The Toxicity of Synthetic Sea Salts and Natural Seawater to the Development of White Sea Urchin (Lytichinus pictus) Larvae

Aquatic Research Laboratory, Marineland, Moorpark,
California 93021

Two bioassay tests were conducted
to evaluate the effects of recently mixed commercial synthetic
sea salts (SSS) and natural seawater (NSW) on the development of
larvae of the White Sea urchin Lytichinus pictus.
Freshly fertilized sea urchin eggs were placed in solutions of
the different SSS and NSW for 72 h at room temperature after
which percent fertilization and larvae development were accessed.
The results showed that no SSS performed significantly better or
worse than any other SSS in terms of percent normal development
of the sea urchin larvae. In one bioassay it was clearly
demonstrated that the use of a natural seawater product (Catalina
Sea Water) resulted in significantly poorer development of newly
fertilized sea urchin larvae. No relationship was found between
decreasing percentage of normal sea urchin development and
increasing amount of total trace element concentration. The
results of these tests show that SSS are not detrimental to newly
fertilized sea urchin eggs, and by extension, marine aquaria.


A frequent question in the marine aquarium hobby in whether
synthetic sea salts are an acceptable substitute for natural
seawater for use in marine fish and coral reef aquaria; the
assumption being that the natural seawater available to hobbyist
is of the best quality. One way to evaluate this question is
through the chemical analysis of major, minor and trace elements.
However, chemical analyses do not always provide a definitive
answer because: 1) the general chemical composition of most SSS
mirrors that of NSW, at least in terms of the concentration of
many major and minor elements, 2) the analysis and chemistry of
trace elements in seawater is difficult to accurately access and
3) certain chemicals and elements could act synergistically and
increase, or decrease, the toxicity of the SSS.


Another, non-chemical but biological, way to evaluate and
compare SSS to each other and NSW is through the conducting of
tests called bioassays. As the name implies, a bioassay is a test
that relies on a biological indicator rather than chemically
determined values. In assaying the toxicity of a substance a
bioassay may yield more valuable information than a chemical
analysis. This is because a bioassay assesses the potential of
the complete substance to cause biological harm rather examining
discrete chemical units.

In general, a bioassay is a test in which a living organism is
exposed to the sample in question for a discrete period of time.
Many times a sample will be tested in a series of bioassays that
are conducted on different life stages of a certain organism or
on organisms from many different phyla spanning several trophic
levels. The question to be answered by the bioassay might be: Can
the eggs of fish species X survive and hatch while exposed to
water from the discharge of an industrial plant? Or will fry of
fish species Z survive 96 hours in the same sample? If the eggs
of fish X do not hatch or fish Z does not live for 96 hours, then
it is a fair assumption that there is a substance or substances
in the sample that are toxic to the fish (the test would include
controls to make sure the fish can survive in water that is known
to be good).

However, a bioassay does not identify the toxic substance(s).
Only through chemical analysis and more thorough testing can the
exact toxic substance(s) be identified.

Sea urchin eggs and larvae have been used in bioassay tests
for over 20 years (Jonczyk 2001). Standard experimental
procedures have been developed and published by the U.S.
Environmental Protection Agency (EPA 2002) that are used by most
laboratories. There are many varieties of these tests that
include acute and sublethal toxicity testing. Sea urchin
bioassays are commonly used to test the toxicity of industrial
effluents into the ocean, to examine sediment pore water, and the
effects of run-off into the ocean, to name only a few.

In this report, sea urchin fertilization bioassays were used
to access the survival and development of newly fertilized sea
urchin eggs in freshly prepared samples of several brands of
synthetic sea salts and samples of natural seawater. A previous
report (Hovanec and Coshland, 2004) detailed the analysis of
fifteen trace elements in these same brands of synthetic sea
salts and samples of natural seawater.

Methods and Materials

The toxicity tests were conducted according to the EPA method
1008.0 Sea Urchin, Arbacia punctulata, Fertilization
Test, Section 15 in Short-Term Methods for Estimating the Chronic
Toxicity of Effluents and Receiving Waters To West Coast Marine
and Estuarine Organisms (EPA 2002). Modifications to this method
are noted below.


