Front Cover
 Table of Contents
 Hair sheep performance in an extensive,...
 Sustainable strategies for reducing...
 Desmanthus, a legume for forage...
 Improving culinary herb production...
 Comparison of tilapia species for...
 Agricultural experiment station...
 Research projects and publicat...
 Back Cover

Group Title: UVI research
Title: UVI research. Volume 5.
Full Citation
Permanent Link: http://ufdc.ufl.edu/CA01300008/00001
 Material Information
Title: UVI research. Volume 5.
Series Title: UVI research
Physical Description: Serial
Language: English
Publisher: Agricultural Experiment Station, University of the Virgin Islands
Publication Date: 1993
 Record Information
Bibliographic ID: CA01300008
Volume ID: VID00001
Source Institution: University of the Virgin Islands
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 31871075


This item has the following downloads:

UVI research ( PDF )

Table of Contents
    Front Cover
        Front Cover
    Table of Contents
        Table of Contents
    Hair sheep performance in an extensive, native pasture system: A five year summary
        Page 1
        Page 2
        Page 3
    Sustainable strategies for reducing papaya disease problems in the U.S. Virgin Islands
        Page 4
        Page 5
    Desmanthus, a legume for forage production in the Caribbean
        Page 6
        Page 7
        Page 8
    Improving culinary herb production with drip irrigation in the Virgin Islands
        Page 9
        Page 10
        Page 11
        Page 12
    Comparison of tilapia species for cage culture in the Virgin Islands
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
    Agricultural experiment station personnel
        Page 18
    Research projects and publications
        Page 19
        Page 20
        Page 21
    Back Cover
        Back Cover
Full Text

Unvestyo th Vigi Isad Agrclua Exeimn Staio Voum 5, 1993


1 Hair Sheep Performance in an Extensive, Native Pasture System: A Five Year Summary
Stephan Wildeus and Joni R. Collins

4 Sustainable Strategies for Reducing Papaya Disease Problems in the U.S. Virgin Islands
Christopher Ramcharan

6 Desmanthus, a Legume for Forage Production in the Caribbean
Martin B. Adjei

9 Improving Culinary Herb Production with Drip Irrigation in the Virgin Islands
Manuel C. Palada, Stafford M.A. Crossman and Charles D. Collingwood

13 Comparison of Tilapia Species for Cage Culture in the Virgin Islands
James E. Rakocy, John A. Hargreaves and Donald S. Bailey

18 AES Personnel

19 Current Research Projects

19 Recent Publications

Orville Kean,
President, University of the Virgin Islands

Darshan S. Padda,
Vice President for Research and Land-Grant Affairs
and Experiment Station Director

James E. Rakocy,
Associate Director

Robin Sterns, Editor

UVI Research is published annually by the University of the Virgin Islands Agricultural Experiment Station, RR 2,
Box 10,000, Kingshill, St. Croix, USVI, 00850.
Contents of this publication constitute public property. The written material may be reprinted if no endorsement of a
commercial product is stated or implied. Please credit the authors, the University of the Virgin Islands Agricultural Experiment
Station and UVI Research.
To avoid overuse of technical terms, trade names of products are occasionally used. No endorsement of these products
or firms is intended, nor is criticism implied of those not mentioned.
The University of the Virgin Islands is committed to the policy that all persons shall have equal access to its programs,
facilities and employment without regard to race, religion, color, sex, national origin, age or veteran status.

Hair Sheep


in an





A Five Year


Stephan Wildeus
Joni R. Collins

The most likely production
environment encountered by hair sheep
in the Caribbean region is that of a
low-input, small scale farming system.
Under these conditions, nutrition,
management and a lack of genetic
diversity become the major constraints
to efficient production. In order for an
animal to be productive in this type of
environment it must successfully cope
with these constraints. However, very
little research data are available on the
productivity of hair sheep under
Caribbean conditions and how factors
such as ewe age and season affect
animal performance.
Among the objectives of the Sheep
Research Facility at the Agricultural
Experiment Station of the University
of the Virgin Islands is to evaluate the
production potential of the local hair
sheep under a management system

typically experienced by these animals.
To this end, animals were grazed
rotationally on native pasture (grass/
legume association) without
supplementation. The rate of rotation
was based on forage availability. All
animals were also confined overnight
to control predation and larceny. Ewes
lambed without help on pasture and
were provided no assistance in
mothering up. Anthelmintics were
administered routinely at two-month
Modifications to the traditional
system included controlled breeding
and weaning (nine weeks after
lambing) to facilitate data collection
and analysis. The breeding system was
based on an eight-month lambing cycle
with ewes divided into an A and B
flock, bred four months apart in a 34-
day single sire mating period. Ewe
body weights were recorded at weekly
intervals, while lamb weights were
recorded within 24 hours of birth and
at weaning (634 days postpartum).
Under this standardized
management system, data were
collected over a five-year period
(1986-1990). Data (390 records) were
analyzed with a statistical model that
included year, season of year, parity,

litter size and flock as main effects.
Parity was assigned into five categories
(1st lambing, 2nd lambing, 3rd
lambing, 4th-9th lambing and 10th-
12th lambing). Season of year was
divided into the periods of January to
April, May to August and September
to December, with lambing
concentrated in February/March, June/
July and October/November. Litter
size was classified as single and
multiple types of birth.

4 Under Caribbean
conditions, nutrition,
management and a lack
of genetic diversity are
the major constraints to
efficient production.

There were significant differences
in ewe and litter weaning weights
between years (Table 1). The most
pronounced change occurred from the
first to the second year of the study
with a general decline in ewe weaning
performance. This reduction is most
likely the result of a change in age
structure of the flock, which started as
a foundation flock of mature stock

Table 1. Variations between years in ewe body weight and Iamb Table 2. Variations between seasons in ewe body weight and lamb

Table 1. Variations between years in ewe body weight and lamb
production in hair sheep ewes on St. Croix.
Criteria Year
1 2 3 4 5
Ewe breeding wt. (Ib) 77.4 80.0 83.3 76.7 84.0
Ewe weaning wt. (Ib) 86.6 77.6 78.5 76.9 79.4
Lambs born/ewe lambing 1.75 1.63 1.60 1.60 1.77
Litter birth wt. (Ib) 11.5 10.8 10.4 10.6 11.2
Lamb survival to weaning (%) 85.5 87.8 88.9 86.2 80.6
Litter weaning wt. (Ib) 45.7 34.8 36.4 39.9 43.7

Table 2. Variations between seasons in ewe body weight and lamb
production in hair sheep ewes on St. Croix.

Criteria Season of Year
Jan-Apr May-Aug Sep-Dec
Ewe breeding wt. (Ib) 76.5 81.6 84.2
Ewe weaning wt. (Ib) 78.7 86.2 81.4
Lambs born/ewe lambing 1.65 1.66 1.69
Litter birth wt. (Ib) 10.5 11.2 10.9
Lamb survival to weaning (%) 87.1 85.9 77.1
Litter weaning wt. (Ib) 39.0 40.3 34.2

recruited from a number of local farms
into which younger replacement
animals were entered in subsequent
years. The younger replacement ewes
had not reached mature production
potential at this point. In the fifth year
of the study ewe weaning performance
had again reached first year levels and
may suggest an overall improvement
in flock performance as a result of
selection and/or changes in native
pasture quality through rotational
Season of year had a significant
impact on ewe performance at
weaning, but not at lambing (Table 2).
Ewe body weights were lower during
the period of January to April
compared to the other two seasons. In
contrast, survival to weaning was

reduced by 10% in lambs during
September to December, compared to
the remaining two seasons. The
reduced lamb survival resulted in a
decrease in litter size at weaning, from
1.45 lambs weaned/ewe lambing
during January to April and May to
August, to 1.31 lambs weaned/ewe
lambing in September to December,
and decreased litter weaning weight by
5-6 lb. The September to December
period generally experiences increased
rainfall and is associated with wet and
muddy conditions, which appears to be
the primary cause of the poor lamb
survival, despite the availability of lush
Ewe performance in relation to
parity is summarized in Table 3. Ewe
body weight at both breeding and

weaning increased gradually
throughout first to fourth parity and
remained stable thereafter. Ewe body
weight at weaning tended to decrease
in the older ewes following the ninth
parturition. Changes in litter size and
litter birth weight with increasing
parity mirrored those of ewe weaning
weights, with the number of lambs
born/ewe lambing increasing from 1.32
at first parturition to 1.83 at a mature
level and litter birth weight increasing
from 7.8 to 12.4 lb for the same stages.
Again, there was a decline in litter size
and birth weight as the ewes exceeded
their ninth parturition. The increase in
litter size at birth with increasing
parity is reflected in the incidence of
multiple births (Figure 1), shifting
from a frequency of 70% single lambs

Figure 1. The effect of parity on the distribution of birth type in
hair sheep ewes on St. Croix.
110 1

Figure 2. Ewe body weight at breeding and subsequent litter
weaning weight in ewes giving birth to singles, twins and triplets.


1st 2nd 3rd

4-9th 10-12th




Ewe wt.


