Kali. 



Denne Bestemmelse, der kræver lang Tid og oin- 

 stændeligt Arbeide, har jeg som for nævnt kun udført i 

 nogle faa Vandprøver. Hertil har jeg anvendt omtrent 50 

 Gram Søvand og fraskilt Kalk og Magnesia efter den af 

 Glassen angivne Methode ' med oxalsur Ammoniak. Alkohol 

 og Eddikesyre. Filtratet blev inddampet til Tørhed. Am- 

 moniaksalteme bortjagede og den tilbageblevne Saltmasse 

 befriet fra Svovlsyre ved gjentagne Glodninger med et Over- 

 skud a f Salmiak. De samlede Klorider blev derpaa opløste 

 i Vand. overførte i Platindobbeltsalte og derpaa behandlede 

 efter den Methode. der af Fresenius foreskrives til Bestem- 

 melsen af Kali og Natron i Mineralvand. Det er imidler- 

 tid meget vanskelig! at faa al Svovlsyre tjernet ved Af- 

 dampning med Salmiak. og det fremstillede Kaliumplatin- 

 klorid indeholdt derfor altid Spor af svovlsure Salte. Der 

 er saaledes en Muligbed for. at de fundne Kalimængder er 

 noget for høie. skjønt Bestemmelserne i de forskjellige 

 Vandprøver stemmer nogenlunde godt overens med hin- 

 anden. 



Hvis man iforveien veier de samlede Klorider og der- 

 fra trækker den erholdte Klorkaliummængde. skulde man 

 altsaa kuune tinde Søvandets Klornatriumgehalt. Men da 

 Kloriderne ialmindelighecl ere forurensede med smaa Mæng- 

 der Svovlsyre og Magnesia, der ikke er blevet fuldstændig 

 fjernede ved de foregaaende Opera tioner. og Kulrester fra 

 Ammoniaksaltenes Afdampning. vil 'man ved denne indirekte 

 Bestemmelse faa altfor hoie og uoverensstemmende Tal for 

 Klornatrium. Man vil paa denne Maade ialmindelighed 

 finde omkring 2.8 pGt. af dette Salt. medens man efter en 

 anden Beregning, som senere skal omtales, neppe faar et 

 Tal der overskrider 2.7 pGt. 



Klor. 



Med Hensyn til disse Bestemmelser, der er udforte 

 ombord ved Titrering. henviser jeg til Hr. Tornøe's Af- 

 handling om Saltmængden i Havet. For at bestemme 

 Våndets samlede Saltgehalt har Hr. Tornøe inddampet en' 

 vis Portion deraf til Tørhed. glødet Residuet i nogle Mi- 

 nuter over en Bunsen's Lampe og derpaa ved Titrering 

 korrigéret tur den tabte Saltsyre. Efterat have udført disse 

 Bestemmelser i nogle Vandprøver, har hau paa Grundlag 



fordel for deu svovlsure Baryts Opløselighed, em den end er for ube- 

 tydelig til at kunne tåges med i Betragtning, naar den kun refererer 

 sig til tu Forsøg. 



I tru svovlsure Baryts Opløselighed i koldt Søvaild er in 

 paaviselig. Ved Tilsætning af en fortyndet Clorbariuraopløsning frem- 

 kommer her Bundfaldet ligesaa hurtig som i enOpløsning af svovlsur 

 Kali i rent Vand. 



Rlormagnesium forekommer i Søvandet i altfor ringe Msengde 



til at kunne have nogen skadelig Indflydelse paa Svovlsyrebestemmel- 



Jeg har desuden ved Forsøg overbevist mig om, at 9eh en 



men koncentreret Opløsning ai dette Salt ikke har nogeu mserkelig" 



Dndvirkning paa I dfældningen af den svovlsure Baryt. 



1 Zeitschrifl fur anal, Chemie 1879, li. I. S. 374 



Potash. 



To determine this comi)ound is a long and laborious 

 ttisk. and. asprevionsly stated. T have performed it in only a 

 few of the samples of sea-water. For this purpose, I pre- 

 cipitated, by Glassen's method. 1 with oxalate of ammonia, 

 alcohol. and acetic. acid. the lime and magnesia in 50'"' of 

 sea-water. The filtered solution was evaporated to dryness, 

 the ammonia compounds volatilized, and the sulphuric acid 

 in the residue got rid of by repeatedly beating to redness, 

 with sal-ammoniac in excess. The several chlorides were 

 then dissolved in water, transformed into double chloride 

 of platinum. and then treated according to the method 

 devised by Fresenius for determining potash and soda in 

 mineral waters. Meanwhile. it is exceedingiv difticult to 

 get rid of all the sulphuric acid by evaporation with sal- 

 ammoniac ; and hence the double chloride of potassium and 

 platinum invariably exhibited traces of sulphates. The 

 amounts found for potash may therefore be a little too 

 high. though the determinations performed with different 

 samples of water agree pretty closely inter se. 



Now. by weighing beforehand the several chlorides, 

 and then subtracting the total a mo unt of chloride of 

 potassium. the proportion of chloride of sodium present in 

 sea-water might accordingly be found. But. unfortunately. 

 the chlorides are as a rule contaminated with small quan- 

 tities of sulphuric acid and magnesia, that were not got 

 rid of in the foregoing operations, along with a residue of 

 carbonized organic matter resulting from the volatilization of 

 the sal-ammoniac : and hence the figures thus indirectly coin- 

 puted for chloride of sodium will be alike incongruous and 

 too high. This method will generally give about 2.S per 

 cent of the salt. whereas the amount by another mode of 

 computation — to be afterwards noticed — does not ex- 

 ceed 2.7 per cent. 



Chlorine. 



As regards these determinations. which were pei*- 

 formed on board by titration, the reader is referred to 

 Mr. Tomoes Memoir on the amount of salt in sea- 

 water. To determine the total amount of salt in ocean- 

 water, Tornøe evaporated a given quantity to dryness. 

 kept the residue for a few minutes at a red-heat over one 

 of Bimsems lamps. and then determined by titration the 

 loss* of hydrochloric acid. After performing these deter- 



of baryta's solubility, though too inconsiderabie to be tak. a, into ac- 

 count, referrihg as it does to only two determinations. 



That sulphate of baryta is also aoluble in cold sea-water, will 

 hardly admjt of proof. On adding a diluted solution of chloride of 

 barium, the precipitate fonns as rapidlj as in a solution of sulphate 

 of potash prepared with pure water. 



Chloride of magnesium occurs far ton spjaringlj in sea-water 



to admit of its disturbing the aecuracy of sulphuric acid determina- 



Besides. experiments have convinced me. that a comparatively 



concentrated Bolution of that salt does not appreciably affect the pre- 



cipitation of sulphate of baryta. 



1 Zeitschrift fur anal. Chemie 1870, 11, 4, pa.-;-. 374. 



