266
R. DAUBENMIRE
control! Normal litter cover had been restored by the end of the third
post-burn s e w n . In this Name region Elwell (1941) considered burning
detrimental, and mowing plus raking a desirable alternative in the pmvention of an accumulation of unused forage.
Where artificial stands of perennial grass are mtablished for seed
production it has been found that peak yields are obtained in 3-4 years,
following which there is a decline with the fields becoming unprofitable
after 4 4 years production (Canode, 1966). Heavy applications of nitrogen sometimes offset the decline, and in some situations burning or
mowing are effective. For example, mechanicd removal of straw and
stubble after harvesting seed of Poa pratensis and F m t w rubra was
found to have the same effect as burning in increaaing subsequent seed
yield (Pumphrey, 1965). Canode (1965) found that running fire over a
stand of Agropyron inter medium after harvest reduced plant density
but increased seed yield the next year. If half the plants are killed
mechanically the same result follows, with the combination of these
treatments having cumulative effect in raising seed yield.
CONCLUSION
Sufficient quantitative data on fire ecology in grasslands have accumulated in recent years to warrant a number of generalizations.
The heat generated aa herbaceous cover burns is determined not only
by the amount of fuel, its dryness, and its initial temperature, but also
by the direction of spread of fire in relation to wind. Backfires, those
moving against the wind, are hotter, and produce their maximum temperatures nearer the ground, in comparison with head fires. Soil surface
temperatures as measured in African savannaa are relatively high (up
t o 72OoC), but elsewhere in grasslands they have seldom risen much
above 100°C, a level much lower than is usual in forest fires. From the
surface downward, the temperature gradient is 80 steep that direct
alteration of soil structure or chemistry by fire is inconsequential in
grasslands.
After a fire, and before a new herbaceous cover develops, the blackened and unshaded soil is appreciably warmer, at least during daytime,
in comparison with unburned areaa. Presumably in consequence of thiE
the foliage usually appears earlier in the firat poet-burn season, and
sometimes the onset of flowering is advanced. Normal litter cover ia
restored in 2-5 years. Whether humus increaaes or decreaaes, in ~ 0 1 ~ 8 8 -
quence of repeated grass fires, or whether erosion is accelerated or not
varies from place to place, but these seem constant for a given region
and vegetation type. I n perennial grasslands soils am drier after a burn.
Plant ash resulting from f i n g graaa cover seems too m n t to be significant, which contraate markedly with the appreciable rekwe of fertility
R. DAUBENMIRE
control! Normal litter cover had been restored by the end of the third
post-burn s e w n . In this Name region Elwell (1941) considered burning
detrimental, and mowing plus raking a desirable alternative in the pmvention of an accumulation of unused forage.
Where artificial stands of perennial grass are mtablished for seed
production it has been found that peak yields are obtained in 3-4 years,
following which there is a decline with the fields becoming unprofitable
after 4 4 years production (Canode, 1966). Heavy applications of nitrogen sometimes offset the decline, and in some situations burning or
mowing are effective. For example, mechanicd removal of straw and
stubble after harvesting seed of Poa pratensis and F m t w rubra was
found to have the same effect as burning in increaaing subsequent seed
yield (Pumphrey, 1965). Canode (1965) found that running fire over a
stand of Agropyron inter medium after harvest reduced plant density
but increased seed yield the next year. If half the plants are killed
mechanically the same result follows, with the combination of these
treatments having cumulative effect in raising seed yield.
CONCLUSION
Sufficient quantitative data on fire ecology in grasslands have accumulated in recent years to warrant a number of generalizations.
The heat generated aa herbaceous cover burns is determined not only
by the amount of fuel, its dryness, and its initial temperature, but also
by the direction of spread of fire in relation to wind. Backfires, those
moving against the wind, are hotter, and produce their maximum temperatures nearer the ground, in comparison with head fires. Soil surface
temperatures as measured in African savannaa are relatively high (up
t o 72OoC), but elsewhere in grasslands they have seldom risen much
above 100°C, a level much lower than is usual in forest fires. From the
surface downward, the temperature gradient is 80 steep that direct
alteration of soil structure or chemistry by fire is inconsequential in
grasslands.
After a fire, and before a new herbaceous cover develops, the blackened and unshaded soil is appreciably warmer, at least during daytime,
in comparison with unburned areaa. Presumably in consequence of thiE
the foliage usually appears earlier in the firat poet-burn season, and
sometimes the onset of flowering is advanced. Normal litter cover ia
restored in 2-5 years. Whether humus increaaes or decreaaes, in ~ 0 1 ~ 8 8 -
quence of repeated grass fires, or whether erosion is accelerated or not
varies from place to place, but these seem constant for a given region
and vegetation type. I n perennial grasslands soils am drier after a burn.
Plant ash resulting from f i n g graaa cover seems too m n t to be significant, which contraate markedly with the appreciable rekwe of fertility