Carbon-isotopic analysis of individual pollen grains from C3 and C4 grasses using a spooling-wire microcombustion interface

Pollen grains from grasses using the C3 and C4 photosynthetic pathways have distinct ranges of d C values that may be used to estimate their relative abundance in paleorecords. We evaluated a spooling-wire microcombustion device interfaced with an isotope-ratio mass spectrometer (SWiM-IRMS) for dC analysis of individual grass-pollen grains. Pollen from four C3 and four C4 grass species was isolated through micromanipulation and analyzed as single grains suspended in water. A carbon yield greater than the 2r range of the carbon content of blanks containing only water was used to distinguish samples containing pollen (‘‘pollen present’’) from those not containing pollen. This criterion resulted in the exclusion of 45% of the 946 samples applied to the wire. The average dC values (±1r) of the remaining samples were 26.9‰ (±6.3‰) and 11.5‰ (±9.6‰) for C3 grasses and C4 grasses, respectively, after blank-correcting the d C data. These results suggest that the SWiM-IRMS system can be used to distinguish C3 from C4 grass pollen. The high variability in measured d C values is likely caused by a combination of factors. These include natural isotopic variability among individual pollen grains; the relatively poor precision that can be obtained when determining dC values of such small samples; and the uncertainty in the magnitude, isotopic composition, and stability of the analytical blank. Nonetheless, high percentages of individual pollen grains were correctly classified as being of either C3 or C4 origin. On average, 90% (range = 78–100%) of pollen grains from C3 grasses had dC values more negative than the cutoff threshold of 19.2‰; while 84% (range = 77–90%) of pollen grains from C4 grasses had d C values more positive than 19.2‰. Compared with analysis using an elemental analyzer interfaced with an IRMS (EA-IRMS), the number of pollen grains required for dC-based evaluation of C3/C4 grass composition is many times lower with the SWiM-IRMS. Additionally, dC data from the SWiM-IRMS does not need to be incorporated into a mixing model to derive estimates of the abundance of C3 and C4 grass pollen. Carbon-isotopic analysis of individual grass-pollen grains using the SWiM-IRMS system may help improve our understanding of the evolutionary and ecological significance of grass taxa in the paleorecord. 2007 Elsevier Ltd. All rights reserved.