Ripening and postharvest behaviour of fruits of two hylocereus species (cactaceae)

Postharvest Biology and Technology 17 (1999) 39–45
Ripening and postharvest behaviour of fruits of two
Hylocereus species (Cactaceae)
Avinoam Nerd
a
, Feiga Gutman
a
, Yosef Mizrahi
a,b,
*
a
The Institutes for Applied Research, Ben-Gurion Uni6ersity of the Nege6, P.O. Box
653
, Beer-She6a
84105
, Israel
b
Department of Life Science, Ben-Gurion Uni6ersity of the Nege6, P.O. Box
653
, Beer-She6a
84105
, Israel
Received 10 February 1999; accepted 26 May 1999
Abstract
Fruit growth and ripening, and the effect on fruit quality of various storage temperatures, were studied with
Hylocereus undatus and Hylocereus polyrhizus plants growing in Beer-Sheva (Israeli Negev desert) under greenhouse
conditions. Fruit growth was sigmoidal with a strong decline in growth rate after the onset of peel colour change. The
first change in peel colour was recorded 24–25 days after anthesis in H. undatus and 26– 27 days in H. polyrhizus.In
both species, the peel turned fully red 4–5 days after the first colour change (mean temperature for the study period
was 26.6 9 2.1°C). The slow growth phase was characterised by a decrease in the proportion of peel and concomitant
increase in that of pulp, increase in the concentration of soluble solids and soluble sugars and a decline in firmness

and the concentration of starch and mucilage. The surge in acidity prior to colour change indicated the beginning of
the ripening processes. For H. polyrhizus, which has a red–violet pulp, the increase in pulp pigment paralleled the
development of peel colour. Fruits were non-climacteric, and when harvested at close to full colour, they retained
market quality for at least 2 weeks at 14°C or 1 week at 20°C. Storage at 6°C is not recommended, because transfer
from that temperature to room conditions caused fruits to lose their firmness and flavour rapidly. © 1999 Elsevier
Science B.V. All rights reserved.
Keywords
:
Hylocereus undatus; Hylocereus polyrhizus; Chemophysical changes; Gas production; Pitaya; Storage
www.elsevier.com/locate/postharvbio
1. Introduction
Several species of climbing cacti of the genus
Hylocereus have recently been developed as fruit
crops (Barbeau, 1990; Reyes-Ramos, 1995;
Mizrahi et al., 1997). The fruit of these species,
known as red pitaya in Latin America, is a
medium– large berry bearing large green or red
scales (Nerd and Mizrahi, 1997). The peel is usu-
ally red, and the pulp varies from purple –red to
white. The pulp is delicate and juicy and contains
numerous small soft seeds. The plants are grown
in the open in tropical areas, but must be pro-
tected from intense solar radiation and subfreez-
ing temperatures when cultivated under
subtropical conditions such those prevailing in
Israel (Mizrahi et al., 1997; Raveh et al., 1997).
* Corresponding author. Tel.: + 972-7-6461966; fax:+972-
7-6472984.
E-mail address
:

(Y. Mizrahi)
0925-5214/99/$ - see front matter © 1999 Elsevier Science B.V. All rights reserved.
PII: S0925-5214(99)00035-6
A. Nerd et al.
/
Posthar6est Biology and Technology
17 (1999) 39 – 45
40
Fruits of cultivated species of Hylocereus are
picked at various stages of peel colour develop-
ment, from first appearance to full colour (Bar-
beau, 1990; Reyes-Ramos, 1995). However little
research has been done on fruit development or
ripening and on the behaviour of the fruit during
or after storage. Studies with yellow pitaya (Se-
lenicereus megalanthus), a related climbing cactus,
show that the duration of fruit development de-
pends on seasonal temperatures and that the
fruits reach the optimal flavour close to full
colour stage (Nerd and Mizrahi, 1998). This stage
is also the best time for harvest for short- or
long-distance markets (Nerd and Mizrahi, 1999).
With the objective of determining ripening
criteria for Hylocereus undatus and Hylocereus
polyrhizus, we examined flavour and a number of
physicochemical parameters during the last stage
of fruit development. In addition, the effect of
storage temperature on fruit quality was deter-
mined for fruits harvested close to full colour.
2. Materials and methods

