p. Groom 31 



latter doubles the deviation the number recorded in the subjoined 

 Table E must be halved in order to give the true facts. The measure- 

 ments recorded from June 21 to July 7 relate to the board during the 

 absorption of water, those subsequent to July 17 to the drying board. 



The diagonal yx . yh runs rather across the grain exposed at the 

 surface and consequently underwent greater elongation and shortening 

 than did y . yxb, which tends more towards parallelism with that grain. 

 The lengths of the diagonals measured were approximately 56 centi- 

 metres. 



In Table E: 



Column 1 represents the time of observation. 



Column 2 records the amoimt of water absorbed subsequent to the 

 immediately preceding period. 



Columns 3 and 7 record the nature of the curve, whether concave (c), 

 or convex (v), or straight (.s) of the whole board along the diagonal 

 and the number of millimetres that the middle is sunk or raised above 

 the true base line as recorded by the instrument. 



