.‘ \ gut-win! ---..-~.. -.<----«.-‘-..-._- "‘7‘. ‘V,._._,..-.. ._»-r‘/rrNo-.Q.r1'V~'Inllh‘V‘Hfi“‘li‘o. PRIMARY PRODUCTION AND METABOLISM OF AN I . , 5g AMMOPHIIA BREVILIGULATA (AMERICAN BEACHGRASS) , ,_ , jj; COMMUNITY ’ Dissertation for the Degree of Ph. D. - ‘ 9 MICHIGAN STATE UNIVERSITY . ERIC NORTON HANSEN , f 1976 ‘ t i. IIIIIIII IIIIIIIIII IIIIIII V 31293 00666 ”A; 12111.. - I :H'v 3-1.1: any c' 39:“: - ‘o- . I ‘ 1 “NF." «1" 1.9.1.134,” 3 I V! as I; l '4' “ R I ... :l ’ I . I . — I . ‘- v 93' 3?.1‘_-3_?._ -f‘,:’ufr~' - .- I b J 9% "4516;; S, . APR 0 7 1999 pl, ABSTRACT PRIMARY PRODUCTION AND METABOLISM OF AN AMMOPHILA BREVILIGULATA (AMERICAN BEACHGRASS) COMMUNITY By Eric N. Hansen Estimates of the net primary production of an American beachgrass community located on a foredune on the east shore of Lake Michigan were determined by harvesting. Gross photosynthetic assimilation, plant respiration, soil respiration and community respiration were estimated from measurements of rates of 002 exchange in light, dark and soil chambers. 002 flux was measured with an infrared gas analysis system. Laboratory measurements of beachgrass photosynthesis were used in the interpretation of annual metabolic variation observed in the field. Interactions of rodents and insects with the beachgrass were also studied. Net primary production of the community was 347 g/m2. This figure was a total of aboveground biomass increase (l83 g/mz), belowground biomass increase (91 g/mz), and aboveground biomass mortality (73 g/mz). The maximum production possible was estimated to be 555 g/m2 and minimum production possible was estimated to be 218 g/m2. At the time of peak aboveground living biomass, the ratio of aboveground living to total belowground biomass was 4.l. The mean caloric values for above- ground living, dead, and total belowground biomass were 4.70 1 0.02, 4.80 i 0.02 and 4.62 i 0.03 Kcal/g dry weight respectively. The Eric Norton Hansen efficiency of beachgrass during the growing season was 0.23% and 0.l4% annually. Maximum rates of gross photosynthetic assimilation occurred in late June and early July, whereas maximum rates of aboveground plant respira- tion occurred in August. Mean daily rates of gross photosynthetic assimilation, aboveground plant respiration and soil respiration during the growing season were 2.64, 1.27 and l.4l g C/mZ/day respectively. Staining aggregations of sand found on the study site revealed a possible vesicular-arbuscular mycorrhizal association of beachgrass with a fungus closely resembling Glomus macrocarpus. The large proportion of juvenile males and the absence of burrows and aboveground nests and runways indicated that the two prevalent rodent species trapped, Peromyscus maniculatus bairdi and Microtus pennsylvanicus were not residents on the study site. The absence of beachgrass in the digestive tracts of these rodents revealed that beach- grass does not contribute to their diets. Only Phyllophaga rugosa larvae appeared to be utilizing beachgrass as their main food source. PRIMARY PRODUCTION AND METABOLISM OF AN AMMOPHILA BREVILIGULATA (AMERICAN BEACHGRASS) COMMUNITY By Eric Norton Hansen A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY _ Department of Botany and Plant Pathology 1976 ACKNOWLEDGMENTS I sincerely thank my major professor, Dr. Peter G. Murphy, for his guidance and support during this study and assistance in the preparation of this manuscript. I also wish to thank the members of my guidance committee, Dr. Rollin H. Baker, Dr. Gerhardt Schneider, and Dr. Stephen N. Stephenson for their suggestions and critical evaluation of the manuscript. Appreciation is extended to Dr. Duane Ullrey for the use of his bomb calorimeter and to Dr. Klause Raschke for the use of his equipment and to Tom Sharkey for technical assistance in the measurement of single leaf metabolism. Special thanks go to Dr. Jerry Hall and Dr. John Fitch for helpful suggestions concerning the rodent portion of this study and to Dr. Roland Fischer for assistance in the identification of the invertebrates. I also wish to thank Mr. John Uhl, Mr. John Luther and Mr. Frank Denison for allowing me to conduct this study on their land. I wish to express my thanks to Dr. Glenn Kroh, Dr. Ken McLeod and Dr. Gene Safir for their help and informative discussions during this study. Special appreciation is given to Dr. John Barko for helpful discussions concerning portions of this study. I would also like to thank the Department of Botany and Plant Pathology of Michigan State University for providing general equipment needs and financial support through graduate assistantships. Finally, I would like to thank my wife, Linda, without whose help and encouragement this study would not have been completed. ii TABLE OF CONTENTS Page INTRODUCTION .............. ....................................... 1 Primary Production in Terrestrial Ecosystems ................ l The Ecology of Ammophila breviligglata ...................... 2 DESCRIPTION OF STUDY SITE ... ............. . ....................... 8 METHODS AND MATERIALS ............................................ 14 Production and Community Metabolism ......................... l4 Biomass Relations ....................................... 14 Community Metabolism .................................... 19 Factors Controlling Photosynthesis ..................... . 27 Estimation of Animal Activity ............................... 29 Vertebrates ............................................. 29 Invertebrates ........................................... 30 RESULTS ........... . ............................ . ................. 31 Productivity and Community Metabolism ....................... 31 Biomass Relations and Community Structure ............... 31 Aboveground ......................................... 3l Belowground ......................................... 34 Decomposition and Mortality ......................... 39 Net Primary Production .............................. 42 Root/Shoot Ratio .................................... 42 Annua1 Patterns of Carbon and Caloric Content ....... 45 Ecological Efficiency ............................... 45 Community Metabolism .................................... 48 Annual Patterns .... ..... ....... .................... . 48 Daily Patterns ...................................... 53 Photosynthetic Assimilation Under Different Conditions ...... 58 Response to Varying Light Intensities . ........... .. ..... 58 Response to Varying C02 Concentrations .................. 61 Vertebrates ............................. . ................... 64 Invertebrates ............................................... 66 Mycorrhizal Fungi ........................................... 67 iii Page DISCUSSION ....................................................... 69 Beachgrass Primary Production ............................... 69 Estimates of Animal Activity ....... ' ......................... 77 Vertebrates ............................................. 77 Invertebrates ........................................... 79 LITERATURE CITED ................................................. 81 APPENDIX ......................................................... 90 iv LIST OF TABLES Table 1. Al. A2. A3 Summary of 1974-1975 climatological data, South Haven and Grand Rapids, Michigan .................................. . Mean caloric values for aboveground living, dead and total belowground biomass during the entire 1974 sampling period .. . Mean daily rates of A, breviligulata community metabolism ... . Daily interrelationships between photosynthesis of A, breviligulata and solar energy. indices of efficiency and daylength ................................................... . Analysis of factors correlated with gross photosynthesis .... . The number of species, sex and age groups of rodents trapped within the study site ..... . ......................... A. breviligulata living biomass harvested during 27 Ju1y T973 from 25 plots on the study site ........................ Living leaf surface area and leaf dry weights used for development of leaf surface area linear prediction equation . . A, breviligulata daily mean harvested dry weights ........... Page 13 48 52 54 55 65 90 91 92 LIST OF FIGURES Figure Page 1. Ammophila brevili ulata on a Lake Michigan sand dune near Saugatuck, M chigan ................................... 4 2. Diagram of the study site .................................. 10 3. Portable metabolism chambers used in gas exchange study .... 21 4. Infrared gas analysis determination of carbon dioxide concentrations .. ........................................... 25 5. Seasonal changes in aboveground living biomass, leaf surface area and the number of living stems of A, breviligulata, 1973-1975 .................................................. 33 6. Seasonal changes in aboveground dead biomass of A, breviligulata, 1973-1975 ................................... 36 7. Season changes in living aboveground and total belowground (0-40 cm) biomass of A, breviligulata, l974-l975 ........... 38 8. Percent mean living leaf mortality and litter disappearance. 41 9. Ratio of belowground biomass to aboveground biomass ........ 44 10. Seasonal changes in the caloric content of A, breviligulata, 1974 .... ......... ... ................. . ..................... 47 ll. A, breviligulata community metabolism ...................... 50 12. Hourly rates of gross photosynthetic assimilation, plant respiration and soil respiration ... ..... . .................. 57 13. Net photosynthetic rates of single leaves of A, breviligulata in relation to light intensity ............... 60 14. Net photosynthetic rates of single leaves of A. breviquulata at a constant light intensity of 252.5 N/mz and different C02 concentrations .... ..... . ................. 63 vi INTRODUCTION Primary_Production Ig_Terrestrial Ecosystems Primary production refers to the new organic material created by photosynthesis in plants and often is expressed as weight or energy (Nestlake, 1969). A not-so-obvious component of primary production is the energy required to maintain the living biomass (respiration). The total primary production of a plant (gross primary production) consists of the accumulated organic material (net primary production) plus that used in respiration. For a more complete estimate of gross primary production, plant material which disappears during the growing season should also be included. Net primary production in terrestrial communities has been measured primarily by harvesting plant biomass (Woodwell and Whittaker, 1968). Improved techniques have been developed for the evaluation and compar- ison of different methods of assessing grassland aerial net primary production and biomass dynamics (Kelley gt_g1,, 1974; Singh gt_gl,, 1975). However, reliable estimates of grassland root production for inclusion with aerial production for complete plant net primary pro- duction are difficult to obtain and frequently are not included in productivity estimates (Kucera g§_gl,, 1967). The separation of roots and rhizomes from the soil is so laborious and time consuming that often it is not practical to get enough replications to insure statistical reliability (Redmann, 1975). Another problem in estimating root production is that the period for root growth corresponds with l 2 that of greatest root decomposition rates so that net observable changes may be negligible. The determination of gross production is done by gas exchange (C02) measurements (Bordeau and Hoodwell, 1965) taken from enclosed portions of a plant or entire plants (Kanemasu and Hiebsch, 1975). By enclosing entire portions of an ecosystem (0dum and Pigeon, 1970) or by indirect estimates of C02 flux (Lemon gt_gl,, 1970) estimates of eco- system metabolism can be obtained. Estimates of gas exchange of grass- land plants have been mainly limited to crop plants (Milthorpe and Moorby, 1974). Ammophila breviligulata Fern.(American beachgrass) (Figure 1) grows as a pure community on sand and has only one seasonal peak. The rate of organic matter decomposition in areas where beachgrass grows is slow (Olson, 1958a). These factors alleviate many of the problems of estimating net primary production encountered in more diverse grassland communities. The Ecology gf_Ammophila Breviligulata A taxonomic description of American beachgrass is given by Fernald (1920) and Hitchcock (1950). Morphologically, beachgrass is adapted to the foredune area where sand accumulation occurs. It has tough but flexible pointed leaves which are 0.6-0.9 m long and are able to survive the force of the wind with little harm. Its vertical growth form, unlike many other early successional dune species (Salisbury, 1938), enables it to withstand up to l m of burial in a year, provided that more than half of the leaf is not buried at a time (Laing, 1954; Ranwell, 1958; Noodhouse and Hanes, 1966). Figure l. Ammophila breviligulata on a Lake Michigan sand dune near Saugatuck, Michigan. 