*3.2. Leaf Growth Variables*

All leaf growth variables were significantly affected by the 'As concentration×rootstock' interaction (Table 2). With the As increase in the nutrient solution, both LN and LA showed the strongest reduction in the grafting combinations "Proteo"/"Dinero" (−43 leaves plant−<sup>1</sup> and <sup>−</sup>34.4 dm<sup>2</sup> plant<sup>−</sup>1, respectively), "Proteo"/"Magnus" (−42 leaves plant−<sup>1</sup> and <sup>−</sup>27.2 dm2 plant<sup>−</sup>1), and "Proteo"/'Strong Tosa' (−27 leaves plant−<sup>1</sup> and <sup>−</sup>19.3 dm2 plant−1) (Table 2). Similarly, LAR and LWR showed a significant reduction passing from As− to As+ solution, with the strongest drops recorded in the grafting combinations "Proteo"/"Dinero" (−7.6 cm2 g−<sup>1</sup> DW and <sup>−</sup>0.054 g g−<sup>1</sup> DW, respectively) and "Proteo"/"Magnus" (−16 cm<sup>2</sup> <sup>g</sup>−<sup>1</sup> DW and <sup>−</sup>0.073 g g−<sup>1</sup> DW) (Table 2).

**Table 2.** Leaf growth variables in melon plant as affected by As concentration in the nutrient solution and rootstock. Different letters within main factors indicate significance at Fisher's LSD test (*p* = 0.05).


LN—number of leaves; LA—leaf area; LAR—leaf area ratio; LWR—leaf weight ratio.

### *3.3. Leaf Relative Chl Content and Gas Exchanges*

Leaf relative Chl content did not differ in relation to rootstock, while in all grafting combinations it increased by 7% (from 55.3 to 59.3 SPAD units) in the As+ solution (Table 3). The highest AN values in As− solution were recorded in "Proteo" grafted onto "Magnus", "RS841", and 'Strong Tosa' (27.4 μmol CO2 m−<sup>2</sup> s−1, on average). Nevertheless under As+ treatment, the grafting combination "Proteo"/"RS841" showed a significant AN reduction (from 27.7 to 22.7 <sup>μ</sup>mol CO2 <sup>m</sup>−<sup>2</sup> <sup>s</sup><sup>−</sup>1, <sup>−</sup>18%), whereas an opposite trend was recorded in self-grafted "Proteo" (from 25.3 to 27.7 μmol CO2 m−<sup>2</sup> s<sup>−</sup>1, +10%) (Table 3). Passing from As− to As+, gs significantly varied in the grafting combinations "Proteo"/"Dinero" (from 273 to 385 μmol CO2 m−<sup>2</sup> s−1, +41%), "Proteo"/"Shintoza" (from 190 to 245 <sup>μ</sup>mol CO2 m−<sup>2</sup> s−1, +29%), and "Proteo"/"Magnus" (from 275 to 207 <sup>μ</sup>mol CO2 m−<sup>2</sup> s−1, <sup>−</sup>25%), with the control showing stable gs values across nutrient solutions (Table 3). "RS841" rootstock maximized the WUE values in As−, but showed the highest decrease in response to As+ (being reduced from 3.5 to 3.0 <sup>μ</sup>mol CO2 <sup>μ</sup>mol−<sup>1</sup> H2O m<sup>−</sup><sup>2</sup> <sup>s</sup><sup>−</sup>1, <sup>−</sup>12%), while an opposite trend was recorded in self-grafted "Proteo" (from 3.0 to 3.4 μmol CO2 μmol−<sup>1</sup> H2O m−<sup>2</sup> s−1, +14%). A similar response was recorded when "Proteo" was grafted onto 'Strong Tosa' (from 2.8 to 3.3 μmol CO2 μmol−<sup>1</sup> H2O m−<sup>2</sup> s<sup>−</sup>1, +17%) and "Magnus" (from 2.8 to 3.6 μmol CO2 μmol−<sup>1</sup> H2O m<sup>−</sup><sup>2</sup> s<sup>−</sup>1, +26%) (Table 3).

**Table 3.** Leaf relative Chl content and gas exchange variables in melon plant as affected by As concentration in the nutrient solution and rootstock. Different letters within main factors indicate significance at Fisher's LSD test (*p* = 0.05). NS—not significant.


AN—net assimilation rate; gs—stomatal conductance; WUE—water use efficiency.

#### *3.4. Arsenic Accumulation in Plant Fractions*

Arsenic accumulation in plant fractions was significantly affected by the main factors and their interaction (Table 4). In As− treatment, the highest Asroot was recorded in the rootstocks "Magnus" and "Proteo" (11.75 mg kg DW<sup>−</sup>1, on average), while the lowest one in "RS841" (6.26 mg kg DW−1). In the As+ treatment, Asroot increased in all the tested rootstocks, with the highest concentration recorded in "Magnus" (1633.57 mg kg−<sup>1</sup> DW, +131 fold than control), while the lowest one in "RS841" (369.10 mg kg−<sup>1</sup> DW, +58 fold) (Table 4). The As+ treatment promoted Asshoot in a rootstock-dependent way too, since passing from As<sup>−</sup> to As+, "RS841" showed the highest Asshoot increase (by 25.61 mg kg−<sup>1</sup> DW, +43 fold than control), whereas 'Strong Tosa' showed the least one (by 21.75 mg kg−<sup>1</sup> DW, +36 fold) (Table 4). By increasing the As concentration in the nutrient solution, Aspulp increased in all the grafting combinations, with the highest rise recorded in "Proteo" grafted onto "Shintoza" and "Dinero" (by 3.09 mg kg−<sup>1</sup> DW, on average, equal to +14 fold than control) and the least one recorded in self-grafted "Proteo" (by 1.86 mg kg−<sup>1</sup> DW, +2 fold) (Table 4).


**Table 4.** Total arsenic accumulation (mg kg−<sup>1</sup> DW) in different organs of melon plant as affected by As concentration in the nutrient solution and rootstock. Different letters within main factors indicate significance at Fisher's LSD test (*p* = 0.05).
