Printer Friendly

Survival and aerenchyma development under flooded conditions of Boltonia decurrens, a threatened floodplain species and Conyza canadensis, a widely distributed competitor.


Boltonia decurrens (Torr. and Gray) Wood (Asteraceae) is a federally threatened, floodplain species (U.S. Fish and Wildlife Service, 1988) which occurs along a 400-km section of the lower Illinois River and nearby parts of the Mississippi River (Schwegman and Nyboer, 1985). Suitable habitat for B. decurrens has decreased due to the increasing use of flood control measures and agricultural practices in the river bottomlands (Bellrose et al., 1979; Schwegman and Nyboer, 1985). Extant populations have become isolated and reduced to human-disturbed, alluvial soil habitats, old fields, roadsides and disturbed bottomland lake shores (Morgan, 1980). Before the record flooding of the Mississippi River and its tributaries in 1993, B. decurrens was found in seven Illinois counties and one Missouri county (Smith, 1991) and was thought to have been extirpated from 15 counties in Illinois and three counties in Missouri (U.S. Fish and Wildlife Service, 1988).

Mature plants of Boltonia decurrens are from 1.5 m to 2 m tall, and flower from July through October. Although these plants are prolific seed producers (ca. 30,000 viable seeds/plant; Smith, 1990), very few seedlings have been found in the field and seedling establishment is expected to be low due to the small size of the achenes and the high light and temperature required for germination (Baskin and Baskin, 1988; Smith, 1990). Much of the population is regenerated each autumn by vegetative ramet production of one or more basal rosettes. Rosettes bolt the following spring and maximum plant height is achieved in July (Schwegman and Nyboer, 1985). Although B. decurrens is a perennial, the overwintering and subsequent spring bolting of the rosettes make it functionally a winter annual (Bazzaz, 1990).

Flooding along the Illinois and Mississippi river basins is highest in the spring (Bellrose et al., 1979), subjecting roots of B. decurrens and other winter annuals to oxygen deficiency and toxic reduction products during their early growth (Ponnamperuma, 1972). Many wetland plants which survive root zone flooding form interconnected gas spaces (aerenchyma) in the cortex of their roots (Justin and Armstrong, 1987; Armstrong and Armstrong, 1988; Laan et al., 1989). Development of aerenchyma allows diffusion of oxygen from aerial shoots to maintain root metabolism (Armstrong, 1971; Crawford, 1978; Laan et al., 1990), while excess oxygen released from roots detoxifies the rhizosphere (Coult, 1963; Armstrong, 1967; Armstrong et al., 1992). These morphological and physiological changes in roots enable the continued uptake of nutrients in flood-adapted plants (Drew and Lynch, 1980; Hook, 1984).

This laboratory experiment compared the survival and growth of Boltonia decurrens with an early successional winter annual, Conyza canadensis (as Erigeron canadensis L.) Cronquist (Asteraceae), during a period of root zone saturation and investigated morphological adaptations which might enhance this survival. A study by Smith (1991) indicated that C. canadensis was a component of the vegetation in the sites where B. decurrens occurred in Illinois and Missouri. Conyza canadensis was chosen for comparison because it co-occurs and, as a winter annual (Regehr and Bazzaz, 1976, 1979), is likely to compete with autumn seedlings and rosettes of B. decurrens. Conyza canadensis is widespread and abundant in Illinois and Missouri, has a similar height, but reaches maturity a few weeks earlier than B. decurrens and also occurs as an early successional weed in disturbed habitats (Keever, 1950; Steyermark, 1963).

As part of a comprehensive study of the biology and ecology of Boltonia decurrens, this work was focused on flooding as a management strategy to preserve the species by eliminating flood-sensitive competitors. We sought to determine how much flooding B. decurrens could tolerate, and also the tolerance to flooding of Conyza canadensis, an abundant, early successional competitor of B. decurrens.


All plants used in this study were obtained from a site near the Melvin Price Locks and Dam, St. Charles County, Missouri (38 [degrees] 52 [minutes] 06 [seconds] N lat, 90 [degrees] 12 [minutes] 22 [seconds] long; elev, 140 m). Basal rosettes of Boltonia decurrens were collected November 1989 and stored in plastic bags at 4 C for ca. 180 days before the study was begun. Small rosettes of Conyza canadensis were collected May 1990 2 wk before the study was begun. Plants of B. decurrens were potted 1 wk before the collection of C. canadensis to allow for greening of the chlorotic rosettes. Plants of each species were potted in 10 cmx 10 cm plastic pots in Pro-Mix A commercial potting soil and were acclimated to potting under greenhouse conditions.

