INTUOIUKTION. 17 



should be made through tlie sail-cloth close to the 

 ring. Through these holes stout cords are passed, and 

 the nets are towed at a distance of a few yards behind 



e/ 



an ordinary rowing-boat. The speed of the boat should 

 not be more than three or four miles an hour, and the 

 net normally takes up a position a foot or eighteen 

 inches below the surface of the water. The net should 

 lie towed for three-quarters of an hour or an hour at a 

 time through the surface-waters of the deepest parts of 

 the lake. It is then drawn in, the water is allowed to 

 drain away, and the sediment is transferred to bottles. 



Sometimes the material can be cleaned by a system 

 of washing and decanting, that is to say, all the 

 heavier material can be removed. Unless, however, the 

 material is very dirty and sandy, we do not recommend 

 cleaning it. If a person wishes to make a real study of 

 Desmids he will do better to examine his collections 

 just as they have been gathered ; he will then become 

 acquainted with the nature of the material among 

 which they occur, and will soon obtain a good idea of 

 what to expect in his collections. 



Nice clean material may be obtained by placing the 

 collections in flat dishes and exposing them to a fairly 

 good light. The Desmids appear at the surface in 

 small gelatinous masses and with care may be removed 

 by means of a small dipping-tube. Another method 

 is to decant off most of the water and cover the sedi- 

 ment with a piece of undyed sieving silk. If the silk 

 is sufficiently coarse the Desmids will make their way 

 through the meshes and can be removed. Such 

 material always contains numerous living Diatoms. 



Living material should always be examined as care- 

 fully as possible, and then it should be preserved for 

 future examination. 



There are several methods of preserving Desmids 

 for subsequent examination. For examining the struc- 

 ture of the cell-contents a 3-4 per cent, solution of 

 formalin is best, but for bring-in s; out the cell-outline 



O o 



and the structure of the cell- wall we find nothing to 



