Influence and implications of the reproductive variability and visitation of insects in four cultivars of <em>Mangifera indica</em> in Puerto Rico

Abstract

In this study, I explored the possible vulnerabilities of mango <em>Mangifera indica</em> L. (Anacardiaceae) crops related to pollination processes. A main question that was asked was what are the potential vulnerabilities of mango yield at the pollination stage and what are the implications of these influences to mango production in Puerto Rico? To address these questions, I studied three mango cultivars of Floridian origin (Keitt, Kent and Tommy Atkins) and West Indies (Julie) origin all of which are part of a larger mango germplasm managed by the Juana Diaz Experimental Station of the University of Puerto Rico in Mayagüez. Through various studies and building upon a rich literature on mango reproduction, this work tackled four main objectives:1) Evaluate the variation of floral and inflorescences characteristics of these cultivars, 2) Conduct experimental studies to evaluate the dependency of these cultivars on local pollinators and the mating systems these cultivars. 3) Monitor pollinator communities and fruit yields over three flowering seasons evaluating the relationships between pollinator diversity and abundance with fruit yields, 4) Conduct field experiments and observations to estimate the pollination efficiency and effectiveness of dominant pollinators. A minor objective was to compare estimates of pollinator diversity and abundance using different field methods (net sweeping and video cameras). Results showed that there is significant difference across cultivars in variables related to inflorescence size (width, length), architecture, flower production and flower sex ratios but that these differences were not necessarily consistent across years, nor across cultivars. Flowers of <em>M. mangifera</em>, reflect color in the UV region, but these patterns are qualitatively different in Keitt which also shows qualitative differences between hermaphroditic and male flowers reflectance in both the UV and visible regions. Mango flower produce more sucrose that glucose and fructose but different cultivars presented different sugar profiles based on the relative percentage of sugar contents. Pollen viability was high for all cultivars, but pollen germination exhibited an optimal temperature that varied depending on the cultivar. Likewise, for all cultivars, pollen from male flowers exhibited higher germination percentages compared to pollen from hermaphroditic flowers at temperatures above optimal values but at temperatures below the optimal, they had a tendency for lower germination percentages relative to hermaphroditic flowers but this pattern was only significant for Kent. Open natural pollinations tended to produce more fruit and yielded seeds with faster development times than artificial pollinations in three of the four cultivars (Julie, Tommy Atkins and Keitt). They also resulted in faster seed germination in two cultivars (Julie and Tommy Atkins). The combined results suggest that not only animal-mediated pollination is needed for these cultivars but that this mode of pollination also results in improved fruit yields, fruit, and seed traits. Over the three years, plants were visited by a combined total of 50 insect species with Diptera being also the most abundant followed by Hymenoptera, Coleoptera and Lepidoptera. The relative abundances of insect communities changed after the passage of Hurricane Maria over the island of Puerto Rico but only one field (hosting Kent) experienced significant species richness declines in 2018 following the hurricane events. Two of the most dominant insects, <em>Palpada vinetorum</em> F. (Diptera: Syrphidae) and <em>Apis mellifera </em> L. (Hymenoptera: Apidae), showed a “reduction-recovery” pattern for in the period of 2018-2019 but not so for <em>Cochliomyia minina</em> (Diptera: Calliphoridae) which was very abundant in 2018 in three out of four cultivars but then returned to pre-hurricane levels in 2019. In 2017, the trees exposed to higher richness and insect abundances experienced higher yields regardless of cultivars but these relationships when present were often weaker in 2018 (post-hurricane) and 2019. Also, not all fields were equally successful at attracting the same levels of diversity and abundance of insects. The visitation rates were studied and estimated with video cameras for each cultivar in 2018 and 2019. When visitation rates were pooled, I found no significant differences in the average total visitation rate (pooled values of all insects) between 2018 and 2019 in any of the cultivars, but the visitation rates of individual species showed different across years. In 2019, a pollen deposition experiment showed no significant differences among insect species in the mean pollen load deposited on stigmas on virgin flowers in any of the cultivars. When both components of pollination effectiveness (pollen deposition and visitation rate in 2019) were combined, there were significant differences in pollinator effectiveness only for Julie where <em>P. vinetorum</em> was expected to deposit more pollen grains than the other species. However, all other pollinators were equally effective in the remaining cultivars. All pollinators showed significant correlations between their visitation rates and fruit yield in at least one cultivar but most significant correlations appeared in 2019 and only one in 2018. Insect diversity and pooled visitation rates estimates from video feeds were not correlated with fruit yields. However, the diversity estimates from video feeds when compared to estimates using net sweeping are significantly lower which suggests that this method as implemented greatly underestimates diversity estimates and that may explain the lack of association between global diversity and fruit yields. As expected, mango reproductive traits related to flower and inflorescence production is highly variable within and between cultivars. During the time period of this study the role of this variability on fruit production in mango seemed secondary to the role of pollinator diversity and abundance in is their relation to agricultural yields. Pollinators are equally effective at depositing pollen and any spatial-temporal fluctuations in the abundance of individual pollinator species is likely to be compensated by the action of other insect species in the pollinator community. Results from pollination experiments suggest that open pollinations may result from cross-pollinations across cultivars which needs further exploration as these result in desirable outcomes in some cultivars (e.g. increased fruit production and faster germination traits) but added variation from open pollinations may also result in added variation in other fruit qualities (e.g. taste, color) not explored in this work. From a management perspective maintaining adequate levels of pollinator diversity would be an important requirement for the stability of fruit yields in this crop system especially in the face of future hurricane events. With expected increases in temperature for the Caribbean basin, a concern would be the potential negative effects of reduced pollen germinability on fruit production.