Scientia Horticulturae 195 (2015) 216225

Contents lists available at ScienceDirect

Scientia Horticulturae
journal homepage: www.elsevier.com/locate/scihorti

Raspberry fruit quality changes during ripening and storage as

assessed by colour, sensory evaluation and chemical analyses
Jon Anders Stavang a , Sabine Freitag b , Alexandre Foito b , Susan Verrall b , Ola M. Heide c ,
Derek Stewart b,d , Anita Snsteby d,
a

Department of Biology, University of Bergen, NO-5008 Bergen, Norway

Environmental and Biochemical Sciences, The James Hutton Institute, Invergowrie, Dundee DD2 5DA, SCT, UK
c
Department of Ecology and Natural Resource Management, Norwegian University of Life Sciences, NO-1432 s, Norway
d
NIBIO, Norwegian Institute for Bioeconomy Research, NO-1431 s, Norway
b

a r t i c l e

i n f o

Article history:
Received 20 February 2015
Received in revised form 26 August 2015
Accepted 28 August 2015
Keywords:
Raspberry
Color
Sensory evaluation
Chemical composition
Ripening
Storage

a b s t r a c t
In order to identify the optimal harvest time and monitor changes in raspberry (Rubus idaeus L. cv. Glen
Ample) fruit quality during ripening and storage, quality was assessed and compared by physical, chemical and sensory fruit quality criteria. Visual classication of fruit colour according to the Natural Colour
System (NCS) chart and by physical measurement of fruit adherence to the receptacle or fruit compression resistance yielded parallel and highly signicant results. The light red colour stage corresponding
to NCS S code 3060-Y90R was identied as the optimal harvest stage for commercial fresh marketing of
the Glen Ample cultivar. Fruit harvested at this stage developed the same chemical and sensory qualities as in situ matured fruits and maintained high sensory quality after 8 days of storage in the dark at
23 C. As the fruits mature, the concentration of titratable acids decreases, whereas the concentrations
of anthocyanins and the sugar:acid ratio increase in parallel with colour development. While correlation
analysis revealed a correlation between sensory traits like sweetness and acidity with sucrose and the
sugar:acid ratio, respectively, the overall fruit tastefulness was not strongly correlated with any specic
phytochemical component, thus illustrating the complex nature of this sensory trait. Due to its ease of
performance, picking raspberry fruits related to a standardised colour chart is recommended for picking
raspberry fruits with optimal quality.
2015 Elsevier B.V. All rights reserved.

1. Introduction
Red raspberry (Rubus idaeus L.) is grown throughout the temperate regions of the world as a commercially important berry crop.
The fruit is highly valued for its avour and high content of potentially important health-benecial constituents (Mullen et al., 2002;
Liu et al., 2002; Anttonen and Karjalainen, 2005; Rao and Snyder,
2010). It is a perishable commodity, and although new cultivars
with rmer fruit have been released (Finn et al., 2008) the shelf life
of raspberry is generally short. Identication of the optimal maturity stage for harvesting and correct post-harvest handling and
storage of the fruit, are therefore, essential for successful marketing
of fresh consumption raspberries.
Fruit colour and adhesion to the receptacle plug are the main
criteria used by the producer for practical assessment of the right
maturity stage for harvesting, while colour is also the main criterion

Corresponding author at:

E-mail address: anita.sonsteby@nibio.no (A. Snsteby).
http://dx.doi.org/10.1016/j.scienta.2015.08.045
0304-4238/ 2015 Elsevier B.V. All rights reserved.

used by the consumer to judge fruit quality. The main contributors to the red colour of the raspberry fruit are the anthocyanins.
Only cyanidin-and pelargonidin-type anthocyanins are present in
red raspberry, the former type predominating (Wang et al., 2009;
Remberg et al., 2010; Mazur et al., 2014). Their synthesis is inuenced by a number of environmental factors in both green leaves
and in fruits. According to Grisebach (1982), light is the most
important factor inuencing anthocyanin biosynthesis in plants
in general, and in vegetative tissues, anthocyanin biosynthesis is
induced by UV light as an important photo-protective mechanism
(Steyn et al., 2002). In Glen Ample raspberry fruit, the concentration of total monomeric anthocyanins (TMA) was not signicantly
affected by post-owering growth temperature in the 1224 C
range or by photoperiod at 18 C (Remberg et al., 2010; Mazur et al.,
2014). It is well known and documented however, that postharvest
colour changes and anthocyanin synthesis take place in immaturely
harvested fruit of red raspberry (Wang et al., 2009) as well as in
other berry species (Kalt et al., 1993; Sachs and Shaw, 1993), thus
demonstrating de novo synthesis in detached fruits. Synthesis can
occur in darkness, but the rate is slightly enhanced in light (Austin

J.A. Stavang et al. / Scientia Horticulturae 195 (2015) 216225

et al., 1960; Wang et al., 2009). While raspberry fruit anthocyanin

concentration increases throughout ripening, the total phenolic
concentration was found to decrease from the green to the lightred (pink) stage, and thereafter to increase again until maturity
(Wang and Liu, 2000). However, because fruit weight increases signicantly during ripening, mainly due to increased water content,
a signicant dilution of all soluble fruit constituents is also taking
place during fruit ripening. This is confounding much of the real
concentration changes in fruit constituents that are taking place.
As pointed out by Remberg et al. (2010), this bias is corrected for
when data are expressed on dry weight basis.
While a number of investigations have examined the effects of
fresh and frozen storage on raspberry fruit chemical composition
(e.g. Kalt et al., 1999; Mullen et al., 2002 and references therein),
experimental attempts to dene and describe maturity stages as
guidelines for selection of the best harvest time for raspberries
are rare. Some studies in raspberries (Krger et al., 2003; Krger
et al., 2011) have been restricted to the analysis of three stages of
development: semi-ripe, ripe and over-ripe. Krger et al. (2003)
observed in four cultivars that the ripening stage had a large effect
on rmness, titratable acidity (TA) and fruit colour, whereas storage
conditions during three days had effects on the content of soluble
solids (SS) and suitability for shipping. A subsequent chemical analysis of red raspberry (cv. Tulameen) at the three different stages of
development, revealed that acidity decreased signicantly, while
total anthocyanins increased signicantly with maturation stage
with cyanidin-3-sophoroside, cyanidin 3-rutinoside and cyanidin
3-O-glucoside levels increasing with maturation but not cyanidin
3-glucosylrutinoside (Krger et al., 2011). Further sensory analysis
revealed that ripening had an effect on odour and taste whereas
the effects of storage were small in comparison (Krger et al.,
2003). Hence, it was suggested that semi-ripe fruits could have
improved suitability for shipping and some sensory assessments
(Krger et al., 2003).
In order to determine how early fruits can be harvested and still
develop acceptable quality, Wang et al. (2009) harvested fruit of
Caroline raspberries at ve maturity stages, arbitrarily classied as
5%, 20%, 50%, 80%, and 100% maturity, and studied their fruit quality
chemical changes during storage in light and darkness at 24/16 C
(day/night) temperature. The authors found that fruit harvested at
5% or 20% maturity never developed the levels of SS, TA and sugars found in ripe berries at harvest, while those harvested at 50%
or 80% maturity attained qualities comparable to in situ matured
berries. Storage in light enhanced sugar and reduced acid content
somewhat of 5% and 20% mature berries, but had negligible effect
on fruit of more advanced maturity The authors conclude that raspberry fruit could be harvested as early as 50% maturity, when the
fruits are rmer and less susceptible to mechanical injury, and still
develop a quality comparable to fully mature fruit. However, no
sensory assessment of fruit quality was provided.
In an attempt to describe more precisely the fruit quality
changes taking place during raspberry fruit maturation and post
harvest storage, and in particular, to dene the right maturity stage
for harvesting, we have assessed and compared physical, chemical
and sensory fruit quality criteria that can be used for this purpose.
It is hoped that this knowledge may benet the fruit industry and
the producers as well as the consumers.

2. Material and methods

2.1. Plant material and cultivation
Long canes of the raspberry cultivar Glen Ample, produced
as described by Snsteby et al. (2009), were cropped in a greenhouse at a commercial growers nursery located on the west coast

217

Fig. 1. The numbered colours selected for classication of the maturity stage at
harvest in Glen Ample raspberry.

of Norway during the spring 2011 (Frekhaug, 60 31 N; 5 14 E).

