U MI MICROFILMED 199S INFORMATION TO USERS
This manuscript has been reproduced frommicrofilm the master. UME films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.
The quality of this reproduction Is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margin*, and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6* x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.
A Bell & Howell information Company 300 North Z eeb Road. Ann Arbor. Ml 48106-1346 USA 313.'761*4700 800/521-0600
NUTRIENT DYNAMICS AND RESORPTION IN FOUR UNDERSTORY
WOODLAND PLANTS AND NOTES ON THE MYCORRHIZAL STATUS OF
SOME TYPICALLY NONMYCORRHIZAL PLANTS
DISSERTATION
Presented in Partial Fulfillment of the Requirements for
the Degree of Doctor of Philosophy in the Graduate School
of The Ohio State University
By t
Brent Gordon DeMars, B.A., M.S.
*****
The Ohio State University
1995
Dissertation Committee: Approved by
R. E. J. Boerner
M. G. Cline
P. S. Curtis \ \ J Advisor Department or Plant Biology E. K. Sutherland UMI Number: 9533957
UNI Microform 9533957 Copyright L995* by UMI Company. All rights reserved.
This microform edition is protected against unauthorized copying under Title 17* United States Code.
UMI 300 North Zeeb Road Ann Arbor* MI 48103 To Brittany
ii ACKNOWLEDGEMENTS
I would like to thank Jin Runkle at Wright state
University for all his assistance in the past and present and Ralph Boerner for his guidance and support throughout this work. Thanks go to Jennifer Brinknan for perfoming all the Nitrogen analyses and the OSU Soil Ecology Lab for equipment sharing. I also thank Dr. Folly Penhale of the
Division of Polar Biology and Medicine/NSF for field work in Antarctica. I also thank my family for their support and love throughout my time at Ohio State.
iii VITA
August 30, 1964 ...... Born Lakewood, Ohio
1986 ...... B.A., Wright State University, Dayton, Ohio
1988 ...... M.S., Wright State University, Dayton, Ohio
1991-1995 G.T.A. Department of Plant Biology, Ohio State University, Columbus, Ohio
PUBLICATIONS
DeMars, B.G. and J.R. Runkle. 1992. Groundlayer vegetation ordination and site-factor analysis of the Wright State University woods (Greene county, Ohio). Ohio Journal of Science 92: 98-106.
DeMars, Brent G. 1994. Star-of-Bethlehem, Ornithogalum umbellatum L. (Liliaceae): an invasive, naturalized plant in woodlands of Ohio. Natural Areas Journal 14: 308-309.
DeMars, B. and R. E. J. Boerner. 1994. Vesicular- arbuscular mycorrhizal fungi colonization in Capsella bursa-pastoris (Brassicaceae). American Midland Naturalist 132: 377-380.
FIELD OF STUDY
Major Field: Plant Biology
iv TABLE OF CONTENTS
DEDICATION ...... ii
ACKNOWLEDGMENTS ...... iii
VITA ...... iv
LIST OF TABLES ...... viii
LIST OF FIGURES...... X
CHAPTER PAGE
I. INTRODUCTION...... 1
II. INTRASTAND VARIATIONS IN FOLIAR NUTRIENT DYNAMICS AND RESORPTION IN NATURALIZED LONICERA MAACKII (CAPRIFOLIACEAE) POPULATIONS IN OHIO, USA
Introduction...... 6 Materials and Methods ...... 7 Results...... 13 Discussion...... 34
III. NUTRIENT RESORPTION IN LONICERA MAACKII (CAPRIFOLIACEAE) ALONG A SOUTH-WESTERN OHIO FOREST CHRONOSEQUENCE
Introduction...... 41 Materials and Methods ...... 42 Results...... 47 Discussion...... 55
v IV. FOLIAR PHOSPHORUS AND NITROGEN RESORPTION IN THREE WOODLAND HERBS OF CONTRASTING PHENOLOGY
Introduction...... 58 Materials and Methods ...... 60 Results ...... 67 Discussion ...... 77
V. ARBUSCULAR MYCORRHIZAL DYNAMICS OF THREE WOODLAND HERBS OF CONTRASTING PHENOLOGY
Introduction...... 83 Materials and Methods ...... 84 Results ...... 90 Discussion ...... 100
VI. ARBUSCULAR MYCORRHIZAL FUNGI COLONIZATION IN CAPSELLA BURSA-PASTORIS (BRASSICACEAE)
Introduction...... 106 Materials and Methods ...... 107 Results ...... 109 Discussion ...... 112
VII. ARBUSCULAR MYCORRHIZAL DEVELOPMENT IN THE BRASSICACEAE IN RELATION TO PLANT LIFE SPAN
Introduction ...... 115 Materials and Methods ...... 120 Results ...... 122 Discussion ...... 142
VIII. ARBUSCULAR MYCORRHIZAL DEVELOPMENT IN AN ANNUAL, BIENNIAL, AND PERENNIAL CRUCIFER
Introduction ...... 145 Materials and Methods ...... 146 Results ...... 149 Discussion ...... 157
vi IX. MYCORRHIZAL STATUS OP DESCHAMP5IA ANTARCTICA DESV. IN THE PALMER STATION AREA, ANTARCTICA
Introduction...... 162 Materials and Methods ...... 163
Results ...... 165 Discussion...... 165
. X. SUMMARY OF MAJOR RESULTS AND CONCLUSIONS .. 168
LIST OF REFERENCES ...... 176
vii LIST OF TABLES
TABLE PAGE
1. Soil characteristics of study sites by topographic position ...... 11
2. Mean summer foliar nutrient concentrations in Lonicera maackii by year by site by topographic position ...... 24
3. Analysis of variance of pooled summer (July September) foliar nutrient concentrations ... 25
4. Comparison of foliar nutrient enrichment (demand:availability quotients) ...... 27
5. Comparison of foliar nutrient resorption in Lonicera maackii during 1992-1994 ...... 31
6. Univariate repeated measures analysis of variance for absolute and proportional P and N resorption for YEAR-repeated measures ...... 32
7. P and N foliar proportional resorption reported for temperate deciduous shrubs of forests ... 40
8. Characteristics of the four study stands in the Wright State University woods (Greene County, Ohio) ...... 44
9. Comparison of summer (July-September) mean foliar nutrient concentrations in Lonicera maackii by successional age per y e a r ...... 51
10. Comparison of foliar nutrient resorption in Lonicera maackii for 1992-1994 within the Wright State University wood's chronosequence .... 52
11. Univariate repeated measures analysis of variance of absolute and proportional foliar P and N resorption for YEAR-repeated measures 53
viii 12. Soil characteristics of study sites by topographic position ...... 65
13. Mean foliar P concentrations determined from presenescent sampling periods when leaves were at maximum leaf area ...... 71
14. Overall analysis of variance for absolute and proportional foliar P resorption...... 73
15. Mean foliar N concentrations determined from presenescent sampling periods when leaves were at maximum leaf area ...... 74
16. Overall analysis of variance for absolute and proportional foliar N resorption...... 76
17. Soil characteristic of study sites by topographic position ...... 89
18. Overall ANOVA for percent of root length colonized by AM fungi in Trillium flexipes and Smilacina racemosa ...... 98
19. Percent AM fungi colonization in Caosella bursa- pastoris ...... 110
20. ANOVA of percent root length colonized by AM fungi in Caosella bursa-oastoris ...... ill
21. Crucifer taxa examined for AM development as reported in the literature...... 117
22. The occurrence of AM development in greenhouse- inoculated crucifers ...... 125
23. Brassicaceae taxa not exhibiting AM fungi colonization following experimental inoculation ...... 132
24. Number of plants colonized by mycorrhizal fungi by harvest date ...... 154
25. ANOVA of mean root length colonized by mycorrhizal fungi in 3 crucifer species .... 155
26. Number of plants colonized by mycorrhizal fungi by harvest d a t e ...... 156
27. Summary of proportional P and N resorption . 170
ix LIST OF FIGURES
FIGURE PAGE 1. Seasonal patterns of Lonicera maackii foliar P within the Wright State University woods during.1992-1994 ...... 17
2 * seasonal patterns of Lonicera maackii foliar P within the Central State University woods during.1992-1994 ...... 19
3. Seasonal patterns of Lonicera maackii foliar N within the Wright State University woods during 1992-1994 ...... 21
4. Seasonal patterns of Lonicera maackii foliar N within the Central State University woods during 1992-1994 ...... 23
5. Mean percent root length colonized by arbuscular mycorrhizal fungi in Cardamine concatenata within the Wright State and Central State forests ...... 92
6. Mean percent root length colonized by arbuscular mycorrhizal fungi in Trillium flexipes within the Wright State and Central State forests ...... 94
7. Mean percent root length colonized by arbuscular mycorrhizal fungi in Smilacina racemosa within the Wright State and Central State forests ...... 96
Percent of mycorrhizal species observed in genera with 2 or more species tested in glasshouse inoculation trials ...... 129
x 9. Percent of mycorrhizal species in genera with 7 or more species tested in relation to the number of species per genus tested in glasshouse inoculation trials ...... 131
10. Mean percent root length colonized in three crucifer species: caosella bursa-pastoris (an annual), Hesperia matronalis (biennial), and Matthiola Icana (perennial) and the matrix treatment grass Sorghum sudanense ...... 153
xi CHAPTER I
INTRODUCTION
Foliar nutrient resorption is the process of nutrient retranslocation from senescent leaves into living storage tissues (usually underground) and prior to abscission
(Killingbeck 1986). Observations from canopy trees of temperate deciduous forests indicate that they can withdraw substantial nutrient capital from foliar tissues during resorption (e.g. Boerner 1984) while understory trees and shrubs usually have lower resorption rates (e.g.
Ralhan and Singh 1987; Zimka and Stachurski 1992;
Minoletti and Boerner 1994), and forest herbs, still lower rates (Boerner 1986). Lower resorption rates in understory plants probably reflect the adaptive syndrome of shade tolerance. As resorption is an energy requiring processes and as the photosynthetic rates of forest understory shrubs (Sparling 1967) and herbs (Sparling
1967; Taylor and Pearcy 1976) are limited by low light intensity, it is probable that insufficient carbon reserves are available to meet the respiration (ATP production) demands of high resorption activity in these
1 2
plants. However, this may not be the case for spring
ephemeral herbs, which have high photosynthetic rates
during the period before canopy leaf out (Taylor and
Pearcy 1976).
Factors other than light may also potentially
influence nutrient resorption. Models of plant nutrient
dynamics suggest that individuals growing in less fertile
sites may have greater nutrient resorption than those
growing in more fertile sites (Chapin 1980; Vitousek 1982;
Shaver and Melillo 1984). Several studies examining
foliar nutrient resorption among temperate deciduous
forest stands support such models for both phosphorus and
nitrogen (Stachurski and Zimka 1975; Zimka and Stachurski
1976; Boerner 1984; Host and Boerner 1985). However,
environmental stresses such as lowered moisture
availability (del Arco fi£ al. 1991; Minoletti and Boerner
1994) may negatively impact this process, especially for
nitrogen (Boerner 1985a). Because moisture and nutrients
vary with topographic position and aspect within a forest
(Whittaker 1956), examination of nutrient dynamics along
intrastand topographic gradients can be used to test hypotheses related to the roles of soil moisture and soil nutrient availability in controlling nutrient resorption.
In this document, hypotheses regarding the nutrient resorption behavior of forest understory plants are examined in chapters 2 through 4. In chapter 2, observations of foliar P and N resorption are reported for the understory shrub Lonicera maackii along contiguous intrastand topographic gradients with varying nutrient availability and moisture. Chapter III reports the results of resorption behavior in this shrub along a successional gradient among stands of several ages within a southwestern Ohio forest. Then, in chapter IV, observations of the nutrient resorption behavior of three forest herbs of contrasting phenology along topographic gradients are reported.
The second component of this document represents research projects involving arbuscular mycorrhizal (AM) associations. In chapter V, I report on the AM dynamics of the three forest herbs along the topographic gradients.
The purpose of this study was to determine if their life history characteristics correlated with any significant differences in AM colonization patterns and to test the hypothesis that AM development would be lower in the moister, bottomland topographic positions, as has previously been shown with spore counts (Anderson al.
1984). Additionally, the hypothesis that AM development would be lower in sampling areas with the highest P availability was tested. * *
4
In the remaining chapters of the document, I examine
AM dynamics in herbaceous plants that have been
traditionally classified as nonmycorrizal (Gerdemann 1968;
Trappe 1987). Chapter VI reports observations of AM
mycorrhizal fungi colonization in the crucifer Capsella
bursa-pastoris from three Ohio sites, each of which where
subdivided into lawn and disturbed habitats. The primary
hypothesis tested in this project was that intact lawn
habitat should support plants with greater AM fungi
colonization than disturbed habitat where fungal inocula
would be low (Miller &1. 1983).
In chapter VII, I examine development of AM in
649 taxa of Brassicaceae using glasshouse inoculation
trials. It has been suggested that life history patterns
may be correlated with AM development in typically
nonmycorrhizal families, and, since many of the first
crucifers examined for AM development were annuals of
early successional and disturbance communities, the
characterization of the Brassicaceae as nonmycorrhizal may
be unwarranted because such communities often have low
mycorrhizal inoculum potential (Moorman and Reeves 1979;
Janos 1980; Biondini fit fll. 1985; Allen 1991). More
specifically, here, I tested the hypothesis that longer-
0 lived perennial crucifers would develop AM at a greater
frequency than shorter-lived annuals and monocarpic perennials (biennials). In chapter VIII, I explored the dynamics of AM development in crucifers over time. Here,
I performed glasshouse inoculation studies on three crucifer species: an annual, a biennial, and a perennial with the intent of examining life history variation.
Finallyj in chapter IX, I report the results of an examination for AM development in another typically nonmycoirhizal species, Deschamosla antarctica. the native grass of Antarctica. Here, both In situ and glasshouse inoculated samples were examined. CHAPTER II
Intrastand Variations in Foliar Nutrient Dynamics
and Resorption in Naturalized Lonicera maackii
(Caprifoliaceae) Populations in Ohio, USA
INTRODUCTION
Models of plant nutrient dynamics suggest that individuals growing in less fertile sites may have greater nutrient use efficiency than those growing in more fertile sites (Chapin 1980; Vitousek 1982; Shaver and Melillo
1984). One measure of nutrient use efficiency in plants is foliar resorption, the process of nutrient translocation from the leaves into storage tissues during senescence and prior to abscission (Killingbeck 1986).
Several studies examining foliar nutrient resorption among temperate deciduous forest stands support such models for nitrogen and phosphorus (Stachurski and Zimka 1975; Zimka and Stachurski 1976; Boerner 1984; Kost and Boerner 1985).
However, environmental stresses such as lowered moisture availability (del Arco &1. 1991; Minoletti and Boerner
1994) and light limitations (Boerner 1986) may negatively impact this process, especially for nitrogen (Boerner
6 1985a).
Because moisture and nutrients vary with topographic
position and aspect within a forest (Whittaker 1956),
examination of nutrient dynamics along intrastand
topographic gradients can be used to test hypotheses
related to the roles of soil moisture and fertility in
nutrient dynamics and conservation. In this study,
nutrient dynamics and nutrient resorption were studied in populations of the non-native understory shrub, Lonicera maackii (Rupr.) Maxim, along contiguous topographic gradients in two mesic deciduous forests. The working hypotheses were 1) that foliar nutrient concentrations would be directly proportional to nutrient availability and 2) that nutrient resorption would inversely proportional to nutrient availability.
MATERIALS and METHODS
Study Species— Lonicera maackii (Caprifoliaceae) is a multi-stemmed deciduous shrub native to northeastern Asia
(Luken and Mattimiro 1991). It is now naturalized in many regions of Canada and the United States (Pringle 1973).
In south central and south-western Ohio the shrub initially invades woodlands (Luken 1988) through bird dispersal (Luken and Thieret 1987), then spreads rapidly by vegetative reproduction. L- maackii often forms a dense, conspicuous shrub layer within the understory of a woodland. In south-western, Ohio the shrub breaks bud in early or mid-April, reaching full leaf expansion by the end of June. While many other woodland shrubs senesce by
September or early October, L* maackii. remains photosynthetic through November, and senescence is often not completed until early to mid-December.
Sample Sites— Sampling was conducted in two forests located in Greene County, Ohio. The first site was the
Wright State University (WSU) woods (84°03'W 39°45'N) in
Bath Township. The site was dominated by Acer saccharum
Harsh., Ouercus rubra L., and Quercus alba L. (DeMars and
Runkle 1992). Common associates included Ulmus amerlcana
L., Carya spp. and Fraxlnus americana L. On the ESE slope we sampled, £. amerlcana and £. alba were more abundant in the uplands, whereas in bottomlands Jualans nigra L. and a. americana were more abundant.
Elevation in the WSU site ranged from 282 m in the uplands to 260 m in the bottomland site, which in turn, was approximately 1.5 m above the normal flow level of the bordering stream. The soils of the WSU forest were
Miamian silt loams, which were well drained in upland and slope positions and moderately drained in the bottomlands
(Garner et al. 1978). These soils were formed on shale and limestone bedrock.
The second site was a forest located 0.6 km west of
Central State University (CSU) in Xenia Township (83°05'W
39°42'N). The site was dominated by Acer saccharum and
Ouercus rubra with the same common associates as the WSU
site. Elevation ranged from 305 m in the uplands to 278 m
in the bottomlands. The CSU soils were Miamian-Hennepin
series soils which were formed on a calcareous bedrock,
and were both shallower and steeper in slope than those of
WSU (Garner fit fll. 1978)
Soil nutrient availability varied among topographic
positions within these sites but not between forest sites
(Table 1). Overall there was greater extractable soil N
in the bottomland topographic positions than in upland or
mid-slope positions, and greater extractable soil P in
uplands and mid-slope positions than in bottomland
positions (Table 1). Moisture gradients based on
gravimetric determinations result from higher percent water content in bottomland soils than in upland or mid
slope position soils (Table 1).
The climate of Greene County is continental (Miller
1969). Annual precipitation (Xenia, OH station) in 1992,
1993, and 1994 was 96, 101, and 106 cm, respectively, which was within the normal 30-yr range. Mean January and
July temperatures were -0.4° C / 23.3° C, 0.9° C / 25.5° C, and 0.6 C / 25.3° for 1992-1994, respectively (NOAA 1992,
1993; 1994 data unpublished).
Field Methods— Within each study site, one 20 m x 20 m plot was established per topographic position (upland, mid-slope and bottomland) along a contiguous topographic gradient (ESE facing slopes). In April 1992, five
Lonicera maackii individuals (> 1.4 m tall) were randomly selected and flagged in each plot. Individuals were only selected if they were at least 3 m from any neighboring canopy tree stem to avoid potential microsite variation due to canopy tree proximity (Boerner and Koslowsky 1989).
Five fresh leaves were harvested from each individual at monthly intervals from April through November during
1992-1994. To avoid crown position affect, leaves were sampled from throughout the crown of each individual.
Freshly abscised litter (10 leaves) was collected under each individual in early December of each year. Table l.— Soil characteristics of study sites by topographic position. Means followed by the sane lower case letter were not significantly different following analysis of variance. Standard errors of the nean for extractable N are given in parentheses. Data are adapted from DeMars and Runkle (1992) and DeMars (unpublished).
Parameter WSU-Dp WSU-Slope WSU-Bottom CStf-Op CSU-Slope CSU-Bottora Extractable Nl (mg N/Kg soil) 7.5a (0.2) 8.Sab (0.4) 12.4c (0.3) 6.9a (0.1) 7.9a (0.3) 11.9c (0.6) Extractable P1 (mg P/Kg soil) 3.6b (0.2) 3.6b (0.7) 2.2a (0.3) 3.lab (0.7) 3.2b (0.5) 1.9a (0.5) Moisture (% water content) 21.2a (1.7)* 21.8a (0.9) 25.3b (0.6) 20.4a (1.5) 20.0a (1.8) 26.0b (0.6) pH 6.02 6.66 6.80 6.57 6.63 7.12
1 Data are from 3 soil samples per site x topographic position for April-November 1992-1994, extracted in 2M KCl (Keeney and Kelson 1982). 1 Data are from 3 composited samples per site x topographic position for April-November 1992-1994, extracted in IK ammonium acetate (Chapman 1965). 12
Laboratory Methods— In the laboratory, leaf area was
measured for each individual sampled using a LI-3100 leaf
area meter (Li-Cor, Inc., Lincoln, Nebraska). Leaves were dried at 70° C for 48-72 h and subsequently digested in
30% Hj02 and concentrated H2S04 for later N and P analyses
(Technicon 1977). Foliar tissue P concentrations were analyzed using the stannous chloride method (A.P.H.A.
1975) and foliar N concentrations were determined by autoanalyzer (QuikChem AE, Lachat). Reliability of N and
P digestions was evaluated by digesting and analyzing
U.S. National Bureau of Standards, standard pine needle material #1585 (National Bureau of Standards 1981).
Foliar nutrient concentrations were calculated on a leaf area basis (mg/dm2 leaf area) for both P and N.
Because foliar nutrient concentrations changed little during the summer, maximum foliar nutrient concentrations were determined from pooled summer (July-September) means.
Absolute resorption was calculated as the difference between each plant's mean summer concentrations and litter foliar concentrations (which occurred in fresh litter for all individuals). Proportional resorption was calculated as the percentage of the mean summer concentration that was actually resorbed prior to abscission. Foliar nutrient enrichment ratios were calculated to estimate of plant nutrient uptake efficiency (Boerner 1986). These 13 values represent the ratio of leaf concentration to soil nutrient availability (Killingbeck and Costigan 1988).
To examine the effects of site and topographic position on nutrient concentrations, summer foliar nutrient concentrations were analyzed by two-way analysis of variance (ANOVA) by sampling year (Statistical Analysis
System 1985). Tukey's studentized multiple range test was used to determine significant differences among means.
Nutrient resorption was analyzed with repeated measures analysis of variance (RMANOVA) because use of the year term in standard ANOVA would violate the requirement of independence among sampling units (Moser al. 1990;
Gumpertz and Brownie 1993). RMANOVA was performed in univariate mode since none of the sphericity tests of repeated measures variance-covariance matrices were rejected (MoBer &1. 1990).
RESULTS
Foliar Nutrient Concentrations— Foliar P concentrations in
Lonicera maackii exhibited marked seasonal variation in both sites and topographic positions during the course of
1992-1994 (Figures 1, 2). In most cases, leaves reached full expansion by the June sampling date, and this was followed by relatively stable foliar P concentrations until senescence in November and early December of each 14 year.
Pooled summer P concentrations ranged from 1.24 mg/dm2 in the WSU bottomlands in 1992 to 1.59 mg/dm2 in the WSU uplands in 1994 (Table 2). Overall, there was a significant effect of topographic position (Table 3), with uplands exhibiting the greatest foliar P concentrations and bottomlands exhibiting the least for each of the years sampled. Significant site differences in summer foliar P concentrations were also observed in 1993 and 1994, but not 1992 (Table 3).
Similarly, N foliar concentrations varied significantly over each growing season (Figures 3, 4).
However, unlike P foliar concentrations, N foliar concentrations tended to decrease consistently from leafout through later months. Additionally, summer foliar
N concentrations were much more similar between sites and years and among topographic positions than were P concentrations (Table 2). Mean summer foliar N ranged only from 22.08 mg/dm2 in the WSU bottomland in 1994 to
24.61 mg/dm2 in the WSU slope in 1994 (Table 2). Overall, there were no significant site effects in the analysis of variances for summer foliar N concentrations (Table 3); however, there was a significant topographic position effect in 1994, as Lonicera maackii plants on slopes had greater foliar N concentrations than plants in bottomlands 15
(Table 3).
Lonicera maackii plants from the relatively low P bottomland topographic positions concentrated P to a greater extent than did the plants from other topographic positions (Table 4). Similarly, plants from the lower N uplands and slopes concentrated more N in their foliar tissues than did the relatively N-rich bottomland plants
(Table 4). Figure 1. Seasonal patterns of Lonicera maackii foliar P
(mg P/dm1 leaf area) within the Wright State University woods during 1992-1994. April and May concentrations are excluded. Standard errors of the mean are represented by vertical bars; where no vertical bars appear, the standard errors were too small to exceed the symbol.
16 FOLIAR PHOSPHORUS (mg/dm ) iue SML MONTH SAMPLE 1 Figure J JASOND J JASONDJ JASOND WSU—LOWLAND 1992 WSU—LOWLAND WSU—UPLAND 1992 WSU—UPLAND WSU—SLOPE 1992 WSU—SLOPE I WSU—LOWLAND 1993 WSU—LOWLAND WSU—UPLAND 1993 WSU—UPLAND WSU—SLOPE 1993 WSU—SLOPE WSU—LOWLAND 1994 WSU—LOWLAND WSU—UPLAND 1994 WSU—UPLAND WSU-SLOPE 1994WSU-SLOPE Figure 2. Seasonal patterns of Lonicera maackii foliar P
(mg P/dma leaf area) within the Central state University woods during 1992-1994. April and May concentrations are excluded. Standard errors of the mean are represented by vertical bars; where no vertical bars appear, the standard errors were too small to exceed the symbol.
18 FOLIAR PHOSPHORUS (mg/dm ) Figure 2 Figure 2 2 2 l 1 1
i i i i i i i i i i i i i i i 1 ! t 1 t 1 ! I 1 1 J J A S O N D J J A S O N D J J A S O N D ------CSU—LOWLAND 1992CSU—LOWLAND CSU—UPLAND 1992 CSU—UPLAND • CSU—SLOPE 1992CSU—SLOPE -- i i i
i t i i i i t i i j ' iiiiiti i i i i i i i i i i • -- SAMPLE MONTH SAMPLE • -- CSU—LOWLAND 1993 CSU—LOWLAND CSU—UPLAND 1993 CSU—UPLAND i ‘ CSU—SLOPE 1993CSU—SLOPE -- • -- • -- • \ iririri (itiiti ii.i i i i i CSU—LOWLAND 1994CSU—LOWLAND CSU—UPLAND 1994CSU—UPLAND CSU—SLOPE 1994CSU—SLOPE Figure 3. Seasonal patterns of Lonicera maackii foliar N
(mg N/draJ leaf area) within the Wright State University woods during 1992-1994. April and May concentrations are excluded, standard errors of the mean are represented by vertical bars; where no vertical bars appear, the standard errors were too small to exceed the symbol.
20 FOLIAR NITROGEN (mg/dm ) Figure 3 Figure 30 - 30 30 40 40 40 20 20 J JASONDJ JASONDJ J A S O N D r WSU—LOWLAND 1992 WSU—LOWLAND WSU-UPLAND 1992 WSU-UPLAND WSU—SLOPE 1992 WSU—SLOPE i i l SAMPLE SAMPLE MONTH j. WSU—LOWLAND 1993 WSU—LOWLAND WSU-UPLAND 1993 WSU-UPLAND WSU—SLOPE 1993 WSU—SLOPE i i i 1 1 i I I 1 l -I I i WSU—LOWLAND 1994 WSU—LOWLAND WSU-UPLAND 1994 WSU-UPLAND L J _ WSU—SLOPE 1994WSU—SLOPE i 1 __ Figure 4. Seasonal patterns of Lonicera naackil foliar N
(mg N/dm3 leaf area) within the Central State University woods during 1992-1994. April and Hay concentrations are excluded. Standard errors of the mean are represented by vertical bars; where no vertical bars appear, the standard errors were too small to exceed the symbol.
