PlantLife Volume 58.8, December 2024. An update on the pollination biology of Barberetta aurea

 

An update on the pollination biology of Barberetta aurea (Haemodoraceae).

Jeremy Midgley

Until recently, Barberetta aurea was relatively unknown because it is quite rare and is somewhat particular in its habitat (shady mid-altitude KZN stream sites). This was until Graham Grieve showed Steve Johnson, Nicci Illing and me some sites in the Weza-Ngele area. Not only were these great sites, but there were abundant insect pollinators visiting the flowers, giving us a chance to figure out the reproductive biology of this rare and interesting plant species. Nicola Illing and Caroline Robertson have introduced B. aurea (PlantLife Volume 53.4, July 2022) and here I report some of our subsequent findings.

Barberetta aurea

Barberetta aurea flowers are enantiostylous: there are two morphs, one with the stigma deflected to the left and one to the right and any particular plant only has one type of deflection. This is referred to as dimorphic enantiostyly. Monomorphic enatiostyly is where both left and right-handed morphs occur on the same individual and is much more common than dimorphic enantiostyly. B. aurea flowers have three anthers: two face in the opposite direction to the stigma and one in the same direction. In some enantiostylous plants, such as the monomorphic enatiostylis Dilatris (also Haemodoraceae), one anther becomes a larger centrally placed feeding anther with two smaller, symmetric, side anthers which are the pollinating anthers. Pollen can be a reward for some visiting insects, such as carpenter bees. A feeding anther is a larger more conspicuous anther which attracts insects but then plays less of a role in pollination.

Fewer than ten species globally are dimorphically enantiostylous, with Wachendorfia in the Cape and Cubanicula (Cuba), both also in the Haemodoraceae, as two of the others. A hypothesis for the evolution of this unusual floral structure is that it causes out-crossing due to the asymmetric placement of pollen on one side of the body of the insect pollinator. The complimentary morph is required to remove pollen. In a right-handed B. aurea flower most pollen would be placed on the left of the insect (where the two pollinating anthers are) and therefore a left-handed flower would be most likely to receive this pollen. Dimorphic enantiostyly would therefore cause cross-pollination rather than self-pollination. Most plants are hermaphrodites with anthers and stigma in the same flower and so there is a danger of self-pollination. Plants try to avoid self-pollination, for instance by keeping anthers and stigma apart in space or time, such as with anthers being far apart from the stigma or anthers ripening before the stigma. Enantiostyly may be another mechanism to prevent selfing in hermaphrodite plants.

We measured B. aurea morph ratios, and they were remarkably close to 1:1 left versus right-handed in B. aurea populations. This even spread suggests that most pollinations are between the different morphs rather than within morphs, which would produce clumps of only one morph. The most common pollinators were nectar-feeding syrphid flies (hoverflies) and interestingly, most of the pollen appeared to be placed on their wings not their bodies. This is further support for the idea that enantiostyly is associated with wing pollination because a straight stigma would not touch the wings to obtain pollen - the stigma would need to be deflected one way or the other to reach the wings rather than the body.  Wing pollination has already been confirmed for Wachendorfia. We crossed (e.g. pollen from left flowers placed on right-handed flowers on the same plant and on different plants) and selfed (e.g. left on left from different individuals as well as within an individual) and found no difference in seed set. We also bagged flowers to prevent access by pollinators to determine whether the plant could set seeds in the absence of a pollinator visit.

We also put fluorescent dye on the anthers of plants to see where it got placed on the syrphid flies and whether it moves on to stigmas of flowers they subsequently visited. What have we learned about enantiostyly and the reproductive biology of B. aurea? Firstly, we found syrphids are crucial as no seeds are set without pollinator visits: it cannot self without a pollinator.  Pollen tracking and balanced morph ratios indicate most pollinations are between morphs (e.g. left pollen to right-handed flowers) rather than within morphs. So enantiostyly in B. aurea really does seem to work to increase out-crossing (that is between the different morphs which only happens between two separate plants because each plant only has one type of flower) and therefore reduce selfing.  

Pollen tracking confirmed that pollen was attached to syrphid wings and that it could move from wings onto the very small stigmas of B. aurea. There is a definite link between enantiostyly and wing pollination.

What are some remaining questions? The genetics of enantiostyly still needs to be analysed. Is it a simple genetics system such as with a simple one dominant and one recessive gene? Does genetic analysis of seeds show that within plant selfing really is as low as we think it is?  What are the costs and benefits of dimorphic enantiostyly, as in B. aurea, versus monomorphic enantiostyly in Dilatris? The answer probably lies between allowing some selfing from pollinator visits between different flowers on the same plant, yet maximising pollen receipt from visits to flowers on different plants. B. aurea still has one anther on the same side as the stigma. This would provide opportunity for some selfing. B. aurea has not taken the extreme step of all anthers being on the opposite side of the stigma; this situation which would prevent self-pollination, but which would make life difficult in small populations without both morphs.

We published these results in the Annals of Botany 132: 1107–1118, 2023 and below are some of Steve Johnson’s great illustrations. In A and B are left and right-handed B. aurea flowers based on where the stigma is facing and with fluorescent dye on anthers. In C and D, a syrphid fly (Melanostoma species) has approached and landed on a flower, where it is probing for nectar. Florescent pollen can be seen on wings and stigma in E and F.

About the author: Jeremy Midgley is an Emeritus Professor in the Department of Biological Sciences at UCT. He enjoys the ecological and evolutionary natural history riches of South Africa, especially where these can be used to test questions in biology.