The widening of xylem vessels from tip-to-base of trees is an adaptation to minimize the hydraulic resistance of a long pathway.  Given that parallel veins of monocot leaves do not branch hierarchically, vessels should also widen basipetally but, in addition to minimizing resistance, should also account for water volume lost to transpiration since they supply water to the lamina along their lengths, i.e. “leakiness”.

We measured photosynthesis, stomatal conductance, and vessel diameter at 5 locations along each leaf of 5 perennial grass species.

We found that the rate of conduit widening in grass leaves was larger than the widening exponent required to minimize pathlength resistance (0.35 vs. ~0.22).  Furthermore, variation in the widening exponent among species was positively correlated with maximal stomatal conductance (r² = 0.20) and net CO₂ assimilation (r² = 0.45).

These results suggest that faster rates of conduit widening (>0.22) were associated with higher rates of water loss. Taken together, our results show that the widening exponent is linked to plant function in grass leaves and that natural selection has favored parallel vein networks that are constructed to meet transpiration requirements while minimizing hydraulic resistance within grass blades.

Plant Material

Six common grass species native to the tallgrass prairie were selected for this study: Andropogon gerardii Vitman (Ag, C4 NADP-ME), Panicum virgatum L. (Pv, C4 NAD-ME), Sorghastrum nutans Nash (Sn, C₄ NADP-ME), Schizachyrium scoparium Nash (Ss, C₄ NADP-ME), Elymus canadensis L. (Ec, C₃) and Bromus inermis Leyss. (Bi, C₃).  C₃ and C₄ species were included in this study to investigate the generality of the patterns across a range of grass species rather than an attempt to characterize differences between these functional groups.  Seeds were collected from Konza Prairie Biological Station, Kansas, USA (KPBS) and stored at 4°C until the initiation of the study.  Seeds of each species were germinated on wet filter paper and then five individuals were transplanted into 1.65 L pots (“Short One” Treepots; Stuewe & Sons, Inc., Tangent, OR, USA) filled with soil taken from KPBS.  This resulted in five replicates per species, which were then randomly distributed within a growth chamber (Conviron PGV 36; Conviron Environments Limited, Winnipeg, Manitoba, Canada).  Air temperature in the growth chamber was maintained at 30°C with a 16 h photoperiod.  The light intensity at the top of the canopy was maintained at ∼1000 µmol m−2.s−1 by adjusting the height of the lights as the canopy grew.  Plants were watered daily to ensure that soil moisture remained near pot-holding capacity.  Plants were grown until they had four to six mature leaves before beginning measurements.  5-6 plants of each species were used to measure rates of gas-exchange and anatomy.  The length of the leaf blade of each individual was measured and gas exchange and anatomy measurements (described below) were centered at 10, 25, 50, 75, and 90% of the leaf length.  The distance from the tip of the leaf to each of these measurement points is reported as dtip.  The gas-exchange measurements were made over 5 days, and 2-3 individuals of each species were randomly selected and measured each day.  After gas-exchange measurements were completed, plant tissue was placed in a chemical fixative and stored until further processing for microscopic analysis.

Gas exchange

The most recently mature leaf blade of each plant was identified and divided into five equal longitudinal sections.  The middle of each section was marked (using a black marker) and all subsequent measurements were centered on this mid-point; the distance from the leaf tip to the middle of each section was measured (d_tip).  Photosynthetic rate (A) and stomatal conductance to water vapor (gs*wv) were measured using an infrared gas exchange system (Li-6400, Li-Cor, Inc., Lincoln, NE, USA) with a small (2 cm²) fluorometer cuvette (6400-40, Li-Cor, Inc., Lincoln, NE, USA), which was chosen to maximize the proportion of the cuvette occupied by leaf.  Conditions inside the cuvette were maintained to match the condition of the growth chamber [∼400 µmol mol⁻¹ CO2, 30 °C, 50 ± 5% relative humidity, 1000 µmol m⁻² s⁻¹ photosynthetically active radiation (PAR)].  Gas exchange measurements were first made on the basal section and then sequentially along the length of the blade toward the apex.  Data from the Li-6400 was logged every 5 s until both A and g_s*wv were stable for >1 min (typically ∼5–10 min of logged data).  Data were imported into Matlab (Mathworks, Inc., Natick, MA, USA), visually inspected to ensure stability and then the mean of the last minute of data (12 logged points) was calculated and corrected for the leaf area in the chamber during the measurement.

Leaf Anatomy

Leaf blade tissue was vacuum infiltrated with a fixative (formalin-acetic acid-alcohol) and placed in 4°C storage for further processing. Each blade section was embedded with paraffin and then double stained with Safranin-O and Fast Green at the Kansas State University Histology Lab.  Images were taken with a digital camera (Leica DFC 290; Leica Microsystems GmbH, Wetzlar, Germany) coupled to a light microscope (Leica DM1000; Leica Microsystems GmbH) and analysed using ImageJ.  We measured xylem lumen diameter of the largest vessel in 3 different vascular bundles: the center bundle in the midrib (there can be multiple bundles in a midrib), the major bundle closest to the leaf edge, and the small vascular bundle that was next to, but closer to, the leaf center.  These measurements were made in the specified vascular bundle in each of the 5 blade sections.  For the midrib, we could ensure that we were measuring the same vessel in all 5 sections.  For the 2 other vascular bundles selected, it was not always possible to identify the same vascular bundle and so some of the measurements were likely made on vessels on opposite sides of the leaf.