Study site.

The study was conducted using mature plants of ‘Elliott’ blueberry located in a field with five other cultivars at the Oregon State University Lewis–Brown Horticultural Research Farm in Corvallis, OR (lat. 44°33′ N, long. 123°13′ W). ‘Elliott’ is a late-season cultivar that typically ripens in August and late September in western Oregon. The field was planted in Oct. 2008. Soil at the site was a Malabon silty clay loam (fine, mixed, superactive, mesic Pachic Ultic Argixerolls) with a pH of 5.5. The plants were spaced 0.8 × 3.0 m apart on 0.4-m-high raised beds and were mulched every other year with a 5-cm-deep layer of douglas fir [Pseudotsuga menziesii Mirb. (Franco)] sawdust. Grass alleyways were planted between the beds and mowed as needed [see Vargas et al. (2015) for complete details regarding establishment of the planting].

Plants were irrigated using two lines of drip tubing (Netafim, Fresno, CA) per row and were fertigated weekly (from mid-April through July) with ammonium sulfate (9N–0P–0K) at a total rate of 168 kg·ha⁻¹ N per year. Urea, monoammonium phosphate, and boric acid were also applied in previous years (Vargas et al., 2015). The drip lines had integrated pressure-compensating emitters (2 L·h⁻¹) every 0.45 m and were located ≈0.45 m from the base of the plants on each side of the row. Weeds were controlled, as needed, by hand-weeding on the top of beds and by applying glyphosate herbicide at the base of beds. No insecticides or fungicides were applied to the field during the study. Calcium fertilizer was never applied to the planting. Leaf tissue analysis (15 Aug. 2016) indicated that nutrients, including Ca, were sufficient at the site and within the ranges recommended for northern highbush blueberry (Hart et al., 2006). The leaves were sampled and analyzed following the standard procedures for blueberry (Strik and Vance, 2015).

Berry dry weight and calcium content.

Berry samples were collected at seven stages of fruit development: petal fall, early green, late green, fruit coloring, 10% blue, 25% blue, and 75% blue (Table 1). The percentage of blue was approximated and represented the proportion of berries that turned blue in the entire field on a given day. At each stage, berries were sampled from three individual plant replicates. The numbers of berries sampled at each stage of berry development are listed in Table 1. Samples were collected randomly from upper and lower parts of the canopy on both sides of the row. During the last stage (75% blue), additional samples (15 berries/replicate) were collected and separated into skin, pulp, and seeds to determine the distribution of dry matter and Ca in the berries. The berries and their components were digested using 70% (v/v) nitric acid, as described by Gavlak et al., (2005), and then analyzed for Ca using an inductively coupled plasma (ICP) optical emission spectrometer (model Optima 3000DV; Perkin Elmer, Wellesley, MA). Other nutrients, including potassium (K) and magnesium (Mg), were also measured in the berries during the ICP analysis; nitrogen (N) was measured using a combustion analyzer (model TruSpec CN; Leco Corp., St. Paul, MN).

Table 1.Berry sampling dates of ‘Elliott’ blueberry in Corvallis, OR in 2016 (study 1).

View Fullscreen

Stomatal conductance.

Stomatal conductance of the berries was measured on sunny days using a steady-state porometer (model LI-1600; LI-COR, Lincoln, NE). Ten berries each were randomly measured from stage 1 through stage 6 (petal fall to 25% blue). Leaves were also measured on 10 plants at the late green stage (stage 3). For each measurement, a single berry was selected from a cluster, and the rest of the berries were removed before turning on the porometer. The porometer was equilibrated to site conditions for 30 min before each measurement, and the null point was set near ambient relative humidity. Each berry was clamped to the head of the porometer on the calyx side (i.e., facing the sensors). The aperture of the clamp was adjusted with different size openings on the foam pads (0.35–0.90 cm²), as needed, to accommodate different size berries. A complete seal and good contact between the instrument’s thermocouple and the berry were ensured before starting the measurements. Each measurement required 5 to 10 min to obtain a stable reading.

Stomatal density and distribution.

The number of stomata on the berries was counted using a light microscope (model DM 2500; Leica, Wetzlar, Germany). To prepare the samples, six berries each from stage 1 through stage 6 (Table 1) were randomly collected and coated with clear lacquer (nail polish). Leaf samples were also collected from plants at the late green stage and coated on the abaxial side with the lacquer. After drying, the lacquer was carefully peeled from the berries and leaves and placed on glass slides. For each berry sample, the berry was divided into the calyx and four equal spherical segments: the distal (calyx) end, distal-middle end, proximal-middle end, and proximal (pedicel) end (Fig. 1A). Under the microscope, six photographs with an area of 1.65 × 1.30 mm were randomly taken from each position at 1800× magnification. The stomata were then counted in the photographs and divided by the area to calculate the stomatal density on each segment of the berries. A total of 1080 photographs were processed. The results were analyzed by one-way analysis of variance (ANOVA) using R version 3.4.2 (R Core Team, 2017). Means were separated using Tukey’s honestly significant difference test (α = 0.05).

View Figure

• Download as Powerpoint
• Download Figure

Stomata on the berries were also examined under a scanning electron microscope (SEM) (Quanta 600F; FEI Company, Hillsboro, OR) with an acceleration voltage of 10 kV. Eight berries from the sun-exposed part of the canopy were collected at three stages, including early green, fruit coloring, and 25% blue. To avoid removing the wax during sampling, each berry was held by inserting a dissection needle through the distal (calyx) end. Skin (epidermis) was carefully removed from the equator of each berry using a razor blade. The skin was immediately submerged into a fixation solution (2.5% glutaraldehyde and 1% paraformaldehyde in 0.1 M sodium cacodylate buffer) and stored overnight at 5 °C. Very little fruit pulp was retained on the skin during removal, and most of it separated from the skin after fixation. Approximately 20 h later, skin samples were gradually washed with a series of 30%, 50%, 70%, 90%, and 100% acetone; to maintain cellular structure, they were dehydrated using a CO₂ critical point dryer (EMS 850; Electron Microscopy Sciences, Hatfield, PA). Once dried, the samples were frozen with liquid N and shattered into small specimens that were mounted on aluminum stubs with double-sided carbon tape and coated with a 15-nm-thick layer of 60% gold and 40% palladium (Cressington 108 Auto Sputter Coater; Cressington, Watford, UK). Images of the specimens were captured at a magnification of 7200×.

Stomatal density and distribution.

Stomata were scarce on the berries and concentrated primarily on the distal end near the calyx (Fig. 1B and C). Density of the stomata was greatest at petal fall, averaging 5 to 108 stomata/mm² from the distal end to the proximal end of the berries (Fig. 2A). In comparison, the mean density on the abaxial surface of the leaves was, on average, 496 stomata/mm². Stomata were wide-open at the early green stage of berry development and had only a slight deposit of wax along the guard cells (Fig. 1D).

View Figure

• Download as Powerpoint
• Download Figure

As the berries expanded during the initial period of rapid growth (stage I), most of the stomata remained near the distal end of the berries, and by the late green stage, almost none was found in the middle and proximal segments (Fig. 2A). The majority of these stomata were completely covered with wax when the berries began to change color and ripen (stages II and III; Fig. 1E). By the time the berries were blue, it was nearly impossible to locate the stomata with the SEM.

Stomatal conductance.

Stomatal conductance of the berries averaged 45 mmol·m⁻²·s⁻¹ at petal fall and rapidly declined as the fruit developed (Fig. 2B). By the fruit coloring stage, conductance was low and remained less than 15 mmol·m⁻²·s⁻¹ throughout the ripening period. Leaf g_S measured at the green stage averaged ≈100 mmol·m⁻²·s⁻¹.

Accumulation and partitioning of dry matter and calcium.

With ‘Elliott’, dry matter accumulated in the berries in a typical double-sigmoid pattern, with an initial period of rapid growth (stage I) from petal fall to fruit coloring, followed by a short lag period of growth (stage II) during fruit coloring, and, finally, a second period of rapid growth (stage III) during fruit ripening and before harvest (Fig. 2C). Calcium also accumulated rapidly during the initial stage of berry development; however, in this case, accumulation slowed considerably between the late green and fruit coloring stages, and it stopped completely during fruit ripening (Fig. 2C). In contrast, K and Mg continued to accumulate throughout development of the berries (Fig. 2C, inset).

Calcium accumulation also typically slowed or stopped by the late green stage in ‘Duke’, ‘Bluecrop’, and ‘Aurora’ (Fig. 3), whereas uptake of K, Mg, and other nutrients such as N, phosphorus (P), and sulfur (S) continued throughout berry development in each cultivar (data not shown). Results of a three-way analysis indicated that the dry weight and Ca content of the berries were significantly affected by location, cultivar, and stage of development, as well as by two-way interactions of cultivar × stage and location × stage (P ≤ 0.05). In general, Ca accumulation appeared to end sooner in the early and midseason cultivars (Duke and Bluecrop) than in the late-season cultivar (Aurora; also see Elliott) (Fig. 2C). However, there were two exceptions. In Burlington, WA, ‘Duke’ continued to accumulate Ca until the berries were 25% blue (Fig. 3A); however, ‘Aurora’ appeared to stop accumulating Ca by the late green stage but then resumed uptake once the berries were 25% blue (Fig. 3C).

View Figure

• Download as Powerpoint
• Download Figure

By harvest (25% and 75% blue stage of the field), the Ca content was less than 0.15 mg/berry, on average, in each of the cultivars examined. However, levels were low in two cultivars in Burlington, WA, averaging only 0.12 and 0.10 mg/berry of Ca each in ‘Bluecrop’ and ‘Aurora’, respectively (Fig. 3B and C). ‘Duke’ also had low Ca in the berries at the organic site in Aurora, OR. Despite a large amount of dry matter (0.33 mg/berry), ‘Duke’ berries from the organic site accumulated only 0.07 mg/berry of Ca by the 75% blue stage, which was less than half the amount measured at the same stage in Corvallis and Burlington (Fig. 3A).

When the berries matured, 18% of the total dry matter in ‘Elliott’ berries was in the skin, 75% was in the pulp, and less than 7% was in the seeds (Table 3). Calcium, however, was approximately three-times more concentrated in the skin and seeds than in the pulp; consequently, nearly half of the Ca in the berries was in the skin (35%) and seeds (13%) (Table 3). Proportionally less dry matter and more Ca were also allocated to the skin and seeds of ‘Duke’, ‘Bluecrop’, and ‘Aurora’ berries (Fig. 4). A two-way analysis revealed that allocation of dry matter and Ca to each tissue was affected by cultivar (P ≤ 0.01), but it was unaffected by the location in which the berries were collected or by the interaction of cultivar × location. In these cultivars, partitioning of dry matter and Ca were similar at each site, although ‘Aurora’ allocated significantly more dry matter and Ca to the skin and less to the seeds than either ‘Duke’ or ‘Bluecrop’.

Table 3.Mean (±se) dry weight and concentration and content of Ca in the skin, pulp, and seeds of ‘Elliott’ blueberries (n = 3).^z

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

Stomatal density of the blueberries declined as the fruit matured.

Grusak and Pomper (1999) observed similar results in snap bean (Phaseolus vulgaris L.) pods. As in the present study, the decline in stomatal density was related to the expansive growth of the fruiting structures. It would appear that blueberries begin with a certain number of stomata that become more diffuse as the berries expand. During expansion, most of the stomata remained near the calyx or distal end of the berry. Eventually, these stomata became covered with wax and were probably no longer functional during the later stages of berry development. Therefore, by the time fruit coloring begins, water loss through the surface of the berries would have to occur via minute cracks in the cuticle (Konarska, 2015). Significant reduction in transpiration is also associated with thickening of epicuticular waxes after véraison in wine grape (Blanke and Leyhe, 1987, 1988; Rogiers et al., 2004).

The rate of g_S measured in the berries after petal fall was approximately half of what was measured in actively transpiring leaves on the plants. Fruit stomata are well-known as essential for gas exchange and C assimilation during early fruit development in many crops (Aschan and Pfanz, 2003; Cipollini and Levey, 1991; Palliotti and Cartechini, 2001). However, rates measured at later stages of development were comparable to leaves with closed stomata (Bryla and Strik, 2006). Cuticular water loss apparently predominates in the berries during later stages of development.

Among the cultivars examined in this study, Ca content was below average in ‘Bluecrop’ and ‘Aurora’ blueberries in Burlington, WA. These anomalies may have been because the berries in Washington were collected from different plants at each stage of development, whereas those in Oregon were sampled from the same plants at each stage. However, the Burlington site was the most northern of the sites sampled in the study, and the climate there is cooler and wetter than at the other two sites sampled in Oregon (Yang et al., 2019). Low temperatures and high humidity reduce vapor pressure deficit; therefore, they might have resulted in a reduction in the cumulative influx of Ca in the xylem sap at this site (Zhang and Keller, 2015). Seeds in the fruit are also involved in fruit Ca accumulation (Buccheri and Di Vaio, 2005). Cultivars that regularly produce low numbers of seeds or are exposed to poor environmental conditions during pollination are expected to have lower fruit Ca levels (Arrington and DeVetter, 2016). Weather conditions were poor during pollination in Washington in 2016 (L. DeVetter, personal observation), which could have reduced the number of seeds per berry and the associated accumulation of Ca in the fruit. In contrast, seed number was consistently high in Aurora, OR, due to the favorable climate and a high potential for cross-pollination among the cultivars at the site (Strik et al., 2017b; Strik and Vance, 2019).

The Ca content was also below average in organic ‘Duke’ blueberries. Many organic fertilizers, including those used in this study, tend to be high in K and low in Ca. Yard debris compost, which is also high in K, was also incorporated in the soil and used as a mulch in the planting (Strik et al., 2017b). Excessive K can inhibit Ca uptake in plants, thereby interfering with Ca movement to the leaves and fruit (Marschner, 2012). ‘Duke’, in particular, demonstrates a genetic tendency toward sensitivity to high K and any negative effects it might have on fruit and leaf Ca (Strik and Vance, 2015; Strik et al., 2017a, 2019).

Although g_S is low in developing blueberries, it appears to be an important mechanism by which Ca is delivered to the fruit. Therefore, to increase Ca and enhance the quality of the fruit, cultivars could be developed with more stomata on the berry surface. Less wax would also increase stomatal functioning, particularly at later stages of berry development, but it might lead to problems such as increased infection by fungi and bacteria, more physical damage, and higher rates of water loss after harvest (Jenks and Ashworth, 1999; Riederer and Schreiber, 2001). Alternatively, cultivars could be developed with a higher proportion of Ca in the skin of the berries, which, as mentioned, increases berry firmness and decreases cracking and postharvest decay of fruit in other crops. Cultural practices such as heavier pruning to increase light exposure or running fans to increase air circulation (e.g., for containerized production of blueberries in tunnels and greenhouses) could also be used as a more immediate means to improve xylem influx and movement of Ca into the berries (Mazzeo et al., 2011, 2013; Montanaro et al., 2006; Rebucci et al., 1997). Finally, applying soil or foliar Ca fertilizers during early stages of berry development can increase Ca in the berries (Gerbrandt et al., 2019; Hanson and Berkheimer, 2004), but the efficacy of uptake may be limited by the rate of application, cultivar, and weather conditions (Arrington and DeVetter, 2017; Vance et al., 2017).