Seed Quality of Lablab Bean ( Lablab purpureus ) as Inﬂuenced by Seed Maturity and Drying Methods

: The lablab bean or ‘Kacang sepat’ is a food crop originating from India and is commonly found in tropical areas such as Indonesia, Thailand and the Philippines due to its adaptability. However, in Malaysia, lablab is grown as a backyard crop, and the area cultivated with this crop is still limited. Lablab has the potential to become a commercial vegetable crop in Malaysia due to its high protein content (18–25%). Therefore, to increase lablab cultivation, good quality seeds are a prerequisite. A key factor in the production of quality seeds is the ideal harvest time of the pods during seed development and maturation and based on identiﬁable characteristics. The ﬁrst part of this study was conducted to assess the physical characteristics of lablab pods and seeds of the cultivar Highworth (MDI 12842). Plants were grown using standard cultural practices. Pods and seeds were collected at ten different maturity stages (5, 10, 15, 20, 25, 30, 35, 40, 45, and 50 days after anthesis (DAA)). Pods and seeds showed a signiﬁcant difference in size and colour at different maturity stages. Seeds harvested at 20 DAA had maximum pod and seed size. The seeds attained physiological maturity (PM) at 30 DAA when the pod colour is light reddish brown and, after being subjected to sun and oven drying, gave maximum germination percentages of 89% and 91%, respectively, while fresh seeds had only 78% germination. Results indicated that the germination percentage of lablab improved, regardless of the drying method used. Therefore, the lablab bean pod colour is a suitable indicator to be used as an easy method for the identiﬁcation of the best time for pod harvesting for seed production.


