10 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 139 



or, since R d = o.S, 



X = 2.3 — O.I + logyf-. 



A plot of x vs. D may now be made, where x is calculated from 

 stations at which hydrophotometer readings which give k and Secchi 

 Disc readings which give D were taken simultaneously. This plot 

 shows a marked variation of U w as the Secchi Disc reading is 

 changed, and is the graph which was used to determine unknown U w 

 values when the S.D. readings were known, both for stations which 

 had hydrophotometer and Secchi Disc readings and for those which 

 had only S.D. data. 



By means of this methodology, then, it was possible to calculate 

 equivalent Secchi Disc readings for each hydrophotometer reading 

 taken. 



DISCUSSION OF DATA 



In the two appended tables, all the data taken on the Elsie Fenimore 

 are tabulated. Table I includes the hydrophotometer and Secchi Disc 

 data presented by seasons and in geographical order from North to 

 South. Winter is considered to include the months of January, Feb- 

 ruary, and March ; spring — April, May, and June ; summer — July, 

 August, and September ; and fall — October, November, and Decem- 

 ber. The various stations may be easily located by number on the 

 series of charts (figs. 1-13, preceding the tables), which show the 

 latitude and longitude of each of the stations mentioned. 



Table 2 includes all the other data taken, utilizing the various de- 

 vices of Dyer plus a few others which were also used. These data 

 are presented in simple geographical order, proceeding from north 

 to south. 



The data as a whole, although being among the most extensive 

 available at the present time, have many limitations and shortcomings, 

 and these should be kept in mind while any attempt at utilization is 

 being made. 



The hydrophotometer readings were taken with utmost care. How- 

 ever, the calibration in air was apparently not standardized, the ad- 

 justment varying from 92 to 96% T in air instead of 92 percent as 

 previously mentioned. This would have the effect of making all read- 

 ings above 90 percent highly suspect since a small change in %T at 

 this end of the scale is associated with a large change in the Secchi 

 Disc reading. 



This is probably also the reason for the significant number of read- 

 ings which are above 100 percent, and hence change from quantitative 

 readings to qualitative. This 92 percent reading in air as being the 



