22 University of California Publications in Zoology [Vol.22 



The most probable explanation of the variation shown is that the 

 incoming tide carried some of the polluted Stockton Channel water up 

 stream when the natural stream flow was very weak. No other tests 

 of the chemical composition of water at any station are available. 



Measurement op Volume 



In order to measure the mass of material secured in the collections, 

 a Bausch and Lomb hand centrifuge was used. The sedimentation 

 tubes were graduated to tenths of a cubic centimeter. This machine 

 makes 23 revolutions to each turn of the crank. After a few trials 

 it was decided to run at 54 turns per minute. While no amount 

 of practice made uniform speed possible, the average was kept at 

 about 1300 revolutions per minute. The time of running was four 

 minutes. Experience indicated that this was a little longer time than 

 necessary, but it was undertaken with the intention of compensating 

 for any possible inaccuracy due to variation in rate. Since this indi- 

 cated only the mass of material held by the net, it was necessary to 

 make some supplementary test to determine as nearly as possible what 

 volume went through. Only two or three such tests were made because 

 of the pressure of other duties, but the indication from filter paper 

 tests is that the net sometimes retains only one-tenth of the mass of 

 material actually present in the water. This material is often com- 

 posed principally of silt. 



The Enumeration 



After considerable experiment, the count of the organisms in the 

 catches was begun in September, 1914. Since the time available was 

 nearly always at night, almost all the counting was done by artificial 

 light from an ordinary 60-watt, frosted, incandescent globe. Only 

 rarely was it possible to make two counts in one day. Hence the enum- 

 eration was not finished until the night of June IS, 1915. 



APPARATUS 



The microscope used throughout was a Spencer with 16 millimeter 

 objective and 8x ocular. It was equipped with a quick screw sub- 

 stage and a Spencer mechanical stage. The Whipple ocular micro- 

 meter and the Rafter counting cell were used. Its largest square cov- 

 ered 1.1 square millimeters with the above lens equipment, so that the 

 labor of calculating was slightly increased by the fractional area. 



