Field Tests for Pollen Viability; a Comparative Approach

Many tests for pollen viability have been proposed. The ability to determine this reproductive state in the field can be very important in botanical research as well as for horticultural purposes. After reviewing a large number of tests, we compared four methods of determining pollen viability. Each of these methods can easily be carried out in the field. Viability tests were carried out on 17 species from Israel and were compared to in vitro germination trials. No one test worked well on all species. The MTT stain was determined as the most effective for a wide number of species, but tests for individual species should be conducted before final determinations are made. INTRODUCTION Information about the ability of pollen grains to germinate when they reach the stigmas of flowers of their own species is valuable both for horticultural purposes and general botanical research. Viability tests provide a means of assessing the potential of pollen to germinate on the stigma. Stone, Thomson and Dent-Acosta (1995) pointed out the need for assessing the viability of pollen used in hand-pollination experiments. In their survey of 283 papers they found that this had been done infrequently. The literature differs on what a “viability test” actually measures. The term viability has been defined as "having the capacity to live, grow, germinate or develop" (Lincoln, Boxshall, and Clark, 1982). However, the term viability has been used to describe the germination of pollen grains on the stigma (Morse, 1987; Preston, 1991; Vaughton and Ramsey, 1991; Niesenbaum, 1992), in vitro germination of pollen grains (Shchori, Goren and Ben-Jaacov, 1992; Beardsell, Knox and Williams, 1993; Lingren et al., 1995), the results of various staining procedures (Bernhardt, Knox, and Calder, 1980; Becker and Ewart, 1990; Mione and Anderson, 1992; Nyman, 1992), and seed set following pollination (Smith-Huerta and Vasek, 1984). Our use of the term viability is based on the definition given by Lincoln, Boxshall, and Clark (1982). Many different methods have been suggested to determine the viability of pollen (Dafni, 1992; Kearns and Inouye 1993). Vital stains, such as cotton blue in lactophenol and lissamine green, have been used by many workers in the past. Although these stains can distinguish between sterile and partially sterile plants, they do not document the loss of viability in pollen grains very well (Hauser and Morrison, 1964; Ockendon and Gates, 1976; Heslop-Harrison et al., 1984; Knox, 1984). Artificially pollinating flowers and assessing seed production might seem to be the most accurate means of determining pollen viability, but the amount of pollen used per stigma (Cruzan, 1986; Young and Young, 1992), the complications of fruit abortion (Stephenson, 1981; Primack, 1987), and the effect of environmental conditions such as temperature, humidity and water availability (Chang and Struckmeyer, 1976; Schoper, Lambert and Vasilas, 1986; Burgos, Egea, and Dicenta, 1991; Gudin, Arine, and Pellegrino, 1991; Shivanna et al., 1991) can alter the results of pollination trials. Additionally, waiting for seed maturation is time consuming and may mean the flowering season for that species is over by the time the results are known. Seed counts are also more qualitative than quantitative in many species because a limited number of pollen grains may be all that is necessary for full seed set (Shivanna and Johri, 1985; Stone, Thomson, and Dent-Acosta, 1995). Also, tests for self-incompatibility mechanisms should be done if seed production is to be a measure of viability. Staining techniques also give no information about incompatibility. In vitro germination, while it does provide a quantitative measure, is time consuming and