506 The American Naturalist. [June, 
the bottle consists of a brass wire (A) attached to a weight (B) — 
from which strips of brass extend upwards terminating in clips 
(C). These brass strips have considerable spring, and serve to 
hold the bottle in place. The frame is supported by the spring 
(F) attached to the sinking rope (E). A flexible cord (G) extends 
from the top of the spring (E) to the stopper (H) of the bottle. 
The length of this cord and the length and stiffness of the 
spring are so adjusted that when the apparatus is suspended 
in the water the flexible cord will bea little slack. In this 
condition it is lowered to the required depth. A sudded jerk 
given to the sinking rope stretehes the spring, produces tension 
on the flexible cord, and pulls out the stopper. A slack safety 
cord (not shown in the figure) extends from E to J to prevent 
too great a stretching of the spring and to guard against loss 
of apparatus in case the spring should break. For great depths, 
where the pressure is too great to allow of the stopper being re- 
moved, a smaller aperture is used. 
The temperature of each sample is iat as at the time of 
collection, and additional series of observations upon the tem- 
perature of the water at various depth are carried on, using for 
this purpose the newly invented Thermophone. These obser- 
vations give useful information regarding the circulation of 
the water. When the samples reach the laboratory, record is 
made of the color (using the platinum-cobalt standard in a 
colorimeter of special design), transparency, amount of sedi-. 
ment, taste, and odor. 
Environmental studies are carried on by personal inspection ~ 
of the watershed, and the collection of numerous extra samples 
from the small brooks and feeders. These inspection trips are 
made whenever anything abnormal appears in the condition 
of the water. 
2. Microseopical.—The microscopical work is the most import- - 
ant part of the laboratory routine. It consists chiefly in the 
quantitative determination of the various micro-organisms 
(except the bacteria) in each sample of water by the Sedgwick- 
Rafter method. At the present time this is carried on as fol- 
lows: 500 c.c. of the water to be examined is filtered through 
a thin layer of quartz sand supported at the bottom of a cylin- 
