GRAZING EFFECTS ON HERBAGE COMPOSITION AND NUTRIENT DISTRIBUTION ON A FLORIDA RANGE FLATWOODS By BURTON J„ SMITH A DISSERTATION PRESENTED TO THE- GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1978 "There is something fascinating about science: One gets such wholesale conjecture out of such a trifling investment of fact." Mark Twain ACKNOWLEDGMENTS I wish to express my appreciation to Dr. G. 0. Mott, chairman of the Supervisory Committee, for his sincere counsel during the research and preparation of this manuscript. Thanks to Dr. L. D. White for guidance during the initial phase of the research; Dr. D. H. Hunter for his assistance during the latter portion of this project; Dr. H. T. Odum for his valuable assistance in the modeling portion; Dr. D. E. McCloud for his counsel and advice, and to Dr. J. K. Loosli for his assistance with the nutritional aspects . Special thanks to Steve Terry, who did much of the leg work obtaining measurements during this study; Janet Burdett and Ron Gilles for assistance during the laboratory and collection portion of this investigation; Miz C, Barbara, John, Pat, Barry, Tom, Bea and all the rest of the range students who x%rittingly or unwittingly were involved in this project. Most sincere thanks and love to my wife and typist, Marty. Finally, I wish to acknowledge all those people, known and unknown, who consciously or unconsciously, willing or unwilling, contributed in the shaping of my life to this point. TABLE OF CONTENTS Page ACKNOWLEDGMENTS ii± LIST OF TABLES vii LIST OF FIGURES xii ABSTRACT xlv CHAPTER 1 INTRODUCTION 1 2 LITERATURE REVIEW 5 3 SITE AND METHODS Physical Site Description 14 Methods 19 4 GENERAL SOIL STUDY OF TOE SITE Introduction 25 Materials and Methods 27 Resul ts 29 Discussion and Conclusions 39 5 CHANGES IN THE CHEMICAL CONTENT OF SOIL SAMPLES Introduction 44 Materials and Methods 45 Results 46 Discussion and Conclusions 51 6 MICRO-VARIABILITY AND ITS EFFECT ON SOIL NUTRIENT CONTENT Introduction 55 Materials and Methods 56 Resul ts 58 Discussion and Conclusions 61 7 IN VITRO DIGESTIBILITIES OF RANGE FORAGES Introduction 64 Materials and Methods 67 Results 70 Discussion and Conclusions 76 8 CHEMICAL COMPOSITION OF RANGE SPECIES Introduction 85 Materials and Methods 88 Results 90 Discussion and Conclusions 107 9 UNDERSTORY HERBAGE PRODUCTION Introduction 119 Materials and Methods , 122 Results 126 Discussion and Conclusions 137 10 HERBIVORES OF THE FLATWOODS Introduction 146 Materials and Methods 153 Results 156 Discussion and Conclusions 167 11 CHEMICAL ANALYSIS AND IN VITRO DIGESTIBILITY OF THE FRONDS OF SERENOA REPENS (SAW PALMETTO) Introduction 173 Materials and Methods 175 Results „ . 175 Discussion and Conclusions 179 12 DECOMPOSITION OF PLANT AND FECAL MATERIAL Introduction 183 Materials and Methods 185 Results 187 Discussion and Conclusions 194 13 NUTRIENT INCREASES DUE TO OUTSIDE INPUTS Introduction 198 Materials and Methods 199 Resul ts 201 Discussion and Conclusions 206 14 NUTRIENT SIMULATION MODEL Introduction 209 Materials and Methods 212 Results 220 Discussion and Conclusions 226 15 CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE RESEARCH 230 v APPENDIX page A SITE AND METHODS DATA 237 B DESCRIPTION OF SOILS FOUND ON STUDY SITE 255 C PROCEDURE FOR THE TWO-STAGE IN VITRO ORGANIC MATTER DIGESTION OF FORAGES 269 D SOIL STUDY DATA 273 E HERBAGE DATA 280 F HERBIVORE DATA 322 G CIRCUIT ENERGY SYMBOLS FOR SIMULATION MODEL 325 LITERATURE CITED 326 BIOGRAPHICAL SKETCH 350 LIST OF TABLES Table Page 4.1. ANOVA analysis, based on the pastures actual physical relation to one another, for average soil data 31 4.2. Comparison of soil data collected in April 1977 with that collected in March 1978 for P, Ca, and K 34 4.3. Analysis of various soil components for the wet (Adams- ville Series) and the dry (Sparr Series) taken near the interface of the two soil groups 37 5.1. pH and some extractable elements of a Sparr sand as a function of drying temperature 47 5.2. pH and some extractable elements of a Sparr sand after wetting and re-drying the original samples 50 5.3. The change from the original analysis (Table 5.1.) and level of significance, as a result of wetting and re- drying the original samples 52 7.1. Effects of different components of the two-stage in vitro technique on digestibilities of Smilax auriculata and Trilisa paniculata 72 7.2. r Effect of inoculum from steer used as standard and range steer on in_ vitro digestibilities of twenty one range species 73 7.3. Nutrient content of forages used for comparison of different inoculum 74 7.4. Seasonal variations in digestibilities of various range plants, in percent 77 8.1. Average foliar phosphorus content (ppm) of eleven species in replications I and II 91 8.2. Average foliar calcium content (ppm) of eleven species in replications I and II 92 8.3. Average foliar potassium content (ppm) of eleven species in replications I and II 93 8.4. Mean separation analysis of treatment and replications for foliar P content at beginning and end of experiment. . 96 8.5. Relative percent change of foliar P from summer 1976 to summer 1977, composite of all species 97 8.6. Change in foliar P content from summer 1976 to summer 1977 for A. stricta (ppm) 98 8.7. Percent of species reflecting increase in chemical content (P, K, Ca) over control treatment (12 month rest) 103 8.8. Average foliar chemical content of Pinus palustris, summer 1977 collection, replication I and II combined 105 8.9. Foliar chemical content of Quercus incana, portions of replications III and II 106 8.10. Average coefficients of variation of each season for IVOMD, P, Ca, and K 108 9.1. Forage production (Kg/ha) based on uniform length of experiment (326 days: 23 August 1976 to 18 July 1977) 129 9.2. Production (Kg/ha) for treatments compared to control samples taken at same time as the particular treatment.... 131 9.3. Relative percent change in species diversity from summer 1976 to summer 1977 132 9.4. Relative percent change in leaf hits on line transects for the average of all treatments from September 1976 to September 1977 133 9.5. Diversity indices from 30 m line transects of control and two-month rest pastures 135 9.6. Overstory characteristics of the study site (all trees)... 136 10.1. Forage quality of range diet based on bite counts for different seasons 157 10.2. Relative percent change in individual cattle weights from date of entry to the site, to date of leaving 161 10.3. Calculations of forage consumed in the two-month rest pasture (July and August 1977) per cow per day by two different methods; estimations of forage removed (Table A.l.) and by eq. 7 (fecal output and average digestibility 163 10.4. List of food items from stomach analysis of deer collect- ed on Ocala Wildlife Management Area, September - February, 1952-1953 164 10.5. Number of different groups of arthropods caught in pit traps, September 1977 165 10.6. Average number of Lycosidae caught during the five day collecting period, by pastures 166 10.7. Average weight and P, Ca, and K content of three groups of insects trapped at the study site 168 11.1. In vitro digestibility and chemical content (P, Ca, and K) of Serenoa repens fronds 177 11.2. Statistical differences of composited fronds of Serenoa repens for in vitro digestibilities and chemical composi- tions (P, Ca, and K) 178 11.3. The Ca:P ratio of Serenoa repens fronds 180 12.1. Change in soil nutrient content, July 1977 to March 1978 in the controls at different depths 188 12.2. Average value of nutrients underneath fecal samples, compared to control, at different depths, for March 1978.. 189 12.3. Average nutrient content of soil under fecal sample (1978) compared to soil nutrient content in 1977 190 12.4. Average nutrient concentration under fecal sample, at the center and 12.7 and 17.8 cm from center at five depths 191 12.5. Phosphorus profile below fecal sample (one sample) 192 13.1. Chemical content and pH of rainwater collected from May through September 1977 at the BRU 202 13.2. Quantity of minerals deposited on study site during five ' month collection period (May - September 1977) and year total (August 1976 - August 1977) 203 13.3. Mineral content of supplement and estimated daily consumption rate of each mineral 204 13.4. Amount of mineral in mineral supplement estimated to have been returned to each pasture and compared to 1977 soil analys is 205 14.1. Definitions of symbols and equations used in simulation model 216 A.l. Dates of sampling, herbage weights (Kg/ha), based on regression equations, percent of each pasture classified as moist (M) , wet (W) , and dry (D) 237 A. 2. Common names of trees and shrubs found at the study site.. 250 A. 3. Common names of legumes, forbs, and miscellaneous plants found at the study site 251 IX A. 4. Common names and characteristics of range grasses at the study site 253 A. 5. 1976 and 1977 monthly total rainfall and average minimum and maximum temperatures for the BRU and the 70 year averages for Gainesville 254 D.l. Average organic matter content and coefficients of variation (%) for soils at three depths, 0 - 10, 10 - 20, and 20 - 30 cm at the BRU 273 D.2. Average pH and coefficients of variation (%) for soils at three depths, 0 - 10, 10 - 20, and 20 - 30 cm, at the BRU 274 D.3. Average phosphorus content (ppm) and coefficients of variation (%) for soils at three depths, 0 - 10, 10 - 20, and 20 - 30 cm, at the BRU 275 D.4. Average potassium content (ppm) and coefficients of variation (%) for soils at three depths, 0 - 10, 10 - 20, and 20 - 30 cm, at the BRU 276 D.5. Average calcium content (ppm) and coefficients of varia- tion (%) for soils at three depths, 0 - 10, 10 - 20, and 20 - 30 cm, at the BRU 277 D.6. Average magnesium content (ppm) and coefficients of variation (%) for soils at three depths, 0 - 10, 10 - 20, and 20 - 30 cm, at the BRU 278 E.l. Digestibilities of range species collected at study site - seasonal collections 280 E.2.r Phosphorus content of range species collected at the BRU - seasonal collections 286 E.3. Calcium content of range species collected at the study site 292 E.4. Potassium content of range species at study site 298 E.5. Average foliar chemical composition (ppm) composited for eleven species for each pasture 304 E.6. Average P foliar content in pasture at beginning of experiment - 1976 305 E.7. Average P foliar content of pasture at the end of experiment - 1977 306 E.8. Average chemical content of Pinus palustris, August 1977.. 307 E.9. Average chemical content of Pinus palustris, March 1977 and March 1978 collections 308 E.10. Functional groupings of biomass data for summer 1976 and summer 1977 309 E.ll. Diversity indices of pastures, summer 1976, based on four 100 foot line transects per pasture 315 E.12. Diversity indices of all pastures, summer 1977, based on four 100 foot line transects per pasture 316 E.13. Frequency of leaf hits on line transects for the average of all treatments for September 1976 and 1977 317 F.l. Cattle weights (lbs) as entered and left study site, showing weight changes 322 LIST OF FIGURES Figure page 3.1. Physical layout at the study site (BRU) 15 4.1. Soil profile of the wet and dry sites along vegetational ecotone 36 4.2. Soil map based on soil cores showing boundary of the two different soil types 38 5.1. Change in chemical analysis for Fe, Mg, and pH as a function of drying temperature 48 5.2. Change in Al, Ca, K, and P as a function of drying temperature 49 6.1. Contour of transect of micro relief study area 59 6.2. Amounts of Ca, P, and water as a function of distance from the center of a small rise in the top 10 cm of soil 60 7.1. In vitro digestibilities of four range plants as a function of time of removal from in vitro process 71 8.1. Phosphorus of selected species as a function of rest 100 8.2'. Calcium content of selected species as a function of rest.. 101 8.3. Potassium content of selected species as a function of rest 102 9.1. Standing crop, all treatments, replication I, for the study period 128 11.1. Symmetrical pattern of defoliation by cattle on Serenoa repens 174 11.2. Collection profile of Serenoa repens 176 14.1. Energy model of grazed flatwoods system 213 14.2. Phosphorus flows in the grazed flatwoods system 214 14.3. Simulated model with equations 222 14.4. Performance of the soil (Q^) , understory (Q„) , and cattle (Cv) as a function of time (initial conditions) 224 XII 14.5. Effects of harvesting trees, without cattle, on the soil and understory (initial conditions) 224 14.6. Twice initial soil phosphorus level, with cattle 225 14.7. Twice initial soil phosphorus level, without cattle 225 14.8. Soil phosphorus level four times initial value 227 14.9. Initial conditions, phosphorus input to cattle equal to initial soil value 227 Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy GRAZING EFFECTS ON HERBAGE COMPOSITION AND NUTRIENT DISTRIBUTION ON A FLORIDA RANGE FLATWOODS By Burton J. Smith August 1978 Chairman: Gerald 0. Mott Major Department: Agronomy The effects of short duration-high intensity grazing systems on a flatwoods site at the Beef Research Unit, Gainesville, Florida, were investi- gated from July 1976 to September 1977. Length of rest was the variable, with four replications and four treatments of two, four, six and twelve (control) months rest. Grazing was regulated to remove 50% of available forage, determined prior to cattle entry. Herbage samples were collected prior to and after grazing and forage production determined. In vitro digestibilities, and chemical determinations for phosphorus, calcium, and potassium were conducted on herbage samples for each collection period. Soils, groundwater, precipitation, insects, fecal and litter decomposition were monitored to obtain a total ecological picture of the effects of one year's grazing on flatwoods. Forage production trends indicate that three to four months rest at 50% utilization levels result in the highest production. Cattle production data indicated that this level was too high and a 30 to 40% utilization of flatwoods ranges is suggested. Clipped forage samples were below minimal requirements for dry cows in phosphorus and energy. High Ca:P ratios exist in the native forages, averaging approximately 5:1 over the year. Cattle were found to be actively selecting the outer one-half bud of Serenoa repens in an apparent attempt to rectify this Ca:P imbalance. This portion of the bud has the lowest Ca:P ratio of any part of the plant. This same imbalance may well have implications for white-tailed deer management in the state. Soils at the site are highly variable in chemical content, especially phosphorus. Phosphorus content of the top 30 cm of soils averaged approxi- mately 12 ppm with individual pasture values of less than 0.5 ppm. This high variability of phosphorus, also noted within plant species, may be indicative of low nutrient sites and is suggested as a method for determining the limit- ing element for production of such sites. Lateral flow of nutrient contain- ing water in the micro-relief of the soil is proposed as one mechanism responsible for the high variations noted in the soil profile. There appeared to be a positive relationship between phosphorus content of the soil and foliar content in plants. Phosphorus input from rainfall was less than one kilogram per hectare per year; however, on phosphorus deficient sites this represents a significant input. Mineral supplementation of cattle considerably increased the phosphorus levels in the soil. The duration of the experiment was too short to note any meaningful changes in foliar chemical content or species composition. However, foliar content of phosphorus in needles of longleaf pine (Pinus palustris) are higher (P < 0.05) in the two-month rest pastures than in controls during the winter. The phosphorus cycle was simulated on the analog computer and under- story plants reacted favorably to additions of phosphorus via cattle supple- ment. Results of the insect study are tentative but indicate that biomass of the insect population may exceed that of cattle grazing the area. The Florida flatwoods exhibit characteristics similar to the tropical areas of the world. Of 21 grasses at the site, 20 were of the C^ pathway. This high preponderance of C plants is indicative of tropical environments rather than temperate ones. The implications of this are that grazing systems developed in the Western or Northern areas of the United States are not necessarily appropriate for Florida. Range management in Florida should address itself to tropical or subtropical areas of the world and research those systems that show promise of being adapted to local conditions. CHAPTER 1 INTRODUCTION Florida has had cattle longer than any other state in the Union but, from the standpoint of future potential and development, it is one of the youngest cattle states (Cunha, 1976). The Southeastern states, Florida in particular, have the largest potential for increased cattle numbers through the increased use of native range, improved varieties of forage, and manage- ment techniques suited to the local conditions than any other group of states in the nation (Little, 1977; Conner, 1977; USDA, 1975). Florida's native vegetation has a production potential exceeding most areas of the United States (White and Carter, 1977). "However, at present much of the woodland in Florida does not fie the image of a managed forest nor does much of the range fit the image of good grazing" (Commissioner D. Conner, 1977). About 6 to 8% of the native forage resource is being managed to its full potential (Yarlett, 1977). Cattle numbers in Florida have been increasing at the rate of 5% per year from 1967 to 1976. This compares to a national rate of increase for the same time span of 1.8% (Florida Department of Agriculture Staff, 1976). It is estimated that by the year 1985 Florida will have increased its beef herd 36% over 1974 and will be producing 82% of the non-fed and 28% of the fed beef that the state consumes (Cunha, 1976) . Florida flatwoods make up about half of the land area in Florida (Pritchett and Smith, 1974). The major forage component of this range consists of wiregrass and saw palmetto. The fact that large areas are covered by this type of vegetation is not a tribute to its desirability by ranch managers, but rather is the result of years of misuse. This situation has resulted from the attendant management practice of burning and immediate grazing that was developed as a necessary feature in order to obtain economical gains from range cattle enterprises. This is the periodic burning of the wiregrass range in order to remove the old herbage and stimulate new growth which is palatable to cattle. For a wiregrass range, Duvall and Hilmon (1965) showed that a three year burn rotation produced substantially more useable nutrients than an unburned range. Even so, a pine/wiregrass range is nutrient poor and generally does not meet the yearly requirements of dry cows without supplementation (Moore, J., 1977; Lewis et al., 1975). The end result of this type of use has been a decrease in the more desirable forage species and an increase in the poorer quality plants. Beef production on native range amounts to about 22 Kg/ha, compared to 336 Kg/ha on improved pastures with good management (Moore, 1977) . Knowledge of forage quality of various range plants in Florida is due primarily from the result of practical experience. Experienced ranchers know which plants are preferred by the grazing animal and which plants result in the best performance; maidencane (Panicum hemitomon)is a case in point (White, 19771; Camp, 1932). Forage quantity has little meaning unless some index of quality is known. When an area is evaluated solely on the basis of available dry matter, as it often is, capacity is generally overstated (Blair et al . , 1977). Except for wiregrass (Killinger, 1948; Hilmon et al . , 1962), nutrient content and digestibilities of native range species in Florida are virtually non existant (Ammerman, 1977). Young (1977) has looked at eight species over a four month period from a flatwoods range at the Beef Research Unit (BRU) . Lewis et al. (1975) have reported on species that are common in Florida but 1. White, L. D. 1977. Assoc. Range Prof., Univ. Fla., Personal communication. were grown in Georgia. Several investigators have reported chemical values for native range, but not by species (Halls, et al . , 1956; Koger et al . , 1961). Forage quality of a species, in terms of nutrient content and digesti- bilities, have little meaning unless some measure of intake by the grazing animal is also known (Abrams, et al . , 1978). A start in determining forage intake has been made by Gumma (1977) for range species in a flatwoods site, by determinations of bite count and microscopic point analysis of an esophageal- f is tula ted cow. The intensity and frequency of grazing will dictate the kinds of grasses that will exist on a native pasture (Yarlett, 1977) . Animals will graze the most palatable species more frequently. This continued cropping causes the plant to lose vigor and results in replacement by a less desirable species. With the knowledge of the growth patterns of the important range forage species, management systems can be developed that will account for the critical periods of a plant's growth. Knowing when to close and open the gates is the key to any grazing management system. In general, transplanting of systems of management that are successful in the West, such as various deferred grazing and deferred rotation, are not successful in the South (Duvall and Hilmon, 1965; Biswell and Foster, 1947) . However, a promising alternative to the practices used in Florida in the past appears to be the high intensity-short duration grazing system. This technique places large numbers of animals into small pastures for short periods of time, followed by long periods of rest (Norris et al., 1975). Such a system was studied by Felts (1976) at the BRU. She concluded, after a one year study, that a four to six month rest, following 50% removal of available forage, appeared to give the most promising results. This system is also being used in South Florida by commercial cattlemen. There exists a need for an integrated study utilizing an ecosystem approach to define the various components that are interacting in the existing system. A need exists for data on the nutrient quality and digestibilities of the range species in a flatwoods site both for wildlife and livestock managers. The short duration grazing system should be researched further to determine the effectiveness under sub tropical conditions. Based on the hypothesis that the short duration grazing system is valid under Florida conditions, the objectives of this study are to: 1. Determine nutrient content and digestibilities of native range plants. 2. Further analyze the short duration-high intensity grazing system. 3. Determine interactive effects of a grazed flatwoods ecosystem. 4. Make specific recommendations for future research and management techniques for best utilization of a flatwoods range ecosystem for livestock production. CHAPTER 2 LITERATURE REVIEW The first cattle to arrive in the United States landed on the Gulf Shore in 1521, when Ponce de Leon brought cattle to supply his projected colony (Bryan and Sharrock, 1941) . Additional stock was brought by DeSoto in the 1530's (Yarlett, 1961). "Some of those animals evidently outran the butcher and escaped into the woods, for later settlers found Indian tribes in possession of half-wild cattle, descendants of which have been kept on the run ever since, frequently consuming more miles than sustenance in the search for forage" (Bryan and Sharrock, 1941, p. 4). Cattle were herded in the vicinity of St. Augustine as early as 1712 by cattlemen supplying beef to the Spanish garrison (Yarlett, 1961) . Great herds were owned by Miccosukee Indians, with Chief Micanopy running a large herd on Payne's Prairie in the 1770's (Buchholz, 1929; Yarlett, 1961). By the late 1820' s many settlers, both Spanish and English, had taken up land in North Florida for the purpose of raising cattle. This settlement was further hastened by the building of the Bellamy Road from St. Augustine to Tallahassee in 1826 (Camp, 1932) . The major shipping points at this time were Tampa and Punta Rassa, with Cuba the main market. Over 30,000 head of cattle were shipped to Cuba from Punta Rassa in 1840 (Camp, 1932). Commercial stock raising received further impetus during the Civil War when stockmen supplied beef to the Confederacy. "It was through this period that such noted range kings as Jake Summer! in, H. T. Lykes, and Ziba King popped their 18-foot whips over big herds that waxed temporarily fat during the spring and early summer" (Bryan and Sharrock, 1941, p. 11). 5 Numerous herds were brought lnt0 Florlda from GeorgIa ^ ^ ^^ between 1830 and 1850; however, the first proved breed was a Devon bull introduced in 1861. ^ flrst „„,_ ^ ^ ^^ ^ ^ ^ four buns were brought to Cedar Key (Camp, 1932). Dodd (1954) disputes this claiming that four crossbred Brahman buHs were introdueed by 1859 ln South Florida. By !900, most of the Engiish breeds had been established in F1orida. The introduced cattle had a host of problems to overcome, aside from the poor ,uality of the range during m„ch of the year. Cattle tick fever too, a high ton. out of a carload of Hereford bulls shipped from Texas, only three were alive nine months later (Camp, 1932) . rhe eradication program for tick fever was begun in 1920 and completed in 1950 (Cunha, 1976). Salt- sick, a nutritional anemia caused by a deficiency of iron, copper, cobalt, or some combination of these elements, was first suspected by MaxweU (1888) Stoekbridge et a! . (1,02) first demonstrated an iron deficiency among cattle on native range. Specific recommendations were not made for correction of these conditions until 1931 (Becker at a!., 1931). Local cattlemen did recognize these problems and got around the, s„me„hat by rotating cattle from pasture- to pasture or, on unfeneed ranges, establishing a hospital farm on a wholesome pasture (Becker et al., 1965). Other common probiems noted by early ranches were paces, !imebided, Mrsh sickness, and falling disease. Lack of copper was the major contributor, but this was not ascertained until reeentiy (Becker et a!., 1965). Screwworms were not brought under contro! until the late 1950's. M„sq„itoes, horseflies, internal parasites, and a host of other pests still piague cattle in spite of various programs of spraying, dipping, or dusting. Florida cattlemen have not been particularly adept at accepting new concepts or ideas. Salting was a well accepted practice in the early 1900's on Western ranges (Doc Smith, 1948, old time Wyoming cattle rancher, personal communication) , however as late as 1932 only 15% of the ranchers in Alachua County salted their cattle (Camp, 1932). Rummell (1956) has also commented on the low level of management in South Florida. One area where ranchers did raise a clamor for new concepts was in the use of improved pastures, beginning about 1920. In 1924 a series of experi- ments were conducted with introduced grasses by the Agricultural Extension of the University of Florida, Seaboard Airline Railway Co., and the Florida East Coast Railway Co. (Bryan and Sharrock, 1941). A mixture of carpetgrass (Axonopus compressus) , Dallisgrass (Paspalum dilatatum) , Bahiagrass (Palpalum no ta turn) , and Japanese clover (Lespedeza spp.) were used on several locations throughout the state; annual carrying capacities were reported from one to two cows per acre. Numerous experiments followed; Ritchey and Henley (1936) looked at different grasses alone and in mixture and Leukel and Stokes (1942) worked at establishing carpetgrass under range conditions. Pangolagrass (Digitaria decumbens) was first introduced to researchers at the University of Florida in 1936 and later released to farmers and ranchers in 1943 (Hodges et al . , 1975) . Breeding programs and testing procedures were developed to test promising new varieties and cultivars of different forages and are continuing to this day (Moore and Ruelke, 1978) . While results of these early tests of improved pastures were made known, Florida, like the rest of the nation was in the midst of the "great depression." It wasn't until 1936 when financial aid was made available through the U. S. Department of Agricultural Adjustment Administration that large scale plant- ing by cattlemen began (Bryan and Sharrock, 1941) . The only hitch was that in order to obtain this aid the land must be fenced. At this time Florida was an open range state in most counties. This allowed the larger, more affluent, cattlemen (being In better financial condition) to take advantage of the new improved pasture grasses, which they did. One rancher in Collier and Hendry Counties bought and fenced 38,000 acres each year; Carey Carlton, with herds in DeSota, Hendry, Glades, and Highlands Counties converted 7000 acres between 1939 and 1940 (Bryan and Sharrock, 1941) . The passage of the "No Fence Law" by the Florida Legislature in 1949 resulted in better quality cattle and higher production of forage (Killinger, 1960) . Beef cattle were confined by fences to more restricted areas which made for a forced type of rotational grazing. Range management had arrived. In a general sense of the word, range management began with the advent of man into the Florida environment some 10,000 years ago (Griffin, 1974). Early man used fire as an aid in hunting, a method of clearing ground, and later to provide forage for his cattle (White, 1975a) . Fire was the first management tool to be used by ranchers in this state and remained about the only one until the mid-1950's (Rummell, 1956). Cattlemen used fire to control the growth of brush and weeds, stimulate new growth, and to remove dead grass. Controlled rotational burning was practiced every two to three years on the flatwoods. Ranchers generally felt that there was nothing to be gained by burning the prairie, hammock, and blackjack ranges where the undergrowth was thin. The indescriminate use of fire was frowned on by most of the early ranchers as most, if not all, had experienced some economic loss from wild or escaped fires (Camp, 1932) . Fire is a controversial subject with many arguments on both sides. Numerous studies were made to determine the effects of fire on the Coastal Plains of Florida and Georgia. Lemon (1946) studied the effect of fire on the Coastal Plain of Georgia; Killinger (1948) looked at the effect of fire on wiregrass ranges in Florida; Rummell (1956) reported stocking rates on burned flatwoods ranges in Florida; and Hughes (1974) noted the effects of fire on wiregrass ranges in South Florida. In general, all agreed that burning increased the nutrient content of the native range, but that this effect was temporary. Hilmon et al. (1962), working on a wiregrass range in South Florida, reported that the nutrient content, protein, and chemical composition was almost doubled in a few months after a fire, but had re- turned to near original levels by the end of a year. Early observation of repeated burned ranges, noted that certain kinds of plants seem to predominate. They were broomsedge (Andropogon virginicus) and wiregrass (Camp, 1932) . Later studies by Hilmon (1968) found that fire increases the range of saw palmetto and Parrott (1967) reported the same effect for wiregrass. This practice of burning the range continued almost unabated until the 1940 's. The end result was a general deterioration of the range, the more palatable and nutritious forage being all but eliminated. The shift to improved pastures brought a much needed respite to the range areas. The Florida range lands of today are in much better condition than they were 50 years ago (White, 1975b) . Range management in Florida can be dated from the publication of "A Study of Range Cattle Management in Alachua County, Florida," by Paul Camp in 1932. This was essentially a survey of existing practices in the ranch- ing community around Gainesville in the early 1930's. Little more was written on the subject, probably due to the interest in improved pastures and World War II, until 1942. The Coastal Plain Experiment Station in Tifton, Georgia, had been conducting research on cattle grazing on the Coastal Plain and a survey of forest grazing and beef cattle production was published by Biswell et al . (1942). Wells (1942) reported on the ecological problems in the Coastal Plains and presented a coherent picture of the ecology of the Coastal Plains. The year 1945 saw Nieland formalize his multiple use concept (Nieland, 1945) , a plan which was apparently some 25 years ahead of 10 its time, and the U. S. Forest Service admitted that grazing was alright for Southern pine ranges (U. S. Forest Service Staff, 1945). The 1948 yearbook of agriculture also devoted a chapter condoning the grazing on Southern ranges (Cassady and Shepherd, 1948). This position was moderated considerably, primarily as a result of grazing studies in the hardwood regions of the North indicating damage to the forest (Morrow, 1955; Williams, 1951) . Also as a result of a study reporting sheep damage to longleaf seedlings (Mann, 1947) and a study in the Coastal Plain by Cassady et al. (1955) which noted cattle damage to pine seedlings. The bias against grazing cattle in Southern forests persists to this day, although the evidence indicating damage under proper management is slight (Lewis, 1977). Rotational grazing on a native Coastal Plains range was studied by Biswell and Foster (1947) in North Carolina. Three different systems were investigated; continuous grazing, rotation at mid-season and a 28-day rotation. They concluded that all three systems gave essentially the same results. Fertilization of wiregrass ranges in conjunction with burning was investi- gated by Killinger (1948) at Gainesville, Florida. At 211 days after treat- ment the unburned, unfertilized plots were higher in dry matter but lower in the elements tested for than the burned plots. The plots that received 2-14-10 at the rate of 672 Kg/ha (600 lb/acre) plus micro-elements had the highest levels of nutrients in the foliar analysis, with P being 0.57%. A survey was conducted in West Florida by Brasington (1949) 2 of the use of the forested range lands. Out of 132,000 head of cattle 41% depended entirely upon the native range for sustenance (without supplementation of any kind) , 32% had below standard winter supplement and the balance utilized the woods on a part time basis. Srasington, J.J. 1949. Unpublished data, So. For. Exp. Sta., New Orleans, La. 11 With the passage of the No Fence Law and depressed cattle prices in the mid-19501 s, a renewed interest in the range was generated. Intensive range research in Florida did not begin until 1953 (Rummell, 1960) . An analysis of range problems in South Florida was published in 1956 (Rummell, 1956) . This report showed that in this area, with more than 10 million acres of native range and a population of over a million head of beef cattle, a very low level of native range and cattle management was practiced. In 1955, the Caloosa Experimental Range became the center for U. S. Department of Agricul- ture range research in South Florida (Rummell, 1960) . With the discovery that litter, rather than shade, was the dominant factor in limiting understory production, particularly with respect to planted clovers (Halls, 1955), efforts were begun to devise methods for measuring vegetation in the pine regions of the Coastal Plains. This work culminated in the publication of a technique and methods manual, the results of a sym- posium held in Tifton, Georgia, in 1958 (Southern and Southeastern Forest Experiment Stations Symposia, 1959) . For the first time a comprehensive description of methods and techniques especially adapted to measuring the understory of the Southern ranges was available. Later, other technical guides have also been available to Soil Conservation Service Staff and field workers (SCS Staff, 1975, 1976). Gatherum (1955) showed a preliminary cost return combination for maxi- mizing profits on management of forest lands for timber and cattle. Yarlett and Moore (1963) investigated the utilization of Gulf Coast salt marshes for cattle production and noted that the dominant invader, Black needlerush (Juncus roemerianus) can be effectively controlled. Halls et al. (1964) published a guide for use of the longleaf-slash pine forest for forage and cattle management. Application of rock phosphate applied to cut over and 12 chopped native ranges was found to double yields when applied at the rate of one ton per acre (Lewis, 1963; Hilmon et al . , 1962). Determinations of native grasses that furnish forage indicated that of 350 species of native grasses found in Florida. 