Introduction
Beans, often known as legumes, are the second most important food source in the human diet. They are a rich source of proteins, carbohydrates, fibre, vitamins and minerals. Long beans, French beans and four-angled beans are among the most popular ones. According to Rudrappa [1], the protein content in the above-mentioned beans ranges between 3 and 12%. Many indigenous and underutilized beans have a much higher protein content, including the lablab bean. According to Al-Snafi [2], lablab beans have many local names, such as Hyacinth bean, Field bean, Indian bean, Sem, Simii and Bian dou. This bean can grow well under tropical and subtropical conditions. It is famous in countries such as India, Bangladesh, Africa, and Indonesia but not widely grown in others such as in Malaysia. The lablab bean is grown in Malaysia only as a backyard crop, with the current area cultivated being very low. To increase lablab bean cultivation to a commercial scale, good quality seeds should be made available locally.
This bean is a good candidate as a commercial vegetable due to the high protein content (18-25%) [3,4]. Therefore, good quality seeds must be produced locally to promote and increase lablab bean cultivation. Based on the literature, there are three cultivars of lablab bean: Rongai, with white flowers and light brown seeds when mature; Highworth, with purple flowers and black seeds; and Endurance, with white flowers and black seeds [5,6].
However, based on Beckett [7], the cultivar Highworth has better commercial potential due to early flowering and maturation characteristics, produces high yield, and pod maturity is more uniform compared with Rongai and Endurance. Fortunately, this cultivar is the most commonly found lablab in Malaysia, making it easier to introduce at the commercial level.
Harvesting seeds at the right time helps to ensure high seed quality. Good quality seeds are an important factor in ensuring maximum seed germination and superior seedling establishment. The lablab bean is a creeping plant with indeterminate flowering characteristics. Hence, they can have flowers, young pods and overly mature pods in one cycle. This complicates seed production, as pods must be harvested multiple times at the right maturity stage. Thus, harvesting at the correct developmental stage must be established based on a clear, simple visual indicator, such as changes in colour based on physiological responses. In addition, days after anthesis (DAA) and dry weight can be used as indicators to achieve high seed quality. Colour changes are widely used as the indicator for farmers, as legumes can change colour from early development until late harvesting. Cianzio and Ortiz [8], who studied seed development in soybean, an indeterminate crop, concluded that pod colour was the most reliable method to determine the correct time for harvesting. Therefore, an understanding of the physiological state of the seed and its association with pod colour must be established.
Seeds, upon reaching the physiological maturity stage, will undergo severance of the funiculus, which triggers a net loss in moisture. This net loss in moisture, or desiccation, is a process that initiates the switch from development to germination, as mentioned by Bewley et al. [9]. Plants with indeterminate growth characteristics will produce seeds in a wide range of maturity stages; thus, it is difficult to collect seeds at a single, optimized time. The chances of collecting relatively immature seeds can be high. Immature seeds must be at a stage that can tolerate desiccation in order for them to survive and have a similar function to mature seeds, particularly in initiating the switch from development to germination. Thus, the quality can be improved by drying, to mimic the natural desiccation of mature seeds.
Desiccation improved seed quality in Vicia sativa according to Samarah [10] and Ceiba pentandra according to Lima et al. [11]. Over-mature seeds (late harvesting) can become dormant. Thus, the objectives of this study were to determine the ideal developmental stage for harvesting pods of indeterminate lablab beans for maximum seed quality and to evaluate the effect of drying on the viability of lablab bean seeds harvested at different maturity stages.
According to Vidigal et al. [12], harvesting seeds at an ideal time can minimize seed losses and reduce the ratio of unripe seeds due to early harvest and can maximize seed germination, vigour and viability. Seed quality is said to be maximum at the end of the seed filling period [13][14][15], termed physiological maturity (PM) by Harington [13]. Beyond this point, the deterioration process begins, which gradually reduces the physiological quality of the seeds. However, numerous evidences is also present to show that seed quality can increase after PM. Since lablab has indeterminate growth habits, identifying the correct time to harvest is needed to avoid seed loss.
Seed development is the process that begins with the formation of the zygote and endosperm; it comprises three phases, i.e., histodifferentiation, seed filling and desiccation. There is a rapid cell division and formation of embryonic cells with a high moisture content at the first phase, histodifferentiation. The second phase is seed filling, categorized by cell expansion and accumulation of storage proteins, lipids and carbohydrates [16,17]. This period is also characterized by a decrease in moisture content, with an increase in seed dry weight. Therefore, when seeds reach the maximum level of seed dry weight, they are physiologically mature. The last phase is the desiccation phase. This phase involves a major loss in water that leads to dry seeds, in preparation for a quiescent period, as mentioned by Angelovici [18].
Drying seeds right after harvest is important in seed production. Yousaf et al. [19] reported that up to 30% losses can occur because of the post-harvest handling and drying Agronomy 2022, 12, 363 3 of 12 of seeds. Seed drying is intended to reduce the seeds' moisture content to safe limits to maintain its viability and vigour during storage [20], which would otherwise deteriorate quickly due to mould growth, heating and enhanced microbial activity [21]. Orthodox seeds can be dried to low moisture content to increase longevity. While recalcitrant seeds such as Cocos nucifera will lose their viability once dried to low moisture content, lablab belongs to the orthodox seed category [22]. There are many drying methods based on the literature, such as sun, shade and vacuum drying. Nevertheless, because of the high cost of drying facilities and maintenance, seed producers in many developing countries prefer sun, shade and oven drying, due to low initial capital and ease of operation, with no complicated technological protocol.
Therefore, it is important to understand the seed development process, particularly in an indeterminate crop species, to identify important stages of seed development. Seeds harvested at the appropriate stages must be subjected to drying; hence, the most suitable and practical drying method must be established for improved seed quality.

Materials and Field Experiments
A completely randomized design experiment with four replications was conducted at Field 15, Faculty of Agriculture, Universiti Putra Malaysia (UPM), to determine the ideal developmental stage for the production of quality lablab bean seeds. The lablab bean cultivar Highworth was obtained from MARDI Seberang Perai, Pulau Pinang, Malaysia. A land area of 5 m × 5 m was ploughed, harrowed and levelled before sowing lablab seeds. Staking with 1 inch × 2 inch wood was constructed, as lablab is a climbing herbaceous crop with a total height of 8 ft. Then, seeds were directly sown in the soil and watered twice daily using a sprinkler system. Plants were arranged in a square planting system, with a planting distance of 30 cm × 50 cm. Silver shine was used to control weed growth around the plants. In order to control pest attacks, plants were sprayed with pesticides 3 times per season. Infected plants were sprayed with Vita Dimethoate (VITACHEM (M) SDN. BHD) (14 mL for 10 L water) to control pests. The flowers that opened were tagged, and this was used to count the number of days after anthesis (DAA). In other words, this is the age of the pods (days) when they were harvested. Flowers were tagged upon anthesis using different coloured thread to indicate different maturity stages. Then, pods having the same-coloured thread at five days intervals-5, 10,15,20,25,30,35,40,45, and 50 DAA-were harvested.

Colour and Seed and Pod Size
Pod and seed colour were recorded as an indicator for seed maturity stages using the Royal Horticultural Society (RHS) Colour Chart, Edition V. Lablab seeds were harvested at different maturity stages, and the pod and seed length and width were measured using Vernier callipers.

Moisture Content and Seed Dry Weight
Seed moisture content and seed dry weight were determined to identify the time when seeds reached physiological maturity. Then, the seeds were dried using two different drying methods; sun and oven drying. Seed moisture content was determined using the high constant temperature oven method, in accordance with ISTA [23].

Moisture content
where W1 = weight of the container with lid W2 = weight of the container with lid and sample before drying W3 = weight of the container with lid and sample after drying

Protein Content
Crude protein content was determined using the Kjedahl method as described by Pearson (1970) [24], with slight modifications. Lablab bean seeds were ground using a Labtac CT410 Grinder and were ground to a fine powder. Seed powder of 0.2 g was weighed and transferred into a digestion tube. The sample was neutralised with one Kjedhal catalyst (5 g K 2 SO 4 + 0.5 g CuSO 4 × 5 H 2 O) tablet and 6 mL of 98% sulphuric acid. The digestion tube was heated moderately in a fume hood at 400 • C for 1.5 h until the solution appeared clear. Then, the solution was allowed to cool, and all the samples were analysed for crude protein content using Kjeltec TM8400 Analyzer Unit (Foss NRISystems, Denmark).

Germination Percentage
Germination tests were conducted using sand as the germination media, with 25 seeds per replicate for each treatment. Germinated seeds were counted every day for up to 10 days. Seeds were considered as germinated when essential structures such as the radicle and plumule were observed. The germination percentage was calculated as per the formula below, as recommended by ISTA [23].
Germination percentage = number of seeds germinated total number of seeds × 100 (2)

Germination Index (GI)
Germination Index (GI) was calculated as specified by Kader [25]. The germination index is a parameter that combines both germination percentage and speed.
where n1, n2, n3, is the number of germinated seeds on the first, second and subsequent days until day 10.

Electrical Conductivity
Electrical conductivity is a test to measure viability and seedling vigour [26]. A total of 10 seeds per replicate were used for each treatment and were placed in a vial containing 80 mL distilled water at 25 • C for 24 h. After 24 h of soaking, seeds were swirled for 10-15 s, and the conductivity meter was dipped into the water in the cell until a stabilized reading was achieved and recorded. The mean of four plastic vials containing distilled water only (control) served as the background reading. An electrical conductivity test was calculated as per the formula below, recommended by ISTA [23].
Conductivity test (µS cm −1 g −1 ) = ( Conductivity reading µS cm −1 − background reading weight of replicate (g) ) (4) pod and seed started to shrink, until 50 DAA. This is supported by pictures showing the physical observation of pod and seed of lablab cultivar Highworth in Table 1. Correlation analyses indicate that there is highly significant correlation between the moisture content with seed width and no significant correlation between seed length, pod length and pod width, as shown in Table 2. seed production [27]. In lablab the size of the pod increased and was a maximum at 20 DAA, with an average pod length of 81 mm and pod width of 15 mm. Maximum seed size was achieved at 15 DAA, 13 mm seed length and 12 mm seed width. Then, both the pod and seed started to shrink, until 50 DAA. This is supported by pictures showing the physical observation of pod and seed of lablab cultivar Highworth in Table 1. Correlation analyses indicate that there is highly significant correlation between the moisture content with seed width and no significant correlation between seed length, pod length and pod width, as shown in Table 2. seed production [27]. In lablab the size of the pod increased and was a maximum at 20 DAA, with an average pod length of 81 mm and pod width of 15 mm. Maximum seed size was achieved at 15 DAA, 13 mm seed length and 12 mm seed width. Then, both the pod and seed started to shrink, until 50 DAA. This is supported by pictures showing the physical observation of pod and seed of lablab cultivar Highworth in Table 1. Correlation analyses indicate that there is highly significant correlation between the moisture content with seed width and no significant correlation between seed length, pod length and pod width, as shown in Table 2. seed production [27]. In lablab the size of the pod increased and was a maximum at 20 DAA, with an average pod length of 81 mm and pod width of 15 mm. Maximum seed size was achieved at 15 DAA, 13 mm seed length and 12 mm seed width. Then, both the pod and seed started to shrink, until 50 DAA. This is supported by pictures showing the physical observation of pod and seed of lablab cultivar Highworth in Table 1. Correlation analyses indicate that there is highly significant correlation between the moisture content with seed width and no significant correlation between seed length, pod length and pod width, as shown in Table 2. seed production [27]. In lablab the size of the pod increased and was a maximum at 20 DAA, with an average pod length of 81 mm and pod width of 15 mm. Maximum seed size was achieved at 15 DAA, 13 mm seed length and 12 mm seed width. Then, both the pod and seed started to shrink, until 50 DAA. This is supported by pictures showing the physical observation of pod and seed of lablab cultivar Highworth in Table 1. Correlation analyses indicate that there is highly significant correlation between the moisture content with seed width and no significant correlation between seed length, pod length and pod width, as shown in Table 2. seed production [27]. In lablab the size of the pod increased and was a maximum at 20 DAA, with an average pod length of 81 mm and pod width of 15 mm. Maximum seed size was achieved at 15 DAA, 13 mm seed length and 12 mm seed width. Then, both the pod and seed started to shrink, until 50 DAA. This is supported by pictures showing the physical observation of pod and seed of lablab cultivar Highworth in Table 1. Correlation analyses indicate that there is highly significant correlation between the moisture content with seed width and no significant correlation between seed length, pod length and pod width, as shown in Table 2. seed production [27]. In lablab the size of the pod increased and was a maximum at 20 DAA, with an average pod length of 81 mm and pod width of 15 mm. Maximum seed size was achieved at 15 DAA, 13 mm seed length and 12 mm seed width. Then, both the pod and seed started to shrink, until 50 DAA. This is supported by pictures showing the physical observation of pod and seed of lablab cultivar Highworth in Table 1. Correlation analyses indicate that there is highly significant correlation between the moisture content with seed width and no significant correlation between seed length, pod length and pod width, as shown in Table 2. seed production [27]. In lablab the size of the pod increased and was a maximum at 20 DAA, with an average pod length of 81 mm and pod width of 15 mm. Maximum seed size was achieved at 15 DAA, 13 mm seed length and 12 mm seed width. Then, both the pod and seed started to shrink, until 50 DAA. This is supported by pictures showing the physical observation of pod and seed of lablab cultivar Highworth in Table 1. Correlation analyses indicate that there is highly significant correlation between the moisture content with seed width and no significant correlation between seed length, pod length and pod width, as shown in Table 2.  seed production [27]. In lablab the size of the pod increased and was a maximum at 20 DAA, with an average pod length of 81 mm and pod width of 15 mm. Maximum seed size was achieved at 15 DAA, 13 mm seed length and 12 mm seed width. Then, both the pod and seed started to shrink, until 50 DAA. This is supported by pictures showing the physical observation of pod and seed of lablab cultivar Highworth in Table 1. Correlation analyses indicate that there is highly significant correlation between the moisture content with seed width and no significant correlation between seed length, pod length and pod width, as shown in Table 2.  seed production [27]. In lablab the size of the pod increased and was a maximum at 20 DAA, with an average pod length of 81 mm and pod width of 15 mm. Maximum seed size was achieved at 15 DAA, 13 mm seed length and 12 mm seed width. Then, both the pod and seed started to shrink, until 50 DAA. This is supported by pictures showing the physical observation of pod and seed of lablab cultivar Highworth in Table 1. Correlation analyses indicate that there is highly significant correlation between the moisture content with seed width and no significant correlation between seed length, pod length and pod width, as shown in Table 2.  seed production [27]. In lablab the size of the pod increased and was a maximum at 20 DAA, with an average pod length of 81 mm and pod width of 15 mm. Maximum seed size was achieved at 15 DAA, 13 mm seed length and 12 mm seed width. Then, both the pod and seed started to shrink, until 50 DAA. This is supported by pictures showing the physical observation of pod and seed of lablab cultivar Highworth in Table 1. Correlation analyses indicate that there is highly significant correlation between the moisture content with seed width and no significant correlation between seed length, pod length and pod width, as shown in Table 2.  seed production [27]. In lablab the size of the pod increased and was a maximum at 20 DAA, with an average pod length of 81 mm and pod width of 15 mm. Maximum seed size was achieved at 15 DAA, 13 mm seed length and 12 mm seed width. Then, both the pod and seed started to shrink, until 50 DAA. This is supported by pictures showing the physical observation of pod and seed of lablab cultivar Highworth in Table 1. Correlation analyses indicate that there is highly significant correlation between the moisture content with seed width and no significant correlation between seed length, pod length and pod width, as shown in Table 2.  seed production [27]. In lablab the size of the pod increased and was a maximum at 20 DAA, with an average pod length of 81 mm and pod width of 15 mm. Maximum seed size was achieved at 15 DAA, 13 mm seed length and 12 mm seed width. Then, both the pod and seed started to shrink, until 50 DAA. This is supported by pictures showing the physical observation of pod and seed of lablab cultivar Highworth in Table 1. Correlation analyses indicate that there is highly significant correlation between the moisture content with seed width and no significant correlation between seed length, pod length and pod width, as shown in Table 2.  seed production [27]. In lablab the size of the pod increased and was a maximum at 20 DAA, with an average pod length of 81 mm and pod width of 15 mm. Maximum seed size was achieved at 15 DAA, 13 mm seed length and 12 mm seed width. Then, both the pod and seed started to shrink, until 50 DAA. This is supported by pictures showing the physical observation of pod and seed of lablab cultivar Highworth in Table 1. Correlation analyses indicate that there is highly significant correlation between the moisture content with seed width and no significant correlation between seed length, pod length and pod width, as shown in Table 2.  seed production [27]. In lablab the size of the pod increased and was a maximum at 20 DAA, with an average pod length of 81 mm and pod width of 15 mm. Maximum seed size was achieved at 15 DAA, 13 mm seed length and 12 mm seed width. Then, both the pod and seed started to shrink, until 50 DAA. This is supported by pictures showing the physical observation of pod and seed of lablab cultivar Highworth in Table 1. Correlation analyses indicate that there is highly significant correlation between the moisture content with seed width and no significant correlation between seed length, pod length and pod width, as shown in Table 2.  seed production [27]. In lablab the size of the pod increased and was a maximum at 20 DAA, with an average pod length of 81 mm and pod width of 15 mm. Maximum seed size was achieved at 15 DAA, 13 mm seed length and 12 mm seed width. Then, both the pod and seed started to shrink, until 50 DAA. This is supported by pictures showing the physical observation of pod and seed of lablab cultivar Highworth in Table 1. Correlation analyses indicate that there is highly significant correlation between the moisture content with seed width and no significant correlation between seed length, pod length and pod width, as shown in Table 2.  seed production [27]. In lablab the size of the pod increased and was a maximum at 20 DAA, with an average pod length of 81 mm and pod width of 15 mm. Maximum seed size was achieved at 15 DAA, 13 mm seed length and 12 mm seed width. Then, both the pod and seed started to shrink, until 50 DAA. This is supported by pictures showing the physical observation of pod and seed of lablab cultivar Highworth in Table 1. Correlation analyses indicate that there is highly significant correlation between the moisture content with seed width and no significant correlation between seed length, pod length and pod width, as shown in Table 2.  seed production [27]. In lablab the size of the pod increased and was a maximum at 20 DAA, with an average pod length of 81 mm and pod width of 15 mm. Maximum seed size was achieved at 15 DAA, 13 mm seed length and 12 mm seed width. Then, both the pod and seed started to shrink, until 50 DAA. This is supported by pictures showing the physical observation of pod and seed of lablab cultivar Highworth in Table 1. Correlation analyses indicate that there is highly significant correlation between the moisture content with seed width and no significant correlation between seed length, pod length and pod width, as shown in Table 2.  seed production [27]. In lablab the size of the pod increased and was a maximum at 20 DAA, with an average pod length of 81 mm and pod width of 15 mm. Maximum seed size was achieved at 15 DAA, 13 mm seed length and 12 mm seed width. Then, both the pod and seed started to shrink, until 50 DAA. This is supported by pictures showing the physical observation of pod and seed of lablab cultivar Highworth in Table 1. Correlation analyses indicate that there is highly significant correlation between the moisture content with seed width and no significant correlation between seed length, pod length and pod width, as shown in Table 2.  seed production [27]. In lablab the size of the pod increased and was a maximum at 20 DAA, with an average pod length of 81 mm and pod width of 15 mm. Maximum seed size was achieved at 15 DAA, 13 mm seed length and 12 mm seed width. Then, both the pod and seed started to shrink, until 50 DAA. This is supported by pictures showing the physical observation of pod and seed of lablab cultivar Highworth in Table 1. Correlation analyses indicate that there is highly significant correlation between the moisture content with seed width and no significant correlation between seed length, pod length and pod width, as shown in Table 2.  seed production [27]. In lablab the size of the pod increased and was a maximum at 20 DAA, with an average pod length of 81 mm and pod width of 15 mm. Maximum seed size was achieved at 15 DAA, 13 mm seed length and 12 mm seed width. Then, both the pod and seed started to shrink, until 50 DAA. This is supported by pictures showing the physical observation of pod and seed of lablab cultivar Highworth in Table 1. Correlation analyses indicate that there is highly significant correlation between the moisture content with seed width and no significant correlation between seed length, pod length and pod width, as shown in Table 2.  The colour of the pod shows changes, from strong yellow-green at 5 DAA and then slowly turning to a light reddish-brown at 30 DAA. For the external visual characteristics, the seeds from the early stage (5-15 DAA) had shiny and smooth surfaces, while seeds at 30-50 DAA were black in colour, with a non-shiny surface as summarized in Table 1. The seed size was smaller after undergoing the desiccation process. The increase in seed size is attributed to the higher food reserve and respiratory rate, as stated by Massimi [28]. Figure 1 shows that seed moisture content decreased with an increment in seed dry weight from early development stages to late maturity stages. An increase in seed size at the early stage and the significant decrease of moisture content at the end of the maturation process is mentioned by Barnwal et al. [29] and also has been reported for okra [30] and Agronomy 2022, 12, 363 6 of 12 soybean seeds [31]. In relation to maturation stages, pod colour can be used as a visual indicator to establish the ideal time for harvesting [32,33]. The pods were green from early development until 20 DAA and then turned to greenish-yellow. Around 30 DAA, the pod colour turned from light reddish-brown to dark greyish yellow-brown. It is presumed that the change in colour could be due to the changes in the fruits' internal tissue, while the brown appearance may be associated with chlorophyll degradation and synthesis of pre-existing pigment, responsible for the characteristic colour for each plant species [34]. Therefore, the pod colour of lablab is suitable to be used as an easy indicator for establishing the ideal harvesting time in seed production: when the colour turned from light reddish-brown (30 DAA) to dark greyish yellow-brown (35 DAA).  The moisture content was 80.78% at 5 DAA. This is due to the mobilization of water at the early phase when water is actively mobilized into the seeds. The potential water binding sites at the molecular level are saturated, and solvent water is available for biochemical activities to allow cell division and growth [9]. Moisture content is high at 80%, because the nucleic acid and protein are still undergoing synthesis and have liquid water being pulled into the capillaries and structure. Thus, it can generate sufficient tension to retain water in seeds, as mentioned by Bewley et al. [9]. At this stage, plant embryogenesis The moisture content was 80.78% at 5 DAA. This is due to the mobilization of water at the early phase when water is actively mobilized into the seeds. The potential water binding sites at the molecular level are saturated, and solvent water is available for biochemical activities to allow cell division and growth [9]. Moisture content is high at 80%, because the nucleic acid and protein are still undergoing synthesis and have liquid water being pulled into the capillaries and structure. Thus, it can generate sufficient tension to retain water in seeds, as mentioned by Bewley et al. [9]. At this stage, plant embryogenesis begins with a globular shape and is mainly filled with liquid endosperm (Figure 1I). At 15 DAA, the embryo shifts into a heart shape and torpedo shape at 25 DAA ( Figure 1II). At 30 DAA, the moisture content drops to 27.3% and continues to drop until 50 DAA, to 6.22%. The seeds attain maximum seed dry weight at 30 DAA, at 4.478 g/10 seeds, and embryos start to elongate the external shape, as shown in Figure 1III. The cotyledon begins to bend, and the embryo elongates, with a complete forming of the mature embryo. Seed moisture content starts to decrease because of the utilization of water in various metabolic activities and the removal of water by the desiccation of the surrounding environment, as stated by Gurusamy and Thiagarajan [35].
The seed dry weight increases gradually, and the peak is obtained at 30 DAA. The highest seed dry weight can be an indicator of physiological maturity (PM) and can be used to obtain high seed quality, as mentioned by Browne [36] and Harington [13]. Seeds at 30 DAA recorded maximum seed dry weight, with no further increase, presumably due to the severance of the funiculus, causing a cessation of the translocation of soluble substances into the seeds. At this stage, the embryo expended the cotyledon without cell division, and became a matured embryo, developing shoot and root apical meristem. According to the study by Das and Fakir [37] on lablab beans (DS-52 and DS 106 genotypes), PM is achieved around 33 days after flowering (DAF), when seed dry weight is 7.61 g/pod for DS-52 genotypes and 7.17 g/pod for DS-106 gentoypes. Therefore, the variation of PM in days could be due to the difference of genotype, location and growing conditions, as stated by Das and Fakir [37].
Protein content showed significant differences at different maturity stages during seed development. The protein was highest at 10 DAA, at 28.5%, and then decreased until 50 DAA, as shown in Figure 2. The percentage of crude protein decreased until 50 DAA due to the utilization of protein for pod growth, in conjunction with the increase of starch and energy content [38,39]. This result is in line with studies by Hill and Breidenbach [40] and Siddique and Wright [41]; they also found that seed protein content increased very rapidly in the early development stage of soybean during seed development and decreased afterward, possibly due to the onset of seed desiccation. Protein content showed significant differences at different maturity stages during seed development. The protein was highest at 10 DAA, at 28.5%, and then decreased until 50 DAA, as shown in Figure 2. The percentage of crude protein decreased until 50 DAA due to the utilization of protein for pod growth, in conjunction with the increase of starch and energy content [38,39]. This result is in line with studies by Hill and Breidenbach [40] and Siddique and Wright [41]; they also found that seed protein content increased very rapidly in the early development stage of soybean during seed development and decreased afterward, possibly due to the onset of seed desiccation.

Effect of Maturity Stages and Different Drying Methods on Germination Percentage and Germination Index
To ensure successful crop production and seedling establishment, germination percentage should be included as an important parameter, as noted by Almansouri [42]. A good drying method is important to ensure seed quality in the form of germination percentage and speed, as presented in Figure 3. Data are means of triplicate measurements, and bars indicate the standard error.

Effect of Maturity Stages and Different Drying Methods on Germination Percentage and Germination Index
To ensure successful crop production and seedling establishment, germination percentage should be included as an important parameter, as noted by Almansouri [42]. A good drying method is important to ensure seed quality in the form of germination percentage and speed, as presented in Figure 3. Data are means of triplicate measurements, and bars indicate the standard error.

Germination Index
To ensure successful crop production and seedling establishment, germinati centage should be included as an important parameter, as noted by Almansouri good drying method is important to ensure seed quality in the form of germinati centage and speed, as presented in Figure 3. Data are means of triplicate measure and bars indicate the standard error. At 20 DAA, seeds were immature and had only 33% germination; howev using an oven improved germination to 82%. Several researchers have reported At 20 DAA, seeds were immature and had only 33% germination; however, drying using an oven improved germination to 82%. Several researchers have reported that drying, particularly slow drying, can enhance the ability to tolerate desiccation and improve germination [17,31]. The improvement in germination is due to the improvement of the cell membrane integrity.
Seeds harvested at 30 DAA naturally developed the ability to germinate, recording germination up to 78%. Similar results have been noted in many crops, with high germination being recorded at physiological maturity. When seeds at this stage of development were subjected to sun and oven drying, they recorded a high percentage, 89% and 91%, respectively. It is very interesting to note that lablab bean seeds can achieve maximum germination percentage and germination index, irrespective of the drying method used. Based on Kader [25], the higher the germination index (GI) value, the higher the per-centage and rate of germination. Even though there was no significant effect on germination percentage and germination index with different drying methods, it was shown that drying types can enhance germination percentage as well as the germination index of lablab bean seeds. From this data, we can see that the drying methods (sun and oven) can be equally good in reducing seed moisture content in maintaining the viability of the seeds. Other crops also showed a similar response to drying, improving seed quality before physiological maturity; these include sorghum [43], peanut [44], and carrot seeds [45].
Seed qualities, such as high germination percentage, viability, vigour, storability and germination speed can be influenced by the harvesting stage. Studies by Wright and Siddique [46] on the viability and vigour of Pisum sativum seeds achieved a maximum germination percentage of 96% when seeds were dried in an oven at 40 • C. The present findings also support Enoch et al. [47], who concluded that the germination percentage of Bambara nut, cowpea and maize seeds that underwent sun drying had improved germination (>90%) at all maturity stages compared with shade and room drying.

Effect of Maturity Stages and Different Drying Methods on Electrical Conductivity (EC)
The lowest value of EC was obtained at 30 DAA with the oven and sun drying method, showing a reading of 94.23 µS cm −1 g −1 and 96.52 µS cm −1 g −1 , respectively, which increased thereafter. Less electrolyte leakage was observed at 30 DAA, which could be attributed to good cell membrane integrity; intactness was able to minimize seed leakage and increase in the deposition of storage reserves [9]. According to Kanmegne et al. [48], electrolyte leakage from tissue can be used to indicate the effectiveness of membranes as barriers to solute diffusion. Electrolyte leakage increasing (oven and sun drying) after 30 DAA might be due to the seeds being dried below the critical moisture content; it reached the maximum value at the lethal moisture content level [49]. This is probably due to cell membrane damage during excessive desiccation. The excessive dehydration of seeds beyond the moisture level will increase the concentration of leaked solutes and cause the seeds to lose their viability. Low values of EC indicate that membranes had greater organization and consequently have high seed vigour. This is also confirmed by increasing germination percentage and positive correlation between EC and germination index, as stated in Table 3, which is an indicator of high seed quality. This finding was similar and consistent with the results obtained by Silva et al. [50] for Jatropha curcas and Solanum melangena L. [51] and Vidigal et al. [12] for Capsicum annuum L. seeds. Table 3. Correlation (r) coefficients between germination percentage, germination index and electrical conductivity of lablab cultivar Highworth.

Conclusions
The study showed that lablab bean pods attained physiological maturity at 30 DAA when the seed moisture content was 27%. At this stage, the pods changed colour from pale greenish-yellow to light reddish-brown and resulted in the highest germination percentage (78%). Drying using an oven at 30 • C for 72 h reduced the moisture content to 14%. This also improved germination from 78 to 91%.
In the present study, harvested immature seeds subjected to oven drying had improved germination percentage because of the enhanced ability to tolerate desiccation. Lablab seeds attained maximum dry weight, germination, and vigour at 30 days after an-thesis (DAA). Based on the DAA, the best physical indicator that can be recommended for lablab beans is when the pods change to a light reddish-brown. Upon harvest, the lablab bean should be subjected to drying using either sun or oven drying to reduce moisture content for the next step, which is storage. Lablab beans have up to 91% germination after oven drying and 89% germination after sun drying. However, in terms of germination speed and membrane integrity (based on electrolyte leakage), seeds harvested at 30 DAA and dried using an oven showed better results compared with sun drying.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author. The data are not publicity available.