CLYDE SEA AREA. 691 



Table XXII. — Quantitative Relation between Rainfall and Salinity. 



Landward Division. 



Jan. 



Feb. 



Mar. 



April. 



May. 



June. 



July. 



Aug. 



Sept. 



Rainfall, 1886, . . . 



6-25 



2-42 



3-51 



2-44 



319 



1-65 



3-28 



2-69 



6-11 



„ 1887, . . . 



4-75 



373 



2-49 



2-59 



1-17 



1-34 



3-89 



3-76 



4-02 



Difference 1886-87, . . 



1-50 



— i*3i 



I '02 



-OT5 



2'02 



0-31 



-0"6i 



— 1 -07 



2-09 



Surface Density, 1886, . 





• • • 





2145 



2240 



2340 



2365 



2390 



2385 



„ „ 1887, . 





• • • 



... 



2170 



2265 



2315 



2365 



2365 



2365 



Difference 1886-87, . . 







... 



—0025 



—OO25 



0025 



0000 



0025 



0020 



Bottom Density, 1886, . 







... 



2455 



2455 



2460 



2470- 



2475 



- 2470 



„ „ 1887, . 









2430 



2450 



2465 



2465 



2465 



2465 



Difference 1886-87, . . 









0025 



0005 



-0005 



0005 



OOIO 



0005 



Seaward Division. 





















Rainfall, 1886, . . . 



4-32 



2-22 



2-95 



1-89 



3-13 



137 



2-84 



2-36 



4-41 



„ 1887, . . . 



3-66 



2-61 



1-82 



2-74 



1-86 



0-94 



2-84 



2-38 



4-09 



Difference 1886-87, . . 



0-66 



-o-39 



ri3 



-0-85 



1-27 



0'43 



O'OO 



— 0'02 



0-32 



Surface Density, 1886, . 



• . . 





... 



2365 



2410 



2470 



2490 



2495 



2460 



,, „ 1887, . 









2450 



2460 



2480 



2480 



2475 



2465 



Difference 1886-87, . . 







... 



-0095 



— 0050 



— OOIO 



OOIO 



0020 



—OO05 



Bottom Density, 1886, . 









2510 



2515 



2515 



2520 



2520 



2515 



1887, . 









2485 



2505 



2530 



2520 



2520 



2520 



Difference 1886-87, . . 









0025 



OOIO 



-0015 



OOOO 



OOOO 



— OO05 



The surface landward curve shows a regular rise and fall of salinity two months after a 

 fall or rise in the rainfall, the bottom landward curve shows a similar effect, but the period 

 is retarded two months in April and lengthens to four months in September. 



The surface seaward curve follows at an interval of about four months, and the 

 bottom seaward curve at an interval of three months at first and four months latterly. 

 This generally confirms the deductions drawn from consideration of the general curves 

 for the whole period, and extends it, showing that when evaporation is at a minimum 

 bottom water is most rapidly affected by rainfall, when evaporation is at a maximum 

 least rapidly ; and that the period is somewhat greater for the seaward than for the land- 

 ward divisions. As to quantitative relations, it appears unsafe to draw any conclusion 

 from the surface data ; and by considering the data for bottom water, the most probable 

 guess seems to be that an increase of 2 inches of rainfall per month would cause a decrease 

 of density of (V00025 within two months. If this be correct, it would follow that an 

 increase of 200 inches of rainfall would reduce the density by 0*02500, or completely 

 freshen the water in two months. It is obvious that to completely freshen the water of 

 the area, at least 25 cubic miles of rain water would be necessary, that being the total 

 volume of the area. If the area were pumped dry of salt water and the fresh water 

 allowed to flow in, this quantity would suffice. But our hypothesis is that the fresh 

 water drives out the salt water and takes its place ; and as this must occur by successive 

 dilution, it is evident that more than the minimum quantity of water is necessary. A 

 fall of 200 inches over 3500 square area miles amounts to only 10 cubic sea miles, so that 



