43 



anotlier composition. The coating slionld be made of a 

 good strong mortar, about one part of lime to two of 

 sand, and after it is apj)lied it should be allowed to set 

 for several days under cover before the prepared slates 

 are made up into series in frames of various sizes and 

 put into the water. Cement alone does not answer the 

 purpose, as it becomes so extremely hard in the course 

 of time that it cannot be readily removed from the sur- 

 faces to which it has been apj)lied, which of course ren- 

 ders it practically useless. The slates after they are 

 prepared were fastened into frames about two feet high, 

 the ends being let into notches about an inch or so 

 apart. In this way a dozen could be made up into one 

 frame, which held them horizontally at a good distance 

 from the bottom and far enough opart to allow the water 

 to pass readily through between the slates, giving the 

 floating embryos a good chance to ii:K: themselves to the 

 mortared surfaces. Other frames were made of an oblong 

 shape, one foot wide and six or eight feet long, the edges of 

 the long side pieces were deex)ly notched with the saw 

 every three or four inches apart to receive the one edge 

 of a slate, which was held in a vertical position when the 

 frame was laid down horizontally. Some of these frames 

 would hold twelve or fourteen slates in position at once. 

 In other cases the slates were simply coated with mortar 

 and laid in rows horizontally upon parallel poles sup- 

 ported above the bottom a few inches. In otlier cases 

 again the slates were suspended upon galvanized wire 

 by the four corners, in a tier of 4 or 5, about an inch 

 a,part. I also suspended them singly by means of gal- 

 vanized wire. I did not however obtain the same results 

 from all of the apparatus. The conditions were such 

 however in most cases as to make it extremely difficult 

 to make a fair comparison. Where tlie slates Avere in a 

 horizontal jjosition more mud seemed to be deposited on 

 the upper side whicli interfered with the attachment of 

 the spat on that side, while the comparative shade below 



