HYPOTHESIS FOR ORIGIN OF GYPSUM DEPOSITS 225 



of the Laramie, using all of the SO4 in CaSO^, gives oo.S parts CaSO^ 

 per million. Eecalculation of an analysis of water taken TO miles below 

 that given above and "more or less contaminated by seepage from the 

 canals/' using all of the Ca and 36/37 of the SO^, gives CaSO^ about 

 220 parts per million. ^^ Eecalculation of the Popo Agie analyses, using 

 all of the SO4 in CaSO^, gives 8.67 parts per million.^® Recalculation 

 of the mean of 29 composites of the North Platte, at North Platte,-'' 

 Nebraska, using all of the SO^ in CaSO^, gives a calcium sulphate con- 

 tent of about 198 per million.-^ (The North Platte analyses are used 

 because the river flows almost entirely on sedimentary beds across Wy- 

 oming and Nebraska.) 



All of these waters are high in calcium carbonate, and in only one 

 analysis is there more than enough calcium to unite with the CO3, but in 

 most cases this union takes place early in the process of concentration 

 and limestone is precipitated. Most of the sulphuric ions would prob- 

 ably unite with sodium and magnesium, as they have done in the pre- 

 cipitates formed in the alkaline lakes of Wyoming, where sodium sulphate 

 and magnesium sulphate make up most of the deposits and calcium sul- 

 phate is present in small amounts.^- The calculated amount of CaSo^ is 

 then much too high, as all of the sulphuric ions should not be used with 

 calcium. From the data available it seems that 50 parts of calcium 

 sulphate per million is more than would precipitate from waters that 

 have run some distance, at low grade, through arid regions, and that the 

 amount furnished by streams as they emerge from regions of high relief 

 would be less than 10 parts per million. 



Analyses of six streams tributary to Great Salt Lake show an average 

 calcium siilpliate content of 120 parts per million, if all of tlio sulphuric 

 ions are used with calcium; but in all but one of these analyses there is 

 not enough calcium to combine with all of the CO3 ions. All of the 

 calcium luis been precipitated fi-oni the lake, much of it as calcium cai'- 

 bonalc. As with the W'vorninu- ri\crs. it seems ])i'()l)al)le that .")0 parts 

 per million calcium sulphate in the water is more than would he pre- 

 cipitatiMl on e\aporation. 



ASSUMPTIONS MADE IN ^VORKINO OUT HYPOTHESIS 



Assumptions foi- origin of 1-foot bed of gypsum over 1(1.00(1 square 

 miles, but with the deposit ion in nianv. nioi'c or less isolated, lakes: 



'* Runoff al)Out 10 i)er cent. 



'" 'riie run-off probably 50 to 70 per cent, but rolialjle data is n(jt availulile. 



-" Kun-offi less than 10 per cent. Data not satisfactory. 



■-■'Clarke: TT. S. Geological Survey, Bull. 4f)l, p. 74. 



•^Wyoming E.xix^rluieut Station, Bull. 40, pp. 110-121. 



