Table 3.

Linear regression coefficients associated with regression lines in Fig. 1

Study‡	Intercept (a)	Slope (b)	PH0: b=1	r2	n
Original studies
FR-Med	0·045 n.s.	1·001***	n.s.	0·69***	44
SA-Med	0·194 n.s.	0·956***	n.s.	0·78***	46
SP-Med	–0·106 n.s.	1·039***	n.s.	0·68***	46
Published studies
EU-1	–0·317 n.s.	1·197***	†	0·83***	27
EU-2	0·209***	0·905***	***	0·70***	632
AR-Cen	0·646*	0·789***	†	0·50***	55
CA-Qué	–0·116 n.s.	1·014***	n.s.	0·65***	33
AU-1	0·186 n.s.	0·939***	n.s.	0·60***	37
AU-2	0·314 n.s.	0·888***	n.s.	0·82***	26
AU-3	0·167 n.s.	0·937***	n.s.	0·82***	70
AU-4	0·278 n.s.	0·882***	n.s.	0·90***	23
All studies combined
	0·094* ± 0·082	0·963*** ± 0·034	n.s.	0·76***	1039

Linear regression coefficients and associated 95 % confidence intervals for all the studies (total). Regression for all studies is shown on Fig. 1C. Linear model: log(LT) = a + b log[(SLA × LDMC)⁻¹].

LT, measured leaf thickness (µm); SLA, specific leaf area (m² kg⁻¹); LDMC, leaf dry matter content (mg g⁻¹).

^***P < 0·001;

^**P < 0·01;

^*P < 0·050.

†: P < 0·100.;

n.s., not significant.

^‡See Table 1 for study names.