Thirty-eight plants (4-10 cm tall) of each species were selected for the experiment. Six plants of each species were randomly assorted into each of five treatment groups: non-flood (control); 2 wk flood (2WK); 4 wk flood (4WK); 6 wk flood (6WK) and 8 wk flood (8WK). Control pots were placed in plastic flats with holes and watered each day until excess water drained from the pots. The plastic pots with flood treatment plants were placed into metal containers in which the water level was maintained daily at the soil surface. 2WK, 4WK and 6WK groups were removed from the metal containers at 14 days, 28 days and 42 days, respectively, to recover from waterlogging while the 8WK group was flooded for the 56-day duration of the experiment. Plants removed from flooding to recover were placed in plastic flats and subsequently watered daily to capacity.

The remaining eight plants of each species were randomly assigned, four of each species, to either the control or flood conditions. These plants were sampled at day 56 for histological studies.

All plants were given 25 ml of full-strength Peter's Plant Food (20-20-20) every 2 wk. The experiment was conducted in a Conviron E-15 environmental chamber maintained at a constant 25 C, 75%RH, 900 [micro]mol photons [m.sup.-2] [s.sup.-1] with a 12-h day/12-h night cycle. Each week individual pots were rearranged within containers to minimize shading effects from other plants and to make space adjustments as pots were removed from the flood treatment and placed in flats to recover from flooding.

Plant height was initially recorded at day 0. Every 14 days plant height and survival were recorded until completion of the experiment at 56 days. Inflorescence development was recorded at the end of the experiment. Plants from the control and 8WK groups were harvested at 56 days, fresh weight (fwt) was recorded, and then plants were dried at 80 C for 5 days, after which dry weight (dwt) was recorded.

Plants removed for histological study had 2-cm sections cut from the midsection of selected roots. Roots were fixed in F.P.A. (5 ml 37% formaldehyde, 5 ml propionic acid and 90 ml 50% EtOH), dehydrated according to Sass (1964) and embedded in paraplast. Transverse sections (8 [[micro]meter]) were mounted on glass slides and stained with 0.05% toluidine-blue O in distilled water (Sakai, 1973). Sections were examined by light microscopy using a Leitz Laborlux S microscope. The composition of stele, cortical cells and aerenchyma (gas space) tissue as a percentage of total root cross-sectional (x.s.) area was determined by making paper tracings from magnified slide projections of the root and comparing total x.s. weight of root traces with the weight of tracing of each individual tissue.

Plant height data were analyzed by ANOVA (3-factor repeated measures at 14-day intervals by species and treatment group), as was tissue composition of roots, and all means were compared using the Newman-Keuls test. Biomass data were compared by Student's t-test (Winer, 1971).


Within 21 days of the beginning of the study, all plants of both species had begun bolting (internode elongation in the rosettes), and all plants bolted within 10 days of the first plant. There was no effect of flood treatment within species on timing of the initiation of bolting; however, Boltonia decurrens initiated bolting a few days earlier than Conyza canadensis (data not shown).

There were significant differences in height within the groupings for the three individual factors (species, treatment and time) and all 2-way and 3-way factor interactions (Table 1). Consequently, comparisons were first made within each of the 10 experimental groups (species x treatment) for increase in height over time. Control plants of Conyza canadensis and all plants of Boltonia decurrens showed significant increases in height (Newman-Keuls, P [less than] 0.001) between day 14 and day 28 and subsequently every 14 days until the experiment ended at 56 days (Table 2). In C. canadensis, height of 2WK and 6WK flood groups at day 56 was significantly greater (P [less than] 0.001) than at day 14. The 4WK and 8WK flood groups had no significant height increase.

Comparisons between the 10 experimental groups (species x treatment) for height at each measurement period showed no significant difference until day 28 (Table 2). Both the control and 2WK groups of Boltonia decurrens and the Conyza canadensis control group had similar mean height after day 56. The 4WK, 6WK and 8WK B. decurrens flood groups [TABULAR DATA FOR TABLE 1 OMITTED] maintained ca. 80% of control height, whereas all C. canadensis flood groups were ca. 12%-16% of control height.

