Losing the thread
The evidence for global climate change is now incontrovertible. Global
temperatures have increased, and are continuing to do so. This is, of course, a
matter of serious concern at several levels. In agriculture and horticulture
the concerns include the possibility that conditions may become more difficult
for crop growth. Within any one species, different aspects of growth and
physiology are differently affected by high temperature and thus the most
vulnerable process will be different in different species. In cotton, for
example, boll formation, retention and yield are affected by high temperature,
as described by a US research team, Kakani et al. (pp. 59–67).
The authors have identified pollen germination and growth as being especially
sensitive to temperature. Typically, pollen is released in the morning before
the temperature reaches its maximum. However, pollen tube growth is relatively
slow and is in progress during the heat of the day with fertilization occurring
between 12 and 24 hours after pollen germination. The authors have worked with
12 cotton cultivars, exhibiting a range of tolerances to stresses such as
drought and high temperature. They collected ripe pollen from these cultivars
and observed pollen germination and growth in vitro on artificial media
at a range of different temperatures. Minimum, optimum and maximum temperatures
for germination and tube growth varied between cultivars and, in general, the
temperature optima for tube growth were lower than for germination. For both
processes, the breadth of the range of ‘permissive’ temperatures varied.
Further, at their optimum temperatures, the cultivars showed a good deal of
variation in tube growth rates. Based on principal component analysis, the
cultivars could be grouped according to their temperature tolerance, with the
most high-temperature-tolerant group having optima for pollen germination above
32 °C. Interestingly, there was no correlation between temperature tolerance of
pollen and membrane thermostability, emphasizing again that different processes
are differently vulnerable.
Professor J. A. Bryant
University of Exeter, UK
j.a.bryant{at}exeter.ac.uk