Sea urchins. Adult White Sea urchins
(Lytichinus pictus) were purchased from Marinus
Scientific (Long Beach, CA) and held in 37 L aquaria filled with
Catalina Seawater and located in a constant cold room
(temperature=10°C) until used in the experiments. Males were held
separately from females once sex was determined. The sea urchins
were constantly supplied kelp collected from the beaches at
Malibu and Venice, CA and Red Ogo purchased from Sea Dwelling
Creatures (Los Angeles, CA). Water conditions in the aquaria were
monitored twice a week for salinity, pH, ammonia, nitrite and
nitrate. Water changes using fresh Catalina Seawater were
performed as needed. Each aquaria had a fluorescent light which
was on a 12 hours on; 12 hours off cycle.

Collection and fertilization of gametes.
Fourteen sea urchins were selected from 10ºC holding tanks and
placed oral side up in a Petri dish lined with several
seawater-moistened paper towels. Urchins were injected with 0.5 M
KCl using a 3 ml syringe with a 27.5 gauge needle through the
soft periostomal membrane of each sea urchin. Each urchin
received between 0.2cc to 0.5cc depending on its size. After each
injection, the needle was rinsed with hot tap water to avoid
accidental injection of male gametes into female gametes and vice
versa. Sea urchins were held oral side up and gently swirled to
induce spawning. The urchins were then placed oral side down in a
tray lined with seawater-moistened paper towels. Gametes were
collected for the first 15 minutes after each urchin started
releasing. If spawning did not occur within 5 or 10 minutes, a
second injection of 3/10cc of 0.5 M KCl was injected. If spawning
still did not occur, the urchin was not used.

(i) Collection of sperm. Semen was collected
dry (directly from the surface of the sea urchin), using a 200 µl
pipette with the end of the tip cut off so that the opening was
at least 2 mm. The semen from each male was then placed into
separate 1.5 ml micro centrifuge tubes and stored in an ice water

(ii) Collection of eggs. The initial release
of eggs was rinsed off with Catalina seawater and disposed of as
the initial release is thought to contain urchin waste. The
remaining eggs were rinsed off from each individual female into
50 ml beakers with Catalina seawater. This is referred to as the
egg solution.

(iii) Checking for sperm viability. In a
counting chamber, a dab of semen was combined with 1 ml of
Catalina water, and then examined at 400x under a Zeiss
microscope in light phase. Semen was deemed viable when at least
90% possessed a tail that moved in a spiral motion causing a
wiggling movement.

(iv) Checking for egg viability. A small
sample of eggs from each female was placed into a counting
chamber and then examined at 100X under a light microscope. Eggs
were considered normal if they were round and revealed a
pronucleus. Eggs were deemed viable when at least 90% were
considered normal.

(v) Verification of fertilization. To check
for fertilization, a sperm solution (a.k.a. dilution) was
necessary. In a 1.5 ml epi 50 µI of sperm from all the males was
combined and vortexed. In a 50 ml falcon tube, 12.5 µL of pooled
sperm was placed into 25 ml of seawater. This solution was then
stored in an ice water bath for no more than 1.5 h. Every 1.5 h a
new sperm solution was made. Fertilization was tested for each
individual egg solution with the sperm solution. In a 1 ml micro
centrifuge tube, 1000 µL of sperm solution and 375 µL of egg
solution were combined. After 20 min fertilization was checked.
Fertilization was positive when a fertilization membrane formed
around the eggs. Sperm-egg pairs were accepted when at least 90%
of the eggs acquired a fertilization membrane.

(vi) Pooling and dilution of eggs. Once the
eggs were deemed viable, they were pooled together and kept at
10ºC. Eggs were placed into a 1 L flask filled to 600 ml with
Catalina seawater. After 2 h of settling, 300 ml of seawater was
siphoned off with a sterile 25 ml pipette and the volume was
brought up to 500 ml with Catalina seawater. This is referred to
as the egg suspension solution.


(vii) Density determination. To determine the
density of the sperm in the solution 25 ml of the 50 ml of
freshly made sperm solution and 1 ml of 10% acetic acid was
placed into a 50 ml volumetric flask. The volume was brought up
to 50ml with Catalina seawater and mixed by inversion. 10 µL of
killed sperm solution was then placed on each side of the
hemocytometer and allowed to settle for 15 minutes. Eighty small
squares on each side of the hemocytometer were counted under 400x
magnification. For the density of the sea urchin eggs, 10ml of
the well mixed egg suspension solution was placed into a 1 liter
graduated cylinder that was brought to 1 L volume with Catalina
water. This is referred to as the egg dilution. One ml of well
mixed egg dilution was placed on a counting chamber and counted
at 100x magnification. The process was repeated and the mean of
two counts was taken.

