Printer Friendly

Comparison of muscle activity and energy cost between various bodyweight squat positions.


* muscle activity varies with depth of squatting

* energy cost of squatting is dependent on net muscle activity

* foot position alters muscle activity and energy cost


Squatting is adopted by people belonging to different cultural backgrounds across the globe in multiple ways to perform activities of daily living, occupational work and in sport and rehabilitation programs. Common variations of squat include partial, parallel and deep squat (1, 23). People in the Indian subcontinent and Asian nations often adopt deep squat for toileting activity and house hold chores (16). Sustained deep squat is adopted during occupation related activity by vegetable vendors, fisherwomen, manual labourers and farmers. Whereas partial squats are utilized by carpenters, plumbers, floor layers and house-maids during various job related activities, universally (12). While parallel squats have been used widely in cruciate ligament injury rehabilitation and muscle strengthening programs (17). Deep squatting offers benefit of providing full range of motion across hip, knee and ankle joints. Being weight bearing in nature it imparts beneficial load, improved proprioception, better contraction of muscles and enhanced balance control (6, 8, 14, 17).

Although various squat forms exist, only partial squat has been studied extensively in terms of lower limb muscle activity, forces acting on joints, moments and power generated in selected groups of musculoskeletal conditions like anterior cruciate ligament injuries and osteoarthritis and in people involved in sporting activities like weight lifters (1, 10, 11, 22). High muscle activation and sound joint stability is noted during squat activity in weight lifters and athletes who often perform parallel and deep squats (1, 10, 17, 23). Biomechanically, low to compressive forces and tibio-femoral compressive moderate shear forces and increasing patella-femoral forces are reported during deep squats and full knee flexion closed kinematic chain exercises respectively (5, 11, 14, 23). Alteration of muscle activity and stability in relation to positional variation of foot has been explored previously (13, 15). However there is a lacuna in literature describing squat symmetry and muscle activity in variations of squat with respect to depth of squat.

Additionally, along with posing biomechanical challenges, the squat is also known to stress cardiopulmonary system (7, 21). Increase in heart rate following squatting exercises is well documented in athletes. However, limited literature is available on energy cost of squatting.

Variants of squat are being increasingly used in rehabilitation protocols for various musculoskeletal derangements (9, 19). Hence, to enable prescription of targeted goal oriented squat protocols this study aimed at comparing muscle activity and energy cost in various squat positions i.e. partial, parallel, deep-squat with heel-off-ground and deep-squat with heel-on-ground.


Following ethical approval from Institutional Ethics Review Committee, 90 healthy female participants, aged 18-25 years, without any history of neurological, musculo-skeletal injuries and/or compromised cardio-pulmonary system were recruited by consequent convenient sampling (Table 1). As per Declaration of Helsinki written informed consent was sought from all consenting participants.

All participants were instructed to practice four squat positions two days prior to testing. Four positions included were--partial-squat, parallel-squat, deep-squat with heel-on-ground, deep-squat with heel-off-ground On the day of test, baseline muscle activity was recorded using surface electromyography (Biograph Infinity, Thought Technology LTD, Quebec, Canada). Skin impedance was reduced by cleaning skin with sterile gauze (24). Triode electrodes placed on vastus lateralis (VL), gastrocnemius (GN), and gluteus maximus(GM) recorded muscle activity during squats. VL electrode was placed on lateral aspect of thigh, hand breadth proximal to patella. GC electrode was on lateral head of gastrocnemius. GM electrode was placed midway between greater trochanter and sacrum (18). Baseline EMG activity was recorded in relaxed standing position and thereafter during each squat activity. Three repetitions were performed for each squat position. Each squat was sustained for 10 s with 30-s pause between trials (5). A three minute rest period was interposed between squat positions to avoid muscle fatigue. Maximum amplitude and mean amplitude were noted.

Energy cost of squat was recorded as oxygen consumption(V02) during activity using Fitmate PRO (COSMED, Italy). A non-leak mask was fitted on participant's face. Polar heart rate monitor strapped on mid-sternum recorded heart rate. Resting oxygen consumption and heart rate were recorded in relaxed sitting position. Three trials of four squat positions were performed in random order with 30-s rest between each trial. Results were averaged across three trials.

Data were analyzed with SPSS 16. No deviations were noted from normal distribution for all variables of maximum amplitude and oxygen consumption. One-way ANOVA was used to compare maximum amplitude (Maxamp), mean amplitude (Meanamp) and oxygen consumption in all squat positions. Linear contrast studied the trend of muscle activity in these squats. Level of significance was set at p [less than or equal to] 0.05 and adjusted for post hoc analysis.


Demographic characteristics of participants are presented in Table 1. Although the squat appears to be a symmetrical activity, higher maximum amplitudes were observed on the right side. In order to analyze symmetry of squatting activity, muscle activity of VL, GC and GM were compared between right and left side. Right side demonstrated higher maximum amplitude and mean amplitude in VL (1-6% and 513% respectively). Mean amplitude of GC and GM were higher on right compared to left side (5-37% and 2-15% respectively).

Although differences in maximum and mean amplitudes were observed between sides, pattern of recruitment of all three muscles was similar on right and left side, hence right side muscle activity was considered for discussion.

Maximum and mean amplitude in VL, GC, and GM varied in four squat positions as confirmed by one way ANOVA (p [less than or equal to] 0.00) (refer Figure 1, Table 2). In all four squats, highest activity was seen in VL, followed by GC and lowest in GM.

Maximum amplitudes of VL in partial, parallel, deep-squat with heel-off-ground, and deep squat with heel-on-ground increased linearly ([VL.sub.Maxamp] = 125.5 [micro]V, [VL.sub.Maxamp] = 293.1 [micro]N, [VL.sub.Maxamp] = 312.2 [micro]N, and [VL.sub.Maxamp] = 324.2 [micro]N, respectively). Mean amplitude of VL was highest in parallel-squat, followed by partial-squat, deep-squat with heel-on-ground and was least in deep-squat with heel-off-ground ([VL.sub.Meanamp] = 139.6 [micro]V, [VL.sub.Meanamp] = 64.4 [micro]N, [VL.sub.Meanamp] = 53.93 [micro]N, and [VL.sub.Meanamp] = 43.5 V, respectively).

Maximum amplitude of GC was highest in deep-squat with heel-off-ground, followed by deep-squat with heel-on-ground, parallel-squat and was least in partial-squat ([GC.sub.Maxamp] = 206.5 [micro]N, [GC.sub.Maxam] = 147.9 [micro]N, [GC.sub.Maxamp] = 83.2 [micro]N, and [GC.sub.Maxamp] = 60.8 [micro]N, respectively). Mean amplitude in GC followed a similar trend ([GC.sub.Meanamp] = 24.3 [micro]N, [GC.sub.Meanamp] = 17.3 V,

[GC.sub.Meanamp] = 16.3 [micro]N, and [GC.sub.Meanamp] = 14.2 [micro]N, respectively).

Highest maximum amplitude in GM was recorded in deep-squat with heel-on-ground, followed by parallel-squat, deep-squat with heel-off-ground and least in partial-squat ([GM.sub.Maxamp] = 76.06 [micro]N, [GM.sub.Maxamp] = 72.2 [micro]N, [GM.sub.Maxamp] = 62.33 [micro]N, and [GM.sub.Maxamp] = 28.22, respectively). Mean amplitude in GM was highest in parallel-squat followed by deep-squat with heel on-ground, deep-squat with heel-off-ground and least in partial-squat ([GM.sub.Meanamp] = 6.8 [micro]N, [GM.sub.Meanamp] = 4.9 [micro]N, [GM.sub.Meanamp] = 4.4 [micro]N, and [GM.sub.Meanamp] = 3.31 [micro]N, respectively).

In terms of energy expenditure, oxygen consumption was significantly different in four squat positions (p [less than or equal to] Among the four squats, oxygen consumption was highest during parallel-squat (mean VO2 = 6.81 ml [min.sup.-1] [kg.sup.-1]) followed by deep-squat with heel-on-ground (mean V[O.sub.2] = 6.49 ml [min.sup.-1] [kg.sup.-1]) and least in partial-squat (mean V[O.sub.2] = 5.85 ml [min.sup.-1] [kg.sup.-1]) (refer Figure 2). Heart rate was highest in parallel-squat (99 beats [min.sup.-1]) and least in deep-squat with heel-off-ground (94 beats [min.sup.-1]) (refer Table 2).