147 are important in furnishing forage to livestock (Yarlett, 1963) . Consciousness that all grass is not wiregrass and should not be managed as such was making itself apparent to ranchers (Yarlett, 1965a) . About this same period of time heavy machinery began to be used for range improvement work. Bulldozers or road patrols, pulling drum choppers or webbers (a V blade, designed to cut off roots just below ground level) were first used to clear land for improved pastures (Reuss, 1958) but were later used on rangelands with promising results. Lewis (1972) noted that webbing was less effective than cross-chopping especially on moist sites for the control of shrub species such as saw palmetto. Estimates of forage quality of native species have depended primarily on " studies conducted in other states. Lewis et al . (1975) reported values for in vitro digestibility (IVOMD) for several groupings of understory plants in Georgia. Earlier, Halls et al. (1956, 1957) had reported values for wiregrass and other species on native ranges in Georgia. Hilmon and Douglas (1967) re- ported that the effects of fertilization on a forested range watershed in South Florida caused changes in the plant composition and that the rate and/or direction of plant succession may be altered. Significant gains in livestock and wildlife foods were also reported in this study. Yarlett (1965b) published a list of important native forage grasses in Florida. Lists of plants native to Florida and their potential were summarized as to ecosystem type by White (1973). An economic study by Hughes (1975) showed returns of $2.50, $2.00, and $1.25 per acre for high, medium, and low rates of stocking in a study at the Caloosa Experimental Range, South Florida. 13 The short duration grazing system was investigated by Felts (1976) . The results were inconclusive but trends indicated that a rest period of between four to six months was best suited after removal of 50% of the available forage. Gumma (1977) obtained the first data on cattle diet selection of native species on a flatwoods range; some plants that were not suspected of being utilized (e.g. grassy leaf golden aster) were found to supply a sizeable portion of the diet. Young (1977) also worked on a short- duration grazing system as affecting diet quality and reported on the IVOMD and chemical content of six range species during four months of growth. Anderson and Hipp (1974), on an economic evaluation of pasture and range operations in Florida, found the highest returns to land were from irrigated pastures, but the highest returns per animal were from, all range operations. With the inactment of the Forest Rangeland Renewable Resource Act of 1974 by the U. S. Congress, national direction was shifted to the concept of multiple use. The assessment made of the nations renewable resources (USDA Staff, 1975) focused on areas of prime interest of the nation as a whole with the Southeast showing the greatest potential. Nieland's (1945) plan became tenable for Florida. Investigators began to look seriously at multiple use systems (White, 1977). This interest resulted in range (for the first time) getting equal billing with forest and wildlife at the 1977 symposium "Mutual Opportunities for Forest, Range, and Wildlife Management" held at Gainesville, Florida. Range management in Florida is at the threshold of making significant contributions to the economic picture of the state. The high cost of ferti- lizer, coupled with the increasing cost of energy, will place the native range in a more favorable light to livestock producers. CHAPTER 3 SITE AND METHODS Physical Site Description The study was conducted at the Beef Research Unit (BRU) , approximately 16 Km north of Gainesville, Florida. The experimental site lies about 1 Km southeast of the farm headquarters, in a native flatwoods area. The site is bounded to the north, west, and south by lanes (roads) , with a lane running north and south through the center of the site. There are 16 pastures vary- ing in size from 0.648 to 0.714 hectares. A small intermittent pond is present in the northwest pasture. A creek approaches the site diagonally at the southeast corner (Figure 3.1.) • Corrals and scales were located adjacent to the northwest pasture. Water tanks were located in all grazed pastures. There is a young slash pine (Pinus elliottii) plantation to the west and a mature longleaf pine (Pinus palustris) plantation to the south. A cultivated field (usually corn) lies to the north and a Mesic hardwoods (Magnolia and Bays) community to the east. The land is mostly level, but slopes to the east and south (approximately 2%) on the eastern side. All replications and most of the pastures are bordered by fire lanes. Abandoned logging roads course the area, starting in the two-month treatment, replication IV, the road forks, one going to the northwest corner and the other to the northeast corner of the site. The site had been burned in February 1974 and again in February 1976. At the initiation of the experiment the soils classification of the area was a Leon fine sand, an Aerie Haplaquod. The site was reclassified 9 May 1978 into four different soils. Approximately 4% of the area is of the Myakka series 14 15 CORRAI CREEK 2,4, 6,& 12 MONTHS REST '.' I, Ml, « IV REPLICATION SOIL SERIES 17 --. WAUCHUA 28 --- ADAMSVILLE 48 ••• MYAKKA 50--- SPARR Figure 3. 1. Physical layout at the study site - BRU , 16 and 12% of the Wauchua series. Both of these soils lie to the extreme east of the site. The Sparr series comprises about 22% of the total area and lies in the central southern area. The remainder (62%) is of the Adamsville series and lies to the north and west (Figure 3.1.). The Adamsville series is a member of the hyperthermic, uncoated family of Aquic Quartzipsamments. These are sandy soils having grayish A horizons and mottled grayish or brownish C horizons. The mean annual soil temperature at depths of 0.5 m below the surface is 22 to 23°C. These soils are on low, broad flats in lower Coastal Plains. Slopes are generally less than 2%. These soils are somewhat poorly drained. Soils range from very strongly acid to mildly alkaline. Permeability is rapid. The water table is at 0.5 to 1.0 m for two to six months during most years. It is at 0.3 to 0.5 m for periods of up to two weeks in some years. It is within depths of 1.5 m for more than nine months in most years. These soils were formerly classified as Regosols intergrading Low-Humic gley soils . Tne Myakka series consists of deep, poorly drained soils formed in sandy marine deposits. These soils have rapid permeability in that A horizons and moderate or moderately rapid permeability in the Bh horizons. Slopes range from 0 to 2%. Combined thickness of the A and Bh horizons is more than 0.0 m. These soils range from extremely acid to slightly acid. The water table is at depths of less than 0.3m for one to four months in most years, and recedes to depths of more than 1.0 m during very dry years. Depressions are covered with standing water for periods of six to nine months or more in most years. These soils were formerly classified in the Leon series as ground water podzols. The Sparr series consists of somewhat poorly drained, moderately permeable soils formed in thick beds of sandy and loamy marine sediments. The subsoil is saturated in the summer, water runs off the surface slowly. Slopes range from 0 to 8%. Solum thickness is 1.5 m or more. Soils range from very strongly acid X to slightly acid in all horizons. The water table in these soils is at depths of 0.5 to 1.0 m for periods of one to four months. The water table is usually perched on the surface of the loamy layers. These soils were formerly classified in the Regosol great soil group. The Wauchula series is a member of the sandy over loamy, siliceous, hyper- thermic family of Ultic Haplaquods. These soils have a sandy dark colored Al horizon and a sandy light colored A2 horizon that total less than 0.8 m thick over a Bh horizon and an underlying Bt horizon with low base saturation. These soils range from very strongly to strongly acid throughout. These soils are on level to nearly level areas of the lower Coastal Plain. They are poorly drained; slow runoff; moderately rapid or moderate permeability. Water table rises to depths of less than 0.3 m for one to four months during most years. It is at depths of about 0.3 to 1.0 m for periods ranging to about six months, but during the driest season it recedes to depths of more than 1.0 m. These soils were formerly included in the Leon series and classified in the Ground-Water Podzol great soil group. A more complete description of these soils is given in Appendix B . Vegetation on the site is a typical flatwoods vegetation type as described by White (1973) . The vegetation associated with a flatwoods is considered to be fire climax, due to the frequent natural occurrence of fire and the presence of fire adapted species (Laessle, 1942) . Pinus palustris was the dominant overstory tree, however Quercus incana is present in some numbers in the southern part of the area. The average basal area and canopy cover for the entire site is 6.47m^ per hectare (ha) and 50.3%, respectively. The predominant understory shrub is Serenoa repens, with an average canopy cover of 4.5%. Aristida stricta is the predominant understory plant with an average frequency of 13.6% for the entire area, however, its occurrence is con- siderably below this figure for most of replication III. There were two distinct vegetation types on the site, a more mesic appearing vegetation was to the north and a more xeric type to the south. The ecotone between these two types was quite narrow lying between three and six meters across. Actual boundaries correspond very well to the soil map that was drawn based on soil cores (Figure 4.1.) • A listing of some species found at the study site with common names and other miscellaneous information is presented in Appendix A, Tables A. 2., A. 3., and A. 4. A weather station is maintained by the University of Florida and IFAS at the BRU. A monthly summary of these data for rainfall, minimum and maximum temperatures are presented in Appendix A., Table A. 5. Average long term weather data is not available for the BRU. The 70 year average for Gainesville will be used for comparison. The average annual temperature and precipitation for the Gainesville area are 21.2°C and 138.8 cm, respectively. This places the Gainesville area in the subtropical moist forest life zone classification system (Holdridge, 1967) . In 1976 the annual precipitation was 85% of normal and in 1977, 63%; the mean temperatures were slightly below normal (1°C) , for both years. While the mean annual temperature was near normal the monthly means were below normal in the winter (January 1977) and above normal in the summer. The distribution of rainfall was similar, with more rain falling in the winter and less in the summer. On a three year comparison (1975 to 1977) , the BRU appears to be about 1°C cooler (yearly average) than Gainesville. The precipitation is more erratic but is about the same as in Gainesville. Man induced inputs are essentially in the form of mineral supplementation to the grazing cattle. This input, especially in the case of P, may represent a significant portion of the total P content in the soil profile (Chapter 13). Field practices used on the cornfield to the north of the site might have resulted in drift of material (fertilizer, lime, and dust) onto the site. There 19 is some evidence to support this (Chapter 14) . Man was also responsible for burning the area in 1974 and 1976. However, since this is a "fire climax" (Laessle, 1942), no substantial changes in community structure should have been introduced by this factor. The fire lanes and the old logging roads did cause vegetational changes that are apparent. These areas have a greater preponderance of invader species (e.g. Rubus spp . and Paspalum notatum) than the main areas (Chapter 9). Other than the above effects, man's impact has been relatively minor. Methods Statistical treatment for this experiment was according to procedures given by Steel and Torrie (1960). A randomized complete block was the basic experimental design. The analysis of variance (ANOVA) was used to determine the "F" value and the means were separated by Duncans New Multiple Range Test. Where only two populations were being compared, a standard t-test was used. The level of probability was P < 0.05 unless otherwise stated. Equations used for fitting data (regressions) were of three types, linear, multiple, and quadratic. The form of these equations are: linear Y = a x + an (1) o 1 multiple Y = aQ + a^z + a-ps. (2) quadratic Y = a + a-, + a£ (3) Closeness of fit was determined by the coefficient of determination (r^) (Steel and Torrie, 1960) . Diversity indices were calculated by the Shannon-Wiener equation (Shannon and Weaver, 1949) : H' =T(ni/N) log(ni/N) (4) ni is the total number for each specie and N is the total number of individuals. 20 The evenness component was calculated by: J' = H'/log s (5) where s is the total number of species present (Pielou, 1975) . The equation used to convert percent in vitro organic matter digesti- bility (IVOMD), percent organic matter content of forage (OM) , animal dry matter intake (IM) to digestible energy intake (DEI), was obtained from Dr. J. Moore, Professor, Animal Science, University of Florida, 1978. The equation for DEI (in Mcal/da/animal) is: DEI = (IM)(M) (IVOMD) (4.4) (6) where IM is in Kg/day/animal, OM and IVOMD are in percent, and 4.4 is the conversion factor from Kg of dry forage matter to Meal. The equation to estimate dry matter intake (DMI) from fecal output (FO) and IVOMD was adapted from equations described by Morris and Kover (1970) and Mott (1973) : DMI = (100)(FO)/(100-%IVOMD) (7) where DMI is in Kg/pasture, and FO is in Kg/pasture. Soil samples were collected, placed in paper sacks, and air dried (25°C) in the University of Florida Range Laboratory, then passed through a 850 micrometer (0.0331 in) screen (U.S. standard testing sieve #20). The sample was then sent to the Soil Testing Laboratory, University of Florida, for chemical analysis. The standard procedure used by the Soil Laboratory is a double acid extract as described by Nelson et al. (1953). Phosphorus was determined by the Murphy-Riley (1962) method. Other elements were determined by the Technicon Auto Analyzer, Atomic Absorption or Flame Emission, depending upon the element requested. Soil water pH was determined by a 1:2 soil to water ratio, stirred to form a slurry just prior to inserting the electrode. The procedures used by the University Soil Testing Laboratory are outlined by Sabbe and Breland (1974) . Chemical analysis of soils are reported in parts per million (ppm) . 21 The in vitro organic matter digestibility (IVOMD) used in this study is the two-stage technique developed by Tilley and Terry (1963) , with modifi- cations as described by Moore and Matt (1974, 1976). This procedure is the same as is used by the Animal Nutrition Laboratory at the University of Florida, a detailed description of the procedure is presented in Appendix C. Briefly, the procedure consists of weighing approximately one half gram of the ground sample into a 50 ml centrifuge tube, innoculation with rumen fluid, mixed with a buffered artificial saliva and incubation for 48 hours at 35°C. The second stage consisted of addition of HC1 and pepsin then another 46 hours of incubation. The contents of the centrifuge tubes were then filtered through glass wool in a Gooch crucible, dried at 70°C and ashed at 500°C. The organic matter content was determined by ignition and the digestibility calculated. The rumen fluid was obtained from a steer maintained by the Animal Nutrition Laboratory, University of Florida. This steer was fed a constant diet of Coastal bermudagrass hay (Cynodon dactylon) , soybean meal, minerals and vitamins, throughout the duration of the experiment. Results of the IVOMD are expressed as percent digestibility. Samples for foliar analysis were placed in open beakers and oven dried at 70°C for 24 hours to remove moisture. A half gram aliquot of the sample was placed in a crucible, ashed in a muffle furnace at 500°C for four hours, cooled and weighed. Ten milliliters of 40% HC1 were added to the crucible, followed by evaporation, cooling, addition of 3 ml of concentrated HC1, evaporation, cooled, addition of 13 ml 0.1 N, HCl and standing for 18 hours. The sample was then transferred and brought up to 25 ml volume, filtered and stored in plastic vials. The sample was then sent to the Soil Testing Labora- tory for analysis. The analysis is basically the same as that described for soil samples after extraction by the double acid process. Chemical values for foliar content are reported in parts per million (ppm) . 22 The amount of forage in each pasture prior to cattle entry was estimated by the use of clipped plots and a Neal Herbage Meter (Model 18-2000). Each pasture was traversed by four equally spaced transects running north and south. Each transect was divided into ten equally spaced sampling points. Each sampling point was permanently marked and numbered. The dimensions of the sampling quad- 2 2 rat and the herbage meter were the same, 0.186 m (2 f t ) . Golf tees were used to mark the position of the legs of the herbage meter on the ground so that accurate positioning of the meter would be possible throughout the course of the experiment. There were 40 permanently marked quadrats. Each quadrat was classified as to wet, dry, or moist, based upon the presence or absence of certain indicator species. The indicator species for wet were: Ilex coriacea, Amphicarpum muhlenbergianum, Panicum hemitonom, and Centella asiatica. The species used for dry site determinations were: Asimina spp. , Anthaenantia villosa, Erecmochloa orphiuroides, Sporobolus junceus, Triplasis americana, Rhychosia spp. , and Diodia teres. In the presence of any of the above species the quadrat was considered as wet or dry, respectively, In the absence of the above species the quadrat was considered moist. The pasture was partitioned into percent wet, dry, or moist, based upon the per- centage of the quadrats in each category. Initially, eight quadrats were to be clipped prior to cattle entry and again after leaving the pasture. This was reduced to four in November 1976 and raised to eight again in June 1977. The procedure for determining the location of the clipped plots was as follows: The herbage meter was first calibrated to bare ground, then a meter reading was obtained for each permanently marked quadrat. The difference be- tween the high and the low meter reading was divided by the number of quadrats to be clipped (four or eight) . This interval (D) was then added to the lowest meter reading, the resulting value was then compared to the list of readings 23 taken for the pasture, the quadrat with closest meter reading to this calcu- lated value was selected. Two intervals (2D) were then added to the lowest meter value, this new value was compared to the list and the closest matchup recorded. This procedure was followed (3D, 4D, etc.) until all quadrats to be clipped had been accounted for. In case two or more meter readings were equally distant from the computed value, selection was made on the basis of having each line represented by at least one clipped plot. This procedure gave a random selection for clipping, over the range of meter readings. The meter was then taken to the selected plot and the vegetative composition noted (ocular) of the permanent quadrat. A similar community was then selected close by, and the meter positioned until the same value was obtained as noted in the permanent quadrat. This plot was flagged and clipped. The permanent quadrats were not clipped during the course of the experiment. Each clipped plot was sorted as to species (standing dead material was included), placed in paper sacks and marked as to specie, date, plot, and pasture location. Litter was also collected with large branches removed. The samples were then taken to the laboratory and dried at 70°C for three days, and weighed. Species identification was double checked at this point. The samples were then stored at room temperature until needed. Two separate tallys were made, species and weights, and a grouping as to grazables, un- grazables, and litter with corresponding weights. Shrubs and trees were generally considered as ungrazables; Aristida stricta, other grasses, and most forbs were considered as grazables. Linear regression equations were developed with the grazable portion, meter readings, and the average available forage calculated for each moisture type in the pasture. Based upon the percentage of each moisture type in each pasture total available forage was determined. Assuming 2.5% forage dry matter required 45.4 Kg (100 lbs) animal live weight, the number of animals 24 and the length of stay in each pasture was determined, assuming 50% utilization of grazable forage. There were eleven animals available for use. Five or six animals were placed in each pasture, in an effort to keep the duration of grazing to six days or less (Mott, 1973) . Consequently, grazing in all replications for a particular treatment was not simultaneous. The cattle were weighed prior to entry into each pasture and again upon leaving. The experiment was designed to run for one full year starting on 7 July 1976, however, due to delays in sampling and other aspects discussed in Chapter 9, the final sample was collected on 14 September 1977. Dates of sampling are presented in Appendix A, Table A.l. CHAPTER 4 GENERAL SOIL STUDY OF THE SITE Introduction Nutrients cycle through ecosystems with the type of cycle depending on the nutrient of interest. There are generally two different cycles, complete and incomplete (Deevey, 1970). A complete cycle is one in which the element exists in all the major components of the earth; the atmosphere, lithosphere, hydrosphere and the biosphere. Nitrogen, carbon, sulfur, and oxygen are examples of complete cycles. Where one of the above components is essentially missing it is an incomplete cycle. Since, for all practical purposes, P does not volatilize it is considered as having an incomplete cycle. Along with energy flows, the cyclic nature of elements is responsible for the continued productivity of the worlds ecosystems and, in particular, allows fertilizers to be used to economic advantage (Curlin, 1970; Deevey, 1970; Golley, 1975) . A major factor in determining the quality and quantity of plants is the soil (Brady, 1974; Odum, E., 1971). However, in Florida, the sandy soils pro- vide minimal nutrients in their natural state (Geraldson, 1977) . In the native longleaf pine (Pinus palustris) forest of Florida nutrients cycle through the site from soil to tree to litter to soil, in a much simplified description of the actual cycle. In a mature stand it takes between eight and twelve years for a forest floor to reach stability with respect to litter buildup and decomp- osition, in the absence of fire (Keyword and Barnette, 1936) . Laessle (1942) recognized that the natural pine flatwoods of Florida were dependent upon fire to maintain a certain stability, and described the forest as a fire climax. 25 26 Fire reduces the decomposition time and, in effect, speeds up the nutrient cycling of the system, thus altering the soil characteristics (Ralston and Hatchell, 1971) . Leithead (1973) has recognized seven woodland suitability groups in Florida that are of major concern. Of these, the flatwoods type soils occupy approximately one-half of the land area in Florida (Smith et al., 1967). These soils contain very low levels of available nutrients (Yuan, 1966; White and Pritchett, 1970; Pritchett and Smith, 1974). Phosphorus is believed to be the most limiting element in these soils (Yuan and Breland, 1969; Ballard and Pritchett, 1974) . Phosphorus does move in the soil, it is just that the amount moving is small, relative to the amount present (Thomas, 1970). The rate of loss by stream flow and leaching is not large; Duffy et al . (1978) reported an average of 0.027 ppm from five different watersheds in Northern Mississippi; Thomas, summarizing from streams flowing over high phosphate containing limestone, reports ranges from 0.09 to 0.22 ppm. The rate of movement of P is also dependent upon the amount of fertilizer applied (Rodulfo and Blue, 1970), and has been shown to be related to amounts of Fe and Al present (Yuan and Breland, 1960; Ballard and Pritchett, 1974) . The loss of P from native systems in Florida can probably be considered minimal unless Fe and Al are in relatively low concentrations. One avenue of loss that has not received too much attention is the loss associated with the removal of timber. Bole wood accounts for about 28% of the total P in slash pine (White and Pritchett, 1970) . The actual amount of P removed would be a function of the total biomass removed from the site. Pulpwood production takes even a higher percentage of the total tree, and if the current trend toward total harvest continues, this figure will rise higher. There is some evidence that second growth pines do not do as well as first growth unless fertilized (Pritchett, 1976). The above line of reasoning would indicate that unless some sort of fertilization is practiced, the fertility of these soils can be expected to decline with continued use. Range grazing systems contribute relatively little to the loss of minerals (Pieper, 1974) and if minerals are supplied to the animals, grazing may repre- sent a net gain (Chapter 13) . Other forms of P input to the system are from parent material and atmospheric inputs. The parent materials in Florida are marine deposits of sand, with some phosphate bearing layers that are presently being mined and shipped to out-of-state locations, however, there is little P that is being added to Florida rangelands from this source. The atmospheric inputs of P are minimal (Chapter 13) . The effect of wind-throw on the redistribution of nutrients in a mature forest has been discussed by Lutz (1940); Goodlett (1954); and Drury and Nisbet (1973) . This effect is probably minimal in Florida since there is little evidence that P accumulates at soil depth. Also the management practice of harvesting young trees for pulpwood rather than older ones for saw timber, would result in less trees susceptible to wind-throw. Materials and Methods T — — Soil samples were collected from each pasture in April 1977. Three equal distant lines were set up in each pasture and five soil samples were taken along each line, at equal intervals. Samples were collected from three depths (0 - 10, 10 - 20, and 20 - 30 cm) with a 2 cm diameter soil core. Five separ- ate cores were taken at each sample point and composited for each depth. A total of 15 composited samples were taken at each depth for each pasture, a total of 45 samples per pasture. The samples were analyzed at the University of Florida Soil Laboratory for pH, P, K, Ca, and Mg. Statistical analysis of the results consisted of taking the average, standard deviation and CV at each depth for each pasture, for each of the six 28 soil parameters. These means were then used in a randomized complete block design and tested for replication and treatment effects. Pasture averages were calculated, using the total of all samples at three depths; analysis of variance (ANOVA) was also computed for each of the six parameters. Two differ- ent groupings of pastures were used in both of these analyses. The first was by grazing treatment and replication, the second was as to physical location of the pasture on the site. In March 1978, additional soil samples were taken in replication IV (the control, and four -month rest) . The entire 30 cm profile was sampled at one time; there were 15 replicated samples per pasture. A t-test was used to compare these results to those from the same pastures in the 1977 sample. All three depths at each sample location were composited in the 1977 samples, thus giving 15 values for the 30 cm soil profile. To check on possible effects of trampling, bulk density determinations were made in February 1977 for replication I (the control and two-month rest) . Fourteen samples were taken in each pasture. Each sample position was ran- domly selected. A t-test was used to analyze the results. In April 1977 ground water wells were placed in all but the six-month rest pastures to measure evapo-transpiration rates and water quality. The wells were dug with a bucket auger and four 10 cm (4 in) diameter PVC pipes were positioned. The depth of the wells varied from pasture to pasture, depending upon the depth to the spodic horizon. The depths of the wells were such that this horizon was penetrated at least 5 cm (2 in) . Three water level recorders were obtained to record fluctuations in the water table. The initial procedure was to have a recorder on each well in a replication for a period of one week; then moved to another replication. Two weeks after installation the wells dried up, except for two wells located on the east side of the site. These two wells went dry about a month later. 