(viii) Making of fertilized egg solution. The
recommended initial sperm to egg ratio for fertilization of eggs
is 500:1. The composition of this solution was determined by
using the equations set forth in the Environmental Protection
Agency method (EPA 2002). The initial fertilized solution was
comprised of 200 ml of the egg suspension, 520 ml of Catalina
Seawater and 18.8 ml of sperm solution. After 20 minutes only 5%
of the eggs were fertilized. Another 10 minutes passed with no
further progress. In an attempt to achieve at least 90%
fertilization, more sperm solution was added into the
fertilization mixture, with negative results. The final attempt
was to transfer the remaining concentrated sperm into the
fertilization mixture and wait 18 minutes. This was successful as
100% of the eggs were fertilized.

Synthetic sea salt treatments. The synthetic
salt salts were purchased from a variety of distributors of fish
and pet supplies. The synthetic salt salts tested in the first
bioassay were BioSea Marine Mix, Coralife, Crystal Seas Marine
Mix Bioassay, Instant Ocean, Red Sea, Reef Crystals, Tropic Marin
and Oceanic Sea Salt. A natural seawater product, Catalina Sea
Water, was purchased from Pete’s Tropical Fish in Simi
Valley, CA. In the second bioassay Instant Ocean, Reef Crystals,
Catalina Sea Water and Oceanic Sea Salt were tested.

One liter of each of the synthetic sea salts was made by
dissolving an appropriate amount of the sea salt in deionized
water (DI) and adjusting the salinity to 33 ppt that matched that
of Catalina Seawater. The pH was not adjusted for any solution
but was measured after each solution was made up and again after
24 h, just prior to the commencement of the test. All sea salt
mixtures were made the day before the experiment began and
allowed to sit overnight at room temperature without

For toxicity testing, 10 ml of each treatment was added to a
glass vial. There were four replicates of each treatment.
Treatment vials were numbered such that researchers determining
larvae development did not know the identity of the actual
treatment in the vial.

Toxicity test procedure. The toxicity of the
sea salts was assayed by adding 250 µl of the well mixed
fertilized egg solution to each vial already containing the
various sea salt treatment solutions using a pipette tip that was
cut to provide an opening of at least a 0.5 mm. The vials were
not mixed after introducing the eggs. All vials were maintained
at room temperature near a closed laboratory window that allowed
natural light into the room for 72 hours. Air temperature was
continuously monitored and ranged from 18.8 to 24.6°C during the
72 hours test period. The vials were exposed to laboratory
fluorescent lighting of approximately 8 h light:16 h

After 72 h sea urchin development was arrested by adding 500
µl of 1.0% glutaraldehyde to each vial. Each vial was tightly
capped and incubated for 1 day at room temperature. The next day
all but about 1 ml of overlying liquid was removed from each
vial. The embryos in the treatments were examined under 100x
magnification by two researchers over the next three days after
the glutaraldehyde was added. The first 100 embryos seen in each
vial were examined and scored as being normal, abnormal or

Copper toxicity testing. A bioassay on the
effect of copper on the development of the sea urchin larvae was
also conducted as part of this experiment. A stock solution of
copper was made by adding 0.4233 g of CuCl2
(anhydrous) to 100 ml of DI water. 150 µl of this stock solution
was then added to 100 ml of DI to make the working stock
solution. Five treatment concentrations (3.0, 6.6, 13.9, 20.4,
and 30.0 µg/L Cu) were made up by adding 100, 220, 460, 680 and
1,000 µl, respectively, of working solution to volumetric flasks
and diluting each with Catalina Sea Water to a total volume of
100 ml. There were four replicates of each treatment. Fertilized
sea urchin eggs were added to each vial as described for the sea
salt experiment. Vials were number coded and mixed in with the
vials of sea salts such that the researchers scoring the vials
did not know the identity of the vials with copper.


Results and Discussion

Of the fourteen sea urchins injected, five were female, seven
were male, and two were not determinable. Four of the five female
urchins were used, as one urchin was unable to produce enough
gametes for use in the test. Sperm from six of the seven male
urchins were used, as one of the urchins did not produce enough
sperm. Fertilization was deemed viable in all of the sperm egg
combinations used for this experiment. Sperm counts were within
80% of each other and the mean was determined to be 19,150,000
sperm/ml. For the sea urchin eggs, the mean of two counts was
3,600 eggs/ml.


