**11. Appendix**

throughout pollen development, this leads to decreased numbers of viable pollen at anthesis. Larger pollen numbers could be obtained in plants with larger anthers. Anther size is a trait that has been used for selection of cold tolerance in rice [297] and the growth hormone gibberellic acid plays an important role in controlling stamen development [298]. Elucidating the mechanism of interrupting or pausing pollen development under unfav‐ ourable conditions may also provide useful information about avoiding stress damage. Understanding these mechanisms will require a better understanding of the signals driving gametophyte development *per se*. The haploid genome of the male gametophyte is derived from the sporophyte, but very little is known about its functionality in regulating pollenspecific development and metabolism. Achieving this challenge is now within reach, thanks

to sensitive new-generation transciptome analysis techniques [29, 71, 299].

728 Abiotic and Biotic Stress in Plants - Recent Advances and Future Perspectives

nisms in sporophyte-gametophyte signalling is also mounting [307–309].

more detailed investigations.

**•** The importance of air relative humidity in pollen development has so far been grossly underestimated. The growing area of staple crops such as cereals is increasingly extending into environments that require different adaptations of pollen development. For instance, tropical rice is grown in temperate climate zones and temperate climate wheat is grown in humid tropical environments [310–312]. Air humidity and climatic conditions modifying atmospheric humidity (rain, fog, cold, heat and drought) have a dramatic effect on plant species producing orthodox and recalcitrant pollen, causing asynchrony and reducing pollen number and fertility. The dynamics of water relations and osmotic regulation in pollen grains and their interactions with the environment are research topics that need urgent attention. Adapting the breeding system of crop species (self- versus cross-pollina‐ tion) may offer opportunities for improved protection of pollen during dispersal, but the trade-offs between chasmogamy and cleistogamy in terms of abiotic stress tolerance require

**•** It is important to understand the signalling mechanisms between mother plant and male gametophyte. Some crucial steps in pollen development (meiosis, tapetal activity and anther dehiscence) are clearly under sporophytic control. The high sensitivity to abiotic stresses of the meiotic, young microspore and anthesis stages indicates that sporophytic signals are critical in controlling male gametophyte development. Stress-proofing crop plants may therefore have to start by understanding the sporophyte signals (sink-source relationships, carbohydrate and hormone signalling, control of PCD during tapetum degeneration and anther dehiscence). It has been known for some time that treatments with one stress or with the stress hormone abscisic acid (ABA) can improve tolerance to another stress – a process called stress 'hardening' or 'priming' [300–305]. More recent studies in rice have shown that stress treatments at the vegetative stage can affect abiotic stress tolerance during flowering and reactive oxygen species (ROS) signalling could play a role in this sporophytic signalling event [306]. But evidence for involvement of genomic imprinting and epigenetic mecha‐



