in the case considered above, such an approach would greatly oversimplify the factors
affecting relative gas exchange under the dynamic, highly variable light regimes prevailing
in natural shaded habitats. Research by Pearcy and his colleagues (Pearcy et al.
1985; Chazdon and Pearcy 1986a, 1986b; Chazdon 1986) on trees growing in tropical
forest understories has elegantly demonstrated the importance of photosynthetic induction
and non-steady-state behaviour in sunflecks for gas exchange under 'shady' conditions.
Finally, Osmond et al. (1980) presented an analysis similar to that given here, but
reached different conclusions based on two inappropriate assumptions. They found that
leaves acclimated to a given irradiance have a greater 24 h carbon balance at that
irradiance than leaves acclimated to another irradiance, regardless of whether photosynthesis
is expressed per unit area or per unit mass. However, they assumed a sinusoidal
variation in irradiance during the photoperiod, rather than the on/ off switch to which
the leaves were actually acclimated. A sinusoidal variation introduces a daytime period
during which leaves acclimated to high irradiance would be below their compensation
point in chambers at 'low' irradiance, and artificially inflates the advantage of leaves
acclimated to low irradiance under such conditions. More importantly, comparisons
were made only between leaves acclimated to low and high irradiances, at only those
two irradiance levels. Thus, Osmond et a[. (1980) did not recognise the crucial complications
introduced by leaves acclimated to intermediate irradiance which-when carbon
balance is expressed per unit leaf area-do better than leaves acclimated to low
irradiance at that irradiance, and worse than leaves acclimated to high irradiance at
intermediate irradiance (Fig. 2a).
Economics of Support
This section addresses three aspects of leaf arrangement likely to be shaped by
tradeoffs involving the economics of support: leaf area index, relative crown width, and
spiral v. distichous phyllotaxis. All three traits involve the integration of individual
leaves into an effective photosynthetic canopy. Analysis of optimal variation in the first
trait requires a whole-plant perspective on how to measure light compensation point and
photosynthetic light response, and bears on the issue of maximum plant height at a given
irradiance. The second trait entails allocation to leaves v. support tissue in dynamic
canopies. The third involves a key trait at the twig, rather than the canopy, level.
Additional constraints on the evolution of phyllotaxis are reviewed by Givnish (1979,
1984).
Leaf Area Index, Effective Compensation Point and Maximum Plant Height
Leaf area index (LAI) is the ratio of a plant's total leaf area to the horizontal area
covered by its canopy. in other words, LA1 is the average number of leaf layers over
each point occupied by a canopy. Horn (1971) presented an early, highly simplified
model for evaluating the relative photosynthetic outputs at a given irradiance from
canopies that differ in their number of leaf layers (see also Blackman and Black 1959).
This model does not incorporate many traits and processes that help determine wholecanopy
photosynthesis and whole-plant carbon gain, including: within-crown variation
in leaf orientation, distribution, temperature, and photosynthetic capacity; movement
of the sun across the sky; shading by adjacent competitors; shading by a plant's own
trunk and branches; costs of building and maintaining support tissue; and eddy transfer
of COz through canopy. Nevertheless, Horn's model has the notable advantage of
exposing and analysing in the clearest possible terms one tradeoff central to adaptive
variation of LAI, involving the balance between number of leaf layers and photosynthesis
per layer. Canopies with more leaf layers have a greater potential for carbon gain
because they contain more photosynthetic tissue, but they also entail more self-shading,