Present findings demonstrate difference in muscle activity and energy cost among four squat positions. Knee flexion angle and heel contact with ground largely determined muscle activity and energy cost of squat. Although, squat appears to be deceitfully symmetrical, a difference of muscle activity existed between right and left side. Pattern of muscle activation was similar on both sides hence right side was considered for ease of discussion.

As depth of knee flexion increased from partial-squat to deep-squat, muscle activity of vastus lateralis, gastrocnemius and gluteus maximus increased. Greater displacement of body during deep-squat and higher flexor torque is known to produce strong VL activity to counterbalance flexor torque in order to maintain balance (8). Similar observations regarding VL activity during knee flexion angles beyond 60[degrees] and 90[degrees] are reported earlier.

Highest maximum amplitude generated in VL reflected on greater demand on knee extensors to control descent of the body during deep-squat. Additionally, deep-squat demands higher activity of VL, GC, and GM to enable ascent against gravity. Coactivation of quadriceps and gastrocnemius is reported during parallel and deep squats in context with closed kinetic chain exercises (4, 9, 19). Likewise, as seen in our study highest maximum amplitude of VL were recorded during deep-squat with heel-on-ground followed by deep-squat with heel-off-ground and parallel-squat. Greater depth of squat challenged stability and demanded high muscle activity in VL, GC, and GM to maintain COG within the base of support.

Least activity in VL, GC, and GM during partial squat confirmed that knee flexion angle was an important determinant of muscle activity during squatting.

In contrast to maximum amplitude, mean amplitude in VL decreased sharply after full depth of squat was attained. At full depth of squat, participant rested comfortably with posterior thigh or buttock making contact with heel or posterior calf; consequently reducing quadriceps moment arm and mean muscle activity (2, 23). This may explain why people with lifestyles that require frequent squatting can maintain such a position for prolonged period (3). In contrast, co-contraction of muscles required to actively maintain parallel and partial squat explains higher mean amplitude recorded in all three muscles.

Secondly, position of ankle-foot complex influenced muscle activity to a great extent. Deep squat with heel-off-ground recruited GC strongly to execute active heel elevation and stabilization of hind foot. Heel elevation is known to influence postural adjustment due to forward shift of centre of gravity (3). Resultant increase in gastrocnemius activity assists to maintain body upright and maintain centre of gravity within the narrow base of support created by lifting heel-off the ground.

In terms of motor control strategies, partial-squat employed ankle strategy with greater activation of GC in order to control small amplitude perturbations created by partial-squat. During parallel-squat, higher maximum and mean amplitude in VL and GC indicated greater co-contraction of VL and GC necessary for maintaining stability due to larger perturbation. During deep-squat with heel-off-ground, mean amplitude of GC was higher than deep squat with heel-on-ground. Cumulative effect of lack of support and a narrowed base of support demanded continuous activity in GC to maintain balance (13, 20). During deep squat with heel-on-ground, mean amplitude of all three muscles was significantly higher suggesting involvement of hip and ankle strategies to maintain COG within the base of support. However, fall in mean amplitude during deep squats suggest requirement of low level of muscle activity in order to sustain balance.

Effect of variants of squat positions on muscle activity may be exploited to place incremental load or to target specific muscle activation during exercise training phases in case of pathologies like osteoarthritis of knee and anterior cruciate ligament repairs where weakness of muscles and altered biomechanics play an important role in functional limitation (20).

In addition to biomechanical determinants like knee angle and foot position, energy cost is another factor guiding implementation of squat in daily living and during sporting activities. Variants of squat demonstrated differences in maximum muscle activity resulting in varied amount of energy expenditure. Highest co-contraction of all three muscles may have led to highest amount of oxygen consumed during parallel-squat and deep-squat with heel-on-ground. (V[O.sub.2] = 6.81 ml [min.sup.-1] [kg.sup.-1], heart rate 99 beats x [min.sup.-1] and V[O.sub.2] = 6.49 ml x [min.sup.-1] [kg.sub.-1], heart rate = 95 beats x [min.sup.-1], respectively). Interplay between VL, GC, and GM led to variations in energy cost of activity with a surge in energy cost, when maximum activity was high in any two of these muscles as seen during deep squats.

