A history of artificial seawater for aquarists by Albert J. Klee

t was shortly before Christmas in 1853, a time when people were counting their blessings and eagerly anticipating the New Year. Philip Henry Gosse, however, was not one of them. The past few years had not been particularly joyous for him. In 1850, Warington was the first one to demonstrate the proof of principle of the balanced freshwater aquarium, and in 1852, he even preceded Gosse by nine months in publication priority for the marine aquarium.

In 1851, Gosse had recommended the system for the marine tanks at the new Fish House in the London Zoo, designed so as to not require replenishing, except for adding freshwater to replace that had evaporated; but by 1853, it was proving to be a failure and the Zoological Society of London had to keep buying expensive additional seawater in order to keep the exhibits alive. Then in July of 1853 when Gosse presented his bill for the marine specimens he had collected for the Fish House, the Society�s Council found that Gosse had also collected and sent specimens to the Surrey Zoological Gardens (a competitor to the London Zoo) during his time collecting in Weymouth but charged the Zoological Society for this time. The Council was incensed and angrily denounced his want of consideration for the Society. Gosse was henceforth persona non grata with the Zoological Society of London.

Parasitic anemone illustrated by Gosse

However, there was an opportunity for him to increase his reputation within the scientific community. He had come across a paper written by Edward G. Schweitzer who presented his results in analyzing the waters in the English Channel off the coast of Brighton (Schweitzer, 1839). Schweitzer, was a German physician and chemist, and Director of the Royal German Spa in Brighton. Schweitzer and Warington, by the way, knew each other. A list of the officers and members of the Chemical Society of London dated June 1843 shows Warington as one of its two Secretaries and Schweitzer as one of its members.I�ll explain Schweitzer�s interest by quoting the final paragraph of his paper:

Philip Henry Gosse (1810-1888) in 1855
I cannot conclude this paper without drawing the attention of medical men to the importance which the brine springs on the Continent have lately acquired, as, for instance, the springs near Kissingen, the Adelheids quelle, near Heilbroun, and above all, the springs of Kreugnach, which have been found highly beneficial in scrophulous diseases when internally administered, their action being dependent entirely on the chlorides, iodides, and bromides they contain. Seawater would afford similar advantages for bathing, and when evaporated to dryness, the residue might be kept in earthen vessels, and thus be conveyed to any distance; and as its constituents are very soluble, sea-water in perfection might be procured at any place. The evaporation of sea-water should be performed with care, and the ingredients kept by chemists. One great advantage would accrue from this method, viz. that sea-water could be had of any degree of concentration which the practitioner might deem necessary. At the baths of Kreugnach, for example, extraordinary effects have been produced when from 40 to 70 quarts of the mother liquor were added to the natural salt-water of that spring, and this mixture used for bathing.

It is amusing to note that the start of the development of artificial seawater for marine aquaria began

Robert Warington (1807-1867)

Page 2

with desire of the Victorians to immerse themselves in a tub full of water containing bath salts, and the more genuine or natural the salts the better!

Now there are a number of reasons why aquarists might be interested in artificial seawater rather than using the real thing. The density of the final solution is controllable, it is more stable, the formula can be varied for specific uses (e.g., for marine plants versus for marine animals) and it is sterile and thus diseases cannot spread. In Gosse�s time, however, the main reason was economics. It was a lot less expensive to reconstitute the seawater with local freshwater than to bring seawater to the aquarium. Water is heavy, its salts are not.

GOSSE�S PLAN

Gosse could have made up the Schweitzer formula, added marine animals and plants and, if everything worked out satisfactorily, he could have written the very first paper on artificial seawater for aquarists. However, based upon his past experiences he wondered if Warington was preparing a paper on this subject; there had been some rumors to the effect that he had experimented with artificial seawater. His concern that Warington might be preparing such a paper is understandable. Warington�s interest was but a pastime. In 1704, the House of Lords gave apothecaries the right to practice medicine, and so they may be viewed as forerunners of present-day family physicians. Furthermore, in 1815 the Apothecaries Act granted the society the power to license and regulate medical practitioners throughout England and Wales. Warington had an important job in Apothecaries� Hall and thus was limited in the time he could spend on aquarium research. For Gosse, on the other hand, a significant source of his income involved writing popular books on nature, and it was important for him to maintain a favorable reputation, not only among the masses, but with the scientific community as well. If he could write and publish a paper on artificial sea

water before Warington did, it would be a very great advantage in this regard.

Therefore, Gosse devised the following plan. He would visit Warington and steer the conversation in such a way that it would not be a consultation but Warington would talk about his experiences. Executing his plan, Gosse visited Warington in 1854 on two occasions at Apothecaries� Hall: on January 16th and again on January 21st. As it turns out, however, Warington did consider Gosse�s visits to be consultations. During these �chats,� Warington had no idea that Gosse already knew about Schweitzer�s paper. However, Gosse did discover that Warington had no plans to publish a paper dealing with artificial seawater. Gosse then went ahead with his plan to write his article and by April 21, 1854 he had devised an artificial seawater formula, made up a tank with the solution and added some marine invertebrates. Everything went well and on June 9th he submitted his paper to The Annals and Magazine of Natural History. On July 1954 it was published (Gosse, 1854) and Gosse had achieved his goal. For once, he had beaten Warington to the printed page and went down in history as the first to write about - and to provide a formula for - artificial seawater for aquarists!

Apothecaries� Hall where Gosse met with Warington on January 16 and 21, 1854, and discussed artificial seawater.

Page 3

For comparison purposes, Schweitzer had included the seawater formula based upon a sample from the Mediterranean Sea devised by G. Laurens. I wanted to find out something about him and why he did his analysis but unfortunately, Gosse had labeled him three times in his paper as �M. Laurent� (the M standing for Monsieur). Gosse wasn�t the only one who made this error, however. In an report made on the action of air and water on cast iron, wrought iron and steel, Robert Mallett (1841) also misspelled the name as �Laurent.� I finally went to Schweitzer�s paper - something I should have done in the first place - and found the correct spelling. I also consulted Laurens� paper and found out that Gosse had made errors in Laurens� calcium sulphate and calcium carbonate numbers, and had also omitted entirely Laurens� entry for magnesium carbonate. Clearly Gosse had been in too much of a hurry to get his paper published.

In any event, in the original source (Laurens, 1834) I found that Laurens was a pharmacy student in Marseille and had sent a note to the Soci�- t� de Pharmacie containing an analysis that he had made of the water of the Mediterranean Sea. The commentary on the article included the following (translated): �In summary, the note from Mr. Laurens, describing the composition of the water of the Mediterranean Sea in a new, well-done analysis, should interest those doctors who prescribe the use of sea bathing.� Thus, Laurens� purpose was also the same as that of Schweitzer�s. Unfortunately, Gosse had included the following paragraph in his paper:

Several scientific friends to whom I mentioned my thoughts, expressed their doubts of the possibility of the manufacture; and one or two went so far as to say that it had been tried, but that it had been found not to answer; that though it looked like sea-water, tasted, smelt, like the right thing, yet it would not support animal life. Still, I could not help saying, with the lawyers,

�If not, why not?� - Experientia docet [i.e., Experience teaches]. I determined to try the matter for myself.

From this, there were unexpected, disastrous consequences. One wonders what Warington really thought when he read Gosse�s paper. He must, at the very least, have been dumfounded but as will be seen, he actually only used the word �surprised.�

In any event, Gosse had miscalculated converting from grains to avoirdupois pounds and ounces, and had mistakenly assumed that the magnesium sulphate in Schweitzer�s paper represented the ordinary crystallized salt, and not the anhydrous sulphate that is always the case in giving analytical results. Furthermore, Gosse had eliminated the following from Schweitzer�s formula:

Magnesium bromide because of its small concentration and because Laurens did not find any in the water of the Mediterranean;
Calcium carbonate because calcium could be found in abundance in the fragments of shell, coral and calcareous material that made up the bottom of the aquarium;
Calcium sulphate because it was difficult to dissolve and because it was found by Laurens in a minute concentration.

Gosse�s decision to eliminate magnesium bromide and calcium sulphate simply because Laurens found them absent or in a minute concentration made no sense. The composition of Mediterranean waters differs from that of the water in the English Channel. The English Channel sits on top of layer of chalk and it was to be expected that its calcium content would be higher. Since Schweitzer�s analysis was appropriate for marine animals found off the southern cost of England where most specimens would be collected for British aquaria, it was the relevant analysis and logically the one that should have been followed.

Page 4

Warington responded to Gosse�s paper thusly (Warington, 1854):

In the �Annals and Magazine of Natural History� for July last, you published a short communication from Mr. Gosse, on the artificial formation of sea water, and having lately had my attention especially directed to this paper by a friend who wished to put the formula given into practice, I was surprised at the difference in the proportions of the ingredients as compared with what I had myself employed in the course of 1853, more particularly from the circumstance, that when Mr. Gosse called upon me in January last, and consulted me on the feasibility of the plan, I told him that there could be no difficulty in the matter, as I had made and had then in use several small quantities artificially produced, and that all that was required was that a good analysis should be taken as the basis for deducing the proportions, and at the same time referred him to the source from which I myself had worked, namely Dr. E. Schweitzer's analysis of the water of the English Channel taken off Brighton. Now, as numerous parties have been inquiring respecting this subject, and the erroneous formula has been copied into other journals, it may prevent much annoyance as well as disappointment if this matter is set right.

Upon reading Warington�s corrections to his paper, Gosse had two reactions, the first being a rather acrimonious one (Gosse, 1855):

If Mr. Warington supposes that I obtained from him one atom of information previously unknown to me, on the subject of making seawater from its constituent salts, he is most thoroughly mistaken. He is no less wrong in saying that I �consulted� him ; since I merely mentioned what was on my mind in familiar conversation. With this, however, the public are of course not concerned, and I shall say no more on that head.

In this regard I fault Gosse. Up to this point Gosse and Warington had been good friends and Gosse should not have kept Warington in the dark about what he knew, In any event, the two never corresponded or met again.

GOSSE DEFENDS HIS FORMULA

The second reaction was a defense of his formula. At one point Gosse states:

That Mr. Warington's calculations are correct I do not at all deny; but that they convict mine of �error� I by no means admit; as my facts will presently show. The �error� (which is of that kind technically called �nidus equ�) [a humorous comment meaning a �mare�s nest, implying that the criticism was a �humbug�] lies altogether on the other side.

Essentially Gosse claimed in his rebuttal that (1) his formula was ��for practical people, to whom minute accuracy is impossible and to whom a chemical formula expressed in quantities of four or five decimals would certainly act as a prohibition,� and (2) by actual experimentation, examples of Crustacea (crabs, etc.), Mollusca (sea snails, etc.), Annelida (worms, etc.) and Zoophyta (sponges, etc.), as well as Ulva and Conferva plants, thrived for over eight months. As can be seen from Table I, the differences between his formula and that of Schweitzer�s were large enough to raise some eyebrows. Ironically, with regard to simplicity it was Warington who suggested the simplest solution of all (Warington, 1854):

There cannot be a question that by far the simplest plan would consist in the evaporation of the sea water itself in large quantities at the source, preserving the resulting salt in closely stopped vessels to prevent the absorption of moisture, and vending it in this form to the consumer; the proportion of this dry saline matter being 56-1/2 oz. to the 10 gallons of water, less the 3 pints. This plan was suggested by Dr. E. Schweitzer himself

Page 5

for the extemporaneous formation of sea water for medicinal baths, and, on inquiry since writing the above, I find that such a preparation is manufactured by Messrs. Brew and Schweitzer of No. 71 East Street, Brighton, under the title of �Marine Salts for the instantaneous production of sea water.� Mr. H. Schweitzer writes me, that he has for many years made this compound in accordance with his cousin�s analysis. The proportion ordered to be used is 6 oz. to the gallon of water and stirred well until dissolved.

Now the major ionic components of seawater are the same as those ions responsible for maintaining the electrical and osmotic balance within the cells of living animals and for the transmission of nerve impulses. What is an ion? Well sometimes atoms gain or lose electrons. The atom then loses or gains a �negative� charge. These atoms are then called ions. For example, when common salt is dissolved the Na (sodium) loses an electron and the Cl (chlorine) gains an electron; the Na becomes Na+ (the sodium ion) and the chlorine becomes Cl- (the chlorine ion). The major ionic components are shown in Table II. The complete absence of any one of these major ions is fatal. A fish placed into an artificial seawater that contained

no calcium, for example, would soon die as a result of the chemical imbalance. Therefore, one of the theoretical problem�s with Gosse�s formula was that he had omitted the calcium. Fortunately for Gosse, laboratory grade chemicals are not 100% pure and technical grade chemicals are far less so, especially in Gosse�s day. Thus Gosse�s formulation not only contained the necessary calcium but it also contained trace elements, such as silica and iron, that known to be biologically active. In support of his assertion that his formula worked, Gosse provided the following postscript to his article:

P.S. Some letters which have been lately published by Mr. W. A. Lloyd in the �Athenaeum�' confirm my experience. Perhaps I may be excused for citing a few words contained in a private letter from the same gentleman to myself: � �In reference to what has recently been published on an improvement (or a supposed improvement) on your receipt for the manufacture of sea water, a friend of mine took me by the button and said, �My dear Sir, Mr. Gosse is altogether wrong; he has not

Table 1 THE SCHWEITCHER, GOSSE AND LAURENS FORMULAS COMPARED The numbers are pounds / 100 lbs of water ( = 10 imperial gallons)
COMPONENT	SCHWEITZER	GOSSE	% increase in (1) over (2)	LAURENS
Sodium Chloride	2.7060	2.1875	24%	2.722
Potassium Chloride	0.0766	0.0563	36%	0.001
Magnesium Chloride	0.3666	0.2813	30%	0.614
Magnesium Bromide	0.0029	-	-	-
Calcium sulphate	0.1407	-	-	0.015
Magnesium sulphate	0.2296	0.1563	47%	0.702
Calcium carbonate	0.0030	-	-	0.009
Magnesium carbonate	0.0030	-	-	0.011

Page 6

TABLE II THE SIX MAJOR IONIC COMPONENTS OF SEAWATER
COMPONENT	mg/l
Chloride	19,000
Sodium	10,500
Sulphate	2,600
Magnesium	1,350
Calcium	400
Potassium	380

salt enough; he has no�.�' To which I replied, pointing to two fine Actinia dianthus [an anemone] in full blow in one of my vessels, � �But if Mr. Gosse is altogether wrong, why do these Actini� flourish ?� This was unanswerable.�

However, except for the Crustacea, all of the animals studied by Gosse and Lloyd were invertebrates; no fish were involved.

OTHER PEOPLE CHIME IN

At first Gosse was criticized in the scientific community for his decisions made when he modified Schweitzer�s formula. George Wilson (1818- 1859), a Scottish chemist and physician, lecturer in the School of Arts, director of the Industrial Museum of Scotland and regius professor of technology in the university of his native city was reported as saying (Wilson, 1854):

Dr. Wilson considers, however, that the less abundant, but still essential, constituents of sea-water�such as carbonate of lime, sulphate of lime, phosphate of lime, fluoride of calcium, silica, iodine, and bromine�should not be absent, as these latter substances are found in marine plants and animals; and it is therefore plainly evident that the medium in which they live ought to contain the same substances.

However, after Gosse had learned of Wilson�s remarks, he sent Wilson two samples of his artificial seawater, both of which had sustained living plants and animals, one for six months and the other for ten months. By the next year, Wilson had completed his analyses of Gosse�s samples. The following is an extract from The Eclectic Magazine, reporting on the meeting of The British Association for the Advancement of Science in Glasgow in 1855 (Anon., 1855):

Professor George Wilson, of Edinburgh, read a paper on �The Chemical Changes undergone by Artificial Sea Water after Ten Months� Use in the Marine Vivarium.� The author stated that the communication which he now made to the section was in continuation of one read to it at the meeting in Liverpool last year. Mr. Gosse, the distinguished naturalist, who has done so much for the improvement of marine vivaria, had given him two specimens of artificial sea water, in which living plants and animals had been kept in full vigor for periods respectively of ten and six months. On analysis, it appeared that, whereas magnesia, sulphuric

George Wilson (1818-1859)

Page 7

acid, potassium, sodium, and chlorine, were the only substances originally present in solution in the artificial sea water; lime, phosphoric acid, silica, iodine, and iron now occurred in it. The lime [i.e., calcium oxide] was probably dissolved by carbonic acid [i.e., the name given to solutions of carbon dioxide in water] evolved from the animals; the phosphoric acid was taken up as phosphate of lime by the same gas, along with water ; the iodine was separated from the sea weeds ; the silica from the Infusoria and fragments of rock within the vivarium; and the iron from many sources. It was further stated, that certain important substances which were likely to be present could not be detected, owing, as the author believed, solely to the small amount , of water which could be spared from the vivaria not permitting a minute amount of such bodies as bromine, fluorine, ammonia, or nitric acid to be discovered. The success of Mr. Gosse�s artificial sea water was shown to be complete.

Wilson, therefore, found trace elements of calcium, phosphorus, silica, iodine and iron in the sample, and offered some ideas on how they arrived there. He neglected, however, to mention the impurities of the constituents and also that trace elements could find their way into the artificial seawater through the foods and the water used to reconstitute the seawater. He did note that most likely other substances, such as bromides, fluorides, ammonia and nitrates, were in the samples but the samples were small, defying detection using the analytical tools available at the time.

Aquarists were satisfied by this report coming from a scientific person of such high stature and began to experiment themselves with Gosse�s formula. However, a year later, Hibberd wrote (1856):

�But artificial water is quite unsuited for animal life of any kind, until it has been brought into condition by means of growing weeds for eight or ten days, and for Crustaceans, Starfishes, and Fishes proper, it is not suitable till it has been in use for many months, and even

then some species lose their health in it, and at last perish��

ARTIFICIAL SEAWATER
GETS A BIG BOOST

In his Hand-Book to the Marine Aquarium (June 23, 1873), Lloyd was still favorably disposed to Gosse�s formula: Such artificial sea-water answers very well: indeed the large aquaria at Hanover and Berlin have no other. It was Lloyd, by the way, who suggested that these two public aquaria use artificial seawater.

However, by 1879 Lloyd had changed his mind. He had secured the post of Superintendent of the Aston Aquarium and since it was located far from the coast where seawater was easily obtained and transportation of the water from the coast would be expensive, Lloyd decided that the use of artificial seawater was justified and selected the Schweitzer formula. Thus was the first use of artificial seawater in a public aquarium in England. In an article titled, �The Artificial Sea-Water at the Aston Aquarium�, William Southall (1879) wrote:

My firm (Southall Bros. and Barclay) was appointed to manufacture the water, and about fifty tons of chemical substances have been used. Each ingredient was subjected to analysis, and allowance made in every case for water of crystallization, hygroscopic moisture and impurities, and the various constituents of the well-water used were also allowed for in calculating the working formula. The analysis of Dr. Schweitzer was taken as a basis, supplemented by our own analysis of water recently taken near Brighton, a mile from the shore; and from the latter the data necessary for the required amounts of iodine, &c., were obtained.

Table III provides a comparison between the Schweitzer and Aston Aquarium formulas, the latter being supplied to Lloyd by H. W. Jones, analyst to Messrs. Southall Brothers and Barclay

Page 8

of Birmingham. Above the dotted line are the concentrations of the salts added to the well-water (as can be seen, the components and concentrations are essentially identical to those of the Schweitzer formula) and below are the components and concentrations of what was found in an analysis of the well water. Added together they are the actual components and concentrations in the Aston tanks.

It is clear from these measurements that the water added to a basic group of components is a major consideration with regard to trace elements. All freshwater contains one or more of those used in the Aston Aquarium, more or less, but usually more. Lloyd pointed out that the Thames at London contains about 0.02 or 0.03 pounds of trace elements instead of the roughly 0.01 lbs. per 100 lbs. of water of the Aston well water (Lloyd,1879). Considering the accuracy required for the correct

reproduction of certain of the minor constituents, the purity of the added salts and of the water used is critical, endangering the final formulation unless the quantity and quality of the impurities are known. For example, if the sodium chloride used is the commercial grade or household salt, the arsenic and barium levels (even though these are in minute quantities and can be measured in single figures in parts per million) can present problems unless accounted for. In addition to these two, sodium chloride of this grade contains as many as twelve other elements in substantial enough quantities to give rise to doubt as to what their final concentrations would be.

FORCHHAMMER'S PRINCIPLE

The concept of salinity was introduced in 1865 by the Danish chemist and mineralogist Johan Georg Forchhammer (1794-1865). Forchhammer

TABLE III THE ASTON FORMULA COMPARED TO THE SCHWEITZER FORUMULA The numbers are pounds / 100 lbs. of water (= 10 imperial gallons)
COMPONENT	SCHWEITZER	ASTON
Sodium Chloride	2.70595	2.72143
Potassium Chloride	0.07655	0.07857
Magnesium Chloride	0.36666	0.37643
Magnesium Bromide	0.00293	0.00400
Calcium Sulphate	0.14063	0.14429
Magnesium Sulphate	0.29578	0.23700
Calcium Carbonate	0.00330	0.00841

Calcium Carbonate		0.00329
Magnesium carbonate	0.00217
Silica		0.00148
Magnesium sulphate		0.00046
Oxide of iron and alumina		0.00022
Sodium Chloride		0.00201
Magnesium nitrate		0.00050
Sodium nitrate		0.00201
Potassium chloride		0.00006
Sodium iodide		traces
Ammoniacal salts & organic matter		slight traces
Those below the dashed line were present in the well water.

Page 9

Johan Georg Forchhammer (1794-1865)

worked under great disadvantages: his samples of water were brought home by seafaring men from different parts of the world in corked bottles, and they were necessarily all taken from the surface or immediately beneath it. Forchhammer did not attempt to determine quantitatively all the elements that occur in sea water, but confined himself to the very accurate estimation of the principal salt components, such as chlorine, sulphuric acid, magnesia, lime, potash and soda. Forchhammer found that the ratio of major salts in samples of seawater from various locations was constant. This constant ratio is known as Forchhammer's Principle, or the Principle of Constant Proportions. One of the most interesting of his scientific work, �On the Constitution of Sea Water at Different Depths and in Different Latitudes� (1863) started a new era in the history of ocean chemistry.

THE H.M.S. CHALLENGER EXPEDITION

The 1870s voyage of H.M.S. Challenger lasted 1,000 days and covered more than 68,000 nautical miles. Many consider it to be the first true oceanographic expedition because it yielded a wealth of information about the marine environment. Those aboard identified many organisms then new to science, and they gathered data at 362 oceanographic

stations on temperature, currents, water chemistry and ocean floor deposits. The scientific results of the voyage were published in a 50-volume, 29,500- page report that took 23 years to compile. Specialists in numerous scientific disciplines studied the collections and data, and helped produce the reports.

Also, the reports written by members of the Challenger expedition provided rich descriptions of the flora, fauna and cultures of the lands visited. Photography - new at the time - was highlighted as well, along with scientific illustration.

The HMS Challenger originally was designed as a British warship - a steam corvette in the Royal Navy - outfitted with 17 guns and an engine capable of over 1,200 horsepower. The 200-foot ship was three-masted, square-rigged and built of wood.

In 1870, Dr. Charles Wyville Thomson, a Scottish natural historian and marine zoologist, suggested that the Royal Society of London ask the British government for the use of one of its ships for an extended research cruise. The government agreed, and the HMS Challenger was modified to conduct oceanic research. Ammunition and 15 of the guns were removed from the ship and replaced with laboratories, workrooms, and storage space. The HMS Challenger used sails rather than the steam engine most of the time to allow for frequent stops when collecting data. The steam engine was used only during dredging operations to collect samples from the depths of the ocean.

At the start of the voyage in 1872, the science and ship crew consisted of six civilian/scientific staff, led by Dr. Thomson. It also included 21 naval officers, including

HMS Challenger, 1872

Page 10

Dr. Charles Wyville Thomson (1830-1882)

Captain George Nares (replaced by Captain Frank Thomson in 1875), and approximately 216 crew. When the voyage ended in 1876, only 144 crew remained on the ship. Seven people had died, five left when Captain Nares did, 26 were left in hospitals or were unable to continue the journey, and several had deserted at the various ports of call. The following are the �Editorial Notes� taken from the report on the scientific results of the voyage:

The physical and chemical investigations conducted by Mr. J. Y. Buchanan, during the three and a half years� cruise of H.M.S. Challenger, are among the most important and valuable of the Expedition.

Mr. Buchanan collected daily, with much care, samples of the surface, water, and determined the specific gravity. At all Stations, a slip water bottle was attached to the sounding line, and the specific gravity of the specimen of bottom

water thus collected was also ascertained. At every Station, where practicable, waters were collected from intermediate depths at 25, 50, 100,200, 300, 400, and 800 fathoms from the surface, with a stop-cock water bottle attached to a separate sounding line, under Mr. Buchanan's personal supervision. The specific gravity of these waters was also determined. The routine chemical work of the Laboratory consisted in boiling out the gases from, and in determining the carbonic acid in, as many samples as possible. A very large number of samples of sea-water were collected from the surface, bottom, and intermediate depths, and preserved in glass stoppered bottles. These were either sent homo along with other collections from various ports touched at during the Expedition, or brought home by the ship. It is difficult for any one, except those who actually witnessed the daily work at sea. to form an adequate idea of the labour, skill, and continuous effort required to carry on these observations in all sorts of weather, and to form, and bring home successfully, collections and observations like those which have resulted from Mr. Buchanan's exertions.

THE DITTMAR ANALYSES

The most comprehensive early study of the composition of seawater was that made in 1884 by Wilhelm Dittmar (1833-1892), a German-born chemist

The Challenger�s Natural History Workroom

Page 11

Wilhelm Dittmar (1833-1892)

renowned as a chemical analyst, on the 77 samples collected by chemist J.Y. Buchanan during the Challenger Expedition.

It should be noted that the analysis of seawater is primarily concerned with its ions, notably chloride, sodium, sulphate, magnesium, calcium and potassium. To construct a formula useful for the manufacture of artificial seawater, however, this information must be translated into specific chemical compounds. This is an arbitrary procedure since it doesn�t make any difference what chemical compounds and their concentrations are chosen as long as the mixture after being dissolved contains the ions in their correct amounts. An example is potassium. Schweitzer added it as potassium chloride and adjusted the amount of chloride elsewhere; Dittmar added it as the sulphate and adjusted the amount of sulphate elsewhere. Thus, combining the acids and bases in the arbitrary mode shown in Table IV, Dittmar held that the seven numbers in the table expressed as percentages of the total salts -

Column (1) - may be taken as holding approximately for any sample of seawater, in other words, the ratios of these components is constant. I quote Dittmar here:

From my analysis (which I do not pretend exhaust the subject), it would appear that the composition of seawater is independent of the latitude and longitude whence the same is taken. Nor can we trace any influence of the depth from which the same comes, if we confine ourselves to the ratio to one another of chlorine, sulphuric acid, magnesia, potash and bromine. I emphasize the bromine because, while present in very small proportion, it is taken up preferably by sea-plants, and consequently must be presumed to be more liable than any of the major components to at least temporary local diminution. And yet my analyses of the three mixtures of Challenger waters, and of the Arran water referred to, gave identical values for the bromine present per 100 of the chlorine. [As shallow shore waters did not occur in the series of Challenger samples, Dittmar also analyzed a water that had been collected for him near Port Arran, Scotland, at a shallow place where there was an abundance of marine vegetation.] But the determination of the lime in the same set of waters make it most highly probable that the proportion of this component increases with the depth.

In order to derive a formula for artificial seawater, one has to know the salinity, i.e., the total amount of the salts in the water. For Table IV, I chose the average ocean salinity of 3.5 grams of salts per 100 grams of seawater. By multiplying the percentage figures in Column (1) of Table IV by this number, I obtained the gram weights for each compound; these are listed in Column (2) as grams per 100 grams of seawater. As Warington showed in his critique of Gosse�s paper, we can consider these numbers as lbs./100 lbs. of water and the Dittmar and Schweitzer formulas, therefore, can be compared directly in Table IV. There are some significant differences, mainly in the magnesium sulfate, calcium carbonate, and magnesium bromide.

Page 12

If Dittmar�s formula may have been of some interest to public aquaria, it had no effect on the aquarium hobby. The interest in marine aquaria had waned because it was expensive and required careful attention on the part of the hobbyist. In this country Stage IV of the aquarium hobby (the Stage we are in now) started just before 1900, but the emphasis was on freshwater and, to a very limited extent, brackish water fishes. Some public aquaria, particularly those situated inland, did use artificial seawater but this had no real effect on the hobby until after World War II ended. This did not , however, stop oceanographers from devising new formulas, such as McClendon et. al in 1917 and Brujewicz (in Subow, 1931). Furthermore, everytime the International Atomic Weights were revised, the formulas were revised also. In 1940, for example,

Lyman and Fleming recalculated Dittmar�s figures using the 1938 Atomic Weights.

THE ORIGIN OF MARINE MIX

The Lilly Ackerland Fleischmann Memorial Aquarium at the Cincinnati Zoo opened on Friday, May 26, 1950. I moved to Cincinnati in June of 1950 and by 1951 I was a consultant to the Aquarium, mostly engaged in identifying species of newly-obtained fishes. The Aquarium was interested in displaying marine life - it was all freshwater at the time - and wanted to set up some experimental marine aquaria. However, at that time there were no artificial seawater mixes on the market, so I was asked to provide a formula. In early 1952 I turned over a copy of what I had come up with to the Chief Aquarist and we sat down for a lengthy discussion over the details. It was decided to go ahead with the experiments and

TABLE IV THE SCHWEITZER AND DITTMAR FORMULAS COMPARED The numbers in column (1) are the % of the salt used in the total of the salts in the sample. The total used in this comparison was the average ocean salinity or 3.5 grams per 100 grams of Water. The numbers in column (2) are in grams/100 grams of Water and they are obtained by multiplying the percentages in column (1) by 3.5; the numbers in column (3) are in pounds/100 lbs of water. Note they they are identical scales.
COMPONENT	(1) SALT %	(2) DITTMAR	(3) SCHWEITZER
Sodium Chloride	77.758	2.7215	2.7060
Magnesium chloride	10.878	0.3807	0.3666
Magnesium sulphate
Calcium sulphate
Potassium Chloride			0.0766
Potassium sulphate	2.465	0.0863
Magnesium bromide	0.217	0.0076	.0029
Calcium carbonate	0.345	0.0121	0.0030
TOTAL	100%

Page 13

so the Aquarium had a quantity of the mixture made up by a local commercial chemist, enough for about 100 gallons of water. In May of that year, however, I entered the United States Army (the Korean War still was in full tilt) so I was not around to see how things turned out. The following, however, is from the Cincinnati Enquirer of Sunday, November 16, 1952:

Something�s Latest addition to the Fleischmann Aquarium at the Cincinnati Zoo is a collection of tropical fish, some of which are pictured on these pages. The first of their kind to be shown in Cincinnati, these specimens were flown here from the gulf of Mexico. Special water prepared from a formula, is required to keep the marine inhabitants who are used to a saltier atmosphere than Cincinnati.

The caption under one of the pictures read: J.F. Heusser, Zoo Director, checks the temperature of the artificial sea water in one of the seven new exhibition tanks for tropical fish in the Fleischmann Aquarium.

After my tour of duty ended (I had been Commanding Officer of the 501st Chemical Depot Company), I spent a year in graduate school and returned to Cincinnati at the end of August 1955. I renewed my acquaintances with the Cincinnati Zoo and Fleischmann Aquarium staff and was able to examine the seven tanks that utilized the formula that I had provided. Everything was fine; the vertebrates, were in good health, as well as were the anemones and other marine creatures. The Lilly Ackerland Fleischmann Memorial Aquarium at the Cincinnati Zoo.

The Lilly Ackerland Fleischmann
Memorial Aquarium at the Cincinnati Zoo

Much to my surprise, I learned that during my absence the Aquarium had decided to sell the salt mixture to aquarists under the name of �Marine Mix.� This was done mainly because it was cheaper to manufacture in large quantities than small. Thus, in 1956 the first artificial seawater mix appeared on the market in this country. It was only on the market briefly since the Fleischmann Aquarium was satisfied that artificial seawater was feasible and so they planned to build an East Wing on the building to house small marine tanks. This took some time to plan, secure the necessary funds and build, but the addition was finally opened in June 1972. By this time, of course, there were a number of improved mixes on the market and the Aquarium did not need to manufacture its own as it did in the 1950s.

ADDENDUM:
THE COMPOSITION OF MARINE MIX

The element strontium, one of the major constituents of seawater, had not been isolated in Dittmar�s time. The usual wet-chemical analysis methods for calcium register any strontium present as an equivalent amount of calcium As a result, the strontium went unreported in the early analysis of seawater. I had briefly looked over several older formulas, but settled on that of Lyman and Fleming (1940) since they had included strontium. Their formula yields a water containing all of the major ionic components in their correct proportions, lacking only in the minor elements.

First, some biographical information. Dr. John Lyman (1915-1977) directed studies of the ocean for three agencies in the course of 20 years with the federal government. Born in Berkeley, Calif., Lyman did undergraduate work there and received his master's and doctorate degrees from Scripps Institution of Oceanography, part of the University of California in Los Angeles. He began his career in Washington in 1946, when he became director of oceanography for the Navy Hydrographic Office. He later became director of oceanography at the National Science Foundation and then chief adviser for oceanographic research at the Bureau

Page 14

Dr. John Lyman (1915-1977) in 1936

of Commercial Fisheries. Before going to Chapel Hill in 1968, he worked for the Office of Naval Research. At the time of his death, at age 62, Lyman was professor emeritus of environmental chemistry at the University of North Carolina, where he had moved in 1968 to head the school's office of marine sciences. He served in the Navy in World War II and was a member of the Washington Academy of Sciences and the American Association for the Advancement of Science.

Richard Howell Fleming (1909-1989), was a Canadian-born American oceanographer who conducted wide-ranging studies in the areas of chemical and biochemical oceanography, ocean currents (particularly those off the Pacific coast of Central America), and naval uses of oceanography. He joined the Scripps Institution in La Jolla, California, in 1931 and served as assistant director from 1946 to 1950, when he left to become professor of oceanography at the University of Washington. Fleming measured the amounts of chemical elements in sea water and in marine life. His work on ocean currents included measurement of tidal current velocities, description of upwelling of seawater, and observations of seasonal variations in currents. He also explored sedimentation in moving

and stagnant bodies of water. With H.U. Sverdrup and Martin W. Johnson, Fleming wrote The Oceans (1942).

Those aquarists not interested in an arithmetic exercise can safely skip this part of the Addendum and go directly to Column (2) of Table V.

Referring to Table V, lines 1-10 in Column (1) comprise the Lyman and Fleming formula; lines 11 -17 are my additions. The sum of lines 1-10 in Column (1) is 34.481 grams. This, by the way, rounded is a salinity of 34.5 (in grams of salts per 1000 grams of seawater). Subtracting this from 1000 grams (1 kilogram) we obtain 965.519 grams of water. (Items 11-17 were not included since they are in quantities too small to have any effect on these calculations.) Since 1 gallon of water = 3.79 kilograms, 965.519/3790 = 0.2547543 gallons. (At this stage in the computations I like to keep as many decimal places as possible in order to minimize rounding errors.)

Now I wanted a formula for 10 gallons of water, not 0.2547543 gallons. Therefore, I multiplied the lines 1-17 in Column (1) by 10/0.2547543 = 39.253508 and obtained the final formula for Marine Mix shown in Column (2). The salinity of this formulation was not affected and remained at 34.5. There has always been controversy over the relative merits of natural and synthetic seawater, especially with regard to trace or minor elements. Most synthetic seawaters have at some stage failed in

Richard Howard Fleming (1909-1989) in 1935

Page 15

certain respects, such as in maintaining a specific marine organism or plant, or in the breeding and rearing of the higher forms of marine life such as the fishes. A list of components in seawater of proven or suspected biological activity, in order of their concentrations in micrograms per liter (μ/l), is shown in Table VI. Copper, Barium and Nitrate (as NO3) are found in Cincinnati water. If I had such a list when I was putting together my formula for Marine Mix, I might have replaced the Barium, Aluminum and Copper with Silicon, Iron and Molybdenum.

BIBLIOGRAPHY

Anon., �A List of the Officers and Members of the Chemical Society of London. June 1843,� The Chemical Society of London, Richard and John E. Taylor, London, 1843.

Anon., �Scientific Pickings from The British Association Papers,� The Eclectic Magazine, p. 1071, December, 1855.

TABLE VI A LIST OF COMPONENTS IN SEAWATER OF PROVEN OR SUSPECTED BIOLOGICAL ACTIVITY.
COMPONENT	CONCENTRATION MICROGRAMS PER LITER (µ/l)
SILICON	500-2900
NITROGEN AS NO₃	500-700
IODINE	45-65
PHOSPHOROUS AS PO₄	1-90
BARIUM	6-90
IRON	1-40
MOLYBDENUM	9-16
ZINC	5-21
MANGANESE	0.4-10
VANADIUM	0.2-7
NICKLE	0.5-6.6
ARSENIC	2.1-2.7
COPPER	0.9-3
COBALT	0.1-0.7
NIOBIUM	o.015
YTTRIUM	0.01-0.03
VITAMIN B₁₂	0.001-0.5
GOLD	0.004-0.1

TABLE V
COMPONENTS AND CALCULATIONS FOR MARINE MIX

COMPONENT	(1) AMT. 1/KG	(2) AMT ADDED TO 10 GAL.
1. Sodium chloride, NaCl	23.476 grams	921.515 grams
2. Magnesium Chloride, MgCl₂	4.981	195.522
3. Sodium sulphate, Na₂SO₄	3.917	153.756
4. Calcium Chloride, CaCl₂	1.102	43.257
5. Potassium chloride, KCl	0.664	26.064
6. Sodium bicarbonate, NaHCO₃	0.192	7.367
7. Potassium bromide, KBr	0.096	3.768
8. Boric acid, H₃BO₃	0.026	1.021
9. Strontium chloride, SrCl₂	0.024	0.942
10. Sodium Flouride, NaF	0.003	0.118
11. Disodium phosphate, Na₂HPO₄	0.250 milligrams	9.813 milligrams
12. Potassium iodide, (KI)	0.077	3.023
13. Alum, (AlK(SO₄)	0.044	0.727
14. Zinc sulphate, ZnSO₄.7H₂O	0.015	0.589
15. Manganese chloride, MnCl₂.4H₂O	0.005	0.186
16. Barium chloride, BaCl₂.2H₂O	0.005	0.196
17. Copper sulphate, CuSo₄.5H₂O	0.002	0.076
18. Water	65.519 grams	10 gallons

Page 16

Dittmar, Wilhelm, �Report on Researches into the Composition of Ocean-Water, collected by H.M.S. Challenger, During the Years 1873-1876,� in Report on the Scientific Results of the Voyage of H.M.S. Challenger During the Years 1873-76, Physics and Chemistry - Vol I., Received January 1884.

Forchhammer, George, �On the Constitution of Sea-Water, at Different Depths, and in Different Latitudes,� Proceedings of the Royal Society of London, Vol. 12, pp. 129-132, 1862-1863.

Gosse, Philip H., �On keeping Marine Animals and Plants alive in unchanged Sea-water,� The Annals and Magazine of Natural History, Vol. X, Second Series, pp. 263-268, October, 1852.

Gosse, Philip H., �On Manufactured Seawater for the Aquarium,� The Annals and Magazine of Natural History, Vol. 14, Second Series, 65- 67, 1854.

Gosse, Philip H., �On Artificial Sea Water ,� Annals and Magazine of Natural History, Vol. XV, Second Series, No. 85, pp. 17-19, 1855.

Hibberd, Shirley, �The Book of the Aquarium and Water Cabinet,� Groombridge & Sons, London, pp. 59-60, 1856.

Lloyd, William Alford, �Sea-Water for Aquaria,� The Boy�s Own Paper, No. 49, Vol. II, p. 190, Saturday, December 20, 1879.

Lyman, John and Fleming, Richard Howell, �Composition of Sea water,� Journal of Marine Research, vol. 3, pp. 134-146, 1940.

Mallet, Robert, �Second Report upon the Action of Air and Water, whether fresh or salt, clear or foul, and at various temperatures, upon Cast Iron, Wrought Iron, and Steel,� Report of the Tenth Meeting of the British Association for the Advancement of Science held at Glasgow in August 1840, p. 223, John Murray, London, 1841.

McClendon, J. F., C. C. Gault, and S. Mulholland, �The hydrogen-ion concentration, CO2− tension, and CO2−content of sea water,� Carnegie Institute of Washington, Publication no. 251, Papers from the Department of Marine Biology, pp. 21-69, 1917.

Schweitzer, Edward G., �Analysis of Seawater as it exists in the English Channel near Brighton,� Philosophical Magazine and Journal, XV, pp. 51- 60, 1839

Schweitzer, Edward G., �Analysis of the Bonnington Water, near Leith, Scotland,� Memoirs and Proceedings of the Chemical Society of London, II, pp. 201-217, 1843.

Southall, William, �The Artificial Sea-Water at the Aston Aquarium,� Read before the Birmingham Natural History and Microscopial Society, September 23, 1879, The Midland Naturalist, Volume II, p. 246-247, 1879.

Subow, N. N., �Oceanographical tables. U.S.S.R.� , Oceanographic Institute, Hydrometeorological Commission, 208 pp., Moscow, 1931.

Warington, Robert, �Notice of Observations on the adjustment of the relationship between the Animal and Vegetable Kingdoms, by which the vital functions of both are permanently maintained.� A paper read before the assembled members of the Chemical Society on the evening of the 4th March, 1850.

Warington, Robert, �The Aquatic Plant Case, or Parlour Aquarium, �The Garden Companion, and Florists� Guide, William S. Orr and Co., Amen Corner, Paternoster Row, p. 7, January, 1852.

Warington, Robert, �On preserving the balance between the animal and vegetable organisms in sea -water,� The Annals and Magazine of Natural History, Second Series, Vol. 12, No. 71, pp. 319-324, 1853.

Warington, Robert, �On Artificial Sea Water,� The Annals and Magazine of Natural History, Vol. 14, No. 84, pp. 419-421, 1854.

Wilson, George, �On the Artificial Preparation of Sea-water for Marine Aquaria,� Report of the Twenty-Fourth Meeting of the British Association for the Advancement of Science, Notices and Abstracts of Miscellaneous Communications to the Sections, p. 77, Liverpool, September, 1854.