Canopy gaps are sites of reduced belowground plant competition in a productive old field James F. Cahill Jr. 1,2, * and Brenda B. Casper 1 1 Department of Biology, University of Pennsylvania, 19104, Philadelphia, USA; 2 Current address: Department of Biological Sciences, University of Alberta, Edmonton, T6G 2E9, Alberta, Canada; *Author for correspondence (e-mail: jc.cahill@ualberta.ca) Received 6 December 2000; accepted in revised form 7 September 2001 Key words: Abutilon theophrasti, Favorable microsite, Root competition, Root gaps, Succession Abstract Whether small gaps in the aboveground vegetation of productive old fields correspond to gaps in belowground plant biomass, and whether such “root gaps” result in a reduction of competition for soil resources is not known. Our study in an abandoned hayfield shows that root biomass within small gaps (< 0.50 m diam) is 20% of that found within intact vegetation, similar to the findings for shoot biomass. Associated with the decrease in root biomass was a 25% reduction in the intensity of belowground competition within gaps compared to the sur- rounding matrix vegetation. These differences could not be attributed to variation in soil properties, as gap and matrix soils did not differ in any of the physical or chemical properties measured. These results indicate that the increased plant growth commonly observed within gaps may be partly due to reduced belowground competition, independent of any advantage gained from increased light availability. By providing areas of low belowground competitive intensity, gaps in this field could allow poor belowground competitors to exist with in old fields, thus increasing community diversity. Introduction In plant communities, local areas of reduced plant biomass (gaps) are common, and are important in community dynamics (Pickett and White 1985). The scale of these gaps can vary dramatically between and within communities, ranging from less than an indi- vidual plant to the size of an entire watershed (Pick- ett and White 1985). In productive herbaceous com- munities, gaps less than 50 cm diameter are most common (Platt 1975; Goldberg and Gross 1988), with gaps not only areas of bare ground, but generally con- taining some reduced level of plant cover (e.g., Brad- shaw and Goldberg (1989)). Many species grow or survive better in gaps than within the surrounding matrix vegetation (Gross and Werner 1982; Hobbs and Mooney 1985; Aguilera and Lauenroth 1995; Morgan 1998), and gaps often have a species compo- sition different from that of the matrix vegetation (Platt 1975; King 1977; Hobbs and Mooney 1985; English and Bowers 1994; McIntyre et al. 1995). Growth benefits from gaps have generally been attrib- uted to increased light availability (Goldberg and Werner 1983; Morgan 1998). The study of gap dynamics has historically been from a canopy perspective, and only recently have researchers attempted to determine if gaps in the aboveground vegetation correspond to gaps in below- ground biomass (Sanford 1990; Silver and Vogt 1993; Wilczynski and Pickett 1993; Campbell et al. 1998; Denslow et al. 1998; Ostertag 1998), and to a lesser extent, what effects gaps in root distributions may have on belowground interactions between plants (Aguilera and Lauenroth 1993, 1995; Ostertag 1998). However, the vast majority of this work has been conducted in forests (e.g., Whendee and Vogt (1993); Campbell et al. (1998); Denslow et al. (1998); Oster- tag (1998)) and semi-arid grasslands (e.g., Coffin and Lauenroth (1998); Aguilera and Lauenroth (1995)). As plant-soil interactions in dry grasslands are funda- 29 Plant Ecology 164: 29–36, 2002. © 2002 Kluwer Academic Publishers. Printed in the Netherlands.