Survival for combined flood groups of Conyza canadensis decreased at each 14-day interval, ending with 42% survival at day 56 [ILLUSTRATION FOR FIGURE 1 OMITTED]. Five plants in the C. canadensis 2WK group, alive upon removal from flood conditions, survived to the end of the experiment, whereas a number of plants in the 4WK and 6WK flood groups died after removal from saturated conditions. All flooded plants of Boltonia decurrens survived the experimental treatments.

Control plants of both species had all begun flower production by the end of the experiment (Table 2). Over 80% of Boltonia decurrens flooded plants and slightly less than 30% of Conyza canadensis flooded plants flowered during the experiment.

Biomass accumulation (Table 3) for the 8WK flood group of Boltonia decurrens was over 50% of control fresh and dry mass, whereas the 8WK flood group of Conyza canadensis [TABULAR DATA FOR TABLE 2 OMITTED] accumulated less than 2% of control fresh and dry mass. Significant reduction in root mass accounted for 77% of the reduction in total fresh mass but only 50% of the reduction in total dry mass for the 8WK group in B. decurrens.

At the time of collection, rosettes of Boltonia decurrens had numerous adventitious roots arising from the stem base ([less than]14 cm long), whereas rosettes of Conyza canadensis had a small tap root ([less than]10 cm long). Plants of B. decurrens, in control or flood conditions, retained their adventitious root network. However, control plants of C. canadensis developed an adventitious root network arising from the stem base, whereas the tap root was reduced or completely disappeared. Flood plants of C. canadensis retained their tap root and were unable to develop adventitious roots during the flood treatment, but those plants which survived after removal from the flood treatment began to develop adventitious roots.

Roots of both species selected for histological study were grouped by size as primary ([greater than]0.4 mm diam) or lateral ([less than]0.2 mm diam) roots. Primary roots of Conyza canadensis control plants and all plants of Boltonia decurrens, regardless of treatment, were adventitious roots [TABULAR DATA FOR TABLE 3 OMITTED] arising from the stem base, while lateral roots arose from the primary adventitious roots. The primary root of C. canadensis flood plants was the tap root while lateral roots arose from the tap root.

Only one flooded plant of Conyza canadensis could be used to measure primary root tissue composition. Because this data could not be accounted for in a 2-factor analysis, a single factor ANOVA was performed on the seven remaining groups (2 species x 2 treatments x 2 root types minus 1 group) for each tissue type as a percentage of total root x.s. area (Table 4). Significant differences were indicated for each tissue type (stele, cortex and aerenchyma) and individual means were compared using the Newman-Keuls test (Winer, 1971).



Both primary and lateral roots of Conyza canadensis control plants had aerenchyma as [less than]10% of root x.s. area. Aerenchyma in the lateral roots of flooded plants increased to 14% with the development of larger lysigenous holes in the cortex, however, the cortical space of primary roots had completely disintegrated under flooded conditions and these roots were apparently dead (Table 5). The primary and lateral roots of Boltonia decurrens had a well-developed lattice of schyzogenous aerenchyma under both drained and flooded conditions. Aerenchyma in control plants of B. decurrens was 26% in both primary and lateral roots. Under flooded conditions, aerenchyma in B. decurrens increased in primary roots to 49% and lateral roots to 37%. The composition of stele tissue was highest in control roots of C. canadensis (the value for the one flooded primary root might be incorrect if the epidermal cells had contracted as the cortical cells decayed). The composition of cortical cells was, in general, greater in lateral roots of both species in control and flood conditions.


