Evaluating indices of body condition in two cricket species

Clint D. Kelly, Brittany R. Tawes, Amy Worthington

Research output: Contribution to journalArticle

16 Citations (Scopus)

Abstract

Body mass components (dry mass, lean dry mass, water mass, fat mass) in each sex correlate strongly with body mass and pronotum length in Gryllus texensis and Acheta domesticus. Ordinary least squares (OLS) regression underestimates the scaling relationship between body mass and structural size (i.e., pronotum length) in both cricket species compared with standard major axis (SMA) regression. Standardized mass components correlate more strongly with scaled mass index (M^i) than with residual body mass (Ri). Ri represents the residuals from an OLS regression of log body mass against log pronotum length. Neither condition index predicts energy stores (i.e., fat content) in G. texensis. Ri is not correlated with energy stores in A. domesticus whereas M^i is negatively correlated. A comparison of condition index methods using published data showed that neither sex nor diet quality affected body condition at adulthood in G. texensis when using the scaled mass index. However, the residual index suggested that sex had a significant effect on body condition. Further, analysis of covariance (ANCOVA) suggested that diet quality significantly affects body mass while statistically controlling for body size (i.e., body condition). We conclude that the statistical assumptions of condition index methods must be met prior to use and urge caution when using methods that are based on least squares in the y-plane (i.e., residual index ANCOVA).

Original languageEnglish (US)
Pages (from-to)4476-4487
Number of pages12
JournalEcology and Evolution
Volume4
Issue number23
DOIs
StatePublished - Dec 1 2014
Externally publishedYes

Fingerprint

cricket
body condition
Gryllidae
body mass
covariance analysis
index method
Gryllus texensis
fat
diet
Acheta domesticus
least squares
nutritional adequacy
water mass
energy
index
body size
gender
adulthood
methodology

All Science Journal Classification (ASJC) codes

  • Ecology, Evolution, Behavior and Systematics
  • Ecology
  • Nature and Landscape Conservation

Cite this

Evaluating indices of body condition in two cricket species. / Kelly, Clint D.; Tawes, Brittany R.; Worthington, Amy.

In: Ecology and Evolution, Vol. 4, No. 23, 01.12.2014, p. 4476-4487.

Research output: Contribution to journalArticle

Kelly, Clint D. ; Tawes, Brittany R. ; Worthington, Amy. / Evaluating indices of body condition in two cricket species. In: Ecology and Evolution. 2014 ; Vol. 4, No. 23. pp. 4476-4487.
@article{ed59e46a1f3c48108f059066c5f9287a,
title = "Evaluating indices of body condition in two cricket species",
abstract = "Body mass components (dry mass, lean dry mass, water mass, fat mass) in each sex correlate strongly with body mass and pronotum length in Gryllus texensis and Acheta domesticus. Ordinary least squares (OLS) regression underestimates the scaling relationship between body mass and structural size (i.e., pronotum length) in both cricket species compared with standard major axis (SMA) regression. Standardized mass components correlate more strongly with scaled mass index (M^i) than with residual body mass (Ri). Ri represents the residuals from an OLS regression of log body mass against log pronotum length. Neither condition index predicts energy stores (i.e., fat content) in G. texensis. Ri is not correlated with energy stores in A. domesticus whereas M^i is negatively correlated. A comparison of condition index methods using published data showed that neither sex nor diet quality affected body condition at adulthood in G. texensis when using the scaled mass index. However, the residual index suggested that sex had a significant effect on body condition. Further, analysis of covariance (ANCOVA) suggested that diet quality significantly affects body mass while statistically controlling for body size (i.e., body condition). We conclude that the statistical assumptions of condition index methods must be met prior to use and urge caution when using methods that are based on least squares in the y-plane (i.e., residual index ANCOVA).",
author = "Kelly, {Clint D.} and Tawes, {Brittany R.} and Amy Worthington",
year = "2014",
month = "12",
day = "1",
doi = "10.1002/ece3.1257",
language = "English (US)",
volume = "4",
pages = "4476--4487",
journal = "Ecology and Evolution",
issn = "2045-7758",
publisher = "John Wiley and Sons Ltd",
number = "23",

}

TY - JOUR

T1 - Evaluating indices of body condition in two cricket species

AU - Kelly, Clint D.

AU - Tawes, Brittany R.

AU - Worthington, Amy

PY - 2014/12/1

Y1 - 2014/12/1

N2 - Body mass components (dry mass, lean dry mass, water mass, fat mass) in each sex correlate strongly with body mass and pronotum length in Gryllus texensis and Acheta domesticus. Ordinary least squares (OLS) regression underestimates the scaling relationship between body mass and structural size (i.e., pronotum length) in both cricket species compared with standard major axis (SMA) regression. Standardized mass components correlate more strongly with scaled mass index (M^i) than with residual body mass (Ri). Ri represents the residuals from an OLS regression of log body mass against log pronotum length. Neither condition index predicts energy stores (i.e., fat content) in G. texensis. Ri is not correlated with energy stores in A. domesticus whereas M^i is negatively correlated. A comparison of condition index methods using published data showed that neither sex nor diet quality affected body condition at adulthood in G. texensis when using the scaled mass index. However, the residual index suggested that sex had a significant effect on body condition. Further, analysis of covariance (ANCOVA) suggested that diet quality significantly affects body mass while statistically controlling for body size (i.e., body condition). We conclude that the statistical assumptions of condition index methods must be met prior to use and urge caution when using methods that are based on least squares in the y-plane (i.e., residual index ANCOVA).

AB - Body mass components (dry mass, lean dry mass, water mass, fat mass) in each sex correlate strongly with body mass and pronotum length in Gryllus texensis and Acheta domesticus. Ordinary least squares (OLS) regression underestimates the scaling relationship between body mass and structural size (i.e., pronotum length) in both cricket species compared with standard major axis (SMA) regression. Standardized mass components correlate more strongly with scaled mass index (M^i) than with residual body mass (Ri). Ri represents the residuals from an OLS regression of log body mass against log pronotum length. Neither condition index predicts energy stores (i.e., fat content) in G. texensis. Ri is not correlated with energy stores in A. domesticus whereas M^i is negatively correlated. A comparison of condition index methods using published data showed that neither sex nor diet quality affected body condition at adulthood in G. texensis when using the scaled mass index. However, the residual index suggested that sex had a significant effect on body condition. Further, analysis of covariance (ANCOVA) suggested that diet quality significantly affects body mass while statistically controlling for body size (i.e., body condition). We conclude that the statistical assumptions of condition index methods must be met prior to use and urge caution when using methods that are based on least squares in the y-plane (i.e., residual index ANCOVA).

UR - http://www.scopus.com/inward/record.url?scp=84919465658&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84919465658&partnerID=8YFLogxK

U2 - 10.1002/ece3.1257

DO - 10.1002/ece3.1257

M3 - Article

AN - SCOPUS:84919465658

VL - 4

SP - 4476

EP - 4487

JO - Ecology and Evolution

JF - Ecology and Evolution

SN - 2045-7758

IS - 23

ER -