29 Each well was to have been sampled for water quality every two weeks after installation; again, the drying up of the wells precluded this. Conse- quently, only three samples were taken on an irregular basis, depending upon presence of water in the wells. Differences noted in the soil profile during construction of the wells prompted a more detailed coring of the area. The main area of concern was the ecotone between the two vegetational sites, the dryer site to the south and the wetter to the north. Cores were taken to a minimum of 1.4 m (4.5 ft) until well within the spodic horizon; 36 cores were taken on both sides of this ecotone. Five representative cores were chosen and samples taken, start- ing with the surface, at 0.3 m (one ft) intervals until the spodic horizon was reached. These samples were analyzed for pH, P, K, Ca, and Al. Results Average soil parameter data for three different depths (0 - 10, 10 - 20, 20 - 30 cm) with coefficients of variations (CV) and results of ANOVA analysis are presented in Appendix D, Tables D.l. through D.6., for organic matter, pH, P, K, Ca, and Mg, respectively. Of interest is the extremely high CV recorded for P, an average of 109%, with ranges from 47 to 195%. Phosphorus had the lowest variation with an average CV of 6.8%. The other parameters were moderately active with average CV's in the 30 to 40% range. The ANOVA analysis failed to show up any significant trend at any of the three depth levels for all components as a function of grazing treatment. A general trend was noted in that replication IV was usually lowest in nutrient content when the different averages were ranked, and replications I and II were generally higher. A different method of grouping for ANOVA analysis was made (not shown); this grouping reflected the actual physical location at the site. Again, no meaningful results were noted. When the averages were ranked, there 30 appeared to be a general decline in all parameters from the north to the south and also from the west to the east. The analysis of the average soil content (all three depths composited for each pasture) was conducted with two different groupings of the pastures (physical location and by treatment and replication). This analysis showed trends similar to that observed for each soil depth. When the pastures were grouped as to their physical location (Table 4.1.), all components except P and pH had higher values along the corn fields to the north. Both Ca and K were different at the P < 0.05 level. No column effects were found to be significant. However, all components except Mg were lowest in those pastures on the east side (nearest the creek) . The ANOVA analysis, based upon grouping the pastures as to grazing treat- ment and replication (not shown), indicated meaningful treatment and replication effects for K and organic matter. The four-month and the control were higher in K than the other treatments. Organic matter was highest in the four -month treatment. Replication I had the highest levels of K and organic matter. The other components (except P) were also higher (not different at the P < 0.05 level) in replication I, generally followed by replication II. Efforts to relate, through linear regression (eq. 1), the different soil components failed to arrive at a meaningful coefficient of determination (r2) . The highest r2 computed was 0.32 for Ca and Mg at the 20 to 30 cm depth. All other comparisons yielded r2 values less than 0.27. Due to the' high variation within and between pastures, an effort was made to determine which of the four replications were most alike, with respect to the different soil parameters. A t-test was used on a replication by repli- cation comparison using the average value of each component for each pasture, and on various groupings of replications. The basis of selection was that there was not a difference at the P < 0.1 level. The result of this comparison 31 Table 4.1. ANOVA analysis, based on the pastures actual physical relation to one another, for average soil data. Column Row Row Row Organic matter (%) 3.7 2.7 2.3 1.7 2.6 2.7 2.0 1.3 1.3 1.8 2.3 1.7 1.7 2.0 1.9 1.7 1.7 1.7 2.0 1.8 2.6 2.0 1.8 1.8 Row F = 2.865 (ns) Column F = 3.115 (ns) pH 4.47 4.33 4.30 4.33 4.36 4.53 4.63 4.77 4.30 4.56 4.53 4.73 4.80 4.13 4.55 4.90 4.43 4.47 4.17 4.49 4.61 4.53 4.59 4.23 Row F = 0.959 (ns) Column F = 3.422 (ns) 1.7 2.7 4.1 1.7 2.6 3.8 8.3 8.4 1.5 5.5 6.7 4.3 4.6 0.5 4.0 12.9 10.9 2.1 0.9 6.7 6.3 6.7 4.8 1.2 Row F = 1.217 (ns) Column F = 2.319 (ns) Note: Numbers followed by same letters or absence of a letter are not significant at the P < 0.05 level. F values followed by ns are not significant at P < 0.05 level, other entries indicate the level of significance. Table 4.1. - continued Column Ca Row X 154 141 120 123 134a 124 109 103 102 110 79 129 102 73 96 111 102 98 98 102 117 120 106 99 w F = 5.249 (. 025) Column F = 1.797 (ns) Row 11.7 9.3 8.0 8.7 9.4a 8.0 6.3 6.0 5.7 6.5 6.0 6.3 8.3 5.0 6.4 5.3 5.7 5.7 6.7 5.9 7.8 6.9 7.0 6.5 Row F = 6.426 (.025) Column F = 0.651 (ns) Mg Row X 30.3 14.7 7.7 8.3 15.3 18.7 11.1 13.7 7.7 12.9 13, 9. 8. 9. 10. 13.7 11.0 9.3 13.7 11.9 19.1 11.5 9.9 9.9 Row F = 3.664 (ns) Column F = 0.810 (ns) 33 (not shown) indicated to this investigator, that replication I and II were the best combination to use for Ca, K, and Mg, and replication I and III were the best combination for P, pH and organic matter. The comparison between the four-month rest and the control in replication IV for P, Ca, and K for the April 1977 collections and the March 1978 collections showed an increase in all three components (Table 4.2.). The bulk density of the two-month treatment in replication I was 1.38 g/cc and that in the control was 1.29. This was not different at the P < 0.05 level. Two water level recorders were set in position on 1 May 1977, in repli- cations II (four-month rest treatment) and in replication IV (control) . Five days later the well in replication II went dry. The well in replication IV went dry on 4 June 1977. The rate of discharge was computed to be 0.7 cm per day for both wells. The water level in both wells was over a meter from ground surface when the recorders were set up. The water level recorder trace made by the well in replication II was a stepwise function, but could not be construed as a daily transpiration pattern due to irregular fluctuations. The trace in replication IV was smooth. On the afternoon of 8 May 1977, 3 cm of rain fell. The exact time is not known since only one collection of weather data is made per day (8:00 a.m.) at the BRU farm. At 4:30 a.m. on 9 May, the water level in replication IV started to rise and 7.2 hours later the level in replication II started to rise. Six days were required before the levels in each well returned to their original levels. The rate of water movement through the soil was calculated to be about 18 m per hour. This same kind of water level increase was noted on two other occasions, 27 May and 25 June. In both instances the amount of precipitation was greater than 2.5 cm and again there appeared to be a lag from the time the rain fell to when the recorder measured an increase. The 34 Table 4. 2. Comparison of soil data collected in April 1977 with that collected in March 1978 for P, Ca , and K. P Ca K 1977 1978 1977 1978 1977 1978 4 month rest X (ppm) 0.83 2.23 97.5 140.0 6.3 12.0 s (ppm) 0.78 0.97 27.2 35.1 2.1 5.9 CV (7.) 94.20 43.70 27.9 25.1 33.6 48.9 Level of Signi f icance .001 .001 010 (P<) Control X (ppm) 0.46 1.77 72.9 113.4 4.9 8.9 s (ppm) 0.41 0.66 13.0 29.3 1.2 2.3 0.46 1 77 0.41 0 66 89.60 37 10 CV (7.) 89.60 37.10 17.8 25.8 23.7 26.1 Level of Significance .001 .001 .001 (P<) 35 well in replication II was abandoned on 23 May and that in replication IV on 30 June 1977. Due to the muddy condition of the water in the well, and the general low level of water, samples were taken for three dates in May only. The P content was less than 0.05 ppm, K was 0.6, and Ca 2.75 ppm. Based on the results of the soil cores, two separate soils were determined to exist on the site. Sample profiles of each type taken approximately 30 m apart are shown in Figure 4.1. Table 4.3. shows the average values for pH, P, Ca, K, Mg, Fe, and Al, for the two types based on depth from the surface. Since the intervals where samples were obtained did not necessarily fall in the same horizon in the two different soils, the only comparison made was with the spodic horizon. The spodic horizon occurred at about 1.5 m (5 ft) in the wet type (soil lying to the north) and about 1.4 m (4.5 ft) in the dry type (soil lying to the south) . The spodic horizon in the wet type was of a sandy nature and that in the dry was more clay-like which is reflected in the high Al content in the spodic horizon of the dry type soil (Table 4.3.). The dry type spodic horizon also had a mottled appearance with visual evidence of iron. A t-test indicated that there was a higher (P < 0.05) Fe and Al content in the spodic horizon of the dry site than the wet site. The content of Mg and K were found to be higher (P < 0.01) in the dry site. P, Ca, and pH were not found to be different at this level. A linear regression (eq. 1) was used to o relate soil P to soil Al and an rL of 0.32 was obtained. A multiple regression 9 (eq. 2) with Al and Fe as the independent variables resulted in an r of .502. Coring on both sides of the ecotone determined that vegetational differ- ences were associated with the different soil types. Figure 4.2. delineates the boundary between the two different soil types. Coring was not conducted on the vegetational differences noted along the far eastern side since this difference was believed to have been due to a moisture gradient rather than soil differences. 36 •Wet Dark Sand Al Light Colored Sand White Sand Cley Send Derk Colored Spodic ■Surface 15cm 30cm 60cm 90cm 120cm 150cm 180cm •30m Dry Dark Sand Al Light Colored Sand White Sand Mottled Spodic Figure 4.1. Soil profile of the wet and dry sites along vegetational ecotone. 37 c -u c o> O J2 a *-» e O U-l o o tn a> 3 '4J O c •H ■H ^ w 01 > je i— > cm cm r-< O r- co r-» co co O m CM co oo o O sfr £> on CM --. u Ci > 38 1,11,111 & !V--- nEPLiCA.iON 2,4,6 a 12---M0NTHS REST Figure 4.2. Soil map based on soil cores showing boundary of the two different soil types. 39 The soil survey that was made available on 9 May 1978 (Chapter 3) de- fines the wet soil as an Adamsville series and that of the dry type a Sparr series. The boundaries drawn from this survey conform to the boundaries determined in this study. Discussion and Conclusions The high variability of the P content of these soils is believed to be associated with the low amount of P noted and is a real phenomenon that occurs on sandy flatwoods soils. A discussion of this variability is delayed until the conclusion of Chapter 6 (micro-variability) . The general trends noted appear to be associated with the physical location of the pastures at the study site, rather than the treatments that were imposed upon them. The major reasons are the field practices used in the cornfield to the north, the presence of the creek to the east, the four different soils present in the site and the manner of distribution, the length of time required for fecal material to move into the soil profile (Chapter 11), the irregular weather pattern that existed during the experimental phase (Chapter 3) , the apparent slowdown of the general decomposition process due to the cold weather and the lack of moisture, and the relatively short period of time the experiment had been in progress prior to soil sampling. There also appear to be effects that are not explained by the data gathered. In particular, the increase noted in the soil chemical content that were sampled in March 1978. This increase was noted in two pastures and was not apparently related to treatment. However, there were only two pastures sampled, and the method of sampling was slightly different; whole cores being used rather than the average of three separate depths. This difference in sampling procedure might account for the differences noted. In the absence of replications, it is impossible to draw a conclusion. 40 Failure to obtain any relation between any of the soil parameters may be due to the parameters that were compared and the method of comparisons rather than a lack of relationship. Had a multiple regression (eq. 2) been used, a better correlation might have been forthcoming. This is, in part, borne out by the comparison of P with Fe and Al (eq. 2) and the resulting r2 of 0.502. This correlation is less than that reported by Yuan and Breland (1960) . These 2 authors reported r values of 0.76 or higher for spodosols in a multiple re- gression with Fe and Al as the independent variables. Only 13 samples were available for comparison in this study. Future studies should report the Fe and Al levels since these have been linked to P retention in Coastal Plain soils (Ballard and Fiskell, 1974) . The results of the soil cores, delineating the boundary of the two vegetational types into two distinct soil types (Adamsville and Sparr) were borne out by the later soil survey of the area. The boundaries determined in this study are somewhat different than those reported by the soil survey. This difference is explained by the sampling techniques used by the survey team. Samples are taken over an area at intervals of about 91.4 m (300 ft) or greater. Consequently, the position of the boundary has this built-in variation. The boundaries defined in this experiment (Figure 4.2.) lie well within the range of precision of the survey team. The recommendations made as to which replications are most alike are very general in nature and are offered as guides for later comparisons. Almost any combination could have been made with some kind of justification, depending upon which soil component was of interest. In the opinion of this investigator, replication I and II are most alike with respect to Ca and K, and replications I and III with respect to P. 41 Considerable effort had gone into the installation of the wells in the hope that evapotranspiration rates could have been measured, along with water quality. The drying up of these wells precluded the accumulation of this type of data. It does appear that ground water recharge does occur from the adjacent creek, and that rainfall of the order of 2.5 cm is required before this effect is noted. There also appears to be a time lag associated with the time of rainfall and the time that it appeared in the soil water table. Observations of the wells indicate that the wells on the western side of the site were the first to go dry, the eastern side last. Chemical analysis of the water samples were in general agreement with those reported in the literature. The dry site (Sparr) has a clay type hardpan or spodic horizon lying above the spodic horizon of the wet (Adamsville) soil. These two effects combine to form the dryer site. Ground water moving from the creek would tend to flow underneath the Sparr soil. Thus, the plants on the Adamsville soil would have water available to them longer than those on the Sparr soil. The clay horizon acts as a barrier to root penetration and upward movement of any water below this depth. This horizon would not be expected to be equally continuous, and fractures or ruptures (possibly caused by wind-throw) would alloxtf areas or islands of wetter vegetation to exist side by side with the dryer types. This island effect was observed on the Sparr soil. The low levels for P at this site are not unique for Florida flatwoods. Pritchett and Smith (1974) reported a P content of 0.2 ppm for a site prepared, but unfertilized, flatwoods site in site in Western Florida. Rudolfo and Blue (1970) reported a P content of 48 ppm on a virgin flatwoods area at the BRU; they did not report on the vegetative cover present at the time of sampling. Koger et al. (1961) report 15.7 Kg/ha of P present in an open young longleaf pine stand, similar in appearance to that at this site, except that the trees are much smaller. Using a bulk density of 1.31 g per cc, and procedures given 42 in Brady (1974), this converts to a 7.9 ppm of soil P. Numerous references exist for the general low content of P in Florida flatwoods soils (Yuan, 1960; White and Pritchett, 1970) . Phosphorus is generally considered to be limiting on most of these flat- woods areas (Yuan and Breland, 1960; White and Pritchett, 1970). Since P is known to increase as the standing biomass increases (Rodin and Basilevich, 1965), there have been efforts to develop equations predicting the amount of nutrients present in trees (White and Pritchett, 1970) . These equations relate the amount of P present as a function of size of the tree. Since size of the tree is a function of age, the amount of P in the over- story is also correlated with age of the stand (Curlin, 1970) . Unless P was being obtained from the parent material or other sources faster than it was being taken up by the vegetation and stored, there would be a net decline in P availability as the stand matures. This would continue until the system attains an equilibrium, as much new biomass and hence P, added as was deposited as litter (Grier and Logan, 1977) . This implies that a native flatwoods site would become progressively less fertile as the trees on it grow to maturity. Harvesting the tree overstory also represents a drain on the available nutrient supply, unless fertilization is practiced. The flatwoods of Florida have been cut over at least once and perhaps more; old stumps in the study site attest to this. There is reason to believe that the flatwoods were once more fertile than they are at present. A possible line of inquiry would be to determine whether if, in fact, soil fertility is related to age of stand of timber and to determine this rate of decline. This potential decline in soil fertility has implications for all aspects of range management. There is reason to believe that the quality of the native plants growing on these sites is not as high as those growing on different soils 43 (Chapter 8) . Forage quality with respect to deer foods indicate that the herd size is much less than would be expected from a consideration of the plant biomass alone (Chapter 10) . The livestock and the timber industry would also be affected. More effort should be directed toward relating soil fertility and forage quality for Florida conditions. CHAPTER 5 CHANGES IN THE CHEMICAL CONTENT OF SOIL SAMPLES Introduction Soil samples are commonly stored in paper sacks and air-dried at room temperatures prior to analysis. Other methods of handling samples include sun-drying, oven-drying, or refrigerating in plastic bags. Extreme methods include oven-drying samples at temperatures in excess of 200°C and even placing samples in frying pans on a heating element to speed the drying process. A sample often is inadvertently re-wetted by rain, dried, and then submitted for analysis. Reports in the literature on changes in chemical properties of soil as related to different treatments of the sample are generally restricted to the analysis of nitrogen and phosphorus. For example, oven-drying and crush- ing soil samples (McKenzie and Kurtz, 1976) increased the accessibility of the energy source for the denitrif ication process. Organic matter (Brady, 1974) influences physical and chemical properties of soil far out of proportion to its low content in the soil. The amount of organic material (Birch, 1959) that is extracted from a soil sample by water increases as it's storage time in the air-dried state lengthens. Lebedjantzev (1924) used the increased availability of nitrogen and phosphorus after wetting a dry soil to explain the apparent growth response of plants following a rain. Stevenson (1965) observed a higher metabolic activity after air-drying and re-wetting a soil sample. Birch (1960) noted that extra nitrogen was available upon moistening the soil sample. 44 45 Mesophiles are most of the soil microflora, while thermophiles are generally restricted to soils that receive additions of composted material (Dickinson, 1974) . Optimum temperatures range from 25°C to 37°C for the growth of mesophiles and from 55°C to 65°C for the growth of thermophiles (Dickinson, 1974) . Mesophiles should undergo stress as environmental temper- atures approach 35°C and experience increasing mortality as temperatures continue to rise. Many proteins are denatured as the temperature approaches 50°C (Leopold and Kriedeman, 1975) . Consequently, various organic substances are released and are available for chemical combination with soil particles as the temperature increases . The objectives of this study were (1) to assess changes in chemical properties of a Sparr sand as affected by treatment of the sample prior to analysis and (2) to define a procedure for handling field-moist samples from this and related soils of Florida. Materials and Methods A composited soil sample, the surface 10 cm of Sparr sand at the BRU, was collected and placed in a large plastic bag. The sample was sieved, mixed and split into 50 sub-samples. Groups of 10 samples each were randomly selected for each of the following treatments: samples refrigerated in plastic bags at 6°C; air-dried in paper sacks at 25°C; oven-dried in beakers at 36, 69, or 105°C for five days. All samples were then placed in plastic containers until analysis. The following components were determined: pH, P, Ca, K, Mg, Fe, and Al . Percent organic matter expressed as air-dry basis, was determined by sieving and mixing ten samples from the original bulk sample and ashing at 500°C. After initial analysis, three groups of samples (air-dried, oven-dried at 69°C, and oven-dried at 105°C) were taken and wetted with distilled water 46 and allowed to stand for 24 hours. Each group contain ing the same ten sub-samples as in the initial run was then dried for five days at the same temperature as it had previously been dried, and analyzed as indicated above. Statistical analysis utilized a completely random design to determine significant differences at the 0.05 level. Duncans New Multiple Range Test was then used to separate the means. Coefficients of determination (r2) and coefficients of variability (CV) were determined. Regression analysis was attempted to determine the best possible fit of the data. Results Results of the analysis for each group of samples are presented in Table 5.1. With the air-dried temperature (25°C) as the control, there was no significant difference at the 36°C drying temperature; phosphorus and iron had significant changes at the 69°C temperature, all elements showed signifi- cant changes at the 105°C temperature and pH, potassium and iron were signifi- cantly different at the 6°C temperature. Quadratic regression (eq. 3) were constructed from the above data with the resulting coefficients of determination (r2) ; 0.969 for phosphorus, 0.965 for potassium, 0.958 for calcium, 0.971 for magnesium, 0.995 for iron, 0.945 for aluminum, and 0.945 for pH . The derivative of each equation, except pH, was taken and set to zero and solved for the maximum or minimum. The pre- dicted minimum for phosphorus was 33°C and that for iron 18°C; the maximums for calcium, magnesium, potassium, and aluminum were 34, 36, 58, and 28°C, respectively. Figures 5.1. and 5.2. illustrate the predicted curves, and the maximums and minimums for each of the above elements. The effects of taking three of the above samples, air-dried (25°C) , oven-dried at 69 and oven-dried at 105°C, wetting and re-drying at the same temperatures as before are given in Table 5.2. With air-dried as the control, all components showed a significant difference when compared to the 105°C 47 Table 5.1. pH and some extractable elements of a Sparr sand as a function of drying temperature. Drying temperature . °c 6 25 36 69 105 PH 4.90 4.63a 4.64a 4.58a 4.10 CV (7.) 1.3 1.9 1.4 1.7 3.5 Phosphorus (ppm) 6.0a 5.6a 5.8a 7.0 23.4 CV (7.) 19.5 7.5 6.4 14.4 5.1 Potassium (ppm) 7.5 9.6ab 9.4a 10.2b 8.2 CV {%) 8.8 6.9 10.9 7.3 4.9 Calcium (ppm) 209.0a 202.0a 211. 0a 207.0a 151.0 CV (7.) 3.9 6.4 5.6 8.0 7.0 Magnesium (ppm) 40.8a 41.1a 42.0a 40.8a 30.2 CV (7.) 3.2 4.2 4.3 8.4 2.9 Aluminum (ppm) 244.0a 222. Oab 254.0a 225.0b 157.0 CV (7.) 5.6 7.4 5.6 7.7 4.6 Iron (ppm) 23.7 26.1a 25.8a 32.7 54.0 CV (7.) 7.4 1.9 1.0 3.0 2.8 Organic matter (7.) 1.5 1.5 1.5 1.5 1.5 CV (7.) 3.2 3.2 3.2 3.2 3.2 1. Organic matter was determined from the original composite sample. Note: Entries followed by the same letter are not significantly different at the (P< 0.05) level. pH = lj..8l|089 - 0.00312X - 0.00003X2 Mg = 39.0705 ■*- 0.l8i;iX - 0.0025X2 Fo = 25.5338 - 0.J439X v- 0.0039X2 48 4.8 4.6 4.4 4.2 x : max/min 60 40 5 20 20 40 60 80 DRYING TEMPERATURE s Deg. C 100 Figure 5.1. Change in chemical analysis for Fe, Mg, and pH as a function of drying temperature. 49 Al a 231.lij.89 + 0.8652X - O.O148X' Ca » 198.2835 *• O.836OX - 0.0120X' K = 6.9693 + 0.112JJC - 0.0010X2 P - 8.3677 - 0.2393X f 0.0036X2 - 250 O - 200 150 20 40 60 80 100 DRYING TEMPERATURE ^ Deg. C Figure 5.2. Change in Al, Ca, K, and P as a function of drying temperature, 50 Table 5.2. pH and some extractable elements of a Sparr sand after wettinj and re-drying the original samples. Characteristic 25 Drying temperature, °C 69 105 pH 4.72 CV (7.) 0.9 Phosphorus (ppm) 7.2 CV (7.) 18.6 Potassium (ppm) 8.8a CV (7.) 4.6 Calcium (ppm) 189.0a CV (7.) 5.9 >iagnesium (ppm) 3 9.7a CV (7.) 2.1 Aluminum (ppm) 230.0 CV (7.) 2.9 Iron (ppm) 21.0 CV (7.) 3.6 4.31 3.0 10.7 10.3 9.5a 12.7 201.0a 17.8 39.4a 8.0 219.0 4.0 32.3 3.9 3.94 1.2 21.6 6.9 7.1 9.9 155.0 8.8 33.6 5.7 165.0 4.9 46.5 3.3 Note: Entries followed by the same letter are not significantly different at the (P< 0.05) level. 51 drying temperature; when compared to the 69°C temperature, pH , phosphorus, aluminum and iron demonstrated a significant difference. Table 5.3. presents levels of significance and amount and direction of change (using the original analysis as base), due to the effects of wetting and re-drying the sample. The change in calcium was not meaningful at the 69 or the 105°C temperature. The other components showed changes at some probability levels at both 69 and 105°C temperatures. Discussion and Conclusions The high coefficients of determination (r ) of fitted quadratic equations for the elements and pH suggest that observed differences at the various drying temperatures are a product of one or more natural mechanisms. The time (five days) that the samples were kept at their respective temperatures allows ample time for biological mechanisms to manifest themselves. There is an interactive effect between decreasing moisture and optimum temperature for growth, but the time lag for moisture depletion appears sufficient for some metabolic activity to occur, especially at the lower temperatures. It is noted that the predicted minimum for phosphorus and for calcium and magnesium occur in the range of optimum mesophile growth (Dickinson, 1974) . Therefore, possibility exists that phosphorus uptake at these temperatures (30 - 37°C) fixes the element in a slightly less soluble form, thereby reducing the amount shown in the analysis. At higher temperatures this would not occur, as death and/or inactivation of the mesophile is more rapid due to dessication and high temperatures. The low phosphorus content of the soil and the small changes noted would tend to magnify any such fixation by various organisms. A Sparr sand is a loamy, siliceous hyperthermic grossarenic paleundult, has a Bh horizon and is poorly drained. Appendix B gives a detailed description of this soil series. At the time of collection the water table was at a depth 52 Table 5.3. The change from the original analysis (Table 5.1) and level of significance, as a result of wetting and re-drying the original samples. Characteristic - Drying temperature, C 25 69 105 pH 0.090 -.270 -.160 Level1 0.020 0.001 0.010 Phosphorus (ppm) 1.600 3.700 -1.800 Level1 0.001 0.001 0.020 Potassium (ppm) -0.800 -0.700 -1.100 Level1 0.010 0.02Q 0.001 Calcium (ppm) -13.000 -10.000 3.000 Level1 0.050 ns ns Magnesium (ppm) -1.400 -1.400 2.400 Level 0.050 0.400 0.001 Aluminum (ppm) 8.500 -6.000 8.000 Level1 0.020 0.400 0.050 Iron (ppm) -5.100 -0.400 -7.500 Level1 0.001 0.400 0.010 1. Refers to level of significance. 53 of two meters, due to the unseasonably dry conditions that existed in Northcentral Florida in 1977. A possible physical scenario to explain the results noted might be as follows: Some of the sulfides should be oxidized to sulfate as the sample is dried. Moreover, the amount of sulfate should be significantly higher in a sample dried at 105°C than in one dried at 69°C. The sulfate should combine with Fe and form hydrolytic acidity in an aqueous suspension. These reactions would explain the decrease in pH and the increase in content of extractable Fe and P (Table 5.1. and 5.2.) between a sample dried at 69°C and at 105°C. Exchange sites are probably destroyed by heating at 105°C and exchangeable cations are possibly converted to insoluble forms. The amount of extrac tables Al, Ca, and Mg (Table 5.1. and 5.2.) is signifi- cantly lower in a sample dried at 105°C than in one dried at 69°C. Apparently, the amount of hydrolytic acidity generated in an aqueous suspension of a sample dried at 105°C was insufficient to dissolve the converted forms of exchangeable Al, Ca, and Mg. The possibility of the oxidation of sulfides to sulfate and subsequent combination with Fe to form a hydrolytic acidity in an aqueous suspension, might explain the differences. Sulfide content of the samples was not determined, however, Neller (1959) reported in an analysis of 18 Florida soils 2 ppm of extractable sulfate in the top 9 cm. Mitchell-^ (1978, unpublished data) has found an average of 7.6 ppm of total sulfur in surface (13 cm) soils around the Gainesville area, all of which lie in the organic mantle. Neller (1959) found that sulfur leaches rapidly downward, due to the extremely low clay content of these Florida sands. Whether the low content of sulfur and related compounds found in Florida soils is sufficient to cause the proposed reactions remains to be tested. 3. Mitchell, R. 1978. Graduate student, Univ. Fla., Soil Sci. Dept, 54 Changes that occurred upon wetting the sample and subsequent drying appear to be a continuation of the above mentioned processes. The water would provide a medium for reactivation of the biological processes at the lower temperatures and dispersal of organic acids. Hydration of the various oxides also occurs along with another mixing prior to drying. A final possibility exists that because this soil contains very low levels of the elements studied, small changes, that might go unnoticed in more fertile soils, take on a significance that might be a result of the sampling or testing procedures. However, these differences are suspected to be manifestations of mechanisms that occur in many soils, but are only more noticeable in soils of low fertility. More experimentation on different soil types are indicated. Different drying temperatures and the wetting and re-drying of soil samples causes significant changes in any subsequent chemical analysis. However, several chemical properties of surface samples of a Sparr sand and related soils in Florida should not be significantly affected by treatment prior to analysis if field-moist samples are transported in closed plastic bags and dried quickly between 25°C and 36°C and analyzed as soon as possible. If the treatment of the sample deviates from this procedure, details of the actual treatment should be reported along with the data. CHAPTER 6 MICRO-VARIABILITY AND ITS EFFECTS ON SOIL NUTRIENT CONTENT Introduction The physical and chemical properties of the soil show small scale variations that may be as significant to the plant as it is unwelcome to those who make measurements on soils (Whittaker, 1975) . The present study was initiated after the main study had been completed and all data collected. In the analysis of the main body of soil data it soon became apparent that there was a great deal of variation associated with the different soil nutrients. The large coefficients of variation did not fit well with the notion that this site was a fire climax site and as such should have a climax soil type. The analysis of different methods of handling soil samples, discussed in the previous chapters, allowed for some variation due to handling methods of samples prior to chemical analysis. However, all soil samples were handled essentially the same in the preceeding chapters. Other possibilities were examined, errors arising in the analysis, various human errors, but all were discarded as either being consistent, or too small to account for the large variations noted. Gravity moves nutrients from higher elevations to lower, this is attested by the general observation that valleys are more fertile than the side hills. There is no reason why this same mechanism does not act similarly on a smaller scale. Water infiltrating into a relatively homogeneous soil is known to have a lateral movement (Amorocho, 1967; Branson et al . , 1972). Sinai and 55 56 Zaslavsky (1977) studied the effects of lateral flow in soils that had been wetted (rainfall and sprinkler) and noted non-uniform patterns of wetting in the soil profile. Simulation models were made and predictions verified on cultivated fields in Israel (Sinai and Zaslavsky, 1977) . The major factor thought to be contributing to the lateral flow was the anisotropic conditions found in cultivated soils as a result of plowing and mixing of the soil pro- file. The water movement appeared to be confined to the top layers of the soil, and was more pronounced in these layers if there was a slope present. The hypotehsis is advanced that part of the variation noted in the soils at the BRU are due to this dilution of nutrients in the small elevations and the accumulation of these nutrients in the depressions. Materials and Methods On 13 March 1978 the control (no grazing) pasture in replication IV was sampled to determine effects of micro-relief on chemical content found in the soil. An area was selected that had visual evidence of relief and a transit positioned and leveled. The line lay 130°E. of N. on an easterly slope of 1.2% fall. A stadia rod was used to measure elevation. The interval of horizontal distance was 15 cm (6 in) , 20 vertical measurements (elevation) were made for a distance of 6 m. At each point vegetative species were noted and soil samples from two depths (0 - 10 and 10 - 20 cm) were collected. The soil group at this particular site was a Wauchula series (Appendix B) . Soil samples were collected directly under the stadia rod by a 2 cm diameter soil core and stored in plastic bags (two cores were taken and composited for each depth) . They were immediately taken to the Range Laboratory, weighed, dried at 70°C for 48 hours, re-weighed and percent moisture determined. Each sample was split into two equal portions. One half was weighed, placed in a muffle furnace at 500°C for four hours, weighed and percent organic matter 57 calculated. The other half was sieved (850 micro-meter mesh) and sent to the University of Florida Soil Testing Laboratory for chemical analysis of P, Ca, and K. The elevations were plotted as a function of horizontal distance, and thij positions of rises and depressions noted. Chemical, organic matter, and moisture data were selected for the high and the low points and compared for the elevations or depressions, at both depths. Different groupings of data as to elevations or depressions were made, based on the elevation plot. These groups of data were compared for differences. The largest mound was selected and data plotted as a function of distance from the center. Quadratic equations (eq. 3) were used to fit the data. A standard t-test was used for comparisons. On 28 April 1978 an experiment was conducted on a Lakeland series soil near Keuka Lake, Florida. This series is characterized by deep sand, 2 m or more in depth. A description of this series may be found in Carlise and Moritz (1978) . One site was on a freshly cultivated plot and the other a native, uncultivated site. No moisture had fallen on either site for 22 days. A sprinkler was set up and 3.8 cm water was applied to the cultivated site and 2.1 cm- to the native site at a rate of 0.8 cm per hour. Small ridges and mounds, 10 cm or less in height were sectioned and visual observations noted. Depth of water penetration was noted on the top of these mounds and in the intervening depressions. On 3 and 4 May 1978 a total of 9 cm of rain fell on the native site. Mounds 5 cm or less (33 locations) and also level areas (34 locations) were sectioned and the presence or absence of dry soil spots noted. The general slope of this area was 1.0%. A site with a steeper slope was selected (3 -4%) and level areas sectioned and presence or absence of dry soil spots noted (11 locations) . 