Media-Dependent Responses of Shallot to a Phytochemical Mixture Under Different Organic Growing Media

2.1 Study site and experimental design

Investigation of the benefits of phytochemicals in urban shallot cultivation was carried out from June to August 2023 in two stages, namely 1) a preliminary bioassay was held at the Regional Research and Innovation Agency of South Sumatra Province, and 2) application testing on field demonstration plots conducted in Kelurahan Srimulya (Figure 1), Palembang City, South Sumatra, Indonesia. Kelurahan Srimulya is a suburban settlement affected by the tides of the Borang River, a tributary of the Musi River. Climatological conditions during the field experiment were: a) rainfall 30 - 120 mm/month; b) relative humidity 80%; c) average daily sunshine duration 8.3 hours; and d) average air temperature 32.2 ℃ [22].

1.png

Figure 1. The field experiment location

Table 1. Summary of experimental design

2.png

Figure 2. Block samples arrangement and polybag layout

The research design and sample sizes at each stage are summarized in Table 1. This study used 30 polybags for each treatment to observe plant growth and yield, as well as disease attacks. For growth and yield analysis, 20 samples were randomly selected from the harvested polybags. For disease attack analysis, the entire population (30 polybags for each treatment) was used. Block sample arrangement and polybag layout are presented in Figure 2.

2.2 Preliminary bioassay: Plant collection, extract preparation, and phytochemical testing method

At the research site (Kelurahan Srimulya), leaves from ten different shrubs (Table 2) were collected at 7:00 a.m. Western Indonesian Time (WIB), 2^nd June 2023, with each sample weighing 100 grams. The leaves were rinsed with clean water to remove any dust and dirt. Phytochemical extracts were prepared as follows:

•Single Phytochemical Extract: To prepare a single extract, 50 grams of crushed leaves from one species were dissolved in 500 mL of distilled water (10% w/v). The mixture was then macerated for 72 hours.

•Mixed Phytochemical Extract: For the mixed extract, 500 grams of a blend of leaves (50 grams from each of the ten species) were dissolved in 5,000 mL of distilled water (10% w/v). This mixture was also macerated for 72 hours.

Phytochemical potential testing was conducted following these steps:

Step 1. Each of the ten pure extracts, along with one mixed extract, was obtained by filtering each maceration solution through a filter cloth (200 mesh).

Step 2. The filtered extracts were poured separately into cups (7 cm in diameter and 10 cm in height) filled with rockwool until it becomes saturated.

Step 3. Nine mung bean (Vigna radiata) seeds, specifically marked for measurement, were soaked in the extract for 10 minutes and then planted in each cup.

Step 4. The cups were placed in a dark room, and the growth of the seedlings was measured at 24 hours and 72 hours post-planting.

Step 5. Distilled water was used as a control to compare the performance of the extracts, ensuring that an equivalent number of seeds, sowing methods, and incubation conditions were maintained.

Table 2. List of plant species used as a source of phytochemical extract

No.	Plant Species	Plant Image	Description#	Phytochemical Substances Based on Literature
1	Ageratum conyzoides L.	t1.png	Herbs, annual, 50–100 cm tall, sometimes less than 10 cm. Stems robust, ca. 4 cm in diam. Leaves often with axillary abortive buds; median leaves ovate, elliptic, or oblong, 3–8 × 2–5 cm; upper leaves gradually smaller, oblong, sometimes all leaves small, ca. 1 × 0.6 cm.	Prescocene I (7-methoxy-2,2-dimethylchromene), Precocene II (6,7-dimethoxy derivative, ageratochromene) [25]
2	Alternanthera sessilis (L.) DC.	t2.png	Herb perennial, 10–45 cm tall. Stem ascending or creeping. Leaf blade linear-lanceolate, oblong-obovate, or ovate-oblong, 1–8 × 0.2–2 cm, glabrous or pilose, base attenuate, margin entire or slightly serrate, apex acute or obtuse.	2″-O-rhamnosylvitexin, Apigenin-6,8-di-C-β-D-glucopyranoside isomer, Kaempferol monosulfate, Kaempferol monosulfate, p-Hydroxybenzoic acid, p-Hydroxycinnamoyl moiety, Protocatechuic acid, Gibberellin, Daidzein, Benzophenone-4 [26]
3	Asystasia gangetica (L.) T.Anderson	t3.png	Herbs to 0.5 m tall, ascending. Stems 4-angled, pilose. Petiole 3-5 mm, pubescent; leaf blade ovate to elliptic, 3–12 × 1–5 cm, glabrous or sparsely pilose, especially on veins, adaxially with numerous cystoliths, base truncate to rounded, margin entire or slightly crenulate, apex acuminate.	Z-(13,14-Epoxy) tetradec-11-en-1-olacetate, Octadecane, 1-(ethenyloxy), Cycloheptano[d]imidazolidine, cis-Vaccenic acid, 15-Hydroxypentadecanoic acid, proanthocyanidin, flavone, spartein, aphylidine and cyanogenic glycoside [27]
4	Cleome rutidosperma DC.	t4.png	Herbs, annual or rarely perennial, 30-100 cm tall. Stems branched, often with decumbent branches, glabrous or glabrescent to slightly scabrous but sometimes glandular pubescent. Leaflets 3; leaflet blades oblanceolate to rhomboid-elliptic, 1–3.5 × 0.5–1.7 cm.	Eugenol, n-Tetradecanol, 1-Heptadecene, n-Hexadecanol, (3Z)-Cembrene, Isopropyl hexadec-anoate, Phyllocladene, n-Octadecanol, (Z)-Phytol, Acetoxy manool, (Z,Z)-6,9-cis-3,4-Epoxynonadecadien [28]
5	Euphorbia hirta L.	t5.png	Herbs, annual, 30–60(–70) cm tall, usually few-branched. Stem branched from middle or above, ascending to erect, rarely prostrate, ca. 3 mm thick. Leaves opposite; stipules membranous, triangular, 0.8-1.7 mm, caducous; leaf blade lanceolate-oblong, long elliptic, or ovate-lanceolate, 10–50 × 3–16 mm,	Afzelin, quercitrin, myricitrin, rutin, euphorbin-A, euphorbin-B, euphorbin-C, euphorbin-D [29]
6	Mikania micrantha Kunth	t6.png	Vines, slender, branched. Stems yellowish or brownish, usually terete. Leaves opposite; petiole 1-6 cm; blade ovate, 3-13 × ca. 10 cm, both surfaces glabrate with numerous glandular spots, base cordate to deeply so, margin entire to coarsely dentate, apex shortly acuminate.	Hydroxycinnamic derivatives and octulopyranosonic acid derivatives, Quercetin O-heterosides, 5-O-caffeoylquinic acid and dicaffeoylquinic acids [30]
7	Phyllanthus amarus Schumach. & Thonn.	t7.png	Herbs, annual, monoecious. Main stems terete, not winged, glabrous. Leaves on main stems spiral, scalelike. Leaves on ultimate branchlets distichous, well developed; blade elliptic-oblong to obovate, 5–11 × 3–6 mm, base obtuse or rounded, apex obtuse to rounded, often apiculate, both surfaces glabrous.	Corilagin, geraniin, gallic acid, phyllanthin, hypophyllanthin, ellagic acid, phyltetralin, niranthin, catechin, quercetin, astragalin, and chebulagic acid [31]
8	Ruellia tuberosa L.	t8.png	Herbs to 45 cm tall, perennial, erect. Roots with elongate tuberlike swellings. Stems slightly swollen above nodes, almost 4-angled, strigulose on angles. Petiole to 8 mm, glabrous; leaf blade oblong-obovate, 4-8 × 1.5–4.2 cm, both surfaces glabrous.	Isobergapten-5-O-β-D-glucopyranoside syringaresinol, catechin, pulmatin, stigmast-4-en-3-one, verbascoside, hydroxymethylfurfural, rutin and homoplantaginin [32]
9	Sida rhombifolia L.	t9.png	Subshrubs erect or prostrate, many branched, to ca. 1 m tall. Branchlets stellate. Stipules spinelike, 3-5 mm; petiole 2–5(–8) mm, stellate puberulent; leaf blade rhombic to oblong-lanceolate or obovate, rarely linear-lanceolate, 1–4.5 × 0.6-2 cm.	Scopoletin, scoporone, ethoxy-ferulate, kaempferol, kaempferol-3-O-β-D-glycosyl-6-α-D-rhamnoside, quindolinone, 11-methoxy-quindoline, quindoline [33]
10	Spigelia anthelmia L.	t10.png	Herb, annual, 15 cm–1.3 m high, the roots usually shallow and fibrous; stems ascending, leaves sessile or with petioles to 10(–20) mm long, the leaves of the terminal whorl often sessile or subsessile; blades 2–18 cm long, 0.8-6.5 cm broad, broadly ovate or rhombic-ovate to lance-oblong or lance-elliptic or lanceolate.	Spiganthine, 20-Deoxyspiganthine, 8a-Hydroxyspiganthine, 20-Norspiganthine-5-carboxylic acid, 20-Hydroxyryanodine, 9-Hydroxy-10-epi-ryanodine, 8,9-Dehydro-10-epi-ryanodine, 8a,9a-Epoxy-10-epi-ryanodine [34]

2.3 Field experiment

2.3.1 Growing media preparation and substrate characteristics

The field experiment used three types of growing media prepared by mixing ultisol soil and chicken manure (CM) with three different organic amendments: CP, RC, or SW. The final composition of each GM was as follows:

•Cocopeat-based (CPB): ultisol soil + CM + CP (2:1:1, v/v)

•Rice husk-based (RCB): ultisol soil + CM + RC (2:1:1, v/v)

•Sawdust-based (SWB): ultisol soil + CM + SW (2:1:1, v/v)

Ultisol soil was collected from the field site at a depth of 0-20 cm. CM was obtained from nearby farms that use concentrated feed. CP and RC were sourced from an agricultural store, while SW was acquired from a local lumberyard. All materials were air-dried and sieved before use. The growing media consisted of a mixture of organic amendments (CP, RC, and SW) and CM. These materials were obtained from different commercial suppliers prior to the experiment. Due to the agricultural origin of these components, their physical and chemical properties are subject to inherent variability, as the nutrient composition of CM, for instance, is influenced by feed type and season, while the quality of plant-based materials may vary depending on harvest period and processing methods. To ensure internal consistency within this study, all materials were produced within the same time frame and thoroughly homogenized before mixing to minimize within-batch variability. Consequently, the findings of this study are strictly representative of the specific batches tested, and conclusions should be interpreted within the context of these particular substrate characteristics, rather than being generalized to all available commercial products. The characteristics of these materials at the time of the research are presented in Table 3.

Table 3. Physical and chemical characteristics of the components used in growing media preparation

The physicochemical properties of the final mixed substrates, including porosity, WHC, and electrical conductivity (EC), were not directly measured in this study. EC was not measured at all due to equipment limitations. Consequently, the media-dependent responses reported herein reflect the compositional characteristics of the raw materials rather than the actual physicochemical conditions of the final mixtures.

2.3.2 Shallot cultivation method

The field experiment used Bima Brebes shallots obtained from the Hadi Sutomo nursery in Brebes, Central Java Province. The seeds used were visually healthy bulbs measuring 1.97–2.2 cm in diameter and weighing 5.8–6.3 g, with a dormant period of at least 60 days before planting. One bulb was planted per polybag (40 cm in diameter × 45 cm in height) at a depth of 3 cm. The cultivation method is described as follows:

•Synthetic chemical inputs:

○Fertilization was applied three times for each polybag. The first was given seven days after planting (DAP), consisting of 1.2 g of urea and 1.2 g of ammonium sulfate. The second fertilization, 2.4 g of nitrogen (N): phosphorus (P): potassium (K) (16:16:16), was applied 21 DAP, and 1.5 g of Kalium Chloride (KCl) on 35 DAP.

○Fungal control using ziram with a concentration of 3 g/L per week, starting from the first week until one week before harvest.

•PM inputs:

○The PM (prepared as described in section 2.2) was diluted with water at a ratio of 10 mL/L and applied as a foliar spray at intervals of 7 DAP.

•Synthetic chemical + PM input:

○Each polybag is treated with the right amount of chemical and phytochemical inputs at the specified intervals as previously described.

•Irrigation, weeding, and other agronomic practices were performed as needed uniformly across all treatments.

2.3.3 Post-harvest handling

The bulbs and leaves were cleaned of dirt and root residue. Each bulb and leaf per clump from each treatment was placed in separate containers. Sun drying was done by placing the containers on a drying rack inside a screenhouse for 7 consecutive days. During the drying period, the average daily sunlight exposure was 5.6 hours, with an average maximum temperature of 34.0 ℃ and a minimum temperature of 25.3 ℃. Air relative humidity was around 80% [22]. Physical observations were made to check for damage to the bulbs and leaves.

2.4 Variables and statistical analysis

Table 4 describes each variable used in the preliminary bioassay and the experiment. Table 5 further outlines the number of discarded replicates from the growth and yield study. These plants were discarded after visual confirmation showed they exhibited critical symptoms (like severe leaf chlorosis, curling, wilting, or root rot) for over a week, leading to their death before the crop could be harvested. All statistical analyses were performed using SPSS software v.22. The preliminary bioassay data were analyzed using the Kruskal-Wallis test. This choice was made because data from the early germination phase showed characteristics of the lag phase, where growth variability between individuals was very high, and the data distribution was not normal due to the presence of non-germinating seeds (value 0). The growth and yield data were analyzed using the Two-way Analysis of Variance (ANOVA) test with a 95% confidence level (α = 0.05). The number of samples selected for the Two-way ANOVA is described in Table 5. The linear model used for the two-way ANOVA in a factorial completely randomized design is as follows:

Y_ijk = μ + M_i + P_j + (MP)_ij + ε_ijk

where,

Y_ijk = observation value at the i-th level of media, j-th level of cultivation treatment, and k-th replication;

μ = general mean;

M_i= effect of the i-th level of planting media (i = 1, 2, 3);

P_j= effect of the j-th level of cultivation treatment (j = 1, 2, 3);

(MP)_ij = interaction effect between planting media and cultivation treatment;

ε_ijk = experimental error.

Table 4. Variable description and measurement method

Table 5. Number and status of replicates on the day of harvest

4.1 Preliminary bioassay validation

The superior radicle elongation observed in Mikania micrantha treatment suggests the presence of bioactive compounds that stimulate early root development. Previous studies identified α-Bisabolol and Eudesma-5,11(13)-dien-8,12-olide in M. micrantha extract, which possess dual bioactivity: promoting root initiation at low concentrations but inhibiting growth at higher concentrations—a characteristic feature of biostimulant compounds [36]. In the 24-hour assay, the positive effect on radicle elongation likely reflects this early-stage promoting activity. In contrast, A. sessilis is rich in water-soluble phenolic compounds (e.g., chlorogenic and gallic acids) responsible for negative growth response effects. These compounds may interfere with germination-related enzymes, inhibit mobilization of seed storage reserves, and disrupt cell membrane permeability, thereby impairing embryo development and reducing radicle elongation [37]. The severely shortened radicle length observed in A. sessilis treatment is consistent with these mechanisms. In conclusion, the bioactivity of phytochemical extracts on V. radiata is highly species-dependent. Mixing multiple species did not enhance—and may have diminished—the beneficial effects observed from individual species, highlighting the importance of species selection and concentration optimization. Nevertheless, the PM used in this study showed a tendency as a biostimulant. Future studies should investigate dose-dependent responses of purified compounds to better understand their biostimulatory and inhibitory thresholds.

4.2 Media-dependent duality: Production and post-harvest performance

The DI pattern observed appears closely related to the interaction between fungicide and WHC. In the absence of synthetic fungicide (PM treatment), high WHC increases DI [38]. When synthetic fungicide is applied (SI and PMSI), high WHC may prolong its residence time, potentially enhancing disease control. However, excessive WHC may trigger uneven fungicide distribution, thereby elevating DI risk. One possible explanation for the observed patterns is that high concentrations of iron (Fe) and manganese (Mn) could increase pathogen fungal infection risk when plants are treated with either synthetic chemicals or phytochemicals alone, but an opposite trend was observed in the PMSI. This observed reversal might be related to protective elements such as calcium (Ca), zinc (Zn), and silicon (Si) in the growing media. Conversely, SWB appears to contain only low levels of these protective elements, which could explain its reduced protection against pathogens. These results align with earlier findings that micronutrient imbalances—especially excess Fe and Mn—can worsen fungal pathogen infections, while adequate Ca, Zn, and Si strengthen plant resistance [39, 40]. The increased DI in RCB upon phytochemical supplementation (PMSI) raises the possibility of reduced Ziram fungicide activity, potentially due to chelation or binding of the Zn ion essential for Ziram's fungicidal activity [41]. Such interaction could render Ziram less effective prior to pathogen exposure. For example, quercetin (present in Euphorbia) contains hydroxyl and carbonyl groups that function as metal-binding sites [42, 43].

RCB demonstrated a higher tiller count than SWB and CPB under SI, which is consistent with its greater phosphorus content [44, 45]. However, this advantage was not observed under PMSI treatment, which may be related to phenolic acid compounds that are known to impede phosphorus absorption, thereby reducing tiller formation [46]. In contrast, the PMSI substantially elevated tiller number in CPB—an effect not observed when either treatment was applied separately—possibly due to CPB's exceptionally high WHC, which could prolong residence time of both synthetic fertilizers and phytochemicals.

Three distinct media-specific patterns emerged with clear implications for production and post-harvest parameters. CPB was the only medium where PM improved bulbs per clump compared to SI. When PM combined with SI, these improvements became substantially larger, consistent with the hypothesis that CPB high WHC may extend residence time of both inputs and enable sustained complementary action. Post-harvest quality in CPB also improved: PM and PMSI reduced weight loss during drying. Phytochemicals led to smaller bulb diameter, which reduced drying surface area per bulb, resulting in better post-drying characteristics and longer storage life [47]. For farmers, two viable options exist: using PM alone to reduce SI dependency, or using PMSI to maximize yield and post-harvest quality.

RCB presented the opposite pattern, where the PM exhibited predominantly negative growth response effects as summarized in Table 10. PM as a substitute or as a complementer of SI reduced key growth parameters below SI levels: bulbs per clump, bulb diameter, and the number of leaves. While negative growth response effects are often mediated by phytochemicals, such a mechanism appears unlikely here. An alternative hypothesis is that phytotoxicity is indirectly facilitated by microbial activity. Decomposition of RC, accelerated by microbiota from CM, might cause release and accumulation of Fe and Mn, which could exert toxic effects on plants [48, 49]. Phytochemicals (e.g., phenolics and flavonoids) are known metal chelators; it is possible that they chelated essential micronutrients or interfered with silicon uptake and translocation. Given silicon's critical role in strengthening cell walls and forming water-resistant outer skin, if disruption of silicon metabolism occurs, it could compromise bulb skin integrity and water retention [50], which may explain the higher weight loss observed during curing. The PM solely increased post-harvest weight loss, and combination with SI (PMSI) reduces this effect. For farmers, the PM offers no advantage and may be detrimental in RCB, consistent with the negative growth response effects summarized in Table 11.

SWB displayed a mixed pattern: PM alone showed positive growth with a negative yield (negative properties), while PMSI acted as a positive catalyst for both growth and yield. PMSI substantially improved both fresh and dry bulb weight parameters and reduced weight loss. However, this benefit must be weighed against elevated disease risk (26.67% under PMSI). This pattern creates a high-risk, high-reward trade-off: exceptional yield and quality performance but with the highest DI observed in the study. Unlike in RCB, phytochemicals in SWB may not have interfered with silicon metabolism, possibly due to SWB's negligible intrinsic silica content. This pattern suggests that in SWB, the mixture is only beneficial when used together with SI—a finding pointing to the unique decomposition dynamics of SWB, as discussed below.

4.3 Mechanisms underlying media-dependent effects: Water holding capacity and decomposition

The observed patterns can be explained by two interacting mechanisms: WHC and decomposition dynamics. The Performance Ratio data suggest that WHC may play a crucial role in determining the effectiveness of PM, even though WHC was not directly measured in this study. In the case of CPB, which is known for its high WHC, PM improved bulbs per clump by 9% compared to SI. However, in RCB, which has low WHC, PM resulted in a decrease in bulbs per clump by 13% and bulb diameter by 21%. SWB, which has intermediate WHC, also showed negative effects under PM; the number of leaves decreased by 46%, bulb diameter by 11%, fresh weight by 2%, and weight loss increased by 2%. These contrasting results suggest that a sufficient level of WHC may be necessary for the mixture to exhibit a positive effect. However, this hypothesis needs to be verified directly through WHC measurement.

The relationship between WHC and DI is more complex. In the case of CPB, high WHC may have acted as a protective factor when synthetic fungicide was present–potentially by extending fungicide residence time–but became a risk factor when fungicide was absent, as persistent moisture may create an ideal environment for fungal proliferation. Thus, high-WHC media may be safe only when fungicide protection is maintained. The pattern is consistent with the observation that DI in CPB decreased from 20.00% (PM) to 3.33% (PMSI) when fungicide was added.

A potential mechanism for the observed disease patterns may involve nitrogen availability. Pathogenic fungi might act as nitrogen scavengers: when nitrogen is plentiful, they could live saprophytically; when nitrogen becomes scarce, they might switch to a pathogenic mode [39]. This mechanism could explain the pattern observed in CPB, where the addition of synthetic nitrogen reduced DI from 20.00% to 3.33%. However, this hypothesis does not fully account for the behavior seen in RCB, where DI remained at 20.00% regardless of nitrogen addition, or in SWB, where nitrogen addition unexpectedly increased DI from 10.00% to 26.67%. In RCB, other factors—such as negative phytochemical–fungicide interaction—may have overridden the nitrogen effect. In SWB, active decomposition dynamics likely favored pathogen fungi regardless of nitrogen availability.

Decomposition dynamics in SWB represent a high-risk, high-reward environment. We hypothesize that SW underwent significant decomposition during the two-month experiment, in contrast to CP, which is largely inert, and RC, which is protected by silica. This hypothesis is based on the known chemical composition of SW, which contains extractable compounds such as resins, terpenes, and phenolics, as well as labile carbohydrates that support microbial activity–though the specific composition of the SW used in this study was not analyzed. Additionally, the distinct growth and disease patterns observed in SWB align with this interpretation. We propose that decomposition may have released bioavailable compounds, such as simple sugars, organic acids, and phenolic fragments, which could have contributed to enhanced root growth, nutrient uptake, and stress tolerance. The presence of PM may have accelerated this decomposition, while synthetic fertilizers provided nitrogen to prevent any immobilization that could limit growth. However, this same process might have also favored fungi: the release of simple sugars could have provided food for pathogens, temporary nitrogen immobilization may have created nitrogen stress, and the loss of natural antimicrobials could have diminished chemical defenses. The net outcome was a high-risk, high-reward environment—offering exceptional growth potential but also increasing the likelihood of disease outbreaks. This hypothesis explains the SWB paradox: while PMSI produced the strongest putative supplementary enhancer for growth, productivity, and yield, it also led to the highest DI (26.67%).

A concerning pattern has emerged in RCB and SWB regarding the compatibility of fungicides. In RCB, DI was identical at 20.00% with or without the application of fungicide. In SWB, PMSI showed a significantly higher DI of 26.67%, compared to 10.00% for PM. This negative interaction suggests that the combination may interfere with the effectiveness of fungicide through several possible mechanisms that warrant further investigation. For example, adsorptive compounds such as tannins and phenolics may bind to the active ingredient; the mixture could stimulate the microbial degradation of the fungicide, or sublethal exposure might induce cross-tolerance in pathogenic fungi. Importantly, this interaction was dependent on the media used, as it did not occur in CPB, where WHC may have allowed fungicide residues to persist long enough to mitigate interference. However, none of these proposed mechanisms were directly tested in this study and should be regarded as hypotheses that require experimental verification.

4.4. Practical recommendations for shallot farmers

Based on the observed data, the following media-specific recommendations are offered. These are derived from a single controlled study and should be validated under commercial conditions:

•CPB: Use PMSI for maximum yield and post-harvest quality while maintaining fungicide; PM alone is an option only if farmers accept moderate disease risk.

•RCB: Conventional SI remains the preferred approach. PM is not advised; if used, expect reduced growth, higher weight loss, and no disease benefit.

•SWB: PMSI is a high-risk, high-reward option. Only adopt if farmers can implement additional disease management (e.g., shorter cycles, fungicide rotation, disease-free planting material, or pre-composting SW). PM alone is not recommended.

In all growing media, PM alone does not suppress disease. Combining PM with fungicide showed media-dependent incompatibility: effective in CPB, ineffective in RCB, and detrimental in SWB. Farmers should not reduce fungicide applications when using PM, especially in RCB and SWB. Economic analyses and multicycle field validations are needed before large-scale adoption.

4.5 Limitations and future directions

Several limitations should be acknowledged when interpreting the findings of this study. These limitations relate to chemical characterization, experimental design, statistical analysis, the scope of mechanistic inference, and lack of etiological confirmation of the observed disease. First, the phytochemical composition of the mixture was not chemically analyzed, so specific bioactive compounds responsible for the observed positive or negative growth response effects remain unknown. Second, another limitation concerns the physicochemical characterization of the final mixed substrates. Parameters such as actual C/N ratio, EC, porosity, WHC, and available nutrient contents of the final growing media (CPB, RCB, and SWB) were not directly measured. In particular, EC was not measured at all due to equipment limitations, which prevented evaluation of substrate salinity as a potential factor influencing plant growth and DI. It is important to acknowledge that the components of the growing media are natural by-products with inherent batch-to-batch variation. Since the materials used in this study were obtained from multiple suppliers but within a single procurement period, the reproducibility of the absolute nutrient values is assured for those specific batches. However, caution should be exercised when generalizing these findings to materials from different suppliers, harvest seasons, or production cycles. Future studies employing substrates from varied sources are recommended to validate the consistency of these results.

Third, the study lacks pathogen isolation, microscopy, or molecular confirmation, although the symptoms were very similar to those typically caused by Fusarium sp. Conclusions regarding disease percentages are descriptive and based solely on macroscopic symptoms, which may limit our interpretative power. Fourth, the study was conducted under controlled conditions and may not fully represent commercial production systems. The two-month duration of the shallot experiment captured only a single growing cycle; longer-term studies are needed to assess the persistence of effects and potential cumulative impacts of repeated phytochemical applications. Fifth, several limitations of the statistical analysis should be acknowledged. While two-way ANOVA revealed significant main and interaction effects for most variables, the effect sizes (Partial Eta Squared) varied considerably, ranging from 0.035 (small effect) to 0.621 (very large effect). Fifth, the proposed mechanisms were not directly measured in this study. We did not measure decomposition rates (e.g., CO₂ evolution, C/N ratio), nitrogen availability over time, fungicide residues, or pathogen fungi population density. Consequently, these mechanisms should be interpreted as hypotheses requiring direct experimental verification rather than established explanations.

To address these limitations, future research should investigate the following priorities: (i) chemical analysis of the PM using LC-MS to identify specific bioactive compounds, followed by dose-response experiments with purified compounds to establish biostimulant versus phytotoxic thresholds; (ii) Future studies should prioritize direct measurement of EC, WHC, porosity, decomposition rates, and nutrient dynamics in each growing medium, including time-series measurements of Fe, Mn, Ca, Zn, and Si concentrations in soil solution on the final substrates to enable more robust interpretation; (iii) Longer-term and multi-cycle experiments to assess the persistence of biostimulant effects, potential accumulation of the phytochemicals or their degradation products, and impacts on soil microbial communities; and (iv) combines comprehensive phytopathological diagnostics, including Koch’s postulates and molecular coding, to definitively identify the pathogen involved and to establish a clear disease etiology. Despite these limitations, this study provides a valuable empirical foundation for understanding the media-dependent effects of a PM on shallot growth, DI, and post-harvest quality. The consistent patterns observed across multiple variables and the clear media-specific responses offer a robust basis for generating testable hypotheses and guiding future research. However, the proposed mechanisms should be treated as hypotheses requiring direct verification, and practical recommendations should be applied with caution pending field validation.

This work is supported by the Regional Research and Innovation Agency of South Sumatra Province, Indonesia and the Regional Research and Innovation Agency of Ogan Ilir Regency, Indonesia.