Characterisation of seed marking types in chickpea (Cicer arietinum L.): Tiger stripe and other blemishes

Desi chickpea is a significant export crop for Australia; Australia being the largest exporter globally. Visual appearance of the seed is an economically significant measure of seed quality by the Indian subcontinent, the major importer of desi chickpea worldwide. Any visual blemish on the seed is considered undesirable, regardless of the cause (biotic or otherwise). Literature on biotic causes of seed blemishes, such as ascochyta blight, are available; however, little could be found on abiotic blemishes. Abiotic seed blemishes caused by physiological plant responses are more commonly known as seed markings. Despite the presence of seed markings being confirmed by several chickpea‐producing countries during personal discussions (India, Canada), no scientific literature has been published. The aim of this study was to proactively seek out and characterise different types of seed marking patterns using a wide genetic pool of desi chickpea across a range of environments in Australia. Thirteen different seed marking patterns were identified in desi chickpea and three in kabuli chickpea, including several rare seed markings that were discovered, photographed, and described. Seed markings (blemishes thought to be caused by physiological plant stress) can be characterised as dark patterns on the testa (seed coat) that do not visually affect the underlying cotyledon. In contrast, other seed blemishes (caused by pests and disease, physical damage, or poor storage) were more likely to affect the cotyledons underlying the testa, but not always. This paper classifies and describes various types of seed markings and blemishes for future reference by the global chickpea industry.

suffer from price penalties or rejection of harvested crops due to visual seed defects (Pulse Australia, 2005).
Visual seed defects include the presence of foreign material and obvious seed damage like sprouting, broken or chipped seeds, insect damage, shrivelled seeds, and green or sappy seeds. These defects are relatively easy to discern. Visual blemishes appearing as darkened areas of the seed can also be classified as defective, with the tolerance differing depending on the size of the blemish and the cause.
Seed markings are one type of seed blemish, appearing as darkened patterns on the seed coat, not affecting the underlying cotyledons and presumably caused by physiological plant stress. Biotic causes of seed blemishes include ascochyta blight and moulds, and the market has a very low or "nil" tolerance for these types of blemishes. Other causes of poor colour and blemishes, including seed markings like "tiger stripe," range in their tolerance allowable under the Standards (Pulse Australia, 2019). Tiger stripe is defined in the Australian Pulse Standards as "dark-coloured lines of striping on the seed coat or kernel" (Pulse Australia, 2019).
Most seed markings are not commonly observed or noticed in commercial crops. However, on grain delivery, routine evaluation of samples can lead to marked seeds being mistaken for disease.
A potential misclassification of a seed marking, such as tiger stripe, into a nil tolerance disease category would unfairly penalise a chickpea grower.
Although no scientific literature on physiological seed markings in chickpea exists, tiger stripe markings have previously been seen in some breeding accessions of desi chickpea in Australia, Canada, and India (Pers. Comm.). As seed markings detract from the visual appearance of the seed, they are undesirable and accessions displaying obvious seed markings are removed from breeding programmes. The aim of this 5-year study was to proactively seek out and characterise different types of seed marking patterns from a diverse genetic pool of desi chickpea across a wide range of environments in Australia. This characterisation of seed marking types is now being used as a tool to correctly identify and differentiate different seed marking types for breeding and research purposes in Australia.

| Samples
Harvested chickpea seed samples were obtained from Australian grain growers (predominantly from Southern NSW to Central Queensland; 2014-2017 seasons) and research agronomy trials and breeding accessions (spanning all the major growing regions of 2013 Striker, Rupali, Sonali, and Yorker). These trials were subject to a wide range of environmental conditions and management treatments. In total, 9,493 chickpea samples were evaluated over five seasons.
Seed sample quantities received ranged between 500 g and 3 kg, depending on seed availability and source. The entire sample was visually evaluated for the presence/absence of seed markings, through individual seed-by-seed examination. All seeds with markings were recorded, and photographs were taken for any new types of markings discovered.

| Microscopy
Mature seeds of desi chickpea identified to have seed markings were soaked overnight in distilled water to facilitate removal of the testa and improve pliability for mounting. The testa of the soaked seeds were removed from the cotyledons manually after incising with a scalpel. Tweezers were used to peel back the lower parenchyma layer of the testa which severs at the hourglass cells (Wood et al., 2011). Both the top layer (palisade layer) and lower layer (parenchyma layer) of the testa and the underlying cotyledons were examined for visual pigmentation and other visual changes.
For mature blotch/tiger-striped seeds of desi chickpea (cv. PBA Pistol), detached testa (as described above) were cut finely with a razor blade and cross sections directly mounted on microscopic slides under a coverslip. Marked and unmarked testa were studied without the need for artificial staining by phase-contrast microscopy (BHA model, Olympus) to ascertain more detailed information about the location of pigmentation in the testa responsible for the markings.

| Visual characterisation
Previously known and named seed markings were observed during our detailed visual evaluation, including black, mosaic, speckled, and T A B L E 1 Seed marking types observed and classified in desi chickpea (Cicer arietinum L.) material tiger stripe. We also named and described several newly discovered markings (Table 1). These seed markings normally only appear on a low percentage of seeds within any given sample.
Tiger stripe was the most commonly observed seed marking throughout the five seasons. A new marking, discovered in 2013, was named "blotch" (Wood, 2013). Initially, attempts were made to classify seeds as having either tiger stripe or blotch. However, it was technically difficult to achieve this differentiation because both marking types could be found in combination on individual seeds. It is noteworthy that when blotching occurred, there were always some seeds displaying the tiger stripe marking within the same sample. It became apparent that the blotch marking often appeared as the severity of the tiger stripe marking in samples increased (i.e., markings becoming larger, darker, and more obvious). We decided that the two markings were intrinsically linked and impossible to differentiate when individual seeds displayed both marking types, hence the two classifications were reassigned into one "blotch/tiger stripe" marking category from 2014 onwards ( Figure 1).
We discovered additional new markings and named them as slit, owl eyes, critters, black beak, scroll, veined, ornate scroll, and painted. Table 1 shows images of all the types of seed markings and the classification names assigned. In addition, several non-uniformly located and irregularly shaped markings were also observed.

| Microscopic characterisation
All seed markings characterised in Table 1 were found to be confined to the outer layers of the testa (Figure 2), known as the palisade layer.
The inner testa layer, known as the parenchyma layer, was opaque and contained no pigments (Figure 2), and no pigmentation or visual changes were identified in the underlying cotyledon.
Microscopy showed that the markings involve an accumulation of amber-coloured pigment within the vacuoles of the macrosclerid cells below the light line of the testa ( Figure 3). This is the same region where normal pigmentation occurs in the nonmarked portion of the testa and in regular desi chickpeas, as previously characterised by (Cubero, 1987). The amber-coloured deposits appear to be phenolic compounds, such as proanthocyanins.

| Seed markings of kabuli chickpea
Of all the thousands of samples evaluated, kabuli chickpeas showed an almost total absence of seed markings, with only several individual marked seeds found once in only three genetically diverse breeding accessions, representing <1% of the kabuli samples evaluated ( Table 2).
The reason for seed markings being extremely rare in chickpeas of the kabuli type may be due to their white-cream testa colour.
Kabuli chickpeas are readily distinguished from desi plants as having white (nonpigmented) flowers and an absence of anthocyanin pigments in their vegetative tissues. Staining of testas for proanthocyanidin confirmed no proanthocyanidins in the kabuli testa, whereas desi chickpeas were abundant in proanthocyanidins as shown through staining green with DMACA and brown with Vanillin ( Figure 4). In addition, a desi chickpea variety, gully, which is characterised by an absence of anthocyanin pigments in the vegetative tissues but produces a light pink flower and light salmon coloured

| Other blemishes of chickpea seeds
We observed other blemishes on seeds that we believe are not physiological seed markings during this study. These blemishes included bruised seeds, ascochyta blight-infected seeds, and several seeds with blemishes reminiscent of symptoms associated with biotic causes in other species (Table 3). These will be discussed in more detail in the discussion section.

| DISCUSSION
Physiological seed markings are known to occur in desi chickpeas throughout breeding programmes of the major producing countries (Pers. Comm.), including India, Canada, and Australia. Moreover, seed markings are likely occurring in most, if not all, countries as chickpea domestication has led to a narrow genetic base globally (von Wettberg et al., 2018). Most chickpea-producing countries have seed evaluation within their breeding programmes that will evaluate and discard germplasm that is affected.
We identified numerous different patterns of seed markings in this study. All of the seed markings identified in Table 1 were confined to the testa (localised in the vacuoles of the macrosclerid cells below the light line) and had no effect on the underlying cotyledons. These markings are believed to be abiotic or genetic in origin and generally not caused by biotic factors. The majority of the markings were very distinct in their patterning and location on the seed. Genetics control the regular patterning of testa in other pulses, like lentils (Vandenberg & Slinkard, 1990) and common bean (Caldas & Blair, 2009). However seed markings that occur when physiologically triggered by environmental factors have not been reported in the literature previously. The frequency of seed marking incidence and the role of genetics and environment will be explored further for some of the more common seed markings in subsequent publications.
We observed and classified seed marking types on desi chickpea seeds as black, mosaic, speckled, blotch/tiger stripe, slit, owl eyes, critters, black beak, scroll, veined, ornate scroll, painted, and non-uniform irregular markings. Many of these seed markings are very rare and were only observed in one or two samples, and sometimes, only on a single seed. Light speckling is the most common marking observed in commercial desi chickpea crops. The next most commonly observed markings were blotch/tiger stripe and mosaic type markings. Although they are much less common in commercial crops of chickpea, when they do occur, it is normally at a rate of less than 3% of seeds affected. At these levels, the markings are often not visually discernible in a sample during normal screening processes. T A B L E 2 Seed marking types observed and classified in kabuli chickpea (Cicer arietinum L.)

Seed marking name Image Description
Lacework A pattern of light spots and slightly darker swirls extending over the entire testa.

Slash
Testa features a brown slit-like marking, reminiscent of a cut but not a physical wound. Occur on the dorsal side toward the base. Muzquiz . In agreement with our staining, chickpea seeds from desi-types contained approximately 2.3-6.7 mg/g total phenolic compounds, of which 0.1-0.9 mg/g is proanthocyanidin, whereas the kabuli-type contains much less: 0.5-0.6 mg/g and 0.0-0.4 mg/g, total phenolics and proanthocyanidin, respectively (Segev et al., 2010;Wood & Grusak, 2007). Due to the large natural diversity of phenolic chemical structures and the complexity of their biochemical interactions, the scientific understanding of colour chemistry in plant biology is still in its infancy. The specific compounds accumulating in the testa that create the visual appearance of seed markings will require further investigation to elucidate, as will the environmental triggers.
We observed seed bruising during our study. Pod restriction can cause seed bruising when there is limited pod expansion during development, causing seeds to be squashed too tightly against each other in the pod ( Figure 5A). Seed bruising was also caused by hail directly hitting pods during the podding stage ( Figure 5B). Both types of seed bruising led to cotyledon damage ( Figure 5B, insert). Bruising can also be caused through mechanical damage sustained during the season or at harvest, particularly when the seeds contain higher than recommended moisture content.
Although bruised seeds may at first appear similar to the blotch/tiger stripe seed marking, bruising almost always caused some damage (necrosis) to the underlying cotyledon, whereas seed markings were always confined to the testa. Therefore, removal of the seed testa on inspection was the best way to identify damaged cotyledons sustained from bruising events.
Biotic causes also can result in seed blemishes, particularly when infection of plants occurs late in the season just prior to, or during, seed development. Ascochyta blight (Ascochyta rabiei) is a fungus that can cause large yield losses in chickpea, and late season infection can result in seed blemishes. Typical ascochyta blight symptoms of mature pods and seeds of desi chickpea are shown in Figure 6. Infected seeds show dark round or irregular cankers, sometimes with a grey centre and bearing pycnidia, whereas the underlying cotyledons of badly affected seeds will also have dark necrotic marks (Pande et al., 2005).
We observed damaged cotyledons due to infection in this study ( Figure 6). However, we observed also that many infected blemished seeds can have normal underlying cotyledons with no visual blemish or necrosis. Infected seeds may also be dark and shrivelled when infection occurred on pods and the seeds were still small and developing (Nene, 1982;Pande et al., 2005).
A substantial amount of research into ascochyta blight occurred due to its significant contribution to yield losses in Australia and other chickpea-producing countries. Ascochyta blemishes on seeds may not damage the underlying cotyledons, and they can appear somewhat similar to blotch markings. This can lead to confusion over classification of seed quality in relation to receival standards. Pods display typical concentric circles, caused by pycnidia ( Figure 6). Pycnidia circles are the best characterising symptom to look for in seed samples that still contain some pod material. Knowledge of paddock history and crop infection will also assist in differentiating between blemishes caused by disease and seed markings caused by physiological effects; however, this information is not always available with harvested seed samples.
We observed a single seed from only one sample that displayed distinct concentric circles (Table 3; Figure 7A). These types of circles are often associated with disease and although ascochyta blight causes similar circular patterns on pods, it does not cause this appearance on seeds ( Figure 6). No literature or images of similar markings could be found in searches conducted by the authors, except for an image of watermelon mosaic virus ( Figure 7B). It is possible that this pattern could be caused by a viral infection; however, we found only one single seed with this pattern in one sample of the thousands of samples examined.
Similarly, one seed was found in a different sample with a marking that appeared similar to tiger stripe but with the addition of "wiggly lines" around the outer rim of the chickpea sides (Table 3; Figure 8A).
We discovered this wiggly lines seed during temporal sampling and evaluation of seeds in a different experiment, and no other seeds on the same plant displayed this marking. The wiggly lines around the seed perimeter (Table 3; Figure 8A) did resemble the pea seed-borne mosaic virus (PSbMV; Genus Potyvirus, Family Potyviridae) symptoms observed on faba bean (Vicia faba L.) seeds shown in Figure 8B (GRDC, 2017). In field pea (Pisum sativum L.), PSbMV symptoms include necrotic rings and line patterns, most obvious on whiteseeded pea varieties, often referred to as "tennis ball" patterns that F I G U R E 4 Testa (seed coats) of mature desi (left) and kabuli (right) stained with DMACA/Methanol/6M HCl stain (top row) and vanillin stain (bottom row). The desi testa were heavily stained green (top left) and brown (bottom left) by both methods, whereas the kabuli testa did not stain green (top right) or brown (bottom right), indicating that desi testa have an abundance of proanthocyanidin whilst the kabuli testa had no proanthocyanidin T A B L E 3 Other blemishes observed and classified in desi chickpea (Cicer arietinum L.)

Seed marking name Image Description
Bruising Bruised seeds. Darkened blemishes of irregular to round shapes. Inset: Cotyledon damage (necrosis) beneath the testa blemish.

Ascochyta
Infected seeds show dark round or irregular cankers, sometimes with a grey centre. Infected seeds can also be dark and shrivelled. Inset: Badly infected seeds exhibit dark necrotic marks on the cotyledon underlying the testa blemish.
Wiggly lines Tiger stripe like marking plus wiggly lines around the outer rim of the chickpea lateral sides particularly near the ventral tiger stripe (only one seed observed).

Concentric circles
Distinct concentric circles all over the seed surface (only one seed observed).

(a) (b)
F I G U R E 5 Seed marks on desi chickpea (Cicer arietinum L.) due to bruising caused by (A) pod restriction and (B) hail (cotyledon damage from bruising; inset) can be blistered and also can cause reduced seed size, splitting, and malformation ( Figure 8C; Coutts, 2018).
There is little literature describing the effect of PSbMV on the seed of chickpea. Latham and Jones (2001) stated that necrotic ring and line pattern symptoms were observed in seed coats from a range of legume species, including chickpea; however, only images of field pea, faba bean, and Lathyrus ochrus seed symptoms were shown. They also tested six varieties of chickpea for PSbMV and classified all as resistant and seed transmission as low. Coutts, Prince, and Jones (2008) infected chickpeas with PSbMV and showed images of infected desi and kabuli chickpea seeds at maturity being shrivelled and darker in appearance ( Figure 9); however, there were no obvious characteristic wiggly lines on those seeds to match this single seed ( Figure 8A). Additionally, the tennis ball patterns of PSbMV-infected field pea ( Figure 8C; Coutts, 2018) also somewhat resemble the ornate scroll markings (Table 1). We are unable to conclude why the green seed had a wiggly lines appearance. We also cannot conclude whether the wiggly lines or ornate scroll markings are symptoms of biotic infection, but due to the low transmission rate and resistant classification of Australian chickpeas to PSbMV, we believe that PSbMV is unlikely to be the cause. Further research in this area will be required to elucidate the answer.
Some images of kabuli seeds from phytoplasma-infected plants were uncovered during the literature search for chickpea seed blemishes caused by diseases (Brier, 2018). The patterns on these seeds were visually similar in appearance to the "veined" seed marking characterised on desi chickpea seeds in Table 1. The images are reproduced in Figure 10 for direct comparison. The biggest difference between these two images is the type of seed affected, desi compared with kabuli.
Phytoplasma is a specialised bacterial species that infects plants and, in northern Australia, a brown leafhopper (Orosius orientalis) is believed to be the likely insect vector (Brier, 2018). Phyllody is a recognised plant symptom of phytoplasma infection, whereby abnormal clumps of tiny leaves or secondary flowering structures grow out of the original flowers (Brier, 2018 Figure 10B reproduced from Brier (2018). Whether all phytoplasma plants infected late in the season produce these veined type of seed blemishes will need to be confirmed; similarly, it will be important to determine whether veined seed markings can be produced by plants that are not infected by phytoplasma. Then, this would provide evidence whether the veined appearance of desi chickpea was caused by phytoplasma, abiotic stress, or both. Furthermore, since the "scroll" and ornate scroll seed markings occur with a similar patterning to the veined marking (Table 1), there is a possibility that these markings also may be caused by phytoplasma.
It is noteworthy that the phytoplasma-infected kabuli seed blemished very heavily despite the testa lacking proanthocyanidins.
Relatively, little is known about the polyphenol mechanism(s) responsible for changes in testa colour and darkening. However, oxidation of proanthocyanidins and resultant formation of complexes with other phenolic compounds were implicated in seed darkening of other species including bean (Phaseolus vulgaris L.), grape (Vitis vinifera L.), sorghum, and Arabidopsis, (Beninger et al., 2005;Beninger & Hosfield, 1999;Dean et al., 2011;Jorgensen, Marin, & Kennedy, 2004). Further research is necessary to elucidate the biochemical pathway that can cause nonpigmented seeds to display these dark blemishes after phytoplasma infection. Such research may assist in understanding potential phenolic biochemical pathways and mechanisms of darkening of other seeds and produce.

| CONCLUSIONS
Thirteen different seed marking patterns of desi chickpea, and three of kabuli chickpea, that are believed to be physiological seed markings were identified and characterised. To achieve this, 9,493 samples collected over 5 years spanning a diverse range of genetics and environments in Australia were examined. Each seed marking type was characterised and photographed, and the new types were assigned names. Similarities of several newly identified seed markings were compared with known disease blemishes and discussed. Sometimes, it can be difficult to differentiate the cause of these seed blemishes, especially when visual blemishes appear similar to disease blemishes. This publication will serve as a reference document for chickpea breeders, researchers, agronomists, and marketers in identifying seed markings and other blemishes of chickpea seeds. Further research is required to understand the genetic, environmental, and biotic factors influencing the expression of seed markings, as well as research on the underlying biosynthetic pathways and regulation within chickpea plants.