E. W. Broivn — Apparatus for Tidal Analysis. 387 



a quarter of an inch apart. This size permits two figures to 

 be written in each square witli no parts of the figures project- 

 ing over the ruled lines."* Its width is to be 19" (72 squares 

 -f an inch overlap) and height at least 8" (32 squares). 



The cardboard should be fairly flexible so that if bent into 

 an arch whose height is about one-sixth of the base, it will not 

 tend to break and will return to its original form when released. 

 The height of the card should be about 12", its width rather 

 less than 9". The latter measurement is to be such that when 

 two sheets of the ruled paper are folded closely over it one 

 edge of the outer sheet shall come accurately over the ruled 

 line on the sheet 18" from that edge. 



A cover to the cardboard is made by folding a sheet of the 

 same kind of paper (ruling is unnecessary) closely over it and 

 pasting the edges together, care being taken that the cover 

 does not stick to the cardboard. If the latter be bent a little 

 the cover can easily be slipped off and on. 



A convenient paper cutter is that used to trim photographs ; 

 it must be large enough to make a 9" cut. 



Four brads are driven into a board so as to form a rectangle 

 about 8" x 6". They should be a little inclined inwards along 

 the direction of the 8" sides of the rectangle. The double- 

 pointed tacks are partly driven in close to the brads with their 

 lengths in the same direction, so that when the sheet of card- 

 board is bent and the edges placed between the pairs of brads 

 it will remain bent and will be slightly raised above the board. 

 Each brad and tack may be replaced by a small wood step 

 nailed to the board. 



In Darwin's scheme for the analysis of a year's observations, 

 hourly heights are used. He also suggested that such units 

 should be adopted (e. g. tenths of a foot or inches) that all 

 heights could be expressed by two digits. It is convenient to 

 describe the use of the apparatus on this basis, although there 

 is room for four digits if necessary. The 24 observations for 

 the first day (day 0) are written in every third square of the 

 top line of the ruled paper beginning with the third square 

 from the left and ending with the 72d. The second and suc- 

 ceeding days are similarly written in the following lines up to 

 the end of the first block, which, for the solar tides, contains 30 

 days. At the end of several of the blocks one day of observa- 

 tions is not used for these tides : it must, however, be inserted. 

 Thus 12 sheets contain all the observations; these may be 

 written in, as they are measured from the tide curve. They 

 are then summed according to Darwin's published instructions, 



* I have found two figures between ruled lines a better guide to the eye 

 than one figure between closer rulings, especially when dealing with numbers 

 of several digits. 



