Effect of breaks on sugarcane growth: relations between glasshouse and field studies

Abstract

The effects of varying types and duration of breaks (other crops, pastures or bare fallows) and soil fumigation on subsequent growth of sugarcane crops were investigated in a series of five field experiments in Australia. Results from the first series of test plantings have shown breaks can produce significant effects on sugarcane growth and yield - at least as large as those from soil fumigation. These responses have generally been characterised by marked differences in shoot establishment and early growth in the first 70-90 days. Glasshouse studies conducted using soil from the same sites showed significant (P<0.05) effects of break history on early shoot growth (35-45 days) at four of the five sites. Positive responses to soil fumigation were also recorded at three of the five sites. Fumigation with methyl bromide produced minimal effects on early growth in soil that had been bare fallowed (-15% to 2%), or soil that had been fumigated in the field (-4% to 29%). Fumigation of soil from continuous cane treatments typically increased early cane growth by 25-30%. Most breaks had effects intermediate between the control and the bare fallow - both in early growth and response to fumigation. Shoot dry matter in unfumigated soil in the short-term glasshouse studies was strongly correlated with field shoot numbers at 70-90 days after planting and with dry matter production after 8 months. Correlations were either much weaker, or had disappeared entirely, by final harvest due to unexplained differences in growth rates among treatments during the final 4 months. Growth in fumigated break soil in the glasshouse was not correlated significantly with growth in the field, indicating that biotic factors were associated with these early growth differences in the field. These findings suggest that short duration pot experiments may be a useful research tool to study the biotic effects on early growth of sugarcane. This will provide more rapid experimental cycles and significant cost savings compared to field studies. However, other evidence suggests that results from such studies will not necessarily be a good indicator of ultimate crop yield.