Determination of whether a sea urchin larva was normal or
abnormal can be subjective. Figures 1 and 2 show normally
developed urchin embryos. They are characterized by:

  1. having lateral arms at half the length of the body,
  2. guts sectioned into three parts,
  3. “belly” portion of the anterior lateral arms
    present although the anterior lateral arms are not completely
  4. overall “arrowhead-like” appearance, and
  5. crisp sharp image.

Figures 3 and 4, on the contrary, show abnormally developed
urchin embryos. They are characterized by:

  1. a visible fertilization membrane,
  2. dark blobs of cells,
  3. missing any section of the gut,
  4. larvae still at their blastula or gastrula stage,
  5. missing lateral arms, and
  6. rods growing in abnormal directions.

Figure 5 shows an embryo that falls in between the clearly
normal or clearly abnormal. These urchins are past their blastula
and gastrula phase, they possess all the characteristics of being
normal but their body parts are not fully developed. Perhaps
these urchins are normal, but just slow in their development.
Because these were difficult to categorize, the “in
between” embryos may have either been counted as normal or
abnormal depending on the researcher.

The pH values for the treatments on the days the salts were
mixed and 24 h later just before the start of the test are
presented in Table 1. Sea urchin larvae have been shown to
experience poor fertilization, development and survival at high
pH values (>9.0). None of the treatments had pH values that
would be considered detrimental to sea urchins.

Table 1. pH values of the synthetic sea salts
used in Bioassay Test 1 one hour after they were mixed and 24 h
later just before the fertilized sea urchin eggs were added

Sea Salt BrandpH on day sea salt madepH 24 h later just before start of bioassay
BioSea Marine Mix8.578.60
Crystal Sea Marine Mix8.518.60
Instant Ocean8.158.18
Red Sea8.688.69
Reef Crystals8.158.19
Tropic Marin8.218.26
Catalina Sea Water*ND**7.96

* A natural seawater product

** not determined

The results (mean of four replicates with standard error
bars) of the copper toxicity test are presented in Fig. 6. The
toxicity trend shows the classic dose response curve indicating
decreased percent normal sea urchin development with increasing
copper concentrations. The importance of this chart is that it
shows that the fertilized sea urchin eggs were viable and do
respond to increased concentrations of specific toxins.


In Fig. 7 the data from Fig. 6 is combined with the results
from two other studies on copper toxicity to sea urchins. One of
the primary reason for the detailed protocols, such as those
published by the EPA, is so that one can compare their results
with others who have previously done the same test. While, of
course, there will be some variation in the results between
tests, researchers and facilities, these types of comparisons are
useful because they allow one to know if their results are
reasonable as the results should be similar to what other
researchers have found. As seen in Fig. 7, the copper toxicity
curve from this study agrees well with the copper toxicity curves
from the two other studies. These results tell us that our
techniques are scientifically acceptable as we compare well with
other laboratories that have run the same types of bioassays.


The mean percentage normal sea urchin larvae development after
72 hours for seven synthetic sea salts and one natural seawater
product with standard error bars (N=4) tested in bioassay 1 is
shown in Fig. 8. A statistical analysis of the data using
analysis of variance (ANOVA) was conducted and determined that
the means of the eight treatments were equal (Table 2). This
indicates that there were no differences between any of the
treatments in terms of mean percentage of normal sea urchin egg

Table 2. ANOVA of all the treatments and replicates in
bioassay test 1 showing no significant differences between any of
the treatments. Note that in this and subsequent ANOVA tables the
P-Value (P stands for probability) for a significant difference
is 0.05 or less. Generally, a level of 5% (0.05) or less is set
by scientists as signifying that the test results show a
significant trend or difference. For instance, in bioassay test 1
the null hypothesis is “The mean percentage of normal sea
urchin development will be equal in all the treatments”. To
reject the null hypothesis, the P-value would need to be 0.05 or
less. The value determined in this test is 0.0944 and therefore
the null hypothesis is accepted – all the treatments are the


However, note that the data in Table 2 shows a large
difference in the standard error values of the eight treatments
(this can also be seen in Fig. 8). Further investigation into the
data revealed a possibility of there being a significant
difference in the scoring between the researchers. The samples
were read by one of two researchers in a blind format. This means
that they did not know what sample they were scoring. Each had
the same guide on larvae development from which to rate the sea
urchins. But was one researcher “tougher” than the


Even though the researchers randomly picked the samples,
samples from each treatment were read by each of them (meaning no
one individual read all four replicates from one treatment). This
allowed a statistical analysis of the results to determine if
there was a significant difference between researchers. Table 3
provides the results of an ANOVA analysis comparing the
researchers to each other. The analysis shows that there was a
significant difference between the two researchers with the
percent normal egg development values of Researcher 2 being
consistently lower than the values determined by Researcher 1. It
seems that Researcher 2 was a tougher scorer. This means that any
bias, if indeed there were a bias, would be towards a lower
percentage of normal sea urchin development.

Table 3. ANOVA of all treatments and replicates in bioassay
test 1 with Researcher (Tech 1 or Tech 2) who scored each sample
taken into consideration. Results show a significant difference
between the scoring of the two researchers, the P-Value is much
less than 0.05.


Next, the set of values for each treatment were tested for
outliers (Anderson, 1987). This analysis determined that four
values (one value from each treatment of Coralife, Crystal Sea
Marine Mix, Red Sea, and Reef Crystals) were outliers. Each of
these values was the lowest value within their treatment set.
These four values were excluded from the data set and the ANOVA
recalculated. The ANOVA determined that the significant
difference between the two research technicians disappeared
(Table 4).

Table 4. ANOVA of all treatments in bioassay test 1 but with
one outlier of each of the treatments of Coralife, Crystal Sea
Marine Mix, Red Sea and Reef Crystals removed with Researcher


Table 5. ANOVA of all treatments in bioassay test 1 but with
one outlier from Coralife, Crystal Sea Marine Mix, Red Sea and
Reef Crystals removed from the analysis.


The ANOVA of the SSS and NSW was also recalculated with the
above four data points removed (Fig. 9). This analysis determined
that the mean normal value of the sea urchin larvae development
was not equal in all the treatments (Table 5). However, this
analysis does not disclose which treatments are different from
the others.


To determine which samples were different, an analysis called
the Tukey Honestly Significant Difference (Tukey HSD) with the
Spjotvoll-Stoline modification due to unequal sample size among
treatments was conducted. This test compares the means of each
treatment in a certain order starting with the largest mean
against the smallest mean. Once those comparisons are finished
the second largest mean is compared with the next smallest mean
and so on (Zar 1974). The results can then be presented as means
that are considered equal.

The result of this analysis splits the treatments into two
groups (Fig. 10). The first group, which are statistically equal,
consists of Reef Crystals, Red Sea, Catalina Seawater, BioSea
Marine Mix, Instant Ocean and Tropic Marin. When Reef Crystals
(the lowest scoring treatment) is not considered in the analysis,
as per the standard procedure of this test, the means of all the
remaining treatments (Red Sea, Catalina Seawater, BioSea Marine
Mix, Instant Ocean, Tropic Marin, Coralife and Crystals Sea
Marine Mix) are considered equal (Fig. 10).

Fig. 10. Graphical summation of the
comparisons of individual group differences using Tukey HSD with
unequal N (Spjotvoll-Stoline test) for the sea salt treatments

Reef Crystals*
Red Sea*
Catalina Seawater
BioSea Marine Mix
Instant Ocean
Tropic Marin
Coralife*Crystal Sea Marine Mix*

* There were only three replicates used in the analysis for
these treatments.

The results from the above experiment demonstrate that SSS are
not “poisonous” to sea urchin larvae. Furthermore, no
one SSS can be considered better, or worse, than the others.
While one might be tempted to consider the sea salts who returned
lower mean values of normal sea urchin development to be
“worse” than others, this view is not supported by the
statistical analysis.

Furthermore, a significant difference could only be determined
if certain values were excluded from the analysis. While this
seems to be justified on the basis that it did eliminate the
difference between the scoring of the two researchers, it also
increases the chances of making a Type I error. In statistics, a
Type I error is when one incorrectly rejects the null hypothesis.
In this experiment, the null hypothesis is that the mean
percentage normal sea urchin development of all the treatments is
equal and the null hypothesis is true if the four replicates are
not removed from the analysis. Overall what this means is that
there is very little, if any, difference in the performance of
the various SSS and NSW in the first bioassay test.

A second bioassay was conducted to investigate Oceanic Sea
Salt a new sea salt that came on the market after the first
bioassay had been completed. This test was run in exactly the
same fashion as the first bioassay except only one researcher
(the second one who was “tougher”) read all the vials.
Three synthetic sea salts (Instant Ocean, Reef Crystals and
Oceanic Sea Salt) and one natural sea salt product (Catalina Sea
Water) were included in this test for a total of four treatments.
A copper toxicity test was also performed as part of this
bioassay (Cu concentrations of 6.6, 13.9 and 30.0 µg/L were

The initial pH values of the four treatments were normal (data
not shown). The means, standard deviation and standard error for
the second bioassay are presented in Table 6. One vial
(replicate) of the Oceanic Sea Salt treatment was dropped during
counting so there were only 3 replicates for this treatment. The
other treatments had four replicates each. Analysis of the
results (ANOVA) showed that the percent normal development in the
three synthetic sea salts (SSS) was significantly different
(better) compared to the Catalina Sea Water but there were no
differences between the three SSS.

Table 6. Mean, standard error, ANOVA and posthoc analysis
(Tukey HSD with Spjotvoll-Stoline modification) of results in
bioassay test 2. N equals 4 (except for Oceanic Sea Salts


The mean percent normal development of sea urchin larvae in
Catalina Sea Water was 9.0. This value was the lowest of all the
treatments in the two bioassays. Further evidence that this batch
of Catalina Sea Water was toxic to sea urchin larvae comes from
the results of the copper toxicity testing. Catalina Sea Water
was used as the sea water solution in the copper toxicity testing
of both bioassays. A comparison of the copper results from the
two bioassays is shown in Fig. 11. The second copper bioassay
with Catalina Sea Water showed very poor development for all
concentrations of copper. This is in contrast to the results of
the first bioassay (Fig. 6) and other bioassays (Fig. 7), which
have demonstrated that at low concentrations of copper sea urchin
larvae development normally. While the results of the second
bioassay clearly show that the second batch of Catalina Sea Water
was toxic to sea urchin eggs/larvae, it does not tell us why.
Trace element or other chemical analyses were not conducted on
this sample so one cannot speculate as to the definitive reason
this sample was toxic.


The statistical results for the three SSS show that they were
not significantly different in terms of percentage normal sea
urchin development (Table 6). For comparison purposes the mean
values for the treatments of bioassay 2 and the values for the
same treatments in bioassay 1 are presented in Fig. 12.


The results of the two bioassays demonstrate that the eight
brands of synthetic sea salts do not significantly retard the
development of sea urchin larvae that have spent their first 72
hours after fertilization in these mixtures. The results for the
natural seawater product Catalina Sea Water were mixed. In one
bioassay, there was no significant effect on the sea urchin
larvae development. However, in the second bioassay there was
almost no normal development of the sea urchin larvae in Catalina
Sea Water.

Of course, the more important questions, which can only be
discussed and not definitively answered, are: How relevant are
sea urchin bioassay results to marine aquaria? How can one
correlate or extrapolate the results of these tests to the
survival of organisms in marine aquaria? As previously mentioned,
many times bioassays are conducted on many different life stages
of an organism and on many different trophic levels in an
environment. Research has shown that fish eggs can be much more
susceptible to certain toxins than adults of the same species,
while the toxicity to fry falls in between the two extremes.
However, generalizations are difficult because much relies on the
actual species and environmental conditions. For example, copper
is known to be much more toxic in soft freshwater (below 50 mg/L
CaCO3) versus hard water (above 150 mg/L CaCO3).


None of the synthetic sea salts examined in this test severely
retarded the development of newly fertilized, in contrast to the
natural seawater product Catalina Sea Water, and thus there is no
evidence to suggest that the synthetic sea salts in these tests
would be detrimental to organisms in marine aquaria including
reef tanks.

Of course, this report only includes the results of two
bioassays and only a portion from each bag of SSS was utilized in
the bioassays. Two potential criticisms might be that:

  1. an investigator should mix up the entire bag of sea salt
    and take a sample from that mixture to run the test and
  2. the bioassays should be conducted multiple times over a
    longer time period of time to cover different production runs
    and the possible variation that comes with any manufacturing

To the first point, modern manufacturing processes are used to
completely blend the SSS on the market these days. Further, many
times only a portion of a bag of SSS is used at one time for
doing small water changes etc. Thus the procedures used in the
bioassays do reflect how SSS are used in many situations.

A survey of the scientific literature can be used to address
the second point regarding conducting multiple bioassays. A
review of a number of seawater toxicity tests published in the
journal of SETAC (Society of Environmental Toxicology and
Chemistry) reveals that between the years of 1996 into early
2004, 43 studies were published that used a synthetic sea salt
(Fig. 14). Of these 43 studies, 35 of them used Instant Ocean
with the next most common SSS being Crystal Sea Marine Mix that
was used in 4 studies. Fig. 14 also presents a breakdown of the
tests by organism group in which Instant Ocean was used. If a SSS
were toxic it would quickly be discovered from all the aquatic
testing that occurs in laboratories from around the world. This
wide range of experiments and organisms used in the tests
conducted in multiple laboratories is ample evidence that SSS,
especially Instant Ocean, are not toxic to aquatic organisms and
the hobbyist can rest assured that they are maintaining their
marine aquaria with a high quality product.


Furthermore, in one of the 43 studies mentioned above, Jonczyk
et al (2001) compared natural seawater to a synthetic sea salt in
a large series of sea urchin fertilization bioassays that spanned
4 years. These investigators used the white sea urchin,
Lytichinus pictus, in a series of bioassays split into
two phases to determine if there were any differences between the
two marine waters when used to adjust salinity in bioassays or
when the tests were conducted in 100% of these waters. They
compared Instant Ocean to Atlantic Ocean water and Pacific Ocean
water and found that when comparing hypersaline brine to dry
salts “it is our view that the dry salts (Instant Ocean)
method should be used for adjusting salinity in these
tests.” When comparing the sea urchin fertilization toxicity
tests when either natural seawater or synthetic seawater is used
to dilute a number of industrial effluents they conclude that
synthetic seawater (Instant Ocean) prepared from dry salts can be
satisfactorily used for control/dilution water is these
tests.” Finally, when examining the results for the
bioassays on copper conducted in natural seawater and synthetic
sea salts (Instant Ocean) they found “that fertilization
rates are significantly higher when using artificial water for
dilution rather than natural seawater.”

In conclusion, the results of this study and many other
studies demonstrate that there is no evidence of toxicity of
freshly mixed synthetic sea salts to sea urchin larvae and no
evidence that using synthetic sea salts in a marine aquarium
would be detrimental to the health of the organisms in the

Literature Cited

  1. Anderson, R.L. 1987. Practical
    Statistics for Analytical Chemists
    . Van Nostrand Reinhold
    New York. 316p.
  2. Chapman, G.A., D.L. Denton, and J.M.
    1995. Short-term methods for estimating the
    chronic toxicity of effluents and receiving waters to West
    Coast marine and estuarine organisms. EPA/600/R-95-136.
    Cincinnati, OH: U.S. Environmental Protection Agency.
  3. EPA. 2002. Short-term Methods for
    Estimating the Chronic Toxicity of Effluents and Receiving
    Waters to Marine and Estuarine Organisms. 3rd
    Edition EPA-821-R-02-014.
  4. Hovanec, T. A. and J. L. Coshland. 2004. A
    chemical analysis of select trace elements in synthetic sea
    salts and natural seawater. Advanced Aquarist Online Sept. 2004.
  5. Jonczyk, E., G. Gilron and B. Zajdlik.

    2001. Sea Urchin Fertilization Assay: An evaluation of
    assumptions related to sample salinity adjustment and use of
    natural and synthetic marine water for testing. Environ.
    Toxicology and Chemistry 20(4): 804-809.

  6. Pilson, M.E.Q. 1998. An Introduction
    to the Chemistry of the Sea
    . Prentice-Hall, Inc. Upper
    Saddle River, N.J. 431p.
  7. Shuman, C.S. 1998. Evaluation of the US
    EPA Purple Sea Urchin, Strongylocentrotus
    purpuratus, Fertilization and Larval Development Test
    Methods. Master of Arts Thesis, University of California, Santa
    Barbara. 43p.
  8. Zar, J.H. 1974. Biostatistical
    . Prentice-Hall, Inc. Englewood Cliffs, N.J.
  Advanced Aquarist

 Timothy A. Hovanec

  (2 articles)

Leave a Reply