10 CCELENTERATA AND 



glass jar or finger bowl so that the light passes through it, 

 small animals can be detected through the sides. Most of 

 the small animals seek that side of the vessel on the surface 

 turned to the light and one can easily find them there. 

 Small glass finger-bowls in which larvae can be raised, can 

 be examined by holding them between the light and the 

 eye. The animals may be picked out by pipettes or tubes. 

 The water in which the animals are first placed, if crowded 

 with life, soon becomes vitiated. When few animals are 

 found they can be left in the pan in the same water in which 

 they were captured. It is a- good plan to add in such cases 

 some pure water, and keep in the pan small genera of 

 bright green algae. 



b. Collecting surface animals by observation in tJie 

 water. 



Although the dip and the drag-net yield the best re- 

 sults, it is often necessary to see the animals in their na- 

 tive habitat, in order to pick out what is wanted. The 

 surface is often so crowded with Salpae, for instance, that 

 the net gets clogged with them, and a person in search of 

 anything else cannot use the net to advantage. 



If the sea is very smooth, very small animals can be de- 

 tected by the eye from the boat. I have used a water-glass 

 with advantage. The fishermen in Villa Franca, southern 

 France, carry a bottle of oil in the boat and use oil to 

 quiet the surface. A blackened plate of tin, lowered in 

 the water, renders it possible to detect very small animals 

 in the water above it. When once detected, it is not dif- 

 ficult to capture the animal with glass dishes or hand nets. 



c. Places for collecting surface animals. 

 The best localities must be learned from observation. 

 Tide eddies are favorable points, and the water in the vi- 

 cinity of floating masses of seaweed is sometimes crowded 



