The struttmc and clicmistry ot iinthcr dehiscence m LycDpcrsicoii csculcutiim (Solanaceiie) has been examined usinj,' a iaTif,'e of microscopic teclmiqLics. No single e\ent or process has been identilnecl as bciny solely responsible for the release of the pollen grains, instead, an integrated programme of development across a number of tissues appears to be involved. The earliest events in this process are changes in tlie intersporangial septa and growth of specific epidermal cells. Large numbers of calcium oxalate (druse) crystals accumulate in the septa and may play a part in the weakening and e\entual enz\mic dissolution of this cell layer. The differential growth of epidermal cells fulfils two roles; firstly it ser\es to define the point of rupture or stomiutn and, secondly the enlarged cells generate and transmit the force recjuired to disrupt the unenlarged stomial cells. Adjacent loculi are unified within tbe anther first by the enzymic degradation of the crystal-filled septal cells, which is followed by similar degradation of the remaining connective tissue cells and the eventual rupture of the remnants of the tapetal walls.Once the epidermal cell system has been established, and adjacent loculi unified, the endotheciutn develops in the anther wall, but only in the region adjacent to the stomium in the distal third of the anther. Opetiing of the anther is achie\ ed by a combination of the mechanical forces generated by the enlarged epidermal, and possibly the endothecial, cells operating on the previously weakened stomial cells, which at the time of rupture have started to desiccate. Following rupture of the atither wall, the stomium is transfortned from a slit to a wide-mouthed pore via the ordered desiccation and contraction of the endothecial and surrounding tisstie. .Althogh these final stages of dehiscence show some sensiti\ity to ambient \'apour pressure differences, there is little dotibt that the process of dehiscence is not purely a desiccatory one, indeed only specific locahsed areas of the antlier degenerate and dehydrate, whilst the bulk of the atither tisstie temaitis turgid and metabolically acti\e.This multicotnponential type of anther dehiscence is discussed in the perspeetive of other models proposeci to explain anther opening in tnembers of the Solanaceae, as well as systems suggested for other plants which clearly po.ssess ver>' different methods of dehiscence.

SUMMARYWater tracers have been used to demonstrate the continued functioning of the anther filament during dehiscence in Lycopersicon esculentum Mill. Since Lycopersicon possesses neither nectar nor nectaries dehiscence cannot be related to sugar secretion, as has been demonstrated for other plants. The anthers seem structurally adapted for water conservation and the only transpirational loss of the flower appears to be through the petals. Transpiration from the anthers themselves thus appears not to be involved in regulating dehiscence. Anther opening is preceded by dehydration of the locule and circumstantial evidence indicates it to be an active process, with water being exported through the filaments to the petals along an osmotic gradient generated by starch/sugar interconversion. Measurement of water potential differentials between the petals and anthers points to the mechanism driving this export of water from the anthers. Interestingly, turgor pressure is maintained in the majority of the anther tissues until senescence, remaining remarkably constant against a background of dramatic changes in osmotic potential. These data, combined with the observation that the hydration level of the anthers falls only to 70 % at senescence, indicates that dehiscence cannot primarily be a desiccatory process. Some domains within the anther do desiccate, but these are strictly localized. Thus, while hydration levels of living tissues remain independent of the environment, those observed to undergo protoplast degeneration and wall collapse exhibit environmentally-linked hygroscopic absorption. Dehiscence thus emerges as an orchestrated programme of structural and physiological events leading to the desiccation of specific domains of the anthers. These findings are discussed in terms of current models for anther dehiscence in fiowering plants.

Plants of Ocimum basilicum L. grown under glass were exposed to short treatments with supplementary UV-B. The effect of UV-B on volatile essential oil content was analysed and compared with morphological effects on the peltate and capitate glandular trichomes. In the absence of UV-B, both peltate and capitate glands were incompletely developed in both mature and developing leaves, the oil sacs being wrinkled and only partially filled. UV-B was found to have two main effects on the glandular trichomes. During the first 4 d of treatment, both peltate and capitate glands filled and their morphology reflected their 'normal' mature development as reported in the literature. During the following days there was a large increase in the number of broken oil sacs among the peltate glands as the mature glands broke open, releasing volatiles. Neither the number of glands nor the qualitative or quantitative composition of the volatiles was affected by UV-B. There seems to be a requirement for UV-B for the filling of the glandular trichomes of basil.

The protective layer between the cell wall and plasmalemma of xylem parenchyma cells has variously been suggested to be involved in protection of the protoplast from attack by autolytic enzymes from neighbouring, dying cells, tylose formation, deep supercooling of xylem, and strengthening of the pit. None of these ideas has universal application to all species in which parenchyma cells possess a protective layer. It is proposed instead, that the protective layer is primarily laid down in order to preserve apoplastic continuity around the protoplast of a lignified cell, bringing the entire plasmalemma surface, and not just that part of it in contact with the porous pit membrane, into contact with the apoplast. If this is so, then other functions may be coincidental, or have arisen secondarily.

No abstract