22 CSU—UPLAND 1992 CSU—UPLAND 1993 CSU—UPLAND 1994 40
30 20 i V _ % i i i i i i t r i i t i i t i i i i i i r CSU—SLOPE 1992 CSU—SLOPE 1993 CSU— SLOPE 1994 40
30
20 % i i i i i t i 1 1 1 t 1 f 1 i CSU—LOWLAND 1992 CSU—LOWLAND 1993 CSU-LOWLAND 1994 40
30
20
i i i i i i « i i i _ i i i i iiiiiii J JASOND J JASONDJ JASOND * 4 SAMPLE MONTH Table 2.--Mean summer foliar nutrient concentrations in Lonicera maackii (mg/dm* leaf area) by year by site by topographic position. Means were calculated from post-leafout and presenescent months (July- September). Standard errors of the means are given in parentheses. Means followed by the same lower case letter within a row were not significantly different at £ > 0.05.
site Fear Upland Slope Bottomland Foliar P WSU 1992 1.46a (0.03) 1.38a (0.02) 1.24b (0.03) 1993 1.39a (0.03) 1.34a (0.02) 1.33a (0.03) 1994 1.59a (0.03) 1.56a (0.05) 1.37b (0.02) CSU 1992 1.55a (0.04) 1.40ab (0.06) 1.29b (0.02) 1993 1.42a (0.02) 1.44a (0.03) 1.34b (0.01) 1994 1.42a (0.03) 1.43a (0.03) 1.45a (0.03)
r9liW.IT WSU 1992 23.62a (0.43) 23.19a (0.42) 23.58a (0.45) 1993 23.51a (0.45) 24.25a (0.43) 23.57a (0.62) 1994 23.77a (0.76) 24.61ab (0.29) 22.08b (0.24) CSU 1992 23.62a (0.82) 22.57a (0.11) 22.79a (0.39) 1993 23.55a (0.92) 22.26a (0.17) 22.62a (0.45) 1994 23.23a (0.97) 24.02ab (0.46) 22.93b (0.47) Table 3.— Analysis of variance of pooled summer (July-September) foliar nutrient concentrations (mg/dm3 leaf area). N=15 per site per topographic position.
Source DF MS £ £ > £ Seasonal Foliar P Concentrations 1992 site 1 0.06 3.36 n.s. Topographic position1 2 0.42 22.65 0.0001 Site‘Topographic Position 2 0.01 0.37 n.s.
1993 site* 1 0.05 5.26 0.05 Topographic position1 2 0.04 4.67 0.05 Site‘Topographic Position 2 0.02 1.75 n.s.
1994 Site4 1 0.13 9.03 0.01 Topographic position3 2 0.08 5.00 0.01 Site‘Topographic Position 2 0.14 9.75 0.001
Seasonal Foliar W Concentrations 1992 site 1 4.24 1.24 n.s. Topographic position 2 . 4.20 1.24 n.s.
Site*Topographic Position 2 1.05 0.30 n.B.
M Ul Table 3. (Continued)
1993 Site 1 15.99 3.47 n.s. Topographic position 2 5.70 0.59 n.s. Site*Topographic Position 1 8.07 1.75 n.s.
1994 Site 1 1.53 0.29 n.s. Topographic position4 2 18.16 3.46 0.05 Site*Topographic Position 2 2.04 0.39 n.s.
* Significant differences: uplands > slopes > bottomlands a Significant differences: HSU > CSU 1 Significant differences: uplands > slopes > bottomlands 4 Significant differences: WSU > CSU * Significant differences: uplands “ slopes > bottomlands 4 Significant differences: slopes > bottomlands > uplands Table 4.— Comparison of foliar nutrient enrichment (demand:availability quotients) calculated as the ratio between pooled summer nutrient concentrations (July-September) to mean soil nutrient availability. Means followed by the same lower case letter within a row were not significantly different at £ > 0.05.
Site Year Upland Slope Bottomland Kean Foliar P Enrichment WSU 1992 4.01a (0.04) 3.98a (0.09) 5.74b (0.17) 1993 3.78a (0.05) 3.80a (0.04) 6.27b (0.14) 1994 4.25a (0.09) 4.19a (0.13) 6.30b (0.04) CSU 1992 4.88a (0.13) 4.62a (0.11) 6.87b (0.07) 1993 4.56a (0.01) 4.49a (0.07) 7.07b (0.04) 1994 4.46a (0.06) 4.27a (0.11) 7.42b (0.05) Kean Foliar N Enrichment WSU 1992 2.87a (0.05) 2.57b (0.06) 1.89c (0.02) 1993 2.89a (0.01) 2.59b (0.01) 1.76c (0.02) 1994 2.86a (0.02) 2.67b (0.03) 1.80c (0.01)
CSU 1992 2.98a (0.01) 2.81b (0.01) 1.82c (0.01) 1993 2.97a (0.01) 2.78b (0.03) 1.83c (0.01) 1994 2.94a (0.02) 2.83b (0.01) 1.84c (0.01) Nutrient Resorption— Mean absolute P resorption ranged
from 0.36 mg/dm3 in the WSU slope position during 1993 to
0.80 mg/dm3 in the CSU bottomland position in 1994 (Table
5). Repeated measures analysis of variance (RMANOVA)
indicated highly significant topographic position and year
first order effects, but no significant site effect (Table
6). For the significant year effect both linear and
quadratic contrasts were significant; however, the
quadratic contrast was highly significant (£ > o.oool)
indicating a complex interyear covariance relationship.
The topographic position effect represents an overall
absolute P resorption rate one and one half times greater
in bottomland positions (0.80 ±0.03) than in either
upland (0.48 ± 0.03) or slope positions (0.48 ± 0.04).
The significant year effect stems from greater overall
absolute P resorption in 1994 (0.62 ± 0.04) than in either
1992 (0.54 ± 0.03) or 1993 (0.54 ± 0.03). In addition to
these significant first order effects, the site x year and the third order interactions were significant. The significant site x year interaction reflects the WSU site's larger increase in foliar P resorption from 1992 and 1993 to 1994 than in the CSU site (Table 6).
Mean proportional P resorption ranged from 26.3% ±
1.4 in the WSU upland position in 1992 to 56.9% ± 1.5 in the CSU bottomland site in 1992 (Table 5). RMANOVA indicated the same significant effects for proportional P resorption as for absolute P resorption (Table 6).
Similar to absolute P resorption, the topographic position effect resulted from a greater overall proportional resorption in the bottomland positions (53.1% + 1.8) relative to upland (36.3% ± 1.9) and slope positions
(33.7% + 2.7). Again, greater resorption occurring in
1994 (42.9% +2.5) and lower resorption occurring in 1992 (40.0% ± 2.1) and 1993 (38.1% + 1.9) reflected the significant year effect. For proportional P resorption all interactions except the site x topographic position interaction were significant.
Absolute foliar N resorption ranged from 6.33 + 0.33 mg/dm3 in the CSU upland position during 1994 to 9.47 ±
0.30 mg/dm3 in the WSU slope position during 1994 (Table
5). RMANOVA indicated both topographic and year significant first order effects and significant interactions except for the site x year effect (Table 6).
Unlike absolute P resorption, the significant topographic effect for absolute N resorption resulted from greater overall mean resorption occurring in slope positions (8.41
+ 0.28) than in either uplands (6.88 + 0.33) or bottomlands (6.82 + 0.26). The significant year effect stemmed from lower overall mean absolute P resorption during 1993 (7.19 + 0.37) than in 1992 (7.48 + 0.23) or 30
X994 (7.43 ± 0.27). The significant site x topographic position reflected, in part, greater mean resorption in the WSU's slope position than in CSU's slope position, whereas the significant site x topographic position effect reflected overall greater resorption occurring in the WSU site for all topographic positions (Table 5).
Proportional foliar N resorption ranged from 28.8% ±
1.2 to 40.4% ± 1.2 in the WSU's bottomland and slope positions during 1994 (Table 5). RMANOVA indicated a highly significant topographic position effect, however, unlike absolute P and N and proportional N resorption the year effect was not significant (Table 6). As for absolute N resorption, the significant topographic position effect for proportional N resorption reflected greater resorption in the slope positions (37.3% ± 1.1) than in uplands (32.8% + 1.4) or bottomlands (30.7% +
1.0). Table 5.--Comparison of foliar nutrient resorption in Lonicera maackii during 1992-1994. Means of both absolute resorption (mg/dm3 leaf area) and proportional resorption are given for N = 5 individuals by site by topographic position by year, standard errors of the means are given in parentheses. Across rows, means followed by the same lower case letter were not significantly different at £ < 0.05. tSSSS888S8881M8888SSSZ888S89S8S: 88888888888888 Site Year Upland Slope Bottomland Absolute Foliar P Resorption WSU 1992 0.37a 0.02) 0.47b (0.05) 0.64b (0.04) 1993 0.41a 0.04) 0.36a (0.03) 0.73b (0.03) 1994 0.64a 0.04) 0.59a (0.06) 0.69a (0.04) CSU 1992 0.51a 0.03) 0.47a (0.04) 0.75b (0.02) 1993 0.45a 0.02) 0.51a (0.03) 0.69b (0.02) 1994 0.51a 0.04) 0.47a (0.04) 0.80b (0.01) Proportional Foliar P Resorption WSU 1992 26.3a 1.8) 33.3a (2.8) 51.1b (2.8) 1993 29.3a 2.4) 26.8a (2.2) 53.1b (1.4) 1994 41.7a 2.2) 39.8a (3.8) 49.8b (3.0)
CSU 1992 33.2a 1.8) 33.1a (2.1) 56.9b (1.5) 1993 32.0a 1.6) 35.4a (2.0) 51.6b (1.7) 1994 36.3a 2.2) 33.5a (3.1) 56.0b (0.7) Absolute Foliar N Resorption WSU 1992 6.75a 0.54) 8.24a (0.30) 8.02a (0.66) 1993 6.84a 0.22) 8.45b (0.29) 6.34a (0.20) 1994 7.75b 0.35) 9.47c (0.30) 6.39a (0.27)
CSU 1992 6.70a 0.35) 8.24b (0.27) 6.95b (0.11) 1993 6.93a 0.17) 7.99b (0.33) 6.59a (0.18) 1994 6.33a 0.33) 8.04b (0.19) 6.60a (0.15) Proportional Follag_W_Resorptlon WSU 1992 31.3a 2.1) 36.5a (1.0) 34.2a (2.4) 1993 31.7a 1.1) 37.2b (1.3) 29.1a (0.8) 1994 36.2b 1.5) 40.4b (1.2) 28.8a (1.2)
CSU 1992 32.5a 1.7) 37.2b (1.3) 31.3a (0.4) 1993 33.8b 0.7) 36.5b (1.1) 30.4a (0.8) 1994 31.3a 1.5) 36.1b (0.9) 30.1a (0.6) h» Table 6.— Univariate repeated measures analysis of variance for absolute and proportional foliar P and N resorption for YEAR-repeated measures. BBn&8Kiss3fiEXBsssBiiBB8Biss±BnssssBn8SBnas9aaasssssaa=3S5S8898«aas3aaaassns83a3Ei3ass ABSOLUTE P RESORPTION Source d£ MS z £ > £ Site 1 0.01 0.57 n.s Topographic Position 2 0.38 34.61 0.0001 Site x Topographic Position 2 0.02 2.07 n.s. Plant 24 0.01 Year 2 0.12 18.44 0.0001 Linear 1 0.05 0.05 0.05 Quadratic 1 0.48 28.35 0.0001 Site x Year 2 0.04 6.52 0.01 Topographic Position x Year 4 0.01 1.77 n.s. Site x Topographic Position x Year 4 0.04 6.50 0.001 Error 48 0.01
PROPORTIONAL P RESORPTION Source df MS £ P > £ Site 1 27.13 0.63 n# s • Topographic Position 2 3070.74 71.51 0.0001 site x Topographic Position 2 54.18 1.26 n.s. Plant 24 42.94 Year 2 270.68 10.46 0.001 Linear * 301.75 6.92 0.05 Quadratic 1 1081.61 17.54 0.05 Site x Year 2 89.27 3.45 0.05 Topographic Position x Year 4 66.99 2.59 0.05 Site x Topographic Position x Year 4 107.82 4.17 0.01 Error 48 25.87 Table 6.— (Continued)
ABSOLUTE N RESORPTION Source df MS Z £ > £ Site 1 0.57 0.68 n.s. Topographic Position 2 34.25 40.72 0.0001 site x Topographic Position 2 3.55 4.22 0.05 Plant 24 0.84 Year 2 1.42 3.37 0.05 Linear 1 0.24 0.26 n.s. Quadratic 1 3.16 4.47 0.05 Site x Year 2 0.20 0.47 n. s. Topographic Position x Year 4 2.88 6.81 0.001 Site x Topographic Position x Year 4 2.80 6.63 0.001 Error 48 0.42
PROPORTIONAL N RESORPTION Source df MS E £ > Z Site 1 0.75 0.06 n.s Topographic Position 2 501.10 42.71 0.0001 Site x Topographic Position 2 34.21 2.92 n.s. Plant 24 11.71 Year 2 11.04 2.00 n.s. Linear 1 2.20 0.18 n.s. Quadratic 1 23.70 2.74 n.s. Site x Year 2 3.75 0.68 n.s. Topographic Position x Year 4 24.11 4.36 0.01 site x Topographic Position x Year 4 32.00 5.79 0.001 Error 48 5.53 DISCUSSION
The first working hypothesis that foliar nutrient concentrations in Lonicera maackii would be directly proportional to soil fertility is generally supported by our analyses for foliar P but not N. maackii plants growing in upland and slope positions had higher overall mean foliar P than bottomland topographic positions.
Other studies reported mixed results for foliar P, with data from Ouercus prinus L. and Ouercus alba L. (Boerner
1984) and the understory tree Hammamelis viroinlana L.
(Boerner 1985a) supporting the direct proportionality hypothesis, but data from Acer rubrum L. or Faous grandifolia Ehrh. (Boerner 1984), the understory tree
Cornus florida L. (Host and Boerner 1985), and the shrub
Viburnum acerifolium L. (Minoletti and Boerner 1994) failing to support it.
In contrast, our data do not support the direct proportionality hypothesis for foliar N. There were no significant overall effects of topography on foliar N concentration, despite significant topographic differences in soil N availability. These data stand in contrast with studies which have shown higher foliar N (Boerner 1985b) in more fertile sites and studies which have shown lower foliar N among sites (Kost and Boerner 1985; Minoletti and
Boerner 1994). 35
The lack of large differences among topographic positions may have resulted from carbon dilution effects
(Chapin 1980) in the more N-fertile bottomland positions.
Increased fertility (along with greater light penetration) may result in higher growth rates, and consequently, greater foliar carbon gain. The lack of significant differences in our study sites may also reflect the narrower, intrastand fertility gradients among topographic positions in this study's sampling sites as opposed to the wider, interstand gradients in fertility in other studies.
Patterns of foliar nutrient resorption in Lonicera maackii support our second working hypothesis that foliar resorption would be greater in lower fertility sites. For
P, both absolute and proportional resorption were significantly greater in the bottomland topographic positions where soil P was lowest. This is consistent with P resorption in other temperate, deciduous understory woody plants including Cornus florida (Kost and Boerner
1985) and Hammaraells virginiana (Boerner 1985a) among sites of varying soil-P fertility, however, not with
Viburnum acerifolium (Minoletti and Boerner 1994) and other shrubs (Killingbeck and Costigan 1988).
The higher resorption of P in bottomland sites may, in part, reflect the means by which Lonicera maackii obtains P. In the Caprifoliaceae, arbuscular mycorrhizae are used to absorb soil P (Trappe 1987). We have shown
that there are significantly lower levels of mycorrhizal
colonization in herbaceous understory plants in the
moister bottomlands of the study sites (DeMars and
Boerner, unpublished). Hence, the combination of low soil
P availability and lower mycorrhizal development probably
necessitated greater P conservation.
Resorption rates for foliar N did not support the
second hypothesis as clearly. Absolute and proportional
resorption were greatest in the slope topographic
positions which had intermediate levels of soil In
fertility. This outcome may have reflected drier
conditions of these ESE facing slopes due to exposure and
more rapid drainage. Much data indicates that nutrient
resorption rates are affected by soil moisture status
{Hocking 1982; Host and Boerner 1985; del Arco s£ al.
1991). For instance, Minoletti and Boerner (1994) showed
that N resorption in Viburnum acerifolium was lower during
a drought year than for an average precipitation year.
This pattern is consistent with N resorption rates in this
study as plants growing on slopes, which were consistently
driest, had greater resorption than the plants from other
topographic positions. The importance of moisture differences among topographic positions in this study, however, remains speculative as there were no 37
collaborative reductions in foliar N tissue concentrations
associated with slopes. This would be expected because
moisture availability affects mass flow of soil N (Barber
1984).
The significant year effect is difficult to explain.
Annual precipitation measured at the Xenia, Ohio station
was within the 30-year normal range for each year.
However, monthly values varied more. There were no
consistent patterns between precipitation values and
resorption. In fact, 1994 was the driest autumn of the 3
years, yet, overall P resorption was greatest during this
time and N resorption was lowest. The lower N resorption may support the hypothesis that resorption is negatively
impacted by lower moisture, but this is not the case for P
resorption. Probably, the variations in resorption would
best be explained by microclimatic and microsite variations in the sampled plants. However, moisture was not specifically measured in the soils underneath each
individual.
Another factor which may have impacted resorption was the continued defoliation of individuals. Nay and
Killingbeck (1992) showed that prevention of resorption by presenescent defoliation significantly reduced P resorption in subsequent years, but did not alter N resorption. If this were applicable to the present study 38 a decline in P resorption should be observed from 1992-
1994. However, this was not observed.
The magnitude of differences foliar resorption rates were consistently less in this study than those reported in others. For example, Boerner (1984) reported a range of 26.5% to 63.7% proportional N resorption in Acer rubrum from high to low N-fertile sites, respectively. The smaller range in the present study probably reflects that intrastand gradients were investigated rather than interstand gradients. Since both stands and topographic positions had the same soil types, general fertility among them would be similar due to the same soil parent material and vegetation. Therefore, significant differences in resorption probably reflect microsite differences in moisture more so than fertility. However, other factors such as light availability (Boerner 1986) may also play a role. More probably, the amount of resorption a particular plant performs is a result of a complex of many environmental factors and physiological adaptations.
Overall, the levels of proportional resorption of P and N in Lonicera maackii tended to parallel those reported in the literature for other deciduous shrubs of temperate forests (Table 7). In general, woody understory species including shrubs have lower resorption rates than canopy species growing in the same areas (e.g. Ralhan and Singh 39
1987; Zimka and Stachurski 1992). This probably reflects
the low light intensities common in forest understories.
One other factor that may be important in explaining
the levels of resorption observed in L» maackii in Ohio populations is their phenology. Unlike other understory
shrubs in the sample sites, L- maackii does not begin senescing until November. Consequently, the shrub is exposed to colder temperatures and an increased probability of frost damage. Since the plant is a non native, these conditions could represent physiological stresses during this time of the year and could result in inefficient resorption. However, it should be noted that inefficient resorption can be compensated for by increased nutrient use efficiency (Killingbeck and Costigan 1988).
Since Lonicera maackii has an extended seasonal phenology, its photosynthetic carbon gains may offset the costs associated with nutrient losses to litter.
In summary, Lonicera maackii foliar P levels and nutrient resorption varied predictably along topographical gradients in response to the underlying fertility gradients. However, underlying soil moisture gradients probably also influenced nutrient dynamics. The results for this species support a generalized model of nutrient dynamics linking soil nutrient availability to foliar nutrient concentrations and resorption. Table 7.— P and N foliar proportional resorption reported for temperate deciduous shrubs of forests.
Species P N Location Citation Corvlus avellana 14.0 39.6 Poland Zimka and Stachurski 1992 Gavlussacia baccata 34.5 34.7 Rhode Island Killingbeck and Costigan 1988 Lonicera maackii 40.2 33.6 Ohio this study Quercus ilicifolia 59.1 70.0 Rhode Island Killingbeck and Costigan 1988 vaccinium vacillans 29.1 25.7 Rhode Island Killingbeck and Costigan 1988 Viburnum acerifolium 37.2 52.6 Ohio Hinoletti and Boerner 1994 Viburnum continifolium 48.5 55.7 India Ralhan and Singh 1987 Mean (w/o L. maackiil 37.1 46.4 CHAPTER III Nutrient Resorption in Lonicera maackii (Caprifoliaceae)
along a Southwestern Ohio Forest Chronosequence
INTRODUCTION
Whereas some hypotheses which deal explicitly with how rates and patterns of nutrient cycling change over the course of succession suggest that nutrient conservation increases with time (Odum 1969), others suggest that conservation is maximal in intermediate stages (Vitousek and ReinerB 1975). One important functional means of nutrient conservation in ecosystems is the foliar resorption of nutrients which occurs prior to leaf abscission (Ryan and Bormann 1982; Killingbeck 1986).
Resorption lowers nutrient losses while also reducing the amount of a nutrient a plant must absorb in subsequent seasons to sustain growth.
The objectives of this study were to determine when during succession nutrient conservation was maximal by quantifying phosphorus (P) and nitrogen (N) foliar dynamics and resorption in an understory shrub along a temperate, deciduous forest chronosequence consisting of a
41 25-, 45-, 65-year old and an older growth stand in
southwestern Ohio, USA. Specifically, we tested the hypothesis that nutrient resorption would increase with
stand age in the forest chronosequence as implied by Odum
(1969).
MATERIALS and METHODS
Study Species— Lonicera maackii (Rupr.) Maxim.
(Caprifoliaceae) is a multi-stemmed deciduous shrub native to northeastern Asia (Luken and Mattimiro 1991). It is now naturalized in many regions of Canada and the United
States (Pringle 1973). In south central and southwestern
Ohio the shrub initially invades woodlands (Luken 1988) through bird dispersal (Luken and Thieret 1987), then spreads rapidly by vegetative reproduction. &. maackii often forms a dense, conspicuous shrub layer within the understory of a woodland. In southwestern Ohio, L* maackii breaks bud in early-to-mid-April, and reaches full leaf expansion by the end of June. While many other woodland shrubs senesce by September or early October,
Ii. maackii remains photosynthetic through November, and senescence is often not completed until early to mid-
December. This species was chosen as it spanned the entire topographic range in both sample sites.
42 Sample site— Sampling was conducted in the Wright State
University (WSU) woods (84°03'W 39°45'N) in Bath Township,
Ohio. The WSU woods encompasses approximately 80 ha and
is comprised of four stands of different ages. The oldest
stand (hereafter, older-growth) has never been clear-cut
but has probably experienced selective cutting and
livestock grazing until 1950 (as evidenced by aerial photographs and historical records). The canopy vegetation fits Gordon's (1969) classification of an oak- sugar maple forest with Acer saccharum dominating stands of all ages (Table 8). The other three stands are recent secondary forests derived from abandoned pasture. These stands are approximately 25, 45, and 65 years old
(determined from aerial photographs back to 1940).
Consequently, these four stands represent a chronosequence from which inferences regarding successional processes can be pursued.
The soils of the WSU woods are Miamian silt loams, which are well drained in upland and slope positions and moderately drained in the bottomlands (Garner £& fll.
1978). These soils were formed on shale and limestone bedrock. Soil nutrient availability varied little among chronosequence stands (Table 8), although the 25-year old stand had significantly greater extractable soil N and soil moisture than the three others stands (Table 8). Table 8.— Characteristics of the four study stands in the Wright State University woods (Greene Co., Ohio). Means followed by the saae lower case letter were not significantly different following analysis of variance. Standard errors of the mean are given in parentheses. Data are adapted from DeMars and RunXle (1992) and DeMars (unpublished).
Parameter Older-Growth 65-Year 45-Year 25-Year Dominant C a n o p y Acer saccharum &. saccharum aaccharum &. saccharum Species Quercus rubra Ultmis amerlcana tj. amerlcana U. amerlcana Fraxlnus amerlcana Roblnla pgeudoacacla JJ. pseudoacacia Extractable N1
(mg N/Kg soil) 7.5a (0.8) 9.3a (1.3) 8.2a (1.1) 11.1b (0.3) Extraetable p*
(mg P/Kg soil) 3.6a (0.2) 3.5a (0.3) 3.4a (0.1) 3.9b (0.2) Moisture
(% water content) 21.2a (1.7) 20.43a (1.4) IB.4a (1.9) 24.2b (0.5)
1 Data are from 3 soil samples per site x topographic position for April-November 1992-1994, extracted in 2M KC1 (Keeney and Kelson 1982). 1 Data are from 3 composited samples per site x topographic position for April-November 1992-1994, extracted in IN ammonium acetate (Chapman 1965). 45
The climate of Greene County is continental (Miller
1969). Annual precipitation (Xenia, OH station) in 1992,
1993 and 1994 was 96, 101, and 106 cm, respectively all of which were within the normal 30-yr range. Mean January and July temperatures were -0.4° C / 23.3° C, 0.9° C /
25.5° C, and 0.6 C / 25.3° for 1992-1994, respectively
(NOAA 1992, 1993; 1994 data unpublished).
Field Methods— Within each study site, a single 20 m x
20 m plot was established within the center of the uplands portion of each stand. In April 1992, five Lonicera macckii individuals (at least breast height) were randomly selected and flagged in each plot. However, individuals were only selected if they were at least 3 m from any neighboring canopy tree stem to avoid potential microsite variation due to canopy tree influences (Boerner and
Koslowsky 1989).
Five fresh leaves were harvested from each individual at monthly intervals from April through November during
1992-1994. To avoid crown position effects, leaves were sampled throughout the crown of each individual. Freshly abscised litter (10 leaves) was collected under each individual in early December of each year. Laboratory Methods— In the laboratory, leaf area was measured for each individual sampled using a LI-3100 leaf area meter (Li-Cor, Inc., Lincoln, Nebraska). Leaves were dried at 70° C for 48-72 h and subsequently digested in
30% H202 and concentrated H3S04 for later N and P analyses
(Technicon 1977). Foliar tissue P concentrations were analyzed using the stannous chloride method (A.P.H.A.
1975) and foliar N concentrations were determined by autoanalyzer (QuikChem AE, Lachat). Reliability of N and
P digestions was evaluated by digesting and analyzing U.S.
National Bureau of Standards, Standard Pine Needle
Material #1585 (National Bureau of Standards 1981).
Foliar nutrient concentrations were calculated on a leaf area basis (mg/dm1 leaf area) for both P and N.
Because foliar concentrations from samples taken from the
July through September varied little, the data for this period were pooled. Absolute resorption was calculated as the difference between the mean summer concentration
(July-September) and minimum foliar concentration (which occurred in litter for all individuals). Proportional resorption was calculated as the percentage of the mean summer concentration that was actually resorbed prior to abscission.
To examine the effects of site and topographic position on nutrient concentrations, summer foliar 47
nutrient concentrations were analyzed by two-way analysis
of variance (ANOVA) by sampling year (Statistical Analysis
System 1985). Tukey's studentized multiple range test was used to determine significant differences among means.
Nutrient resorption was analyzed with repeated measures analysis of variance (RMANOVA) because use of the year term in standard ANOVA would violate the requirement of
independence among sampling units (Moser al. 1990;
Gumpertz and Brownie 1993). RMANOVA was performed in univariate mode since all sphericity tests of repeated measures variance-covariance matrices were not rejected
(Moser g& &1. 1990).
RESULTS
Foliar Nutrient Concentrations— Foliar P and N concentrations in Lonicera maackii varied significantly over the growing season. Mean foliar P was maximum at initial bud break and decreased steadily through June then remained relatively stable until senescence. Overall, mean summer foliar P ranged from 1.39 mg/dm2 in the older- growth stand in 1993 to 1.94 mg/dm1 in the 25-year old stand (Table 9). Leaves of £. maackii from the 25-year old stand had significantly higher mean summer P concentrations than leaves from the other stands for all three years. In contrast, mean summer foliar N 48 concentrations were lowest in plants from the 65-year old stand for each year (Table 9), and foliar N did not vary significantly among leaves from the other three stands in any year.
Foliar Nutrient Resorption— Mean absolute P resorption varied significantly among stand ages and over the three sampling years (Table 10). Absolute P resorption was greatest in the 25-year old stand in 1994, greater in the
25- and 65-year old stands in 1993, and did not vary among stands in 1992 (Table 10). Univariate repeated measures analysis of variance (RMANOVA) indicated a significant stand age effect, year effect, and second-order interaction for absolute P resorption (Table 11). The stand age effect represented a pattern of differences in overall amount of absolute resorption, with the greatest resorption in the 25-year old stand (0.62 ± 0.05) and the least resorption in the older growth and 45-year old stands (0.46 + 0.03 and 0.45 + 0.05, respectively). The significant year effect represented an increase in absolute P resorption over the sampling period: 0.44 ±
0.06 mg/dm3 in 1992, 0.54 ± 0.04 in 1993, and 0.60 ± 0.04 in 1994.
