FINDING A SENSE OF SELF: the evolution from outbreeding to selfing in morning glories and its impact on agriculture

stephen dimaria

duke university


One of the most common changes in flowering plants is to start reproducing with itself, known as selfing. The default mode of reproduction for plants is outbreeding, which involve the male gametes (pollen) being moved by water, wind, or animal pollinators to the female gametes of another individual. On the other hand, if the male and female gametes are produced at the same time by a plant, selfing can occur.1 Selfing is the fusion of male and female gametes from a single organism. As a result of the transition from outbreeding to selfing, research has delved into studying plant sexual diversity involving the fitness consequences of outcrossing and selfing as early as 1957, where Stebbins observed selfing and explained the results that come with selfing, who thought selfing was an evolutionary “dead-end.”2


Theorizing why selfing occurs


The most widely accepted model to describe the why selfing occurs describes a setting where selective pressures include limited pollinator and/or mate availability.3-4 In this model, since flowering plants lack an adequate amount of pollinators to help them reproduce, selfing proves to be evolutionarily beneficial. This phenomenon confers an evolutionary advantage, as first discussed by Darwin, via reproductive assurance, which is the ability to reproduce when pollinators or potential mates are limited.5 Additionally, as described by Fisher, there is an intrinsic advantage to selfing. The advantage is to have an ovule fertilized and to fertilize another ovule - i.e. to produce two offspring, one as a mother and one as a father. For outcrossing plants, two copies of its genome are passed down to the next generation: one in the ovule that it receives pollen (from another plant), and one in the pollen that the plant itself exports to another plant. On the other hand, a selfing plant is able to use its own pollen to fertilize that ovule and exports pollen, thus passing on three copies of the genome instead of two. Thus, selfing increases the amount of progeny; however, only of typically short-lived persistence.6-7 In summary, selfing guarantees reproduction for the plant, but over time, selfing leads to reduced diversity and an increase in harmful mutations in the plant’s genes. As a result, selfing has been thought of as an evolutionary “dead-end” since in the long-term, selfing is harmful to the plant, albeit being beneficial in the short-term.


Effect of Morning Glories on Agriculture


Ipomoea lacunosa is a selfing species that is commonly researched, due to its vast agricultural implications. The genus Ipomoea is often referred to as “morning glories,” which is a typical kind of weed. The genus Ipomoea is characterized by rapid generation time, ease of growth in the field, and excessive seed production, making it especially noxious to agricultural crops. In addition, morning glories can have severe economic impact. Specifically, competition between morning glories and crops can reduce yield by as much as 80-88 percent.8-9 To further complicate the issue, morning glory infestations cause declines in crop quantity and quality, which leads to agricultural problems and reduced harvesting efficiency. This would certainly have deleterious consequences and can lead to complete loss of crop.10     

Farmers utilize various methods to help monitor weed infestations, typically involving the use of herbicides. The most ubiquitous herbicide is RoundUp Ready Technology, which has caused a surge in the use of glyphosphate, the active chemical in RoundUp. Despite the increased use of RoundUp, morning glory species have been shown to be notoriously difficult to control with this herbicide.11


Identifying selfing genes in plants to monitor our food supply

Identifying the genes involved in selfing is especially important and has vast agricultural implications. Specifically, finding what genes are present in outcrossers but are absent in selfers can be one way to secure the food supply. For example, due to the domestication of crops, as a human species, we have caused the evolution of crops for thousands of years. In crops such as rice, a change from outbreeding to selfing has been associated with domestication. This change occurred since self-pollination is favored in agriculture as it increases the short-term response to selection and preserves gene combinations present in selected individuals.12 By determining genes involved in outcrossing and those in selfing, we can prevent selfing, which is essential for food security, implications of which are discussed next.  

Despite the prevalence of selfing, offspring produced by selfing are generally less fit than outbred offspring. This reduced fitness is generally due to inbreeding depression - or the reduced survival and fertility of offspring of related individuals - usually as a result of the expression of bad genes. Furthermore, selfing limits the genetic diversity of the population, reducing its ability to adapt to a changing environment. Despite these negative consequences, it has been shown that these are not negative enough, in the sense that they are not so detrimental to cause the counteract the short-term advantages of selfing.13 However, selfing as a mode of reproduction leads to the accumulation of harmful mutations. These harmful mutations reduce crop yield. Thus, by controlling plant fertility (i.e. how they reproduce), food supplies would be more secure and plant breeding would be done more efficiently (as there are two types of reproduction, outbreeding and selfing).


Morphological changes associated with selfing

The change from outbreeding to selfing is often associated with a typical set of changes to the appearance and function of flowers, termed the selfing syndrome.14-15 It has been shown in large sample sizes that predominantly selfing species open less (have smaller corolla length and width) (Fig. 1), produce less pollen, less nectar, and have closer anther (the part of the stamen where pollen is produced) and stigma (position on the pistil where pollen germinates) positioning.16-17 The transition to produce less pollen is an especially important trait since it is a good sign of the transition to selfing.  A predominant model describing these morphological observations theorizes that the energy used to build large corollas and provide much nectar as a reward for pollination would be allocated to other processes in a selfing flower, which does not rely on pollinators. As a result of the corolla closing, the anther and stigma become closer together, causing stigma length to decrease.18

Mimulus, or monkeyflowers.

Mimulus, or monkeyflowers.

Leptosiphon, or linanthus.

Leptosiphon, or linanthus.

Relationship of pollen production with floral morphology

Genes that control pollen production are independent from the genes that are in control of nectar volume production, stigma length, corolla width, and corolla length in Ipomoea lacunosa. With little correlation amongst pollen count and the respective phenotypes, it was shown that these traits are controlled by separate genes/genetic mechanisms. This independence amongst the genes sheds light onto the genetic basis and evolutionary path of the selfing syndrome. First, it demonstrates that the genes that control these traits are unlinked, meaning they do not show a tendency to be inherited together. This finding shows that there are many genes responsible for pollen production. The observed distributions are consistent with many loci affecting the floral traits, each with small to moderate effects, which is similar to the findings in Mimulus, or monkeyflowers20-21 and Leptosiphon, or Linanthus22-23. It has been shown that the genes that affect pollen production, nectar volume, stigma length, corolla length, and corolla width are probably controlled by many genes. More research into the genes that are involved with selfing would provide great insight onto how to control plant reproduction, which is very useful for food security as it would increase crop yield and a more secure food supply.


[1] Raduski, A. R. et al. Evolution. 2012, 66, 1275–83.

[2] Stebbins G.L. Am. Nat, 1957, 91, 337–354.

[3] Baker, H.G. Evolution. 1955, 9, 347-349.                                                                                              

[4] Lloyd, D.G. Int. J. Plant Sci. 1992, 153, 370-380.

[5] Darwin, C.R. The Effects of Cross and Self-fertilization in the Vegetable Kingdom John Murray: London, 1876.

[6] Fisher, R. A. A. Eug. 1941, 11, 53–63.

[7] Igic, B.; Busch, J. W. New Phytol. 2013, 198, 386-397.

[8] Howe, O.W.; Oliver L.R. Weed Sci. 1987, 35, 185–193.

[9] Stoller E.W. et al. Rev. Weed Sci. 1987, 3, 155–181.

[10] Baucom, R. S. et al. Heredity. 2011, 107, 377-385.

[11] Culpepper, A.S. Weed Technol. 2006, 20, 277–281.

[12] Olsen, Kenneth M. et al. Genetics. 2006, 173, 975-983.

[13] Rausher M.D.; Chang S.M. Am Nat. 1999, 154, 242–248.

[14] Ornduff, R. Taxon. 1969, 18, 121-133.

[15] Wilcock. C. C. J. Trop. Ecol. 1987, 3, 279-280

[16] Foxe, J. P. et al. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 5241-5245.

[17] Sicard, A. et al. Plant Cell. 2011, 23, 3156-3171.

[18] Brunet, J. Trends Ecol. Evol. 1992, 7, 79-84.

[19] Runions, C. J.; Geber, M. A. Am. J. Bot. 2000, 87, 1439-1451.

[20] Lin, J. Z.; Ritland, K. Genetics. 1997, 146, 1115-1121.

[21] Fishman, L. et al. Evolution. 2002, 56, 2138-2155

[22] Goodwillie, C. et al. Evolution. 2006, 60, 491-504

[23] Georgiady, M. et al. Genetics. 2002, 344, 333-344.