5 The rhizomes are able to penetrate sand and withstand the sand's abrasive action since the apices are enclosed by a hard, spine-like sheathing leaf (Salisbury, 1952). The rhizomes possess potentially un- limited horizontal growth (Cowles, 1899; Ranwell, 1972). Roots do not penetrate significantly deeper than about 1 m (Ranwell, 1972), although individual roots or vertical rhizomes can be much longer (Laing, 1954). American beachgrass is found on sand dunes along the Atlantic coast from Newfoundland to North Carolina and on the shores of all the Great Lakes (Hitchcock, 1950; Voss, 1972). The Great Lakes beachgrass is thought to have spread from the Atlantic coast since the retreat of the Wisconsin ice sheet (McLaughlin, 1933; Peattie, 1922). The Great Lakes and Atlantic coast beachgrass populations are ecotypes of the same species (Seneca and Cooper, 1971). Beachgrass communities are estab- lished, by seeds or through regeneration from rhizome fragments carried by water from eroded dunes, on dunes with active sand deposition. The flowering heads begin to emerge in late July and the seeds are usually mature by late August to mid-September (Laing, 1958L. However, in a study on the Baltic coast, Kubien (1970) found that only 4% of the seeds that could develop did. This was due to either abnormal development in the anthers and ovules or a breakdown in the later stages of endosperm development. 0f the florets from the Indiana dunes, studied by Laing (1958) 95% had little or no seed development, 4% had shrunken seeds (possibly an endosperm deficiency) and 1% had developed seeds. Seedling establishment is sporadic because of death caused primarily by heat, dessication and sand burial (Gemmell gt_g1,, 1953; Jagschitz and Bell, (1966; Laing, 1958; Waterman, 1919; Noodhouse and Hanes, 1966). Estab- lishment of seedlings is generally confined to leeward slopes of eroding 6 dunes, moist hollows and occasionally beaches (Laing, 1958). Rhizome fragments from eroded foredunes are transported laterally along the shore by currents and are eventually cast up on shore where new plants may be established. Laing (1958) feels that this may be more important than seed dispersal in the establishment of new communities. An established beachgrass community increases its area primarily vegetatively by horizontal rhizomes which, in Michigan, are only pro- duced under long day conditions (Seneca and Cooper, 1971). To remain vigorous, beachgrass needs continual sand accumulation. Once the dune surface becomes stabilized there is a decline in vigor. Because of its sand binding ability, beachgrass (American and European) has been introduced into a number of temperate areas for sand stabilization (Anon., 1957, Barbour gt_gl,, 1975; Breckton and Barbour, 1974; Keet, 1936; Michell, 1966; and others). Most of the previous literature on beachgrass is concerned with planting and fertilizing it fbr this use (Brown and Hafenrichter, 1948a,b,c; Coaldrake gt 11,, 1973; Coaldrake and Beattie, 1974; Lehotsky, 1939; Jagschitz and Bell, 1966; Stoesz and Brown, 1957; Hestgate, 1904; Hoodhouse and Hanes, 1966; and others). Although many grasses have been tried, none has proven as effective in binding sand as beachgrass. Both the above and below ground portions of beachgrass contribute to its sand binding ability. Its aerial portions reduce the wind speed below that needed to move sand (about 4 m/sec or 9 mph) (Olson, 1958b) and its extensive roots and rhizomes provide some protection to the slopes of eroded windward foredunes (Patton, 1934). However there have been no studies of the .annual productivity of beachgrass. There are a variety of insects, spiders, etc. associated with the 7 succession of higher plants progressing inland from the front dune (Shelford, 1913). The succession of tiger beetles (Cicindelidae) is an example of this (Shelfbrd, 1907). Most studies of invertebrates in association with beachgrass fbredune communities have been limited to classification of individuals deposited on the beach by wave action during and after storms, or life cycles of the insects living on the shore (Herms, 1907; Needam, 1900, 1904; Shelford, 1909, Snow, 1902). Whether beachgrass is utilized by any of these insects has not been determined. Few animals are found on or associated with the shore and early successional plants. Toads, skunks and birds may inspect material that has washed ashore (Shelford, 1913) and Laing (1958) found some rodent consumption of young flowering heads. The amount of beachgrass that makes up the diet of rodents in the study area is not known. The general objectives of this study were to estimate the pro- ductivity of beachgrass, and to determine the relationship of animals to beachgrass. The specific study objectives were: 1. to follow the seasonal change in above and belowground biomass; 2. to estimate net and gross primary production and community respiration; 3. to determine the extent of seasonal changes in the photosynthesis of beachgrass leaves; and 4. to evaluate the extent of interactions of rodents and insects with beachgrass. DESCRIPTION OF THE STUDY SITE The study site was located on the foredune north of the Kalamazoo River near Saugatuck, Allegan County, Michigan (Figure 2). Prior to the selection of this study site in 1973 the east shore of Lake Michigan from Charlevoix to Michigan City was surveyed by boat and through the use of aerial photographs for areas with sufficient beachgrass for a study of this type. Besides the study site, stands of beachgrass were also found on Warren dunes, the dunes around Silver Lake, Sleeping Bear dunes, and South Manitou Island dunes. These four areas were either 'State or Federal parks. Suitable stands of beachgrass did not exist along the rest of the coast due to erosion or housing developments. From talks with people who have spent their lives near the shore of Lake Michigan, from descriptions by earlier writers (Cowles, 1899; Waterman, 1922) and from the way the beachgrass was growing in the parks, I feel the area chosen was similar to beachgrass communities which existed along the Lake Michigan shore prior to the 1940's. The study site foredune was approximately 370 m long (N-S) and 50 to 60 m wide (E-W) but only 300 m by 40 m was used for the study. Be- cause of the high lake level during recent years, a cliff 2-3 m high formed the front edge of the foredune. The foredune was bordered to the north, south and partially to the west by bare sand caused by prior dune buggy use; the bare sand to the south had started to form a blowout. All traveling from one sampling area to another was confined Figure 2. Diagram of the study site. -——--——-encloses study site 4“ Aé Ammophila brevilifllata 0 young Populus deltoides young Salix glaucophylloides mature ngulus deltoides early successional forest early successional vegetation (denser vegetation than in following category) early successional vegetation colonizing sand beach (sand) dune pond dune buggy road breakwall 10 _. v .5 . s, J Km._,.m”.;._.,w. . 4.. fl.1#:_.w;~ fl,“ ....muwa, ._ .. .. .; .... Lm> ..m OONmEm Pav— I‘III ...... .. _ ”...—f. ... 3... ,.,. _. — .— .. IA fi::. ’ m _\_. .. ...~ . .. ‘ —. ... ..\~ _.‘ . _ ... ...... a..._ \__._,. ~ ... k _. —_. ._ ~ . n .... .c \ a _... . : \ t 9 ... -.._. x: \\ m / .... _ ._ a c . s \.~.\_\\\s . m a; . \\\ \ ~_\_ f \ _‘ v .. ; . .. .. 1.» . \z. U low 1" \ 0 ~_ re. F..._ . , .. .. C, .. . C .. .—.q\ .. ‘ n N m a ’. ... ‘ u. ..nnUu 11 to paths outside the study site or to a path which penetrated the center of the study site. The site contained predominately Ammophila breviligulata plus four small areas of low woody vegetation, two of Salix glaucophylloides (blue willow) and two of Populus deltoides (cottonwood). The areas of shrubs consisted of only young vegetation and averaged l-l.5 m in height and 2-3 m in diameter. Just to the west of the study site was a row of Populus deltoides 8-10 m tall. The area behind the study site graded from pure Ammophila breviligulata to mixed vegetation dominated by either Populus deltoides, Prunus pgmila (sand cherry) or Calamovilfa longifolia (reedgrass). From aerial photographs and discussions with Mr. John Uhl and Mr. John Luther (owners of a portion of the land where the study was con- ducted), it was learned that there was little vegetation present on the study site prior to 1956. During Lake Michigan's low water level of the late 1950's and early 1960's sand was deposited on the area and the existing beachgrass developed into the community now present. During the current high water level the foredune has lost approximately 20 m from its lakeward side. The vegetation on the foredune was burned in 1972. However Jaschitz and Bell (1966) in Rhode Island found several sections of their beachgrass study area had been burned over with little effect on subsequent growth. Based on their findings, the 1972 burn was not considered a deterrent in the selection of this area as a study site. Weather data collected by the United States Weather Bureau located at the South Haven Agricultural Experimental Station were used to .characterize the study area. The station monitors temperature and precipitation (United States Weather Bureau, 1973-1975a) and is located 12 28 Km SSW of the study site and within several meters of the lake shore. The cloud cover and wind speed were measured at the United States Weather Bureau in Grand Rapids located 61 Km ENE of the study site (United States Weather Bureau, 1973-1975b). Monthly climatological data from both of these stations are summarized in Table l. The mean annual temperatures during this study differed by no more than 0.3° C from the 78 year mean annual temperature of 9.8° C. The climate during the 1973-1975 period was fairly uniform except for pre- cipitation. Precipitation during the growing season in 1973 was 7.7 cm above the 29 year norm. In 1974 it was 14.8 cm below the norm. Pre- cipitation during the 1975 growing season was 21.7 cm above the norm. The wind during the growing season was predominately from the SW. Table 1. Grand Rapids, Michigan 13 Summary of 1973-1975 climatological data, South Haven and 1973 1974 1975 Mean annual temperature 10.4°C 9.2°C 10.8°C Mean growing season(]) temperature 17.0°C 19.9°C l7.0°C Mean of the coldest month Feb -2.1°C Feb -3.4°C Feb -2.4°C Mean of the warmest month Aug 21.8°C July 21.8°C Ju1y 21.1°C Total annual precipitation 89.6 cm 78.0 cm 101.3 cm Total precipitation during growing season 55.2 cm 36.9 cm 67.4 cm Total evaporation during growing season 81.5 cm 81.3 cm 74.6 cm Growing season evaporation as % of total annual precipitation 90% 104% 74% Growing season evaporation as % of growing season precipitation 148% 220% 111% Mean % possible sunshine during growing season 52% 58% 54% Mean annual wind speed 4.5 m/sec 4.2 m/sec 4.0 m/sec (1) April - September METHODS AND MATERIALS Productivity And Communitnyetabolism Biomass Relations. Seasonal changes in beachgrass biomass and pro- ductivity were estimated by periodic harvesting. 0f the two principal methods of estimating productivity by harvesting, the less complex method is to measure the total weight of standing vegetation at the end of the growing season (Hadley and Kieckhefer, 1963; Kucera gt_gl,, 1967). However, this method underestimates total biomass production due to turnover during the growing season. A second method is to harvest the standing vegetation more frequently and equate the season's peak value with above ground net production as done by Bray gt gl,, (1959) and 9 ‘ , “ as .. ‘- ‘ l“"“' ‘1- : Malone (1968). This method also underestimates total biomass produc- E tion unless estimates of living material which senesced and dead II material which decomposed were included (Wiegert and Evans, 1964). I: Kelley gt 21, (1974) and Singh gt_gl, (1975) have reviewed earlier I: solutions to this problem and present results of different techniques E; applied to the same data sets in order to more completely resolve the ,4 problem. The second harvesting method was chosen for this study along with estimates of senescence and decomposition. To facilitate the harvesting of beachgrass, the study site was divided into 20 x 20 m squares with stakes prior to the beginning of the study. Since beachgrass vigor appears to be directly related to I sand accumulation (Hope-Simpson and Jefferies, 1966; Lain, 1954; Marshall, 1965), five graduated stakes were placed at various positions in the 14 15 study site during this time, and were checked periodically for any change in sand level. 'The number and size of samples to be collected was determined at the beginning of the study in July 1973. From the literature review of grassland primary production measurement techniques, by Milner and Hughes (1970), two sample sizes were selected (0.5 and 1 m2) and used as nested quadrats. The study site was mapped and five points were chosen at 60 m intervals along the lake side edge of the site. Five more points were placed at 9 m intervals back from each point on the lake side edge (total 25 plots). Graig-Smith's (1964) graphical method of deciding when an acceptable number of samples has been obtained was followed. Ten 1 m2 samples were determined to be adequate for this study site. The dry weight of the living harvested beachgrass from the front, back, and sides of the study site (Appendix Table l) were com- pared using the Student's t test and found not to be significantly different (a = .05). The study site was mapped and plots to be harvested were selected using a random numbers table. Plots were not harvested adjacent to a previously harvested plot or to the trail which penetrated the center of the study site to insure a lack of recent disturbance of study plots. The plots to be sampled were located in the field by reference to the permanent stakes. I Above ground biomass was harvested at three week intervals from July 1973 through July 1975 except during the winter months (December through February). Each plot was clipped as close to the ground as - possible and all biomass and litter were removed and placed in plastic bags. The biomass was later separated into living and dead fractions, l6 and the living fraction was inspected for evidence of insect damage. To follow more completely the changes in beachgrass biomass, estimates of belowground biomass were also made. In order to determine the best depth for sampling belowground biomass, all roots and rhizomes in three 1 m2 plots were harvested to a depth of 1 m. The greatest concentration of root and rhizome biomass (85%) occurred in the top 40 cm. Similar results have been found for roots of a tall grass prairie and Tennessee Old fields (Dahlman and Kucera, 1965; Kelly, 1975). Each additional 20 cm from 40 to 100 cm added only 5% more beachgrass root biomass. Because of these findings, belowground biomass (living and dead or total belowground) was sampled by removing all sand and plant biomass to a depth of 40 cm. I During each sampling period in 1973 one of the ten plots to be harvested was randomly chosen for belowground sampling. Since this single sample did not produce any recognizable trends, the belowground biomass of three plots, rather than one, was harvested at each sampling period during the rest of the study. The belowground biomass was separated from the sand by shaking through a screen with a mesh of 6 x 6 mm in the field. It was then placed in plastic bags and later the bio- mass and residual sand were further separated by drying completely, shaking, washing and floating in water, and drying again. To obtain dry weight estimates, the above and belowground samples were dried for 48 hours in a forced air oven at 100°C and weighed on a triple beam balance. These samples were then ground in a Wiley Mill to pass a 0.5 mm mesh. The belowground samples were again separated from .any imbedded sand using a South Dakota Seed Blower. All samples were then subsampled and combusted in a muffle furnace at 550°C and reweighed 17 on a torsion balance to obtain ash free dry weight estimates. The caloric content of living and dead aboveground and belowground biomass harvested during the 1974 sampling period (March through November) was determined using a Parr adiabatic oxygen bomb calorimeter. All samples were run in at least triplicate and on only one occasion did sample replicates differ by more than 0.5%. This variation is within the level of acceptance established by Golley (1961). Since this community had only one plant species which had no annual living aboveground carry over, only one living biomass peak during the growing season, and negligible herbivory (see results), the peak stand- ing crop determined by sampling throughout the growing season was a valid estimation of the aboveground net primary production (Wiegert and Evans, 1964; Odum, 1960). Also, since all belowground biomass be- longed to the one species present, unlike most other grassland studies where multiple root systems exist (Malone, 1968), the measured below- ground biomass gave a valid approximation of net belowground primary production. However the estimate of total net primary production was still low until corrected for mortality and disappearance of dead material. Many of the studies of grassland productivity have neglected the death and disapperance of plant material during the growing season (Singh et_§l,, 1975). Golley's (1965) procedure of estimating the transfer of live to dead biomass and adding this to the estimated net production for a broom sedge community was adopted. However the disappearance of dead material from the plots must also be included for -a complete estimate. This disappearance of dead material was measured according to the method of Weigert and Evans (1964) which used paired 18 plots and litter bags. Their formulas were used to estimate both the disappearance of dead material and green material mortality. Five pairs of paired plots were established at the beginning of each of the following periods: 17 November 1973 - 6 April 1974, 6 April 1974 - 28 June 1974, 28 June 1974 - 1 September 1974 and 1 September 1974 - 2 November 1974. Because of the clumped growth habit of beach- grass, the second plot of each pair was chosen from an area in the immediate vicinity which was most similar to the first plot. The second plot of each pair was harvested at the end of each period. These sampling periods corresponded to the overwinter and pregrowth period, early rapid vegetative growth period, slower growth period associated with the rest of the growing season, and the post growing season period respectively. Thus there were 25 two plot pairs, with the change in five pairs studied during each interval. The nylon litter bags used for sampling litter loss were 38 by 38 cm, with a mesh of 7 by 5 mm, and were filled with approximately 20 g (oven dry weight) of beachgrass leaves. On 24 November 1973, 16 bags were placed on the sand surface and 16 were buried a few cm below the surface. Four aboveground and four belowground litter bags were ran- domly chosen and removed after 140 days, 235 days, 281 days and 344 days in the field, corresponding to the dates at the end of each period when the paired plots were harvested. The dynamics of the aerial shoots were further followed by counting the number of living stems and determining the living leaf area for each plot that was harvested. Laing (1958) estimated there were 50 vegetative shoots/m2 present on his site in the Indiana dune region. However it is not known if the number increased during the growing season. To help 19 determine this, a 1 m2 quadrat, divided into 25 sections, was placed over each clipped plot and the living stems in each section were counted during 1974 and 1975. The total leaf area of all living leaves in a l m2 plot harvested in July 1974 was determined. The plot was divided into 25 sections and each was harvested separately. The living and dead leaves were separated and the living leaves were stored moist at 3°C until fully expanded, at which time they were traced on paper and their areas computed. The living leaves and stems were then oven dried and weighed. The dry weights of the leaves (Appendix Table 2) were regressed on their respective leaf areas (r = .98) to develop a linear prediction equation (y = mx + b) that was used to calculate the leaf areas of all harvested plots: Leaf area (cmz) = 0.6178 x (living dry weight in grams) + 0.1889 Communitngetabolism. Daily rates of gross photosynthesis, community respiration, plant respiration and soil respiration were estimated from C02 samples collected in the field and measured with an infrared gas analyzer (Beckman, Model 215 A) in the laboratory, using a method adapted from that developed by Barko (1975). Three portable chambers were used to enclose portions of the beach- grass community for analysis of CO2 exchange (Figure 3). Two of the chambers were constructed from 0.32 cm thick clear plexiglass, and were 61 cm tall and 16 cm in diameter. Both had 0.64 cm inlet and outlet ports at 55.6 and 17.5 cm from the base respectively. One chamber (dark chamber) was covered with several layers of extra heavy aluminum foil . excluding all light, and the other chamber (light chamber) was trans- parent. The third chamber (soil chamber) was 20 cm tall and 16 cm in 20 Figure 3. Portable metabolism chambers used in gas exchange study. In a clockwise direction starting in upper center: Light chamber (A), soil chamber (8), hand operated pressure- vacuum pump (C) and dark chamber (0). 22 diameter and had 0.64 cm inlet and outlet ports at 15 and 7.3 cm from the base respectively. This chamber was also covered with aluminum foil to exclude light. Determinations of community metabolism were made at three week intervals from June 1974 to July 1975 except during December 1974 through March 1975 when the intervals were extended. Community meta- bolism was measured using six areas with approximately equal numbers of beachgrass plants between 0900 and 2400 hr during a sampling day. The same six areas were used at each sampling time. Measurements usually started at approximately 900, 1230, 1545 and 2330 hr and were completed after 60 to 90 minutes. The light chamber was placed over three of these areas one at a time and the dark chamber over the other three in a similar manner. The soil chamber was placed on three different patches of bare sand close to the plants being monitored. The chambers were pushed into the sand and sand was poured around each base to close large holes. Each chamber was left in place for 15 minutes and then moved to its next site. During this time three samples of ambient air were taken 1.5 m above the ground upwind of the sample area. Gas exchange of the plants used with the light chamber was not measured at night. The biomass beneath and between clumps of beachgrass was sampled to provide an indication of whether the carbon flux between clumps was a valide estimate for that under a clump. A minimum of 20 biomass samples was removed at two different times with a soil core that removed a core 25.3 cm deep by 16.3 cm in diameter. Roots and rhizomes sampled between the clumps and under the clumps were not significantly different (a = .05) from each other and averaged 8.6 and 10.3 g/core sample respectively. 23 The duration of the incubation period did not inhibit the rate of apparent photosynthesis. Gas samples were removed after 5, 10 and 15 minutes of incubation. Between 5 and 10 minutes and between 10 and 15 minutes there was a mean change of 30.64 ppm C02 and 30.92 ppm CO2 respectively, on 11 July 1975 based on three replications. Temperature in all chambers and ambient air were monitored at 4 to 5 minute intervals with a six channel YSI telethermometer. The tem- perature of the light chamber was maintained close to the ambient air temperature by circulating the chamber air 2.3 times per minute with a Millipore pressure-vacuum pump powered by a portable generator through 42.19 m of copper tubing (0.31 cm inside diamter) surrounded by ice in a styrofoam ice chest. The dark and soil chambers remained at their reSpective ambient air temperatures and therefore were not cooled. Gas samples were removed from the dark and soil chambers and the ambient air with a hand operated pressure-vacuum pump (910 ml capacity) causing a sight positive pressure in the chambers. The Millipore pressure-vacuum pump removed the gas samples from the light chamber. The gas samples passed through a column containing filters and drierite to prevent contamination due to microbial activity under humid conditions before entering the 500 m1 glass collecting flasks. An infrared gas analyzer, coupled to a strip chart recorder (Honeywell, Model 193), was calibrated with known gases in 20 cc samples injected into a stream of nitrogen flowing through the analyzer. This produced a series of normal curves on the recorder (Figure 4). The areas under the curves of the calibration gases, determined by plani- metry, were regressed on their respective CO2 concentrations to develop linear prediction equations (y = mx + b). These equations were used to 24 Figure 4. Infrared gas analysis determination of carbon dioxide concentrations. The top curves are from calibrated gases while the bottom curves are from field samples. 25 a 25mm m 35mm a 353 m 353 ew< eonsmcu Lansmgu smasmgu 26.2.5 :8 3:8 22... .ON .om can mm_ Egg mom sag mmw sag com '4 g a CF Jom 26 calculate the CO2 concentration (in ppm) of the sample gases. The three ambient air samples from each 60 — 90 minute sample period were averaged and subtracted from the CO2 concentration in each dark and soil chamber sample and added to each light chamber sample. All samples were then converted to mg C evolved in one hour: mg C/hr = 60 {$96 (273°)x (12 x103mg/mole |5 pmmin2 24 liteerole/X vol of system ‘06 PPM/liter in liters where: CT = chamber temperature in K The carbon flux in the soil chamber was expanded to a square meter basis. The three estimates for soil respiration were averaged and sub- tracted from the carbon flux estimated for each dark chamber and added to each light chamber estimation. The extimation of the carbon flux in the light and dark chamber was then expressed on a square meter basis: new > 6 where: =the weight of living beachgrass in each chamber. FL = the average weight of living beachgrass per square meter harvested during the same sampling period. The three flux estimations from the light chamber were averaged as were the estimates from the dark chamber. Gross photosynthesis was calculated as an hourly rate by summing the carbon flux from the average values estimated in the light and dark chambers. The mean hourly rates were expanded to daily rates by multiplying by the number of hours between sunrise and sunset. Above- ground plant respiration was taken as the carbon flux in the dark 27 chamber minus the soil respiration (including root and microbial respira- tion) as the carbon flux in the soil chamber. The carbon flux through the sand was determined by placing the soil chamber over sand from which the belowground biomass had been removed 24 days earlier. This was compared with the carbon flux measured by placing the soil chamber be— tween undisturbed beachgrass clumps. The carbon flux from the disturbed areas was 45% of that measured on the undisturbed site. Community respiration was taken as the sum of aboveground plant respiration and soil respiration. Daily rates of plant respiration, soil respiration and community respiration were plotted on curves. The growing season rates, sampling period rates and annual mean rates of respiration were determined by integrating the areas under their respective curves with respect to time. Plant respiration plus net primary production (from biomass estimates) were combined to give estimates of gross production. The daily pattern of solar radiation at the study site during each 24 hr sampling period was determined with a Belfort recording pyrhelio- graph. During the entire study period, solar radiation was determined with a recording Eppley pyrheliometer, located at the Kellogg Biological station, approximately 65 Km from the study site. It was assumed that 50% of the total solar radiation (wavelengths between 390 and 760 nannometers) could be used in photosynthesis (Bray, 1961; Talling, 1961). Factors Controlling_Photosynthesis. The effects of varied light inten- sities and the influence of different C02 concentrations on net photo- synthesis of beachgrass were investigated in the laboratory using single leaf chambers. This apparatus is in the laboratory of Dr. Klaus Raschke and has been described (Raschke 9191., 1975). 28 Measurements of young leaves were taken on 16 May 1975 and of mature leaves on 29 July 1975. Several clumps of beachgrass with roots and rhizomes attached were taken from the field the evening before testing. They were placed in containers with sand, watered,taken to the laboratory and placed in a 3°C cold room over night. The following morning, assimilation chambers were clamped to the upper and lower sur- faces 0f fully expanded detached leaves. Light intensities were main- tained with neutral density filters. Leaf chamber temperatures were not controlled and closely approximated temperatures recorded in the field under similar light intensities: leaf chamber temperatures averaged 25°C under lower light intensities and up to 30°C under full light intensity. Rates of net photosynthesis were measured with four replications at varied light intensities of 62.5, 135.0, 252.5, and 500.0 N/m2 (151, 264, 479 and 947 x 102 lux respectively) at a constant CO2 level of 300 ppm. Another measure of photosynthesis, also with four replica- tions, was made at varied CO2 levels 0, 100, 200, 300, 400 and 600 ppm at a constant light intensity of 252.5 W/mz. Finally, chambers were covered with a dark cloth to give an estimation of dark respiration. Since the only measurable gas exchange occurred on the lower sur- face of beachgrass leaves, the location of stomates on leaves was determined by making leaf skeletons (Dilcher, 1974). Leaf pieces were heated in 20% potassium hydroxide for 48 hours and bleached with a 5% solution of hydrogen peroxide. The location of stomates was noted and the number per unit area counted. 29 Estimation of Animal Activity Although tracks of rodents were seen around the study site, the species present were unknown. It was also unknown whether they were feeding on the beachgrass. In order to clarify these points, rodents were trapped and the contents of their stomachs inspected. Whether the invertebrates present in the study site were likely to be feeding on beachgrass was also investigated by identification of the most numerous kinds present. Vertebrates. Rodents were trapped with groups of three Victor snap traps placed at 20 meter intervals along two lines 5 and 35 m in from the lake side border of the study site (total, 96 traps). The traps were baited with peanut butter. During July 1973, it was found that trapping more than one night brought in fewer rodents on later nights as compared to the trapping results on the first night. This has also been found by Gentry and Odum (1957) under similar conditions. Trapping was conducted on the study site for one night at three week intervals corresponding to beachgrass harvesting dates from July 1973 to August 1974 except for the period of November 1973 through March 1974. Rodents were identified according to Burt (1957) and verified by John Fitch (Michigan State University), sexed, and divided into age groups (Blair, 1940a,b,c). To determine if beachgrass contributed to the rodents' diet, their stomachs were opened, the contents removed and placed in afine mesh net, and then washed in water, separating the particles as much as possible. The staining technique was modified from one developed by J. Hall (University of Georgia, personal communication). 1. 50% ethanol 30 seconds 2. 1% Cholozol Black E in 70% ethanol 35 seconds 3O 3. 95% ethanol 60 seconds 4. 95% ethanol (rinse) 30 seconds 5. 95% ethanol (rinse) 30 seconds 6. Absolute ethyl alcohol 30 seconds Sections were brought from absolute ethyl alcohol to xylene and permatized with a 60% Harleco resin solution in xylene. Two slides were made from each stomach. Since the stomach contents were homo- genously mixed, ten locations were randomly located under low micro- scopic power and the contents were classified as plant, invertebrate or unknown (Baumgartner and Martin, 1939; Brusven and Mulkern, 1960; Williams, 1962). The specific identification of beachgrass leaf cells was accomplish- ed by making permanent reference slides. Beachgrass leaf pieces were macerated by hand or separated by Jeffrey's Method (Sass, 1951), and stained and mounted as described previously. Epidermal characteristics present in these reference slides were used for possible identification of beachgrass plant cells present in the rodent stomachs (Baumgartner and Martin, 1939). Invertebrates. Insects were collected in the sample area by sweeping the study site during the growing seasons from July 1973 to August 1974. AiDe Vac was also used during this time. Pitfall traps 8 cm deep by 10 cm in diameter were placed in the sand with the tops at ground level, and filled with 2 to 3 cm of ethylene glycol and a small amount of detergent. Using these different collecting methods, the most numerous kinds of insects present in the study site could be deter- mined (Delong, 1932; Gray and Treloar, 1933; Smalley, 1960). Once the kinds of insects were known their potential for beachgrass herbivory was evaluated. RESULTS Productivity and Community Metabolism Biomass Relations and Community Structurez' Aboveground. Beachgrass began growth in late March, living biomass peaked in mid September, and then commenced to die back (Figure 5). From late March to early June living biomass increased by an average of 1.6 g/mzlday. This rate declined to an average of 1.3 g/m2/day for the period of June through late September. Maximum aboveground living bio- mass production was 230 g/m2 in 1973 and 183 g/m2 in 1974 (Appendix Table 3). Based on the 1973 through 1975 growth rates it was estimated that over 250 g/m2 would have been produced by September 1975. Beach- grass died back at an average of 2.3 g/mzlday from late September into November, although as late as mid November an average of 58 g/m2 of living beachgrass was still present on the study site (28% of the total living biomass). The number of living stems increased annually from April to early June and then decreased through November (Figure 5). Although the trends in 1974 and 1975 were similar, the number of living stems pre- sent was not. The maximum mean number of living stems present per m2 was 280 in 1974 and 233 in 1975. The photosynthetic leaf surface area increased during each growing season until mid September and then declined (Figure 5). The trends were similar to those of the aboveground living biomass. The maximum 31 32 Figure 5. Seasonal changes in aboveground living biomass, leaf surface area and the number of living stems of A, breviligulata, 1973-1975. Stem Number/M2 Leaf Surface Area (cm2)/M2 Grams (dry wt)/M2 300i 250i ZOOI 150. 100* 50‘ 1504 100i 50 l 250. 200. 150I 100‘ 0'1 0 33 Number of Living Stems :7 J1 Leaf Surface Area ,__/’”\. / Living Aboveground 1973 (NW/fl ‘J 'A 1s ' 0‘ N'”rN' A' M r0' 0 'A' s 1974 1975 34 1974 living leaf surface area of 113 cmzlm2 was a 25% decrease from the 1973 maximum of 151 cmzlmz. The aboveground dead biomass has increased by 373 g/m2 from 261 g/m2 to 634 g/m2 over the course of the study (Figure 6, Appendix Table 3). During the growing season in 1974 (April through mid September), dead biomass increased by an average of 0.3 g/m2/day, (total 56 g/mz). From mid September through November in 1973 and 1974, when the living beachgrass was dying back, the dead biomass increased at an average rate of 2.0 g/mzlday. During the entire study, excluding the periods ’between December through February, the dead biomass increased by an average of 0.8 g/m2/day. Also during this period the number of dead stems increased 70% from a mean of 60 stems/m2 in April 1974 to a mean of 197 stems/m2 in late June and July of 1975. Comparing the change in dead biomass during the winter months (December through February) for 1973-1974 and 1974-1975, it appeared that no measurable amounts were lost. Belowground. Total belowground biomass exhibited seasonal changes, but the overall trend was toward increased belowground biomass (Figure 7, Appendix Table 3). The biomass decreased 209 g/m2 (2.9 g/m2/day) from a high of 795 g/m2 during the spring of 1975 to 586 g/m2 (March-May 1974) and then increased 191 g/m2 (8.3 g/mzlday) to a mean of 777 g/m2 (June - July 1974). The initial spring decrease in biomass in 1975 occurred later in the season than during 1974, although it decreased by a similar amount (208 g/mz). This spring decrease corresponded to an increase in living aboveground biomass and living stem numbers. In August 1974, there was a second biomass decrease of 154 g/m2 followed by an increase to 921 g/m2. From March to mid September 1974, there was a net increase 35 Figure 6. Seasonal changes in aboveground dead biomass of A, breviligulata, 1973-1975. 36 mmmEoMm ucaogmm>oa< name 4 R cum? .41.. Eff sPI. IdfiI 2 P o mm _ H MI mnmp m._ on<0 559.38g 32:52.80 553.552 osmficzmopoza .396 O 6 r4 6 .. com . coop .oomp [PO/zN/D 5“! w ooom r comm 4 gen . comm 51 sampling period except in early spring. As stated earlier, it is felt that beachgrass growth during the 1975 growing season would exceed that measured during the 1974 growing season. This was also shown by the estimate of gross photosynthesis on 24 June 1975 which exceeded that on 28 June 1974 by 0.68 g C/mzlday. Peaks of 1.95 and 3.32 g C/mzlday in gross photosynthesis occurred in late June 1974 and in early July 1975, respectively. By the time living aboveground biomass reached its peak (22 September 1974), the rate of gross photosynthesis had decreased 0.71 g C/mzlday from the June high. The mean daily rates for the respective portions of community meta- bolism based on the growing season sampling period, and annual means are presented in Table 3. The ratio of the aboveground plant respiration to gross production during the growing season and annually was 48% and 66%, respectively. Gross production when expressed as a percent of community respiration was 98% during the growing season and 71% during the entire year. The ratios of net production to gross production during the growing season and annually were similar (35% and 34%, respectively). Community respiration followed the general patterns of soil and aboveground plant respiration. It peaked in late June 1974 and early July 1975 at a mean of 3.41 g C/mzlday (24 hr). The ratio of plant respiration to community respiration averaged 47% during the growing season. During October through November 1974 and March through April 1975 the ratio averaged 30%. Aboveground plant respiration averaged 0.27 g C/mz/day (24 hr) during late fall and early spring, increased till August 1974 (1.54 g C/m2/day) and decreased throughout the fall. Soil respiration exceeded plant respiration throughout most of the 52 Table 3. Mean daily rates of A, breviligulata community metabolism. Growing season Sampling period Annual mean(3) mean(]) mean(2) . (g C/mz/day) (g C/mZ/day) (g C/mzlday) Gross primary production 2.640 1.724 1.306 Net primary production 0.926 0.622 0.446 Aboveground plant respiration 1.268 1.102 0.860 Soil respiration 1.409' 1.220 0.968 Community respiration 2.676 2.343 1.828 (1) Based on 179 days from June to September 1974 and April to June 1975. (2) Based on 262 days from June to December 1974 and April to June 1975. (3) Based on 365 days from June 1974 to June 1975. 53 sampling period. Soil respiration increased from February to June (mean of 1.64 g C/m2/day) (24 hr) and then decreased during the rest of the sampling period. On 20 March 1975 soil respiration was estimated to be 0.58 g C/mz/day. At the time of this estimate portions of sand were frozen in the belowground biomass samples. Since the respiration of soil organisms such as fungi, bacteria, and various invertebrates, plus beachgrass roots and rhizomes were included in soil respiration, it was not possible to separate out that portion contributed only by beachgrass. Daily patterns. Daily interrelationships between gross photosynthesis, solar energy, gross photosynthetic efficiency and day length are given in Table 4 for the sampling periods between 28 June 1974 and 11 July 1975. Gross photosynthetic efficiency generally increased during the growing season, however many of the dates had fairly similar efficiencies. The highest efficiencies occurred during 13 October 1974 and 24 June 1975 when a heavy cloud cover was present most of the day. The estimates of plant metabolism on 11 July 1975 were greater, as computed on a per m2 or per gram basis, than estimates on prior sampling dates. An analysis of possible correlative factors affecting gross photosynthesis showed little relationship between gross photosynthesis and day length, solar energy or chamber temperature (Table 5). The highest correlation (r = 0.60) was with solar energy, as would be expected. The measuring of C02 exchange rates at three to four different periods during each diurnal measurement cycle provided an opportunity to observe any diurnal rate fluctuations. Hourly rates of gross photo- synthetic assimilation, plant respiration and soil respiration measured during four days which represent the general trends during spring, 54 Table 4. Daily relationships between gross photosynthesis of A, breviligulata and solar energy, indices of efficiency and daylength. Sampling Gross . (1) Solar Ffficiencyfz) Day— (3) photosyntheSis energy index (x10 3) length 9 C/mzlday g cal/cmzlday 28 June 1974 1.95 434.27 4.49 14.75 28 July 1.80 ‘ 440.19 4.09 13.33 9 Aug. 1.63 343.26 4.74 12.75 1 Sep. 1.53 304.66 5.01 10.83 22 Sep. 1.24 286.14 4.61 10.33 13 Oct. 0.78 83.07 9.39 8.75 2 Nov. 0.34 182.50 1.85 8.25 23 Nov. 0.32 78.46 4.09 8.00 22 Apr. 1975 0.24 307.12 0.80 11.75 11 May 1.09 261.42 4.16 13.25 3 June 1.72 423.03 4.05 13.75 24 June 2.63 216.95 12.12 14.00 11 July 3.32 427.80 7.76 13.50 (1) Calculated by multiplying mean uptake by hours (2) Gross photosynthesis/solar energy. (3) Hours between sunrise and sunset. of daylight. 55 Table 5. Analysis of factors correlated with gross photosynthesis(1). Correlation coefficient (r) Gross photosynthesis and daylength (2) 0.53 Gross photosynthesis and solar radiati0n(3) 0.60 Gross photosynthesis and chamber temperature 0.54 (1) Linear regression analysis (2) Hours between sunrise and sunset (3) g Cal/cmzlday (50% of total) summer and fall are presented in Figure 12. Rates of gross photosynthetic assimilation during April and May increased from morning to early afternoon and then decreased during late afternoon. During this time the greatest rates per gram of plant tissue were recorded. During the rest of the sampling period, early afternoon rates did not exhibit any consistent trends. At the time the late afternoon measurements were made there was a trend for the rate of gross photosynthetic assimilation to return to that which had been measuredirithe morning. During the period of June through July, the morning, early afternoon and late afternoon rates generally decreased for each consecutive sampling date. There was no consistent trend for early afternoon rates. Late afternoon rates tended to return to rates measured in the morning regardless of whether early afternoon rates had increased or decreased. .The efficiency indices (gross photosynthesis/solar radiation) increased from early afternoon to morning to late afternoon. During 56 Figure 12. Hourly rates of gross photosynthetic assimilation, plant respiration and soil respiration. mg C/g (dry wt)/hr Gross Photosynthesis and Plant Respiration (11 May 1975 0 Gross Photosynthesis 3 0 ZSPlant Respiration EJSoil Respiration 2.0, 1.0.I 11 July 1975 > 2.0. '2 1.0. 1 p 1 1 September 1974 W .1 0.5. 2 November 1974 0.5. Tif— 000 600 Soil Respiration mg C/14.5 cmz/hr 58 late afternoon measured insolation was usually decreased by increased cloud cover. The daily and hourly efficiencies indicate that beach- grass was light saturated at least during the early afternoon period and possibly most of the day during much of the year when skies were clear (Zelitch, 1971). The highest rates of plant respiration per gram of plant tissue per hour were recorded during the period of March through May. From March through August, plant respiration rates increased during the day and at night decreased to a point lower than the morning rates. During the period of June through August plant respiration generally decreased at each consecutive sampling period: however, there was much fluctuation. The rates decreased from morning to early afternoon and increased to a high in the late afternoon during the period of September through November. During this period the night rates were at their lowest levels. Hourly soil respiration rates did not exhibit any general day time trends. During the period of June through September, night rates were lower than the mean day rates. Rates measured at night during March and during the period of September through November were lower than any hourly day rates. Photosynthetic Assimilation Under Different Conditions Response tg_varyinglight intensities. The results in this section and the next are based only on responses measured with assimilation chambers on the lower leaf surfaces, since little to no gas exchange was detected with upper leaf surface assimilation chambers. Young leaves photo- synthesized more rapidly than mature leaves (Figure 13): The mature 59 Figure 13. Net photosynthetic rates of single leaves of A, breviligulata in relation to light intensity. 60 12‘ :I i 1 SE 10, Young Leaves 84 e4... Mature Leaves 24 ‘05 206 1 300 406 50.0 L1°9ht Energy (w/MZ) 61 2 leaves reached light saturation at 252.5 N/m2(479 x 10 lux), whereas the young leaves were not saturated at 500 H/m2 (947 x 102 lux), the highest measurement used. Stomatal oscillations occurred when mature leaves were exposed to a light intensity of 500 Wm2 (947 x 102 lux). Rates of photosynthesis fluctuated between an average low of 5.8 i 0.6 mg C02/dm2/hr to an aver- age high of 7.3 i 0.7 mg C02/dm2/hr. Internal C02 and conductance were in phase with each other, with periods of 30 minutes. Response to varying,C02 concentrations. Under a light intensity of 252.5 11m2 (479 x 102 lux) and different levels of CO2 concentration, the rate of photosynthesis of the young leaves was again greater than that of mature leaves (Figure 14). The young and mature leaves had graphically the same C02 compensation point of 44 ppm. Using the slope of the curves between the compensation point and 200 ppm as a relative comparison of the efficiency of the leaves for photosynthesis, the young leaves proved to be more efficient than the mature leaves. The points at 0 ppm represent a minimum estimate of photorespiration (Zeltich 1971). The greater photorespiration by the young leaves was increased by a larger rate of dark reSpiration. The ratio of photo- reSpiration to dark respiration for young leaves was 1.24 and mature leaves was 1.11. The slight gas exchange from the upper leaf surface during the pre- vious varied conditions was due to the lack of stomates. Stomates occurred in rows on the abaxial surface of the leaves between protruding vascular bundles and averaged 27 stomates per 1.5 x 10'3 mmz. The C02 compensation point (44 ppm) and the crossectional leaf anatomy were similar to a group of plants collectively called low 62 Figure 14. Net photosynthetic rates of single leaves of A, breviligulata at a constant light intensity of 252.5 W/m2 and different C02 concentrations. 16‘ 1411 12 i 10‘ 84 61 4.1 mg COzldmzlhr 2'1 63 1.]. T Young Leaves __I 1. Mature Leaves 0 100 200 A 300 PPM co2 400 500 w 600 64 photosynthetic capacity plants (Black, 1971) or "C3" plants (Downton and Tregunna, 1968). Vertebrates Four species of rodents Peromyscus maniculatus bairdi (prairie deer-mouse), Microtus pennsylvanicus (meadow vole), Zapus hudsonius (meadow jumping mouse), and Mus musculus (house mouse) totaling 54 individuals, were trapped during the 1973 through 1974 trapping period (Table 6). Juvenile male 3, maniculatus comprised the largest group of individuals caught. 0f the 54 mice trapped, 76% were either juveniles, subadults or young adults; of these 74% were males. In the mornings numerous rodent tracks were observed in the sand of the old dune buggy road and the small blowout just south of the study site. Progressing from the south end toward the north end of the study site, 34 rodents (63% of the total) were trapped in the first 40 m, 8 (15%) from 40 to 140 m, 4 (7%) from 140 to 240 m, and 8 (15%) from 240 to 300 m. Examples of Peromyscus, Zapus and Mp§_were trapped throughout the study site, whereas those of Microtus were predominantly trapped within the 0 to 40 m area. Trapping during the periods of August through October 1973 and July through August 1974 yielded the most rodents. Although trapping was not done during the winter, rodent tracks were not seen in the snow on the study site either, but were observed further inland. The stomach contents from these catches were classified as belong- ing in one of three categories: Insects (including all chitinous ‘ material), plants and unknown. The stomachs of Peromyscus contained predominantly insect remains along with some plant material. This 65 pm m o m N mN ¢ m up wugwmn mspmpzuPcmE msumzsogma .eeoe e_=e< e_=eee=m e_eee>=a .eeoe e_=e< p_=eee=m aFeee>=n N N p — o o o mapsomze ms: m P o F N p P mzwcomuzn maqu mp N m e FF e N maowcw>~xmccum mzuoguPz _epoe eF=u< p.388 mesa» .eooe p_=e< “Page acne» Page» mpmsmm mpg: mmwumam .muwm xvsgm mew cwguwz tenancy mpcmvoc we mqaogm mom can xom .mmwuoam mo Leeann ask .0 open» 66 predominance of insects in their diets is characteristic (Hamilton, 1941; Jameson, 1952; Whitaker, 1966). The identifiable stomach contents of Ap§_and Zapp§_were equally divided between plant and insect materials. Although plants predominated in the diet of Microtus, some insects were also consumed, an observation also made by Thompson (1965) and Zimmerman (1965). In only four Microtus stomachs were there any cells which resembled beachgrass. The content of rodent stomachs, in total, consisted of 38% insect parts, 35% plant parts and 27% unknown. Identifiable beachgrass was present in only three of 54 stomachs examined. Invertebrates Various species of invertebrates were collected during the sampling period. Species of Diptera formed the largest group of individuals pre- sent. Other orders represented were: 'Hymenoptera, Coleoptera, Orthoptera, Hemiptera, Homoptera, Lepidoptera, I50ptera, and Odonata. Species of Phalangida and Araneida were also present. 0f the preceding, only species of the families Acrididae, Cicadellidae, Chrysomelidae and Scarabaeidae may possibly be feeding on the beachgrass, and except for one species of Scarabaeidae (Phyllophaga rugosa) only small numbers of individuals of these families were found at any one time. Larvae, pupae and adults of Phyllophaga rugosa (June bug) were found when separating the sand from the belowground biomass. It was assumed that the larvae were feeding on the roots and/or rhizomes of the beachgrass. It is known that Phyllophaga ppgp§p_larvae annually .have done considerable damage to cereal and forage crops in this manner (Anon. 1959; Forbes, 1907; Teetes, 1973). The larvae were found 67 predominantly in areas where the most vigorous beachgrass growth was occurring. During the 1973 sampling period, larvae were found only during August. At this time eight adults, six larvae and four pupae 2 were uncovered in three 1 m samples. From March into June 1974 an average of two adults and two larvae were found per 1 m2 to a depth of 40 cm. By the end of June through July 1974 none were found in the belowground samples, although adults were seen and heard at night around the Populus deltoides behind the study site. During the rest of 1974 there was an average of one larva/m2. Larvae were always found in concentrations of at least two or they were entirely absent from below- ground samples. During June and into July 1975, larvae averaged seven/ m2 and corresponded with sites of active beachgrass growth. Mycorrhizal Fungi, Throughout the sampling period, balls of tightly compacted sand grains were found when separating the belowground biomass from the sand. The occurrence of small balls of sand has been reported elsewhere (Brown, 1958; Koske gt_a1,, 1975; Olsen, 1958a). Koske §t_al,, (1975) have reported an endomycorrhizae of the family Endogonaceae associated with many plant species colonizing dunes on the eastern shores of Lake Huron. Their physical description of the balls of sand is similar to those found in this study. Staining the "sand balls" with acid fuchsin in lactophenol revealed the presence of a large number of spores. The spores are of a vesicular- arbuscular mycorrhizal fungus and closely resemble those produced by , Glomus macrocarpus Tul. and Tul. (J. N. Gerdemann, University of Illinois, personal communication). Dr. Gerdemann suspects that the 68 “sand balls" may be sporocarps although they may must be dense clusters of chlamydospores enclosing a mass of sand. Staining a small clump of roots showed fungal mycelia in close proximity to beachgrass roots. While positively identified penetration points could not be found, Dr. Gerdemann thinks there is little doubt that the fungus is mycorrhizal on Ammophila breviligulata. DISCUSSION Beachgrass Primary_Production The net primary production of the beachgrass community during 1974 was 347 g/m2 based on aboveground and belowground accumulation and aboveground senescence. Net primary production for 1975 was estimated to be 550 g/m2. This estimate was based on the extrapolation of measurements made during the period of March through July 1975 to the end of the growing season, using the trends established during 1973 and 1974. The estimated greater productivity during 1975 relative to 1974 would probably be related to two factors: sand accumulation and high amounts of spring precipitation. The growth of beachgrass is known to increase with sand deposition (Hope-Simpson and Jeffereies, 1966; Laing, 1958; Marshall, 1965). During the growing seasons of 1973 and most of 1974, there was no measurable sand deposition on the study site, but between September 1974 and June 1975, 2-3 m of the foredune west of the study had eroded and by July 1975 4-5 cm of sand had accumulated behind this eroded area. Year-to-year differences in precipitation during May and June account for 74% of the variation of forage production in short- grass prairie (Smoliak, 1956). Precipitation on the study site during the period of April through July 1975 was 14.0 cm above the 29 year norm of 33.4 cm, and 52% above that measured during a similar period of time in 1974. 69 70 Estimates of beachgrass net primary production were comparable to the average (500 g/m2) annual temperate grassland net primary production which ranges from 150 to 1500 g/m2 (Whittaker, 1970). However, since estimates of senescence were not included in the temperate grassland estimates, their values could be up to twice as high, based on the re- sults of Bradbury and Hofstra (1976). Niegert and Evans (1964) estimated the net primary production of an upland Michigan old field, including estimates of above and belowground production and living aboveground senescence, to be 1991 g/mz, a substantially greater amount than the beachgrass community. The productivity of beachgrass can be increased with fertilization. Previously, artificially planted and fertilized l- to 3-yr-old stands of beachgrass when fertilized have pro- duced up to 380 g/m2 of aboveground living biomass (Noodhouse and Hanes, 1966) compared to the 183 g/m2 produced during the 1974 growing season of this study. However Woodhouse and Hanes did not determine above- ground senescence or belowground production. The growing season net production efficiency of the beachgrass community (0.23%) was similar to those of desert (0.19%) (Chew and Chew, 1965) and xeric alpine (0.20%) (Scott and Billings, 1964) regions (as cited in Jordan, 1971), but was less than that of a Missouri tall- grass prairie (0.54%) (Kucera gt_al,, 1967). This difference from the Missouri tall grass prairie would be expected because of the clumped growth habit of beachgrass with areas of open sand between clumps instead of a more uniform distribution of leaves which would more completely intercept incident light. The difference was also partly due . to the greater nutrient levels present in the prairie soil compared to the dune sand. 71 The rate of gross photosynthesis increased with increasing above- ground biomass and increasing photosynthetic leaf surface area. Once the beachgrass leaves were fully expanded, there was a gradual decrease in the photosynthetic rates for the remainder of the growing season. The decline in the rate of photosynthesis was probably related to the shading of leaves and the lower photosynthetic rates of mature leaves which were present in a higher proportion in the beachgrass community during the late part of the growing season. Laboratory investigations indicated a decline in net photosynthetic potential with leaf age. This decline also has been described in other species (Hardwick §t_gl,, 1968; Treharne gt_al,, 1968; Noledge, 1971; and others). In this study the young beachgrass leaves photosynthesized at a more rapid rate than the mature leaves and were not light saturated at 500.0 Wm2 (947 2 lux) whereas the mature leaves were light saturated at 252.5 N/m2 (479 x 102 x 10 lux). The decreased rate of photosynthesis could be partially attributable to the increased rate of plant respiration measured in the field, since respiration also decreases in older leaves but at a slower rate than photosynthesis (Zelitch, 1971). The rate of gross photosynthesis when expressed as g C/g (living beachgrass enclosed in a chamber)/hr also decreased during the season. Generally the morning and afternoon rates of gross photosynthesis de- creased throughout the annual study period. The noon rates generally also decreased during the annual study period but the comparison of noon rates to morning and afternoon rates did not show any consistent seasonal trend. Midday fluctuations in the rate of gross photosynthesis appeared to be correlated with light intensity and air temperature. Because 72 winds off Lake Michigan at times moderated air temperatures, high light intensities and high air temperatures did not necessarily occur together. During the periods of April through May and September the trends of in- creasing light intensity and air temperatures from morning through mid- day and then declined in the afternoon. During the period of June through August the mi-day rates decreased (compared to morning and afternoon rates) when light intensities approached 10,000 foot candles (measured with a Heston Electric Corp. footcandle meter) and also when air temperatures exceeded 30°C. During the times of depressed rates of gross photosynthesis the midday rates of plant respiration (dark chamber) increased while those of photosynthesis (light chamber) decreased when compared to morning and afternoon rates. The increased rates of respiration were likely due to increased temperatures and the lower rates of photosynthesis were likely due to increased photorespiration with increasing tempera- tures. Zelitch (1971) has reviewed different studies relating to these occurrences. The phenomenon of photorespiration, well known in terrestrial plants, has recently been reviewed by Chollet and Orgren (1975). The gas exchange methods for estimating photosynthesis in the field did not distinguish between mitochondrial (dark) respiration and photorespira- tion. Because estimates of photorespiration were not included in this study, gross photosynthesis was underestimated, but to what extent is un known . The close interrelationship of living above and belowground portions of the beachgrass community can be seen when their respective biomass estimates are compared. The living aboveground beachgrass exhibited a 73 rapid spring increase, a slower summer rate of increase and a fall die- back. In contrast, the belowground biomass showed an inverse pattern of biomass increase except during June and July. This inverse pattern may in part represent seasonal root decomposition, and/or root to shoot and shoot to root carbohydrate translocation. Since results from the belowground litter experiment and the work of Olson (1958a) indicate the rate of decomposition and organic matter build-up in sandy soils to be a very slow process, the former may not be involved in this case. However, Poa pratensis shows a similar spring and fall inverse root and shoot biomass relationship (Bernard, 1974), and Bernard attributes this to translocation of material between above and belowground portions. Carbohydrate concentrations in roots have been shown to decrease during periods of rapid aboveground growth. The carbohydrate concentrations again increased in the belowground portions when the aboveground growth rate slowed down and the photosynthetic leaf area increased (McCarty, 1935). During the fall the belowground biomass once more increased as the living aboveground biomass decreased. This could possibly represent a shoot to root carbohydrate translocation. Riemann (1940) and McCarty (1935) have shown that in Chloris gaygpg, Elymus ambigus and Muhlengeria pracilis the root carbohydrate content in the fall increased while the aboveground carbohydrate content decreased an equivalent amount. The proportion of the living aboveground biomass that was incor- porated into the aboveground dead and belowground biomass between mid ‘ September and mid November of 1974 was determined by measuring weight changes in the respective biomass portions. During this period 134 g/m2 74 of living aboveground biomass died and the belowground biomass and the dead biomass increased 39 g/m2 and 108 g/m2, respectively. This com- bined increase of 147 g/m2 exceeded that lost by the living aboveground biomass by 13 g/m2 but was within 1 SE for both the dead above and total belowground biomass. The large amount of belowground biomass compared to living above- ground biomass was evident from the root/shoot ratio. However, how much of the belowground biomass was living and functioning was not determined. The ratio of belowground to aboveground production for 1974 was 0.5. Bray (1963) compiled root/shoot ratio data from a large number of sources and showed that the ratio of belowground to above- ground production for certain grasses increased as xeric conditions in- creased. In Bray's study, plants with a production ratio similar to that of beachgrass were considered to be mesic. Chapman (1964) states that the dry character of a sand dune habitat may be more apparent than real. He states that water tables under dunes tend to be dome shaped. Together with this and the fact that McLeod (1974) found water to be permanently available for plant usage at depths of 25 cm (at a site 2 Km south of the present site) it was felt that the root/shoot production ratio of 0.5 illustrated that either roots were efficient or that water was not a limiting factor in the growth of beachgrass. Soil respiration increased in the spring until June and then de- creased for the remainder of the season. Soil respiration of the beach- grass community, when compared to estimates of other communities includ- ing pine oak forest, cedar stand, wet tropical secondary growth and Grape Peninsula grass community as reviewed by Kucera and Kirkham (1971), was most similar to that of a Missouri prairie. However, the total soil 75 carbon flux of the beachgrass community was less than that of the prairie, as expected, since the prairie had a greater belowground bio- mass (Dahlman and Kucara, 1965). Rates of soil respiration have been found to decrease with de- creasing soil moisture (Kucera and Kirkham, 1971; Wiant, 1967a) and to increase with increasing soil temperature (Kucera and Kirkman, 1971; Reiners, 1968; Niant, 1967b). Monteith gt_§1, (1964) have suggested that the availability and depletion of substrate material for micro- organisms may also be contributing to the seasonal change in soil respiration. Two other factors which may be involved in the soil respiration of a stabilized dune are decomposition of material from previous years and the possible diffusion of C02 within the dune. The microbial population of developing dunes is low and will remain low until the dune has become stabilized (Brown, 1958). This is due to the continual deposition of sand. The conditions associated with this are unfavorable to microorganisms (Nebley §t_gl,, 1963). Once a dune has become stabilized, the beachgrass root system has been shown to be associated with increased numbers of microorganisms (Kisiel, 1970; Marchant, 1970; Nicolson, 1960; Hahlrab gt a1,, 1963). The soil respiration on a recently stabilized dune (such as this study site) may represent decomposition of roots and rhizomes that have been in existence several years but were only recently being used as a substrate for microbial activity. The measurement of soil respiration may also be complicated by C02 diffusion from different areas within the dune. .To test this, the below- . ground biomass was removed from a square meter plot to a depth of 40 cm and the sand was replaced, thus destroying the sites for microbial 76 activity. However, 24 days later the carbon flux from this area was 45% of that measured on undisturbed sites. Possibly the soil respira- tion measured on the harvested site was due to the diffusion of CO2 from areas of decomposition and respiration. Another possibility was that by removing the belowground biomass and then replacing the sand favorable conditions for short term microbial activity were created. These are two factors which may enhance beachgrass growth where there is no sand accumulation. The first is the slow decomposition of dead aboveground biomass and the second is the apparant association of beachgrass roots with a vesicular-arbuscular mycorrhizal fungus. The region of root initiation in beachgrass communities without sand accumulation is near the surface (Olson, 1958a) where soil tenperatures increase and soil moisture decreases (McLeod, 1974). Dur- ing the period of June through August noon soil (bare sand) temperatures were frequently observed to exceed 30°C during this study and that of McLeod's (1974). Although optimum soil temperatures for root initiation and elongation vary with different species, soil temperatures above 30°C have been found to be detrimental (Hansen, 1971). The slowly decomposing material would provide a physical means of cooling the soil through shading. The ability of beachgrass to cool sand by shading has been previously demonstrated by Salisbury (1934). The accumulation of dead aboveground plant material may also be a means of retaining mineral nutrients in a rapidly leaching sand dune community, where, by its slow rate of decomposition it would release nutrients over a long period of time. Gerdemann (1968) cited evidence that vesicular-arbuscular mycorrhizal infected plants had enhanced growth compared to non-infected 77 plants. Although A, arenaria has been found to have a mycorrhizal association (Nicolson, 1960), such an association has not been previous- 1y reported for A, breviligulata. Howevec such an association would be expected since conditions of high light intensity and a deficiency of available nitrogen or phosphorus, both present on dunes, make plants more susceptible to mycorrhizal infection (Bjorkman, 1942). The mycorrhizal association of A, arenaria occurred only after the dune surface had become stabilized (Nicolson, 1960). This would also appear to apply to A, breviligulata. Estimates pf_Animal Activity Vertebrates. The presence of only three meadow jumping mice and three house mice in the trapping samples of 1973 and 1974 suggest that these species were only transient and would be found infrequently in the area of the study site. The large majority of juvenile and sub-adult (87%) prairie deer-mice and of young-adult (65%) meadow voles raises the question of whether, in fact, the beachgrass stand supports any per- manent rodent population. I do not believe that it does. The litera- ture contains many examples of young mice wandering for a time after leaving the nest (Blair, l940a,b, 1951; Burt, 1940; Howard, 1949; and others). Only one pregnant mouse (a house mouse) was trapped during the sampling period. If the adult mice represented a resident population it would be expected that more pregnant females would have been trapped. During the entire beachgrass sampling no rodent holes, run- ways or aboveground nests were found on the study site. Also a resident rodent population would be expected to support some form of predators, none of which were caught. The lack of tracks or trapped predators does 78 not disprove their presence, however if present, they appear to be scarce. The presence of the meadow voles was surprising since they have previously been reported to prefer more mesic areas in Michigan (Blair, 1940a); and New Jersey (Shure, 1970). Blair (1940b) reports a contin- uous flow of transients moving through an area without any taking up residence. 0f the meadow voles, 83% were trapped within 40 m of the old road along the south side of the study site. Occasionally beach- grass stems showing signs of feeding by meadow voles as defined by Bailey (1924) and Summerhayes (1941) were found in this area of the study site. Meadow voles feed by a continuous process of trial and error (John Fitch, Michigan State University, personal communication). The absence of beachgrass in meadow vole stomachs indicates little to none was in fact eaten. Since at times large amounts of marsh vegeta- tion carried by the Kalamazoo River into Lake Michigan were deposited on the shore and fringes of the SN corner of the study site, it is possible that the meadow voles were feeding on this and occasionally on insects. Zimmerman (1965) has found insect remains in meadow vole stomachs. Even though prairie deer-mice were trapped throughout the study site, 62% were trapped within 40 m of the old road on the south. This area contained some of the densest beachgrass clumps and it was likely that it contained the highest concentrations of June bug larvae. Several times in this area small holes were found in the soil approxi- mately 10 cm deep. Upon excavation there were no noticeable chewed beachgrass roots or rhizomes. It is felt that praire deer-mice dug these looking for June bug larvae, pupae or adults. Deer mice appeared 79 to utilize whatever was available such as insects (larvae and adults), spiders, seeds and occasionally portions of plants as also seen by Hamilton (1941), Jameson (1952), Whitaker (1963) and Williams (1959). Invertebrates. Although a large variety of insects were present in and around the beachgrass community, only June bug larvae were found to be residents of the community and feeding on it. The concentration of 0.4 larvae/m2 in 1974 and 1.1 larvae/m2 in 1975 would be classified as a light infestation compared to a severe infestation of 63 to 125 larvae/m2 (Hammond, 1948). The herbivorous insects present were thought not to be residents of the beachgrass community but blown into the community by on or offshore winds. The beachgrass community does support a population of spiders and insects (ants, tiger-beetles, wasps, etc) which are predators and scavengers that utilize these transient invertebrate populations (Shelford, 1913). In conclusion it is felt that beachgrass has a major structural and perhaps minimal nutrient relation to the majority of vertebrates and invertebrates found in its community. Beachgrass not only binds sand but also catches material blown from the shore which may belater consumed by the vertebrates and gives insects resting and hiding places. These insects in turn support a population of predators and opportunists (spiders, ants and rodents). These predators and opportunists are in turn hidden by beachgrass from other predators (e.g. wasps and whatever might prey on the rodents). Invertebrates survive on dunes by being able to tolerate great temperature extremes or avoiding them by only short exposures (Chapman gt_a1,, 1926). Beachgrass undoubtedly also aids insects by creating an environment of lower temperatures by shading (Salisbury, 1934). Beachgrass, though 80 not contributing directly to most invertebrate and vertebrate diets does provide a structural benefit within which they are able to exist. LITERATURE CITED LITERATURE CITED Anon. £957. Reclamation of sand dunes in New Zealand. Wld. Crop 9: 09. Anon. 1959. Control of common white grubs. U.S.D.A., Farmers' Bulletin, No. 1798. 13 pp. Bailey, V. 1924. Breeding, feeding and other life habits of meadow mice (Microtus). J. Agr. Research. 27:523-535. Barbour, M. G., T. M. DeJong and A. F. Johnson. 1975. Additions and corrections to a review of North American Pacific coast beach vegetation. Madrono. 23:130-134. Barko, J. W. .1975. Primary production and ecosystem metabolism in a Lake Michigan dune pond. Ph.D. dissertation. Michigan State Univ. E. Lansing. 96 pp. Baumgartner, L. L. and A. C. Martin. 1939. Plant histology as an aid in squirrel food-habit studies. J. Wildl. Mgmt. 3:266-268. Bernard, J. M. 1974. Seasonal changes in standing crop and primary production in a sedge wetland and an adjacent dry old-field in central Minnesota. Ecology. 55:350-359. BjBrkman, E. 1942. Uber die Bedingungen der Mykorrhizabildung bei Kiefer und Fichte. Symbolae Botan. Upsalienses. 6:1-190. As cited in Gerdemann, 1968. Black, C. C. 1971. Ecological implications of dividing plants into groups with distinct photosynthetic production capacities. Adv. Ecol. Research. 7:87-114. Blair, W. F. 1940a. A study of prairie deer-mouse populations in southern Michigan. Am. Midl. Nat. 24:273-305. Blair, W. F. 1940b. Home ranges and populations of the meadow vole in southern Michigan. J. Wild. Mgmt. 4:149-161. Blair, W. F. 1940c. Home ranges and populations of the jumping mouse. Am. Midl. Nat. 23:244-250. Bordeau, F. F. and G. M. Woodwell. 1965. Measurements of plant carbon dioxide exchange by infrared absorption under controlled conditions in the field. In Methodology of Plant Eco-Physiology, Proceedings of the Montsellier Symposium, UNESCO. Paris. pp. 283-269. 81 82 Bradbury, I. K. and G. Hofstra. 1976. Vegetation death and its importance in primary production measurements. Ecology. 57:209-211. Bray, J. R. 1961. An estimate of a minimum quantum yield of photo- synthesis based on ecologic data. Plant Physiol. 36:371-373. Bray, J. R. 1963. Root productivity and the estimation of net productivity. Can. J. Bot. 41:65-72. Bray, J. R., O. B. Lawrence and L. C. Pearson. 1959. Primary production in some Minnesota terrestrial communities for 1957. Oikos. 10:38-49. Breckon, G. J. and M. G. Barbour. 1974. Review of North American Pacific coast beach vegetation. Madrono. 22:333-360. Brown, J. C. 1958. Soil fungi of some .British sand dunes in relation to soil type and succession. J. Ecol. 46:641-664. Brown, R. L. and A. L. Hafenrichter. 1948a. Factors influencing the production and use of beachgrass and dunegrass clones for erosion control. I. Effect of date of planting. Jour. Amer. Soc. Agron. 40:512-521. Brown, R. L. and A. L. Hafenrichter. 1948b. Factors influencing the production and use of beachgrass and dunegrass clones for erosion control. II. Influence of density of planting. Jour. Amer. Soc. Agron. 40:603-609. Brown, R. A. and A. L. Hafenrichter. 1948c. Factors influencing the production and use of beachgrass and dunegrass clones for erosion control. III. Influence of kinds and amounts of fertilizer on production. Jour. Amer. Soc. Agron. 40:677-684. Brusven, M. A. and G. B. Mulkern. 1960. The use of epidermal characteristics for the identification of plants recovered in fragmentary condition from the crops of grasshoppers. North Dakota Agr. Exp. Sta. Res. Rep. No. 3. 11 pp. Burt, W. H. 1940. Territorial behavior and populations of some small mammals in southern Michigan. Misc. Publ. Mus. Zool. Univ. Mich. NO. 4551-580 Burt, W. H. 1957. Mammals of the Great Lakes Region. The University of Michigan Press. Ann Arbor, Michigan. 246 pp. Chapman, R. H., C. E. Mickel, J. R. Parker, G. E. Miller and E. G. Kelly. 1926. Studies in the ecology of sand dune insects. Ecology. 7:416-426. Chapman, V. J. 1964. Coastal Vegetation. The MacMillan Co. New York, N. Y. 245 pp. 83 Chew, R. M. and A. E. Chew. 1965. The primary production of a desert- shrub (Larrea tridentata) community. Ecol. Monogr. 35:355-375. Chollet, R. and W. L. Ogren. 1975. Regulation of photorespiration in C3 and C4 species. Bot. Rev. 41:138-179. Coaldrake, J. E., M. McKay and P. A. Roe. 1973. Annotated Biblio- graphy on the Ecology and Stabilization of Coastal Sand Dunes, Mining Spoils and other Disturbed Areas. CSIRO. Div. Plant Industry, Canberra, Australia. 158 pp. Coaldrake, J. E. and K. J. Beattie. 1974. Annotated Bibliography on the Ecology and Stabilization of Coastal Sand Dunes, Mining Spoils and other Distrubed Areas. Sup. No. 1. CSIRO. Div. Plant Industry, Canberra, Australia. 103 pp. Cowles, H. C. 1899. The ecological relations of the vegetation on the sand dunes of Lake Michigan. Bot. Gaz. 27:95-117, 167-202, 281-308. 361-391. Dahlman, R. C. and C. L. Kucera. 1965. Root productivity and turnover in native prairie. Ecology 46:84-89. DeLong, D. M. 1932. Some problems encountered in the estimation of gnsect populations by the sweep method. Ann. Entomol. Soc. Am. 5:13-17. Dilcher, D. L. 1974. Approaches to the identification of angiosperm leaf remains. Bot. Rev. 40:1-157. Downton, W. J. S. and E. B. Tregunna. 1968. Carbon dioxide compensation - its relation to photosynthetic carboxylation reactions, systemat- ics of the Gramineae, and leaf anatomy. Can. J. Bot. 46:206-215. Fernald, M. L. 1920. The American Ammophila. Rhodora. 22:70-71. Forbes, S. A. 1907. On the life history, habits and economic relations of the white-grubs and may-beetles. Ill. Agr. Exp. Sta. Bul. 116-445-480. Gemmell, A. R., P. Greig-Smith, and G. H. Gimingham. 1953. A note on the behaviour of Ammophilia arenaria (L.) Link. in relation to sand-dune formation. Trans. Proc. Bot. Soc. Edinb. 36:132-136. Gentry, J. B. and E. P. Odum. 1957. The effect of weather on the winter activity of old-field rodents. J. Mamm. 38:72-77. Gerdemann, J. W. 1968. Vesicular-arbuscular mycorrhiza and plant growth. Ann. Rev. of Phytopathology. 6:397-418. Golley, F. B. 1961. Energy values of ecological materials. Ecology. 42:581-584. 84 Golley, F. B. 1965. Structure and function of an old-field broom- sedge community. Ecol. Monogr. 35:113-137. Gray, H. E. and A. E. Treloar. 1933. On the enumeration of insect p0pu1ations by the method of net collection. Ecol. 14:356-367. Grieg-Smith, P. 1964. Quantitative Plant Ecology. Butterworths, London. 256 pp. Hadley, E. B. and B. J. Kieckhefer. 1963. Productivity of two prairie grasses in relation to fire frequency. Ecology. 44:389-395. Hamilton, W. J. Jr. 1941. The food of small forest mammals in eastern United States. J. Mammal. 22:250-263. Hammond, G. H. 1948. The distribution, life-history and control of Ph 110 ha a anxia Lec. in Quebec and Ontario. Scientific Agr. 48:403-416. Hansen, E. H. 1971. The short term effect of temperature on root elongation and initiation. M.S. dissertation. Michigan State Univ. E. Lansing. 97 pp. Hardwick, K., M. Wood and H. W. Woolhouse. 1968. Photosynthesis and respiration in relation to leaf age in Perilla frutescens. New Phytol. 6:79-86. Herms, W. B. 1907. An ecological and experimental study of the Sarcophagidae with relation to lake beach debris. Jour. Exp. 2001. 4:45-89. Hitchcock, A. J. 1950. Manual of the Grasses of the United States. U.S. Dept. of Agr. Misc. Pub. No. 200. 1051 pp. Hope-Simpson, J. F. and R. L. Jefferies. 1966. Observations relating to vigour and dzebility in marram grass (Amnophilia arenaria (L.) Link). J. Ecol. 54:271-274. Howard, W. E. 1949. Disperal, amount of inbreeding and longevity in a local population of prairie deer mice on the George Reserve, Southern Michigan. Univ. of Mich. Lab. Vert. Biol. Contrib. 43. 50 pp. Jagschitz, J. A. and R. S. Bell. 1966. American beachgrass (establish- ment - fertilization - seedling). Rhode Island Agr. Exp. Sta. Bul. No. 383. 43 pp. Jameson, E. W. Jr. 1952. Food of deer mice Peromyscus maniculatus and E, boylei in the northern Sierra Nevada, california. J. Mann. 33:50-60. Jordan, C. F. 1971. Productivity of a tropical forest and its relation to a world pattern of energy storage. J. Ecol. 59:127-142. 85 Kanemasu, E. T. and C. K. Hiebsch. 1975. Net carbon dioxide exchange of wheat, sorghum, and soybean. Can. J. Bot. 53:382-389. Keet, J. D. M. 1936. Report on drift sands in South Africa. For. Ser. Un. S. Afr. No. 9. Kelley, J. M. 1975. Dynamics of root biomass in two eastern Tennessee old-field communities. Am. Midl. Nat. 94:54-61. Kelley, J. M., G. M. Van Dyne and W. F. Harris. 1974. Comparison of three methods of assessing grassland productivity and biomass dynamics. Am. Midl. Nat. 92:357-369. Kisiel, M. 1970. Studies on the ecology of the nematodes inhabiting Ammophila arenaria plant communities on beaches and dunes of Baltic seashores. Akademia Rolnicza. 34:111-150. Kosek, R. E., J. C. Sutton and B. R. Sheppard. 1975. Ecology of Endogone in Lake Huron sand dunes. Can. J. Bot. 53:87-93. Kubien, E. 1970. Embryological studies in Ammophila arenaria (L.) Link. Acta. Biol. Cracov., Ser. Bot. 13:1110. Kucera, C. L., R. C. Dahlman, and M. R. Koelling. 1967. Total net productivity and turnover on an energy basis for tallgrass prairie. Ecology 48:536-541. Kucera, C. L. and D. R. Kirkham. 1971. Soil respiration studies in tallgrass prairie in Missouri. Ecology. 52:912-915. Laing, C. C. 1954. The ecological life history of Ammo hila breviligulata community on Lake Michigan dunes. PE 0. dissertation. Univ. of Chicago. Chicago. 108 pp. Laing, C. C. 1958. Studies in the ecology of Ammophila breviligulata I. Seedling survival and its relation to population increase and dispersal. Bot. Gaz. 119:208-216. Lehotsky, K. 1939. Sand dune fixation in Michigan. J. Forestry. 39:998-1004. Lemon, E., L. H. Allen, Jr. and L. Muller. 1970. Carbon dioxide exchange of a tropical rain forest. Part II. Bio. Science. 20:1054-1059. Malone, C. R. 1968. Determination of peak standing crop biomass of herbaceous shoots by the harvest method. Am. Midl. Nat. 79:429-435. Marchant, R. 1970. The root surface of Ammophila arenaria as a sub- strate for micro-organisms. Trans. Br. Mycol. Soc. 54:479-482. Marshall, J. K. 1965. Corynephorous canescens (L.) P. Beauv. as a model for the Ammophila probTem. J. Ecol. 53:447-463. 86 McCarty, E. C. 1935. Seasonal march of carbohydrates in Elymus ambiguus and Muhlenbepgia_gracilis and their reaction under moderate grazing use. Plant Phys. 10:727-738. McLaughlin, W. T. 1932. Atlantic coastal plain plants in the sand barrens of northwestern Wisconsin. Ecol. Monogr. 2:335-383. McLeod, K. W. 1974. Survival strategy of Ptelea trifoliata during establishment on Lake Michigan sand dunes. Ph.D. dissertation. Michigan State Univ. E. Lansing. 97 pp. Milner, C. and R. E. Hughes. 1970. Methods for the Measurement of the Primary Production of Grasslands. IBP Handbook. No. 6. Blackwell Scientific Publ., Oxford. 70 pp. Milthorpe, F. L. and J. Moorby. 1974. An Introduction to Crop Physiology. Cambridge University Press. New York, N. Y. 202 pp. Mitchell, A. 1966. The role for soil and water conservation in Australia. Proc. Ninth Int. Grassld. Congr. 1:577-581. Monteith, J. L., G. Szeicz and K. Yabuki. 1964. Crop photosynthesis and the flux of carbon dioxide below the canopy. J. Appl. Ecol. L:321-337. Needham, J. G. 1900. Insect drift on the shore of Lake Michigan. Occasional Memoirs of the Chicago Ent. Soc. 1:19-26. Needham, J. G. 1904. The beetle drift on Lake Michigan. Can. Entomol. 36:294-296. Nicolson, T. H. 1960. Mycorrhiza in the gramineae II. Development in different habitats, particularly sand dunes. Trans. Brit. Mycol. Soc. 43:132-145. Odum, E. P. 1960. Organic production and turnover in old field succession. Ecology 41:34-39. Odum, H. T. and R. F. Pigeon (eds.). 1970. A tropical rain forest. A study of irradiation and ecology at El Verde, Puerto Rico. Nat. Tech. Info. Service. Springfield, Va. 1678 pp. Olson, J. S. 1958a. Rates of succession and soil changes on southern Lake Michigan sand dunes. Bot. Gaz. 119:125-170. Olson, J. S. 1958b. Lake Michigan dune development. 1. Wind- velocity profiles. Jour. Geology. 66:254-263. Patton, R. T. 1934. Ecological studies in Victoria. Part III. Coastal sand dunes. Proc. R. Soc. Vict. 47:135-157. Peattie, D. C. 1922. The Atlantic coastal plain element in the flora of the Great Lakes. Rhodora. 24:57-70, 80-88. 87 Ranwell, D. S. 1958. Movement of vegetated sand dunes at Newborough Warren, Angley. J. Ecol. 46:83-100. Ranwell, D. S. 1972. Ecology of Salt Marshes and Sand Dunes. Chapman and Hall, London. 258 pp. Raschke, K., R. O. Firn and M. Pierce. 1975. Stomatal closure in response to xanthoxin and abscisic acid. Planta. 125:149-160. Redmann, R. E. 1975. Production ecology of grassland plant communities in western North Dakota. Ecol. Monogr. 45:83-106. Reiners, W. A. 1968. Carbon dioxide evolution from the floor of three Minnesota forests. Ecology. 49:471-483. Salisbury, E. J. 1934. On the day temperatures of sand dunes in rela- tion to the vegetation at Blakeney Point, Norfolk. Trans. Norfolk Norwich. Nat. Soc. 13:333-355. Salisbury, E. J. 1938. Plants of the sand dune and why they grow \J/ there. Nature, Lond. 141:814-818. Salisbury, E. J. 1952. Downs and dunes, their plant life and its environment. Bell and Sons Ltd. London. 328 pp. Sass, J. E. 1951. Botanical Microtechnique. Iowa State College Press, Ames, Iowa. 228 pp. Scott, D. and W. D. Billings. 1964. Effects of environmental factors on standing crop and productivity of an alpine tundra. Ecol. Mongr. 34:243-270. Seneca, E. D. and A. W. Cooper. 1971. Germination and seedling response to temperature, daylength and salinity by Ammpphila breviligulata from Michigan and North Carolina. Bot. Gaz. 132:203-215. Shelford, V. E. 1907. Preliminary note on the distribution of the tiger beetles (Cicindela) and its relation to plant succession. Biol. Bull. 14:9-14. Shelford, V. E. 1909. Life histories and larval habitats of the tiger beetles (Cicindelidae). Linn. Soc. Jour. Zool. 30:157-184. Shelford, V. E. 1913. Animal Communities in Temperate America. University of Chicago Press. Chicago, 111. 362 pp. Shure, D. J. 1970. Ecological relationships of small mammals in a New Jersey barrier beach habitat. J. Mammal. 51:267-278. Singh, J. S., W. K. Lauenroth and R. K. Steinhorst. 1975. Review and assessment of various techniques for estimating net aerial primary production in grasslands from harvest data. Bot. Rev. 41:181-232. 88 Smalley, A. E. 1960. Energy flow of a salt marsh grasshopper popula- tion. Ecology. 41:672-677. Smoliak, S. 1956. Influence of climatic conditions on forage produc- tion of shortgrass rangeland. J. Range Mgnt. 6:89-91. Snow, L. M. 1902. The microcosm of the drift line. Am. Nat. 36: 855-864. Stoesz, A. D. and R. L. Brown. 1957. Stabilizing sand dunes. in Yearbook of Agriculture 1957, (ed) A. Stefferud. 321-326. Summerhayes, V. S. 1941. The effect of voles (Microtus agrestis) on vegetation. J. Ecol. 29:14-48. ’ Talling, J. F. 1961. Photosynthesis under natural conditions. Ann. Rev. Plant Physiol. 12:133-154. Teetes, G. L. 1973. Phyllophaga crinita: Damage assessment and control in grain sorphum and wheat. J. Econ. Ent. 66:773-776. Treharne, K. J., J. P. Cooper and T. H. Taylor. 1968. Growth response of orchardgrass (Dactylis glpmerata L.) to different light and temperature environments. II. Leaf age and photosynthetic activity. Crop Sci. 8:441-445. Thompson, 0. Q. 1965. Food preferences of the meadow vole (Microtus pennsylvanicus) in relation to habitat affinities. Am. Mi 1. Nat. 74:76-86. United States Weather Bureau. 1973-1975a. Climatological Data for Michigan. Washington, D. C. United States Weather Bureau. 1973-1975b. Local Climatological Data for Grand Rapids, Michigan. Washington, D. C. Voss, E. G. 1972. Michigan Flora, Part I. Gymnosperms and Monocots. Cranbrook Institute of Science, Bloomfield Hills, Michigan. 488 pp. Wassink, E. C. 1959. Efficiency of light energy conversion in plant growth. Plant Physiol. 34:356-361. Waterman, W. G. 1919. Development of root systems under dune conditions. Bot. Gaz. 68:22-53. Waterman, W. G. 1922. Development of plant communities of a sand ridge region in Michigan. Bot. Gax. 74:1-31. Webley, D. M., D. J. Eastwood and C. H. Gimingham. 1952. Development of a soil microflora in relation to plant succession on sand dunes, including the 'rhizosphere' flora associated with colonizing species. J. Ecol. 40:168-178. 89 Weinmann, H. 1940. Storage of root reserves in rhodes grass. Plant Physiol. 15:467-484. Westlake, D. F. 1969. Theoretical aspects of the comparability of productivity data. In C. R. Goldman (ed). Primary productivity in aquatic environments. Men. Inst. Ital. Idrobiol. 18 suppl. Univ. Cal. Press, Berkeley, Cal. p 313-322. Westgate, J. M. 1904. Reclamation of Cape Cod sand dunes. U.S. Dept. of Agric. Bur. Plant Ind. Bull. No. 65. Whitaker, J. 0. Jr. 1963. Food of 120 Peromyscus leucopus from Ithaca, New York. J. Mann. 44:418-419. Whittaker, R. H. 1970. Communities and Ecosystems. Macmillan Company. Toronto, Canada. 162 pp. Wiant, H. V., Jr. 1967a. Influence of temperature on the rate of soil respiration. J. Forestry. 65:489-490. Wiant, H. V., Jr. 1967b. Influence of moisture content on soil respiration. J. Forestry 65:902-903. Wiegert, R. G. and F. C. Evans. 1964. Primary production and the disappearance of dead vegetation on an old-field in south- eastern Michigan. Ecology 45:49-63. Williams, 0. 1962. A technique for studying microtine food habits. J. Mamm. 43:365-368. Wohlrab, G., R. W. Tuveson and C. E. Olmsted. 1963. Fungal populations from early stages of succession in Indiana dune sand. Ecology. 44:734-740. Woledge, J. 1971. The effect of light intensity during growth on the subsequent rate of photosynthesis of leaves of tall fescue (Festuca arundinacea Schreb.). Ann. Bot. 35:311-322. Woodhouse, W. W., Jr. and R. E. Hanes. 1966. Dune stabilization with vegetation on the outer banks of North Carolina. North Carolina University, State College of Agric. and Eng., Soils Information Series. vol. 8. 50 pp. I Woodwell, G. M. and R. H. Whittaker. 1968. Primary production in terrestrial ecosystems. Am. 2001. 8:19-30. Zelitch, I. 1971. Photosynthesis, Photorespiration and Plant Productivity. Acadmeic Press, New York, N. Y. 347 pp. APPENDIX 90 Appendix Table l. A, breviligulata living biomass(]) harvested during 27 July 1973 from 25 plots on the study site. .9 m mu 0):!- 3m :5"? 37 120 123 204 ‘ 104 112 3 5'13 28 178 128 116 107 143 ct.$ A» 19 128 113 114 150 157 £325 (”A10 154 165 121 161 211 8 k 32 o 142 209 113 129 170 W: 25.8 262 202 142 82 22 Distance (m) from the south boundary of the study site (1) dry weight ((3)/m2 91 Appendix Table 2. Living leaf surface area and leaf dry weights used for development of leaf surface area linear prediction equation. Leaf dry weight (9) Leaf surface area (mmz) 407.4 831.1 293.9 590.3 500.5 578.3 296.1 366.5 492.5 584.3 705.5 528.5 1534.0 474.2 1104.0 1248.6 907.2 1368.3 400.3 831.5 537.3 886.2 431.4 593.4 1645.0 d _a (”Mk-hm Voom-a-owoswwoommmooxoooom-boo . o O o I o o o o O O o O O o o o o 0 o o duo—acaci—am—Iao-hooomowmwhwwxoma- N 92 Appendix Table 3. A, breviligulata daily mean harvested dry weights. Date Aboveground Aboveground Total belowground living dead (1) (2) 7/27/73 138 (18) 261 (32) 8/11/73 151 (16) 246 (21) 8/30/73 201 (23) 339 (41 9/17/73 230 (30) 312 (19 10/6/73 194 (16) 318 (33) 10/27/73 183 (18) 338 (30) 11/17/73 6 ( 7) 396 (48) 3/7/74 0 392 51) 795 (47) 4/6/74 5 E 1) 381 33) 749 45) 4/27/74 23 2) 425 37 655 (67) 5/18/74 48 ( 4) 380 27 586 (68) 6/10/74 106 (10) 457 49) 783 (54) 6/28/74 118 ( 8) 373 (34) 778 (50 7/17/74 131 ( 7) 359 (27) 771 (53) 8/9/74 142 ( 7) 382 (29) 701 (75) 9/1/74 162 (13) 415 (31) 623 (64) 9/21/74 183 (18) 437 (44) 886 (74) 10/12/74 159 ( 6) 471 (21) 891 (89) 11/2/74 97 ( 8) 468 (27) 981 (91) 11/23/74 49 ( 4) 545 (39) 925 (86) 3/20/75 0 573 (50 954 (54) 4/22/75 1) 584 (55 944 (61) 5/11/75 2) 502 (41) 970 (63) 6/3/75 (4) 547 (62) 876 (60) 6/24/75 1430(13) 587 (28) 748 (53) 7/11/75 166 17) 634 (65) 955 (51) (1) dry weight (g)/m2 (2) i 1 SE