Table 3. Effect of parity on ewe body weight and Iamb production in

Table 3. Effect of parity on ewe body weight and lamb production in
hair sheep ewes on St. Croix
Criteria Parity
1st 2nd 3rd 4th 5th
Ewe breeding wt. (Ib) 65.3 74.7 81.6 87.7 87.5
Ewe weaning wt (Ib) 71.2 77.6 81.6 87.7 85.3
Lambs born/ewe lambing 1.32 1.56 1.69 1.83 1.60
Litter birth wt (Ib) 7.8 10.0 10.7 12.4 10.7
Lamb survival to weaning (%) 86.6 82.8 87.1 81.6 73.3
Litter weaning wt. (Ib) 32.0 37.0 40.8 39.7 28.4

at first parity to 29% in mature ewes. The corresponding
increase in multiple births in the mature ewes was made up
of 53% twin and 12% triplet births. No triplet births were
recorded in ewes that exceeded their 9th parturition.
Parity in the younger and mature ewes had no effect on
lamb survival to weaning which ranged from 82 to 87%
(mean 84.5%). However, lamb survival declined by 11%
from this mean in the older ewes exceeding nine parturitions.
This decline in lamb survival in the older ewes is also
reflected in a litter weaning weight 11.5 lb lower than that
of mature ewes. Mature ewe weaning performance was
achieved with third parity (Table 3).
Ewe body weight at time of breeding had a significant
effect on subsequent litter size (Figure 2). To a large degree
this effect would be an indirect result of changes in age
and parity, particularly in the case of single vs. multiple
litters. However, differences in ewe body weight at breeding
between twin-bearing ewes (85.3 lb) compared to triplet-
bearing ewes (92.3 Ib) may indicate a direct effect of body
weight/condition on ovulation rate and subsequently litter

size. There was a positive relationship between litter size
and litter weaning weight (Figure 2), but differences
between single and twin litters were more pronounced (11.6
lb) than between twin and triplet litters (2.9 lb). In contrast,
other research at this station indicated that short term
supplementation immediately prior to breeding ("flushing")
had no effect on ovulation rate in hair sheep under these
production conditions.
The average ewe body weight at breeding and weaning
was 81 and 82 lb, respectively, when accounting for
differences in year, season of year and parity structure of
the flock. Similarly, overall flock lambing performance was
1.67 lambs born/ewe lambing, with an average litter birth
weight of 10.8 lb. Average ewe weaning performance was
1.40 lambs weaned/ewe lambing, with a total litter weaning
weight of 40.2 lb. The litter size at weaning reflected a
lamb survival of 83.7% from birth to weaning, with
approximately 4-5% of these lambs born dead and the
remainder dying during the period between birth and
The overall productivity of the flock was high,
considering the limited nutritional and managerial input,
and ewes weaned approximately 50% of their body weight
at breeding at nine weeks postpartum. The data further
indicate that ewes reached their mature production potential
following their third parturition and that productivity
declined after ten parturitions. The latter would be linked
to the relatively harsh production environment and the short
lambing interval of eight months, although other data from
this station indicate that the ewes resume cyclic activity
within one to two months postpartum and thus
reproductively support this production system.
There was very little evidence of seasonal effects on
productivity of the flock with the exception of poor lamb
survival to weaning during wet season
conditions. Ovulation rate and
rebreeding were not affected by season,
which is known to influence the
performance of sheep under more
temperate production environments.
Subsequent work at this station has
indicated some seasonal effects on the
onset of puberty and the length of the
postpartum interval in hair sheep under
the environmental conditions on St.
Croix, but these would not be
expressed under the production system
applied here.
The authors would like to thank Victor
Callas, Jr., Allan Schuster and Kim
Traugott for their assistance in the
collection of the data. This research was
supported in part by the U.S. Department
of Agriculture under Hatch Grant No. 787.








Problems in

the U.S.




In the U.S. Virgin Islands, papaya
(Carica papaya L.) is a unique tropical
fruit with good domestic and export
market potential to the U.S. mainland,
Canada, Europe and Japan. Excellent
potential for development exists locally
with an expanding tourist market.
There is also a certain amount of
versatility in the use of papaya. Besides
consumption as a fresh fruit, several
processed papaya products can be
developed, and papayas are used in both
the ripe and green forms in various
ethnic cuisines.
Since the early 1960s when a small
but ambitious program for the
production and export of papaya was
initiated on St. Croix, disease problems
have decimated production and
inhibited export of the fruit. This early
program, however, was based solely on

the Hawaiian Solo cultivar, which
performs well only within a narrow
ecological niche.
Major Disease Problems
The two primary diseases affecting
papaya in the U.S.V.I. are Papaya
Ringspot Virus (PRV) and Bacterial
Stem Canker (BSC). While the former
is probably of worldwide importance,
BSC has been identified only in certain
regions, most of which possess climatic
and soil conditions similar to those on
St. Croix. For example, while BSC is
prevalent on islands like Antigua which
have agroecosystems similar to St.
Croix's, the disease never occurs in
papaya plants of the geographically
closer but ecologically different island
of Puerto Rico. Soil type and rainfall
patterns of Puerto Rico differ greatly from
St. Croix's, without the consequent
environmental stress conditions.
Infected plants are characterized by
oily translucent spots or lesions on leaf
petioles or stem of the top third of the
plant. These enlarge or coalesce to form
a stem canker, weakening the stem
which may eventually become
decapitated. The incidence of BSC also
appears to be stress-related, particularly
drought and soil-nutrient stresses, with
certain cultivars such as the Hawaiian
Solo being more susceptible than the
Barbados Solo. The combination of
water stress and high soil pH-induced
nutrient deficiencies also make
otherwise resistant cultivars susceptible
to BSC.
Papaya Ringspot Virus is a more
ubiquitous disease problem and occurs

in ecosystems as diverse as those in the
Hawaiian islands to those in the U.S.
Virgin Islands. The major cause for this
widespread incidence is that PRV is
aphid-transmitted by the vector Aphis
gossypii. Several other agricultural crop
species, notably the cucurbits, are hosts
for PRV. The symptoms caused by PRV
include mottled and distorted leaves,
ring spots on the fruit, and water-soaked
streaks on stems and petioles.
Unlike BSC, the PRV-infected
papaya plant can often withstand or
sometimes even outgrow the effects of
the virus, depending on cultivar and
plant vigor (Cariflora cv. is an example).
In most cultivars, however, infected
plants have an unhealthy appearance,
gradually decline and produce infected
and blemished fruits with drastically
reduced market value. While one or two
harvests can sometimes be obtained
from PRV-infected plants, BSC usually
terminates all production and results in
stem collapse and death of the plant.
Control Strategies
In-depth research at UVI-AES in
1985-86 elucidated an Erwina sp. of
bacteria as the pathogen causing BSC.
Succulent-stem type cultivars such as
the Hawaii Sunrise Solo, when
cultivated under dry, windy and high pH
soil conditions, become easy targets for
this pathogen (cultivars such as the
Barbados Solo with a more woody stem
exhibit greater tolerance to BSC by
usually outgrowing initial leaf or stem
infections). Under these environmental
stress conditions, cellular and plant
tissue breakdown of succulent cultivars

in', !

GUL',ILn .


Table 1. Evaluation and rating of selected papaya cultivars under
an integrated production system.

Cultivar PRV1 BSC1 High Fruit Comments3
pH1 bearing2

Cariflora R T T E Early low-bearing,
short SL, good
processing potential.

Criolla R T T L Tall, late-bearing,
large non-uniform
fruits, good SL.

Guanica S T T M Drought tolerant,
prolific, pear-shaped,
pink-fleshed fruit.

PR6-65 FT T T M Pear-shaped, yellow-
fleshed firm fruit,
good SL.

PR6-65 X T T T M Pink-fleshed, both
Cariflora ovoid and elongate
fruits, good SL.

Solo SR X FT T T M Solo-shaped, pink-
Cariflora fleshed, prolific.

S-64 S T T E Low-bearing, very
large pear-shaped
fruit, few seeds.

Barbados FT T T M Pear-shaped firm
Solo fruit, good SL.

Palau S S S M Small, elongate firm
orange-fleshed fruit.

1. R = Resistant; T = Tolerant; FT =
2. E = Early; M = Medium; L = Late.
3. SL = Shelf Life.

Fairly tolerant; S = Susceptible.

occur more easily and rapidly, thus facilitating bacterial entry
and infection.
Research conducted by UVI-AES in 1986 on container-
grown papaya plants under greenhouse conditions
demonstrated conclusively that bacterial infections occurred
only when Erwina inoculum was sprayed onto bruised leaves.
This clearly indicated that plant injury, whether physically
or physiologically induced, was essential for the bacterial
infection causing BSC in papaya. Field trials conducted during
the same years indicated that polyculture or intercropping
papaya with other crops, such as pigeon pea, banana and
cassava, significantly reduced disease incidence.
More recent studies initiated in 1989 involving
intercropped banana and papaya cultivars adapted to local
soil and climatic conditions have shown no incidence of BSC.
Intercropping with plantain and the use of Moringa (horse
radish tree) windbreaks have also delayed and reduced the
incidence of PRV. These two sustainable cultural practices
apparently act by reducing actual wind-induced physical
damage to plants, filtering off bacterial inoculum and deterring
the buildup of virus-carrying aphids. These are in addition to
the physiological advantages of reducing ambient moisture
stress, thus decreasing excessive transpirational water loss
in papaya plants and increasing water use efficiency. In a
broad-leaved plant such as papaya, this is critical to
maintaining turgidity in leaves and essential for reducing
stomatal resistance and increasing photosynthetic activity.
An ongoing UVI-AES trial with ten papaya cultivars
for evaluating production methods has incorporated even more
sustainable cultural practices, including minumum tillage, a
permanent orchard sod cover, judicious use of herbicides and
Leucaena andMoringa windbreaks. After almost four years
of continuous production, no papaya trees have been lost
to BSC, and only in the last year has there been a gradual
buildup of PRV infection. The cultivars with the best
overall performance under such an integrated system are
included in Table 1.
While a system of intercropping with the use of
windbreaks and integration of certain cultural practices has
virtually eliminated the incidence of BSC in papaya in the
U.S.V.I. (Table 1), PRV continues to be a major problem.
The selection, breeding and evaluation of virus-tolerant
cultivars will continue to be major strategies for controlling
PRV. Recently, biotechnology has been used to genetically
engineer virus-resistant plants to the Hawaiian strain of PRV.
This offers an exciting new phase in plant protection, and,
to this end, the recently established biotechnology laboratory
at UVI-AES has as one of its major objectives to genetically
engineer papaya plants with resistance to the Caribbean strain
of PRV.

This research was funded in part by Hatch Project No.

t' .C *a~a~..
~ x, C` wt.
~j. ~ (, i~-'~tg: .
*C ~~ +, "'" ~$
;ZrF-;Cr" .~~

Table 1. Desmanthus virgatus accessions
subjected to clipping defoliation on St. Croix
VI accession Origin
1 St. Croix
2 Venezuela
3 Venezuela
4 Venezuela
5 Venezuela
6 Venezuela
7 Venezuela
9 St. Croix
11 Dominican Republic
12 Dominican Republic
14 Dominican Republic
15 Dominican Republic
16 Dominican Republic
17 Dominican Republic
18 Dominican Republic
20 St. Croix
22 Antigua
23 Venezuela
24 Venezuela
25 Venezuela
26 Mexico (CPI 92802)
27 Mexico (CPI 92803)
28 St. Croix


a Legume

for Forage


in the


Martin B. Adjei

In the Caribbean, predominant
tropical grasses such as guinea grass
(Panicum maximum) produce large
quantities of biomass that are usually
low in protein. Investments in fertilizer,
especially N, P, and K, can help improve
the protein status of tropical grasses.
However, tropical grasses seldom
exceed 10-12% protein levels, and these
levels usually occur at young
physiological ages when yields are very
low. Confined livestock feeding
operations, dairies and feedlots within
the region need feed stuff that is high in
protein. This need would be reduced if
high-yielding legumes adapted to the
environment were selected.
The genus Desmanthus, of the
Leguminosae subfamily Mimosoideae,
includes tropical species with a center
of origin in the Caribbean basin.
Ecotypes range from 2 m high D.
virgatus shrubs to the prostrate D.
depressus. They are adapted to alkaline
clay soils and occur naturally on many
Caribbean grassland sites where they
contribute to the diets of grazing
livestock. Desmanthus is characterized
by bipinnate leaves with leaflets less
than 10 mm in length. Although in the
same subfamily of legumes as
Leucaena, Desmanthus is a much
smaller plant with very little aggressive
tendency to invade and dominate an

entire pasture. Progress is being made
in the selection of Desmanthus varieties
with superior agronomic traits for
forage production at the Virgin Islands
Agricultural Experimental Station on St.
The initial screening ofDesmanthus
for their forage potential on St. Croix
began in 1988. It included 35 collections
of two species (D. virgatus and D.
depressus) representing a wide range in
area of origin and growth habit.
Seedlings were transplanted on 1 m
centers and evaluated as spaced plants.
Although most accessions grew
vigorously prior to defoliation, only five
accessions continued to maintain
vigorous growth following five
successive harvests at 90-day intervals.
In 1989, growth from the initial
planting was accumulated and subjected
to mob-grazing three times (April, mid-
July and late October) to gauge their
acceptability to sheep and response to
grazing. Herbage of the erect accessions
near shoulder high was grazed first.
Prostrate accessions were grazed later.
Animals readily and selectively grazed
leaves, immature stems and green pods.

Results obtained so far
indicate Desmanthus
to be an outstanding
forage crop with good
quality when grown under
natural Caribbean
conditions. ??

Two accessions-VI 26 (also
identified as CPI 92802 or CF 495) and
VI 27 (CPI 92803 or CF 543), both
originating from the Yucatan of
Mexico-were superior in initial
growth rating, herbage yield and final
persistence. Accessions VI 26 and VI
27 were also the most vigorous
following grazing.
From May, 1990, to March, 1992,
plots of 23 Desmanthus virgatus
accessions (Table 1), established from
reseeding of the original planting, were
further subjected to prolonged clipping

Figure 1. Mean annual total dry matter and edible dry leaf yields of
Desmanthus accessions from three harvests on St Croix (1990-92).
< 14. Leaf fraction
p 13 Stem fraction
-J 11
w 8
f 7
< 6
r 5
> 4
o 3

1 2 3 4 5 6 7 9 111214151617182022232425262728

Figure 2. Canopy ground coverage on 2 m2 plots of Desmanthus
accessions at the final (March, 1992) of six successive harvests on
St. Croix.
w 80
o 70
. 50
z 40
o 30
0 1 I I
1 2 3 4 5 6 7 9 1112 14151617182022232425262728

management to evaluate their forage
yield potential, forage quality and stand
persistence. This study was conducted
on Fredensborg silty clay loam which
had a pH of 7.8. There were three
replicates of each entry arranged in a
randomized complete block design. As
'a low-input trial, plots were neither
irrigated nor fertilized. Six successive
harvests (three per year) were removed
from plots (1 m x 2 m) whenever the
median of all accessions attained the
early-pod stage of maturity. Harvested
subsamples were separated into leaf
(including young stems and green pods)
and stem components, dried and
weighed. Selected accessions were
analyzed for crude protein (CP) and in
vitro organic matter digestibility

(IVOMD). Prior to the final harvest in
March, 1992, the canopy groundcover
for each plot was estimated with a
divided 1 m quadrat (100 divisions).
Accessions differed significantly in
total dry biomass and edible leaf forage
yield (Figure 1). Three accessions (VI
6, VI 7 and VI 23), originating from
Venezuela, had the highest mean annual
dry, leafy forage yield of 4-5 tons/ha.
Most of the remaining accessions
produced 3 tons/ha leafy forage, yearly,
except for VI 11 which yielded less than
2 tons/ha. However, the two earlier
selections from the Yucatan of Mexico,
VI 26 and VI 27, were among the most
persistent entries as determined by
canopy ground coverage of
approximately 95% at the final harvest

(Figure 2). Crude protein content of
Desmanthus leaves was consistently
high, ranging from 18 to 21% of dry
matter (Figure 3). However, crude
protein content of stems averaged only
8%. This suggests a necessity to select
and manage Desmanthus forage crop for
a higher leafy component in future
studies. The leaf IVOMD of all
accessions evaluated was above 80%
(Figure 4). The data indicate the
possibility to select accessions with a
potential production of 3 to 5 tons/ha
dry, edible leafy forage annually, when
Desmanthus is harvested at the early
pod stage of maturity. Such forage
should be highly nutritious to livestock,
containing approximately 20% CP and
80% IVOMD.
The results obtained so far on
Desmanthus indicate it to be an
outstanding forage crop with good
quality when grown under natural
Caribbean conditions. Our preliminary
feeding trials also suggest a more
favorable hair sheep growth response to
diets derived from grass/Desmanthus
combination than from grass alone.
Desmanthus is readily established
from seed that is commercially available
at the Caribbean Agricultural Research
& Development Institute (CARDI) in
Antigua. The initial seed multiplication
by CARDI was contracted jointly by the
University of the Virgin Islands and
University of Florida with Desmanthus
selections from St. Croix. Hard-
seededness is a characteristic of the
genus. The outcome of evaluation of
methods of scarification on St. Croix is

Figure 3. Crude protein as a percentage of dry matter in stems and leaves Figure 4. In vitro organic matter digestibility of stems and leaves of
of Desmanthus accessions on St Croix (1990-92). Desmanthus accessions on St. Croix (1990-92).
25 100loo
25 ILeaf O-Stem IMLeaf O-Stem
w 20 80
I- 70
) 15 60
z 5o 5 i

w 25 20

1 2 3 4 5 6 7 9 12 14 15 16 17 18 22 23 24 26 27 1 2 3 4 5 6 7 9 12 14 15 16 17 18 22 23 24 26 27

that Desmanthus seed should be soaked
in 70 oC water for one minute and then
dried to improve germination. Planting
should be done in well-prepared, moist
seed bed at a seeding rate of 6 kg/ha.
Good soil moisture availability during
the initial two post-planting months
ensures quick establishment and
minimum weed problems. Three annual
harvests at the early-pod stage of
maturity are recommended for cut-and-
carry forage management. Grazing
management strategies for sustained
Desmanthus pasture production are
currently being investigated on St. Croix.
A high level of interest in
Desmanthus forage for alkaline, clay
soils continues to be shown among
researchers and graziers in tropical
America, Australia and Africa. Present
collections of Desmanthus in
repositories include almost 300
accessions at CSIRO, Australia; over
150 accessions at CIAT, Colombia; and
smaller collections at ILCA, Ethiopia;
CARDI, Antigua; and the USDA.
Recently, three cultivars were
commercially released by Wright
Stephenson Seed Co. of Queensland,
Australia, at the 1993 International
Grassland Congress. Clipping and
grazing defoliation management
information being generated by UVI-
AES on St. Croix will contribute to
Caribbean forage programs.

This research was supported in part
by the U.S. Department of Agriculture
under Hatch Project No. 0220530.





with Drip

Irrigation in

the Virgin


Manuel C. Palada,
Stafford M.A.
and Charles D.

Culinary herbs are important
horticultural crops in the Virgin Islands.
An informal survey conducted in 1988
revealed that sales of herbs and spices
constitute a major source of income for
many small-scale growers in St. Thomas
and St. Croix.
Herb production in the Virgin Islands
mainly supplies the local market. A small
percentage is exported to the U.S., while
most of the herbs consumed in the U.S.
are imported from the Mediterranean
region, Africa and Latin America.
Improving field production of herbs in the
Virgin Islands will increase its potential for
the export market and at the same time
improve the local economy.
Although most culinary herb species
and varieties are easily adapted to the
tropical climate of the Virgin Islands,
field production is constrained by
environmental factors and management

practices. Water is a major limiting
resource for agriculture in the Virgin
Islands, and without supplemental
irrigation, commercial production of
high value crops such as herbs is not
economically feasible. The traditional
practice of growing herbs in the Virgin
Islands requires frequent watering with
sprinkler cans and garden hoses. This
practice is inefficient and involves high
labor and water use.
Drip irrigation was first introduced
to the Virgin Islands in the early 1980s.
Studies at UVI-AES indicated that drip
irrigation has more advantages compared
to the traditional practice of supplying
water to the crops, in terms of labor,
water use, efficiency and cost. Many
growers are now considering a drip
irrigation system an essential component
of their vegetable production enterprise.
At UVI-AES, a research project was
initiated in 1988 to improve field
production of herbs through efficient
water use and optimum fertilizer
application. The application of drip
irrigation technology was subsequently
extended to herb production throughout
the Virgin Islands. This report presents
the results of drip irrigation experiments
with basil (Ocimum basilicum) and
thyme (Thymus vulgaris).
Two field experiments were
conducted to 1) determine the response
of thyme to irrigation methods and rates
and 2) determine the water requirements
of thyme. In the first experiment, thyme
seedlings were started in seedling trays
and grown in the greenhouse in March,
1991. At 45 days after sowing, seedlings

were transplanted in field plots that were
9.4 feet long and 7.4 feet wide. Each plot
contained three rows 30 inches apart.
Seedlings were planted at a spacing of
12 inches in the row.
Irrigation treatments consisted of
two methods (drip versus micro-
sprinkler) and two rates based on pan
evaporation (40% PE versus 60% PE).
Pan evaporation is the amount of water
that is lost through evaporation from a
standard Class A circular pan measuring
46.5 inches in diameter and 10 inches
deep. Half of the plots in treatment under
drip irrigation were covered with 1 mil
(0.025 mm) black plastic mulch. None
of the plots under micro-sprinkler
irrigation were mulched.
The drip irrigation system
consisted of main and submain lines
made of 0.6-inch (15 mm) polyethylene
hose, and the laterals were made of 0.6-
inch (15 mm) drip strip with laser-
drilled orifices 12 inches apart. The
micro-sprinkler system consisted of
0.6-inch polyethylene hose with micro-
sprinklers attached to the laterals in
center rows. The experiment was
arranged in a randomized complete
block design with six treatments and
three replications. The experiment was
conducted from March to September,
Data presented in Table 1 show that
under drip irrigation, fresh yield of thyme
in plots without plastic mulch was higher
than with mulch. The difference in yield
between the two rates of irrigation (40%
PE vs. 60% PE) was not statistically
significant either in mulch or no mulch

Table 1. Yield of thyme under two methods and rates of irrigation.
UVI-AES, 1990.

Irrigation Irrigation Rate Mulch Fresh leaf yield1
system (% PE2) (t/ac)

Drip 40 No 3.57a
Drip 40 Yes 1.34c
Drip 60 No 3.75a
Drip 60 Yes 2.32bc
Sprinkler 40 No 2.05bc
Sprinkler 60 No 3.04ab

'Values within a column followed by the same letters are not significantly
different by Duncan's Multiple Range Test, (P<0.05).
2Pan evaporation.
Table 2. Plant height, total plant fresh and dry matter yield of
thyme grown at three levels of drip irrigation with and without
mulch. UVI-AES, 1992.

Irrigation Mulch Plant Plant Yield1
level height1 Fresh Dry
(cb) (in) (Ib/ac) (Ib/ac)

20 No 8.03a 4436a 1432a
40 No 7.87a 2647b 884b
60 No 7.60ab 3121b 1044ab
20 Yes 5.98b 2013bc 597bc
40 Yes 6.42ab 1799bc 571 bc
60 Yes 6.73ab 963c 296c
Rainfed No 6.57ab 2479b 876b

'Values within a column followed by the same letters are
different by Duncan's Multiple Range Test, (P<0.05).

not significantly

Table 3. Estimated irrigation water use and efficiency of thyme at
three levels of irrigation with and without mulch. UVI-AES, 1992.

level Mulch Total water use WUE1
(cb) (gal/plt) (ft3/ac) (Ib/ft3)

20 No 1.55 7036 0.63
40 No 1.19 5406 0.49
60 No 0.48 2160 1.44
20 Yes 1.55 7036 0.29
40 Yes 1.19 5406 0.33
60 Yes 0.48 2160 0.45
Rainfed No 4.332 196212 0.132
'WUE= water use efficiency in pounds dry matter per cu. ft.
2Represents amount of rainfall based (1 inch rain = 27,000 gallons/acre).

Table 4. Estimated cost of irrigation water and efficiency in thyme
production with drip irrigation. UVI-AES, 1992.

Irrigation Irrigation Returns to
level Mulch water cost' irr. water2
(cb) ($/ac) ($/$)

20 No 844 70.62
40 No 649 54.80
60 No 259 161.93
20 Yes 844 32.04
40 Yes 649 37.57
60 Yes 259 44.95

'Based on water cost of $0.12/cu.ft.
2Dollar return to every dollar spent in irrigation water.

treatment. Under micro-sprinkler irrigation, fresh yield of
thyme at an irrigation rate of 60% PE was numerically higher
than at 40% PE. However, the yield at 60% PE was similar
with those obtained from drip-irrigatedtreatments without mulch.
The lower yields from drip-irrigated plots with black
plastic mulch and those under micro-sprinkler irrigation were
attributed to the high incidence of fungal diseases. Wet soil
in plots with plastic mulch and frequent wetting of foliage in
sprinkler-irrigated plots may have contributed to favorable
development of soil-borne fungus diseases including
Phytophthora sp., Phythium sp., and Rhizoctonia solani.
Results of this experiment indicate that improved yield
of thyme was achieved under drip irrigation without mulch.
This method has the added advantage of allowing growers to
cultivate and mound the soil around the plant for better plant
growth. The micro-sprinkler method is not only inefficient, but
also requires high volumes of irrigation water since part of the
water applied is lost through evaporation and spray drift.
The second experiment was conducted to determine the
irrigation water use and requirement of thyme with and
without mulch. The treatments consisted of three irrigation
levels corresponding to soil moisture tension of 20, 40 and
60 in centibars (cb). A centibar is a unit used to measure the
amount of pressure by which the soil particles hold water or
moisture. The higher the pressure the lower the amount of
water or moisture available to plants. Soil moisture tension
is measured by using a device called a soil tensiometer. Half
of the plots were mulched using a black plastic, water-
permeable weed barrier ground cover. A rainfed control (no
mulch) was included as a treatment.
Thyme seedlings were transplanted on March 20, 1992,
in three-row plots arranged in a randomized complete block
design with four replications. Plot size measured 3 feet x 8
feet. Plants were spaced 12 inches between rows and 8 inches
between plants within rows. The plants were fertilized with
nitrogen, phosphorus and potassium at the rate of 45, 90 and
90 lb/ac, respectively.
The fertilizer was applied one week after transplanting.
The drip irrigation system installed in plots was similar to
that in the first experiment. Soil tensiometers were installed
at a 6-inch depth in each treatment on two replications to
monitor soil moisture tension. The tensiometers were read
daily and irrigation was turned on when readings were above
a prescribed centibar treatment. A water meter and timer were
also installed for each treatment. Data were collected on plant
height at harvest, plant fresh weight yield and dry matter
weight. Total irrigation water use was determined over a
period of nine weeks.
High rainfall during the growing season confounded the
results of the trial. Over a period of three months (April to
June) total rainfall was 13 inches (Figure 1). May was the
wettest month with rainfall of 11 inches. Despite the high
rainfall, differences in yield among treatments were significant
(Table 2). Total plant fresh yield was highest (4436 lb/ac)

Figure 1. Rainfall and potential evapotranspiration at UVI-AES,
Rainfall, inches I Pan ET, inches

10 -


Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Table 5. Plant fresh yield of drip-irrigated basil grown with organic
and synthetic mulches. UVI-AES, 1991.

Mulch 1st 2nd 3rd 4th Total
(yield, t/ac)1

Compost 2.2 5.2 10.5a 10.4a 28.4a
Grass straw 1.8 3.8 5.7b 8.2ab 19.6b
Black plastic 2.5 5.3 9.2ab 7.lab 24.2ab
Weed barrier 2.4 5.0 6.2b 4.6b 18.3b
Control (no mulch) 1.9 4.8 8.4ab 5.0b 20.3b
'Values within each column followed by the same letter are not significantly
different by Duncan's Multiple Range Test, (P<0.05).

from the treatment with an irrigation
regime maintained at 20 cb without
mulch. Dry matter yield was also highest
for this treatment, but not significantly
different from the treatment under an
irrigation regime of 60 cb. Generally, all
treatments with mulch and the control
(rainfed) had lower yields than irrigation
treatments without mulch. This result is
consistent with those obtained in the first
experiment and would indicate that
mulching is not beneficial for thyme.
Furthermore, when soil moisture is not
a limiting factor because of high rainfall,
mulching is no longer an advantage.
Total irrigation water use was
highest in treatments with an irrigation
regime of 20 cb (Table 3). Treatments
under 60 cb had the least water use.
However, only the treatment at 60 cb
without mulch resulted in more efficient
use of irrigation water at the least cost
(Table 4). This result would suggest that
under high rainfall, supplemental
irrigation can be reduced by maintaining
the soil moisture at 60 cb. At this level
the minimum water requirement ofthyme
can be met. However, during seasons of
low rainfall this level may change.
Although thyme does not respond
favorably to mulching as shown in the
two experiments, integrating drip
irrigation with mulching can further
conserve water by reducing water use,
evapotranspiration and weed control and
by increasing overall water use efficiency
for other crops. The choice of mulching
materials may also influence the
response of herb species to drip

In the summer of 1991 an experiment
was initiated to 1) determine the effect
of various mulching materials on yield
and water use of basil under drip
irrigation and 2) compare water use
efficiency of basil grown with organic
and synthetic mulches.
Basil was grown in plots with
organic (compost or grass straw) and
synthetic (black plastic or weed barrier
ground cover) mulches. The compost
was prepared using a 1:1 ratio of sheep
manure and vegetable crop residues. The
grass straw was made of dry pangola
grass. The compost and grass straw
mulches were applied at 2 inches thick.
A control plot (no mulch) was also
included as a treatment. Plot size was 4
feet x 7 feet with rows 16 inches apart.
Plants were planted at 12 inches spacing
within rows. The experiment was

arranged in a randomized complete block
design with four replications. All plots
were drip-irrigated to maintain soil
moisture at 30 cb. The drip irrigation
system installed was similar to that of
the 1991 experiment on thyme. The
amount of irrigation water used was
monitored with water meters installed for
each treatment.
Data on Table 5 show that plant
fresh yield did not vary significantly
among treatments during the first and
second harvests. Differences in yield
were observed in the third and fourth
harvests. At the third harvest, basil
grown with compost mulch produced the
highest yield of almost 10.5 t/ac. This
was followed by plants with black plastic
mulch and the control. Basil grown with
grass straw and weed barrier ground
cover produced lower yields. At the


Table 6. Irrigation water use of basil grown with organic and
synthetic mulches. UVI-AES, 1991.

Mulch Total fresh yield Total water use WUE1
(t/ac) (ft3/ac) (Ib/ft3)

Compost 28.4 22719 1.25
Grass straw 19.6 21091 0.93
Black plastic 24.2 24877 0.97
Weed barrier 18.3 18390 1.00
Control 20.3 21634 0.94
'WUE = water use efficiency (Ib fresh basil per cu. ft. water).

"Although basil grown with compost
mulch used more water than other
treatments, yield and water use
efficiency were the highest, resulting
in higher gross returns.

fourth harvest, basil grown with compost mulch maintained
the highest yield. Total yields of basil grown with compost
mulch were significantly higher than the other treatments with
the exception of the black plastic mulch.
Differences observed in plant fresh yield were also
reflected in leaf fresh yields. The treatment with compost
mulch produced the highest fresh leaf yield, followed by the
treatment with black plastic mulch (Figure 2).
Total water use was highest in plots with black plastic
mulch (15% more than the control), followed by compost
mulch which was 5% more than the control (Table 6). Grass
straw and weed barrier ground cover reduced water use by 3
and 15%, respectively.
The high water use in plots with black plastic mulch may
be associated with high soil temperature (data not shown)
which may have induced rapid loss of soil moisture. In
compost-mulched plots large water use may be associated
with high water absorbing capacity of compost material which
acted like a sponge. Low irrigation water use in the control

Figure 2. Fresh leaf yield of basil grown with organic and synthetic
mulches. UVI-AES, 1991.
7 Harvest 1st
6 ~ 3rd
_ 5 4th
Co 4 -
0 3
2 -
Compost Straw Black Plastic Weed Barrier Control

Table 7. Yield and economic returns from drip-irrigated basil grown
with organic and synthetic mulches. UVI-AES, 1991.

Mulch Total yield Gross value1 Irr. water Returns to
(t/ac) ($/acx1000) cost2 irr. water3
($/ac) ($/$)

Compost 28.4 113.6 2726 41.6
Grass straw 19.6 78.4 2531 31.0
Black plastic 24.2 96.8 2985 32.4
Weed barrier 18.3 73.2 2207 33.2
Control 20.3 81.2 2596 31.3
1Fresh basil at $4.00/lb.
irrigation water at $0.12/cu.ft.
3Dollar return for every dollar spent on irrigation water.

plots can be explained by the addition of rainfall which totalled
13 inches during the five-month growing season (Marchto July).
The higher yield from the treatment with compost mulch
can be partly explained by possible leaching of nutrients from
compost down to the root zone. During decomposition of
compost materials, nutrients are released slowly and become
available to plants. Nutrient contribution from the compost
material may have increased towards the latter part of the
growing season as shown by the higher yield of the final two
Although basil grown with compost mulch used relatively
more water than other treatments, yield and water use
efficiency were also the highest. This resulted in higher gross
returns and returns on irrigation water (Table 7). At a $4.00/
lb wholesale price of fresh basil, treatment with compost
mulch gave the highest gross return of $114,000/ac. If herb
growers have to pay for irrigation water at $0.12/cu ft., using
compost mulch will give them the highest returns on irrigation
water of $42 for every dollar spent.
This study has shown that compost mulch increased basil
yield, improved water use efficiency and boosted economic
returns to irrigation water. Integrating compost mulch with
drip irrigation for basil production offers some economic
benefit for small-scale herb growers in the Virgin Islands.
Organic mulches such as compost and grass straw are locally
available, thus saving herb growers the cost of importing
expensive synthetic mulches.
The only drawback of using organic mulches is the
potential weed problem associated with either compost or
grass straw material. If the mulch is made from materials
containing viable weed seeds, weed growth can be a problem.
The results of experiments onthyme and basil also suggest
that the response of herb species to mulching is variable. Basil
responded favorably to mulching while thyme did not.
Additional studies are being carried out at UVI-AES to
investigate the response of other herb species to drip irrigation
and mulching.

This research was supported in part by Hatch Regional
Project No. 247 and CBAG Project No. 472.


of Tilapia

Species for


Culture in

the Virgin


James E. Rakocy,
John A. Hargreaves
and Donald S. Bailey

Tilapia is a tropical fish native to
Africa and the Middle East. Since tilapia
is a hardy fish that feeds low on the food
chain and is relatively easy to culture, it
has been distributed widely throughout
tropical countries to increase protein
supplies. Tilapia is also being cultured
in many temperate countries in heated
or geothermal water because its
delicious taste and excellent nutritional
qualities make it an attractive food item.
At least 77 species oftilapia are found
in the African continent, but only nine are
considered to be suitable for aquaculture.
These species differ in a number of
important culture traits. For example,
Oreochromis aureus is the most cold-
tolerant (tilapia generally die at
temperatures below 500F) while O.
mossambicus is tolerantto high salinities.
Aquaculturists have increased the
diversity within tilapia genera by crossing
species to create hybrids or through
programs of selective breeding to develop
new strains. Many new strains have
resulted from accidental crossing
(contamination) of pure species. Strains
also develop naturally within a species

when there is geological isolation. An
example of significance to aquaculture
is the occurrence of three distinct
species ofO. niloticus, the Ghana, Ivory
Coast and Egypt strains, with varying
culture characteristics such as cold
tolerance and growth rate.
One major breakthrough in the
development of new strains was the
appearance of a red mutant O.
mossambicus in Taiwan in 1968. This
single fish was crossed with 0. niloticus
to begin a strain of red tilapia, which led
to a second wave of tilapia dispersal to
tilapia-culturing countries where the red
strain was crossed with other species to
create new strains of red tilapia, some
containing genes from as many as four
species. Selective breeding of red tilapia
strains has improved their appearance
from an initial mottled look with blotches
of dark wild color to a blend ofred, orange,
pink and white. Red tilapia now resemble
colorful ocean fish such as red snapper,
which increases their marketability.
Wide variability among cultured
tilapia species and strains causes
uncertainty during stock selection.
Farmers want stock that exhibit fast
growth, high survival, efficient feed
conversion, attractive coloration and good
body conformation under actual culture
conditions. Other desirable traits include
resistance to disease, high dressing

percentage and high quality flesh. To aid
perspective Virgin Islands fish farmers in
selecting the best available stock, two
species comparison experiments were
conducted at the University of the Virgin
Islands Agricultural Experiment Station
(UVI-AES). In experiment 1, the
performance of 0. niloticus, 0. aureus
and three strains of red tilapia were
evaluated in freshwater cages. In
experiment 2, the performance of O.
mossambicus and "Florida" red tilapia
were evaluated in saltwater cages.
In experiment 1, tilapia stocks came
from the following sources:
1) Oreochromis niloticus (Ivory Coast
strain). The stock of O. niloticus originated
in West Africa and was shipped to Auburn
University via Brazil in 1974. Fish were
subsequently shipped from Auburn
University to the Tennessee Valley
Authority in Muscle Shoals, Alabama, and
from there to UVI-AES in 1981.
2) Oreochromis aureus. The stock of O.
aureus originated in Israel from which fish
were shipped to Auburn University in
1957. Fish were subsequently shipped from
Auburn University to the University of
Puerto Rico, Mayaguez, and from there to
UVI-AES in 1974.
3) "Florida" red tilapia (FRT). FRT was
developed in the late 1970s at Natural
Systems in Palmetto, Florida, by
crossing wild (black) 0. urolepis

Figure 1. Average growth rate of 0. niloticus, 0. aureus, and "Florida," "Aurea" and
"Taiwan" red tilapia cultured in 1-m3 cages for 143 days.

hornorum females with red 0.
mossambicus males to produce F, (first
generation) hybrids exhibiting variable
skin coloration ranging from white to
red/orange with melanistic (black)
mottling. After an extensive selective
breeding program, initially at South
Florida Fisheries in Lantana, Florida,
and later by Sunfish Hatcheries in
Jamaica, F15 or F, fish were obtained
by UVI-AES in 1986 from the Jamaican
hatchery and used in experiment 1.
4) "Aurea" red tilapia (ART). Male F1
"Florida" red hybrids (0. urolepis
hornorum X 0. mossambicus) were
crossed with female 0. niloticus and
female 0. aureus in the early 1980s at
Tennessee Valley Authority. The red
progeny of these crosses were crossed
again to form a "four-way line"
containing 25% of the genetic material
of each of the parent species. South
Florida Fisheries crossed wild 0. aureus
females with four-way line males to
produce F2 ART. This process was
repeated until an F4 ART was obtained
that consisted of 95% 0. aureus
heritage. UVI-AES received F4 ART
from Sunfish Hatcheries in 1986 and
used them in experiment 1.
5) "Taiwan" red tilapia (TRT). This strain
was developed in Taiwan from 1968-1972
by crossing red female 0. mossambicus
with male 0. niloticus. In 1983, the
University of Puerto Rico provided UVI-
AES with TRT which were derived from
a founder stock of 40 fish from Panama.
The freshwater cage culture study was
conducted in Chimney Bush pond, a five-
acre watershed pond on St. Croix, in 1-m3

(35ft3) cages. The cages were 3.5 feet in
diameter, four feet deep and made from
semi-rigid plastic screen with 0.75-inch
mesh. The cages were individually
anchored to concrete blocks and placed in
six rows of five cages each. The rows of
cages were perpendicular to the pond
bank and roughly parallel with the
prevailing wind direction (east,
northeast). Cages were placed
approximately six feet apart within rows
and 20 feet apart between rows.
A demand (self) feeder was secured
to the top of each cage. The feeders were
checked every day at 10:00 a.m. and 3:00
p.m. and refilled when empty with 12 lb
of floating, pelleted feed containing a
nutritionally-complete diet (36% protein,
Purina #5144).

The fingerlings in this study were fed
a male sex hormone as fry to convert the
female fry to males. Sex-reversed male
fingerlings of each tilapia group were
stocked at a rate of 300 fish per cage into
six cages. Tilapia groups were assigned
to cages according to a randomized
complete block design with each row of
cages representing a block. Three of the
30 cages were lost during the
experiment due to the theft of one cage
from each red strain. At intervals of 28
feeding days, approximately 20% of the
fish from each cage were weighed,
counted and returned to the cage. The
experiment was terminated after 143
feeding days, at which time all of the
fish were weighed and counted.
During the harvest, ten fish were

selected from each tilapia group and
weighed individually to the nearest gram.
Scales, gills and viscera were removed
from each fish, which was then reweighed
to determine dressing percentage with head.
After removing the head, the carcass was
weighed again to determine dressing
percentage without head.
Ten additional fish were selected from
each tilapia group and weighed
individually to the nearest gram. After
scaling the fish, two fillets were excised
from each fish and weighed. The fillets
were reweighed following removal of the
skin. The percentage ofthe fillet yield with
and without skin was calculated.
A gross proximate analysis of body
composition was conducted on three
individuals of each tilapia group.
Moisture was determined by drying the
fish in an oven. Crude protein was
determined by Kjeldahl-sulfuric acid
digestion. Lipid (fat) was determined by
the modified Babcock method.
Results of this study showed that O.
niloticus was the fastest-growing species
(3.77g/day) and grew to the largest size
(616 g, 1.37 lbs.) within the 20-week
culture period (Figure 1, Table 1). Among
the red varieties, the "Florida" red tilapia
grew at the highest rate (3.50 g/day) to
the largest size (577 g, 1.27 lbs.). As cage
biomass increased, growth rates declined
for TRT after week 8, for 0. aureus after
week 12, and for FRT and ART after week
16 (Figure 1). Although 0. niloticus grew
at its highest rate from week 4 through 8,
it continued to grow at a steady and
substantial rate through week 20.
Net production reflected the growth
rate with 0. niloticus ranking first and
FRT ranking second (Table 2). Total
production ranged from 119 kg (262
lbs.)/m3 (35ft3) for 0. aureus to 182 kg
(400 lbs.)/m3 (35ft3) for 0. niloticus.
There was little difference in total
production per cage between FRT (163
kg, 359 lbs.) and ART (161 kg, 354 lbs.).
Survival ranged from 93.3% for FRT
to 98.4% for 0. aureus.
The feed conversion ratio (FCR),
calculated by dividing feed weight by fish
weight gain, ranged from 1.29 for 0.
aureus to 1.68 for TRT (Table 2). This

Table 1. Average initial and final body weight, daily weight gain and specific growth
rate of 0. niloticus, 0. aureus and "Florida," "Aurea," and "Taiwan" red tilapia cultured
in 1.0-m3 cages for 143 days.

Tilapia Number Initial Final Daily Specific
group of cages body weight body weight weight gain growth rate
(g/fish) (g/fish) (g) (%/day)

0. niloticus 6 73c1 616a2 3.77a2 1.45a2
0. aureus 6 77b 403d 2.28d 1.16d
"Florida" red 5 70d 577b 3.50b 1.41ab
"Aurea" red 5 83a 551b 3.31b 1.38b
"Taiwan" red 5 85a 470c 2.75c 1.27c

'Average values within a column followed by the same letter are not significantly different (P>0.05)
as determined by pairwise t-tests.
2These average values are "least square means," the values that would have occurred if the initial
body weights were equal.

Table 2. Average initial and final fish biomass, net production, feed conversion ratio and
survival rate of 0. niloticus, 0. aureus and "Florida," "Aurea," and "Taiwan" red tilapia
cultured in 1.0-m3 cages for 143 days.

Tilapia Number Initial Final Net Feed Survival
group of cages biomass biomass production conversion rate
(kg/m3) (kg/m3) (kg/m3) ratio (%)

0. niloticus 6 22c1 182a2 159a2 1.30c2 97.7a2,3
0. aureus 6 23b 119c 96c 1.29c 98.4a
"Florida" red 5 21d 163b 140b 1.45b 93.3b
"Aurea" red 5 25a 161b 138b 1.31bc 98.3a
"Taiwan" red 5 25a 132c 109c 1.68a 94.9ab

'Average values within a column followed by the same letter are not significantly different (P>0.05)
as determined by pairwise t-tests.
2These average values are "least square means," the values that would have occurred if the initial
body weights were equal.
3Survival growth rates were arcsine-transformed prior to analysis.

Table 3. Average condition factor (K) and dressing percentage of 0. niloticus, 0. aureus
and "Florida," "Aurea," and "Taiwan" red tilapia1.

Tilapia K2 Dressed Dressed Fillet Fillet
group with head without head with skin without skin
(%) (%) (%) (%)

0. niloticus 3.11b3 85.6a4 65.6a4 39.0a4 33.4a4
0. aureus 3.00c 85.2a 64.7a 38.4a 33.8a
"Florida" red 3.21a 84.9a 64.0bc 38.6ab 34.3a
"Aurea" red 2.87d 85.6a 64.3ab 38.3a 33.5a
"Taiwan" red 2.85d 86.0a 63.4c 36.4b 31.1b

1Ten individuals were analyzed from each group for dressing percentage with and without head; ten
individuals were analyzed from each group for dressing percentage of fillet with and without skin.
2K = body weight (g) x 105/total length (mm)3.
3Average values within a column followed by the same letter are not significantly different (P>0.05)
as determined by pairwise t-tests.
4These average values are "least square means," the values that would have occurred if the initial
body weights were equal to 456g for dressed with and without head and 571g for fillet with and
without skin. All percentage data was arcsine-transformed prior to analysis.

Table 4. Average moisture, protein and lipid content of 0. niloticus,
0. aureus and "Florida," "Aurea," and "Taiwan" red tilapia'.

Tilapia Moisture Protein Lipid
group (%) (%) (%)

0. niloticus 78.2a2 18.5cd 1.4b
0. aureus 75.3c 21.0a 1.4b
"Florida" red 76.4bc 20.1b 3.6a
"Aurea" red 78.0ab 17.7d 1.2b
S "Taiwan" red 78.5a 19.1c 1.2b

'Three individuals were analyzed from each group.
2Average values within a column followed by the same letter are not
significantly different (P>0.05) as determined by pairwise t-tests.

means that 1.29 lb of feed were required
to produce a one-pound weight gain in O.
niloticus compared to 1.68 lb of feed for a
one-pound gain in TRT. The FCR values
of 0. niloticus (1.30) andART (1.31) were
exceptionally low for such large fish. The
use of demand feeders appears to be a very
efficient method of feeding caged
tilapia.The FCR of FRT was numerically
intermediate (1.45).
The condition factor is a number that
relates weight to length and indicates how
robust (or emaciated) a fish is. FRT
displayed the highest (best) condition
factor (3.21), followed by 0. niloticus
(3.11) (Table 3). The lowest condition
factors occurred among TRT (2.85) and
ART (2.87). These strains were visibly
thinner than the other tilapia groups, and
approximately 15% of TRT exhibited
some degree of caudal fin rot.
The dressing percentage with head
(scales, gills and viscera removed) was
similar for all the tilapia groups, ranging
from 84.9% for FRT to 86.0% for TRT
(Table 3). With the removal of the head,
some statistically significant changes
occurred in the ranking. TRT displayed
the lowest dressing percentage (63.4%),
indicating that its head size was relatively
large compared to its body, while O.
niloticus exhibited the highest dressing
percentage (65.6%), suggesting that its
head size was relatively small. Small head
size is a desirable trait because meat yield
increases as head size decreases. This
relationship was confirmed bythe results
of fillet yield with skin, which was lowest
for TRT (36.4%) and highest for 0.
niloticus (39.0%). With the removal of

skin, 0. niloticus dropped to fourth in the
ranking of fillet yield (33.4%), indicating
that its skin may be thicker than the skin
of the other tilapia groups. The fillet skin
ranged from 4.3% (FRT) to 5.6% (0.
niloticus) of total body weight. The
differences in dressing percentages among
the tilapia groups were relatively minor
except for those of TRT, which were
significantly lower in most categories.
Gross proximate analysis of body
composition revealed that from 75.3% (0.
aureus) to 78.5% (TRT) of total body
weight consisted of moisture (Table 4).
Protein levels ranged from 17.7% (ART)
to 21.0% (0. aureus). Lipid content,
which was similar among four of the
groups (1.2-1.4%), was significantly
higher (3.6%) for FRT. Fat deposits were
visible in the flesh of FRT and in a thick
layer over the swim bladder. FRT
registered the highest condition factor
(Table 3), another indication that it was
the fattest fish. FRT may require a
restricted feeding regime or a diet with a
lower energy-to-protein ratio to reduce fat
content. The most nutritious species was
0. aureus, which registered the lowest
moisture content, the highest protein level
and a moderate lipid level.
By nearly all measures, the culture
performance of O. niloticus was superior
to the other tilapia groups. Dressing
characteristics of 0. niloticus were similar
to or better than the other groups, although
protein levels were comparatively low. If
color is not a factor in marketing tilapia
products, such as skinless fillets, 0.
niloticus is the best species for freshwater,
cage culture in the Virgin Islands. After

completing this study, UVI-AES replaced
the Ivory Coast strain of 0. niloticus with
the Egypt strain, which grows faster
according to a recent study. The other pure
species in this study, 0. aureus, grew at a
low rate and should not be considered for
culture, although all other growth and
dressing characteristics were favorable.
Marketing studies have shown that
Virgin Islanders prefer red tilapia because
they resemble brightly-colored ocean fish.
Among the red strains, FRT performed
best in most categories, except survival,
feed conversion and fat content. ART
performed nearly was well as FRT, and
most of the differences between ART and
FRT were not statistically significant.
However, production of F4 ART
fingerlings involves the maintenance of
separate breeding lines and is much more
complicated than using a single FRT strain
to generate seed stock. Since ART offers
no clear growth or dressout advantage,
FRT is the recommended strain of red
tilapia for the Virgin Islands. TRT
performed poorly in growth, feed
conversion and meat yield and is therefore
unsuitable for the Virgin Islands.
In experiment 2, 0. mossambicus
were derived from stocks captured in the
Salton Sea in California and transported
to the University of Puerto Rico and from
there to UVI-AES in 1988. FRT came
from the same source as described in
experiment 1.
The study was conducted on St.
Croix in Salt River Bay, a protected
embayment which is open to the sea and
receives freshwater runoff only during
periods of intense rainfall. During the

study, salinity ranged from 35 to 38 parts per thousand (ppt).
Three cross-shaped rafts (wooden walkways) were anchored
in approximately nine feet of water at a distance of 30 feet from
a fringe of red mangrove in a small arm of the bay. Two 2-m3
(70ft3) cages were secured to each of the three rafts. The
cages were constructed of galvanized mesh (0.75-inch) formed
into a rectangular shape measuring five feet long by four feet
wide by four feet deep. The mesh was treated with a blue
antifouling paint containing tributyltin fluoride.
A feeding ring was placed in each cage. Initially, 0.12-
inch plastic mesh was tied to rubber inner tubes. Many inner
tubes deflated due to degradation of the rubber by salt water
and sunlight. These were replaced by encapsulated
polystyrene rings measuring 16 inches (inner diameter) by
24 inches (outer diameter) by six inches (height). The same
0.12-inch plastic mesh was tied to the outside of the rings
and extended 12 inches into the water.
Sex-reversed male fingerlings of each tilapia group were
stocked at a rate of 40 fish per cage (20 fish/m3, 35ft3) into
three cages, one cage from each raft. The cages were stocked
at low rates due to a shortage of fingerlings and the logistics of
acclimating the fingerlings to seawater. The fingerlings had been
raised in fresh water and were acclimated to seawater over a
five-day period by direct transfer to aerated tanks of increasing
salinity: 15 ppt (day 1), 20 ppt (day 2), 25 ppt (day 3), 30 ppt
(day 4), and 35 ppt (day 5).
The fish were fed a floating, nutritionally complete diet
(36% protein, Purina #5144) at 3.0% of body weight daily once
per day in the morning. The feeding rate was adjusted weekly
based on an assumed feed conversion ratio of 2. The fish were
not sampled during the experiment to avoid injury and
subsequent infection, as tilapia are not as resistant to disease in
full-strength seawater as they are in fresh water. After 50 feeding
days, the fish began to show signs of bacterial infection which
led to low rates of mortality. After 64 days, feed treated with
oxytetracycline (Terramycin) was offered to the fish at a
rate of 80 mg/kg of fish body weight for 14 days. Fish
resumed active feeding following treatment with medicated
feed, although feed quantity was restricted to that which
could be consumed within a short period. The experiment
was terminated after 112 feeding days, at which time all of
the fish from each cage were weighed and counted.
The culture performance of FRT was superior to that of O.
mossambicus, indicating that FRT is more suitable for culture
in marine cages (Table 5). FRT grew at a faster rate (2.56g/
day) and reached a larger size (439 g, 0.97 lbs.) than O.
mossambicus (2.08 g/day and 381 g, 0.84 lbs.) Final cage
biomass and survival were higher for FRT (16.6 kg, 36.6
lbs. and 94.2%) than 0. mossambicus (13.0 kg, 28.6 lbs.
and 83.3%). The feed conversion ratio was lower for FRT
(2.52) than for 0. mossambicus (3.56).
The superior performance of FRT was probably due to its
heritage from two saltwater-tolerant species (0. urolepis
hornorum and 0. mossambicus) and to a 10-year selective

Table 5. Mean + standard error of initial and final body weight,
specific growth rate, weight gain, final cage biomass and survival
rate of "Florida" red tilapia and 0. mossambicus cultured in 2-m3
cages for 112 days in seawater.

"Florida" red 0. mossambicus

Initial body weight 152 3 148 4
Final body weight 439 2 381 21
Specific growth rate 0.95 0.01 0.84 0.04
Weight gain 2.56 0.01 2.08 0.17
Final cage biomass 8.3 0.3 6.5 1.5
Survival rate 94.2 3.0 83.3 16.7

breeding program for improved growth and color, whereas O.
mossambicus originated from an unselected (wild) population.
FRT grew at a slower rate in saltwater cages (2.56 g/day)
than it did in freshwater cages (3.61 g/day), but this may
have resulted from the different feeding regimes (once a

day feeding in salt water compared to continuous feeding in
fresh water). The bay was too rough for demand feeders.
The underlying cause of the bacterial disease episode may
have been poor nutrition. In an effort to obtain maximum growth
rates, the fish were initially fed at a rate beyond satiation,
contributing to the high feed conversion ratios. Large quantities
of feed remained on the surface for several hours and became
saturated with seawater, possibly leaching water-soluble
nutrients. Ingestion of this water-soaked feed of reduced
nutritional quality may have stressed the fish to the extent
that their resistance to marine pathogens was weakened.
Upon completion of the antibiotic treatment, the fish
remained free of diseases at feeding rates that allowed them
to consume the feed quickly.
Based on these studies, in the Virgin Islands 0. niloticus
and "Florida" red tilapia are recommended for freshwater
cage culture and "Florida" red tilapia is recommended for
saltwater cage culture. Tilapia farmers should be aware that
there are other species, strains and crosses that have not been
tested here, and that tilapia breeding programs throughout
the world are continually striving to improve growth
characteristics. New breeds should periodically be tested
under Virgin Islands conditions. Furthermore, individual
farmers must establish their own breeding programs to
prevent inbreeding and maintain the quality of their stocks.

We wish to express our appreciation to Dr. John
Kubaryk, Department of Marine Sciences, University of
Puerto Rico, for conducting the proximate analysis and to
the Virgin Islands Food Fish Farm, Inc. for allowing us to
use their facility. This research was supported in part by
Hatch Project No. 80100.

-i ~

IrI jj'f

, fr ,

Agricultural Experiment Station Personnel


Darshan S. Padda....Vice President for Research and Land Grant Affairs & Director
James Rakocy....Associate Director
Raquel Santiago Silver.... Assistant to the Vice President for Research and Land Grant
Yvonne Horton.... Administrative Assistant III
Audrey Valmont Schuster.... Administrative Assistant III
Francis Diaz....Trades Leader


Martin Adjei....Research Assistant Professor, Agronomy
Terry Gentry....Research Analyst I
Osvaldo Lopez.... Agricultural Aide III
Antonio Rodriguez.... Agricultural Aide II

Animal Science

Robert W. Godfrey....Research Assistant Professor, Animal Science
Mark Gray....Research Specialist II
Joni Rae Collins....Research Analyst III
Victor Callas.... Agricultural Aide II


James Rakocy....Research Professor, Aquaculture
William Cole....Research Specialist II
Donald Bailey....Research Specialist II
Kurt Shultz....Research Analyst II
Ezekiel Clarke.... Agricultural Aide II


Manuel Palada....Research Assistant Professor
Stafford Crossman....Research Specialist III
Jacqueline Kowalski... .Research Analyst I
Paulino Perez....Research Assistant I
Nelson Benitez.... Agricultural Aide II
Reinardo Vasquez.... Agricultural Aide II

Horticulture-Ornamentals, Fruits, Forestry and

Christopher Ramcharan....Research Assistant Professor
Thomas Zimmerman....Research Specialist III
James O'Donnell....Research Specialist II
Aberra Bulbulla....Research Analyst II
Jeremiah Hassan.... Agricultural Aide II
Agustin Ruiz.... Agricultural Aide II

Current Research Projects

Evaluation of Forage Conservation Systems in the Caribbean.
Evaluation of Integrated Mechanical and Chemical Control of
Casha (Acacia spp.) on Native Pasture.
Improving Forage Feeding Value by Urea Treatment.
Breeding and Biotechnology for Forage Yield, Quality and
Persistence ofPennistems.
Evaluation of Native Pasture and Agro-By-Product-Based Systems
for Market Lamb Production.
Herbage Allowance and Pasture Rotation Systems for Animal and
Forage Production on Tropical Pasture.
Increased Efficiency of Sheep Production.
Reducing Effects of Heat Stress on Reproduction in Dairy Cattle.
Evaluation of Tropical Adaptation of Non-Zebu Cattle Germplasm.
Studies on the Production of Tilapia in Marine Cages.
Evaluation of the Culture Potential of Selected Caribbean Marine
Integration of Tilapia and Hydroponic Vegetable Production in
Recirculating Systems.
Economic Analysis of Integrated Recirculating Systems.
Integrating Tilapia Culture in Tanks with Field Production of
Vegetable Crops.
Micro-Irrigation of Horticultural Crops in Humid Regions.
Evaluation of Saline Water for Irrigating Vegetable Crops in the
U.S. Virgin Islands.

Horticultural and Economic Evaluation of Vegetable Varieties in
the U.S. Virgin Islands.
Alley Cropping Systems for Sustainable Vegetable Production in
the U.S. Virgin Islands.
Improving Crop Management Systems for the Production of
Culinary Herbs in the U.S. Virgin Islands.
Evaluation of Horticultural Practices for Enhancing Root Crop
Production in the Virgin Islands.
Evaluation of Cultural Practices for Sweet Potato Weevil Control.
Evaluation of Integrated Production Methods for Tropical Fruit
Evaluation of the Effects of Tissue Culture on Somatic Variation
and Propagation of Breadfruit.
Evaluation of Minor Tropical and Subtropical Fruits and Nuts for
Production in the U.S. Virgin Islands.
Evaluation of Trees for Agroforestry in the U.S. Virgin Islands.
Potential for Ornamental Pot Crops in the Virgin Islands Using
Growth Regulators.
Transformation and Regeneration of Hibiscus and Bougainvillea.
Effects of Bioherbicides on Competitive Ability of Nutsedge.
Biochemical Basis of Resistance ofNutsedge Biotypes and Species
to Nutsedge Rust.

Recent Publications

Adjei, M.B. 1992. Sorghum silage production from an alley
cropping system. UVI Research. Agricultural Experiment
Station, University of the Virgin Islands 4:14-16.

Adjei, M.B., K. Albrecht and C.L. Wildeus. 1992. Performance of
Desmanthus virgatus accessions in the Caribbean. Proceedings
of the International Grasslands Congress, New Zealand and
Queensland, Australia. (in press).

Adjei, M.B., P. Mislevy and W. Chasen. 1992. Seed yield of
bahiagrass in response to sward management by phenology.
Agronomy Journal 84(2):599-603.

Adjei, M.B. and W.D. Pitman. 1993. Response of Desmanthus to
clipping on a phosphatic clay mine-spoil. Tropical Grasslands
27. (in press).

Bailey, D.S. and J.E. Rakocy. 1992. Economics oftilapia culture
in freshwater cages. UVI Research. Agricultural Experiment
Station, University of the Virgin Islands 4:6-11.

Cole, W.M., J.E. Rakocy, D.S. Bailey and J.A. Hargreaves. 1992.
The effect of three pelleted diets on the survival, growth and
feed conversion ofjuvenile white grunts (Haemulon plumeri).
Proceedings of the Gulf and Caribbean Fisheries Institute-44,
Merida, Mexico. (in press).

Collingwood, C.D., S.M.A. Crossman and M.C. Palada. 1992.
Tomato germplasm evaluation for growth and productivity in
the Virgin Islands. Proceedings of the Caribbean Food Crops
Society-28, Santo Domingo, Dominican Republic. (in press).

Courtney, C.H. and S. Wildeus. 1992. Epidemiology of parasitic
gastroenteritis of sheep on St. Croix. Proceedings of the
American Association of Veterinary Parasitologists 37:53.

Crossman, S.M.A. 1992. Status of sweet potato weevil problems
in the U.S. Virgin Islands. Proceedings of the International
Conference on Sweet Potato Pest Management, Miami,
Florida. (in press).

Crossman, S.M.A. 1992. Sweet potato germplasm evaluation in
the U.S. Virgin Islands. UVI Research. Agricultural
Experiment Station, University of the Virgin Islands 4:4-5.

Crossman, S.M.A. 1993. Sweet potato: a new look at an old crop.
Virgin Islands Agriculture and Food Fair Bulletin No. 7: 33-

Crossman, S.M.A. and C. George. 1993. Sweet potato production
in the Virgin Islands. Growers Factsheet No. 2, Cooperative
Extension Service, University of the Virgin Islands. 4 pp.

Crossman, S.M.A. and C. George. 1993. Vegetable varieties
recommended for the Virgin Islands. Growers Factsheet No.
1, Cooperative Extension Service, University of the Virgin
Islands. 8 pp.

Crossman, S.M.A. and C.D. Collingwood. 1992. Herb production
in the Virgin Islands. Celebrating Twenty Years of Learning
and Growing. UVI Cooperative Extension Service, St.
Thomas/St.John District Agriculture and Food Fair Bulletin,

Crossman, S.M.A., C.D. Collingwood and M.C. Palada. 1993.
Effect of physical barriers on sweet potato weevil control.
Proceedings of the Caribbean Food Crops Society-29, Fort de
France, Martinique. (in press).

Crossman, S.M.A., A.A. Navarro and C.D. Collingwood. 1992.
The potential for year-round tomato production in the Virgin
Islands. UVI Research. Agricultural Experiment Station,
University of the Virgin Islands 4:11-13.

Crossman, S.M.A., M.C. Palada and C.D. Collingwood. 1992.
Yield evaluation of sweet potato cultivars in the U.S. Virgin
Islands. Proceedings of the Caribbean Food Crops Society-
28, Santo Domingo, Dominican Republic. (in press).

Hammond, A.C. and S. Wildeus. 1993. Effects of coconut meal or
fish meal supplementation on performance and carcass
characteristics in whole diet apparent digestibility of growing
St. Croix lambs fed a tropical grass-based diet. Small
Ruminant Research. (in press).

Locasio, S.J., G.A. Clark, A.A. Csinszky, C.D. Stanley, S.M.
Olson, F. Rhoads, A.G. Smajstrala, G. Vellidis, R.D. Edling,
H.Y. Hanna, M.G. Goyal, S.M.A. Crossman and A.A.
Navarro. 1992. Water and Nutrient Requirements for Drip-
Irrigated Vegetables in Humid Regions. Southern Cooperative
Series Bulletin 363. 17 pp.

*Losordo, T.M., M.P. Masser and J.E. Rakocy. 1992. Recirculating
aquaculture tank production systems: an overview of critical
considerations. Southern Region Aquaculture Center
Publication No. 451, Delta Branch Experiment Station,
Stoneville, Mississippi. 8 pp.

*Losordo, T.M., J.E. Rakocy and M.P. Masser. 1992. Recirculating
aquaculture tank production systems: component options.
Southern Region Aquaculture Center Publication No. 453,
Delta Branch Experiment Station, Stoneville, Mississippi.
12 pp.

*Masser, M.P., J.E. Rakocy and T.M. Losordo. 1992. Recirculating
aquaculture tank production systems: management of
recirculating systems. Southern Region Aquaculture Center
Publication No. 452, Delta Branch Experiment Station,
Stoneville, Mississippi. 12 pp.

Mislevy, P., M.B. Adjei, G.M. Prine and F.G. Martin. 1992.
Energycane response to harvest management. Soil and Crop
Science Society of Florida Proceedings 51:79-84.

O'Donnell, J.J. 1993. Mahogany (Swietenia spp.) response to
increasing water deficits. Proceedings of the Caribbean Food
Crops Society-29, Fort de France, Martinique. (in press).

O'Donnell, J.J. and M.B. Adjei. 1992. Symbiotic effectiveness of
inoculated and indigenous rhibozia with six Leucana varieties.
Agronomy Abstracts 264.

O'Donnell, J.J. and M.C. Palada. 1993. Establishment and growth
of four hedgerow species for alley cropping systems on St.
Croix. Agronomy Abstracts 59.

O'Donnell, J.J., J.P. Muir, G.E. Aiken, M.B. Adjei and W.D.
Pitman. 1991. Response of 'Florida' Carpon Desmodium to
inoculation with VA mycorrhizal fungi and rhizobia.
Agronomy Abstracts: 273.

O'Donnell, J.J., D.M. Silvia, WD. Pitman and J.E. Rechcigl. 1992.
Inoculation of Vigna parkeri with mycorrhizal fungi in an
acid Florida spodosol. Tropical Grasslands 26:120-129.

Palada, M.C., S.M.A. Crossman and C.D. Collingwood. 1992.
Effect of organic and synthetic mulches on yield of basil under
drip irrigation. Hortscience 27(6):99. (abstract).

*Palada, M.C., B.T. Kang and S.L. Claassen. 1992. Effect of alley
cropping with Leucana leucocephala and fertilizer application
on yield of vegetable crops. Agroforestry Systems 19:139-147.



Palada, M.C. 1993. Composting: an appropriate method of soil
conservation for the U.S. Virgin Islands. Virgin Islands
Agriculture and Food Fair Bulletin No. 7:36-39.

Palada, M.C., S.M.A. Crossman and C.D. Collingwood. 1993.
Irrigation water use and yield of thyme in the Virgin Islands.
Proceedings of the Caribbean Food Crops Society-29, Fort de
France, Martinique. (in press).

Palada, M.C., S.M.A. Crossman and C.D. Collingwood. 1993.
Yield Performance of Vegetable Varieties in the U.S. Virgin
Islands: 1988-1992. Technical Bulletin No. 5. Agricultural
Experiment Station, University of the Virgin Islands. 28 pp.

Palada, M.C., S.M.A. Crossman, C.D. Collingwood and J.A.
Kowalski. 1993. Water use and yield of bell peppers in
hedgerow intercropping with drip irrigation. Agronomy
Abstracts. (in press).

Phelps, R.P, W. Cole and T. Katz. 1992. Effect offluoxymesterone
on sex ratio and growth of Nile tilapia, Oreochromis niloticus
(L.). Aquaculture and Fisheries Management 23:405-410.

*Rakocy, J.E. 1992. Feasibility of using vegetable hydroponics to
treat aquaculture effluents. Pages 347-350 in J. Blake, J.
Donald and W. Magette, editors. Proceedings of the National
Livestock, Poultry and Aquaculture Waste Management
Workshop. Kansas City, Missouri.

Rakocy, J.E. 1992. Waste management in integrated recirculating
systems. Twenty-first Symposium of the United States-Japan
Natural Resources Panel, 1992, Kyoto, Japan. (in press).

*Rakocy, J.E., T.A. Losordo and M.P. Masser. 1992. Recirculating
aquaculture production systems: integrating fish and plant
culture. Southern Regional Aquaculture Center Publication
No. 454, Delta Branch Experiment Station, Stoneville,
Mississippi. 8 pp.

Rakocy, J.E. and J.A. Hargreaves. 1993. Integration of vegetable
hydroponics with fish culture: a review. Pages 112-136 in J.-
K. Wang, editor. Proceedings of the Aquaculture Engineering
Conference on Techniques for Modern Aquaculture.
Aquacultural Engineering Group, American Society of
Agricultural Engineers.

Rakocy, J.E., J.A. Hargreaves and D.S. Bailey. 1993. Nutrient
accumulation in a recirculating aquaculture system integrated
with hydroponic vegetable production. Pages 148-158 in J.-
K. Wang, editor. Proceedings of the Aquaculture Engineering
Conference on Techniques for Modern Aquaculture.
Aquacultural Engineering Group, American Society of
Agricultural Engineers.

Ramcharan, C. 1992. A low cost chemigator for the irrigation of
fruit crops. Proceedings of the Caribbean Food Crops Society-
28, Santo Domingo, Dominican Republic. (in press).

Ramcharan, C. 1992. The potential of Christmas Snowflake as a
Christmas pot crop using growth retardants. UVI Research.
Agricultural Experiment Station, University of the Virgin
Islands 4:17-18, 20.

Wildeus, S. 1993. Age-related changes in scrotal circumference,
testis size and sperm reserves in bulls of the tropically-adapted
Senepol breed. Animal Reproduction Science. (in press).

Wildeus, S. and C.H. Courtney. 1992. Influence of gastrointestinal
parasite control on body weight, blood parameters and fecal
egg counts in hair sheep ewes and lambs in the dry tropics.
Journal of Animal Science 70:585. (abstract).

Wildeus, S. and A.C. Hammond. 1993. Testicular, semen, and
blood parameters in adapted and nonadapted Bos taurus bulls
in the semi-arid tropics. Theriogenology 40:345-355.

Wildeus, S. and A.C. Hammond. 1993. Effects of supplementation
on body weight change and postpartum interval in lactating
hair sheep ewes grazing tropical pastures. Journal of Animal
Science 71(Supp. 1):25. (abstract).

Wildeus, S. and'G. Willock, Jr. 1993. Physiological response of three
sheep breeds to summer temperatures on St. Croix. Virgin Islands
Agriculture and Food Fair Bulletin No. 7:51-53.

Wildeus, S., J.R. Collins and M.L. Gray. 1993. Lambing
performance and postpartum interval in hair and wool sheep
ewes under tropical conditions. Journal of Animal Science
71(Supp. 1):242. (abstract).

Wildeus, S., M.L. Gray and J.R. Collins. 1993. Seasonal effects
on pubertal characteristics in hair and wool sheep ewe lambs
under tropical conditions. Journal of Animal Science
71(Supp.1):119. (abstract).

*Wildeus, S., L.R. McDowell and J.R. Fugle. 1992. Season and
location effects on serum and liver mineral concentrations of
Senepol cattle on St. Croix. Tropical Animal Health &
Production 24:223-230.

Wright, D.W, H.D. Hupp, R.J. Deans and I.L. Mao. 1992. Growth
characteristics of Senepol bulls fed various energy levels over
different time periods, under tropical conditions. Tropical
Agriculture (Trinidad) 69(2):181-185.

Wright, D.W, S. Wildeus, C.J. Brown and Z.B. Johnson. 1992.
Expected progeny differences for weaning weight in the
Senepol breed. UVI Research. Agricultural Experiment
Station, University of the Virgin Islands 4:2-4.

Zimmerman, T.W 1993. Dwarfing plants through applied biology.
1993. Proceedings of the Caribbean Food Crops Society-29,
Fort de France, Martinique. (in press).

*An asterisk in front of an entry indicates that it has been
previously listed, but was in press before. Now the entry is

3 3138l0019 iii8
3 3138 00191 8043


University of the Virgin Islands
Agricultural Experiment Station
RR 2, Box 10,000 Kingshill
St. Croix, USVI 00850

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs