Why Early Season Weed Control Is Important in Maize Eric R. Page, Diego Cerrudo, Philip Westra, Mark Loux, Kenneth Smith, Chuck Foresman, Harold Wright, and Clarence J. Swanton* Control of early-emerging weeds is essential to protect the yield potential of maize. An understanding of the physiological changes that occur as a result of weed interference is required to address variability in yield loss across sites and years. Field trials were conducted at the University of Guelph (UG), the Ohio State University (OSU), and Colorado State University (CSU) during 2009 and 2010. There were six treatments (season-long weedy and weed-free, and weed control at the 1st-, 3rd-, 5th-, and 10th-leaf-tip stages of maize development) and 20 individual plants per plot were harvested at maturity. We hypothesized that, as weed control was delayed, weed interference in the early stages of maize development would increase plant-to-plant variability in plant dry-matter accumulation, which would result in a reduction of grain yield at maturity. The onset of the critical period for weed control (CPWC) occurred on average between the third and fifth leaf tip stages of development (i.e., V1 to V3, respectively). Rate of yield loss following the onset of the CPWC ranged from 0.05 MG ha 21 d 21 at UG 2009 to 0.22 MG ha 21 d 21 at CSU 2010 (i.e., 0.5 and 1.6% d 21 , respectively). On average, reductions in kernel number per plant accounted for approximately 65% of the decline in grain yield as weed control was delayed. Biomass partitioning to the grain was stable through early weed removal treatments, increased and peaked at the 10th-leaf-tip time of control, and decreased in the season-long weedy treatment. Plant-to-plant variability in dry matter at maturity and incidence of bareness increased as weed control was delayed. As weed control was delayed, the contribution of plant-to-plant variability at maturity to the overall yield loss was small, relative to the decline of mean plant dry matter. Nomenclature: Atrazine; glyphosate; mesotrione; S-metolachlor; maize, Zea mays L. Key words: Zea mays, corn, weed interference, yield loss, harvest index, kernel number, kernel weight, plant-to-plant variability, reproductive allometry. Weed interference remains one of the major limitations to crop productivity in North America (Rajcan and Swanton 2001; Subedi and Ma 2009). It has been estimated that, at their peak, yield losses from weed interference can range from 0.03 to 0.21 MG ha 21 d 21 in maize (Hall et al. 1992). Timing of weed emergence relative to the crop, weed density, and weed competitive ability are variables that will influence the onset and rate of yield loss (Kropff and Spitters 1991; O’Donovan et al. 1985; Swanton et al. 2008). The onset and severity of yield losses from weed interference, however, can vary across sites and years, often with little apparent connection to the level of weed pressure or to the timing of weed management practices (Evans et al. 2003a; Hall et al. 1992). To address this variability, integrated weed manage- ment strategies have often incorporated knowledge of the critical period for weed control (CPWC; Swanton et al. 2008). The CPWC has provided generalized guidelines for the timing of weed control practices based on the mean yield losses observed during several site years (Hall et al. 1992; Knezevic et al. 1994, 2002; Swanton et al. 1999; Van Acker et al. 1993). Although these studies have made significant contributions toward optimizing the timing of weed manage- ment practices, they have often overlooked the critical information required to explain the inconsistency in yield losses that they have summarized. In a review of the concept and analysis of the CPWC, Knezevic et al. (2002) suggested that there is a minimum amount of data in addition to grain yield that should be collected in order to address the inherent variability in crop– weed interference relationships. The authors suggested that data on important variables, such as weed species and density, date of weed and crop emergence, and weekly staging and height measurements, should be collected in order to quantify the competitive environment and extrapolate the results beyond the scope of the experiment in question. We advocate that, in order to take a further step toward understanding the physiological mechanisms underlying the observed yield losses, weed control studies should also collect data on crop yield components (i.e., seed number and weight) and biomass partitioning at physiological maturity. To date, only a few weed control studies have collected such data (Cox and Cherney 2010; Cox et al. 2006; Evans et al. 2003a,b; Tollenaar et al. 1997). Although yield losses from weed interference are reported on per-unit-area basis, they are in fact the direct result of changes in biomass accumulation and partitioning of the individuals that comprise the crop stand. Therefore, an understanding of how resources are allocated and yield is formed by individuals of a given crop species is required in order to identify commonalities in the patterns of yield losses observed across sites and years. In maize, the proportion of total aboveground biomass at maturity allocated to grain (i.e., harvest index [HI]) is relatively stable for large and midsize plants, but then declines rapidly in smaller, stressed individuals (Vega et al. 2000). The variation in HI among smaller individuals with similar biomasses is indicative of a breakdown in reproductive allometry, where reproductive growth is uncoupled from vegetative growth (Vega and Sadras 2003). The decline in HI for individuals with low plant dry matter (PDM) is often associated with a reduction in kernel number per plant (KNP). Kernel number is associated with the rate of plant dry-matter accumulation and partitioning to Weed Science wees-60-03-23.3d 9/5/12 20:53:41 1 Cust # WS-D-11-00183 DOI: 10.1614/WS-D-11-00183.1 * First author: Research Scientist, Agriculture and Agri-Food Canada, Greenhouse and Crops Processing Centre, 2585 County Road 20, Harrow, Ontario, Canada, N0R 1G0; second and eighth authors: Research Associate and Professor, Department of Plant Agriculture, University of Guelph, 50 Stone Road East, Guelph, Ontario N1G 2W1, Canada; third author: Professor, Department of Bioagricultural Sciences and Pest Management, Colorado State University, 1177 Campus Delivery, Fort Collins, CO 80523; fourth author: Professor, Department of Horticulture and Crop Science, The Ohio State University, Columbus, OH 43210; fifth author: Professor, Department of Crop Soils and Environmental Sciences, University of Arkansas, 1366 West Altheimer Drive, Fayetteville, AR 72704; sixth author: Syngenta USA; seventh author: Syngenta Crop Protection Canada. Corresponding author’s E-mail: cswanton@uoguelph.ca Weed Science 2012 60:000–000 Page et al.: Yield loss in maize N 0