58 Results Figure 6.1. is a plot of the transect showing elevations and distance. Two groups of elevations are noted: A large one with an average elevation of 7.5 cm and numerous smaller elevations with an average height of 1 cm or less. The large mound showed higher contents of moisture, P, and Ca in the depression than in the higher portion of the mound (P < 0.001), for the surface 10 cm of soil. Potassium and organic matter did not show a difference between elevations at the P < 0.05 level. Organic matter was higher in the depressions (P < 0.1) but K showed no difference. At the deeper soil level only moisture showed a difference (P < 0.01) with the depressions having the higher value. There were 11 degrees of freedom for the above comparisons. No differences were noted at the P < 0.1 level for any of the smaller elevations. Linear equations (eq. 1) relating the different components to soil moisture in the top 10 cm (13 sample points) gave r^ values of 0.67 for P and Ca, 0.14 for organic matter, and 0.02 for K. Fitting the data (top 10 cm of soil) to distance from the center of the large mound with a quadratic equation gave r2 values of 0.68, 0.63, 0.52, 0.50, and 0.04 for moisture, Ca, P, organic matter, and K, respectively. There were 22 sample locations used for these calculations. These data for moisture, Ca, and P were plotted and are presented in Figure 6.2. No relationship was in evidence in the plant communities growing on the mound and adjacent depressions. Sample size is believed to be the major reason. The data gathered in the second portion of this study on the Lakeland soil are qualitative rather than quantitative in nature. Sectioning the small mounds (1 to 10 cm in height) revealed that the depth of water penetration was about 0.4 to 0.7 cm on top of the mounds and to depths of 10 cm or greater 59 o < I- — o (iuo) — N0I1VA3T3 60 350h 250 Ca (ppm) 150 50 Y --369 70 4 -80.552 X+ 6. 04lXJ H20 (%) 12 10 8 50 30 P (pprn) 10 Y = 14.036 -1.603X + 0.1 19X2 Y =58. 057-10. 628X + 0.786X2 180 120 60 0 60 120 DISTANCE FROM CENTER — (cm) 180 Figure 6.2. Amounts of Ca, P, and water as a function of distance from the center of a small rise in the top 10 cm of soil. 61 in the adjacent depressions. This same effect was observed in the irrigated native plot. It was also noted that water appeared to follow roots or small sticks that were lodged in the soil. The depth of water penetration around such material was much deeper than in adjacent areas. Leaves and litter laying on the surface also affected water penetration; dry area under leaves were ob- served extending down to about 1 cm, the actual depth was dependent upon the size of the leaf. Five hours after the two day intermittent rain ceased, the native area was sampled for dry spots. On the flat area (less than 1% slope) 24% of the sampled (level, no elevation) locations showed dry areas lying 1 era or less beneath the surface. On the same flat area 76% of the mounds (1 to 5 cm elevation) had one or more dry areas. On the steeper slope (3 - 4%) , sampling was confined to flat areas with 36% of these locations having dry areas. No surface obstructions were present on any of the above sample locations. Discussion and Conclusions One robin does not make a spring. The results of this investigation are offered as providing one mechanism to explain the degree of variation noted throughout the soils portion of this research. The lack of replication places any conclusions in the realm of speculation; considerable more research on the same and different soils would be needed before any reasonable firm conclusions could be drawn. The following discussion is based on the assumption that the results are repeatable. There is some basis for the above assumption. Sinai and Zaslavsky (1977) studied this effect of lateral flow in Israeli soils that had been wetted. They noted similar effects from visual observation on sand dunes and clay soils. The anisotropic condition of the soil would appear to be the main argument against such a condition existing in a virgin soil. 62 There is reason to believe that forest soils under a climax vegetation are not uniform. Goodlett (1954) mentions the effects of wind-throw in pro- ducing a high degree of variability over a long period of time, to depths of 90 cm. Lutz (1940) , studying a soil in Southern New Hampshire noted irregular and discontinuous horizons, with material from upper and lower horizons intimate- ly mixed. Lutz and Griswold (cited by Goodlett, 1954), suggest that all soils which bear, or have in the past, borne forest stands, have been more or less disturbed. This same sentiment is shared by Drury and Nisbet (1973). Aside from the mixing effects of wind-throw causing an anisotropic con- dition other factors could contribute to creating this condition in the soil. Animal movement through the soil and the consequent mixing of materials would be one factor. Another, perhaps more important contributor, is the water insoluble fats, oils, waxes, and resins found in plants and plant litter. Plants contain from 5 to 15% by weight of these components (Gray and Biddleston, 1974) . Humic acid, composed of flavonoids and aromatic products of lignin decomposition, are also present in soils, as are small particles of chitin. These so-called end products of decomposition tend to accumulate in soils although not indefinite- ly since there exist populations of soil microbes that appear to be able to degrade almost anything (Satchell, 1974) . The components migrate through the soil and generally come to rest in the B2 horizon (Brady, 1974) . The fact that many of these substances are water insoluble would tend to inhibit downward moving water. This would be especially true in areas that had non-uniform deposits of litter. Such areas might exhibit a hydrophobic effect, thus shunting downward moving water into horizontal paths or lateral flow. Observation on the Lakeland soil tends to substantiate the premise set forth by Sinai and Zaslavsky (1977) as lateral flow existing in a natural state. The observation at the BRU gives some insight as to the effects of this lateral 63 water movement and helps to explain the high variability noted for the various nutrients taken from samples obtained a short distance from one another. In sands it might be expected that nutrient particulate movement would also occur, as well as by solution. The volume of water flowing in the top surface layer would be considerably larger, due to the reduced channel size, than would be expected if water movement were uniform. This increased flow would move particulate material and deposit it, much as a stream does, when flow is reduced upon reaching a level fetch, thus building up concentrations of nutrients in the depressions at the expense of the higher elevations. Two aspects of soil variability need to be mentioned. The effects of different methods of handling can cause differences in chemical analysis (Chapter 5). However, all samples taken in this study were handled the same. The other aspect is attempting to define a large surface area on the basis of 5 g of material. The problem of obtaining an accurate picture of the soil component for fertilizer recommendations has been resolved by recommending that 15 to 25 cores should be taken from a similar soil and composited (Smith, 1977; Sanchez, 1976). Soil variability does not have the same meaning in a natural ecosystem, where the nutrient quality of the particular micro-site may be quite significant to a seed that happens to fall on that site. Since each species has a center of nutrient preference for maximum seedling survival (Whittaker, 1975), species composition may be intimately related to soil variability. The above assumptions, if proved correct, would help to explain the varia- tions that exist in a natural ecosystem. Research should be directed in verify- ing the existence of lateral flow in a natural setting, and what effects, if any, these dryer sites have on community structure. CHAPTER 7 IN VITRO DIGESTIBILITIES OF RANGE FORAGES Introduction In vitro digestibility determinations of range species found in North Central Florida are virtually non-existent. Some studies were made by Young (1977) on ten species. In vitro digestibilities and chemical constituents were presented for the months of July, August, and September. Other investi- gators working in different states have determined digestibilities of species that are found in Florida (e.g. Lewis et al., 1975, in Georgia; Campbell et al., 195/4, in Louisiana; Voigt, 1958, in Texas). This work was initiated to provide an information base for comparison in future investigations and also to provide nutritional data which will be of use to the various rangeland users. In vivo digestibility trials have been the mainstay of evaluating forages for animal consumption for many years (Schneider and Flatt, 1975). However, feeding trials are expensive to operate in terms of time, money, and utili- zation of facilities. Feeding trials also have some additional limitations when used for range forage evaluation, primarily due to the large numbers of species that may be selected by the animal when grazing a rangeland, and the inherent difficulty of duplicating day-to-day diets. Different indicator methods have been developed for estimation of the in vivo parameters under grazing conditions. Harris et al . (1967) outlines proceduress for use of in vitro estimates combined with estimates of fecal output. However, this method does have some inherent problems (Short, 1970) . 64 65 The use of two-stage in vitro technique developed by Tilley and Terry (1963) , for the routine screening of forages is a method that is increasingly being used, when large numbers of different forages are to be analyzed (Moore and Mott, 1973; Pearson, 1970; Milchunas, 1977). The two-stage in vitro technique is considered to be more accurate in predicting in vivo digesti- bility than the Weende system (Henneberg, 1859) , or other proximate methods such as proposed by Cramp ton and Maynard (1938) , or the comprehensive system of feed analysis developed by Van Soest (1967), (Minson et al., 1976; Moore and Mott, 1973; Moore, 1973) . This is not too surprising since bacteria are sensitive to undetermined factors influencing the extent of digestion (Van Soest, 1967) . Barnes (1973) , in a comprehensive review of literature of the different laboratory methods of predicting digestibility favors the two-stage method as giving the highest correlation to in vivo results than either the one-stage, or other laboratory methods. The primary purpose of this study was to gather preliminary data on various range species as an aid in evaluating ranges and rangeland potential. To this end, only in vitro digestibilities and mineral analysis (P, Ca, and K) , (Chapter 8) were determined on various species at different seasons. Because of differences in morphological and anatomical characteristics between range species and the type of forages (grasses and legumes) that are usually associated with in vitro analysis, some earlier experiments were dupli- cated, in theory, if not in actual methodology. Pearson (1967) studied the effects of using inoculums from animals grazing range species to animals grazing pasture species and noted no overall differences. However, Nelson et al. (1972) found highly significant differences in digestibilities of the same forages when different inoculums from different diets were used. It was felt by this investigator that the rumen microflora of an animal grazing 66 a native range should be different than that of an animal fed a uniform diet, and consequently, the digestibilities of species (compared with the different inoculums) would be different. This hypothesis \^as tested. Pearson (1967) also investigated the effect of the different components of the in vitro technique on a range species (chaparral) in Arizona. A similar investigation was also conducted on two different range species found in Florida. Effects of length of fermentation time have been investigated by Tilley et al. (1960) and Barnes (1966) ; three Florida species were used for this aspect of the study. Fineness of grind has been shown to increase in vitro digestibilities if ground very fine (Minson and Milford, 1967), while having the opposite effect in in vivo studies (Moore, 1964) . This effect was not investigated and the recommendations of Van Dyne (1962), grinding with a 1 mm mesh screen, were followed. Other methods of in vitro techniques for predicting forage quality, utilizing the different animals than ruminants, have been attempted with varying degrees of success. Asay et al. (1975) used adult crickets on genotypes of Festuca arundinaceae and noted no correlation. However, Pfander et al. (1964), using cricket nymphs, obtained the same ranking of forages as when fed to sheep. Caswell and Reed (1976) used ten species of grasshoppers to test the unavailability of nutrients in C, plants, no ranking of forages was presented. It has been well documented that plants with the C, type pathways (Hatch- Slack cycle) are generally less nutritious for herbivores than C, type (Calvin- Bensen cycle) plants (Caswell et al., 1973). It has also been documented that tropical forages are, in general, less digestible than temperate forages (Moore and Mott, 1973; Ludlow, 1976). Since numerous range species found in Florida are known or suspected of having the C, pathway, primarily in the family Cramineae, a discussion of this aspect is presented in the conclusion. 67 Materials and Methods Samples were collected from each pasture just prior to cattle entry throughout the course of the study. A complete description of sampling methods are given in Chapter 3.2.; dates of collections are found in Appendix A, Table A.l. Each sample represented a total clip of each specie found in the quadrat with standing dead material included. Samples were then dried at 70°C for three days and stored at room temperature in paper sacks. Pearson (1970) reported no changes in sample characteristics dried at 40, 100°C, or freeze drying temperatures. The 70°C temperature was selected since there have been some reports of undesirable changes occurring at higher drying temperatures (Van Soest, 1965; Urness et al . , 1977). The in vitro analysis was begun in September 1977 and concluded in February 1978. Samples were ground (1 mm mesh screen) and stored in plastic (Whirl-Pac) bags at room temperature until needed (large woody material was removed prior to grinding) . The two-stage in vitro analysis was conducted at the University of Florida Range Laboratory. To determine the rates of digestion occurring in the in vitro process, ten uniform subsamples of four species (Smilax auriculata, Centella asiatica, Trilisa paniculata and Andropogon capillipes) were prepared and In vitro analysis initiated. These species were selected as representing a wide range of plant types found at the BRU. Two samples of each species were removed after 5, 24, 48, 72, and 96 hours (completion of the run). Upon removal from the i_n vitro sequence the samples were dried, ashed, and digestibilities calculated. A quadratic equation (eq. 3) was used to fit the data as a function of time. investigation as to the effects of the various components that make up the two-stage in vitro technique were conducted using two species (Smilax 68 auriculata and Trilisa paniculata) . A solution containing rumen fluid, artificial saliva, and a solution of HC1 and pepsin (40 ml of distilled water was added to insure proper wetting of the samples) were prepared. These solutions were added to six uniform subsamples of each specie then placed in the incubator. Two samples of each species were removed at 6, 24, and 48 hours. Effects of saliva alone were determined by adding 40 ml of artificial saliva to six uniform subsamples of the above two species, placed in the incubator, and two samples of each specie removed at 5, 10, and 15 minute intervals. Effects of water alone were determined as for the saliva test, except that S_. auriculata samples were removed at 10, 20, and 50 minute intervals and T. paniculata samples removed at 5, 10, and 15 minute intervals. To test the possible effects of different donor animals on digestibilities, rumen fluid was collected from the steer maintained at the Animal Nutrition Laboratory and from a steer that had been grazing the study site three weeks prior to the collection date (20 February 1978) . In vitro analysis was con- ducted simultaneously with both inoculums on 22 species selected for this comparison. These species represented a mixture of collecting dates and pastures. Each species was run in duplicate for both inoculums. Foliar con- tent of P, Ca, K, and protein were also determined for these samples. The test for differences due to treatment effects (length of rest) was conducted by collecting samples of 12 different species from all 16 pastures in September 1977. These species were selected since they were common to all pastures in the study. The species selected were: S_. auriculata, Galactia spp. , Tephrosia spp. , Q. pumila, V^. myrsinities, C^. nlctitans, C. americana, Desmodium spp . , S^. repens, A. stricta, H. graminifolia and T_. paniculata. With the exception of A. stricta, five plants, randomly selected from each pasture, were composited for each specie. The samples of A. stricta were not 69 composited; each plant was analyzed separately. A healthy vigorous condition was the main criterion for selection of all species with standing dead material included . Seasonal digestibilities of different species were computed by taking the average of all samples collected in that season (e.g. Winter, December 21 to March 21) . All samples from all pastures were used for this determin- ation, regardless of treatment. Irrespective of the number of plants that comprised the composited samples, all samples were given equal weight. Grasses that occurred on the study site were divided, where possible, into warm or cool season, depending on the time of flowering and/or amount of growth that occurred in each season. References used for this classification were: Yarlett (1965); White (1973); Leithead et al. (1971); Gould (1968); Hitchcock (1951) ; and Grelen and Duvall (1966) . Division of the grasses into metabolic pathways (C or C/) were made according to determination by Downton (1975), Moore (1977), and Teeri and Stowe (1976), or by a comparison of growth habits and related species within the same genera. Classification of non-grass species into normally associated habitat types (wet, moist, or dry) were made from reference to Halls (1977); Duncan and Foote (1975); Oefinger and Halls (1974); and Halls (1977). Ranking herbi- vore preference with respect to livestock (cattle) , deer (Odocolleus virgin- ianus) and insects is meant to serve only as a guide as to what plant species have been found in the respective herbivore diets. This is not to imply that other consumer species (e.g. birds, other wildlife, and livestock, etc.) do not utilize these plants. This ranking reflects the three consumer groups that were of concern in this study. References used were Halls and Ripley (1961); Harlow and Jones (1965); Schopmeyer (1974); Harlow and Hooper (1971); White (1973); Leithead et al . (1971); and Yarlett (1965). 70 Results Of the four species that were used to determine rates of digestion during the two-stage in vitro process, three (Smilax auriculata, Centella asiatica, and Trilisa paniculata) were over 96% complete at the end of 48 hours. Andropogon capillipes was 88% complete (Figure 7.1.). A fit of the data to a quadratic equation (eq. 3) gave coefficients of determinations (r^) of 0.95 or higher for each of the four species. The effects of the different components of the two-stage in vitro technique are presented in Table 7.1. for Smilax auriculata and Trilisa paniculata. With the results obtained at the end of a standard in vitro run (96 hr) , as a comparison, percent of the expected digestibility were calculated for various lengths of time that the samples were in the respective mediums. Both species displayed a large amount of water soluble contents (74% for Smilax and 39% for Trilisa) , that were removed in a very short period of time (5 to 7 minutes) . The use of siliva did not appreciably change the percentages noted for water alone. The microbial portion of the process (saliva plus rumen fluid) extracted a higher percentage of the ex- pected value than did the acid-pepsin portion at the end of 48 hours of digestion. The effects of using inoculums from different donor animals was quite pronounced (P < 0.01) with the digestibilities of the range steer only 69.2% of the laboratory steer (Table 7.2.). Crude protein (6.25 times the percent of nitrogen), phosphorus, calcium, and potassium were also determined for this group of samples (Table 7.3.). Linear regressions (eq. 1) were calcu- lated, using different chemical data, in an attempt to predict digestibilities resulted in r^ values of 0.24 or lower. When a multiple linear regression 2 (eq. 2) with two variables (Ca and K) , was used the r rose to 0.44, These were not different at the P < 0.05 level. 71 % — QI/MOAI 72 1 m o u ^ « i-> 3 GJ ej C •,-i J-J C cr. <<-< u w RJ C XJ C.I CI y r-l OJ 7J U-l (J lM •H •r-l M T) D « 0 :•: ed 0) i-i •i-' o E o t/i n5 ^ o o X. X o o WlHl o o wlHl CT\ CM .0 B c c COlHl CT\ CM i-l CM c c B B principal herbivore (cattle, deer or insects) for some of the trees, shrubs and forbs are presented in Appendix A, Table A. 2. and A. 3. The grasses (Appendix A, Table A. 3.) have, in addition to the above, a compilation of probable metabolic pathway (C3 and C/) and considerations as to whether warm or cool season species. An interesting aspect of this listing of the grasses as to probable metabolic pathway is the large number (20) of C, type species found at the site compared to the single Co specie. 76 Discussion and Conclusions The rates of digestion for the four species (Figure 7.1.) compared well with data from Pearson (1970) . The majority of the digestion occurred during the microbial portion of the process. Two of the grasses in the above study by Pearson were C^ species and both of these had lower digestibilities than the other C, species that were tested. The rates of digestion were also lower, being 82 and 87% complete at the end of the microbial portion (48 hours) . This compares to the 88% obtained in this study for Andropogon capillipes (a sus- pected C,) for the same time period. Gumma (1977), working with an esophageal-f istulated animal, comparing diet selections obtained by the microscopic point analysis and bit count observations, noted a fairly large discrepancy in S^. auricula ta. The microscopic point analy- sis being 67% less than the bite count observations. With S^. auriculata being 74% water soluble (Table 7.1.) » this difference becomes a reasonable extension of the process of biting, masticating, and the length of time the material re- mained in the collection bag before being dried with the water soluble portion adhering to the bag. The use of inoculums from different donor animals resulted in the labora- tory animal having considerably higher digestibilities than the range steer (P < 0.001). It was expected that there would be a difference between the two animals since it was felt that the rumen microflora of the range steer would be better adapted to digesting range forage than the microflora of the laboratory steer. However, the laboratory animal had the higher digestibility in the com- parison. The major reason for this discrepancy appears to lie with the physical condition of the range steer. Throughout the study he had consistently lost weight whenever placed on the study site. A companion cow that was with him during the three weeks prior to the collection date was in excellent condition. It would have been of interest to have been able to compare rumen fluid from 77 Tabic 7.4. Seasonal variations in digestibilities of various range plants^ in percent. Speci es 1 Summer Fall Winter Spring Total # of 1976 1976 1977 1977 Samples Andropogon virginicus 15.4 24.5 21.1 26.5 15 Aristida stricta 17.6 15.2 13.3 15.5 102 Ca Hi car pa americana (berries) 39.4 36.2 - - 3 Cassia nic titans 19.3 22.6 - 28. 2 31 Centrosema spp. 39.0 36.9 - 15.7 15 Ctenium aroma ti cum 25.1 23.5 21.6 24.5 63 Eragrostis spectabilis 38.7 30.9 24.7 31.5 20 Galactia spp. 34.1 27.1 - 41.2 43 Heterotheca graminifolia 36.0 34.2 40.7 41.2 95 Panicum anceps 51.1 38.1 39.1 54.7 19 Paspalun.i notatum 48.6 44 . 2 36.9 57.5 9 Quercus incana 22 . 9 22.3 22.8 21.9 72 Quercus pumila 20.9 19.1 17.6 18.4 22 Schizachyrium stol cnifer 31.6 30.1 27.3 34.2 75 Serenoa repens 18.3 16.6 23.6 22.0 30 Soii lax aur icalata 30.2 34.4 27.9 33.2 23 Sorgha strum nutans 34.6 19.5 17.7 26.9 15 Sporobolus curtissi i 26.4 22.8 21.5 18.8 32 Tephrosia spp. 32.1 35.8 - 40.9 19 Vacciniuui iryrsinit.es 33.9 31.7 22.6 39.0 27 1. All species italicized, 78 this animal to that of the laboratory steer. Unfortunately, the steer was the only fistulated animal available. Health is an important criterion when using inoculum from an animal (Barnes, 1973) . The range steer was not diseased, but was undernourished and had lost close to 45 Kg (100 lbs) prior to collection of the inoculum. Pearson (1970) suggests using inoculum from animals grazing the forage to be tested as providing the most reliable information. Another possible reason for the poor performance of the range steer may be a result of the low P content of the forages and the consequent low P level in the diet. This low dietary P intake level might result in the rumen bacteria not having sufficient P for normal growth and development. This in turn would result in lower efficiencies in the digestion of the forages. Natur- al selection pressures would also be operating on the rumen bacteria, favoring those capable of functioning at low P levels, which in all probability would not be the ones capable of rapid assimilation of foodstuffs. Even though P is added to the inoculum, via the artificial saliva, the amount may not be sufficient, or if sufficient the composition of the resident rumen bacteria might be such as to preclude efficient digestion in the normal duration of an i_n vitro run. Due to the poor condition of the range steer, no conclusion can be drawn as to effects of using inoculum from different animals in jLn vitro digestibilities of Florida range species. More research is needed to delineate this suspected difference . It is of interest that the use of two variables (Ca and K) used in pre- dicting digestibility in a multiple regression equation (eq. 2) resulted in a coefficient of determination of 0.44 compared to a single regression equation (eq. 1) where r2 was 0.15 for Ca and 0.24 for K. Van Soest (1967), when dis- cussing the use of single chemical constituents (i.e. crude fiber, acid 79 detergent extract, etc.) to predict in vivo digestibility, cautions against the use of just one constituent and proposes the use of a summative equation to predict digestibilities. The use of elemental chemicals in attempting to predict digestibility apparently has not been seriously considered. Calcium has some interesting properties in its association with livestock and livestock feeds. It has long been known that oxalic acids is one of the most abundant and strongest acids found in plants (Gallaher, 1975) . Schimper (cited by Gallaher, 1975) suggested that one of the principal functions of Ca is to precipitate oxalic acid as Ca-oxalate, thus preventing possible injurious effects to the plant. A suggested treatment from oxalate poisoning in animals is the administration of Ca (Merck, 1967) . Calcium, when fed to cattle (kiln dust in feed rations), causes a significant increase in gain in weight (Ma ugh, 1978) . Whether the effects noted in the feeding trials are due to the buffer- ing action that takes place in the rumen or interactive effects with other feed components, the mechanism is not understood at this time. The use of CaCl in the in vitro buffer solution may tend to mask the effect of Ca that is noted in animal research. More effort needs to be directed alont this line of research. The failure to obtain a consistent meaningful difference in in vitro digestibility as a function of length of rest for selected range species does not necessarily imply that no differences exist, only that the experi- mental sample size was not sufficiently large enough to detect those differ- ences (Chew, 1978) . All of the 12 species that were selected had been grazed to some extent during the course of the experiment. Consequently, they could expect to be under some grazing stress, particularly in the two-month rest treatment. Different species react to grazing pressures in different ways, and there is no reason to assume that the response of all species would be similar. In this respect it would have been desirable if one or more species 80 had been selected that were generally considered unpalatable (pines, large oaks), and these tested for differences in in vitro digestibility, as was done for the chemical analysis. Another factor that is probably confounding the results is the difference in soil characteristics (Chapter 4) . When all species were summed for each treatment and all replications used, low r2 values were obtained. This low value compared to the higher r2 value when only replication I and II were used, points up to a possible soil-chemical; *" plant-digestibility interaction. As mentioned in Chapter 4, replications I and II are the most similar, soilwise, of the four. In the computation of seasonal digestibilities of the various species no multiple comparison procedures were used to test for differences in seasons. Multiple comparisons tests are almost never appropriate for experiments in which the treatments are graded levels of a quantitative variable (Peterson, (1977) . Seasons seemingly fall into this category, particularly so since the samples were collected at different times in each season, depending upon dates of collection (Appendix A, Table A.I.). For pasture grasses in vitro digestibilities have been shown to decline with increasing maturity (Moore, 1973; Abrams et al., 1978). This same effect was assumed to occur in range species, however Quercus incana appeared to re- main fairly constant over the year. Since this is a deciduous species more variation would be expected as a result of regrowth affects. The relatively low values of the coefficient of variation and the reasonable large number of samples involved would seem to indicate that this trend is real, and should be investigated further. The large ratio of C^ to C3 (20 to 1) for the grasses found at the study site has some profound implications for grazing management in Florida. Of 75 species of grasses listed by Yarlett (1965), 21 species have the C, pathway according to Down ton (1975) . An additional 42 species are strongly suspect 81 based on Gould's (1968) classification of grasses into tribes (Teeri and Stowe, 1976) . Thus some 84% of the grasses listed as important for range conservation in Florida either have a C, pathway or are strongly suspected of having it. There are three major classifications of plants based upon their method of fixing carbon dioxide; the Calvin-Bensen RuDP carboxylation (or Co) , the dicarboxylic (C, ) , and the Crassulacean acid metabolism (CAM) (Moore, 1967) . The CAM cycle apparently has been evolved primarily as a survival mechanism, a method of "hanging-on" until conditions get better. The C, plants are generally considered to be less efficient from an energy standpoint than Co plants. However, when the cost of photorespiration are added for the C plants the two pathways are about the same (Black, 1973) . The Co and the C^ plants are of concern here because of the numerous differences, biochemically and morpho- logically. Of primary interest is the seemingly lower forage quality of the C species compared to C species. The C, pathway has been found in at least 13 plant families, with Graminacea having the largest portion. In addition, different species within the same genus exhibit different pathways (e.g. sub- genus Dicanthelium is primarily Co and Eupanicum primarily C.) (Downton, 1975) . It has been known for some time that tropical grasses are generally lower quality than comparable temperate grasses (Patterson, 1935) . Whyte (1962) argues that native tropical grasslands have neither the potential quality nor the yield required to achieve improved animal performance. This low quality of tropical grasses is a major problem, however other factors (environment, nutrition, management, etc.) must be considered (Moore and Mott, 1973). In simultaneous comparisons of two tropical species (Panicum and Setaria) and two temperate cultivars of Lolium perenne Wilson and Ford (1971) noted an increase in cell wall constituents and a decrease in in vitro dry matter digestion as day and night temperatures increased. Fertilization studies (nitrogen and phosphorus) by Wilson and Haydock (1971) , on a wide variety of temperate and 82 tropical grasses indicated a general increase in in vitro dry matter digestibility as levels of fertilization increased. Other authors have noted similar increases of digestibility with increased levels of fertilization, especially nitrogen (Ludlow, 1976; Rogler and Lorenz, 1973; Salih and Burzlaff, 1977). Silica has also been demonstrated to affect forage quality. Van Soest and Jones (1968) reported that an increase of silica content of 1% dry matter decreased dry matter digestibility by three percentage units. Smith et al . (1971) reported an increase of 1% of silica in dry matter decreased digestibility one percentage unit. However, Minson (1971) could find no relationship between silica and OM digestibility. When the species that were used are sorted as to metabolic pathway, the two studies that showed a correlation with silica did so with predominantly C, plants. The study by Minson utilized several panicum species which contained both types of pathways. The failure to separate the species as to metabolic pathway might be one reason for the conflicting claims as to effects of silica that are found in the literature. Other factors such as the presence of alkaloids may also reduce in vitro digestibility in certain pasture grasses (Bush et al . , 1972). In a literature comparison of the crude protein content of 22 C species to 44 C, species, Caswell et al. (1973) noted that the C3 plants had higher levels than the C, (14.5 and 8.7%, respectively). Numerous insect feeding trials have demonstrated the inability of Ca species to support the numbers of herbivores that Co species support (Shade and Wilson, 1967; Kroh and Beaver, as cited by Caswell and Reed, 1976). Caswell and Reed (1976), in a study utilizing 10 species of grasshoppers from different areas of the United States, concluded that nutrient material in the bundle sheathes of C, plants was at least partially unavailable to herbivores. 