Clinical Implications

Present findings can be useful for patients with conditions affecting lower limb muscle activity to guide return to lifestyle involving squatting activities, prescription of appropriate squatting exercises designed for health promotion, strengthening of specific muscle groups and ergonomic application in musculoskeletal and neurological conditions.

The present study did not include males, therefore squat performance across; gender could not be studied. Secondly, as all participants were young female adults of 18-25 years, hence effect of age on muscle adaptations and energy cost during squat needs further exploration.


Muscle activity and oxygen consumption varied between four squat variants viz partial-squat, parallel-squat, deep-squat with heel-on-ground, and deep-squat with heel of-ground. Maximum and mean amplitudes were highest in vastus lateralis followed by gastrocnemius and gluteus maximus in all four squat positions. Muscle activity in all three muscles was least in partial-squat and greatest in deep-squat. Activity in VL increased with rise in depth of knee flexion. However, energy cost was dependent on total muscle activity of all three muscles.


(1.) Anthony, C., Raymond, M., Thomas, P., Katherine, W., et al., The effect of back squat on the EMG activity of 4 superficial hip and thigh muscles, J Strength Cond Res, 16(3):428-432, 2002.

(2.) Sriwarno, A.B., Iwanaga, K., Shimomura, Y., and T. Katsuura. The effects of heel elevation on postural adjustment and activity of lower extremity muscles during deep squatting-to-standing movement in normal subjects, J Phys Ther Sci, 20(1):31-38, 2008.

(3.) Blaszczyk, J.W., Hansen, P.D., and D.L. Lowe,. Evaluation of the postural stability in man: movement and posture interaction. Acta Neurobiol Exp (Wars). 53(1):155-60, 1993.

(4.) Barton C.J., Kennedy, A., Twycross-Lewis, R., Woledge, R., Malliaras, P., and D. Morrissey. Gluteal muscle activation during isometric phase of squatting exercises with and without Swiss ball. Phys Ther Sport. 15(1):39-46, 2014.

(5.) Robertson, D.G.E., Wilson, J-M., and T.A. St. Pierre. Lower extremity muscle functions during full squats. J Appl Biomech. 24(4):333-339, 2008.

(6.) Escamilla, R.F., Fleisig, G.S., Zheng, N., Barrentine, S.W., Wilk, K.E., and J.R. Andrews. Biomechanics of the knee during closed kinetic chain and open kinetic chain exercises. Med Sci Sports Exerc. 30(4):556-69, 1998.

(7.) Escamilla, R.F. Knee biomechanics of the dynamic squat exercise Med Sci Sports Exerc. 33(1): 127-41, 2001.

(8.) Escamilla, R.F., Fleisig, G.S., Zheng, N., Lander, J.E., Barrentine, S.W., Andrews, J.R., Bergemann, B.W., and C.T. Moorman. Effect of technique variations on knee biomechanics during squat and leg press. J Orthop Res. 13(6):930-5, 1995.

(9.) Escamilla, R.F., Zheng, N., Macleod, T.D., Edwards, W.B., Imamura, R., Hreljac, A., Fleisig, G.S., Wilk, K.E., Moorman, C.T., and J.R. Andrews. Patellofemoral joint force and stress during the wall squat and one-leg squat. Med Sci Sports Exerc. 41(4):879-88, 2009.

(10.) Macrum, E., Bell, D.R., Boling, M., Lewek, M., and D. Padua. Effect of limiting ankle dorsiflexion range of motion on lower extremity kinematics and muscle activation patterns during a squat. J Sports Rehabil. 21(2):144-50, 2012.

(11.) Fleming, B.C., Ohlen, G., Renstrom, R.A., Peura, G.D., Beynnon, B.D., and G.J. Badger. The effects of compressive load and knee joint torque on peak anterior cruciate ligament strains. Am J Sport. 31(5):701-707, 2003.

(12.) Schaub, P.A. and T.W. Worrell. EMG activity of six muscles and VMO: VL ratio determination during a maximal squat exercise. J Sports Rehabil. 4(3):195-202, 1995.

(13.) Gholami, J., Mansournia, M.A., Davatchi, F., Mohammad, K., Hosseini, H, and R. Majdzadeh. Are daily physical activities risk factors for knee osteoarthritis? Int J Rheum Dis. 19(3):241-7, 2016.

(14.) Philips, J.C., and A.J. Scheen. Squatting test: A posture to study and counteract cardiovascular abnormalities associated with autonomic dysfunction. Auton Neurosci. 5(162), 3-9, 2011.

(15.) Mark Rippetoe, Starting Strength, 3rd ed. The Aasgaard Company, Wicheta Falls, TX, 2013.

(16.) Mohr, M., Nann, M., von Tscharner, V., Eskofier, B., and B.M. Nigg. Task-dependent intermuscular motor unit synchronization between medial and lateral vastii muscles during dynamic and isometric squats. PLoS One. 10(11):e0142048, 2015. doi: 10.1371/journal.pone.0142048.

(17.) Mulholland, S.J. and U.P. Wyss. Activities of daily living in non-Western cultures: range of motion requirements for hip and knee joint implants. Int J Rehabil Res. 24(3):191 -8, 2001.

(18.) Nagura, T., Matsumoto, H., Kiriyama, Y., Chaudhari, A., and T.P. Andriacchi. Tibiofemoral joint contact force in deep knee flexion and its consideration in knee osteoarthritis and joint replacement. J Appl Biomech. 22(4):305-13, 2006.

(19.) Murray, N., Cipriani, D., O'Rand, D., and R. Reed-Jones. Effects of foot position during squatting on the quadriceps femoris: an electromyographic study. Int J Exerc Sci. 6(2): 114-125, 2013.

(20.) Powers, C.M., Ho, K.Y., Chen, Y.J., Souza, R.B., and S. Farrokhi. Patellofemoral joint stress during weight-bearing and non-weight-bearing quadriceps exercises. J Orthop Sports Phys Ther. 44(5):320-7, 2014.

(21.) Roos, P.E., Button, K., and R.W. van Deursen. Motor control strategies during double leg squat following anterior cruciate ligament rupture and reconstruction: an observational study. J Neuroeng Rehabil. 11:19, 2014. doi: 10.1186/1743-0003-11-19.

(22.) Rosenberger, A., Liphardt, A.M., Bargmann, A., Muller, K., Beck, L., Mester, J., and J. Zange. EMG and heart rate responses decline within 5 days of daily whole-body vibration training with squatting. PLoS One. 9(6):e99060, 2014. doi: 10.1371/journal.pone.0099060.

(23.) Flanagan, S., Salem, G.J., Wang, M.Y., Sanker, S.E., and G.A. Greendale. Squatting exercises in older adults: Kinematic and kinetic comparison. Med Sci Sports Exerc. 35(4):635-643, 2003.

(24.) Smith, S.M., Cockburn, R.A., Hemmerich, A., Li, R.M., and U.P. Wyss. Tibiofemoral joint contact forces and knee kinematics during squatting. Gait Posture. 27(3):376-386, 2007.

(25.) Dioniso, V.C., Almeida, G.L., Duarte, M., and R.P. Hirata. Kinematic, kinetic and EMG patterns during downward squatting. J Electromyog Kines. 18(1): 134-143, 2008.


We wish to acknowledge MGM Institute's University Department of Physiotherapy, Navi Mumbai, India for providing support and internally funding the project. We wish to thank all the participants of the study. We would like to acknowledge Vijendra Rajguru for providing support during measurement of energy cost of activity.


Bela M. Agarwal

MGM Institute?s University Department of Physiotherapy

MGM Institute of Health Sciences

Sector 1, Kamothe, Navi Mumbai--410209,India

email ID :

Mobile Number: 91-9819000674

Shreya S. Sahasrabudhe, MPT, Bela M. Agarwal, MSc (PT), Rajani P. Mullerpatan, Ph.D.

MGM Institute's University Department of Physiotherapy, MGM Institute of Health Sciences,Mumbai, India

Caption: Figure 1. Maximum muscle activity in vastus lateralis, gastrocnemius and gluteus maximus in four squat positions.
Table 1. Demographic characteristics of 90 female participants.

Characteristics      Mean (SD )

Age (yr)             20.8(1.8)
Height (cm)          159.2(6.9)
Weight (kg)          55.6(9.4)
BMI (kg [m.sup.2])   21.9(3.5)

Table 2. Comparison of muscle activity and oxygen consumption of four
squat variants: partial squat, parallel squat,
deep squat with heel-off-ground and deep squat with heel-on-ground.

                           Partial-squat    Parallel-squat

Variables                   Mean     SD      Mean      SD

VL                    RT   64.08    7.94    57.09     6.32
Median frequency      LT   64.65    10.25    56.8     7.17
[VL.sub.Maxamp]       RT   125.57   61.67   293.13   116.11
[micro]V              LT   117.35   64.27   268.66   121.37
[VL.sub.Meanamp]      RT   64.49    25.99   139.6    51.29
[micro]V              LT   59.32    24.79   130.11   54.39
[GC.sub.Maxamp]       RT   60.84    37.62   83.21    42.57
[micro]V              LT   58.19    37.84   79.26    46.25
[GC.sub.Meanamp]      RT   14.23    6.68    16.34     5.89
[micro]V              LT   11.58    6.75    13.24     6.16
[GM.sub.Maxamp]       RT   28.22    33.14   72.26    119.74
[micro]V              LT   19.73    20.91   47.06    41.08
[GM.sub.Meanamp]      RT    3.31    2.53     6.88     6.48
[micro]V              LT    2.84    2.06     6.03     3.81
Oxygen consumption          5.85    0.63     6.81     0.59
  ml x [min.sup.-1]
Heart rate beats x         95.13    10.92   99.61    10.95

                      Deep-squat with   Deep-squat with
                      heel-off-ground   heel-on-ground

Variables              Mean      SD      Mean      SD

VL                    50.24     5.97    55.65     6.48
Median frequency      51.63     5.22    56.48     7.32
[VL.sub.Maxamp]       312.2    142.78   324.2    165.51
[micro]V              317.34   165.49   319.73   159.07
[VL.sub.Meanamp]      43.53    29.95    53.93     33.7
[micro]V              49.19    42.23    56.06    46.01
[GC.sub.Maxamp]       206.5    301.68   147.9    131.63
[micro]V              214.62   303.57   184.88   292.35
[GC.sub.Meanamp]      24.38    54.15    17.13     5.97
[micro]V               15.2    16.72    16.18    22.34
[GM.sub.Maxamp]       62.33    59.98    76.06    72.11
[micro]V              90.24    187.12   79.64    88.55
[GM.sub.Meanamp]       4.4      3.25     4.9      3.47
[micro]V               4.29     3.44     4.51     2.56
Oxygen consumption     6.17     0.64     6.49     2.10
  ml x [min.sup.-1]
Heart rate beats x    94.57    10.38    95.38    10.234

                      p value     p value
                      (between    (linear
                      groups)    contrast)


VL                     0.00 *     0.00 *
Median frequency       0.00 *     0.00 *
[VL.sub.Maxamp]        0.00 *     0.00 *
[micro]V               0.00 *     0.00 *
[VL.sub.Meanamp]       0.00 *     0.00 *
[micro]V               0.00 *     0.00 *
[GC.sub.Maxamp]        0.00 *     0.00 *
[micro]V               0.00 *     0.00 *
[GC.sub.Meanamp]        0.08       0.20
[micro]V                0.15       0.02
[GM.sub.Maxamp]        0.00 *     0.00 *
[micro]V               0.00 *     0.00 *
[GM.sub.Meanamp]       0.00 *      0.25
[micro]V               0.00 *      0.02
Oxygen consumption     0.00 *      0.13
  ml x [min.sup.-1]
Heart rate beats x     0.00 *      0.39

Figure 2. Oxygen consumption during four squat positions.

Partial squat                     5.85
Parallel squat                    6.81
Deep squat with heel off-ground   6.17
Deep squat with heel on ground    6.49

Note: Table made from bar graph.
COPYRIGHT 2017 American Kinesiotherapy Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Sahasrabudhe, Shreya S.; Agarwal, Bela M.; Mullerpatan, Rajani P.
Publication:Clinical Kinesiology: Journal of the American Kinesiotherapy Association
Article Type:Report
Geographic Code:9INDI
Date:Jun 22, 2017
Previous Article:Effects of dual task constraints on intra-limb coordination during treadmill walking in people with chronic stroke.
Next Article:Perceptions of coping with an injury in sport at the NCAA Division I level: perceptual continuity between student athletes and their athletic...

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters