Genome size and metabolic intensity in tetrapods: a tale of two lines - PubMed (original) (raw)

Figure 1

Regression of heart index (ratio of heart mass to body mass) on genome size (circles, birds; squares, reptiles; triangles, mammals; diamonds, amphibians). Both lines are approximated by second-order polynomial regression (note that even in this case the regression for reptiles–birds looks linear). If approximated by linear regression, for reptiles–birds: _r_=−0.89, p<10−8, _n_=72; for amphibians–mammals: _r_=−0.92, _p_<10−8, _n_=66 (the difference between slopes of regression lines is highly significant, _p_<10−8). The regression holds within separate phylogenetic groups: birds, _r_=−0.58, _p_<10−4, _n_=53 (Spearman _r_=−0.53, _p_<10−4); reptiles, _r_=−0.82, _p_<10−4, _n_=19 (Spearman _r_=−0.75, _p_<10−3); mammals, _r_=−0.76, _p_<10−6, _n_=40 (Spearman _r_=−0.73, _p_<10−4); amphibians, _r_=−0.88, _p_<10−6, _n_=26 (Spearman _r_=−0.90, _p_<10−4); anurans, _r_=−0.61, _p_<0.01, _n_=16 (Spearman _r_=−0.73, _p_<0.01); caudata, _r_=−0.71, _p_<0.02, _n_=10 (Spearman _r_=−0.59, _p_<0.05). The significance levels for pair-wise differences between the slopes of regression lines: birds versus reptiles, _p_>0.8; mammals versus amphibians, p<0.01; anurans versus caudata, _p_>0.9. (Dotted lines, confidence intervals for _p_=0.95.) If controlled for body mass using multiple regression analysis, the picture was similar.