The plants were tipped at a height of 180 cm and grown in 3.5 L
pots in rows with 5 plants per running m. The distance between
the rows was 2.20 m. The plants were fertilized with a complete
nutrient solution containing a 2:3 mixture of CalcinitTM (15.5% N,
19% Ca) and SuperbaTM Red (7422% NPK + micronutrients) (Yara
International, Oslo, Norway). The media electric conductivity (EC)
was 1.31.6 mS cm1 . The production started in mid February 2011
and the rst berries were harvested the rst week of May 2011. The
berries used in this experiment were harvested on 17 May 2011.
2.2. Physical characterization of maturing raspberries
In order to pick berries with different degrees of maturity,
ve berry colours along an assumed maturity gradient were identied and used as a reference during the picking. The colours
ranged from orange/red (colour one), light red (colour two), red
(colour three), dark red (colour four) and dark red/lilac (colour
ve). These colours were visually classied according to the natural colour system (NCS; http://www.ncscolour.com/en/naturalcolour-system/; see Fig. 1). Raspberries of the respective colour
classes were handpicked, 23 berries from each plant, from two
rows in the greenhouse, and placed in black, capped plastic containers (200 g). When full, the container with berries was transported to
a cold room (23 C) within 30 min. Four containers were collected
for each colour class and used in the further analysis.
For further characterization of berries with different colours, we
also measured the force needed to remove the berry from the receptacle in relation to colour. To measure the pull force, we rst placed
a piece of 15 cm duct tape around an individual berry so that the
ends met surrounding the edge of the berry and the tape sticking
to the side walls of the fruit. The loop of the tape was open to allow
connection to the hook of a Pesola precision spring scale (Pesola
AG, Baar, Switzerland; Medio Line 300 g scale for colour class 13
and Light Line 100 g scale for colour class 4 and 5). With the duct
tape connected to the berry, it was slowly pulled off the receptacle. Maximum force monitored during the pull was recorded and
this was replicated 12 times for each colour class. We also recorded
berry weight in relation to colour by weighing three replicates of
20 berries each on a technical balance (0.1 g).
In order to measure fruit rmness at harvest, we placed 50
raspberries from each maturation stage into a 1-litre cylinder and
applied a load of 500 g with an adaptor tting exactly the cylinder diameter on top of the berries. The volume of the berries was
recorded before the load was added and after 6 min. From the volume change, %-compression was calculated. These measurements
were done within 30 min after harvest. After 8 d of storage this
compression method was not applicable as the berries became too
soft and the compression method resulted in a fruit pulp.

218

J.A. Stavang et al. / Scientia Horticulturae 195 (2015) 216225

At the end of the day, the cooled berries were transported by

car in electric portable coolers to a cold storage room at the Bioforsk Ullensvang research station. Here, berries were stored for 1
d or 8 d in the dark at 23 C until used for sensory evaluation, or
frozen at 40 C for subsequent chemical analysis. To prepare the
berries for targeted chemical analyses, a pooled sample of 55195 g
of berries from each colour class was freeze dried, ground up and
homogenized in a mortar, and shipped in vacuum sealed bags to
James Hutton Institute, Scotland.
2.3. Sensory evaluation
At day 1 and 8, berries were put into electric portable coolers
and transported by car to the Oslo head ofce of BAMA, Norways
largest private distributor of fruit and vegetables. There, staff of the
administration made up a sensory team with 9 members. The sensory panel followed a standard protocol routinely used in BAMAs
sensory evaluation of fruits. First, the berries were allowed to reach
room temperature. Then the panel gathered in a meeting room
and the judges of the panel individually evaluated taste of acidity,
sweetness, freshness, and bitterness, overall tastefulness and visual
attractiveness of the berries, and gave the berries scores from 1 to
9 for each character. Each judge wrote their evaluation on a paper
that was collected for data analysis after the end of the evaluation.
A score of 9 for instance for acidity, implied a very acid berry, and
a score of 1 would imply that the berries were not acid at all.
2.4. Phytochemical content
2.4.1. Soluble solids and titratable acidity
At Bioforsk Ullensvang, raspberry juice prepared from frozen
raspberries from 2 replicates of 10 homogenized and ltered raspberries of each colour class, was measured for% SS by a table
refractometer (Atago DBX-50; Atago, Tokyo, Japan). An aliquot
(5 mL) of this raspberry juice was further diluted with distilled H2 O
1:6 and assessed for TA with a Titramaster 85 (Radiometer analytical, Villeurbanne Cedex, France). Results for TA are expressed as%
citric acid equivalents.
2.4.2. Individual sugar, organic acid and polyphenol
quantication
All chemicals used for analysis were of HPLC grade: Acetonitrile,
methanol (HIPerSOLV, Chromanorm, VWR, UK), acetic acid (Chromanorm, Prolabo, UK), formic acid (Sigma Aldrich, Dorset UK) and
ultrapure water (Elga, UK). Polyphenolic standards for compounds
found in raspberry extracts were purchased from Extrasynthese
Ltd. (Genay, France) including cyanidin-3-O-glucoside chloride,
cyandidin-3-O-rutinoside chloride, cyanidin-3-O-sambubioside
chloride, cyanidin-3-O-sophoroside chloride, cyanin chloride,
quercetin-3-O-glucuronide and hyperoside (quercetin-3-Ogalactoside). Morin, sugars (glucose, fructose and sucrose) and
organic acid standards (quinic acid, oxalate, succinic acid and citric
acid) were all purchased from SigmaAldrich (Dorset, UK). All
chemical analysis data were expressed on dry weight basis.
2.4.2.1. Extraction and quantication of individual sugars and
organic acids. Freeze-dried material was weighed (60 2 mg) into
a 2 mL Eppendorf vial and extracted at room temperature for
30 min as described by Mazur et al. (2014). In summary, samples were extracted in 1.5 mL of extraction solvent (50:49:1
methanol:dionised water:formic acid) at room temperature for
30 min. After centrifugation 1 mL of the supernatant was heated
at 80 C for 10 min. After a centrifugation step, 500 L of the supernatant was evaporated and resuspended in 1 mL of deionised water.
Organic acids (citrate and malate) were quantied following extract

dilution (1:20) by anion exchange HPLC on a Dionex IonPac AS11HC 4 250 mm column [Dionex (UK) Ltd. Camberley, UK] tted
with a 4 50 mm guard column utilising gradient of NaOH in 10%
methanol (see supplementary Table 1) as described in Mazur et al.
(2014). Sugars were quantied following extract dilution (1:500)
by anion exchange chromatography on a Dionex Carbopac PA-100
250 4 mm column [Dionex (UK) Ltd. Camberley, UK] utilising an
isocratic elution with 200 mM NaOH prepared in degassed water
at a ow rate of 1 mL min1 for 15 min, as described by Mazur et al.
(2014).
2.4.2.2. Extraction and quantication of individual polyphenols. For
extraction and quantication of individual polyphenols 100 2 mg
of freeze dried berry powder was weighted into 15 mL amber glass
vials sealed with a screw cap containing a PTFE liner (SULPELCO,
SigmaAldrich, UK) and extracted with 3 mL of a water, acetonitrile and acetic acid mixture in the ratio of 60:40:1 at 20 C for
1-h in a rotary shaker as described by Mazur et al. (2014). The
supernatant was diluted in 1:10 ratio utilising extraction solved and
subsequently transferred into lter vials and sealed with a 45 m
PTFE line screwcap (Thomson Instrument Company, London, UK).
The chemical analysis of the berry extracts was performed on
a high performance liquid chromatography (HPLC) system consisting of a quaternary pump (Agilent 1260), a DAD (Agilent 1260), a
column temperature control device (Agilent 1260) and an autosampler Thermostat (Agilent 1290) coupled to a Triple Quadrupole
Mass Spectrometer (Agilent Technologies, Santa Clara, CA, USA)
with the instrument settings described in Mazur et al. (2014).
The chromatography was performed on a Phenomenex C18(2)
2 150 mm (4 m) column tted with a C18 4 2 mm Security
GuardTM cartridge (Phenomenex, Torrance, CA, USA) at ow rate
of 0.3 mL min1 utilising a gradient consisting of 3 phases (see
supplemental table X). Individual polyphenols were quantied by
reference to an external calibration curve generated using purchased standards.
2.5. Data analysis
Using GenStat for Windows, 16th Edition [16.2.11713 (64-bit
edition) VSN International Ltd., Hemel Hempstead, UK], an analysis of variance with linear model was applied to the data set which
tested for a linear relationship to colour. For the sensory, sugar,
organic acid and polyphenol data a two-factor model was used
to test for signicant differences between the storage treatment,
colour class and the interaction. For adhesion, weight and compression data a single factor model was applied to test for signicant
difference in the colour variable.
For the sensory analysis the judges were treated as a random
effect, and all the data for colour 5 was excluded in the statistical
analysis of the sensory evaluations of acidity, sweetness, bitterness,
freshness and overall tastefulness as the data for this colour was
missing for berries stored for 8 d.
Pearsons correlation analysis and regression analysis were
statistical methods of choice for the analysis of phytochemical
properties. Results for regression analysis gave the following information: (a) whether the slope for regression on the colour was
signicantly different to zero when combining the two treatments
(e.g. Colour at harvest and Storage) (b) whether the intercept of 1
or 8 d storage for individual compounds was signicantly different, while having the same slope and (c) whether slopes for short
and long term storage for individual compounds were signicantly
different from each other. Furthermore, correlation analysis was
applied in order to investigate the internal relationship between
organic acids, sugars and avonoids. Finally, correlation analysis
was applied in order to analyse the relationship between sensory
scores and fruit chemical content.

J.A. Stavang et al. / Scientia Horticulturae 195 (2015) 216225

219

Table 1
Physical traits of Glen Ample raspberry fruits harvested at ve dened colour stages. Results from the ANOVA with linear model are presented, rstly the ANOVA test, then
the test for linearity given as the signicance of the slope in relation to colour. Different letters indicate a signicant difference.
Colour at
harvest

Pull force
(g) SD
(n = 12)

Berry weight
(g FW) SD
(n = 3, 10 in each)

Compression (%) SD
(3) (n = 3)

1
2
3
4
5

183.4 35a
138.5 44b
94.7 35c
53.7 16d
33.2 15d

4.9 0.30a
5.3 0.06ab
5.6 0.03b
6.5 0.30c
6.7 0.50c

30.74a*
28.30 4.3a
32.70 1.7ab
34.69 1.1ab
38.84 4.3b

Source of variation (ANOVA)

P - Value colour
s.e.d. colour
Linear t to colourSignicance of slope
Slope
s.e. slope

<0.001
13.35
<0.001
38.5
2.990

<0.001
0.238
<0.001
0.488
0.053

0.046
2.749
0.008
2.26
0.615

Only one measurement.

3. Results

3.2. Sensory evaluation

3.1. Physical characterization of maturing raspberries

Acidity as judged by the sensory panel was found to be dependent on both colours at harvest and length of storage (P = 0.016,
standard error of difference (s.e.d) 0.539 and P = 0.008 s.e.d. 0.381,
respectively), with no signicant interaction term. Thus, lightcoloured berries were judged to be more acid than darker coloured
berries, and this was also the case after 8 d of storage (Table 2).
However, the longer storage period resulted in higher scores for
acidity from the judges for each colour class.
Likewise, the taste of sweetness was also dependent on both
storage (P = 0.007, s.e.d. 0.359) and colour at harvest (P = 0.028, s.e.d.
0.508). Berries stored for 1 d scored on average higher for sweetness
than berries stored for 8 d, with colour four scoring the highest
(Table 2). There was no signicant interaction between colour at
harvest and storage regarding scores for sweetness.
The sensory panel gave generally low scores for bitterness
(Table 2). There was a trend that bitterness was reduced after 8
d of storage (the average was reduced from 2.2 to 1.7), but this
trend was not signicant (P = 0.151).
Overall tastefulness scores were dependent on the storage treatment (P < 0.001) and interaction between colour and storage time
(P = 0.039). It was evident that the sensory panel preferred colour

The ve harvest colours selected for the experiment represent

a unidirectional scale of maturation of Glen Ample raspberries. These colours were found to match the following NCS S
codes: colour one = 3060-Y80R; colour two = 3060-Y90R; colour
three = 3060-R; colour four = 4050R; and the darkest colour
ve = 4050 R-10B (Fig. 1).
The mean force needed to pull the berries off their receptacle for
berries with colour one was 183 g, and for colour two, three, four
and ve; 139 g, 95 g, 54 g and 33 g, respectively (Table 1). When
picking the berries by hand, one could also feel a difference in
fruit rmness for the different fruit colours. The cylinder compression test conrmed that rmness decreased with coloration and
maturity stage (Table 1). Finally, fresh weight of the berries was
recorded. Berries with colour one had an average weight of 4.9 g,
the weight increasing on average by 0.49 g per colour stages, with
the berries with colour ve weighing on average 6.8 g (Table 1).
There were signicant differences among the ve maturity stages
for all 3 parameters, and the vast majority of this was described by
a simple linear regression (Table 1).

Table 2
Means of sensory traits SE after one or eight days of storage of Glen Ample raspberry fruits harvested at ve dened colour stages. In addition, a Sweetness/Acidity ratio
where calculated for each judge based on the scores that were given in the sensory evaluation. ANOVA results are presented with P - value and s.e.d. of each model factor.
Storage

Colour at harvest

Overall tasteSE

AciditySE

SweetnessSE

BitternessSE

FreshnessSE

AttractivenessSE

Sweetness/AciditySE

1
d

1
2
3
4
5
Mean

4.33 0.39
4.67 0.55
5.44 0.80
6.00 0.58
4.00 0.67
5.11

6.11 0.77
5.33 0.67
4.89 0.73
5.22 0.66
3.56 0.53
5.39

4.11 0.63
3.89 0.54
5.22 0.70
6.22 0.52
5.11 0.56
4.86

1.78 0.55
2.22 0.55
2.22 0.54
2.44 0.60
2.00 0.47
2.17

5.22 0.40
5.22 0.40
5.33 0.57
5.67 0.55
3.67 0.37
5.36

5.67 0.76
6.78 0.47
6.89 0.33
5.67 0.50
4.33 0.47
5.87

0.73 0.12
0.81 0.14
1.17 0.20
1.42 0.33
1.57 0.21
1.14

8
d

1
2
3
4
5
Mean

3.56 0.50
4.56 0.44
3.44 0.65
3.18 0.41
n.a.
3.68

7.78 0.28
6.44 0.58
6.22 0.62
5.29 0.53
n.a.
6.43

3.44 0.50
4.11 0.45
3.56 0.47
4.28 0.62
n.a.
3.85

1.78 0.43
1.56 0.34
1.67 0.44
1.83 0.49
n.a.
1.71

6.33 0.55
6.22 0.36
5.44 0.58
4.74 0.53
n.a.
5.69

6.89 0.54
7.00 0.53
5.44 0.60
4.13 0.61
1.00 0.00
4.89

0.46 0.08
0.72 0.13
0.67 0.15
0.96 0.26
n.a.
0.70

<0.001
0.351
0.508
0.497
0.039
0.702

0.008
0.381
0.016
0.539
0.486
0.762

0.007
0.359
0.028
0.508
0.142
0.719

0.151
0.315
0.878
0.446
0.867
0.631

0.306
0.314
0.522
0.444
0.091
0.627

0.003
0.318
<0.001
0.503
<0.001
0.711

Source of variation (ANOVA)

P - Value, Storage (S)
s.e.d. (S)
P - Value colour (C)
s.e.d. (C)
P - Value S C
s.e.d. S C

Data in italic were omitted in the statistical analysis due to lack of data for colour 5 after 8 d of storage.
n.a. - not applicable.

220

J.A. Stavang et al. / Scientia Horticulturae 195 (2015) 216225

Table 3
Means SD of SS, TA and SS/TA ratio after one or eight days of storage of Glen
Ample raspberry fruits harvested at ve dened colour stages. ANOVA results are
presented with P - value and s.e.d. of each model factor.
Storage

Colour at harvest

SS (%) SD

TA (%) SD

SS/TA SD

1
d

1
2
3
4
5
Mean
1
2
3
4
5
Mean

9.25 0.1
9.95 0.1
9.38 0.2
10.00 0.0
9.70 0.0
9.66
9.60 0.0
9.45 0.1
10.10 0.0
9.80 0.0
9.00 0.0
9.59

2.8 0.06
2.7 0.04
2.4 0.05
2.1 0.06
1.9 0.01
2.4
2.7 0.02
2.4 0.01
2.1 0.04
1.8 0.01
1.6 0.05
2.1

3.3 0.05
3.7 0.09
4.0 0.16
4.7 0.14
5.2 0.02
4.2
3.5 0.03
3.9 0.02
4.8 0.10
5.4 0.02
5.5 0.17
4.6

0.058
0.030
<0.001
0.048
<0.001
0.068

<0.001
0.019
<0.001
0.029
0.071
0.041

<0.001
0.043
<0.001
0.068
<0.001
0.096

8
d

Source of variation (ANOVA)

P - Value, Storage (S)
s.e.d. (S)
P - Value colour (C)
s.e.d. (C)
P - Value S C
s.e.d. S C

term colour at harvest x storage (P < 0.001) were highly significant factors inuencing the rating of the berries (Table 2). After
1 d of storage, berries of colour two and three scored the highest,
while light coloured berries (colour one), and darker berries (colour
four and ve), scored the least (Table 2). After 8 d of storage, however, there was a shift in the preferences towards berries that were
harvested with lighter colours, i.e. colour one and two scored the
highest. Dark-coloured berries were judged as less attractive than
bright-coloured berries, regardless of storage time. In this regard, it
is important to note that the berries increased their pigmentation
during storage, and that the berries with the lightest colour at harvest darkened the most. After 8 d of storage, berries with colour one
and two at harvest could be classied to colour three, and berries
with colour three, four and ve at harvest could be classied to
colour four and ve, respectively.

3.3. Phytochemical content

3.3.1. General analyses of SS and TA
The levels of SS were stable over the 8 d storage period which
is reected by no signicant effect of storage treatment in the total
sugar levels. However, there was a signicant statistical effect of
colour at harvest (P < 0.001) and of interaction between storage and
colour at harvest (P < 0.001) for the values of SS. While the SS for
colour one and three increased with storage, SS for colour two, four,
and ve was reduced. However, all changes in SS as measured with
the refractometer were less than 7% and, although signicant, these
changes were rather minor (Table 3).

four after 1 d of storage, while they preferred colour two for berries
stored for 8 d. Furthermore, the maximum score for 8 d of storage
was signicantly lower than the maximum score for 1 d storage
(Table 2).
In the judgement of visual attractiveness both the storage treatment (P = 0.003), colour at harvest (P < 0.001) and the interaction

Table 4
Concentrations of phytochemicals in raspberry Glen Ample harvested at ve different colour stages as determined in samples stored for 1 d and 8 days.
Colour stage
Phytochemicals
Organic acids (g mg1 dry weight)
Malate (1 d)
Malate (8 d)
Citrate (1 d)
Citrate (8 d)
Oxalate (1 d)
Oxalate (8 d)

16.5
4.7
253.6
186.8
0.5
0.6

9.9
6.3
206.3
243.3
0.5
0.6

4.6
3.0
175.5
158.5
0.5
0.5

3.1
2.0
193.0
146.0
0.5
0.5

0.4
1.0
118.3
165.0
0.5
0.4

Sugars (g mg1 dry weight)

Fructose (1 d)
Fructose (8 d)
Glucose (1 d)
Glucose (8 d)
Sucrose (1 d)
Sucrose (8 d)

137.3
125.1
112.2
103.9
100.8
72.1

117.2
182.0
99.5
144.9
122.7
108.7

126.0
125.0
111.9
101.1
111.3
91.9

123.0
173.3
114.4
145.7
133.9
129.1

126.9
191.4
120.4
175.7
141.3
86.2

Anthocyanins (ng mg1 dry weight)

Cyanidin-3-O-sambubioside (1 d)
Cyanidin-3-O-sambubioside (8 d)
Cyanidin-3-(2-xylosyl) rutinoside (1 d)
Cyanidin-3-(2-xylosyl) rutinoside (8 d)
Cyanidin-3-O-sophoroside (1 d)
Cyanidin-3-O-sophoroside (8 d)
Pelargonidin 3-sophoroside (1 d)
Pelargonidin 3-sophoroside (8 d)
Pelargonidin 3-(glucosyl) rutinoside (1 d)
Pelargonidin 3-(glucosyl) rutinoside (8 d)
Cyanidin-3-O-rutinoside (1 d)
Cyanidin-3-O-rutinoside (8 d)
Cyanidin-3-glucoside (1 d)
Cyanidin-3-glucoside (8 d)
Cyanin (1 d)
Cyanin (8 d)

18.2
43.3
12.9
19.1
1170.5
2543.6
6.3
30.9
4.9
852.9
265.0
322.1
281.0
360.3
39.8
64.5

22.4
43.6
17.3
23.6
1378.0
2433.9
9.1
33.0
8.1
1015.9
240.2
349.9
253.0
382.4
39.8
56.7

34.5
37.7
30.9
30.8
1893.9
2111.9
17.0
31.0
18.1
1197.8
349.4
320.7
341.8
324.7
53.0
55.7

52.0
58.0
52.0
56.6
2449.5
2829.9
35.9
82.3
49.3
2042.0
568.9
657.2
469.3
480.2
80.0
78.6

60.2
61.3
72.0
65.5
2581.8
2841.3
62.3
88.2
118.4
2400.6
1032.9
846.4
631.7
564.3
83.4
77.2

Flavonols (ng mg1 dry weight)

Hyperoside (1 d)
Hyperoside (8 d)
Quercetin-3-O-glucuronide (1 d)
Quercetin-3-O-glucuronide (8 d)

4.0
5.6
1.5
2.5

2.8
4.8
1.0
1.7

5.4
3.9
2.7
2.1

4.3
4.1
1.7
1.6

7.6
5.5
3.2
2.7

J.A. Stavang et al. / Scientia Horticulturae 195 (2015) 216225

221

Table 5
Probability levels of signicance (P - value) and standard errors (s.e.) are presented for regression analyses of organic acids, sugars, anthocyanins and avonols in raspberry
samples harvested at different colour (maturity) stages and stored for 1 d or 8 d in darkness at 23 C.

Phytochemicals

Colour at harvest

Storage

Colour at harvest Storage

Malate
s.e.
Cyanidin-3-O-sambubioside
s.e.
Cyanidin-3-2-xylosyl-rutinoside
s.e.
Cyanidin-3-O-sophoroside
s.e.
Pelargonidin-3-O-sophoroside
s.e.
Oxalate
s.e.
Pelargonidin-3-glucosyl-rutinoside
s.e.
Cyanidin-3-O-rutinoside
s.e.
Cyanidin-3-O-glucoside
s.e.
Cyanin
s.e.
Citrate
s.e.
Glucose
s.e.
Sucrose
s.e.
Hyperoside
s.e.
Quercetin-3-O-glucuronide
s.e.
Fructose
s.e.

0.008
3.23
0.005
9.57
<0.001
6.09
n.s.
478
0.007
19.0
n.s.
0.045
<0.001
21.9
<0.001
135
0.002
67.6
0.006
10.6
0.011
29.1
n.s.
22.2
n.s.
20.9
n.s.
1.25
n.s.
0.662
n.s.
28.5

0.008
2.74
0.004
7.68
<0.001
6.39
0.008
327
0.001
12.3
0.014
0.032
0.002
21.6
0.005
145
0.008
70.4
0.018
10.4
0.046
30.6
n.s.
19.5
n.s.
16.9
n.s.
1.33
n.s.
0.706
n.s.
23.0

<0.001
1.60
0.002
5.96
<0.001
6.44
0.003
233
0.005
13.1
0.015
0.028
0.01
23.2
0.016
149
0.014
66.1
0.012
8.6
n.s.
30.3
n.s.
18.4
n.s.
17.9
n.s.
1.15
n.s.
0.675
n.s.
21.4

n.s. - not signicant.

TA expressed as citric acid equivalents, varied with fruit colour at

harvest. This same trend was maintained in berries during storage,
and storage itself further reduced the amount of TA (Table 3). Thus,
both main factors (colour at harvest and storage time) were highly
signicant (P < 0.001), while there was no signicant interaction
between them.
Due to the reduction of acidity with increasing colour at harvest and extended storage, and small parallel changes in SS, the
sugar:acid ratio increased steadily with both colour at harvest and
length of storage (Table 3). To test whether the sensory panel recognized this change in sugar:acid ratio, we calculated the sugar:acid
ratio from the scores for sweetness and acidity that were given in
the sensory evaluation. This calculation demonstrated that also the
sensory panel judged the sugar:acid ratio to increase with colour at
harvest (Table 2). However, while the chemical analysis revealed a
slight but signicantly higher sugar:acid ratio in berries stored for
8 d compared to berries stored for 1 d (Table 3), the results of the
sensory panel indicated the opposite (Table 2). Apparently, the sensation of sweetness and acidity was inuenced by other parameters
than measurable sugar and acid concentrations.

3.3.2. Targeted analyses

The phytochemical analyses were focused on the concentrations
of organic acids, sugars and avonoids (including anthocyanins and
avonols). The concentration values for individual measurements
for each maturation level and length of storage are shown in Table 4.
The obtained results for samples stored for 1 d demonstrate that
citrate is the predominant organic acid in raspberry Glen Ample
(average 190 g mg1 ), followed by malate (7 g mg1 ) independently of maturity stage. Oxalate appeared only in very small
concentrations (0.5 g mg1 ). While concentrations for citrate and
malate decreased with increasing maturity levels, oxalate concen-

tration appeared to remain constant throughout all ve maturation

stages (Table 4).
The sugars fructose, glucose and sucrose appeared in similar
concentration ranges (120 g mg1 ). For fructose and glucose, no
trend was visible in dependence of the maturation stage at harvest.
Sucrose levels increased with colouration, however, this trend was
not signicant (Table 5). Furthermore, the total amount of sucrose,
fructose and glucose were higher for colour four and ve than they
were for colour one and two.
The most prominent anthocyanin in raspberry Glen Ample was
by far cyanidin-3-O-sophoroside (average 1900 ng mg1 ), followed
by cyanidin-3-O-rutinoside (500 ng mg1 ) and cyanidin-3-Oglucoside (400 ng mg1 ). For all anthocyanins, concentration
increased with maturity at harvest. In contrast to anthocyanins,
the concentrations of the avonols hyperoside and quercetin3-O-glucuronide were low, with averages of 5 and 2 ng mg1 ,
respectively. In both cases, the highest avonol concentrations
were found in berries with the darkest colour (Table 4).
Similar, but weaker trends were apparent for samples stored for
8 d in comparison to samples stored for just 1 d at 23 C. Generally,
the concentrations of all anthocyanins increased markedly during
the 8 d of storage, the increase being the largest in fruits harvested
at early maturity stages. A particularly large increase was noticeable
in the case of pelargonidin 3-(glucosyl) rutinoside, which in fruits
harvested at maturity stages one and two, increased by more than
a hundredfold during 8 d of storage (Table 4).

3.3.3. Phytochemical correlation and regression criteria for

targeted chemical analysis
Figs. 3 and 4 demonstrate the internal relationships between
organic acids, sugars, anthocyanins and avonols determined in
berries of ve different maturation stages and highlight how these

J.A. Stavang et al. / Scientia Horticulturae 195 (2015) 216225

0.99
0.99
0.98
0.99
0.86
0.74

0.99
0.99
0.99
0.83
0.71

-0.85
-0.76
-0.61
0.52
-0.01
0.78
0.81
0.89
0.86
0.91
0.96
0.99
0.99
0.96
1.00

-0.83
-0.67
-0.66
0.04
0.24
0.85
0.48
0.70
0.71
0.75
0.82
0.86
0.83
0.81
0.86
0.87

-0.81
-0.69
-0.76
-0.14
0.19
0.78
0.36
0.61
0.68
0.68
0.74
0.74
0.71
0.72
0.72
0.74
0.96

0.96
0.96 1.00
0.81 0.86 0.87
0.72 0.72 0.74 0.96

-0.33
-0.70
-0.80
0.63
0.79
0.91
0.22
0.88
0.88
0.98
0.93

-0.55
-0.81
-0.79
0.48
0.69
0.80
0.38
0.94
0.85
0.98
0.98
0.95

0.94
0.73
0.88
0.85
0.72
0.80
0.83
0.42
0.14

0.92
0.98
0.98
0.91
0.95
0.96
0.26
0.14

0.93
0.98
1.00
0.99
0.98
0.03
0.15

0.95
0.92
0.96
0.98
0.33
0.25

0.98
0.98
0.98
0.09
0.10

-0.61
-0.88
-0.85
0.46
0.65
0.77
0.35
0.86
0.72
0.91
1.00
0.92
0.98

-0.54
-0.85
-0.87
0.57
0.68
0.81
0.26
0.89
0.80
0.95
0.99
0.96
0.98
0.99

Querc Gluc

-0.58
-0.87
-0.86
0.49
0.67
0.79
0.33
0.86
0.73
0.92

Hyperoside

-0.40
-0.70
-0.74
0.54
0.75
0.85
0.33
0.95
0.94

Cy-3-O-Rut

-0.25
-0.49
-0.57
0.54
0.67
0.75
0.26
0.93

FL
Pel-Gluc-Rut

Pel-3-O- Soph

0.93
0.95
0.86
0.88
0.94
0.86
0.89
0.89
0.22
0.17

Cy-3-O-Gluc

-0.58
-0.75
-0.73
0.54
0.52
0.65
0.27

Cy -Gluc- Rut

-0.26 0.20 0.04 -0.07

-0.52 -0.22 -0.41 -0.08
-0.80 -0.34 -0.56 0.19
0.20 0.42 -0.61
0.20
0.97 0.50
0.42 0.97
0.35
-0.61 0.50 0.35
0.54 0.52 0.65 0.27
0.54 0.67 0.75 0.26
0.54 0.75 0.85 0.33
0.49 0.67 0.79 0.33
0.63 0.79 0.91 0.22
0.48 0.69 0.80 0.38
0.46 0.65 0.77 0.35
0.57 0.68 0.81 0.26
0.61 0.72 0.85 0.23
0.81 0.17 0.28 -0.68
0.90 -0.17 0.04 -0.88

Cy-3-O-Samb

Fructose

0.98
0.99
1.00
0.96
0.96
0.82
0.74

Querc Gluc

0.99
0.96
0.96
0.99
0.91
0.91
0.75
0.68

-0.86
-0.77
-0.62
0.54
-0.05
0.75
0.83
0.88
0.85
0.91
0.96
0.98
0.99
0.96

Hyperoside

0.99
0.97
0.92
0.92
0.97
0.85
0.86
0.71
0.68

-0.87
-0.89
-0.73
0.56
-0.15
0.76
0.87
0.97
0.97
0.99
1.00
0.99
0.99

Cy-3-O-Rut

Pel-3-O- Soph
-0.87
-0.83
-0.67
0.57
-0.10
0.76
0.86
0.93
0.92
0.96
0.99
0.99

Pel-Gluc-Rut

Cy-3-O-Gluc
-0.83
-0.81
-0.65
0.51
-0.01
0.83
0.81
0.95
0.92
0.96
0.98

Cy-Xyl-Rut

Cy -Gluc- Rut
-0.89
-0.90
-0.76
0.54
-0.17
0.76
0.86
0.96
0.97
0.99

Cy-3-O-Soph

0.88 0.65
0.91
0.91
-0.52 -0.80
-0.22 -0.34
-0.41 -0.56
-0.08 0.19
-0.75 -0.73
-0.49 -0.57
-0.70 -0.74
-0.87 -0.86
-0.70 -0.80
-0.81 -0.79
-0.88 -0.85
-0.85 -0.87
-0.81 -0.86
0.07 -0.31
-0.33 -0.62

Quinate

Oxalate

Malate

Citrate
0.88
0.65
-0.26
0.20
0.04
-0.07
-0.58
-0.25
-0.40
-0.58
-0.33
-0.55
-0.61
-0.54
-0.47
0.29
-0.19

-0.84
-0.91
-0.76
0.57
-0.20
0.74
0.87
0.99
0.99

FL

Anthocyanins

Cyanin

Citrate
Malate
Oxalate
Quinate
Frucose
Glucose
Sucrose
Cyanin
Cy-3-O-Soph
Cy-3-O-Samb
Cy -Gluc- Rut
Cy-3-O-Gluc
Pel-3-O- Soph
Cy-Xyl-Rut
Pel-Gluc-Rut
Cy-3-O-Rut
Hyperoside
Querc Gluc

-0.83
-0.94
-0.80
0.54
-0.25
0.70
0.85
0.98

Sugars

Sucrose

OA
Sugars
Anthocyanins
Fl

8 d storage

0.98
0.99
0.96
0.95
0.93
0.97
0.88
0.89
0.70
0.61

Cy-3-O-Samb

-0.75
-0.86
-0.67
0.59
-0.14
0.76
0.84

Cy-3-O-Soph

-0.38 0.35 -0.53 -0.78

-0.47 0.47 -0.51 -0.83
-0.21 0.47 -0.40 -0.64
-0.58 0.04 0.86
-0.58
0.50 -0.55
0.04 0.50
0.36
0.86 -0.55 0.36
0.59 -0.14 0.76 0.84
0.54 -0.25 0.70 0.85
0.57 -0.20 0.74 0.87
0.54 -0.17 0.76 0.86
0.51 -0.01 0.83 0.81
0.57 -0.10 0.76 0.86
0.56 -0.15 0.76 0.87
0.54 -0.05 0.75 0.83
0.52 -0.01 0.78 0.81
0.04 0.24 0.85 0.48
-0.14 0.19 0.78 0.36

OA

Glucose

Fructose

Quinate

Oxalate

0.91 0.90
0.95
0.95
-0.47 -0.21
0.47 0.47
-0.51 -0.40
-0.83 -0.64
-0.86 -0.67
-0.94 -0.80
-0.91 -0.76
-0.90 -0.76
-0.81 -0.65
-0.83 -0.67
-0.89 -0.73
-0.77 -0.62
-0.76 -0.61
-0.67 -0.66
-0.69 -0.76

Cyanin

0.91
0.90
-0.38
0.35
-0.53
-0.78
-0.75
-0.83
-0.84
-0.89
-0.83
-0.87
-0.87
-0.86
-0.85
-0.83
-0.81

Anthocyanins

Sucrose

Citrate
Malate
Oxalate
Quinate
Fructose
Glucose
Sucrose
Cyanin
Cy-3-O-Soph
Cy-3-O-Samb
Cy -Gluc- Rut
Cy-3-O-Gluc
Pel-3-O- Soph
Cy-Xyl-Rut
Pel-Gluc-Rut
Cy-3-O-Rut
Hyperoside
Querc Gluc

Malate

1 d storage

Sugars

Glucose

Fl

Anthocyanins

Sugars

OA

Citrate

OA

Cy-Xyl-Rut

222

-0.47
-0.81
-0.86
0.61
0.72
0.85
0.23
0.89
0.83
0.96
0.98
0.98
0.98
0.97
1.00

0.29
0.07
-0.31
0.81
0.17
0.28
-0.68
0.22
0.42
0.26
0.03
0.33
0.09
-0.01
0.15
0.22

-0.19
-0.33
-0.62
0.90
-0.17
0.04
-0.88
0.17
0.14
0.14
0.15
0.25
0.10
0.13
0.22
0.25
0.76

0.99
0.97 1.00
-0.01 0.15 0.22
0.13 0.22 0.25 0.76

Fig. 2. Correlation matrices of phytochemical parameters of raspberry fruit stored for (A) 1 d and (B) 8 d at 23 C in the dark. The correlation statistic ranges from 1 through
0 to 1, indicating a perfect negative correlation, no correlation, and perfect positive correlation. Correlations of statistical signicance (P < 0.05) are highlighted in red. (For
interpretation of the references to colour in this gure legend, the reader is referred to the web version of this article.)

relationships were altered during storage. The effect of storage on

phytochemical correlations is emphasized in the Venn-Diagram
(Fig. 3), which shows that 42 correlations are unique to samples
stored for 1 d, whereas samples stored for 8 d had 14 unique
correlations. Short-term storage and long-term storage shared 48
correlations (cut off points r < 0.75 and r > 0.75). This means that
the relationships between these pairs of metabolites collected at
different stages of development are similar between fresh and
stored samples. Thus, a large proportion of relationships between
metabolites remain unchanged during these storage conditions,

and colour at harvest inuence the nal metabolite content of the

raspberries.
Both malic and citric acid were strongly negatively correlated (r < 0.75) with all quantied anthocyanins and avonols in
berries stored for 1 d (Fig. 2A). Signicant negative correlations
(P < 0.05) were, however, only observed for citrate and cyanidin-3 (2
glucosyl) rutinoside, and in terms of malic acid for cyanidine-3-Osophoroside, cyanidine-3-O-sambubioside, cyanidin-3 (2 glucosyl)
rutinoside and cyanidine-3 (2-xylosyl) rutinoside. This strong negative relationship between citrate and anthocyanins weakened in

J.A. Stavang et al. / Scientia Horticulturae 195 (2015) 216225

Fig. 3. Venn-Diagram representing the number of unique metabolitemetabolite

correlations for berries stored for 1 or 8 d and the respective number of correlations
that were shared between the two storage treatments. Cut off point r < 0.75 and
r > 0.75.

berries stored for 8 d (r < 0.75, Fig. 2B), with no signicant correlation present.
Citrate was also strongly negatively correlated with the
avonols hyperoside and quercetin-3-O-glucoside (r < 0.75,
Fig. 2A, P < 0.05) in raspberry samples stored for 1 d, but the effect
was not statistically signicant. In contrast, no signicant relationship for those phytochemical parameters was observed in samples
stored for 8 d (Fig. 2B).
Glucose and sucrose were strongly positively correlated with
anthocyanins (r > 0.75) in berries stored for 1 d, while only slightly
weak negative correlations were observed between fructose and
all anthocyanins (Fig. 2A, P > 0.05). In contrast, fructose showed
stronger positive correlations (r > 0.5) with all anthocyanins in
berries stored for 8 d (Fig. 2B). The relationship between glucose and all anthocyanins remained similar for both 1 d and 8
d stored berries. Only one signicant correlation was observed
between glucose and cyanidin-3-O-glucoside in samples stored
for 8 days. However, only very weak positive and non-signicant
correlations between sucrose and anthocyanins were observed in
samples stored for 8 d. Furthermore, sucrose and fructose only
showed weak positive and non-signicant correlations with the
avonols hyperoside and quercetin-3-O-glucoside, while glucose
was strongly positively correlated with both avonols after 1 d of
storage (r > 0.75, Fig. 2A). A different result was found for samples
stored for 8 d: While no correlation was found for fructose and glucose with either avonol, a strong negative correlation was present
for sucrose and hyperoside (r = 0.68, P > 0.05) and quercetin-3-Oglucoside (r = 0.88, P < 0.05).
Sucrose was also strongly negatively correlated with both citric and malic acid (r < 0.75, P > 0.05, Fig. 2). A similar but weaker
trend was observed for glucose and organic acids in berry samples
stored for 1 d. No trend for those phytochemical parameters was
observed in samples stored for 8 d. Furthermore, correlation analysis revealed a stronger positive and also signicant relationship
between glucose and fructose (r > 0.75, P < 0.05) in samples stored
for 8 d in comparison to samples stored for only 1 d (Fig. 2).
All regression analysis results of individually quantied phytochemicals are summarised in Table 5. First, the relationship
between chemical compound concentration and maturity stage
was tested across the two storage times. Linear regression analysis revealed that the slopes of all organic acids and anthocyanins
were signicantly different from zero (P < 0.05), thus revealing a
linear relationship between the concentrations of the respective
organic acids or anthocyanins and the maturity stage at harvest.
On the other hand, no signicant linear relationships were observed
between maturity stage at harvest and the concentrations of sugars,
or avonols (Table 5).

223

In order to investigate if storage of 1 and 8 days inuenced

the content of individual compounds, we set the regression slopes
of each to be the same and tested if the intercept was signicantly different. As can be seen from Table 5, signicant
differences in intercepts for the two storage treatments were
observed for malate (P < 0.01), oxalate (P < 0.01), cyanidin-3-Osambubioside (P < 0.02), cyanidin-3-O-sophoroside (P < 0.001), and
pelargonidin-3-O-sophoroside (P < 0.02). In the case of cyanidin3-O-sophoroside, it was revealed that concentrations of this
anthocyanin were signicantly increased in samples stored for 8
d. The opposite was the case for malate, the concentrations of
which decreased signicantly during storage (Table 4). Finally, to
test whether the individual compounds in berries with colour 15
changed between 1 and 8 days of storage, we tested if the slopes of
the regressions for the compounds were different at day 1 and day
8. Signicant differences in the slopes for 1 and 8 d storage were
observed for malate and cyanidin-3-O-sophoroside, and reect a
decrease and increase, respectively, in these compounds in raspberries harvested with colour one-three during further storage of
these berries.
3.3.4. Correlation analysis between sensory and biochemical data
In addition to the metabolitemetabolite correlation analysis,
a further analysis was performed to assess correlations between
metabolites and sensory characteristics of the sampled fruit (Fig. 4).
Several trends were observed. The freshness trait was positively
correlated with the concentrations of citric and oxalic acids (r = 0.64
and 0.62 respectively) and signicantly negatively correlated with
the concentrations of sucrose (r = 0.69, P < 0.05) and several anthocyanins. Unsurprisingly, sweetness was signicantly positively
correlated with fruit concentrations of sucrose (r = 0.72, P < 0.05),
while fructose and glucose showed no such signicant relationship.
Also the concentrations of several anthocyanins were positively
correlated with the sweetness trait while the levels of oxalate
were negatively correlated. Sensory scores for acidity were positively correlated with the measured concentrations of citrate
(P > 0.05) and oxalate (P < 0.05), but signicantly negatively correlated with the measured concentrations of sucrose and some of the
anthocyanins (P < 0.05). The correlation analyses also indicated a
negative relationship between bitterness taste and fruit concentrations of oxalate (P < 0.05) and fructose (P > 0.05), whereas sucrose
and quinate tended to have the opposite effect. However, since the
bitterness trait was generally low, and did not vary signicantly
with fruit maturity stage or storage treatment (Table 2), these statistical relations should be interpreted cautiously. Furthermore,
the visual attractiveness of the fruit was signicantly negatively
correlated with the concentrations of most anthocyanin pigments
(P < 0.05) and of sucrose (P > 0.05, not signicant), while citrate concentration showed the opposite relationship. Finally, overall fruit
tastefulness was not strongly correlated with any specic phytochemical component, thus illustrating the complex nature of this
sensory trait.
Fig. 4 also presents the correlations between measured phytochemical components and some compositional quality parameters.
These reveal a strong signicant positive correlation between the
concentrations of major organic acids (citrate and malate) and TA,
while the concentrations of most anthocyanins were signicantly
negatively correlated with TA. The opposite situation was found for
the sugar:acid ratio, which was signicantly negatively correlated
with the concentration of the major organic acids and signicantly
positively correlated with all anthocyanins.
4. Discussion
Determination of fruit maturity by visual assessment of fruit
colour, and physical measurements of fruit adherence to the recep-

J.A. Stavang et al. / Scientia Horticulturae 195 (2015) 216225

Sweetness

Acidity

Bitterness

Tastefulness

Visual Attractiveness

Soluble Solids

Titratable Acids

Citrate
Malate
Oxalate
Quinate
Frucose
Glucose
Sucrose
Cyanin
Cy-3-O-Soph
Cy-3-O-Samb
Cy -Gluc- Rut
Cy-3-O-Gluc
Pel-3-O- Soph
Cy-Xyl-Rut
Pel-Gluc-Rut
Cy-3-O-Rut
Hyperoside
Querc Gluc

Chemic al Scores

Freshness
FL

Anthocyanins

Sugars

OA

Se ns ory Sc ores

0.64
0.25
0.62
-0.13
0.13
-0.10
-0.69
-0.36
-0.05
-0.32
-0.67
-0.59
-0.47
-0.70
-0.76
-0.76
-0.40
-0.32

-0.22
-0.33
-0.64
0.25
-0.13
0.18
0.72
0.53
0.22
0.43
0.60
0.55
0.21
0.62
0.43
0.52
0.28
0.21

0.52
0.31
0.75
-0.13
0.16
-0.13
-0.88
-0.36
-0.04
-0.32
-0.69
-0.58
-0.32
-0.72
-0.65
-0.70
-0.35
-0.27

-0.15
-0.10
-0.83
0.53
-0.62
-0.39
0.52
0.13
-0.18
-0.01
0.21
0.10
-0.20
0.25
0.06
0.15
-0.04
0.02

0.34
0.14
-0.49
0.15
-0.28
-0.14
0.36
-0.08
-0.27
-0.17
-0.06
-0.08
-0.45
-0.03
-0.25
-0.11
-0.09
-0.09

0.63
0.38
0.26
-0.08
-0.15
-0.32
-0.53
-0.62
-0.38
-0.57
-0.76
-0.68
-0.76
-0.78
-0.83
-0.76
-0.24
-0.15

-0.47
-0.50
-0.08
0.45
-0.30
-0.25
0.25
0.33
0.27
0.30
0.34
0.15
0.28
0.35
0.25
0.17
-0.30
-0.20

0.81
0.83
0.14
-0.02
-0.22
-0.44
-0.60
-0.80
-0.70
-0.83
-0.92
-0.78
-0.85
-0.91
-0.86
-0.78
-0.35
-0.33

619
620

Sugar: Acid Ratio

224

-0.82
-0.80
-0.15
0.10
0.17
0.38
0.59
0.78
0.68
0.80
0.90
0.75
0.86
0.89
0.86
0.77
0.27
0.26

Fig. 4. Pearsons correlation analysis of phytochemical components versus sensory and chemical scores characteristics as determined in raspberry fruits. Correlation is based
on both 1 d and 8 d storage data compiled together. Signicant (P < 0.05) are highlighted in red. (For interpretation of the references to colour in this gure legend, the reader
is referred to the web version of this article.)

tacle, or fruit compression resistance, yielded highly signicant and

comparable results (Table 1). Because of its ease of performance, the
colour method presents itself as the method of choice. Maturity,
as determined by visual assessment of fruit colour is also the main
method used by the fruit industry today. In a study on Korean black
raspberries, Hyun et al. (2014) also notes that anthocyanins are
powerful markers for ripening and organoleptic quality. However,
in order to become a standardized quantitative method, the colour
stages must be related to a standardised colour system such as the
Natural Colour System (NCS) used in this work (Fig. 1), or better,
such a standard must be developed particularly for raspberries and
made commercially available. Since the various cultivars vary both
qualitatively and quantitatively in their anthocyanin composition,
which are the main determinants of raspberry colour (e.g. Bradish
et al., 2011; Krger et al., 2011), the respective colour stages will
be cultivar specic. The same would apply to physical assessment
methods such as fruit/receptacle adherence and compression resistance which are also highly cultivar specic. We conclude from this
study with Glen Ample raspberry that, for this cultivar, the optimal
maturation stage and fruit colour at harvest for commercial fresh
marketing was the light red (colour two) described here which is
matching the NCS S code 3060-Y90R. Fruit harvested at this stage
attained all the chemical and sensory qualities of in situ matured
berries and maintained high sensory quality after 8 days of storage
at 23 C. This concurs closely with the results of Wang et al. (2009)
with the cultivar Caroline. Fruits for immediate consumption or for
freezing should, however, be harvested at a later maturity stage, in
case of Glen Ample at our colour stage three corresponding to NCS
S code 3060R.
The results documented substantial sensory and chemical
changes during the nal stages of raspberry ripening, and demonstrated that these changes are parallel in fruit matured in situ
and in harvested fruit stored in the dark at 23 C. As the fruit
matures, the concentrations of organic acids decrease and so does

TA, whereas the concentrations of anthocyanins increase while the

fruit colour develops (see Table 4 and Fig. 2). In parallel with these
changes, the sugar:acid ratio increases, mainly due to decreases in
TA while the parallel increase in sucrose was only minor (Table 3).
This is in agreement with the ndings of Haffner et al. (2002),
Wang et al. (2009) and Krger et al. (2011), where fruits were
stored in an unaltered atmosphere. It was evident however, that
the sensory evaluation of the sugar:acid ratios were different from
the measured ratios in berries stored for 8 d. Similarly, the visual
attractiveness judgement was signicantly negatively correlated
with the concentrations of several anthocyanins, and in particular
anthocyanins that are synthesized during the later stages of fruit
maturation (see Table 4), and positively with the concentrations of
some organic acids such as citric acid. These ndings indicate that
consumers tend to score more negatively both the visual attractiveness and the sweetness of fruits with higher levels of anthocyanins,
which usually correspond to fruits in later stages of development
(Fig. 4; Table 4). Therefore, the visual attractiveness appears to be
intrinsically linked to the colour and the maturity stage of the fruit.
Furthermore, there is a risk that the taste sensation for regular
consumers might be inuenced by the visual attractiveness of the
fruit and thus cultivars with dark fruit colours are likely to yield
lower taste scores than fruits with bright colours of the same real
quality. Bearing these considerations in mind, future sensory evaluations should be performed under both red and white light to
distinguish potential sensory biases caused by colour preferences
of the evaluators. In combination with a colour scale, this would also
make results more generally applicable since in parts of Europe, the
darker types of raspberries are preferred.
In this connection, it was interesting to note that the measured
concentrations of sucrose (Fig. 2A) as well as the sugar:acid ratios
(Fig. 4) were closely positively correlated with the concentrations
of most anthocyanins in the fruit samples. Furthermore, for some of

J.A. Stavang et al. / Scientia Horticulturae 195 (2015) 216225

the anthocyanins there was a similar but weaker correlation with

sensory sweetness (Fig. 4).
The correlation analysis results shown in Fig. 4 demonstrate that
a wide range of phytochemical components was signicantly and
diversely correlated with one or more of the sensory traits. Prominent in this respect were sucrose and the organic acids, which were
signicantly and oppositely correlated with several sensory traits,
thus demonstrating a potential relationship of these components
for fruit taste in general. Also, the correlation results between a
wide range of sensory traits and oxalate which was present in the
fruit in minor concentrations only was particularly surprising (Fig.
4; Table 4). However, it should be kept in mind that oxalic acid is a
strong acid with a particularly sharp taste. The results in Fig. 4 also
demonstrate a strikingly similar relationship between the concentrations of the various chemical components and the sensation of
freshness, acidity and the visual attractiveness of the fruits, which
also varied in parallel with the measured TA, but inversely with the
measured sugar:acid ratios.
We conclude that changes in raspberry fruit maturity, which
is closely related to changes in a number of phytochemical components, can be precisely assessed by visual observation of fruit
colour and by physical measurements of fruit/receptacle adherence
or fruit compression resistance. Furthermore, the storage capacity
of the fruits are also determined by colour and maturity at the time
of harvest and thus colour assessment when related to a standardised colour system such as the Natural Colour System (NCS), will
secure picking of berries with optimum quality.
Acknowledgements
Thanks to Iren Lunde Knutsen and Sigrid Flatland for technical
assistance and to the staff at BAMAs head ofce that performed
the sensory evaluation. We gratefully acknowledge nancial support of this work by funding from the Regional Research Funds in
Norway (RFF-Vestlandet) through Project ID: 203935, and from the
Research Council of Norway through project No. 234312/E50. Furthermore, the James Hutton Institute acknowledges funding by IVB
as part of the NSR program: ClimaFruit Project No. 35-05-09.
Appendix A. Supplementary data
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.scienta.2015.
08.045.
References
Anttonen, M.J., Karjalainen, R.O., 2005. Environmental and genetic variation of
phenolic compounds in red raspberry. J. Food Comp. Anal. 18, 759769.
Austin, M.E., Shutak, V.G., Christopher, E.P., 1960. Colour changes in harvested
strawberry fruit. J. Am. Soc. Hortic. Sci. 75, 382386.
Bradish, C.M., Perkins-Veazie, P., Fernandez, G.E., Xie, G., Jia, W., 2011. Comparison
of avonoid composition of red raspberries (Rubus idaeus L:) grown in the
Southern United States. J. Agric. Food Chem 60, 57795786.

225

Finn, C.E., Moore, P.P., Kempler, C., 2008. Raspberry cultivars: Whats new? Whats
succeeding? Where are breeding programs headed? Acta Hortic. 777, 3340.
Grisebach, H., 1982. Biosynthesis of anthocyanins. In: Markakis, P. (Ed.),
Anthocyanins as Food Colours. Academic Press, NY, USA, pp. 4767.
Haffner, K., Rosenfeld, H.J., Skrede, G., Wang, L., 2002. Quality of red raspberry
Rubus idaeus L: cultivars after storage in controlled and normal atmospheres.
Postharvest Biol. Technol. 24, 279289.
Hyun, T.K., Lee, S., Rim, Y., Kumar, R., Han, X., Lee, S.Y., Lee, C.H., Kim, J.Y., 2014.
De-novo RNA sequencing and metabolite proling to identify genes involved
in anthocyanin biosynthesis in Korean black raspberry (Rubus coreanus
Miquel). PLoS One 9 (2), e88292, http://dx.doi.org/10.1371/journal.pone.
0088292, eCollection.
Kalt, W., Prange, R.K., Lidster, P.D., 1993. Postharvest colour development of
strawberries inuence of maturity, temperature and light. Can. J. Plant Sci. 73,
541548.
Kalt, W., Forney, C.F., Martin, A., Prior, P.L., 1999. Antioxidant capasity, vitamin C,
phenolics, and anthocyanins after fresh storage of small fruits. J. Agric. Food
Chem. 47, 46384644.
Krger, E., Schopplein, E., Rasim, S., Cocca, G., Fischer, H., 2003. Effects of ripening
stage and storage time on quality parameters of red raspberry fruit. Eur. J. Hort.
Sci. 68, 176182.
Krger, E., Dietrich, H., Schopplein, E., Rasim, S., Kurbel, P., 2011. Cultivar, storage
conditions and ripening effects on physical and chemical qualities of red
raspberry fruit. Postharvest Biol. Technol. 60, 3137.
Liu, M., Li, X.Q., Weber, C., Lee, C.Y., Brown, J., Liu, R.H., 2002. Antioxidant and
antiproliferative activities of raspberries. J. Agric. Food Chem. 50, 29262930.
Mazur, S.P., Snsteby, A., Wold, A.-B., Foito, A., Verrall, S., Conner, S., Stewart, D.,
Heide, O.M., 2014. Post-owering photoperiod has marked effects on fruit
chemical composition in red raspberry (Rubus idaeus L.). Ann. Appl. Biol. 165,
454465.
Mullen, W., Stewart, A.J., Lean, M.E.J., Gardner, P., Duthie, G.G., Crozier, A., 2002.
Effects of freezing and storage on the phenolics, ellagitannins, avonoids, and
antioxidant capacity of red raspberries. J. Agric. Food. Chem. 50, 51975201.
Rao, A.V., Snyder, D.M., 2010. Raspberries and human health: a review. J. Agric.
Food Chem. 58, 38713883.
Remberg, S.F., Snsteby, A., Aaby, K., Heide, O.M., 2010. Inuence of postowering
temperature on fruit size and chemical composition of Glen Ample raspberry
(Rubus idaeus L.). J. Agric. Food Chem 58, 91209128.
Sachs, E.J., Shaw, D.V., 1993. Colour change in fresh strawberry fruit of seven
genotypes stored at 0 C. HortScience 28, 209210.
Steyn, W., Wand, S., Holcroft, D., Jacobs, G., 2002. Anthocyanins in vegetative
tissue: a proposed unied function in photoprotection. New Phytol. 155,
349361.
Snsteby, A., Myrheim, U., Heiberg, N., Heide, O.M., 2009. Production of
high-yielding raspberry long canes in a Northern climate. Sci. Hortic. 121,
289297.
Wang, S.Y., Liu, H.S., 2000. Antioxidant activity in fruit and leaves of blackberry,
raspberry, and strawberry varies with cultivar and developmental stage. J.
Agric. Food Chem. 48, 140146.
Wang, S.Y., Chen, C.-T., Wang, C.Y., 2009. The inuence of light and maturity on
fruit quality and avonoid content of red raspberries. Food Chem. 112,
676684.