2
.
1
. Plant material and growth conditions
Studies were conducted in 1997 and 1998 in an
orchard of climbing cacti grown in a ventilated
greenhouse in Beer-Sheva (northern Israeli Negev
desert). The plants had been established from
cuttings in 1992. Clone B of H. undatus (red peel
and white pulp) and clone C of H. polyrhizus (red
peel and red–purple pulp; Weiss et al., 1994) were
used for the study. Each clone sample consisted of
nine plants. Spaces were 1.5 m within the row and
2.5 m between rows. Plants were trained on a 1.5
m-high trellis system for support, and the soil was
a loamy loess. Since these plants cannot grow in
Israel under full sunlight, about 50% shade was
provided by black net. Average monthly tempera-
ture and relative humidity during the study peri-
ods (July–October 1997 and the same months in
1998) ranged from 25 to 30°C and 49 –87% RH,
respectively. Plants were drip-irrigated once a
week, to provide5lofwater per plant in the
summer and 2.5 l in the winter. Fertiliser contain-
ing 23% N, 3% P, 20% K at a concentration of 70
mg l
−1
was applied with the irrigation water.
Since both clones were self-incompatible (Weiss
et al., 1994), the nocturnal flowers, which stayed

open for 1 night only, were hand cross-pollinated
in the night with pollen of the other species.
Pollinated flowers were tagged to allow calcula-
tion of the number of days elapsed from anthesis
to various stages of fruit growth and
development.
2
.
2
. Fruit growth and ripening
Length and mid-length diameter were measured
with a caliper during fruit development in four
fruits of each clone, set at the beginning of Au-
gust 1997. Physicochemical parameters and
flavour were determined in fruits of both clones
set between mid-July and mid-August of the same
year. Fruits were harvested at random at 3 –4-day
intervals between 25 and 41 days after anthesis
(DAA) for a total of six fruits per clone at each
date. At the first date the fruits were either green
or on the verge of first colour; 8 –10 days before
the last date they were all fully red.
2
.
3
. Storage
Fruits were picked in September 1998, either at
first colour stage or 2 –3 days after appearance of
first colour (close to full colour). They were
washed with tap water to remove the sugars ex-

creted from the scales during fruit development,
dried with tissue paper and distributed randomly
among three dark chambers held at 69 1.6°C,
149 1.4°C or 20 9 1°C, and at 65–75% RH. Sixty
fruits of each species were stored at each tempera-
ture. Physicochemical parameters and flavour
were examined once a week, until the appearance
of visible disorders such as scale wilting. Fruits
stored at 6°C were also analysed after 1 week of
further storage at 20°C. Five fruits were analysed
at each sampling date.
2
.
4
. Fruit analysis
Colour values of peel and pulp at the equator
of the fruit were determined with a Minolta
Chroma Meter CR-200 (Ramsey, NJ) and ex-
pressed as hue angles (McGuire, 1992). Fruit firm-
A. Nerd et al.
/
Posthar6est Biology and Technology
17 (1999) 39 – 45
41
ness was determined at two opposite points on
the equator of each fruit with a penetrometer
(Faccini, Alfonsine, Italy) fitted with a 2.5-mm
plunger. Peel and pulp were separated and
weighed, and a sample of each was weighed
both when fresh and after drying at 70°C for

determination of water content. The concentra-
tion of red pigment in H. polyrhizus pulp which
belongs to the betacyanin group (Forni et al.,
1992), was determined in 1g samples obtained
from the inner part of the pulp. Tissue was ho-
mogenised in 80% ethanol and filtered through
Whatman paper no. 41. The filtrate was mixed
with 0.1 M acetate buffer at pH 4.5, and ab-
sorbance determined at 538 nm (u
max
) with a
Uvikon 810 spectrophotometer (Kontron, Mi-
lano, Italy). The concentration of pulp pigments
was expressed as betanin using A
538
(1%)= 1120
(Forni et al., 1992). Acid concentration was de-
termined from 10 g of pulp macerated in dis-
tilled water and titrated with 0.05 M NaOH to
pH 8.0. Ascorbic acid concentration was deter-
mined in extracts of fresh pulp by titration with
2,6-dichloro-indophenol (AOAC, 1990). Soluble
solids concentration (SSC) was measured with a
refractometer (PR-100, Atago, Japan) in sap
pressed from the pulp. The concentrations of
total soluble sugars, starch, and mucilage were
determined in dry samples of pulp according to
a procedure previously reported (Nerd and
Mizrahi, 1999). Flavour of fruits was assessed
by a panel of ten on a hedonic scale with 1

indicating extreme dislike and 10 indicating
strong liking.
2
.
5
. Ethylene and CO
2
production rates
Fruits picked at first colour and 3 and 6 days
later (three per species at each date) were
weighed and then enclosed individually in jars
held at 20°C and under a light regime of 12 h
fluorescent light/12 h dark. A continuous air
flow at approximately 5 ml min
−1
was passed
through the jar. Ethylene and CO
2
concentra-
tions in the effluent air stream were analysed
with a gas chromatograph (Varian 3300, Sugar-
land, TX) once a day.
3. Results and discussion
3
.
1
. Fruit growth and ripening
In both H. undatus and H. polyrhizus, growth of
fruits attached to the vine exhibited a sigmoid
pattern (Fig. 1). Peel colour was developed during

the last phase which was characterised by a slow
growth rate. First colour (area between the scales
turned faint red) appeared 24–25 DAA in H.
undatus and 26 –27 DAA in H. polyrhizus. Fruits
of both species turned fully red 4–5 days after first
colour. Like the fruit dimensions, fruit fresh weight
reached 80% of the final maximum weight, before
appearance of first colour (Fig. 2(A)). Significant
increase in pulp content (percent of fruit fresh
weight) occurred several days before, and
throughout colour change (Fig. 2(B)), while fruit
firmness dropped to less than 1 kg cm
−2
in that
time (Fig. 2(C)). During the final stage, percentage
of water in the peel decreased (more markedly in
H. undatus) while remaining almost constant in the
pulp (Fig. 2(D)). Calculations based on data
Fig. 1. Growth (length and diameter) during fruit development
on the vine in H. undatus and H. polyrhizus. Values are
means9SE (n=4).
A. Nerd et al.
/
Posthar6est Biology and Technology
17 (1999) 39 – 45
42
Fig. 2. Whole fruit fresh weight (A), percentage of peel or pulp
in fruit by weight (B), firmness (C), and water percent (FW
basis) (D), during the last stage of fruit development in H.
undatus and H. polyrhizus. Values are means9 SE (n=6).

Fig. 3. Colour changes in H. undatus (red peel and white pulp)
and H. polyrhizus (red peel and red-violet pulp) during the last
stage of fruit development. Values are means9 SE (n=6).
of colour change was low in fruits of both species,
ranging from 7.0 to 11.4 mg 100 g
−1
fresh weight.
Changes in the concentration of soluble solids and
a number of non-structural carbohydrate compo-
nents of the pulp occurred during fruit develop-
ment. Both soluble solids and soluble sugars
concentrations increased significantly during
colour development, reaching 16–17 and 8–9%,
respectively, at full colour stage (Fig. 5(A, B)).
Since changes in the two parameters were highly
correlated, SSC appears to be a reliable indicator
for maturity in the investigated species. The con-
centrations of starch and mucilage decreased in
relation to the accumulation of soluble sugars
presented in Fig. 2 show that during the slow
growth period (25– 41 DAA), pulp dry weight
increased significantly from 36 to 62 g in H.
undatus and from 21 to 48 g in H. polyrhizus,
while peel dry weight (around 18 g in both species
at 25 DAA) decreased slightly by4ginthat time.
Hence the peel although accounting for a high
proportion of fruit weight 20–25 DAA, could not
have been an important source of assimilates to
the developing pulp which presumably received
assimilates from the stem.

The decreasing values of the peel hue angle
during maturation (Fig. 3) reflect development of
peel colour. In H. polyrhizus, which has a red –vi-
olet pulp, the development of peel colour was
accompanied by an increase in the content of
water-soluble pigment in the pulp (Fig. 3). The
maximum concentration of pigment (expressed as
betanin) in fruits with fully red peel was 0.24 mg
g
−1
fresh weight. Titratable acidity of pulp
showed a short-lived surge at the start of colour
change, followed by a decline later on to  30
mmol H
+
kg
−1
in H. undatus and  45 mmol
H
+
kg
−1
in H. polyrhizus (equivalent to 0.22 and
0.32% citric acid, respectively; Fig. 4). Ascorbic
acid concentration of the pulp during the period
Fig. 4. Concentration of titratable acidity (FW basis) during
the last stage of fruit development in H. undatus and H.
polyrhizus. Values are means9 SE (n=6).
A. Nerd et al.
/

Posthar6est Biology and Technology
17 (1999) 39 – 45
43
Fig. 5. Concentration (FW basis) of soluble solids (SSC) (A),
soluble sugars (B), starch (C), and mucilage (D), during the
last stage of fruit development in H. undatus and H.
polyrhizus. Values are means9 SE (n=6).
and SSC and soluble sugars were high. The ratio
of percentage of soluble sugars to acidity (in terms
of citric acid) in such fruits was  40 in H.
undatus and  22 in H. polyrhizus. Thus for both
red pitayas, chemical changes during ripening,
colour development and palatability were found
to coincide, as reported for the yellow pitaya S.
megalanthus (Nerd and Mizrahi, 1998). In the
cactus pear (Opuntia ficus indica), however, opti-
mum eating quality was reached at first colour
stage (Nerd and Mizrahi, 1997).
3
.
2
. Gas production
Production of ethylene and CO
2
in fruits har-
vested at various colour stages was low and did
not peak during 6 days at 20°C. The production
rate of ethylene ranged from 0.025 to 0.091 ml
kg
−1

h
−1
and that of CO
2
from 0.52 to 0.78 ml
kg
−1
h
−1
. Hence H. undatus and H. polyrhizus
may be defined as non-climacteric, similar to
other cactus fruit crops that have been investi-
gated, namely cactus pears (Opuntia species) and
yellow pitaya (S. megalanthus; Nerd and Mizrahi,
1997, 1998).
3
.
3
. Storage
Fruits were picked close to full colour and
examined at the end of 3 consecutive weeks in
storage at different temperatures. Distinct disor-
ders were observed at the end of week 2 for fruits
held at 20°C and at the end of week 3 for fruits
held at 14°C. Symptoms consisted of extreme
softening, difficulty in separation of peel from the
pulp, and wilting and browning of scales (these
phenomena appeared usually together). The con-
centration of soluble solids and soluble sugars
remained fairly constant throughout storage at all

the storage temperatures (Table 2). High storage
temperatures resulted in lower fruit firmness, wa-
ter content, acidity and flavour. The two species
responded to storage in a similar manner, though
fruits of H. polyrhizus tended to maintain a higher
level of acidity at 14 and 20°C. This higher acidity
may explain the higher preference rating of H.
polyrhizus recorded under these conditions. Some
of the fruits stored at 6°C were transferred at
(Fig. 5(C, D)). Maximal concentration of starch
was less than half that of the soluble sugars in
ripe fruit, and accumulation of soluble sugars
during ripening cannot be therefore attributed to
starch degradation alone. The two species differed
with regard to timing of changes in chemical
constituents and other maturation parameters
which coincided with the timing of first colour
appearance.
Organoleptic testing indicated that fruits were
most palatable 33 –37 DAA (Table 1). At that
stage they had turned fully red, acidity was low
Table 1
Flavour rating of fruits of H. undatus and H. polyrhizus
harvested between 25 and 41 DAA
a
DAA Flavour (rating)
H. undatus H. polyrhizus
25 1.3 d
b
1.4 d

3.3 c3.7 c29
5.7 b33 7.0 a
37 6.5 a 6.4 a
5.8 b 5.3 b41
a
Flavour was assessed by a panel of ten according to a
hedonic scale with 1 indicating extreme dislike and 10 indicat-
ing strong liking.
b
Mean separation within column at P50.05 by Duncan’s
multiple range test.
A. Nerd et al.
/
Posthar6est Biology and Technology
17 (1999) 39 – 45
44
Table 2
Effect of storage temperatures on fruit attributes in H. undatus and H. polyrhizus
a
Storage Water loss SSCFirmness Soluble sugars Acidity mmol H
+
Flavour
(%)(kg cm
−2
)kg
−1
(%) (rating)(%)
Weeks°C
H. undatus
14.0 a 6.6 a2.4 a

b
93 aInitial 6.2 a
6 2.4 a1 1.8 c 13.6 a 6.6 a 62 b 5.2 a
2.3 a 2.8 c 13.7 a2 6.7 a6 61 b 5.9 a
36 0.3 b 5.8 a 13.7 a 6.7 a 65 b 5.9 a
2.4 a 2.7 c 13.3 a 6.0 a 37 c14 5.9 a1
2.0 a 4.4 b 12.2 a2 5.4 a14 17 d 2.8 b
20 1.8 a1 4.2 b 13.4 a 6.2 a 16 d 2.6 b
H. polyrhizus
Initial 13.2 bcd2.5 a 6.3 ab 93 ab 6.1 a
2.7 a 2.2 d 12.7 cd1 5.2 b6 100 a 5.8 a
2.4 a 4.0 c 13.0 bcd6 6.3 ab2 84 abc 6.4 a
0.4 c 5.4 a 14.5 a3 5.4 b6 81 bc 5.6 ab
114 1.8 b 2.5 d 13.4 bc 6.8 a 71 cd 6.9 a
1.5 b 4.5 bc 12.0 d2 5.9 ab14 51 e 5.3 ab
120 1.7 b 4.9 ab 13.9 ab 6.8 a 58 de 4.4 b
a
Fruits were harvested close to full colour and analysed prior to appearance of distinct disorders such as extreme softening and
scale wilting. Values are means9 SE (n=5).
b
Mean separation within column at P50.05 by Duncan’s multiple range test.
different times to 20°C and their attributes exam-
ined 1 week later (data not shown). Fruits trans-
ferred at the end of week 1 were similar to those
stored immediately after harvest for 1 week at 20°C
(Table 2). However fruits transferred to 20°C at the
end of week 2 or 3 at 6°C became soft (B 0.3 kg
cm
−2
), had high water loss (8–11%), low acidity

(B 25 mol H
+
kg
−1
) and poor flavour (B 2.5).
Fruits of H. undatus developed chilling injury
symptoms such as wilting and darkening of the
scales and browning of the outer layer of the pulp
upon transfer to 20°C after 2 weeks at 6°C.
To summarise, fruit harvested close to full colour
kept their visual acceptance, i.e. marketing quality
for at least 3 weeks at 6°C, 2 weeks at 14°C or 1
week at 20°C. Fruits stored at 6°C maintained their
eating quality (flavour) for a longer period but lost
quality rapidly when transferred to room tempera-
ture. The effect of harvesting at an earlier ripening
stage (first colour) on fruit quality during or after
storage should be examined with a view to extend-
ing the shelf life of these pitayas.
Acknowledgements
The authors thank the Fleischer Foundation and
the Israel Ministry of Agriculture for a partial
financial support, Dorot Imber for editing the
manuscript and Josef Mouyal and Eyal Naus for
their technical assistance.
References
AOAC, 1990. Official Methods of Analysis, 15th ed. Associa-
tion of Official Analytical Chemist, Arlington, VA.
Barbeau, G., 1990. La pitahaya rouge, un nouveau fruit
exotique. Fruits 45, 141–147.

Forni, G., Polesello, A., Montefiori, D., Maestrelli, A., 1992.
High-performance liquid chromatographic analysis of the
pigments of blood-red prickly pear (Opuntia ficus indica).
J. Chromatogr. 593, 177–183.
McGuire, R.G., 1992. Reporting the objective colour measure-
ments. HortScience 27, 1254–1255.
Mizrahi, Y., Nerd, A., Nobel, P.S., 1997. Cacti as crops. Hort.
Rev. 18, 292–320.
Nerd, A., Mizrahi, Y., 1997. Reproductive biology of fruit
cacti. Hort. Rev. 18, 322–346.
A. Nerd et al.
/
Posthar6est Biology and Technology
17 (1999) 39 – 45
45
Nerd, A., Mizrahi, Y., 1998. Fruit development and ripening
in yellow pitaya. J. Am. Soc. Hort. Sci. 123, 560–562.
Nerd, A., Mizrahi, Y., 1999. The effect of ripening stage on
fruit quality after storage of yellow pitaya. Postharvest
Biol. Technol. 15, 99–105.
Raveh, E., Nerd, A., Mizrahi, Y., 1997. Responses of two
hemiepiphytic fruit-crop cacti to different degrees of
shade. Sci. Hort. 53, 115–122.
Reyes-Ramos, V., 1995. El Cultivo de las Pitahayas y sus
Perspective de Desarrollo en Mexico. Reyes-Ramos,
Tabasco, Mexico.
Weiss, J., Nerd, A., Mizrahi, Y., 1994. Flowering behavior
and pollination requirements in climbing cacti with fruit
crop potential. HortScience 29, 1487–1492.
.