This item has the following downloads:

VIP 2(2) ( PDF )

    Front Cover
        Front Cover 1
        Front Cover 2
    Table of Contents
        Table of Contents
    From the director
        Page i
    International honor award
        iPage ii
        iPage iii
    Determination of the minimum irrigation requirements of tomatoes
        Page 24
        Page 25
        Page 26
        Page 27
    Bull fertility on St. Croix: Effects of breed, age and season
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
    Feeding practices for caged blue tilapia
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
    Alfalfa production in the U.S. Virgin Islands
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
    AES publications
        Page 45
        Page 46
    Back Cover
        Back Cover 1
        Back Cover 2

Virgin Islands Perspective


Agricultural
Research Notes


Summer Fall 1987
Vol. 2, No. 2


Agrclua Exeimn Station


I.rha S.*da-vc rsdn
Arthur~ ~ ~ ~ A.Rcad PeietRsarhadLn-GatPorm


I I




LIBRARY
UiNIVE-fSITY OF THE VIRGIN ISLANDS
ST. CROIX








Virgin Islands Perspective

Agricultural
Research Notes
Summer Fall 1987
Vol. 2, No. 2

Table of Contents

Determination of the Minimum Irrigation
Requirements of Tomatoes ..........................24
Adriano A. Navarro

Bull Fertility on St. Croix:
Effects of Breed, Age and Season ....................28
Stephan Wildeus

Feeding Practices for Caged Blue Tilapia ................34
John Hargreaves

Alfalfa Production in the U.S. Virgin Islands .............. 40
Cyndi L. Wildeus

Recent Publications ............................... .45



James Rakocy Carrol B. Fleming
Assistant Director Editor

Published by the University of the Virgin. Islands
Agricultural Experiment Station
RR2, Box 10,000 Kingshill
St. Croix, U.S. Virgin Islands 00850







From the Director


This is the third issue of the "Virgin Islands Perspective," the
Agricultural Experiment Station's semiannual report of selected
research results. The articles in this issue report results obtained
in four out of the eighteen projects presently underway at the
Station.
Alfalfa was examined as a possible forage under the environ-
mental conditions of the Virgin Islands. To improve cattle
management on St. Croix, bull fertility was characterized in
relation to breed, age, and season. Pan evaporation rates were
used to determine the optimum drip irrigation rates for maxi-
mum tomato production. Demand (self) feeders were developed
and tested in comparison with manual feeding for the cage
culture of tilapia.
These projects have been designed to generate new knowl-
edge of advanced technologies in alternative agricultural enter-
prises, to develop better adapted and more productive strains of
plants and animals, to improve farming practices and farm
management methods. All of these resultant technologies
should produce higher economic returns to farmers and
improve the quality of food for Virgin Islands consumers.
I urge all Virgin Islands residents to review the research infor-
mation reported in this publication, and contact us for more
information. The Agricultural Experiment Station and the
Cooperative Extension Service have generated many other
publications which are available to the community at no cost.
Call or write us for information and assistance.


Darshan S. Padda
Vice President
Research and Land-Grant Programs









International

Honor Award


c*~E~, `
I,~~ 1. -

i~~~l~45 X u ., ,~1 l

ik *A l


,d


~r
rl
1
"
r;

r" -ri


Dr. Darshan S. Padda was the recipient of the 1987 International
Honor Award from USDA. UVI President Arthur A. Richards, who
participated in the ceremonies, proudly joined the other USDA
dignitaries in congratulating Dr. Padda. Shown in the picture, from
left to right, are Daniel G. Amstutz, Undersecretary for Interna-
tional Affairs and Commodity Programs, Deputy Secretary Peter C.
Myers, Dr. Padda, President Richards and OICD Administrator
Joan S. Wallace.


--s


rl


r.
Yt:













































I.


I
t-
ccs~irr(
-;
i
r-- ` B


Tomatoes have specific water requirements.


:~~` '


r
I


`





ra.








Determination of the Minimum
Irrigation Requirements
of Tomatoes
By Adriano A. Navarro
Olericulture Program Leader
Tomatoes are one of the world's most popular vegetables.
Among the twenty-two important vegetables in the world, toma-
toes are ranked second in value, next to Irish potatoes. They are
a good source of vitamins A and C, and the number of ways they
can be used to improve the flavor of other foods is seemingly
endless. In the U.S. Virgin Islands, as in other parts of the world,
tomatoes are one of, if not the most important, cash crops. They
are well adapted to the VI climate and soil, and the local market
for fresh tomatoes is excellent.
In a dry climate such as we have in the U.S. Virgin Islands, it is
almost impossible to produce a good crop of tomatoes without
irrigation. Because of low and often erratic rainfall, water is not
an abundant resource in the U.S. Virgin Islands. The under-
ground water that is suitable for irrigation is also very limited. For
these reasons, the use of water for irrigation can be justified from
an economic standpoint only if used efficiently on high value
crops such as tomatoes.
Determination of the mininum water requirements of toma-
toes would provide valuable information in applying irrigation
water more efficiently and economically. At the University of the
Virgin Islands Agricultural Experiment Station, one of the
research areas being pursued is the determination of the least
amount of water that can be applied to various crops without any
significant reduction in yield or impairment of product quality.
Drip irrigation studies to determine minimum water require-
ments of tomatoes were started in 1980. In the first experiment,
different rates of water were applied using automatic switching
tensiometers and solenoid valves. The tensiometers were set at
suction pressures of 20, 40 and 60 centibars. Treatments set at
lower pressures received more water than those at higher
pressures. With the switching tensiometers, moisture level in the
different treatments was maintained automatically. Irrigation
water was turned on or off with the change in the degree of wet-
ness or dryness of the soil. Royal Chico, a determinate tomato
variety, was used in this experiment.
Funding Source: CBAG Project 572








The data (Table 1) show that the highest yield was obtained
from the treatment which received water at the rate of 6.6 mm
per week or 4.5 liters of water per plant per week. The second
highest yield was taken from the treatment with irrigation appli-
cation at the rate of 6.4 mm or 4.4 liters of water per plant. Since
the difference in yield between the two treatments is not statis-
tically significant, the lower rate of irrigation of 4.4 liters per plant
per week is the preferable rate of application.

Table 1
Effect of varying irrigation application rates on the yield
of tomatoes cv. Royal Chico. (1980)

Average
Weekly
Plant Irrigation Precipi- Mean'
Treatment Density Per Week station Yield
(centibar) (plants/ha) (mm) (I/plant) (mm) (t/ha)

60 14,375 6.4 4.4 16.8 25.6 a
40 14,375 6.6 4.5 16.8 28.7 a
20 14,375 17.3 11.7 16.8 20.7 b

'Differences between means subscripted with different letters are signif-
icant at the 5% level.

In the following year, a similar experiment was conducted,
using Tropic, an indeterminate tomato variety. The highest yield
occurred in the treatment which was irrigated at a rate of 14.7
mm of water per week or 5.4 liters of water per plant per week
(Table 2). The yield in this treatment was significantly (P <0.05)
higher than the yields at the lower rates of irrigation. A com-
parison of the irrigation rates shows an increase of 93% in the
weekly irrigation rate (2.8 to 5.4 liters/plant/week) resulted in an
increase of approximately 26% in the yield (from 30.2 to 38.0
tons). These data demonstrate that water was used much more
efficiently at the lower rate of irrigation.
In 1986, another experiment on the determination of minimum
water requirements of tomatoes was conducted using the inde-
terminate variety N-69. The results show that yields increased by
9.2 tons/hectare as the irrigation rate increased from 3.1 to 5.9
liters/plant (Table 3).
Analysis of the water use efficiency (unit of production per unit
of water used) showed that irrigation water was more efficiently








Table 2
Effect of varying water application rates on the yield
of tomatoes cv. Tropic (1981)
Average
Weekly
Plant Irrigation Precipi- Mean'
Treatment Density Per Week station Yield
(centibar) (plants/ha) (mm) (I/plant) (mm) (t/ha)

60 26,953 7.6 2.8 17.5 30.2 a
40 26,953 9.6 3.6 17.5 30.9 a
20 26,953 14.7 5.4 17.5 38.0 b

'Differences between means superscripted with different letters are
significant at the 5% level.


used at the lower rate of irrigation. An increase of 74 percent in
water application (from 3.1 to 5.9 liters/plant/week) increased
the yield by only 20 percent (from 45.4 to 54.6 tons).
The three experiments have revealed some important infor-
mation on the water requirements of tomatoes. The first experi-
ment indicated that tomatoes are quite specific in their water
requirements. Yields will increase up to a certain level of water
application, but beyond that level a reduction in yields can be
expected. The second and third experiments also revealed that
although the yield can be increased with an increase in water
application (up to a point) the utilization of water becomes less
efficient beyond a certain level of application and may no longer
be economical, particularly where water is a limited and expen-
sive resource. The data further showed that water use efficiency
was highest at irrigation rates of 4.4,2.8 and 3.1 liters/plant/week
in the first, second and third experiments, respectively, or an
average for the three experiments of 3.4 liters/plant/week.
The experiments also provided some information concerning
the extent to which plant density and variety affected yields. The
higher yields in the 1986 experiment can be attributed partly to
variety and planting density. Planting at higher densities and
using proven varieties can increase total yields and possibly
improve water use efficiency. Finally, the data showed, as gen-
erally expected, that the amount of precipitation during the
cropping period affects the irrigation requirements of crops. The
amount of supplemental irrigation was lowest in the 1981








Table 3
Effect of two irrigation rates on the yield
of tomatoes cv. N-69. (1986)

Average
Weekly
Plant Irrigation Precipi- Mean'
Treatment Density Per Week station Yield
(centibar) (plants/ha) (mm) (I/plant) (mm) (t/ha)

40-50 17,968 5.5 3.1 9.3 45.4 a
20-30 17,968 10.4 5.9 9.3 54.6 b

'Differences between means is significant at 5% level.


experiment because of high precipitation during that cropping
season.


Accurate water metering devices and soil moisture sensing
devices such as tensiometers (not shown) can make irrigation
more efficient.








Bull Fertility on St. Croix:
Effects of Breed, Age and Season
By Stephan Wildeus
Research Animal Scientist
The reproductive performance of bulls contributes signifi-
cantly to the overall fertility of a herd, particularly under conditions
of natural mating. As part of natural mating under pasture condi-
tions the bull is exposed to environmental stressors, is required
to display a high level of physical fitness and may have to service
large numbers of estrous cows over relatively short periods of
time. These demands become more pronounced in a tropical
setting, in which the animal is exposed to periodic nutritional
stress and high environmental temperatures, both known to
interfere with proper testicular function. Hence, bull fertility and
management should play an important role in the overall
management of a cattle operation.
On St. Croix the predominant breeds of cattle are the Senepol
and the Holstein. Apart from the contrasting production stress
encountered by the two breeds in a beef and dairy setting,
respectively, these genotypes also differ in their degree of adap-
tation to the local environment. While the Senepol cattle were
developed and selected to perform in a low input system under
the local tropical conditions, the Holstein breed has a back-
ground as a high input dairy animal developed in a temperate
climate. The following information was generated to provide an
estimate of differences in bull performance of these two cattle
breeds under St. Croix conditions.
In a first experiment two groups of mature, sexually-rested
Senepol (10) and Holstein (9) bulls were examined for physical
characteristics and semen quality during the summer months on
St. Croix. The bulls of both breeds were located on adjacent
farms on the southeastern shore of St. Croix, and exposed to
grazing on native pasture without supplementation. Measure-
ments on all bulls were made on the same day (82F, 81.5% rela-
tive humidity) and by the same operator to reduce measurement
error. Semen was collected by electroejaculation and analyzed
for both quality and quantity.
The information obtained in this first study is presented in
Table 1. Both groups were of similar age and body weight, but
the Holsteins had a lower body condition score. Testis size, as
Funding Source: Regional Hatch Project #791








indicated by scrotal circumference, and testis tone, a measure of
the firmness of the testicular tissue, were lower in the Holstein
bulls. Under more ideal production conditions these values are
reported to be generally higher for the Holstein breed and would
suggest a somewhat reduced reproductive function under local
conditions. Other indications of the physiological stress in the
dairy bulls were elevated rectal temperatures and lower blood
packed cell volume and hemoglobin values compared to the
Senepol bulls.
Some of these physical differences were also reflected in
ejaculate characteristics. Ejaculate volume and total number of
spermatozoa per ejaculate were considerably higher in the
Senepol bulls. In contrast, sperm quality was not affected to a
similar degree. Sperm motility and seminal fructose did not differ
significantly, due to greater variation between individual bulls
rather than the two breeds. However, abnormal sperm tails and
Table 1
Performance and semen characteristics (mean+SEM) in two groups of
Senepol and Holstein bulls during the summer months on St. Croix.

Characteristics Senepol Holstein

Age (months) 36 2+2.9 38 3+2.7
Weight (Ibs) 1545+64 1606+119
Condition score (1-10) 58+0.13 5.0+0.5
Scrotal circumference (cm) 37.7+1.0 34.7+1.5
Testis tone (1-9) 7.7+0 21 5.1+0.4
Body temperature (OF) 102 8+0.2. 104.0+0 3
Packed cell volume (%) 53.0+0.8 45.6+1.8
Hemoglobin (g/dl) 13.0+0.3 12.3+0.7

Sperm motility (%) 71.7+3.1 63.6+5.9
Ejaculate volume (ml) 11.1+0.8 2.6+0.6
Concentration (mill/ml) 350+86 333+119
Sperm/ejaculate (bill) 4.14+1.11 0.92+0.5
Seminal fructose (mg/dl) 251+36 309+59
Sperm morphology (%) 87.3+1.9 72.3+8.1
abnormal heads 6.3+1.8 6 4+1.5
abnormal tails 5.20.9 17.95.7
cystoplasmic droplets 1.1+0 6 3.441.4
intact acrosomes 75.6+3 7 48.9+11 1








































Portable lab for analyzing semen in the field.


missing acrosomes (the cap covering the sperm head) were
present in larger numbers in the Holstein bulls.
In a second study these preliminary observations were re-
evaluated in a larger number of Senepol and Holstein bulls. Bulls
were examined on three beef and five dairy farms located
throughout the island. Samples were collected in association
with the end of the dry/warm (August, September) and cool/wet
season (February, March) to determine possible seasonal dif-
ferences. These dates were chosen in order for the animals to

30







display the long-term effects of the season on the animals'
physiological characteristics. Apart from testicular and semen
measurements, blood glucose and blood urea nitrogen levels
were determined to estimate the nutritional status of the bulls in
this study.
Scrotal circumference and testis tone were higher in the
Senepol bulls in this study and are in agreement with the findings
of the earlier experiment. Season only affected seminal fructose
concentrations significantly, which were lower following the
wet/cool season, but sperm motility showed a similar tendency
to be lower following the wet/cool season (Table 2). In contrast,
the percentage of morphologically normal spermatozoa was
lower during the dry/warm season. However, measurements of
sperm motility and morphology varied greatly between indi-
vidual bulls. Sperm motility may have been influenced by the
ambient temperature at the day of collection. Blood metabolite
levels were not affected by either breed or season.
Bulls were further classified as young (less than 37 months),
mature (37 to 72 months) and old (73 months and older) in order
to examine the effects of age within the breed on reproductive
performance. Scrotal circumference increased with age in both
breeds but reached a greater mature dimension in the Senepol
compared to the Holstein bulls (Figure 1). In contrast, testis tone
significantly decreased with age in both breeds, but was con-

Table 2
Reproductive characteristics and blood metabolites (meanSEM) in
Senepol and Holstein bulls during the wet and dry season on St. Croix

wet/cool season dry/warm season
Characteristics Senepol Holstein. Senepol Holstein

Scrotal cir. (cm) 35.2+0.4 35.2+0.7 36.8+0.6 35.70.6
Testis tone (1-9) 7.690.12 6.95+0.23 7.880.08 7.00+0.22
Sperm motility (%) 59.3+3.6 48.8+8.6 65.02.7 55.2+5.2
Normal sperm (%) 82.72.6 88.3+3.5 80.1+2.3 83 1+3.3
Semen fructose (mg/dl) 101+20 54+18 186+25 9427
BUN' (mg/dl) 13.70.9 15.0+3.7 14.1+3.1 10.6+0.9
Blood glucose (mg/dl) 68.31.2 67.9+1.5 67.7+1.8 65.5+1.9

'BUN = Blood urea nitrogen







sistently lower in the Holstein bulls within age groups. Neither
the percentage of motile or morphologically normal sperma-
tozoa nor seminal fructose levels changed with age, while blood
glucose levels tended to decline in the older bulls (Table 3). The
breakdown into age-categories resulted in relatively small
numbers of animals in the 'old' classification and thus accounts
for the considerable variation and missing measurements in
some of the variables in this group. Age in the older bulls appar-
ently had no effect on ejaculate characteristics and should impair
bull reproduction function only indirectly through reduced
physical fitness.
42


young mature old
Figure 1. Distribution by age of scrotal circumference and testis tone
(mean+SEM) in Senepol and Holstein bulls (young: <37 mo, mature: 37to
72 mo, old: >72 mo).







Table 3
Semen characteristics and blood metabolites (meanSEM) in
Senepol and Holstein bulls in three age categories.


Characteristics Young (<37mo) Mature (37-72mo) Old (>72mo)
Senepol Holstein Senepol Holstein Senepol Holstein

% Motile 634+2 3 56 7+5 1 62 65 4 41 4+9 8 60 0+12.3 62.5+12 5
% Normal 80 7+2 4 82 5+3 4 81 82 7 91 01 7 84 3+5 2 86.0+0.0
Fructose (mg/dl) 152+22 70+18 15630 49+28
BUN (mg/dl) 149+23 127+25 107+1 5 13 01 5 209+129 8 8+20
Glucose (mg/dl) 69 2+1 1 67 4+1 8 673+1 9 65 4+2 3 60 2+10 5 633+2 1


The results obtained in these studies point to a markedly
lowered reproductive function in Holstein compared to Senepol
bulls on St. Croix. This difference appears to be consistent
throughout the year, with only minimal variations between
seasons. The lack of seasonal differences would point to high
environmental temperatures, present on St. Croix throughout
the year, as the primary source of stress on the dairy bulls,
though the inabilityy to cope with nutritional deficiencies and
high parasite burdens is likely to contribute to the variation in
performance between the two breeds. Since sperm quantity
rather than sperm quality tended to be affected more severely,
some benefits in dairy bull management may be derived from
reducing the numbers of cows exposed per bull and providing
the bulls with adequate sexual rest.

The technical assistance of Ms. Joni Rae Fugle and Ms. Kim
Traugott is gratefully acknowledged. This study was possible due to the
cooperation of the following farms: Annaly, Castle Nugent, Corn Hill,
Estate Granard, Estate Sight, Mon Bijou, Petronella and Windsor. The
author would like to thank Dr. A. Hammond, USDA-ARS, for the
analysis of the blood metabolites.







Feeding Practices for
Caged Blue Tilapia
By John Hargreaves
Research Specialist Aquaculture
The UVI-AES aquaculture program has been conducting re-
search on the cage culture of the blue tilapia (Tilapia aurea) since
1977. Cage culture offers many advantages over the open pond
culture of fish1, as caged fish are more conveniently stocked, fed,
sampled, and harvested. Perhaps most importantly, the culture
of tilapia in cages will prevent successful spawning as eggs will fall
through the cage mesh. The initial capital investment for cage
materials and construction is low, resulting in a relatively quick
payback period.
One of the main facets of this cage culture research has been
the evaluation of various feeds and feeding practices. Since
caged tilapia are usually confined at high densities (200 400
fish/m3), the contribution of naturally-occurring food items to
fish production is minimal. The use of a complete ration with a
high level of crude protein (32 40%) and vitamin and mineral
supplements is essential to achieve fish growth rates (2.5 3.5
g/day) and feed conversion ratios (<2:1) which will yield a favor-
able return on investment to the fish farmer. Feed costs consti-
tute approximately 50 60% of the total costs of a cage culture
operation.
Fish culturists typically feed fish their daily ration as a percent-
age of body weight, which is higher for smaller fish and decreases
as the fish grow. Adjustments are made based on a sample of the
fish population or an assumed feed conversion efficiency over
the sample interval. However, feeding fish on a fixed schedule
based on feeding rate tables does not take account of the inter-
active complex of water quality and other variables which affect
fish feeding response, digestion, and assimilation.
'Freshwater resources on St. Croix are limited and generally fully exploited.
However, freshwater ponds are multi-use resources that have not been devel-
oped to their optimum potential. These ponds are used for livestock watering,
crop irrigation, temporary water storage for recharge of aquifers, and as a soil
conservation or flood prevention measure. Between 1920 and 1975, 225 ponds
were constructed or renovated on St. Croix. Most of these ponds are 0.5 to 1
ha in surface area, irregular in shape and bottom profile, subject to wide fluctua-
tions in water level, and difficult or impossible to drain completely. As such,
they are not suitable for the open pond culture of fish.
Funding Source: Hatch Project #011






































































































Demand (self) feeder for cage culture of tilapia.


-Ij,


C1


;








Tilapia have a digestive system adapted to their generally
herbivorous feeding habit: the intestine is long relative to body
length and the food passage rate is rapid (2.5 3.0 hrs at 300C).
Consumption of natural food items is continuous throughout the
daylight hours. Increasing the feeding frequency of cultured
tilapia may allow digestion to proceed more continuously than
less frequent feedings and thereby improve the efficiency of
digestion and assimilation. Demand (self) feeders allow fish to
regulate feed consumption and thereby eliminate the need for a
feeding rate table, feeding rate adjustments, and the labor re-
quired for regular fish sampling.
Studies were conducted during 1983 85 in St. Croix farm
ponds to evaluate the growth, feed conversion ratios, and
marketable production of blue tilapia fed according to a feeding


Key:
A demand feeder
B PVC support structure
C cover
D plastic bucket
E feed chamber
F brass nut and bolt


G plastic funnel
H horizontal brass rod
I key ring swivel
J brass nuts
K plexiglass plate
L vertical brass rod


Figure 1. Cross-sectional view of demand feeder with side view of support
structure superimposed.







rate table or from demand feeders. The 1-m3 cylindrical cages
utilized in the studies were constructed of semi-rigid, 19-mm
plastic mesh tied onto steel hoops.
Demand feeders were constructed with an 18.9-L plastic
bucket into which a polyethylene funnel was inserted (Fig. 1).
The feeder was mounted on the cage top and would release
feed when the fish hit a rod that was suspended vertically from
the funnel into the water. The feeders, which had a capacity of
5.5 kg of pelleted feed, were checked twice daily during fixed
schedule feedings. Demand feeders were filled only when com-
pletely empty to determine the minimum labor required for
feeding. Fish were cultured for 20 28 weeks depending upon
seasonal water temperatures and other water quality factors.
Fish growth in all ponds and experiments was nearly linear and
was not different between the two feeding regimes. Feed conver-
sion ratios of fish obtaining feed from demand feeders (1.55)
were lower than those of fish fed according to a fixed schedule
(2.30) for all ponds and experiments, although these differences
were not statistically significant (Table 1).
Marketable fish production in cages was affected by feeding
regime, stocking rate, number of fish from the original cage stock
which grew to market size, cage recruitment from tilapia popu-
lations in ponds, differential mortality and possible fish escape

Table 1
Production parameters from one experiment in which
caged Tilapia aurea were fed according to fixed-schedule
or demand feeding regimes.


Feeding Regimes

Parameter Fixed Schedule Demand Feeding

Growth rate (g/day) 1.8 2.1
Specific growth rate (%/mo) 38 41
Feed conversion ratio
(wet wt) 2.30 1.55
Marketable production (kg)
Small 45.7 51.6
Large 77.9 94.0
Total 123.6 145.6
Feedings (no) 285 34







following runoff events. Fish 19 cm were considered market-
able. Marketable production was 91 95% (by weight) of total
production.
Interference from open pond fish populations can be a serious
limitation of cage culture. Tilapia <9 g were able to pass through
19-mm mesh. These cage recruits competed for feed and space
with the stocked population and were unable to escape after a
period of growth. Most of these fish did not reach marketable
size during the culture period and therefore represented a re-
duction in feed utilization efficiency. Cage recruitment ac-
counted for a relatively small proportion (3.8 11.0%) of caged
fish production by weight, but could account for more than one
third of the mean number of fish harvested.
Control of cage recruitment of tilapia by stocking a piscivorous
fish species into the open pond or by chemical eradication may
be necessary. Use of a smaller mesh size (13 mm) would restrict
fish entry into cages, but would also limit water circulation which
could affect production.
Rapid changes in water quality following storm runoff gen-
erally have a deleterious effect on caged fish production. Small
ponds (0.1 0.5 ha) are particularly sensitive to runoff-induced
water quality changes. Storm runoff can lead to a depletion of
dissolved oxygen in several ways. First, the flush of fresh water
will mix anaerobic bottom layers of water with surface layers of
water, causing a rapid decline in dissolved oxygen levels as the
organic material breaks down in the presence of oxygen. The
decomposition and mineralization of re-suspended organic sedi-
ments will release nutrients to the water which may result in the
development of a dense phytoplankton bloom. Should this algae
bloom "crash", oxygen production by algal photosynthesis will
cease and severe caged fish mortality may occur. Second, in-
creases in pond water level will submerge and thereby suffocate
shoreline vegetation. The decay of this plant material will further
depress oxygen levels. Vigilant observations of water quality,
particularly dissolved oxygen concentrations, following a runoff
event are necessary to inform the culturist of the potential re-
quirement for emergency aeration.
Cage culture is likely to be a part-time activity, secondary to
the main enterprise of the farm or ranch. The quantity of labor
allocated to feeding fish in cages must be minimized if successful
integration with the other farming activities is to be realized.
Demand feeders can reduce the labor required for feeding by 88

























Cage culture of tilapia in a golf course pond on St. Croix.


- 94%. Despite significant savings in labor for feeding, regular
observations of feeding response and fish health should be
maintained.
In summary, the results of the feeding studies indicated that
differences in growth, feed conversion, and marketable produc-
tion of fish fed according to a feeding rate table or by demand
feeders were slight. The use of demand feeders significantly
reduced the quantity of labor required for feeding fish. The
importance of this factor should be evaluated in terms of the
opportunity costs for labor in a commercial culture operation,
with due consideration of the local employment situation.


The UVI-AES Aquaculture Program extends its appreciation to
Annaly Farm, Mon Bijou Farm, Windsor Farm, Bethlehem Farm,
Fountain Valley Golf Course, and the V.I. Department of Agriculture
for use of their farm ponds.







Alfalfa Production
in the U.S. Virgin Islands
By Cyndi L. Wildeus
Agronomy Research Technician II

Alfalfa (Medicago sativa) is a highly nutritious, palatable per-
ennial forage crop that is widely adapted to areas that have well
drained soils but no highly developed hardpans. In such soils it
develops a deep root system which makes it drought tolerant, an
important consideration when selecting a forage crop for the
semi-arid tropics. Alfalfa is a good forage for all types of livestock
and may be grazed, ensiled, used as green chopped feed or made
into hay. It is chosen by dairymen particularly for its favorable
effect on milk production. For these reasons the following
experiment was designed to examine the feasibility of growing
alfalfa under St. Croix conditions.
The experiment was conducted on a Fredensborg clay soil
with a 2 to 5% slope. The soil pH ranged from 7.4 to 8.4. Average
annual rainfall for the area is 35 to 40 inches (875 to 1000 mm),
and the average temperature is 78 to 800F (25.5 to 26.60C). A
randomized block design with three replications was used for the
study. On December 6, 1985, twelve varieties of alfalfa were
innoculated and broadcast by hand at a rate of 20 Ib/a (22 kg/ha)
onto individual plots 5 feet wide and 12 feet long (1.5 m by 3.6 m).
The plots were rolled with a manual roller. Triple superphos-
phate fertilizer (0-46-0) was broadcast at the time of planting at
the rate of 150 Ib/a (168 kg/ha). Plots were harvested in strips 3
feet wide and 8 feet in length (0.9 m by 2.4 m) with a pick-up flail
mower. A total of five harvests were made at the physiological
stage of 10% bloom and a height of 2 to 4 inches (5.0 to 10 cm).
The fall armyworm (Spodoptera frugiperda) is a major pest
affecting alfalfa production on St. Croix. A regular spraying
schedule was followed to insure the insect population did not
increase rapidly and cause extensive damage. It was found that
insect damage may occur over a relatively short period of time (2
to 3 days). Therefore, the stand was checked frequently to deter-
mine when spraying was necessary. Previous research indicated
that fall armyworm populations increased rapidly following rain-
fall when lush new growth occurred. Therefore, frequent field
checks were made during the rainy season.
Funding Source: Hatch Project #014








Without the use of a pre- or post-emergence herbicide, several
hand weedings were necessary to remove broadleaf weeds.
Hand weeding post-emergence was necessary due to the
unavailability of herbicide and plots were periodically sprayed
with Lannate (brand, active ingredient methomyl) insecticide at
a rate of 10 ml Lannate/3.8 liters of water until the entire plot area
was treated. Post-emergence, approximately 60% of the stand
was infested with broadleaf weeds. Weeding was a necessary
step to keep the weeds from competing with the newly emerged
seedlings. In the established stand, broadleaf weeds were no
longer a factor. However, there was a slow invasion of grasses,
especially when the stand was stressed due to lack of moisture or
insect damage.
The varieties tested were selected from previous trials con-
ducted in South Texas. In St. Croix the top three producing
varieties, on a dry matter basis, were Arc, WL 318, and WL
Southern Special, respectively (Table 1), but were not signif-
icantly different from the other varieties (P>0.05). However, the
lowest yielding variety, Team, was significantly different from all


Table 1
Alfalfa harvest data taken on St. Croix.

Total Dry Rank by
yield Height matter maturity
Variety (kg/ha) (cm) (%) (% bloom)

Arc 9124 34 33 7
WL 318 8806 31 34 8
WL South. Sp. 8661 35 34 3
Cimmaron 8562 38 33 4
Baron 8120 37 35 2
Classic 7648 37 34 6
Florida 77 7302 41 35 1
Raidor 6712 33 34 5
Pioneer 555 6215 31 35 12
HiPhy 5967 31 35 10
Weevlcheck 5876 28 35 11
Team 5559 30 34 9





































Harvesting with flail mower.
other varieties. The earliest varieties to reach the physiological
stage for harvesting (10% bloom) were Florida 77, Baron, and
WL Southern Special. These data suggest that variety WL
Southern Special is the optimum choice under these growing
conditions if a high yield and early maturity is desired. Early
maturity is desirable for the following reason: Ten percent bloom
is the optimum stage to harvest alfalfa. If harvested prior to 10%
bloom, root reserves will become depleted resulting in poor
recovery, and eventually the loss of the stand. Therefore, the
earlier the stand matures, the sooner the grower can utilize the
forage. Height measurements were taken, but did not appear to
correlate well with the amount of dry matter produced. Dry
matter ranged from 33% to 35%.

42








In the U.S. Virgin Islands, water is the most limiting factor in
agricultural production. The data indicate that the amount of dry
matter produced correlates with the rainfall. Rainfall received
during the course of this study is shown in Figure 1. Due to its
poor distribution, rainfall appears to be a limiting factor in alfalfa
production. To compound the problem, the water retention
capacity of Fredensborg clay is relatively low, evaporation is high
and exceeds rainfall during some months of the year. Therefore,
drought periods of 3 or more weeks appear to exert a marked
effect on the amount of dry matter produced.


10o ----


5000 1


4000-


3000-


2000-


1000-


0 50 100


150 200
Days from planting


Figure 1. Yield of alfalfa in relation to rainfall.


U'\\ -


250 300


[ i i


P

















Alfalfa is a labor intensive crop.
Another long-term problem that must be addressed is the
apparent ability of alfalfa to deplete soil moisture reserves that
may not be replenished under natural rainfall conditions. Reduc-
tion of soil moisture under dryland conditions could affect sub-
sequent crops and a period of fallow may be necessary before
planting a subsequent crop.
This study, due to the intensive manual weeding, frequent
invasions of the armyworm and moisture stress was terminated
after one year of data collection. While specific herbicides were
economically feasible-in most cases, the price of shipping to
this location doubled their cost-their use was necessary to
prevent encroachment of broadleaf weeds and grasses. Without
herbicide usage to insure a good weed free stand, weeds became
damaging competitors during the dry season when alfalfa growth
is slow. This resulted in significantly reduced yields due to the
shading out of new growth. If manual weeding was not done
under these circumstances, the extended life of the stand could
not be maintained and costly reseeding would be necessary. For
these reasons, alfalfa is not recommended to the grower on St.
Croix who is unable to fertilize, plant, and maintain it properly. It
was found to be a labor intensive crop and is not feasible for
those who do not have the necessary equipment to maintain a
healthy stand. Therefore, it is recommended that those with
limited experience in alfalfa production and limited mechanical
resources should plant only a small acreage to determine how it
will grow on their farm.

The author wishes to acknowledge Stanley Simicek (Texas A&M
University) for his help in obtaining seed, Dr. W.R. Ocumpaugh (Texas
Agricultural Experiment Station) for his invaluable assistance in plan-
ning the experiment, and Dr. M.W. Michaud (University of the Virgin
Islands) for the use of his field personnel.


Aik









Recent AES Publications


Hupp, H.D. 1987. Milk production of Senepol cattle under Virgin
Islands conditions. Pages 25-30 in S. Wildeus, ed. Proceedings -
International Senepol Research Symposium. University of the
Virgin Islands, St. Croix, U.S.V.I.
Hupp, H.D., and A.R. Williams. 1987. Development and genetic history
of the Senepol cattle. Pages 9-13 in S. Wildeus, ed. Proceedings -
International Senepol Research Symposium. University of the
Virgin Islands, St. Croix, U.S.V.I.
Hupp, H.D., and D.W. Wright. 1987. Growth performance and carcass
characteristics of Senepol bulls under Virgin Islands conditions.
Pages 97-103 in S. Wildeus, ed. Proceedings International
Senepol Research Symposium. University of the Virgin Islands,
St. Croix, U.S.V.I.
McMillan, J.P. 1987. Ciguatera fish poisoning in the Eastern Caribbean.
Virgin Islands Perspective Agricultural Research Notes 2:12-15.
McMillan, J.P., P.A. Hoffman, and R.H. Granade. 1986. Gambierdiscus
toxicus from the Caribbean: A source of toxins involved in Cigua-
tera. Marine Fisheries Review 48:48-52.
Michaud, M.W. 1986. The establishment of improved pastures in the
U.S. Virgin Islands. Pages 118-27 in R.T. Paterson, ed. Proceed-
ings of a Workshop on Pasture Research and Development in the
Eastern Caribbean. Caribbean Agricultural Research and Devel-
opment Institute, St. John's, Antigua.
Michaud, M.W., and P.J. Michaud. 1987. The forage situation in St.
St. Croix. Pages 97-111 in Conferencia Sobre Produccion,
Manejo y Utilizacion de Pastos y Forrajes. Universidad de Puerto
Rico, Mayaguez, P.R.
Michaud, M.W., and P.J. Michaud. 1987. Naturally-occurring legumes
in the pastures of St. Croix. Virgin Islands Perspective Agricultural
Research Notes 2:7-11.
Padda, D.S. 1987. Development of Senepol Cattle: A Collaborative
Research Story. Pages 1-4 in S. Wildeus, ed. Proceedings -
International Senepol Research Symposium, University of the
Virgin Islands, St. Croix, U.S.V.I.
Petersen, A.C. 1987. Yield results of vegetable varietal evaluation trials
in the U.S. Virgin Islands. Virgin Islands Perspective Agricultural
Research Notes 2:16-18.








Rakocy, J.E., and A. Nair. 1987. Integrating fish culture and vegetable
hydroponics: problems and prospects. Virgin Islands Perspective
Agricultural Research Notes 2:19-23.
Wildeus, S. ed. 1987. Senepol Cattle: Proceedings International
Senepol Research Symposium. University of the Virgin Islands,
St. Croix, U.S.V.I.
Wildeus, S. 1987. Performance of Senepol cows at two locations on St.
Croix. Pages 15-23 in S. Wildeus, ed. Proceedings International
Senepol Research Symposium. University of the Virgin Islands,
St. Croix, U.S.V.I.
Wildeus, S. 1987. Growth and reproductive characteristics in a flock of
V.I. White (St. Croix) hair sheep. Virgin Islands Perspective -
Agricultural Research Notes 2:2-6.
Wildeus, S., and J.R. Fugle. 1987. Reproductive performance of
Senepol bulls on St. Croix. Pages 31-38 in S. Wildeus, ed. Pro-
ceedings International Senepol Research Symposium. University
of the Virgin Islands, St. Croix, U.S.V.I.
Wildeus, S., J.R. Fugle, and D.W. Wright. 1987. Scrotal circumference,
testis size and sperm production in bulls of the Senepol breed.
Journal of Animal Science 65:424 (Supplement 1).
Wildeus, S., and D.W. Wright. 1987. Growth and age at first calving in
Senepol x Charolais cross cattle on St. Croix. Pages 91-96 in S.
Wildeus, ed. Proceedings International Senepol Research Sym-
posium. University of the Virgin Islands, St. Croix, U.S.V.I.




UNIVERSITY OF VIRGIN ISLANDS

SIIIIII III IIIIIIIIIII IIii
3 3138 00175 2574































i. U'SRiTY OF THE VIR6!N lHANMr
















































Disclaimer Statement


V.I. Perspective is published biannually by the University of the Virgin Islands Agricultural Expet ir.t
Station. Contents of this publication constitute public property. The written material maybe reprii d if
no endorsement of a commercial product is stated or implied Please credit the University of the ,rgm
Islands Agricultural Experiment Station. Trade names or products occasionally are printed. No
endorsement of products or firms is intended, nor is criticism implied of those not mentioned.
The University of the Virgin Islands, including the Agricultural Experiment Station, 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, handicap, age or veteran status.


-i



0.



ii''


All