Negative Effects of the Western Thatching Ant (Formica Obscuripes) on Spiders (Araneae) Inhabiting Big Sagebrush (Artemisia Tridentata)
Total Page:16
File Type:pdf, Size:1020Kb
Great Basin Naturalist Volume 59 Number 4 Article 11 10-15-1999 Negative effects of the western thatching ant (Formica obscuripes) on spiders (Araneae) inhabiting big sagebrush (Artemisia tridentata) Michael W. Heikkinen Utah State University Follow this and additional works at: https://scholarsarchive.byu.edu/gbn Recommended Citation Heikkinen, Michael W. (1999) "Negative effects of the western thatching ant (Formica obscuripes) on spiders (Araneae) inhabiting big sagebrush (Artemisia tridentata)," Great Basin Naturalist: Vol. 59 : No. 4 , Article 11. Available at: https://scholarsarchive.byu.edu/gbn/vol59/iss4/11 This Article is brought to you for free and open access by the Western North American Naturalist Publications at BYU ScholarsArchive. It has been accepted for inclusion in Great Basin Naturalist by an authorized editor of BYU ScholarsArchive. For more information, please contact [email protected], [email protected]. Grent Basin Nilturulist 59(4), el999, pp. 380-383 NEGATIVE EFFECTS OF THE WESTERN THATCHING ANT (FORMICA OBSCUillPES) ON SPIDERS (ARANEAE) INHABITING BIG SAGEBRUSH (ARTEMISIA TRIDENTATA) Michael W. Heikkinen 1 An.\1:UACT.-Th<.~ effect of thatching ants (Formica obscuripes) on number of spiders found on sagebrush <Artemisia tritl.crua:a) was determined by comparing spider abundance on shrubs at diffcl'li'nt distances from ant mounw. Spider ahundance was inversely related to proximity to mounds. Spider abundance on shrubs around abandoned mounds revealed no effccts of mound proximity. Two mounds in which ants were exterminated showed a declining effect of mound proximity 1 month after treabnent. One yera after treatment there was no effect. Kay WOt-d..~: Amneae. Formica obscnripes, Artemisia tridentata, rompetit:icn. Ants exert significant influences on their F. obscuripes suggested that ants depress spi environments (Petal 1978) by limiting plant der abundances. In this study the effect of F growth (Ryti 1992) or increasing heterogeneity obscuri,pes on spider abundance on sagebrush ofplant communities (Beattie and Culver 1977). was measured at varying distances from active "In the forests ofFinland ants are the premier ant mounds, abandoned mounds, and mounds predators, scavengers, and turners of soil" from which anL<; were exterminated, (Holldobler and Wilson 1994:6). Many ants sig nificantly rednm populations ofarthropods with METHODS in their foraging area because of their inten~ sive foraging activity, social life, and polyph Surveys were conducted at 2 sites in the agy (Petal 1978, Bolldobler and Wilson 1990, Cache National Forest, Cache County, Utah: McIver et al. 1997). High Creek, which is northeast of Richmond Thatching ant, (Formica obscuripes) produce at 41°55'5.5"N, 111°4(}'20"1.v, and Cherry Creek, large mounds that are covered by small piece.s 4.8 km south of High Creek at 41°58'43"N, of plant material, often at the base of a shrub 111°44'44"W. Both are in shrub-steppe habitat (Allred 1982). They prey on insects, forage on at 1600 m elevation. dead insects, consume honeydew which they Three treatments were studied: active ant harvest from aphids, and occasionally eatplant mounds, abandoned ant mounds, and mounds tissue (Wheeler and Wheeler 1963, Beattie in which an ts were removed by treating with and Culver 1977, Petal 1978, Douglas and 225 g granular Diazanon per mound watered Sudd 1980, H611dobler and WIlson 1994). in with 2 gallons ofwater. No si!,'Uificant differences were found in Surveys of spiders on sagebrush near 2 the spider fauna of areas inside aud outside active aot mounds at High Creek were con the hunting area of the red wood ant, Fom'ica ducted in June and Angust 1995 and August polyctelw (Bruning 1991), and spider species 1996. Spiders on sagebrush near 4 active composition was not influenced by presence mounds at Cheny Creek were surveyed in of ant mounds in the Netherlands (Van der August 1996. Surveys of spiders on sagebrush Aart and DeWit 1971). Yet, an experimental near 3 abandoned ant mounds at Cherry Creek removal of ants from canopies of Douglas-fir were conducted during Angust 1996. Ants were (Psem:Wtsuga, menziesii) increased the number experimentally eliminated from 2 mOlll1ds at ofhunting spiders (Halaj et al. 1997). Cheny Creek in August 1997, and the subse Casual observations of spider abuodance quent effect on spider abundance was mea on shrubs at varying distances from mounds of sured in September 1997 and July 1998. 380 1999] NEGATIVE EFFECfS OF A"rrs ON SPIDERS 381 TABLE 1. Variables associated with each thatching ant motmd and results of the regression of distance from mound and spider abundance (transformed to square root ofspider abundance). Year Treatment Date Mound P ¥- intercept Slope 1995 Active 30 June HCla 0.004 0.7 0.195 Active 13 Aug HC2 0.048 2.3 0.048 1996 Active 15 Aug HCl 0.000 1.8 0.488 Active 15 Aug HC2 0.000 2.2 0.494 Active 1 Aug CCl" 0.008 1.5 0.268 Active 1 Aug CC2 0.003 0.1 0.305 Active 6 Aug CC3 0.004 0.4 0.400 Active 6 Aug CC4 0.001 -0.5 0.629 Abandoned 17 Aug CCAI 0.303 4.6 0.155 Abandoned 17 Aug CCA2 0.289 4.8 0.107 Abandoned 17 Aug CCA3 0.491 4.7 -0.130 Treated 11 Sept CC2 0.014 2.4 0.154 'freated II Sept CC3 0.029 2.4 0.160 1997 1reated 24 July CC2 0.237 2.5 0.090 ueated 24 July CC3 0.092 2.4 0.111 -He", High C.....dt. hcc '" CI'oefT)' Creek. Surveys were conducted by establishing would no longer increase as distance from the thre<l12-m lines from each mound and collect mound increased. ing spiders on each shrub along these lines. To identify the distance where the effect of The 3 lines were '~sually selected so they ants was not significant, a series of regression would intersect the greatest number of shrubs equations was produced for each active mound. and still be relatively evenly spaced around In the 1st step the data were divided into 2 the mound. Spiders were collected using a groups based on a fixed distance from the beating sheet (Southwood 1978), after which mound, and a regression line was fitted to they were preserved in 70% ethanol and taken each set of data. This was done repeatedly by to the laboratory for sorting and counting. The regrouping the data at 1-m intervals from 2 to distance from the mound to eacb sbrub was 9 m to find the point that divided the data into measured to the nearest 0.1 m. 2 groups in which the group closer to the mound had a significant positive slope and the RESULTS more distant group had a slope that was not significantly different from zero. Wben a dis Analysis was conducted for individual tance was identified, the analysis was refined mounds because size of mound, size of shrub, by repeating the process by dividing the dis and time of year can affeet number of spiders found on shrubs. Regression analysis was used tance into groups based on O.l-m increments. to examine the relationship of spider abun The distance that met the criteria listed above dance on sagebrush and proximity to thatch and had the lowest sum of squared residuals ing ant mounds. Spider numbers were trans was identified and used in the next procedure. formed using the square root of spider abun Once a point was identified, a new single dance per shrub to reduce ditferences in vari regression equation was computed in which ance. Results are shown in Table 1. the slope of the regression line was assumed to Regression analyses for surveys of spider change at the point identified. Both slopes were abundance near the 3 abandoned mounds then forced through a single point at the iden found no significant relationship between dis tified distance. Results are shown in Table 2. tance of shrubs from the mound and spider ldentified points ranged from 6.0 to 8.6 m abundance per shrub (P = 0.289--0.491). Eight from mounds. Curves produced by this proce regression analyses for surveys of active dure significantly fit the data (P = 0.001-0.039). mounds were significant (P = 0.001-0.048); Y-intercepts ranged from - 0.10 to 1.80. Since however, it was not logical that the number of the data were transformed, the number of spi spiders would rise continuously. There had to ders at zero distance from the mound ranged be a point at which the number of spiders from -0.01 to 3.2. 382 GREAT BASIN NATUJ\ALlST [Volume 59 TAUI.J~ Z. Purameters derived from 2 slope models developt'.d to predict tile effect of proximity to thatching ant mounds on !lvidcr llulindam..-e on sagebrush. PI'edictw Pre<]jcted D spider spider abl1fl~ (distance ubund.1.nce Slope dam..-e wht::re to slope pror at 0.0 m lor 0.0 slope cMnges Year Mound change) model from mound toD to 7.t:t'O 1995 HeIR 8.0 0.004 0.16 0.28 2.7 HC2 8.6 0.039 3.24 0.32 4.4 1996 HCI 7.0 0.001 1.04 0.74 6.2 IlC2 7.7 0.002 0.0004 0.55 5.9 cel" 8.0 0.012 -0.0004 0.62 4.8 Cc,'2 6.0 0.000 -0.0001 0.61 4.0 CC3 7.8 0.009 0.25 0.50 4.4 CC4 8.0 0.002 -O.Ql 0.36 2.8 CC2 (t,ent",l) 8.1 NS CC3 (treated) 8.0 NS Bile = I fiJ:h Cnd.