Mean proportional P resorption also varied over the chronosequence, but not in the same manner as absolute P resorption (Table 10). On a proportional basis, the greatest P resorption occurred in the 65-year old stand in
1992 and in the 65- and 25 year old stands in 1993. There were no significant differences in proportional P resorption in 1994 (Table 11). As was the case for absolute P resorption, RMANOVA indicated significant effects of stand age and year on proportional P resorption. However, unlike absolute resorption, the significant stand age effect on proportional P resorption resulted from a pattern with greatest resorption in the
65-year old stand (43.6% ±3.1) followed by the 25-year old stand (33.2% ± 2.1), and then the older growth (31.9%
± 2.0) and 45-year old stands (28.6% ± 2.8). Again, the significant year effect was represented by an increase in overall mean resorption throughout the three sample years
(30.1% ± 3.3, 34.8% ± 2.0, and 38.1% ± 2.2 for 1992-1994, respectively).
Mean absolute foliar N resorption in &. maackii varied considerably over the chronosequence and among years, from 4.07 ± 0,15 mg/dm* in the 65-year old stand in
1994 to 10.33 ± 0.38 mg/dm* in the 25-year old stand in
1994 (Table 10). RMANOVA indicated a significant stand age effect and stand age by year interaction, but no first-order year effect (Table 11). As was the case for absolute P resorption, the significant stand age effect 50 for absolute N resorption reflected greatest overall mean resorption in plants from the 25-year old stand (9.66 +
0.49 mg/dm*) and lowest mean resorption in plants from the
65-year old stand (4.51 mg/dm2 ± 0.24).
Proportional foliar N resorption ranged from 20.6 +
1.5 in the 65-year old stand in 1992 to 51.4 ± 1.2 in the
25-year old stand in 1993 (Table 10). RMANOVA indicated no significant year effect (Table 11). In contrast, there was a significant stand age effect with the greatest mean resorption of 47.9% ± 2.2 and the in the 25-year old stand and the lowest in the 65-year old stand (23.2% + 1.1)* a difference of 52%. Table 9.— Comparison of summer (July-September) mean foliar nutrient concentrations (mg/dm3 leaf area) in Lonicera maackii by successional age per year. Standard errors of the means are given in parentheses. Means followed by the same lower case letter within a row were not significantly different at E > 0.05. Year 25-Year Old 45-Year Old 65-Year Old Older-Growth Foliar P 1992 1.67b (0.03) 1.63a (0.04) 1.45a (0.06) 1.46a (0.03) 1993 1.94b (0.02) 1.57a (0.04) 1.50a (0.06) 1.39a (0.02) 1994 1.90c (0.03) 1.55ab (0.03) 1.42a (0.04) l.S9b (0.03)
Foliar H 1992 23.55b (0.60) 24.73b (0.32) 20.55a (0.42) 23.62b (0.42) 1993 22.21b (0.72) 23.97c (0.26) 20.01a (0.30) 23.51bc (0.45) 1994 22.95b (0.63) 123.2b (0.14) 18.89a (0.20) 23.77b (0.76) Table 10.— Comparison of foliar nutrient resorption in Lonicera maackii for 1992- 1994 within the Wright state University wood's chronosequence. Means of both absolute resorption (mg/dm3 leaf area) and proportional resorption are given for N = 5 individuals per successional age per year. Standard errors of the means are given in parentheses. Across rows, means followed by the same lower case letter were not significantly different at £ < 0.05.
Year 25-Year Old 45-Year Old 65-Year Old Older-Growth Absolute p Resorption 1992 0.40a (0.06) 0.42a (0.07) 0.60a (0.07) 0.37a (0.03) 1993 0.73b (0.04) 0.44b (0.03) 0.61b (0.05) 0.39a (0.04) 1994 0.75b (0.05) 0.50a (0.04) 0.50a (0.02) 0.64ab (0.03) Proportional P Resorption 1992 21.5a (2.8) 25.4a (4.0) 47.6b (4.6) 25.8a (1.6) 1993 38.4b (1.2) 27.8a (1.4) 44.6b (2.7) 28.3a (2.6) 1994 39.6a (2.2) 32.7a (2.9) 38.6a (1.8) 41.7a (2.0)
Absolute N Resorption 1992 8.42b (0.89) 7.65b (0.24) 4.12a (0.33) 6.75ab (0.54) 1993 10.23c (0.20) 6.14ab (0.26) S.34a (0.24) 6.84b (0.22) 1994 10.33d (0.38) 5.48b (0.34) 4.07a (0.15) 7.74c (0.34) Proportional W Resorption
1992 42.0c (4.0) 32.1b (1.0) 20.6a (1.5) 31.3b (2.1) 1993 51.3c (1.2) 26.9a (1.1) 27.1b (1.1) 31.7b (1.1) 1994 50.1c (1.5) 25.6a (1.5) 22.0a (0.7) 36.2b (1.5)
MUi Table 11.-"-Univariate repeated measures analysis of variance of absolute and proportional foliar P and H resorption for YEAR-repeated measures. ABSOLUTE P RESORPTION Source d£ MS z Z > £ Stand Age 3 0.31 8.98 0.001 Plant 16 0.01 Tear 2 0.11 10.91 0.001 Linear 1 0.17 9.07 0.01 Quadratic 1 0.06 3.60 n.e. Stand Age x Tear 6 0.07 6.96 0.0001 Error 32 0.01
PROPORTIONAL P RESORPTION Source df MS z £ > £ Stand Age 3 626.64 13.93 0.0001 Plant 16 44.98 Year 2 328.76 10.66 0.001 Linear 1 445.86 9.07 0.01 Quadratic 1 224.23 4.53 0.05 Stand Age x Year 6 241.10 7.82 0.0001 Error 32 30.83 TABLE 11.— (Continued) ABSOLUTE N RESORPTION Source df MS z £ > Z Stand Age 3 68.40 82.92 0.0001 Plant 16 0.82 Year 2 0.87 1.16 n.B. Linear 1 3.44 2.69 n.s. Quadratic 1 1.18 1.32 n.s. Stand Age x Year 6 5.08 6.79 0.0001 Error 32 0.75 PROPORTIONAL N RESORPTION Source d£ MS Z £ > Z Stand Age 3 1696.05 105.43 0.0001 Plant 16 16.09 Year 2 40.70 2.84 n.s. Linear 1 152.46 7.10 0.05 Quadratic 1 11.49 0.61 n.s. Stand Age x Year 6 80.98 5.65 0.001 Error 32 14.34 DISCUSSION
Our results do not support the hypothesis that nutrient resorption in Lonicera maackii increases with increasing stand age in this forest chronosequence in southwestern Ohio. We found P and N resorption occurring at equal or greater amounts in successional stands as in the older growth stand. These results parallel those of
Potter &£ fll. (1987) who found equal or greater P and N proportional resorption in Acer rubrum, Quercus prlrma, and Cornus florida in a successional southern Appalachian forest than in a nearby older growth forest. Furthermore, our data for L» maackii do not provide support for Odum's hypothesis (1969) of more closed nutrient cycling with increasing successional age. However, because of resprouting, advanced regeneration, and smaller changes in soil properties, secondary forest succession begins at a more advanced stage than does old field succession following abandonment from agriculture.
Although our data show that P and N foliar resorption in Lonicera maackii were linearly or directly correlated with stand age, we can not apply these conclusions to the ecosystems as a whole without assessing the resorption behavior of the broader canopy community, since shrubs and understory trees, in general, often have lower resorption rates than do canopy trees within the same forests (e.g. Ralhan and Singh 1987; Potter fll. 1987; Zimka and
Stachurski 1992). For instance, Potter fit fll. (1987) reported lower P and N resorption in Cornus florida (an understory tree) than in the canopy trees Acer rubrum and
Ouercus prinus in both older growth and successional stands at Coweeta. Similarly, Boerner (1985) reported mean proportional P and N resorption of €5.3% and 55.9%, respectively among four canopy tree species in an Ohio mixed oak forest during an average precipitation and temperature year, while mean proportional P and N resorption for the understory shrub Viburnum acerifolium was only 29.9% and 47.8%, during a subsequent average precipitation year (Minoletti and Boerner 1994).
An alternative explanation for the resorption behavior of L* maackii along this chronosequence may involve the differences in nutrient availability among stands, current models of plant nutrient dynamics suggest that individuals growing in less fertile sites may have greater nutrient resorption than those growing in more fertile sites (Chapin 1980; Vitousek 1982; Shaver and
Melillo 1984). However, in this study, P and N resorption was inversely proportional to soil nutrient availability, as the greatest absolute P resorption and greatest absolute and proportional N resorption occurred in the 25- year old stand where nutrient availability was greatest. Higher light levels in the understory due to a more open
canopy in the 25-year old stand nay help explain this
pattern. Since resorption is an energetically denanding
process, the higher light environnent of the 25-year old
stand nay provide additional carbon resources with which
to fuel higher nutrient resorption. This nay be
especially inportant in young forest stands, as Boring
al. (1981) showed that rapid increases in plant growth
during the early stages of forest succession can pronote
high nutrient denand.
In sumnary, resorption of both P and N was greatest
in the youngest aged stand overall. This is inconsistent
with Odum's (1969) hypothesis of increasing nutrient
conservation with succession but consistent with Vitousek
and Reiners' (1975) hypothesis of greatest conservation in
intermediate stages of succession. Overall, resorption
patterns also do not support a general model (Chapin 1980)
of plant nutrient dynamics linking soil nutrient availability and nutrient resorption in a nonambiguous manner. Most probably, the high rates of nutrient resorption observed in the youngest forest stand suggest a
link between resorption magnitude and light and higher growth rates. CHAPTER IV
Foliar Phosphorus and Nitrogen Resorption in Three
Woodland Herbs of Contrasting Phenology
INTRODUCTION
Foliar nutrient resorption is the process of retranslocation from the leaves into storage tissues during senescence and prior to abscission (Killingbeck
1986). Shrubs of deciduous forest understories typically display lower resorption amounts than canopy trees (e.g.
Ralhan and Singh 1987; Zimka and Stachurski 1992;
Minoletti and Boerner 1994), while forest herbs have even lower resorption rates (Boerner 1986). Lower resorption rates in understory plants probably reflect the adaptive syndrome of shade tolerance. As resorption is an energy requiring processes and as the photosynthetic rates of forest understory herbs are limited by low light intensity
(Sparling 1967; Taylor and Pearcy 1976), it is probable that insufficient carbon reserves are available to meet the respiration demands of high resorption activity.
However, this may not be the case for spring ephemerals, which have high photosynthetic rates during the period
58 59 before canopy leaf out (Taylor and Pearcy 1976), although data with which to test this hypothesis are currently lacking.
Factors other than light nay also potentially influence nutrient resorption. Models of plant nutrient dynamics suggest that individuals growing in less fertile sites may have greater nutrient resorption than those growing in more fertile sites (Chapin 1980; Vitousek 1982;
Shaver and Melillo 1984). Several studies examining foliar nutrient resorption among temperate deciduous forest stands support such models for both phosphorus and nitrogen (Stachurski and Zimka 1975; Zimka and Stachurski
1976; Boerner 1984; Kost and Boerner 1985). However, environmental stresses such as lowered moisture availability (del Arco &1. 1991; Minoletti and Boerner
1994) may negatively impact this process, especially for nitrogen (Boerner 1985a). Because moisture and nutrients vary with topographic position and aspect within a forest
(Whittaker 1956), examination of nutrient dynamics along intrastand topographic gradients can be used to test hypotheses related to the roles of soil moisture and soil nutrient availability in controlling nutrient resorption.
In this study, nutrient resorption was studied in three understory woodland herbs of contrasting phenology
(a spring ephemeral, a spring herb, and a summer herb) 60
along contiguous topographic gradients in two mesic
deciduous forests. These herbs were chosen to provide a
basis for evaluating the various phenological patterns
exhibited by forest understory herbs on nutrient resorption while sampling along topographic gradients was pursued to examine resorption in plots of varying nutrient and moisture availability. The working hypotheses were 1) that foliar nutrient resorption would be greatest in the spring ephemeral and lowest in the summergreen herb and 2) that nutrient resorption would be inversely proportional to soil nutrient availability.
MATERIALS and METHODS
Study Species— Three species of contrasting phenology and which occurred over the entire range of topography in our study areas were sampled. Cardamlne concatenata (Michx.)
O. Schwarz. (Brassicaceae) or cut leaf toothwort is a true spring ephemeral of rich, moist woodlands throughout the eastern United States and southeastern Canada (Rollins
1993). In Ohio, it is active aboveground from March to
May, prior to canopy closure in late May and early June; it flowers during April. Trillium flexioes Raf.
(Liliaceae) or nodding trillium is a spring and early summer herb of rich, moist, and often calcareous woodland soils (Gleason and Cronquist 1991). In Ohio, it is active 61
aboveground from late April to June and flowers in May.
Smllacina racemosa (L.) Desf. (Liliaceae) or false
Solomon's seal is a rhizomatous summer herb which emerges
in late April or early May and senesces in late August to
mid-September.
Sample Sites— Sampling was conducted in two forests
located in Greene County, Ohio. The first site was the
Wright State University (WSU) woods (84°03'W 39°45'N) in
Bath Township. The site was dominated by Acer saccharum
Marsh., Quercus rubra L., and Quercus alba b. (DeMars and Runkle 1992). Common associates included Ulmus americana
L., Carva spp. and Fraxlnus americana L. On the ESE slope
we sampled, £. americana and Q. alba were more abundant in
. the uplands, whereas in bottomlands Juolans nigra L. and
IJ. americana were more abundant.
Elevation in the WSU site ranged from 282 m in the
uplands to 260 m in the bottomland site, which in turn,
was approximately 1.5 m above the normal flow level of the
bordering stream. The soils of the WSU forest were
Miamian silt loams, which were well drained in upland and
slope positions and moderately drained in the bottomlands
(Garner si, 1978). These soils were formed on shale
and limestone bedrock. 62
The second site was a forest located 0.6 km west of
Central State University (CSU) in Xenia Township (83°05'W
39°42'N). The site was dominated by Acer saccharum and
Quercus rubra with the same common associates as the WSU
site. Elevation ranged from 305 m in the uplands to 278 m
in the bottomlands. The CSU soils were Miamian-Hennepin
series soils which were formed on a calcareous bedrock, and were both shallower and steeper in slope than those of
WSU.
Soil nutrient availability varied among topographic positions within these sites but not between forest sites
(Table 12). Overall there was greater extractable soil N in the bottomland topographic positions than in upland or mid-slope positions, and greater extractable soil P in uplands and mid-slope positions than in bottomland positions (Table 12). Moisture gradients based on gravimetric determinations result from higher moisture content in bottomland soils than in upland or mid-slope position soils (Table 12).
The climate of Greene County is continental (Miller
1969). Annual precipitation (Xenia, OH station) in 1992,
1993, and 1994 was 96, 101, and 106 cm, respectively, which was within the normal 30-yr range. Mean January and
July temperatures were -0.4° c/ 23.3° C, 0.9° C/25.5° C, and 0.6 C/25.3° for 1992-1994, respectively (NOAA 1992, 63
1993; 1994 data unpublished).
Field Methods— Within each study site, three 20 in x 20 n
plots were established randomly in each of 3 positions
along a contiguous topographic gradient: an upland, a
midslope (ESE-facing), and a bottomland stream terrace.
The plots were used to evaluate the variation within a
topographic position and since little intrapositlonal
variation was demonstrated, the data from the three plots
within each site by topographic position were subsequently
pooled for analysis.
Because of their differing phenologies, the three
herbs were sampled at different times in the season.
£. concatenata was sampled six times: twice in Harch,
April and Hay. X* flexipes was sampled six times also:
twice in April, May and June. £. racemosa was sampled
once monthly from April through August and twice in
September during senescence. Each species was sampled
over two year periods. £. concatenata and X» flexipes
were sampled during 1992 and 1993 while £. racemosa was
sampled during 1993 and 1994. On each sampling date, two
individuals of each species were excavated carefully from
each of the three plots for a total of six individuals per site per topographic position. In each case individuals were only sampled if they were at least 1.5 m from any neighboring canopy tree stem to reduce potential microsite variation from canopy tree proximity (Boerner and
Koslowsky 1989). I
Table 12.— Soil characteristics of study sites by topographic position. Means followed by the same lover case letter were not significantly different following analysis of variance. Standard errors of the mean for extractable N are given in parentheses. Data are adapted from DeMars and Runkle (1992) and DeMars (unpublished). ■sB SB B BB M aK m ssuK SssssB ssvisssB ssuB m ssass& nssB sssssssaviissB saaB ssassaitsasm ssaassssauB ism snsiasisa Parameter WSU-Up WSU-Slope HSO-Bottora CSU-Op CSU-Slope CSU-Bottom Extractable Nl (mg K/Kg soil) 7.5a (0.2) 8.8ab (0.4) 12.4c (0.3) 6.9a (0.1) 7.9a (0.3) 11.9c (0.6) Extractable P1 (mg P/Kg soil) 3.6b (0.2) 3.6b (0.7) 2.2a (0.3) 2.lab (0.7) 3.2b (0.5) 1.9a (0.5) Moisture 21.2a (1.7) 21.8a (0.9) 25.3b (0.6) 20.4a (1.5) 20.0a (1.8) 26.0b (0.6) (% water content) pH 6.02 6.66 6.80 6.57 6.63 7.12 1 Data are from 3 soil samples per site x topographic position for April-November 1992-1994, extracted in 2M KCl (Keeney and Nelson 1982). 3 Data are from 3 composited samples per site x topographic position for April-November 1992-1994, extracted in 1M ammonium acetate (Chapman 1965).
at ui Laboratory Methods— In the laboratory, leaves were dried at 70° C for 48-72 h and then digested in 30% H202 and concentrated H2S04 for later N and P analyses (Technicon
1977). Foliar tissue P concentrations were analyzed using the stannous chloride method (A.P.H.A. 1975) and foliar N concentrations were determined by autoanalyzer (QuikChem
AG, Lachat). Reliability of N and P digestions was evaluated by digesting and analyzing Standard Pine Needle
Material #1585 from the U.S. National Bureau of Standards
(National Bureau of standards 1981).
Foliar nutrient concentrations were determined on a mass basis (mg nutrient/g dry foliar tissue) for both P and N. To calculate nutrient resorption, presenescent foliar nutrient concentrations were determined from the mean foliar nutrient concentration of individuals after they had reached maximum leaf area. For £. concatenata. this corresponded to the April samples; for £• flexipes. the May samples, and for £. racemosa. the June and July samples. Absolute resorption was calculated as the difference between the mean presenescent nutrient concentration and postsenescent foliar litter concentrations. Proportional resorption was calculated as the percentage of the mean presenescent nutrient concentration that was actually resorbed prior to abscission (Boerner 1984). 67
To determine whether herbs of differing phenology exhibit different resorption patterns and to examine the effects of site and topographic position on nutrient resorption, data were subjected to two-way analysis of variance (Statistical Analysis System 1985). Tukey's studentized multiple range test was used to determine significant differences among means, one-way analysis of variance was performed where interactions effects were significant to further clarify resorption patterns.
RESULTS
P Foliar Resorption— Overall, absolute P resorption varied from -0.13 ± 0.12 mg/g in Smilacina racemosa in the WSU uplands during 1993 to 2.14 + 0.06 mg/g in Cardamine concatenata in the CSU bottomland position during 1992
(Table 13). Analysis of variance (ANOVA) indicated highly significant species and topographic position first order effects and significant species x year, species x site, and species x topographic position interactions (Table
14). £. concatenata had the greatest overall mean absolute P resorption of 1.79 ± 0.09 mg/g, followed by X. flexipes with 0.65 ± 0.05 mg/g, then £. racemosa with 0.34
±0.05 mg/g. Overall, resorption was greatest in plants in the bottomlands, with mean absolute P resorption of
1.19 ± 0.06 mg/g, followed by upland plants (0.80 ± 0.07 68 mg/g), then slope plants (0.79 ± 0.05 mg/g). The significant overall topographic effect reflected greater absolute P resorption in plants from bottomland topographic positions regardless of species, site or sample year (Table 13) and intermediate resorption levels for both £. concatenata and I. flexipes individuals in upland topographic positions while lowest resorption occurred in plants growing in slope positions. For
£. racemosa. upland plants consistently had significantly lower absolute P resorption than other species (Table 13) reflecting the significant species x topographic position interaction as well. The significant species x site interaction stemmed from greater resorption in
£. concatenata and £. racemosa at CSU than HSU, and the reverse for X* flexipes while the significant species x year interactions resulted from greater absolute P resorption by £. concatenata in 1993 than in 1992, and
S. racemosa in 1994 than in 1993 (Table 13).
Overall, proportional P resorption varied from -8.3%
±7.6 for £. racemosa in the WSU upland site during 1993 to 68.1% ± 2.0 for £. concatenata in the WSU bottomland position in 1992 (Table 13). ANOVA indicated the same overall trends for proportional P resorption as for absolute P resorption (Table 14). £. concatenata had the greatest overall mean proportional P resorption (58.8% ± 3.1), followed by 2. flexipes (23.9% ± 1.6), then
£. racemosa (17.6% ± 2.6). As for absolute P resorption, the significance of the topographic effect reflected consistently greater proportional P resorption in plants of bottomland positions regardless of site or year.
Overall, plants growing in bottomlands had the greatest mean proportional P resorption (42.3% ± 2.1), followed by plants in slope positions (33.0% + 2.3), then plants in upland positions (25.0% ± 3.0).
N Foliar Resorption— Absolute foliar N resorption ranged from 11.25 ± 0.58 mg/g in Smilacina racemosa in the WSU slope position during 1994 to 37.71 ± 4.11 mg/g in
Cardamine concatenata in the WSU upland position during
1992 (Table 15). ANOVA indicated significant species and topographic position effects as well as a significant species x topographic position interaction (Table 16).
As was the caBe for absolute P resorption, absolute N resorption was greatest in £. concatenata (30.58 ± 1.26
®9/g)t intermediate in 2* flexipes (16.33 + 0.64 mg/g), and lowest in £. racemosa (14.73 + 0.69 mg/g). Overall, plants from bottomlands and uplands had nearly equal absolute N resorption (22.41 + 0.97 and 21.93 ± 1.10 mg/g, respectively) which was significantly greater than that of plants from slope positions (17.30 ± 0.52 mg/g). The significant species x topographic position interaction reflected consistently lower absolute N resorption by
£. concatenata and £. racemosa in slope positions and greater resorption by individuals in upland and bottomland positions but not for T. flexipes (Table 15).
Proportional foliar N resorption varied from 23.3% ±
1.0 in I. flexipes in the CSU upland position during 1993 to 62.4% ± 6.8 in £. concatenata in the WSU upland position during 1992 (Table 15). ANOVA indicated the same significant effects as for absolute N resorption.
Overall, the mean proportional N resorption was greatest in £. concatenata (58.2% + 2.3), and lower and similar in both £. racemosa (29.3% + 1.4) and I. flexipes (28.7% ±
1.1). Again, as for absolute N resorption, plants from bottomlands and uplands resorbed similar proportions of foliar N (39.7% + 1.7 and 39.5% ± 2.1, respectively), while plants from slopes resorbed a significantly lower proportion (37.1% ± 1.0). The significant interaction of species x topographic position reflected the same resorption patterns as observed for absolute N resorption. Table 13.-- Mean foliar P concentrations (mg P/g dry leaf mass) determined from presenescent sampling periods when leaves were at maximum leaf area, mean absolute p resorption, and mean proportional P resorption. Standard errors of the mean are listed in parentheses. Values followed by different lower case letters within a site x year x topographic position indicated significant differences at £ > 0.05. B U B ssnsB usicsB SsniiB sm ninm sssraB SSsacitknansussnsasvm ftsaussM isiissanssaasiaM nasiasaiM aM Species site Year Topographic Presenescent P Absolute P Proportional P position concentration resorption resorption C. concatenata WSD 1992 Upland 3.32 (0.15) 1.59ab (0.20) 48.0a (6.0) Slope 2.11 (0.20) 1.05b (0.10) 49.8a (4.9) Bottomland 3.36 (0.20) 2.01a (0.19) 59.7a (5.6) 1993 Upland 3.38 (0.05) 1.72ab (0.15) 51.1a (4.4) Slope 2.44 (0.09) 1.42b (0.05) 58.2a (2.3) Bottomland 3.25 (0.16) 2.05a (0.12) 63.1a (3.7) CSU 1992 Upland 3.32 (0.16) 1.88ab (0.13) 56.8b (4.0) Slope 2.67 (0.12) 1.68b (0.05) 62.8ab (1-7) Bottomland 3.15 (0.19) 2.14a (0.06) 68.1a (2.0) 1993 Upland 3.42 (0.13) 2.00a (0.02) 58.7b (0.5) Slope 2.80 (0.04) 1.75b (0.03) 62.4ab (0.9) Bottomland 3.21 (0.17) 2.12a (0.05) 66.2a (1.5)
T. flexioes HSU 1992 Upland 2.88 (0.04) 0.93a (0.07) 32.4a (2.3) Slope 2.84 (0.04) 0.67b (0.05) 25.5b (1.8) Bottomland 2.73 (0.05) 1.03a (0.02) 37.7a (0.6)
1993 Upland 2.82 (0.05) 0.55b (0.04) 19.4b (1.4) Slope 2.69 (0.02) 0.48b (0.04) 17.9b (1-4) Bottomland 2.60 (0.02) 0.75a (0.07) 28.8a (2.6)
CSU 1992 Upland 2.72 (0.05) 0.54b (0.06) 19.8b (2.1) Slope 2.66 (0.04) 0.43b (0.03) 16.1b (1.2) Bottomland 2.79 (0.07) 0.80a (0.05) 28.6a (1.8)
1993 Upland 2.69 (0.02) 0.43b (0.03) 16.1b (1.3) Slope 2.63 (0.02) 0.29b (0.06) 10.9b (2.2) Bottomland 2.74 (0.02) 0.93a (0.03) 33.9a (1.0) Table 13.— (Continued) g. racemoaa HSU 1993 Upland 1.57 (0.03) - 0.13b (0.12) -8.3b (7.6) Slope 1.97 (0.02) 0.48a (0.10) 24.3a (4.9) Bottomland 2.00 (0.04) 0.63a (0.02) 32.0a (1.0) 1994 Upland 1.56 (0.02) 0.04c (0.05) 2.8c (2.9) Slope 1.90 (0.02) 0.48b (0.03) 25.2b (1.4) Bottomland 2.03 (0.09) 0.73a (0.02) 36.1a (1.0) CSU 1993 Upland 1.50 (0.03) - 0.01c (0.02) -0.6b (1.3) Slope 1.79 (0.03) 0.37b (0.02) 20.9a (1.0) Bottomland 1.93 (0.03) 0.50a (0.04) 26.2a (2.3) 1994 Upland 1.61 (0.03) 0.05b (0.03) 3.3b (2.1) Slope 1.88 (0.02) 0.40a (0.07) 21.4a (3.6) Bottomland 1.96 (0.02) 0.54a (0.03) 27.5a (1.6)
to Table 14.— Overall analysis of variance for absolute and proportional foliar P resorption. Kean Squares were determined from Type ill sums of squares.
ABSOLUTE P RESORPTION Source df MS z > z Species 2 32.76 804.08 0.0001 Year 2 0.05 1.18 n.s. Site 1 0.01 0.11 n.s. Topographic Position 2 2.74 67.33 0.0001 Species x Year 1 0.27 17.80 0.0001 Species x Site 2 2.00 24.55 0.0001 Species x Topographic Position 4 2.21 13.54 0.0001 Year x Site 2 0.05 0.59 n.s. Year x Topographic Position 4 0.07 0.45 n.s. Site x Topographic Position 2 0.07 0.81 n.s. Error 193 0.04
PROPORTIONAL P RESORPTION Source df MS Z £ > E Species 2 29209.05 527.03 0.0001 Year 2 163.98 2.96 n.s. Site 1 28.65 0.52 n.s. Topographic Position 2 3747.27 67.61 0.0001 Species x Year 1 565.98 10.21 0.01 Species x Site 2 842.45 15.20 0.0001 Species x Topographic Position 4 935.57 16.88 0.0001 Year x Site 2 41.49 0.75 n.s. Year x Topographic Position 4 46.05 0.83 n.s. Site x Topographic Position 2 40.13 0.72 n.s. Error 193 55.42 Table 15.— Mean foliar N concentrations (mg N/g dry leaf mass) determined from presenescent sampling periods when leaves were at maximum leaf area, mean absolute N resorption, and mean proportional N resorption. Standard errors of the mean are listed in parentheses. Values followed by different lower case letters within a site x year x topographic position indicated significant differences at £ > 0.05.
Species Site Year Topographic Presenescent N Absolute N Proportional N position concentration resorption resorption C. concatenata WSU 1992 Upland 60.42 (2.62) 37.71a 4.11) 62.4a (6.8) Slope 46.60 (0.91) 24.32b 0.68) 52.2a (1.5) Bottomland 58.14 (2.17) 34.96a 1.95) 60.1a (3.4) 1993 Upland 57.26 (0.70) 35.20a 1.54) 61.5a (2.7) Slope 39.87 (2.20) 19.65b 0.11) 49.3b (0.3) Bottomland 58.14 (2.17) 34.96a 1.92) 60.0a (3.2) CSU 1992 Upland 56.11 (1.93) 34.21a 1.09) 61.0a (1.9) Slope 41.71 (1.41) 21.83b 0.50) 52.3b (1.2) Bottomland 52.42 (2.97) 33.31a 1.31) 63.6a (2.5) 1993 Upland 56.76 (0.46) 35.79a 0.69) 63.1a (1.2) Slope 41.59 (1.51) 21.71c 0.57) 52.2b (1.4) Bottomland 54.50 (2.39) 33.28b 0.61) 61.1a (1.1) T. flexipes WSU 1992 Upland 57.19 (0.47) 16.96a 1.11) 29.7a (1.9) slope 56.11 (0.61) 16.94a 0.82) 30.2a (1.5) Bottomland 57.69 (0.56) 15.53a 0.60) 26.9a (1.0) 1993 Upland 58.27 (0.54) 16.21a 1.03) 27.8b (1.8) Slope 56.31 (0.27) 18.15a 0.50) 32.2a (0.9) Bottomland 58.20 (0.31) 18.02a 0.35) 31.0ab (0.6) CSU 1992 Upland 57.87 (0.43) 14.98b 0.81) 26.2b (1.4) Slope 55.86 (0.81) 17.63a 0.34) 31.4a (0.6) Bottomland 56.08 (0.47) 14.41b 0.45) 25.7b (0.8) 1993 Upland 55.45 (0.27) 12.91c 0.55) 23.3c (1.0) Slope 57.40 (0.18) 18.88a 0.68) 32.9a (1.2) Bottomland 56.66 (0.20) 15.37b 0.40) 27.1b (0.7) Table 15.— (Continued)
S. racemosa WSU 1993 Upland 47.77 (0.30) 14.73a (0.91) 30.8a (1.9 Slope 42.17 (1.06) 11.93b (0.34) 28.3a (0.8 Bottomland 58.14 (2.17) 15.36a (0.50) 26.4a (0.9 1994 Upland 48.31 (0.31) 12.43b (0.85) 25.7b (1.8 slope 40.10 (0.56) 11.25b (0.58) 28.0ab (1.4 Bottomland 56.71 (1.09) 19.12a (1.08) 33.7a (1.9 CSU 1993 Upland 51.82 (0.39) 16.79a (0.49) 32.4a (0.9 Slope 44.78 (1.08) 12.41b (0.30) 27.7a (0.7 Bottomland 58.05 (0.89) 17.48a (1.10) 30.1a (1.9 1994 Upland 50.30 (0.57) 15.26ab (0.03) 30.3a (1.6 Slope 45.23 (0.51) 12.89b (0.81) 28.5a (1.3 Bottomland 56.92 (1.13) 17.02a (1-34) 29.9a (2.4
u* Table 16.— Overall analysis of variance for absolute and proportional foliar N resorption. Mean Squares were determined from Type III suns of squares. H BtStBm nnBSnH Sim BSgn8M SBBSa888B«|U 8BttSaSSn8SaS9S899»Sa3SS«S38SSSS9U SSSS«*sa9 a b s o l u t e n r e s o r p t i o n
Source a MS £ E > Z Species 24531.72 582.17 0.0001 Year 2 0.97 0.12 n.s. site 1 11.66 1.50 n.s. Topographic Position 2 442.23 56.81 0.0001 Species x Year 1 19.48 2.50 n.s. Species x Site 2 15.39 1.98 n.s. Species x Topographic Position 4 485.43 62.36 0.0001 Year x Site 2 5.61 0.72 n.s. Year x Topographic Position 4 15.14 1.94 Q*8e Site x Topographic Position 2 13.40 1.72 n.s. Error 193 7.78
PROPORTIONAL N RESORPTION Source df MS £ E > £ Species 2 20487.14 867.84 0.0001 Year 2 0.07 0.00 n.s. Site 1 0.87 0.04 n.s. Topographic Position 2 146.78 6.22 0.01 Species x Year 1 18.54 0.79 n.s. Species x Site 2 53.51 2.27 n.s. Species x Topographic Position 4 431.89 18.29 0.0001 Year x Site 2 3.13 0.13 n.s. Year x Topographic Position 4 42.79 1.81 n.s. Site x Topographic Position 2 6.31 0.27 n.s. Error 193 23.61 77
DISCUSSION
Phenoloalcal Patterns— Our first working hypothesis: that
foliar nutrient resorption would be greatest in the spring ephemeral and lowest in the summergreen herb, is supported as both P and N absolute and proportional resorption were highest for Cardamlne concatenata. In all cases except proportional N resorption, the summer herb Smilacina racemosa had the lowest resorption levels, while the spring herb Trillium flexipes had intermediate levels which were not consistently significantly different from
£. racemosa. The high level of P resorption observed in
Cardamine concatenata (58.7%) is similar to that reported for Anemone nemorosa. a European spring ephemeral (Ernst
1983). Likewise, the high level of N resorption in
£. concatenata (58.2%) is similar to that of 61.1% observed for Ervthronium americanum. an eastern North
American spring ephemeral (Muller 1978). For the spring and summergreen understory forest herbs X. flexipes and
£. racemosa. the overall mean proportional resorption rates parallel those reported for two other summergreen herbs Polvaonatum pubescens (16.8% for P and 36% for N) and Geranium roaculatum (20.9% for P and 39.6% for N) in two Ohio forests (Boerner 1986).
Plant species living in the understories of forests where light levels are low may possess several adaptive 78
strategies (Hicks and Chabot 1985). In temperate
deciduous forests, one adaptive strategy is the spring
ephemeral syndrome, in which relatively small herbs
complete their aboveground activities in the narrow window
of opportunity afforded by the spring months, prior to
canopy leaf out. such plants characteristically have high
absolute photosynthetic rates and high light compensation
and saturation levels (Taylor and Pearcy 1976; Mahall and
Bormann 1978). Moving toward the other end of the
spectrum, Sparling (1967) has classified spring and
summergreen forest plants into intermediate and shade
tolerant species based on their phenology and activity in
the understory. Such species have correspondingly lower
photosynthetic rates and light compensation and saturation
levels (Sparling 1967; Taylor and Pearcy 1976).
Because nutrient resorption is an energy demanding
process, shade tolerant plants may have to trade off
increased nutrient conservation by foregoing or reducing
resorption. Such tradeoffs may be offset by greater
nutrient use efficiency in shade tolerant plants, and,
indeed Smilacina racemosa had the greatest nutrient use
efficiency of the species we examined (based on the
inverse of tissue concentrations (Chapin 1980) while
£. concatenata had the lowest. Such a mechanism has been also suggested as the reason for lower observed resorption 79 rates in shrubs (Killingbeck and Costigan 1988).
Topographical Patterns— The second working hypothesis- that nutrient resorption would be inversely proportional to soil nutrient availability- is generally supported by the data. The bottomland topographic positions had the lowest extractable soil P levels and plants of all three species displayed the greatest absolute and proportional P resorption in these topographic positions. This is consistent with P resorption reported in temperate, deciduous woody understory plants including Cornus florlda
(Host and Boerner) and Hammamells vlrainiana (Boerner
1985a) among sites of varying soil-P fertility, and
Lonicera maackii in our study sites (DeMars and Boerner, unpublished), however, not with Viburnum acerifolium
(Minoletti and Boerner 1994) and other shrubs (Killingbeck and Costigan 1988).
In this study, slopes and uplands had very similar soil P availability, but slopes consistently had lower resorption rates for all three species. This may reflect the drier conditions of these ESE facing slopes due to exposure and more rapid drainage. Much data indicate that nutrient resorption rates are sensitive to soil moisture status with drier soils or drought-affected soils resulting in lower resorption by plants (Hocking 1982; 60
Kost and Boerner 1984; del Arco al. 1991; Mlnolettl and
Boerner 1994). A phenological attribute related to soil moisture status that may facilitate greater resorption rates in spring ephemerals is seasonal variations in soil moisture availability. Muller (1978) suggested that
"exploitation of the vernal niche" by spring ephemerals not only provides benefits from increased photosynthetic capacity, but also from the generally greater soil moisture present during spring. Increased soil moisture from spring precipitation, snow melt, and low evapotranspiration may favor increased nutrient resorption. In contrast, decreased available soil moisture during the later aboveground period for spring and summer herbs coincides with higher temperatures and evapotranspiration rates.
Resorption rates for foliar N did not support the second hypothesis as clearly. Overall, plants from bottomlands and uplands did not have significantly different N resorption rates even though they had significantly different soil N availability. Plants from slopes had significantly lower N resorption rates than either uplands or bottomlands even though slopes had nearly equal soil N availability as uplands. Again, this may have reflected soil moisture status, as soluble soil N would be less available over time in slope positions with 81 greater drainage than in upland positions which are less well drained. This would be expected as moisture availability affects mass flow of soil N (Barber 1984).
Only one other study has reported foliar nutrient resorption in temperate forest herbs along fertility gradients. Boerner (1986) reported a direct relationship between fertility and foliar nutrient resorption. This is in contrast to our overall observations of an inverse relationship. However, the inverse relationship we observed was based on much greater resorption by the spring ephemeral Cardamine concatenata (indicated by the significant species by topographic position interactions), while Boerner (1986) examined only summergreen herbs.
Although our data still suggest an inverse relationship for the summergreen Smilacina racemosa. the relationship for this species alone is not as strong.
In summary, the results of this study suggest that the adaptive strategies employed by these forest understory herbs affect the magnitude of resorption they can perform during senescence. The spring ephemeral,
Cardamine concatenata is capable of high photosynthetic rates, and coupled with its low biomass, may provide a mechanism within its adaptive syndrome to provide energy reserves necessary for fueling levels of resorption as high as those observed in many forest canopy trees. In 82 contrast, the summergreen Smilacina racemosa> which possesses a shade tolerant syndrome, is incapable of fueling high levels of resorption while the spring herb with intermediate phenology, Trillium flexipes displays resorption levels only slightly higher than the summergreen herb. Additionally, resorption in these herbs may be affected by soil nutrient availability, as our data indicated that overall P resorption was greatest in the bottomland sites where available P was lowest. However, the relationship between resorption and soil N availability was not as clear because overall N resorption was not significantly different between N-rich bottomlands and P-poorer bottomlands, but was different between slopes. Differences in soil moisture and more specifically drainage patterns may explain this observation. CHAPTER V
Arbuscular Mycorrhizal Dynamics of Three Woodland
Herbs of Contrasting Phenology
INTRODUCTION
Arbuscular mycorrhizae (AM) are ubiquitous in terrestrial ecosystems (Harley and Smith 1983), including the eastern deciduous forest biome of North America
(Brundrett and Kendrick 1988). However, few studies have examined AH dynamics in deciduous forest herbs in relation to either seasonal or topographic variation. Because AM may influence a plant's ability to acquire adequate amounts of phosphorus and other nutrients (Hayman 1983) studies examining AH dynamics of forest understory herbs along temporal and spatial gradients of soil nutrient availability may shed light on the relative importance of this mutualism in these ecosystems.
Previous research has shown that AM fungal spore counts vary over topographic gradients, with lower counts in lower, moister topographic positions (Anderson al.
1984). Such differences may have reflected variations in
P availability (Mosse and Phillips 1971; Menge e£ al.
83 84
1978; Nelsen £fc al. 1981) or soil moisture (Rabatin 1979).
Similarly, Boerner (1986) showed that AM development was greatest in low P availability sites than in high P availability sites for two summergreen forest perennial herbs, but did not explicitly consider the possible influences of topographic variations or sample species with other phenologies.
Thus, the purpose of this study was to examine AM dynamics in three deciduous forest herbs of differing phenology over several years and over intrastand topographic gradients to determine if their life history characteristics correlated with any significant differences in AM colonization patterns. Additionally, we tested the working hypotheses that AM development would be lower in lower topographic positions, as has previously been shown with spore counts (Anderson a£ al» 1984) and that AM development would be lower in sampling sites with the highest P availability.
MATERIALS and METHODS
Study Species— Three species of contrasting phenology and which occurred over the entire range of topography in our study areas were studied. Cardamine concatenata (Michx.)
0. Schwarz. (Brassicaceae) is a true spring ephemeral of rich, moist woodlands throughout the eastern United States and southeastern Canada (Rollins 1993). In Ohio, it is
active aboveground from March to May, prior to canopy
closure; it flowers during April. The perennating organ
is a tuberous rhizome. The roots of £. concatenata are
fine, with diameters ranging from 0.2 mm to 0.7 mm.
Trillium flexipes Raf. (Liliaceae) is a spring and
early summer herb of rich, moist, and often calcareous woodland soils (Gleason and Cronguist 1991)* In Ohio, it
is active aboveground from late April to June and flowers
in May. The perennating organ is a short rhizome bearing up to 24 coarse roots with diameters ranging from 0.9 mm
to 2.6 mm. Smilacina racemosa (L.) Desf. (Liliaceae) is a summer herb of rich woodland soils (Gleason and Cronguist 1991).
It emerges in late April or early May and senesces in early to mid-September. Its perennating organ is a rather
long rhizome measuring up to 1.7 dm. Roots are produced during summer on the previous years' rhizome growth
(Brundrett and Kendrick 1988) and are coarse, with diameters ranging from 0.6 mm to 2.1 mm.
Sample Sites— Sampling was conducted in two forests located in Greene County, Ohio. The first site was the
Wright State University (WSU) woods (84°03'W 39®45'N) in
Bath Township. The site was dominated by Acer saccharum 86
Marsh., Quercus rubra L., and Quercus alfea L. (DeMars and
Runkle 1992). Common associates included Ulmus amerlcana
L., Carva spp. and Fraxinus americana L. On the ESE slope we sampled, £. amerlcana and Q. alba were more abundant in the uplands, whereas in bottomlands Jualans nigra L. and
II. amerlcana were more abundant.
Elevation in the WSU site ranged from 282 m in the uplands to 260 m in the bottomland site which, in turn, was approximately 1.5 m above the normal flow level of the bordering stream. The soils of the WSU forest were
Miamian silt loams, which were well drained in upland and slope positions and moderately drained in the bottomlands
(Garner &!• 1978). These soils were formed on shale and limestone bedrock.
The second site was a forest located 0.6 km west of
Central State University (CSU) in Xenia Township (83°05'W
39°42'N). The site was dominated by Acer saccharum and
Quercus rubra with the same common associates as the WSU site. Topography ranged from 305 m in the uplands to 278 m in the bottomlands. The CSU soils were Miamian-Hennepin series soils which were formed on a calcareous bedrock and were both shallower and steeper in slope than those of WSU
(Garner g£ &1. 1978)
Soil nutrient availability varied among topographic positions within these sites but not between forest sites 87
(Table 17). Overall there was greater extractable soil N in the bottomland topographic positions than in upland or mid-slope positions, and greater extractable soil P in uplands and mid-slope positions than in bottomland positions (Table 17). Moisture gradients based on gravimetric determinations result from higher moisture content in bottomland soils than in upland or mid-slope position soils (Table 17).
The climate of Greene County is continental (Miller
1969). Annual precipitation (Xenia, OH station) in 1992,
1993, and 1994 was 96, 101, and 106 cm, respectively, which was within the normal 30-yr range. Mean January and
July temperatures were -0.4° C/ 23.3° C, 0.9° C/25.50 C, and 0.6 C/25.3° for 1992-1994, respectively (NOAA 1992,
1993; 1994 data unpublished).
Field Methods— Within each study site, three 20 m x 20 m plots were established randomly in each of 3 positions along a contiguous topographic gradient: an upland, a midslope (ESE-facing), and a bottomland stream terrace.
On each sampling date, 6 individuals of each species were excavated carefully from each plot to obtain intact underground organs. Cardamine concatenata and Trillium flexipes were sampled during 1992 and 1993 while Smillclna racemosa was sampled during 1993 and 1994. 88
Plants were sampled at least 8 times during each
year. In addition to samples taken during the active
growth and reproductive period, at least 4 sets of samples
were taken during the period of aboveground senescence and
months of aboveground inactivity.
Laboratory Methods— In the laboratory, roots were
separated, cleared and stained with trypan blue (Phillips
and Hayman 1970) and stored in plastic tissue cassettes
with FAA (formalin:acetic acid:ethanol) preservative
solution. AM development was examined in 5 random 2-3 cm
root segments per plant, and was defined as the presence
of any internal non-septate hyphae, vesicles, arbuscules
or chlamydospores. Aborted appresoria were not included
in this definition. The percentage of root length
colonized (PRLC) by mycorrhizal fungi was estimated for
each segment (Giovanetti and Mosse 1980) and the mean for
the five segments computed. Roots without cortex or roots with obvious damage to the cortex were not used. PRLC was
analyzed using analysis of variance (ANOVA) and
significant differences among class variables determined with Tukey's studentized multiple range test (Statistical
Analysis System 1985). Table 17.— Soil characteristics of study sites by topographic position. Means followed by the same lower case letter were not significantly different following analysis of variance. Standard errors of the mean for extractable N are given in parentheses. Data are adapted from DeMars and Runkle (1992) and DeMars (unpublished).
Parameter WSU-Up WSO-Slope KSU-Bottom CSU-Up CSU-Slope CSU-Bottom Extractable ft1 {mg N/Kg soil) 7.5a (0.2) 8.8ab (0.4) 12.4c (0.3) 6.9a (0.1) 7.9a (0.3) 11.9c (0.6) Extractable PJ (mg P/Rg soil) 3.6b (0.2) 3.6b (0.7) 2.2a (0.3) 3.lab (0.7) 3.2b (0.5) 1.9a (0.5) Moisture (% content) 21.2a (1.7) 21.8a (0.9) 25.3b (0.6) 20.4a (1.5) 20.0a (1.8) 26.0b (0.6) pH 6.02 6.66 6.80 6.57 6.63 7.12
1 Data are from 3 soil samples per site x topographic position for Aprll-November 1992-1994, extracted in 2M KCl (Keeney and Nelson 1982). * Data are from 3 composited samples per site x topographic position for April-November 1992-1994, extracted in 1M ammonium acetate (Chapman 1965).
V)03 90
RESULTS
Cardamine concatenta— During both sample years, roots of
£. concatenata only exhibited significant AM fungal colonization during the period in May when the above ground parts of the plants were senescing (Figure 5).
Analysis of variance revealed no significant effects of sample year, topographic position, or sample site on AM infection, either overall or on the single date on which significant AM development was present.
The mean percent root length colonized (PRLC) at this point in the season was 18.1% (+ 17.7); however, no arbuscules were observed in any of the 3456 £. concatenata root segments examined. There was also considerable variation in PRLC within populations, with some individuals approaching 100% colonization and others entirely uninfected. As no roots were present on
£. concatenata rhizomes on the dates of our June and
August samples, root abscission must have followed closely after senescence of aboveground tissues. Figure 5. Mean percent root length colonized by arbuscular mycorrhizal fungi in Cardamlne concatenata within the
Wright State and Central State forests. Results are pooled as there were no significant site, year, or topographic position effects. Bars represent standard errors of the mean.
91 Figure 5 Figure MEAN PERCENT ROOT LENGTH COLONIZED 1992 ot bscission ab Root ^ I j_JL D A T ra n c atnata a ten ca n co ine ardam C J 1993 ot abscission Root A 0 D F M VO Figure 6. Mean percent root length colonized by arbuscular mycorrhizal fungi in Trillium flexlpes within the Wright
State and Central state forests. Individual plots represent site x topographic position results for 1992 and
1993. Bars represent standard errors of the mean.
93 Figure Figure
MEAN PERCENT ROOT LENGTH COLONIZED 30 40 50 60 50 70 80 60 6
A J A O D F A J A O D F J » t t i i i i » H. i i i i i i t t i i i H r i i -- i i I t i i r i i i i i H i i i i r t i i i i II t t i i i -J-1 1992
Trillium flexipes Trillium 1993 WSU—LOWLAND WSU—UPLAND WSU-SLOPE JL 1992
1993 CSU-LOWLAND CSU—UPLAND CSU-SLOPE Figure 7. Mean percent root length colonized by arbuscular mycorrhizal fungi in Smllacina racemosa within the Wright
State and Central State forests. Individual plots represent site x topographic position results for 1993 and
1994. Bars represent standard errors of the mean.
95 80 Smilacina racemosa 70
60
5 0 WSU—UPLAND CSU-UPLAND
i r i f i i i i I I 1 I I 1 I • I I 3 90
X 80
g 60 WSU-SLOPE * CSU—SLOPE w 5 0 5 70
< 60
WSU-LOWLAND CSU-LOWLAND
AJAODFAJAO Figure 7 1993 1994 1994 Trillium flexlues— significant AM development was present in the roots of 2* flexipes on all sampling dates during both years (Figure 6). Internal hyphae, vesicles, and arbuscules were common in the roots of both 2* flexipes and Smilaclna racemosa. and only 13 of 3240 (or 0.4%)
2. flexipes root segments and 24 of 3060 (or 0.8%)
£. racemosa root segments lacked AM mycorrhizae. ANOVA revealed significant first-order effects of topographic position and sample date on the PRLC in 2. flexipes (Table
18). Overall, the mean PRLC of 2* flexipes roots decreased in the order: slope plots (66.5% ± 8.2) > uplands (60.2% + 7.6) > bottomlands (47.3% ± 6.0).
In general, AM infection levels in 2* flexipes were greatest in the summer and lowest in winter and early spring (Figure 6), with the greatest changes in infection occurring between April and June (increase) and November and February (decrease). The only exception to this general pattern was the population growing in the bottomlands at the CSU site, in which both April-June increase and the November-February decrease were less apparent in 1993. Thus, despite sufficient variation among sampling dates and years to produce significant interactions between sampling date and topography/year in the analysis of variance (Table 18), the same general patterns of variation among seasons and topographic Table 18.— Overall ANOVA for percent of root length colonized by AM fungi in Trillium flexipes and Smilacina racemosa.
Trillium flexipes
Source DF Sum Souares F-value P > F Yr 1 0.3 0.1 0.9182 Sample Site 1 86.2 3.2 0.0748 Topographic Position 2 41035.1 757.9 0.0001 □ate 8 12369.6 57.1 0.0001 Sample Site*Yr 1 29.9 1.1 0.2937 Sample Site*Topography 2 88.1 1.6 0.1972 Sample Site*Date 8 283.9 10.5 0.0001 Topography*Yr 2 1395.9 25.8 0.0001 Topography*Date 16 1617.5 3.7 0.0001 Yr*Date 8 616.1 2.8 0.0041 Total (model) 647 75698.9 44.9 0.0001
Smilacina racemosa
Source DF Sum Souares F-value P > F Yr 1 48.4 0.8 0.3706 Site 1 305.4 5.1 0.0248 Topographic Position 2 40570.1 336.5 0.0001 Date 8 11655.4 24.2 0.0001 Sample site*Yr 1 167.9 2.8 0.0957 Sample Site*Topographic Posn . 2 21.8 0.2 0.8345 Sample Site*Date 8 211.9 0.4 0.8974 Topographic Position*Yr 2 357.3 3.0 0.0524 Topographic Position*Date 16 6451.1 6.7 0.0001 Yr*Date 7 652.2 9.3 0.0001
Total (model) 611 103571.7 24.1 0.0001 99
positions were apparent in 11 of the 12 combinations of
year, site, and topographic position (Figure 6).
Smilacina racemosa— The roots of £. racemosa also
exhibited considerable AM development on all sample dates
(Figure 7). As was the case for X. flexipes. analysis of
variance revealed significant effects of topographic
position and date on AM infection (Table 18). Mean PRLC
decreased in the order: slope positions (71.0% ±7.0) >
uplands (64.0% ± 8.0) > bottomlands (49.1% ± 5.6), with
the means for all three topographic positions differing
significantly from the others. There was also a
significant difference in PRLC between forest sites,
although the actual difference between site means (WSU:
62.0% ± 12.3 vs CSU: 60.1% ± 12.1) was small.
£. racemosa plants growing in the upland and slope
positions exhibited the same cyclic pattern of infection
intensity as X. flexipes: however, unlike I. flexipes. the
spring increase in PRLC for £. racemosa continued into
August (Figure 7). In the bottomland positions, however,
the cyclic infection pattern was not present, and PRLC
varied little between September 1993 and November 1994
(Figure 7). This difference in the temporal pattern of AM
infection among sites led to the significant site-by-date variance component in the analysis of variance. 100 nscussioN
The predominantly ncnmycorrhizal roots of Cardamine
concatenata are consistert with the presumed
nonmycotrophic status of the Brassicaceae (Trappe 1987).
The colonization of AM ftngi that was observed in May
probably reflects events triggered by structural and
defense changes associated with root senescence (Tester at
al. 1987). As roots senesce, they lose their physical and
chemical defensive capabilities, thus allowing AM fungi to
colonize. The lack of arbuscules in the 3456 root
segments examined further suggests the opportunistic
nature of this fungal infection. Although numerous i studies have reported suah "apparent" AM development in
some species of the Brassicaceae (e.g., Medve 1983; Harley
and Harley 1987; DeMars and Boerner 1994), Brundrett and
Kendrick (1988) did not observe this phenomena in the
£, concatenata population they studied.
Significant seasonal variation in mycorrhizal
infection has been observed in several ecosystem types
(Rabatin 1979; Gay £& al. 1982; Giovannetti 1985; Dhillion et al. 1988; Gemma and Koske 1988). Mayr and Goday (1989) observed moderate seasonal variation in some herbs of
European deciduous forests. However, in Ohio, Boerner
(1986) observed no significant seasonal variation in the forest summer herbs Geranium maculatum L. and Polvqonatum pubescens Pursh, and in an Ontario forest Brundrett and
Kendrick (1988) observed high mean PRLC in Trillium
orandiflorum (Michx.) Salisb. and Trillium erectum L.
throughout the year without significant seasonal
variation. In contrast, Brundrett and Kendrick (1988) did
report a PRLC pattern for Smilicina racemosa that was
similar to that observed in the present study, i.e.
somewhat lower colonization in early spring and higher in
summer. Although seasonal variations in AM development may not be ubiquitous in these forests, where seasonal variation does exist, it is likely a response to temporal variations of environmental factors, or to phenological patterns of root production and turnover in the plants, or both.
The variations in AM among species, dates, and sites may have been due to interactions among several factors.
First, the lower levels of AM colonization in winter and early spring may have resulted from cooler soil temperatures during this time of year. Hayman (1983),
Chilvers and Daft (1982), and Anderson sfc al. (1987) have shown that mycorrhizal activity is limited at temperatures as low as 10° C, while Smith and Bowen (1979) showed that low temperatures inhibit AM initiation. Another factor potentially contributing to lower PRLC during early spring is the pulse of soil p released during this time. In our 1 0 2
study sites extractable soil P was highest during
February, March, and April for all topographic positions
in all years. Many authors have reported reduced AM
development with higher available P (e.g., Menge al.
1978; Thompson fit al. 1986; Amijee fit fil* 1989) or
increased AM colonization with relatively lower soil P
(Koide and Li 1990).
Despite the observed seasonal patterns in mean PRLC, colonization was moderate to high throughout the year (at
least 50% in uplands and slopes and 40% in bottomlands).
To some degree, this probably reflected root morphology, as both Trillium flexipes and Smlllclna racemosa have coarse, magnolioid root types which are generally highly dependent upon endomycorrhizal symbioses (Baylis 1975).
Root turnover and root phenology in these species may also contribute to seasonal variations in PRLC. The roots of X* flexipes and £. racemosa persist for at least 3 and
5 years, respectively (Brundrett and Kendrick 1988). Once the root is colonized, AM fungi probably remain in it for the duration of its existence, although the root length colonized may vary within it through time. Specialized root structures may also contribute to greater PRLC.
£. racemosa roots have longitudinal air channels which help AM hyphae spread within the root (Brundrett and
Kendrick 1990a). 103
Phenologically, new root production In both species occurs In early to mid-summer and roots elongate throughout the summer (Brundrett and Kendrick 1990b).
Summer root elongation might contribute to the relatively higher PRLC during these months as roots will intercept more AM inoculum as they move through soils.
Although there were statistically significant differences in AM development in plants from slope vs upland topographic positions, the magnitude of those differences was small and then probably not biologically significant. In contrast, the significant differences in
AM development between slope and upland vs bottomlands were considerably larger and likely great enough to suggest biological significance. Soils of the bottomland sites remain moister in summer (DeMars and Runkle 1992).
Laboratory experiments have demonstrated that available moisture has the greatest single effect on root colonization of all environmental variables considered
(Redhead 1975; Reid and Bowen 1979; Iqbal and Tauqir 1982) with high moisture levels contributing to lower mycorrhizal counts. Field studies have also shown lower
PRLC in moister months and in moister sites (Gerdemann
1974; Rabatin 1979). However, Dickman et al. (1984) did not observe this pattern. 104
Although direct root colonization by AM fungi has
not, heretofore, been studied along topographic gradients,
AM spore counts have. Anderson al. (1984) showed that
AM spore counts were lower in moister regions of a prairie
topographic gradient, and that some plants growing across
the entire gradient were nonmycorrhizal in the wet zone
and mycorrhizal in the dry zone, similarly, Anderson §£
al. (1983) showed decreasing spore counts in sites (not
associated with topographic differences) with greater soil
moisture. Although spore counts may not always correlate
directly with AM root colonization patterns (Allen and
Allen 1980), these studies of spore counts also support
our initial hypothesis that AM development would be more
important in higher, drier sites than lower, moister ones.
The observation that PRLC was significantly lower in
the bottomland sites is, however, counter to what would be
expected from observed extractable soil P from this site.
As previously mentioned, experiments altering soil P
levels generally have shown increasing root colonization with decreasing P. Evidently, the effect of soil moisture
(and perhaps other unknown factors) in our sites must have played a more significant role than P availability in regulating the PRLC.
In conclusion, our observations support the hypothesis that AM development is lower in moister 105 bottomland topographic positions than in better drained topographic positions. However, our results do not support the hypothesis that AM development would increase with decreasing P availability. In our sites, therefore, moisture appeared more important than P in regulating AM development. Also, we have shown that there is considerable seasonal variation in mean PRLC for both
I. flexipes and £. racemosa. especially between cold and warm periods of the year, probably reflecting interactions between species phonological patterns, seasonal temperature and moisture variations, and root morphology. CHAPTER VI
Arbuscular Mycorrhizal Fungi Colonization in
Caosella bursa-pastoris (Brassicaceae)
INTRODUCTION
Historically, several plants families including the
Brassicaceae and Chenopodiaceae have been considered to be
nonmycorrhizal (Gerdemann 1968). Previous studies have
shown that arbuscular mycorrhizae development in typically
nonmycorrhizal plant species can be induced by AH fungi
colonization from nearby mycorrhizal species when grown in
pot culture or sampled under field conditions. This
phenomenon has been observed in members of the
Chenopodiaceae (Miller et al. 1983), Brassicaceae (Ocampo
et al. 1980) and the nonmycorrhizal legume genus Luoinus
(Trinick 1977) when grown in pot cultures. Field studies
on Atriolex confertifolla. a typically non-mycorrhizal
species, have also shown greater AM development in
individuals located in sampling plots containing grass and
other mycorrhizal species (Miller fit al. 1983). Similar
results were reported for Luoinus cosentinli when growing
in the presence of clover (Trinick 1977). To date similar plant association effects have not been reported for any
106 107
Brassicaceae under field conditions.
The purpose of this study was to determine whether AM fungi colonize Capsella bursa-pastoris (L.) Medikus.
(Brassicaceae) in the field, and whether colonization is influenced by the presence of a neighboring mycorrhizal root matrix. To accomplish this, we sampled £. bursa- pastoris from two habitat types - lawn and waste areas.
We chose £. bursa-pastoris because previous studies indicated that this species might develop AM relationships
(Kruckelmann 1975; Saif and Iffat 1976; Saif al. 1977).
We sampled from April to July to determine if there were temporal differences in AM fungal presence during the growing season.
MATERIALS and METHODS
Capsella bursa-pastoris is a cosmopolitan ubiquitous annual or winter-annual (Gleason and Conquist 1991). In
Ohio, it begins blooming in February and continues blooming and setting fruit until autumn. We sampled
£. bursa-pastoris from lawn and disturbed-waste habitats in three sites located in Franklin County, Greene County and Cuyahoga County, Ohio. The Franklin Co. site was located on the Ohio State University campus. The lawn habitat sampled was the grassy field east of Mount Hall and adjacent to an agricultural field. The waste habitat sampled was a heavily compacted portion of an agricultural
field which had been fallow for two years. The Greene Co.
site was located on the Wright State University campus.
The lawn habitat sampled was the grassy field located
south of the "K" parking lot. The waste habitat sampled
was the adjacent former construction area which supported
a robust weed community. The Cuyahoga Co. site was
located on the Cuyahoga Community College Western Campus
on the western portion of the property. Here, a large
population of £. bursa-pastoris existed in both a lawn-
field border and an adjacent bulldozed, compacted area.
Ten 2m x 2m plots were established in each habitat-site in
April 1992. Three plants from each plot were harvested in
late April, May, June and July (30 plants per habitat).
In the laboratory, roots were separated, cleared and
stained with trypan blue (Phillips and Hayman 1970). Root
samples were stored in plastic tissue cassettes and
preserved in FAA (formalin:acetic acid:ethanol). Roots were cut into approximately 2-cm segments and the percent
of segments infected was determined (Giovannetti and Mosse
1980). AM fungi colonization was estimated by examining
10 random root segments per plant. Infection was defined as the presence of any internal hyphae, vesicles, arbuscules or chlamydospores. Roots without cortex or roots with severe damage were not used. Two-way analysis 109
of variance (Statistical Analysis System 1985) was used to
examine whether plot, habitat, site or date (and their
interactions) had affected AM fungi colonization in this
species.
RESULTS
Caosella bursa-pastoris plants were colonized by AM
fungi in all three sites on all four sample dates.
Internal hyphae, vesicles, extramatrical hyphae and
chlamydospores were common. In contrast, no arbuscules
were observed in any of the 7200 root segments examined.
Percent colonization ranged from 10% to 43% (Table
19). Mean colonization was significantly greater in lawn
habitats than in waste habitats (Lawn mean = 35%; waste
area mean » 15%; £ < 0.0001; Table 20). There were also
significant temporal differences in colonization (£ <
0.05) in lawn habitats, the greatest colonized percentage was in April (mean « 27%) and the least was in July (mean=
21%). Mean colonization was significantly lower (£ <
0.0001) in plants from the Cuyahoga County site than in plants from the other two sites; thus, there were significant site-by-date and site-by-habitat interactions
in the analysis of variance (ANOVA). Table 19.— Percent AM fungi colonization in Capsella bursa-pastoris. N=30 for each habitat x date combination; standard deviation of the mean is given in parentheses.
Franklin Co. Greene Co. CuyahogaL Co. Month Lawn Waste Lawn Waste Lawn Waste
April 43 (23) 18 (21) 43 (31) 19 (19) 30 (17) 11 (15) May 31 (23) 18 (16) 39 (23) 15 (16) 28 (20) 14 (13) June 38 (15) 16 (16) 46 (25) 15 (14) 31 (17) 10 (12)
July 33 (19) 19 (12) 32 (18) 13 (13) 21 (16) 10 (10) 110 Table 20.— ANOVA of percent AM fungi colonization in Capsella bursa-pastoris in three Ohio sites.
SOURCE df Sum of F-value
squares
SITE 2 105.4 15.8** DATE 3 37.3 3.7*
HABITAT 1 706.1 212.3" PLOT 9 49.0 1.6 DATE*HABITAT 3 28.5 2.9*
SITE*HABITAT 2 22.2 3.3*
* signifies at £ > 0.05; • • signifies at E > 0.0001 Ill 112
DISCUSSION
These results Indicate that AM fungi colonized
Capsella bursa-pastorls in the populations sampled in
Ohio. Also, our data showed that association of £. bursa- pastorls with typically mycorrhizal plants in lawn habitat increased the probability of AM fungal colonization.
Conflicting reports of AM development in this species have been reported elsewhere. Saif and Iffat (1976) and Saif et al. (1977) reported AM development in Pakistan populations of £. bursa-pastorls. whereas Medve (1983) observed no AM development in Pennsylvania or California populations. AM development in other members of the
Brassicaceae has been reported from field and laboratory studies (Hirrel &I. 1978; Medve, 1983; Peterson al.
1985; Tommerup 1984). In addition, Newman and Reddell
(1987) reported that of the 63 Brassicaceae species examined for AM development, 13 have been reported capable of developing AM associations. Thus, sporadic establishment of AM fungal colonization may be widespread in the family.
The observation that AM fungi colonization in
£. bursa-pastoris is higher in lawn habitat is consistent with similar studies on other typically nonmycorrhizal plants. Miller g£ al. (1983) reported that Atrlolex confertifolla had a five times greater percent infection 113 when sampled in grass community matrix than when sampled
from plots where it grew solitary. Trinick (1977) reported similar increases in nonmycorrhizal Luoinus spp. when grown in association with red clover* similarly,
Ocampo al* (1980) reported increased AM development in several agricultural Brassicaceae taxa when grown in the presence of a mycorrhizal host under greenhouse conditions. Clearly, the results of these and the present study demonstrate that AM fungi colonization is affected by the community matrix in which a plant grows.
AM fungi colonization may result from interaction between roots and chlamydospores, colonized living roots, or colonized dead roots (Smith and Gianinazzi-Pearson
1988). The presence of a dense mycorrhizal root matrix, such as in a grass community, would greatly increase the probability of interaction between typically nonmycorrhizal plant roots and AM fungi inoculum. This suggests that the low development of AM in nonmycorrhizal species is due to an external phenomenon such as low root exudation (Tester ££ &1. 1987). However, internal factors must also be considered as AM fungi penetration can usually be observed in typically nonmycorrhizal species, only to abort later (Allen e£ sJL. 1989).
The significance of AM development in £. bursa- pastoris and other typically nonmycorrhizal plants is not 114 known. Kolde and Schreiner (1992) would classify this relationship as a nonfunctional symbiosis because all root segments in our samples lacked arbuscules. Hence, the development of AM in £. bursa-pastorls may only represent a plant association artifact. Research examining the functional significance of AM in typically nonmycorrhizal species is required to address this question more clearly. CHAPTER VII
Arbuscular Mycorrhizal Development in the Brassicaceae in
Relation to Plant Life Span
INTRODUCTION
Most herbaceous plant species are capable of forming arbuscular mycorrhizae (AM). However, some families, including the Brassicaceae (Cruciferae), Chenopodiaceae, and Caryophyllaceae are described as being nonmycorrhizal
(Gerdemann 1968; Trappe 1987). The lack of AM development in species in these families may reflect chemical barriers, including fungicidal excretions such as glucosinolates (Schreiner and Koide 1993a; 1993b), low root exudation (Schwab fit slL. 1983), or lack of signaling compounds (Koide and Schreiner 1992). Cell wall structure and physiology may also play a role (Tester fit £l1« 1987).
Despite the categorization of the Brassicaceae as nonmycorrhizal, AM in crucifers has been reported from 80 taxa (Table 21). However, because AM development does not occur consistently in any one species, considerable uncertainty still exists concerning the mycorrhizal status of this family (Harley and Harley 1987).
115 116
Life history patterns nay be correlated with development of AM in those species from nonmycorrhizal families which do form AM. For instance, many of the first crucifers examined for AM development have been annuals characteristic of early successional and disturbance communities. Such communities often have low mycorrhizal inoculum potential (Moorman and Reeves 1979;
Janos 1980; Biondini fit fil. 1985; Allen 1991) and, consequently, would not give rise to high levels of mycorrhizal colonization. Hence, the early characterization of the Brassicaceae as nonmycorrhizal has been based primarily on examination of individuals growing in sites where AM would be expected to be low.
The purpose of this study was to examine the hypothesis that life history patterns play an important role in determining the AM status of various crucifers.
More specifically we examined the hypothesis that shorter- lived taxa (annuals, winter annuals, or monocarpic perennials) are predominantly nonmycorrhizal or, at the least, have less AM development than longer-lived perennials. To test this hypothesis, we combine an extensive survey of data from the literature with experimental glasshouse inoculation studies on 649 taxa from the Brassicaceae. Table 21.— Crucifer taxa examined for AM development as reported in the literature1 B»tggflciiMcgKgggB«iBK«»ggttKBBa8Bg3gtt»cgsg«ggssaBagag8agaaaB3Bgaaag3sgaggaa»»«iiflga3gaaa8«»a«33a«»aa»iaa«a»a Taxon AM* Life* Citation* Alliaria petiolata (M. Bleb.) Carva and Grande - a/b Medve 1983 Alyaatut bertolonii Deav. - P Lioi and Giovannetti 1989 Arabidopaia thaliana (L.) Heynh. + a sense Harley and Harley 1987 Arabia alplna L. - P sensu Barley and Harley 1987 Arabia hirsute (L.) Scop. - b aenau Barley and Harley 1987 Arabia holboelli Hornem. — P Currah and Van Dyk 1986 Arabia laevigata (Kulenb. ex Willd.) Poiret - b Medve 1983 Arabia lyrata L. — b/p Medve 1983 Arabia nipponlca H. Boies. — P Kim et al. 1989 Armoracia ruaticana (Lam.) Gaertner + P sensu Harley and Harley 1987 Barbarea vulgaris R. Br. + b Arnold and Kapustka 1987 Barteroalla maximovlczii (Palib.) O. E. Schulz + b Kim et al. 1989 Biacutella laevigata L. + P sensu Harley and Harley 1987 Brassica campestrla L. + a sensu Tester et al. 1987 Braaalca caulorapa Paaq. - a Medve 1983 Braaaica juncea (L.) Czern. - a Harinikumar and Bagyaraj 1988 Braaalca aapobraaalca Mill. - a Medve 1983 Braaaica napua L. + a sensu Harley and Harley 1987 Braaalca nigra (L.) Koch. + a Medve 1983 Braaaica oleracea L. var, oleracea + b/p sensu Harley and Harley 1987 Braaalca oleracea L. var. botrytia — a Medve 1983 Braaaica oleracea L. var. capitata - a Medve 1983 Braaaica rapa L. — a aensu Harley and Harley 1987 Braaaica toumefortii Gouan + a sensu Tester et al. 1987 Cakile edentula (Bigelow) Hook. — a Medve 1983 Cakile maritime Scop. + a sensu Harley and Harley 1987 Capaella bursa-pastorls (L.) Medic. + a/b aensu Harley and Harley 1987 Cardamine amara L. — P sensu Harley and Harley 1987 Cardamine bulbifera (L.) Crantz + P sensu Harley and Harley 1987 Cardaaine concatenate (Michx.) O. Schwarz* — P Medve 1983 Cardaoine diphylla (Michx.) Alph. Hood* - P Medve 1983 Cardamine douglaaaii Britton — P Medve 1983 Cardamine flexuoaa With. + a/b sensu Harley and Harley 1987 Cardamine hirauta L. + a sensu Harley and Harley 1987 Cardamine impatiena L. - a/b sensu Harley and Barley 1987 117 Table 21. (Continued)
Taxon Life* Citation4 Cardamine leucantha (Tauach) O. E. Schulz + P Kim et si. 1989 Cardamine penaylvanica Huhlenb. — b Hedve 1983 Cardamine pratenais L. — P sensu Harley and Harley 1987 Cardamine trifolia b. — P sensu Harley and Harley 1987 Choriapora tenella (Pallas) DC. — a sensu Hedve 1983 Cochlearia anglica b. + b sensu Harley and Harley 1987 Cochlearia danica L. + a/b sensu Harley and Harley 1987 Cochlearia officinalis L. + b sensu Harley and Harley 1987 Descurainia pinnata (Halt.) Britt. — a sensu Medve 1983 Diplotaxia erucoidea (L.) DC. + a/w sensu Harley and Harley 1987 Diplotaxia muralis (L.) DC. + a/b sensu Harley and Harley 1987 Draba oligoaperma Hook. — P Currah and Van Dyk 1986 Dr aba reptana (Lam.) Pern. — a Hedve 1983 Erophila vema (L.) Chevall. - a sensu Harley and Harley 1987 Eryaimm cheirantholdea b. — a/w sensu Harley and Harley 1987 Heliophila africana (1*.) Karlas — a Allsopp and Stock 1983 Beliophila arenaria Sond. — a Allaopp and Stock 1983 Heaperia matronalia b. — b Hedve 1983 Iberia amara b. + a sensu Harley and Harley 1987 Isatie tinctoria b. — b sensu Harley and Harley 1987 bepidium bidentatum Kontin + P Kooks et ai. 1992 bepidium campeatre (L.) R. Br. — a/b Hedve 1983 bepidium perfoliatum L. - a/b sensu Hedve 1983 bepidium ruderale L. — a/b sensu Harley and Harley 1987 bepidium Berra H. Mann — P Koske et al. 1992 bepidium virginicum b. + a/b sensu Tester et al. 1987 bobularia maritime (L.) Desv. + P Hedve 1983 Lunar la annua I>. + a Medve 1983 Matthiola £1avida Bo Ib s . - P sensu Medve 1983 Nasturtium officinale R. Br. — P sensu Harley and Harley 1987 Phyaaria acutifolia Rydb. — P sensu Hedve 1983 Phyaaria didymocarpa (Hook.) A. Gray — P Currah and Van Dyk 1986 Phyaaria floribunda Rydb. — P sensu Medve 1983 Raphanus acanthiformis Horel — a/b sensu Hedve 1983 Raphanus aativua L. + a/b sensu Harley and Harley 1987 118 Table 21. (Continued)
Taxon fin1 Life9 Citation4 Raphnua raphanlstrum L. — a/b aensu Harley and Harley 1987 Rorippa lndica (L.) Hiern. + a/b aensu Hedve 1983 Rorippa paluatrls (L.) Besaer - any aensu Harley and Harley 1987 Slnapla alba L. — a senau Harley and Harley 1987 Sisymbrium loeselli L. — a/b sensu Harley and Harley 1987 Sisymbrium officinale (L.) Scop. + a aenau Harley and Harley 1987 Sisymbrium orientals L. + a/b aenau Teeter et 1987 Snelouskia calyclna (Stephen.) C. A. Hey. — P Currah and Van Dyk 1986 Stenopetalnm lineare R. Br. ex DC. + a aenau Tester et al. 1987 Thlaspi arvenae L. - a/b aenau Harley and Harley 1987
1 Includes both field and greenhouse-inoculation reports 9 Arbuscular mycorrhizae + (reported as present) - (reported as negative) 9 Life history a= annual; w» winter annual; b* biennial; p= perennial
4 Where conflicting reports of AH occur for a taxon, only positive reports are scored; where multiple reports occur, the moat recent citation is given 9 formerly Dentaria laclnlata Kuhl. and Dentaria diphylla Michx., respectively. 119 120
MATERIALS and METHODS
Seeds from a total of 1233 crucifer taxa were
obtained from sources representing all continents except
Antarctica. During 1992-1994, sets of seeds of each taxa
were sown in sterilized sand and inoculated with a Glomus
Intraradices Schenck and Smith root-based inoculum in 6 cm
x 6 cm x 7 cm plastic potting containers with 4 replicate
pots per taxon. Resulting seedlings were thinned to two
or three individuals per container following germination.
Taxa that did not germinate were cold-treated and resown.
A total of 649 crucifer taxa germinated and produced
sufficient root mass for examination of AM development.
In the greenhouse, plants were watered as needed for
5 weeks, during which time no fertilizer was applied.
Roots were then harvested, carefully washed, and examined
to ensure all initial root inoculum was removed. This
step was crucial since contamination of our crucifers
roots with inoculum-containing roots could have confounded
assessment of AM development.
In the laboratory, roots were separated, cleared and
stained with trypan blue (Phillips and Hayman 1970) and
stored in plastic tissue cassettes with FAA
(formalin:acetic acid:ethanol) preservative solution. AM development was examined in 5 random 2-3 cm root segments per plant, and was defined as the presence of any internal 121 non-septate hyphae, vesicles, arbuscules or chlamydospores. Aborted appresoria were not included in this definition. A taxon was scored as positive if more than one examined root segment possessed any of the AM structures. Roots without cortex and roots with obvious damage to the cortex were discarded.
A chi-square (X3) test was performed on all taxa scored as positive to examine whether life span patterns affected AM development. Shorter-lived taxa were tested against longer-lived taxa for AM development or lack thereof. Taxa which spanned the entire life history range excluded from this analysis. Plant longevity patterns were determined from regional floras, and nomenclature follows those who supplied the seeds.
A second X3 test was performed on all taxa reported to have been examined for AM in the literature.
Literature reports of AM development include both field collections and greenhouse-inoculation studies. In several instances conflicting reports of AM development exist. In such cases, the taxa have been scored as mycorrhizal. Where duplicate reports existed, only the most recent citation was listed (Table 21).
Finally, a third X3 test was performed on the data presented for the Brassicaceae by Harley and Harley
(1987), who reported 96 instances where AM development has 122 been assessed in 40 taxa of crucifers in the United
Kingdom. This test incorporated multiple and conflicting reports for the same taxa.
RESULTS
Of the 649 crucifer taxa we experimentally inoculated, 123 (19.0%) showed some degree of AM development (Tables 22 and 23). The typical AM development we observed included both internal hyphae and vesicle development; however, no arbuscules were observed in any of the 3245 root sections we examined.
Of the 83 genera in which we were able to test more than one species, 19.0% of the species showed some AM development and 81.0% showed none (Tables 22 and 23). Of those 83 genera, 37 (44.6%) showed no AM development. AM development in all tested species within a genus occurred only three (3.6%) times: Bunias (2 species tested),
Chorisoora (2 species tested) and Goldbachia (2 species tested).
AM development was detected in a total of 46 genera.
In these genera an average of 33.3% of the species within a genus were mycorrhizal, and in most all of these genera the percentage of species tested showing some AM development was <50% (Figure 8). Of the 24 genera in which we were able to test at
least seven species, 19.7% of the species showed some AM
development. Interestingly, only two genera in this
subset (8.3%) had no species with detectable AM
development: Braya and Morlcandla (both with 7 species
tested). Because the proportion of apparently
nonmycorrhizal genera was so much lower in this subset of
more intensively-tested genera, we thought a relationship might exist between the number of species tested and the
likelihood that mycorrhizal species would be found.
However, there was no significant correlation of the number of species tested with the percentage that developed AM (Figure 9), with an overall average of 21.5% of the species tested within a genus being mycorrhizal.
Thus, no matter how we stratified the data, we arrived at an overall percentage range of 19.0%-19.7% of crucifer taxa showing at least some AM development, and within genera with at least one AM species, a range of 21.5%-
33.3% of species within a genus being mycorrhizal.
To test the hypothesis that AM development would be greater in longer-lived species, we compared the percentage of species showing AM development between 348 shorter-lived species (annuals and shorter-lived monocarpic perennials) and 292 longer-lived taxa
(perennials). Among the shorter-lived taxa, 58 of 348 124
(16.6%) were mycorrhizal, compared with 64 of 292 (21.9%) of the longer-lived taxa (X2=2.606, p>0.10). We then restricted the analysis to those species in which we observed significant AM development (Table 23). Of those,
58 (47.2%) were shorter-lived and 64 (52.8%) were longer- lived. Similarly, of the 520 taxa in which we observed no
AM development, 55.5% were shorter-lived species and 43.0% were longer-lived. I Our review of the published literature indicated that
31 of the 80 species (38.8%) examined to date had been reportedJ as mycorrhizal at least once. Of the 53 with relatively short lifespans, 24 (45.3%) were mycorrhizal, as were 7 of the 26 (26.9%) longer-lived species
(X*m2.466, p>0.11). Of the 95 crucifers assessed by
Harley and Harley (1987), 21 of the 71 shorter-lived taxa
(29.6%) and 8 of the 24 longer-lived taxa (33.3%) were mycorrhizal (Jf*“0.0l2, p>0.73). 125
Table 22.— The occurrence of AM development In greenhouse- inoculated crucifers. Zflssn Llfe-hlBtorv Aethlonema arablcum (L.) Andrz. ex DC. a Aethlonema cordatum (Desf.) Boise. p Aethlonema gcandiflocum Boise. £ Hohsn. p Alyaaum paterl Nyar p Alyeeum marltlmum Lam. p Alyaeum montanum L. p Alyaaum murale Waldst. £ Kit. p Alyaaum atrictum Willd. a Anchonlum elichryBifolium (DC.) Boles. p Arabia alplna L. p Arabia crucieetoBa Canstance £ Rollins p Arabia miaaourlenala E. L. Oreone b Arabia perennans S. Hats. p Aubrieta caneacena (Boies.) Bocnm. p Aubrieta deltoidea (L.) DC. p Barbaraa vulgaris R. Br. b Blacutella Intermedia Couan p Blacutella laevigata L. p Braaalca elongate Ehrh. b Braealca napua L. var. olelfera a Braealca repanda (Willd.) DC. p Braealca tournefortil Couan a Bunlaa erucago L. a/b Buniaa orlentalia L. b Camellna rumellca Velen. a Capaella bursa-pastorls (L.) Medic. b Caradamine anguatata O. B. Schulz p Cardamine brewer1 L. p Cardamine bulboaa Schreber ex Muhlenb. p Cardamine bulbifera (L.) crantz p Cardamine callfornica (Nutt ex Torrey fi A. Gray) E. L. Greene p Cardamine flexuoaa With. a/b Cardamine hirsute L. a Cardamine lyrata Bunge p Cardamine obligua Kochet. ex Richard p Cardamine penBylvanica Huhlenb. ex Hilld. b Cardamine pratenala L. p Cardamine trlfolla L. p Cardamine uliginose M.B. p Cardaminopaia hlspida (Myg.) Hay p Cardaria draba (L.) Desv. p Carrichtera annua (L.) DC. a Caulantbua flavescens (Hook.) Payson a Chorispora elegans Camb. p Choriapora tenella (Pallas) DC. a Clypeola jonthlaapi L. a Cochlearia officinalis L. b Conringla orlentalia (L.) Dumort a/w Crambe filiformia Jacg. p Cryptoapora falcate Kar. £ Kir. a Deacuralnia argentine 0. E. Schulz a Deacurainia callfornica (A. Gray) 0. E. Schultz b Deacuralnia inciaa (Engelm. ex A. Gray) Britton a Deacurainia pinnata (Halt.) Brit. a Diplotaxia griffithii (Hook. f. £ Thoms.) BoIbs. a 126
Table 22. (Continued) Draba alzoldea L. P Draba aoprella B. L. greene P Draba cualckll Robinson ex O. E. Schulz P Draba helleriana B. L. Greene b/p Draba hiapanica Boies. P Draba muralia L. a/b Draba norveglca Gunnerue P Draba payaonli J. F. Hacbr. P Draba verna L. a Dryopetalon runclnatum A. Gray any Enarthrocarpue lyatua (Forsk.) DC. a Bramobium aegyptlacum (Spreng.) Boies. P Erucaria craaaifolla (Foreek.) Del. a Brucaria hiapanica (L.) Druce a Erucaria mlcrocarpa Bo JLb s . a Eryalmum anguatatum Rydb. P Erysimum chelrantholdea L. a/w Eryalmum dlffuaum Bhrh. b/p Erysimum helveticum (Jacg.) DC. P Eryalmum llnlfolium (Pourr. ex Pare.) Gay P Eryalmum myrlophyllum Lange P Eryalmum repandum L. a Euclidlum ayrlacum (L.) R. Br. a Euromodendron bouraaanum Coeeon P Flbigla auffrutlcoaa (Vent.) Sweet P Fortuynia bungai Boise. P Geococcus puaillua J. L. Drumm a Ooldbachla laalgata DC. a Ooldbachla verrucosa Komar. a Hellophila atrlcana (L.) Karais a Hallophlla elata Sond. a Hesperia kotachyana Fenzl. b/p Hesperia matronalla L. b/p Haaparla rupeatrla Bornm. P Horvoodla dicfcsoniae Turrill a Iberia amara L. a/b Iberia saxatllla Torner p Iberia aempervlrana L. P laatia djurjuraa Coss. & Dur. P Koniga maritime (Daev.) R.Br. P Lepldlum draba L. P Lepldium papillosum F. Huell a Lepldlum aatlvum L. a LoJbularia maritime (L.) Desv. P Lunarla redivia L. P Halcolmia oralnlana (Ten.) Ten. a Mareaia nana (DC.) Batt. a Hatthiola icana (L.) R. Br. P Ifattftiola longipatala (Vent.) DC a Hatthiola tricuspidata (L.) R. Br. a Menkea auatraliaa Lehm. a Moriatta caneacena Boiss. P Healla paniculate (L.) Desv. a Peltaria perennia (Ard.) Mgf. P Ptilotrichum lapeyrousianum (Jordan) Jordan P Rorippa curvialllqua (Hook.) Bessey ex Britton a/b Rorippa BtyloBa (Pera.) Kanaf. et Rothm. a Sinapia alba L. a 127
Table 22. (Continued)
Siaymbrlum auatriacum Jacq. Sisymbrium hirautum Lag. ex DC. Sisymbrium atvlctiaaimum L. stanopatalum tilifolium Benth. Taaadalla nudlcaulia (L.) R. Br. Thlaapl arvense L. Thlaapl montanum L. Valla paeudocytiaua L. Waaabla japonica (Miq.) Matsum. (Hook. f. & Thoms.) Korah. Winklara allalfolla tr» ’O'O'O’O »n3 » cr a" annual; w" winter annual; b>biennial; p«perennial Figure 8. Percent of mycorrhizal species observed in genera with 2 or more species tested in glasshouse inoculation trials.
128 129
GENERA WITH TWO OR MORE SPECIES TESTED
0 1-25 26-50 51-75 76-100 PERCENT OF MYCORRHIZAL SPECIES
Figure 8 Figure 9. Percent of mycorrhizal species in genera with 7 or more species tested in relation to the number of species per genus tested in glasshouse inoculation trials.
The best fit regression line and 95% confidence intervals are indicated.
130 Figure 9 Figure
PERCENT OF 0 s tS3 CO O 10 CJ O 40 W 30 j 0 5 o c O w < h
NUMBER OF SPECIES/GENUS TESTED OF SPECIES/GENUS NUMBER GENERA WITH GENERA SEVEN OR MORE SPECIESMORE TESTED 20 30 40 50 60 131 132
Table 23.‘—Brassicaceae taxa not exhibiting AM fungi colonization following experimental inoculation.
Taxon Life scan Aethlonema carneum (Banks & soland.) B. Fedtsch. a Aethlonema elongatum Boise. P Aethlonema flmbrlatum Boiss P Aethlonema saxatlle (L.) R. Br. b Aethlonema trlnervlum (DC.) Boise. P Alllarla petlolata (H. Bieb.} Carva & Grande w/b Allyaum dasycarpum Steph. ex Willd. a Alyaaoldea utrlcvlata (L.) Hoench P Alyaaum alyaaoldea (L.) L. a/b Alyssum condensation Boiss. & Hausskn. P Alyaaum contemptum Schott & Ky. a Alyaaum daaycarpum Steph. ex Willd. a Alyssum doaortomm Stapf a Alyssum heterotrlchum Boiss. a Alyssum linlfollum Steph. ex Willd. a A lyaaum marginatum Steud. ex Boiss. a Alyssum menlocoidoa Boise. a A lyaaum minua (L.) Rothm. a Alyssum mlnutum Schlecht ex DC. a Alyssum obovatum (C. A. Hiller) Turc2 . P Alyaaum aaxatile L. P Alyssum aerpylllfolium Deaf. P Alyssum atvlgoaum Banks & Sol. a Alyaaum axowltaianum Fisch & C. A. Hey a Amoracla lapathifolla Gilib. P Anaatatlca hlerocuntica L. a Anelsonla eurycarpa (A. Gray) J. F. Hacbr. & Payson P Aphragmus eachacholtaianua Andrz. ex DC P Arabidopala pumila (Steph.) N. Busch. a Arabidopala aalauglnea (Pallas) N. Busch a/b Arabidopala thaliana (L.) Heynh. a Arabidopala walllchll (Hook. f. Thoms.) B. Busch. b Arahls aculeolata E. L. Greene P Arahls albida (Stev.) Poech P Arahls blepharophylla Hook, fi Arn. P Arahls brewer1 S. Wats, P Arahls canadensis L. b Arabia caucaalca Willd. P Arabia cobrenaia H. E. Jones P Arabia cualckll S. Wats. P Arabia demiaaia E. L. Greene P Arabia drummondil A. Gray b/p Arabia fendleri (S. Wats.) E. L. Grenne P Arabia fernaldiana Rollins p Arabia fruoticoaa A. Nela. P Arabia iurcata S. Wats P Arabia georgiana Harper b Arahls gerardil Besser P Arahls glabra (L.) Bernh. b/p Arabia gracilipea E. L. Greene P Arabia hirsute (L.) Scop. b Arahls holboelli Hornem. b/p Arabia koehlerl T. J. Howell P Arahls laevigata (Hulenb. ex Willd.) Poiret b Arabia lemmonil S. Wats P 133
Table 23. (Continued)
Arabia lyallll S. Hats. p Arabia lyrata L. b/p Arabia microphylla Nutt, ex Torrey £ A. Gray p Arabia mlcroaparma RollinB b/p Arabia nova Vill. P Arabia nuttallll Robinson P Arabia patens Sullivant b Arabia pendulina E. L. Greene p Arabia paretallata B. Braun P Arabia platyaperma A. Gray p Arabia puberula Nutt, ex Torrey £ A. Gray b/p Arabia pulchra H.E. Jones ex S. Hats. p Arabia ractlaalma B. L. Greene b Arabia repanda S. Hats P Arabia aalbyi Rydb. p Arabla aerotina Steele b Arabia aerrata Pr. £ Sav. any Arabia ahortll (Fern.) Gleason b Arabia aparaiflora Nutt, ex Torrey fi A. Gray p Arabia auffruteacena S. Hats P Arabia tibetica Hook. f. £ Thoms. a Arabia turrita L. b Arabia varna (L.) R. Br. a Arabia x dlvarlcarpa A. Nels. b/p Aubrleta columnae Guss. p Aurinia petraea (ard.) Schur. p Aurinia aaxatiiis (L.) Desv. P Barbarea arcuata Rchb. b Barbarea intermedia Boreau b Barbarea minor C. Koch P Barbarea orthoceraa Ledeb. b/p Barbarea atrlcta Andrz. b Barbarea verna (Hiller) Asch. b Berteroa incana (L.) DC. a/b Berteroa mutabills (Vent.) DC. b/p Berteroella maximowlcxii (Palib.) O. E. Schulz b Biacutolla aurlculata L. a Blacutella dldyma L. a Blacutella fruteacena Cosson p Biscutella valentina (L.) Heywood p Blennodla pteroaperma (J. Black) J. Black a Braaaica adpreaaa Boiss. a/b firassica balearica Pars. p Braaaica barreiieri (L.) Janka a Braaaica cretica Lam. p Braaaica deflexa Boiss. a Braaaica irutlculoaa Cyr. b Braaaica juncea (L.) Czern. a Braaaica nigra (L.) Koch a Braaalca oleracae L. var. gongylodea L. b Braaaica oleracea L. var. acephala DC. b Braaaica oleracea L. var. botrytla L. f. b Braaalca oleracea L. var. buliata DC. f. b Braaalca oleracea L. var. capitate L. b Braaalca pubeacena L. P Braaaica rape L. subsp. chinenaia (L.) Oost a Braaalca rape L. eubsp. ollfera DC. a Braaaica rapa L. subsp. peklnenaia (Lour.) Oost a 134
Table 23. (Continued)
Braaaica. rapa L. subsp. aylveatrla (L.) Janchen a Braya alpina Sternb. & Hoppe P Braya farnaldll Abbe P Braya glabella Richardson P Braya humllla (C. A. Heyer) Robinson P Braya longll Fern. P Braya oxycarpa (Hokk. f. & Thoms.) Jafri P Braya thorlld-vulffil Ostenf. P BroaBardia papyaceae Boiss. P Cakile Arablca Velen. fi Bornm. Cakile conatricta Rodman Cakile ecfentula (Bigelow) Hook. Cakile lanceolate (Willd.) O. E. Schulz Cakile maritime Scop. Calepina irregularis (Asso) Thell. /b Camelina alyaaum (Miller) Thell. Camelina hiapida Boiss. /b Camel in a microcarpa Andrz. Camelina piloaa N. Camelina native (L.) Crantz /w Capaella peCraea Fr. Capaella procumbena (L.) Fries Capaella rubella Reut. i/b Cardamine africana L. P Cardamine amara L. P Cardamine angulata Hook. P Cardamine auriculata S. Wats. a Cardamine bellldlfolla L. P Cardamine clematltua Shuttlew. ex A. Gray P Cardamine concatenata (Michx.) 0. Schwarz P Cardamine cordlfolla A. Gray P Cardamine douglaaali Britton P Cardamine tlaccida Cham, fi Schlecht P Cardamine praca L. a Cardamine gunnil Hewson P Cardamine AeptapAylla (Vill.) O. E. Schulz P Cardamine impatlena L. a/b Cardamine lllacina Hook. any Cardamine macrocarpa Brandegee a/w Cardamine maxima (Nutt.) Alph. Wood P Cardamine nuttallll E. L. Greene P Cardamine parvltlora L. a Cardamine paucijuga Turcz. a Cardamine pentaphyllos (L.) Crantz P Cardamine rotundifolla Michx. P Cardamine trichocarpa Hochst. a Cardaminopala petraea (L.) Hiit. P Caulanthua glaucus S. Wats. P Caulanthua heterophyllua (Nutt ex Torrey & A. Gray) a CaulantAus laalophyllua (Hook. G Arn.) Payson a Caulanthua major (M. E. Jones) Payson P Chalcanthua renlfoliua (Boiss. & Hohen.) Boiss. P Chamira clrcaeoldea (L.f.) A. Zahlbr. a Chelranthua cheiri L. P Clausia turkeatanica Lipsky P Clypeola aapera (Grauer) Turrill a Clypeola cyolodontea Delile a Clypeola microcarpa Moris a Table 23. (Continued) Cochlearia angllca L. Cochlearia eaxatllle L. Colncya motional a (L.) Greuter & Burdet Coluteocarpua veraicarla (L.) Holmboe Conrlnga perelca Boiaa. Conrlnga planlalllqua Finch. & C.A. Hey. Conrlngia auatriaca (Jacq.) Sweet Conrlngla perfoliate (C. A. May) Busch. Corispora lberlca (H.B.) DC. CoronopuB dldymuB (L.) Sm. Coronopua integrlfoliue (DC.) Prantl. Coronopue procumbene 01lib. Coronopua squamatua (ForsBk.) ABChera. Crambe abyaainlca Hochat. Crambo hiapanica L. Crambe kotachyana Boiaa. Crambe maritime L. Crambe orlentalia L. Crambe plnnatifida R. Br. Crambo tataria Sebeok Cuphonotua andraeanua (F. Muell) S. Shaw Cuphonotua humlatratua (F. Huall) 0. Schulz Cualcklella douglaall (A. Oray) Rollina Deacurainia artemlalodea Svent. Deacuralnia gilva Svont. Deacuralnia hartvgiana (Fourn.) Britton DoBcuralnia incana (Barnh. ex Flacher & C. A. Deacurainia 1email Bramwell Deacurainia millefolia (Jacq.) Webb fi Berth. Deacurainia obcuaa (B. L. Greene) O. E. Schulz Deacurainia aophla (L.) Webb Deacurainia aophioldea (Flacher) 0. E. Schulz Deacurainia atreptocarpa (Fourn.) 0. E. Schulz Dldeamua aogyptlua (L.) Deav. Dldeamua bipinnatua (Deaf.) DC. Didymophyaa aucheri Boiaa. Dlmorphocarpa candlcana (Raf.) Rollina Diplotaxia catholica (L.) DC. Diplotaxia erucoidea (L.) DC. Diplotaxia harra (Forak.) Boiaa. Diplotaxia muralla (L.) DC. Diplotaxia allfolla G. Kunze Diplotaxia tenuifolla (L.) DC. Diplotaxia viminea (L.) DC. Dlptychocarpua atrictua (Fisher) Trautv. Dlthyrea callfornica Harvey Draba brachycarpa Nutt, ex Torrey & A. Gray Draba broveri S. Wata Draba brunlifolla Stev. Draba caii/ornica (Jepaon) Rollina & Price Draba corrugate S. Wata. Draba cuneifolia Nutt, ex Torrey & A. Gray Draba fladnlxenaia Wulfen Draba glabella Purah Draba incana L. Draba incerta Payaon Draba lanceolate Royle Draba lonchocarpa Rydb. 136
Table 23. (Continued)
Draba ollgoaperma Hook. P Draba oreadea Schrenk p Draba reptana (Lam.) Fern. a Draba apectabllla E. L. Greene p Draba atenocarpa Hook. f. & Thome. a Draba atreptocarpa A. Gray p Draba tlbetica Hook. f. £ Thome. p Draba tomentoaa clairv. p Draba trlnervla O. E. Schulz p Drabopala varna C. Koch a Dryopatalon crenatum (Brandegee) Rollins Bnarthrocarpua arcuatua Labi11. Ermanla hlmalayenala (Camb.) 0. E. Schulz Bropbila minima C. A. Hey Brophlla verna (L.) Chevall. Bruca pinnatlfida (Deef.) Pomel Eruca aatlva Hiller Eruca vesicaria (L.) Cav. Brucarla caklloldaa (DC.) 0. E» Schulz Erucaria erucarloidea (Cobb. & Dur.) HueHer Brucaria leucanthemum (Steph.) B. Fedtech Brucarla olllvlerl Haire Brucarla pinnaea (Viv.) Tackh. & Bouloe BrucaBtrum galllcum (Willd.) 0. E. Schulz a/w Erucaatrum naaturtiifolium (Poirot) o. E. Schulz b/p Bryalmum ammophllum A. A. Heller b Erysimum conclnnum Eaetw. b Bryalmum craaalcaule Boiee. b Bryalmum craaalpaa Flech. £ Hey. p Bryalmum cuapidatum (H. Bieb.) DC. a/b Bryalmum grandlflorum Deaf. p Eryaimum hlaracifollum L. b Eryaimum lnconapicuum (S. Wats.) HacHillan p Bryalmum ochroleucum DC. p Eryaimum odoratum Ehrh. a/b Bryalmum pallaall (Pureh) Fern. p Eryaimum alaymbrioldea C. A. Hey a Eryaimum aylvaatra (Crantz) Scop. p Euclidlum tenuiaaimum (Pall.) B. Fedtech. a Eutrama penlandli Rolline p Faraatla aagyptla Turra p Faraatla jacquemontll Hook. £. £ Thome. p Faraatla longiatyla Bak. a FarBatla ramoBiaaima Hochet. ex Boiee. a Faraatla atanoptera Hochet. a Fibigia clypeata (L.) HedicuB p Flbigia eriocarpa (DC.) BoieB. p Fibigia macrocarpa (Boies.) Boies. p Giastaria glaatlfolia (DC.) 0. Kuntze a Oraallaia aaxifragifolla (DC.) Boies. p Hallmoloboa barlandiari (Fourn.) 0. E. Schulz b Halimoloboa hiapidula (DC.) 0. E. Schulz b Hallmoloboa laaioloba (Link) 0. E. Schulz b Hellophlla aranaria Sond. a Heliopbila crithmlfolia Willd. a Hellophlla diffusa (Thunb.) DC. a Hellophlla digitata L.f. a Hellophlla puailla L. f. a 137
Table 23. (Continued)
Hellophila seaellfolla Burch, ex DC. a Hesperia blcuapldata (Willd.) Poiret p Hesperia laciniata All. b/p Heeperie peraica Boiaa. b/p Hesperia eylveetrie Crantz. p Heeperie trietie L. b Hirechfeldia icana (L.) Lagreze-Foasat a Hornungia petraea (L.) Reich. a HutchlnBla alpina (L.) R. Br. p HutchlnBla petraea (L.) R. Br. a Hutchineia procumbena (L.) Deav. a Hymenolobue pauciflorua (Koch) Schinz fi Thell. a Hymenolobue procumbena (L.) Nutt, ex Torre/ fi Gray a Iberia attica Jord. a Iberia crenata Lam. a Iberia intermedia Gueraent any Iberia linifolia L. p Iberia odorata L. a Iberia pinnate L. b Iberia procumbena Lange p Iberia pruitii Tineo p Iberia aampaiana Franco fi P. Silva a Iberia aemperflorena L. p Iberia taurica DC. a/b Iberia umbellate L. a/b Xcfahoa acapigora (Hook.) A. Nela fi J. F. Hacbr. a lodanthua acuminatua Rolline p Iodanthua pinnatifidua (Hichx.) Steudel p lonopaidium abulenae (Pau) Rothm, a Xonopsirfium acaule (Deaf.) Reichenb. a lonopaidium prolongoi (Boiaa.) Batt. a Xaatis cappadodca Deav. P laatla emarginata Kar. fi Kir. a laatia gaube Bornm. p laatia glauca Aucher ex Boiaa. p Xaatis indigotica Fort. p laatia kotachyana Boiaa. fi Hohen. p laatia luaitanica L. a Xaatis minima Bge. a Xaatis steveniana (Trautv.) Davie p Xaatis tinctoria L. b Lachnoloma lehmanni Bge. ' a Leavenvortbia aurea Torrey w Leavenvortbia exigua Rollins w Leavenvortbia uniflora (Hichx.) Britton w Lepidium alyaaoidea A. Gray P Lepidium apetalum Willd. a/b Lepidium aucheri Boiaa. a Lepidium bonarienae L. a Lepidium campeatre (L.) R. Br. a/b Lepidium cartilagineum (J. Hey) Thell. p Lepidium catapycnon Hewaon p Lepidium denaiflorum Schrader a Lepidium flavum Torrey a Lepidium foliaum Deav. p Lepidium Xremontii S. Wats. p Lepidium geniatoidea Hewaon p Lepidium gramlnifollum L. p 138
Table 23. (Continued)
Lepidium heteropbyllum Bentham. p Lepidium hirtum (L.) Sm. p Lepidium integrlfollum Nutt, ex Torrey fi A. Gray p Lepidium latiiollum L. p Lepidium montanum Nutt, ex Torrey fi A. Gray p Lepidium perfollatum L. a/b Lepidium peraicum Boies. p Lepidium phlebopetalum {? Huell.) P. Huell p Lepidium pbolldogynum P. Huell. a Lepidium pinnatifidum Ledeb. a Lepidium ramoBiaaimum A. Nels w/b Lepidium rotundum (Deav.) DC. p Lepidium ruderale L. a/b Lepidium veaicarium L. a/b Lepidium virginicum L. a/b Leptaleum filifoilium (Willd.) DC. a Leaquerella aeenoaa (Richardeon) Rydb. p Leaquerella argyraea (A. Gray) S. Kate p Leaquerella denaipila Rolline a Leaquerella globoaa (Daev.) S. Wats b/p Leaquerella gracilia (Hook.) S. Wats a/b Lesguerella kingll (S. Wats.) S. Wats. p Leaquerella occldentalie (S. Wata) s. Wata p Leaquerella ovalifolia Rydb. p Lobularla intermedia Webb fi Berth. a Lobularia lybica (Viv.) Webb G Berth. a Lunaria annua L. a/b Lyrocarpa coulter! Hook, fi Harvey p Nalcolmia aegyptiaca Sprang a Nalcolmia africana (L.) R. Br. a Halcolmia chia (L.) DC. a Nalcolmia cranulata (DC.) Boise. a Halcolmia flexuoaa (Sm.) Sm. a Nalcolmia granditlora (Bgl.) Ktse. a Nalcolmia lacera (L.) DC. a Nalcolmia littorea (L.) R. Br. a Nalcolmia maritima (L.) R. Br. a Nalcolmia nana (DC.) Boies. a Nalcolmia ramoaiaaima (Desf.) Thell. a Natbhiola flavida Boies. p Natthiola arabica Boies. p Nattbiola cbenopodiifolia Pisch & C. A. Hey p Nattbiola farinoaa Bge. ex Boiss. p Nattbiola fruticuloaa (L.) Haire p Nattbiola lunata DC. a Nattbiola ovatlfolla (Boiss.) Boise. p Hatthiola parvlflora (Schousboe) R. Br. a Nattbiola ainuata (L.) R. Br. b Nlcrolepidlum plloaulum P. Huell a Nicroaisymbrium minutiflorum (Hook. f. fi Thoms.) O. B. Schulz a Noricandia arvenaia (L.) DC. p Noricandla foatlda Bourgeau p Noricandia foleyii Batt. p Noricandia moricandlodea (Boiss.) Heywood p Noricandia nitena (Viv.) Durd. & Barr. p Noricandia ainaica (Boiss.) BoIsb. any Noricandia spinosa Pomel p Nurbecklella buetli (Boiss.) Rothm. p 139
Table 23. (Continued) Nurbecklella plnnatlfida (Lam. at DC.) Rothm. p Nyagrum perfollatum L. a Nasturtium mlcrophyllum (Boernm.) Reichenb. p Nasturtium officinale R. Br. p Nerlsyrenia llnearlfolla (S. Hats) E. L. Oreene p Neslla aplculata Fisch. a Notoceras blcorne (Alton) Amo a Octhodium aegyptiacum (L.) DC. b Octocerus lehmannlanum Bge. a Ornlthocarpa torulosa Rollins p Pachyterglum brevlpes Bga. a Pachyterglum multlcaule (Kar. fi Kir.) Bge. a Parrya nudlcaulls (L.) Regal p Pennellla longlfolla (Benth.) Rollins b Pennellla patens (O. E. Schulz) Rollins b Petrocallls pyrenalca (L.) R. Br. p Phegmatospermum cochlearlnum (P. Huell) 0. Schulz a Phoenlcaulis chelrantholdes Nutt, ax Torrey fi A. Cray p Physarla acutlfolla Rydb. p Physarla brasalcoldes Rydb. p Physarla didymocarpa (Kook.) A Cray p Physarla florlbund Rydb. p Polyctenlum fremontll (S. Hats.) E. L. Greena p Paedudofortuynla eafandiari Hedge p Pseuderucarla teretifolla (Deaf.) Pomel p Pseudocamellna glaucophylla (Boise.) N. Busch p Ptllotrlchum longicaula (Boise.) Boiss. p Ptllotrlchum spinosum (L.) Bols. p Raphanua rasphanlatrum L. a/b Raphanus satlvus L. a/b Raplatrum perenne (L.) All. p Rapistrum rugosum (L.) All. a Rlcotia aucherl (Boiss.) B. L. Burtt p Romanachulala coatarlcenBla (Standley) Rollins p Rorlppa amphibia (L.) Bess. p Rorlppa auatrlaca (Crantz) Besser p Rorlppa cantonlenala (Lour.) Ohwi a/b Rorlppa gambellii (S. Hats.) Rollins fi Al-Shehbaz p Rorlppa glgantea (J. D. Hook.) Garnock-Jones a Rorlppa hisplda (DC.) Britton b/p Rorlppa indica (L.) Hiern a/p Rorlppa Islandlca (Oeder) Borbs any Rorlppa mlcrophylla (Boenn. ex Reichenb.) Hylander p Rorlppa naaturtlum-aquaticum (L.) Hayek p Rorlppa paluatria (L.) Besser p Rorlppa prostrata (Berg.) Schinz et Thellg. p Rorippa seesillflora (Nutt, ex Torey fi A Gray) A. S. Hitch. a Rorlppa sylvestrls (L.) Besser p Rorlppa teres (Hichx.) Stuckey a Rorrlpa curvipes E. L. Greene a Rynchosinapsia chelranthos (Vill.) Dandy a Stunerarla armena (L.) Desv. a Samerarla nuaaularia Bor run. a Samerarla atylophora (Jaub. fi Spach) Boiss. a Savlgnya parvlflora (Del.) Hebb a Schimpera arabica Koschst & Steud. a Schoenocrambe llnearlfolla (A. Gray) Rollins p Schoenocranbe linifolia (Nutt.) E. L. Greene p Table 23. (Continued) Schouwla thebalea Webb Scorploldee ep. (Bge.) Boiee. Selenla aurea Nutt. Slbara deaertll (K. S. Jonee) Rolline Slllcularla polygaloidea (Schltr.) MaraiB Slnapla arvenala L. Slnapla aucherl (Boiee.) O. B. Schulz Slnapla bolvlnll Baillargeon Slnapla flexuoaa Poirot Slnapla puboBcana L. Sisymbrium altiaaimum L. Slaymbrlum eryalmold&a Beef. Sisymbrium Irlo L. Slaymbrlum loaaalll L. Slaymbrlum lutaum (Maxim.) O. E. Schulz Slaymbrlum officinale (L.) Scop. Slaymbrlum orientals L. Slaymbrlum polyceratlum L. Slaymbrlum raboudianum Verlot Slaymbrlum aaptulatum L. Slaymbrlum volganaa Bieb. ex E. Fourn. Smalowakla calycina (Stephen) C. A. Meyer Spryginla w inkier! (Regel) M. Pop. Spryglnloldea Botach. & Vved. Stanleys alata M. E. Jones Stanleys pinnata (Pureh) Britton Stanleys vlrldlflora Nutt, ex Torrey & Gray Stenopetaium daclpiena E. Shaw Stenopetalum llneraa R. Br. ex DC. stenopetaium nutana F. Huell Stenopetaium velutinum F. Muell Sterigmostemum acanthocarpus Fiech & Hey sterlgmoatemum Icanum Bieb. Sterigmostemum sulpursus (Banks fi Soland.) Borum. Streptanthella longiroatria (S. Wats) Rydb. Streptanthus cordatus Nutt, ex Torrey & A. Gray Streptanthus glanduloaua Hook. Streptanthus tortuosus Kellogg Subularia aquatics L. Succowia balearlca (L.) Medicue synthlipaia greggli A. Gray Tauscheria laalocarpa Fiech. ex DC. Teeadalia coronopi/olia (Bergeret) Thell. Teeadaliopala conferta (Lag.) Rothm. Tetracme quadricormia (Steph. in Willd.) Bge. Tbelypodiopaia ambigua (S. Wate.) Al-Shehbaz Thelypodiopaia slogans (M. E. Jonee) Rybd. Thelypodlum hovellli S. Wate. Thelypodlum integrlfolium (Nutt, ex Torrey fi A. Gray) Endl. Thlaapi alliacaum L. Thlaapi alpestre L. Thlaapi ceratocarpum (Pallas) Hurray Thlaapi kotachyanum Boise, fi Hohen. Thlaapi nevadenae Boiss. & Reuter. Thlaapi perfoliatum L. Thlaapi umballatum (Stev. ex) DC. Toruiaria torulosa (Deaf.) O. E. Schulz Turritis glabra L. 141
Table 23 (Continued) Turritis laxa (Sm.) Hayak a/b Valia anrhemerica (Lit. & Maira) Gomez-Campo Oi Qi B Valla Bpinosa Boiaa. Warea cuneifolia (Huhlenb. ex Nutt.) Nutt. Q
Waaabla tenuis (Miq.) Haatum i 0
Zilla macroptera Coaaon i Q
Zilla spinosa (L.) Prantl. i a - annual w ■ winter annual b - biennial p*perennlal 142
DISCUSSION
Our initial hypothesis was that AM development would be greater in longer-lived taxa than in shorter-lived, ruderal ones. Neither our experimental inoculations nor analysis of the existing literature produced results with which we could reject the null hypothesis that AM development in crucifers is independent of lifespan. This finding contrasts that of Peat and Fitter (1993), who showed that genera composed primarily of annuals were more likely to be nonmycorrhizal than were genera containing primarily perennials in the United Kingdom.
The three ways we calculated the percentage of crucifer species which would form AM arrived at a percentage of approximately 19%. This estimate matches well the estimates of 13% of Newman and Redell (1987) and
20% of Harley and Harley (1983). However, our estimate of the percentage of species capable of forming AM was considerably greater than the 8% estimate of Medve (1983) which was based on a considerably smaller number of taxa.
The design of our experiment did not allow for the assessment of the physiological functionality of the AM we observed. Koide and Schreiner (1992) would have classified the crucifer AM development we observed as non functional because of the complete lack of arbuscules.
However, arbuscule development in the crucifer genus 143
Brasalca has been reported elsewhere (Ross and Harper
1973; Tommerup 1984), and some genera within other weakly mycorrhlzal families may develop AH In the presence of mycorrhlzal hosts (e.g. Trlnlck 1977; Ocampo £& &1. 1980;
Miller fll. 1983; DeMars and Boerner 1994), and some species in weakly mycorrhlzal families (e.g.
Caryophyllaceae and Solanaceae) may respond positively to
AM inoculation (Boerner 1992).
Hirrel fll. (1978) and Tommerup (1984) have hypothesized that AM development in crucifers is limited to older, senescent roots. Thus this development would reflect more the loss of defensive capabilities in senescent tissues than it would the development of a symbiotic relationship. This hypothesis is supported by unpublished experimental data of DeMars and Boerner who demonstrate a direct correlation between AM development and plant age/developmental stage in three crucifers.
Despite the considerable attention that has been paid to this issue in the past, in the absence of specific experimental information on transfers of carbon and phosphorus (or lack thereof) between crucifers and the AM fungi they may harbor, the issue of functionality remains unresolved.
Several studies should be performed to address these issues more specifically. First, studies should be performed which address the functionality of roots prior
to clearing and staining. Fluorescein diacetate staining may be able to distinguish functional roots from senescent roots in crucifers (Cooper 1984) and this could be
followed by mycorrhlzal assessment. Second, tests examining the functionality of the apparent AN development in crucifers are needed. Should AM development correlate with increased biomass or tissue nutrient levels, evidence would be available to suggest that AM in crucifers can operate mutualistically without arbuscules.
In conclusion, given the results of our study and others, and in the absence of direct experimental data demonstrating non-functionality, we suggest that the mycorrhlzal development in the Brassicaceae is a random phenomenon, about 20% of roots taken from any sample of plants is likely to show some (probably non-functional) AM fungi colonization. This may explain the numerous conflicting literature reports of AM development in crucifer species. CHAPTER VIII
Arbuscular Mycorrhlzal Development in an Annual,
Biennial, and Perennial Crucifer
INTRODUCTION
Although the family Brassicaceae is typically considered to be nonmycotrophic (Gerdemann 1968; Trappe
1987), reports of arbuscular mycorrhlzal (AM) development in crucifers are common (Medve 1983; Harley and Harley
1987; DeMars and Boerner 1994) from both field collected and laboratory-inoculated specimens. Despite such reports, the functionality of mycorrhizas which develop in species from this family remains unknown, especially since few authors have observed arbuscules in the root segments examined. The lack of arbuscules suggests that any mycorrhlzal development in the family may be non functional since these structures serve as the interface for symbiotic nutrient transfer.
Numerous hypotheses have been advanced to explain the general nonmycotrophic nature of the Brassicaceae (Schwab et al. 1983; Tester al. 1987; Schreiner and Koide
1993b), as have hypotheses which purport to explain the
145 146 occasional apparent mycorrhlzal occurrences (Hirrel £& al. 1978; Ocampo al. 1980; Miller £& al* 1983; Peat and Fitter 1993; DeMars and Boerner 1994). One explanation commonly suggested for the sporadic occurrence of AM in this family is the induction of AM development when typically nonmycotrophic taxa are exposed to roots of normally mycotrophic plants. However, Hirrel a£ al*
(1978) were first to caution that such apparent mycorrhlzal development in typically nonmycotrophic plants may result from the difficulty in distinguishing physiologically active roots from older, senescing roots.
The primary objective of this study was to examine apparent mycorrhlzal development in three crucifers of different life histories over a sufficient period of time for observation of arbuscules, should they develop.
Additionally, we indirectly examined the hypothesis that apparent mycorrhlzal development in the Brassicaceae is a result of the inability to distinguish between functional and senescent roots.
MATERIALS and METHODS
Three species of Brassicaceae were selected to represent the range of life span in the family. Capsella bursa-pastoris (L.) Medikus is a cosmopolitan annual,
Hesperis matronalis L. is a common biennial and Matthiola 147
icana (L.) R. Br. is an often woody-based perennial. All three are native to Eurasia and naturalized in North
America (Rollins 1993).
Several seeds of each crucifer were sown into acid- washed sand in 512 cm3 plastic potting containers. In addition, to determine the effect of having living roots of mycotrophic species in the vicinity of the crucifer roots, one half of the containers were oversown with seeds of sudan grass (Sorghum sudanense (Piper) Stapf.). All containers were inoculated with a root inoculum of Glomus intraradices Schenck and Smith at planting. The root inoculum was prepared from Sorghum sudanense roots which were grown in sand cultures. The original fungal culture was obtained from J. H. Gerdemann at the University of
Illinois and were isolated from agricultural field crops.
On day ten, the containers were thinned to one crucifer per container, and for those oversown with sudan grass, three grass seedlings per container. There were
120 containers per crucifer species, 60 of which also had grass seedlings in them.
The planted containers were placed in trays and randomly allocated to three glasshouse benches (blocks), and watered three times per week through the 90 day growing period. In addition, 250 ml of a low P (1.0 mgP/1) Ruakura solution (Smith al. 1983) was added to 148 each container on days 0 and 45. No other nutrients were added during the experiment.
Plants were grown under ambient glasshouse light and temperature conditions through the spring and summer. On days 10, 25, 40, 60, and 90 four replicates from each block by crucifer species by root matrix treatment (i.e. presence or absence of living sudan grass in the container) were harvested. Sudan grass roots were also harvested from each pot with grass to facilitate comparison of the level of mycorrhizal development in the grass with that of the crucifers.
In the laboratory, roots were separated, cleared and stained with trypan blue (Phillips and Hayman 1970) and stored in plastic tissue cassettes in FAA (formalin:acetic acid:ethanol) preservative solution. Mycorrhizal colonization was determined microscopically in 5 random 2 to 3 cm root segments per plant and was defined as the presence of internal non-septate hyphae, vesicles, arbuscules or chlamydospores. The percentage of root length colonized by mycorrhizal fungi was estimated for each segment (Giovanetti and Mosse 1980) and the mean for the five segments computed. Roots without cortex or roots with obvious damage to the cortex were not used.
The experimental design was a randomized complete block, with 3 blocks (glasshouse benches) by 3 crucifer 149 species by 2 root matrix treatments (with and without sudan grass in the pot) by 5 sample dates. Each combination was replicated four times for a total of 360 pots. The mean percentage root length colonized
(Giovanetti and Mosse 1980) for each plant within treatment combinations were analyzed by analysis of variance (ANOVA) with root matrix treatment, crucifer species, and harvest date and their second order interactions as main effects (Statistical Analysis System
1985). F-values were computed using the mean squares of the block interactions as a divider term.
RESULTS
By day 90, the Caosella bursa-oastoris plants had flowered, set seed, and begun to senesce. In contrast,
Hssperig matronal is. MatthiPlfl isana, and ffgrgtlHTB sudanense remained vegetative and actively growing throughout the 90 day growth period. The number of plants in each treatment group exhibiting AM development varied with crucifer species, root matrix treatment, and harvest dates (Table 24). The frequency of AM development in plants was greater in £. bursa-oastoris and H. matronalis than in lcana. especially in the later harvests, and the frequency of AM development was greater in plants grown with sudan grass than in plants grown alone in eight 150 of nine species-by-harvest date comparisons for days 40,
60, and 90 (Table 24). AM development in crucifer roots was limited to hyphae and vesicles; no arbuscules were observed in any of the 1800 root segments.
The dynamics of mycorrhizal development expressed as the percent root length colonized also varied by species, root matrix treatment, and harvest dates (Figure 10). The percent of root length colonized increased through day 90 in all treatment combinations except the £. bursa-pastoris grown without sudan grass (Figure 10). ANOVA revealed significant effects of harvest date, root matrix treatment, and the interaction of harvest date and matrix treatment on percent root length colonized (Table 25).
The species effect was not statistically significant (£ >
0.08). However, among the three crucifer species, maximum mean colonization ranged from a low of 14.2% ± 2.8 for
H* matronalis grown alone to a high of 31.3% +5.0 for
£. bursa-pastoris grown with sudan grass (Table 25).
Overall, the crucifers grown with sudan grass had greater
AM colonization than plants grown alone (27.1% +5.5 and
16.0 + 3.5, respectively). However, the time lag before which this matrix effect appeared differed among crucifers, such that the difference between plants grown with and without sudan grass appeared as early as day 25 for H. matronal is. on day 40 for ]J. icana. and not until 151 day. 90 for £. bursa-pastoris (Figure 10; Table 26). These differences in tine lag were responsible for the significant root matrix treatnent-by-harvest date interaction in the ANOVA (Table 25).
In contrast to the relatively low percent root length colonized in the crucifers, considerable AM development in the sudan grass was already apparent at day 10 (Figure
10). AM development in sudan grass reached a peak of
68.2% + 4.0 of root length colonized on day 60, and did not increase thereafter. Again, in contrast to the crucifers, arbuscules were common in the sudan grass roots. Figure 10. Mean percent root length colonized in three crucifer species: Caosella bursa-pastoris (an annual),
Hesoeris matronalis (biennial), and Matthiola icanfl
(perennial) and the matrix treatment grass Sorghum sudanense. Triangles represent the grass (£. sudanense) matrix treatment and open squares represent non-grass matrix treatment. Circles refer to the mean root length colonized in £. sudanense in pots containing crucifers.
152 Mean Percent Root Length Colonized 30 20 30 20 Figure 10 Figure 0 0 0 0 0 0 0 0 90 80 70 60 50 40 30 20 10 asla bursa—pastoris Capsella eprs matronalis Hesperis 0 0 0 0 0 0 0 90 80 70 60 50 40 30 20 DAY 30- Matthiola30- icana 60 45 30- 10 ogu sudanense Sorghum 0 0 0 0 0 0 80 70 60 50 40 30 20 20
0 0 0 0 0 0 90 80 70 60 50 40 30 153 Table 24.— Number of plants colonized by mycorrhizal fungi by harvest date (N-12 for each crucifer species by root matrix treatment by harvest date combination).
PRY OF HARVEST
Species x treatment 10 25 40 60 90
C. bursa-pastoris
Without Sudan Grass 0 5 8 10 10 With Sudan Grass 0 7 9 11 11 H- matronalis Without Sudan Grass 1 7 6 10 10
With Sudan Grass 0 7 9 10 11 £. jcana
Without Sudan Grass 0 0 2 4 8
With Sudan Grass 0 0 3 8 9 155
Table 25.--AN0VA of mean root length colonized by mycorrhlzal fungi In 3 crucifer species.
Source df SS £ E > E
Glasshouse Bench (Block) 2 55.4 0.2 0.7927
Crucifer Species 2 2476.3 10.2 0.0B40
Root matrix 1 2533.4 20.9 0.0296 Harvest Date 4 20426.0 42.2 0.0001 Species * Root matrix 4 296.0 1.2 0.2934 Species * Harvest Date B 1411.5 1.5 0.1679
Root matrix * Harvest Date 4 1213.5 2.5 0.0409
Model 23 28412.4 10.27 0.0001 Error 336 40406.9 Table 26.— Percent Root Length Colonized by mycorrhizal fungi by harvest date (N=12 for each crucifer species by root matrix treatment by harvest date combination). Standard errors of the mean are indicated within parentheses. Values followed by different lower case letters within a crucifer species by matrix treatment by harvest date indicate significant differences at £ > 0.05.
DMT OF HARVEST Species x treatment 10 25 40 60 90
£. bursa-oaBtoris Without Sudan Grass 0.0b (0.0) 8.3ab (3.6) 11.7ab (3.0) 18.8a (3.9) 17.5a (3.1) With Sudan Grass 0.0c (0.0) 8.3bc (2.6) 14.6b (3.5) 20.4ab (3.9) 31.3a (5.0) H* matronalis Without Sudan Grass 0.8b (0.8) 0.0b (0.0) 4.6b (1.8) 13.8a (3.2) 14.2a (2.8) With Sudan Grass 0.0b (0.0) 12.lab (3.7) 15.8a (4.3) 20.4a (4.1) 24.2a (5.4) K. icana Without Sudan GraBS 0.0b (0.0) 0.0b (0.0) 1.7b (1.1) 4.6ab (2.1) 16.3a (4.5)
With Sudan Grass 0.0b (0.0) 0.0b (0.0) 6.3b (3.9) 12.5b (4.0) 25.8a (6.2) 157
DISCUSSION
Our data suggest two possible interpretations of AM development in these crucifers. First, an argument can be made that true AM development occurred. In this case, development was sporadic, and was influenced by the heavily mycorrhizal root matrix offered by the living sudan-grass and its mycorrhizae (Ocampo &1. 1980;
Miller al* 1983; DeMars and Boerner 1994).
Alternatively, an argument could be made which supports the hypothesis that the observed AM development is essentially an artifact of our inability to distinguish between senescent roots and active roots of these crucifers (Hirrel at al* 1978), with fungal colonization being linked to senescent roots no longer able to effectively defend themselves against fungi.
Although reports of AM development in crucifers are common, only Ross and Harper (1973) and Tommerup (1984) have reported arbuscule development. Since no arbuscules were observed in any of the root segments we examined here or in a comparison study of 649 taxa of crucifers (DeMars and Boerner, unpublished), it is probable that the normal mechanisms leading to the establishment of a functional mutualism are not present in these plants and, hence, the 158
AM development should be classified as "apparent" only.
Other evidence for the nonfunctional interpretation of crucifer mycorrhizal development exists. Glenn al.
(1985) showed that hyphae from several Glomus spp. (but not Glaaspora spp.) robustly penetrated the roots of
Brasslca spp., provided that there were dead epidermal or cortical cells present. In these cases arbuscules never formed, but vesicles were common and all internal hyphae were intercellular. Reports of "host-non-host" mycelial interconnections between mycotrophic sorghum and nonmycotrophic cabbage also support this nonfunctional view (Ocampo 1986). Although, mycorrhizal development in non-host plants has been observed when grown in the presence of a mycorrhizal host (Hirrel al. 1978; Ocampo et al. 1980), Ocampo (1986) showed that no nutrient transfer occurred between the mycotrophic and typically nonmycotrophic individuals. These results and ours also suggest that the apparent AM development in crucifers may be parasitic (Anderson 1988).
In the present study, apparent AM development consistently increased over time in crucifers growing in the presence of living sudan grass roots. In contrast, AM colonization leveled off in crucifers grown without a sudan grass matrix. Furthermore, the rate of increase in the percent of root length colonized in these crucifers 159
through day 90 suggested to us that colonization was
likely to continue to increase. We suspect that this
pattern is the result of two interacting factors: inoculum
availability and fine root turnover.
Although estimates of the rate of fine root turnover
for herbaceous species are scant, Garwood (1967) showed
that root turnover in perennials can be as high as 81% per
year. Thus, it is quite likely that some of the roots of
both U. matronalis and ft. icana began senescing during the
study period. Since the study plants were grown in
initially sterile sand cultures, low rates of
decomposition of senesced roots would be expected, even
though fine root decomposition can be rapid in the field
(Fahey and Hughes 1994). Thus, the amount of senescent
(and even dead) roots in the crucifer pots should have
increased over time.
The mycorrhizal inoculum potential of the sand in the
pots in which crucifers were grown alone was supplied
solely by the spores and hyphae added at the beginning of
the experiment. In these pots, the inoculum potential
should decrease over time, as spores germinate and
sporlings die for lack of suitable hosts. In contrast,
the soil inoculum potential in pots with living sudan grass roots should continue to increase throughout the
experiment as the hyphal network from the grass roots 160 explores more of the soil volume. Thus, in pots with crucifers and sudan grass, both the availability of senescent roots and the inoculum potential increase, leading to a steady increase in the percent length of colonized roots. Whereas in pots without the grass, the steadily decreasing inoculum potential leads to a leveling off in the rate of development (except for incana) despite a continuing increase in senescent roots over time.
Unpublished data (DeMars and Boerner) from the woodland crucifer Cardamine concatenata (Michx.) Scharz also support the hypothesis that apparent mycorrhizal development is due to colonization of senescent roots.
During the active growth period in March and April,
£. concatenata roots showed no indication of AM development. At senescence (prior to root shedding), in contrast, roots had a mean percent root length colonized of 18%.
This study demonstrates that future studies of apparent AM development in crucifers must incorporate analysis of root turnover and inoculum potential to fully understand the dynamics of AM development in this family.
Furthermore, studies of the functionality of these AM structures in crucifer roots using techniques such as fluorescein diacetate staining (Cooper 1984) or 161 radiolabelling must be done to assess functionality of the mycorrhizae formed in this family. CHAPTER IX
Mycorrhizal status of Descharopsia antarctica Desv.
in the Palmer Station Area, Antarctica.
INTRODUCTION
Arbuscular mycorrhizae (AM) of vascular plants are the most common and widely distributed type of mycorrhizae among terrestrial ecosystems. Until recently, AM had been thought to be absent from Antarctica (Freckman si. al. 1988; Cabello si &1. 1994) . Earlier studies of the only two vascular plants native to Antarctica, Deschampsia antarctica Desv. and Colobanthus auitensis Bartl. had found no AM development in populations on the coastal islands off the Palmer Peninsula of Antarctica (Christie and Nicholson 1983). However, AM fungal spores had been reported from continental Antarctica (Freckman &£ fil.
1988), and earlier this year came the first report of AM development in Antarctic plants (Cabello si 1994). The purpose of this study was to extend the geographical range for which AM-forming fungi and mycorrhizae have been searched for in Antarctica. To do this, areas of the
Palmer Peninsula not covered by prior studies were
162 163 sampled. We also assessed the ability of £• antarctica to develop AM under glasshouse conditions through inoculation.
MATERIALS and METHODS
Clumps of fi. antarctica were collected from the
Palmer Station area, Gamage Point, Anvers Island (64°46'S;
64°3'W), the nearby Old Palmer Station area, and Torgerson
Island in February 1994. At the Palmer Station, the clumps were separated and approximately 50 individuals from each sampling area were preserved in 70% ethanol.
Another 25 individuals were air-dried with soil intact.
In the laboratory, 25 preserved individuals from each location were separated and washed. The distil portion of three fine roots (2-3 cm in length) was removed from each plant. These root segments were then cleared and stained with trypan blue (Phillips and Hayman 1970) and stored in plastic tissue cassettes with FAA (formalin:acetic acid:ethanol) preservative solution. Potential AM development was examined microscopically.
Mycorrhizal soil infectivity of dried soil samples was determined by growing Sorghum sudanense (Piper) Stapf.
(sudan grass) in 14 cm diameter, 1 1 pots containing sterile sand and a subsurface layer of the soil
(approximately 2 cm thick) in a glasshouse located at Ohio 164
State University. At the tine of sowing, 250 ml volume of a low P (1.0 ng P/1) Ruakura solution (Smith al. 1983) was applied. Plants were grown under ambient light and temperature for 60 days, then harvested and examined for
AH development as above. Spore production was assessed by wet sieving 250 ml of sand from each pot (Daniels and
Skipper 1982).
To assess the ability of £. antarctica to develop AM,
30 plants were inoculated in 14 cm diameter, 1 1 pots containing sterile sand and a 2 cm deep layer of inoculum.
The inoculum was composed of sand and £. sudanense roots infected with either Glomus intraradices Schenck and Smith or Glomus etunicatum Becker and Gerdemann. Both fungal cultures were originally obtained from J. H. Gerdemann at the University of Illinois and were isolated from field crops. The £• intraradices culture had originally been identified as £. fasclculatum. but was subsequently re identified (S. B. Rabatin, personal communication). At the time of sowing, 250 ml of low P Ruarkura solution was added to each. After 90 days, the plants and their vegetative progeny were harvested and examined for AM as above. 165
RESULTS
No mycorrhizae in any of 75 antarctica individuals collected from the Palmer Station area were observed. In most cases, root segments were clear of any internal symbionts or colonizers; however, occasionally fungi were observed, including a fruiting ascomycete which resembled the fungus of the ascomycete-hepatic mycorrhizae of
Antarctica (Williams fit al. 1994). However, the ascomycete was only observed on the surface of
Q. antarctica roots in this study. Similarly, there was no mycorrhizal soil infectivity associated with the soil samples collected and no spores were obtained by wet- sieving of the soils.
Inoculated glasshouse 12* antarctica specimens formed extensive AM. All inoculated individuals were mycorrhizal, and those treated with Glomus lntraradices had a higher mean root length colonized (58.6% ±7.8) than those treated with Glomus etunleaturn (49.2% ± 12.9).
Arbuscules, vesicles, and internal hyphae were observed in the roots of all plants examined.
DISCUSSION
Although AM were not observed in any of the specimens sampled from the Palmer Station area, our greenhouse trials show that Deschamosia antarctica can form AM under 166 glasshouse conditions. This confirms Cabello &£ alfs.
(1994) observation of fi. antarctica's ability to form AM with endophytic mycorrhizal fungi, including Glomus antarcticum from the Danco Coast of Palmer Peninsula,
Antarctica. The lack of mycorrhizal soil infectivity in our soil samples suggests that AM do not commonly form in fi. antarctica in the Palmer Station area due to lack of inoculum. Since B. antarctica is mycorrhizal in subantarctic islands (Christie and Nicholson 1983) a source of wind-dispersed inoculum is available. However, wind dispersal of AM fungal spores over long distances may be rare as Mosse (1973) suggested that spores of AM fungi are too large to be dispersed by wind currents (however, see Warner al. 1987). This probably limits potential dispersal to maritime birds (Corner 1971). It has been speculated that cape pigeons and various gulls are involved with dispersal of the plant (Edwards 1972), and perhaps, they may also disperse AM fungi in soils or in a co-dispersing manner similar to that described by Koske and Gemma (1990).
Other factors, including extreme cold, poor soil development, and scarcity of potential hosts also would lower the potential for AM development in Antarctic plants. The primary habitat of £. antarctica includes 167 rock ledges, cracks In rocks, and gravelly ground (Greene and Holton 1971) with a cryosapristic soil base which is nearly 80% organic natter (Leonardi fit fil- 1987).
Additionally, these soils nay be water-logged or very noist throughout the growing season (Edwards 1972).
The finding that fi. antarctica is capable of foming
Glonalean nycorrhizae under glasshouse conditions and the recent discovery of a potentially new Glonus species fron the Peninsula (Cabello fit fll* 1994) suggest several future studies that could shed light on AM developnent in
Antarctica and AM dispersal and dynanics in general.
Although non-native fungi cannot be used in situ inoculation studies (due to the Antarctic Conservation
Act), mycorrhizal inoculum could be extracted from the populations described from the Danco Coast (Cabello fi£ fll.
1994). The dynamics of AM development in £. antarctica patches could then be studied in heretofore uninfected areas. Such studies would be beneficial to understanding the functionality of AM in Antarctic populations and ascertaining the ecological variables regulating the mutualism in this extreme environment. CHAPTER X
Summary of Major Results and Conclusions
Foliar Nutrient Resorption
Nutrient resorption was examined in four understory woodland plants of differing life form and phenology. Of these four species, the spring ephemeral Cardamine concatenta is a high-light adapted organism while the other three are shade-adapted organisms (Sparling 1967).
The primary working hypothesis tested in this series of studies relative to species differences suggested that the spring ephemeral would exhibit the greatest levels of resorption, followed by the shrub, the spring herb, and then the summer herb. This hypothesis was formulated based upon previously published resorption magnitudes.
Since resorption is an energetically demanding process, organisms with low light compensation points and photosynthetic and saturation rates will have to invest a greater overall proportion of respiratory carbon into resorption than plants with a high-light adaptation syndrome. For this reason, plants possessing the shade tolerance syndrome, probably trade off high levels of resorption in lieu of other nutrient conservation
168 169
mechanisms.
The data presented in these studies supported the
hypothesis. Overall, the spring ephemeral possessed
resorption rates of 58.8% for P and 58.2% for N. These
rates were followed by the shrub (40.2% / 33.6%) and the
spring (23.9% / 28.7%) and summer (17.6% / 29.3%) herbs.
The resorption rates for the spring herb and summer herb
suggest that these two species physiologically are
similar. Sparling (1967) and Taylor and Pearcy (1976)
report very similar light compensation points and photosynthetic rates for these two species.
The other aspect of these resorption studies examined variations among plantB along intrastand topographic gradients. Previous studies have shown that resorption rates may decrease with increasing site fertility (e.g.
Boerner 1984). The working hypothesis tested in my studies was that resorption would be indirectly proportional to extractable soil nutrient levels. In both study sites, extractable soil P was lowest in the bottomland topographic positions and greatest in the slope and upland positions. The gradient associated with extractable soil N was reversed, with greater N in the bottomlands and lower N in the slopes and uplands.
Consequently, the hypothesized results for P resorption were greater levels in the bottomlands and lower levels in 170 the slopes and uplands; for N resorption the hypothesized results were greater levels in the bottomlands and lower in the slopes and uplands.
Table 27.--Summary of P and N proportional resorption in the shrub and herbs and AM percent root length colonized in mycotrophic herbs for different topographic positions in the Wright State and Central State forests.
BtSeSSBBSSCSSSBSSSSSBBBS&setSSSSSBBSSSSSSSSSBSS&SSSSBBSS&BSSB
Plant Type Uplands Slopes Bottom.
P Resorption
Shrub 33.1% 33.7% 53.1%
Herbs 25.0% 33.0% 42.3%
N R w r p t l o n Shrub 32.8% 37.3% 30.7%
Herbs 39.5% 37.1% 39.7%
AM Development fPRLCl
I. flexipes 60.2% 66.5% 47.3%
£. racemosa 64.0% 71.0% 49.1%
The results clearly supported the hypothesis for P resorption, however, the data for N resorption were ambiguous (Table 27). The variation among topographic 171 positions was also smaller for N resorption. This lower range may suggest that such differences are not biologically significant and may reflect that N is not limiting in these sites.
Another effect in these studies was statistically significant inter-year differences. These differences occurred for proportional P resorption in the shrub along the topographic gradient and chronosequence, but not for N proportional resorption in the shrub or resorption in the herbs. Previous studies (Minolettl and Boerner 1994) have shown that resorption levels may be substantially lower in drought years. Although total annual precipitation was not lower than the 30-year normal range over the study period, during autumn 1994 there was substantially lower monthly precipitation than the previous two years. This was marked by negative Palmer Hydrologic Drought values an order of magnitude greater than those during summer months. The impact of the drier autumn of 1994, however, did not negatively impact p resorption. In fact, proportional P resorption was greatest in 1994 for both shrub individuals along the topographic gradient and chronosequence. consequently, the significant inter-year differences probably represent inter-year variations in microclimate and microsite characteristics associated with these sites and individuals. 172
A further possibility affecting the levels of
resorption along the topographic gradients is the
interaction of resorption with mycorrhizal colonization
levels. Since both resorption and maintenance of AH
require carbon expenditure, and given the light
limitations of £. flexloes and £. racemosa (Sparling
1967), it is probable that the relatively low levels of
resorption in these plants represents, in part, a tradeoff
with AM carbon demands (Snellgrove fi£ &1> 1982). The
overall data (Table 27) support this hypothesis as P
resorption was greatest in plants from bottomland
positions where AM development was lowest. Upland and
slope plants which had very high levels of PRLC displayed
relatively low P resorption (Table 27). This inverse
relationship between AM and resorption has also been
observed by Boerner (1986) and may also explain, in part,
the high levels of resorption observed in the
nonmycotrophic spring ephemeral.
In conclusion, these data have shown that resorption
rates vary among plants of differing life form and
phenology, that soil P availability may be an important
factor associated with the level of resorption a plant performs, and that mycorrhizal interactions (at least for herbs) may play a role. Most probably, the actual resorption rate a plant experiences results from multiple 173 interactions among intrinsic biological and extrinsic environmental factors. Experimental field studies which manipulate soil nutrient availability would be useful in confirming these generalizations.
AM Development in the Brassicaceae
Four studies in this document presented data which indicated limited development of AM in members of the
Brassicaceae. The data indicated that sporadic infections may arise; however, the functionality of such mycorrhizas remains unknown. In the 24,436 root segments examined in these four studies, no arbuscules were observed. Koide and Schreiner (1992) would classify such AM development as nonmycorrhizal. Since arbuscules are the nutrient transfer structures in an endomycorrhiza, the lack of arbuscules probably indicates that no nutrient exchanges occur.
Although arbuscules were not present, internal hyphae and occasional vesicles were observed. This indicates that AM fungi have the capability to colonize roots of mustards. However, as Hirrel ai. (1978) first suggested, such colonization may result from infection of senescent roots. Data from the studies presented here support Hirrel al's (1978) hypothesis. Cardamine concatenata only displayed AM development during its 174
period of senescence in Kay. since roots are abscised
during this period, the roots displaying AM infection had
to be senescent. Additionally, in the inoculation tine
trials performed on the three nustards, AM root
colonization increased with time and then leveled off.
The colonization was greatest in Capsella bursa-pastoris.
the annual, which completed its life cycle within the 90
day experiment. In this species, roots also had to have
become senescent during this period. The low colonization
in all species during the early part of the experiment
also supports the contention that primarily senescent
roots are infected.
The relatively low levels of percent root length
colonized in the mustards of these studies and the
relatively small percentage of mustards infected at all,
also support the contention that only senescent roots are
infected. In general mean root length colonized by
mycorrhizal fungi (where AM development was observed) was
less than 20% in mustards while in mycorrhizal species
such as Trillium flexipes mean colonization levels reached more than 60%.
Additionally, the observations that mustards of differing life spans did not possess significantly different probabilities of obtaining AM fungi infections also support the contention that mustards do not normally form AM. In these instances, AM development can be interpreted as being a sporadic occurrence, probably developing in roots which became senescent. Finally, the observation that greater AM development in mustards can be
"induced" by a dense mycorrhizal matrix also suggests that only senescent roots are infected. An increased inoculum density surrounding mustard roots increases the probability that senescent root tissue patches will come into contact with an inoculum source. If mustards possess the functional capabilities to become mycorrhizal in a normal, consistent fashion, even a low inoculum potential would eventually lead to higher percentages of root colonization.
Since senescent roots lose their ability to defend against fungal invasion (Tommerup 1984), the proported development of AM in mustard roots probably represents an artifact of our inability to adequately distinguish between senescent and nonsenescent roots. Consequently, such AM development should be termed "apparent" until further research demonstrates experimentally any functionality of this phenomenon. LIST OF REFERENCES
Allen, E. B. and M. F. Allen. 1980. Natural reestablishment of vesicular-arbuscular mycorrhizae following stripmine reclamation in Wyoming. Journal of Applied Ecology 17: 139-147.
Allen M. F., E. B. Allen and C. F. Friese. 1989. Responses of the non-mycotrophic plant Salsola kali to invasion by vesicular-arbuscular mycorrhizal fungi. New Phytologist 111: 45-49.
Allen, M. F. 1991. The ecology of mycorrhizae. Cambridge University Press, New York.
Allsopp, N. and W. D. Stock W. D. 1983. Mycorrhizal status of plants growing in the Cape Floristic Region, South Africa. Bothalia 23: 91-104.
American Public Health Association. 1976. Standard methods for examination of water and wastewater, 14th edn. American Public Health Assoc., New York.
Amijee, F., P. B. Tinker, and D. P. Stribley. 1989. The development of endomycorrhizal root systems. VII. A detailed study of effects of soil phosphorus on colonization. New Phytologist 111: 435-446.
Anderson, C. P., E. I. Sucoff, and R. K. Dixon. 1987. The influence of low soil temperatures on the growth of vesicular-arbuscular mycorrhizal Fraxinus pennsylvanica. Canadian Journal of ForeBt Research 17: 951-956.
Anderson, A. C., A. E. Liberta, and L. A. Dickman. 1984. Interaction of vascular plants and vesicular- arbuscular mycorrhizal fungi across a soil moisture- nutrient gradient. Oecologia 64: 111-117.
Anderson, A. 1988 Mycorrhizae-host specificity and recognition. Phytopathology 78: 375-378.
176 177
Anderson, R. C., A. E. Llberta, L. A. Dickman, and A. J. Katz. 1983. Spatial variation in vesicular-arbuscular mycorrhiza spore density. Bulletin of the Torrey Botanical Club 100: 519-525.
Arco, J. M. del, A. Escudero, and M. V. V. Garrido. 1991. Effects of site characteristics on nitrogen retranslocation from senescing leaves. Ecology 72: 701-708. Arnold, P. T. and L. A. Kapustka. 1987. VA mycorrhizal colonization and spore populations in an abandoned agricultural field after five years of sludge application. Ohio Journal of Science. 87: 112-114.
Barber, s. A. 1984. Soil Nutrient Bioavailability. John Wiley and Sons, New York, New York. 398 Pp.
Baylis, G. T. S. 1975. The magnolioid mycorrhiza and mycotrophy in root systems derived from it, pp. 373- 389, in Endomycorrhizas, F. E. Sanders, B. Mosse, and P. B. Tinker (eds.), Academic Press, New York.
Biondini, M. E., C. D. Bonham, and E. F Redente. 1985. Secondary successional patterns in a sagebrush fArtemesia tridentata) community as they relate to soil disturbance and soil biological activity. Vegetatio 60: 25-36.
Boerner, R. E. J. 1984. Foliar nutrient dynamics and nutrient use efficiency of four deciduous tree species in relation to nutrient fertility. Journal of Applied Ecology 21: 1029-1040.
Boerner, R. E. J. 1985a. Foliar nutrient dynamics, growth, and nutrient use efficiency of Hammamelis virqiniana in three forest microsites. Canadian Journal of Botany 63: 1476-1481.
Boerner, R. E. J. 1985b. Interrelations among soil nutrient availability, N and P cycling, and N and P use efficiency in four hardwood forest stands. In. Proceedings, Fifth Hardwood Forest Conference (edB J. 0. Dawson and K. A. Majerus), Pp. 125-137. University of Illinois Press, Champaign-Urbana, Illinois. 178
Boerner, R. E. J. 1986. Seasonal nutrient dynamics, nutrient resorption, and mycorrhizal infection intensity of two perennial forest herbs. American Journal of Botany 73i 1249-1257.
Boerner, R. E. J. and S. 0. Koslowsky. 1989. Microsite variations in soil chemistry and nitrogen mineralization in a beech-maple forest. Soil Biology and Biochemistry 21: 795-801.
Boerner, R. E. J. 1990. Role of mycorrhizal fungus origin in growth and nutrient uptake by Geranium robertianum. American Journal of Botany 73: 483-489.
Boerner, R. E. J. 1992. Plant life span and response to inoculation with vesicular-arbuscular mycorrhizal fungi. II. Species from weakly mycotrophic genera. Mycorrhiza l: 163-167.
Boring, L. R., C. D. Monk, and W. T. Swank. 1981. Early regeneration of a clear-cut southern Appalachian forest. Ecology 62: 1244-1253.
Brundrett, M. and B. Kendrick. 1988. The mycorrhizal status, root anatomy, and phenology of plants in a sugar maple forest. Canadian Journal of Botany 66: 1153-1173.
Brundrett, M. and B. Kendrick. 1990a. The root and mycorrhizas of herbaceous woodland plants. II. structural aspects of morphology. New Phytologist 114: 469-479.
Brundrett, M. and B. Kendrick. 1990b. The root and mycorrhizas of herbaceous woodland plants. I. Quantitative aspects of morphology. New Phytologist 114: 457-468.
Cabello, M., L. Gaspar, and R. Pollero. 1994. Glomus antarcticum sp. nov., a vesicular-arbuscular mycorrhizal fungus from Antarctica. Mycotaxon 51: 123-128.
Chapin III, F. s. 1980. The mineral nutrition of wild plants. Annual Review of Ecology and Systematics 11: 233-260.
Chapman, H. D. 1965. Cation exchange capacity. In: C. A. Black (ed), Methods of Soil Analysis, American Society of Agronomy, Madison, WI, pp. 891-895. 179
Chilvers, M. T. and M. J. F. Daft. 1982. Effects of low temperature on development of the vesicular - arbuscular mycorrhizal association between Glomus caledonlum and Allium ceoa. Transactions of the British Mycological Society 79: 153-157.
Christie, P. and T. Nicolson. 1983. Are mycorrhizas absent from the Antarctic? Transactions of the British Mycological Society 80: 557-560.
Cooper, K. M. 1984. Physiology of VA mycorrhizal associations. In: Powell, C. L. and D. J. Bagyaraj (eds), VA Mycorrhizae. CRC Press, Boca Raton, FL, pp. 155-186.
Corner, R. W. M. 1971. Studies in Colobanthus auitensis (Kunth) Bartl. and Deschampsia antarctica Desv.: IV. Distribution and reproductive performance in the Argentine Islands. British Antarctic Survey Bulletin 26: 41-50.
Currah, R. s. and M. Van Dyk. 1986. A survey of some perennial vascular plant species native to Alberta (Canada) for occurrence of mycorrhizal fungi. Canadian Field Naturalist 100: 330-342.
Daniels, B. A. and H. D. Skipper. 1982. Methods for the recovery and quantitative estimation of propagules from soil. In: Methods and Principles of Mycorrhizal Research. Ed. N. C. Schenck. American Phytopathological Society, St. Paul, MN, pp. 29-36.
DeMars, B. G. and J. R. Runkle. 1992. Groundlayer vegetation ordination and site-factor analysis of the Wright State University woods (Greene County, Ohio). Ohio Journal of Science 92: 98-106.
DeMars, B. G. and R. E. J. Boerner. 1994 Vesicular- arbuscular mycorrhizal fungi colonization in Capsella bursa-pastoris (Brassicaceae). American Midland Naturalist 132: 377-380.
Dhillion, S. S., R. C. Anderson, and A. E. Liberta. 1988. Effect of fire on the mycorrhizal ecology of little bluestem (Schizachvrium scopariuml. Canadian Journal of Botany 66: 706-713. 180
Dickman, L. A., A. E. Liberta, and R. C. Anderson. 1984. Ecological Interaction of little bluestem and vesicular-arbuscular mycorrhizal fungi. Canadian Journal of Botany 62: 2272-2277.
Edwards, J. A. 1972. studies in Colobanthus aultensls (Kunth) Bartl. and Deschampsla antarctica Desv.: V. Distribution, ecology and vegetative performance on Signy Island. British Antarctic Survey Bulletin 28: 11-28.
Fahey, T. J. and J. W. Hughes. 1994. Fine root dynamics in a northern hardwood forest ecosystem, Hubbard Brook Experimental Forest, NH. Journal of Ecology 82: 533-548.
Freckman, D. W., M. F. Allen, and R. A. Wharton. 1988. Hematodes and VA mycorrhizae in the Antarctic Dry Valley. Exobiology and Future Mars Missions. NASA, Ames, Mountain View, California (Abstract).
Garner, D. E., A. Ritchie, and V. L. Siegenthaler. 1978. Soil survey of Greene County, OH. U.S. Dept, of Agric., Soil Conservation Serv., Washington, DC.
Garwood, E. A. 1967. Seasonal variation in appearance and growth of grass roots. Journal of the British Grasslands Society 22: 121.
Gay, P. E., P. J. Grubb, and H. J. Hudson. 1982. Seasonal changes in the concentrations of nitrogen, phosphorus and potassium, and in the density of mycorrhiza, in biennial and matrix-forming perennial species of closed chalkland turf. Journal of Ecology 70: 571-593.
Gemma, J. N. and R. E. Koske. 1988. Seasonal variation in spore abundance and dormancy of Gigaspora gigantea and in mycorrhizal inoculum potential of a dune soil. Mycologia 80: 211-216.
Gerdemann, J. W. 1968. Vesicular-arbuscular mycorrhiza and plant growth. Annual Review of Phytopathology 6: 397-418.
Gerdemann, J. W. 1974. Mycorrhizae, in The plant root and its environment, Carson, E. W. (ed.). Univ. of Virginia Press, Charlottesville. 181
Giovanetti M. and B. Mosse. 1980. An evaluation of techniques for measuring vesicular-arbuscular mycorrhizal infection in roots. New Phytologist 84: 489-500.
Giovanetti, M. 1985. Seasonal variations of vesicular- arbuscular mycorrhizas and endogonaceous spores in a maritime sand dune. Transactions of the British Mycological Society 84: 679-684.
Gleason, H. A. and A. Cronquist. 1991. Manual of Vascular Plants of Northeastern United States and Adjacent Canada. New York Botanical Garden, Bronx, New York.
Glenn, M. g ., F. S. Chew, and P. H. Williams. 1985. Hyphal penetration of Brassica (Cruciferae) roots by a vesicular-arbuscular mycorrhizal fungus. New Phytologist 99: 463-472.
Gordon, R. B. 1969. The natural vegetation of Ohio in pioneer days. Bulletin of the Ohio Biological survey 3: 1-109.
Greene, D. M., A. Hoitom. 1971. Studies in Colobanthus quitensis (Kunth) Bartl. and Deschampsia antarctica Desv.: III. Distribution, habitats and performance in the Antarctic botanical zone. British Antarctic Survey Bulletin 26: 1-29.
Gumpertz, M. L. and C. Brownie. 1993. Repeated measures in randomized block and split-block experiments. Canadian Journal of Forest Research 23: 625-639.
Harinikumar, K. M. and D. J. Bagyaraj. 1988. Effect of crop rotation on native vesicular-arbuscular mycorrhizal propagules in soil. Plant and Soil 110: 77-80.
Harley, J. L. and S. E. Smith. 1983. Mycorrhizal symbiosis. Academic Press, Toronto.
Harley J. L. and E. L. Harley. 1987. A check-list of mycorrhiza in the British flora. New Phytologist (Suppl.) 105: 1-102.
Hayman, D. S. 1983. The physiology of vesicular-arbuscular endomycorrhizal symbiosis. Canadian Journal of Botany 61: 944-963. 182
Hicks, D. J. and B. F. Chabot. 1985. Deciduous forest. In: B. F. Chabot and H. A. Mooney (eds), Physiological ecology of North American plant communities, Chapman and Hall, New York, NY, pp 257-277.
Hirrel MC, H. Mehravaran, and J. W. Gerdemann. 1978. Vesicular-arbuscular mycorrhizae in the Chenopodiaceae and Cruciferae: do they occur? Canadian Journal of Botany 56: 2813-2817.
Hocking, P. J. 1982. Effect of water stress on redistribution of nutrients from leaflets of narrow leaved lupin I'Lupinus anaustifolius L.) Annals of Botany 49: 541-543.
Iqbal, S. H. and S. Tauqir. 1982. The influence of soil moisture contents on VA endophytes of mycorrhizal cereals and subsequent production of endogenous spore by them. Bulletin of Mycology 2: 59-68.
Janos, D. P. 1980. Mycorrhizae influence tropical succession. Biotropica 12 (Suppl.): 56-64.
Keeney, D. R. and D. W. Nelson. 1982. Nitrogen-inorganic forms. In: A. L. Page, R. H. Miller and D. R. Keeney (eds) Methods of Soil Analysis, Part 2. Chemical and microbiological properties Soil Science Society of America, Madison, WI, pp. 643-698.
Killingbeck, K. T. 1986. Litterfall dynamics and element use efficiency in a Kansas gallery forest. American Midland Naturalist 116: 180-189.
Killingbeck, K. T. 1986. The terminological jungle revisited: making the case for use of the term resorption. Oikos 46: 263-264.
Killingbeck, K. T. and Costigan, S. A. 1988. Element resorption in a guild of understory shrub species: niche differentiation and resorption thresholds. Oikos 53: 366-374.
Kim, C. K., D. M. Choe, and H. T. Mun. 1989. Vesicular- arbuscular mycorrhizae in some plants (IV). Korean Journal of Mycology 17: 214-222.
Koide, R* T. and M. Li. 1990. On host regulation of the vesicular-arbuscular mycorrhizal symbiosis. New Phytologist 114: 59-64. 183
Koide, R. T. and R. P. Schreiner. 1992. Regulation of the vesicular-arbuscular mycorrhizal symbiosis. Annual Review of Plant Physiology and Plant Molecular Biology 43: 557-581.
Koske, R. E. and J. N. Gemma. 1990. VA mycorrhizae in strand vegetation of Hawaii: evidence for long distance codispersal of plants and fungi. American Journal of Botany 77: 466-474.
Koske, R. E., J. N. Gemma, and T. Flynn. 1992. Mycorrhizae in Hawaiian Angiosperms: a survey with implications for the origin of the native flora. American Journal of Botany 79: 853-862.
Kost, J. A. and R. E. J. Boerner. 1985. Foliar nutrient dynamics and nutrient use efficiency in Cornus florida. Oecologia 66: 602-606.
Kruckelmann, H. W. 1975. Effects of fertilizers, soils, soil tillage, & plant species on the frequency of Endogone chlamydospores & mycorrhizal infection in arable soils. In: Sanders, F., B. Mosse, and P. Tinker (eds) Endomycorrhizas. Academic Press, London, pp 511-525.
Leonardi, J. M., C. Marchetti, and L. Monticelli. 1987. Characterizacion preliminar de un Histol Antartico bajo gramineas. Contr. 340 del Instit. Antartico Argentino, Buenos Aires.
Lioi, L. and M. Giovannetti. 1989. Vesicular-arbuscular mycorrhizae and species of the Endogonaceae in an Italian serpentine soil. Giornale Botanico Italia 123: 1-8.
Luken, J. 0. and J. W. Thieret. 1987. Sumac-directed patch succession on northern Kentucky (USA) roadside embankments. Transactions of the Kentucky Academy of Sciences 48: 51-54.
Luken, J. 0. 1988. Population structure and biomass allocation in the naturalized shrub Lonicera roaackii (Rupr.) Maxim, of forest and open habitats. American Midland Naturalist 119: 258-267.
Luken, J. 0. and D. T. Mattimiro. 1991. Habitat-specific resilience of the invasive shrub amur honeysuckle (Lonicera raaackii) during repeated clipping. Ecological Applications 1: 104-109. 184
Mahall, B. E. and F. H. Bormann. 1978. A quantitative description of the vegetative phenology of herbs in a northern hardwood forest. Botanical Gazette 139: 467-481.
May, J. D. and K. T. Killingbeck. 1992. Effects of preventing nutrient resorption on plant fitness and foliar nutrient dynamics. Ecology 73: 1868-1878.
Mayr, R. and R. Godoy. 1989. Seasonal patterns in vesicular-arbuscular mycorrhiza in melic beech forest. Agricultural Ecosystems and Environments 29: 281-288.
Medve, R. J. 1983. The mycorrhizal status of the Cruciferae. American Midland Naturalist 109: 406-408.
Menge, J. A., D. Steirle, D. J. Bagyaraj, E. L. Johnson, and R. T. Leonard. 1978. Phosphorus concentrations in plants responsible for inhibition of mycorrhizal infection. New Phytologist 80: 575-578.
Miller, M. E. 1969. Climatic guide for selected locations in Ohio. State of Ohio, Dept. Natural Resources, Columbus.
Miller, R. M., T B. Moorman, and S. K. Schmidt. 1983. Interspecific plant association effects on vesicular- arbuscular mycorrhiza occurrence in Atriplex confertifolia. New Phytologist 95: 241-246.
Minoletti, M. L. and R. E. J. Boerner. 1994. Drought and site fertility effects on foliar nitrogen and phosphorus dynamics and nutrient resorption by the forest understory shrub Viburnum acerifolium L. American Midland Naturalist 131: 109-119.
Moorman, T. and R. B. Reeves. 1979. The role of endomycorrhizae in revegetation practices in the semi-arid West: II. A bioassay to determine the effect of land disturbance on endomycorrhizal populations. American Journal of Botany 66: 14-18.
Moser, E. B., A. M. Saxton, and S. R. Pezeshki. 1990. Repeated measures analysis of variance: application to tree research. Canadian Journal of Forest Research 20: 524-535. 185
Mosse, B. and J. M. Phillips. 1971. The influence of phosphate and other nutrients on the development of vesicular-arbuscular mycorrhiza in culture. Journal of General Microbiology 69: 157-166.
Mosse, B. 1973. Advances in the study of vesicular- arbuscular mycorrhiza. Annual Review of Phytopathology 11: 171-196.
Muller, R. N. 1978. The phenology, growth and ecosystem dynamics of Ervthronium americanum in the northern hardwood forest. Ecological Monographs 48: 1-20.
National Bureau of Standards. 1981. Standard Reference Materials Catalog. NBS Special Publication 260. National Bureau of Standards, Washington, DC.
NOAA. 1992. Climatological data annual summary, Ohio. Vol. 97 (13). U.S. Government Printing Office, Washington, DC.
NOAA. 1993. Climatological data annual summary, Ohio. Vol. 98 (13). U.S. Government Printing Office, Washington, DC.
Nelsen, C. E., N. C. Bolgiano, S. C. Furutane, and G. R. Safir. 1981. The effect of soil phosphorus levels on mycorrhizal infection of field-grown onion plants and on mycorrhizal reproduction. Journal of the American Society of Horticultural Science 106: 786-788.
Newman, E. I. and P. Reddell. 1987. The distribution of mycorrhizas among families of vascular plants. New Phytologist 106: 745-751.
Ocampo, J. A., J. Martin, and D. S. Hayman. 1980. Influence of plant interactions an vesicular- arbuscular mycorrhizal infections. I. Host and non host plants grown together. New Phytologist 84: 27-35.
Ocampo, J. A. 1986. Vesicular-arbuscular mycorrhizal infection of "host" and "non-host" plants: effect on the growth responses of the plants and competition between them. Soil Biology and Biochemistry 18: 607-610.
Odum, E. P. 1969. The strategy of ecosystem development. Science 164: 262-270. Peat, H. J. and A. M. Fitter. 1993. The distribution of arbuscular mycorrhizas in the British flora. New Phytologist 125: 845-854.
Peterson, R. L., A. E. Ashford and W. G. Allaway. 1985. Vesicular-arbuscular mycorrhizal associations of vascular plants on Heron Island, a Great Barrier Reef coral cay. Australian Journal of Botany 33: 669-676.
Philips, J. M. and D. S. Hayman. 1970. Improved procedures for clearing roots and staining parasitic and vesicular-arbuscular mycorrhizal fungi for rapid assessment of infection. Transactions of the British Mycological Society 55: 158-161.
Potter, C. S., H. L. Ragsdall, and C W. Berish. 1987. Resorption of foliar nutrients in a regenerating southern Appalachian forest. Oecologia 73: 268-271.
Pringle, J. S. 1973. Lonicera maackii (Caprifoliaceae) adventive in Ontario. Canadian Field Naturalist 87: 54-55.
Rabatin, S. C. 1979. Seasonal and edaphic variation in vesicular-arbuscular mycorrhizal infection of grasses by Glomus tenuis. New Phytologist 83: 95-102.
Ralhan, P. K. and S. P. Singh. 1987. Dynamics of nutrients and leaf mass in central Himalayan forest trees and shrubs. Ecology 68: 1974-1983.
Redhead, J. F. 1975. Endotrophic mycorrhizas in Nigeria: some aspects of the ecology of the endotrophic mycorrhizal association of Khayfl qrandifoliola DC., In: Sanders, F. E., Mosse, B, and Tinker, P. B. (eds), Endomycorrhizas, Academic Press, New York.
Reid, C. P. P. and G. D. Bowen. 1979. Effects of soil moisture on V/A mycorrhiza formation and root development in Medicago. In: The soil-root interface, J. L. Harley and R. S. Russell (eds), Academic Press, New York, pp. 211-219.
Rollins, R. C. 1993. The cruciferae of continental North America. Stanford University Press, Stanford.
Ross, J. P. and J. A. Harper. 1973. Hosts of vesicular- arbuscular endogone species. Journal of the Elisha Mitchell Science Society 89: 1-3. 187
Ryan, D.F. and F. H. Bormann. 1982. Nutrient resorption In northern hardwood forests. BloBclence 32: 29-32.
Saif, S. R. and N. Iffat. 1976. Vesicular-arbuscular mycorrhizae in plants and endogonaceous spores in the soil of the northern areas of Pakistan. I. Hunza, Nagar and Gilgit. Pakistan Journal of Botany 8: 163-179.
Saif, S. R., I. All and A. Zaidi. 1977. Vesicular- arbuscular mycorrhizae In plants and endogonaceous spores in the soil of northern areas of Pakistan. III. Dir & Chitral. Pakistan Journal of Botany 9: 129-148.
Schreiner, R. P. and R. T. Koide. 1993a. Antifungal compounds from the roots of mycotrophic and non- mycotrophic plant species. New Phytologist 123: 99-105.
Schreiner, R. P. and R. T. Koide. 1993b. Mustards, mustard oils and mycorrhizas. New Phytologist 123: 107-113.
Schwab, S. M., J. A. Menge, and R. T. Leonard. 1983. Quantitative and qualitative effects of phosphorus on extracts and exudates of sudangrass roots in relation to vesicular-arbuscular mycorrhiza formation. Plant Physiology 73: 761-765.
Shaver, G. R. and J. M. Melillo. 1984. Nutrient budgets of marsh plants: efficiency concepts and relation to availability. Ecology 68: 1491-1510.
Smith, G. S., C. M. Johnston, and I. s. Cornforth. 1983. Comparison of nutrient solutions for growth of plants in sand culture. New Phytologist 94: 537-548.
Smith, s. E. and G. D. Bowen. 1979. Soil temperature, mycorrhizal infection and nodulation in Medicaao truncatula and Trifolium subterraneum. Soil Biology and Biochemistry 11: 469-473.
Smith, S. E. and V. Gianinazzi-Pearson. 1988. Physiological interactions between symbionts in vesicular-arbuscular mycorrhizal plants. Annual Review of Plant Physiology and Plant Molecular Biology 39: 221-244. 188
Snellgrove, R. C., W. E. Splitttoesser, D. B. Stribley, and P. B. Tinker. 1982. The distribution of carbon and the demand of the fungal symbiont in leek plants with vesicular-arbuscular mycorrhizas. New Phytologist 92: 75-87.
Sparling, J. H. 1967. Assimilation rates of some woodland herbs in Ontario. Botanical Gazette 128: 160-168.
Stachurski, A. and J. R. Zimka. 1975. Methods of studying forest ecosystems: leaf area, leaf production and withdrawl of nutrients from leaves of trees. Ekologia Polska 33: 637-648.
Statistical Analysis System. 1985. SAS User's Guide: Statistics, 1985 edition. SAS Institute, Cary, North Carolina.
Taylor, R. J. and R. W. Pearcy. 1976. Seasonal patterns of CO, exchange characteristics of understory plants from a deciduous forest. Canadian Journal of Botany 54: 1094-1103.
Technicon. 1977. Individual/simultaneous determination of nitrogen and/or phosphorus in BD acid digests industrial methods 329-74W/B. Technicon Industrial Systems, Tarrytown, New York.
Tester, M., S. E. Smith, and F. A. Smith. 1987. The phenomenon of "nonmycorrhizal" plants. Canadian Journal of Botany 65: 419-431.
Thompson, B. D., A. D. Robson, and L. K. Abbott. 1986. Effects of phosphorus on the formation of mycorrhizas by Gigaspora calosoora and Glomus fasciculatum in relation to root carbohydrates. New Phytologist 103: 751-765.
Tommerup, I. C. 1984. Development of infection by a vesicular-arbuscular mycorrhizal fungus in Brasstea napus L. and Trifolium subterraneum L. New Phytologist 98: 487-495.
Trappe, J. M. 1987. Phylogenetic and ecologic aspects of mycotrophy in the angiosperms from an evolutionary standpoint. In: Safir, G. R. (ed) Ecophysiology of VA Mycorrhizal plants. CRC Press, Boca Raton, FL, pp 5-25. 189
Trinick, M. J. 1977. Vesicular-arbuscular infection and soil phosphorus utilization in Lupinus spp. New Phytologist 78: 297-304.
Vitousek, P. M. and W. A. Reiners. 1975. Ecosystem succession and nutrient retention: A hypothesis. Bioscience 25: 376-381.
Vitousek, P. 1982. Nutrient cycling and nutrient use efficiency. American Naturalist 119: 553-572.
Warner, N. J., M. F. Allen, and J. A. MacMahon. 1987. Dispersal agents of vesicular-arbuscular mycorrhizal fungi in a disturbed arid ecosystem. Mycologia 79: 721-730.
Whittaker, R. H. 1956. Vegetation of the Great Smoky Mountains. Ecological Monographs 26: 1-80.
Williams, P. G., D. J. Roser, and R. D. Seppelt. 1994. Mycorrhizas of hepatics in continental Antarctica. Mycological Research 98: 34-36.
Zimka, J. R. and A. Stachurski. 1976. Vegetation as a modifier of carbon and nitrogen transfer to soil in various types of forest ecosystems. Ekologia Polska 24: 493-514.
Zimka, J. R. and A. Stachurski. 1992. Intensity of retranslocation of macro- and microelements from ageing foliage of deciduous forest vegetation. Ekologia Polska 40: 333-351.