Our results show that Boltonia decurrens is a flood-tolerant species with all flood treatment plants surviving the 56-day experiment [ILLUSTRATION FOR FIGURE 1 OMITTED]. Conyza canadensis was sensitive to complete soil saturation, as plant survival steadily decreased throughout the experiment. Survival of C. canadensis was 80% when flooded only 14 days, so for flooding to be used to eliminate competitors of B. decurrens, the flood duration would need to be maintained for at least 28 days to sufficiently reduce survival of species which are as flood-sensitive as C. canadensis. Biomass of Boltonia decurrens flood plants was reduced by 50% of control plants (Table 3), and over 80% of these plants flowered by the end of the experiment (Table 2). While the onset of inflorescence production in B. decurrens was a few weeks earlier than expected from field observations, this onset occurred on taller plants which had grown to the limits of the growth cabinet. The few smaller plants without inflorescences would probably also have begun flowering as they grew closer to those limits. Flooded plants of Conyza canadensis amassed less than 2% of the biomass (Table 3) and less than 30% of flowering of control plants (Table 2). The few surviving plants never grew close to the growth cabinet limits, but most had begun flowering at this small size, possibly due to a poststress response which would allow some level of seed production prior to the plant's death (Kinet et al., 1985).

Plants of Boltonia decurrens, flooded for 28 days or more, reached nearly 80% of control height, whereas flooded plants of Conyza canadensis reached less than 20% of control height, although a few individual plants grew to almost 40% of control height (Table 2). The ability of B. decurrens to survive and increase in height when flooded 56 days gives it a distinct competitive advantage for light during early growth compared to C. canadensis and other flood-sensitive species. A previous study of B. decurrens by Smith (1990) showed positive correlation of the number of inflorescences (therefore the number of seeds) produced with both plant height and biomass. Consequently, with a 20% reduction in height and 50% less dry mass under flooded field conditions, total seed production of B. decurrens might be expected to decline to some degree. If this correlation also applied to C. canadensis (80% loss in height, 98% loss of dry mass), its seed production under flooded conditions would be very negatively affected, thereby greatly reducing the regeneration of this annual species. Conversely, because B. decurrens regenerates both by seedling establishment on saturated soils and vegetative ramet reproduction, under flooded conditions it could replace C. canadensis and other flood-sensitive species as they die, even though the overall seed production of B. decurrens is reduced.

Roots of control and flooded plants of Boltonia decurrens had a similar adventitious growth from the stem base with extensive lateral root development from the primary roots. Control roots showed some lignification while flooded roots were more spongy and had less mass than control plants. Aerenchyma was extensive in roots of B. decurrens from control conditions and increased under flooded conditions whereas the composition of stele and cortical tissue was reduced under flooded conditions (Table 5). The structural changes in roots of B. decurrens under flooded conditions would increase oxygen diffusion while lowering the respiratory demand. Plants of Conyza canadensis grown under control conditions developed adventitious roots from the stem base and these control roots showed little defined aerenchyma development. Plants grown in flooded conditions were unable to develop adventitious roots, whereas the tap root (primary) and lateral roots became waterlogged. Lateral roots of flooded plants had some increased development of cortical air space (14%) which appeared to form from cell rupture rather than cell wall separation. These lysigenous holes in the lateral roots would probably deteriorate further under continued anaerobic conditions to the complete disintegration of the cortex, as was evident in the primary roots. Roots of C. canadensis showed more secondary growth in the cortex than B. decurrens. This continued growth would tend to keep porosities low and respiratory demand high for C. canadensis and may likely contribute to the differential flood tolerance of the two species (Justin and Armstrong, 1987).

The high degree of aerenchyma formation (ca. 43%) in Boltonia decurrens compares well with other flood-tolerant species. Three aquatic macrophytes, Carex lacustris, Phalaris arundinacea and Typha latifolia, all thrive in continuous flood conditions and have ca. 45% aerenchyma (Bedford et al., 1991). Laan et al. (1989) report that Rumex maritimus, a terrestrial species like B. decurrens, has roots which increase aerenchyma from [less than]20% of x.s. area in nonflood conditions to above 40% when soils are saturated. Increasing internal air space allows for the increased diffusion of oxygen within the plant while reducing the total cellular [O.sub.2] demand (Armstrong, 1979). Brix (1988) found Phalaris australis had an oxygen gradient from 21% in the aerial stems to 3.6% in the rhizomes and also reported that internal oxygen varied with the highest concentrations occurring during daytime. We are currently studying B. decurrens to measure the composition of aerenchyma in aerial shoots and to determine if porous connections exist in the root-shoot transition. If aerenchyma is present in both the shoot and transitional tissue, it would indicate that an interconnected system for oxygen diffusion already exists in B. decurrens.

The morphological characteristics of Boltonia decurrens make it well adapted to the wet prairies and shallow marshes in which it originally flourished (Schwegman and Nyboer, 1985). However, habitat destruction and modification of hydrological conditions have essentially eliminated these areas within its range (Bellrose et al., 1979). The remaining populations of B. decurrens are found in the flood transition areas of backwater lake shores, the moist-soil regions behind levees and on bottomland field margins (U.S. Fish and Wildlife Service, 1988). The extensive man-made levee system of the Illinois River prevents many of the smaller spring floods from reaching areas with extant populations of B. decurrens. Because these areas are less frequently flooded, many species less tolerant to flooding, such as Conyza canadensis, establish during the dry periods. A recent study by Smith el al. (1993) indicates that B. decurrens requires high light in order to maintain optimal plant growth through to reproduction, with low light lowering plant survival and inhibiting regeneration of basal rosettes. With natural succession in these drier areas, faster-growing plants would overtop and shade B. decurrens after a few years. If a few plants of B. decurrens survived until the next flood occurrence, they could quickly re-establish in areas opened as flood-sensitive species died. This trend has been observed for B. decurrens in the past decade with large increases in population size occurring in the years immediately following large floods (Smith, 1991). Because of the extensive 1993 flood, the population of B. decurrens in Illinois that had previously declined has re-established into a few areas newly opened by flood disturbance. However, a large stand in Missouri was completely lost due to the long duration and complete submergence of plants under flood waters well above 2 m (Smith, 1994).

The ability of Boltonia decurrens to increase aerenchyma and thereby enhance oxygen diffusion to its roots under conditions of soil anaerobiosis enables it to withstand extended periods of soil saturation with only a slight reduction in plant height. Thus, flood tolerance allows B. decurrens to exist in open floodplain habitats where there is less competition for the light necessary for optimal growth and reproduction. The use of controlled floods as a management tool could help maintain viable habitat for this species.

Acknowledgments. - This work was supported by funds from the U.S. Army Corps of Engineers (USA-CE) and Illinois Department of Conservation (IDOC). The authors gratefully thank Tom Keevin, USA-CE and John Schwegman, IDOC, for their support during our investigations. We also thank Dr. Richard Keating and Sue Eder for their help with histological techniques.


ARMSTRONG, J. AND W. ARMSTRONG. 1988. Phragmites australis - a preliminary study of soil-oxidizing sites and internal gas transport pathways. New Phytol., 108:373-382.

-----, ----- AND P.M. BECKETT. 1992. Phragmites australia. Venturi- and humidity- induced pressure flows enhance rhizome aeration and rhizosphere oxidation. New Phytol., 120:197-207.

ARMSTRONG, W. 1967. The oxidising activity of roots in waterlogged soils. Physiol. Plant., 20:920-926.

-----. 1971. Radial oxygen losses from intact rice roots as affected by distance from the apex, respiration and waterlogging. Physiol. Plant., 25:192-197.

-----. 1979. Aeration in higher plants, p. 225-332. In: H. W. W. Woolhouse (ed.). Advances in botanical research, vol. 7. Academic Press, New York.

BASKIN, C. C. AND J. M. BASKIN. 1988. Germination ecophysiology of herbaceous plant species in a temperate region. Am. J. Bot., 75:286-305.

BAZZAZ, F. A. 1990. Plant-plant interactions in successional environments, p. 239-263. In: J. B. Grace and D. Tillman (eds.). Perspectives on plant competition. Academic Press, New York.

BEDFORD, B., D. BOULDIN AND B. BELIVEAU. 1991. Net oxygen and carbon-dioxide balances in solutions bathing roots of wetland plants. J. Ecol., 79:943-959.

BELLROSE, F. C., F. L. PAVEGLIO, JR. AND D. W. STEFFECK. 1979. Waterfowl and the changing environment of the Illinois River Valley. Ill. Nat. Hist. Surv. Bull., 32:1-54.

BRIX, H. 1988. Light-dependent variations in the composition of the internal atmosphere of Phragmites australis (Cav.) Trin. ex Steudel. Aquat. Bot., 30:319-329.

COULT, D. A. 1963. Observations on gas movement in the rhizome of Menyanthes trifoliata L., with comments on the role of the endodermis. J. Exp. Bot., 15:205-218.

CRAWFORD, R. M. M. 1978. Metabolic adaptations to anoxia, p. 119-136. In: D. D. Hook and R. M. M. Crawford (eds.). Plant life in anaerobic environments. Ann Arbor Science, Ann Arbor.

DREW, M. C. AND J. M. LYNCH. 1980. Soil anaerobiosis, microorganisms, and root function. Annu. Rev. Phytopathol., 18:37-66.

HOOK, D. D. 1984. Adaptations to flooding with fresh water, p. 265-294. In: T. T. Kozlowski (ed.). Flooding and plant growth. Academic Press, New York.

JUSTIN, S. H. F. W. AND W. ARMSTRONG. 1987. The anatomical characteristics of roots and plant response to soil flooding. New Phytol., 105:465-495.

KEEVER, C. 1950. Causes of succession on old fields of the Piedmont, North Carolina. Ecol. Monogr., 20:230-250.

KINET, J. M., R. M. SACHS AND G. BERNIER. 1985. The physiology of flowering, vol. 3. CRC Press, Boca Raton, Florida. 274 p.

LAAN, P., M. J. BERREVOETS, S. LYTHE, W. ARMSTRONG AND C. W. P. M. BLOM. 1989. Root morphology and aerenchyma formation as indicators of the flood-tolerance of Rumex species. J. Ecol., 77:693-703.

-----, M. TOSSERAMS, C. W. P.M. BLOM AND B. W. VEEN. 1990. Internal oxygen transport in Rumex species and its significance for respiration under hypoxic conditions. Plant Soil, 122:39-46.

MORGAN, S. 1980. Status report on Boltonia asteroides var. decurrens in Illinois. Report to the U.S. Fish & Wildlife Service by the Missouri Department of Conservation, Jefferson City. 11 p.

PONNAMPERUMA, F. N. 1972. The chemistry of submerged soils. Adv. Agron., 24:29-95.

REGEHR, D. L. AND F. A. BAZZAZ. 1976. Low temperature photosynthesis in successional winter annuals. Ecology, 57:1297-1303.

----- AND -----. 1979. The population dynamics of Erigeron canadensis, a successional winter annual. J. Ecol., 67:923-933.

SAKAI, W. S. 1973. Simple method for differential staining of paraffin embedded plant material using Toluidine Blue O. Stain Technol., 48:247-249.

SASS, J. E. 1964. Botanical microtechnique, 3rd ed. Iowa State Univ. Press, Ames. 228 p.

SCHWEGMAN, J. E. AND R. W. NYBOER. 1985. The taxonomic and population status of Boltonia decurrens (Torr. and Gray) Wood. Castanea, 50:112-115.

SMITH, M. 1990. Report on basic life history characteristics of Boltonia decurrens (decurrent false aster). Report to the U.S. Army Corps of Engineers, St. Louis, Missouri. 14 p.

-----. 1991. Life history research for decurrent false aster. Report to the Illinois Department of Conservation, Springfield. 26 p.

-----. 1994. Effects of the flood of 1993 on the decurrent false aster (Boltonia decurrens). Report to the U.S. Army Corps of Engineers, Rock Island, Illinois. 14 p.

-----. Y. Wu AND O. GREEN. 1993. Effect of light and water-stress on photosynthesis and biomass production in Boltonia decurrens (Asteraceae), a threatened species. Am. J. Bot., 80:854-864.

STEYERMARK, J. A. 1963. Flora of Missouri. Iowa State Univ. Press, Ames. 1725 p.

U.S. FISH AND WILDLIFE SERVICE. 1988. Endangered and threatened wildlife and plants, determination of threatened status for Boltonia decurrens (decurrent false aster). Fed. Regist, 53:45858-45861.

WINER, B. J. 1971. Statistical principles in experimental design, 2nd ed. McGraw-Hill, New York. 907 p.
COPYRIGHT 1995 University of Notre Dame, Department of Biological Sciences
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1995 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Stoecker, M.A.; Smith, M.; Melton, E.D.
Publication:The American Midland Naturalist
Date:Jul 1, 1995
Previous Article:Differential postdispersal seed predation in disturbed and intact temperate forest.
Next Article:Effects of rhizosphere inundation on the growth and physiology of wet and dry-site Acer rubrum (red maple) populations.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters