Before it became the fashion to end all plant family names with –aceae, the Apiaceae was generally called the Umbelliferae. As described here, the description of the family includes the Hydrocotylaceae, Angelicaceae, Daucaceae and Umbellaceae of some authors. This seems to represent the current trend. Depending on the reference consulted, the family consists of between about 250 and 300 genera and between approximately 2800 to a little more than 3000 species. The family generally inhabits the temperate and boreal regions of the Northern Hemisphere, and tropical representatives are frequently found in mountainous areas. About 25% of the genera and 10% of the species are native to the U.S. The family is generally made up of biennial or perennial herbs, occasionally woody plants, but only rarely, trees. The stems are often stout, furrowed and hollow between the junctures of leaves or branches (internodes). The alternately placed leaves are usually compound, often pinnately compound, and the leaf stems (petioles) commonly form sheathes at their base. Members of the family are often aromatic. The individual flowers are usually small to minute, usually bisexual and radially symmetrical. There are usually 5 sepals, petals and stamens (5-merous), though sometimes petals are absent. When present, the petals are usually yellowish or whitish. The ovary is in the inferior position1, composed of two carpels2, each with one ovule3 and a style. The inflorescence usually takes the form of a compound umbel4. There is also an inflorescence type that forms a somewhat cylindrical or conical head. For those of us who have associated the family with plants like Queen Anne’s lace, this form has a tendency to raise our eyebrows and cause us to shake our heads! The fruit is a schizocarp5. The family has considerable economic importance. It contains numerous food plants (carrots, parsley, parsnips, and celery), spices/seasonings (coriander, caraway, anise, fennel and dill), and ornamentals (eryngo, rattlesnake master, sea holly, angelica and cow parsnip). Some have been used for medicinal purposes in the Western world and probably still are, to a small extent. They seem to be used more extensively for medicinal purposes in other parts of the world. Some cause a dermatitis in sensitive individuals (examples:wild parsnip and cow parsnip). Depending on how sensitive the person is to the plant, the dermatitis can be quite severe. There are also some deadly poisonous plants in the family (examples: poison hemlock and water hemlock). One should not even taste members of the family unless there is absolute certainty about its identification.[2, 13 & 21] Carrot, (wild forms: wild carrot, Queen-Anne’s lace, queen’s lace, devil’s-plague) Scientific name: Daucus carota Origin: Carrots are thought by some to have come from a natural variant of wild Queen Anne’s lace in Afghanistan, and were domesticated there and in adjacent regions of Russia, Iran, India, Pakistan and Anatolia, a process that may have started 2000-3000 years ago[12]. Plant description: There are two groups of carrots, an annual tropical type and a biennial temperate zone type. Carrots are generally erect herbs with finely cut pinnately compound6 leaves with long stems that often form sheaths at their base. In its vegetative state (first year of the biennial form) it can be 20 to 50 cm (7.9 to 20 inches) in height, and in its flowering state can be 120-150 cm (47 to 59 inches). The root of the commercial carrot has a straight compact cylindrical or conical form about 2-5 cm (0.79-2 inch) in diameter at the top, and is 5-50 cm (2-20 inches) in length. Wild types often have a more branched root form. The root portion of the carrot comes in a variety of colors that range through white, black, purple, red, green, yellow and orange. The orange carrots, to which we in the US have become accustomed, were derived from a mutation in Netherlands during the 17th century, and, therefore, have a narrow genetic base, since they have been derived from only a few 18th century cultivars. The main carrot inflorescence consists of a 5 to 6 inch diameter terminal primary compound umbel consisting of small white flowers. Below this, there is a series of lesser umbels that decrease in size the farther down on the plant they are located. The different orders of umbels usually flower successively at 8-12 day intervals with the flowering of a single umbel lasting about 7-9 days. The outer umbellets (subumbels) of each umbel are first to flower and this is followed by the flowering of more central subumbels. The pollen at the center of an umbellet is larger than the pollen from the flowers on the periphery of the umbellet, and is also frequently also more fertile than that from the umbellet’s edge[12]. The individual flowers are usually perfect, but in the lower parts of the plant there is a tendency to produce male flowers. The ovary of the individual florets is inferior, and the pollen is released before the stigma becomes receptive. In normal flowers, after the pollen is released, the stamens are quickly shed. In the production of hybrids, the stamens of the male-sterile flowers either become petals (petaloid type) in which case they are not shed, or the anthers degenerate and wither before the flower opens (brown-anther type). A normal bisexual flower has five functional stamens and two styles, each leading to a locule7 of the ovary. Each locule has a single ovule and, therefore, each flower can produce two seeds. Nectar is produced on the surface of a swollen disc located on the upper surface of the ovary and is easily available to a variety of insects. It apparently both reflects and fluoresces under ultraviolet (UV) light. The anthers are also UV reflective[6]. Because bees can “see” UV light, these properties may be part of the guidance system that directs insects to the pollen and nectar resources of the flower. The center of the top umbel of the wild Queen Anne’s lace frequently, but not always, has a small central reddish floret. Its significance is unknown to me. It is sometimes suggested that it attracts pollinators. Distribution: Nearly all the carrots grown in the U S. are biennials, meaning they will not bloom and produce flowers until they have experienced a winter (vernalization). Carrots that are grown for eating, therefore, will be of no benefit to bees. Originally carrot seed production was carried out primarily in California. Sometimes, however, the winters were not cold enough for vernalization to occur or the summers were too hot for good seed development. As a result, much of the carrot seed production industry has moved to parts of Washington, Oregon, and Idaho. While hybrid carrot seed production is possible, for a variety of reasons it is easier, cheaper and more productive to produce open-pollinated carrot seed. As a result, carrot seed production is split into two production systems; hybrid seed production for the fresh market where quality and appearance are paramount, and open-pollinated production for the processing industry. Approximately 85% of the hybrid carrot seed production is carried out on the eastern side of the Cascade Mountains of Oregon in Jefferson, and surrounding counties [27 & 28]. The distribution indicated by the accompanying map represents that of the wild Queen Anne’s lace. Blooming period: Burgett et al.[4] provide a blooming date for Oregon as late June to mid August. Importance as a honey plant: Oertel[18], from his questionnaires, found carrot to be important in CA and ME. “Wild carrot” was reported as important in CA. Ayers and Harman[1], from their questionnaires, found carrot to be important in WA, OR and ID. In each instance, the crop was reported to provide commercial pollination opportunities. The difference between the two reports apparently reflects the shift in location of commercial carrot seed production described under Distribution. Both John Lovell[15] and Pellett[19] state that wild carrot produces nectar only occasionally. Lovell defines occasionally as about once every ten years. Pellett cites what appears to be a personal letter from a Pennsylvania beekeeper indicating that he had never seen a bee on wild carrot until 1921 when it “yielded freely” for a time in his locality. Milum[17] places wild carrot it in his minor honey plants list. Burgett et al.[4] state that it is worked mainly for pollen. Honey potential: Actual honey potential data for carrot appears to be scarce. While it doesn’t appear to be among the best of honey plants, it can divert bees from onion pollination (see September 2010 issue). On the other hand, Franklin[8] describes an incident in 1961 where four strong hives of bees located near an approximately half acre carrot plot could not be made to work the carrots despite redistribution of the hives around the field or the herculean attempt of spraying trails of carrot honey from the hive’s entrances to the carrot plot. Honey: There seems to be a difference of opinion about the quality of carrot honey. Harvey Lovell[14] states, “The honey is usually described as white with a mild flavor.” This statement isn’t changed in the Goltz[10] edition of this work. On the other hand, a number of authors describe it quite differently. John Lovell[15] describes it as “Yielding a white honey with the flavor of the foliage, which granulates in a few months.” After sampling a few carrot leaves, I have concluded that I wouldn’t consider honey with that taste to be of high quality. Pellett[19] describes a sample he received from a California beekeeper as, “very dark in color with a strong flavor.” He also provides the following quote from Tom Davis of Sacramento, CA, “(the) honey is dark amber and strong (in) flavor. It is somewhat turgid and granulates within two weeks after extraction. It foams quite a bit when heated denoting the presence of yeasts....” Vansell, in his 1931 Nectar and Pollen Plants of California[25], describes the honey as being “of light amber color”. This is changed in the 1941 Vansell and Eckert[26] edition to, “The honey is dark in color and strong in flavor.” Burgett et al.[4] state that the honey from cultivated carrot is “strong flavored” and is used for colony winter feed or by the bakery trade. Pollen: Carrot flowers are protandrous8, the pollen being released on the first or second day after the flower opens with the stigma becoming receptive on the third or fourth day, and then remaining receptive for week or more[9]. The pollen is available to a wide variety of insects, perhaps most notably members of the superfamily Apòidea9[2]. Additional information: Carrot pollinators Bohart and Nye[3] studied the insect pollinators of carrot in Utah over a three-year period in three different habits. In these studies they found 334 insect species from 71 families that were, to varying degrees, important in the pollination of carrots. They devised an efficiency rating system for the insects that most commonly visited their carrot plots based on the amount of loose pollen carried on their bodies, their size, flightiness (indicating tendency to cross pollinate) and their contact with stamens and stigmas as they moved across the flower heads. Pollen collecting honey bees received their highest rating (6). About this group they say, “In the case of honey bees the pollen collectors literally wade across the heads, swinging their abdomens back and forth and scraping the pollen from stamens with their forelegs.” Nectar-collecting honey bees received only a rating of 4 because they, “stand higher on the flowers, move about less, and lap up droplets from the exposed nectaries.” Their categorization of the various groups they found associated with carrot flowers is presented in Table 1. It’s an interesting group when one speculates why they are there. Some of them are there for the floral resources of nectar and pollen. That’s clearly why honey bees and bumble bees are there. The solitary Apòidea provision their young with pollen. Other visitors probably provision their young with “insect burger” and/or themselves are predators. In both cases, they are there for the kill. Some are probably primarily there for nectar to be used as “flight fuel” so they can go on about their daily routines. Some of the males might have been there to find mates. Some are there to feed on the internal juices (not nectar) of the plants; others to directly feed on the flowers. The researchers come to an interesting, and to me at first startling conclusion, that because all insect visitors feed on available resources at the flowers, large numbers of inefficient pollinators tend to reduce, rather than increase pollination, because they remove the resources that attract the efficient pollinators. In some cases they may even remove the pollinators. Pollination indices were developed by multiplying the various individual efficiency ratings by the corresponding numbers of the groups seen in the carrot fields. Honey bees were next to the bottom of the index ratings, just above the tiny flies. The top-rated sphecid wasps probably were there for “flight fuel” so they could go on about their daily predatory routines, or they might have been using the flowers as a “killing field” in order to provision their young with “insect burger”. The larger flies placed just under the sphecid wasps. Many of them masquerade as bees or wasps and were probably there mainly for the nectar and pollen, though the males may have been there, at least in part, to find mates. The researchers recommend that carrots raised for seed should be grown in diverse habitats that would provide greater numbers of natural pollinators. The immature stages of many of these larger flies live in wet and decaying matter and they suggest that providing sources of this type of material near carrot fields would probably improve pollination, and the best way to raise the honey bee pollination index values would be to eliminate competing bee forage and avoid competing crops. Genetics of Hybrid Carrot seed production The genetics of male sterility is similar to that of the onion (see this column September 2010). In the September 2010 column I used the word epistatic to indicate the “on and off” effect genes can have on other genes in a different location. While that is basically the definition of epistatic, I see it now as a bad choice of words. When the genetic material that controls male sterility is in both the cytoplasm and nucleus the term cytoplasmic male sterility (CMS) is used. It results from an interaction between the genes in the nucleus and a male sterility factor in the cytoplasm. The cytoplasmic factor controls male sterility and is inherited only from the female plants (male sterile plants). Genes in the nucleus can turn off the cytoplasmic control, and if the cytoplasmic factor is not such that it would produce male sterility, the nuclear genes essentially have no effect on pollen production, and the plant produces pollen. Successive inbreeding in carrots quickly reduces plant vigor and, therefore, seed production of carrot hybrids are normally the result of the three-way cross, (AxB) x C, because this procedure produces considerably more seed than the single AxB hybrid cross [12] using inbred A and B lines10. Pollination challenges related to honey bee behavior In the beginning, breeding hybrid carrots was pretty challenging. To be effective bees must transfer pollen from male-fertile to male-sterile plants. Erickson and Peterson[6 & 7] in studies done between 1974 and 1977 and working with petaloid type male sterility[20] in cages, found the following: (1) Male-fertile lines usually, but not always, bloomed ahead of the male-sterile lines and rarely was the bloom synchronized, the difference being as much as 30 days, (2) Five distinctly different aromas were generally noted among various male-sterile plants while only a single aroma was typical in nearly all male-fertile lines, (3) Flowers from male-sterile lines ranged in color from white to green, while all the male-fertile lines produced white flowers, (4) Generally the male-sterile plants produced less nectar than the pollen donor11, and there was an indication that differences in percentages of nectar solids might also exist. All of these differences would allow the bees to show considerable fidelity to particular lines. As an example, in their 1973 and 1974 studies only 8% of bees initially marked on the white male-fertile line moved to the white male-sterile line and only 5% moved to a green-male sterile line. In their 1976 study 8.5 % crossed over from fertile line-1, 23.6% crossed over from fertile line-2 and 33% crossed over from fertile line-312. Clearly carrot breeders need to pay close attention to developing parental lines that are quite similar. The results of Rodet et al.[20] in a caged study, using the brown-anther type of male sterility, are somewhat different than those obtained by Erickson and Peterson who apparently used the petaloid type. As one might expect, there were more pollen gatherers on the male-fertile lines than on the male-sterile lines, but in these studies the bees were equally distributed on the different carrot lines. On the male-fertile lines, the pollen gatherers had 14,446 pollen grains while the nectar-gatherers had 9582 grains. On the male-sterile lines the bees had 916 grains, which suggests a relatively small crossover rate between the lines. The researchers seem to argue that this low crossover rate resulted not so much from a fidelity to a particular line, but to a fidelity to a foraging area and “competition between foragers of one colony for the same foraging space”. If that is the case, one would expect that fields with landmarks (field corners, buildings and roads etc.) would be particularly troublesome. Clearly as we began to develop hybrid carrots, there were many problems to be solved. Apparently many of these problems have been largely solved. As mentioned above, the US fresh carrot market is dominated by hybrids. While the older literature exploring the problems is relatively easy to find, I think I see a trend that the published solutions are becoming less frequent. I speculate that this to some extent is accounted for by a change in whom is doing the research . Once much of the plant breeding was done by, or at least done in conjunction with, university personnel. Back then, the mantra at these institutions was “publish or perish”. Today much of the plant breeding is done by the seed companies, and in some cases, large pesticide companies, and I speculate that the information that is collected is now considered proprietary. Need for isolating carrot breeding fields Carrots are highly protandrous, and therefore, likely to have a high level of cross pollination if other varieties are available. To investigate this, Thompson[23] planted plots of orange rooted carrots as seed parents surrounded by a white rooted variety. Many of these transplants didn’t survive to produce flowers, and the ratio of white to orange transplants can’t be determined from his article, but there were many more white rooted than orange rooted plants involved. They found the percentage of self-pollination ranged between 0.0 and 2.5 % with an average of 1.1%. The researchers also reported on a similar earlier experiment by another researcher where the percentage of selfing was 2.4%. Thompson went so far as to suggest that hybrids could be made by planting small numbers of one variety along with larger numbers of another variety, and then harvesting the seed of the smaller planting. He pointed out, however, that this would not be economically practical for commercial hybrid seed production because of the “wasted” space occupied by the pollen parent. He also pointed out that other tests needed to be done because all carrot varieties might not act the same. Thompson’s research does suggest, however, that some degree of isolation would be necessary to produce seed that would provide pure plantings of a given carrot type. Franklin[8] proposed a series of isolation distances similar to his proposal for onions (see September 2010 column) for the production of both orange open-pollinated varieties, as well as for orange hybrids. In general the recommended distances for open-pollinated varieties were less than for hybrid seed production, and within each system the distances recommended increased as the differences in carrot type, shape and/or color increased. The recommended distances for open-pollinated varieties ranged from 1 mile down to 1/4 mile while those for hybrid seed production ranged from, 3 miles down to 1/2 mile. Kalia[12] states that the isolation distance for fields producing commercial seed crops should be 1000m (0.62 miles) and this should be increased to 1600m (0.99 miles) for “basic seed” production. Presumably, isolation should also include at least some isolation from wild carrots (Queen Anne’s lace). Spacing carrot lines for hybrid seed production Franklin studied the effect of distance between the male-fertile plants to male-sterile plants during 1963 and 1964. The field was laid out with 21 rows of male-sterile plants between a two single rows of male-fertile plants. The 288 ft long rows were spaced 3 ft apart, and the blooming dates of the two varieties coincided closely. In 1963, the two outside male-sterile rows averaged 21.5 lbs of seed, while the two central rows averaged yields of only 12.1 lbs of seed. In the 1964 study, the two outside male-sterile rows averaged yields of 9.4 lbs of seed, while the two central rows averaged 8.9 lbs of seed. The authors concluded that in 1963 there was a diminishing seed yield toward the center of the field, but that this didn’t happen during 1964. I found no explanation for the poorer overall yields of 1964 as compared to 1963. Takahashi[22] states in Japan the ratio of male-sterile to male-fertile plants ranges between 2 to 4 male-sterile plants to 1 male-fertile plant. Kalia[12] reports that a 4 to 1 ratio is common and this is often grown in an 8 to 2 arrangement. Number of bees recommended for pollination In a four-year study (1954-1957) Hawthorn et al.[11] studied pollination in: (1) uncaged carrot plots, (2) cages with honey bees, (3) cages that admitted only “tiny” insects and (4) cages that were designed to essentially be insect free. They found the carrot seed yield of the carrots caged with honey bees and the yields from the open field situation to not be statistically different except for the year 1957, when the honey bee populations in the field were exceptionally low, which was reflected by exceptionally low seed yields from the open plots. Based on the lowest average yearly population of caged bees over the four years, they estimated that 8 bees per square yard “is apparently as high as the plants can use to advantage”. They also speculated that smaller populations would probably work just as well, but admitted they hadn’t the data to back such a claim. In addition to producing lower seed yields, with the exception of 1957 when the honey bee population in the field was exceptionally low, the open-field plots and the plots caged with honey bees provided significantly higher quality seed than the other caged plots. During 1957 the open field plots also produced high levels of poor quality seed. McGregor[16] doesn’t suggest a number of colonies per se, but based on the research by Hawthorn et al. just reported above, suggests at least 8 bees per square yard, regardless of the number of colonies needed to provide that number. Delaplane and Mayer[5] provide references for recommendations for open-pollinated carrot seed production of 2 to 4 hives per acre (5-10 hives per ha) with an estimated literature average of 3 hives per acre (7.5 hives per ha). For hybrid seed production they recommend 4-6 hives per acre (10-14.8 hives per ha). Alternatively, they suggest 6-8 bees per square yard (7.2- 9.6 bees per square meter) and provide an estimated literature average of 7.3 bees per square yard (8.8 bees per square meter). References 1.Ayers, G. S. and J. R. Harman. 1992. Bee Forage of North America and the Potential for Planting for Bees. In The Hive and the Honey Bee (J. M. Graham, Ed.), Dadant and Sons. Hamilton, L. 2.Baumgardt, J. P. 1982. How to Identify Flowering Plant Families--A Practical Guide for Horticulturists and Plant Lovers. Timber Press. Portland, Oregon. 3.Bohart, G. E. and W. P. Nye. 1960. Insect pollinators of carrots in Utah. Utah Agricultural Experiment Station Bulletin 419. 4.Burgett, D. M., B. A. Stringer and L. D. Johnston. 1989. Nectar and Pollen Plants of Oregon and the Pacific Northwest. Honeystone Press. Blodgett, OR. 5.Dellaplane, K. S. and D. E. Mayer 2000. Crop Pollination by Bees. CABI Publishing. New York. 6.Erickson, E. H. and C. E. Peterson. 1979a. A synchrony of floral events and other differences in pollinator foraging stimuli between fertile and male-sterile carrot inbreds. Journal American Society of Horticultural Science 104: 639-643. 7.Erickson, E. H. and C. E. Peterson. 1979b. Problems encountered in the pollination of cytoplasmically male-sterile hybrid carrot seed parents. Proceedings of the Fourth International Symposium on Pollination, Maryland, (1978): 59-63. 8.Franklin, D. F. 1970. Problems in the production of vegetable seed. In: The Indispensible Pollinators, A Report of the Ninth Pollination Conference, Hot Springs, Ark, Oct 12-15, 1970. University Arkansas Agricultural Extension Service MP 127: 112-141. 9.Free, J. B. 1993. Insect Pollination of Crops (2nd Edition). Academic Press. Ltd. London. 10.Goltz, L. R. 1977. Honey Plants--A revised edition of the original Honey Plants Manual of Harvey B. Lovell, Ph.D. Gleanings in Bee Culture 1977. 11.Hawthorn, L. R., G. E. Bohart and E. H. Toole, W. P. Nye and M. D. Levin. 1960 Carrot Seed Production as Affected by Insect Pollination. Utah Agricultural Experiment Station Bulletin 422. 12.Kalia, P. 2004. Root Vegetable Crops. In: Hybrid Vegetable Development. Singh, P. K, S. K. Dasgupta, and S. K. Tripathi Eds. Foods Products Press. New York. 13.Liberty Hyde Bailey Hortorium Staff. 1976. Hortus Third. A Concise Dictionary of Plants Cultivated in the United States and Canada. Macmillan Publishing Co. Inc. New York. 14.Lovell, H. 1966. Honey Plants Manual: A Practical Field Handbook for Identifying Honey Flora. A. I. Root Co. Medina, OH 15.Lovell, J.H. 1926. Honey Plants of North America. A. I. Root Co. Medina, OH. 16.McGregor. S. E. 1976. Insect Pollination of Cultivated Crop Plants. Agriculture Handbook No. 496, Agricultural Research Service. United States Department of Agriculture. Washington D. C. This publication is being updated and is available on the web at gears.tucson.ars.ag.gov/book 17.Milum, V. G. 1957. Illinois Honey and Pollen Plants. Contributions from the Department of Horticulture, University of Illinois. Urbana, IL. 18.Oertel, E. 1939. Honey and Pollen Plants of the United States. U. S. D. A. Circular 554 U. S. Government Printing Office. Washington D.C. 19.Pellett, F. C. 1978. American Honey Plants. Dadant and Sons, Hamilton, IL. 20.Rodet, G., J. P Grosse and A. Bonnet 1991. Foraging behavior of Apis mellifera L. on male-sterile and male-fertile inbred lines of carrot (Daucus carota L) in gridded enclosures. The 6th International Symposium on Pollination, Tilburg, The Netherlands, August 1990. Acta Horticulturae 288: 371-375. 21.Smith, J. P. 1977. Vascular Plant Families. Mad River Press. Eureka, CA. 22.Takahashi, O. 1987. Utilization and seed production of hybrid vegetable varieties in Japan. In: Hybrid Production of selected cereal oil and vegetable crops (Eds. Feistritzer, W. P. and A. F. Kelly. FAO Plant Production and Protection Paper 82. pp 313-328. 23.Thompson, D. J. 1962 Natural cross-pollination in carrots. Proceedings Amcerican Society of Horticultural Science 81: 332-334. 24.USDA, NRCS. The PLANTS Database, Version 3.5 (http://plants.usda.gov). National Plant Data Center, Baton Rouge, LA 70874-4490 USA 25.Vansell, G. H. 1931. Nectar and Pollen Plants of California. University of California Agricultural Experiment Station Bulletin 517. Berkley, CA. 26.Vansell, G. H. and J. E. Eckert. 1941. Nectar and Pollen Plants of California. University of California Agricultural Experiment Station Bulletin 517 (1941 Revision), Berkley, CA. 27.http://www.ag.ohio-state.edu/~seedsci/vsp02.html 28.http://extension.oregonstate.edu/catalog/pdf/em/em8879-e.pdf