STUDIES ON THE CEREAL ECOSYSTEM
135
and crop senescence and the rate of increase of apterae during that
time. I n 1970 the peak density of Metopolophium dirhodum was
measured in 15 fields of spring-sown barley. The peak density ( y ) was
found to be described by the equation: y = 58x+230 (P
60% of the variation in peak numbers-with no allowance for error in
the remaining 40%. The principal factor appeared to be operating
during the early part of the essentially logistic curve and therefore
when numbers were small. It had little or nothing to do with the size
of the immigration of alates into the crop. For example, the immigration of Gramineae-infesting alatae in 1971 was about eight times that of
1970-but apterous populations were far larger in 1970 than in 1971.
The alate immigration was even more productive in 1969.
3. Factors governing the rate of increase of apterous aphids
I n 1969 the senior author noticed that there were often fewer aphids
in cereal fields which were weedy compared with weed-free ones, but
there was insufficient sweep-net data to pursue the matter. In 1970
the same effect was again noticed, this time with the vacuum insect net,
but the mean peak density of aphid apterae in weedy areas was only 30%
lower than in the clean fields-and the general abundance of other
insects during the immigration and early build-up phase (early June)
was also important. In that year the general abundance of insects
in early June was determined mostly by the weed density, and in
spring-sown barley very high rates of increase of aphids did not occur
when other insects were abundant in early June (Potts, 1970b).
However, it seemed likely that neither the density of weeds nor the
early abundance of arthropods was causally related to the rate of
increase of the aphid apterae, because the correlation in wintersown cereals was of a different kind. In the latter caae the apterae
other arthropod ratio was better correlated to outbreak tendency than
the number of arthropods alone.
A different approach was therefore tried in 1971; this resolved most
of these diiiiculties.
4. The relationship between faunal diversity and the number of apterous
aphids
An index of faunal diversity ( u ) was calculated for each sample
obtained with the vacuum insect net. The index of diversity ( u ) used
was again that of Williams (1944, 1946) (see p. 120). The diversity
index (a) of Williams is particularly useful since it is independent of
sample size and is a property of the community. The greater the value
135
and crop senescence and the rate of increase of apterae during that
time. I n 1970 the peak density of Metopolophium dirhodum was
measured in 15 fields of spring-sown barley. The peak density ( y ) was
found to be described by the equation: y = 58x+230 (P
60% of the variation in peak numbers-with no allowance for error in
the remaining 40%. The principal factor appeared to be operating
during the early part of the essentially logistic curve and therefore
when numbers were small. It had little or nothing to do with the size
of the immigration of alates into the crop. For example, the immigration of Gramineae-infesting alatae in 1971 was about eight times that of
1970-but apterous populations were far larger in 1970 than in 1971.
The alate immigration was even more productive in 1969.
3. Factors governing the rate of increase of apterous aphids
I n 1969 the senior author noticed that there were often fewer aphids
in cereal fields which were weedy compared with weed-free ones, but
there was insufficient sweep-net data to pursue the matter. In 1970
the same effect was again noticed, this time with the vacuum insect net,
but the mean peak density of aphid apterae in weedy areas was only 30%
lower than in the clean fields-and the general abundance of other
insects during the immigration and early build-up phase (early June)
was also important. In that year the general abundance of insects
in early June was determined mostly by the weed density, and in
spring-sown barley very high rates of increase of aphids did not occur
when other insects were abundant in early June (Potts, 1970b).
However, it seemed likely that neither the density of weeds nor the
early abundance of arthropods was causally related to the rate of
increase of the aphid apterae, because the correlation in wintersown cereals was of a different kind. In the latter caae the apterae
other arthropod ratio was better correlated to outbreak tendency than
the number of arthropods alone.
A different approach was therefore tried in 1971; this resolved most
of these diiiiculties.
4. The relationship between faunal diversity and the number of apterous
aphids
An index of faunal diversity ( u ) was calculated for each sample
obtained with the vacuum insect net. The index of diversity ( u ) used
was again that of Williams (1944, 1946) (see p. 120). The diversity
index (a) of Williams is particularly useful since it is independent of
sample size and is a property of the community. The greater the value