TABLE 1.

Description of the 16 environments selected for the analysis.

Environment	Groupa	Correlation HD/GY	$\widehat{\mathrm{\alpha }}$b	$\widehat{\mathrm{\alpha }}$c	$\widehat{\mathrm{\beta }}$d
Rea13WW	Low	–0.06	–0.34	–0.33	–0.32
Cle13LN	Low	0.19	0.14	0.10	–0.20
Sau13HN	Low	0.25	0.17	0.17	0.01
Gre12WW	Low	–0.28	–0.64	–0.64	–0.39
Mon12LN	Medium	0.50	0.81	0.83	–0.27
Cap12LN	Medium	0.51	0.59	0.68	–1.03
Gre13WD	Medium	–0.54	–0.82	–0.82	0.01
Mon13LN	Medium	0.59	1.03	1.03	–0.19
Vra13LN	High	0.61	0.73	0.74	0.06
Gre14WD	High	–0.64	–0.81	–0.82	–0.90
Gre12WD	High	–0.68	–0.99	–0.98	–0.39
Rec13LN	High	0.76	1.91	1.95	0.64
Cap12HN	Quadratic	0.38	0.18	0.36	–1.56
Coi12WW	Quadratic	0.34	0.50	0.18	–1.58
Lou12LN	Quadratic	0.35	0.82	0.39	–1.91
All14LN	Quadratic	0.19	0.20	0.19	–2.20

^aThe sixteen environments were grouped in four classes according to the kind of relationship between HD and GY: “low,” “medium,” or “high” correlation, or “quadratic” relationship. ^bα estimates in the 16 environments using model EG_HD. ^cα and ^dβ estimates in the 16 environments using model EG_HD². α and β correspond to the coefficients for the linear and quadratic regressions of GY on HD in each environment. The estimates of α and β presented here were obtained with the full dataset.