188
G. R. POTTS AND 0. P. VICKERMAN
nests of a population of great tits (Parus major L.) was inversely
proportional to the general abundance of small mammals. Pearson
(1966) has demonstrated that predation was a very important cause of
mortality in a vole population and has taken these arguments a step
further by suggesting that carnivore predation is an essential part of the
regular cycles of abundance of lemmings, Microtw and other microtines.
Considerable evidence has recently accumulated to show that the
ten-year cycle of grouse in the North American boreal forest ecosystem
is produced by a “predator shift t o snowshoe hares which increases
grouse survival and hence numbers as hare densities rise; predator
densities eventually reach levels where frequencies of random or
fortuitous contact with grouse are sufficient to again depress survival
and hence grouse populations” (Dr. Lloyd B. Keith, Univ. Wisconsin).
So far in this paper we have only considered the cereal ecosystem
processes but now we have extended the work to grassland. We can
already conclude that cereal ecosystems are modified grassland
ecosystems; indeed, as we have seen earlier in this paper, many species
migrate into cereals from grasses. The invertebrate fauna associated
with traditional ley farming areas appears to have a higher predatorprey ratio than other arable systems. Indeed, much of the present
stability of the cereal fauna may result from its association with the
relatively stable and complex grassland fauna. Relatively little pesticide
is yet used on grassland, and Strickland (1965) considered that although
pest control might show an average increment in yield of 5-10yo,
including new grass, a 20-25y0 yield increment would probably be
needed to tempt growers to take active control measures. Recent
small-scale trials (Henderson, 1971) have shown that the yield of grass
can be increased by as much as 30% with the application of pesticides.
It would surely be prudent to investigate the structure of these
ecosystems before adopting the large-scale use of more pesticides and
further intensification of cropping. We hope that the results of our
future studies will help to provide data which will allow further increases
in agricultural productivity without unnecessarily violating sound
ecological processes and without undue environmental costs.
A C K N O W L E D G E M E N T S
This paper is dedicated to the late Christopher Hunt, who made it all
possible.
We are particularly grateful for the continued encouragement and
willing help of all kinds from everyone at North Farm, especially
J. G . Goring, B. P. W. Morris, N. C. M. Brine, F. P. Allen, R. E. Largen
and P. B. Adam-Smith, and in the rest of our study areas, especially
G. R. POTTS AND 0. P. VICKERMAN
nests of a population of great tits (Parus major L.) was inversely
proportional to the general abundance of small mammals. Pearson
(1966) has demonstrated that predation was a very important cause of
mortality in a vole population and has taken these arguments a step
further by suggesting that carnivore predation is an essential part of the
regular cycles of abundance of lemmings, Microtw and other microtines.
Considerable evidence has recently accumulated to show that the
ten-year cycle of grouse in the North American boreal forest ecosystem
is produced by a “predator shift t o snowshoe hares which increases
grouse survival and hence numbers as hare densities rise; predator
densities eventually reach levels where frequencies of random or
fortuitous contact with grouse are sufficient to again depress survival
and hence grouse populations” (Dr. Lloyd B. Keith, Univ. Wisconsin).
So far in this paper we have only considered the cereal ecosystem
processes but now we have extended the work to grassland. We can
already conclude that cereal ecosystems are modified grassland
ecosystems; indeed, as we have seen earlier in this paper, many species
migrate into cereals from grasses. The invertebrate fauna associated
with traditional ley farming areas appears to have a higher predatorprey ratio than other arable systems. Indeed, much of the present
stability of the cereal fauna may result from its association with the
relatively stable and complex grassland fauna. Relatively little pesticide
is yet used on grassland, and Strickland (1965) considered that although
pest control might show an average increment in yield of 5-10yo,
including new grass, a 20-25y0 yield increment would probably be
needed to tempt growers to take active control measures. Recent
small-scale trials (Henderson, 1971) have shown that the yield of grass
can be increased by as much as 30% with the application of pesticides.
It would surely be prudent to investigate the structure of these
ecosystems before adopting the large-scale use of more pesticides and
further intensification of cropping. We hope that the results of our
future studies will help to provide data which will allow further increases
in agricultural productivity without unnecessarily violating sound
ecological processes and without undue environmental costs.
A C K N O W L E D G E M E N T S
This paper is dedicated to the late Christopher Hunt, who made it all
possible.
We are particularly grateful for the continued encouragement and
willing help of all kinds from everyone at North Farm, especially
J. G . Goring, B. P. W. Morris, N. C. M. Brine, F. P. Allen, R. E. Largen
and P. B. Adam-Smith, and in the rest of our study areas, especially