TY - JOUR

T1 - Analyses of developmental rate isomorphy in ectotherms: introducing the Dirichlet regression

AU - Boukal, D.S.

AU - Ditrich, T.

AU - Kutcherov, D.

AU - Sroka, P.

AU - Dudová, P.

AU - Papá ek, M.

PY - 2015

Y1 - 2015

N2 - Temperature drives development in insects and other ectotherms because their metabolic rate and growth depends directly on thermal conditions. However, relative durations of successive ontogenetic stages often remain nearly constant across a substantial range of temperatures. This pattern, termed ‘developmental rate isomorphy’ (DRI) in insects, appears to be widespread and reported departures from DRI are generally very small. We show that these conclusions may be due to the caveats hidden in the statistical methods currently used to study DRI. Because the DRI concept is inherently based on proportional data, we propose that Dirichlet regression applied to individual-level data is an appropriate statistical method to critically assess DRI. As a case study we analyze data on five aquatic and four terrestrial insect species. We find that results obtained by Dirichlet regression are consistent with DRI violation in at least eight of the studied species, although standard analysis detects significant departure fr

AB - Temperature drives development in insects and other ectotherms because their metabolic rate and growth depends directly on thermal conditions. However, relative durations of successive ontogenetic stages often remain nearly constant across a substantial range of temperatures. This pattern, termed ‘developmental rate isomorphy’ (DRI) in insects, appears to be widespread and reported departures from DRI are generally very small. We show that these conclusions may be due to the caveats hidden in the statistical methods currently used to study DRI. Because the DRI concept is inherently based on proportional data, we propose that Dirichlet regression applied to individual-level data is an appropriate statistical method to critically assess DRI. As a case study we analyze data on five aquatic and four terrestrial insect species. We find that results obtained by Dirichlet regression are consistent with DRI violation in at least eight of the studied species, although standard analysis detects significant departure fr

U2 - 10.1371/journal.pone.0129341

DO - 10.1371/journal.pone.0129341

M3 - Article

VL - 10

SP - e0129341

JO - PLoS ONE

JF - PLoS ONE

SN - 1932-6203

IS - 6

ER -