83 If the hypothesis of Caswell et al . (1973) that herbivores tend to avoid feeding on C plants is correct, then the ramifications go beyond the ecological implications of interspecific competition mentioned by Black (1971) . If select- ive grazing is being directed at C, forage species this would imply an eventual replacement by a C^ species. Evans and Tisdale (1972) reported the invasion and apparent takeover of an Idaho grassland, originally dominated by Agropyron spicatum (CO , by Aristida longieseta (C.) . They attribute this to the low palatability of A. longiseta and the consequent selective grazing of A. spicatum by cattle, sheep, and mule deer. The above discussion has centered on the general lower quality of C, 4 species and the apparent inability of herbivores to extract nutrients from these plants as being part of a natural biological process. Man, at least in Florida, has also played a role in changing the relative proportion of C„ to C plants. Rightmire and Hanshaw (1973), studying the carbonate chemistry of the aquifer system in Central Florida (stable carbon isotope analysis method) , reported that one hundred years ago this area was predominantly of the Cq vegetation type. This type of analysis utilizes the known ratio of ^C/ C that exist for C~ (-25 + 5%) and C (-5.6 to -18.6%) plants (Smith and Epstein, 3 4 1971) . The total forage acreage, which includes all pastures (xv'oodland and range pastures, forage type crops, etc.) for Florida is approximately 12.2 million acres (IFAS, 1975). Approximately 3.9 million acres of this is planted in C^ crops (Johnson, 1974) . Most of this acreage has come into existence through recent agronomic practices within the last 100 years, and in 1974 represented 32% of the total forage acreage. The acrenge that is currently in the C, plant category is projected to increase IT'! by the year 1985 (IFAS, 1975) . 4. vlohnson, J. T., unpublished data compiled from a survey of County Extension Directors . 84 Thus, man is a significant contributor to the increase of C plants in 4 Florida. The implications of the change in vegetative types and what meaning this has for the insect and wildlife populations is highly speculative and beyond the scope of this current investigation. The livestock industry is faced with a situation where most of the important range grasses, and virtually all of the improved pasture grasses (except for winter annuals) are of the C, 4 type. In effect, this places Florida in a situation that is common to the more tropical areas of the world. Therefore, Florida range management systems should be geared to this tropical situation. Research needs to be directed toward identification of the more palatable and digestible species, with con- comitant management techniques for better utilization. Animal selection is also a vital criterion, particularly when native range is to be utilized (Chapter 10) . Range management in Florida should look to the tropics for understanding and management criteria for its tropical vegetation and cease to consider itself as an extension of the western range. CHAPTER 8 CHEMICAL COMPOSITION OF RANGE SPECIES Introduction The nutrient quality of the soil is expected to influence the quality of the plants growing upon it (Brady, 1974; Loneragan, 1973), thus the basis for soil testing (Melsted and Peck, 1973) . While a great deal is known about the nutrient requirements of agricultural plants, little is known about the requirements of wild plants (Larcher, 1975) . Nutrients cycle through eco- systems and on this basis certain general predictions can be made concerning particular aspects of the system. If an element is in a tight circulation pattern (deficient) , the amount present in the plant is related to the amount in the soil, and the addition of the nutrient to the soil will increase the productivity of the system (Whitaker, 1970; Loneragan, 1973) . This concept has led to the development of the technique of foliar analysis. Again, this technique has been developed to a high degree in various agricultural crops (Smith, 1962; Bates, 1971) and forage crops (Martin and Matocha, 1973). Re- lating foliar nutrient content to growth or yield is usually accomplished through use of the Mitscherlich-Bray growth function (Melsted and Peck, 1977) , with expanded versions of this function to predict nutrient application rates (Malavola and DaC'ruz, 1971) . However, Ovington (1968) has questioned whether a meaningful relationship can be obtained between the total nutrient content of the plants and the measured nutrient content of the soil. Part of the difficulty may be that perennials (trees) may rely more on primary minerals and less on readily extractable nutrients (Voigt, 1958) . The relationship of 86 yield to tissue nutrient content becomes even further confused due to the Steenbjerg effect (C-shaped response curve) . This effect explains the erratic responses of plants to the addition of a particular nutrient when grown in an overall nutrient poor site (Bates, 1971) . This difficulty in interpretation of the soil test results is still the subject of discussion based on two differ- ent concepts: sufficiency levels of available nutrients (SLAN) and basic cation saturation ratios (BCSR) (McLean, 1977). While little is known about the nutrient requirements of the various range plants that make up the understory of the study site, considerable information is available on plantation pines, particularly P_. elliottii which is extensive- ly grown in Florida. A rather large body of literature concerning its growth, yields, nutrient requirements, etc. has been developed. The literature on Z.' palustris is not nearly as extensive. However, since the two species are very similar genetically and phenotypically, what generally holds for one is valid for the other, with a few notable exceptions. Mineral cycling is an area where it is felt that the two species behave similarly (Fisher, 1978)' . In studies where both foliar and soil analysis have been used to predict growth response to P fertilizer, foliar analysis has generally proved to be the more effective (Pritchett, 1968; Wells et al., 1973). Several studies have been made to determine the critical level of foliar P in pines (Jahromi, 1967; Richards and Bevage, 1972; Ballard, 1974). Generally, if foliar P values are below 0.08% for 30 year old P_. elliottii, there is over a 90% chance that there will be a growth response to applications of P (Richards and Bevage, 1972) . Slightly higher values are given for P_. taeda (0.13). Those cases where foliar P diagnosis fails can probably be attributed to factors such as other nutrients 5. Fisher, R. F., 1978. Asst. Prof. Forestry, Univ. Fla., Personal communication. 87 or site characteristics limiting production or differences in sampling procedures (Leaf, 1973) . Movement of P into the foliage from applications of fertilizer can be quite short (one day), depending on rainfall (Mead and Pritchett, 1975). This is due to an extensive root system with many small diameter roots lying in the F horizon (Lyford and Pritchett, 1976) . Because of this characteristic and the size of the root system, pines are fairly efficient at picking up nutrients that fall to the forest floor. Animals may be very effective agents in the recirculation of nutrients j.n a system (Mott, 1974; Heady, 1975) . Mineral supplementation to livestock would be expected to add to the nutrient pool of the site with the animals providing the dispersal mechanisms. The competitive advantage of the over- story, through canopy cover (shading) and litter cast determines the composi- tion of the understory (Halls, 1955). The amount of nutrients in circulation and their turnover rates determine the production capabilities of an ecosystem (Odum, 1969) . Stresses introduced from grazing, drought, low soil fertility, climate, and the frequency of occurrence are all reflected in the community structure of the ecosystem. Foliar chemical analysis is one method of measuring the community "health". Before an intelligent program of livestock or wildlife management can be implemented it is essential to know the quality of forage on the site and not just assume that because it is green it is good. Quality of the diet is more important than quantity (Lay, 1964) and to base carrying capacity solely on the basis of quantity is to generally overestimate (Blair et al . , 1977). Foliar chemical analysis is one way to measure forage quality. This aspect is the major thrust of this chapter. Materials and Methods In general, all samples that were run by the in vitro analysis are also represented by chemical analysis. However, due to accidents or experimental errors in this or the in vitro procedure, some samples were lost or destroyed. Chemical determinations were also made for some species that were not analyzed for in vitro digestibilities. In addition, some samples were represented by such a small amount of material that only a chemical determination was possible. Consequently, the numbers of samples used for the ±n_ vitro analysis for a particular species, may or may not be the same as the numbers used for the chemical analysis. The sampling period represents six consecutive seasons, starting in the summer of 1976 and concluding in the fall of 1977. For ease of presentation samples of the selected species were composited for the season of their collection. Averages, standard deviation, and coefficient of variation were calculated of each species for each of the nutrients monitored. To test for differences in chemical content due to treatment effects (length of rest) 11 different species were collected from all 16 pastures in September 1977. These species selected occurred in all pastures. Five plants randomly selected from each pasture were composited and analyzed for P, Ca, and K, for each of the 11 species. A healthy vigorous condition was the major criterion for selection of all species; standing dead material was included. The species selected were: A. auriculata, Galactia spp. , Tephrosia spp . , Q. pumila, V. myrsinities, C_. nictitan, C_. americana, Desmodium spp. , Serenoa repens, P. palustris, and A. stricta. A. stricta samples were not composited, but analyzed separately. To check the possibility that the 11 species selected did not constitute a representative chemical sampling of the total species present, all samples ^59 that had been collected after the last grazing application were pooled for each pasture. Phosphorus levels were determined for each pasture and an AN OVA analysis conducted to determine treatment effects. This procedure was also used for all samples collected at the beginning of the study, summer and fall of 1976. The effects of length of rest on the chemical composition of the overstory foliage was tested by collected samples from Pinus palustris and Quercus incana. Collections were made of P_. palustris in March and August 1977 and March 1978. Samples were obtained from the top sun needles of the dominant trees and only first flush growth needles were taken. Since the trees were quite tall, a rifle was used to shoot the desired clusters of needles from the tree. A preliminary collection (March 1977) consisted of five samples each from the control and the two-month rest pastures of replication I. Phosphorus was the only nutrient determined. The second collection (August 1977) consisted of five samples of each pasture, all replications and treatments represented. The nutrients analyzed for were P, Ca, K, Mg, Al, and Fe. The third collection (March 1978) consisted of five samples each, of all treatments in replication I; also the control and four-month rest pasture in replication IV. Money pre- cluded a complete analysis at this date. Nutrients determined were P, Ca, and K. An ANOVA analysis was used to test for treatment and replication effects with all four replications and replications I and II only. Five samples each were taken of _Q. incana from the top sun leaves in November 1977, from the control, two-month and six -month rest, replication III, and from the two- and six-month rest, replication IV. These were the only pastures where large Q. incana occurred. Statistical analysis of the results were similar to that for P_. palustris . To test for possible relationships between soil chemical content and foliar chemical levels linear regression (eq. 1) were used. The nutrients 90 compared were P, Ca, and K levels found in the soil (Chapter 4). The chemical content of the soil was compared to the foliar content of P. palus_tris, Q. incana (trees), the 11 representative forage samples, and the composite forage samples on an element by element comparison. Results Compiled data of chemical composition for various species in the different seasons are presented in Appendix E, Tables E.2., E.3., and E.4. for P, Ca, and K, respectively. Species are listed in alphabetically order, with the mean, standard deviation, CV, and the number of samples collected for each season. The test of the effects of grazing on A. striata, utilizing samples collected just prior to and after cattle had grazed the area resulted in no difference (P < 0.1) between the phosphorus content of the samples before and after being grazed. Each of the chemicals that were monitored (P, Ca, and K) were averaged by grazing treatment for each species. These averages were then summed and a new composite average computed for each pasture along with the CV (Appendix E, Table E.5.). No meaningful treatment effects (completely random design) were noted for any of the three elements. There was, however, considerable varia- tion in chemical content among species as reflected in the high CV's. An average CV of 46.6, 52.3, and 61.0 was calculated for P, Ca, and K, respectively. With the randomized complete block design for the ANOVA test, block (repli- cation) effects were noted at the P< 0.1 level while treatment effects were not different. Because replications I and II are most similar with respect to all chemi- cal elements except P (Chapter 4) averages of foliar chemical content were made over these pastures and are presented in Tables 8.1., 8.2., and 8.3. for 91 Tabic 8.1. Average foliar phosphorus content (ppm) of eleven species in replications I and IT. c • 1 Species Months of rest 2 4 6 12 Aristida stricta 1030 680 640 650 Smilax auriculata 1450 1670 1250 1061 Galactia spp. 2220 1470 1630 1630 Tephrosia spp. 3580 1670 1820 1310 Desmodium spp. 2770 2510 1560 1700 Cassia nictitans 2570 2360 2580 2050 Quercus pumila 1370 1250 1470 1450 Callicarpa americana 2570 2530 2650 2760 Vaccinium myrsinites 720 610 710 880 Serenoa repens 1200 1120 1150 1090 Pinus palustris 470 410 390 410 All species are italicized. 92 Table 8.2. Average foliar calcium content (ppm) of eleven species in replications I and II, c • 1 opecies Months of rest 2 4 6 12 Aristida stricta 2732 2110 2020 1770 Smilax auriculata 7820 7220 8650 10070 Galactia spp. 12140 11080 11850 11450 Tephrosia spp. 11690 9590 9890 10550 Desmodium spp. 10900 9600 9420 9120 Cassia nictitans 8960 8620 7750 6970 Quercus pumila 12250 12110 7550 9460 Callicarpa americana 5760 5590 5430 4260 Vaccinium myrsinities 8810 9720 10400 9170 Serenoa repens 2290 1240 1660 1690 Pinus palustris 1090 1090 1220 1080 1. All species are italicized. 93 Table 8.3. Average foliar potassium replications I and II. content (ppm) of eleven species in Species^ Months of rest 2 4 6 12 Aristida stricta 3380 3350 3400 2620 Stnilax auriculata 9770 10560 6700 6220 Galactia spp. 10270 10580 9390 9560 Tephrosia spp. 5570 5770 5940 4837 Desmodium spp. 6800 12620 9070 6790 Cassia nictitans 7750 11770 9700 8300 Quercus pumila 3380 4800 4430 3900 Callicarpa americana 15720 18230 16770 13860 Vaccinium myrsinities 3250 2500 2750 2970 Serenoa repens 11360 9430 9900 10950 Pinus palustris 1760 1950 1790 1970 1, All species are italicized. 9 k P, Ca, and K, respectively. The ANOVA analysis resulted in no different block or treatment effects for any of the three nutrients (P < 0.05). Three pastures contained large percentages of both wet and dry sites (Figure 4.2.). They are II-2, III-4, and IV-'.. Samples were collected in these pastures representing both wet and dry sites. These samples were com- pared for phosphorus content by a t-test . No difference (P < 0.1) was noted. From the soils map (Figure 4.2.) two pastures were selected, one that lies entirely within the wet area (1-2) and another that lies in the dry site (IV-2) . The above pastures contained 59 and 84 foliar samples for the wet and dry sites, respectively. Phosphorus content of the dry site was 1220 ppm and 1590 ppm for the wet site, difference at the P< 0.001 level. Linear regressions (eq. 1) relating foliar to soil chemical content were found to be weakly correlated for Ca and K (r^ less than 0.2) for all 9 plants and plant groups. The r for P was less than 0.2 for P_. palustris, 0.41 for Q. incana , 0.51 for the average 11 representative samples and 0.63 for the composite of all samples collected for each pasture after the last grazing period. Regression equations derived only from pastures that had the o same grazing treatment had r values above 0.5 for both plant groupings (11 species and all samples collected after the last grazing period) . The no grazing or complete rest pasture showed high correlation (r = 0.94) between soil and foliar levels of P. The coefficients of the equation were 23.6 for the slope and 1304.6 for the intercept. Foliar P for each pasture was averaged at the beginning (1976) and the end (1977) of the study and are presented in Appendix E, Tables E.6. and E.7., along with standard deviations (s) and CV's. These data were analyzed by a randomized complete block design and tested for differences between treatment menas using ANOVA. The 1976 data showed no treatment effects (P < 0.1); 95 however, the block (replication) effects were different at the P < 0.005 level. When only replications I and II were used, there were no differences (P ^0.1) for either treatment or replication. The 1977 data (all replications) showed a treatment effect at P < 0.1 and a block effect at P < 0.005. When only replications I and II were used, there was a treatment effect and no repli- cation effect (P < 0.1). When all replications were averaged with respect to treatment, no differences were noted (P ( 0.05), but there were differences when all treatments were averaged with respect to replication (Table 8.4.). Relative percent change of foliar P (all species composited) was cal- culated by taking the difference for each pasture between the summer averages for 1976 and 1977 data, and dividing this by the summer 1976 average (Table 8.5.). Statistical analysis indicated no different F value for either treat- ment or replication effects, when all replications were used or when repli- cations I and II were used. However, when replications I and II were used (most alike with respect to soil P) there was a treatment effect (P < 0.025); replication effects were not different. The relative ranking of foliar P as a function of length of rest, lists the six-month rest highest, followed by the two, four, and no grazing treatments. This same order may be noted in the 1977 data (Table 8.4.). The same procedure (except that 200 was added to the differences to make them all positive) was used for wiregrass. This was the only species where enough samples were collected to have sufficient data points for both years in all but replication III. The ranking of the means was the same as the above composited data. The means separate out with no difference between twelve and four; four and 2; and two and 6; at the P j 001 — ssvwoia 129 Table 9.1. Forage production (Kg/ha) based on uniform length of experiment (326 days: 23 August 1976 to 18 July 1977). Replication Trea tir.ent (months res t) 4 12 I II III IV X 2987.2 3305.9 3074.8 2418.5 2 946.6a 2120.9 1649.0 1176.2 1979.4 1731.4ab 2098.6 1984.3 1302.6 3648.4 2258. 5ab 1116.5 1402.4 1070.1 2377.7 1491.7b Trt F = 3.860 significant cMr^Ocxir--mi— ■ ■ — i O P- o in r| m m •— ioo>ooin* r i icm 00C00^ON00Cr\00CT>CCCr. CTv m vd r- co o> CM '.£> I I CM I I O I cicioociinococcoic") ^-irirOt-ivDOC^iOOC o^ l-i CI T-1 ■i-i co h 4-1 4-J to o C U) rd <> a co w 0 Cx •H :-. « W) CJ V' c ni 0. o T! w r-i CO hi a •H o CO fH 0 3 o 4J ■r-l OJ u Vj W r~t V) 4-J 4-1 CJ T' •^ l-t cr. c C CD r u crl c: OJ 01 4.J < < t_> CJ u u o int-4inincMoorocnr~r~- f-iin^i-iooO'-<'-tOO en C O O M (0 4-J « co B en r-l 3 1-1 w 0> D- CI r r-l 3 C 3 C 3 a. 1-1 --^ OJ 4-1 IT) •r-4 -r-l 0) o o U- r^TD ■H o o u E >-! D. 'r-l t CO e !) 1-4 C c c 3 • >v a) 1-1 3 CO c (1) Q ■r-i a ax i-i 3 U 3 CO B -h r F Q- O CO 4-1 r-l •r-l 3 4J F 3 CO co to co w CO O CO r. 6T 3 r- 1 1 3 N o X CO -CJ o C eg t-i si m O CJ CO -r-l c CO .c o r-l •r-l J-l V a. 1-1 rl 3 rN OJ r-l 6C V-i -C o o 4-J 41 C en tu OX £ 1-1 •r-4 Vj o D- cj H flJ 1-4 ffl a 3 3 O o h 0 (X 01 K3 PC. 1-4 p-> Pj aa« w CO co CO CO H > 3 >> cr u a; td Pi R r-l CO -i 0J 01 R 4-J OJ. •r-l 4-1 •r-l C CI •a •r-l b U O -r-l 4-1 •r-l U-J rJ e tj rO 0 (J ■r4 a 4-1 c 4-1 4-J J-l -C 0 r n c Wj -> i£ o C\ :* r^r-ivocncNOOcNO O CN CM CN i-l i-l vO CO 5 O CN r». vD O O in co o m i-H CN moo o \o r^ CN in in cm cn N m cn sj CNOCNO>OOONOi—l -j-CNcn^occoonco cncNcncNi— vjcncsi ocooo^ i in in r— o i o o> n a> h c- cn On CO On 00 CO CO CK ffi Ol CO CTi CO ON CO Cn cn o M CO ■I-I •r-l cn i- w a co 01 4-1 CO •r-l -r-l Ol C) a a >>T3 4-1 jj 4-1 cn o a. ■H 01 o 0 CJ E r-l a. •r-l r CI R W C Cfl cn CJ to co r-l cn cn C) r-l C c c 3 • r^ 0J r-l 3 cn r. T) > G ex •H h CO •r-i 01 rO to ■r-l a a. £, S-i 3 ^ 3 CI! •r-l CU <1 tu eg H t- CO oj 4-J cn X. cn h (X CJ cn 4-1 I-l 3 rJ 0 •d CtJ in h cn •r-l 4-J h 3 to W W CO ca cn c. cn . 1 •r^ u. •H cj Cfl i — i 0 3 cO 4-. 0 tuj 3 r— i 3 3 N o X CO •C r CO 3 r* 0 4-1 •H •H cy r-i •r-i rJ CI r-i O B) o CJ CO -r-l c cn -C 0 v-< 4-1 CT n M 01 i — l cn 4.J 4-1 c tJU W 0) X -r-l CX U V-i 3 >% a M M rC C) O CD C) ■i; •r-t cn c c o o r-- i 4-1 CJ c cn a) CJ rD JC u •r-l rJ O D. C) H tr< (J n « CO CJ cj 4J rJ cfl UJ cd 01 3 3 3 CJ CJ H 0 a 0) CI — I LT~> ^1" ' — ! ' — 1^ H H O ooooooor-im or^\DmrsJorocNio NnovfNcou-icoo ^D CO rovocovDmcsir-t-iu-ir-.cMOr-icMv)- r^-^OvOcor^m ro--icMcoOcoOt-ivOCMr-ir^.r^roo i 0-( OOOOOOOOt-(CNJOO^-iOO mvDvOf. ocoOHvfNon(»ovDNi«i uirvlCN-j-oinOOCMroOc>-)ronO O ON CI 0\ r-4 O CXI m co w co i^ co en CMvOOCNCMOCNIOO vDr-tr-Hci^ccxcioo OvOLnr-IOCNr-tCOCO N»D-Cfvf^vJONinOvD vDCNrgmovDOOo*l»£i'--i, a u CX ■r-l h co •rJ CJ ,o en •r-i ex ex .c u -J i-i c a cu 4-J CO ^ c\t h ex C) w ...; 1—* CO H CO ■r-l 4-J r— 1 H 3 CO w co n CO o en O 3 ct. 4-> O 60 3 r- < 1 3 t-0 C) y •H ■ri cu u •r-l M o u O ctj c> CJ CO •.H c r : to 4-J 4-1 C 60 CO CJ X •i-i ex )-l U 3 r^ o CO C c CU CJ r-l 4-> cu c CO CD cj xi x: Vj el CO flj a 4J Vj trt OJ . — I CD CO 3 3 3 O CJ fi U U u u u w O K n (X, ex- aapj to CO CO CU X & CO C 3 ftK-rl O u a ^^a w • E co E 60 3 CO C T3 > Cl 3 « E 'rl CI < 4-1 r-l -H 3 |J^ CO O 1(1 »t H rl CO X O C CO 3 IS X O )-i -H 4-1 D* O 60 l-i X CJ O CU O U O CX cj H BJ vD vO M m O o O O CO c o m ct\ Nn O cvj m n rj cm O oo m M(nMMriNnntn-iCCC3 M C3 v< C u eg u ctl £ en O-oro •-i— ig (X-HUnJi^O 3 to 4-1 O 60 3 O 4J .r-l -.-I 01 |-l -r-l U O l-J CJ r-l w 4-1 4-1 c 60 en QJ CO r. c 01 o 4-1 « Hi 0) CD 4-1 r4 to CD u u u u o W O rc e a o 3 ig to it it r-l 3 3 N CO U U 10 'H X -r-l D. Vj Vj 3 >-> oi c to m ci rO x. r-4 CO CO 3 3 3 O h ri< fd a ff a: w 3 U CO C 3 ex >-l -r-l tJ cx t^TD h CO H D CO c r43rJ3caE-r4 01lH 0 X to -Q O C to c co x: c u -r4 4-i 01 r-l 60 r4 XL O O V4 T-l r4 O CX O H o E o a oj u J) M O) CO H > 3 S tr o 01 o 160 161 re oj r-i 4J Ci T3 O U-l LO m cm r^- vo cr> i— t m o cm m vo m •* rucn }HI.1 CM vD i— I C") I I I I o\ vo vo vo vd r-» Is- r~~- Is* f»s r-» r- co cr> .-< r-i ■— i tM -J vO N co 162 the eleven times on the site. The other animals showed a weight increase two or three times out of the possible eleven grazing periods. The calculations of forage intake for the two-month rest treatments for the July and August grazing periods from Appendix A, Table A.l. and from equation 7, (fecal output and average digestibility) are presented in Table 10.3, The amount of dry matter consumed per animal per day (calculated from the fecal output data) is low compared to minimal dry matter requirements obtained from NAS (1976). The values computed from Appendix A, Table A.l. are more reason- able except for replication I, where the regression equations indicated more herbage present after grazing the pasture than before grazing. A sample diet for deer at the Ocala Wildlife Management Area, for late fall and early winter is presented in Table 10.4. Over 51% of the diet is composed of species that were not covered by this study. Consequently, no comparable diet quality analysis was attempted for deer, as was made for cattle. There were a total of 4317 arthropods captured during the collection per- iod. Of this number 80% were either Formicidae (ants) or Trombiculidae (chiggers) . A ranking of the number of arthropods caught for each type group, along with common names, is presented in Table 10.5. Total arthropod numbers and numbers of families were not found to be different with respect to the different grazing treatments. Of the various families that were analyzed for differences in population levels as a function of the different grazing treatments imposed, only the family Lycosidae (wolf spiders) were found in higher numbers (P < 0.05) in the two-month rest pastures than in the ungrazed (Table 10.6.). The mark and re-capture analysis resulted in 17 wolf spiders, 51 crickets, and 11 beetles marked. There were five spiders and one cricket re-captured. 163 Table 10.3. Calculations of forage consummed in the 2 month rest pasture (July and August 1977) per cow per day by two different methods; estimations of forage removed (Table A.l) and by eq. 7 (fecal output and average digestibility). Replication II III IV Cow days 25.00 60.00 40.50 22.50 Forage removed Table 1 Kg -424.30 570.70 404.10 354.90 Avg. daily consumption per cow (Kg) - 9.50 10.00 15.40 Fecal matter at end of grazing period (Kg) 30.28 31.81 38.69 36.50 Din intake per pasture (Kg) (eq. 2) 44.50 46.70 56. SO 53.56 Avg. daily consumption per cow (Kg) 1.78 0.78 1.40 2.38 1. Average cummer digestibility equal to 33.9/1 by methods of Pearson (1967a) 164 Table 10.4. List of food items from stomach analysis of deer collected on Ocala Wildlife Management Area, September - February, 1952 - 1953. Food item 1 Part eaten 7. Total volume Sibal etonia Fruits 31.10 Basidiomycetes Entire 25.15 Quercus spp. Acorns 12.51 Serenoa repens Drupes 8.90 Galactia Stems - Leaves 1.50 Vaccinium myrsinites Leaves - Twigs 1.34 Gramineae Stems - Blades 0.38 Mixed legumes Leaves - Stems 0.21 Smilax spp. Leaves - Vine 0.16 Misc. - 18.75 Note; Adapted from Harlow (1961) 1. All species italicized. 165 Table 10.5. Number of different groups of Arthropods caught in pit traps September 1977. Grouping Common name # Caught Formicidae Trombiculidae Lycos idae Acarina Gryllidae Arichnids Cuvculicnidae Carabidae Mutillidae Scarabaeidaa Chilopoda Melandryidae Staphylinidae Tettigoniidae Acrida-; Opilcnes Reduviidae Bostrichidae Chrysomelidae Trogidae Ccccinellidae Elateridae l.epidoptera Phasmididae Homoptera Biptera Phalacridae Erotylidae Eurytomidae Blattidae Diplopoda Passalidae Tenebr ionids Os tomatid&e Thysanoptera Scorpiones Nabidae Cydnidae Ants 2195 Chiggers 1280 Wolf spiders 249 Mites 186 Crickets 174 Misc. spiders 57 Snout beetles 27 Ground beetles 21 Velvet ants 17 Dung beetles 1.7 Centipedes 14 Bark beetles 12 Rove beetles 12 Camel crickets 10 Grasshoppers 7 Daddy-longlegs 4 Assassin bugs k Power-post beetles 3 Leaf beetles 3 Skin beetles 3 Lady bugs 2 Click beetles 2 Misc. moths, butterflies 2 Walking sticks 2 Misc. Leafhoppers, treehoppers , etc. 2 Misc. Flies, mosquitoes, etc. 2 Shining flower beetles Pleasing fungus beetles Straw worms Cockroaches Mil li pedes Darkling beetles Bark-gnawing beetles Thrips Scorpions Damsel bugs Burrowing bugs Total 4317 All groupings are italicized. 166 Table 10.6. Average number of Lycosidae caught during the five day collecting period, by pasture. Grazing treatment Replication (month rest) I II III IV X 2 14.6 14.0 12.1 13.0 13.4 4 24.7 17.6 18.0 10.1 17.6ab 6 19.2 20.0 22.0 18.9 20.1b 12 22.7 19.5 26.0 19.0 21.8b X 20.3 17.8 20.3 15.3 Trt F = 5.398 Blk F = 2.059 167 This gives population estimates of 847 spielers and 8874 crickets for the areas around the pitfall traps. There was no differences noted for any of the V traps that would indicate any type of movement pattern for any of the differ- ent groupings. Of all the crickets captured, 87% belonged to the species Gryllus uvisopus, the remaining were G^. fultoni. Identification to specie was made with the assistance of T. L. Walker . G_. ovisopus is a mute, sedentary cricket (Walker, 1973) , x^ith males the most mobile, moving about 14 m per night . Chemical analysis for P, Ca, and K and individual weights are presented in Table 10.7. for crickets, spiders, and beetles. Assuming that the traps were 50% efficient and that the average radius of motion was 14 m, then calculations indicate that there would be about 2 x 10 crickets per pasture (0.8 ha average) . Further assuming that the ratio of 1:10 (trophic level conversion) exists for spiders, the primary predators of Gryllus, then there would be 2 x 10 spiders per pasture. This 1:10 ratio is borne out by the results of the mark and recapture data. From data in Table 10.7. there would be about 10 Kg dry weight of these two arthro- pods present in east pasture. At an 80% moisture content this represents about 50 Kg biomass of these two groups per pasture. The chemical content of this amount is about 0.8, 0.04, and 0.9 Kg of P, Ca, and K, respectively, for each pasture . Discussion and Conclusions The emergence of H. graminif olia as a major contributor to range cattle diets was first mentioned by Gumma (1977) . Gumma also noted that the select- ivity ratio (percent in the diet/percent available) was less than 1.0, which 10. Walker, T. L. 1977. Prof. Dept. Fntomology, Univ. Fla . , Personal communication. Table 10.7. Average weight and P, Ca, and K content of three groups of insects trapped at the study site. 168 Avg. wt. (g) P(ppm) Ca(pprn) K(ppm) Beetles Crickets Spiders .0328 5840 1360 6630 .0300 9370 2430 9420 .1501 12250 2470 10200 169 indicates that the animals were not actively seeking out this species. H. gramini folia was also relatively high in digestibility and nutrient content (Chapters 7 and 8) . This plant may play a more important role in range cattle diets than has heretofore been suspected and should be researched further. Setenoa repens, A. stricta, and Q. incana are also consumed in fairly large quantities (25 - 30%) at different seasons, however they have low digestibilities (20%) and are low in P . It is doubtful that they are pre- ferred foods (Gumma, 1977) but should be considered as emergency, or more likely, stuffing. With the exception of the first grazing period, the cattle generally lost weight each time they were placed on the site. There are several explan- ations for this observation. The cattle were being taken from a good pasture with adequate supplements and placed on a poor quality range at irregular intervals. Cattle lose weight when the diet is shifted (Ensminger, 1976), when they are moved (Heady, 1975; Wagonner et al . , 1960; Snapp and Newmann, 1960) , or almost any time anything is done to them out of the ordinary (Hafez, 1969; Bonsraa, 1965), particularly in warm climates (McDowell, 1972; Bonsma, 1965) . The process of driving the cattle out of the pasture and weighing them would also cause a weight loss. Differences in the amount of fill would also be a factor for the loss of weight noted. The length of time the animals were on the site was, with two exceptions, not long enough to gain an accurate weight picture of the animal performance. The two longest periods the animals were on the site were the initial and final grazing periods, 18 and 28 days, respectively. In the first period all but two animals showed a net weight gain. In the last grazing period only three did (four, counting the calf, however, his mother showed the greatest weight loss). On the basis of these two observations it would appear that animal selection 170 is a major factor for a successful range operation. This observation is in agreement with this investigators own practical experience with range animals in Utah and Nevada. In the study of the comparative digestive efficiencies between Bos taurus (Hereford) and Bos indicus (Brahman) cattle, Howes (1964) found that on a low level of protein intake, Brahman cattle digested more protein and consumed more dry matter than did Herefords. It is strongly suspected that had Brahman or Brahman-crossed cattle been used, cattle performance on the study site would have been improved. Based on the quality diet determinations and the lackluster performance of the animals, energy would appear to be the main limiting factor. The calcu- lated DE available is in all seasons less than half of the required energy. It is extremely doubtful that the actual diet was this low compared to the minimum maintenance requirement. As mentioned in the introduction, animals appear able to select a higher quality forage than is indicated by sampling techniques. This is further attested to by the relatively low loss of weight the animals experienced on the site, which was about 5%. Weight losses in this range could be explained by other factors, as mentioned above. Forage intake determinations based on the average digestibility of the forage and the amount of fecal output, did not give values that could be reasonably interpreted as minimal dry forage intake (approximately 2% of body weight). This was found to be generally true for all pastures. The values calculated from the regression equations were, in general, better but still showed considerable variations. For the above reasons, no data for forage intake are presented other than that given in this chapter (Figure 10.3.). The quality of range foods for cattle and, by analogy, deer in the Florida flatwoods are low. Digestible energy requirements are, in general, not being met for English breed cattle. Deer, being adapted to the general location, 17 1 appear to be able to select higher quality foods than domestic cattle. How- ever, the imbalance of the Ca:P ratio (5:1) in flatwoods forages nay well be one reason why there are not more deer found in these regions. Selection of the proper type of cattle to use on a range area would be a fruitful area of research. Tests at the range cattle experiment station in Florida on native range showed that crossbred Brahman calves gained more than European breeds (Peacock et al., 1966). It is suspected that a Brahman or Brahman-cross would provide the best point of departure, followed by rigor- ous selection. Experiments of the above nature coupled with investigations of forage intake and nutrient quality would provide much needed answers for the range cattle industry of Florida. The stocking rate of cattle in this study was approximately 0.361 cows per hectare per year. The average weight of the animals on the pasture was about 383 Kg per pasture or around 93.3 Kg of cow per pasture per year. The average biomass of the two groups of arthropods was estimated to be about 100 Kg per pasture during one week in the latter part of September. There is considerable room for doubt as to the validity of the esti- mates of arthropod biomass. However, the fact remains that this estimate only represents two groups of the crawling arthropods and the soil and aerial populations were not sampled. Gryllus species are primarily omnivorous and the spiders are predators, so their impact on the vegetative portion of the site may not be as great as the biomass numbers would indicate. Other herbi- vorous insects were noted at different times in rather large numbers, grass- hoppers in particular, in the late spring. One sampling does not adequately sample the insect population over the course of the year, but it does point out that the biomass of the arthropod populrtion may be quite significant with respect to the biomass of the larger herbivores of the area. This area should be researched further. 172 Specific recommendations for use of native flatwood ranges in Florida based on the results of this investigation are: Use a Brahman or Zebu type animal, and/or follow a rigorous selection program. The native range should be used primarily for dry cows, supplemented with energy (molasses with protein) . Phosphorus should be supplied in order to rectify the Ca:P imbalance of the native feeds. Calcium should not be fed, or should be at low levels with respect to P, if put on offer. CHAPTER 11 CHEMICAL ANALYSTS AND IN VITRO DIGESTIBILITY OF THE FRONDS OF SERENOA REPENS (SAW PALMETTO) Introduction Saw palmetto (Serenoa repens) is a common component of the undurstory of much of the Coastal Plains of the Southeast. Saw palmetto is considered to be an invader species (although the rate of invasion is slow) and was probably held in check by the competitive effects of other native species and periodic wildfires prior to the advent of modern man. With the control of fires, the heavy grazing pressure on the range areas and the establishment of plantation tree farming, crown cover of saw palmetto has increased to about 20%, especially in Southern Florida (Hilmon, 1968) . The main economical value of saw palmetto is in the use by the honey industry, though wildlife (deer, turkey, bear) eat the drupes, and in the case of the Florida bear, may at certain times of the year provide a substantial portion of the diet (Halls, 1977) . Presently, mechanical control appears to be the best method of eradi- cation, with roller chopping giving good results (Carter, 1973) . It was widely believed that cattle would only graze saw palmetto when forced to eat it through lack of other suitable forage (Hilmon, 1968) . How- ever, it was noted on the flatwoods at the BRU that cattle grazed saw palmetto, even in the presence of adequate forage, often grazing it when they were first turned into the pasture. In view of the striking symmetrical pattern of graz- ing of the fronds (Figure 11.1.), it was surmised that the top portion of the buds were being grazed; this was later substantiated by direct observation. 173 174 Figure 11.1. Symmetrical pattern of defoliation by cattle on Serenoa repens, 175 Cattle, when initially placed on rangeland, will usually sample many differ- ent species at first, and later will tend to concentrate on only a few pre- ferred species (Stoddart and Smith, 1955). In the case of saw palmetto, it was observed that cattle would first eat the outer portion of the fronds, but after a short adjustment period (one or two days) would concentrate on the buds and only occasionally graze the fronds. In order to investigate the reason for the preference for the bud, especially the top portion, the following study was conducted . Materials and Methods The collection site was located on an ungrazed plot in replication IV. This site had been burned in February 1974 and again in February 1976. Collect- ions were made from five plants randomly selected from within the pasture in October 1977. In addition to the bud, four other fronds were selected in des- cending order down the plant stem (Figure 11.2.). Each frond was divided into an outer and inner portion, with respect to the petiole, with only the frond material sampled. Each sample was ground and split into two portions, one for chemical analysis (P , Ca , and K) and the other for i_n vitro digestibility. Analysis of the data was with a randomized complete block design for A NOVA . Results The chemical content between the outer and inner portion of the fronds was not found to be different at the P< 0.05 level (Table 11.1.). However, there was a strong trend for the outer portion to be higher in chemical con- tent (all three elements) than the inner portion. The IVOMD and chemical values for the inner and outer portions of the frond were composited and set up as a randomized complete block design, with the fronds as the treatment and the plant as the blocks. Table 11.2. lists the results of this analysis. The difference for the table was calculated at 176 Bud Figure 11.2. Collection profile of Serenoa repens . 177 Ix Ix Ix V) 01 01 .c tn •u o u r-i fl o 14-1 T> ■o c c Ix cd tf x: JC a ai c c o o a) a) H £ CO E o - CJ P-i c •^ -< O u a. m h 178 co is ■u c C/3 CO 179 the P< 0.05 level, however the difference between the bud and the rest of the fronds was found to be different at the P < 0.01 level, for both chemical and in vitro analysis. The Ca:P ratio was calculated for each of the fronds, both inner and outer portions (Table 11.3.). The lowest ratio occurred in the outer portion of the bud, (.78) and increased with the age of the frond. Discussion and Conclusio n Hilmon (1968) reported that it requires between three and seven years after a fire for saw palmetto to return to an average frond production of five per year, with an average of seven to eight fronds being produced the first year, and three the second year. Consequently, the samples taken in descending order along the stem represent a loose aging criterion with the oldest being the lowest on the stem. As noted from Table 11.2., phosphorus and potassium decreased with frond age and calcium and in vitro digestibility increased , Charley (1977) points out that generally up to 90X of phosphorus may be returned to the shoot; potassium is more variable in behavior but is usually withdrawn in most species and calcium often increases with age of the leaf. Ovington (1968) report;; that potassium leaches out from plant tissue with relative ease, calcium more moderately and phosphorus with difficulty. The data for the chemical composition of the fronds as a function of age are not out of line with that reported In the literature, even though saw palmetto was not specifically studied in these reports. The interesting aspect of increased digestibility with increasing age is not so readily explainable. A linear re- gression (eq. 1) was performed in an attempt to relate increase of calcium 2 9 content with that of digestibility (r = 0.394). A higher r was determined with a multiple regression (eq. 2) with phosphorus and calcium as the indepen- dent variables. This demonstrates Van Soest (1967) arguments concerning using Table 11.3. The Ca:P ratio of Serenoa repens fronds. 180 Frond position Ca:P Bud A Bud B 1 A 1 B 2 A 0.80 0.78 1.08 1.37 2.54 3.07 2.59 3.12 3.26 5.96 181 more than one component to predict digestibility. Obviously, there are numer- ous factors affecting digestibility and to select any one factor as the predomi- nant one is to imply a relationship that may or may not exist. Young leaves of plants, under optimal moisture conditions, generally have higher rates of photosynthesis than do older leaves (Moore, 1977). The fact that more P and K were found in the outer portion of the tip and leaves (Table 11.1.) would tend to support the contention that the tips were more metabolically active than the inner portion of the fronds, and hence more pa latable. Nutrient removal, either via translocation or leaching, would explain the differences noted in saw palmetto fronds as the age of the frond increases. The increase of in vitro digestibility from the bud to the older fronds may be due, in part, to the increased calcium content of the more mature fronds, however, more research is needed in this area. The preference by cattle for the top portion of the bud may be due to the location of the bud (the bud is about head high, with respect to cattle), the configuration (a small package, compared to the more leafy frond), or perhaps the lessor amount of fiber in the bud than in the more mature fronds. However, this preference is more likely due to some inherent ability of the animal to recognize a source rich in phosphorus and satisfy its requirements in this manner. The ability of cattle to select proper rations of minerals to satisfy nutritional requirements has been claimed by some to have been lost through domestication (Coppock, et a., 1976). Specific appetites for certain sub- stances has been noted in rats (Scott, 1950), though phosphorus was not one of the nutrients investigated. Coleman and Barth (1972) noted that animals selected forages higher in protein and digestibilities than were observed in the avail- able forage sample. Leigh (1961) found that livestock accepted grass cultivars 182 highest in phosphorus and potassium before those with low contents of these minerals. The fact that the Ca:P ratio of most of the forages tested is high, (sample diets were found to have a ratio of 5:1, Chapter 10), would make any forage with a low ratio attractive to the animal. With the possible exception of an undetected exotic compound that makes the outer portion of the bud more desirable to cattle, there appears to be little reason for the preference noted, other than the low Ca:P ratio. More research is needed to determine if this ratio exists year round or just in the fall, when these samples were collected. Since the cattle diet has the greatest proportion of S_. repens in the winter (31%) , over 16% in the spring and fall and only 7% in the summer, either S_. repens has the lowest Ca:P ratio during fall, winter and spring months or the animals are selecting plants in the summer (possibly legumes) that have a higher P content in an effort to maintain a favorable Ca:P ratio. This entire area of animal selection should be investigated further. The observed fact that cattle were selecting exactly that portion of the frond that had the lowest Ca:P ratio would substantiate the argument for an inherent ability to select foods high in needed nutrients. Basiciomycetes comprise a large portion (up to 60%) of the deer diet (Harlow, 1961; Harlow and Jones, 1965) and these plants have a higher level of P content (1%), (Mulder et al . , 1969) than most range plants. This high utilization of fungi might be a response of the deer to the high Ca:P ratio found in most of the other components of their diet. CHAPTER 12 DECOMPOSITION OF PLANT AND FECAL MATERIAL Introduc tion In the absence of herbivores, minerals cycle from the soil to plants, to litter, and back to the soil. Grazing adds more pathways since minerals are shunted to animals and deposited as dung or urine, which decompose at different rates than unaltered plant material (Heady, 1975). In a study of herbivory in grasslands, Grant and French (1975) reported that the turnover tine of minerals entering the soil was considerably reduced by the presence of herbivores. The rate of turnover time was increased by as much as 2.8 times over the rate for a system without herbivory. Petersen et al. (1956a) determined that a mature cow produced 25 Kg of dung and 9 Kg of urine daily. Most of the voided P occurred in the dung and the urine was richest in N and K. Cattle deposit their excreta haphazardly while grazing. However, since they tend to bunch up at night there is usually a higher fecal count on the bedgrounds, or for that matter, any place where cattle tend to congregate (Hafez et al., 1969). "'. imerous investigators have reported on the distri- bution patterns of grazing cattle (Petersen et al . , 1956a; Richards and Wolton, 1976). | Changes in botanical composition are believed to be due, at least in part, to the deposition of dung and urine/ Sears (1956) reported that dung and urine were responsible for an increase noted in the grass component and a decrease in the clovers. Other investigators have reported similar conclusions (Heady, 1975; MacDiarmid and Watkin, 1971). 183 184 Petersen et al. (1956b) suggest that the greatest benefits froia excretal return for P and K are under conditions of high stocking rates. MacDiarmid and Watkin (1972) suggest that the loss due to fecal contamination and avoidance by cattle are not as great as it would seem. They noted that cattle do not graze areas around fecal droppings or urine spots as close to the ground as in uncontaminated areas. However, because of the increased plant growth around the dropping and later, in it, the actual amount of forage removed is the same or greater than if the dropping had not been present. The same effect from other herbivores would differ in amount rather than in principle. Most of the studies involving nutrient cycling through cattle have been done in humid pasture environments and very few under range conditions. Effects of fertilization and the movement of fertilizers in the soil have been studied in Florida (Chapter 4) but the effects of grazing on the rate of decomposition has not been investigated. This would be a fertile field of research. Different plants do not decompose at the same rate even under tie same environmental conditions (Williams and Gray, 1974) . An experiment conducted in England, using two different size mesh litter bags, noted a higher rate of loss for the larger mesh, this difference was attributed to the presence of small invertebrates (Bocock, 1964) . The bottleneck in mineral cycling rest in the slow decomposition of organic matter. Management practices to increase this rate would result in a faster turnover rate, hence an increase in annual production , This present study was designed to provide an indication of the rates of clung decomposition and the consequent movement of minerals into the soil. Also of interest was the decomposition of a highly digestible range grass and one of the more common plants found in Florida, S_. repens, and the role that insects play in the decomposition process. 18 5 Materials and Methods The fecal decomposition study was begun 27 July 1977 and concluded 13 March 1978, a total of 228 days. The control pasture in replication IV was selected as the study site. The study plot was selected so it would be in the open, not subject to canopy effects. A grid was laid out 1.22 m on a side, nine evenly spaced sampling positions were selected, with one position in each corner of the grid. Soil samples were taken, at each (5 cm intervals) to a depth of 25 cm. After sampling these were plugged with soil and marked with golf tees. Three positions in this grid were randomly selected for controls. Fecal material was collected from the feed yard at the BRU , thoroughly mixed, and three samples removed for percent water, P, Ca , and K determinations Trie fecal material was shaped into circular pies, 20.3 cm in diameter and 4.4 cm in height, and weights of each sample recorded. The pies were placed on the grid, slightly off to one side of the golf tees. A second group of samples was positioned in two rows 0.7 m apart. Five samples were placed in each row 0.7 m apart. In one row the sample was ]aid on a nylon mesh, the ends folded over and stapled together. Six soil samples were taken, three to a row at intervals of 5 cm to a depth of 25 cm. The samples were collected on 13 March 1978, weighed, dried, and percent moisture determined. Three samples were analyzed for P, Ca , and K. Soil samples were taken at five depths (5 cm intervals), under each sample at least 2 cm away from the golf tee, and analyzed for P, Ca , and K. Amounts of P, Ca , and K present in the original sample were determined from the dry weight of the sample and the percent of the nutrients present. At the conclusion of the experiment three fecal samples were collected, sand removed, dry weights and percent of each nutrient determined. The difference between these two values represents the amount of each nutrient that has been removed from the original sample. 186 Soil samples were taken on one unraeshed pie in the two rows of five samples each. The sampling format was in the form of a cross. One soil core (5 cm interval, depth to 25 cm) was taken directly under the center of the pie, four cores taken at the edge of the pie (12.7 cm from the center) and four cores 17.8 cm from the center of the pie. Any surface manure was removed prior to sampling. On 15 May 1977 a quantity of Panicum anceps was collected from the 6 month pasture in replication 1 and dried at 70°C for three days. Aliquots of approximately one-half gram were taken and placed in nylon (1 mm) mesh bags measuring 10 by 15 cm. Eighty samples were placed in bags and the ends stapled together (control bags) and additional 80 samples were placed in bags with the ends held open by two wire stays (treatment bags). Five bags of each type were placed in pairs in each pasture in an X position; one pair in the center and the other pairs 26 m away, toward the corner of each pasture. The two bags (treatment and control) were placed 15 cm apart and the position marked. Twenty closed litter bags containing S_. repens were placed in replication IV, five to 3 pasture, each sample placed within 20 cm of the P. anceps samples. All bags were in position on 24 May 1977 and collected 13 March 1978 (292 days). Upon collection of the sample excess sand was brushed off, dried at 70°C for 48 hours and the contents weighed. Ten bags were composited and ashed in a muffle furnace at 500°C for four hours. The light ash was floated off with water, the remaining sand was dried and weighed. Percent of the P_. anceps remaining was computed with allowances made for the residual sand. Chemical analysis of the P. anceps was conducted for P, Ca , and K. A t-t:est was the method of statistical analysis used for this study, when P < 0,1. The results are considered non-significant (ns). Comparisons made were between the initial nutrient level of the site (1977) and the control 187 (no fecal material) at the end (1978), soils under the samples and the controls, nutrient content under the samples compared to the same site at the initiation of the experiment, and the treatment bags and the controls. Results The soils of the study site show a general increase in nutrient content from July 1977 to March 1978. An increase was noted at all depths for P and all but 5 - 10 cm depth for K. This increase was noted for Ca, only in the deepest layer. These increases were generally found to be different (P < 0.1) for P and K (Table 12.1.). A comparison of the nutrient levels under positions in the nine sample grid indicate differences only for P and K in the upper two levels of the soil profile (Table 12.2.). Comparing the soil cores taken in 197 6 to the cores taken at these same locations under the fecal samples (1977) shows increases for P and K (P < 0.1). Only the top two soil layers show an increase for Ca (Table 12.3.). Sampling in a cross mode, at the center of the pie, the edge and slightly beyond the edge, gave a generalized picture of the movement of the nutrients into the soil. Since only one sample point was available the results were averaged for each equal radius and are presented in Table 12.4. To show the uneven movement of nutrients into the soil, P content of the soil at the five depths was listed for the north-south and the east-west arm of the cross (Table 12.5.). The area has a gentle slope to the south and east, this is reflected by the apparent concentration of P in these directions. The composition of the initial fecal material is compared to the analysis of the material at the completion of the experiment (expressed as ppm, 1976 presented first followed by 1977 analysis for each nutrient): 22657:]1677, P; 20319:25086, Ca; 4714:680, K. These figures represent a percent change of -48.5, 19.0, and -85.6, for P, Ca, and K, respectively. 188 o c > -H QJ C ►J 60 » cc p. o 189 Qj n o <4-l TD a t-i c x: eo O C3 ■w c o « :-< .-< -H -H ^1 00 AJ r~- C C-N U r-l M 4" J.) O 3 V-" C w ■'-' 0 Q) X- o *J Us c 0 Cu o r-l •r^ 0) '-i-J I> •H 190 .,-1 G\ O (3 o .-I -M > -H o <; (J 191 o XI > u •h B CM CT> v£> cm r-~ n vt n t— I k— 1 i — I N ri OO o o o o o o 00 r^ o r-l O CM CO vO CT\ r-l o o r-~ G\ u~i t-i o o n r^ n r-< O i-i <3- r-~ r-t a\ r-i r- o • 0J C >.<-< tx V-i -u nj ~0 O 0 U o e & SJ CO a, 01 cd rO Tt M <1> (1) •H .C M-l •H ^ cn 60 rn C cn •H H H O a H CI) cri M (fl 3 iw Ml en S S Q £ S Q O S S Q O E 13 Q O 241 0! ■"-, CM On vD 1^ tO 6C to U, co —J vD c— 1 Xi w \D in m in \~t r^ T-l o^ a cj F: T3 60 6C B -a m in o CM M 1T| O C^ r-l o cr\ iJ c^ m i-i vD c r* intnw o o O CM b CM r-i en -1 S & Q O 242 m oo vj CO CNI cm m 00 LT> O CN f-t CO v£> •~o O N tO o O CM CSI CO r- O m r» co oo vO N N r-l CO CO O CN r-l co r-- CN CO r-l CM o o o o m o m ft. « E m in r-l m r-l O CO l S Q o 243 a> \ CO o m C^ M £4 • 03 ^i r-» CM r- CO 00 r-l CO i-H CO c vD r>- r-» r- c o r^ r^ r-» r^ r-4 on in oo m r^ r-i co o o in r~- in t-i co CD ^~- ON vO CM r*- r-l CS CO co in oo o on r- 60 b£ X. « W U n o no «* 00 CM r-l o 00 o co r-~ ,D v-' B 00 r-l ON CO r-l <^ n w-4 co v£> U-l r-l C~) oo cm . o O oo •<)- r-i vo cm hianoo o i— i m r» r> i-i oo \o m i— i <4-< tg o in m mom O o r>» cm t-- o cm i< ro C\ f~- CM S S O o - r- m m cm r- S & Q o SSQO 6 ~a 246 fr-J ct) i-l CO G-. • I • cr> o on x: o o o r-t CO o VD O CM CM r-< CM CM fc o SSQH £ 2= O H SSOH £ IS Q H 6 247 id h ri o W r^ O ^o co rv. in r-l o~> r-l vD VJD N (M O CO CO CM o c t-- CM m m 00 CM VO CO O CO o o ■u cu •r-i >\ W t->\ x :* q o s ^ S^QO ^^OO 248 C i-l CTv r-» <}• 00 CM CM <)" CO CO CM ON On 00 LO U~> CO r-- in !— 1 CO O m VD CM >-< r-^ o ct> o CM 00 250 ►j ,j j JtJtJ MOOQQQQO Qr-tr-IMWr-lr-IM a i q o s q o C c 01 a" TO r~J O •,M <4-4 > X. CI O fr. H CL TO -r-l o o « h CJ r-l to 0) R •r-l ,c 1— 1 ■r-l u r-l Cfl W M T3 0) CD 4-1 V) Vj TO C E 3 4-J u QJ oj O TO n >-. PS (X, to rJ CO co S S ►j e ^ r-i g ^ I-t CD r-H C3 0) r-l CD U to to r-H F. 4J o c c •r-l C/J u ctl w ;r; ^— ' to TO •r-l « •h c en *^> r-l r-l •r4 ■r-l w 4-1 0 M ■r-l •r-l TO CM C4-I r-l 4-1 l- c C to •r-l •r-l 4-1 4-1 •r-4 w 1-4 4-1 0 cr, H CI WJ r-l •r-l 3 oi c •r-l >, .C O r-4 CO 3i4tiaoioiwB H >. 3 3 3 3 ex rn « a o o o IhOSSWMMI-i cuconcoioicuci) c.) -4 -H -r-i 3 3 3 3 , C/l r-l c CO 3 cvl r-l 3 4-1 m .C to ^ X) u u Vj V-i >■ •r-l 4) <4-l 0) x> 0J OJ F u JZ U 60 r-l u 4-1 iil a. CO r-l r-H 3 4-J X a o r* to to TO o TO < o p X 3 cd rC r-J O -J [--. s M 3 >» U f*, rl U 3 r-ITOWr-JCUJ-lCllCUr-l O r-l £ r-l XI r» ^ M HI rH {J CD rJ C J3 TO rO CO r-l XI 0) ti ai cxci) c ,*: c CCO ^ CJ ,1 3 ■UXC-r-l«50cu«0 3jCr-4r-J^3CLV- pi w fl m u cytn o < •r-l < ^* TO fcl J3 C C cn TO TO o TO r-l •r-l CJ cn 3 •r-l r-'J O a- o rJ O H ■■— ' U-l a 3 •r-4 t M 13 TO 4J CO O 1-4 C rJ C •■"! 00 CO 3 0> D -a >w -r* t4 y — 0»4-> C C TO V CO CO -r-l -r-l g r-4 O- 3 OCOTO— w- S ^ TO 01 U R a •H 3 4-1 4J cn v4 TO CD c > U r-4 O cn r^jO r-l cn (-4 >1 TO H TO •r-l H R bi) 3 •J c o o cr; CJ M TO TO TO 3 O o cn •r-l c C •r-l •r-l cn 3 •-4 •r-l O u cn c 3 1—1 •r-4 -r-4 n r-l >. t—i X X f- e ■r-l r-l CO 3 CJ rO 1—1 o a •H u >> o QJ o ^ QJ 3 XJ u rJ C) C) cn «t .n TO I— 1 t3 >. r^ 3 JC 3 0) 0 4-J TO a < o u o r-l r-4 W r-J S O* OS PS cyj cn CO > > S S Q 251 >1 c 'n > M •H h X •H Vj H 01 PL, " n q m HHhJJJiJfliJ JhJMMMWhJM an o o n o i i x xj i i i i eg eg 01 OJ a a T) crt T3 OJ CJ >-, •r-l OJ CO C) CJ CO r-l M OJ C3 N \i J~j XJ M-l Rl :* OJ CJ 1-J •r-l 1-1 r— * J-1 txx> u i u 4J 4J 60 JrJ CI. r-l -L' h u o Ki-i hih a) eg 3»joj-hpjoo g ^-S ^N ^ 05 X X «~- C X 3 ^v 3 • O •U n) 0) « Vj O C > C X 3 W 01 /-v to J: O O Jj PQ rJ O O -r-l i-3 (U to to cl d. a to cl cotw-JcaNio:-^ W CO TH -rJ 01 O tfl ta o r-i -a 4-j x> x" o •--I V-i CO O O 0) o u tO C O 10 r-l (/) ^ o. to cj n o rj CJ X CJ UUOQOrJaiH rJ (JOMp-J m Om mmm MWQMQMMrJMMMMQMMMQQQ Q S3 Q Q I Q I I I £ 5= S rS s s QSQaSS • S S i i i i S i •u vj a -u S a e m r-l •r^ Rl 4J CJ ■u 1—1 0 r-l 0 60 r-l >-l o r^ 60 o O OJ ffl c o ■U U-l U-l S o X r-l CO 3 CJ j_i 0) XI to CO 3 CJ CJ XI Rj t-i 3 C > 0 T— 1 •r-l OJ 4J 15 C >j 4J ft .c 611 w *i CO, -V •r-l CJ 0) c o a M r-l to XJ 0J u XI ■o XI u ■r-l r-l o « 3 to i—l a It X! eg ■ — i •rl CJ ffl r-l CJ cx u u •r-l u 1-1 eg o 3 rx C) n "J w M c/i o 00 ►* i_> pa Cd S w en Q CJ JH X 3 « X XI •— \ 3 o t-l (0 •--1 CJ r-N X S 4-1 r-l O i—l ^S -r-l en c3 to& i—i 3 0 to B u-l E rJ U O 3 •H 3 v^ oj J-l -r-l c X c fi 3 XJ 3 c- 3 >. X W X o S— ' o •r-i « r. H j_i 3 tr. ai -ri o eg •r-i c -i-i S= U U V: 0) tO 13 CO rJ •H E r-l to 3 t— i eg O — I CD -rl U W 4J XJ OJ 0 rJ R ►J OJ c o rJ <"- tH r ^-^ C eg •rl c ^-^ o •H l-i M u o 4J y 61! X) CJ •r-l >> w u a CJ- •rl 3 CL c w 3 Cl 03 Cl 1-J iw rr C3 3 r^ to •i-l t/; o •r-l to i~i a t) X; o r-l o h ffl OJ M iw O h 1—1 rfl •u 3 ■H X 3 c 3 3 H C •r-l .IJ ■u 0 C3 •r-l CO co UJ ^ u •r-l CJ tg X; tJ eg X •r-l CJ l< u 3 •r-l •r-l 0 CL C X 0.' CJ ■u Cl r-l V-i X 0) r>^ CL JJ CL tj Cl OJ OJ OJ 1—1 Vl D a !>•> 51 •r-l 4.J •u X w (•: w PC PC t-i r-l P-. D-. « *j e oj cl eg -r-i pa C D. r-i 60^-' t-t t» 3 Vi —i o ti a o eg -r-l > a ■h l-i -h c 60 to -tJ crj cd i-l D. -H X XJ H CO r-i TO to c3 -1 IS 10 H IJ 'rl O ll O h h O >i CO H H S X M Q Q ►J a q x ■ i i X X s e eg 01 01 01 01 (-1 o o cd TO H D- "- 1 •H « P .c 1-. » .o 01 CJJ 0 c •t-i l-t "> a CO S & 252 •h "a a QJ o D 'D 01 c v^ w n> h 0 •H 4-1 3 M 01 o •h T3 ^ f ;>: C 253 rt -u JJ co cu o-i CO « O Q O O CO. Q pq ;.; Q ?J i. & S Q 3= » „ a i i i s « S3 Js S O Q ^^SCSOS^SSS&^SOSSSSUSO E 60 a B in cn i-i & cO to h B) CI to »-i rt cn cu O o co 60 :-j CO H 4-1 CO o * »n 60 rt co I-I (j 01 >n Ci o >>X CO M 60 Q> f\i- —4 3 CU a; 4-1 co C > H-J c CD 1— 1 tO r-i en QJ -r-l O a (-j co ■H CU •r4 B -h a M r-4 II 1! II II « 1 O i-l Q rq r-4 rt /-s rt X QJ rH On e- ^-^ 4-J ^--. 3 Vh /~s CO a 3 cu rt ^, r?H rJ 3 4-1 CU ^~s r-4 CU « u X rt ra 4-J C CO iw^ V^* CJ ' « 3 60X iw E CO > o C) co ^D •r4 X to rt i—4 co cn CO -j X 60 CJ •r4 rt rt ^w •rH « N-r* CJ CTn CO CO o rt e rt •i-i cu a w CJ :d •rH' c rJ !2 •r4 r-4 CU 3 & 1-4 TJ a X C/J r-H e o •r4 cd CM Vr' a. o r^-. CJ •r4 -r-i o •~^ Ph 1-1 •r4 rt ■H •i-4 O D. -r-l rt V— J ^s B X O •r4 N— ' 0 rt CO CO CO 60 4-J in •H c C/. 3 rt rt rJ CJ S, E B 4_l 4-J B CO "3 rt r-l C i-l o> 1-4 -r-l o ra CO o CO 4-J c >J o B 3 CO ra rt •rH CJ u ^> QJ CJ rt r-4 r-4 4-1 ■r4 O CJ o 3 QJ 4J •i-l CJ kJ X •r4 CX VJ I— 1 O c: 4-J r-l 0J X 1 — 1 to o 4-1 CJ E 3 3 U 3 rH N-' c o j.j rt -h -.-4 rt r-4 ■a Q. 3 a 4-J rt rt 3 a B 3 3 a CU co QJ ►W o > > r-4 14-4 E •r4 CO O u ClJ 4-1 4-J •H •H CJ E • X CJ 4-1 CM o > •1-1 CJ E o QJ rH 3 E « D. i-l 1-1 TJ o E c rt CO rt i-i CO rt o E CJ c CO >, QJ •J to CO CL 3 o U CJ o o •»-i rt rt •H c rt rt X X 60 r4 3 CO CN » E CH C X 60 62 4-J 13 CO •r-l 4-J E E CJ 4-J r-J •rH CO cm u o o B TJ 3 B u to X E E E 3 3 rt « CO O C to CJ co rt r^TJ 4-J a. o- CD •r4 a, 3 rt c o 3 3 3 i— i 1-4 N ■r4 rt X X rt -^ •rJ t4 4-1 « •r4 G O «a 4J o •H 4-1 r-l o o CJ CJ rt « u X o c r-4 ■rH X rH -rH Oh K U U X cn B c •r-4 60 E •r4 •r4 •rH a Oh rS « 60 u u D. I— 1 O CU rH TJ TJ 4J •i-l O OJ 60 O CU B E B CO CO X rH o o •rH CO O- r-4 o C C C M X 4-J •r4 U u rt rt rt ti rt o QJ O a. a u CO >s O X < < < < < u a w Ul p-i Oh Oh Oh Oh CO to CO CO CO H rt W CO 4-1 i-4 C-J CU X CU CO 4-> «-s & « & lo O X o r^ 4-) j-( 0> CO I 60 O CU 3 QJ Q H CO 60 254 ~-: CJ a M) ij 01 o >J il C.) > V s CD Cii ao 6 tu cd .-< ON r-l ■-< I— 1 o c R) 0) ffl J-i t-i CO CJ CJ P-, > -■;' H ge 9613 5812.5 60.5 Total 23 Ca s s i a nictitans Summer 1976 5827 2533.8 43.5 4 Fall 1976 3909 1300.7 33.3 12 Spring 1977 8921 12783.4 31.2 6 Summer 1977 8703 2129.3 24.5 20 Fall 1977 6391 2151.2 33.7 3 ■verage 6750 2096.4 31.1 Total 43 299 Table E. 4. - continued Species Season X (ppm) s (ppm) CV a) # Samples Centella asiatica Centror.ema s PP . Ctenium aroma ti cum Desmocj ium SRPr. Elephantopus tomentosus Eragrostis spectabi. lis Galactia spp. Fall 1976 7433 Winter 1976 5680 Fall 1977 8116 Summer 1976 6161 Fall 1976 6585 Spring 1977 6680 Summer 1977 4955 Fall 1977 8400 Average 6576 Average 2493 Fall 1977 11368 Average 3603 1984.6 163.3 220.7 1810.3 1405.8 1369.2 1239.9 472.6 1408.6 26.7 2.0 3.6 27.1 28.2 16.3 18.9 Total Summer 1976 2884 959 1 33 3 Fall 1976 2127 617 6 29 0 Winter 1976 2773 585 1 21 1 Spring 1977 2902 - - Summer 1977 2540 1238 0 48 7 Fall 1977 1730 543 2 31 4 19.0 Total Summer 1976 9552 2224.5 23.3 Summer 1977 7564 3203.4 42.4 Summer 1976 1948 352 6 18 1 Fall 1976 6139 1172 5 19 1 Winter 1976 3437 - - Spring 1977 2785 632 2 22 7 Summer 1977 3496 821 6 23 5 Fall 1977 3810 582 9 15 3 39.1 Total Summer 1976 7608 6365 8 83 7 Fall 1976 4868 4785 3 98 3 Spring 1977 7997 3127 9 39 1 Summer 1977 9181 18652 1 203 2 Fall 1977 5454 1701 0 31 2 Averag 7022 1806 5 25 7 3 1 2 3 2 11 7 9 2 1 5 5 29 3 17 2 3 1 4 3 2 15 24 17 k 89 5 Total 139 300 Table E. 4. - continued Species Season X (ppm) (ppm) CV (Z) # Samples Heterotheca gramini folia Ilex glabra Lespedeza sppT~ Lyonia lucida Pant cum aciculare Pas pal urn no ta turn Pteridium aquilinum Quercus Summer 1976 8549 Fail 1976 8602 Winter 1976 9622 Spring 1977 10539 Summer 1977 6366 Fall 1977 11073 A vera < Fall 1977 912 5 Summer 1977 1560 Fall 1977 3154 Spring 1977 2650 Summer 1977 2964 Fall 1977 1969 Summer 1977 3921 Fall 1977 3644 Summer 1976 4960 Fall 1976 9977 Winter 1976 4258 Spring 1977 12114 Summer 1977 7125 Fall 1977 9415 Average 7975 8071 Summer 1977 1682 Fall 1977 3007 Summer 1976 4382 Fall 1976 3779 Winter 1976 2633 Spring 1977 4778 Summer 1977 4591 Fall 1977 3993 4086 2 47 8 2893 0 33 6 3522 5 36 6 3974 7 37 7 2213 0 34 8 3995 0 3 6 1 1688.7 138.6 677.4 736.7 11.3 3060.7 1766.0 38 51 24 23 15 27 18.5 Total 178 8.9 21.5 27.8 0.6 1781 3 4 5 4 2145 4 58 9 2209 2 44 5 3460 9 34 7 1467 0 34 5 3791 5 53 2 3350 3 35 6 38.4 Total 58.7 1699 2 38 8 1341 8 35 5 425 4 16 2 1921 1 40 2 2448 2 53 3 2052 2 51 4 6 3 3 1 2 4 19 28 24 6 10 11 13 301 Table E. 4. - continued Species Season X (ppm) s (ppm) CV (X) # Samples Quercus incgna Quercus pumila Averag 4026 776.4 19 3 Summer 1976 3888 828.1 21 3 Fall 1976 3140 571.5 18 2 Winter 1976 2478 473.3 19 9 Spring 1977 4497 957.9 21 3 Summer 1977 4463 1357.7 30 4 Fall 1977 4556 2246.4 49 3 Average 3837 857.8 22.4 Total 92 4 5 4 & 43 20 82 Rhus_ copallina Fall 1977 Rhynch.isia SPP- Summer 1976 Rubus spp. Sunvner 1977 Fall 1977 Schizachyrium stolonifer Summer 1976 Scleria mulhenbergia Serenoa repcns Average 8334 5325 2990 7606 4360 2380.7 922.9 31.3 Summer 1976 4586 2037 7 44 4 Fall 1976 5879 4666 0 79 4 Winter 1976 4204 2622 8 62 h Spring 1977 3406 860 8 25 3 Summer 1977 3424 1764 9 51 5 Fall 1977 4660 1394 0 29 9 21.2 Total Summer 1976 2680 6.4 0 2 Winter 1976 8620 3068 7 35 6 Summer 1976 6677 3513 4 52 6 Fall 1976 4116 2126 9 51 7 Winter 1976 5372 3132 6 58 3 Spring 1977 2471 839 6 34 0 S umme r 1977 10079 5317 2 52 8 Fall 1977 5082 4055 0 79 8 17 11 3 4 6 13 54 10 10 8 10 43 9 Average i633 2589.5 46.0 Total 90 302 Table E. 4. - continued Species Season X (ppm) s (ppm) CV (7o) # Samples Saul ax auricula ta Summer 1976 6007 2871.7 47.8 7 Fall 1976 2808 1297.3 46.2 4 Winter 1976 3674 1406.2 38.3 4 Spring 1977 4753 4487.5 94.4 5 Summer 1977 8046 16838.1 209.3 82 Fall 1977 5771 3115.1 54.0 4 Average 5177 1861.4 36.0 Total 106 Sorghastruiti nutans Summer 1976 5485 3414.3 62.2 6 Fall 1976 6530 2696.9 41.3 3 Winter 1976 4856 1850.1 38.1 2 Spring 1977 2504 - _ 1 Summer 1977 4370 843.4 19.3 2 Fall 1977 4762 1504.8 31.6 2 Average 4751 1334.8 28.1 Total 15 Sporobolus cur tissii Summer 1976 2716 897.1 33.0 8 Fall 1976 2219 400.4 18.0 10 Winter 1976 1369 207.0 15.1 3 Spring 1977 4208 5652.4 134.3 4 Summer 1977 2198 738.5 33.6 5 Fall 1977 1882 215.1 11.4 3 Average 2432 976.5 40.2 Total 33 Sporobolus iunceus Fall 1977 2030 1398.8 .9 Stillingia sylvatica Fall 1977 1812 Tephrosia SPP . Summer 1976 7592 3667.4 48.3 5 Fall 1976 3289 316.8 9.6 2 Spring 1977 5521 1706.0 30.9 4 Su.t imer 1977 5511 1111.9 20.2 17 Fall 1977 5513 1576.7 28.6 3 Average 5485 1521.9 27.7 Total 31 Tillandsia usenoides Fall 1977 4256 422.1 9.9 303 Table E. 4. - continued Species Season X s CV # Samples (ppm) (ppm) (7o) Woodwardia virginica Xyris spp- Summer 1976 2527 Fall 1976 2705 Winter 1976 2426 Spring 1977 2280 Summer 1977 2637 Fall 1977 2637 Average 2535 Spring 1977 1678 Fall 1977 252 2 699 6 27 7 1048 7 38 8 1004 6 41 4 183 0 8 0 625 7 23 7 1114 5 42 3 Trilisa pan icu lata Summer 1976 8270 2207.9 26.7 3 Fall 1977 6343 200.1 3.2 2 Triplasia americana Fall 1977 4044 - 1 Vaccinium myrsinitps Summer 1976 2527 699.6 27.7 4 6 5 4 22 7 159.2 6.3 Total 48 Grand Total 1306 301 Table E.5. Average foliar chemical composition (ppm) composited for eleven species for each pasture. Replication Treatment (mon ths rest) 2 4 6 12 Phosphorus I CV(7.) 1573 44.4 1448 43.0 1628 43.5 1363 47.5 II CV(%) 1822 48.3 1668 46.5 1409 49.9 1351 42.3 III CV(%) 1605 50.2 1539 43.7 1730 42.8 1616 44.1 IV GV(70) 1340 49.1 1265 55.6 1456 43.0 1329 52.4 Avg. CV 46.67. Calcium I CV(Z) 7510 60.7 7570 48.7 7270 47.5 6540 56.6 II CV(7.) 7650 50.0 7600 52.4 7490 46.7 7580 55.5 Ill CV(7.) 7300 48.1 7840 40.8 7770 48.4 7540 48.1 IV CV(%) 7280 56.9 6790 53.1 6920 65.9 7530 57.6 Avg. CV 52.37o Potassium I CV(%) 6650 55.9 8780 53.6 7640 53.0 6340 58.0 II CV(X) 7810 60.3 8900 59.5 7640 56.1 6750 60.0 III CV(7.) 7630 62.4 6950 56.4 7270 63.9 7810 54.2 IV CV(7) 6610 63.8 5830 75.1 7710 77.9 6320 66.6 Avg. CV 61. 07. 305 Table E.6. Average P foliar content in pasture at beginning of experiment 1976. Months rest Replication I Replication II Replication III Replication IV 12 X 1067.0 1098.0 956.0 1110.0 ft 412.0 400.0 289.0 404.0 cv 38.6 36.5 30.3 36.4 # Samples 31 13 15 48 X 1038.0 1017.0 972.0 885.0 s 320.0 247.0 301.0 254.0 CV 30.8 24.3 31.0 28.7 # Samples 31 13 16 37 X 1071.0 1361.0 1238.0 1450.0 3 450.0 304.0 419.0 351.0 CV 42.0 22.4 33.9 24.2 # Samples 28 8 11 23 X 976.0 692.0 828.0 819.0 s 259.0 213.0 269.0 214.0 CV 26.5 30.8 32.5 26.1 # Samples 35 14 14 20 Total samples 357 ANOVA Analya is All replications Replication I and II Trt F = 0.200 Trt F = 0.741 Blk F = 9.075 Blk F = 2.323 Trt df » 3 ns (? P< 0.1 Trt df = 3 ns @ P q cm m o CM O (M n o o o o lO N N N CM CM CM i-l CM O O O r^ r^ m in cm o CM vD \X o o o o CM oo in CM m vo m \D CO 00 CN i— I r-l i— I CM o o o o r-~ -j- r» oo o o o o o n o oo OS tO OMfl oo r» O o> r^. o r^ m CM r-< •— i o o o o o «* - 49.3a 43.5a 50.3a 46.5a Trt F = 0.528 Blk F = 1.893 ?. 4 6 12 41 44 53 47 46.3a 48 41 49 45 45.8a 49 47 51 48 48.8a 59 42 48 46 48.8a 0.663 0.424 0.728 0.727 0.636a 0.790 0.656 0.819 0.746 0.753a 0.676 0.628 0.849 0.687 0.710a 0.761 0.622 0.800 0.680 0.716a 0.723a 0.583b 0.799a 0.710a Trt F = 3.043 Blk F = 10.178* * Significant at the P < 0.01 level. Note: Numbers followed by same letter are not significant at the P< 0.05 level. 317 o\ i— i r~ oa oo <]- o r-l i^> o> o m O CM vf vD N vD H N IS H o m i-< o o 4J •i-j eel H ■u > O 03 h 0 r>r> r-i to •H 3 0 C •r-l >, CJ CO rH H bo •r-l c r 0) 0 r-l CO r-l (-1 01 CX CO CO rn CO 4J r'1 3 3 3 3 ex to CO CJ 0 0 0 CO 3 3 Vj t-i (-1 r-l O 0 P C O CJ CJ -l r-l CO •H Q> .O O •-< e o e fco a. a ra 03 o o CO co ra x. ra 1-1 1— 1 ra ■u ra CJ r-l c 0 3 T3 m •H 0 u ra CO E D ra >-i u O Vi a 0) •r-l rU •r-l 3 CJ CX • a J-. rJ ra .C r-l 0 CX ex O r-l w ex Ul ra O tJj ra ra ra 3 co 0 •r-l •r-l CJ 0 ro C X X h c ■r-l \l CO 3 0 CJ 0 0 rJ cj 3 rO u 1—1 r-l ra ►> >^ 3 .c 1 CI M M K^ r-l JSJ C_Y C4 K CO 318 S-. >! tr, CO fc= 3 H r S-. 3 ;: JC CO c C ■H •rJ o CJ o o u Cf) M re •.-1 E 3 J-> C & X c> CO to C/> rH 3 1-1 •r-l 4J 01 e -h O CO > Ph S VO M O CM N H .J o ^ N On oo o t-i r-~ vo in m vo in cm in ,-i o m cri in o r-i en n oo O- o\ <}-cno>-IC0Oi-l ci CO > (-1 a <4-i r- •r-l CO o ;> cu •r-l 4-1 4-J CX •H •rl O F E 4-1 CX U-l ' > •H u Fi CX C) 01 CX r-l n F d 3 C U C CO CO CO CO I-i CO rcl c; c CD ex c; CO co >-, fi) 3 CX O u O -r4 « C!l •r4 O c C(i -C CO n cm rJ 1 rl l-i 60 MJ 00 4-J CO tn CO ••-J 4-J i-l F F a 4-J 1-1 •r-l ct) 0 o o c •D '0 a h Vj CO X. H K Fi F n n i crt Cfl C/i 0 n r>n o a. ex CX 0) •r-i a. 3 <-) C) C) 3 3 3 n i-i — i t— i N •r-l rt .n n .,-1 O 0 O CO 4-J 4J o •r-1 4-1 l-i o o O C) U Crt crt crt •r4 1-^ rr o o 1—1 X- u M U X. U) CO c C ■rJ fell H •r-l •r-l •r-l CX a. ex >-. en M r-l Vt 0. CX TD •u ~0 4-1 •r-l o o WJ C) Cl) r= C C r. CO CO CO n 4-1 rJ o o •r-l e c iZ c c u 1-1 >: 4J •r^ 1-1 l-i crt crt W cs crt n crt CJ 0,1 o CX CX u •h < < < < < < < < U w w W Cx, tu U-l p-l CX: P-, a, tO CO C/j O) to H X i— ir^r-irooONONr-Jor^-oo OOOCNOi-iOO o^ --o.co •^(-icoOcJOicjVj • CO 4J 4J E CO D. 11 J3 CJ . a, en eoa.r4ajcoa>.n-i CO )-. u >-l CO •r-l o • O CD o cu E « a. ,o 4-1 u 4-) •H X 3 R f= CO 0 ,Q 4-1 o Cfl CO CO IW 03 )-4 u •H eO CO ■H O u M t-* •f-l en •a rC CD >Xi ^ >, •r-l ■M < uowpstcpsHmss APPENDIX F HERBIVORE DATA i— i co co i m m cm m h vJO \D vD \D I CM in o co on o in o o o in in in o in o oi vD vO vD vO <}■ mo O in ON o o o CO CM i-l r-l in <}• 05 * • r^ o 00 O0 CO CO i— 1 00 00 r^ i r-~ 00 ON 00 o o o cm m o o> o> O O iA in iTivo » i m mom m m o CO CM • CO M M O O O S~S o o o M O O B~S 322 o m o o o C co JC i-i JC U «0 o oo 4-1 £> CM 323 in in o m o m m o in m on in oo \o • vD m I ON m m o m o CO CM oo r^ i cm J3 CO E co in in J? ■* vO v£3 52 CM 00 CO m in o ^d co o co oo oo I co m o m on cr. oo i— i i — r — i ■ — i o in m in in m o> O- <* • on m o vD 1 vD MD M0 o m m in m m o> co in ■-) r~- r>- r» r-- r- i on m o m o m cm en vD (^ en O r^ r-^ \0 vO r-i ro o m o o O co co cr> ■— i r-~. on on oo on i CO o o o m m oo r^- i co on co oo co i i N n n h. r-» r~~ r-~ r». m m o rH r^ co -d- ON ON I o CO in o m o m m m -4 o> m cr\ cr\ on o> oo i -o en m o o m cm m m o in o in cm r-- cm o r-~ o cm n i^. od r-- co en m o o m co cm o\ o\ en o> co co i en o o m m o o v£> <± vO O CM v£> CO CO CO ct\ o in o m o i-i m in o in o LO oo <}• o- C x: 00 CYi CO o% « u ■u APPENDIX G CIRCUIT ENERGY SYMBOLS FOR SIMULATION MODEL Pathway, whose flow is proportional to the quantity in the -> storage or source upstream. Outside source, material or energy, a forcing function. I I Storage, a compartment within the system, a state variable. j Heat sink, loss of potential energy from further use in the — system. K \ Interaction or work gate, intersection of two pathways coupled ^ — ■* to produce an outflow through a mathematical relationship. ) ( Switch, indicates one or more switching actions or choicer j ■>. Producer, self limiting energy and/or material receiver, I ) e.g. plants. o Consumer, self maintaining unit, e.g. herbivores. 325 LITERATURE CITED Abrams, S. M., J. E. Moore and W. R. Ocumpaugh. 1978. Digestibility and intake of cultivars of bermudagrass , digitgrass, and bahiagrass fed to sheep. In Florida beef cattle research rep. IFAS , Univ. Fla., Gaines- ville, Fla. p. 65-66. Ackerly, J. T. 1977. Container Corporation of America, Empire Tract, Brad- ford County, Fla. Paper presented 1977 Spring Symposium. Mutual opportunities for For. Range and Wildl. Mgmt., School of For. Res. and Cons. Resources Rep. $3, IFAS, Univ. Fla. p. 43. Alexander, M. 1964. Introduction to soil micro-biology. John Wiley & Son N.Y. Ammerman, C. B. 1977. Mineral supplementation for the beef cow. 26th Ann. Beef Cattle Short Course, IFAS, Univ. Fla. p. 111-119, Amorocho, J. 1967. Role of infiltration in nonlinear v?atershed analysis processes. Am. Soc. Ag. Eng. Trans. 10:396-399. Anthony, R. G. and N. S. Smith. 1977. Ecological relationships between Mule deer and White-tailed deer in Southeastern Arizona. Ecol. Mono. 47:255-277. Asay, K. II., T. R. Minnick, G. B. Garner, and B. W. Harmon. 1975. Use of crickets in bioassay of forage quality in tall fescue. Crop Sci. 15:585-588 Ballard, R. 1974. Soil testing as a guide to phosphorus fertilization of Slash Pine (Pinus palustris) . PhD thesis, School of For. and Res. Conser., Univ. Fla. Ballard, R. and J. G. A. Fiskell. 1974. Phosphorus retention in coastal plain soils: I. Relationship to soil properties. Soil Sci. Soc. Proc. 38:250-255. Ballard, R. and W. L. Pritchett. 1974. Phosphorus retention in coastal plain soils: II. Significance to forest fertilization. Soil Sci. Soc. Fla Proc. 3S(2) :363-366. Barnes, R. F. 1966. The development and application of in vitro rumen ferment- ation technique. Proc. X Int. Grassl. Congr., Helsinki, Finland, Sect 2 Pap. 19:434-438. Barnes, R. F. 1973. Laboratory methods of evaluating feeding value of her- bage. In chemistry and biocheriis try of herbage, Vol. 3, (ed.) G. W. Butler and R. W. Bailey, Acad. Press, N.Y. p. 179-213. 326 327 Bates, J. E. 1971. Factors affecting critical nutrient concentrations in plants and their evaluation: A review. Soil Sci. 112:116-130. Becker, R. B., P. T. D. Arnold, W. G. Kirk, G. K. Davis and R. W. Kidder. 1957. Minerals for beef cattle. Fla. Agr. Exp. Sta. Bull. #513R. Becker, R. B., J. R. Henderson and R. B. Leighty. 1965. Mineral malnutrition in cattle. Ag. Exp. Sta. Tech. Bull. #699. IFAS, Univ. Fla. 54 p. Becker, R. B., W. M. Neal and A. L. Shealy. 1931. I. Salt sick: its cause and correction. II. Mineral supplements for cattle. Ag. Exp. Sta. Bull. #231, Univ. Fla. Bell, M. K. 1974. Decomposition of herbaceous litter. In biology of plant litter decomposition, Vol. I. (ed . ) C. H. Dickinson and G. J. F. Pugh, Acad. Press N.Y. p. 37-68. Bell, R. H. V. 1971. A grazing ecosystem in the Serengeti. Sci. Am. 225(1): ( 95-104. Eeneke, R. R. and R. Winterboer . 1973. Linear programming application to agriculture. Iowa St. Univ. Press. Ames, Io. Bennett, F. A. and J. L. Clutter. 1968. Multiple-product estimates for un- thinned slash pines plantations - - pulpwood , saw timber, grass. U.S. For. Ser. Pap. #SE-35. USDA For. Ser. S.E. For. Exp. Sta., Asheville, N. Carolina. Birch, H. F. 1959. Further observations on humis decomposition and nitrifi- cation. Plant Soil 9(3) :262-286. Birch, H. F. 1960. Nitrification in soils after different periods of dryness. Plant Soil 12(1) :81-96. Biswell, H. H. and J. E. Foster. 1947. Is rotational grazing on native range practical? Ag. Exp. Sta. Bull. #360, N.C. State Coll., Raleigh, N.C. \S Biswell, H. H., B. L. Southwell, J. W. Stevenson and W. 0. Shepherd. 1942. Forest grazing and beef cattle production in the coastal plains of Georgia./- Ga. Coastal Plain Exp. Sta. Cir. #8, Univ. Ga., Tifton, Ga. Black, C. A. 1968. Soil plant relationship. John Wiley and Son. N.Y. p. 558-653, Black, C. C. 1971. Ecological implications of dividing plants into groups with distinct photosynthetic production capacities. In advances in ecolog- ical research, Vol. 7 (ed.) J. B. Crogg, Acad. Press, N.Y. p 87-114. Black, C. C. 1973. Photosynthetic carbon fixation in relation to net C0? uptake. Ann. Rev. Plant Physiol. 24:253-286. Blair, R. M. and E. A. Epps , Jr. 1969. Seasonal distribution of nutrients in plants of seven browse species in Louisiana. S. For. Exp. Sta. Res. ** Pap. SO-51. USDA, New Orleans, La. . 328 Blair, R. M., H. L. Short, and E. A. Epps, Jr. 1977. Seasonal nutrient yield and digestibility of deer forages from a young pine plantation "* J. Wildl. Mgmt. 41(4) :667-676. Blue, W. G. 1970. The recovery of autumn and winter applied potassium by a warm-season grass from Leon fine sand. Soil and Crop Sci Fla Proc 30:9-15. Bocock, K. L. 1964. Changes in the amounts of dry matter, nitrogen, carbon and energy in decompos ting woodland leaf litter in relation to the activities of the soil fauna. J. Ecol. 52:273-284. Bonsma, J. D. 1965. Wortham lectures in animal ssience. Texas A&M Univ. Press, College Sta. Tex. Bornemissza, C. F. 1960. Could dung eating insects improve our pastures? J. Aust. Inst. Agr. Sci. 26:54-56. Bowers, D. M. and B. A. Haws. 1978. Rangeland insects: A significant factor in range management. Paper presented 31st Ann. Mtg. Soc. Range Mgmt., San Antonia, Tex. Abstracts, p. 40. Brady, N. C. 1974. The nature and properties of soils (8th ed. ) . MacMilli.in Co., N.Y., N.Y. Branson, F. A., G. F. Gifford and J. R. Owen. 1972. Rangeland hydrology. Range Sci. Ser. #1. Soc. Range Mgmt. Denver, Colo. Brown, J. W. and J. L. Schuster. 1969. Effects of grazing on a hardland site in the southern high plains, J. Range Mgmt. 22:418-423. Bryan, L. M. and E. M. Sharrock. 1941. Better pastures for Florida. Fla. St Dept. Ag. Bull. #108. Buchholz, L. W. 1929. History of Alachua County, Florida. Record Pub. Co. St. Augustine, Fla. p. 102-166. ' ^ Bush, L. P., J. A. Bollng, G. Allen and R. C. Buckner. 1972. Inhibitory effects of perloline to rumen fermentation in vitro. Crop Sci 12: 277-279. Camp, P. D. 1932. A study of range cattle management in Alachua County Fla Ag. ExF. Sta. Bull. #248, Univ. Fla. Campbell, R. S. and H. H. Biswell. 1945. Better management oh southern coastal forest ranges. For. Ser. Bull. #AIS-17. USDA , For. Ser Wash ""' D.C. .... Campbell, R. S., E. A. Epps, Jr., C. C. Moreland, J. L. Farr and F. Bonner. 1954. Nutritive values of native plants on forest range in central Louisiana. Ag. and Mech. Col. Ag. Exp. Sta. Bull. #488. Louisiana St. ' Univ. 329 Carlisle, V. W. and F. L. Moritz. 1978. Florida soil identification handbook: Thermic and hyperthermic temperature zones including selected interpreta- tions. Soil Sci. Dept. Univ. Fla. p. 123. Carpenter, L. H., 0. C. Wallmo, and M. J. Morris. 1973. Effect of woody stems on estimating herbage weights with a capacitance meter. J. Range Mgmt. 26(2) : 151-152. Carter, C. W. 1973. Management of undesirable native plant species. In range resources of the southeastern United States, (ed.) R. S. Campbell and W. Keller. ASA spec. pub. #21, American Society of Agronomy, p. 35-40. Cassady, J. T., W. Hopkins and L. B. Whitaker. 1955. Cattle grazing damage to pine seedlings. S. For. Exp. Sta. Occ. Pap. #141. For. Ser. USDA. Cassady, J. T. and W. 0. Shepherd. 1948. Grazing on forested lands. In 1948 yearbook of agriculture. USDA Wash., D.C. p. 468-471. Caswell, K. and F. C. Reed. 1976. The indigestibili ty of C, bundle sheath cellj- by grasshoppers. Oecologia 26:151-156. Caswell, H. , F. Reed, S. N. Stephenson and P. A. Werner. 1973. Photosynthetic pathway and selective herbivory; A hypothesis. Am. Natur. 107(956)465-480. Charley, J. L. 1977. Mineral cycling in rangeland ecosystems. In rangeland plant physiology, (ed . ) R. E. Sosebee., Soc. Range Mgmt. Co. p. 213-256. Chev:, V. 1973. Statistical test of significance are informationless . Paper presented Southern Branch A-n. Soc. Agron. Houston, Tex. Abstracts of Tech. Papers #5. p. 12. Christian, J. J., V. Flyger and D. E. Davis. 1960. Factors in the mass mortality of a herd of sika deer, Cervus neppon. Chesapeak Sci. 1:79-95. Cochran, W. G. and G. M. Cox. 1957. Experimental designs, 2nd ed . J. Wiley and Sons, N.Y. Cole, V. C, G. S. Innis and J. W. R. Stewart. 1977. Simulation of phosphorus cycling in semiarid grasslands. Ecol. 58(1):1-15. Coleman, S. W. and K. M. Barth. 1972. Quality of diets selected by grazing animals and its relation to available forage and species composition of * pasture. J. Anim. Sci. 36 (4) :754-761. Conner, D. 1977. Fla. State Agric. Comm. Introduction to the 1977 Beef Cattle Short Course., Gainesville, Fla. Cook, W. C. 1954. Common use summer range by sheep and cattle. J. Range Mgmt. 7(1): 10 -13. Cook, W. C., R. D. Child and L. L. Larson. 1977. Digestible protein in range forages -is an index to nutrient content and animal response. Sci. Ser. #29, Colo. St. Univ. Fort Collins, Colo. 330 Coppock, E. E., R. W. Everett and R. L. Belyea. 1976. Effect of low calcium or low phosphorus diets on free choice consumption of dicalcium phosphate by lactating dairy cows. J. Dairy Sci . Vol. 59(3) :571-580. Coupland, R. T. 1975. Productivity of grassland ecosystems. In productivity of world ecosystem. Natl. Acad. Sci., Wasl. D.C. p. 44-49. Cramp ton, E. W. and L. A. Maynard. 1938. The relation of cellulose and lignin content to the nutritive value of animal feeds. J. Nutrition 15:383-395. Cubillos, G. 1976. Some management factors in the production of tropical grasses. Tropical Agr. Res. and Training C t. Bull. 2014 #1, Turriabla/ Costa Rica. Cunha, T. J. 1976. Introduction. In beef cattle in Fla. Bull. #28, IFAS, p. 1-11. Curlin, J. W. 1970. Nutrient cycling as a factor in site productivity and forest fertilization. In tree growth and forest soils (ed.) C. T. Young- berg and C. B. Davey, Proc. 3rd N. Am. For. Soils Conf. N.C. State Univ. p. 313-325. Currie, P. 0., M. J. Morris and D. L. Neal. 1973. Uses and capabilities of electronic capacitance instruments for estimating standing herbage. Part II. Sown ranges. J. Brit. Grassld. Soc. 28:155-160. Debellevue, E. B., R. J. Gutierrez and S. Bayley. 1975. Actions and research alternatives for pasture systems. In carrying capacity for man and nature, in South Florida. Center for Wetlands, Univ. Fla., Gainesville, p. 441-447. Deevey, Jr., E. S. 1970. Mineral cycles., Sci. Am. 223 (3) : 149-158. Dickinson, C. H. 1974. Decomposition of litter in soil. In C. H. Dickinson and G. J. F. Pugh (ed.) biology of plant litter decomposition. Vol. 2, Acad. Press, N.Y. p. 633-658. Dodd, D. 1954. Popular brahman cattle first introduced into Florida in 1859. Fort Myers, Fla., News-Press, Eeb. 17. Downton, W. J. S. 1975. The occurrence of C^ photosynthesis among plants. Photosynthetica 9(1):96-105. Drury, W. H. and I. C. T. Nisbet. 1973. Succession. J. Ar ' Id . Arbor. 54(3) :331-368. Duffy, P. D., J. D. Schrieber, D. C. McClurkin and L. L. McDowell. 197S. Aqueous and sediment phosphorus yields from fine southern pine water- sheds. J. Envir. Qual. 7(l):45-50. Duncan, W. H. and L. E. Foote. 1975. Wild flowers of the southeastern United States. Univ. Ga. Press, Athens, Ga. 331 Duvall, V. L. and J. B. Hi]mon. 1965. New grazing research programs for , southern forest ranges. J. Range Mgmt. 18(3) :132-136. Ensminger, M. E. 1976. Beef cattle science. 5th ed. Interstate, Danville, 111. Etherington, J. R. 1975. Environment and plant ecology. J. Wiley and Sons, N.Y. Evans, G. C. and E. W. Tisdale. 1972. Ecological characteristics of Aristida longiseta and Agropyron s pica turn in West Central Idaho. Ecol. 53:137-142. Evans, L. T. 1975. Crop physiology: some case histories. Cambridge Univ. Press , N.Y. Ewel, K. C, J. F. Gamble and A. E. Lugo. 1975. Aspects of mineral-nutrient cycling in a southern mixed -hardwood forest in North Central Florida. In mineral cycling in southeastern ecosystems, (ed.) F. G. Howell, J. B. Gentry and M. H. Smith. Tech. Inf. Cent. U.S. Energy Res. and Devel. Ad. p. 700-714. Felts, B. L. 1976. Short duration grazing in North Florida flatwoods. MS , thesis. Sch. For. Res. and Conser., Univ. Fla. Florida Department of Agriculture Staff. 1976. Florida agricultural statis- tics: Livestock summary 1976. Fla. Dept. Ag. Tallahassee, Fla. Forest Service Staff. 1976. Volume, yield and stand tables for second-growth pines. For. Ser. Misc. Pub. #50. USDA. Fried, M. and H. Broeshart. 1967. The soil-plant system in relation to inorganic nutrition. Acad. Press. N.Y. Gallaher, R. N. 1975. The occurrence of calcium in plant tissue as crystals of calcium oxalate. Comrnun. in Soil Sci. and Plant Anal. 6(3) :315-330. Gatherum, G. E. 1955. An analytical approach to the management of forest lands for beef cattle and timber production. Iowa St. J. Sci. 34(4): ' 565-574. Geng, S. and F. J. Hills. 1978. A procedure for determining numbers of experimental and sampling units. Agron. J. 70(3) :441-444. Geraldson, C. M. 1977. Nutrient intensity and balance. In M. Stelly (ed.) Soil testing. ASA Spec. Pub. #29. p. 75-84. Gillard, P. 1967. Coprophagous beetles in pasture ecosystems. J. Aust. Inst. Agr. Sci. 33:30-34. Given, B. B. 1966. Biological control of weeds and insect pest in New Zealand, In natural enemies in the Pacific area: Biological controls. Mushi 39, Suppl. p. 131-136. 33: Golley, F. B. 1975. Productivity and mineral cycling in tropical forests. In productivity of world ecosystems. NAS Wash., D.C. p. 106-115. Goodlett, J. C 1954. Vegetation adjacent Lo the border of the Wisconsin drift in Potte County, Pennsylvania. Harvard For. Bull. #25. Gould, F. W. 1968. Grass systematics. McGraw-Hill Book Co., N.Y. Grant, W. E., N. R. French. 1975. Functional role of small mammals in grassland ecosystems. IBP Grassland Biome. Natural Res. Ecol. Lab. Colo. St. Univ., Ft. Collins, Colo. Gray,,K. R. and A. J. Biddleston. 1974. Decomposition of urban waste. In 'biology of plant litter decomposition (ed.) E. H. Dickinson and G. J. F. Pugh, Vol. 2, Acad. Press, N.Y. p. 743-775. Grelen H. E. and V. L. Duvall. 1966. Common plants of longleaf pine blue- stem range. U.S. For. Ser., Res. Paper SO-23. S. for Exp. Sta. USDA. Grier, C. C. and R. S. Logan. 1977. Old-growth Pscudotsuga menziesii communities of a Western Oregon watershed: Biomass distributions and production budgets. Ecol. Mono. 47 (4) :373-400. Griffin, J. W. 1974. Archeology and environment in South Florida. In environments of South Florida: Past and present. Miami Geol. Soc. Memoir 2. (ed.) D. J. Gleason, Miami Geol. Soc. p. 342-346. Gumma, A. N. 1977. Cattle diet species determination as affected by rest treatments and forage availability. MS thesis. Sch. of For. Res. Consv. Univ. Fla. Gutierrez, R. J. 1978. Energy analysis and computer simulations of pastures in Florida. MS thesis. College Engr., Univ. Fla., Gainesville. Gutierrez, R. J., E. DeBellevue and M. Sell. 1975. Agriculture. In water ouality models for understanding potential eutrophication in Lake Okeechobee, Florida. Ctr. for Wetlands, Univ. Fla., Gainesville, p. 534-590. Hafez, E. S. E., M. W. Schein and R. Eubank. 1969. The behavior of cattle, in the behavior of domestic animals, (ed.) E. S. E. Hafez. Williams and Wilkins Co., Baltimore, Md . p. 235-295. Hall, C. A. S. and J. W. Day Jr. 1977. Systems and models: Terms and basic 'principles. In ecosystem modeling in theory and practice, (ed.) C. A. S. Hall' and J. W. Day Jr., John Wiley and Sons, N.Y. p. 5-36. Hall, C. A. S.,'J: W. Day Jr., and H. T. Odum. 1977. A circuit language for 'energy and matter. In ecosystem modeling in theory and practice. (ed . ) C. A. S. Hall and J. W. Day Jr., John Wiley and Sons, N.Y. p. 38-48. Halls, L. K. 1955. Grass production under dense longleaf -slash pine canopies^ USDA S. E. Fo-r. Ser. Res. Note #83 . Asheville, N.C. 333 Halls, L. K. 1977. Southern fruit-producing woody plants used by wildlife. S. For. Exp. Sta,, USDA For. Ser. Gen. Tech. Rep. SO-16 . Halls, L. K., D. M. Hale and F. E. Knox. 1957. Seasonal variation in grazing use, nutritive content, and digestibility of wiregrass range. Ag. Exp. ^ Sta. Tech. Bull. //N.5.U. Univ. Ga„ Halls, L. K., D. M. Hale and B. L. Southwell. 1956. Crazing capacity of wiregrass-pine ranges in Georgia. Ag. Exp. Sta. Tech. Bull. //N.S.2, <-' Univ. Ga. Halls, L. K., R. H. Hughes, R. S. Rumme., and B. L. Southwell. 1964. Forage and cattle management in longleaf-slash pine forest. Farmer Bull. #2199, / USDA, Wash. D.C. Halls, L. K. and T. H. Ripley. 1961. Deer browse plants of Southern forest. S. and S.E. For. Exp. Sta. USDA. Harcombe, P. A. and P. L. Marks. 1977. Understory structure of a mesic forest in Southern Texas. Ecol. 58(5) :1144-1151 . Harlow, R. F. 1961. Fall and winter foods of Florida Whi te-tailed deer. Quart. J. Fla. Acad. Sci. 24(l):19-38. Harlow, R. F. and R. G. Hooper. 1971. Forages eaten by deer in the south- east. Proc. 25th Ann. Conf. S.E. Assoc. Game and Fish Comm. p. 18-46. Harlow, R. F. and F. K. Jones. 1965. The White-tailed deer in Florida. Tech. Bull. #9, Fla. Game and Fresh Water Fish Comm. Harris, L. D. 1972. An ecological description of a semi-arid East African ecosystem. Range Sci. Dept. Sci. Ser. //ll, Colo. St. Univ. Harris, L. E., G. P. Lofgreen, C. J. Kercher, R. J. Raleigh and V. R. Bohmon. (1967). Utah Ag . Exp. Sta. Bull. #471. Heady, H. F. 1975. Rangeland management. McGraw-Hill Book Co., N.Y. Henneberg, W. 1859. Veber den Heuwert der Futterstof fe. J. F. Landwirt- schaft 7:B (III) Henzell, E. F. and D. 0. Norris. 1962. Review of nitrogen in the tropics with particular reference to pastures. Symposium, Div. Trop. Post. C.S.I.R.D. Commw. Bur. Post Fd . Crops, Bull. #46. Hurley, Engl. p. 1-18. Henzell, E. F. and P. J. Ross. 1973. The nitrogen cycle of pasture eco- systems. In chemistry and biochemistry of herbage. Vol. 2. (ed.) G.W. Butler and R. W. Bailey, Acad. Press. N .Y . p 228-247. Hewitt, G. B., E. W. Huddleston, R. J. Lavigne, D. N. Jeckert and J. G. Watts. 1974. Rangeland entomology. Range Sci. Ser. #2, Soc. Range Mgmt., Denver, Colo. 334 Heyword, F. and R. M. Barnette. 1936. Field characteristics and partial chemical analysis of the humus layer of longleaf pine forest soils. Ag. Exp. Sta. Bull. #302, Univ. Fla. Hilmon, J. G. 1968. Autecology of Saw Palmetto (Serenoa repens). PhD thesis, y-- Dept. of Bot., Duke Univ. Hilmon, J. B. and J. E. Douglas. 1967. Potential impact of forest fertili- zation on range, wildlife and watershed management. In forest fertili- K zation: Theory and practice symposium. T.V.A. Gainesville, Fla. p. 197-202. Hilmon, J. B., C. E. Lewis and J. E. Bethune. 1962. Highlights of recent results of range research in Southern Florida. Proc. Soc. Am. For., v' Atlanta, Ga . p. 73-76. Hitchcock, A. S. 1951. Manual of the grasses of the United States, 2nd ed. (Revised by A. Chase), USDA Misc. Pub. #200. Hodges, E. M., G. B. Killinger, J. E. McCaleb, 0. C. Ruelke, R. J. Allen, Jr., S. C. Schank and A. E. Kretschmer, Jr. 1975. Pangola digitgrass. Ag. Exp. Sta. Bull. #718a, IFAS , Univ. Fla. Holdridge, L. R. 1967. Lifezone ecology. Tropical Sci. Ctr., San Jose, Costa Rica. Hormay, A. I,. 1970. Principles of rest-rotation grazing and multiple-use land management. B.M., USDA and For. Ser. USDA #TT-4 (2200). Horn, H. S. 1976. Succession. In theoretical ecology: Principle and Practices, (ed.) R. M. May, W. B. Saunders & Co., Toronto, p. 187-204. Howes, J. R. 1964. The comparative physiology of Bos taurus and Bos indicus cattle on two nutritional levels. PhD thesis, Dept. Anim. Sci., Univ. Fla. p. 100-114. Hughes, R. H. 1974. Management and utilization of pineland threeawn range ^ in South Florida. J. Range Mgmt. 27 (3) :186-191 . Hundley, L. R. 1956. The available nutrients in selected deer browse species growing on different soils. PhD thesis, Dept. Biol. Virginia Polytechnic Institute, Blacksburg. Hungate, R. E. 1966. The rumen and its microbes. Acad. Press., N.Y. Hutton, J. B., K. E. Jury and E. B. Davies. 1967. Studies of the nutritive value or New Zealand dairy pasture. 5. The intake and utilization of potassium, sodium, calcium, phosphorus and nitrogen in pasture herbage by lactating dairy cattle. N. Z. J. Agr . Res. 10:367-388. IFAS. 1975. Agricultural growth in an urban age. IFAS, Univ. Fla. Jahromi , S. T. 1967. The effect of nitrogen and phosphorus on seedling growth of Pinus brut La. MS thesis, Univ. Fla. Jameson, D. A., S. D'Aquino and E. T. Bartlett. 1974. Economics and manage- ment planning of range ecosystems. A. A. Balkema, Rotterdam, Netherlands. 335 Jansen, H. C. 1977. Range RAM user's manual part I: User's guide and instructions. Pac. S.W. For. and Range Exp. Sta. For. Ser. USDA, Berkeley, Calif. Jaques, H. E. 1947. The insects. 2nd ed . How to know series. Wm. C. Brown Co. , Dubuque, Io. Killinger, G. B. 1948. Effect of burning and fertilization of wiregrass on ... pasture establishment. J. Am. Soc. Agron. 40(5) :l-5. Killinger, G. B. 1960. Results of five years research at "Beef Research Unit". In report of the 17th southern pasture and forage crop improve- ment conference, Univ. Ark., Fayetteville, Ark. USDA., GCR. p. 16-23. Kirk, W. G., R. L. Shirley, E. M. Hodges, G. K. Davis, F. M. Peacock, J. F. Easley and F. G. Martin. 1970. Production performance and blood and bone composition of cows grazing Pangolagrass pastures receiving different phosphate fertilizer. Ag. Exp. Sta. Tech. Bull. #735, IFAS, Univ. Fla. Koger, M., W. G. Blue, G. B. Killinger, R. E. L. Greene, H. C. Harris, J. M. Myers, A. C. Warnick and N. Gammon Jr. 1961. Beef production, soil and forage analyses, and economic returns from eight pasture programs in North Central Florida., Fla. Agr. Exp. Sta., Tech. Bull. #631. Laessle, A. M. 1942. The plant communities of the Welaka area. Univ. Fla. / Biol. Sci. Ser. #4. Lamprey, H. F. 1963. Ecological separation of the large mammal species in the Tarangire Game Reserve, Tanganyika., E. Afr. Wildl. J. 1:63-92. Larcher, W. 1975. Physiological plant ecology. Springer-Verlag, N.Y. Lay, D. U. 1956. Some nutritional problems of deer in southern pine type. Proc. S.E. Assn. of Game and Fish Comm. 10:53-58. Lay, D. U. 1964. The importance of variety to southern deer. Proc. 18th Ann. Conf. S.E. Assn. of Game and Fish Comm. Clearwater, Fla. p. 57-62. Leaf, A. L. 1973. Plant analysis as an aid in fertilizers: II. A petro- graphic study of their alteration in soils. Soil Sci. Soc. Amer. Proc. 22:24-32. Lebedjantzev, A. N. 1924. Drying of soil, as one of the natural factors in maintaining soil fertility. Soil Sci. 18:419-447. Leigh, J. H. 1961. The relative palatability of various varieties of weep- ing lovegrass (Eragrostis curvula) . J. Br. Grass 1. Soc. 16:135-140/ 336 Leithead, H, L. 1973. Soil surveys - guide to forage quantity and quality. In range resources of the Southeastern United States, (ed.) R. S. Camp- .-' bell, ASA Spec. Pub. #21. Am. Soc. Agron. Madison, Wis. p. 15-34. Leithead, H. L., L. L. Yarlett and T. N. Shiflet. 1971. 100 native forage grasses in 11 southern states. USDA Soil Conser. Ser. Ag. Handbook #389. Lemon, P. C. 1946. Prescribed burning in relation to grazing in the long- 1 leaf-slash pine type. J. For. 44(2) :115-117. Leopold, A. C. and P. E. Kriedeman. 1975. Plant growth and development. • . McGraw-Hill Book Co., N.Y. Leopold, A., L. K. Sowles and D. L. Spencer. 1947. A survey of overpopulated deer ranges in the U. S. J. Wildl. Mgmt. 11 (2) :162-177. Leukel, W. A. and W. E. Stokes. 1942. Establishing carpetgrass under range conditions by controlled burning and seeding. Ag. Exp. Sta. Press Bull/ #571., Univ. Fla. Levi, H. and L. R. Levi. 1968. A guide to spiders. Golden Press, N.Y. Lewis, C. E. 1963. Early study results show application of rock phosphate boosts native forage. Fla. Cattlemen and Livestock J., June 1967. Lewis, C. E. 1972. Chopping and webbing control saw-palmetto in South ^ Florida. USDA For. Ser. Res. Note. SE-177, S.E. For. Exp. Sta, Asheville, N.C. Lewis, C. E. 1977. Principles and status of integrated management from the range viewpoint. In mutual opportunities for forest, range, and wildlife management, (ed.) A. S. Jensen, Sch. For. Res. and Consv. , Res. Rep. #4, IFAS, Univ. Fla. p. 2 6-34. Lewis, C. E., R. S. Lowrey, W. G. Monson and F. E. Knox. 1975. Seasonal trends in nutrients and cattle digestibility of forage on pine-wiregrass ; range. J. Anim. Sci. 41(1)208-212. Likens, G. E., F. H. Bormann, N. M. Johnson and R. S. Pierce. 1967. The calcium, magnesium, potassium and sodium budgets for a small forested ecosystem. Ecol. 48:772-785. Littell, R. C. and G. 0. Mott. 1974. Computer assisted design and analysis of response surface experiments in agronomy. Soil and Crop Sci. Fla. Proc. 34:94-97. Little, C. 0. 1977. Capabilities of various physiographic regions, dis- cussion. In forage-fed beef; production and marketing alternatives in the South, (ed.) J. A. Stredemann, D. L. Huffman, J. C. Purcell, and 0. L. Walker, S. Coop. Ser. Bull. #220. p. 449-450. 337 Loewer, 0. J., E. M. Smith, G. Benock, T. C. Bridges, L. Wells, N. Gay, S. Burgess, L. Springate and D. Debertin. 1978. A simulation model'for assessing alternative strategies of beef production with land, energy and economic constraints. Paper 1978 S.E. - 3. W. Reg. Mtg. Am. Soc. Ag. Eng., Houston, Tex. Lohrey, R. E. 1974. Growth and yield in longleaf pine plantations. In symposium on management of young pines. Charleston, S. C., S. and S E * For. Exp. Sta. USDA. p. 88-96. Loneragan, J. F. 1973. Mineral absorption and its relation to the mineral composition of herbage. In chemistry and biochemistry of herbage, Vol. 2. (ed.) G. W. Butler and R. W. Bailey, Acad. Press, N.Y. p. 103-125! Ludlow, M. M. 1976. Physiology of growth and chemical composition. In ^ tropical pasture research: principles and methods, (ed.) N. H. Shaw and W. W. Bryan., Bull. #51. Commonwealth Agric. Bureaux, Hurley Enel p. 251-276. Lusby, G. C. 1970. Hydrologic and biotic effects of grazing vs. no-grazing near Grand Junction, Colo., J. Range Mgmt. 23:256-260. Lutz, H. J. 1940. Disturbance of forest soil resulting from the uprooting of trees. Yale Sch. For. Bull. #45. Lyford, V. H. and W. L. Pritchett. 1976. Slash pine root systems. CRIFF Pub. #76-1., Univ. Fla. MacDiarmid, B. N. and B . R. Watkin. 1971. The cattle dung patch: I. effect of dung patches on yield on botanical composition of surrounding and underlying pasture. J. Br. Grassl. Soc. 26:239-245. MacDiarmid, B. N. and B. R. Watkin. 1972. The cattle dung patch on the chemical status of the soil, and ammonia nitrogen losses from the patch J. Br. Grassl. Soc. 27:43-47. Mackie, R. J. 1970. Range ecology and relations of mule deer, elk, and cattle in the Missouri River Breaks, Montana, Wildl. Mono. #20. p. 1-78. Malavola, E. and V. F. DaCruz. 1971. A meaning for foliar diagnosis. In present advances in plant nutrition, Vol. 1. (ed.) R. M. Samish. Gordon & Breach Sci. Pub. N.Y. p. 1-13. Mann, W. F., Jr. 1947. Sheep damage to longleaf pine seedlings. In southern forestry notes. Southern For. Exp. Sta. Bull. #52. New Orleans. Margalef, R. 1963. Perspectives in ecological theory. Univ. Chi. Press. Martin, W. E. and J. E. Matocha. 1973. Plant analysis as an aid in the fertilization of forage crops. In soil testing and plant analysis (rev. ed.) (ed.) L. M. Walsh and J. D. Beaton. Soil Sci. Soc. Am., Madison, Wise. p. 393-426. 338 Matches, A. G. 1970. Pasture research methods. Proc. Natl. Conf. For. Qual. Eval., Util. Nebr. Ctr. Cont. Ed., Lincoln, p. 11-132. Mattson, W. J. and N. D. Addy. 1975. Phytophagous insects as regulators of forest primary productivity. Sci. 190(4214) :515-521. Maugh, T. H. II. 1978. The fatted calf (II): the concrete truth about beef Sci. Vol. 199(4327) :413. Maxwell, G. T. 1888. Veterinary and physiology section. Fla. Ag. Exp. Sta Bull. #2. Maynard, L. A. and J. K. Loosli. 1969. Animal nutrition, 6th ed., McGraw- Hill Book Co. , N.Y. McCarty;, M. K. and A. P. Mazurak. 1976. Soil composition in eastern Nebraska after 25 years of cattle grazing management and weed control. J, Range Mgtnt. 29(5):384-386. McDowell, R. E. 1972. Improvement of livestock production in warm climates. W. H. Freeman and Co., San Francisco, Calif. McEwen, L. C. 1957. Nutrient requirements of the white-tailed deer. Trans. N. Am. Wildl. Conf., 22:119-131. McFee, W. W., W. H. Smith and J. Gooding. 1977. Seasonal trends in foliar N used to predict nitrogen fertilizer response in slash pines. Presented Amer. Soc. of Agron. Mtg. , Nov. Los Angeles, Agron. Abstracts. McKenzie Jr., E. and L. T. Kurtz. 1976. Effect of pre-treatment on loss of ni trogen-15-labelled fertilizer nitrogen from waterlogged soil during incubation., Soil Sci. Am. Proc. 40- (4) : 534-537. McLean, E. 0. 1977. Contrasting concepts in soil test interpretations: sufficiency levels of available nutrients versus basis cation saturation ratios. In soil testing: correlation and interpreting the analytical results, (ed.) M. Stelly. ASA Spec. Pub. #29., Madison, Wise. p. 39-54. Mead, D. J. and W. L. Pritchett. 1975. Fertilizer movement in a slash pine ecosystem. I. Uptake of N and P and N movement in the soil. Plant and Soil 43:451-465. Melsted, S. W. and T. R. Feck. 1973. The principles of soil testing. In soil testing and plant analysis. (ed.) L. M. Walsh and J. D. Beaton Soil Sci. Soc. Amer. Madison, Wise. p. 13-22. Melsted, S. V. and T. R. Peck. 1977. The Mitscherlich-Bray growth function. In soil testing: correlating and interpreting the analytical results, (ed.) M. Stelly. ASA Spec. Pub. #29, Madison, Wise. p. 1-18. Merck. 1967. The Merck veterinary manual. 3rd ed . , Merck and Co. Pub. Merrill, L. B. 1959. Heavy grazing lowers range carrying capacity. Tex. Ag. Prog. 5(2) :18. 339 Merrill, L. B., P. 0. Reardon and C. L. Leinweber. 1966. Cattle, sheep, goats - mix'em up for higher gains. Texas Ag. Prog. 12(4):13-14. Merrill, L. B., J. G. Teer and 0. C. Wallmo, 1957. Reaction of deer popula- tions to grazing practices. Tex. Ag. Prog. 3(5):10-12. Merrill, L. B. and V. A. Young. 1952. Range management studies on the ranch experiment station. Tex. Ag. Exp. Sta. Prog. Rep. #1449. Texas A&M, College Sta., Tex. Milchunas, D. G. 1977. In vivo - in vitro relationships of Colorado mule deer forages. MS thesis, Colo. St. Univ., Ft. Collins. Minson, D. J. 1971. Influence of lignin and silicon on summative system for assessing the organic matter digestibility of Panicum. Aust. J. Ag. Res. 22:559-598. Minson, D. J. and R. Milford. 1967. In vitro and fecal nitrogen technique for prediction the voluntary intake of Chloris gayana . J. Brit. Grassl. Soc. 22:170-175. Minson, D. J., T. H. Stobbs, M. T. Hegarty and M. J. Playne. 1976. Measuring the nutritive values of pasture plants. In tropical pasture research: principles and methods (ed.) N. H. Shaw and W. W. Bryan. Bull. #51. Commonwealth Ag. Bur. Hunley, Engl., p. 308-337. Mitsch, W. J. 1975. Systems analysis of nutrient disposal in cypress wet- lands and lake ecosystems in Florida. PhD thesis, Sch. of Env. Eng. Sci. Univ. Fla. p. 32-40. Moore, J. E. 1973. Effects of maturity on the energy and protein value of Bahia, Bermuda, and Pangola. , 22nd Ann. Beef Cattle Short Course, Univ. Fla. Moore, J. E. 1977. Limitations of Florida forages for meeting the nutritional needs of the cow. 26th Ann. Beef Cattle Short Course, Univ. Fla. p. 91-103, Moore, J. E. and G. 0. Mott. 1973. Structural inhibitors of quality in tropical grasses. In antiquality components of forages (ed.) A. G. Matches. CSSA Spec. Pub. //4. Crop Sci. Am., Madison, Wise, p. 53-98. Moore, J. E. and G. 0. Mott. 1974. Recovery of residual organic matter from in vitro digestion of forages. J. Dairy Sci. 57:1258-1259. Moore, J. E. and G. 0. Mott. 1976. Fermentation tubes for in vitro digestion of forages. J. Dairy Sci. 59:167-169. Moore, J. E. and 0. C. Ruelke. 1978. Composition and quality of 'Florida 66' alfalfa, ' Suwannee ' bermudagrass and 'Pangola' digitgrass hays. Fla. Beef Cattle Res. Rep., p. 61-64. Moore, L. A. 1964. Symposium on forage utilization: nutritive value of forage as affected by physical form. I. general principle involved with ruminants and effect of feeding pelleted or wafered forage to dairy cattle. J. Anim. Sci. 23:230-238. Moore, R. M. 1966. Interaction between the grazing animal and its environment. 9th Internatl. Cong. Anim. Prod. Oliver and Boyd, N.Y. p. 188-195. 340 Moore, R. T. 1977. Gas exchange in photosynthetic pathways in range plants. In rangeland plant physiology, (ed.) R. E. Sosebee., Range Sci. Ser. #u[ Soc. Range Mgmt. , Denver, Colo. p. 1-46. Moore, W. H. and R. B. Manney. 1962. Deer browse plants in Loblolly and Water Oak - Gum types of Piedmont Georgia. S.E. For. Exp. Sta Res Note #172. USDA. Morris, M. J. and J. L. Kovner. 1970. Forage quality and animal production: Some statistical problems. In range and wildlife habitat evaluation: A research symposium. For. Ser. Misc. Pub. #1147. USDA. p. 32-36. Morrow, R. R. 1955. The case against woodlot grazing. Cornell Ext. Bull #939, N. Cornell Univ. Mott, G. 0. I960. Grazing pressure and the measurement of pasture production. „ Proc. 8th Int. Grassl. Cong., Reading. Engl. p. 601-611. Mott, G. 0. 1973. Evaluating forage production. In forages: the science of grassland agriculture, 3rd ed . (ed.) M. E. Heath, D. S. Metcalfe and R. F. Barnes, Io. St. Univ. Press, Ames, Io. p. 126-135. Mott, C. 0. 1974. Nutrient recycling in pastures. In forage fertilization, (ed,) D. Mays, Am. Soc. Agron. Madison, Wise. p. 323-339. Mulder, E. G., T. A. Lie and J. W. Woldendrup. 1969. Soil biology reviews of research., UNESCO, Paris, p. 188-192. Murphy, D. A. and J. A. Coates. 1966. Effects of dietary protein on deer. Trans. 31st N.A. Wildl. and Nat. Res. Conf. Pittsburg, Pa. p. 129-139. Murphy, J. and J. P. Riley. 1962. A modified single solution method for the determination of phosphate in natural waters. Anal. Chem. Acta. 27:31-36. NAS. 1962. Basic problems and techniques in range research. Natl. Acad. Sci. Natl. Res. Council #890. Wash., D.C. p. 63. NAS. 1976. Nutrient requirements of beef cattle., 5th ed. Natl. Acad. Sci. Natl. Res. Council. WasB. , D. C. Neal, D. L. and J. L. Neal. 1973. Uses and capabilities of electronic capaci- tance instruments for estimating standing herbage. Part I. history and development. J. Brit. Grassl. Soc. 28:81-89. Neller, J. R. 1959. Extractable sulfate-sulfur in soils of Florida in rela- tion to amount of clay in the profile. Soil Sci. Soc. Am. Proc. 23 (5): 346-343. Nelson, B. D., H. D. Ellzey, C. Montgomery and E. B. Morgan. 1972. Factors affecting the variability of an i_n vitro rumen fermentation technique for estimating forage quality. J. Dairy Sci. 55:358-366. 341 Nelson, W. L. , A. Mehlich and E. Winters. 1953. The development, evaluation, and use of soil test for phosphorus availability. In soil and fertilizer- phosphorus, (ed.) W. H. Pierce and A. G. Norman. Vol. IV. Agronomy, A series of monographs , Acad. Press., N.Y. p. 153-188. Nicol, A. A. 1936. The experimental feeding of deer. Trans. 1st NA Wildl Conf. p. 403-410. Nieland, L. T. 1945. Timber-grazing-game : A suggested combined program for uncultivated lands. Ag. Ext. Ser. Bull. #127., Univ. Fla. Norris, J., A. Wilhite and G. Westmoreland. 1975. Evaluation of short dura-, tion grazing systems. Advisory range TX-1, Soil Consv. Ser. USDA, Temple Tex. Odum, E. P. 1969. The strategy of ecosystem development. Sci. 164:262-269. Odum, E. P. 1971. Fundamentals of ecology. W. B. Saunders Co Phila Penn . Odum, H. T. 1971. Environment, power, and society. Wiley-interscience, N.Y. Odum, H. T. 1972. An energy circuit language for ecological and social systems: Its physical basis. In systems analysis and simulation in ecology. Vol. II. (ed.) B. C. Patton. Acad. Press. N.Y. Odum, H. T. and E. C. Odum. 1976. Energy basis for man and nature. McGraw- Hill Co. Oefinger, Jr. S. W. and L. K. Halls. 1974. Identifying woody plants valuable to wildlife in southern states. S. For. Exp. Sta . , USDA For Ser Res Paper SO-92. 0'J.son, J. S. 1975. Productivity of forest ecosystems. In productivity of vorld ecosystems. Natl. Acad. Sci. Wash., D. C. p. 33-43. Orr, H. K. 1975. Recovery from soil compaction on bluegrass ran°e in the v Black Hills. ASAE Trans. 18(6)1076-1081. Ovington, J. D. 1965. The form, weights, and productivity of tree species grown in close stands. New Phytol. 55:289-304. Ovington, J. D. 1968. Some factors affecting nutrient distribution within ecosystems. In functioning of terrestrial ecosystems at the primary production level, (ed . ) F. E. Echardt. UNESCO Sym. Carmorine, S. A., Liege, Belgium, p. 95-105. Parrott, R. T. 1967. A study of wiregrass (Aristida stricta Michx.) with particular reference to fire. MS thesis, Dept. of Bot. Duke Univ. Patterson, D. D. 1935. The growth, yeild, and composition of certain tropical fodders. J. Agr. Sci. 25:369-394. 342 Peacock, F. M., W. G. Kirk and M. Koger. 1965. Factors affecting the weaning weight of range calves. Ag. Exp. Sta. Tech. Bull, #578a., IFAS, Univ. Fla. Pearson, H. A. 1967a. Cattle diet digestibilities determined for components. J. Range Mgmt. 20:405-406. Pearson, H. A. 1967b. Effects of delays in innoculum collections on artifi- cial rumen digestibilities. J. Range Mgmt. 20:332-333. Pearson, H. A. 1970. Digestibility trials: in vitro techniques. In range and wildlife habitat evaluation: a research symposium. For. Ser, Misc. Publ. #1147, USDA. p. 85-92. Petersen, R. G. 1977. Use and misuse of multiple comparison procedures, Agron. J. Vol. 69(2) .-205-208. Petersen, R. G. , H. L. Lucas and W. W. Woodhouse, Jr. 1956a. The distribution of excreta by freely grazing cattle and its effect on pasture fertility: I. excretal distribution. Agron. J. 48:440-444. Petersen, R. G., H. L. Lucas and W. W. Woodhouse, Jr. 1956b. The distribution of excreta by freely grazing cattle and its effect on pasture fertility: II. effect of returned excreta on the residual concentration of some fertilizer elements. Agron. J. 48:444-449. Petrusewicz, K. and W. L. Grodzinski. 1975. The role of herbivore consumers in various ecosystems. In productivity of world ecosystems. Natl. Acad. Sci. Wash, D. C., p. 64-70. Pfander, W. H. , W. A. Hargus , W. Tyree and P. C. Stone, 1964. The use of crickets to evaluate ruminant ration. Federation and Society Symoosia. 23:135. Pielou, E. C. 1975. Ecological diversity. Wiley-interscience Pub., N.Y. Fieper, R. D. 1974. Effects of herbivores on nutrient cycling and distri- butions. Proc. 2nd. U.S . -Australia Range Workshop. Pritchett, W. L. 1968. Progress in the development and standards for soil and foliar diagnosis of phosphorus deficiency in slash pines. In forest fertilization: theory and practice. TVA, Muscle Shoals, Ala. p. 81-87. Pritchett, W. L. 1976. Forest soils: properties and management. IFAS, Univ. Fla . Pritchett, W. L. and J. W. Gooding. 1975. Fertilizer recommendations for pines in the southeastern coastal plain of the U.S., IFAS Bull. #774. Univ. Fla. Pritchett, W. L. and W. K. Smith. 1974. M3nagement of wet savanna forest soils for pine production. Ag. Exp. Sta. Tech. Bull. #762., IFAS, Univ. Fla, 343 Quinn, J. A. and D. F. Hervey. 1970. Trampling losses and travel by cattle on sand hill range. J. Range Mgmt. 23(1)25-26. Ralston, C. W. and G. E. Hatchell. 1971. Effects of prescribed burning on physical properties of soil. In prescribed burning symposium proceedings. S. E. For. Exp. Sta., USDA, Charleston, S. C. p. 6Sr83. Range Term Glossary Committee. 1974. A glossary of terms used in range management, 2nd ed . , (ed.) M. M. Kluthmann, Soc. Range Mgmt., Denver, Colo. Read, R. A. 1957. Effect of livestock concentrations on surface-soil porosity with shelter belts. J. For. 55:529-530. Reed, M. J. and R. A. Peterson. 1961. Vegetation, soil, and cattle responses to grazing on Northern Great Plains range. USDA Tech. Bull. #1252. Reuss, L. A. 1958. Cost of clearing land and establishing improved pastures in Central Florida. Ag. Exp. Sta. Bull. #600, Univ. Fla. Reynolds, H. G. and P. E. Packer. 1963. Effects of trampling on soil and vegetation. USDA Misc. Pub. #940:116-122. Richards, B. N. and D. I. Bevage. 1972. Principles and practice of foliar analysis as a basis for crop-logging, in pine plantations. II. determin- ations of critical phosphorus levels. Plant and Soil, 37:159-169. Richards, I. R. and K. M. Wolton. 1976. The spatial distribution of excreta under intensive cattle grazing. J. Br. Grassl. Soc. 31:89-92. Richardson, C. J. and J. A. Lund. 1975. Effects of clear-cutting on nutrient losses in aspen forest on three soil types in Michigan. In mineral cycling in southeastern ecosystems, (ed . ) F. G. Howell, S. B. Gentrv, and M. II. Smith, Tech. Inf. Ser. Cen. U. S. Energy Res. and Devel. Ad. p. 673-686. Rightmire, C. T. and B. B. Handshaw. 1973. Relationship between the carbon isotope composition of soil C02 and dissolved carbonate species in ground water. Water Resources Research, 9(4) :958-967. Ritchey, G. E. and W. W. Henley. 1936. Pasture value of different grasses alone and in mixture. Ag. Exp. Sta. Bull. #289. Univ. Fla. Rodin, L. W. and N. I. Basilevich. 1965. Production and mineral cycling in terrestrial vegetation. Scripta Technica Ltd. trans, (ed.) G. E. Fogg, Oliver and Boyd Pub., London, Engl. Rodulfo, S. and W. G. Blue. 1970. The availability to forage plants of accumulated phosphorus in Leon fine sand. Soil and Crop Sci. Soc. Fla Proc. 30:167-174. Rogler, G. A. and R. J. Lorenz. 1973. Fertilization for range improvement. In range research and range problems, (ed . ) M. J. Wright. CSSA Spec. Pub. #3. Crop Sci. Soc. Am. Madison, Wise. p. 81-86. 344 Rosenberg, N. J. 1974. Microclimate: the biological environment. S. Wiley and Sons, N.Y. Ross, H. H. 1965. A textbook of entomology, 3rd ed., J. Wiley and Sons, N.Y. Rummeil, R. S. 1956. Range management research in South Florida: A project analysis. S.E. For. Exp. Sta. Pap. #71, USDA, Asheville, N.C. Rummeil, R. S. 1960. An abridged history of southern range research., Iowa St. J. of Sci. 34(4):749-760. Sabbe, W. E. and H. L. Breland . 1974. Procedures used by state soil testing laboratories in the southern region of the United States. Southern Coop Ser. Bull. #190. Salih, M. S. and D. F. Burzlaff. 1977. Improvement of crude protein content and in vitro digestibility of range grasses with nitrogen fertilization. Paper presented at 30th Ann. Mtg . of Soc. for Range Mgmt., Portland, Ore., Abstract of Papers, p. 27. Sanchez, P. A. 1976. Properties and management of soils in the tropics. J. Wiley and Sons, N.Y. Sanders, H. L. 1968. Marine benthic diversity: a comparative study. Am. Natur. 102(925) :243-282. Satchell, J. E. 1974. Introduction: litter-interface of animate/inanimate matter. In biology of plant litter decomposition, (ed.) C. H. Dickinson and G. J. F. Pugh, Vol. 1, p. XIII-XLIV. Schneider, B. H. and W. P. Flatt. 1975. The evaluation of feeds through digestibility experiments. Univ. Ga. Press, Athens, Ga. Schopmeyer, C. S. 1974. Seeds of woody plants in the United States. Ag. Handbook #450, For. Ser. USDA. Scott, E. M., E. L. Verney and P. D. Morissey. 1950. Self selection of diet. XI. appetites for calcium, magnesium and potassium. J. Nutr., 41:187. Sears, P. D. 1956. The effect of the grazing animal on pasture. Proc . 7th Int. Grassl. Congr. Palmerston, North N.Z. p. 92-103. Shade, R. G, and M. C. Wilson. 1967. Leaf-vein spacing as a factor affecting larval feeding behavior of the cereal leaf beetle Oulema melanopa. Ann. Eutomal. Soc. Amer. 60:494-496. """ ""' Shannon, C. E. and W. Weaver. 1949. The mathematical theory of communication, Univ. 111. Press. Sharp, D. 1970. Insects, Vol. 2., Dover Pub., N.Y. 345 Shaw, N. H., L. ' tMannette, R. M. Jones and R. J. Jones. 1976. Pasture measurements. In tropical pasture research: principles and methods, (ed.) N. H. Shaw and W. W. Bryan, Bull. #51, Commonwealth Agric. Bureaux Rurley, Engl. p. 235-250. Short, H. L. 1970. Determining forage quality. In range and wildlife habitat evaluation: a research symposium. For. Ser. Misc. Pub. #1147 USDA p. 79-84. Shore, H. L. 1971. Forage digestibility and diet of deer on southern upland range. J. Wildl. Mgmt. 35 (4) :698-706. Short, H. L. 1972. Ecological framework for deer management. J. For 70(4): 200-203. Sinai, G. and 0. Zaslavsky. 1977. Factors affecting water distribution after uniform irrigation. Paper presented 1977 winter meeting Am. Soc. Ag. Eng Chi. 111. Smith, B. J. 1976. A new application in the pitfall trapping of insects. Trans. Ky. Acad. Sci. 37 (3-4) : 94-97 . Smith, B. N. and S. Epstein. 1971. Two categories of C13/C12 ratios for higher plants. Plant Physical. 47:380-384. Smith, C. M. 1977. Interpreting inorganic nitrogen soil test: sample depth, soil water, climate, and crops. In soil testing: correlating and inter- preting the analytical results, (ed.) M. Stelly. ASA Spec. Pub. #29., Am. Soc. Agron. Madison, Wise. p. 85-98. Smith, F. B., R. G. Leighty, R. E. Caldwell, V. W. Carlisle, L. G. Thompson, Jr., and T. C. Mathews. 1967. Principle soil areas of Florida. Fla. Ag. Exp. Sta. Bull. #717. Smith, P. F. 1962. Mineral analysis of plant tissues. Ann. Rev. Plant Physio 13:81-107. Snapp, R. R. and A. L. Neumann. 1960. Beef cattle. 5th ed . John Wiley and Sons, N.Y. Soil Conservation Service Staff. 1975. Potential natural vegetative community interpretation for rangelands. Tech. guid., Section HE. - Descriptions for native grazing lands. SCS Fla. Field Of c . Tech. Guide. SCS USDA, Wash. , D.C. Soil Conservation Service Staff. 1976. National range handbook, USDA, Soil Conservation Service, Wash., D.C. Sollins, P. and G. Drewry. 1970. Electrical conductivity and flow rate of water, through the forest canopy. In a tropical rain forest (ed . ) H. T. Gdum and R. F. Pigeon, USAEC. p. H137-H154. 346 Southern and Southeastern Forest Experimental Stations Symposium. 1959. Techniques and methods of measuring understory vegetation. Proc. of Symposium, Tifton, Ga . 1958. Spedding, C. R. W. 1971. Grassland ecology. Oxford Clarendon Press. Steel, R. G. D. and J. 11. Torrie. 1960. Principles and procedures of statistics. McGraw-Hill Book Co., N.Y. Stevenson, I. L. 1965. Some observations on the nicrobial activity in remoistened air dried soil. Plant Soil. 8(2) :170-182. Stockbridge, H. E., W. E. French and J. E. Ennis. 1901. Salt sickness. Fla. Ag. Exp. Sta. Ann. Rpt. p. 43-54. Stcddart, L. A. and A. D. Smith. 1955. Range management, 2nd ed., McGraw- Hill Book Co. , N.Y. Teeri, J. A. and L. G. Stowe. 1976. Climatic patterns and distribution of C^ grasses in North America. Oecologia, 23:1-12. Terman, G. L. 1977. Variability in nutrient uptake based on analysis of plant samples from individual pots versus replicate composites. Commun . in Soil Sci. and plant analysis, 8 (4) :313-326. Terman, G. L. and L. A. Nelson. 1976. Letter to the editor. Agron. J 68:148-150. Thomas, G. W. 1970. Soil and climatic factors which affect nutrient mobility. In nutrient mobility in soils: Accumulation and losses, (ed.) 0. P. Engelstad. SSSA Spec. Pub. #4. Soil Sci. Soc. Am. Madison, Wise. p. 1-20. Thorsland, 0. A. 1966. Nutritional analysis of selected deer foods in S. Carolina. Proc. 20th Ann. Conf. S.E. Assn. Game and Fish Cornm. , Ashe- ville, N.C. p. 84-104. measure- Tilley, J. M. A., R. E. Deriaz and R. A. Terry. 1960. The in vitro ment of herbage digestibility and assessment of nutritive value. Proc. 8th Int. Grassl. Congr., p. 533-637. Tilley, J. M. A. and R. A. Terry. 1963. A two stage technique for the in vitro digestion of forage crops. J. Brit. Grassl. Soc. 18:104-111. Timmons, D. R., R. F. Holt and J. J. Latterell. 1970. Leaching of crop residues as a source of nutrients in surfacr runoff water. V/ater Resource5 Res. 6(5) :1367-1375. Urnsss, P. J., A. D. Smith and K. K. Watkins. 1977. Comparison of in vivo and in vitro dry matter digestibility of mule deer forages. J. Ranpf> Memt 30(2) :119-121. 347 USDA Staff. 1975. The nations renewable resources. For. Ser. USDA Wash., D.C. U. S. Forest Service Staff. 1945. Better management on Southern Coastal forest ranges. Bull. #AIS-17. USDA, Wash., D.C. Van Dyne, G. M. 1962. Micro-methods for nutritive evaluation of range forages. J. Range Mgmt. 15:303-314. Van Soest, P. J. 1965. Use of detergent in analysis of fibrous feeds. III. Study of effects of heating and drying on yield of fiber and lignin in forest. J. Assn. Ofc. Agr. Chem. 48:785-790. Van Soest, P. J. 1967. Development of a comprehensive system of feed analysis its applications. J. Anim. Sci. 26:119-128. Van Soest, P. J. and L. H. P. Jones. 1968. Effect of silica in forages upon digestibility. J. Dairy Sci. 51:1644. Varley, G. C. 1967. The effects of grazing by animals on plant productivity. In secondary productivity of terrestrial ecosystems: Principles and methods. (ed.) K. Petrusewicz. Vol. II. Polish Acad. Sci., Warsaw, p. 773-777. Vitousek, D. M. and W. A. Reiners . 1975. Ecosystem succession and nutrient retention: A hypothesis. Bio. Sci. 25(6) :376-381 . Voigt, G. K. 1958. Plant soil factors in chemical soil analysis. For. Soil Conf. 1:31-41. Voigt, P. W. , L. I. Croy and F. P. Horn. 1978. Digestibility and palatability of eragrostis selections. Paper presented 31st Ann. Mtg. Soc. Range Mgmt., San Antonio, Tex. Abstract, p. 34. Waggoner, K. A., R. Albaugh and G. T. Hart. 1960. Beef cattle production, McicMillan Co. , N.Y. Walker, T. J. 1973. Gryllus ovisopis N. SP. : A tactiturn cricket with a life cycle suggesting allochronic speciation. Fla. Entomol. 57(l):13-22. Wells, B. W. 1942. Ecological problems in the Southeastern United States Coastal Plain. Bot. Rev. 8 (8) :533-561. Wells, C. G., D. M. Crutchfield, N. M. Berenyi and C. B. Davey. 1973. Soil and foliar guidelines for phosphorus fertilization of loblolly pines. S.E. For. Exp. Sta. USDA For. Ser. Res. Paper, SE-110. White, E. H. and W. L. Pritchett. 1970. Water table control and fertilization for pine production in the flatwoods. IFAS Tech. Bull. #743. Univ. Fla. White, L. D. 1973. Native forage resources and their potential. In range resource of the Southeastern United States, (ed . ) R. S. Campbell and W. Keller., ASA Spec. Pub. #21, Madison, Wise. p. 2-16. 348 White, L. D. 1975a. Ecosystem analysis of Paynes Prairie. Sch. For. Res and Consv., Res. Rep. #24., Univ. Fla. White, L. D. 1975b. Intensive forest management practices on understory plant production. In forest soils of the Southeastern Coastal Plains Sch. of For. Res. and Consv., p. 138-15?. White, L. D. 1975c. Increasing cattle production on Florida range resources 24th Ann. Beef Short Course, IFAS, Univ. Fla., Gainesville. White L. D. 1977. Importance of Florida's forested rangelands. Ranpe report #77-1, IFAS, Univ. Fla. ^ White, L. D. and C. Carter. 1977. Beef resource management trends for beef cow production in Florida. In mutual opportunities for forest, range, and wildlife management. IFAS and Soc. Am. For. Fla. Sec. Res! Rep. H,* p . 5— 9 . Whittaker, R. G. 1970. Communities and ecosystems. Current concepts in Biology Series., MacMillian Co., N.Y. Whittaker, R. G. 1975. The design and stability of plant communities. In unifying concepts in ecology, (ed.) W. H. Van Dobben, and R. H. Lowe- McConnell. Dr. W. Junk B. V. Pub. Hague, Wageningen. p. 169-181. Whyte, R. 0. 1962. The myth of tropical grasslands. Trop. Agr . Trinidad 39:1-11. Wilkinson, S. R. and R. W. Lowrey. 1973. Cycling of mineral nutrients in pasture ecosystems. In chemistry and biochemistry of herbage, Vol. II, (ed.) G. W. Butler and R. Basley., Acad. Press, N.Y. p. 247-315. Williams, L. 1972. Florida deer facts. Fla. Game and Fresh Water Fish Coram. Tallahassee, Fla. Williams, S. T. and T. R. Gray. 1974. Decomposition of litter on the soil surface. In biology of plant litter decomposition Vol. 2, (ed.) C. H. Dickinson and G. J. F. Pugh, Acad. Press, N.Y. p. 611-632. Williams, W. K. 1951. Protect hardwood stands from grazing, USDA For Ser Leaflet #86. Wilson, J. R. and C. W. Ford. 1971. Temperature influences in the growth, digestibility, and carbohydrate composition of two tropical grasses, I^icum maximium var . trichoglume and Selaria sphacelata, and two culti- vares of the temperate grass Lolium perenne. Aust. J. Ag. Res. 22:563-571. Wilson, J. R. and K. P. Haydock. 1971. The comparative response of tropical and temperate grasses to varying levels on nitrogen and phosphorus nutrition Aust. J. Ag. Res. 22:573-587. 349 Wolcott, G. N. 1937. An animal census in Northern New York, Ecol Mono 7(l):l-96. Woodwell, G. M. and R. H. Whittaker. 1968. Primary production in terrestrial ecosystems. Am. Zool. , 8:19-30. Yarlett, L. L. 1961. Range management in Florida. Soil Consv. 25 (6) :130-133 . , Yarlett, L. L. 1963. Some important and associated native grasses on Central and South Florida ranges. J. Range Mgmt. 16(l):25-27. Yarlett, L. L. 1955a. The range resource and management opportunities. Paper presented 14th Ann. Mtg. S. Sec. Am. Soc. Range Mgmt., Kissimmee, Fla. ^ Yarlett, L. L. 1965b. Important native grasses for range conservation in Florida. USDA Soil Consv. Ser. p. 163. *" Yarlett, L. L. 1977. Management of rangelands for maximum production. Int. 26th Proc. Beef Cattle Short Course, IFAS , Univ. Fla. p. 128-133. Yarlett, L. L. and J. R. Moore. 1963. Management of gulf coast salt marshes J. Soil and Water Consv. 18(4) . -166-167. Young, S. A. 1977. Quality of cattle diet under short duration grazing in north Florida flatvoods. MS thesis., Sch. of For. and Res. Consv Univ Fla. Yuan, T. L. 1966. Characteristics of surface and spodic horizons of some sporiosols., Soil and Crop Sci. Soc. Fla. Proc. 26:163-174. Yuan, T. L. and H. L. Breland. 1969. Correlation of Al and Fe as extracted by different reagents with phosphate retention in several soil groups. Crop Sci. Soc. Fla. Proc. 29:78-86. BIOGRAPHICAL SKETCH Burton J. Smith was born in Douglas Wyoming in 1932. After serving in the U.S. Navy he received a bachelors degree in mathematics from San Diego State in 1959. He was a research engineer in the aero-space industry, until he left and pruchased a dairy and diversified farm in Yost, Utah. In 1966 he moved to Eureka Nevada, where he was the owner-operator of a 400 head desert cattle ranch. In 1973 he moved to Lee Kentucky, where he raised corn, soybeans, cattle and hogs. He entered Western Kentucky in the summer of 1974 and received a MS degree in biology a year later. In January 1976, he enrolled at the University of Florida and began work on his PhD. He is married to a very understanding woman and has three children. 350 I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor" of Philosophy. (/ . ■--, ,, Gerald 0. Mott, Chairman Professor of Agronomy I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. John K. Loosli Visiting Professor of Animal Science I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation^and is fully adequate, in scope and quality, as a dissertation for the degree' ""of^Do'ctor of Philosophy. %J -'X/'/.V. Dennis H. Hunter Assistant Professor of Range. Ecosystem Management I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Darell E. McCloud Professor of Agronomy I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philsosphy. Howard T. Odum Graduate Research Professor of Environ- mental Sciences This dissertation was submitted to the Graduate Faculty of the College of Agriculture and to the Graduate Council, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. August, 1978 l r 4 Dean// College of Agriculture Dean, Graduate School UNIVERSITY OF FLORIDA 3 1262 08553 2785