Tiedemann and others (1984). For the few samples 

 with insufficient material for all analyses, fat and 

 occasionally other parameters were not determined. 

 Moisture content could not be determined in a few 

 samples because of drying problems. 



Soil samples were also collected and analyzed. Six 

 upland samples and six lowland samples collected from 

 each of the three study areas were analyzed for pH, 

 electrical conductivity (EC), sodium adsorption ratio 

 (SAR), sand, silt, clay, calcium, magnesium, sodium, 

 nitrogen, phosphorus, potassium, zinc, iron, and cop- 

 per using the methods of Page (1982). The analyses 

 for calcium, magnesium, and sodium were for the 

 water-soluble forms used in the SAR analysis. Soils 

 of the study areas have been classified as Winkel-Rock 

 land, Pintura-Toquerville-Dune land. Cave, and Rock 

 Outcrop-Rock land associations (Mortensen and others 

 1977). These are shallow, gravelly, fine sandy loams 

 (Winkle-Rock and Cave), deep loamy fine sands and 

 fine sands (Pintura-Toquerville-Dune), and shallow 

 soils over bedrock (Rock Outcrop-Rock). The City 

 Creek site includes Winkel-Rock, Pintura-Toquerville- 

 Dune, and Rock Outcrop-Rock soil associations. The 

 Woodbury-Hardy and Littlefield sites have Cave and 

 Rock Outcrop-Rock soil associations (Mortensen and 

 others 1977). The rocky soil associations in our study 

 areas were sandy. 



Statistical analyses, including analysis of variance, 

 Tukey means comparison tests, t tests, and correlation 

 coefficients, were performed using SAS (1985). Data 

 are presented proportionally, in percent, for all param- 

 eters except zinc, iron, manganese, and copper, which 

 are given in parts per million (ppm). Differences were 



considered significant at the P < 0.05 level, but actual 

 significance probability values are presented in some 

 cases. All proportional data were arcsine transformed 

 for analyses; they have been converted to real values 

 for presentation here. In comparing cured (quiescent) 

 and green (growing) samples, those with moisture con- 

 tent of 0.23 or greater were considered "green" and 

 those with moisture content less than 0.23 were con- 

 sidered "cured"; 0.23 is the arcsine transfer value of 

 0.50. This was an arbitrary value that effectively sepa- 

 rated green plant tissue (which fell above the value) 

 and cured plant material (which fell below the value). 

 Climatic data are from National Oceanic and Atmo- 

 spheric Administration (NOAA 1988-91). Botanical 

 nomenclature follows Welsh and others (1987) and 

 Baird (1990). 



This study is part of a larger project designed to assay 

 habitat and biological characteristics of desert tortoises 

 at the northern extent of their distribution (Esque and 

 others 1991; McArthur and Sanderson 1992a,b). 



RESULTS 



The appendixes list the species, times of collection, 

 and seasonal means with significant differences for 

 those species collected intensively for such comparisons. 



Table 2 lists mean values of all nutrients and min- 

 erals by location. These values are in the normal 

 range for semiarid sites for most parameters (Jones 

 and Hanson 1985; Kincaid 1988; Miller and Ramsey 

 1988; Rains 1976). However, the level for sodium is 

 abnormally low, the levels for sulfur and iron are low, 

 and the level for phosphorus is in the low end of the 



Table 2 — Nutrient and mineral values for each study area. Different letters in the same row 

 indicate a significant difference in mean value 







City Creeic 



Woodbury-Hardy 



Littlefield 



Nutrient 



Units 



(n=182) 



(n = 184) 



(n = 236) 



Moisture 



Percent' 



38.08 



A 



41.50 



AB 



46.96 B 



Nitrogen 



Percent 



1.58 



AB 



1.73 



B 



1.50 A 



Phosphorus 



Percent 



0.13 



B 



0.10 



A 



0.14 B 



Potassium 



Percent 



1.29 



A 



1.39 



A 



1.71 B 



Zinc 



ppm 



15.13 



A 



16.30 



AB 



18.56 B 



Iron 



ppm 



283.41 



A 



289.34 



A 



371.12 A 



Manganese 



ppm 



48.64 



AB 



39.54 



A 



49.14 B 



Copper 



ppm 



6.00 



A 



5.79 



A 



6.76 B 



Calcium 



Percent 



1.18 



A 



1.43 



AB 



1.60 B 



Magnesium 



Percent 



0.30 



A 



0.40 



B 



0.46 B 



Sulfur 



Percent 



0.29 



A 



0.36 



A 



0.40 B 



Sodium 



Percent 



0.010 



A 



0.010 



A 



0.012 B 



ADF2 



Percent 



36.34 



A 



35.66 



A 



34.52 A 



TNC3 



Percent 



7.32 



A 



7.27 



A 



8.56 A 



FaH 



Percent 



7.53 



A 



7.48 



A 



8.07 A 



'Based on fresh weight; all other values are based on dry weight. 



^Acid detergent fiber. 



^Total nonstructural carbohydrates. 



^Sample numbers for fat were reduced to 153 (City Creek), 152 (Woodbury-Hardy), and 217 (Littlefield). 



3 



