**1. Introduction**

The eggplant (*Solanum melongena* L.) is one of the top ten vegetables that originated from Southeast Asia; it has a high antioxidant capability and nutrient value. Soil-borne diseases, and biotic and abiotic stresses have limited the yield in many commercial eggplant plantations. Vegetable grafting is considered to be a rapid alternative way to slow breeding due to the absence of resistance genotype [1]. Grafting has been found effective not only for resistance advantages but also for the improvement of production and some quality traits of the fruit [2,3]. Eggplant grafting was started in the early 1950s by using Scarlet eggplant as rootstocks [4].

Several studies have already investigated different eggplant scion-rootstock combinations [5–8]. For instance, *Solanum paniculatum* increased vigour and fruit yield, but did not have any positive effect on fruit quality and composition [6]. *Solanum incanum* induced tolerance to water and temperature stress [9], while *Solanum aethiopicum* and *Solanum macrocarpon* were tolerant to *Fusarium oxysporum* and resistant to *Ralstonia solanacearum* [10]. *Solanum sisymbriifolium* and *Solanum integrifolium* have been reported effective in controlling bacterial wilt and resulted in higher yield [11]. *Solanum torvum* is the most commonly used rootstock for eggplant, which has been reported to be resistant to soil-borne diseases; but due to lack of rapid and homogeneous seed germination, there is a need to find other alternative rootstocks [12]. Some tomato rootstocks can be valuable, as well, to improve

eggplant growth and production. Grafting eggplant onto tomato rootstocks (*Solanum lycopersicum* L. × *Solanum habrochaites*) indicated a positive result on yield and the appearance of fruits, but not equally, and it largely depended on the combination and environment [13]. Overall, some rootstock-scion combinations are moderately compatible, and unfavourable effects may occur; hence, the selection of appropriate rootstock-scion combinations is crucial.

Fruit quality of grafted vegetables can be measured by sensory evaluation and instrumental methods as well [14]. Grieneisen et al. [15] conducted a review of 202 different rootstocks and 1023 experimental treatments related to tomato grafting, and they concluded that fruit quality data based on sensory tests were rare among the published studies. Sensory evaluation of grafted plants of the members of the *Cucurbitaceae* family, i.e., melon [16], watermelon [17,18], and cucumber [19], and those of the *Solanaceae* family [20,21] have been conducted, but there is relatively few data for eggplant.

The majority of previous studies investigated eggplant and tomato rootstock compatibility with eggplant scion individually. The objective of the present study was to compare the effect of tomato and eggplant rootstocks on fruit yield and certain quality parameters of eggplant cv. 'Madonna' in soilless pot culture in an unheated plastic greenhouse. Moreover, current research is aimed at identifying the best rootstock type to maximize yield and fruit quality, conducting not only laboratory testing, but sensory evaluation as well.

### **2. Materials and Methods**

### *2.1. Experimental Site, Plant Material, and Management Practices*

The experiment was conducted from March 2019 until October 2019 at the Experimental and Research Farm, Faculty of Horticultural Science, Szent István University, Budapest, in unheated plastic house (47◦23 49 N, 19◦09 10 E, 120 m a.s.l.). The study was arranged based on completely randomized design (CRD) to understand the effect of different rootstocks suitable for eggplant grafting, including *Solanum grandiflorum* × *Solanum melongena* (SH), *Solanum torvum* (ST), *Solanum melongena* × *Solanum integrifolium* (SI), and *Solanum integrifolium* (A) and tomato rootstocks (*Solanum lycopersicum*), including cv. 'Optifort' (O) and cv. 'Emperador' (E). Moreover, self-grafted (SG) cv. 'Madonna' and self-rooted (SR) plants, as control, were used with four replications and five plants in each replication.

The used, popular eggplant cultivar 'Madonna' has evenly coloured, dark-violet fruits. Due to slow seed germination, all four eggplant rootstock seeds (SH, A, ST, and SI) were sown 10 days earlier than cv. 'Madonna' and tomato rootstocks seeds (E and O) (which germinated and developed at the same rate); thus, a uniform stem diameter for grafting was obtained. Eggplant seedlings (with the adequate diameter and 4–5 leaves) were grafted onto the rootstocks by cleft grafting method and kept in high humidity and low light conditions for one week, and then acclimatized to the natural conditions. Three weeks after grafting, all seedlings were transplanted into 10 L pots containing peat substrate, in an unheated polyethylene greenhouse. Temperature and relative humidity inside the greenhouse were recorded by the Flower Power Sensor data logger, as shown in Figure 1. All phytotechnical work recommended for eggplant greenhouse cultivation was performed uniformly. Irrigation was applied twice a week with a commercial fertilizer solution.

Mature fruits were harvested once a week between June and October 2019 (15 times) according to fruit size, colour, and glossiness. Right after picking, the weight of each fruit was recorded by using a digital scale, and the marketable or unmarketable fruits had been sorted. Fruit width (at maximum fruit diameter) and fruit length (from stalk to end of fruit) had been measured by a tape line, and fruit index was calculated based on fruit width/fruit length.

**Figure 1.** Temperature and relative humidity at the greenhouse during cultivation.

#### *2.2. Instrumental Measurements*

The fruit flesh acidity (pH) was measured from two sides of each fruit in all combinations by a pH meter (Hanna HI 98128). The pulp of three fruits from each replication were blended using a kitchen blender. The total soluble solids of the fruit (TSS %) was measured by using a refractometer (PAL-1 Brix 0–53% Digital Hand) from the fruit juice. Fruit firmness from each replication (two fruits) were evaluated by a small hand-operated penetrometer. The pressure value was measured in kg/cm3.

The skin colour of three fruits from each replication was measured on two sides of the skin by using a Minolta Chroma CR-400 colourimeter (Minolta Corporation, Ltd., Osaka, Japan). Fruit chromaticity was expressed in L\*, a\*, b\* colour space coordinates. Chroma (C\*) and Hue angles (H◦) were calculated according to the following formula 1:

$$\mathbf{C}^\* = (\mathbf{a}^{\*2} + \mathbf{b}^{\*2})^{1/2}, \mathbf{H}^\circ = \tan^{-1} \text{ (b}^\*/\text{a}^\*). \tag{1}$$

The same colourimeter was used to determine the whitening index (DW), oxidation potential (OP), and colour differences (CD) of the fruit pulp. Two fruits of each replication were cut longitudinally with a straight edge plastic knife. The pulp was measured quickly after being cut (L0) and after 30 min (L30) in the central and lateral zone. Colour space had been divided into a three-dimensional (L, a and b) so that L (lightness; 0 black and 100 white); a (red to green); and b (blue to yellow). The distance of pure white (DW) was measured as Euclidean distance of the colour coordinates to the pure white coordinates (L\* = 100 a\* = 0 b = 0) using the formula 2 [22]:

$$\text{LDW} = ((100 - \text{L})^2 + \text{a}^{\ast 2} + \text{b}^{\ast 2})^{0.5} \tag{2}$$

The colour differences (CD) were measured as a Euclidean distance between the colour coordinates at 0 and 30 min and calculated by using formula 3:

$$\text{CD} = \left[ (\text{L}^\* \text{30} - \text{L}^\* \text{0}) + (\text{a}^\* \text{30} - \text{a}^\* \text{0}) + (\text{b}^\* \text{30} - \text{b}^\* \text{0}) \right] \tag{3}$$

Moreover, the oxidation potential (OP) was measured by International Commission on Illumination (CIE) L\*a\*b\* values using the following formula 4 [23]:

$$\text{OP} = \text{L}^\* \text{\text{\textdegree}}\_{\text{\textdegree}0} - \text{L}^\* \text{\text{\textdegree}}\_0 \tag{4}$$

Ten independent slices from the equatorial region of fruits were cut and incubated at 20 ◦C to induce seed browning and facilitate seed boundaries identification. Images were used to determine the number of seeds by using the software ImageJ (Version 1.8.0\_172; Research Services Branch, National Institute of Mental Health, Bethesda, MD, USA).
