount) are seedless if grown in an isolated block, even when bees are present. Some varieties (including several promising new varieties) are seedless in a mixed variety planting even when bees are present. The current most popular seedless mandarin varieties, however, require isolation from compatible pollen sources or else they will become seedy (if bees are present). All of this was known when Sun Pacific started planting mandarins but growers felt as long as they could isolate the problem varieties the seed problem would be solved. Here is where a single word, isolation, assumes critical importance and the definition of that word even greater, or paramount, importance. An improper definition of the word eight years ago is the genesis (seed, if you will) of the current imbroglio.

Instead of relying solely on University of California expertise, Sun Pacific hired one or more consultants from South Africa who had some experience in mandarin cultivation. South Africans enjoy a deserved reputation as superior agriculturists and their English skills, work ethic and affability make them ideal collaborators in any venture. South Africans, in short, are great people and when these self-assured people look you in the eye and tell you this is the way to do things it is difficult, if not impossible, to question their advice.

In contrast, University of California personnel, on the whole, usually couch their recommendations in qualifying language: “ this is what our data show, but it may not apply to your situation” or “additional studies are needed before we can make a recommendation” or “ we simply don’t have enough data to answer your question.” Such responses can cause impatience and frustration for someone chomping at the bit to plant mandarins. The UC extension service held meetings throughout the San Joaquin Valley when the mandarin craze started about 8 years ago and I attended almost all (maybe all) of these meetings. Kern county farm advisor Craig Kallsen presided over the meetings I attended in Bakersfield. Drs. Tracy Kahn and Thomas Chao, UC, Riverside, were speakers at some of the meetings. Sun Pacific and Paramount Citrus personnel were at some of the meetings and one or more South African consultants attended an occasional meeting. The question inevitably arose at every meeting: “How far do Clementines need to be isolated to be seedless?” The UC personnel would usually reply (paraphrasing), “At present, we don’t know; we don’t have sufficient data at this time to answer that question” a prudent (and correct) reply. In her Power-point presentation, Dr. Kahn included a slide giving the South African Guidelines, which many in the room dutifully wrote down: At least 10 rows of a buffer cultivar or 1640 feet of bare ground between 2 cross-pollinating cultivars.

Sun Pacific (and Paramount via Sun Pacific) likely made their planting arrangements based on input from the South Africans. They may not have followed the exact Guidelines presented above, but likely at least some permutation of them. The South Africans also exchanged information with UC personnel – a common practice among those studying the same subject. South African Etienne Rabe, currently employed by Sun Pacific, did his graduate work at UC Riverside and in 1999 in the UC publication Subtropical Fruit News wrote a two page article titled Considerations Relative to Establishment of Clementine Mandariins (Vol. 7, No. 1, 1999). In this article, Dr. Rabe addresses the seediness problem thusly: “The Clementine is a self-incompatible variety which does not have the capacity to pollinate itself. It therefore sets seedless fruit in solid-block plantings. This characteristic of self-incompatibility, producing seedless fruit is a sought-after characteristic.” There is no mention of buffer zones between solid-block plantings of different varieties and the word “isolation” does not appear in the article. Apparently in 1999, Dr. Rabe, and other South Africans, did not consider seediness a significant problem. Note: Many of those at the UC meetings mentioned above were Pest Control Advisors (or consultants like myself) who attended to keep up to speed on all facets of citrus culture (and to attain education credits for their PCA licenses) since the meetings also included segments on pest control, etc. I know of no one at the UC meetings that based their plantings on the South African Guidelines.

The South Africans may have arrived at the Guidelines via South African G.H. Barry’s 1995 M.S. Thesis (A prediction model to determine the cross-pollination ability of Citrus spp.) cited by Dr. Chao, et al. in a 2005 paper. (Note: Barry may have also served as a consultant for Sun Pacific). This South African model in turn may have been based on studies, including studies done in almonds, which showed that bees are very territorial – worker honey bees return to the same relatively small area on every trip from the hive. A 10 row buffer or a 1640 foot separation between mandarin varieties would seem more than ample. Further bolstering this position is that a roughly 500 feet separation is required between varieties of alfalfa seed plantings to assure pure seed; and producers of genetically modified (GM) seed and crops would agree that the South African Guidelines would be more than adequate to prevent gene transfer to “normal” crops and weeds. Unfortunately, we are now finding GM crops need a much wider separation than previously thought. The GM people, and possibly the South Africans, neglected the fact that honey bees can transfer pollen, bee to bee inside the hive. Thus a bee that is territorial to one variety of mandarins can transfer its pollen to a bee that is fixed on another variety, and vice versa. The pioneering work of Gloria DeGrandi-Hoffman (currently head of the USDA Bee Lab in Tucson, Arizona) showed this in-hive pollen transfer and subsequent workers have confirmed her work; all this is in published studies going back to 1986. DeGrandi-Hoffman ran a model to predict apple set and came up with a set of 1%, far short of a commercial crop; when she allowed for in-hive transfer of pollen, the model worked. Apparently the bees weren’t consulted when the initial mandarin plantings were made.

The mandarin-seed history in Morocco and Spain also brings into question the South African Guidelines – a 2 to 3 mile buffer zone has been found necessary in both countries. If Sun Pacific had received more input from Spain or Morocco, their planting layouts would likely have been quite different and beekeepers would be the better for it. This is not in any way to imply that the South Africans are poor or careless scientists – far from it. Their initial studies, esp. Barry’s work, represent superior science; these studies just weren’t carried far enough. All research is a continuum with one worker building on the work of another. Dr. Chao built on Barry’s work and determined that much more isolation was necessary than previously thought to get seedless mandarins. The South Africans may well have arrived at the same conclusion had they continued their work. Although Dr. Chao’s work serves as the foundation for the article you are currently reading (this article would not have been written without it) others will undoubtedly build on his work and shed even more light on the seediness problem in mandarins.

Sun Pacific planted their first mandarins around 1998, apparently using the South African Guidelines (or a form of the Guidelines) to separate different varieties. The first uh-oh moment came around 2003 when significant seed numbers were found in some plantings. In 2004, Sun Pacific contacted landowners and beekeepers in the Maricopa area to clear the area of bees. They got beekeepers to vacate their bee locations by promising them other citrus locations on Sun Pacific’s holdings further north. Sun Pacific kept its promises to beekeepers and successfully turned the Maricopa citrus area into a ghost town for honey bees. I was personally involved in the Maricopa “cleansing” having secured locations in the area for two beekeepers and I met with Sun Pacific representative, Don Berry, to resolve the situation. In our “frank” discussions, (and in his discussions with the two beekeepers) Mr. Berry was a gentleman at all times. He never raised his voice, but put forth Sun Pacific’s case in a forceful but calm and courteous manner; there was never any threat, veiled or overt, of litigation. Mr. Berry’s diplomacy undoubtedly led to a smooth resolution of the “Maricopa problem.” Although they probably wouldn’t have worked because of the vast acreage involved (see below) Mr. Berry’s diplomatic skills were never utilized (to my knowledge) in the current problem; instead many beekeepers were blindsided by The Letter.

Sun Pacific formed a working relationship with Paramount Citrus around 2000, advising Paramount on aspects of mandarin growing, including variety layout. Most of Paramount’s citrus is in the citrus belt of the San Joaquin Valley, the east sides of Kern, Tulare, Fresno and Madera counties. This vast citrus area grows over 200,000 acres of citrus including about 130,000 acres of pollen-sterile navels, 50,000 acres of Valencias, 5,000 acres of lemons and 18,000 acres of tangerines (including mandarins) Just as the South African Guidelines likely influenced Sun Pacific’s plantings in the Maricopa area, they also likely played a part in Paramount’s planting arrangements in the citrus belt (beekeepers might say that the Guideline virus was transmitted to the citrus belt) the major difference being the huge difference in citrus acreage between the two areas, the Maricopa area being much, much smaller in size. Note: When Clementines were first considered in California, Valencias were thought to be the pollen monster because of their extensive acreage. In an April, 2000 trip to Morocco, Dr. Chao (and co-author Guy Whitney) reported that Valencia orange planted with Clementine mandarin did not cause the resulting mandarins to have seeds (UC Subtropical Fruit News, 2000); the strains of Valencias were not specified. In the same M.S. thesis referred to above, and in a superior piece of work, Graham Berry showed that there were significant differences in the ability of pollen from different varieties to set seeds in other varieties and that some of the Valencia strains he tested had a very low ability to set seeds in Clementines. UC Riverside has temporarily (pending further study) categorized the two main California strains of Valencias, Olinda and Frost Nucellar, as moderate in their ability to set seeds. UCR has shown that W. Murcott Afourer was one of the highest, if not the highest, in its ability to set seeds in other varieties (i.e., that pollen from W. Murcott Afourer is one of the most virulent pathogens if seediness is considered a disease).

Probably emboldened by their success in clearing the Maricopa area of bees, Sun Pacific and Paramount used the same tactics to establish a 2-mile bee-free zone around their citrus belt holdings. A circle with a radius of 2 miles has an area of 8038 acres. Since many orchards are half a mile long, a two and a half- mile bee-free circle would be needed; such a circle contains 12,560 acres. Paramount representative James Jordan gave beekeepers 8 mandarin orchard locations targeted to be bee-free – two in Kern County, four in Tulare County, one in Fresno County and one in Madera County. Three of the four Tulare County locations are within 2 miles of each other and the two Kern County locations are within four miles of each other so the total acreage to be cleared (of bees) is less than if the locations were separated by four miles. Still, establishing a two-mile bee-free zone around these eight sites means clearing about 70,000 acres of bees. To accomplish this Herculean task, Sun Pacific and Paramount again offered beekeepers citrus locations away from their mandarins but these locations had lost much or all of their value because they were already over-crowded with bees. In an aside, Mr. Jordan mentioned that he had some Clementine trees at this house and intended to establish a bee-free zone around them; hopefully, he was kidding — if not, add another 8,000 acres to the Off Limits to Beekeepers list. If smaller growers of problem mandarin varieties climb on Paramount’s bee-free bandwagon, many more acres of bee locations would be potentially lost to beekeepers.

As the seediness problem grew in importance, Paramount Citrus, to their everlasting credit, cooperated with UC researcher Dr. Thomas Chao in a 2004 study to determine the parentage of seeds on one of their Kern County orchards. Another orchard in Madera County was used for the same purpose. In a superior piece of work (and one that will be cited by the anti-GM crowd for years) Chao determined that pollen was transferred across over 90 buffer rows at the Madera site and across 1430’ of open ground or 91 rows of pollen-sterile navels at the Kern County site. In Paramount’s Kern County orchard 100% of Clementine seeds tested came from W. Murcott Afourer pollen (there was a Valencia orchard of unspecified size located 3608’ or 1.1 km south of the Clementines; no Valencia parentage was found in the Clementine seeds). Chao concluded that “Growers in California, as well as in other places, need to consider a large distance of isolation, up to several km or more than 116 rows of buffer trees, to prevent cross-pollination by honeybees.” Chao’s study put a fork in the Guidelines (may they rest in peace).

The two most popular seedless mandarin varieties are Clementine Nules and W. Murcott Afourer and the latter is an especially effective seed setter in the former as Chao showed in his study. Paramount Citrus farms about 20,000 acres of citrus in the San Joaquin Valley and manages another 10,000 acres with its subsidiary S&J Farm Management. With all that acreage, much of it pollen-sterile navels, Paramount (and Sun Pacific) should have easily been able to isolate problem varieties from problem pollen sources. The reason they didn’t probably lies in the harvesting windows for mandarin varieties – Clementine Nules ripens in October-November, W. Murcott January-March. Less popular varieties (Fina Sodea, Caffin) probably fall in between. Fruit in the San Joaquin valley ripens from south to north with southern Kern County as much as 2 weeks earlier than Madera County in some years. Planting a combination of varieties on a single ranch makes sense from a marketing standpoint but can be devastating from a seed standpoint..

By mixing mandarin varieties (mainly W. Murcott Afourer and Clementine Nules) on single ranches, Paramount (and Sun Pacific) fouled their own nest. They planted trees that produced a known pathogen (W. Murcott pollen on some of their plantings) far too close to a pristine seedless variety (Clementine Nules, which can also cause seeds in W. Murcott). Paramount is now focusing almost 100% of its energy in the seediness problem to the transmitter of the pathogen (mainly honey bees) and little or no attention to the pathogen (pollen). Shouldn’t Paramount be suing those that gave them planting advice rather than threatening to sue beekeepers?

It takes considerable gumption for someone to admit he made a mistake, but Paramount has shown no indication that they are willing to admit they made a mistake in the varietal distribution on their mandarin plantings. Instead, they are exerting extreme pressure on beekeepers to neutralize their initial planting error.

Paramount’s mandarin plantings could be considered an attractive nuisance for bees, just as an unfenced swimming pool is considered an attractive nuisance for youngsters (at least in California) and pool owners can be sued if a youngster in the neighborhood drowns in their pool and the pool was unfenced. That Paramount Citrus knew it had a pollen problem with mandarins is evidenced by the fact they had buffer zones between varieties. These fences just weren’t constructed well enough to prevent pollen transfer by bees. From a legal standpoint, Paramount would probably have been better off by planting different varieties side by side with no buffer zone between them.

Paramount Citrus and its sister company Paramount Farming are class organizations. I have worked with Paramount Farming and know they are class people from top to bottom. I have had very little interaction with Paramount Citrus but feel it is also run by class people because the owner of both companies, Stewart and Lynda Resnick appear to be quality people. Stewart Resnick is the ultimate decision maker in this dispute (Sun Pacific is heavily involved but is staying on the sidelines and letting Paramount do the heavy lifting). Mr. Resnick usually stays removed from major decisions by the Paramount companies because he knows competent people run them. Apparently no one in either of the Paramount companies has informed Mr. Resnick that he risks being badly bloodied if this dispute goes to litigation.

Paramount Citrus can (and eventually probably will) solve this problem by topworking problem pollen sources (and problem pollen recipients) to other varieties. It will be costly to do so but it doesn’t take long for topworked trees to return to full production. It is highly unlikely that Paramount will succeed in pushing beekeepers off long-held locations. The longer the delay in transitioning their problem varieties, the longer their seeded mandarins will suffer in the marketplace. Legal battles sometimes last for years before a resolution but time is not on Paramount’s side in this one. Each year that goes by without a decision means more money lost in the marketplace for Paramount. If the ultimate decision goes against Paramount they will not only have lost precious time in re-designing their orchards but will look foolish for having waited. To topwork trees, however, would be an admission of error, and to date Paramount Citrus has shown no inclination to admit making a mistake.

There are no “bad guys” in this matter (although beekeepers might argue that Paramount attorney Andrew Asch, author of The Letter, deserves the “bad guy” label). The three major mandarin participants – Paramount, Sun Pacific and the South Africans – are all good people and all of them acted with the best of intentions. As the oft-quoted Robert Burns said: The best laid plans o’ mice and men gang aft agley. The best intentions of the mandarin participants went awry and beekeepers are under threat of paying a heavy penalty as a result.

Both Paramount and Stewart Resnick are likely getting conflicting advice on this. It is unlikely that Sun Pacific will encourage Paramount to admit to a mistake. On the other hand, there are likely some (although I don’t know of any) in Paramount Citrus or Paramount Farming that feel that Paramount Citrus should admit its planting error, cut their losses and move on with life. Stewart Resnick holds the knife that currently resides in the belly of the US bee industry – he can either remove it, or give it a twist. Only time will tell if Mr. Resnick makes an informed and proper decision.


Chao, T., J. Fang and P. Devanand. Long Distance Pollen Flow in Mandarin Orchards Determined by AFLP Markers – Implications for Seedless Mandarin Production J. Amer. Soc. Hort. Sci 130(3):374-380 (2005).

DeGrandi-Hoffman, G, R.D. Hoopingarner and K. Lomparenf. The Influence of Honey Bee (Hymenoptera Apidae) In-Hive Pollen Transfer on Cross-Pollination and Fruit Set in Apple. Environmental Entomology,15:723-725(1986).

Postscript: UC Riverside recently released what is essentially a seedless W. Murcott Afourer, calling it the Tango variety. Not only is Tango seedless in the presence of bees (and nearby compatible pollen) but if does not make seeds in other mandarins. UCR is currently working on a seedless Clementine Nules. Spain reportedly already has a seedless Clementine Nules (called Nulesin) that may be available in the US in the future. There are indications that fruit set on the new seedless varieties can be improved by the addition of bees (note: seedless watermelons require bees to set a commercial crop). If so, what a switch that would be – mandarin growers begging beekeepers to place bees on or near their orchards! Many in the citrus industry feel that in five years, virtually all mandarins in California will bee the new seedless varieties (along with the old seedless Satsuma mandarins). If (when) this happens, the current dispute between beekeepers and Paramount will be remembered (if it is remembered at all) as a brief but unpleasant episode in the annals of California beekeeping.